diff options
1267 files changed, 32201 insertions, 15582 deletions
@@ -339,7 +339,7 @@ W: http://tomas.nocrew.org/ D: dsp56k device driver N: Ross Biro -E: bir7@leland.Stanford.Edu +E: ross.biro@gmail.com D: Original author of the Linux networking code N: Anton Blanchard diff --git a/Documentation/00-INDEX b/Documentation/00-INDEX index 72dc90f8f4a..8de8a01a247 100644 --- a/Documentation/00-INDEX +++ b/Documentation/00-INDEX @@ -12,8 +12,6 @@ Following translations are available on the WWW: 00-INDEX - this file. -BK-usage/ - - directory with info on BitKeeper. BUG-HUNTING - brute force method of doing binary search of patches to find bug. Changes diff --git a/Documentation/BK-usage/00-INDEX b/Documentation/BK-usage/00-INDEX deleted file mode 100644 index 82768784ea5..00000000000 --- a/Documentation/BK-usage/00-INDEX +++ /dev/null @@ -1,51 +0,0 @@ -bk-kernel-howto.txt: Description of kernel workflow under BitKeeper - -bk-make-sum: Create summary of changesets in one repository and not -another, typically in preparation to be sent to an upstream maintainer. -Typical usage: - cd my-updated-repo - bk-make-sum ~/repo/original-repo - mv /tmp/linus.txt ../original-repo.txt - -bksend: Create readable text output containing summary of changes, GNU -patch of the changes, and BK metadata of changes (as needed for proper -importing into BitKeeper by an upstream maintainer). This output is -suitable for emailing BitKeeper changes. The recipient of this output -may pipe it directly to 'bk receive'. - -bz64wrap: helper script. Uncompressed input is piped to this script, -which compresses its input, and then outputs the uu-/base64-encoded -version of the compressed input. - -cpcset: Copy changeset between unrelated repositories. -Attempts to preserve changeset user, user address, description, in -addition to the changeset (the patch) itself. -Typical usage: - cd my-updated-repo - bk changes # looking for a changeset... - cpcset 1.1511 . ../another-repo - -csets-to-patches: Produces a delta of two BK repositories, in the form -of individual files, each containing a single cset as a GNU patch. -Output is several files, each with the filename "/tmp/rev-$REV.patch" -Typical usage: - cd my-updated-repo - bk changes -L ~/repo/original-repo 2>&1 | \ - perl csets-to-patches - -cset-to-linus: Produces a delta of two BK repositories, in the form of -changeset descriptions, with 'diffstat' output created for each -individual changset. -Typical usage: - cd my-updated-repo - bk changes -L ~/repo/original-repo 2>&1 | \ - perl cset-to-linus > summary.txt - -gcapatch: Generates patch containing changes in local repository. -Typical usage: - cd my-updated-repo - gcapatch > foo.patch - -unbz64wrap: Reverse an encoded, compressed data stream created by -bz64wrap into an uncompressed, typically text/plain output. - diff --git a/Documentation/BK-usage/bk-kernel-howto.txt b/Documentation/BK-usage/bk-kernel-howto.txt deleted file mode 100644 index b7b9075d291..00000000000 --- a/Documentation/BK-usage/bk-kernel-howto.txt +++ /dev/null @@ -1,283 +0,0 @@ - - Doing the BK Thing, Penguin-Style - - - - -This set of notes is intended mainly for kernel developers, occasional -or full-time, but sysadmins and power users may find parts of it useful -as well. It assumes at least a basic familiarity with CVS, both at a -user level (use on the cmd line) and at a higher level (client-server model). -Due to the author's background, an operation may be described in terms -of CVS, or in terms of how that operation differs from CVS. - -This is -not- intended to be BitKeeper documentation. Always run -"bk help <command>" or in X "bk helptool <command>" for reference -documentation. - - -BitKeeper Concepts ------------------- - -In the true nature of the Internet itself, BitKeeper is a distributed -system. When applied to revision control, this means doing away with -client-server, and changing to a parent-child model... essentially -peer-to-peer. On the developer's end, this also represents a -fundamental disruption in the standard workflow of changes, commits, -and merges. You will need to take a few minutes to think about -how to best work under BitKeeper, and re-optimize things a bit. -In some sense it is a bit radical, because it might described as -tossing changes out into a maelstrom and having them magically -land at the right destination... but I'm getting ahead of myself. - -Let's start with this progression: -Each BitKeeper source tree on disk is a repository unto itself. -Each repository has a parent (except the root/original, of course). -Each repository contains a set of a changesets ("csets"). -Each cset is one or more changed files, bundled together. - -Each tree is a repository, so all changes are checked into the local -tree. When a change is checked in, all modified files are grouped -into a logical unit, the changeset. Internally, BK links these -changesets in a tree, representing various converging and diverging -lines of development. These changesets are the bread and butter of -the BK system. - -After the concept of changesets, the next thing you need to get used -to is having multiple copies of source trees lying around. This -really- -takes some getting used to, for some people. Separate source trees -are the means in BitKeeper by which you delineate parallel lines -of development, both minor and major. What would be branches in -CVS become separate source trees, or "clones" in BitKeeper [heh, -or Star Wars] terminology. - -Clones and changesets are the tools from which most of the power of -BitKeeper is derived. As mentioned earlier, each clone has a parent, -the tree used as the source when the new clone was created. In a -CVS-like setup, the parent would be a remote server on the Internet, -and the child is your local clone of that tree. - -Once you have established a common baseline between two source trees -- -a common parent -- then you can merge changesets between those two -trees with ease. Merging changes into a tree is called a "pull", and -is analagous to 'cvs update'. A pull downloads all the changesets in -the remote tree you do not have, and merges them. Sending changes in -one tree to another tree is called a "push". Push sends all changes -in the local tree the remote does not yet have, and merges them. - -From these concepts come some initial command examples: - -1) bk clone -q http://linux.bkbits.net/linux-2.5 linus-2.5 -Download a 2.5 stock kernel tree, naming it "linus-2.5" in the local dir. -The "-q" disables listing every single file as it is downloaded. - -2) bk clone -ql linus-2.5 alpha-2.5 -Create a separate source tree for the Alpha AXP architecture. -The "-l" uses hard links instead of copying data, since both trees are -on the local disk. You can also replace the above with "bk lclone -q ..." - -You only clone a tree -once-. After cloning the tree lives a long time -on disk, being updating by pushes and pulls. - -3) cd alpha-2.5 ; bk pull http://gkernel.bkbits.net/alpha-2.5 -Download changes in "alpha-2.5" repository which are not present -in the local repository, and merge them into the source tree. - -4) bk -r co -q -Because every tree is a repository, files must be checked out before -they will be in their standard places in the source tree. - -5) bk vi fs/inode.c # example change... - bk citool # checkin, using X tool - bk push bk://gkernel@bkbits.net/alpha-2.5 # upload change -Typical example of a BK sequence that would replace the analagous CVS -situation, - vi fs/inode.c - cvs commit - -As this is just supposed to be a quick BK intro, for more in-depth -tutorials, live working demos, and docs, see http://www.bitkeeper.com/ - - - -BK and Kernel Development Workflow ----------------------------------- -Currently the latest 2.5 tree is available via "bk clone $URL" -and "bk pull $URL" at http://linux.bkbits.net/linux-2.5 -This should change in a few weeks to a kernel.org URL. - - -A big part of using BitKeeper is organizing the various trees you have -on your local disk, and organizing the flow of changes among those -trees, and remote trees. If one were to graph the relationships between -a desired BK setup, you are likely to see a few-many-few graph, like -this: - - linux-2.5 - | - merge-to-linus-2.5 - / | | - / | | - vm-hacks bugfixes filesys personal-hacks - \ | | / - \ | | / - \ | | / - testing-and-validation - -Since a "bk push" sends all changes not in the target tree, and -since a "bk pull" receives all changes not in the source tree, you want -to make sure you are only pushing specific changes to the desired tree, -not all changes from "peer parent" trees. For example, pushing a change -from the testing-and-validation tree would probably be a bad idea, -because it will push all changes from vm-hacks, bugfixes, filesys, and -personal-hacks trees into the target tree. - -One would typically work on only one "theme" at a time, either -vm-hacks or bugfixes or filesys, keeping those changes isolated in -their own tree during development, and only merge the isolated with -other changes when going upstream (to Linus or other maintainers) or -downstream (to your "union" trees, like testing-and-validation above). - -It should be noted that some of this separation is not just recommended -practice, it's actually [for now] -enforced- by BitKeeper. BitKeeper -requires that changesets maintain a certain order, which is the reason -that "bk push" sends all local changesets the remote doesn't have. This -separation may look like a lot of wasted disk space at first, but it -helps when two unrelated changes may "pollute" the same area of code, or -don't follow the same pace of development, or any other of the standard -reasons why one creates a development branch. - -Small development branches (clones) will appear and disappear: - - -------- A --------- B --------- C --------- D ------- - \ / - -----short-term devel branch----- - -While long-term branches will parallel a tree (or trees), with period -merge points. In this first example, we pull from a tree (pulls, -"\") periodically, such as what occurs when tracking changes in a -vendor tree, never pushing changes back up the line: - - -------- A --------- B --------- C --------- D ------- - \ \ \ - ----long-term devel branch----------------- - -And then a more common case in Linux kernel development, a long term -branch with periodic merges back into the tree (pushes, "/"): - - -------- A --------- B --------- C --------- D ------- - \ \ / \ - ----long-term devel branch----------------- - - - - - -Submitting Changes to Linus ---------------------------- -There's a bit of an art, or style, of submitting changes to Linus. -Since Linus's tree is now (you might say) fully integrated into the -distributed BitKeeper system, there are several prerequisites to -properly submitting a BitKeeper change. All these prereq's are just -general cleanliness of BK usage, so as people become experts at BK, feel -free to optimize this process further (assuming Linus agrees, of -course). - - - -0) Make sure your tree was originally cloned from the linux-2.5 tree -created by Linus. If your tree does not have this as its ancestor, it -is impossible to reliably exchange changesets. - - - -1) Pay attention to your commit text. The commit message that -accompanies each changeset you submit will live on forever in history, -and is used by Linus to accurately summarize the changes in each -pre-patch. Remember that there is no context, so - "fix for new scheduler changes" -would be too vague, but - "fix mips64 arch for new scheduler switch_to(), TIF_xxx semantics" -would be much better. - -You can and should use the command "bk comment -C<rev>" to update the -commit text, and improve it after the fact. This is very useful for -development: poor, quick descriptions during development, which get -cleaned up using "bk comment" before issuing the "bk push" to submit the -changes. - - - -2) Include an Internet-available URL for Linus to pull from, such as - - Pull from: http://gkernel.bkbits.net/net-drivers-2.5 - - - -3) Include a summary and "diffstat -p1" of each changeset that will be -downloaded, when Linus issues a "bk pull". The author auto-generates -these summaries using "bk changes -L <parent>", to obtain a listing -of all the pending-to-send changesets, and their commit messages. - -It is important to show Linus what he will be downloading when he issues -a "bk pull", to reduce the time required to sift the changes once they -are downloaded to Linus's local machine. - -IMPORTANT NOTE: One of the features of BK is that your repository does -not have to be up to date, in order for Linus to receive your changes. -It is considered a courtesy to keep your repository fairly recent, to -lessen any potential merge work Linus may need to do. - - -4) Split up your changes. Each maintainer<->Linus situation is likely -to be slightly different here, so take this just as general advice. The -author splits up changes according to "themes" when merging with Linus. -Simultaneous pushes from local development go to special trees which -exist solely to house changes "queued" for Linus. Example of the trees: - - net-drivers-2.5 -- on-going net driver maintenance - vm-2.5 -- VM-related changes - fs-2.5 -- filesystem-related changes - -Linus then has much more freedom for pulling changes. He could (for -example) issue a "bk pull" on vm-2.5 and fs-2.5 trees, to merge their -changes, but hold off net-drivers-2.5 because of a change that needs -more discussion. - -Other maintainers may find that a single linus-pull-from tree is -adequate for passing BK changesets to him. - - - -Frequently Answered Questions ------------------------------ -1) How do I change the e-mail address shown in the changelog? -A. When you run "bk citool" or "bk commit", set environment - variables BK_USER and BK_HOST to the desired username - and host/domain name. - - -2) How do I use tags / get a diff between two kernel versions? -A. Pass the tags Linus uses to 'bk export'. - -ChangeSets are in a forward-progressing order, so it's pretty easy -to get a snapshot starting and ending at any two points in time. -Linus puts tags on each release and pre-release, so you could use -these two examples: - - bk export -tpatch -hdu -rv2.5.4,v2.5.5 | less - # creates patch-2.5.5 essentially - bk export -tpatch -du -rv2.5.5-pre1,v2.5.5 | less - # changes from pre1 to final - -A tag is just an alias for a specific changeset... and since changesets -are ordered, a tag is thus a marker for a specific point in time (or -specific state of the tree). - - -3) Is there an easy way to generate One Big Patch versus mainline, - for my long-lived kernel branch? -A. Yes. This requires BK 3.x, though. - - bk export -tpatch -r`bk repogca bk://linux.bkbits.net/linux-2.5`,+ - diff --git a/Documentation/BK-usage/bk-make-sum b/Documentation/BK-usage/bk-make-sum deleted file mode 100755 index 58ca46a0fcc..00000000000 --- a/Documentation/BK-usage/bk-make-sum +++ /dev/null @@ -1,34 +0,0 @@ -#!/bin/sh -e -# DIR=$HOME/BK/axp-2.5 -# cd $DIR - -LINUS_REPO=$1 -DIRBASE=`basename $PWD` - -{ -cat <<EOT -Please do a - - bk pull bk://gkernel.bkbits.net/$DIRBASE - -This will update the following files: - -EOT - -bk export -tpatch -hdu -r`bk repogca $LINUS_REPO`,+ | diffstat -p1 2>/dev/null - -cat <<EOT - -through these ChangeSets: - -EOT - -bk changes -L -d'$unless(:MERGE:){ChangeSet|:CSETREV:\n}' $LINUS_REPO | -bk -R prs -h -d'$unless(:MERGE:){<:P:@:HOST:> (:D: :I:)\n$each(:C:){ (:C:)\n}\n}' - - -} > /tmp/linus.txt - -cat <<EOT -Mail text in /tmp/linus.txt; please check and send using your favourite -mailer. -EOT diff --git a/Documentation/BK-usage/bksend b/Documentation/BK-usage/bksend deleted file mode 100755 index 836ca943694..00000000000 --- a/Documentation/BK-usage/bksend +++ /dev/null @@ -1,36 +0,0 @@ -#!/bin/sh -# A script to format BK changeset output in a manner that is easy to read. -# Andreas Dilger <adilger@turbolabs.com> 13/02/2002 -# -# Add diffstat output after Changelog <adilger@turbolabs.com> 21/02/2002 - -PROG=bksend - -usage() { - echo "usage: $PROG -r<rev>" - echo -e "\twhere <rev> is of the form '1.23', '1.23..', '1.23..1.27'," - echo -e "\tor '+' to indicate the most recent revision" - - exit 1 -} - -case $1 in --r) REV=$2; shift ;; --r*) REV=`echo $1 | sed 's/^-r//'` ;; -*) echo "$PROG: no revision given, you probably don't want that";; -esac - -[ -z "$REV" ] && usage - -echo "You can import this changeset into BK by piping this whole message to:" -echo "'| bk receive [path to repository]' or apply the patch as usual." - -SEP="\n===================================================================\n\n" -echo -e $SEP -env PAGER=/bin/cat bk changes -r$REV -echo -bk export -tpatch -du -h -r$REV | diffstat -echo; echo -bk export -tpatch -du -h -r$REV -echo -e $SEP -bk send -wgzip_uu -r$REV - diff --git a/Documentation/BK-usage/bz64wrap b/Documentation/BK-usage/bz64wrap deleted file mode 100755 index be780876849..00000000000 --- a/Documentation/BK-usage/bz64wrap +++ /dev/null @@ -1,41 +0,0 @@ -#!/bin/sh - -# bz64wrap - the sending side of a bzip2 | base64 stream -# Andreas Dilger <adilger@clusterfs.com> Jan 2002 - - -PATH=$PATH:/usr/bin:/usr/local/bin:/usr/freeware/bin - -# A program to generate base64 encoding on stdout -BASE64_ENCODE="uuencode -m /dev/stdout" -BASE64_BEGIN= -BASE64_END= - -BZIP=NO -BASE64=NO - -# Test if we have the bzip program installed -bzip2 -c /dev/null > /dev/null 2>&1 && BZIP=YES - -# Test if uuencode can handle the -m (MIME) encoding option -$BASE64_ENCODE < /dev/null > /dev/null 2>&1 && BASE64=YES - -if [ $BASE64 = NO ]; then - BASE64_ENCODE=mimencode - BASE64_BEGIN="begin-base64 644 -" - BASE64_END="====" - - $BASE64_ENCODE < /dev/null > /dev/null 2>&1 && BASE64=YES -fi - -if [ $BZIP = NO -o $BASE64 = NO ]; then - echo "$0: can't use bz64 encoding: bzip2=$BZIP, $BASE64_ENCODE=$BASE64" - exit 1 -fi - -# Sadly, mimencode does not appear to have good "begin" and "end" markers -# like uuencode does, and it is picky about getting the right start/end of -# the base64 stream, so we handle this internally. -echo "$BASE64_BEGIN" -bzip2 -9 | $BASE64_ENCODE -echo "$BASE64_END" diff --git a/Documentation/BK-usage/cpcset b/Documentation/BK-usage/cpcset deleted file mode 100755 index b8faca97dab..00000000000 --- a/Documentation/BK-usage/cpcset +++ /dev/null @@ -1,36 +0,0 @@ -#!/bin/sh -# -# Purpose: Copy changeset patch and description from one -# repository to another, unrelated one. -# -# usage: cpcset [revision] [from-repository] [to-repository] -# - -REV=$1 -FROM=$2 -TO=$3 -TMPF=/tmp/cpcset.$$ - -rm -f $TMPF* - -CWD_SAVE=`pwd` -cd $FROM -bk changes -r$REV | \ - grep -v '^ChangeSet' | \ - sed -e 's/^ //g' > $TMPF.log - -USERHOST=`bk changes -r$REV | grep '^ChangeSet' | awk '{print $4}'` -export BK_USER=`echo $USERHOST | awk '-F@' '{print $1}'` -export BK_HOST=`echo $USERHOST | awk '-F@' '{print $2}'` - -bk export -tpatch -hdu -r$REV > $TMPF.patch && \ -cd $CWD_SAVE && \ -cd $TO && \ -bk import -tpatch -CFR -y"`cat $TMPF.log`" $TMPF.patch . && \ -bk commit -y"`cat $TMPF.log`" - -rm -f $TMPF* - -echo changeset $REV copied. -echo "" - diff --git a/Documentation/BK-usage/cset-to-linus b/Documentation/BK-usage/cset-to-linus deleted file mode 100755 index d28a96f8c61..00000000000 --- a/Documentation/BK-usage/cset-to-linus +++ /dev/null @@ -1,49 +0,0 @@ -#!/usr/bin/perl -w - -use strict; - -my ($lhs, $rev, $tmp, $rhs, $s); -my @cset_text = (); -my @pipe_text = (); -my $have_cset = 0; - -while (<>) { - next if /^---/; - - if (($lhs, $tmp, $rhs) = (/^(ChangeSet\@)([^,]+)(, .*)$/)) { - &cset_rev if ($have_cset); - - $rev = $tmp; - $have_cset = 1; - - push(@cset_text, $_); - } - - elsif ($have_cset) { - push(@cset_text, $_); - } -} -&cset_rev if ($have_cset); -exit(0); - - -sub cset_rev { - my $empty_cset = 0; - - open PIPE, "bk export -tpatch -hdu -r $rev | diffstat -p1 2>/dev/null |" or die; - while ($s = <PIPE>) { - $empty_cset = 1 if ($s =~ /0 files changed/); - push(@pipe_text, $s); - } - close(PIPE); - - if (! $empty_cset) { - print @cset_text; - print @pipe_text; - print "\n\n"; - } - - @pipe_text = (); - @cset_text = (); -} - diff --git a/Documentation/BK-usage/csets-to-patches b/Documentation/BK-usage/csets-to-patches deleted file mode 100755 index e2b81c35883..00000000000 --- a/Documentation/BK-usage/csets-to-patches +++ /dev/null @@ -1,44 +0,0 @@ -#!/usr/bin/perl -w - -use strict; - -my ($lhs, $rev, $tmp, $rhs, $s); -my @cset_text = (); -my @pipe_text = (); -my $have_cset = 0; - -while (<>) { - next if /^---/; - - if (($lhs, $tmp, $rhs) = (/^(ChangeSet\@)([^,]+)(, .*)$/)) { - &cset_rev if ($have_cset); - - $rev = $tmp; - $have_cset = 1; - - push(@cset_text, $_); - } - - elsif ($have_cset) { - push(@cset_text, $_); - } -} -&cset_rev if ($have_cset); -exit(0); - - -sub cset_rev { - my $empty_cset = 0; - - system("bk export -tpatch -du -r $rev > /tmp/rev-$rev.patch"); - - if (! $empty_cset) { - print @cset_text; - print @pipe_text; - print "\n\n"; - } - - @pipe_text = (); - @cset_text = (); -} - diff --git a/Documentation/BK-usage/gcapatch b/Documentation/BK-usage/gcapatch deleted file mode 100755 index aaeb17dc7c7..00000000000 --- a/Documentation/BK-usage/gcapatch +++ /dev/null @@ -1,8 +0,0 @@ -#!/bin/sh -# -# Purpose: Generate GNU diff of local changes versus canonical top-of-tree -# -# Usage: gcapatch > foo.patch -# - -bk export -tpatch -hdu -r`bk repogca bk://linux.bkbits.net/linux-2.5`,+ diff --git a/Documentation/BK-usage/unbz64wrap b/Documentation/BK-usage/unbz64wrap deleted file mode 100755 index 4fc3e73e9a8..00000000000 --- a/Documentation/BK-usage/unbz64wrap +++ /dev/null @@ -1,25 +0,0 @@ -#!/bin/sh - -# unbz64wrap - the receiving side of a bzip2 | base64 stream -# Andreas Dilger <adilger@clusterfs.com> Jan 2002 - -# Sadly, mimencode does not appear to have good "begin" and "end" markers -# like uuencode does, and it is picky about getting the right start/end of -# the base64 stream, so we handle this explicitly here. - -PATH=$PATH:/usr/bin:/usr/local/bin:/usr/freeware/bin - -if mimencode -u < /dev/null > /dev/null 2>&1 ; then - SHOW= - while read LINE; do - case $LINE in - begin-base64*) SHOW=YES ;; - ====) SHOW= ;; - *) [ "$SHOW" ] && echo "$LINE" ;; - esac - done | mimencode -u | bunzip2 - exit $? -else - cat - | uudecode -o /dev/stdout | bunzip2 - exit $? -fi diff --git a/Documentation/Changes b/Documentation/Changes index caa6a5529b6..57542bc25ed 100644 --- a/Documentation/Changes +++ b/Documentation/Changes @@ -357,14 +357,14 @@ Quota-tools ---------- o <http://sourceforge.net/projects/linuxquota/> -Jade ----- -o <ftp://ftp.jclark.com/pub/jade/jade-1.2.1.tar.gz> - DocBook Stylesheets ------------------- o <http://nwalsh.com/docbook/dsssl/> +XMLTO XSLT Frontend +------------------- +o <http://cyberelk.net/tim/xmlto/> + Intel P6 microcode ------------------ o <http://www.urbanmyth.org/microcode/> diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index a221039ee4c..e69b3d2e788 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -7,10 +7,9 @@ # list of DOCBOOKS. DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \ - kernel-hacking.xml kernel-locking.xml via-audio.xml \ - deviceiobook.xml procfs-guide.xml tulip-user.xml \ - writing_usb_driver.xml scsidrivers.xml sis900.xml \ - kernel-api.xml journal-api.xml lsm.xml usb.xml \ + kernel-hacking.xml kernel-locking.xml deviceiobook.xml \ + procfs-guide.xml writing_usb_driver.xml scsidrivers.xml \ + sis900.xml kernel-api.xml journal-api.xml lsm.xml usb.xml \ gadget.xml libata.xml mtdnand.xml librs.xml ### @@ -42,14 +41,16 @@ MAN := $(patsubst %.xml, %.9, $(BOOKS)) mandocs: $(MAN) installmandocs: mandocs - $(MAKEMAN) install Documentation/DocBook/man + mkdir -p /usr/local/man/man9/ + install Documentation/DocBook/man/*.9.gz /usr/local/man/man9/ ### #External programs used KERNELDOC = scripts/kernel-doc DOCPROC = scripts/basic/docproc -SPLITMAN = $(PERL) $(srctree)/scripts/split-man -MAKEMAN = $(PERL) $(srctree)/scripts/makeman + +XMLTOFLAGS = -m Documentation/DocBook/stylesheet.xsl +#XMLTOFLAGS += --skip-validation ### # DOCPROC is used for two purposes: @@ -96,45 +97,44 @@ $(obj)/procfs-guide.xml: $(C-procfs-example2) # Rules to generate postscript, PDF and HTML # db2html creates a directory. Generate a html file used for timestamp -quiet_cmd_db2ps = DB2PS $@ - cmd_db2ps = db2ps -o $(dir $@) $< +quiet_cmd_db2ps = XMLTO $@ + cmd_db2ps = xmlto ps $(XMLTOFLAGS) -o $(dir $@) $< %.ps : %.xml - @(which db2ps > /dev/null 2>&1) || \ - (echo "*** You need to install DocBook stylesheets ***"; \ + @(which xmlto > /dev/null 2>&1) || \ + (echo "*** You need to install xmlto ***"; \ exit 1) $(call cmd,db2ps) -quiet_cmd_db2pdf = DB2PDF $@ - cmd_db2pdf = db2pdf -o $(dir $@) $< +quiet_cmd_db2pdf = XMLTO $@ + cmd_db2pdf = xmlto pdf $(XMLTOFLAGS) -o $(dir $@) $< %.pdf : %.xml - @(which db2pdf > /dev/null 2>&1) || \ - (echo "*** You need to install DocBook stylesheets ***"; \ + @(which xmlto > /dev/null 2>&1) || \ + (echo "*** You need to install xmlto ***"; \ exit 1) $(call cmd,db2pdf) -quiet_cmd_db2html = DB2HTML $@ - cmd_db2html = db2html -o $(patsubst %.html,%,$@) $< && \ - echo '<a HREF="$(patsubst %.html,%,$(notdir $@))/book1.html"> \ +quiet_cmd_db2html = XMLTO $@ + cmd_db2html = xmlto xhtml $(XMLTOFLAGS) -o $(patsubst %.html,%,$@) $< && \ + echo '<a HREF="$(patsubst %.html,%,$(notdir $@))/index.html"> \ Goto $(patsubst %.html,%,$(notdir $@))</a><p>' > $@ %.html: %.xml - @(which db2html > /dev/null 2>&1) || \ - (echo "*** You need to install DocBook stylesheets ***"; \ + @(which xmlto > /dev/null 2>&1) || \ + (echo "*** You need to install xmlto ***"; \ exit 1) @rm -rf $@ $(patsubst %.html,%,$@) $(call cmd,db2html) @if [ ! -z "$(PNG-$(basename $(notdir $@)))" ]; then \ cp $(PNG-$(basename $(notdir $@))) $(patsubst %.html,%,$@); fi -### -# Rule to generate man files - output is placed in the man subdirectory - -%.9: %.xml -ifneq ($(KBUILD_SRC),) - $(Q)mkdir -p $(objtree)/Documentation/DocBook/man -endif - $(SPLITMAN) $< $(objtree)/Documentation/DocBook/man "$(VERSION).$(PATCHLEVEL).$(SUBLEVEL)" - $(MAKEMAN) convert $(objtree)/Documentation/DocBook/man $< +quiet_cmd_db2man = XMLTO $@ + cmd_db2man = if grep -q refentry $<; then xmlto man $(XMLTOFLAGS) -o $(obj)/man $< ; gzip -f $(obj)/man/*.9; fi +%.9 : %.xml + @(which xmlto > /dev/null 2>&1) || \ + (echo "*** You need to install xmlto ***"; \ + exit 1) + $(call cmd,db2man) + @touch $@ ### # Rules to generate postscripts and PNG imgages from .fig format files diff --git a/Documentation/DocBook/kernel-api.tmpl b/Documentation/DocBook/kernel-api.tmpl index 1bd20c86028..757cef8f849 100644 --- a/Documentation/DocBook/kernel-api.tmpl +++ b/Documentation/DocBook/kernel-api.tmpl @@ -49,13 +49,33 @@ !Iinclude/asm-i386/unaligned.h </sect1> -<!-- FIXME: - kernel/sched.c has no docs, which stuffs up the sgml. Comment - out until somebody adds docs. KAO <sect1><title>Delaying, scheduling, and timer routines</title> -X!Ekernel/sched.c +!Iinclude/linux/sched.h +!Ekernel/sched.c +!Ekernel/timer.c </sect1> -KAO --> + <sect1><title>Internal Functions</title> +!Ikernel/exit.c +!Ikernel/signal.c + </sect1> + + <sect1><title>Kernel objects manipulation</title> +<!-- +X!Iinclude/linux/kobject.h +--> +!Elib/kobject.c + </sect1> + + <sect1><title>Kernel utility functions</title> +!Iinclude/linux/kernel.h +<!-- This needs to clean up to make kernel-doc happy +X!Ekernel/printk.c + --> +!Ekernel/panic.c +!Ekernel/sys.c +!Ekernel/rcupdate.c + </sect1> + </chapter> <chapter id="adt"> @@ -81,7 +101,9 @@ KAO --> !Elib/vsprintf.c </sect1> <sect1><title>String Manipulation</title> -!Ilib/string.c +<!-- All functions are exported at now +X!Ilib/string.c + --> !Elib/string.c </sect1> <sect1><title>Bit Operations</title> @@ -98,6 +120,25 @@ KAO --> !Iinclude/asm-i386/uaccess.h !Iarch/i386/lib/usercopy.c </sect1> + <sect1><title>More Memory Management Functions</title> +!Iinclude/linux/rmap.h +!Emm/readahead.c +!Emm/filemap.c +!Emm/memory.c +!Emm/vmalloc.c +!Emm/mempool.c +!Emm/page-writeback.c +!Emm/truncate.c + </sect1> + </chapter> + + + <chapter id="ipc"> + <title>Kernel IPC facilities</title> + + <sect1><title>IPC utilities</title> +!Iipc/util.c + </sect1> </chapter> <chapter id="kfifo"> @@ -114,6 +155,10 @@ KAO --> <sect1><title>sysctl interface</title> !Ekernel/sysctl.c </sect1> + + <sect1><title>proc filesystem interface</title> +!Ifs/proc/base.c + </sect1> </chapter> <chapter id="debugfs"> @@ -127,6 +172,10 @@ KAO --> <chapter id="vfs"> <title>The Linux VFS</title> + <sect1><title>The Filesystem types</title> +!Iinclude/linux/fs.h +!Einclude/linux/fs.h + </sect1> <sect1><title>The Directory Cache</title> !Efs/dcache.c !Iinclude/linux/dcache.h @@ -142,13 +191,31 @@ KAO --> !Efs/locks.c !Ifs/locks.c </sect1> + <sect1><title>Other Functions</title> +!Efs/mpage.c +!Efs/namei.c +!Efs/buffer.c +!Efs/bio.c +!Efs/seq_file.c +!Efs/filesystems.c +!Efs/fs-writeback.c +!Efs/block_dev.c + </sect1> </chapter> <chapter id="netcore"> <title>Linux Networking</title> + <sect1><title>Networking Base Types</title> +!Iinclude/linux/net.h + </sect1> <sect1><title>Socket Buffer Functions</title> !Iinclude/linux/skbuff.h +!Iinclude/net/sock.h +!Enet/socket.c !Enet/core/skbuff.c +!Enet/core/sock.c +!Enet/core/datagram.c +!Enet/core/stream.c </sect1> <sect1><title>Socket Filter</title> !Enet/core/filter.c @@ -158,6 +225,14 @@ KAO --> !Enet/core/gen_stats.c !Enet/core/gen_estimator.c </sect1> + <sect1><title>SUN RPC subsystem</title> +<!-- The !D functionality is not perfect, garbage has to be protected by comments +!Dnet/sunrpc/sunrpc_syms.c +--> +!Enet/sunrpc/xdr.c +!Enet/sunrpc/svcsock.c +!Enet/sunrpc/sched.c + </sect1> </chapter> <chapter id="netdev"> @@ -194,11 +269,26 @@ X!Ekernel/module.c !Iarch/i386/kernel/irq.c </sect1> + <sect1><title>Resources Management</title> +!Ekernel/resource.c + </sect1> + <sect1><title>MTRR Handling</title> !Earch/i386/kernel/cpu/mtrr/main.c </sect1> <sect1><title>PCI Support Library</title> !Edrivers/pci/pci.c +!Edrivers/pci/pci-driver.c +!Edrivers/pci/remove.c +!Edrivers/pci/pci-acpi.c +<!-- kerneldoc does not understand to __devinit +X!Edrivers/pci/search.c + --> +!Edrivers/pci/msi.c +!Edrivers/pci/bus.c +!Edrivers/pci/hotplug.c +!Edrivers/pci/probe.c +!Edrivers/pci/rom.c </sect1> <sect1><title>PCI Hotplug Support Library</title> !Edrivers/pci/hotplug/pci_hotplug_core.c @@ -223,6 +313,14 @@ X!Earch/i386/kernel/mca.c !Efs/devfs/base.c </chapter> + <chapter id="sysfs"> + <title>The Filesystem for Exporting Kernel Objects</title> +!Efs/sysfs/file.c +!Efs/sysfs/dir.c +!Efs/sysfs/symlink.c +!Efs/sysfs/bin.c + </chapter> + <chapter id="security"> <title>Security Framework</title> !Esecurity/security.c @@ -233,6 +331,61 @@ X!Earch/i386/kernel/mca.c !Ekernel/power/pm.c </chapter> + <chapter id="devdrivers"> + <title>Device drivers infrastructure</title> + <sect1><title>Device Drivers Base</title> +<!-- +X!Iinclude/linux/device.h +--> +!Edrivers/base/driver.c +!Edrivers/base/class_simple.c +!Edrivers/base/core.c +!Edrivers/base/firmware_class.c +!Edrivers/base/transport_class.c +!Edrivers/base/dmapool.c +<!-- Cannot be included, because + attribute_container_add_class_device_adapter + and attribute_container_classdev_to_container + exceed allowed 44 characters maximum +X!Edrivers/base/attribute_container.c +--> +!Edrivers/base/sys.c +<!-- +X!Edrivers/base/interface.c +--> +!Edrivers/base/platform.c +!Edrivers/base/bus.c + </sect1> + <sect1><title>Device Drivers Power Management</title> +!Edrivers/base/power/main.c +!Edrivers/base/power/resume.c +!Edrivers/base/power/suspend.c + </sect1> + <sect1><title>Device Drivers ACPI Support</title> +<!-- Internal functions only +X!Edrivers/acpi/sleep/main.c +X!Edrivers/acpi/sleep/wakeup.c +X!Edrivers/acpi/motherboard.c +X!Edrivers/acpi/bus.c +--> +!Edrivers/acpi/scan.c +<!-- No correct structured comments +X!Edrivers/acpi/pci_bind.c +--> + </sect1> + <sect1><title>Device drivers PnP support</title> +!Edrivers/pnp/core.c +<!-- No correct structured comments +X!Edrivers/pnp/system.c + --> +!Edrivers/pnp/card.c +!Edrivers/pnp/driver.c +!Edrivers/pnp/manager.c +!Edrivers/pnp/support.c + </sect1> + </chapter> + + <chapter id="blkdev"> <title>Block Devices</title> !Edrivers/block/ll_rw_blk.c @@ -250,7 +403,23 @@ X!Earch/i386/kernel/mca.c <chapter id="snddev"> <title>Sound Devices</title> +!Iinclude/sound/core.h !Esound/sound_core.c +!Iinclude/sound/pcm.h +!Esound/core/pcm.c +!Esound/core/device.c +!Esound/core/info.c +!Esound/core/rawmidi.c +!Esound/core/sound.c +!Esound/core/memory.c +!Esound/core/pcm_memory.c +!Esound/core/init.c +!Esound/core/isadma.c +!Esound/core/control.c +!Esound/core/pcm_lib.c +!Esound/core/hwdep.c +!Esound/core/pcm_native.c +!Esound/core/memalloc.c <!-- FIXME: Removed for now since no structured comments in source X!Isound/sound_firmware.c --> @@ -258,6 +427,7 @@ X!Isound/sound_firmware.c <chapter id="uart16x50"> <title>16x50 UART Driver</title> +!Iinclude/linux/serial_core.h !Edrivers/serial/serial_core.c !Edrivers/serial/8250.c </chapter> @@ -310,9 +480,11 @@ X!Isound/sound_firmware.c <sect1><title>Frame Buffer Memory</title> !Edrivers/video/fbmem.c </sect1> +<!-- <sect1><title>Frame Buffer Console</title> -!Edrivers/video/console/fbcon.c +X!Edrivers/video/console/fbcon.c </sect1> +--> <sect1><title>Frame Buffer Colormap</title> !Edrivers/video/fbcmap.c </sect1> diff --git a/Documentation/DocBook/stylesheet.xsl b/Documentation/DocBook/stylesheet.xsl new file mode 100644 index 00000000000..e14c21dda40 --- /dev/null +++ b/Documentation/DocBook/stylesheet.xsl @@ -0,0 +1,5 @@ +<?xml version="1.0" encoding="UTF-8"?> +<stylesheet xmlns="http://www.w3.org/1999/XSL/Transform" version="1.0"> +<param name="chunk.quietly">1</param> +<param name="funcsynopsis.style">ansi</param> +</stylesheet> diff --git a/Documentation/DocBook/tulip-user.tmpl b/Documentation/DocBook/tulip-user.tmpl deleted file mode 100644 index 6520d7a1b13..00000000000 --- a/Documentation/DocBook/tulip-user.tmpl +++ /dev/null @@ -1,327 +0,0 @@ -<?xml version="1.0" encoding="UTF-8"?> -<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN" - "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []> - -<book id="TulipUserGuide"> - <bookinfo> - <title>Tulip Driver User's Guide</title> - - <authorgroup> - <author> - <firstname>Jeff</firstname> - <surname>Garzik</surname> - <affiliation> - <address> - <email>jgarzik@pobox.com</email> - </address> - </affiliation> - </author> - </authorgroup> - - <copyright> - <year>2001</year> - <holder>Jeff Garzik</holder> - </copyright> - - <legalnotice> - <para> - This documentation is free software; you can redistribute - it and/or modify it under the terms of the GNU General Public - License as published by the Free Software Foundation; either - version 2 of the License, or (at your option) any later - version. - </para> - - <para> - This program is distributed in the hope that it will be - useful, but WITHOUT ANY WARRANTY; without even the implied - warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - See the GNU General Public License for more details. - </para> - - <para> - You should have received a copy of the GNU General Public - License along with this program; if not, write to the Free - Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, - MA 02111-1307 USA - </para> - - <para> - For more details see the file COPYING in the source - distribution of Linux. - </para> - </legalnotice> - </bookinfo> - - <toc></toc> - - <chapter id="intro"> - <title>Introduction</title> -<para> -The Tulip Ethernet Card Driver -is maintained by Jeff Garzik (<email>jgarzik@pobox.com</email>). -</para> - -<para> -The Tulip driver was developed by Donald Becker and changed by -Jeff Garzik, Takashi Manabe and a cast of thousands. -</para> - -<para> -For 2.4.x and later kernels, the Linux Tulip driver is available at -<ulink url="http://sourceforge.net/projects/tulip/">http://sourceforge.net/projects/tulip/</ulink> -</para> - -<para> - This driver is for the Digital "Tulip" Ethernet adapter interface. - It should work with most DEC 21*4*-based chips/ethercards, as well as - with work-alike chips from Lite-On (PNIC) and Macronix (MXIC) and ASIX. -</para> - -<para> - The original author may be reached as becker@scyld.com, or C/O - Scyld Computing Corporation, - 410 Severn Ave., Suite 210, - Annapolis MD 21403 -</para> - -<para> - Additional information on Donald Becker's tulip.c - is available at <ulink url="http://www.scyld.com/network/tulip.html">http://www.scyld.com/network/tulip.html</ulink> -</para> - - </chapter> - - <chapter id="drvr-compat"> - <title>Driver Compatibility</title> - -<para> -This device driver is designed for the DECchip "Tulip", Digital's -single-chip ethernet controllers for PCI (now owned by Intel). -Supported members of the family -are the 21040, 21041, 21140, 21140A, 21142, and 21143. Similar work-alike -chips from Lite-On, Macronics, ASIX, Compex and other listed below are also -supported. -</para> - -<para> -These chips are used on at least 140 unique PCI board designs. The great -number of chips and board designs supported is the reason for the -driver size and complexity. Almost of the increasing complexity is in the -board configuration and media selection code. There is very little -increasing in the operational critical path length. -</para> - </chapter> - - <chapter id="board-settings"> - <title>Board-specific Settings</title> - -<para> -PCI bus devices are configured by the system at boot time, so no jumpers -need to be set on the board. The system BIOS preferably should assign the -PCI INTA signal to an otherwise unused system IRQ line. -</para> - -<para> -Some boards have EEPROMs tables with default media entry. The factory default -is usually "autoselect". This should only be overridden when using -transceiver connections without link beat e.g. 10base2 or AUI, or (rarely!) -for forcing full-duplex when used with old link partners that do not do -autonegotiation. -</para> - </chapter> - - <chapter id="driver-operation"> - <title>Driver Operation</title> - -<sect1><title>Ring buffers</title> - -<para> -The Tulip can use either ring buffers or lists of Tx and Rx descriptors. -This driver uses statically allocated rings of Rx and Tx descriptors, set at -compile time by RX/TX_RING_SIZE. This version of the driver allocates skbuffs -for the Rx ring buffers at open() time and passes the skb->data field to the -Tulip as receive data buffers. When an incoming frame is less than -RX_COPYBREAK bytes long, a fresh skbuff is allocated and the frame is -copied to the new skbuff. When the incoming frame is larger, the skbuff is -passed directly up the protocol stack and replaced by a newly allocated -skbuff. -</para> - -<para> -The RX_COPYBREAK value is chosen to trade-off the memory wasted by -using a full-sized skbuff for small frames vs. the copying costs of larger -frames. For small frames the copying cost is negligible (esp. considering -that we are pre-loading the cache with immediately useful header -information). For large frames the copying cost is non-trivial, and the -larger copy might flush the cache of useful data. A subtle aspect of this -choice is that the Tulip only receives into longword aligned buffers, thus -the IP header at offset 14 isn't longword aligned for further processing. -Copied frames are put into the new skbuff at an offset of "+2", thus copying -has the beneficial effect of aligning the IP header and preloading the -cache. -</para> - -</sect1> - -<sect1><title>Synchronization</title> -<para> -The driver runs as two independent, single-threaded flows of control. One -is the send-packet routine, which enforces single-threaded use by the -dev->tbusy flag. The other thread is the interrupt handler, which is single -threaded by the hardware and other software. -</para> - -<para> -The send packet thread has partial control over the Tx ring and 'dev->tbusy' -flag. It sets the tbusy flag whenever it's queuing a Tx packet. If the next -queue slot is empty, it clears the tbusy flag when finished otherwise it sets -the 'tp->tx_full' flag. -</para> - -<para> -The interrupt handler has exclusive control over the Rx ring and records stats -from the Tx ring. (The Tx-done interrupt can't be selectively turned off, so -we can't avoid the interrupt overhead by having the Tx routine reap the Tx -stats.) After reaping the stats, it marks the queue entry as empty by setting -the 'base' to zero. Iff the 'tp->tx_full' flag is set, it clears both the -tx_full and tbusy flags. -</para> - -</sect1> - - </chapter> - - <chapter id="errata"> - <title>Errata</title> - -<para> -The old DEC databooks were light on details. -The 21040 databook claims that CSR13, CSR14, and CSR15 should each be the last -register of the set CSR12-15 written. Hmmm, now how is that possible? -</para> - -<para> -The DEC SROM format is very badly designed not precisely defined, leading to -part of the media selection junkheap below. Some boards do not have EEPROM -media tables and need to be patched up. Worse, other boards use the DEC -design kit media table when it isn't correct for their board. -</para> - -<para> -We cannot use MII interrupts because there is no defined GPIO pin to attach -them. The MII transceiver status is polled using an kernel timer. -</para> - </chapter> - - <chapter id="changelog"> - <title>Driver Change History</title> - - <sect1><title>Version 0.9.14 (February 20, 2001)</title> - <itemizedlist> - <listitem><para>Fix PNIC problems (Manfred Spraul)</para></listitem> - <listitem><para>Add new PCI id for Accton comet</para></listitem> - <listitem><para>Support Davicom tulips</para></listitem> - <listitem><para>Fix oops in eeprom parsing</para></listitem> - <listitem><para>Enable workarounds for early PCI chipsets</para></listitem> - <listitem><para>IA64, hppa csr0 support</para></listitem> - <listitem><para>Support media types 5, 6</para></listitem> - <listitem><para>Interpret a bit more of the 21142 SROM extended media type 3</para></listitem> - <listitem><para>Add missing delay in eeprom reading</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.11 (November 3, 2000)</title> - <itemizedlist> - <listitem><para>Eliminate extra bus accesses when sharing interrupts (prumpf)</para></listitem> - <listitem><para>Barrier following ownership descriptor bit flip (prumpf)</para></listitem> - <listitem><para>Endianness fixes for >14 addresses in setup frames (prumpf)</para></listitem> - <listitem><para>Report link beat to kernel/userspace via netif_carrier_*. (kuznet)</para></listitem> - <listitem><para>Better spinlocking in set_rx_mode.</para></listitem> - <listitem><para>Fix I/O resource request failure error messages (DaveM catch)</para></listitem> - <listitem><para>Handle DMA allocation failure.</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.10 (September 6, 2000)</title> - <itemizedlist> - <listitem><para>Simple interrupt mitigation (via jamal)</para></listitem> - <listitem><para>More PCI ids</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.9 (August 11, 2000)</title> - <itemizedlist> - <listitem><para>More PCI ids</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.8 (July 13, 2000)</title> - <itemizedlist> - <listitem><para>Correct signed/unsigned comparison for dummy frame index</para></listitem> - <listitem><para>Remove outdated references to struct enet_statistics</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.7 (June 17, 2000)</title> - <itemizedlist> - <listitem><para>Timer cleanups (Andrew Morton)</para></listitem> - <listitem><para>Alpha compile fix (somebody?)</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.6 (May 31, 2000)</title> - <itemizedlist> - <listitem><para>Revert 21143-related support flag patch</para></listitem> - <listitem><para>Add HPPA/media-table debugging printk</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.5 (May 30, 2000)</title> - <itemizedlist> - <listitem><para>HPPA support (willy@puffingroup)</para></listitem> - <listitem><para>CSR6 bits and tulip.h cleanup (Chris Smith)</para></listitem> - <listitem><para>Improve debugging messages a bit</para></listitem> - <listitem><para>Add delay after CSR13 write in t21142_start_nway</para></listitem> - <listitem><para>Remove unused ETHER_STATS code</para></listitem> - <listitem><para>Convert 'extern inline' to 'static inline' in tulip.h (Chris Smith)</para></listitem> - <listitem><para>Update DS21143 support flags in tulip_chip_info[]</para></listitem> - <listitem><para>Use spin_lock_irq, not _irqsave/restore, in tulip_start_xmit()</para></listitem> - <listitem><para>Add locking to set_rx_mode()</para></listitem> - <listitem><para>Fix race with chip setting DescOwned bit (Hal Murray)</para></listitem> - <listitem><para>Request 100% of PIO and MMIO resource space assigned to card</para></listitem> - <listitem><para>Remove error message from pci_enable_device failure</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.4.3 (April 14, 2000)</title> - <itemizedlist> - <listitem><para>mod_timer fix (Hal Murray)</para></listitem> - <listitem><para>PNIC2 resuscitation (Chris Smith)</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.4.2 (March 21, 2000)</title> - <itemizedlist> - <listitem><para>Fix 21041 CSR7, CSR13/14/15 handling</para></listitem> - <listitem><para>Merge some PCI ids from tulip 0.91x</para></listitem> - <listitem><para>Merge some HAS_xxx flags and flag settings from tulip 0.91x</para></listitem> - <listitem><para>asm/io.h fix (submitted by many) and cleanup</para></listitem> - <listitem><para>s/HAS_NWAY143/HAS_NWAY/</para></listitem> - <listitem><para>Cleanup 21041 mode reporting</para></listitem> - <listitem><para>Small code cleanups</para></listitem> - </itemizedlist> - </sect1> - - <sect1><title>Version 0.9.4.1 (March 18, 2000)</title> - <itemizedlist> - <listitem><para>Finish PCI DMA conversion (davem)</para></listitem> - <listitem><para>Do not netif_start_queue() at end of tulip_tx_timeout() (kuznet)</para></listitem> - <listitem><para>PCI DMA fix (kuznet)</para></listitem> - <listitem><para>eeprom.c code cleanup</para></listitem> - <listitem><para>Remove Xircom Tulip crud</para></listitem> - </itemizedlist> - </sect1> - </chapter> - -</book> diff --git a/Documentation/DocBook/via-audio.tmpl b/Documentation/DocBook/via-audio.tmpl deleted file mode 100644 index 36e642147d6..00000000000 --- a/Documentation/DocBook/via-audio.tmpl +++ /dev/null @@ -1,597 +0,0 @@ -<?xml version="1.0" encoding="UTF-8"?> -<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN" - "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []> - -<book id="ViaAudioGuide"> - <bookinfo> - <title>Via 686 Audio Driver for Linux</title> - - <authorgroup> - <author> - <firstname>Jeff</firstname> - <surname>Garzik</surname> - </author> - </authorgroup> - - <copyright> - <year>1999-2001</year> - <holder>Jeff Garzik</holder> - </copyright> - - <legalnotice> - <para> - This documentation is free software; you can redistribute - it and/or modify it under the terms of the GNU General Public - License as published by the Free Software Foundation; either - version 2 of the License, or (at your option) any later - version. - </para> - - <para> - This program is distributed in the hope that it will be - useful, but WITHOUT ANY WARRANTY; without even the implied - warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - See the GNU General Public License for more details. - </para> - - <para> - You should have received a copy of the GNU General Public - License along with this program; if not, write to the Free - Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, - MA 02111-1307 USA - </para> - - <para> - For more details see the file COPYING in the source - distribution of Linux. - </para> - </legalnotice> - </bookinfo> - -<toc></toc> - - <chapter id="intro"> - <title>Introduction</title> - <para> - The Via VT82C686A "super southbridge" chips contain - AC97-compatible audio logic which features dual 16-bit stereo - PCM sound channels (full duplex), plus a third PCM channel intended for use - in hardware-assisted FM synthesis. - </para> - <para> - The current Linux kernel audio driver for this family of chips - supports audio playback and recording, but hardware-assisted - FM features, and hardware buffer direct-access (mmap) - support are not yet available. - </para> - <para> - This driver supports any Linux kernel version after 2.4.10. - </para> - <para> - Please send bug reports to the mailing list <email>linux-via@gtf.org</email>. - To subscribe, e-mail <email>majordomo@gtf.org</email> with - </para> - <programlisting> - subscribe linux-via - </programlisting> - <para> - in the body of the message. - </para> - </chapter> - - <chapter id="install"> - <title>Driver Installation</title> - <para> - To use this audio driver, select the - CONFIG_SOUND_VIA82CXXX option in the section Sound during kernel configuration. - Follow the usual kernel procedures for rebuilding the kernel, - or building and installing driver modules. - </para> - <para> - To make this driver the default audio driver, you can add the - following to your /etc/conf.modules file: - </para> - <programlisting> - alias sound via82cxxx_audio - </programlisting> - <para> - Note that soundcore and ac97_codec support modules - are also required for working audio, in addition to - the via82cxxx_audio module itself. - </para> - </chapter> - - <chapter id="reportbug"> - <title>Submitting a bug report</title> - <sect1 id="bugrepdesc"><title>Description of problem</title> - <para> - Describe the application you were using to play/record sound, and how - to reproduce the problem. - </para> - </sect1> - <sect1 id="bugrepdiag"><title>Diagnostic output</title> - <para> - Obtain the via-audio-diag diagnostics program from - http://sf.net/projects/gkernel/ and provide a dump of the - audio chip's registers while the problem is occurring. Sample command line: - </para> - <programlisting> - ./via-audio-diag -aps > diag-output.txt - </programlisting> - </sect1> - <sect1 id="bugrepdebug"><title>Driver debug output</title> - <para> - Define <constant>VIA_DEBUG</constant> at the beginning of the driver, then capture and email - the kernel log output. This can be viewed in the system kernel log (if - enabled), or via the dmesg program. Sample command line: - </para> - <programlisting> - dmesg > /tmp/dmesg-output.txt - </programlisting> - </sect1> - <sect1 id="bugrepprintk"><title>Bigger kernel message buffer</title> - <para> - If you wish to increase the size of the buffer displayed by dmesg, then - change the <constant>LOG_BUF_LEN</constant> macro at the top of linux/kernel/printk.c, recompile - your kernel, and pass the <constant>LOG_BUF_LEN</constant> value to dmesg. Sample command line with - <constant>LOG_BUF_LEN</constant> == 32768: - </para> - <programlisting> - dmesg -s 32768 > /tmp/dmesg-output.txt - </programlisting> - </sect1> - </chapter> - - <chapter id="bugs"> - <title>Known Bugs And Assumptions</title> - <para> - <variablelist> - <varlistentry><term>Low volume</term> - <listitem> - <para> - Volume too low on many systems. Workaround: use mixer program - such as xmixer to increase volume. - </para> - </listitem></varlistentry> - - </variablelist> - - </para> - </chapter> - - <chapter id="thanks"> - <title>Thanks</title> - <para> - Via for providing e-mail support, specs, and NDA'd source code. - </para> - <para> - MandrakeSoft for providing hacking time. - </para> - <para> - AC97 mixer interface fixes and debugging by Ron Cemer <email>roncemer@gte.net</email>. - </para> - <para> - Rui Sousa <email>rui.sousa@conexant.com</email>, for bugfixing - MMAP support, and several other notable fixes that resulted from - his hard work and testing. - </para> - <para> - Adrian Cox <email>adrian@humboldt.co.uk</email>, for bugfixing - MMAP support, and several other notable fixes that resulted from - his hard work and testing. - </para> - <para> - Thomas Sailer for further bugfixes. - </para> - </chapter> - - <chapter id="notes"> - <title>Random Notes</title> - <para> - Two /proc pseudo-files provide diagnostic information. This is generally - not useful to most users. Power users can disable CONFIG_SOUND_VIA82CXXX_PROCFS, - and remove the /proc support code. Once - version 2.0.0 is released, the /proc support code will be disabled by - default. Available /proc pseudo-files: - </para> - <programlisting> - /proc/driver/via/0/info - /proc/driver/via/0/ac97 - </programlisting> - <para> - This driver by default supports all PCI audio devices which report - a vendor id of 0x1106, and a device id of 0x3058. Subsystem vendor - and device ids are not examined. - </para> - <para> - GNU indent formatting options: - <programlisting> --kr -i8 -ts8 -br -ce -bap -sob -l80 -pcs -cs -ss -bs -di1 -nbc -lp -psl - </programlisting> - </para> - <para> - Via has graciously donated e-mail support and source code to help further - the development of this driver. Their assistance has been invaluable - in the design and coding of the next major version of this driver. - </para> - <para> - The Via audio chip apparently provides a second PCM scatter-gather - DMA channel just for FM data, but does not have a full hardware MIDI - processor. I haven't put much thought towards a solution here, but it - might involve using SoftOSS midi wave table, or simply disabling MIDI - support altogether and using the FM PCM channel as a second (input? output?) - </para> - </chapter> - - <chapter id="changelog"> - <title>Driver ChangeLog</title> - -<sect1 id="version191"><title> -Version 1.9.1 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - DSP read/write bugfixes from Thomas Sailer. - </para> - </listitem> - - <listitem> - <para> - Add new PCI id for single-channel use of Via 8233. - </para> - </listitem> - - <listitem> - <para> - Other bug fixes, tweaks, new ioctls. - </para> - </listitem> - - </itemizedlist> -</sect1> - -<sect1 id="version1115"><title> -Version 1.1.15 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Support for variable fragment size and variable fragment number (Rui - Sousa) - </para> - </listitem> - - <listitem> - <para> - Fixes for the SPEED, STEREO, CHANNELS, FMT ioctls when in read & - write mode (Rui Sousa) - </para> - </listitem> - - <listitem> - <para> - Mmaped sound is now fully functional. (Rui Sousa) - </para> - </listitem> - - <listitem> - <para> - Make sure to enable PCI device before reading any of its PCI - config information. (fixes potential hotplug problems) - </para> - </listitem> - - <listitem> - <para> - Clean up code a bit and add more internal function documentation. - </para> - </listitem> - - <listitem> - <para> - AC97 codec access fixes (Adrian Cox) - </para> - </listitem> - - <listitem> - <para> - Big endian fixes (Adrian Cox) - </para> - </listitem> - - <listitem> - <para> - MIDI support (Adrian Cox) - </para> - </listitem> - - <listitem> - <para> - Detect and report locked-rate AC97 codecs. If your hardware only - supports 48Khz (locked rate), then your recording/playback software - must upsample or downsample accordingly. The hardware cannot do it. - </para> - </listitem> - - <listitem> - <para> - Use new pci_request_regions and pci_disable_device functions in - kernel 2.4.6. - </para> - </listitem> - - </itemizedlist> -</sect1> - -<sect1 id="version1114"><title> -Version 1.1.14 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Use VM_RESERVE when available, to eliminate unnecessary page faults. - </para> - </listitem> - </itemizedlist> -</sect1> - -<sect1 id="version1112"><title> -Version 1.1.12 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - mmap bug fixes from Linus. - </para> - </listitem> - </itemizedlist> -</sect1> - -<sect1 id="version1111"><title> -Version 1.1.11 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Many more bug fixes. mmap enabled by default, but may still be buggy. - </para> - </listitem> - - <listitem> - <para> - Uses new and spiffy method of mmap'ing the DMA buffer, based - on a suggestion from Linus. - </para> - </listitem> - </itemizedlist> -</sect1> - -<sect1 id="version1110"><title> -Version 1.1.10 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Many bug fixes. mmap enabled by default, but may still be buggy. - </para> - </listitem> - </itemizedlist> -</sect1> - -<sect1 id="version119"><title> -Version 1.1.9 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Redesign and rewrite audio playback implementation. (faster and smaller, hopefully) - </para> - </listitem> - - <listitem> - <para> - Implement recording and full duplex (DSP_CAP_DUPLEX) support. - </para> - </listitem> - - <listitem> - <para> - Make procfs support optional. - </para> - </listitem> - - <listitem> - <para> - Quick interrupt status check, to lessen overhead in interrupt - sharing situations. - </para> - </listitem> - - <listitem> - <para> - Add mmap(2) support. Disabled for now, it is still buggy and experimental. - </para> - </listitem> - - <listitem> - <para> - Surround all syscalls with a semaphore for cheap and easy SMP protection. - </para> - </listitem> - - <listitem> - <para> - Fix bug in channel shutdown (hardware channel reset) code. - </para> - </listitem> - - <listitem> - <para> - Remove unnecessary spinlocks (better performance). - </para> - </listitem> - - <listitem> - <para> - Eliminate "unknown AFMT" message by using a different method - of selecting the best AFMT_xxx sound sample format for use. - </para> - </listitem> - - <listitem> - <para> - Support for realtime hardware pointer position reporting - (DSP_CAP_REALTIME, SNDCTL_DSP_GETxPTR ioctls) - </para> - </listitem> - - <listitem> - <para> - Support for capture/playback triggering - (DSP_CAP_TRIGGER, SNDCTL_DSP_SETTRIGGER ioctls) - </para> - </listitem> - - <listitem> - <para> - SNDCTL_DSP_SETDUPLEX and SNDCTL_DSP_POST ioctls now handled. - </para> - </listitem> - - <listitem> - <para> - Rewrite open(2) and close(2) logic to allow only one user at - a time. All other open(2) attempts will sleep until they succeed. - FIXME: open(O_RDONLY) and open(O_WRONLY) should be allowed to succeed. - </para> - </listitem> - - <listitem> - <para> - Reviewed code to ensure that SMP and multiple audio devices - are fully supported. - </para> - </listitem> - - </itemizedlist> -</sect1> - -<sect1 id="version118"><title> -Version 1.1.8 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Clean up interrupt handler output. Fixes the following kernel error message: - </para> - <programlisting> - unhandled interrupt ... - </programlisting> - </listitem> - - <listitem> - <para> - Convert documentation to DocBook, so that PDF, HTML and PostScript (.ps) output is readily - available. - </para> - </listitem> - - </itemizedlist> -</sect1> - -<sect1 id="version117"><title> -Version 1.1.7 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Fix module unload bug where mixer device left registered - after driver exit - </para> - </listitem> - </itemizedlist> -</sect1> - -<sect1 id="version116"><title> -Version 1.1.6 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Rewrite via_set_rate to mimic ALSA basic AC97 rate setting - </para> - </listitem> - <listitem> - <para> - Remove much dead code - </para> - </listitem> - <listitem> - <para> - Complete spin_lock_irqsave -> spin_lock_irq conversion in via_dsp_ioctl - </para> - </listitem> - <listitem> - <para> - Fix build problem in via_dsp_ioctl - </para> - </listitem> - <listitem> - <para> - Optimize included headers to eliminate headers found in linux/sound - </para> - </listitem> - </itemizedlist> -</sect1> - -<sect1 id="version115"><title> -Version 1.1.5 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Disable some overly-verbose debugging code - </para> - </listitem> - <listitem> - <para> - Remove unnecessary sound locks - </para> - </listitem> - <listitem> - <para> - Fix some ioctls for better time resolution - </para> - </listitem> - <listitem> - <para> - Begin spin_lock_irqsave -> spin_lock_irq conversion in via_dsp_ioctl - </para> - </listitem> - </itemizedlist> -</sect1> - -<sect1 id="version114"><title> -Version 1.1.4 -</title> - <itemizedlist spacing="compact"> - <listitem> - <para> - Completed rewrite of driver. Eliminated SoundBlaster compatibility - completely, and now uses the much-faster scatter-gather DMA engine. - </para> - </listitem> - </itemizedlist> -</sect1> - - </chapter> - - <chapter id="intfunctions"> - <title>Internal Functions</title> -!Isound/oss/via82cxxx_audio.c - </chapter> - -</book> - - diff --git a/Documentation/RCU/RTFP.txt b/Documentation/RCU/RTFP.txt index 12250b342e1..9c6d450138e 100644 --- a/Documentation/RCU/RTFP.txt +++ b/Documentation/RCU/RTFP.txt @@ -108,8 +108,9 @@ year saw a paper describing an RCU implementation of System V IPC 2004 has seen a Linux-Journal article on use of RCU in dcache [McKenney04a], a performance comparison of locking to RCU on several different CPUs [McKenney04b], a dissertation describing use of RCU in a -number of operating-system kernels [PaulEdwardMcKenneyPhD], and a paper -describing how to make RCU safe for soft-realtime applications [Sarma04c]. +number of operating-system kernels [PaulEdwardMcKenneyPhD], a paper +describing how to make RCU safe for soft-realtime applications [Sarma04c], +and a paper describing SELinux performance with RCU [JamesMorris04b]. Bibtex Entries @@ -341,6 +342,17 @@ Dipankar Sarma" ,pages="18-26" } +@techreport{Friedberg03a +,author="Stuart A. Friedberg" +,title="Lock-Free Wild Card Search Data Structure and Method" +,institution="US Patent and Trademark Office" +,address="Washington, DC" +,year="2003" +,number="US Patent 6,662,184 (contributed under GPL)" +,month="December" +,pages="112" +} + @article{McKenney04a ,author="Paul E. McKenney and Dipankar Sarma and Maneesh Soni" ,title="Scaling dcache with {RCU}" @@ -373,6 +385,9 @@ in Operating System Kernels" ,school="OGI School of Science and Engineering at Oregon Health and Sciences University" ,year="2004" +,note="Available: +\url{http://www.rdrop.com/users/paulmck/RCU/RCUdissertation.2004.07.14e1.pdf} +[Viewed October 15, 2004]" } @Conference{Sarma04c @@ -385,3 +400,13 @@ Oregon Health and Sciences University" ,month="June" ,pages="182-191" } + +@unpublished{JamesMorris04b +,Author="James Morris" +,Title="Recent Developments in {SELinux} Kernel Performance" +,month="December" +,year="2004" +,note="Available: +\url{http://www.livejournal.com/users/james_morris/2153.html} +[Viewed December 10, 2004]" +} diff --git a/Documentation/RCU/UP.txt b/Documentation/RCU/UP.txt index 551a803d82a..3bfb84b3b7d 100644 --- a/Documentation/RCU/UP.txt +++ b/Documentation/RCU/UP.txt @@ -2,11 +2,11 @@ RCU on Uniprocessor Systems A common misconception is that, on UP systems, the call_rcu() primitive -may immediately invoke its function, and that the synchronize_kernel +may immediately invoke its function, and that the synchronize_rcu() primitive may return immediately. The basis of this misconception is that since there is only one CPU, it should not be necessary to wait for anything else to get done, since there are no other CPUs for -anything else to be happening on. Although this approach will sort of +anything else to be happening on. Although this approach will -sort- -of- work a surprising amount of the time, it is a very bad idea in general. This document presents two examples that demonstrate exactly how bad an idea this is. @@ -44,14 +44,14 @@ its arguments would cause it to fail to make the fundamental guarantee underlying RCU, namely that call_rcu() defers invoking its arguments until all RCU read-side critical sections currently executing have completed. -Quick Quiz: why is it -not- legal to invoke synchronize_kernel() in +Quick Quiz: why is it -not- legal to invoke synchronize_rcu() in this case? Summary Permitting call_rcu() to immediately invoke its arguments or permitting -synchronize_kernel() to immediately return breaks RCU, even on a UP system. +synchronize_rcu() to immediately return breaks RCU, even on a UP system. So do not do it! Even on a UP system, the RCU infrastructure -must- respect grace periods. diff --git a/Documentation/RCU/checklist.txt b/Documentation/RCU/checklist.txt index b3a568abe6b..8f3fb77c9cd 100644 --- a/Documentation/RCU/checklist.txt +++ b/Documentation/RCU/checklist.txt @@ -32,7 +32,10 @@ over a rather long period of time, but improvements are always welcome! them -- even x86 allows reads to be reordered), and be prepared to explain why this added complexity is worthwhile. If you choose #c, be prepared to explain how this single task does not - become a major bottleneck on big multiprocessor machines. + become a major bottleneck on big multiprocessor machines (for + example, if the task is updating information relating to itself + that other tasks can read, there by definition can be no + bottleneck). 2. Do the RCU read-side critical sections make proper use of rcu_read_lock() and friends? These primitives are needed @@ -89,27 +92,34 @@ over a rather long period of time, but improvements are always welcome! "_rcu()" list-traversal primitives, such as the list_for_each_entry_rcu(). - b. If the list macros are being used, the list_del_rcu(), - list_add_tail_rcu(), and list_del_rcu() primitives must - be used in order to prevent weakly ordered machines from - misordering structure initialization and pointer planting. + b. If the list macros are being used, the list_add_tail_rcu() + and list_add_rcu() primitives must be used in order + to prevent weakly ordered machines from misordering + structure initialization and pointer planting. Similarly, if the hlist macros are being used, the - hlist_del_rcu() and hlist_add_head_rcu() primitives - are required. + hlist_add_head_rcu() primitive is required. - c. Updates must ensure that initialization of a given + c. If the list macros are being used, the list_del_rcu() + primitive must be used to keep list_del()'s pointer + poisoning from inflicting toxic effects on concurrent + readers. Similarly, if the hlist macros are being used, + the hlist_del_rcu() primitive is required. + + The list_replace_rcu() primitive may be used to + replace an old structure with a new one in an + RCU-protected list. + + d. Updates must ensure that initialization of a given structure happens before pointers to that structure are publicized. Use the rcu_assign_pointer() primitive when publicizing a pointer to a structure that can be traversed by an RCU read-side critical section. - [The rcu_assign_pointer() primitive is in process.] - 5. If call_rcu(), or a related primitive such as call_rcu_bh(), is used, the callback function must be written to be called from softirq context. In particular, it cannot block. -6. Since synchronize_kernel() blocks, it cannot be called from +6. Since synchronize_rcu() can block, it cannot be called from any sort of irq context. 7. If the updater uses call_rcu(), then the corresponding readers @@ -125,9 +135,9 @@ over a rather long period of time, but improvements are always welcome! such cases is a must, of course! And the jury is still out on whether the increased speed is worth it. -8. Although synchronize_kernel() is a bit slower than is call_rcu(), +8. Although synchronize_rcu() is a bit slower than is call_rcu(), it usually results in simpler code. So, unless update performance - is important or the updaters cannot block, synchronize_kernel() + is important or the updaters cannot block, synchronize_rcu() should be used in preference to call_rcu(). 9. All RCU list-traversal primitives, which include @@ -155,3 +165,14 @@ over a rather long period of time, but improvements are always welcome! you -must- use the "_rcu()" variants of the list macros. Failing to do so will break Alpha and confuse people reading your code. + +11. Note that synchronize_rcu() -only- guarantees to wait until + all currently executing rcu_read_lock()-protected RCU read-side + critical sections complete. It does -not- necessarily guarantee + that all currently running interrupts, NMIs, preempt_disable() + code, or idle loops will complete. Therefore, if you do not have + rcu_read_lock()-protected read-side critical sections, do -not- + use synchronize_rcu(). + + If you want to wait for some of these other things, you might + instead need to use synchronize_irq() or synchronize_sched(). diff --git a/Documentation/RCU/listRCU.txt b/Documentation/RCU/listRCU.txt index bda6ead69bd..f8a54fa0d8a 100644 --- a/Documentation/RCU/listRCU.txt +++ b/Documentation/RCU/listRCU.txt @@ -32,6 +32,7 @@ implementation of audit_filter_task() might be as follows: enum audit_state state; read_lock(&auditsc_lock); + /* Note: audit_netlink_sem held by caller. */ list_for_each_entry(e, &audit_tsklist, list) { if (audit_filter_rules(tsk, &e->rule, NULL, &state)) { read_unlock(&auditsc_lock); @@ -55,6 +56,7 @@ This means that RCU can be easily applied to the read side, as follows: enum audit_state state; rcu_read_lock(); + /* Note: audit_netlink_sem held by caller. */ list_for_each_entry_rcu(e, &audit_tsklist, list) { if (audit_filter_rules(tsk, &e->rule, NULL, &state)) { rcu_read_unlock(); @@ -139,12 +141,15 @@ Normally, the write_lock() and write_unlock() would be replaced by a spin_lock() and a spin_unlock(), but in this case, all callers hold audit_netlink_sem, so no additional locking is required. The auditsc_lock can therefore be eliminated, since use of RCU eliminates the need for -writers to exclude readers. +writers to exclude readers. Normally, the write_lock() calls would +be converted into spin_lock() calls. The list_del(), list_add(), and list_add_tail() primitives have been replaced by list_del_rcu(), list_add_rcu(), and list_add_tail_rcu(). The _rcu() list-manipulation primitives add memory barriers that are -needed on weakly ordered CPUs (most of them!). +needed on weakly ordered CPUs (most of them!). The list_del_rcu() +primitive omits the pointer poisoning debug-assist code that would +otherwise cause concurrent readers to fail spectacularly. So, when readers can tolerate stale data and when entries are either added or deleted, without in-place modification, it is very easy to use RCU! @@ -166,6 +171,7 @@ otherwise, the added fields would need to be filled in): struct audit_newentry *ne; write_lock(&auditsc_lock); + /* Note: audit_netlink_sem held by caller. */ list_for_each_entry(e, list, list) { if (!audit_compare_rule(rule, &e->rule)) { e->rule.action = newaction; @@ -199,8 +205,7 @@ RCU ("read-copy update") its name. The RCU code is as follows: audit_copy_rule(&ne->rule, &e->rule); ne->rule.action = newaction; ne->rule.file_count = newfield_count; - list_add_rcu(ne, e); - list_del(e); + list_replace_rcu(e, ne); call_rcu(&e->rcu, audit_free_rule, e); return 0; } diff --git a/Documentation/RCU/rcu.txt b/Documentation/RCU/rcu.txt index 7e0c2ab6f2b..eb444006683 100644 --- a/Documentation/RCU/rcu.txt +++ b/Documentation/RCU/rcu.txt @@ -43,7 +43,9 @@ o If I am running on a uniprocessor kernel, which can only do one o How can I see where RCU is currently used in the Linux kernel? - Search for "rcu_read_lock", "call_rcu", and "synchronize_kernel". + Search for "rcu_read_lock", "rcu_read_unlock", "call_rcu", + "rcu_read_lock_bh", "rcu_read_unlock_bh", "call_rcu_bh", + "synchronize_rcu", and "synchronize_net". o What guidelines should I follow when writing code that uses RCU? diff --git a/Documentation/aoe/aoe.txt b/Documentation/aoe/aoe.txt index 43e50108d0e..3a4dbe4663c 100644 --- a/Documentation/aoe/aoe.txt +++ b/Documentation/aoe/aoe.txt @@ -4,6 +4,16 @@ The EtherDrive (R) HOWTO for users of 2.6 kernels is found at ... It has many tips and hints! +The aoetools are userland programs that are designed to work with this +driver. The aoetools are on sourceforge. + + http://aoetools.sourceforge.net/ + +The scripts in this Documentation/aoe directory are intended to +document the use of the driver and are not necessary if you install +the aoetools. + + CREATING DEVICE NODES Users of udev should find the block device nodes created @@ -35,14 +45,15 @@ USING DEVICE NODES "echo eth2 eth4 > /dev/etherd/interfaces" tells the aoe driver to limit ATA over Ethernet traffic to eth2 and eth4. AoE traffic from - untrusted networks should be ignored as a matter of security. + untrusted networks should be ignored as a matter of security. See + also the aoe_iflist driver option described below. "echo > /dev/etherd/discover" tells the driver to find out what AoE devices are available. These character devices may disappear and be replaced by sysfs - counterparts, so distribution maintainers are encouraged to create - scripts that use these devices. + counterparts. Using the commands in aoetools insulates users from + these implementation details. The block devices are named like this: @@ -66,7 +77,8 @@ USING SYSFS through which we are communicating with the remote AoE device. There is a script in this directory that formats this information - in a convenient way. + in a convenient way. Users with aoetools can use the aoe-stat + command. root@makki root# sh Documentation/aoe/status.sh e10.0 eth3 up @@ -89,3 +101,23 @@ USING SYSFS e4.7 eth1 up e4.8 eth1 up e4.9 eth1 up + + Use /sys/module/aoe/parameters/aoe_iflist (or better, the driver + option discussed below) instead of /dev/etherd/interfaces to limit + AoE traffic to the network interfaces in the given + whitespace-separated list. Unlike the old character device, the + sysfs entry can be read from as well as written to. + + It's helpful to trigger discovery after setting the list of allowed + interfaces. The aoetools package provides an aoe-discover script + for this purpose. You can also directly use the + /dev/etherd/discover special file described above. + +DRIVER OPTIONS + + There is a boot option for the built-in aoe driver and a + corresponding module parameter, aoe_iflist. Without this option, + all network interfaces may be used for ATA over Ethernet. Here is a + usage example for the module parameter. + + modprobe aoe_iflist="eth1 eth3" diff --git a/Documentation/aoe/status.sh b/Documentation/aoe/status.sh index 6628116d4a9..751f3be514b 100644 --- a/Documentation/aoe/status.sh +++ b/Documentation/aoe/status.sh @@ -14,10 +14,6 @@ test ! -d "$sysd/block" && { echo "$me Error: sysfs is not mounted" 1>&2 exit 1 } -test -z "`lsmod | grep '^aoe'`" && { - echo "$me Error: aoe module is not loaded" 1>&2 - exit 1 -} for d in `ls -d $sysd/block/etherd* 2>/dev/null | grep -v p` end; do # maybe ls comes up empty, so we use "end" diff --git a/Documentation/dontdiff b/Documentation/dontdiff index 734fcc87db2..9a33bb94f74 100644 --- a/Documentation/dontdiff +++ b/Documentation/dontdiff @@ -1,137 +1,140 @@ -.* +*.a +*.aux +*.bin +*.cpio +*.css +*.dvi +*.eps +*.gif +*.grep +*.grp +*.gz +*.html +*.jpeg +*.ko +*.log +*.lst +*.mod.c +*.o +*.orig +*.out +*.pdf +*.png +*.ps +*.rej +*.s +*.sgml +*.so +*.tex +*.ver +*.xml +*_MODULES +*_vga16.c +*cscope* *~ +.* +.cscope +53c700_d.h 53c8xx_d.h* -*.a +BitKeeper +COPYING +CREDITS +CVS +ChangeSet +Kerntypes +MODS.txt +Module.symvers +PENDING +SCCS +System.map* +TAGS aic7*reg.h* -aic7*seq.h* aic7*reg_print.c* -53c700_d.h +aic7*seq.h* aicasm aicdb.h* asm asm_offsets.* autoconf.h* -*.aux bbootsect -*.bin bin2c binkernel.spec -BitKeeper bootsect bsetup btfixupprep build bvmlinux bzImage* -ChangeSet classlist.h* -compile.h* comp*.log +compile.h* config config-* config_data.h* conmakehash consolemap_deftbl.c* -COPYING -CREDITS -.cscope -*cscope* +crc32table.h* cscope.* -*.out -*.css -CVS defkeymap.c* devlist.h* docproc dummy_sym.c* -*.dvi -*.eps +elfconfig.h* filelist fixdep fore200e_mkfirm fore200e_pca_fw.c* gen-devlist -gen_init_cpio -gen_crc32table -crc32table.h* -*.cpio gen-kdb_cmds.c* -gentbl +gen_crc32table +gen_init_cpio genksyms -*.gif -*.gz -*.html +gentbl ikconfig.h* initramfs_list -*.jpeg +kallsyms kconfig kconfig.tk -Kerntypes keywords.c* ksym.c* ksym.h* -kallsyms -mk_elfconfig -elfconfig.h* -modpost -pnmtologo -logo_*.c -*.log lex.c* +logo_*.c logo_*_clut224.c logo_*_mono.c lxdialog make_times_h map +maui_boot.h +mk_elfconfig mkdep -*_MODULES -MODS.txt +mktables +modpost modversions.h* -Module.symvers -*.mod.c -*.o -*.ko -*.orig -*.lst -*.grp -*.grep +offsets.h oui.c* -mktables -raid6tables.c -raid6int*.c -raid6altivec*.c -wanxlfw.inc -maui_boot.h -pss_boot.h -trix_boot.h -*.pdf parse.c* parse.h* -PENDING +pnmtologo ppc_defs.h* promcon_tbl.c* -*.png -*.ps -*.rej -SCCS +pss_boot.h +raid6altivec*.c +raid6int*.c +raid6tables.c setup -*.s -*.so -*.sgml sim710_d.h* sm_tbl* split-include -System.map* tags -TAGS -*.tex times.h* tkparse -*.ver +trix_boot.h version.h* -*_vga16.c vmlinux -vmlinux.lds vmlinux-* +vmlinux.lds vsyscall.lds +wanxlfw.inc +uImage zImage diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 56627c1546d..d3c52dd24a2 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -48,3 +48,18 @@ When: April 2005 Why: Replaced by ->compat_ioctl in file_operations and other method vecors. Who: Andi Kleen <ak@muc.de>, Christoph Hellwig <hch@lst.de> + +--------------------------- + +What: RCU API moves to EXPORT_SYMBOL_GPL +When: April 2006 +Files: include/linux/rcupdate.h, kernel/rcupdate.c +Why: Outside of Linux, the only implementations of anything even + vaguely resembling RCU that I am aware of are in DYNIX/ptx, + VM/XA, Tornado, and K42. I do not expect anyone to port binary + drivers or kernel modules from any of these, since the first two + are owned by IBM and the last two are open-source research OSes. + So these will move to GPL after a grace period to allow + people, who might be using implementations that I am not aware + of, to adjust to this upcoming change. +Who: Paul E. McKenney <paulmck@us.ibm.com> diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index a934baeeb33..1045da582b9 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -219,8 +219,12 @@ This may also be done to avoid internal deadlocks, but rarely. If the filesytem is called for sync then it must wait on any in-progress I/O and then start new I/O. -The filesystem should unlock the page synchronously, before returning -to the caller. +The filesystem should unlock the page synchronously, before returning to the +caller, unless ->writepage() returns special WRITEPAGE_ACTIVATE +value. WRITEPAGE_ACTIVATE means that page cannot really be written out +currently, and VM should stop calling ->writepage() on this page for some +time. VM does this by moving page to the head of the active list, hence the +name. Unless the filesystem is going to redirty_page_for_writepage(), unlock the page and return zero, writepage *must* run set_page_writeback() against the page, diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index cbe85c17176..6c98f2bd421 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -909,16 +909,6 @@ nr_free_inodes Represents the number of free inodes. Ie. The number of inuse inodes is (nr_inodes - nr_free_inodes). -super-nr and super-max ----------------------- - -Again, super block structures are allocated by the kernel, but not freed. The -file super-max contains the maximum number of super block handlers, where -super-nr shows the number of currently allocated ones. - -Every mounted file system needs a super block, so if you plan to mount lots of -file systems, you may want to increase these numbers. - aio-nr and aio-max-nr --------------------- diff --git a/Documentation/i386/zero-page.txt b/Documentation/i386/zero-page.txt index 67c053a099e..df28c741678 100644 --- a/Documentation/i386/zero-page.txt +++ b/Documentation/i386/zero-page.txt @@ -79,6 +79,6 @@ Offset Type Description 0x22c unsigned long ramdisk_max 0x230 16 bytes trampoline 0x290 - 0x2cf EDD_MBR_SIG_BUFFER (edd.S) -0x2d0 - 0x600 E820MAP -0x600 - 0x7ff EDDBUF (edd.S) for disk signature read sector -0x600 - 0x7eb EDDBUF (edd.S) for edd data +0x2d0 - 0xd00 E820MAP +0xd00 - 0xeff EDDBUF (edd.S) for disk signature read sector +0xd00 - 0xeeb EDDBUF (edd.S) for edd data diff --git a/Documentation/networking/DLINK.txt b/Documentation/networking/DLINK.txt index 083d24752b8..55d24433d15 100644 --- a/Documentation/networking/DLINK.txt +++ b/Documentation/networking/DLINK.txt @@ -178,10 +178,9 @@ Released 1994-06-13 7. ACKNOWLEDGMENTS. These drivers wouldn't have been done without the base - (and support) from Ross Biro <bir7@leland.stanford.edu>, - and D-Link Systems Inc. The driver relies upon GPL-ed - source from D-Link Systems Inc. and from Russel Nelson at - Crynwr Software <nelson@crynwr.com>. + (and support) from Ross Biro, and D-Link Systems Inc. + The driver relies upon GPL-ed source from D-Link Systems Inc. + and from Russel Nelson at Crynwr Software <nelson@crynwr.com>. Additional input also from: Donald Becker <becker@super.org>, Alan Cox <A.Cox@swansea.ac.uk> diff --git a/Documentation/networking/netdevices.txt b/Documentation/networking/netdevices.txt index 1509f3aff96..3c0a5ba614d 100644 --- a/Documentation/networking/netdevices.txt +++ b/Documentation/networking/netdevices.txt @@ -51,6 +51,8 @@ dev->hard_start_xmit: set_multicast_list Context: BHs disabled Notes: netif_queue_stopped() is guaranteed false + Interrupts must be enabled when calling hard_start_xmit. + (Interrupts must also be enabled when enabling the BH handler.) Return codes: o NETDEV_TX_OK everything ok. o NETDEV_TX_BUSY Cannot transmit packet, try later diff --git a/Documentation/pci.txt b/Documentation/pci.txt index 67514bf87cc..62b1dc5d97e 100644 --- a/Documentation/pci.txt +++ b/Documentation/pci.txt @@ -279,6 +279,7 @@ pci_for_each_dev_reverse() Superseded by pci_find_device_reverse() pci_for_each_bus() Superseded by pci_find_next_bus() pci_find_device() Superseded by pci_get_device() pci_find_subsys() Superseded by pci_get_subsys() +pci_find_slot() Superseded by pci_get_slot() pcibios_find_class() Superseded by pci_get_class() pci_find_class() Superseded by pci_get_class() pci_(read|write)_*_nodev() Superseded by pci_bus_(read|write)_*() diff --git a/Documentation/power/pci.txt b/Documentation/power/pci.txt index c85428e7ad9..35b1a7dae34 100644 --- a/Documentation/power/pci.txt +++ b/Documentation/power/pci.txt @@ -165,40 +165,9 @@ Description: These functions are intended for use by individual drivers, and are defined in struct pci_driver: - int (*save_state) (struct pci_dev *dev, u32 state); - int (*suspend) (struct pci_dev *dev, u32 state); + int (*suspend) (struct pci_dev *dev, pm_message_t state); int (*resume) (struct pci_dev *dev); - int (*enable_wake) (struct pci_dev *dev, u32 state, int enable); - - -save_state ----------- - -Usage: - -if (dev->driver && dev->driver->save_state) - dev->driver->save_state(dev,state); - -The driver should use this callback to save device state. It should take into -account the current state of the device and the requested state in order to -avoid any unnecessary operations. - -For example, a video card that supports all 4 states (D0-D3), all controller -context is preserved when entering D1, but the screen is placed into a low power -state (blanked). - -The driver can also interpret this function as a notification that it may be -entering a sleep state in the near future. If it knows that the device cannot -enter the requested state, either because of lack of support for it, or because -the device is middle of some critical operation, then it should fail. - -This function should not be used to set any state in the device or the driver -because the device may not actually enter the sleep state (e.g. another driver -later causes causes a global state transition to fail). - -Note that in intermediate low power states, a device's I/O and memory spaces may -be disabled and may not be available in subsequent transitions to lower power -states. + int (*enable_wake) (struct pci_dev *dev, pci_power_t state, int enable); suspend diff --git a/Documentation/s390/cds.txt b/Documentation/s390/cds.txt index d9397170fb3..f0be389c711 100644 --- a/Documentation/s390/cds.txt +++ b/Documentation/s390/cds.txt @@ -56,12 +56,16 @@ read_dev_chars() read device characteristics read_conf_data() +read_conf_data_lpm() read configuration data. ccw_device_get_ciw() get commands from extended sense data. ccw_device_start() +ccw_device_start_timeout() +ccw_device_start_key() +ccw_device_start_key_timeout() initiate an I/O request. ccw_device_resume() @@ -197,19 +201,21 @@ The read_dev_chars() function returns : operational. -read_conf_data() - Read Configuration Data +read_conf_data(), read_conf_data_lpm() - Read Configuration Data Retrieve the device dependent configuration data. Please have a look at your device dependent I/O commands for the device specific layout of the node -descriptor elements. +descriptor elements. read_conf_data_lpm() will retrieve the configuration data +for a specific path. -The function is meant to be called with an irq handler in place; that is, +The function is meant to be called with the device already enabled; that is, at earliest during set_online() processing. The function may be called enabled or disabled, but the device must not be locked -int read_conf_data(struct ccw_device, void **buffer, int *length, __u8 lpm); +int read_conf_data(struct ccw_device, void **buffer, int *length); +int read_conf_data_lpm(struct ccw_device, void **buffer, int *length, __u8 lpm); cdev - the ccw_device the data is requested for. buffer - Pointer to a buffer pointer. The read_conf_data() routine @@ -263,6 +269,25 @@ int ccw_device_start(struct ccw_device *cdev, unsigned long intparm, __u8 lpm, unsigned long flags); +int ccw_device_start_timeout(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + unsigned long flags, + int expires); +int ccw_device_start_key(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + __u8 key, + unsigned long flags); +int ccw_device_start_key_timeout(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + __u8 key, + unsigned long flags, + int expires); cdev : ccw_device the I/O is destined for cpa : logical start address of channel program @@ -272,7 +297,12 @@ user_intparm : user specific interrupt information; will be presented particular I/O request. lpm : defines the channel path to be used for a specific I/O request. A value of 0 will make cio use the opm. +key : the storage key to use for the I/O (useful for operating on a + storage with a storage key != default key) flag : defines the action to be performed for I/O processing +expires : timeout value in jiffies. The common I/O layer will terminate + the running program after this and call the interrupt handler + with ERR_PTR(-ETIMEDOUT) as irb. Possible flag values are : @@ -327,6 +357,13 @@ current (last) I/O request. In case of a delayed status notification no special interrupt will be presented to indicate I/O completion as the I/O request was never started, even though ccw_device_start() returned with successful completion. +The irb may contain an error value, and the device driver should check for this +first: + +-ETIMEDOUT: the common I/O layer terminated the request after the specified + timeout value +-EIO: the common I/O layer terminated the request due to an error state + If the concurrent sense flag in the extended status word in the irb is set, the field irb->scsw.count describes the numer of device specific sense bytes available in the extended control word irb->scsw.ecw[0]. No device sensing by diff --git a/MAINTAINERS b/MAINTAINERS index 4333b69e56b..5b8483334de 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1699,7 +1699,9 @@ P: Pavel Roskin M: proski@gnu.org P: David Gibson M: hermes@gibson.dropbear.id.au -W: http://www.ozlabs.org/people/dgibson/dldwd +L: orinoco-users@lists.sourceforge.net +L: orinoco-devel@lists.sourceforge.net +W: http://www.nongnu.org/orinoco/ S: Maintained PARALLEL PORT SUPPORT @@ -2113,6 +2115,13 @@ M: perex@suse.cz L: alsa-devel@alsa-project.org S: Maintained +TPM DEVICE DRIVER +P: Kylene Hall +M: kjhall@us.ibm.com +W: http://tpmdd.sourceforge.net +L: tpmdd-devel@lists.sourceforge.net +S: Maintained + UltraSPARC (sparc64): P: David S. Miller M: davem@davemloft.net @@ -332,9 +332,7 @@ KALLSYMS = scripts/kallsyms PERL = perl CHECK = sparse -NOSTDINC_FLAGS = -nostdinc -isystem $(shell $(CC) -print-file-name=include) CHECKFLAGS := -D__linux__ -Dlinux -D__STDC__ -Dunix -D__unix__ -CHECKFLAGS += $(NOSTDINC_FLAGS) MODFLAGS = -DMODULE CFLAGS_MODULE = $(MODFLAGS) AFLAGS_MODULE = $(MODFLAGS) @@ -531,6 +529,10 @@ endif include $(srctree)/arch/$(ARCH)/Makefile +# arch Makefile may override CC so keep this after arch Makefile is included +NOSTDINC_FLAGS := -nostdinc -isystem $(shell $(CC) -print-file-name=include) +CHECKFLAGS += $(NOSTDINC_FLAGS) + # warn about C99 declaration after statement CFLAGS += $(call cc-option,-Wdeclaration-after-statement,) @@ -1188,8 +1190,8 @@ cmd_TAGS = $(all-sources) | etags - quiet_cmd_tags = MAKE $@ define cmd_tags rm -f $@; \ - CTAGSF=`ctags --version | grep -i exuberant >/dev/null && echo "-I __initdata,__exitdata,EXPORT_SYMBOL,EXPORT_SYMBOL_GPL"`; \ - $(all-sources) | xargs ctags $$CTAGSF -a --extra=+f + CTAGSF=`ctags --version | grep -i exuberant >/dev/null && echo "-I __initdata,__exitdata,EXPORT_SYMBOL,EXPORT_SYMBOL_GPL --extra=+f"`; \ + $(all-sources) | xargs ctags $$CTAGSF -a endef TAGS: FORCE diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 0c79b9d95f7..f7c96635d3b 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -280,6 +280,10 @@ config ISA (MCA) or VESA. ISA is an older system, now being displaced by PCI; newer boards don't support it. If you have ISA, say Y, otherwise N. +config ISA_DMA_API + bool + default y + config PCI bool depends on !ALPHA_JENSEN diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index b5d0fd2bb10..64e450dddb4 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -457,22 +457,6 @@ osf_getdomainname(char __user *name, int namelen) return 0; } -asmlinkage long -osf_shmat(int shmid, void __user *shmaddr, int shmflg) -{ - unsigned long raddr; - long err; - - err = do_shmat(shmid, shmaddr, shmflg, &raddr); - - /* - * This works because all user-level addresses are - * non-negative longs! - */ - return err ? err : (long)raddr; -} - - /* * The following stuff should move into a header file should it ever * be labeled "officially supported." Right now, there is just enough diff --git a/arch/alpha/kernel/ptrace.c b/arch/alpha/kernel/ptrace.c index d0058316157..bbd37536d14 100644 --- a/arch/alpha/kernel/ptrace.c +++ b/arch/alpha/kernel/ptrace.c @@ -14,6 +14,7 @@ #include <linux/user.h> #include <linux/slab.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -335,7 +336,7 @@ do_sys_ptrace(long request, long pid, long addr, long data, /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -365,7 +366,7 @@ do_sys_ptrace(long request, long pid, long addr, long data, case PTRACE_SINGLESTEP: /* execute single instruction. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; /* Mark single stepping. */ child->thread_info->bpt_nsaved = -1; diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S index 3864b33562e..05212088287 100644 --- a/arch/alpha/kernel/systbls.S +++ b/arch/alpha/kernel/systbls.S @@ -227,7 +227,7 @@ sys_call_table: .quad sys_semop .quad osf_utsname .quad sys_lchown - .quad osf_shmat + .quad sys_shmat .quad sys_shmctl /* 210 */ .quad sys_shmdt .quad sys_shmget diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 4055115ae0e..bf397a9f8ac 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -85,6 +85,7 @@ choice config ARCH_CLPS7500 bool "Cirrus-CL-PS7500FE" select TIMER_ACORN + select ISA config ARCH_CLPS711X bool "CLPS711x/EP721x-based" @@ -96,6 +97,7 @@ config ARCH_CO285 config ARCH_EBSA110 bool "EBSA-110" + select ISA help This is an evaluation board for the StrongARM processor available from Digital. It has limited hardware on-board, including an onboard @@ -120,13 +122,16 @@ config ARCH_INTEGRATOR config ARCH_IOP3XX bool "IOP3xx-based" + select PCI config ARCH_IXP4XX bool "IXP4xx-based" select DMABOUNCE + select PCI config ARCH_IXP2000 bool "IXP2400/2800-based" + select PCI config ARCH_L7200 bool "LinkUp-L7200" @@ -155,6 +160,8 @@ config ARCH_RPC config ARCH_SA1100 bool "SA1100-based" + select ISA + select DISCONTIGMEM config ARCH_S3C2410 bool "Samsung S3C2410" @@ -165,6 +172,9 @@ config ARCH_S3C2410 config ARCH_SHARK bool "Shark" + select ISA + select ISA_DMA + select PCI config ARCH_LH7A40X bool "Sharp LH7A40X" @@ -252,8 +262,6 @@ config ARM_AMBA config ISA bool - depends on FOOTBRIDGE_HOST || ARCH_SHARK || ARCH_CLPS7500 || ARCH_EBSA110 || ARCH_CDB89712 || ARCH_EDB7211 || ARCH_SA1100 || ARCH_MX1ADS - default y help Find out whether you have ISA slots on your motherboard. ISA is the name of a bus system, i.e. the way the CPU talks to the other stuff @@ -263,12 +271,13 @@ config ISA config ISA_DMA bool - depends on FOOTBRIDGE_HOST || ARCH_SHARK + +config ISA_DMA_API + bool default y config PCI bool "PCI support" if ARCH_INTEGRATOR_AP - default y if ARCH_SHARK || FOOTBRIDGE_HOST || ARCH_IOP3XX || ARCH_IXP4XX || ARCH_IXP2000 help Find out whether you have a PCI motherboard. PCI is the name of a bus system, i.e. the way the CPU talks to the other stuff inside @@ -296,7 +305,7 @@ menu "Kernel Features" config SMP bool "Symmetric Multi-Processing (EXPERIMENTAL)" - depends on EXPERIMENTAL && n + depends on EXPERIMENTAL #&& n help This enables support for systems with more than one CPU. If you have a system with only one CPU, like most personal computers, say N. If @@ -336,8 +345,7 @@ config PREEMPT config DISCONTIGMEM bool - depends on ARCH_EDB7211 || ARCH_SA1100 || (ARCH_LH7A40X && !LH7A40X_CONTIGMEM) - default y + default (ARCH_LH7A40X && !LH7A40X_CONTIGMEM) help Say Y to support efficient handling of discontiguous physical memory, for architectures which are either NUMA (Non-Uniform Memory Access) diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S index c0e7aff3dec..7c7f475e213 100644 --- a/arch/arm/boot/compressed/head.S +++ b/arch/arm/boot/compressed/head.S @@ -18,48 +18,30 @@ * Please select one of the following when turning on debugging. */ #ifdef DEBUG -#if defined(CONFIG_DEBUG_DC21285_PORT) - .macro loadsp, rb - mov \rb, #0x42000000 - .endm - .macro writeb, rb - str \rb, [r3, #0x160] - .endm -#elif defined(CONFIG_DEBUG_ICEDCC) + +#include <asm/arch/debug-macro.S> + +#if defined(CONFIG_DEBUG_ICEDCC) .macro loadsp, rb .endm - .macro writeb, rb - mcr p14, 0, \rb, c0, c1, 0 - .endm -#elif defined(CONFIG_FOOTBRIDGE) - .macro loadsp, rb - mov \rb, #0x7c000000 + .macro writeb, ch, rb + mcr p14, 0, \ch, c0, c1, 0 .endm - .macro writeb, rb - strb \rb, [r3, #0x3f8] +#else + .macro writeb, ch, rb + senduart \ch, \rb .endm -#elif defined(CONFIG_ARCH_RPC) + +#if defined(CONFIG_FOOTBRIDGE) || \ + defined(CONFIG_ARCH_RPC) || \ + defined(CONFIG_ARCH_INTEGRATOR) || \ + defined(CONFIG_ARCH_PXA) || \ + defined(CONFIG_ARCH_IXP4XX) || \ + defined(CONFIG_ARCH_IXP2000) || \ + defined(CONFIG_ARCH_LH7A40X) || \ + defined(CONFIG_ARCH_OMAP) .macro loadsp, rb - mov \rb, #0x03000000 - orr \rb, \rb, #0x00010000 - .endm - .macro writeb, rb - strb \rb, [r3, #0x3f8 << 2] - .endm -#elif defined(CONFIG_ARCH_INTEGRATOR) - .macro loadsp, rb - mov \rb, #0x16000000 - .endm - .macro writeb, rb - strb \rb, [r3, #0] - .endm -#elif defined(CONFIG_ARCH_PXA) /* Xscale-type */ - .macro loadsp, rb - mov \rb, #0x40000000 - orr \rb, \rb, #0x00100000 - .endm - .macro writeb, rb - strb \rb, [r3, #0] + addruart \rb .endm #elif defined(CONFIG_ARCH_SA1100) .macro loadsp, rb @@ -70,65 +52,22 @@ add \rb, \rb, #0x00010000 @ Ser1 # endif .endm - .macro writeb, rb - str \rb, [r3, #0x14] @ UTDR - .endm -#elif defined(CONFIG_ARCH_IXP4XX) - .macro loadsp, rb - mov \rb, #0xc8000000 - .endm - .macro writeb, rb - str \rb, [r3, #0] -#elif defined(CONFIG_ARCH_IXP2000) - .macro loadsp, rb - mov \rb, #0xc0000000 - orr \rb, \rb, #0x00030000 - .endm - .macro writeb, rb - str \rb, [r3, #0] - .endm -#elif defined(CONFIG_ARCH_LH7A40X) - .macro loadsp, rb - ldr \rb, =0x80000700 @ UART2 UARTBASE - .endm - .macro writeb, rb - strb \rb, [r3, #0] - .endm -#elif defined(CONFIG_ARCH_OMAP) - .macro loadsp, rb - mov \rb, #0xff000000 @ physical base address - add \rb, \rb, #0x00fb0000 -#if defined(CONFIG_OMAP_LL_DEBUG_UART2) || defined(CONFIG_OMAP_LL_DEBUG_UART3) - add \rb, \rb, #0x00000800 -#endif -#ifdef CONFIG_OMAP_LL_DEBUG_UART3 - add \rb, \rb, #0x00009000 -#endif - .endm - .macro writeb, rb - strb \rb, [r3] - .endm #elif defined(CONFIG_ARCH_IOP331) .macro loadsp, rb mov \rb, #0xff000000 orr \rb, \rb, #0x00ff0000 orr \rb, \rb, #0x0000f700 @ location of the UART .endm - .macro writeb, rb - str \rb, [r3, #0] - .endm #elif defined(CONFIG_ARCH_S3C2410) - .macro loadsp, rb + .macro loadsp, rb mov \rb, #0x50000000 add \rb, \rb, #0x4000 * CONFIG_S3C2410_LOWLEVEL_UART_PORT .endm - .macro writeb, rb - strb \rb, [r3, #0x20] - .endm #else #error no serial architecture defined #endif #endif +#endif .macro kputc,val mov r0, \val @@ -734,7 +673,7 @@ puts: loadsp r3 1: ldrb r2, [r0], #1 teq r2, #0 moveq pc, lr -2: writeb r2 +2: writeb r2, r3 mov r1, #0x00020000 3: subs r1, r1, #1 bne 3b diff --git a/arch/arm/common/rtctime.c b/arch/arm/common/rtctime.c index c397e71f938..72b03f201eb 100644 --- a/arch/arm/common/rtctime.c +++ b/arch/arm/common/rtctime.c @@ -141,10 +141,10 @@ void rtc_next_alarm_time(struct rtc_time *next, struct rtc_time *now, struct rtc next->tm_sec = alrm->tm_sec; } -static inline void rtc_read_time(struct rtc_ops *ops, struct rtc_time *tm) +static inline int rtc_read_time(struct rtc_ops *ops, struct rtc_time *tm) { memset(tm, 0, sizeof(struct rtc_time)); - ops->read_time(tm); + return ops->read_time(tm); } static inline int rtc_set_time(struct rtc_ops *ops, struct rtc_time *tm) @@ -163,8 +163,7 @@ static inline int rtc_read_alarm(struct rtc_ops *ops, struct rtc_wkalrm *alrm) int ret = -EINVAL; if (ops->read_alarm) { memset(alrm, 0, sizeof(struct rtc_wkalrm)); - ops->read_alarm(alrm); - ret = 0; + ret = ops->read_alarm(alrm); } return ret; } @@ -283,7 +282,9 @@ static int rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd, break; case RTC_RD_TIME: - rtc_read_time(ops, &tm); + ret = rtc_read_time(ops, &tm); + if (ret) + break; ret = copy_to_user(uarg, &tm, sizeof(tm)); if (ret) ret = -EFAULT; @@ -424,15 +425,15 @@ static int rtc_read_proc(char *page, char **start, off_t off, int count, int *eo struct rtc_time tm; char *p = page; - rtc_read_time(ops, &tm); - - p += sprintf(p, - "rtc_time\t: %02d:%02d:%02d\n" - "rtc_date\t: %04d-%02d-%02d\n" - "rtc_epoch\t: %04lu\n", - tm.tm_hour, tm.tm_min, tm.tm_sec, - tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday, - rtc_epoch); + if (rtc_read_time(ops, &tm) == 0) { + p += sprintf(p, + "rtc_time\t: %02d:%02d:%02d\n" + "rtc_date\t: %04d-%02d-%02d\n" + "rtc_epoch\t: %04lu\n", + tm.tm_hour, tm.tm_min, tm.tm_sec, + tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday, + rtc_epoch); + } if (rtc_read_alarm(ops, &alrm) == 0) { p += sprintf(p, "alrm_time\t: "); diff --git a/arch/arm/configs/ixdp2800_defconfig b/arch/arm/configs/ixdp2800_defconfig index d36f9919296..7be3521f91f 100644 --- a/arch/arm/configs/ixdp2800_defconfig +++ b/arch/arm/configs/ixdp2800_defconfig @@ -133,7 +133,7 @@ CONFIG_ALIGNMENT_TRAP=y # CONFIG_ZBOOT_ROM_TEXT=0x0 CONFIG_ZBOOT_ROM_BSS=0x0 -CONFIG_CMDLINE="console=ttyS0,9600 root=/dev/nfs ip=bootp mem=64M@0x0 pci=firmware" +CONFIG_CMDLINE="console=ttyS0,9600 root=/dev/nfs ip=bootp mem=64M@0x0" # CONFIG_XIP_KERNEL is not set # diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index 07a56ff6149..4a2af55e134 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -31,8 +31,3 @@ head-y := head.o obj-$(CONFIG_DEBUG_LL) += debug.o extra-y := $(head-y) init_task.o vmlinux.lds - -# Spell out some dependencies that aren't automatically figured out -$(obj)/entry-armv.o: $(obj)/entry-header.S include/asm-arm/constants.h -$(obj)/entry-common.o: $(obj)/entry-header.S include/asm-arm/constants.h \ - $(obj)/calls.S diff --git a/arch/arm/kernel/asm-offsets.c b/arch/arm/kernel/asm-offsets.c index 99d43259ff8..c1ff4d1f1bf 100644 --- a/arch/arm/kernel/asm-offsets.c +++ b/arch/arm/kernel/asm-offsets.c @@ -64,6 +64,26 @@ int main(void) DEFINE(TI_VFPSTATE, offsetof(struct thread_info, vfpstate)); DEFINE(TI_IWMMXT_STATE, (offsetof(struct thread_info, fpstate)+4)&~7); BLANK(); + DEFINE(S_R0, offsetof(struct pt_regs, ARM_r0)); + DEFINE(S_R1, offsetof(struct pt_regs, ARM_r1)); + DEFINE(S_R2, offsetof(struct pt_regs, ARM_r2)); + DEFINE(S_R3, offsetof(struct pt_regs, ARM_r3)); + DEFINE(S_R4, offsetof(struct pt_regs, ARM_r4)); + DEFINE(S_R5, offsetof(struct pt_regs, ARM_r5)); + DEFINE(S_R6, offsetof(struct pt_regs, ARM_r6)); + DEFINE(S_R7, offsetof(struct pt_regs, ARM_r7)); + DEFINE(S_R8, offsetof(struct pt_regs, ARM_r8)); + DEFINE(S_R9, offsetof(struct pt_regs, ARM_r9)); + DEFINE(S_R10, offsetof(struct pt_regs, ARM_r10)); + DEFINE(S_FP, offsetof(struct pt_regs, ARM_fp)); + DEFINE(S_IP, offsetof(struct pt_regs, ARM_ip)); + DEFINE(S_SP, offsetof(struct pt_regs, ARM_sp)); + DEFINE(S_LR, offsetof(struct pt_regs, ARM_lr)); + DEFINE(S_PC, offsetof(struct pt_regs, ARM_pc)); + DEFINE(S_PSR, offsetof(struct pt_regs, ARM_cpsr)); + DEFINE(S_OLD_R0, offsetof(struct pt_regs, ARM_ORIG_r0)); + DEFINE(S_FRAME_SIZE, sizeof(struct pt_regs)); + BLANK(); #if __LINUX_ARM_ARCH__ >= 6 DEFINE(MM_CONTEXT_ID, offsetof(struct mm_struct, context.id)); BLANK(); diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index bb27c317d94..4eb36155dc9 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -14,12 +14,12 @@ * it to save wrong values... Be aware! */ #include <linux/config.h> -#include <linux/init.h> -#include <asm/thread_info.h> #include <asm/glue.h> -#include <asm/ptrace.h> #include <asm/vfpmacros.h> +#include <asm/hardware.h> /* should be moved into entry-macro.S */ +#include <asm/arch/irqs.h> /* should be moved into entry-macro.S */ +#include <asm/arch/entry-macro.S> #include "entry-header.S" @@ -118,7 +118,7 @@ __dabt_svc: @ @ IRQs off again before pulling preserved data off the stack @ - disable_irq r0 + disable_irq @ @ restore SPSR and restart the instruction @@ -198,7 +198,7 @@ __und_svc: @ @ IRQs off again before pulling preserved data off the stack @ -1: disable_irq r0 +1: disable_irq @ @ restore SPSR and restart the instruction @@ -232,7 +232,7 @@ __pabt_svc: @ @ IRQs off again before pulling preserved data off the stack @ - disable_irq r0 + disable_irq @ @ restore SPSR and restart the instruction @@ -269,6 +269,12 @@ __pabt_svc: add r5, sp, #S_PC ldmia r7, {r2 - r4} @ Get USR pc, cpsr +#if __LINUX_ARM_ARCH__ < 6 + @ make sure our user space atomic helper is aborted + cmp r2, #VIRT_OFFSET + bichs r3, r3, #PSR_Z_BIT +#endif + @ @ We are now ready to fill in the remaining blanks on the stack: @ @@ -316,7 +322,7 @@ __dabt_usr: @ @ IRQs on, then call the main handler @ - enable_irq r2 + enable_irq mov r2, sp adr lr, ret_from_exception b do_DataAbort @@ -418,7 +424,7 @@ call_fpe: movcss r7, r5, lsr #(TIF_USING_IWMMXT + 1) bcs iwmmxt_task_enable #endif - enable_irq r7 + enable_irq add pc, pc, r8, lsr #6 mov r0, r0 @@ -472,7 +478,7 @@ fpundefinstr: __pabt_usr: usr_entry abt - enable_irq r0 @ Enable interrupts + enable_irq @ Enable interrupts mov r0, r2 @ address (pc) mov r1, sp @ regs bl do_PrefetchAbort @ call abort handler @@ -499,8 +505,12 @@ ENTRY(__switch_to) mra r4, r5, acc0 stmia ip, {r4, r5} #endif +#if defined(CONFIG_HAS_TLS_REG) + mcr p15, 0, r3, c13, c0, 3 @ set TLS register +#elif !defined(CONFIG_TLS_REG_EMUL) mov r4, #0xffff0fff - str r3, [r4, #-3] @ Set TLS ptr + str r3, [r4, #-15] @ TLS val at 0xffff0ff0 +#endif mcr p15, 0, r6, c3, c0, 0 @ Set domain register #ifdef CONFIG_VFP @ Always disable VFP so we can lazily save/restore the old @@ -519,11 +529,209 @@ ENTRY(__switch_to) ldmib r2, {r4 - sl, fp, sp, pc} @ Load all regs saved previously __INIT + +/* + * User helpers. + * + * These are segment of kernel provided user code reachable from user space + * at a fixed address in kernel memory. This is used to provide user space + * with some operations which require kernel help because of unimplemented + * native feature and/or instructions in many ARM CPUs. The idea is for + * this code to be executed directly in user mode for best efficiency but + * which is too intimate with the kernel counter part to be left to user + * libraries. In fact this code might even differ from one CPU to another + * depending on the available instruction set and restrictions like on + * SMP systems. In other words, the kernel reserves the right to change + * this code as needed without warning. Only the entry points and their + * results are guaranteed to be stable. + * + * Each segment is 32-byte aligned and will be moved to the top of the high + * vector page. New segments (if ever needed) must be added in front of + * existing ones. This mechanism should be used only for things that are + * really small and justified, and not be abused freely. + * + * User space is expected to implement those things inline when optimizing + * for a processor that has the necessary native support, but only if such + * resulting binaries are already to be incompatible with earlier ARM + * processors due to the use of unsupported instructions other than what + * is provided here. In other words don't make binaries unable to run on + * earlier processors just for the sake of not using these kernel helpers + * if your compiled code is not going to use the new instructions for other + * purpose. + */ + + .align 5 + .globl __kuser_helper_start +__kuser_helper_start: + +/* + * Reference prototype: + * + * int __kernel_cmpxchg(int oldval, int newval, int *ptr) + * + * Input: + * + * r0 = oldval + * r1 = newval + * r2 = ptr + * lr = return address + * + * Output: + * + * r0 = returned value (zero or non-zero) + * C flag = set if r0 == 0, clear if r0 != 0 + * + * Clobbered: + * + * r3, ip, flags + * + * Definition and user space usage example: + * + * typedef int (__kernel_cmpxchg_t)(int oldval, int newval, int *ptr); + * #define __kernel_cmpxchg (*(__kernel_cmpxchg_t *)0xffff0fc0) + * + * Atomically store newval in *ptr if *ptr is equal to oldval for user space. + * Return zero if *ptr was changed or non-zero if no exchange happened. + * The C flag is also set if *ptr was changed to allow for assembly + * optimization in the calling code. + * + * For example, a user space atomic_add implementation could look like this: + * + * #define atomic_add(ptr, val) \ + * ({ register unsigned int *__ptr asm("r2") = (ptr); \ + * register unsigned int __result asm("r1"); \ + * asm volatile ( \ + * "1: @ atomic_add\n\t" \ + * "ldr r0, [r2]\n\t" \ + * "mov r3, #0xffff0fff\n\t" \ + * "add lr, pc, #4\n\t" \ + * "add r1, r0, %2\n\t" \ + * "add pc, r3, #(0xffff0fc0 - 0xffff0fff)\n\t" \ + * "bcc 1b" \ + * : "=&r" (__result) \ + * : "r" (__ptr), "rIL" (val) \ + * : "r0","r3","ip","lr","cc","memory" ); \ + * __result; }) + */ + +__kuser_cmpxchg: @ 0xffff0fc0 + +#if __LINUX_ARM_ARCH__ < 6 + +#ifdef CONFIG_SMP /* sanity check */ +#error "CONFIG_SMP on a machine supporting pre-ARMv6 processors?" +#endif + + /* + * Theory of operation: + * + * We set the Z flag before loading oldval. If ever an exception + * occurs we can not be sure the loaded value will still be the same + * when the exception returns, therefore the user exception handler + * will clear the Z flag whenever the interrupted user code was + * actually from the kernel address space (see the usr_entry macro). + * + * The post-increment on the str is used to prevent a race with an + * exception happening just after the str instruction which would + * clear the Z flag although the exchange was done. + */ + teq ip, ip @ set Z flag + ldr ip, [r2] @ load current val + add r3, r2, #1 @ prepare store ptr + teqeq ip, r0 @ compare with oldval if still allowed + streq r1, [r3, #-1]! @ store newval if still allowed + subs r0, r2, r3 @ if r2 == r3 the str occured + mov pc, lr + +#else + + ldrex r3, [r2] + subs r3, r3, r0 + strexeq r3, r1, [r2] + rsbs r0, r3, #0 + mov pc, lr + +#endif + + .align 5 + +/* + * Reference prototype: + * + * int __kernel_get_tls(void) + * + * Input: + * + * lr = return address + * + * Output: + * + * r0 = TLS value + * + * Clobbered: + * + * the Z flag might be lost + * + * Definition and user space usage example: + * + * typedef int (__kernel_get_tls_t)(void); + * #define __kernel_get_tls (*(__kernel_get_tls_t *)0xffff0fe0) + * + * Get the TLS value as previously set via the __ARM_NR_set_tls syscall. + * + * This could be used as follows: + * + * #define __kernel_get_tls() \ + * ({ register unsigned int __val asm("r0"); \ + * asm( "mov r0, #0xffff0fff; mov lr, pc; sub pc, r0, #31" \ + * : "=r" (__val) : : "lr","cc" ); \ + * __val; }) + */ + +__kuser_get_tls: @ 0xffff0fe0 + +#if !defined(CONFIG_HAS_TLS_REG) && !defined(CONFIG_TLS_REG_EMUL) + + ldr r0, [pc, #(16 - 8)] @ TLS stored at 0xffff0ff0 + mov pc, lr + +#else + + mrc p15, 0, r0, c13, c0, 3 @ read TLS register + mov pc, lr + +#endif + + .rep 5 + .word 0 @ pad up to __kuser_helper_version + .endr + +/* + * Reference declaration: + * + * extern unsigned int __kernel_helper_version; + * + * Definition and user space usage example: + * + * #define __kernel_helper_version (*(unsigned int *)0xffff0ffc) + * + * User space may read this to determine the curent number of helpers + * available. + */ + +__kuser_helper_version: @ 0xffff0ffc + .word ((__kuser_helper_end - __kuser_helper_start) >> 5) + + .globl __kuser_helper_end +__kuser_helper_end: + + /* * Vector stubs. * - * This code is copied to 0x200 or 0xffff0200 so we can use branches in the - * vectors, rather than ldr's. + * This code is copied to 0xffff0200 so we can use branches in the + * vectors, rather than ldr's. Note that this code must not + * exceed 0x300 bytes. * * Common stub entry macro: * Enter in IRQ mode, spsr = SVC/USR CPSR, lr = SVC/USR PC @@ -544,7 +752,7 @@ vector_\name: @ mrs r13, cpsr bic r13, r13, #MODE_MASK - orr r13, r13, #MODE_SVC + orr r13, r13, #SVC_MODE msr spsr_cxsf, r13 @ switch to SVC_32 mode and lr, lr, #15 @@ -552,6 +760,7 @@ vector_\name: movs pc, lr @ Changes mode and branches .endm + .globl __stubs_start __stubs_start: /* * Interrupt dispatcher @@ -686,37 +895,24 @@ vector_addrexcptn: .LCsabt: .word __temp_abt + .globl __stubs_end __stubs_end: - .equ __real_stubs_start, .LCvectors + 0x200 + .equ stubs_offset, __vectors_start + 0x200 - __stubs_start -.LCvectors: + .globl __vectors_start +__vectors_start: swi SYS_ERROR0 - b __real_stubs_start + (vector_und - __stubs_start) - ldr pc, __real_stubs_start + (.LCvswi - __stubs_start) - b __real_stubs_start + (vector_pabt - __stubs_start) - b __real_stubs_start + (vector_dabt - __stubs_start) - b __real_stubs_start + (vector_addrexcptn - __stubs_start) - b __real_stubs_start + (vector_irq - __stubs_start) - b __real_stubs_start + (vector_fiq - __stubs_start) - -ENTRY(__trap_init) - stmfd sp!, {r4 - r6, lr} - - mov r0, #0xff000000 - orr r0, r0, #0x00ff0000 @ high vectors position - adr r1, .LCvectors @ set up the vectors - ldmia r1, {r1, r2, r3, r4, r5, r6, ip, lr} - stmia r0, {r1, r2, r3, r4, r5, r6, ip, lr} - - add r2, r0, #0x200 - adr r0, __stubs_start @ copy stubs to 0x200 - adr r1, __stubs_end -1: ldr r3, [r0], #4 - str r3, [r2], #4 - cmp r0, r1 - blt 1b - LOADREGS(fd, sp!, {r4 - r6, pc}) + b vector_und + stubs_offset + ldr pc, .LCvswi + stubs_offset + b vector_pabt + stubs_offset + b vector_dabt + stubs_offset + b vector_addrexcptn + stubs_offset + b vector_irq + stubs_offset + b vector_fiq + stubs_offset + + .globl __vectors_end +__vectors_end: .data diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index 53a7e0dea44..3f8d0e3aefa 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -9,19 +9,10 @@ */ #include <linux/config.h> -#include <asm/thread_info.h> -#include <asm/ptrace.h> #include <asm/unistd.h> #include "entry-header.S" -/* - * We rely on the fact that R0 is at the bottom of the stack (due to - * slow/fast restore user regs). - */ -#if S_R0 != 0 -#error "Please fix" -#endif .align 5 /* @@ -30,11 +21,19 @@ * stack. */ ret_fast_syscall: - disable_irq r1 @ disable interrupts + disable_irq @ disable interrupts ldr r1, [tsk, #TI_FLAGS] tst r1, #_TIF_WORK_MASK bne fast_work_pending - fast_restore_user_regs + + @ fast_restore_user_regs + ldr r1, [sp, #S_OFF + S_PSR] @ get calling cpsr + ldr lr, [sp, #S_OFF + S_PC]! @ get pc + msr spsr_cxsf, r1 @ save in spsr_svc + ldmdb sp, {r1 - lr}^ @ get calling r1 - lr + mov r0, r0 + add sp, sp, #S_FRAME_SIZE - S_PC + movs pc, lr @ return & move spsr_svc into cpsr /* * Ok, we need to do extra processing, enter the slow path. @@ -49,7 +48,7 @@ work_pending: mov r0, sp @ 'regs' mov r2, why @ 'syscall' bl do_notify_resume - disable_irq r1 @ disable interrupts + disable_irq @ disable interrupts b no_work_pending work_resched: @@ -59,12 +58,19 @@ work_resched: */ ENTRY(ret_to_user) ret_slow_syscall: - disable_irq r1 @ disable interrupts + disable_irq @ disable interrupts ldr r1, [tsk, #TI_FLAGS] tst r1, #_TIF_WORK_MASK bne work_pending no_work_pending: - slow_restore_user_regs + @ slow_restore_user_regs + ldr r1, [sp, #S_PSR] @ get calling cpsr + ldr lr, [sp, #S_PC]! @ get pc + msr spsr_cxsf, r1 @ save in spsr_svc + ldmdb sp, {r0 - lr}^ @ get calling r1 - lr + mov r0, r0 + add sp, sp, #S_FRAME_SIZE - S_PC + movs pc, lr @ return & move spsr_svc into cpsr /* * This is how we return from a fork. @@ -116,9 +122,26 @@ ENTRY(ret_from_fork) .align 5 ENTRY(vector_swi) - save_user_regs + sub sp, sp, #S_FRAME_SIZE + stmia sp, {r0 - r12} @ Calling r0 - r12 + add r8, sp, #S_PC + stmdb r8, {sp, lr}^ @ Calling sp, lr + mrs r8, spsr @ called from non-FIQ mode, so ok. + str lr, [sp, #S_PC] @ Save calling PC + str r8, [sp, #S_PSR] @ Save CPSR + str r0, [sp, #S_OLD_R0] @ Save OLD_R0 zero_fp - get_scno + + /* + * Get the system call number. + */ +#ifdef CONFIG_ARM_THUMB + tst r8, #PSR_T_BIT @ this is SPSR from save_user_regs + addne scno, r7, #__NR_SYSCALL_BASE @ put OS number in + ldreq scno, [lr, #-4] +#else + ldr scno, [lr, #-4] @ get SWI instruction +#endif arm710_bug_check scno, ip #ifdef CONFIG_ALIGNMENT_TRAP @@ -126,14 +149,14 @@ ENTRY(vector_swi) ldr ip, [ip] mcr p15, 0, ip, c1, c0 @ update control register #endif - enable_irq ip + enable_irq str r4, [sp, #-S_OFF]! @ push fifth arg get_thread_info tsk ldr ip, [tsk, #TI_FLAGS] @ check for syscall tracing bic scno, scno, #0xff000000 @ mask off SWI op-code - eor scno, scno, #OS_NUMBER << 20 @ check OS number + eor scno, scno, #__NR_SYSCALL_BASE @ check OS number adr tbl, sys_call_table @ load syscall table pointer tst ip, #_TIF_SYSCALL_TRACE @ are we tracing syscalls? bne __sys_trace @@ -144,8 +167,8 @@ ENTRY(vector_swi) add r1, sp, #S_OFF 2: mov why, #0 @ no longer a real syscall - cmp scno, #ARMSWI_OFFSET - eor r0, scno, #OS_NUMBER << 20 @ put OS number back + cmp scno, #(__ARM_NR_BASE - __NR_SYSCALL_BASE) + eor r0, scno, #__NR_SYSCALL_BASE @ put OS number back bcs arm_syscall b sys_ni_syscall @ not private func @@ -190,7 +213,7 @@ ENTRY(sys_call_table) @ r5 = syscall table .type sys_syscall, #function sys_syscall: - eor scno, r0, #OS_NUMBER << 20 + eor scno, r0, #__NR_SYSCALL_BASE cmp scno, #__NR_syscall - __NR_SYSCALL_BASE cmpne scno, #NR_syscalls @ check range stmloia sp, {r5, r6} @ shuffle args diff --git a/arch/arm/kernel/entry-header.S b/arch/arm/kernel/entry-header.S index 4039d8c120b..a3d40a0e2b0 100644 --- a/arch/arm/kernel/entry-header.S +++ b/arch/arm/kernel/entry-header.S @@ -1,24 +1,11 @@ -#include <linux/config.h> /* for CONFIG_ARCH_xxxx */ +#include <linux/config.h> +#include <linux/init.h> #include <linux/linkage.h> #include <asm/assembler.h> #include <asm/constants.h> #include <asm/errno.h> -#include <asm/hardware.h> -#include <asm/arch/irqs.h> -#include <asm/arch/entry-macro.S> - -#ifndef MODE_SVC -#define MODE_SVC 0x13 -#endif - - .macro zero_fp -#ifdef CONFIG_FRAME_POINTER - mov fp, #0 -#endif - .endm - - .text +#include <asm/thread_info.h> @ Bad Abort numbers @ ----------------- @@ -29,113 +16,44 @@ #define BAD_IRQ 3 #define BAD_UNDEFINSTR 4 -#define PT_TRACESYS 0x00000002 - -@ OS version number used in SWIs -@ RISC OS is 0 -@ RISC iX is 8 @ -#define OS_NUMBER 9 -#define ARMSWI_OFFSET 0x000f0000 - +@ Most of the stack format comes from struct pt_regs, but with +@ the addition of 8 bytes for storing syscall args 5 and 6. @ -@ Stack format (ensured by USER_* and SVC_*) -@ -#define S_FRAME_SIZE 72 -#define S_OLD_R0 68 -#define S_PSR 64 - -#define S_PC 60 -#define S_LR 56 -#define S_SP 52 -#define S_IP 48 -#define S_FP 44 -#define S_R10 40 -#define S_R9 36 -#define S_R8 32 -#define S_R7 28 -#define S_R6 24 -#define S_R5 20 -#define S_R4 16 -#define S_R3 12 -#define S_R2 8 -#define S_R1 4 -#define S_R0 0 #define S_OFF 8 - .macro set_cpsr_c, reg, mode - msr cpsr_c, \mode +/* + * The SWI code relies on the fact that R0 is at the bottom of the stack + * (due to slow/fast restore user regs). + */ +#if S_R0 != 0 +#error "Please fix" +#endif + + .macro zero_fp +#ifdef CONFIG_FRAME_POINTER + mov fp, #0 +#endif .endm #if __LINUX_ARM_ARCH__ >= 6 - .macro disable_irq, temp + .macro disable_irq cpsid i .endm - .macro enable_irq, temp + .macro enable_irq cpsie i .endm #else - .macro disable_irq, temp - set_cpsr_c \temp, #PSR_I_BIT | MODE_SVC + .macro disable_irq + msr cpsr_c, #PSR_I_BIT | SVC_MODE .endm - .macro enable_irq, temp - set_cpsr_c \temp, #MODE_SVC + .macro enable_irq + msr cpsr_c, #SVC_MODE .endm #endif - .macro save_user_regs - sub sp, sp, #S_FRAME_SIZE - stmia sp, {r0 - r12} @ Calling r0 - r12 - add r8, sp, #S_PC - stmdb r8, {sp, lr}^ @ Calling sp, lr - mrs r8, spsr @ called from non-FIQ mode, so ok. - str lr, [sp, #S_PC] @ Save calling PC - str r8, [sp, #S_PSR] @ Save CPSR - str r0, [sp, #S_OLD_R0] @ Save OLD_R0 - .endm - - .macro restore_user_regs - ldr r1, [sp, #S_PSR] @ Get calling cpsr - disable_irq ip @ disable IRQs - ldr lr, [sp, #S_PC]! @ Get PC - msr spsr_cxsf, r1 @ save in spsr_svc - ldmdb sp, {r0 - lr}^ @ Get calling r0 - lr - mov r0, r0 - add sp, sp, #S_FRAME_SIZE - S_PC - movs pc, lr @ return & move spsr_svc into cpsr - .endm - -/* - * Must be called with IRQs already disabled. - */ - .macro fast_restore_user_regs - ldr r1, [sp, #S_OFF + S_PSR] @ get calling cpsr - ldr lr, [sp, #S_OFF + S_PC]! @ get pc - msr spsr_cxsf, r1 @ save in spsr_svc - ldmdb sp, {r1 - lr}^ @ get calling r1 - lr - mov r0, r0 - add sp, sp, #S_FRAME_SIZE - S_PC - movs pc, lr @ return & move spsr_svc into cpsr - .endm - -/* - * Must be called with IRQs already disabled. - */ - .macro slow_restore_user_regs - ldr r1, [sp, #S_PSR] @ get calling cpsr - ldr lr, [sp, #S_PC]! @ get pc - msr spsr_cxsf, r1 @ save in spsr_svc - ldmdb sp, {r0 - lr}^ @ get calling r1 - lr - mov r0, r0 - add sp, sp, #S_FRAME_SIZE - S_PC - movs pc, lr @ return & move spsr_svc into cpsr - .endm - - .macro mask_pc, rd, rm - .endm - .macro get_thread_info, rd mov \rd, sp, lsr #13 mov \rd, \rd, lsl #13 @@ -165,18 +83,3 @@ scno .req r7 @ syscall number tbl .req r8 @ syscall table pointer why .req r8 @ Linux syscall (!= 0) tsk .req r9 @ current thread_info - -/* - * Get the system call number. - */ - .macro get_scno -#ifdef CONFIG_ARM_THUMB - tst r8, #PSR_T_BIT @ this is SPSR from save_user_regs - addne scno, r7, #OS_NUMBER << 20 @ put OS number in - ldreq scno, [lr, #-4] - -#else - mask_pc lr, lr - ldr scno, [lr, #-4] @ get SWI instruction -#endif - .endm diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index 171b3e811c7..4733877296d 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -19,6 +19,7 @@ #include <asm/procinfo.h> #include <asm/ptrace.h> #include <asm/constants.h> +#include <asm/thread_info.h> #include <asm/system.h> #define PROCINFO_MMUFLAGS 8 @@ -131,7 +132,7 @@ __switch_data: .long processor_id @ r4 .long __machine_arch_type @ r5 .long cr_alignment @ r6 - .long init_thread_union+8192 @ sp + .long init_thread_union + THREAD_START_SP @ sp /* * The following fragment of code is executed with the MMU on, and uses diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 26eacd3e5de..8f146a4b475 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -256,8 +256,6 @@ static unsigned long *thread_info_head; static unsigned int nr_thread_info; #define EXTRA_TASK_STRUCT 4 -#define ll_alloc_task_struct() ((struct thread_info *) __get_free_pages(GFP_KERNEL,1)) -#define ll_free_task_struct(p) free_pages((unsigned long)(p),1) struct thread_info *alloc_thread_info(struct task_struct *task) { @@ -274,17 +272,16 @@ struct thread_info *alloc_thread_info(struct task_struct *task) } if (!thread) - thread = ll_alloc_task_struct(); + thread = (struct thread_info *) + __get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER); -#ifdef CONFIG_MAGIC_SYSRQ +#ifdef CONFIG_DEBUG_STACK_USAGE /* * The stack must be cleared if you want SYSRQ-T to * give sensible stack usage information */ - if (thread) { - char *p = (char *)thread; - memzero(p+KERNEL_STACK_SIZE, KERNEL_STACK_SIZE); - } + if (thread) + memzero(thread, THREAD_SIZE); #endif return thread; } @@ -297,7 +294,7 @@ void free_thread_info(struct thread_info *thread) thread_info_head = p; nr_thread_info += 1; } else - ll_free_task_struct(thread); + free_pages((unsigned long)thread, THREAD_SIZE_ORDER); } /* @@ -350,7 +347,7 @@ copy_thread(int nr, unsigned long clone_flags, unsigned long stack_start, struct thread_info *thread = p->thread_info; struct pt_regs *childregs; - childregs = ((struct pt_regs *)((unsigned long)thread + THREAD_SIZE - 8)) - 1; + childregs = ((struct pt_regs *)((unsigned long)thread + THREAD_START_SP)) - 1; *childregs = *regs; childregs->ARM_r0 = 0; childregs->ARM_sp = stack_start; @@ -447,15 +444,17 @@ EXPORT_SYMBOL(kernel_thread); unsigned long get_wchan(struct task_struct *p) { unsigned long fp, lr; - unsigned long stack_page; + unsigned long stack_start, stack_end; int count = 0; if (!p || p == current || p->state == TASK_RUNNING) return 0; - stack_page = 4096 + (unsigned long)p->thread_info; + stack_start = (unsigned long)(p->thread_info + 1); + stack_end = ((unsigned long)p->thread_info) + THREAD_SIZE; + fp = thread_saved_fp(p); do { - if (fp < stack_page || fp > 4092+stack_page) + if (fp < stack_start || fp > stack_end) return 0; lr = pc_pointer (((unsigned long *)fp)[-1]); if (!in_sched_functions(lr)) diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index efd7a341614..cd99b83f14c 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -19,6 +19,7 @@ #include <linux/user.h> #include <linux/security.h> #include <linux/init.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -693,7 +694,7 @@ static int do_ptrace(int request, struct task_struct *child, long addr, long dat case PTRACE_SYSCALL: case PTRACE_CONT: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -728,7 +729,7 @@ static int do_ptrace(int request, struct task_struct *child, long addr, long dat */ case PTRACE_SINGLESTEP: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; child->ptrace |= PT_SINGLESTEP; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index 7ba6342cf93..f897ce2ccf0 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -227,18 +227,6 @@ asmlinkage int sys_ipc(uint call, int first, int second, int third, } } -asmlinkage long sys_shmat(int shmid, char __user *shmaddr, int shmflg, - unsigned long __user *addr) -{ - unsigned long ret; - long err; - - err = do_shmat(shmid, shmaddr, shmflg, &ret); - if (err == 0) - err = put_user(ret, addr); - return err; -} - /* Fork a new task - this creates a new program thread. * This is called indirectly via a small wrapper */ @@ -314,7 +302,7 @@ long execve(const char *filename, char **argv, char **envp) "b ret_to_user" : : "r" (current_thread_info()), - "Ir" (THREAD_SIZE - 8 - sizeof(regs)), + "Ir" (THREAD_START_SP - sizeof(regs)), "r" (®s), "Ir" (sizeof(regs)) : "r0", "r1", "r2", "r3", "ip", "memory"); diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 6e31718f600..14df16b983f 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -218,7 +218,8 @@ NORET_TYPE void die(const char *str, struct pt_regs *regs, int err) tsk->comm, tsk->pid, tsk->thread_info + 1); if (!user_mode(regs) || in_interrupt()) { - dump_mem("Stack: ", regs->ARM_sp, 8192+(unsigned long)tsk->thread_info); + dump_mem("Stack: ", regs->ARM_sp, + THREAD_SIZE + (unsigned long)tsk->thread_info); dump_backtrace(regs, tsk); dump_instr(regs); } @@ -450,13 +451,17 @@ asmlinkage int arm_syscall(int no, struct pt_regs *regs) case NR(set_tls): thread->tp_value = regs->ARM_r0; +#if defined(CONFIG_HAS_TLS_REG) + asm ("mcr p15, 0, %0, c13, c0, 3" : : "r" (regs->ARM_r0) ); +#elif !defined(CONFIG_TLS_REG_EMUL) /* - * Our user accessible TLS ptr is located at 0xffff0ffc. - * On SMP read access to this address must raise a fault - * and be emulated from the data abort handler. - * m + * User space must never try to access this directly. + * Expect your app to break eventually if you do so. + * The user helper at 0xffff0fe0 must be used instead. + * (see entry-armv.S for details) */ - *((unsigned long *)0xffff0ffc) = thread->tp_value; + *((unsigned int *)0xffff0ff0) = regs->ARM_r0; +#endif return 0; default: @@ -493,6 +498,44 @@ asmlinkage int arm_syscall(int no, struct pt_regs *regs) return 0; } +#ifdef CONFIG_TLS_REG_EMUL + +/* + * We might be running on an ARMv6+ processor which should have the TLS + * register but for some reason we can't use it, or maybe an SMP system + * using a pre-ARMv6 processor (there are apparently a few prototypes like + * that in existence) and therefore access to that register must be + * emulated. + */ + +static int get_tp_trap(struct pt_regs *regs, unsigned int instr) +{ + int reg = (instr >> 12) & 15; + if (reg == 15) + return 1; + regs->uregs[reg] = current_thread_info()->tp_value; + regs->ARM_pc += 4; + return 0; +} + +static struct undef_hook arm_mrc_hook = { + .instr_mask = 0x0fff0fff, + .instr_val = 0x0e1d0f70, + .cpsr_mask = PSR_T_BIT, + .cpsr_val = 0, + .fn = get_tp_trap, +}; + +static int __init arm_mrc_hook_init(void) +{ + register_undef_hook(&arm_mrc_hook); + return 0; +} + +late_initcall(arm_mrc_hook_init); + +#endif + void __bad_xchg(volatile void *ptr, int size) { printk("xchg: bad data size: pc 0x%p, ptr 0x%p, size %d\n", @@ -578,9 +621,19 @@ EXPORT_SYMBOL(abort); void __init trap_init(void) { - extern void __trap_init(void); + extern char __stubs_start[], __stubs_end[]; + extern char __vectors_start[], __vectors_end[]; + extern char __kuser_helper_start[], __kuser_helper_end[]; + int kuser_sz = __kuser_helper_end - __kuser_helper_start; - __trap_init(); + /* + * Copy the vectors, stubs and kuser helpers (in entry-armv.S) + * into the vector page, mapped at 0xffff0000, and ensure these + * are visible to the instruction stream. + */ + memcpy((void *)0xffff0000, __vectors_start, __vectors_end - __vectors_start); + memcpy((void *)0xffff0200, __stubs_start, __stubs_end - __stubs_start); + memcpy((void *)0xffff1000 - kuser_sz, __kuser_helper_start, kuser_sz); flush_icache_range(0xffff0000, 0xffff0000 + PAGE_SIZE); modify_domain(DOMAIN_USER, DOMAIN_CLIENT); } diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index a39c6a42d68..ad2d66c93a5 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -5,6 +5,7 @@ #include <asm-generic/vmlinux.lds.h> #include <linux/config.h> +#include <asm/thread_info.h> OUTPUT_ARCH(arm) ENTRY(stext) @@ -103,7 +104,7 @@ SECTIONS __data_loc = ALIGN(4); /* location in binary */ . = DATAADDR; #else - . = ALIGN(8192); + . = ALIGN(THREAD_SIZE); __data_loc = .; #endif diff --git a/arch/arm/mach-clps711x/Kconfig b/arch/arm/mach-clps711x/Kconfig index f6e676322ca..45c930ccd06 100644 --- a/arch/arm/mach-clps711x/Kconfig +++ b/arch/arm/mach-clps711x/Kconfig @@ -10,6 +10,7 @@ config ARCH_AUTCPU12 config ARCH_CDB89712 bool "CDB89712" + select ISA help This is an evaluation board from Cirrus for the CS89712 processor. The board includes 2 serial ports, Ethernet, IRDA, and expansion @@ -26,6 +27,8 @@ config ARCH_CLEP7312 config ARCH_EDB7211 bool "EDB7211" + select ISA + select DISCONTIGMEM help Say Y here if you intend to run this kernel on a Cirrus Logic EDB-7211 evaluation board. diff --git a/arch/arm/mach-footbridge/Kconfig b/arch/arm/mach-footbridge/Kconfig index 1090c680b6d..324d9edeec3 100644 --- a/arch/arm/mach-footbridge/Kconfig +++ b/arch/arm/mach-footbridge/Kconfig @@ -5,6 +5,9 @@ menu "Footbridge Implementations" config ARCH_CATS bool "CATS" select FOOTBRIDGE_HOST + select ISA + select ISA_DMA + select PCI help Say Y here if you intend to run this kernel on the CATS. @@ -13,6 +16,9 @@ config ARCH_CATS config ARCH_PERSONAL_SERVER bool "Compaq Personal Server" select FOOTBRIDGE_HOST + select ISA + select ISA_DMA + select PCI ---help--- Say Y here if you intend to run this kernel on the Compaq Personal Server. @@ -42,6 +48,9 @@ config ARCH_EBSA285_HOST bool "EBSA285 (host mode)" select ARCH_EBSA285 select FOOTBRIDGE_HOST + select ISA + select ISA_DMA + select PCI help Say Y here if you intend to run this kernel on the EBSA285 card in host ("central function") mode. @@ -51,6 +60,9 @@ config ARCH_EBSA285_HOST config ARCH_NETWINDER bool "NetWinder" select FOOTBRIDGE_HOST + select ISA + select ISA_DMA + select PCI help Say Y here if you intend to run this kernel on the Rebel.COM NetWinder. Information about this machine can be found at: diff --git a/arch/arm/mach-imx/Kconfig b/arch/arm/mach-imx/Kconfig index ec85813ee5d..cddd194ac6e 100644 --- a/arch/arm/mach-imx/Kconfig +++ b/arch/arm/mach-imx/Kconfig @@ -4,6 +4,7 @@ menu "IMX Implementations" config ARCH_MX1ADS bool "mx1ads" depends on ARCH_IMX + select ISA help Say Y here if you are using the Motorola MX1ADS board diff --git a/arch/arm/mach-imx/generic.c b/arch/arm/mach-imx/generic.c index 54377d0f578..41e5849ae8d 100644 --- a/arch/arm/mach-imx/generic.c +++ b/arch/arm/mach-imx/generic.c @@ -26,6 +26,7 @@ #include <linux/init.h> #include <linux/kernel.h> #include <linux/module.h> +#include <asm/arch/imxfb.h> #include <asm/hardware.h> #include <asm/mach/map.h> @@ -228,6 +229,14 @@ static struct platform_device imx_uart2_device = { .resource = imx_uart2_resources, }; +static struct imxfb_mach_info imx_fb_info; + +void __init set_imx_fb_info(struct imxfb_mach_info *hard_imx_fb_info) +{ + memcpy(&imx_fb_info,hard_imx_fb_info,sizeof(struct imxfb_mach_info)); +} +EXPORT_SYMBOL(set_imx_fb_info); + static struct resource imxfb_resources[] = { [0] = { .start = 0x00205000, @@ -241,9 +250,16 @@ static struct resource imxfb_resources[] = { }, }; +static u64 fb_dma_mask = ~(u64)0; + static struct platform_device imxfb_device = { .name = "imx-fb", .id = 0, + .dev = { + .platform_data = &imx_fb_info, + .dma_mask = &fb_dma_mask, + .coherent_dma_mask = 0xffffffff, + }, .num_resources = ARRAY_SIZE(imxfb_resources), .resource = imxfb_resources, }; diff --git a/arch/arm/mach-integrator/core.c b/arch/arm/mach-integrator/core.c index 86c50c3889b..bd17b515431 100644 --- a/arch/arm/mach-integrator/core.c +++ b/arch/arm/mach-integrator/core.c @@ -216,7 +216,9 @@ integrator_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) write_seqlock(&xtime_lock); - // ...clear the interrupt + /* + * clear the interrupt + */ timer1->TimerClear = 1; timer_tick(regs); @@ -264,7 +266,7 @@ void __init integrator_time_init(unsigned long reload, unsigned int ctrl) timer1->TimerValue = timer_reload; timer1->TimerControl = timer_ctrl; - /* + /* * Make irqs happen for the system timer */ setup_irq(IRQ_TIMERINT1, &integrator_timer_irq); diff --git a/arch/arm/mach-integrator/integrator_cp.c b/arch/arm/mach-integrator/integrator_cp.c index 68e15c36e33..3b948e8c275 100644 --- a/arch/arm/mach-integrator/integrator_cp.c +++ b/arch/arm/mach-integrator/integrator_cp.c @@ -420,7 +420,22 @@ static struct clcd_panel vga = { */ static void cp_clcd_enable(struct clcd_fb *fb) { - cm_control(CM_CTRL_LCDMUXSEL_MASK, CM_CTRL_LCDMUXSEL_VGA); + u32 val; + + if (fb->fb.var.bits_per_pixel <= 8) + val = CM_CTRL_LCDMUXSEL_VGA_8421BPP; + else if (fb->fb.var.bits_per_pixel <= 16) + val = CM_CTRL_LCDMUXSEL_VGA_16BPP; + else + val = 0; /* no idea for this, don't trust the docs */ + + cm_control(CM_CTRL_LCDMUXSEL_MASK| + CM_CTRL_LCDEN0| + CM_CTRL_LCDEN1| + CM_CTRL_STATIC1| + CM_CTRL_STATIC2| + CM_CTRL_STATIC| + CM_CTRL_n24BITEN, val); } static unsigned long framesize = SZ_1M; diff --git a/arch/arm/mach-integrator/leds.c b/arch/arm/mach-integrator/leds.c index 9d182b77b31..d2c0ab21150 100644 --- a/arch/arm/mach-integrator/leds.c +++ b/arch/arm/mach-integrator/leds.c @@ -37,7 +37,7 @@ static void integrator_leds_event(led_event_t ledevt) unsigned long flags; const unsigned int dbg_base = IO_ADDRESS(INTEGRATOR_DBG_BASE); unsigned int update_alpha_leds; - + // yup, change the LEDs local_irq_save(flags); update_alpha_leds = 0; diff --git a/arch/arm/mach-integrator/time.c b/arch/arm/mach-integrator/time.c index 20729de2af2..1a844ca139e 100644 --- a/arch/arm/mach-integrator/time.c +++ b/arch/arm/mach-integrator/time.c @@ -40,25 +40,32 @@ static int integrator_set_rtc(void) return 1; } -static void rtc_read_alarm(struct rtc_wkalrm *alrm) +static int rtc_read_alarm(struct rtc_wkalrm *alrm) { rtc_time_to_tm(readl(rtc_base + RTC_MR), &alrm->time); + return 0; } -static int rtc_set_alarm(struct rtc_wkalrm *alrm) +static inline int rtc_set_alarm(struct rtc_wkalrm *alrm) { unsigned long time; int ret; - ret = rtc_tm_to_time(&alrm->time, &time); + /* + * At the moment, we can only deal with non-wildcarded alarm times. + */ + ret = rtc_valid_tm(&alrm->time); + if (ret == 0) + ret = rtc_tm_to_time(&alrm->time, &time); if (ret == 0) writel(time, rtc_base + RTC_MR); return ret; } -static void rtc_read_time(struct rtc_time *tm) +static int rtc_read_time(struct rtc_time *tm) { rtc_time_to_tm(readl(rtc_base + RTC_DR), tm); + return 0; } /* @@ -69,7 +76,7 @@ static void rtc_read_time(struct rtc_time *tm) * edge of the 1Hz clock, we must write the time one second * in advance. */ -static int rtc_set_time(struct rtc_time *tm) +static inline int rtc_set_time(struct rtc_time *tm) { unsigned long time; int ret; diff --git a/arch/arm/mach-ixp2000/ixdp2800.c b/arch/arm/mach-ixp2000/ixdp2800.c index c4683aaff84..aec13c7108a 100644 --- a/arch/arm/mach-ixp2000/ixdp2800.c +++ b/arch/arm/mach-ixp2000/ixdp2800.c @@ -65,19 +65,102 @@ static struct sys_timer ixdp2800_timer = { /************************************************************************* * IXDP2800 PCI *************************************************************************/ +static void __init ixdp2800_slave_disable_pci_master(void) +{ + *IXP2000_PCI_CMDSTAT &= ~(PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY); +} + +static void __init ixdp2800_master_wait_for_slave(void) +{ + volatile u32 *addr; + + printk(KERN_INFO "IXDP2800: waiting for slave NPU to configure " + "its BAR sizes\n"); + + addr = ixp2000_pci_config_addr(0, IXDP2X00_SLAVE_NPU_DEVFN, + PCI_BASE_ADDRESS_1); + do { + *addr = 0xffffffff; + cpu_relax(); + } while (*addr != 0xfe000008); + + addr = ixp2000_pci_config_addr(0, IXDP2X00_SLAVE_NPU_DEVFN, + PCI_BASE_ADDRESS_2); + do { + *addr = 0xffffffff; + cpu_relax(); + } while (*addr != 0xc0000008); + + /* + * Configure the slave's SDRAM BAR by hand. + */ + *addr = 0x40000008; +} + +static void __init ixdp2800_slave_wait_for_master_enable(void) +{ + printk(KERN_INFO "IXDP2800: waiting for master NPU to enable us\n"); + + while ((*IXP2000_PCI_CMDSTAT & PCI_COMMAND_MASTER) == 0) + cpu_relax(); +} + void __init ixdp2800_pci_preinit(void) { printk("ixdp2x00_pci_preinit called\n"); - *IXP2000_PCI_ADDR_EXT = 0x0000e000; + *IXP2000_PCI_ADDR_EXT = 0x0001e000; + + if (!ixdp2x00_master_npu()) + ixdp2800_slave_disable_pci_master(); - *IXP2000_PCI_DRAM_BASE_ADDR_MASK = (0x40000000 - 1) & ~0xfffff; *IXP2000_PCI_SRAM_BASE_ADDR_MASK = (0x2000000 - 1) & ~0x3ffff; + *IXP2000_PCI_DRAM_BASE_ADDR_MASK = (0x40000000 - 1) & ~0xfffff; ixp2000_pci_preinit(); + + if (ixdp2x00_master_npu()) { + /* + * Wait until the slave set its SRAM/SDRAM BAR sizes + * correctly before we proceed to scan and enumerate + * the bus. + */ + ixdp2800_master_wait_for_slave(); + + /* + * We configure the SDRAM BARs by hand because they + * are 1G and fall outside of the regular allocated + * PCI address space. + */ + *IXP2000_PCI_SDRAM_BAR = 0x00000008; + } else { + /* + * Wait for the master to complete scanning the bus + * and assigning resources before we proceed to scan + * the bus ourselves. Set pci=firmware to honor the + * master's resource assignment. + */ + ixdp2800_slave_wait_for_master_enable(); + pcibios_setup("firmware"); + } } -int ixdp2800_pci_setup(int nr, struct pci_sys_data *sys) +/* + * We assign the SDRAM BARs for the two IXP2800 CPUs by hand, outside + * of the regular PCI window, because there's only 512M of outbound PCI + * memory window on each IXP, while we need 1G for each of the BARs. + */ +static void __devinit ixp2800_pci_fixup(struct pci_dev *dev) +{ + if (machine_is_ixdp2800()) { + dev->resource[2].start = 0; + dev->resource[2].end = 0; + dev->resource[2].flags = 0; + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IXP2800, ixp2800_pci_fixup); + +static int __init ixdp2800_pci_setup(int nr, struct pci_sys_data *sys) { sys->mem_offset = 0x00000000; @@ -129,22 +212,47 @@ static int __init ixdp2800_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin) } else return IRQ_IXP2000_PCIB; /* Slave NIC interrupt */ } -static void ixdp2800_pci_postinit(void) +static void __init ixdp2800_master_enable_slave(void) { - struct pci_dev *dev; + volatile u32 *addr; - if (ixdp2x00_master_npu()) { - dev = pci_find_slot(1, IXDP2800_SLAVE_ENET_DEVFN); - pci_remove_bus_device(dev); - } else { - dev = pci_find_slot(1, IXDP2800_MASTER_ENET_DEVFN); - pci_remove_bus_device(dev); + printk(KERN_INFO "IXDP2800: enabling slave NPU\n"); + + addr = (volatile u32 *)ixp2000_pci_config_addr(0, + IXDP2X00_SLAVE_NPU_DEVFN, + PCI_COMMAND); + + *addr |= PCI_COMMAND_MASTER; +} +static void __init ixdp2800_master_wait_for_slave_bus_scan(void) +{ + volatile u32 *addr; + + printk(KERN_INFO "IXDP2800: waiting for slave to finish bus scan\n"); + + addr = (volatile u32 *)ixp2000_pci_config_addr(0, + IXDP2X00_SLAVE_NPU_DEVFN, + PCI_COMMAND); + while ((*addr & PCI_COMMAND_MEMORY) == 0) + cpu_relax(); +} + +static void __init ixdp2800_slave_signal_bus_scan_completion(void) +{ + printk(KERN_INFO "IXDP2800: bus scan done, signaling master\n"); + *IXP2000_PCI_CMDSTAT |= PCI_COMMAND_MEMORY; +} + +static void __init ixdp2800_pci_postinit(void) +{ + if (!ixdp2x00_master_npu()) { ixdp2x00_slave_pci_postinit(); + ixdp2800_slave_signal_bus_scan_completion(); } } -struct hw_pci ixdp2800_pci __initdata = { +struct __initdata hw_pci ixdp2800_pci __initdata = { .nr_controllers = 1, .setup = ixdp2800_pci_setup, .preinit = ixdp2800_pci_preinit, @@ -155,8 +263,21 @@ struct hw_pci ixdp2800_pci __initdata = { int __init ixdp2800_pci_init(void) { - if (machine_is_ixdp2800()) + if (machine_is_ixdp2800()) { + struct pci_dev *dev; + pci_common_init(&ixdp2800_pci); + if (ixdp2x00_master_npu()) { + dev = pci_find_slot(1, IXDP2800_SLAVE_ENET_DEVFN); + pci_remove_bus_device(dev); + + ixdp2800_master_enable_slave(); + ixdp2800_master_wait_for_slave_bus_scan(); + } else { + dev = pci_find_slot(1, IXDP2800_MASTER_ENET_DEVFN); + pci_remove_bus_device(dev); + } + } return 0; } diff --git a/arch/arm/mach-ixp2000/pci.c b/arch/arm/mach-ixp2000/pci.c index 831f8ffb6b6..5ff2f2718c5 100644 --- a/arch/arm/mach-ixp2000/pci.c +++ b/arch/arm/mach-ixp2000/pci.c @@ -37,7 +37,7 @@ static int pci_master_aborts = 0; static int clear_master_aborts(void); -static u32 * +u32 * ixp2000_pci_config_addr(unsigned int bus_nr, unsigned int devfn, int where) { u32 *paddress; @@ -208,15 +208,15 @@ ixp2000_pci_preinit(void) * use our own resource space. */ static struct resource ixp2000_pci_mem_space = { - .start = 0x00000000, + .start = 0xe0000000, .end = 0xffffffff, .flags = IORESOURCE_MEM, .name = "PCI Mem Space" }; static struct resource ixp2000_pci_io_space = { - .start = 0x00000000, - .end = 0xffffffff, + .start = 0x00010000, + .end = 0x0001ffff, .flags = IORESOURCE_IO, .name = "PCI I/O Space" }; diff --git a/arch/arm/mach-ixp4xx/common-pci.c b/arch/arm/mach-ixp4xx/common-pci.c index 94bcdb933e4..aa92e370883 100644 --- a/arch/arm/mach-ixp4xx/common-pci.c +++ b/arch/arm/mach-ixp4xx/common-pci.c @@ -502,15 +502,6 @@ pci_set_dma_mask(struct pci_dev *dev, u64 mask) } int -pci_dac_set_dma_mask(struct pci_dev *dev, u64 mask) -{ - if (mask >= SZ_64M - 1 ) - return 0; - - return -EIO; -} - -int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask) { if (mask >= SZ_64M - 1 ) @@ -520,7 +511,6 @@ pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask) } EXPORT_SYMBOL(pci_set_dma_mask); -EXPORT_SYMBOL(pci_dac_set_dma_mask); EXPORT_SYMBOL(pci_set_consistent_dma_mask); EXPORT_SYMBOL(ixp4xx_pci_read); EXPORT_SYMBOL(ixp4xx_pci_write); diff --git a/arch/arm/mach-pxa/generic.c b/arch/arm/mach-pxa/generic.c index b1575b8dc1c..a45aaa115a7 100644 --- a/arch/arm/mach-pxa/generic.c +++ b/arch/arm/mach-pxa/generic.c @@ -220,6 +220,30 @@ static struct platform_device stuart_device = { .id = 2, }; +static struct resource i2c_resources[] = { + { + .start = 0x40301680, + .end = 0x403016a3, + .flags = IORESOURCE_MEM, + }, { + .start = IRQ_I2C, + .end = IRQ_I2C, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device i2c_device = { + .name = "pxa2xx-i2c", + .id = 0, + .resource = i2c_resources, + .num_resources = ARRAY_SIZE(i2c_resources), +}; + +void __init pxa_set_i2c_info(struct i2c_pxa_platform_data *info) +{ + i2c_device.dev.platform_data = info; +} + static struct platform_device *devices[] __initdata = { &pxamci_device, &udc_device, @@ -227,6 +251,7 @@ static struct platform_device *devices[] __initdata = { &ffuart_device, &btuart_device, &stuart_device, + &i2c_device, }; static int __init pxa_init(void) diff --git a/arch/arm/mach-pxa/sleep.S b/arch/arm/mach-pxa/sleep.S index 16cad2c2497..5786ccad938 100644 --- a/arch/arm/mach-pxa/sleep.S +++ b/arch/arm/mach-pxa/sleep.S @@ -18,6 +18,11 @@ #include <asm/arch/pxa-regs.h> +#ifdef CONFIG_PXA27x // workaround for Errata 50 +#define MDREFR_KDIV 0x200a4000 // all banks +#define CCCR_SLEEP 0x00000107 // L=7 2N=2 A=0 PPDIS=0 CPDIS=0 +#endif + .text /* @@ -28,7 +33,9 @@ ENTRY(pxa_cpu_suspend) +#ifndef CONFIG_IWMMXT mra r2, r3, acc0 +#endif stmfd sp!, {r2 - r12, lr} @ save registers on stack @ get coprocessor registers @@ -61,14 +68,23 @@ ENTRY(pxa_cpu_suspend) @ prepare value for sleep mode mov r1, #3 @ sleep mode - @ prepare to put SDRAM into self-refresh manually + @ prepare pointer to physical address 0 (virtual mapping in generic.c) + mov r2, #UNCACHED_PHYS_0 + + @ prepare SDRAM refresh settings ldr r4, =MDREFR ldr r5, [r4] + + @ enable SDRAM self-refresh mode orr r5, r5, #MDREFR_SLFRSH - @ prepare pointer to physical address 0 (virtual mapping in generic.c) - mov r2, #UNCACHED_PHYS_0 +#ifdef CONFIG_PXA27x + @ set SDCLKx divide-by-2 bits (this is part of a workaround for Errata 50) + ldr r6, =MDREFR_KDIV + orr r5, r5, r6 +#endif +#ifdef CONFIG_PXA25x @ Intel PXA255 Specification Update notes problems @ about suspending with PXBus operating above 133MHz @ (see Errata 31, GPIO output signals, ... unpredictable in sleep @@ -100,6 +116,18 @@ ENTRY(pxa_cpu_suspend) mov r0, #0 mcr p14, 0, r0, c6, c0, 0 orr r0, r0, #2 @ initiate change bit +#endif +#ifdef CONFIG_PXA27x + @ Intel PXA270 Specification Update notes problems sleeping + @ with core operating above 91 MHz + @ (see Errata 50, ...processor does not exit from sleep...) + + ldr r6, =CCCR + ldr r8, [r6] @ keep original value for resume + + ldr r7, =CCCR_SLEEP @ prepare CCCR sleep value + mov r0, #0x2 @ prepare value for CLKCFG +#endif @ align execution to a cache line b 1f @@ -111,6 +139,7 @@ ENTRY(pxa_cpu_suspend) @ All needed values are now in registers. @ These last instructions should be in cache +#if defined(CONFIG_PXA25x) || defined(CONFIG_PXA27x) @ initiate the frequency change... str r7, [r6] mcr p14, 0, r0, c6, c0, 0 @@ -118,14 +147,27 @@ ENTRY(pxa_cpu_suspend) @ restore the original cpu speed value for resume str r8, [r6] - @ put SDRAM into self-refresh - str r5, [r4] + @ need 6 13-MHz cycles before changing PWRMODE + @ just set frequency to 91-MHz... 6*91/13 = 42 + + mov r0, #42 +10: subs r0, r0, #1 + bne 10b +#endif + + @ Do not reorder... + @ Intel PXA270 Specification Update notes problems performing + @ external accesses after SDRAM is put in self-refresh mode + @ (see Errata 39 ...hangs when entering self-refresh mode) @ force address lines low by reading at physical address 0 ldr r3, [r2] + @ put SDRAM into self-refresh + str r5, [r4] + @ enter sleep mode - mcr p14, 0, r1, c7, c0, 0 + mcr p14, 0, r1, c7, c0, 0 @ PWRMODE 20: b 20b @ loop waiting for sleep @@ -188,7 +230,9 @@ resume_after_mmu: bl cpu_xscale_proc_init #endif ldmfd sp!, {r2, r3} +#ifndef CONFIG_IWMMXT mar acc0, r2, r3 +#endif ldmfd sp!, {r4 - r12, pc} @ return to caller diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig index 5b670c9ac5e..c4fc6be629d 100644 --- a/arch/arm/mm/Kconfig +++ b/arch/arm/mm/Kconfig @@ -409,3 +409,24 @@ config CPU_BPREDICT_DISABLE depends on CPU_ARM1020 help Say Y here to disable branch prediction. If unsure, say N. + +config TLS_REG_EMUL + bool + default y if (SMP || CPU_32v6) && (CPU_32v5 || CPU_32v4 || CPU_32v3) + help + We might be running on an ARMv6+ processor which should have the TLS + register but for some reason we can't use it, or maybe an SMP system + using a pre-ARMv6 processor (there are apparently a few prototypes + like that in existence) and therefore access to that register must + be emulated. + +config HAS_TLS_REG + bool + depends on CPU_32v6 + default y if !TLS_REG_EMUL + help + This selects support for the CP15 thread register. + It is defined to be available on ARMv6 or later. If a particular + ARMv6 or later CPU doesn't support it then it must omc;ide "select + TLS_REG_EMUL" along with its other caracteristics. + diff --git a/arch/arm/mm/abort-ev6.S b/arch/arm/mm/abort-ev6.S index 38b2cbb89be..8f76f3df7b4 100644 --- a/arch/arm/mm/abort-ev6.S +++ b/arch/arm/mm/abort-ev6.S @@ -1,5 +1,6 @@ #include <linux/linkage.h> #include <asm/assembler.h> +#include "abort-macro.S" /* * Function: v6_early_abort * @@ -13,11 +14,26 @@ * : sp = pointer to registers * * Purpose : obtain information about current aborted instruction. + * Note: we read user space. This means we might cause a data + * abort here if the I-TLB and D-TLB aren't seeing the same + * picture. Unfortunately, this does happen. We live with it. */ .align 5 ENTRY(v6_early_abort) mrc p15, 0, r1, c5, c0, 0 @ get FSR mrc p15, 0, r0, c6, c0, 0 @ get FAR +/* + * Faulty SWP instruction on 1136 doesn't set bit 11 in DFSR. + * The test below covers all the write situations, including Java bytecodes + */ + bic r1, r1, #1 << 11 | 1 << 10 @ clear bits 11 and 10 of FSR + tst r3, #PSR_J_BIT @ Java? + movne pc, lr + do_thumb_abort + ldreq r3, [r2] @ read aborted ARM instruction + do_ldrd_abort + tst r3, #1 << 20 @ L = 0 -> write + orreq r1, r1, #1 << 11 @ yes. mov pc, lr diff --git a/arch/arm/mm/mm-armv.c b/arch/arm/mm/mm-armv.c index f5a87db8b49..585dfb8e20b 100644 --- a/arch/arm/mm/mm-armv.c +++ b/arch/arm/mm/mm-armv.c @@ -411,9 +411,10 @@ static void __init build_mem_type_table(void) mem_types[MT_MEMORY].prot_sect &= ~PMD_BIT4; mem_types[MT_ROM].prot_sect &= ~PMD_BIT4; /* - * Mark cache clean areas read only from SVC mode - * and no access from userspace. + * Mark cache clean areas and XIP ROM read only + * from SVC mode and no access from userspace. */ + mem_types[MT_ROM].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE; mem_types[MT_MINICLEAN].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE; mem_types[MT_CACHECLEAN].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE; } diff --git a/arch/arm26/Kconfig b/arch/arm26/Kconfig index 3955de5af4c..6caed90661f 100644 --- a/arch/arm26/Kconfig +++ b/arch/arm26/Kconfig @@ -89,6 +89,10 @@ config PAGESIZE_16 machine with 4MB of memory. endmenu +config ISA_DMA_API + bool + default y + menu "General setup" # Compressed boot loader in ROM. Yes, we really want to ask about diff --git a/arch/arm26/kernel/ptrace.c b/arch/arm26/kernel/ptrace.c index 2a137146a77..8a52124de0e 100644 --- a/arch/arm26/kernel/ptrace.c +++ b/arch/arm26/kernel/ptrace.c @@ -18,6 +18,7 @@ #include <linux/ptrace.h> #include <linux/user.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -591,7 +592,7 @@ static int do_ptrace(int request, struct task_struct *child, long addr, long dat case PTRACE_SYSCALL: case PTRACE_CONT: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -626,7 +627,7 @@ static int do_ptrace(int request, struct task_struct *child, long addr, long dat */ case PTRACE_SINGLESTEP: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; child->ptrace |= PT_SINGLESTEP; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/arm26/mm/small_page.c b/arch/arm26/mm/small_page.c index 77be86cca78..30447106c25 100644 --- a/arch/arm26/mm/small_page.c +++ b/arch/arm26/mm/small_page.c @@ -92,8 +92,7 @@ static unsigned long __get_small_page(int priority, struct order *order) page = list_entry(order->queue.next, struct page, lru); again: #ifdef PEDANTIC - if (USED_MAP(page) & ~order->all_used) - PAGE_BUG(page); + BUG_ON(USED_MAP(page) & ~order->all_used); #endif offset = ffz(USED_MAP(page)); SET_USED(page, offset); @@ -141,8 +140,7 @@ static void __free_small_page(unsigned long spage, struct order *order) goto non_small; #ifdef PEDANTIC - if (USED_MAP(page) & ~order->all_used) - PAGE_BUG(page); + BUG_ON(USED_MAP(page) & ~order->all_used); #endif spage = spage >> order->shift; diff --git a/arch/cris/arch-v10/kernel/ptrace.c b/arch/cris/arch-v10/kernel/ptrace.c index da15db8ae48..581ecabaae5 100644 --- a/arch/cris/arch-v10/kernel/ptrace.c +++ b/arch/cris/arch-v10/kernel/ptrace.c @@ -10,6 +10,7 @@ #include <linux/errno.h> #include <linux/ptrace.h> #include <linux/user.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -184,7 +185,7 @@ sys_ptrace(long request, long pid, long addr, long data) case PTRACE_CONT: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { @@ -219,7 +220,7 @@ sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/frv/kernel/ptrace.c b/arch/frv/kernel/ptrace.c index 2a0efb739ad..cbe03cba9f0 100644 --- a/arch/frv/kernel/ptrace.c +++ b/arch/frv/kernel/ptrace.c @@ -20,6 +20,7 @@ #include <linux/user.h> #include <linux/config.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -239,7 +240,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -267,7 +268,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); ptrace_enable(child); diff --git a/arch/h8300/kernel/ptrace.c b/arch/h8300/kernel/ptrace.c index 5f19d774a28..05c15e86977 100644 --- a/arch/h8300/kernel/ptrace.c +++ b/arch/h8300/kernel/ptrace.c @@ -24,6 +24,7 @@ #include <linux/ptrace.h> #include <linux/user.h> #include <linux/config.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -171,7 +172,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data >= _NSIG) + if (!valid_signal(data)) break ; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -202,7 +203,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); child->exit_code = data; diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 17a0cbce6f3..e382f32d435 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -183,7 +183,7 @@ config M386 - "Winchip-C6" for original IDT Winchip. - "Winchip-2" for IDT Winchip 2. - "Winchip-2A" for IDT Winchips with 3dNow! capabilities. - - "MediaGX/Geode" for Cyrix MediaGX aka Geode. + - "GeodeGX1" for Geode GX1 (Cyrix MediaGX). - "CyrixIII/VIA C3" for VIA Cyrix III or VIA C3. - "VIA C3-2 for VIA C3-2 "Nehemiah" (model 9 and above). @@ -311,12 +311,10 @@ config MWINCHIP3D stores for this CPU, which can increase performance of some operations. -config MGEODE - bool "MediaGX/Geode" +config MGEODEGX1 + bool "GeodeGX1" help - Select this for a Cyrix MediaGX aka Geode chip. Linux and GCC - treat this chip as a 586TSC with some extended instructions - and alignment reqirements. + Select this for a Geode GX1 (Cyrix MediaGX) chip. config MCYRIXIII bool "CyrixIII/VIA-C3" @@ -368,7 +366,7 @@ config X86_L1_CACHE_SHIFT int default "7" if MPENTIUM4 || X86_GENERIC default "4" if X86_ELAN || M486 || M386 - default "5" if MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE + default "5" if MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODEGX1 default "6" if MK7 || MK8 || MPENTIUMM config RWSEM_GENERIC_SPINLOCK @@ -387,7 +385,7 @@ config GENERIC_CALIBRATE_DELAY config X86_PPRO_FENCE bool - depends on M686 || M586MMX || M586TSC || M586 || M486 || M386 || MGEODE + depends on M686 || M586MMX || M586TSC || M586 || M486 || M386 || MGEODEGX1 default y config X86_F00F_BUG @@ -417,7 +415,7 @@ config X86_POPAD_OK config X86_ALIGNMENT_16 bool - depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODE + depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 default y config X86_GOOD_APIC @@ -442,7 +440,7 @@ config X86_USE_3DNOW config X86_OOSTORE bool - depends on (MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MGEODE) && MTRR + depends on (MWINCHIP3D || MWINCHIP2 || MWINCHIPC6) && MTRR default y config HPET_TIMER @@ -578,7 +576,7 @@ config X86_VISWS_APIC config X86_TSC bool - depends on (MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MGEODE) && !X86_NUMAQ + depends on (MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MGEODEGX1) && !X86_NUMAQ default y config X86_MCE @@ -653,6 +651,24 @@ config I8K Say Y if you intend to run this kernel on a Dell Inspiron 8000. Say N otherwise. +config X86_REBOOTFIXUPS + bool "Enable X86 board specific fixups for reboot" + depends on X86 + default n + ---help--- + This enables chipset and/or board specific fixups to be done + in order to get reboot to work correctly. This is only needed on + some combinations of hardware and BIOS. The symptom, for which + this config is intended, is when reboot ends with a stalled/hung + system. + + Currently, the only fixup is for the Geode GX1/CS5530A/TROM2.1. + combination. + + Say Y if you want to enable the fixup. Currently, it's safe to + enable this option even if you don't need it. + Say N otherwise. + config MICROCODE tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support" ---help--- @@ -1155,6 +1171,10 @@ source "drivers/pci/pcie/Kconfig" source "drivers/pci/Kconfig" +config ISA_DMA_API + bool + default y + config ISA bool "ISA support" depends on !(X86_VOYAGER || X86_VISWS) diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 314c7146e9b..1c36ca332a9 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -14,7 +14,7 @@ # 19990713 Artur Skawina <skawina@geocities.com> # Added '-march' and '-mpreferred-stack-boundary' support # -# Kianusch Sayah Karadji <kianusch@sk-tech.net> +# 20050320 Kianusch Sayah Karadji <kianusch@sk-tech.net> # Added support for GEODE CPU LDFLAGS := -m elf_i386 @@ -54,8 +54,8 @@ cflags-$(CONFIG_MVIAC3_2) += $(call cc-option,-march=c3-2,-march=i686) # AMD Elan support cflags-$(CONFIG_X86_ELAN) += -march=i486 -# MediaGX aka Geode support -cflags-$(CONFIG_MGEODE) += $(call cc-option,-march=pentium-mmx,-march=i586) +# Geode GX1 support +cflags-$(CONFIG_MGEODEGX1) += $(call cc-option,-march=pentium-mmx,-march=i486) # -mregparm=3 works ok on gcc-3.0 and later # @@ -123,7 +123,7 @@ AFLAGS += $(mflags-y) boot := arch/i386/boot .PHONY: zImage bzImage compressed zlilo bzlilo \ - zdisk bzdisk fdimage fdimage144 fdimage288 install + zdisk bzdisk fdimage fdimage144 fdimage288 install kernel_install all: bzImage @@ -145,8 +145,9 @@ zdisk bzdisk: vmlinux fdimage fdimage144 fdimage288: vmlinux $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@ -install: - $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@ +install: vmlinux +install kernel_install: + $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) install prepare: include/asm-$(ARCH)/asm_offsets.h CLEAN_FILES += include/asm-$(ARCH)/asm_offsets.h diff --git a/arch/i386/boot/bootsect.S b/arch/i386/boot/bootsect.S index ba9fe14db6a..011b7a4993d 100644 --- a/arch/i386/boot/bootsect.S +++ b/arch/i386/boot/bootsect.S @@ -83,7 +83,7 @@ bugger_off_msg: .ascii "\n" .ascii "Remove disk and press any key to reboot . . .\r\n" .byte 0 - + # Kernel attributes; used by setup diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c index fa67045234a..cedc55cc47d 100644 --- a/arch/i386/boot/compressed/misc.c +++ b/arch/i386/boot/compressed/misc.c @@ -12,7 +12,6 @@ #include <linux/linkage.h> #include <linux/vmalloc.h> #include <linux/tty.h> -#include <video/edid.h> #include <asm/io.h> /* diff --git a/arch/i386/boot/setup.S b/arch/i386/boot/setup.S index a934ab32bf8..caa1fde6904 100644 --- a/arch/i386/boot/setup.S +++ b/arch/i386/boot/setup.S @@ -164,7 +164,7 @@ ramdisk_max: .long (-__PAGE_OFFSET-(512 << 20)-1) & 0x7fffffff trampoline: call start_of_setup .align 16 # The offset at this point is 0x240 - .space (0x7ff-0x240+1) # E820 & EDD space (ending at 0x7ff) + .space (0xeff-0x240+1) # E820 & EDD space (ending at 0xeff) # End of setup header ##################################################### start_of_setup: @@ -333,9 +333,9 @@ jmpe820: # sizeof(e820rec). # good820: - movb (E820NR), %al # up to 32 entries + movb (E820NR), %al # up to 128 entries cmpb $E820MAX, %al - jnl bail820 + jae bail820 incb (E820NR) movw %di, %ax diff --git a/arch/i386/boot/video.S b/arch/i386/boot/video.S index 925d3f5a382..0587477c99f 100644 --- a/arch/i386/boot/video.S +++ b/arch/i386/boot/video.S @@ -1924,36 +1924,36 @@ skip10: movb %ah, %al ret store_edid: - pushw %es # just save all registers - pushw %ax + pushw %es # just save all registers + pushw %ax pushw %bx pushw %cx pushw %dx pushw %di - pushw %fs + pushw %fs popw %es movl $0x13131313, %eax # memset block with 0x13 movw $32, %cx movw $0x140, %di cld - rep - stosl + rep + stosl - movw $0x4f15, %ax # do VBE/DDC + movw $0x4f15, %ax # do VBE/DDC movw $0x01, %bx movw $0x00, %cx movw $0x01, %dx movw $0x140, %di - int $0x10 + int $0x10 - popw %di # restore all registers + popw %di # restore all registers popw %dx popw %cx popw %bx popw %ax - popw %es + popw %es ret # VIDEO_SELECT-only variables diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile index aacdae6f372..0fbcfe00dd8 100644 --- a/arch/i386/kernel/Makefile +++ b/arch/i386/kernel/Makefile @@ -23,6 +23,7 @@ obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-$(CONFIG_X86_MPPARSE) += mpparse.o obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o +obj-$(CONFIG_X86_REBOOTFIXUPS) += reboot_fixups.o obj-$(CONFIG_X86_NUMAQ) += numaq.o obj-$(CONFIG_X86_SUMMIT_NUMA) += summit.o obj-$(CONFIG_KPROBES) += kprobes.o diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index 53eb5cfd5b6..848bb97af7c 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -650,7 +650,7 @@ acpi_find_rsdp (void) */ rsdp_phys = acpi_scan_rsdp (0, 0x400); if (!rsdp_phys) - rsdp_phys = acpi_scan_rsdp (0xE0000, 0xFFFFF); + rsdp_phys = acpi_scan_rsdp (0xE0000, 0x20000); return rsdp_phys; } diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c index e3879f7625c..d509836b70c 100644 --- a/arch/i386/kernel/apic.c +++ b/arch/i386/kernel/apic.c @@ -1265,8 +1265,6 @@ int __init APIC_init_uniprocessor (void) setup_local_APIC(); - if (nmi_watchdog == NMI_LOCAL_APIC) - check_nmi_watchdog(); #ifdef CONFIG_X86_IO_APIC if (smp_found_config) if (!skip_ioapic_setup && nr_ioapics) diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c index 8d182e875cd..16dbc4151be 100644 --- a/arch/i386/kernel/cpu/amd.c +++ b/arch/i386/kernel/cpu/amd.c @@ -24,7 +24,7 @@ __asm__(".align 4\nvide: ret"); static void __init init_amd(struct cpuinfo_x86 *c) { -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_SMP int cpu = c == &boot_cpu_data ? 0 : c - cpu_data; #endif u32 l, h; @@ -198,7 +198,7 @@ static void __init init_amd(struct cpuinfo_x86 *c) c->x86_num_cores = 1; } -#ifdef CONFIG_SMP +#ifdef CONFIG_X86_SMP /* * On a AMD dual core setup the lower bits of the APIC id * distingush the cores. Assumes number of cores is a power diff --git a/arch/i386/kernel/cpu/mtrr/cyrix.c b/arch/i386/kernel/cpu/mtrr/cyrix.c index 933b0dd62f4..9027a987006 100644 --- a/arch/i386/kernel/cpu/mtrr/cyrix.c +++ b/arch/i386/kernel/cpu/mtrr/cyrix.c @@ -218,12 +218,12 @@ typedef struct { mtrr_type type; } arr_state_t; -static arr_state_t arr_state[8] __initdata = { +static arr_state_t arr_state[8] __devinitdata = { {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL} }; -static unsigned char ccr_state[7] __initdata = { 0, 0, 0, 0, 0, 0, 0 }; +static unsigned char ccr_state[7] __devinitdata = { 0, 0, 0, 0, 0, 0, 0 }; static void cyrix_set_all(void) { diff --git a/arch/i386/kernel/cpu/mtrr/generic.c b/arch/i386/kernel/cpu/mtrr/generic.c index 9f7a7ea6388..f468a979e9a 100644 --- a/arch/i386/kernel/cpu/mtrr/generic.c +++ b/arch/i386/kernel/cpu/mtrr/generic.c @@ -124,8 +124,8 @@ int generic_get_free_region(unsigned long base, unsigned long size) return -ENOSPC; } -void generic_get_mtrr(unsigned int reg, unsigned long *base, - unsigned int *size, mtrr_type * type) +static void generic_get_mtrr(unsigned int reg, unsigned long *base, + unsigned int *size, mtrr_type * type) { unsigned int mask_lo, mask_hi, base_lo, base_hi; diff --git a/arch/i386/kernel/cpu/mtrr/main.c b/arch/i386/kernel/cpu/mtrr/main.c index 54999e4c55f..e1c2042b9b7 100644 --- a/arch/i386/kernel/cpu/mtrr/main.c +++ b/arch/i386/kernel/cpu/mtrr/main.c @@ -72,17 +72,21 @@ void set_mtrr_ops(struct mtrr_ops * ops) static int have_wrcomb(void) { struct pci_dev *dev; + u8 rev; if ((dev = pci_get_class(PCI_CLASS_BRIDGE_HOST << 8, NULL)) != NULL) { - /* ServerWorks LE chipsets have problems with write-combining + /* ServerWorks LE chipsets < rev 6 have problems with write-combining Don't allow it and leave room for other chipsets to be tagged */ if (dev->vendor == PCI_VENDOR_ID_SERVERWORKS && dev->device == PCI_DEVICE_ID_SERVERWORKS_LE) { - printk(KERN_INFO "mtrr: Serverworks LE detected. Write-combining disabled.\n"); - pci_dev_put(dev); - return 0; + pci_read_config_byte(dev, PCI_CLASS_REVISION, &rev); + if (rev <= 5) { + printk(KERN_INFO "mtrr: Serverworks LE rev < 6 detected. Write-combining disabled.\n"); + pci_dev_put(dev); + return 0; + } } - /* Intel 450NX errata # 23. Non ascending cachline evictions to + /* Intel 450NX errata # 23. Non ascending cacheline evictions to write combining memory may resulting in data corruption */ if (dev->vendor == PCI_VENDOR_ID_INTEL && dev->device == PCI_DEVICE_ID_INTEL_82451NX) { diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c index 4f28eba7fb8..7323c19f354 100644 --- a/arch/i386/kernel/cpu/proc.c +++ b/arch/i386/kernel/cpu/proc.c @@ -25,7 +25,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", "pbe", /* AMD-defined */ - "pni", NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL, NULL, NULL, NULL, "mp", "nx", NULL, "mmxext", NULL, NULL, "fxsr_opt", NULL, NULL, NULL, "lm", "3dnowext", "3dnow", diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 3c73dc865ea..a991d4e5edd 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -260,11 +260,9 @@ restore_nocheck: .section .fixup,"ax" iret_exc: sti - movl $__USER_DS, %edx - movl %edx, %ds - movl %edx, %es - movl $11,%eax - call do_exit + pushl $0 # no error code + pushl $do_iret_error + jmp error_code .previous .section __ex_table,"a" .align 4 @@ -516,8 +514,6 @@ debug_stack_correct: xorl %edx,%edx # error code 0 movl %esp,%eax # pt_regs pointer call do_debug - testl %eax,%eax - jnz restore_all jmp ret_from_exception /* @@ -598,8 +594,6 @@ ENTRY(int3) xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_int3 - testl %eax,%eax - jnz restore_all jmp ret_from_exception ENTRY(overflow) @@ -658,296 +652,6 @@ ENTRY(spurious_interrupt_bug) pushl $do_spurious_interrupt_bug jmp error_code -.data -ENTRY(sys_call_table) - .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ - .long sys_exit - .long sys_fork - .long sys_read - .long sys_write - .long sys_open /* 5 */ - .long sys_close - .long sys_waitpid - .long sys_creat - .long sys_link - .long sys_unlink /* 10 */ - .long sys_execve - .long sys_chdir - .long sys_time - .long sys_mknod - .long sys_chmod /* 15 */ - .long sys_lchown16 - .long sys_ni_syscall /* old break syscall holder */ - .long sys_stat - .long sys_lseek - .long sys_getpid /* 20 */ - .long sys_mount - .long sys_oldumount - .long sys_setuid16 - .long sys_getuid16 - .long sys_stime /* 25 */ - .long sys_ptrace - .long sys_alarm - .long sys_fstat - .long sys_pause - .long sys_utime /* 30 */ - .long sys_ni_syscall /* old stty syscall holder */ - .long sys_ni_syscall /* old gtty syscall holder */ - .long sys_access - .long sys_nice - .long sys_ni_syscall /* 35 - old ftime syscall holder */ - .long sys_sync - .long sys_kill - .long sys_rename - .long sys_mkdir - .long sys_rmdir /* 40 */ - .long sys_dup - .long sys_pipe - .long sys_times - .long sys_ni_syscall /* old prof syscall holder */ - .long sys_brk /* 45 */ - .long sys_setgid16 - .long sys_getgid16 - .long sys_signal - .long sys_geteuid16 - .long sys_getegid16 /* 50 */ - .long sys_acct - .long sys_umount /* recycled never used phys() */ - .long sys_ni_syscall /* old lock syscall holder */ - .long sys_ioctl - .long sys_fcntl /* 55 */ - .long sys_ni_syscall /* old mpx syscall holder */ - .long sys_setpgid - .long sys_ni_syscall /* old ulimit syscall holder */ - .long sys_olduname - .long sys_umask /* 60 */ - .long sys_chroot - .long sys_ustat - .long sys_dup2 - .long sys_getppid - .long sys_getpgrp /* 65 */ - .long sys_setsid - .long sys_sigaction - .long sys_sgetmask - .long sys_ssetmask - .long sys_setreuid16 /* 70 */ - .long sys_setregid16 - .long sys_sigsuspend - .long sys_sigpending - .long sys_sethostname - .long sys_setrlimit /* 75 */ - .long sys_old_getrlimit - .long sys_getrusage - .long sys_gettimeofday - .long sys_settimeofday - .long sys_getgroups16 /* 80 */ - .long sys_setgroups16 - .long old_select - .long sys_symlink - .long sys_lstat - .long sys_readlink /* 85 */ - .long sys_uselib - .long sys_swapon - .long sys_reboot - .long old_readdir - .long old_mmap /* 90 */ - .long sys_munmap - .long sys_truncate - .long sys_ftruncate - .long sys_fchmod - .long sys_fchown16 /* 95 */ - .long sys_getpriority - .long sys_setpriority - .long sys_ni_syscall /* old profil syscall holder */ - .long sys_statfs - .long sys_fstatfs /* 100 */ - .long sys_ioperm - .long sys_socketcall - .long sys_syslog - .long sys_setitimer - .long sys_getitimer /* 105 */ - .long sys_newstat - .long sys_newlstat - .long sys_newfstat - .long sys_uname - .long sys_iopl /* 110 */ - .long sys_vhangup - .long sys_ni_syscall /* old "idle" system call */ - .long sys_vm86old - .long sys_wait4 - .long sys_swapoff /* 115 */ - .long sys_sysinfo - .long sys_ipc - .long sys_fsync - .long sys_sigreturn - .long sys_clone /* 120 */ - .long sys_setdomainname - .long sys_newuname - .long sys_modify_ldt - .long sys_adjtimex - .long sys_mprotect /* 125 */ - .long sys_sigprocmask - .long sys_ni_syscall /* old "create_module" */ - .long sys_init_module - .long sys_delete_module - .long sys_ni_syscall /* 130: old "get_kernel_syms" */ - .long sys_quotactl - .long sys_getpgid - .long sys_fchdir - .long sys_bdflush - .long sys_sysfs /* 135 */ - .long sys_personality - .long sys_ni_syscall /* reserved for afs_syscall */ - .long sys_setfsuid16 - .long sys_setfsgid16 - .long sys_llseek /* 140 */ - .long sys_getdents - .long sys_select - .long sys_flock - .long sys_msync - .long sys_readv /* 145 */ - .long sys_writev - .long sys_getsid - .long sys_fdatasync - .long sys_sysctl - .long sys_mlock /* 150 */ - .long sys_munlock - .long sys_mlockall - .long sys_munlockall - .long sys_sched_setparam - .long sys_sched_getparam /* 155 */ - .long sys_sched_setscheduler - .long sys_sched_getscheduler - .long sys_sched_yield - .long sys_sched_get_priority_max - .long sys_sched_get_priority_min /* 160 */ - .long sys_sched_rr_get_interval - .long sys_nanosleep - .long sys_mremap - .long sys_setresuid16 - .long sys_getresuid16 /* 165 */ - .long sys_vm86 - .long sys_ni_syscall /* Old sys_query_module */ - .long sys_poll - .long sys_nfsservctl - .long sys_setresgid16 /* 170 */ - .long sys_getresgid16 - .long sys_prctl - .long sys_rt_sigreturn - .long sys_rt_sigaction - .long sys_rt_sigprocmask /* 175 */ - .long sys_rt_sigpending - .long sys_rt_sigtimedwait - .long sys_rt_sigqueueinfo - .long sys_rt_sigsuspend - .long sys_pread64 /* 180 */ - .long sys_pwrite64 - .long sys_chown16 - .long sys_getcwd - .long sys_capget - .long sys_capset /* 185 */ - .long sys_sigaltstack - .long sys_sendfile - .long sys_ni_syscall /* reserved for streams1 */ - .long sys_ni_syscall /* reserved for streams2 */ - .long sys_vfork /* 190 */ - .long sys_getrlimit - .long sys_mmap2 - .long sys_truncate64 - .long sys_ftruncate64 - .long sys_stat64 /* 195 */ - .long sys_lstat64 - .long sys_fstat64 - .long sys_lchown - .long sys_getuid - .long sys_getgid /* 200 */ - .long sys_geteuid - .long sys_getegid - .long sys_setreuid - .long sys_setregid - .long sys_getgroups /* 205 */ - .long sys_setgroups - .long sys_fchown - .long sys_setresuid - .long sys_getresuid - .long sys_setresgid /* 210 */ - .long sys_getresgid - .long sys_chown - .long sys_setuid - .long sys_setgid - .long sys_setfsuid /* 215 */ - .long sys_setfsgid - .long sys_pivot_root - .long sys_mincore - .long sys_madvise - .long sys_getdents64 /* 220 */ - .long sys_fcntl64 - .long sys_ni_syscall /* reserved for TUX */ - .long sys_ni_syscall - .long sys_gettid - .long sys_readahead /* 225 */ - .long sys_setxattr - .long sys_lsetxattr - .long sys_fsetxattr - .long sys_getxattr - .long sys_lgetxattr /* 230 */ - .long sys_fgetxattr - .long sys_listxattr - .long sys_llistxattr - .long sys_flistxattr - .long sys_removexattr /* 235 */ - .long sys_lremovexattr - .long sys_fremovexattr - .long sys_tkill - .long sys_sendfile64 - .long sys_futex /* 240 */ - .long sys_sched_setaffinity - .long sys_sched_getaffinity - .long sys_set_thread_area - .long sys_get_thread_area - .long sys_io_setup /* 245 */ - .long sys_io_destroy - .long sys_io_getevents - .long sys_io_submit - .long sys_io_cancel - .long sys_fadvise64 /* 250 */ - .long sys_ni_syscall - .long sys_exit_group - .long sys_lookup_dcookie - .long sys_epoll_create - .long sys_epoll_ctl /* 255 */ - .long sys_epoll_wait - .long sys_remap_file_pages - .long sys_set_tid_address - .long sys_timer_create - .long sys_timer_settime /* 260 */ - .long sys_timer_gettime - .long sys_timer_getoverrun - .long sys_timer_delete - .long sys_clock_settime - .long sys_clock_gettime /* 265 */ - .long sys_clock_getres - .long sys_clock_nanosleep - .long sys_statfs64 - .long sys_fstatfs64 - .long sys_tgkill /* 270 */ - .long sys_utimes - .long sys_fadvise64_64 - .long sys_ni_syscall /* sys_vserver */ - .long sys_mbind - .long sys_get_mempolicy - .long sys_set_mempolicy - .long sys_mq_open - .long sys_mq_unlink - .long sys_mq_timedsend - .long sys_mq_timedreceive /* 280 */ - .long sys_mq_notify - .long sys_mq_getsetattr - .long sys_ni_syscall /* reserved for kexec */ - .long sys_waitid - .long sys_ni_syscall /* 285 */ /* available */ - .long sys_add_key - .long sys_request_key - .long sys_keyctl +#include "syscall_table.S" syscall_table_size=(.-sys_call_table) diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index d273fd74619..e966fc8c44c 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -380,6 +380,7 @@ rp_sidt: ALIGN ignore_int: cld +#ifdef CONFIG_PRINTK pushl %eax pushl %ecx pushl %edx @@ -400,6 +401,7 @@ ignore_int: popl %edx popl %ecx popl %eax +#endif iret /* diff --git a/arch/i386/kernel/i386_ksyms.c b/arch/i386/kernel/i386_ksyms.c index 14ec354bec9..903190a4b3f 100644 --- a/arch/i386/kernel/i386_ksyms.c +++ b/arch/i386/kernel/i386_ksyms.c @@ -169,10 +169,6 @@ EXPORT_SYMBOL(rtc_lock); EXPORT_SYMBOL_GPL(set_nmi_callback); EXPORT_SYMBOL_GPL(unset_nmi_callback); -#undef memcmp -extern int memcmp(const void *,const void *,__kernel_size_t); -EXPORT_SYMBOL(memcmp); - EXPORT_SYMBOL(register_die_notifier); #ifdef CONFIG_HAVE_DEC_LOCK EXPORT_SYMBOL(_atomic_dec_and_lock); diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 5e0d55be543..7a324e8b86f 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -2175,7 +2175,6 @@ static inline void check_timer(void) disable_8259A_irq(0); setup_nmi(); enable_8259A_irq(0); - check_nmi_watchdog(); } return; } @@ -2198,7 +2197,6 @@ static inline void check_timer(void) add_pin_to_irq(0, 0, pin2); if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); - check_nmi_watchdog(); } return; } diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index 67168165924..59ff9b45506 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -217,6 +217,13 @@ static void resume_execution(struct kprobe *p, struct pt_regs *regs) *tos &= ~(TF_MASK | IF_MASK); *tos |= kprobe_old_eflags; break; + case 0xc3: /* ret/lret */ + case 0xcb: + case 0xc2: + case 0xca: + regs->eflags &= ~TF_MASK; + /* eip is already adjusted, no more changes required*/ + return; case 0xe8: /* call relative - Fix return addr */ *tos = orig_eip + (*tos - copy_eip); break; diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index 2f89d000f95..2c0ee9c2d02 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -102,20 +102,21 @@ int nmi_active; (P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT| \ P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE) -int __init check_nmi_watchdog (void) +static int __init check_nmi_watchdog(void) { unsigned int prev_nmi_count[NR_CPUS]; int cpu; - printk(KERN_INFO "testing NMI watchdog ... "); + if (nmi_watchdog == NMI_NONE) + return 0; + + printk(KERN_INFO "Testing NMI watchdog ... "); for (cpu = 0; cpu < NR_CPUS; cpu++) prev_nmi_count[cpu] = per_cpu(irq_stat, cpu).__nmi_count; local_irq_enable(); mdelay((10*1000)/nmi_hz); // wait 10 ticks - /* FIXME: Only boot CPU is online at this stage. Check CPUs - as they come up. */ for (cpu = 0; cpu < NR_CPUS; cpu++) { #ifdef CONFIG_SMP /* Check cpu_callin_map here because that is set @@ -139,6 +140,8 @@ int __init check_nmi_watchdog (void) return 0; } +/* This needs to happen later in boot so counters are working */ +late_initcall(check_nmi_watchdog); static int __init setup_nmi_watchdog(char *str) { diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index b2203e21acb..96e3ea6b17c 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -400,11 +400,6 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long esp, int err; childregs = ((struct pt_regs *) (THREAD_SIZE + (unsigned long) p->thread_info)) - 1; - *childregs = *regs; - childregs->eax = 0; - childregs->esp = esp; - - p->thread.esp = (unsigned long) childregs; /* * The below -8 is to reserve 8 bytes on top of the ring0 stack. * This is necessary to guarantee that the entire "struct pt_regs" @@ -415,7 +410,13 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long esp, * "struct pt_regs" is possible, but they may contain the * completely wrong values. */ - p->thread.esp0 = (unsigned long) (childregs+1) - 8; + childregs = (struct pt_regs *) ((unsigned long) childregs - 8); + *childregs = *regs; + childregs->eax = 0; + childregs->esp = esp; + + p->thread.esp = (unsigned long) childregs; + p->thread.esp0 = (unsigned long) (childregs+1); p->thread.eip = (unsigned long) ret_from_fork; @@ -611,8 +612,8 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas * Save away %fs and %gs. No need to save %es and %ds, as * those are always kernel segments while inside the kernel. */ - asm volatile("movl %%fs,%0":"=m" (*(int *)&prev->fs)); - asm volatile("movl %%gs,%0":"=m" (*(int *)&prev->gs)); + asm volatile("mov %%fs,%0":"=m" (prev->fs)); + asm volatile("mov %%gs,%0":"=m" (prev->gs)); /* * Restore %fs and %gs if needed. diff --git a/arch/i386/kernel/ptrace.c b/arch/i386/kernel/ptrace.c index b2f17640cef..e34f651fa13 100644 --- a/arch/i386/kernel/ptrace.c +++ b/arch/i386/kernel/ptrace.c @@ -16,6 +16,7 @@ #include <linux/security.h> #include <linux/audit.h> #include <linux/seccomp.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -511,7 +512,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -543,7 +544,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); set_singlestep(child); @@ -682,24 +683,18 @@ void do_syscall_trace(struct pt_regs *regs, int entryexit) /* do the secure computing check first */ secure_computing(regs->orig_eax); - if (unlikely(current->audit_context)) { - if (!entryexit) - audit_syscall_entry(current, regs->orig_eax, - regs->ebx, regs->ecx, - regs->edx, regs->esi); - else - audit_syscall_exit(current, regs->eax); - } + if (unlikely(current->audit_context) && entryexit) + audit_syscall_exit(current, AUDITSC_RESULT(regs->eax), regs->eax); if (!(current->ptrace & PT_PTRACED)) - return; + goto out; /* Fake a debug trap */ if (test_thread_flag(TIF_SINGLESTEP)) send_sigtrap(current, regs, 0); if (!test_thread_flag(TIF_SYSCALL_TRACE)) - return; + goto out; /* the 0x80 provides a way for the tracing parent to distinguish between a syscall stop and SIGTRAP delivery */ @@ -714,4 +709,9 @@ void do_syscall_trace(struct pt_regs *regs, int entryexit) send_sig(current->exit_code, current, 1); current->exit_code = 0; } + out: + if (unlikely(current->audit_context) && !entryexit) + audit_syscall_entry(current, AUDIT_ARCH_I386, regs->orig_eax, + regs->ebx, regs->ecx, regs->edx, regs->esi); + } diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c index 3d7e994563d..6dc27eb70ee 100644 --- a/arch/i386/kernel/reboot.c +++ b/arch/i386/kernel/reboot.c @@ -13,6 +13,7 @@ #include <asm/uaccess.h> #include <asm/apic.h> #include "mach_reboot.h" +#include <linux/reboot_fixups.h> /* * Power off function, if any @@ -348,6 +349,7 @@ void machine_restart(char * __unused) /* rebooting needs to touch the page at absolute addr 0 */ *((unsigned short *)__va(0x472)) = reboot_mode; for (;;) { + mach_reboot_fixups(); /* for board specific fixups */ mach_reboot(); /* That didn't work - force a triple fault.. */ __asm__ __volatile__("lidt %0": :"m" (no_idt)); diff --git a/arch/i386/kernel/reboot_fixups.c b/arch/i386/kernel/reboot_fixups.c new file mode 100644 index 00000000000..1b183b378c2 --- /dev/null +++ b/arch/i386/kernel/reboot_fixups.c @@ -0,0 +1,56 @@ +/* + * linux/arch/i386/kernel/reboot_fixups.c + * + * This is a good place to put board specific reboot fixups. + * + * List of supported fixups: + * geode-gx1/cs5530a - Jaya Kumar <jayalk@intworks.biz> + * + */ + +#include <asm/delay.h> +#include <linux/pci.h> + +static void cs5530a_warm_reset(struct pci_dev *dev) +{ + /* writing 1 to the reset control register, 0x44 causes the + cs5530a to perform a system warm reset */ + pci_write_config_byte(dev, 0x44, 0x1); + udelay(50); /* shouldn't get here but be safe and spin-a-while */ + return; +} + +struct device_fixup { + unsigned int vendor; + unsigned int device; + void (*reboot_fixup)(struct pci_dev *); +}; + +static struct device_fixup fixups_table[] = { +{ PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_LEGACY, cs5530a_warm_reset }, +}; + +/* + * we see if any fixup is available for our current hardware. if there + * is a fixup, we call it and we expect to never return from it. if we + * do return, we keep looking and then eventually fall back to the + * standard mach_reboot on return. + */ +void mach_reboot_fixups(void) +{ + struct device_fixup *cur; + struct pci_dev *dev; + int i; + + for (i=0; i < (sizeof(fixups_table)/sizeof(fixups_table[0])); i++) { + cur = &(fixups_table[i]); + dev = pci_get_device(cur->vendor, cur->device, 0); + if (!dev) + continue; + + cur->reboot_fixup(dev); + } + + printk(KERN_WARNING "No reboot fixup found for your hardware\n"); +} + diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index fd36d2f65f8..cbea7ac582e 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -1089,9 +1089,6 @@ static void __init smp_boot_cpus(unsigned int max_cpus) } } - if (nmi_watchdog == NMI_LOCAL_APIC) - check_nmi_watchdog(); - smpboot_setup_io_apic(); setup_boot_APIC_clock(); diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S new file mode 100644 index 00000000000..6cd1ed311f0 --- /dev/null +++ b/arch/i386/kernel/syscall_table.S @@ -0,0 +1,291 @@ +.data +ENTRY(sys_call_table) + .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ + .long sys_exit + .long sys_fork + .long sys_read + .long sys_write + .long sys_open /* 5 */ + .long sys_close + .long sys_waitpid + .long sys_creat + .long sys_link + .long sys_unlink /* 10 */ + .long sys_execve + .long sys_chdir + .long sys_time + .long sys_mknod + .long sys_chmod /* 15 */ + .long sys_lchown16 + .long sys_ni_syscall /* old break syscall holder */ + .long sys_stat + .long sys_lseek + .long sys_getpid /* 20 */ + .long sys_mount + .long sys_oldumount + .long sys_setuid16 + .long sys_getuid16 + .long sys_stime /* 25 */ + .long sys_ptrace + .long sys_alarm + .long sys_fstat + .long sys_pause + .long sys_utime /* 30 */ + .long sys_ni_syscall /* old stty syscall holder */ + .long sys_ni_syscall /* old gtty syscall holder */ + .long sys_access + .long sys_nice + .long sys_ni_syscall /* 35 - old ftime syscall holder */ + .long sys_sync + .long sys_kill + .long sys_rename + .long sys_mkdir + .long sys_rmdir /* 40 */ + .long sys_dup + .long sys_pipe + .long sys_times + .long sys_ni_syscall /* old prof syscall holder */ + .long sys_brk /* 45 */ + .long sys_setgid16 + .long sys_getgid16 + .long sys_signal + .long sys_geteuid16 + .long sys_getegid16 /* 50 */ + .long sys_acct + .long sys_umount /* recycled never used phys() */ + .long sys_ni_syscall /* old lock syscall holder */ + .long sys_ioctl + .long sys_fcntl /* 55 */ + .long sys_ni_syscall /* old mpx syscall holder */ + .long sys_setpgid + .long sys_ni_syscall /* old ulimit syscall holder */ + .long sys_olduname + .long sys_umask /* 60 */ + .long sys_chroot + .long sys_ustat + .long sys_dup2 + .long sys_getppid + .long sys_getpgrp /* 65 */ + .long sys_setsid + .long sys_sigaction + .long sys_sgetmask + .long sys_ssetmask + .long sys_setreuid16 /* 70 */ + .long sys_setregid16 + .long sys_sigsuspend + .long sys_sigpending + .long sys_sethostname + .long sys_setrlimit /* 75 */ + .long sys_old_getrlimit + .long sys_getrusage + .long sys_gettimeofday + .long sys_settimeofday + .long sys_getgroups16 /* 80 */ + .long sys_setgroups16 + .long old_select + .long sys_symlink + .long sys_lstat + .long sys_readlink /* 85 */ + .long sys_uselib + .long sys_swapon + .long sys_reboot + .long old_readdir + .long old_mmap /* 90 */ + .long sys_munmap + .long sys_truncate + .long sys_ftruncate + .long sys_fchmod + .long sys_fchown16 /* 95 */ + .long sys_getpriority + .long sys_setpriority + .long sys_ni_syscall /* old profil syscall holder */ + .long sys_statfs + .long sys_fstatfs /* 100 */ + .long sys_ioperm + .long sys_socketcall + .long sys_syslog + .long sys_setitimer + .long sys_getitimer /* 105 */ + .long sys_newstat + .long sys_newlstat + .long sys_newfstat + .long sys_uname + .long sys_iopl /* 110 */ + .long sys_vhangup + .long sys_ni_syscall /* old "idle" system call */ + .long sys_vm86old + .long sys_wait4 + .long sys_swapoff /* 115 */ + .long sys_sysinfo + .long sys_ipc + .long sys_fsync + .long sys_sigreturn + .long sys_clone /* 120 */ + .long sys_setdomainname + .long sys_newuname + .long sys_modify_ldt + .long sys_adjtimex + .long sys_mprotect /* 125 */ + .long sys_sigprocmask + .long sys_ni_syscall /* old "create_module" */ + .long sys_init_module + .long sys_delete_module + .long sys_ni_syscall /* 130: old "get_kernel_syms" */ + .long sys_quotactl + .long sys_getpgid + .long sys_fchdir + .long sys_bdflush + .long sys_sysfs /* 135 */ + .long sys_personality + .long sys_ni_syscall /* reserved for afs_syscall */ + .long sys_setfsuid16 + .long sys_setfsgid16 + .long sys_llseek /* 140 */ + .long sys_getdents + .long sys_select + .long sys_flock + .long sys_msync + .long sys_readv /* 145 */ + .long sys_writev + .long sys_getsid + .long sys_fdatasync + .long sys_sysctl + .long sys_mlock /* 150 */ + .long sys_munlock + .long sys_mlockall + .long sys_munlockall + .long sys_sched_setparam + .long sys_sched_getparam /* 155 */ + .long sys_sched_setscheduler + .long sys_sched_getscheduler + .long sys_sched_yield + .long sys_sched_get_priority_max + .long sys_sched_get_priority_min /* 160 */ + .long sys_sched_rr_get_interval + .long sys_nanosleep + .long sys_mremap + .long sys_setresuid16 + .long sys_getresuid16 /* 165 */ + .long sys_vm86 + .long sys_ni_syscall /* Old sys_query_module */ + .long sys_poll + .long sys_nfsservctl + .long sys_setresgid16 /* 170 */ + .long sys_getresgid16 + .long sys_prctl + .long sys_rt_sigreturn + .long sys_rt_sigaction + .long sys_rt_sigprocmask /* 175 */ + .long sys_rt_sigpending + .long sys_rt_sigtimedwait + .long sys_rt_sigqueueinfo + .long sys_rt_sigsuspend + .long sys_pread64 /* 180 */ + .long sys_pwrite64 + .long sys_chown16 + .long sys_getcwd + .long sys_capget + .long sys_capset /* 185 */ + .long sys_sigaltstack + .long sys_sendfile + .long sys_ni_syscall /* reserved for streams1 */ + .long sys_ni_syscall /* reserved for streams2 */ + .long sys_vfork /* 190 */ + .long sys_getrlimit + .long sys_mmap2 + .long sys_truncate64 + .long sys_ftruncate64 + .long sys_stat64 /* 195 */ + .long sys_lstat64 + .long sys_fstat64 + .long sys_lchown + .long sys_getuid + .long sys_getgid /* 200 */ + .long sys_geteuid + .long sys_getegid + .long sys_setreuid + .long sys_setregid + .long sys_getgroups /* 205 */ + .long sys_setgroups + .long sys_fchown + .long sys_setresuid + .long sys_getresuid + .long sys_setresgid /* 210 */ + .long sys_getresgid + .long sys_chown + .long sys_setuid + .long sys_setgid + .long sys_setfsuid /* 215 */ + .long sys_setfsgid + .long sys_pivot_root + .long sys_mincore + .long sys_madvise + .long sys_getdents64 /* 220 */ + .long sys_fcntl64 + .long sys_ni_syscall /* reserved for TUX */ + .long sys_ni_syscall + .long sys_gettid + .long sys_readahead /* 225 */ + .long sys_setxattr + .long sys_lsetxattr + .long sys_fsetxattr + .long sys_getxattr + .long sys_lgetxattr /* 230 */ + .long sys_fgetxattr + .long sys_listxattr + .long sys_llistxattr + .long sys_flistxattr + .long sys_removexattr /* 235 */ + .long sys_lremovexattr + .long sys_fremovexattr + .long sys_tkill + .long sys_sendfile64 + .long sys_futex /* 240 */ + .long sys_sched_setaffinity + .long sys_sched_getaffinity + .long sys_set_thread_area + .long sys_get_thread_area + .long sys_io_setup /* 245 */ + .long sys_io_destroy + .long sys_io_getevents + .long sys_io_submit + .long sys_io_cancel + .long sys_fadvise64 /* 250 */ + .long sys_ni_syscall + .long sys_exit_group + .long sys_lookup_dcookie + .long sys_epoll_create + .long sys_epoll_ctl /* 255 */ + .long sys_epoll_wait + .long sys_remap_file_pages + .long sys_set_tid_address + .long sys_timer_create + .long sys_timer_settime /* 260 */ + .long sys_timer_gettime + .long sys_timer_getoverrun + .long sys_timer_delete + .long sys_clock_settime + .long sys_clock_gettime /* 265 */ + .long sys_clock_getres + .long sys_clock_nanosleep + .long sys_statfs64 + .long sys_fstatfs64 + .long sys_tgkill /* 270 */ + .long sys_utimes + .long sys_fadvise64_64 + .long sys_ni_syscall /* sys_vserver */ + .long sys_mbind + .long sys_get_mempolicy + .long sys_set_mempolicy + .long sys_mq_open + .long sys_mq_unlink + .long sys_mq_timedsend + .long sys_mq_timedreceive /* 280 */ + .long sys_mq_notify + .long sys_mq_getsetattr + .long sys_ni_syscall /* reserved for kexec */ + .long sys_waitid + .long sys_ni_syscall /* 285 */ /* available */ + .long sys_add_key + .long sys_request_key + .long sys_keyctl diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c index 4d75b373f90..a0dcb7c87c3 100644 --- a/arch/i386/kernel/time.c +++ b/arch/i386/kernel/time.c @@ -441,7 +441,7 @@ static void __init hpet_time_init(void) set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec); - if (hpet_enable() >= 0) { + if ((hpet_enable() >= 0) && hpet_use_timer) { printk("Using HPET for base-timer\n"); } diff --git a/arch/i386/kernel/time_hpet.c b/arch/i386/kernel/time_hpet.c index 244a31b04be..10a0cbb88e7 100644 --- a/arch/i386/kernel/time_hpet.c +++ b/arch/i386/kernel/time_hpet.c @@ -26,6 +26,7 @@ static unsigned long hpet_period; /* fsecs / HPET clock */ unsigned long hpet_tick; /* hpet clks count per tick */ unsigned long hpet_address; /* hpet memory map physical address */ +int hpet_use_timer; static int use_hpet; /* can be used for runtime check of hpet */ static int boot_hpet_disable; /* boottime override for HPET timer */ @@ -73,27 +74,30 @@ static int hpet_timer_stop_set_go(unsigned long tick) hpet_writel(0, HPET_COUNTER); hpet_writel(0, HPET_COUNTER + 4); - /* - * Set up timer 0, as periodic with first interrupt to happen at - * hpet_tick, and period also hpet_tick. - */ - cfg = hpet_readl(HPET_T0_CFG); - cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | - HPET_TN_SETVAL | HPET_TN_32BIT; - hpet_writel(cfg, HPET_T0_CFG); - - /* - * The first write after writing TN_SETVAL to the config register sets - * the counter value, the second write sets the threshold. - */ - hpet_writel(tick, HPET_T0_CMP); - hpet_writel(tick, HPET_T0_CMP); + if (hpet_use_timer) { + /* + * Set up timer 0, as periodic with first interrupt to happen at + * hpet_tick, and period also hpet_tick. + */ + cfg = hpet_readl(HPET_T0_CFG); + cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | + HPET_TN_SETVAL | HPET_TN_32BIT; + hpet_writel(cfg, HPET_T0_CFG); + /* + * The first write after writing TN_SETVAL to the config register sets + * the counter value, the second write sets the threshold. + */ + hpet_writel(tick, HPET_T0_CMP); + hpet_writel(tick, HPET_T0_CMP); + } /* * Go! */ cfg = hpet_readl(HPET_CFG); - cfg |= HPET_CFG_ENABLE | HPET_CFG_LEGACY; + if (hpet_use_timer) + cfg |= HPET_CFG_LEGACY; + cfg |= HPET_CFG_ENABLE; hpet_writel(cfg, HPET_CFG); return 0; @@ -128,12 +132,11 @@ int __init hpet_enable(void) * However, we can do with one timer otherwise using the * the single HPET timer for system time. */ - if ( #ifdef CONFIG_HPET_EMULATE_RTC - !(id & HPET_ID_NUMBER) || -#endif - !(id & HPET_ID_LEGSUP)) + if (!(id & HPET_ID_NUMBER)) return -1; +#endif + hpet_period = hpet_readl(HPET_PERIOD); if ((hpet_period < HPET_MIN_PERIOD) || (hpet_period > HPET_MAX_PERIOD)) @@ -152,6 +155,8 @@ int __init hpet_enable(void) if (hpet_tick_rem > (hpet_period >> 1)) hpet_tick++; /* rounding the result */ + hpet_use_timer = id & HPET_ID_LEGSUP; + if (hpet_timer_stop_set_go(hpet_tick)) return -1; @@ -202,7 +207,8 @@ int __init hpet_enable(void) #endif #ifdef CONFIG_X86_LOCAL_APIC - wait_timer_tick = wait_hpet_tick; + if (hpet_use_timer) + wait_timer_tick = wait_hpet_tick; #endif return 0; } diff --git a/arch/i386/kernel/timers/timer_hpet.c b/arch/i386/kernel/timers/timer_hpet.c index 713134e7184..f778f471a09 100644 --- a/arch/i386/kernel/timers/timer_hpet.c +++ b/arch/i386/kernel/timers/timer_hpet.c @@ -79,7 +79,7 @@ static unsigned long get_offset_hpet(void) eax = hpet_readl(HPET_COUNTER); eax -= hpet_last; /* hpet delta */ - + eax = min(hpet_tick, eax); /* * Time offset = (hpet delta) * ( usecs per HPET clock ) * = (hpet delta) * ( usecs per tick / HPET clocks per tick) @@ -105,9 +105,12 @@ static void mark_offset_hpet(void) last_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; rdtsc(last_tsc_low, last_tsc_high); - offset = hpet_readl(HPET_T0_CMP) - hpet_tick; - if (unlikely(((offset - hpet_last) > hpet_tick) && (hpet_last != 0))) { - int lost_ticks = (offset - hpet_last) / hpet_tick; + if (hpet_use_timer) + offset = hpet_readl(HPET_T0_CMP) - hpet_tick; + else + offset = hpet_readl(HPET_COUNTER); + if (unlikely(((offset - hpet_last) >= (2*hpet_tick)) && (hpet_last != 0))) { + int lost_ticks = ((offset - hpet_last) / hpet_tick) - 1; jiffies_64 += lost_ticks; } hpet_last = offset; diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c index a685994e5c8..7926d967be0 100644 --- a/arch/i386/kernel/timers/timer_tsc.c +++ b/arch/i386/kernel/timers/timer_tsc.c @@ -477,7 +477,7 @@ static int __init init_tsc(char* override) if (cpu_has_tsc) { unsigned long tsc_quotient; #ifdef CONFIG_HPET_TIMER - if (is_hpet_enabled()){ + if (is_hpet_enabled() && hpet_use_timer) { unsigned long result, remain; printk("Using TSC for gettimeofday\n"); tsc_quotient = calibrate_tsc_hpet(NULL); diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 6c0e383915b..00c63419c06 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -451,6 +451,7 @@ DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS) DO_ERROR(11, SIGBUS, "segment not present", segment_not_present) DO_ERROR(12, SIGBUS, "stack segment", stack_segment) DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0) +DO_ERROR_INFO(32, SIGSEGV, "iret exception", iret_error, ILL_BADSTK, 0) fastcall void do_general_protection(struct pt_regs * regs, long error_code) { @@ -642,16 +643,15 @@ void unset_nmi_callback(void) } #ifdef CONFIG_KPROBES -fastcall int do_int3(struct pt_regs *regs, long error_code) +fastcall void do_int3(struct pt_regs *regs, long error_code) { if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) - return 1; + return; /* This is an interrupt gate, because kprobes wants interrupts disabled. Normal trap handlers don't. */ restore_interrupts(regs); do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL); - return 0; } #endif diff --git a/arch/i386/kernel/vm86.c b/arch/i386/kernel/vm86.c index 2f3d52dacff..ec0f68ce688 100644 --- a/arch/i386/kernel/vm86.c +++ b/arch/i386/kernel/vm86.c @@ -222,7 +222,7 @@ asmlinkage int sys_vm86(struct pt_regs regs) goto out; case VM86_PLUS_INSTALL_CHECK: /* NOTE: on old vm86 stuff this will return the error - from verify_area(), because the subfunction is + from access_ok(), because the subfunction is interpreted as (invalid) address to vm86_struct. So the installation check works. */ @@ -294,8 +294,8 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk */ info->regs32->eax = 0; tsk->thread.saved_esp0 = tsk->thread.esp0; - asm volatile("movl %%fs,%0":"=m" (tsk->thread.saved_fs)); - asm volatile("movl %%gs,%0":"=m" (tsk->thread.saved_gs)); + asm volatile("mov %%fs,%0":"=m" (tsk->thread.saved_fs)); + asm volatile("mov %%gs,%0":"=m" (tsk->thread.saved_gs)); tss = &per_cpu(init_tss, get_cpu()); tsk->thread.esp0 = (unsigned long) &info->VM86_TSS_ESP0; @@ -717,12 +717,12 @@ static irqreturn_t irq_handler(int intno, void *dev_id, struct pt_regs * regs) irqbits |= irq_bit; if (vm86_irqs[intno].sig) send_sig(vm86_irqs[intno].sig, vm86_irqs[intno].tsk, 1); - spin_unlock_irqrestore(&irqbits_lock, flags); /* * IRQ will be re-enabled when user asks for the irq (whether * polling or as a result of the signal) */ - disable_irq(intno); + disable_irq_nosync(intno); + spin_unlock_irqrestore(&irqbits_lock, flags); return IRQ_HANDLED; out: @@ -754,17 +754,20 @@ static inline int get_and_reset_irq(int irqnumber) { int bit; unsigned long flags; + int ret = 0; if (invalid_vm86_irq(irqnumber)) return 0; if (vm86_irqs[irqnumber].tsk != current) return 0; spin_lock_irqsave(&irqbits_lock, flags); bit = irqbits & (1 << irqnumber); irqbits &= ~bit; + if (bit) { + enable_irq(irqnumber); + ret = 1; + } + spin_unlock_irqrestore(&irqbits_lock, flags); - if (!bit) - return 0; - enable_irq(irqnumber); - return 1; + return ret; } diff --git a/arch/i386/oprofile/nmi_timer_int.c b/arch/i386/oprofile/nmi_timer_int.c index b2e462abf33..c58d0c14f27 100644 --- a/arch/i386/oprofile/nmi_timer_int.c +++ b/arch/i386/oprofile/nmi_timer_int.c @@ -36,7 +36,7 @@ static void timer_stop(void) { enable_timer_nmi_watchdog(); unset_nmi_callback(); - synchronize_kernel(); + synchronize_sched(); /* Allow already-started NMIs to complete. */ } diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index e07589d04f6..d6598da4b67 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -495,6 +495,8 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH6_1: case PCI_DEVICE_ID_INTEL_ICH7_0: case PCI_DEVICE_ID_INTEL_ICH7_1: + case PCI_DEVICE_ID_INTEL_ICH7_30: + case PCI_DEVICE_ID_INTEL_ICH7_31: case PCI_DEVICE_ID_INTEL_ESB2_0: r->name = "PIIX/ICH"; r->get = pirq_piix_get; diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 33fcb205fcb..3ad2c4af099 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -46,6 +46,10 @@ config GENERIC_IOMAP bool default y +config SCHED_NO_NO_OMIT_FRAME_POINTER + bool + default y + choice prompt "System type" default IA64_GENERIC @@ -217,6 +221,16 @@ config IA64_SGI_SN_SIM If you are compiling a kernel that will run under SGI's IA-64 simulator (Medusa) then say Y, otherwise say N. +config IA64_SGI_SN_XP + tristate "Support communication between SGI SSIs" + depends on MSPEC + help + An SGI machine can be divided into multiple Single System + Images which act independently of each other and have + hardware based memory protection from the others. Enabling + this feature will allow for direct communication between SSIs + based on a network adapter and DMA messaging. + config FORCE_MAX_ZONEORDER int default "18" @@ -261,6 +275,15 @@ config HOTPLUG_CPU can be controlled through /sys/devices/system/cpu/cpu#. Say N if you want to disable CPU hotplug. +config SCHED_SMT + bool "SMT scheduler support" + depends on SMP + default off + help + Improves the CPU scheduler's decision making when dealing with + Intel IA64 chips with MultiThreading at a cost of slightly increased + overhead in some places. If unsure say N here. + config PREEMPT bool "Preemptible Kernel" help @@ -329,7 +352,7 @@ menu "Power management and ACPI" config PM bool "Power Management support" - depends on IA64_GENERIC || IA64_DIG || IA64_HP_ZX1 || IA64_HP_ZX1_SWIOTLB + depends on !IA64_HP_SIM default y help "Power Management" means that parts of your computer are shut diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig index bfeb952fe8e..6ff7107fee4 100644 --- a/arch/ia64/configs/sn2_defconfig +++ b/arch/ia64/configs/sn2_defconfig @@ -574,6 +574,8 @@ CONFIG_SERIAL_NONSTANDARD=y # CONFIG_N_HDLC is not set # CONFIG_STALDRV is not set CONFIG_SGI_SNSC=y +CONFIG_SGI_TIOCX=y +CONFIG_SGI_MBCS=m # # Serial drivers diff --git a/arch/ia64/configs/tiger_defconfig b/arch/ia64/configs/tiger_defconfig index 99830e8fc9b..9086b789f6a 100644 --- a/arch/ia64/configs/tiger_defconfig +++ b/arch/ia64/configs/tiger_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.11-rc2 -# Sat Jan 22 11:17:02 2005 +# Linux kernel version: 2.6.12-rc3 +# Tue May 3 15:55:04 2005 # # @@ -10,6 +10,7 @@ CONFIG_EXPERIMENTAL=y CONFIG_CLEAN_COMPILE=y CONFIG_LOCK_KERNEL=y +CONFIG_INIT_ENV_ARG_LIMIT=32 # # General setup @@ -21,24 +22,27 @@ CONFIG_POSIX_MQUEUE=y # CONFIG_BSD_PROCESS_ACCT is not set CONFIG_SYSCTL=y # CONFIG_AUDIT is not set -CONFIG_LOG_BUF_SHIFT=20 CONFIG_HOTPLUG=y CONFIG_KOBJECT_UEVENT=y CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y +# CONFIG_CPUSETS is not set # CONFIG_EMBEDDED is not set CONFIG_KALLSYMS=y CONFIG_KALLSYMS_ALL=y # CONFIG_KALLSYMS_EXTRA_PASS is not set +CONFIG_PRINTK=y +CONFIG_BUG=y +CONFIG_BASE_FULL=y CONFIG_FUTEX=y CONFIG_EPOLL=y -# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set CONFIG_SHMEM=y CONFIG_CC_ALIGN_FUNCTIONS=0 CONFIG_CC_ALIGN_LABELS=0 CONFIG_CC_ALIGN_LOOPS=0 CONFIG_CC_ALIGN_JUMPS=0 # CONFIG_TINY_SHMEM is not set +CONFIG_BASE_SMALL=0 # # Loadable module support @@ -85,6 +89,7 @@ CONFIG_FORCE_MAX_ZONEORDER=18 CONFIG_SMP=y CONFIG_NR_CPUS=4 CONFIG_HOTPLUG_CPU=y +# CONFIG_SCHED_SMT is not set # CONFIG_PREEMPT is not set CONFIG_HAVE_DEC_LOCK=y CONFIG_IA32_SUPPORT=y @@ -135,6 +140,7 @@ CONFIG_PCI_DOMAINS=y # CONFIG_PCI_MSI is not set CONFIG_PCI_LEGACY_PROC=y CONFIG_PCI_NAMES=y +# CONFIG_PCI_DEBUG is not set # # PCI Hotplug Support @@ -152,10 +158,6 @@ CONFIG_HOTPLUG_PCI_ACPI=m # CONFIG_PCCARD is not set # -# PC-card bridges -# - -# # Device Drivers # @@ -195,9 +197,10 @@ CONFIG_BLK_DEV_CRYPTOLOOP=m CONFIG_BLK_DEV_NBD=m # CONFIG_BLK_DEV_SX8 is not set # CONFIG_BLK_DEV_UB is not set -CONFIG_BLK_DEV_RAM=m +CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=4096 +CONFIG_BLK_DEV_INITRD=y CONFIG_INITRAMFS_SOURCE="" # CONFIG_CDROM_PKTCDVD is not set @@ -313,7 +316,6 @@ CONFIG_SCSI_FC_ATTRS=y # CONFIG_SCSI_BUSLOGIC is not set # CONFIG_SCSI_DMX3191D is not set # CONFIG_SCSI_EATA is not set -# CONFIG_SCSI_EATA_PIO is not set # CONFIG_SCSI_FUTURE_DOMAIN is not set # CONFIG_SCSI_GDTH is not set # CONFIG_SCSI_IPS is not set @@ -325,7 +327,6 @@ CONFIG_SCSI_SYM53C8XX_DEFAULT_TAGS=16 CONFIG_SCSI_SYM53C8XX_MAX_TAGS=64 # CONFIG_SCSI_SYM53C8XX_IOMAPPED is not set # CONFIG_SCSI_IPR is not set -# CONFIG_SCSI_QLOGIC_ISP is not set CONFIG_SCSI_QLOGIC_FC=y # CONFIG_SCSI_QLOGIC_FC_FIRMWARE is not set CONFIG_SCSI_QLOGIC_1280=y @@ -336,6 +337,7 @@ CONFIG_SCSI_QLA22XX=m CONFIG_SCSI_QLA2300=m CONFIG_SCSI_QLA2322=m # CONFIG_SCSI_QLA6312 is not set +# CONFIG_SCSI_LPFC is not set # CONFIG_SCSI_DC395x is not set # CONFIG_SCSI_DC390T is not set # CONFIG_SCSI_DEBUG is not set @@ -358,6 +360,7 @@ CONFIG_DM_CRYPT=m CONFIG_DM_SNAPSHOT=m CONFIG_DM_MIRROR=m CONFIG_DM_ZERO=m +# CONFIG_DM_MULTIPATH is not set # # Fusion MPT device support @@ -386,7 +389,6 @@ CONFIG_NET=y # CONFIG_PACKET=y # CONFIG_PACKET_MMAP is not set -CONFIG_NETLINK_DEV=y CONFIG_UNIX=y # CONFIG_NET_KEY is not set CONFIG_INET=y @@ -446,7 +448,6 @@ CONFIG_DUMMY=m # CONFIG_BONDING is not set # CONFIG_EQUALIZER is not set # CONFIG_TUN is not set -# CONFIG_ETHERTAP is not set # # ARCnet devices @@ -484,7 +485,6 @@ CONFIG_NET_PCI=y # CONFIG_DGRS is not set CONFIG_EEPRO100=m CONFIG_E100=m -# CONFIG_E100_NAPI is not set # CONFIG_FEALNX is not set # CONFIG_NATSEMI is not set # CONFIG_NE2K_PCI is not set @@ -566,25 +566,6 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768 # CONFIG_INPUT_EVBUG is not set # -# Input I/O drivers -# -CONFIG_GAMEPORT=m -CONFIG_SOUND_GAMEPORT=m -# CONFIG_GAMEPORT_NS558 is not set -# CONFIG_GAMEPORT_L4 is not set -# CONFIG_GAMEPORT_EMU10K1 is not set -# CONFIG_GAMEPORT_VORTEX is not set -# CONFIG_GAMEPORT_FM801 is not set -# CONFIG_GAMEPORT_CS461X is not set -CONFIG_SERIO=y -CONFIG_SERIO_I8042=y -# CONFIG_SERIO_SERPORT is not set -# CONFIG_SERIO_CT82C710 is not set -# CONFIG_SERIO_PCIPS2 is not set -CONFIG_SERIO_LIBPS2=y -# CONFIG_SERIO_RAW is not set - -# # Input Device Drivers # CONFIG_INPUT_KEYBOARD=y @@ -602,6 +583,24 @@ CONFIG_MOUSE_PS2=y # CONFIG_INPUT_MISC is not set # +# Hardware I/O ports +# +CONFIG_SERIO=y +CONFIG_SERIO_I8042=y +# CONFIG_SERIO_SERPORT is not set +# CONFIG_SERIO_PCIPS2 is not set +CONFIG_SERIO_LIBPS2=y +# CONFIG_SERIO_RAW is not set +CONFIG_GAMEPORT=m +# CONFIG_GAMEPORT_NS558 is not set +# CONFIG_GAMEPORT_L4 is not set +# CONFIG_GAMEPORT_EMU10K1 is not set +# CONFIG_GAMEPORT_VORTEX is not set +# CONFIG_GAMEPORT_FM801 is not set +# CONFIG_GAMEPORT_CS461X is not set +CONFIG_SOUND_GAMEPORT=m + +# # Character devices # CONFIG_VT=y @@ -615,6 +614,8 @@ CONFIG_SERIAL_NONSTANDARD=y # CONFIG_SYNCLINK is not set # CONFIG_SYNCLINKMP is not set # CONFIG_N_HDLC is not set +# CONFIG_SPECIALIX is not set +# CONFIG_SX is not set # CONFIG_STALDRV is not set # @@ -635,6 +636,7 @@ CONFIG_SERIAL_8250_SHARE_IRQ=y # CONFIG_SERIAL_CORE=y CONFIG_SERIAL_CORE_CONSOLE=y +# CONFIG_SERIAL_JSM is not set CONFIG_UNIX98_PTYS=y CONFIG_LEGACY_PTYS=y CONFIG_LEGACY_PTY_COUNT=256 @@ -670,6 +672,12 @@ CONFIG_HPET=y # CONFIG_HPET_RTC_IRQ is not set CONFIG_HPET_MMAP=y CONFIG_MAX_RAW_DEVS=256 +# CONFIG_HANGCHECK_TIMER is not set + +# +# TPM devices +# +# CONFIG_TCG_TPM is not set # # I2C support @@ -705,7 +713,6 @@ CONFIG_MAX_RAW_DEVS=256 # CONFIG_VGA_CONSOLE=y CONFIG_DUMMY_CONSOLE=y -# CONFIG_BACKLIGHT_LCD_SUPPORT is not set # # Sound @@ -715,6 +722,8 @@ CONFIG_DUMMY_CONSOLE=y # # USB support # +CONFIG_USB_ARCH_HAS_HCD=y +CONFIG_USB_ARCH_HAS_OHCI=y CONFIG_USB=y # CONFIG_USB_DEBUG is not set @@ -726,8 +735,6 @@ CONFIG_USB_DEVICEFS=y # CONFIG_USB_DYNAMIC_MINORS is not set # CONFIG_USB_SUSPEND is not set # CONFIG_USB_OTG is not set -CONFIG_USB_ARCH_HAS_HCD=y -CONFIG_USB_ARCH_HAS_OHCI=y # # USB Host Controller Drivers @@ -736,6 +743,8 @@ CONFIG_USB_EHCI_HCD=m # CONFIG_USB_EHCI_SPLIT_ISO is not set # CONFIG_USB_EHCI_ROOT_HUB_TT is not set CONFIG_USB_OHCI_HCD=m +# CONFIG_USB_OHCI_BIG_ENDIAN is not set +CONFIG_USB_OHCI_LITTLE_ENDIAN=y CONFIG_USB_UHCI_HCD=y # CONFIG_USB_SL811_HCD is not set @@ -751,12 +760,11 @@ CONFIG_USB_UHCI_HCD=y # CONFIG_USB_STORAGE=m # CONFIG_USB_STORAGE_DEBUG is not set -# CONFIG_USB_STORAGE_RW_DETECT is not set # CONFIG_USB_STORAGE_DATAFAB is not set # CONFIG_USB_STORAGE_FREECOM is not set # CONFIG_USB_STORAGE_ISD200 is not set # CONFIG_USB_STORAGE_DPCM is not set -# CONFIG_USB_STORAGE_HP8200e is not set +# CONFIG_USB_STORAGE_USBAT is not set # CONFIG_USB_STORAGE_SDDR09 is not set # CONFIG_USB_STORAGE_SDDR55 is not set # CONFIG_USB_STORAGE_JUMPSHOT is not set @@ -800,6 +808,7 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_PEGASUS is not set # CONFIG_USB_RTL8150 is not set # CONFIG_USB_USBNET is not set +# CONFIG_USB_MON is not set # # USB port drivers @@ -824,6 +833,7 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_PHIDGETKIT is not set # CONFIG_USB_PHIDGETSERVO is not set # CONFIG_USB_IDMOUSE is not set +# CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_TEST is not set # @@ -867,7 +877,12 @@ CONFIG_REISERFS_FS_POSIX_ACL=y CONFIG_REISERFS_FS_SECURITY=y # CONFIG_JFS_FS is not set CONFIG_FS_POSIX_ACL=y + +# +# XFS support +# CONFIG_XFS_FS=y +CONFIG_XFS_EXPORT=y # CONFIG_XFS_RT is not set # CONFIG_XFS_QUOTA is not set # CONFIG_XFS_SECURITY is not set @@ -945,7 +960,7 @@ CONFIG_NFSD_V4=y CONFIG_NFSD_TCP=y CONFIG_LOCKD=m CONFIG_LOCKD_V4=y -CONFIG_EXPORTFS=m +CONFIG_EXPORTFS=y CONFIG_SUNRPC=m CONFIG_SUNRPC_GSS=m CONFIG_RPCSEC_GSS_KRB5=m @@ -1042,8 +1057,10 @@ CONFIG_GENERIC_IRQ_PROBE=y # # Kernel hacking # +# CONFIG_PRINTK_TIME is not set CONFIG_DEBUG_KERNEL=y CONFIG_MAGIC_SYSRQ=y +CONFIG_LOG_BUF_SHIFT=20 # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set # CONFIG_DEBUG_SPINLOCK is not set @@ -1077,6 +1094,7 @@ CONFIG_CRYPTO_MD5=m # CONFIG_CRYPTO_SHA256 is not set # CONFIG_CRYPTO_SHA512 is not set # CONFIG_CRYPTO_WP512 is not set +# CONFIG_CRYPTO_TGR192 is not set CONFIG_CRYPTO_DES=m # CONFIG_CRYPTO_BLOWFISH is not set # CONFIG_CRYPTO_TWOFISH is not set diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c index 017c9ab5fc1..b8db6e3e5e8 100644 --- a/arch/ia64/hp/common/sba_iommu.c +++ b/arch/ia64/hp/common/sba_iommu.c @@ -1,9 +1,9 @@ /* ** IA64 System Bus Adapter (SBA) I/O MMU manager ** -** (c) Copyright 2002-2004 Alex Williamson +** (c) Copyright 2002-2005 Alex Williamson ** (c) Copyright 2002-2003 Grant Grundler -** (c) Copyright 2002-2004 Hewlett-Packard Company +** (c) Copyright 2002-2005 Hewlett-Packard Company ** ** Portions (c) 2000 Grant Grundler (from parisc I/O MMU code) ** Portions (c) 1999 Dave S. Miller (from sparc64 I/O MMU code) @@ -459,21 +459,32 @@ get_iovp_order (unsigned long size) * sba_search_bitmap - find free space in IO PDIR resource bitmap * @ioc: IO MMU structure which owns the pdir we are interested in. * @bits_wanted: number of entries we need. + * @use_hint: use res_hint to indicate where to start looking * * Find consecutive free bits in resource bitmap. * Each bit represents one entry in the IO Pdir. * Cool perf optimization: search for log2(size) bits at a time. */ static SBA_INLINE unsigned long -sba_search_bitmap(struct ioc *ioc, unsigned long bits_wanted) +sba_search_bitmap(struct ioc *ioc, unsigned long bits_wanted, int use_hint) { - unsigned long *res_ptr = ioc->res_hint; + unsigned long *res_ptr; unsigned long *res_end = (unsigned long *) &(ioc->res_map[ioc->res_size]); - unsigned long pide = ~0UL; + unsigned long flags, pide = ~0UL; ASSERT(((unsigned long) ioc->res_hint & (sizeof(unsigned long) - 1UL)) == 0); ASSERT(res_ptr < res_end); + spin_lock_irqsave(&ioc->res_lock, flags); + + /* Allow caller to force a search through the entire resource space */ + if (likely(use_hint)) { + res_ptr = ioc->res_hint; + } else { + res_ptr = (ulong *)ioc->res_map; + ioc->res_bitshift = 0; + } + /* * N.B. REO/Grande defect AR2305 can cause TLB fetch timeouts * if a TLB entry is purged while in use. sba_mark_invalid() @@ -570,10 +581,12 @@ not_found: prefetch(ioc->res_map); ioc->res_hint = (unsigned long *) ioc->res_map; ioc->res_bitshift = 0; + spin_unlock_irqrestore(&ioc->res_lock, flags); return (pide); found_it: ioc->res_hint = res_ptr; + spin_unlock_irqrestore(&ioc->res_lock, flags); return (pide); } @@ -594,36 +607,36 @@ sba_alloc_range(struct ioc *ioc, size_t size) unsigned long itc_start; #endif unsigned long pide; - unsigned long flags; ASSERT(pages_needed); ASSERT(0 == (size & ~iovp_mask)); - spin_lock_irqsave(&ioc->res_lock, flags); - #ifdef PDIR_SEARCH_TIMING itc_start = ia64_get_itc(); #endif /* ** "seek and ye shall find"...praying never hurts either... */ - pide = sba_search_bitmap(ioc, pages_needed); + pide = sba_search_bitmap(ioc, pages_needed, 1); if (unlikely(pide >= (ioc->res_size << 3))) { - pide = sba_search_bitmap(ioc, pages_needed); + pide = sba_search_bitmap(ioc, pages_needed, 0); if (unlikely(pide >= (ioc->res_size << 3))) { #if DELAYED_RESOURCE_CNT > 0 + unsigned long flags; + /* ** With delayed resource freeing, we can give this one more shot. We're ** getting close to being in trouble here, so do what we can to make this ** one count. */ - spin_lock(&ioc->saved_lock); + spin_lock_irqsave(&ioc->saved_lock, flags); if (ioc->saved_cnt > 0) { struct sba_dma_pair *d; int cnt = ioc->saved_cnt; - d = &(ioc->saved[ioc->saved_cnt]); + d = &(ioc->saved[ioc->saved_cnt - 1]); + spin_lock(&ioc->res_lock); while (cnt--) { sba_mark_invalid(ioc, d->iova, d->size); sba_free_range(ioc, d->iova, d->size); @@ -631,10 +644,11 @@ sba_alloc_range(struct ioc *ioc, size_t size) } ioc->saved_cnt = 0; READ_REG(ioc->ioc_hpa+IOC_PCOM); /* flush purges */ + spin_unlock(&ioc->res_lock); } - spin_unlock(&ioc->saved_lock); + spin_unlock_irqrestore(&ioc->saved_lock, flags); - pide = sba_search_bitmap(ioc, pages_needed); + pide = sba_search_bitmap(ioc, pages_needed, 0); if (unlikely(pide >= (ioc->res_size << 3))) panic(__FILE__ ": I/O MMU @ %p is out of mapping resources\n", ioc->ioc_hpa); @@ -664,8 +678,6 @@ sba_alloc_range(struct ioc *ioc, size_t size) (uint) ((unsigned long) ioc->res_hint - (unsigned long) ioc->res_map), ioc->res_bitshift ); - spin_unlock_irqrestore(&ioc->res_lock, flags); - return (pide); } @@ -950,6 +962,30 @@ sba_map_single(struct device *dev, void *addr, size_t size, int dir) return SBA_IOVA(ioc, iovp, offset); } +#ifdef ENABLE_MARK_CLEAN +static SBA_INLINE void +sba_mark_clean(struct ioc *ioc, dma_addr_t iova, size_t size) +{ + u32 iovp = (u32) SBA_IOVP(ioc,iova); + int off = PDIR_INDEX(iovp); + void *addr; + + if (size <= iovp_size) { + addr = phys_to_virt(ioc->pdir_base[off] & + ~0xE000000000000FFFULL); + mark_clean(addr, size); + } else { + do { + addr = phys_to_virt(ioc->pdir_base[off] & + ~0xE000000000000FFFULL); + mark_clean(addr, min(size, iovp_size)); + off++; + size -= iovp_size; + } while (size > 0); + } +} +#endif + /** * sba_unmap_single - unmap one IOVA and free resources * @dev: instance of PCI owned by the driver that's asking. @@ -995,6 +1031,10 @@ void sba_unmap_single(struct device *dev, dma_addr_t iova, size_t size, int dir) size += offset; size = ROUNDUP(size, iovp_size); +#ifdef ENABLE_MARK_CLEAN + if (dir == DMA_FROM_DEVICE) + sba_mark_clean(ioc, iova, size); +#endif #if DELAYED_RESOURCE_CNT > 0 spin_lock_irqsave(&ioc->saved_lock, flags); @@ -1021,30 +1061,6 @@ void sba_unmap_single(struct device *dev, dma_addr_t iova, size_t size, int dir) READ_REG(ioc->ioc_hpa+IOC_PCOM); /* flush purges */ spin_unlock_irqrestore(&ioc->res_lock, flags); #endif /* DELAYED_RESOURCE_CNT == 0 */ -#ifdef ENABLE_MARK_CLEAN - if (dir == DMA_FROM_DEVICE) { - u32 iovp = (u32) SBA_IOVP(ioc,iova); - int off = PDIR_INDEX(iovp); - void *addr; - - if (size <= iovp_size) { - addr = phys_to_virt(ioc->pdir_base[off] & - ~0xE000000000000FFFULL); - mark_clean(addr, size); - } else { - size_t byte_cnt = size; - - do { - addr = phys_to_virt(ioc->pdir_base[off] & - ~0xE000000000000FFFULL); - mark_clean(addr, min(byte_cnt, iovp_size)); - off++; - byte_cnt -= iovp_size; - - } while (byte_cnt > 0); - } - } -#endif } @@ -1928,43 +1944,17 @@ sba_connect_bus(struct pci_bus *bus) static void __init sba_map_ioc_to_node(struct ioc *ioc, acpi_handle handle) { - struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; - union acpi_object *obj; - acpi_handle phandle; unsigned int node; + int pxm; ioc->node = MAX_NUMNODES; - /* - * Check for a _PXM on this node first. We don't typically see - * one here, so we'll end up getting it from the parent. - */ - if (ACPI_FAILURE(acpi_evaluate_object(handle, "_PXM", NULL, &buffer))) { - if (ACPI_FAILURE(acpi_get_parent(handle, &phandle))) - return; - - /* Reset the acpi buffer */ - buffer.length = ACPI_ALLOCATE_BUFFER; - buffer.pointer = NULL; + pxm = acpi_get_pxm(handle); - if (ACPI_FAILURE(acpi_evaluate_object(phandle, "_PXM", NULL, - &buffer))) - return; - } - - if (!buffer.length || !buffer.pointer) + if (pxm < 0) return; - obj = buffer.pointer; - - if (obj->type != ACPI_TYPE_INTEGER || - obj->integer.value >= MAX_PXM_DOMAINS) { - acpi_os_free(buffer.pointer); - return; - } - - node = pxm_to_nid_map[obj->integer.value]; - acpi_os_free(buffer.pointer); + node = pxm_to_nid_map[pxm]; if (node >= MAX_NUMNODES || !node_online(node)) return; diff --git a/arch/ia64/ia32/ia32_signal.c b/arch/ia64/ia32/ia32_signal.c index 19b02adce68..ebb89be2aa2 100644 --- a/arch/ia64/ia32/ia32_signal.c +++ b/arch/ia64/ia32/ia32_signal.c @@ -460,10 +460,9 @@ __ia32_rt_sigsuspend (compat_sigset_t *sset, unsigned int sigsetsize, struct sig sigset_t oldset, set; scr->scratch_unat = 0; /* avoid leaking kernel bits to user level */ - memset(&set, 0, sizeof(&set)); + memset(&set, 0, sizeof(set)); - if (memcpy(&set.sig, &sset->sig, sigsetsize)) - return -EFAULT; + memcpy(&set.sig, &sset->sig, sigsetsize); sigdelsetmask(&set, ~_BLOCKABLE); diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index a8e99c56a76..72dfd9e7de0 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -779,7 +779,7 @@ acpi_map_iosapic (acpi_handle handle, u32 depth, void *context, void **ret) union acpi_object *obj; struct acpi_table_iosapic *iosapic; unsigned int gsi_base; - int node; + int pxm, node; /* Only care about objects w/ a method that returns the MADT */ if (ACPI_FAILURE(acpi_evaluate_object(handle, "_MAT", NULL, &buffer))) @@ -805,29 +805,16 @@ acpi_map_iosapic (acpi_handle handle, u32 depth, void *context, void **ret) gsi_base = iosapic->global_irq_base; acpi_os_free(buffer.pointer); - buffer.length = ACPI_ALLOCATE_BUFFER; - buffer.pointer = NULL; /* - * OK, it's an IOSAPIC MADT entry, look for a _PXM method to tell + * OK, it's an IOSAPIC MADT entry, look for a _PXM value to tell * us which node to associate this with. */ - if (ACPI_FAILURE(acpi_evaluate_object(handle, "_PXM", NULL, &buffer))) - return AE_OK; - - if (!buffer.length || !buffer.pointer) - return AE_OK; - - obj = buffer.pointer; - - if (obj->type != ACPI_TYPE_INTEGER || - obj->integer.value >= MAX_PXM_DOMAINS) { - acpi_os_free(buffer.pointer); + pxm = acpi_get_pxm(handle); + if (pxm < 0) return AE_OK; - } - node = pxm_to_nid_map[obj->integer.value]; - acpi_os_free(buffer.pointer); + node = pxm_to_nid_map[pxm]; if (node >= MAX_NUMNODES || !node_online(node) || cpus_empty(node_to_cpumask(node))) diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 0272c010a3b..81c45d44739 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -728,12 +728,8 @@ ENTRY(ia64_leave_syscall) mov f8=f0 // clear f8 ;; ld8 r30=[r2],16 // M0|1 load cr.ifs - mov.m ar.ssd=r0 // M2 clear ar.ssd - cmp.eq p9,p0=r0,r0 // set p9 to indicate that we should restore cr.ifs - ;; ld8 r25=[r3],16 // M0|1 load ar.unat - mov.m ar.csd=r0 // M2 clear ar.csd - mov r22=r0 // clear r22 + cmp.eq p9,p0=r0,r0 // set p9 to indicate that we should restore cr.ifs ;; ld8 r26=[r2],PT(B0)-PT(AR_PFS) // M0|1 load ar.pfs (pKStk) mov r22=psr // M2 read PSR now that interrupts are disabled @@ -756,11 +752,15 @@ ENTRY(ia64_leave_syscall) mov f7=f0 // clear f7 ;; ld8.fill r12=[r2] // restore r12 (sp) + mov.m ar.ssd=r0 // M2 clear ar.ssd + mov r22=r0 // clear r22 + ld8.fill r15=[r3] // restore r15 +(pUStk) st1 [r14]=r17 addl r3=THIS_CPU(ia64_phys_stacked_size_p8),r0 ;; -(pUStk) ld4 r3=[r3] // r3 = cpu_data->phys_stacked_size_p8 -(pUStk) st1 [r14]=r17 +(pUStk) ld4 r17=[r3] // r17 = cpu_data->phys_stacked_size_p8 + mov.m ar.csd=r0 // M2 clear ar.csd mov b6=r18 // I0 restore b6 ;; mov r14=r0 // clear r14 @@ -782,7 +782,7 @@ GLOBAL_ENTRY(ia64_ret_from_ia32_execve) st8.spill [r2]=r8 // store return value in slot for r8 and set unat bit .mem.offset 8,0 st8.spill [r3]=r0 // clear error indication in slot for r10 and set unat bit -END(ia64_ret_from_ia32_execve_syscall) +END(ia64_ret_from_ia32_execve) // fall through #endif /* CONFIG_IA32_SUPPORT */ GLOBAL_ENTRY(ia64_leave_kernel) @@ -1417,7 +1417,7 @@ sys_call_table: data8 sys_msgrcv data8 sys_msgctl data8 sys_shmget - data8 ia64_shmat + data8 sys_shmat data8 sys_shmdt // 1115 data8 sys_shmctl data8 sys_syslog diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S index 0d8650f7fce..4f3cdef7579 100644 --- a/arch/ia64/kernel/fsys.S +++ b/arch/ia64/kernel/fsys.S @@ -611,8 +611,10 @@ GLOBAL_ENTRY(fsys_bubble_down) movl r2=ia64_ret_from_syscall ;; mov rp=r2 // set the real return addr - tbit.z p8,p0=r3,TIF_SYSCALL_TRACE + and r3=_TIF_SYSCALL_TRACEAUDIT,r3 ;; + cmp.eq p8,p0=r3,r0 + (p10) br.cond.spnt.many ia64_ret_from_syscall // p10==true means out registers are more than 8 (p8) br.call.sptk.many b6=b6 // ignore this return addr br.cond.sptk ia64_trace_syscall diff --git a/arch/ia64/kernel/head.S b/arch/ia64/kernel/head.S index 105c7fec8c6..8d3a9291b47 100644 --- a/arch/ia64/kernel/head.S +++ b/arch/ia64/kernel/head.S @@ -15,6 +15,8 @@ * Copyright (C) 1999 Don Dugger <Don.Dugger@intel.com> * Copyright (C) 2002 Fenghua Yu <fenghua.yu@intel.com> * -Optimize __ia64_save_fpu() and __ia64_load_fpu() for Itanium 2. + * Copyright (C) 2004 Ashok Raj <ashok.raj@intel.com> + * Support for CPU Hotplug */ #include <linux/config.h> @@ -29,6 +31,139 @@ #include <asm/processor.h> #include <asm/ptrace.h> #include <asm/system.h> +#include <asm/mca_asm.h> + +#ifdef CONFIG_HOTPLUG_CPU +#define SAL_PSR_BITS_TO_SET \ + (IA64_PSR_AC | IA64_PSR_BN | IA64_PSR_MFH | IA64_PSR_MFL) + +#define SAVE_FROM_REG(src, ptr, dest) \ + mov dest=src;; \ + st8 [ptr]=dest,0x08 + +#define RESTORE_REG(reg, ptr, _tmp) \ + ld8 _tmp=[ptr],0x08;; \ + mov reg=_tmp + +#define SAVE_BREAK_REGS(ptr, _idx, _breg, _dest)\ + mov ar.lc=IA64_NUM_DBG_REGS-1;; \ + mov _idx=0;; \ +1: \ + SAVE_FROM_REG(_breg[_idx], ptr, _dest);; \ + add _idx=1,_idx;; \ + br.cloop.sptk.many 1b + +#define RESTORE_BREAK_REGS(ptr, _idx, _breg, _tmp, _lbl)\ + mov ar.lc=IA64_NUM_DBG_REGS-1;; \ + mov _idx=0;; \ +_lbl: RESTORE_REG(_breg[_idx], ptr, _tmp);; \ + add _idx=1, _idx;; \ + br.cloop.sptk.many _lbl + +#define SAVE_ONE_RR(num, _reg, _tmp) \ + movl _tmp=(num<<61);; \ + mov _reg=rr[_tmp] + +#define SAVE_REGION_REGS(_tmp, _r0, _r1, _r2, _r3, _r4, _r5, _r6, _r7) \ + SAVE_ONE_RR(0,_r0, _tmp);; \ + SAVE_ONE_RR(1,_r1, _tmp);; \ + SAVE_ONE_RR(2,_r2, _tmp);; \ + SAVE_ONE_RR(3,_r3, _tmp);; \ + SAVE_ONE_RR(4,_r4, _tmp);; \ + SAVE_ONE_RR(5,_r5, _tmp);; \ + SAVE_ONE_RR(6,_r6, _tmp);; \ + SAVE_ONE_RR(7,_r7, _tmp);; + +#define STORE_REGION_REGS(ptr, _r0, _r1, _r2, _r3, _r4, _r5, _r6, _r7) \ + st8 [ptr]=_r0, 8;; \ + st8 [ptr]=_r1, 8;; \ + st8 [ptr]=_r2, 8;; \ + st8 [ptr]=_r3, 8;; \ + st8 [ptr]=_r4, 8;; \ + st8 [ptr]=_r5, 8;; \ + st8 [ptr]=_r6, 8;; \ + st8 [ptr]=_r7, 8;; + +#define RESTORE_REGION_REGS(ptr, _idx1, _idx2, _tmp) \ + mov ar.lc=0x08-1;; \ + movl _idx1=0x00;; \ +RestRR: \ + dep.z _idx2=_idx1,61,3;; \ + ld8 _tmp=[ptr],8;; \ + mov rr[_idx2]=_tmp;; \ + srlz.d;; \ + add _idx1=1,_idx1;; \ + br.cloop.sptk.few RestRR + +#define SET_AREA_FOR_BOOTING_CPU(reg1, reg2) \ + movl reg1=sal_state_for_booting_cpu;; \ + ld8 reg2=[reg1];; + +/* + * Adjust region registers saved before starting to save + * break regs and rest of the states that need to be preserved. + */ +#define SAL_TO_OS_BOOT_HANDOFF_STATE_SAVE(_reg1,_reg2,_pred) \ + SAVE_FROM_REG(b0,_reg1,_reg2);; \ + SAVE_FROM_REG(b1,_reg1,_reg2);; \ + SAVE_FROM_REG(b2,_reg1,_reg2);; \ + SAVE_FROM_REG(b3,_reg1,_reg2);; \ + SAVE_FROM_REG(b4,_reg1,_reg2);; \ + SAVE_FROM_REG(b5,_reg1,_reg2);; \ + st8 [_reg1]=r1,0x08;; \ + st8 [_reg1]=r12,0x08;; \ + st8 [_reg1]=r13,0x08;; \ + SAVE_FROM_REG(ar.fpsr,_reg1,_reg2);; \ + SAVE_FROM_REG(ar.pfs,_reg1,_reg2);; \ + SAVE_FROM_REG(ar.rnat,_reg1,_reg2);; \ + SAVE_FROM_REG(ar.unat,_reg1,_reg2);; \ + SAVE_FROM_REG(ar.bspstore,_reg1,_reg2);; \ + SAVE_FROM_REG(cr.dcr,_reg1,_reg2);; \ + SAVE_FROM_REG(cr.iva,_reg1,_reg2);; \ + SAVE_FROM_REG(cr.pta,_reg1,_reg2);; \ + SAVE_FROM_REG(cr.itv,_reg1,_reg2);; \ + SAVE_FROM_REG(cr.pmv,_reg1,_reg2);; \ + SAVE_FROM_REG(cr.cmcv,_reg1,_reg2);; \ + SAVE_FROM_REG(cr.lrr0,_reg1,_reg2);; \ + SAVE_FROM_REG(cr.lrr1,_reg1,_reg2);; \ + st8 [_reg1]=r4,0x08;; \ + st8 [_reg1]=r5,0x08;; \ + st8 [_reg1]=r6,0x08;; \ + st8 [_reg1]=r7,0x08;; \ + st8 [_reg1]=_pred,0x08;; \ + SAVE_FROM_REG(ar.lc, _reg1, _reg2);; \ + stf.spill.nta [_reg1]=f2,16;; \ + stf.spill.nta [_reg1]=f3,16;; \ + stf.spill.nta [_reg1]=f4,16;; \ + stf.spill.nta [_reg1]=f5,16;; \ + stf.spill.nta [_reg1]=f16,16;; \ + stf.spill.nta [_reg1]=f17,16;; \ + stf.spill.nta [_reg1]=f18,16;; \ + stf.spill.nta [_reg1]=f19,16;; \ + stf.spill.nta [_reg1]=f20,16;; \ + stf.spill.nta [_reg1]=f21,16;; \ + stf.spill.nta [_reg1]=f22,16;; \ + stf.spill.nta [_reg1]=f23,16;; \ + stf.spill.nta [_reg1]=f24,16;; \ + stf.spill.nta [_reg1]=f25,16;; \ + stf.spill.nta [_reg1]=f26,16;; \ + stf.spill.nta [_reg1]=f27,16;; \ + stf.spill.nta [_reg1]=f28,16;; \ + stf.spill.nta [_reg1]=f29,16;; \ + stf.spill.nta [_reg1]=f30,16;; \ + stf.spill.nta [_reg1]=f31,16;; + +#else +#define SET_AREA_FOR_BOOTING_CPU(a1, a2) +#define SAL_TO_OS_BOOT_HANDOFF_STATE_SAVE(a1,a2, a3) +#define SAVE_REGION_REGS(_tmp, _r0, _r1, _r2, _r3, _r4, _r5, _r6, _r7) +#define STORE_REGION_REGS(ptr, _r0, _r1, _r2, _r3, _r4, _r5, _r6, _r7) +#endif + +#define SET_ONE_RR(num, pgsize, _tmp1, _tmp2, vhpt) \ + movl _tmp1=(num << 61);; \ + mov _tmp2=((ia64_rid(IA64_REGION_ID_KERNEL, (num<<61)) << 8) | (pgsize << 2) | vhpt);; \ + mov rr[_tmp1]=_tmp2 .section __special_page_section,"ax" @@ -64,6 +199,12 @@ start_ap: srlz.i ;; /* + * Save the region registers, predicate before they get clobbered + */ + SAVE_REGION_REGS(r2, r8,r9,r10,r11,r12,r13,r14,r15); + mov r25=pr;; + + /* * Initialize kernel region registers: * rr[0]: VHPT enabled, page size = PAGE_SHIFT * rr[1]: VHPT enabled, page size = PAGE_SHIFT @@ -76,32 +217,14 @@ start_ap: * We initialize all of them to prevent inadvertently assuming * something about the state of address translation early in boot. */ - mov r6=((ia64_rid(IA64_REGION_ID_KERNEL, (0<<61)) << 8) | (PAGE_SHIFT << 2) | 1) - movl r7=(0<<61) - mov r8=((ia64_rid(IA64_REGION_ID_KERNEL, (1<<61)) << 8) | (PAGE_SHIFT << 2) | 1) - movl r9=(1<<61) - mov r10=((ia64_rid(IA64_REGION_ID_KERNEL, (2<<61)) << 8) | (PAGE_SHIFT << 2) | 1) - movl r11=(2<<61) - mov r12=((ia64_rid(IA64_REGION_ID_KERNEL, (3<<61)) << 8) | (PAGE_SHIFT << 2) | 1) - movl r13=(3<<61) - mov r14=((ia64_rid(IA64_REGION_ID_KERNEL, (4<<61)) << 8) | (PAGE_SHIFT << 2) | 1) - movl r15=(4<<61) - mov r16=((ia64_rid(IA64_REGION_ID_KERNEL, (5<<61)) << 8) | (PAGE_SHIFT << 2) | 1) - movl r17=(5<<61) - mov r18=((ia64_rid(IA64_REGION_ID_KERNEL, (6<<61)) << 8) | (IA64_GRANULE_SHIFT << 2)) - movl r19=(6<<61) - mov r20=((ia64_rid(IA64_REGION_ID_KERNEL, (7<<61)) << 8) | (IA64_GRANULE_SHIFT << 2)) - movl r21=(7<<61) - ;; - mov rr[r7]=r6 - mov rr[r9]=r8 - mov rr[r11]=r10 - mov rr[r13]=r12 - mov rr[r15]=r14 - mov rr[r17]=r16 - mov rr[r19]=r18 - mov rr[r21]=r20 - ;; + SET_ONE_RR(0, PAGE_SHIFT, r2, r16, 1);; + SET_ONE_RR(1, PAGE_SHIFT, r2, r16, 1);; + SET_ONE_RR(2, PAGE_SHIFT, r2, r16, 1);; + SET_ONE_RR(3, PAGE_SHIFT, r2, r16, 1);; + SET_ONE_RR(4, PAGE_SHIFT, r2, r16, 1);; + SET_ONE_RR(5, PAGE_SHIFT, r2, r16, 1);; + SET_ONE_RR(6, IA64_GRANULE_SHIFT, r2, r16, 0);; + SET_ONE_RR(7, IA64_GRANULE_SHIFT, r2, r16, 0);; /* * Now pin mappings into the TLB for kernel text and data */ @@ -142,6 +265,12 @@ start_ap: ;; 1: // now we are in virtual mode + SET_AREA_FOR_BOOTING_CPU(r2, r16); + + STORE_REGION_REGS(r16, r8,r9,r10,r11,r12,r13,r14,r15); + SAL_TO_OS_BOOT_HANDOFF_STATE_SAVE(r16,r17,r25) + ;; + // set IVT entry point---can't access I/O ports without it movl r3=ia64_ivt ;; @@ -211,12 +340,13 @@ start_ap: mov IA64_KR(CURRENT_STACK)=r16 mov r13=r2 /* - * Reserve space at the top of the stack for "struct pt_regs". Kernel threads - * don't store interesting values in that structure, but the space still needs - * to be there because time-critical stuff such as the context switching can - * be implemented more efficiently (for example, __switch_to() + * Reserve space at the top of the stack for "struct pt_regs". Kernel + * threads don't store interesting values in that structure, but the space + * still needs to be there because time-critical stuff such as the context + * switching can be implemented more efficiently (for example, __switch_to() * always sets the psr.dfh bit of the task it is switching to). */ + addl r12=IA64_STK_OFFSET-IA64_PT_REGS_SIZE-16,r2 addl r2=IA64_RBS_OFFSET,r2 // initialize the RSE mov ar.rsc=0 // place RSE in enforced lazy mode @@ -993,4 +1123,98 @@ END(ia64_spinlock_contention) #endif +#ifdef CONFIG_HOTPLUG_CPU +GLOBAL_ENTRY(ia64_jump_to_sal) + alloc r16=ar.pfs,1,0,0,0;; + rsm psr.i | psr.ic +{ + flushrs + srlz.i +} + tpa r25=in0 + movl r18=tlb_purge_done;; + DATA_VA_TO_PA(r18);; + mov b1=r18 // Return location + movl r18=ia64_do_tlb_purge;; + DATA_VA_TO_PA(r18);; + mov b2=r18 // doing tlb_flush work + mov ar.rsc=0 // Put RSE in enforced lazy, LE mode + movl r17=1f;; + DATA_VA_TO_PA(r17);; + mov cr.iip=r17 + movl r16=SAL_PSR_BITS_TO_SET;; + mov cr.ipsr=r16 + mov cr.ifs=r0;; + rfi;; +1: + /* + * Invalidate all TLB data/inst + */ + br.sptk.many b2;; // jump to tlb purge code + +tlb_purge_done: + RESTORE_REGION_REGS(r25, r17,r18,r19);; + RESTORE_REG(b0, r25, r17);; + RESTORE_REG(b1, r25, r17);; + RESTORE_REG(b2, r25, r17);; + RESTORE_REG(b3, r25, r17);; + RESTORE_REG(b4, r25, r17);; + RESTORE_REG(b5, r25, r17);; + ld8 r1=[r25],0x08;; + ld8 r12=[r25],0x08;; + ld8 r13=[r25],0x08;; + RESTORE_REG(ar.fpsr, r25, r17);; + RESTORE_REG(ar.pfs, r25, r17);; + RESTORE_REG(ar.rnat, r25, r17);; + RESTORE_REG(ar.unat, r25, r17);; + RESTORE_REG(ar.bspstore, r25, r17);; + RESTORE_REG(cr.dcr, r25, r17);; + RESTORE_REG(cr.iva, r25, r17);; + RESTORE_REG(cr.pta, r25, r17);; + RESTORE_REG(cr.itv, r25, r17);; + RESTORE_REG(cr.pmv, r25, r17);; + RESTORE_REG(cr.cmcv, r25, r17);; + RESTORE_REG(cr.lrr0, r25, r17);; + RESTORE_REG(cr.lrr1, r25, r17);; + ld8 r4=[r25],0x08;; + ld8 r5=[r25],0x08;; + ld8 r6=[r25],0x08;; + ld8 r7=[r25],0x08;; + ld8 r17=[r25],0x08;; + mov pr=r17,-1;; + RESTORE_REG(ar.lc, r25, r17);; + /* + * Now Restore floating point regs + */ + ldf.fill.nta f2=[r25],16;; + ldf.fill.nta f3=[r25],16;; + ldf.fill.nta f4=[r25],16;; + ldf.fill.nta f5=[r25],16;; + ldf.fill.nta f16=[r25],16;; + ldf.fill.nta f17=[r25],16;; + ldf.fill.nta f18=[r25],16;; + ldf.fill.nta f19=[r25],16;; + ldf.fill.nta f20=[r25],16;; + ldf.fill.nta f21=[r25],16;; + ldf.fill.nta f22=[r25],16;; + ldf.fill.nta f23=[r25],16;; + ldf.fill.nta f24=[r25],16;; + ldf.fill.nta f25=[r25],16;; + ldf.fill.nta f26=[r25],16;; + ldf.fill.nta f27=[r25],16;; + ldf.fill.nta f28=[r25],16;; + ldf.fill.nta f29=[r25],16;; + ldf.fill.nta f30=[r25],16;; + ldf.fill.nta f31=[r25],16;; + + /* + * Now that we have done all the register restores + * we are now ready for the big DIVE to SAL Land + */ + ssm psr.ic;; + srlz.d;; + br.ret.sptk.many b0;; +END(ia64_jump_to_sal) +#endif /* CONFIG_HOTPLUG_CPU */ + #endif /* CONFIG_SMP */ diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index c15be5c38f5..88b014381df 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -79,6 +79,7 @@ #include <linux/smp.h> #include <linux/smp_lock.h> #include <linux/string.h> +#include <linux/bootmem.h> #include <asm/delay.h> #include <asm/hw_irq.h> @@ -98,19 +99,30 @@ #define DBG(fmt...) #endif +#define NR_PREALLOCATE_RTE_ENTRIES (PAGE_SIZE / sizeof(struct iosapic_rte_info)) +#define RTE_PREALLOCATED (1) + static DEFINE_SPINLOCK(iosapic_lock); /* These tables map IA-64 vectors to the IOSAPIC pin that generates this vector. */ -static struct iosapic_intr_info { +struct iosapic_rte_info { + struct list_head rte_list; /* node in list of RTEs sharing the same vector */ char __iomem *addr; /* base address of IOSAPIC */ - u32 low32; /* current value of low word of Redirection table entry */ unsigned int gsi_base; /* first GSI assigned to this IOSAPIC */ - char rte_index; /* IOSAPIC RTE index (-1 => not an IOSAPIC interrupt) */ + char rte_index; /* IOSAPIC RTE index */ + int refcnt; /* reference counter */ + unsigned int flags; /* flags */ +} ____cacheline_aligned; + +static struct iosapic_intr_info { + struct list_head rtes; /* RTEs using this vector (empty => not an IOSAPIC interrupt) */ + int count; /* # of RTEs that shares this vector */ + u32 low32; /* current value of low word of Redirection table entry */ + unsigned int dest; /* destination CPU physical ID */ unsigned char dmode : 3; /* delivery mode (see iosapic.h) */ unsigned char polarity: 1; /* interrupt polarity (see iosapic.h) */ unsigned char trigger : 1; /* trigger mode (see iosapic.h) */ - int refcnt; /* reference counter */ } iosapic_intr_info[IA64_NUM_VECTORS]; static struct iosapic { @@ -126,6 +138,8 @@ static int num_iosapic; static unsigned char pcat_compat __initdata; /* 8259 compatibility flag */ +static int iosapic_kmalloc_ok; +static LIST_HEAD(free_rte_list); /* * Find an IOSAPIC associated with a GSI @@ -147,10 +161,12 @@ static inline int _gsi_to_vector (unsigned int gsi) { struct iosapic_intr_info *info; + struct iosapic_rte_info *rte; for (info = iosapic_intr_info; info < iosapic_intr_info + IA64_NUM_VECTORS; ++info) - if (info->gsi_base + info->rte_index == gsi) - return info - iosapic_intr_info; + list_for_each_entry(rte, &info->rtes, rte_list) + if (rte->gsi_base + rte->rte_index == gsi) + return info - iosapic_intr_info; return -1; } @@ -167,33 +183,52 @@ gsi_to_vector (unsigned int gsi) int gsi_to_irq (unsigned int gsi) { + unsigned long flags; + int irq; /* * XXX fix me: this assumes an identity mapping vetween IA-64 vector and Linux irq * numbers... */ - return _gsi_to_vector(gsi); + spin_lock_irqsave(&iosapic_lock, flags); + { + irq = _gsi_to_vector(gsi); + } + spin_unlock_irqrestore(&iosapic_lock, flags); + + return irq; +} + +static struct iosapic_rte_info *gsi_vector_to_rte(unsigned int gsi, unsigned int vec) +{ + struct iosapic_rte_info *rte; + + list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, rte_list) + if (rte->gsi_base + rte->rte_index == gsi) + return rte; + return NULL; } static void -set_rte (unsigned int vector, unsigned int dest, int mask) +set_rte (unsigned int gsi, unsigned int vector, unsigned int dest, int mask) { unsigned long pol, trigger, dmode; u32 low32, high32; char __iomem *addr; int rte_index; char redir; + struct iosapic_rte_info *rte; DBG(KERN_DEBUG"IOSAPIC: routing vector %d to 0x%x\n", vector, dest); - rte_index = iosapic_intr_info[vector].rte_index; - if (rte_index < 0) + rte = gsi_vector_to_rte(gsi, vector); + if (!rte) return; /* not an IOSAPIC interrupt */ - addr = iosapic_intr_info[vector].addr; + rte_index = rte->rte_index; + addr = rte->addr; pol = iosapic_intr_info[vector].polarity; trigger = iosapic_intr_info[vector].trigger; dmode = iosapic_intr_info[vector].dmode; - vector &= (~IA64_IRQ_REDIRECTED); redir = (dmode == IOSAPIC_LOWEST_PRIORITY) ? 1 : 0; @@ -221,6 +256,7 @@ set_rte (unsigned int vector, unsigned int dest, int mask) iosapic_write(addr, IOSAPIC_RTE_HIGH(rte_index), high32); iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); iosapic_intr_info[vector].low32 = low32; + iosapic_intr_info[vector].dest = dest; } static void @@ -237,18 +273,20 @@ mask_irq (unsigned int irq) u32 low32; int rte_index; ia64_vector vec = irq_to_vector(irq); + struct iosapic_rte_info *rte; - addr = iosapic_intr_info[vec].addr; - rte_index = iosapic_intr_info[vec].rte_index; - - if (rte_index < 0) + if (list_empty(&iosapic_intr_info[vec].rtes)) return; /* not an IOSAPIC interrupt! */ spin_lock_irqsave(&iosapic_lock, flags); { /* set only the mask bit */ low32 = iosapic_intr_info[vec].low32 |= IOSAPIC_MASK; - iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); + list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, rte_list) { + addr = rte->addr; + rte_index = rte->rte_index; + iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); + } } spin_unlock_irqrestore(&iosapic_lock, flags); } @@ -261,16 +299,19 @@ unmask_irq (unsigned int irq) u32 low32; int rte_index; ia64_vector vec = irq_to_vector(irq); + struct iosapic_rte_info *rte; - addr = iosapic_intr_info[vec].addr; - rte_index = iosapic_intr_info[vec].rte_index; - if (rte_index < 0) + if (list_empty(&iosapic_intr_info[vec].rtes)) return; /* not an IOSAPIC interrupt! */ spin_lock_irqsave(&iosapic_lock, flags); { low32 = iosapic_intr_info[vec].low32 &= ~IOSAPIC_MASK; - iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); + list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, rte_list) { + addr = rte->addr; + rte_index = rte->rte_index; + iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); + } } spin_unlock_irqrestore(&iosapic_lock, flags); } @@ -286,6 +327,7 @@ iosapic_set_affinity (unsigned int irq, cpumask_t mask) char __iomem *addr; int redir = (irq & IA64_IRQ_REDIRECTED) ? 1 : 0; ia64_vector vec; + struct iosapic_rte_info *rte; irq &= (~IA64_IRQ_REDIRECTED); vec = irq_to_vector(irq); @@ -295,10 +337,7 @@ iosapic_set_affinity (unsigned int irq, cpumask_t mask) dest = cpu_physical_id(first_cpu(mask)); - rte_index = iosapic_intr_info[vec].rte_index; - addr = iosapic_intr_info[vec].addr; - - if (rte_index < 0) + if (list_empty(&iosapic_intr_info[vec].rtes)) return; /* not an IOSAPIC interrupt */ set_irq_affinity_info(irq, dest, redir); @@ -318,8 +357,13 @@ iosapic_set_affinity (unsigned int irq, cpumask_t mask) low32 |= (IOSAPIC_FIXED << IOSAPIC_DELIVERY_SHIFT); iosapic_intr_info[vec].low32 = low32; - iosapic_write(addr, IOSAPIC_RTE_HIGH(rte_index), high32); - iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); + iosapic_intr_info[vec].dest = dest; + list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, rte_list) { + addr = rte->addr; + rte_index = rte->rte_index; + iosapic_write(addr, IOSAPIC_RTE_HIGH(rte_index), high32); + iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); + } } spin_unlock_irqrestore(&iosapic_lock, flags); #endif @@ -340,9 +384,11 @@ static void iosapic_end_level_irq (unsigned int irq) { ia64_vector vec = irq_to_vector(irq); + struct iosapic_rte_info *rte; move_irq(irq); - iosapic_eoi(iosapic_intr_info[vec].addr, vec); + list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, rte_list) + iosapic_eoi(rte->addr, vec); } #define iosapic_shutdown_level_irq mask_irq @@ -422,6 +468,34 @@ iosapic_version (char __iomem *addr) return iosapic_read(addr, IOSAPIC_VERSION); } +static int iosapic_find_sharable_vector (unsigned long trigger, unsigned long pol) +{ + int i, vector = -1, min_count = -1; + struct iosapic_intr_info *info; + + /* + * shared vectors for edge-triggered interrupts are not + * supported yet + */ + if (trigger == IOSAPIC_EDGE) + return -1; + + for (i = IA64_FIRST_DEVICE_VECTOR; i <= IA64_LAST_DEVICE_VECTOR; i++) { + info = &iosapic_intr_info[i]; + if (info->trigger == trigger && info->polarity == pol && + (info->dmode == IOSAPIC_FIXED || info->dmode == IOSAPIC_LOWEST_PRIORITY)) { + if (min_count == -1 || info->count < min_count) { + vector = i; + min_count = info->count; + } + } + } + if (vector < 0) + panic("%s: out of interrupt vectors!\n", __FUNCTION__); + + return vector; +} + /* * if the given vector is already owned by other, * assign a new vector for the other and make the vector available @@ -431,19 +505,63 @@ iosapic_reassign_vector (int vector) { int new_vector; - if (iosapic_intr_info[vector].rte_index >= 0 || iosapic_intr_info[vector].addr - || iosapic_intr_info[vector].gsi_base || iosapic_intr_info[vector].dmode - || iosapic_intr_info[vector].polarity || iosapic_intr_info[vector].trigger) - { + if (!list_empty(&iosapic_intr_info[vector].rtes)) { new_vector = assign_irq_vector(AUTO_ASSIGN); printk(KERN_INFO "Reassigning vector %d to %d\n", vector, new_vector); memcpy(&iosapic_intr_info[new_vector], &iosapic_intr_info[vector], sizeof(struct iosapic_intr_info)); + INIT_LIST_HEAD(&iosapic_intr_info[new_vector].rtes); + list_move(iosapic_intr_info[vector].rtes.next, &iosapic_intr_info[new_vector].rtes); memset(&iosapic_intr_info[vector], 0, sizeof(struct iosapic_intr_info)); - iosapic_intr_info[vector].rte_index = -1; + iosapic_intr_info[vector].low32 = IOSAPIC_MASK; + INIT_LIST_HEAD(&iosapic_intr_info[vector].rtes); } } +static struct iosapic_rte_info *iosapic_alloc_rte (void) +{ + int i; + struct iosapic_rte_info *rte; + int preallocated = 0; + + if (!iosapic_kmalloc_ok && list_empty(&free_rte_list)) { + rte = alloc_bootmem(sizeof(struct iosapic_rte_info) * NR_PREALLOCATE_RTE_ENTRIES); + if (!rte) + return NULL; + for (i = 0; i < NR_PREALLOCATE_RTE_ENTRIES; i++, rte++) + list_add(&rte->rte_list, &free_rte_list); + } + + if (!list_empty(&free_rte_list)) { + rte = list_entry(free_rte_list.next, struct iosapic_rte_info, rte_list); + list_del(&rte->rte_list); + preallocated++; + } else { + rte = kmalloc(sizeof(struct iosapic_rte_info), GFP_ATOMIC); + if (!rte) + return NULL; + } + + memset(rte, 0, sizeof(struct iosapic_rte_info)); + if (preallocated) + rte->flags |= RTE_PREALLOCATED; + + return rte; +} + +static void iosapic_free_rte (struct iosapic_rte_info *rte) +{ + if (rte->flags & RTE_PREALLOCATED) + list_add_tail(&rte->rte_list, &free_rte_list); + else + kfree(rte); +} + +static inline int vector_is_shared (int vector) +{ + return (iosapic_intr_info[vector].count > 1); +} + static void register_intr (unsigned int gsi, int vector, unsigned char delivery, unsigned long polarity, unsigned long trigger) @@ -454,6 +572,7 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery, int index; unsigned long gsi_base; void __iomem *iosapic_address; + struct iosapic_rte_info *rte; index = find_iosapic(gsi); if (index < 0) { @@ -464,14 +583,33 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery, iosapic_address = iosapic_lists[index].addr; gsi_base = iosapic_lists[index].gsi_base; - rte_index = gsi - gsi_base; - iosapic_intr_info[vector].rte_index = rte_index; + rte = gsi_vector_to_rte(gsi, vector); + if (!rte) { + rte = iosapic_alloc_rte(); + if (!rte) { + printk(KERN_WARNING "%s: cannot allocate memory\n", __FUNCTION__); + return; + } + + rte_index = gsi - gsi_base; + rte->rte_index = rte_index; + rte->addr = iosapic_address; + rte->gsi_base = gsi_base; + rte->refcnt++; + list_add_tail(&rte->rte_list, &iosapic_intr_info[vector].rtes); + iosapic_intr_info[vector].count++; + } + else if (vector_is_shared(vector)) { + struct iosapic_intr_info *info = &iosapic_intr_info[vector]; + if (info->trigger != trigger || info->polarity != polarity) { + printk (KERN_WARNING "%s: cannot override the interrupt\n", __FUNCTION__); + return; + } + } + iosapic_intr_info[vector].polarity = polarity; iosapic_intr_info[vector].dmode = delivery; - iosapic_intr_info[vector].addr = iosapic_address; - iosapic_intr_info[vector].gsi_base = gsi_base; iosapic_intr_info[vector].trigger = trigger; - iosapic_intr_info[vector].refcnt++; if (trigger == IOSAPIC_EDGE) irq_type = &irq_type_iosapic_edge; @@ -494,6 +632,13 @@ get_target_cpu (unsigned int gsi, int vector) static int cpu = -1; /* + * In case of vector shared by multiple RTEs, all RTEs that + * share the vector need to use the same destination CPU. + */ + if (!list_empty(&iosapic_intr_info[vector].rtes)) + return iosapic_intr_info[vector].dest; + + /* * If the platform supports redirection via XTP, let it * distribute interrupts. */ @@ -565,10 +710,12 @@ int iosapic_register_intr (unsigned int gsi, unsigned long polarity, unsigned long trigger) { - int vector; + int vector, mask = 1; unsigned int dest; unsigned long flags; - + struct iosapic_rte_info *rte; + u32 low32; +again: /* * If this GSI has already been registered (i.e., it's a * shared interrupt, or we lost a race to register it), @@ -578,19 +725,45 @@ iosapic_register_intr (unsigned int gsi, { vector = gsi_to_vector(gsi); if (vector > 0) { - iosapic_intr_info[vector].refcnt++; + rte = gsi_vector_to_rte(gsi, vector); + rte->refcnt++; spin_unlock_irqrestore(&iosapic_lock, flags); return vector; } + } + spin_unlock_irqrestore(&iosapic_lock, flags); + + /* If vector is running out, we try to find a sharable vector */ + vector = assign_irq_vector_nopanic(AUTO_ASSIGN); + if (vector < 0) + vector = iosapic_find_sharable_vector(trigger, polarity); + + spin_lock_irqsave(&irq_descp(vector)->lock, flags); + spin_lock(&iosapic_lock); + { + if (gsi_to_vector(gsi) > 0) { + if (list_empty(&iosapic_intr_info[vector].rtes)) + free_irq_vector(vector); + spin_unlock(&iosapic_lock); + spin_unlock_irqrestore(&irq_descp(vector)->lock, flags); + goto again; + } - vector = assign_irq_vector(AUTO_ASSIGN); dest = get_target_cpu(gsi, vector); register_intr(gsi, vector, IOSAPIC_LOWEST_PRIORITY, - polarity, trigger); + polarity, trigger); - set_rte(vector, dest, 1); + /* + * If the vector is shared and already unmasked for + * other interrupt sources, don't mask it. + */ + low32 = iosapic_intr_info[vector].low32; + if (vector_is_shared(vector) && !(low32 & IOSAPIC_MASK)) + mask = 0; + set_rte(gsi, vector, dest, mask); } - spin_unlock_irqrestore(&iosapic_lock, flags); + spin_unlock(&iosapic_lock); + spin_unlock_irqrestore(&irq_descp(vector)->lock, flags); printk(KERN_INFO "GSI %u (%s, %s) -> CPU %d (0x%04x) vector %d\n", gsi, (trigger == IOSAPIC_EDGE ? "edge" : "level"), @@ -607,8 +780,10 @@ iosapic_unregister_intr (unsigned int gsi) unsigned long flags; int irq, vector; irq_desc_t *idesc; - int rte_index; + u32 low32; unsigned long trigger, polarity; + unsigned int dest; + struct iosapic_rte_info *rte; /* * If the irq associated with the gsi is not found, @@ -627,54 +802,56 @@ iosapic_unregister_intr (unsigned int gsi) spin_lock_irqsave(&idesc->lock, flags); spin_lock(&iosapic_lock); { - rte_index = iosapic_intr_info[vector].rte_index; - if (rte_index < 0) { - spin_unlock(&iosapic_lock); - spin_unlock_irqrestore(&idesc->lock, flags); + if ((rte = gsi_vector_to_rte(gsi, vector)) == NULL) { printk(KERN_ERR "iosapic_unregister_intr(%u) unbalanced\n", gsi); WARN_ON(1); - return; + goto out; } - if (--iosapic_intr_info[vector].refcnt > 0) { - spin_unlock(&iosapic_lock); - spin_unlock_irqrestore(&idesc->lock, flags); - return; - } + if (--rte->refcnt > 0) + goto out; - /* - * If interrupt handlers still exist on the irq - * associated with the gsi, don't unregister the - * interrupt. - */ - if (idesc->action) { - iosapic_intr_info[vector].refcnt++; - spin_unlock(&iosapic_lock); - spin_unlock_irqrestore(&idesc->lock, flags); - printk(KERN_WARNING "Cannot unregister GSI. IRQ %u is still in use.\n", irq); - return; - } + /* Mask the interrupt */ + low32 = iosapic_intr_info[vector].low32 | IOSAPIC_MASK; + iosapic_write(rte->addr, IOSAPIC_RTE_LOW(rte->rte_index), low32); - /* Clear the interrupt controller descriptor. */ - idesc->handler = &no_irq_type; + /* Remove the rte entry from the list */ + list_del(&rte->rte_list); + iosapic_intr_info[vector].count--; + iosapic_free_rte(rte); - trigger = iosapic_intr_info[vector].trigger; + trigger = iosapic_intr_info[vector].trigger; polarity = iosapic_intr_info[vector].polarity; + dest = iosapic_intr_info[vector].dest; + printk(KERN_INFO "GSI %u (%s, %s) -> CPU %d (0x%04x) vector %d unregistered\n", + gsi, (trigger == IOSAPIC_EDGE ? "edge" : "level"), + (polarity == IOSAPIC_POL_HIGH ? "high" : "low"), + cpu_logical_id(dest), dest, vector); + + if (list_empty(&iosapic_intr_info[vector].rtes)) { + /* Sanity check */ + BUG_ON(iosapic_intr_info[vector].count); + + /* Clear the interrupt controller descriptor */ + idesc->handler = &no_irq_type; + + /* Clear the interrupt information */ + memset(&iosapic_intr_info[vector], 0, sizeof(struct iosapic_intr_info)); + iosapic_intr_info[vector].low32 |= IOSAPIC_MASK; + INIT_LIST_HEAD(&iosapic_intr_info[vector].rtes); + + if (idesc->action) { + printk(KERN_ERR "interrupt handlers still exist on IRQ %u\n", irq); + WARN_ON(1); + } - /* Clear the interrupt information. */ - memset(&iosapic_intr_info[vector], 0, sizeof(struct iosapic_intr_info)); - iosapic_intr_info[vector].rte_index = -1; /* mark as unused */ + /* Free the interrupt vector */ + free_irq_vector(vector); + } } + out: spin_unlock(&iosapic_lock); spin_unlock_irqrestore(&idesc->lock, flags); - - /* Free the interrupt vector */ - free_irq_vector(vector); - - printk(KERN_INFO "GSI %u (%s, %s) -> vector %d unregisterd.\n", - gsi, (trigger == IOSAPIC_EDGE ? "edge" : "level"), - (polarity == IOSAPIC_POL_HIGH ? "high" : "low"), - vector); } #endif /* CONFIG_ACPI_DEALLOCATE_IRQ */ @@ -724,7 +901,7 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, (polarity == IOSAPIC_POL_HIGH ? "high" : "low"), cpu_logical_id(dest), dest, vector); - set_rte(vector, dest, mask); + set_rte(gsi, vector, dest, mask); return vector; } @@ -750,7 +927,7 @@ iosapic_override_isa_irq (unsigned int isa_irq, unsigned int gsi, polarity == IOSAPIC_POL_HIGH ? "high" : "low", cpu_logical_id(dest), dest, vector); - set_rte(vector, dest, 1); + set_rte(gsi, vector, dest, 1); } void __init @@ -758,8 +935,10 @@ iosapic_system_init (int system_pcat_compat) { int vector; - for (vector = 0; vector < IA64_NUM_VECTORS; ++vector) - iosapic_intr_info[vector].rte_index = -1; /* mark as unused */ + for (vector = 0; vector < IA64_NUM_VECTORS; ++vector) { + iosapic_intr_info[vector].low32 = IOSAPIC_MASK; + INIT_LIST_HEAD(&iosapic_intr_info[vector].rtes); /* mark as unused */ + } pcat_compat = system_pcat_compat; if (pcat_compat) { @@ -825,3 +1004,10 @@ map_iosapic_to_node(unsigned int gsi_base, int node) return; } #endif + +static int __init iosapic_enable_kmalloc (void) +{ + iosapic_kmalloc_ok = 1; + return 0; +} +core_initcall (iosapic_enable_kmalloc); diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index 5ba06ebe355..4fe60c7a2e9 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -63,20 +63,30 @@ EXPORT_SYMBOL(isa_irq_to_vector_map); static unsigned long ia64_vector_mask[BITS_TO_LONGS(IA64_NUM_DEVICE_VECTORS)]; int -assign_irq_vector (int irq) +assign_irq_vector_nopanic (int irq) { int pos, vector; again: pos = find_first_zero_bit(ia64_vector_mask, IA64_NUM_DEVICE_VECTORS); vector = IA64_FIRST_DEVICE_VECTOR + pos; if (vector > IA64_LAST_DEVICE_VECTOR) - /* XXX could look for sharable vectors instead of panic'ing... */ - panic("assign_irq_vector: out of interrupt vectors!"); + return -1; if (test_and_set_bit(pos, ia64_vector_mask)) goto again; return vector; } +int +assign_irq_vector (int irq) +{ + int vector = assign_irq_vector_nopanic(irq); + + if (vector < 0) + panic("assign_irq_vector: out of interrupt vectors!"); + + return vector; +} + void free_irq_vector (int vector) { diff --git a/arch/ia64/kernel/mca_asm.S b/arch/ia64/kernel/mca_asm.S index cf3f8014f9a..ef3fd7265b6 100644 --- a/arch/ia64/kernel/mca_asm.S +++ b/arch/ia64/kernel/mca_asm.S @@ -110,46 +110,19 @@ .global ia64_os_mca_dispatch_end .global ia64_sal_to_os_handoff_state .global ia64_os_to_sal_handoff_state + .global ia64_do_tlb_purge .text .align 16 -ia64_os_mca_dispatch: - - // Serialize all MCA processing - mov r3=1;; - LOAD_PHYSICAL(p0,r2,ia64_mca_serialize);; -ia64_os_mca_spin: - xchg8 r4=[r2],r3;; - cmp.ne p6,p0=r4,r0 -(p6) br ia64_os_mca_spin - - // Save the SAL to OS MCA handoff state as defined - // by SAL SPEC 3.0 - // NOTE : The order in which the state gets saved - // is dependent on the way the C-structure - // for ia64_mca_sal_to_os_state_t has been - // defined in include/asm/mca.h - SAL_TO_OS_MCA_HANDOFF_STATE_SAVE(r2) - ;; - - // LOG PROCESSOR STATE INFO FROM HERE ON.. -begin_os_mca_dump: - br ia64_os_mca_proc_state_dump;; - -ia64_os_mca_done_dump: - - LOAD_PHYSICAL(p0,r16,ia64_sal_to_os_handoff_state+56) - ;; - ld8 r18=[r16] // Get processor state parameter on existing PALE_CHECK. - ;; - tbit.nz p6,p7=r18,60 -(p7) br.spnt done_tlb_purge_and_reload - - // The following code purges TC and TR entries. Then reload all TC entries. - // Purge percpu data TC entries. -begin_tlb_purge_and_reload: +/* + * Just the TLB purge part is moved to a separate function + * so we can re-use the code for cpu hotplug code as well + * Caller should now setup b1, so we can branch once the + * tlb flush is complete. + */ +ia64_do_tlb_purge: #define O(member) IA64_CPUINFO_##member##_OFFSET GET_THIS_PADDR(r2, cpu_info) // load phys addr of cpu_info into r2 @@ -230,6 +203,51 @@ begin_tlb_purge_and_reload: ;; srlz.i ;; + // Now branch away to caller. + br.sptk.many b1 + ;; + +ia64_os_mca_dispatch: + + // Serialize all MCA processing + mov r3=1;; + LOAD_PHYSICAL(p0,r2,ia64_mca_serialize);; +ia64_os_mca_spin: + xchg8 r4=[r2],r3;; + cmp.ne p6,p0=r4,r0 +(p6) br ia64_os_mca_spin + + // Save the SAL to OS MCA handoff state as defined + // by SAL SPEC 3.0 + // NOTE : The order in which the state gets saved + // is dependent on the way the C-structure + // for ia64_mca_sal_to_os_state_t has been + // defined in include/asm/mca.h + SAL_TO_OS_MCA_HANDOFF_STATE_SAVE(r2) + ;; + + // LOG PROCESSOR STATE INFO FROM HERE ON.. +begin_os_mca_dump: + br ia64_os_mca_proc_state_dump;; + +ia64_os_mca_done_dump: + + LOAD_PHYSICAL(p0,r16,ia64_sal_to_os_handoff_state+56) + ;; + ld8 r18=[r16] // Get processor state parameter on existing PALE_CHECK. + ;; + tbit.nz p6,p7=r18,60 +(p7) br.spnt done_tlb_purge_and_reload + + // The following code purges TC and TR entries. Then reload all TC entries. + // Purge percpu data TC entries. +begin_tlb_purge_and_reload: + movl r18=ia64_reload_tr;; + LOAD_PHYSICAL(p0,r18,ia64_reload_tr);; + mov b1=r18;; + br.sptk.many ia64_do_tlb_purge;; + +ia64_reload_tr: // Finally reload the TR registers. // 1. Reload DTR/ITR registers for kernel. mov r18=KERNEL_TR_PAGE_SHIFT<<2 diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c index ab478172c34..abc0113a821 100644 --- a/arch/ia64/kernel/mca_drv.c +++ b/arch/ia64/kernel/mca_drv.c @@ -132,8 +132,7 @@ mca_handler_bh(unsigned long paddr) spin_unlock(&mca_bh_lock); /* This process is about to be killed itself */ - force_sig(SIGKILL, current); - schedule(); + do_exit(SIGKILL); } /** @@ -439,6 +438,7 @@ recover_from_read_error(slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_chec psr2 = (struct ia64_psr *)&pmsa->pmsa_ipsr; psr2->cpl = 0; psr2->ri = 0; + psr2->i = 0; return 1; } diff --git a/arch/ia64/kernel/mca_drv_asm.S b/arch/ia64/kernel/mca_drv_asm.S index bcfa05acc56..2d7e0217638 100644 --- a/arch/ia64/kernel/mca_drv_asm.S +++ b/arch/ia64/kernel/mca_drv_asm.S @@ -10,6 +10,7 @@ #include <asm/asmmacro.h> #include <asm/processor.h> +#include <asm/ptrace.h> GLOBAL_ENTRY(mca_handler_bhhook) invala // clear RSE ? @@ -20,12 +21,21 @@ GLOBAL_ENTRY(mca_handler_bhhook) ;; alloc r16=ar.pfs,0,2,1,0 // make a new frame ;; + mov ar.rsc=0 + ;; mov r13=IA64_KR(CURRENT) // current task pointer ;; - adds r12=IA64_TASK_THREAD_KSP_OFFSET,r13 + mov r2=r13 + ;; + addl r22=IA64_RBS_OFFSET,r2 + ;; + mov ar.bspstore=r22 ;; - ld8 r12=[r12] // stack pointer + addl sp=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2 ;; + adds r2=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13 + ;; + st1 [r2]=r0 // clear current->thread.on_ustack flag mov loc0=r16 movl loc1=mca_handler_bh // recovery C function ;; @@ -34,7 +44,9 @@ GLOBAL_ENTRY(mca_handler_bhhook) ;; mov loc1=rp ;; - br.call.sptk.many rp=b6 // not return ... + ssm psr.i + ;; + br.call.sptk.many rp=b6 // does not return ... ;; mov ar.pfs=loc0 mov rp=loc1 diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 71147be3279..71c101601e3 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -480,14 +480,6 @@ typedef struct { #define PFM_CMD_ARG_MANY -1 /* cannot be zero */ typedef struct { - int debug; /* turn on/off debugging via syslog */ - int debug_ovfl; /* turn on/off debug printk in overflow handler */ - int fastctxsw; /* turn on/off fast (unsecure) ctxsw */ - int expert_mode; /* turn on/off value checking */ - int debug_pfm_read; -} pfm_sysctl_t; - -typedef struct { unsigned long pfm_spurious_ovfl_intr_count; /* keep track of spurious ovfl interrupts */ unsigned long pfm_replay_ovfl_intr_count; /* keep track of replayed ovfl interrupts */ unsigned long pfm_ovfl_intr_count; /* keep track of ovfl interrupts */ @@ -514,8 +506,8 @@ static LIST_HEAD(pfm_buffer_fmt_list); static pmu_config_t *pmu_conf; /* sysctl() controls */ -static pfm_sysctl_t pfm_sysctl; -int pfm_debug_var; +pfm_sysctl_t pfm_sysctl; +EXPORT_SYMBOL(pfm_sysctl); static ctl_table pfm_ctl_table[]={ {1, "debug", &pfm_sysctl.debug, sizeof(int), 0666, NULL, &proc_dointvec, NULL,}, @@ -1273,6 +1265,8 @@ out: } EXPORT_SYMBOL(pfm_unregister_buffer_fmt); +extern void update_pal_halt_status(int); + static int pfm_reserve_session(struct task_struct *task, int is_syswide, unsigned int cpu) { @@ -1319,6 +1313,11 @@ pfm_reserve_session(struct task_struct *task, int is_syswide, unsigned int cpu) is_syswide, cpu)); + /* + * disable default_idle() to go to PAL_HALT + */ + update_pal_halt_status(0); + UNLOCK_PFS(flags); return 0; @@ -1374,6 +1373,12 @@ pfm_unreserve_session(pfm_context_t *ctx, int is_syswide, unsigned int cpu) is_syswide, cpu)); + /* + * if possible, enable default_idle() to go into PAL_HALT + */ + if (pfm_sessions.pfs_task_sessions == 0 && pfm_sessions.pfs_sys_sessions == 0) + update_pal_halt_status(1); + UNLOCK_PFS(flags); return 0; @@ -1576,7 +1581,7 @@ pfm_read(struct file *filp, char __user *buf, size_t size, loff_t *ppos) goto abort_locked; } - DPRINT(("[%d] fd=%d type=%d\n", current->pid, msg->pfm_gen_msg.msg_ctx_fd, msg->pfm_gen_msg.msg_type)); + DPRINT(("fd=%d type=%d\n", msg->pfm_gen_msg.msg_ctx_fd, msg->pfm_gen_msg.msg_type)); ret = -EFAULT; if(copy_to_user(buf, msg, sizeof(pfm_msg_t)) == 0) ret = sizeof(pfm_msg_t); @@ -3695,8 +3700,6 @@ pfm_debug(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) pfm_sysctl.debug = m == 0 ? 0 : 1; - pfm_debug_var = pfm_sysctl.debug; - printk(KERN_INFO "perfmon debugging %s (timing reset)\n", pfm_sysctl.debug ? "on" : "off"); if (m == 0) { @@ -4212,7 +4215,7 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs) DPRINT(("cannot load to [%d], invalid ctx_state=%d\n", req->load_pid, ctx->ctx_state)); - return -EINVAL; + return -EBUSY; } DPRINT(("load_pid [%d] using_dbreg=%d\n", req->load_pid, ctx->ctx_fl_using_dbreg)); @@ -4714,16 +4717,26 @@ recheck: if (task == current || ctx->ctx_fl_system) return 0; /* - * if context is UNLOADED we are safe to go + * we are monitoring another thread */ - if (state == PFM_CTX_UNLOADED) return 0; - - /* - * no command can operate on a zombie context - */ - if (state == PFM_CTX_ZOMBIE) { - DPRINT(("cmd %d state zombie cannot operate on context\n", cmd)); - return -EINVAL; + switch(state) { + case PFM_CTX_UNLOADED: + /* + * if context is UNLOADED we are safe to go + */ + return 0; + case PFM_CTX_ZOMBIE: + /* + * no command can operate on a zombie context + */ + DPRINT(("cmd %d state zombie cannot operate on context\n", cmd)); + return -EINVAL; + case PFM_CTX_MASKED: + /* + * PMU state has been saved to software even though + * the thread may still be running. + */ + if (cmd != PFM_UNLOAD_CONTEXT) return 0; } /* @@ -4996,13 +5009,21 @@ pfm_context_force_terminate(pfm_context_t *ctx, struct pt_regs *regs) } static int pfm_ovfl_notify_user(pfm_context_t *ctx, unsigned long ovfl_pmds); - + /* + * pfm_handle_work() can be called with interrupts enabled + * (TIF_NEED_RESCHED) or disabled. The down_interruptible + * call may sleep, therefore we must re-enable interrupts + * to avoid deadlocks. It is safe to do so because this function + * is called ONLY when returning to user level (PUStk=1), in which case + * there is no risk of kernel stack overflow due to deep + * interrupt nesting. + */ void pfm_handle_work(void) { pfm_context_t *ctx; struct pt_regs *regs; - unsigned long flags; + unsigned long flags, dummy_flags; unsigned long ovfl_regs; unsigned int reason; int ret; @@ -5039,18 +5060,15 @@ pfm_handle_work(void) //if (CTX_OVFL_NOBLOCK(ctx)) goto skip_blocking; if (reason == PFM_TRAP_REASON_RESET) goto skip_blocking; + /* + * restore interrupt mask to what it was on entry. + * Could be enabled/diasbled. + */ UNPROTECT_CTX(ctx, flags); - /* - * pfm_handle_work() is currently called with interrupts disabled. - * The down_interruptible call may sleep, therefore we - * must re-enable interrupts to avoid deadlocks. It is - * safe to do so because this function is called ONLY - * when returning to user level (PUStk=1), in which case - * there is no risk of kernel stack overflow due to deep - * interrupt nesting. - */ - BUG_ON(flags & IA64_PSR_I); + /* + * force interrupt enable because of down_interruptible() + */ local_irq_enable(); DPRINT(("before block sleeping\n")); @@ -5064,12 +5082,12 @@ pfm_handle_work(void) DPRINT(("after block sleeping ret=%d\n", ret)); /* - * disable interrupts to restore state we had upon entering - * this function + * lock context and mask interrupts again + * We save flags into a dummy because we may have + * altered interrupts mask compared to entry in this + * function. */ - local_irq_disable(); - - PROTECT_CTX(ctx, flags); + PROTECT_CTX(ctx, dummy_flags); /* * we need to read the ovfl_regs only after wake-up @@ -5095,7 +5113,9 @@ skip_blocking: ctx->ctx_ovfl_regs[0] = 0UL; nothing_to_do: - + /* + * restore flags as they were upon entry + */ UNPROTECT_CTX(ctx, flags); } diff --git a/arch/ia64/kernel/perfmon_default_smpl.c b/arch/ia64/kernel/perfmon_default_smpl.c index 965d2900455..344941db0a9 100644 --- a/arch/ia64/kernel/perfmon_default_smpl.c +++ b/arch/ia64/kernel/perfmon_default_smpl.c @@ -20,24 +20,17 @@ MODULE_AUTHOR("Stephane Eranian <eranian@hpl.hp.com>"); MODULE_DESCRIPTION("perfmon default sampling format"); MODULE_LICENSE("GPL"); -MODULE_PARM(debug, "i"); -MODULE_PARM_DESC(debug, "debug"); - -MODULE_PARM(debug_ovfl, "i"); -MODULE_PARM_DESC(debug_ovfl, "debug ovfl"); - - #define DEFAULT_DEBUG 1 #ifdef DEFAULT_DEBUG #define DPRINT(a) \ do { \ - if (unlikely(debug >0)) { printk("%s.%d: CPU%d ", __FUNCTION__, __LINE__, smp_processor_id()); printk a; } \ + if (unlikely(pfm_sysctl.debug >0)) { printk("%s.%d: CPU%d ", __FUNCTION__, __LINE__, smp_processor_id()); printk a; } \ } while (0) #define DPRINT_ovfl(a) \ do { \ - if (unlikely(debug_ovfl >0)) { printk("%s.%d: CPU%d ", __FUNCTION__, __LINE__, smp_processor_id()); printk a; } \ + if (unlikely(pfm_sysctl.debug > 0 && pfm_sysctl.debug_ovfl >0)) { printk("%s.%d: CPU%d ", __FUNCTION__, __LINE__, smp_processor_id()); printk a; } \ } while (0) #else @@ -45,8 +38,6 @@ MODULE_PARM_DESC(debug_ovfl, "debug ovfl"); #define DPRINT_ovfl(a) #endif -static int debug, debug_ovfl; - static int default_validate(struct task_struct *task, unsigned int flags, int cpu, void *data) { diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index 91293388dd2..ebb71f3d6d1 100644 --- a/arch/ia64/kernel/process.c +++ b/arch/ia64/kernel/process.c @@ -3,6 +3,7 @@ * * Copyright (C) 1998-2003 Hewlett-Packard Co * David Mosberger-Tang <davidm@hpl.hp.com> + * 04/11/17 Ashok Raj <ashok.raj@intel.com> Added CPU Hotplug Support */ #define __KERNEL_SYSCALLS__ /* see <asm/unistd.h> */ #include <linux/config.h> @@ -49,7 +50,7 @@ #include "sigframe.h" void (*ia64_mark_idle)(int); -static cpumask_t cpu_idle_map; +static DEFINE_PER_CPU(unsigned int, cpu_idle_state); unsigned long boot_option_idle_override = 0; EXPORT_SYMBOL(boot_option_idle_override); @@ -172,7 +173,9 @@ do_notify_resume_user (sigset_t *oldset, struct sigscratch *scr, long in_syscall ia64_do_signal(oldset, scr, in_syscall); } -static int pal_halt = 1; +static int pal_halt = 1; +static int can_do_pal_halt = 1; + static int __init nohalt_setup(char * str) { pal_halt = 0; @@ -180,16 +183,20 @@ static int __init nohalt_setup(char * str) } __setup("nohalt", nohalt_setup); +void +update_pal_halt_status(int status) +{ + can_do_pal_halt = pal_halt && status; +} + /* * We use this if we don't have any better idle routine.. */ void default_idle (void) { - unsigned long pmu_active = ia64_getreg(_IA64_REG_PSR) & (IA64_PSR_PP | IA64_PSR_UP); - while (!need_resched()) - if (pal_halt && !pmu_active) + if (can_do_pal_halt) safe_halt(); else cpu_relax(); @@ -200,27 +207,20 @@ default_idle (void) static inline void play_dead(void) { extern void ia64_cpu_local_tick (void); + unsigned int this_cpu = smp_processor_id(); + /* Ack it */ __get_cpu_var(cpu_state) = CPU_DEAD; - /* We shouldn't have to disable interrupts while dead, but - * some interrupts just don't seem to go away, and this makes - * it "work" for testing purposes. */ max_xtp(); local_irq_disable(); - /* Death loop */ - while (__get_cpu_var(cpu_state) != CPU_UP_PREPARE) - cpu_relax(); - + idle_task_exit(); + ia64_jump_to_sal(&sal_boot_rendez_state[this_cpu]); /* - * Enable timer interrupts from now on - * Not required if we put processor in SAL_BOOT_RENDEZ mode. + * The above is a point of no-return, the processor is + * expected to be in SAL loop now. */ - local_flush_tlb_all(); - cpu_set(smp_processor_id(), cpu_online_map); - wmb(); - ia64_cpu_local_tick (); - local_irq_enable(); + BUG(); } #else static inline void play_dead(void) @@ -229,20 +229,31 @@ static inline void play_dead(void) } #endif /* CONFIG_HOTPLUG_CPU */ - void cpu_idle_wait(void) { - int cpu; - cpumask_t map; + unsigned int cpu, this_cpu = get_cpu(); + cpumask_t map; + + set_cpus_allowed(current, cpumask_of_cpu(this_cpu)); + put_cpu(); - for_each_online_cpu(cpu) - cpu_set(cpu, cpu_idle_map); + cpus_clear(map); + for_each_online_cpu(cpu) { + per_cpu(cpu_idle_state, cpu) = 1; + cpu_set(cpu, map); + } - wmb(); - do { - ssleep(1); - cpus_and(map, cpu_idle_map, cpu_online_map); - } while (!cpus_empty(map)); + __get_cpu_var(cpu_idle_state) = 0; + + wmb(); + do { + ssleep(1); + for_each_online_cpu(cpu) { + if (cpu_isset(cpu, map) && !per_cpu(cpu_idle_state, cpu)) + cpu_clear(cpu, map); + } + cpus_and(map, map, cpu_online_map); + } while (!cpus_empty(map)); } EXPORT_SYMBOL_GPL(cpu_idle_wait); @@ -250,7 +261,6 @@ void __attribute__((noreturn)) cpu_idle (void) { void (*mark_idle)(int) = ia64_mark_idle; - int cpu = smp_processor_id(); /* endless idle loop with no priority at all */ while (1) { @@ -261,12 +271,13 @@ cpu_idle (void) while (!need_resched()) { void (*idle)(void); + if (__get_cpu_var(cpu_idle_state)) + __get_cpu_var(cpu_idle_state) = 0; + + rmb(); if (mark_idle) (*mark_idle)(1); - if (cpu_isset(cpu, cpu_idle_map)) - cpu_clear(cpu, cpu_idle_map); - rmb(); idle = pm_idle; if (!idle) idle = default_idle; diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c index 55789fcd721..907464ee727 100644 --- a/arch/ia64/kernel/ptrace.c +++ b/arch/ia64/kernel/ptrace.c @@ -17,6 +17,7 @@ #include <linux/user.h> #include <linux/security.h> #include <linux/audit.h> +#include <linux/signal.h> #include <asm/pgtable.h> #include <asm/processor.h> @@ -1481,7 +1482,7 @@ sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data) case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if (data > _NSIG) + if (!valid_signal(data)) goto out_tsk; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -1520,7 +1521,7 @@ sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data) /* let child execute for one instruction */ case PTRACE_SINGLEBLOCK: ret = -EIO; - if (data > _NSIG) + if (!valid_signal(data)) goto out_tsk; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -1595,20 +1596,25 @@ syscall_trace_enter (long arg0, long arg1, long arg2, long arg3, long arg4, long arg5, long arg6, long arg7, struct pt_regs regs) { - long syscall; + if (test_thread_flag(TIF_SYSCALL_TRACE) + && (current->ptrace & PT_PTRACED)) + syscall_trace(); if (unlikely(current->audit_context)) { - if (IS_IA32_PROCESS(®s)) + long syscall; + int arch; + + if (IS_IA32_PROCESS(®s)) { syscall = regs.r1; - else + arch = AUDIT_ARCH_I386; + } else { syscall = regs.r15; + arch = AUDIT_ARCH_IA64; + } - audit_syscall_entry(current, syscall, arg0, arg1, arg2, arg3); + audit_syscall_entry(current, arch, syscall, arg0, arg1, arg2, arg3); } - if (test_thread_flag(TIF_SYSCALL_TRACE) - && (current->ptrace & PT_PTRACED)) - syscall_trace(); } /* "asmlinkage" so the input arguments are preserved... */ @@ -1619,7 +1625,7 @@ syscall_trace_leave (long arg0, long arg1, long arg2, long arg3, struct pt_regs regs) { if (unlikely(current->audit_context)) - audit_syscall_exit(current, regs.r8); + audit_syscall_exit(current, AUDITSC_RESULT(regs.r10), regs.r8); if (test_thread_flag(TIF_SYSCALL_TRACE) && (current->ptrace & PT_PTRACED)) diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index f05650c801d..b7e6b4cb374 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -4,10 +4,15 @@ * Copyright (C) 1998-2001, 2003-2004 Hewlett-Packard Co * David Mosberger-Tang <davidm@hpl.hp.com> * Stephane Eranian <eranian@hpl.hp.com> - * Copyright (C) 2000, Rohit Seth <rohit.seth@intel.com> + * Copyright (C) 2000, 2004 Intel Corp + * Rohit Seth <rohit.seth@intel.com> + * Suresh Siddha <suresh.b.siddha@intel.com> + * Gordon Jin <gordon.jin@intel.com> * Copyright (C) 1999 VA Linux Systems * Copyright (C) 1999 Walt Drummond <drummond@valinux.com> * + * 12/26/04 S.Siddha, G.Jin, R.Seth + * Add multi-threading and multi-core detection * 11/12/01 D.Mosberger Convert get_cpuinfo() to seq_file based show_cpuinfo(). * 04/04/00 D.Mosberger renamed cpu_initialized to cpu_online_map * 03/31/00 R.Seth cpu_initialized and current->processor fixes @@ -296,6 +301,34 @@ mark_bsp_online (void) #endif } +#ifdef CONFIG_SMP +static void +check_for_logical_procs (void) +{ + pal_logical_to_physical_t info; + s64 status; + + status = ia64_pal_logical_to_phys(0, &info); + if (status == -1) { + printk(KERN_INFO "No logical to physical processor mapping " + "available\n"); + return; + } + if (status) { + printk(KERN_ERR "ia64_pal_logical_to_phys failed with %ld\n", + status); + return; + } + /* + * Total number of siblings that BSP has. Though not all of them + * may have booted successfully. The correct number of siblings + * booted is in info.overview_num_log. + */ + smp_num_siblings = info.overview_tpc; + smp_num_cpucores = info.overview_cpp; +} +#endif + void __init setup_arch (char **cmdline_p) { @@ -356,6 +389,19 @@ setup_arch (char **cmdline_p) #ifdef CONFIG_SMP cpu_physical_id(0) = hard_smp_processor_id(); + + cpu_set(0, cpu_sibling_map[0]); + cpu_set(0, cpu_core_map[0]); + + check_for_logical_procs(); + if (smp_num_cpucores > 1) + printk(KERN_INFO + "cpu package is Multi-Core capable: number of cores=%d\n", + smp_num_cpucores); + if (smp_num_siblings > 1) + printk(KERN_INFO + "cpu package is Multi-Threading capable: number of siblings=%d\n", + smp_num_siblings); #endif cpu_init(); /* initialize the bootstrap CPU */ @@ -459,12 +505,23 @@ show_cpuinfo (struct seq_file *m, void *v) "cpu regs : %u\n" "cpu MHz : %lu.%06lu\n" "itc MHz : %lu.%06lu\n" - "BogoMIPS : %lu.%02lu\n\n", + "BogoMIPS : %lu.%02lu\n", cpunum, c->vendor, family, c->model, c->revision, c->archrev, features, c->ppn, c->number, c->proc_freq / 1000000, c->proc_freq % 1000000, c->itc_freq / 1000000, c->itc_freq % 1000000, lpj*HZ/500000, (lpj*HZ/5000) % 100); +#ifdef CONFIG_SMP + seq_printf(m, "siblings : %u\n", c->num_log); + if (c->threads_per_core > 1 || c->cores_per_socket > 1) + seq_printf(m, + "physical id: %u\n" + "core id : %u\n" + "thread id : %u\n", + c->socket_id, c->core_id, c->thread_id); +#endif + seq_printf(m,"\n"); + return 0; } @@ -533,6 +590,14 @@ identify_cpu (struct cpuinfo_ia64 *c) memcpy(c->vendor, cpuid.field.vendor, 16); #ifdef CONFIG_SMP c->cpu = smp_processor_id(); + + /* below default values will be overwritten by identify_siblings() + * for Multi-Threading/Multi-Core capable cpu's + */ + c->threads_per_core = c->cores_per_socket = c->num_log = 1; + c->socket_id = -1; + + identify_siblings(c); #endif c->ppn = cpuid.field.ppn; c->number = cpuid.field.number; diff --git a/arch/ia64/kernel/signal.c b/arch/ia64/kernel/signal.c index 6891d86937d..499b7e5317c 100644 --- a/arch/ia64/kernel/signal.c +++ b/arch/ia64/kernel/signal.c @@ -224,7 +224,8 @@ ia64_rt_sigreturn (struct sigscratch *scr) * could be corrupted. */ retval = (long) &ia64_leave_kernel; - if (test_thread_flag(TIF_SYSCALL_TRACE)) + if (test_thread_flag(TIF_SYSCALL_TRACE) + || test_thread_flag(TIF_SYSCALL_AUDIT)) /* * strace expects to be notified after sigreturn returns even though the * context to which we return may not be in the middle of a syscall. diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index 5318f0cbfc2..0d5ee57c986 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -1,14 +1,25 @@ /* * SMP boot-related support * - * Copyright (C) 1998-2003 Hewlett-Packard Co + * Copyright (C) 1998-2003, 2005 Hewlett-Packard Co * David Mosberger-Tang <davidm@hpl.hp.com> + * Copyright (C) 2001, 2004-2005 Intel Corp + * Rohit Seth <rohit.seth@intel.com> + * Suresh Siddha <suresh.b.siddha@intel.com> + * Gordon Jin <gordon.jin@intel.com> + * Ashok Raj <ashok.raj@intel.com> * * 01/05/16 Rohit Seth <rohit.seth@intel.com> Moved SMP booting functions from smp.c to here. * 01/04/27 David Mosberger <davidm@hpl.hp.com> Added ITC synching code. * 02/07/31 David Mosberger <davidm@hpl.hp.com> Switch over to hotplug-CPU boot-sequence. * smp_boot_cpus()/smp_commence() is replaced by * smp_prepare_cpus()/__cpu_up()/smp_cpus_done(). + * 04/06/21 Ashok Raj <ashok.raj@intel.com> Added CPU Hotplug Support + * 04/12/26 Jin Gordon <gordon.jin@intel.com> + * 04/12/26 Rohit Seth <rohit.seth@intel.com> + * Add multi-threading and multi-core detection + * 05/01/30 Suresh Siddha <suresh.b.siddha@intel.com> + * Setup cpu_sibling_map and cpu_core_map */ #include <linux/config.h> @@ -58,6 +69,37 @@ #define Dprintk(x...) #endif +#ifdef CONFIG_HOTPLUG_CPU +/* + * Store all idle threads, this can be reused instead of creating + * a new thread. Also avoids complicated thread destroy functionality + * for idle threads. + */ +struct task_struct *idle_thread_array[NR_CPUS]; + +/* + * Global array allocated for NR_CPUS at boot time + */ +struct sal_to_os_boot sal_boot_rendez_state[NR_CPUS]; + +/* + * start_ap in head.S uses this to store current booting cpu + * info. + */ +struct sal_to_os_boot *sal_state_for_booting_cpu = &sal_boot_rendez_state[0]; + +#define set_brendez_area(x) (sal_state_for_booting_cpu = &sal_boot_rendez_state[(x)]); + +#define get_idle_for_cpu(x) (idle_thread_array[(x)]) +#define set_idle_for_cpu(x,p) (idle_thread_array[(x)] = (p)) + +#else + +#define get_idle_for_cpu(x) (NULL) +#define set_idle_for_cpu(x,p) +#define set_brendez_area(x) +#endif + /* * ITC synchronization related stuff: @@ -90,6 +132,11 @@ EXPORT_SYMBOL(cpu_online_map); cpumask_t cpu_possible_map; EXPORT_SYMBOL(cpu_possible_map); +cpumask_t cpu_core_map[NR_CPUS] __cacheline_aligned; +cpumask_t cpu_sibling_map[NR_CPUS] __cacheline_aligned; +int smp_num_siblings = 1; +int smp_num_cpucores = 1; + /* which logical CPU number maps to which CPU (physical APIC ID) */ volatile int ia64_cpu_to_sapicid[NR_CPUS]; EXPORT_SYMBOL(ia64_cpu_to_sapicid); @@ -124,7 +171,8 @@ sync_master (void *arg) local_irq_save(flags); { for (i = 0; i < NUM_ROUNDS*NUM_ITERS; ++i) { - while (!go[MASTER]); + while (!go[MASTER]) + cpu_relax(); go[MASTER] = 0; go[SLAVE] = ia64_get_itc(); } @@ -147,7 +195,8 @@ get_delta (long *rt, long *master) for (i = 0; i < NUM_ITERS; ++i) { t0 = ia64_get_itc(); go[MASTER] = 1; - while (!(tm = go[SLAVE])); + while (!(tm = go[SLAVE])) + cpu_relax(); go[SLAVE] = 0; t1 = ia64_get_itc(); @@ -226,7 +275,8 @@ ia64_sync_itc (unsigned int master) return; } - while (go[MASTER]); /* wait for master to be ready */ + while (go[MASTER]) + cpu_relax(); /* wait for master to be ready */ spin_lock_irqsave(&itc_sync_lock, flags); { @@ -345,7 +395,6 @@ start_secondary (void *unused) { /* Early console may use I/O ports */ ia64_set_kr(IA64_KR_IO_BASE, __pa(ia64_iobase)); - Dprintk("start_secondary: starting CPU 0x%x\n", hard_smp_processor_id()); efi_map_pal_code(); cpu_init(); @@ -384,6 +433,13 @@ do_boot_cpu (int sapicid, int cpu) .done = COMPLETION_INITIALIZER(c_idle.done), }; DECLARE_WORK(work, do_fork_idle, &c_idle); + + c_idle.idle = get_idle_for_cpu(cpu); + if (c_idle.idle) { + init_idle(c_idle.idle, cpu); + goto do_rest; + } + /* * We can't use kernel_thread since we must avoid to reschedule the child. */ @@ -396,10 +452,15 @@ do_boot_cpu (int sapicid, int cpu) if (IS_ERR(c_idle.idle)) panic("failed fork for CPU %d", cpu); + + set_idle_for_cpu(cpu, c_idle.idle); + +do_rest: task_for_booting_cpu = c_idle.idle; Dprintk("Sending wakeup vector %lu to AP 0x%x/0x%x.\n", ap_wakeup_vector, cpu, sapicid); + set_brendez_area(cpu); platform_send_ipi(cpu, ap_wakeup_vector, IA64_IPI_DM_INT, 0); /* @@ -552,19 +613,70 @@ void __devinit smp_prepare_boot_cpu(void) cpu_set(smp_processor_id(), cpu_callin_map); } +/* + * mt_info[] is a temporary store for all info returned by + * PAL_LOGICAL_TO_PHYSICAL, to be copied into cpuinfo_ia64 when the + * specific cpu comes. + */ +static struct { + __u32 socket_id; + __u16 core_id; + __u16 thread_id; + __u16 proc_fixed_addr; + __u8 valid; +}mt_info[NR_CPUS] __devinit; + #ifdef CONFIG_HOTPLUG_CPU -extern void fixup_irqs(void); -/* must be called with cpucontrol mutex held */ -static int __devinit cpu_enable(unsigned int cpu) +static inline void +remove_from_mtinfo(int cpu) { - per_cpu(cpu_state,cpu) = CPU_UP_PREPARE; - wmb(); + int i; - while (!cpu_online(cpu)) - cpu_relax(); - return 0; + for_each_cpu(i) + if (mt_info[i].valid && mt_info[i].socket_id == + cpu_data(cpu)->socket_id) + mt_info[i].valid = 0; +} + +static inline void +clear_cpu_sibling_map(int cpu) +{ + int i; + + for_each_cpu_mask(i, cpu_sibling_map[cpu]) + cpu_clear(cpu, cpu_sibling_map[i]); + for_each_cpu_mask(i, cpu_core_map[cpu]) + cpu_clear(cpu, cpu_core_map[i]); + + cpu_sibling_map[cpu] = cpu_core_map[cpu] = CPU_MASK_NONE; +} + +static void +remove_siblinginfo(int cpu) +{ + int last = 0; + + if (cpu_data(cpu)->threads_per_core == 1 && + cpu_data(cpu)->cores_per_socket == 1) { + cpu_clear(cpu, cpu_core_map[cpu]); + cpu_clear(cpu, cpu_sibling_map[cpu]); + return; + } + + last = (cpus_weight(cpu_core_map[cpu]) == 1 ? 1 : 0); + + /* remove it from all sibling map's */ + clear_cpu_sibling_map(cpu); + + /* if this cpu is the last in the core group, remove all its info + * from mt_info structure + */ + if (last) + remove_from_mtinfo(cpu); } +extern void fixup_irqs(void); +/* must be called with cpucontrol mutex held */ int __cpu_disable(void) { int cpu = smp_processor_id(); @@ -575,9 +687,10 @@ int __cpu_disable(void) if (cpu == 0) return -EBUSY; + remove_siblinginfo(cpu); fixup_irqs(); local_flush_tlb_all(); - printk ("Disabled cpu %u\n", smp_processor_id()); + cpu_clear(cpu, cpu_callin_map); return 0; } @@ -589,12 +702,7 @@ void __cpu_die(unsigned int cpu) /* They ack this in play_dead by setting CPU_DEAD */ if (per_cpu(cpu_state, cpu) == CPU_DEAD) { - /* - * TBD: Enable this when physical removal - * or when we put the processor is put in - * SAL_BOOT_RENDEZ mode - * cpu_clear(cpu, cpu_callin_map); - */ + printk ("CPU %d is now offline\n", cpu); return; } msleep(100); @@ -602,11 +710,6 @@ void __cpu_die(unsigned int cpu) printk(KERN_ERR "CPU %u didn't die...\n", cpu); } #else /* !CONFIG_HOTPLUG_CPU */ -static int __devinit cpu_enable(unsigned int cpu) -{ - return 0; -} - int __cpu_disable(void) { return -ENOSYS; @@ -637,6 +740,23 @@ smp_cpus_done (unsigned int dummy) (int)num_online_cpus(), bogosum/(500000/HZ), (bogosum/(5000/HZ))%100); } +static inline void __devinit +set_cpu_sibling_map(int cpu) +{ + int i; + + for_each_online_cpu(i) { + if ((cpu_data(cpu)->socket_id == cpu_data(i)->socket_id)) { + cpu_set(i, cpu_core_map[cpu]); + cpu_set(cpu, cpu_core_map[i]); + if (cpu_data(cpu)->core_id == cpu_data(i)->core_id) { + cpu_set(i, cpu_sibling_map[cpu]); + cpu_set(cpu, cpu_sibling_map[i]); + } + } + } +} + int __devinit __cpu_up (unsigned int cpu) { @@ -648,21 +768,26 @@ __cpu_up (unsigned int cpu) return -EINVAL; /* - * Already booted.. just enable and get outa idle lool + * Already booted cpu? not valid anymore since we dont + * do idle loop tightspin anymore. */ if (cpu_isset(cpu, cpu_callin_map)) - { - cpu_enable(cpu); - local_irq_enable(); - while (!cpu_isset(cpu, cpu_online_map)) - mb(); - return 0; - } + return -EINVAL; + /* Processor goes to start_secondary(), sets online flag */ ret = do_boot_cpu(sapicid, cpu); if (ret < 0) return ret; + if (cpu_data(cpu)->threads_per_core == 1 && + cpu_data(cpu)->cores_per_socket == 1) { + cpu_set(cpu, cpu_sibling_map[cpu]); + cpu_set(cpu, cpu_core_map[cpu]); + return 0; + } + + set_cpu_sibling_map(cpu); + return 0; } @@ -690,3 +815,106 @@ init_smp_config(void) ia64_sal_strerror(sal_ret)); } +static inline int __devinit +check_for_mtinfo_index(void) +{ + int i; + + for_each_cpu(i) + if (!mt_info[i].valid) + return i; + + return -1; +} + +/* + * Search the mt_info to find out if this socket's cid/tid information is + * cached or not. If the socket exists, fill in the core_id and thread_id + * in cpuinfo + */ +static int __devinit +check_for_new_socket(__u16 logical_address, struct cpuinfo_ia64 *c) +{ + int i; + __u32 sid = c->socket_id; + + for_each_cpu(i) { + if (mt_info[i].valid && mt_info[i].proc_fixed_addr == logical_address + && mt_info[i].socket_id == sid) { + c->core_id = mt_info[i].core_id; + c->thread_id = mt_info[i].thread_id; + return 1; /* not a new socket */ + } + } + return 0; +} + +/* + * identify_siblings(cpu) gets called from identify_cpu. This populates the + * information related to logical execution units in per_cpu_data structure. + */ +void __devinit +identify_siblings(struct cpuinfo_ia64 *c) +{ + s64 status; + u16 pltid; + u64 proc_fixed_addr; + int count, i; + pal_logical_to_physical_t info; + + if (smp_num_cpucores == 1 && smp_num_siblings == 1) + return; + + if ((status = ia64_pal_logical_to_phys(0, &info)) != PAL_STATUS_SUCCESS) { + printk(KERN_ERR "ia64_pal_logical_to_phys failed with %ld\n", + status); + return; + } + if ((status = ia64_sal_physical_id_info(&pltid)) != PAL_STATUS_SUCCESS) { + printk(KERN_ERR "ia64_sal_pltid failed with %ld\n", status); + return; + } + if ((status = ia64_pal_fixed_addr(&proc_fixed_addr)) != PAL_STATUS_SUCCESS) { + printk(KERN_ERR "ia64_pal_fixed_addr failed with %ld\n", status); + return; + } + + c->socket_id = (pltid << 8) | info.overview_ppid; + c->cores_per_socket = info.overview_cpp; + c->threads_per_core = info.overview_tpc; + count = c->num_log = info.overview_num_log; + + /* If the thread and core id information is already cached, then + * we will simply update cpu_info and return. Otherwise, we will + * do the PAL calls and cache core and thread id's of all the siblings. + */ + if (check_for_new_socket(proc_fixed_addr, c)) + return; + + for (i = 0; i < count; i++) { + int index; + + if (i && (status = ia64_pal_logical_to_phys(i, &info)) + != PAL_STATUS_SUCCESS) { + printk(KERN_ERR "ia64_pal_logical_to_phys failed" + " with %ld\n", status); + return; + } + if (info.log2_la == proc_fixed_addr) { + c->core_id = info.log1_cid; + c->thread_id = info.log1_tid; + } + + index = check_for_mtinfo_index(); + /* We will not do the mt_info caching optimization in this case. + */ + if (index < 0) + continue; + + mt_info[index].valid = 1; + mt_info[index].socket_id = c->socket_id; + mt_info[index].core_id = info.log1_cid; + mt_info[index].thread_id = info.log1_tid; + mt_info[index].proc_fixed_addr = info.log2_la; + } +} diff --git a/arch/ia64/kernel/sys_ia64.c b/arch/ia64/kernel/sys_ia64.c index 3ac216e1c8b..a8cf6d8a509 100644 --- a/arch/ia64/kernel/sys_ia64.c +++ b/arch/ia64/kernel/sys_ia64.c @@ -93,20 +93,6 @@ sys_getpagesize (void) } asmlinkage unsigned long -ia64_shmat (int shmid, void __user *shmaddr, int shmflg) -{ - unsigned long raddr; - int retval; - - retval = do_shmat(shmid, shmaddr, shmflg, &raddr); - if (retval < 0) - return retval; - - force_successful_syscall_return(); - return raddr; -} - -asmlinkage unsigned long ia64_brk (unsigned long brk) { unsigned long rlim, retval, newbrk, oldbrk; diff --git a/arch/ia64/kernel/unwind.c b/arch/ia64/kernel/unwind.c index d494ff647ca..2776a074c6f 100644 --- a/arch/ia64/kernel/unwind.c +++ b/arch/ia64/kernel/unwind.c @@ -1943,23 +1943,30 @@ EXPORT_SYMBOL(unw_unwind); int unw_unwind_to_user (struct unw_frame_info *info) { - unsigned long ip, sp; + unsigned long ip, sp, pr = 0; while (unw_unwind(info) >= 0) { - if (unw_get_rp(info, &ip) < 0) { - unw_get_ip(info, &ip); - UNW_DPRINT(0, "unwind.%s: failed to read return pointer (ip=0x%lx)\n", - __FUNCTION__, ip); - return -1; - } unw_get_sp(info, &sp); - if (sp >= (unsigned long)info->task + IA64_STK_OFFSET) + if ((long)((unsigned long)info->task + IA64_STK_OFFSET - sp) + < IA64_PT_REGS_SIZE) { + UNW_DPRINT(0, "unwind.%s: ran off the top of the kernel stack\n", + __FUNCTION__); break; - if (ip < FIXADDR_USER_END) + } + if (unw_is_intr_frame(info) && + (pr & (1UL << PRED_USER_STACK))) return 0; + if (unw_get_pr (info, &pr) < 0) { + unw_get_rp(info, &ip); + UNW_DPRINT(0, "unwind.%s: failed to read " + "predicate register (ip=0x%lx)\n", + __FUNCTION__, ip); + return -1; + } } unw_get_ip(info, &ip); - UNW_DPRINT(0, "unwind.%s: failed to unwind to user-level (ip=0x%lx)\n", __FUNCTION__, ip); + UNW_DPRINT(0, "unwind.%s: failed to unwind to user-level (ip=0x%lx)\n", + __FUNCTION__, ip); return -1; } EXPORT_SYMBOL(unw_unwind_to_user); diff --git a/arch/ia64/lib/flush.S b/arch/ia64/lib/flush.S index 29c802b1966..a1af9146cfd 100644 --- a/arch/ia64/lib/flush.S +++ b/arch/ia64/lib/flush.S @@ -1,8 +1,8 @@ /* * Cache flushing routines. * - * Copyright (C) 1999-2001 Hewlett-Packard Co - * Copyright (C) 1999-2001 David Mosberger-Tang <davidm@hpl.hp.com> + * Copyright (C) 1999-2001, 2005 Hewlett-Packard Co + * David Mosberger-Tang <davidm@hpl.hp.com> */ #include <asm/asmmacro.h> #include <asm/page.h> @@ -26,7 +26,7 @@ GLOBAL_ENTRY(flush_icache_range) mov ar.lc=r8 ;; -.Loop: fc in0 // issuable on M0 only +.Loop: fc.i in0 // issuable on M2 only add in0=32,in0 br.cloop.sptk.few .Loop ;; diff --git a/arch/ia64/lib/memcpy_mck.S b/arch/ia64/lib/memcpy_mck.S index 6f26ef7cc23..6f308e62c13 100644 --- a/arch/ia64/lib/memcpy_mck.S +++ b/arch/ia64/lib/memcpy_mck.S @@ -75,6 +75,7 @@ GLOBAL_ENTRY(memcpy) mov f6=f0 br.cond.sptk .common_code ;; +END(memcpy) GLOBAL_ENTRY(__copy_user) .prologue // check dest alignment @@ -300,7 +301,7 @@ EK(.ex_handler, (p[D]) st8 [dst1] = t15, 4*8) add src_pre_mem=0,src0 // prefetch src pointer add dst_pre_mem=0,dst0 // prefetch dest pointer and src0=-8,src0 // 1st src pointer -(p7) mov ar.lc = r21 +(p7) mov ar.lc = cnt (p8) mov ar.lc = r0 ;; TEXT_ALIGN(32) @@ -524,7 +525,6 @@ EK(.ex_handler, (p17) st8 [dst1]=r39,8); \ #undef B #undef C #undef D -END(memcpy) /* * Due to lack of local tag support in gcc 2.x assembler, it is not clear which diff --git a/arch/ia64/lib/memset.S b/arch/ia64/lib/memset.S index bd8cf907fe2..f26c16aefb1 100644 --- a/arch/ia64/lib/memset.S +++ b/arch/ia64/lib/memset.S @@ -57,10 +57,10 @@ GLOBAL_ENTRY(memset) { .mmi .prologue alloc tmp = ar.pfs, 3, 0, 0, 0 - .body lfetch.nt1 [dest] // .save ar.lc, save_lc mov.i save_lc = ar.lc + .body } { .mmi mov ret0 = dest // return value cmp.ne p_nz, p_zr = value, r0 // use stf.spill if value is zero diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index 6daf15ac894..91a055f5731 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -61,7 +61,8 @@ show_mem (void) printk("%d reserved pages\n", reserved); printk("%d pages shared\n", shared); printk("%d pages swap cached\n", cached); - printk("%ld pages in page table cache\n", pgtable_cache_size); + printk("%ld pages in page table cache\n", + pgtable_quicklist_total_size()); } /* physical address where the bootmem map is located */ diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 3456a9b6971..c0071092939 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -582,7 +582,8 @@ void show_mem(void) printk("%d reserved pages\n", total_reserved); printk("%d pages shared\n", total_shared); printk("%d pages swap cached\n", total_cached); - printk("Total of %ld pages in page table cache\n", pgtable_cache_size); + printk("Total of %ld pages in page table cache\n", + pgtable_quicklist_total_size()); printk("%d free buffer pages\n", nr_free_buffer_pages()); } diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c index da859125aae..4174ec999dd 100644 --- a/arch/ia64/mm/fault.c +++ b/arch/ia64/mm/fault.c @@ -209,10 +209,13 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re } no_context: - if (isr & IA64_ISR_SP) { + if ((isr & IA64_ISR_SP) + || ((isr & IA64_ISR_NA) && (isr & IA64_ISR_CODE_MASK) == IA64_ISR_CODE_LFETCH)) + { /* - * This fault was due to a speculative load set the "ed" bit in the psr to - * ensure forward progress (target register will get a NaT). + * This fault was due to a speculative load or lfetch.fault, set the "ed" + * bit in the psr to ensure forward progress. (Target register will get a + * NaT for ld.s, lfetch will be canceled.) */ ia64_psr(regs)->ed = 1; return; diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index 65cf839573e..547785e3cba 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -39,6 +39,9 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); +DEFINE_PER_CPU(unsigned long *, __pgtable_quicklist); +DEFINE_PER_CPU(long, __pgtable_quicklist_size); + extern void ia64_tlb_init (void); unsigned long MAX_DMA_ADDRESS = PAGE_OFFSET + 0x100000000UL; @@ -50,27 +53,53 @@ struct page *vmem_map; EXPORT_SYMBOL(vmem_map); #endif -static int pgt_cache_water[2] = { 25, 50 }; - -struct page *zero_page_memmap_ptr; /* map entry for zero page */ +struct page *zero_page_memmap_ptr; /* map entry for zero page */ EXPORT_SYMBOL(zero_page_memmap_ptr); +#define MIN_PGT_PAGES 25UL +#define MAX_PGT_FREES_PER_PASS 16L +#define PGT_FRACTION_OF_NODE_MEM 16 + +static inline long +max_pgt_pages(void) +{ + u64 node_free_pages, max_pgt_pages; + +#ifndef CONFIG_NUMA + node_free_pages = nr_free_pages(); +#else + node_free_pages = nr_free_pages_pgdat(NODE_DATA(numa_node_id())); +#endif + max_pgt_pages = node_free_pages / PGT_FRACTION_OF_NODE_MEM; + max_pgt_pages = max(max_pgt_pages, MIN_PGT_PAGES); + return max_pgt_pages; +} + +static inline long +min_pages_to_free(void) +{ + long pages_to_free; + + pages_to_free = pgtable_quicklist_size - max_pgt_pages(); + pages_to_free = min(pages_to_free, MAX_PGT_FREES_PER_PASS); + return pages_to_free; +} + void -check_pgt_cache (void) +check_pgt_cache(void) { - int low, high; + long pages_to_free; - low = pgt_cache_water[0]; - high = pgt_cache_water[1]; + if (unlikely(pgtable_quicklist_size <= MIN_PGT_PAGES)) + return; preempt_disable(); - if (pgtable_cache_size > (u64) high) { - do { - if (pgd_quicklist) - free_page((unsigned long)pgd_alloc_one_fast(NULL)); - if (pmd_quicklist) - free_page((unsigned long)pmd_alloc_one_fast(NULL, 0)); - } while (pgtable_cache_size > (u64) low); + while (unlikely((pages_to_free = min_pages_to_free()) > 0)) { + while (pages_to_free--) { + free_page((unsigned long)pgtable_quicklist_alloc()); + } + preempt_enable(); + preempt_disable(); } preempt_enable(); } @@ -523,11 +552,14 @@ void mem_init (void) { long reserved_pages, codesize, datasize, initsize; - unsigned long num_pgt_pages; pg_data_t *pgdat; int i; static struct kcore_list kcore_mem, kcore_vmem, kcore_kernel; + BUG_ON(PTRS_PER_PGD * sizeof(pgd_t) != PAGE_SIZE); + BUG_ON(PTRS_PER_PMD * sizeof(pmd_t) != PAGE_SIZE); + BUG_ON(PTRS_PER_PTE * sizeof(pte_t) != PAGE_SIZE); + #ifdef CONFIG_PCI /* * This needs to be called _after_ the command line has been parsed but _before_ @@ -564,18 +596,6 @@ mem_init (void) num_physpages << (PAGE_SHIFT - 10), codesize >> 10, reserved_pages << (PAGE_SHIFT - 10), datasize >> 10, initsize >> 10); - /* - * Allow for enough (cached) page table pages so that we can map the entire memory - * at least once. Each task also needs a couple of page tables pages, so add in a - * fudge factor for that (don't use "threads-max" here; that would be wrong!). - * Don't allow the cache to be more than 10% of total memory, though. - */ -# define NUM_TASKS 500 /* typical number of tasks */ - num_pgt_pages = nr_free_pages() / PTRS_PER_PGD + NUM_TASKS; - if (num_pgt_pages > nr_free_pages() / 10) - num_pgt_pages = nr_free_pages() / 10; - if (num_pgt_pages > (u64) pgt_cache_water[1]) - pgt_cache_water[1] = num_pgt_pages; /* * For fsyscall entrpoints with no light-weight handler, use the ordinary diff --git a/arch/ia64/sn/include/pci/pcibr_provider.h b/arch/ia64/sn/include/pci/pcibr_provider.h index b1f05ffec70..1cd291d8bad 100644 --- a/arch/ia64/sn/include/pci/pcibr_provider.h +++ b/arch/ia64/sn/include/pci/pcibr_provider.h @@ -123,9 +123,11 @@ pcibr_lock(struct pcibus_info *pcibus_info) } #define pcibr_unlock(pcibus_info, flag) spin_unlock_irqrestore(&pcibus_info->pbi_lock, flag) +extern int pcibr_init_provider(void); extern void *pcibr_bus_fixup(struct pcibus_bussoft *); -extern uint64_t pcibr_dma_map(struct pcidev_info *, unsigned long, size_t, unsigned int); -extern void pcibr_dma_unmap(struct pcidev_info *, dma_addr_t, int); +extern dma_addr_t pcibr_dma_map(struct pci_dev *, unsigned long, size_t); +extern dma_addr_t pcibr_dma_map_consistent(struct pci_dev *, unsigned long, size_t); +extern void pcibr_dma_unmap(struct pci_dev *, dma_addr_t, int); /* * prototypes for the bridge asic register access routines in pcibr_reg.c diff --git a/arch/ia64/sn/kernel/Makefile b/arch/ia64/sn/kernel/Makefile index 6c7f4d9e8ea..4351c4ff984 100644 --- a/arch/ia64/sn/kernel/Makefile +++ b/arch/ia64/sn/kernel/Makefile @@ -4,9 +4,15 @@ # License. See the file "COPYING" in the main directory of this archive # for more details. # -# Copyright (C) 1999,2001-2003 Silicon Graphics, Inc. All Rights Reserved. +# Copyright (C) 1999,2001-2005 Silicon Graphics, Inc. All Rights Reserved. # obj-y += setup.o bte.o bte_error.o irq.o mca.o idle.o \ huberror.o io_init.o iomv.o klconflib.o sn2/ obj-$(CONFIG_IA64_GENERIC) += machvec.o +obj-$(CONFIG_SGI_TIOCX) += tiocx.o +obj-$(CONFIG_IA64_SGI_SN_XP) += xp.o +xp-y := xp_main.o xp_nofault.o +obj-$(CONFIG_IA64_SGI_SN_XP) += xpc.o +xpc-y := xpc_main.o xpc_channel.o xpc_partition.o +obj-$(CONFIG_IA64_SGI_SN_XP) += xpnet.o diff --git a/arch/ia64/sn/kernel/bte.c b/arch/ia64/sn/kernel/bte.c index ce0bc4085ea..647deae9bfc 100644 --- a/arch/ia64/sn/kernel/bte.c +++ b/arch/ia64/sn/kernel/bte.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. */ #include <linux/config.h> @@ -170,10 +170,6 @@ retry_bteop: /* Initialize the notification to a known value. */ *bte->most_rcnt_na = BTE_WORD_BUSY; - /* Set the status reg busy bit and transfer length */ - BTE_PRINTKV(("IBLS = 0x%lx\n", IBLS_BUSY | transfer_size)); - BTE_LNSTAT_STORE(bte, IBLS_BUSY | transfer_size); - /* Set the source and destination registers */ BTE_PRINTKV(("IBSA = 0x%lx)\n", (TO_PHYS(src)))); BTE_SRC_STORE(bte, TO_PHYS(src)); @@ -188,7 +184,7 @@ retry_bteop: /* Initiate the transfer */ BTE_PRINTK(("IBCT = 0x%lx)\n", BTE_VALID_MODE(mode))); - BTE_CTRL_STORE(bte, BTE_VALID_MODE(mode)); + BTE_START_TRANSFER(bte, transfer_size, BTE_VALID_MODE(mode)); itc_end = ia64_get_itc() + (40000000 * local_cpu_data->cyc_per_usec); @@ -429,10 +425,16 @@ void bte_init_node(nodepda_t * mynodepda, cnodeid_t cnode) mynodepda->bte_recovery_timer.data = (unsigned long)mynodepda; for (i = 0; i < BTES_PER_NODE; i++) { + u64 *base_addr; + /* Which link status register should we use? */ - unsigned long link_status = (i == 0 ? IIO_IBLS0 : IIO_IBLS1); - mynodepda->bte_if[i].bte_base_addr = (u64 *) - REMOTE_HUB_ADDR(cnodeid_to_nasid(cnode), link_status); + base_addr = (u64 *) + REMOTE_HUB_ADDR(cnodeid_to_nasid(cnode), BTE_BASE_ADDR(i)); + mynodepda->bte_if[i].bte_base_addr = base_addr; + mynodepda->bte_if[i].bte_source_addr = BTE_SOURCE_ADDR(base_addr); + mynodepda->bte_if[i].bte_destination_addr = BTE_DEST_ADDR(base_addr); + mynodepda->bte_if[i].bte_control_addr = BTE_CTRL_ADDR(base_addr); + mynodepda->bte_if[i].bte_notify_addr = BTE_NOTIF_ADDR(base_addr); /* * Initialize the notification and spinlock diff --git a/arch/ia64/sn/kernel/bte_error.c b/arch/ia64/sn/kernel/bte_error.c index fd104312c6b..fcbc748ae43 100644 --- a/arch/ia64/sn/kernel/bte_error.c +++ b/arch/ia64/sn/kernel/bte_error.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. */ #include <linux/types.h> @@ -33,48 +33,28 @@ void bte_error_handler(unsigned long); * Wait until all BTE related CRBs are completed * and then reset the interfaces. */ -void bte_error_handler(unsigned long _nodepda) +void shub1_bte_error_handler(unsigned long _nodepda) { struct nodepda_s *err_nodepda = (struct nodepda_s *)_nodepda; - spinlock_t *recovery_lock = &err_nodepda->bte_recovery_lock; struct timer_list *recovery_timer = &err_nodepda->bte_recovery_timer; nasid_t nasid; int i; int valid_crbs; - unsigned long irq_flags; - volatile u64 *notify; - bte_result_t bh_error; ii_imem_u_t imem; /* II IMEM Register */ ii_icrb0_d_u_t icrbd; /* II CRB Register D */ ii_ibcr_u_t ibcr; ii_icmr_u_t icmr; ii_ieclr_u_t ieclr; - BTE_PRINTK(("bte_error_handler(%p) - %d\n", err_nodepda, + BTE_PRINTK(("shub1_bte_error_handler(%p) - %d\n", err_nodepda, smp_processor_id())); - spin_lock_irqsave(recovery_lock, irq_flags); - if ((err_nodepda->bte_if[0].bh_error == BTE_SUCCESS) && (err_nodepda->bte_if[1].bh_error == BTE_SUCCESS)) { BTE_PRINTK(("eh:%p:%d Nothing to do.\n", err_nodepda, smp_processor_id())); - spin_unlock_irqrestore(recovery_lock, irq_flags); return; } - /* - * Lock all interfaces on this node to prevent new transfers - * from being queued. - */ - for (i = 0; i < BTES_PER_NODE; i++) { - if (err_nodepda->bte_if[i].cleanup_active) { - continue; - } - spin_lock(&err_nodepda->bte_if[i].spinlock); - BTE_PRINTK(("eh:%p:%d locked %d\n", err_nodepda, - smp_processor_id(), i)); - err_nodepda->bte_if[i].cleanup_active = 1; - } /* Determine information about our hub */ nasid = cnodeid_to_nasid(err_nodepda->bte_if[0].bte_cnode); @@ -101,7 +81,6 @@ void bte_error_handler(unsigned long _nodepda) mod_timer(recovery_timer, HZ * 5); BTE_PRINTK(("eh:%p:%d Marked Giving up\n", err_nodepda, smp_processor_id())); - spin_unlock_irqrestore(recovery_lock, irq_flags); return; } if (icmr.ii_icmr_fld_s.i_crb_vld != 0) { @@ -120,8 +99,6 @@ void bte_error_handler(unsigned long _nodepda) BTE_PRINTK(("eh:%p:%d Valid %d, Giving up\n", err_nodepda, smp_processor_id(), i)); - spin_unlock_irqrestore(recovery_lock, - irq_flags); return; } } @@ -146,6 +123,51 @@ void bte_error_handler(unsigned long _nodepda) ibcr.ii_ibcr_fld_s.i_soft_reset = 1; REMOTE_HUB_S(nasid, IIO_IBCR, ibcr.ii_ibcr_regval); + del_timer(recovery_timer); +} + +/* + * Wait until all BTE related CRBs are completed + * and then reset the interfaces. + */ +void bte_error_handler(unsigned long _nodepda) +{ + struct nodepda_s *err_nodepda = (struct nodepda_s *)_nodepda; + spinlock_t *recovery_lock = &err_nodepda->bte_recovery_lock; + int i; + nasid_t nasid; + unsigned long irq_flags; + volatile u64 *notify; + bte_result_t bh_error; + + BTE_PRINTK(("bte_error_handler(%p) - %d\n", err_nodepda, + smp_processor_id())); + + spin_lock_irqsave(recovery_lock, irq_flags); + + /* + * Lock all interfaces on this node to prevent new transfers + * from being queued. + */ + for (i = 0; i < BTES_PER_NODE; i++) { + if (err_nodepda->bte_if[i].cleanup_active) { + continue; + } + spin_lock(&err_nodepda->bte_if[i].spinlock); + BTE_PRINTK(("eh:%p:%d locked %d\n", err_nodepda, + smp_processor_id(), i)); + err_nodepda->bte_if[i].cleanup_active = 1; + } + + if (is_shub1()) { + shub1_bte_error_handler(_nodepda); + } else { + nasid = cnodeid_to_nasid(err_nodepda->bte_if[0].bte_cnode); + + if (ia64_sn_bte_recovery(nasid)) + panic("bte_error_handler(): Fatal BTE Error"); + } + for (i = 0; i < BTES_PER_NODE; i++) { bh_error = err_nodepda->bte_if[i].bh_error; if (bh_error != BTE_SUCCESS) { @@ -165,8 +187,6 @@ void bte_error_handler(unsigned long _nodepda) spin_unlock(&err_nodepda->bte_if[i].spinlock); } - del_timer(recovery_timer); - spin_unlock_irqrestore(recovery_lock, irq_flags); } diff --git a/arch/ia64/sn/kernel/huberror.c b/arch/ia64/sn/kernel/huberror.c index 2bdf684c506..5c39b43ba3c 100644 --- a/arch/ia64/sn/kernel/huberror.c +++ b/arch/ia64/sn/kernel/huberror.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1992 - 1997, 2000,2002-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1992 - 1997, 2000,2002-2005 Silicon Graphics, Inc. All rights reserved. */ #include <linux/types.h> @@ -38,8 +38,11 @@ static irqreturn_t hub_eint_handler(int irq, void *arg, struct pt_regs *ep) if ((int)ret_stuff.v0) panic("hubii_eint_handler(): Fatal TIO Error"); - if (!(nasid & 1)) /* Not a TIO, handle CRB errors */ - (void)hubiio_crb_error_handler(hubdev_info); + if (is_shub1()) { + if (!(nasid & 1)) /* Not a TIO, handle CRB errors */ + (void)hubiio_crb_error_handler(hubdev_info); + } else + bte_error_handler((unsigned long)NODEPDA(nasid_to_cnodeid(nasid))); return IRQ_HANDLED; } diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index 001880812b7..9e07f5463f2 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -11,14 +11,15 @@ #include <asm/sn/types.h> #include <asm/sn/sn_sal.h> #include <asm/sn/addrs.h> -#include "pci/pcibus_provider_defs.h" -#include "pci/pcidev.h" +#include <asm/sn/pcibus_provider_defs.h> +#include <asm/sn/pcidev.h> #include "pci/pcibr_provider.h" #include "xtalk/xwidgetdev.h" #include <asm/sn/geo.h> #include "xtalk/hubdev.h" #include <asm/sn/io.h> #include <asm/sn/simulator.h> +#include <asm/sn/tioca_provider.h> char master_baseio_wid; nasid_t master_nasid = INVALID_NASID; /* Partition Master */ @@ -34,6 +35,37 @@ struct brick { int sn_ioif_inited = 0; /* SN I/O infrastructure initialized? */ +struct sn_pcibus_provider *sn_pci_provider[PCIIO_ASIC_MAX_TYPES]; /* indexed by asic type */ + +/* + * Hooks and struct for unsupported pci providers + */ + +static dma_addr_t +sn_default_pci_map(struct pci_dev *pdev, unsigned long paddr, size_t size) +{ + return 0; +} + +static void +sn_default_pci_unmap(struct pci_dev *pdev, dma_addr_t addr, int direction) +{ + return; +} + +static void * +sn_default_pci_bus_fixup(struct pcibus_bussoft *soft) +{ + return NULL; +} + +static struct sn_pcibus_provider sn_pci_default_provider = { + .dma_map = sn_default_pci_map, + .dma_map_consistent = sn_default_pci_map, + .dma_unmap = sn_default_pci_unmap, + .bus_fixup = sn_default_pci_bus_fixup, +}; + /* * Retrieve the DMA Flush List given nasid. This list is needed * to implement the WAR - Flush DMA data on PIO Reads. @@ -142,6 +174,12 @@ static void sn_fixup_ionodes(void) if (status) continue; + /* Attach the error interrupt handlers */ + if (nasid & 1) + ice_error_init(hubdev); + else + hub_error_init(hubdev); + for (widget = 0; widget <= HUB_WIDGET_ID_MAX; widget++) hubdev->hdi_xwidget_info[widget].xwi_hubinfo = hubdev; @@ -179,10 +217,6 @@ static void sn_fixup_ionodes(void) sn_flush_device_list; } - if (!(i & 1)) - hub_error_init(hubdev); - else - ice_error_init(hubdev); } } @@ -201,6 +235,7 @@ static void sn_pci_fixup_slot(struct pci_dev *dev) struct sn_irq_info *sn_irq_info; struct pci_dev *host_pci_dev; int status = 0; + struct pcibus_bussoft *bs; dev->sysdata = kmalloc(sizeof(struct pcidev_info), GFP_KERNEL); if (SN_PCIDEV_INFO(dev) <= 0) @@ -241,6 +276,7 @@ static void sn_pci_fixup_slot(struct pci_dev *dev) } /* set up host bus linkages */ + bs = SN_PCIBUS_BUSSOFT(dev->bus); host_pci_dev = pci_find_slot(SN_PCIDEV_INFO(dev)->pdi_slot_host_handle >> 32, SN_PCIDEV_INFO(dev)-> @@ -248,10 +284,16 @@ static void sn_pci_fixup_slot(struct pci_dev *dev) SN_PCIDEV_INFO(dev)->pdi_host_pcidev_info = SN_PCIDEV_INFO(host_pci_dev); SN_PCIDEV_INFO(dev)->pdi_linux_pcidev = dev; - SN_PCIDEV_INFO(dev)->pdi_pcibus_info = SN_PCIBUS_BUSSOFT(dev->bus); + SN_PCIDEV_INFO(dev)->pdi_pcibus_info = bs; + + if (bs && bs->bs_asic_type < PCIIO_ASIC_MAX_TYPES) { + SN_PCIDEV_BUSPROVIDER(dev) = sn_pci_provider[bs->bs_asic_type]; + } else { + SN_PCIDEV_BUSPROVIDER(dev) = &sn_pci_default_provider; + } /* Only set up IRQ stuff if this device has a host bus context */ - if (SN_PCIDEV_BUSSOFT(dev) && sn_irq_info->irq_irq) { + if (bs && sn_irq_info->irq_irq) { SN_PCIDEV_INFO(dev)->pdi_sn_irq_info = sn_irq_info; dev->irq = SN_PCIDEV_INFO(dev)->pdi_sn_irq_info->irq_irq; sn_irq_fixup(dev, sn_irq_info); @@ -271,6 +313,7 @@ static void sn_pci_controller_fixup(int segment, int busnum) struct pcibus_bussoft *prom_bussoft_ptr; struct hubdev_info *hubdev_info; void *provider_soft; + struct sn_pcibus_provider *provider; status = sal_get_pcibus_info((u64) segment, (u64) busnum, @@ -291,16 +334,22 @@ static void sn_pci_controller_fixup(int segment, int busnum) /* * Per-provider fixup. Copies the contents from prom to local * area and links SN_PCIBUS_BUSSOFT(). - * - * Note: Provider is responsible for ensuring that prom_bussoft_ptr - * represents an asic-type that it can handle. */ - if (prom_bussoft_ptr->bs_asic_type == PCIIO_ASIC_TYPE_PPB) { - return; /* no further fixup necessary */ + if (prom_bussoft_ptr->bs_asic_type >= PCIIO_ASIC_MAX_TYPES) { + return; /* unsupported asic type */ + } + + provider = sn_pci_provider[prom_bussoft_ptr->bs_asic_type]; + if (provider == NULL) { + return; /* no provider registerd for this asic */ + } + + provider_soft = NULL; + if (provider->bus_fixup) { + provider_soft = (*provider->bus_fixup) (prom_bussoft_ptr); } - provider_soft = pcibr_bus_fixup(prom_bussoft_ptr); if (provider_soft == NULL) { return; /* fixup failed or not applicable */ } @@ -339,6 +388,17 @@ static int __init sn_pci_init(void) return 0; /* + * prime sn_pci_provider[]. Individial provider init routines will + * override their respective default entries. + */ + + for (i = 0; i < PCIIO_ASIC_MAX_TYPES; i++) + sn_pci_provider[i] = &sn_pci_default_provider; + + pcibr_init_provider(); + tioca_init_provider(); + + /* * This is needed to avoid bounce limit checks in the blk layer */ ia64_max_iommu_merge_mask = ~PAGE_MASK; diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c index 3be44724f6c..0f4e8138658 100644 --- a/arch/ia64/sn/kernel/irq.c +++ b/arch/ia64/sn/kernel/irq.c @@ -13,8 +13,8 @@ #include <asm/sn/addrs.h> #include <asm/sn/arch.h> #include "xtalk/xwidgetdev.h" -#include "pci/pcibus_provider_defs.h" -#include "pci/pcidev.h" +#include <asm/sn/pcibus_provider_defs.h> +#include <asm/sn/pcidev.h> #include "pci/pcibr_provider.h" #include <asm/sn/shub_mmr.h> #include <asm/sn/sn_sal.h> @@ -82,20 +82,9 @@ static void sn_ack_irq(unsigned int irq) nasid = get_nasid(); event_occurred = HUB_L((uint64_t *) GLOBAL_MMR_ADDR(nasid, SH_EVENT_OCCURRED)); - if (event_occurred & SH_EVENT_OCCURRED_UART_INT_MASK) { - mask |= (1 << SH_EVENT_OCCURRED_UART_INT_SHFT); - } - if (event_occurred & SH_EVENT_OCCURRED_IPI_INT_MASK) { - mask |= (1 << SH_EVENT_OCCURRED_IPI_INT_SHFT); - } - if (event_occurred & SH_EVENT_OCCURRED_II_INT0_MASK) { - mask |= (1 << SH_EVENT_OCCURRED_II_INT0_SHFT); - } - if (event_occurred & SH_EVENT_OCCURRED_II_INT1_MASK) { - mask |= (1 << SH_EVENT_OCCURRED_II_INT1_SHFT); - } + mask = event_occurred & SH_ALL_INT_MASK; HUB_S((uint64_t *) GLOBAL_MMR_ADDR(nasid, SH_EVENT_OCCURRED_ALIAS), - mask); + mask); __set_bit(irq, (volatile void *)pda->sn_in_service_ivecs); move_irq(irq); diff --git a/arch/ia64/sn/kernel/mca.c b/arch/ia64/sn/kernel/mca.c index 857774bb2c9..6546db6abdb 100644 --- a/arch/ia64/sn/kernel/mca.c +++ b/arch/ia64/sn/kernel/mca.c @@ -37,6 +37,11 @@ static u64 *sn_oemdata_size, sn_oemdata_bufsize; * This function is the callback routine that SAL calls to log error * info for platform errors. buf is appended to sn_oemdata, resizing as * required. + * Note: this is a SAL to OS callback, running under the same rules as the SAL + * code. SAL calls are run with preempt disabled so this routine must not + * sleep. vmalloc can sleep so print_hook cannot resize the output buffer + * itself, instead it must set the required size and return to let the caller + * resize the buffer then redrive the SAL call. */ static int print_hook(const char *fmt, ...) { @@ -47,18 +52,8 @@ static int print_hook(const char *fmt, ...) vsnprintf(buf, sizeof(buf), fmt, args); va_end(args); len = strlen(buf); - while (*sn_oemdata_size + len + 1 > sn_oemdata_bufsize) { - u8 *newbuf = vmalloc(sn_oemdata_bufsize += 1000); - if (!newbuf) { - printk(KERN_ERR "%s: unable to extend sn_oemdata\n", - __FUNCTION__); - return 0; - } - memcpy(newbuf, *sn_oemdata, *sn_oemdata_size); - vfree(*sn_oemdata); - *sn_oemdata = newbuf; - } - memcpy(*sn_oemdata + *sn_oemdata_size, buf, len + 1); + if (*sn_oemdata_size + len <= sn_oemdata_bufsize) + memcpy(*sn_oemdata + *sn_oemdata_size, buf, len); *sn_oemdata_size += len; return 0; } @@ -98,7 +93,20 @@ sn_platform_plat_specific_err_print(const u8 * sect_header, u8 ** oemdata, sn_oemdata = oemdata; sn_oemdata_size = oemdata_size; sn_oemdata_bufsize = 0; - ia64_sn_plat_specific_err_print(print_hook, (char *)sect_header); + *sn_oemdata_size = PAGE_SIZE; /* first guess at how much data will be generated */ + while (*sn_oemdata_size > sn_oemdata_bufsize) { + u8 *newbuf = vmalloc(*sn_oemdata_size); + if (!newbuf) { + printk(KERN_ERR "%s: unable to extend sn_oemdata\n", + __FUNCTION__); + return 1; + } + vfree(*sn_oemdata); + *sn_oemdata = newbuf; + sn_oemdata_bufsize = *sn_oemdata_size; + *sn_oemdata_size = 0; + ia64_sn_plat_specific_err_print(print_hook, (char *)sect_header); + } up(&sn_oemdata_mutex); return 0; } diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index f0306b516af..4fb44984afe 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1999,2001-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1999,2001-2005 Silicon Graphics, Inc. All rights reserved. */ #include <linux/config.h> @@ -29,6 +29,7 @@ #include <linux/sched.h> #include <linux/root_dev.h> #include <linux/nodemask.h> +#include <linux/pm.h> #include <asm/io.h> #include <asm/sal.h> @@ -72,6 +73,12 @@ EXPORT_SYMBOL(sn_rtc_cycles_per_second); DEFINE_PER_CPU(struct sn_hub_info_s, __sn_hub_info); EXPORT_PER_CPU_SYMBOL(__sn_hub_info); +DEFINE_PER_CPU(short, __sn_cnodeid_to_nasid[MAX_NUMNODES]); +EXPORT_PER_CPU_SYMBOL(__sn_cnodeid_to_nasid); + +DEFINE_PER_CPU(struct nodepda_s *, __sn_nodepda); +EXPORT_PER_CPU_SYMBOL(__sn_nodepda); + partid_t sn_partid = -1; EXPORT_SYMBOL(sn_partid); char sn_system_serial_number_string[128]; @@ -353,6 +360,14 @@ void __init sn_setup(char **cmdline_p) screen_info = sn_screen_info; sn_timer_init(); + + /* + * set pm_power_off to a SAL call to allow + * sn machines to power off. The SAL call can be replaced + * by an ACPI interface call when ACPI is fully implemented + * for sn. + */ + pm_power_off = ia64_sn_power_down; } /** @@ -364,11 +379,11 @@ static void __init sn_init_pdas(char **cmdline_p) { cnodeid_t cnode; - memset(pda->cnodeid_to_nasid_table, -1, - sizeof(pda->cnodeid_to_nasid_table)); + memset(sn_cnodeid_to_nasid, -1, + sizeof(__ia64_per_cpu_var(__sn_cnodeid_to_nasid))); for_each_online_node(cnode) - pda->cnodeid_to_nasid_table[cnode] = - pxm_to_nasid(nid_to_pxm_map[cnode]); + sn_cnodeid_to_nasid[cnode] = + pxm_to_nasid(nid_to_pxm_map[cnode]); numionodes = num_online_nodes(); scan_for_ionodes(); @@ -468,7 +483,8 @@ void __init sn_cpu_init(void) cnode = nasid_to_cnodeid(nasid); - pda->p_nodepda = nodepdaindr[cnode]; + sn_nodepda = nodepdaindr[cnode]; + pda->led_address = (typeof(pda->led_address)) (LED0 + (slice << LED_CPU_SHIFT)); pda->led_state = LED_ALWAYS_SET; @@ -477,15 +493,18 @@ void __init sn_cpu_init(void) pda->idle_flag = 0; if (cpuid != 0) { - memcpy(pda->cnodeid_to_nasid_table, - pdacpu(0)->cnodeid_to_nasid_table, - sizeof(pda->cnodeid_to_nasid_table)); + /* copy cpu 0's sn_cnodeid_to_nasid table to this cpu's */ + memcpy(sn_cnodeid_to_nasid, + (&per_cpu(__sn_cnodeid_to_nasid, 0)), + sizeof(__ia64_per_cpu_var(__sn_cnodeid_to_nasid))); } /* * Check for WARs. * Only needs to be done once, on BSP. - * Has to be done after loop above, because it uses pda.cnodeid_to_nasid_table[i]. + * Has to be done after loop above, because it uses this cpu's + * sn_cnodeid_to_nasid table which was just initialized if this + * isn't cpu 0. * Has to be done before assignment below. */ if (!wars_have_been_checked) { @@ -571,8 +590,7 @@ static void __init scan_for_ionodes(void) brd = find_lboard_any(brd, KLTYPE_SNIA); while (brd) { - pda->cnodeid_to_nasid_table[numionodes] = - brd->brd_nasid; + sn_cnodeid_to_nasid[numionodes] = brd->brd_nasid; physical_node_map[brd->brd_nasid] = numionodes; root_lboard[numionodes] = brd; numionodes++; @@ -593,8 +611,7 @@ static void __init scan_for_ionodes(void) root_lboard[nasid_to_cnodeid(nasid)], KLTYPE_TIO); while (brd) { - pda->cnodeid_to_nasid_table[numionodes] = - brd->brd_nasid; + sn_cnodeid_to_nasid[numionodes] = brd->brd_nasid; physical_node_map[brd->brd_nasid] = numionodes; root_lboard[numionodes] = brd; numionodes++; @@ -605,7 +622,6 @@ static void __init scan_for_ionodes(void) brd = find_lboard_any(brd, KLTYPE_TIO); } } - } int @@ -614,7 +630,8 @@ nasid_slice_to_cpuid(int nasid, int slice) long cpu; for (cpu=0; cpu < NR_CPUS; cpu++) - if (nodepda->phys_cpuid[cpu].nasid == nasid && nodepda->phys_cpuid[cpu].slice == slice) + if (cpuid_to_nasid(cpu) == nasid && + cpuid_to_slice(cpu) == slice) return cpu; return -1; diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c index 197356460ee..833e700fdac 100644 --- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c +++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c @@ -28,6 +28,7 @@ #include <linux/vmalloc.h> #include <linux/seq_file.h> #include <linux/miscdevice.h> +#include <linux/utsname.h> #include <linux/cpumask.h> #include <linux/smp_lock.h> #include <linux/nodemask.h> @@ -43,6 +44,7 @@ #include <asm/sn/module.h> #include <asm/sn/geo.h> #include <asm/sn/sn2/sn_hwperf.h> +#include <asm/sn/addrs.h> static void *sn_hwperf_salheap = NULL; static int sn_hwperf_obj_cnt = 0; @@ -81,26 +83,45 @@ out: return e; } +static int sn_hwperf_location_to_bpos(char *location, + int *rack, int *bay, int *slot, int *slab) +{ + char type; + + /* first scan for an old style geoid string */ + if (sscanf(location, "%03d%c%02d#%d", + rack, &type, bay, slab) == 4) + *slot = 0; + else /* scan for a new bladed geoid string */ + if (sscanf(location, "%03d%c%02d^%02d#%d", + rack, &type, bay, slot, slab) != 5) + return -1; + /* success */ + return 0; +} + static int sn_hwperf_geoid_to_cnode(char *location) { int cnode; geoid_t geoid; moduleid_t module_id; - char type; - int rack, slot, slab; - int this_rack, this_slot, this_slab; + int rack, bay, slot, slab; + int this_rack, this_bay, this_slot, this_slab; - if (sscanf(location, "%03d%c%02d#%d", &rack, &type, &slot, &slab) != 4) + if (sn_hwperf_location_to_bpos(location, &rack, &bay, &slot, &slab)) return -1; for (cnode = 0; cnode < numionodes; cnode++) { geoid = cnodeid_get_geoid(cnode); module_id = geo_module(geoid); this_rack = MODULE_GET_RACK(module_id); - this_slot = MODULE_GET_BPOS(module_id); + this_bay = MODULE_GET_BPOS(module_id); + this_slot = geo_slot(geoid); this_slab = geo_slab(geoid); - if (rack == this_rack && slot == this_slot && slab == this_slab) + if (rack == this_rack && bay == this_bay && + slot == this_slot && slab == this_slab) { break; + } } return cnode < numionodes ? cnode : -1; @@ -153,11 +174,36 @@ static const char *sn_hwperf_get_slabname(struct sn_hwperf_object_info *obj, return slabname; } +static void print_pci_topology(struct seq_file *s, + struct sn_hwperf_object_info *obj, int *ordinal, + u64 rack, u64 bay, u64 slot, u64 slab) +{ + char *p1; + char *p2; + char *pg; + + if (!(pg = (char *)get_zeroed_page(GFP_KERNEL))) + return; /* ignore */ + if (ia64_sn_ioif_get_pci_topology(rack, bay, slot, slab, + __pa(pg), PAGE_SIZE) == SN_HWPERF_OP_OK) { + for (p1=pg; *p1 && p1 < pg + PAGE_SIZE;) { + if (!(p2 = strchr(p1, '\n'))) + break; + *p2 = '\0'; + seq_printf(s, "pcibus %d %s-%s\n", + *ordinal, obj->location, p1); + (*ordinal)++; + p1 = p2 + 1; + } + } + free_page((unsigned long)pg); +} + static int sn_topology_show(struct seq_file *s, void *d) { int sz; int pt; - int e; + int e = 0; int i; int j; const char *slabname; @@ -169,11 +215,44 @@ static int sn_topology_show(struct seq_file *s, void *d) struct sn_hwperf_object_info *p; struct sn_hwperf_object_info *obj = d; /* this object */ struct sn_hwperf_object_info *objs = s->private; /* all objects */ + int rack, bay, slot, slab; + u8 shubtype; + u8 system_size; + u8 sharing_size; + u8 partid; + u8 coher; + u8 nasid_shift; + u8 region_size; + u16 nasid_mask; + int nasid_msb; + int pci_bus_ordinal = 0; if (obj == objs) { - seq_printf(s, "# sn_topology version 1\n"); + seq_printf(s, "# sn_topology version 2\n"); seq_printf(s, "# objtype ordinal location partition" " [attribute value [, ...]]\n"); + + if (ia64_sn_get_sn_info(0, + &shubtype, &nasid_mask, &nasid_shift, &system_size, + &sharing_size, &partid, &coher, ®ion_size)) + BUG(); + for (nasid_msb=63; nasid_msb > 0; nasid_msb--) { + if (((u64)nasid_mask << nasid_shift) & (1ULL << nasid_msb)) + break; + } + seq_printf(s, "partition %u %s local " + "shubtype %s, " + "nasid_mask 0x%016lx, " + "nasid_bits %d:%d, " + "system_size %d, " + "sharing_size %d, " + "coherency_domain %d, " + "region_size %d\n", + + partid, system_utsname.nodename, + shubtype ? "shub2" : "shub1", + (u64)nasid_mask << nasid_shift, nasid_msb, nasid_shift, + system_size, sharing_size, coher, region_size); } if (SN_HWPERF_FOREIGN(obj)) { @@ -181,7 +260,7 @@ static int sn_topology_show(struct seq_file *s, void *d) return 0; } - for (i = 0; obj->name[i]; i++) { + for (i = 0; i < SN_HWPERF_MAXSTRING && obj->name[i]; i++) { if (obj->name[i] == ' ') obj->name[i] = '_'; } @@ -221,6 +300,17 @@ static int sn_topology_show(struct seq_file *s, void *d) seq_putc(s, '\n'); } } + + /* + * PCI busses attached to this node, if any + */ + if (sn_hwperf_location_to_bpos(obj->location, + &rack, &bay, &slot, &slab)) { + /* export pci bus info */ + print_pci_topology(s, obj, &pci_bus_ordinal, + rack, bay, slot, slab); + + } } if (obj->ports) { @@ -397,6 +487,9 @@ static int sn_hwperf_map_err(int hwperf_err) break; case SN_HWPERF_OP_BUSY: + e = -EBUSY; + break; + case SN_HWPERF_OP_RECONFIGURE: e = -EAGAIN; break; @@ -549,6 +642,7 @@ sn_hwperf_ioctl(struct inode *in, struct file *fp, u32 op, u64 arg) r = sn_hwperf_op_cpu(&op_info); if (r) { r = sn_hwperf_map_err(r); + a.v0 = v0; goto error; } break; diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c new file mode 100644 index 00000000000..ab9b5f35c2a --- /dev/null +++ b/arch/ia64/sn/kernel/tiocx.c @@ -0,0 +1,552 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2005 Silicon Graphics, Inc. All rights reserved. + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/version.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/proc_fs.h> +#include <linux/device.h> +#include <linux/delay.h> +#include <asm/uaccess.h> +#include <asm/sn/sn_sal.h> +#include <asm/sn/addrs.h> +#include <asm/sn/io.h> +#include <asm/sn/types.h> +#include <asm/sn/shubio.h> +#include <asm/sn/tiocx.h> +#include <asm/sn/l1.h> +#include <asm/sn/module.h> +#include "tio.h" +#include "xtalk/xwidgetdev.h" +#include "xtalk/hubdev.h" + +#define CX_DEV_NONE 0 +#define DEVICE_NAME "tiocx" +#define WIDGET_ID 0 +#define TIOCX_DEBUG 0 + +#if TIOCX_DEBUG +#define DBG(fmt...) printk(KERN_ALERT fmt) +#else +#define DBG(fmt...) +#endif + +struct device_attribute dev_attr_cxdev_control; + +/** + * tiocx_match - Try to match driver id list with device. + * @dev: device pointer + * @drv: driver pointer + * + * Returns 1 if match, 0 otherwise. + */ +static int tiocx_match(struct device *dev, struct device_driver *drv) +{ + struct cx_dev *cx_dev = to_cx_dev(dev); + struct cx_drv *cx_drv = to_cx_driver(drv); + const struct cx_device_id *ids = cx_drv->id_table; + + if (!ids) + return 0; + + while (ids->part_num) { + if (ids->part_num == cx_dev->cx_id.part_num) + return 1; + ids++; + } + return 0; + +} + +static int tiocx_hotplug(struct device *dev, char **envp, int num_envp, + char *buffer, int buffer_size) +{ + return -ENODEV; +} + +static void tiocx_bus_release(struct device *dev) +{ + kfree(to_cx_dev(dev)); +} + +struct bus_type tiocx_bus_type = { + .name = "tiocx", + .match = tiocx_match, + .hotplug = tiocx_hotplug, +}; + +/** + * cx_device_match - Find cx_device in the id table. + * @ids: id table from driver + * @cx_device: part/mfg id for the device + * + */ +static const struct cx_device_id *cx_device_match(const struct cx_device_id + *ids, + struct cx_dev *cx_device) +{ + /* + * NOTES: We may want to check for CX_ANY_ID too. + * Do we want to match against nasid too? + * CX_DEV_NONE == 0, if the driver tries to register for + * part/mfg == 0 we should return no-match (NULL) here. + */ + while (ids->part_num && ids->mfg_num) { + if (ids->part_num == cx_device->cx_id.part_num && + ids->mfg_num == cx_device->cx_id.mfg_num) + return ids; + ids++; + } + + return NULL; +} + +/** + * cx_device_probe - Look for matching device. + * Call driver probe routine if found. + * @cx_driver: driver table (cx_drv struct) from driver + * @cx_device: part/mfg id for the device + */ +static int cx_device_probe(struct device *dev) +{ + const struct cx_device_id *id; + struct cx_drv *cx_drv = to_cx_driver(dev->driver); + struct cx_dev *cx_dev = to_cx_dev(dev); + int error = 0; + + if (!cx_dev->driver && cx_drv->probe) { + id = cx_device_match(cx_drv->id_table, cx_dev); + if (id) { + if ((error = cx_drv->probe(cx_dev, id)) < 0) + return error; + else + cx_dev->driver = cx_drv; + } + } + + return error; +} + +/** + * cx_driver_remove - Remove driver from device struct. + * @dev: device + */ +static int cx_driver_remove(struct device *dev) +{ + struct cx_dev *cx_dev = to_cx_dev(dev); + struct cx_drv *cx_drv = cx_dev->driver; + if (cx_drv->remove) + cx_drv->remove(cx_dev); + cx_dev->driver = NULL; + return 0; +} + +/** + * cx_driver_register - Register the driver. + * @cx_driver: driver table (cx_drv struct) from driver + * + * Called from the driver init routine to register a driver. + * The cx_drv struct contains the driver name, a pointer to + * a table of part/mfg numbers and a pointer to the driver's + * probe/attach routine. + */ +int cx_driver_register(struct cx_drv *cx_driver) +{ + cx_driver->driver.name = cx_driver->name; + cx_driver->driver.bus = &tiocx_bus_type; + cx_driver->driver.probe = cx_device_probe; + cx_driver->driver.remove = cx_driver_remove; + + return driver_register(&cx_driver->driver); +} + +/** + * cx_driver_unregister - Unregister the driver. + * @cx_driver: driver table (cx_drv struct) from driver + */ +int cx_driver_unregister(struct cx_drv *cx_driver) +{ + driver_unregister(&cx_driver->driver); + return 0; +} + +/** + * cx_device_register - Register a device. + * @nasid: device's nasid + * @part_num: device's part number + * @mfg_num: device's manufacturer number + * @hubdev: hub info associated with this device + * + */ +int +cx_device_register(nasid_t nasid, int part_num, int mfg_num, + struct hubdev_info *hubdev) +{ + struct cx_dev *cx_dev; + + cx_dev = kcalloc(1, sizeof(struct cx_dev), GFP_KERNEL); + DBG("cx_dev= 0x%p\n", cx_dev); + if (cx_dev == NULL) + return -ENOMEM; + + cx_dev->cx_id.part_num = part_num; + cx_dev->cx_id.mfg_num = mfg_num; + cx_dev->cx_id.nasid = nasid; + cx_dev->hubdev = hubdev; + + cx_dev->dev.parent = NULL; + cx_dev->dev.bus = &tiocx_bus_type; + cx_dev->dev.release = tiocx_bus_release; + snprintf(cx_dev->dev.bus_id, BUS_ID_SIZE, "%d.0x%x", + cx_dev->cx_id.nasid, cx_dev->cx_id.part_num); + device_register(&cx_dev->dev); + get_device(&cx_dev->dev); + + device_create_file(&cx_dev->dev, &dev_attr_cxdev_control); + + return 0; +} + +/** + * cx_device_unregister - Unregister a device. + * @cx_dev: part/mfg id for the device + */ +int cx_device_unregister(struct cx_dev *cx_dev) +{ + put_device(&cx_dev->dev); + device_unregister(&cx_dev->dev); + return 0; +} + +/** + * cx_device_reload - Reload the device. + * @nasid: device's nasid + * @part_num: device's part number + * @mfg_num: device's manufacturer number + * + * Remove the device associated with 'nasid' from device list and then + * call device-register with the given part/mfg numbers. + */ +static int cx_device_reload(struct cx_dev *cx_dev) +{ + device_remove_file(&cx_dev->dev, &dev_attr_cxdev_control); + cx_device_unregister(cx_dev); + return cx_device_register(cx_dev->cx_id.nasid, cx_dev->cx_id.part_num, + cx_dev->cx_id.mfg_num, cx_dev->hubdev); +} + +static inline uint64_t tiocx_intr_alloc(nasid_t nasid, int widget, + u64 sn_irq_info, + int req_irq, nasid_t req_nasid, + int req_slice) +{ + struct ia64_sal_retval rv; + rv.status = 0; + rv.v0 = 0; + + ia64_sal_oemcall_nolock(&rv, SN_SAL_IOIF_INTERRUPT, + SAL_INTR_ALLOC, nasid, + widget, sn_irq_info, req_irq, + req_nasid, req_slice); + return rv.status; +} + +static inline void tiocx_intr_free(nasid_t nasid, int widget, + struct sn_irq_info *sn_irq_info) +{ + struct ia64_sal_retval rv; + rv.status = 0; + rv.v0 = 0; + + ia64_sal_oemcall_nolock(&rv, SN_SAL_IOIF_INTERRUPT, + SAL_INTR_FREE, nasid, + widget, sn_irq_info->irq_irq, + sn_irq_info->irq_cookie, 0, 0); +} + +struct sn_irq_info *tiocx_irq_alloc(nasid_t nasid, int widget, int irq, + nasid_t req_nasid, int slice) +{ + struct sn_irq_info *sn_irq_info; + int status; + int sn_irq_size = sizeof(struct sn_irq_info); + + if ((nasid & 1) == 0) + return NULL; + + sn_irq_info = kmalloc(sn_irq_size, GFP_KERNEL); + if (sn_irq_info == NULL) + return NULL; + + memset(sn_irq_info, 0x0, sn_irq_size); + + status = tiocx_intr_alloc(nasid, widget, __pa(sn_irq_info), irq, + req_nasid, slice); + if (status) { + kfree(sn_irq_info); + return NULL; + } else { + return sn_irq_info; + } +} + +void tiocx_irq_free(struct sn_irq_info *sn_irq_info) +{ + uint64_t bridge = (uint64_t) sn_irq_info->irq_bridge; + nasid_t nasid = NASID_GET(bridge); + int widget; + + if (nasid & 1) { + widget = TIO_SWIN_WIDGETNUM(bridge); + tiocx_intr_free(nasid, widget, sn_irq_info); + kfree(sn_irq_info); + } +} + +uint64_t tiocx_dma_addr(uint64_t addr) +{ + return PHYS_TO_TIODMA(addr); +} + +uint64_t tiocx_swin_base(int nasid) +{ + return TIO_SWIN_BASE(nasid, TIOCX_CORELET); +} + +EXPORT_SYMBOL(cx_driver_register); +EXPORT_SYMBOL(cx_driver_unregister); +EXPORT_SYMBOL(cx_device_register); +EXPORT_SYMBOL(cx_device_unregister); +EXPORT_SYMBOL(tiocx_irq_alloc); +EXPORT_SYMBOL(tiocx_irq_free); +EXPORT_SYMBOL(tiocx_bus_type); +EXPORT_SYMBOL(tiocx_dma_addr); +EXPORT_SYMBOL(tiocx_swin_base); + +static void tio_conveyor_set(nasid_t nasid, int enable_flag) +{ + uint64_t ice_frz; + uint64_t disable_cb = (1ull << 61); + + if (!(nasid & 1)) + return; + + ice_frz = REMOTE_HUB_L(nasid, TIO_ICE_FRZ_CFG); + if (enable_flag) { + if (!(ice_frz & disable_cb)) /* already enabled */ + return; + ice_frz &= ~disable_cb; + } else { + if (ice_frz & disable_cb) /* already disabled */ + return; + ice_frz |= disable_cb; + } + DBG(KERN_ALERT "TIO_ICE_FRZ_CFG= 0x%lx\n", ice_frz); + REMOTE_HUB_S(nasid, TIO_ICE_FRZ_CFG, ice_frz); +} + +#define tio_conveyor_enable(nasid) tio_conveyor_set(nasid, 1) +#define tio_conveyor_disable(nasid) tio_conveyor_set(nasid, 0) + +static void tio_corelet_reset(nasid_t nasid, int corelet) +{ + if (!(nasid & 1)) + return; + + REMOTE_HUB_S(nasid, TIO_ICE_PMI_TX_CFG, 1 << corelet); + udelay(2000); + REMOTE_HUB_S(nasid, TIO_ICE_PMI_TX_CFG, 0); + udelay(2000); +} + +static int tiocx_btchar_get(int nasid) +{ + moduleid_t module_id; + geoid_t geoid; + int cnodeid; + + cnodeid = nasid_to_cnodeid(nasid); + geoid = cnodeid_get_geoid(cnodeid); + module_id = geo_module(geoid); + return MODULE_GET_BTCHAR(module_id); +} + +static int is_fpga_brick(int nasid) +{ + switch (tiocx_btchar_get(nasid)) { + case L1_BRICKTYPE_SA: + case L1_BRICKTYPE_ATHENA: + return 1; + } + return 0; +} + +static int bitstream_loaded(nasid_t nasid) +{ + uint64_t cx_credits; + + cx_credits = REMOTE_HUB_L(nasid, TIO_ICE_PMI_TX_DYN_CREDIT_STAT_CB3); + cx_credits &= TIO_ICE_PMI_TX_DYN_CREDIT_STAT_CB3_CREDIT_CNT_MASK; + DBG("cx_credits= 0x%lx\n", cx_credits); + + return (cx_credits == 0xf) ? 1 : 0; +} + +static int tiocx_reload(struct cx_dev *cx_dev) +{ + int part_num = CX_DEV_NONE; + int mfg_num = CX_DEV_NONE; + nasid_t nasid = cx_dev->cx_id.nasid; + + if (bitstream_loaded(nasid)) { + uint64_t cx_id; + + cx_id = + *(volatile int32_t *)(TIO_SWIN_BASE(nasid, TIOCX_CORELET) + + WIDGET_ID); + part_num = XWIDGET_PART_NUM(cx_id); + mfg_num = XWIDGET_MFG_NUM(cx_id); + DBG("part= 0x%x, mfg= 0x%x\n", part_num, mfg_num); + /* just ignore it if it's a CE */ + if (part_num == TIO_CE_ASIC_PARTNUM) + return 0; + } + + cx_dev->cx_id.part_num = part_num; + cx_dev->cx_id.mfg_num = mfg_num; + + /* + * Delete old device and register the new one. It's ok if + * part_num/mfg_num == CX_DEV_NONE. We want to register + * devices in the table even if a bitstream isn't loaded. + * That allows use to see that a bitstream isn't loaded via + * TIOCX_IOCTL_DEV_LIST. + */ + return cx_device_reload(cx_dev); +} + +static ssize_t show_cxdev_control(struct device *dev, char *buf) +{ + struct cx_dev *cx_dev = to_cx_dev(dev); + + return sprintf(buf, "0x%x 0x%x 0x%x %d\n", + cx_dev->cx_id.nasid, + cx_dev->cx_id.part_num, cx_dev->cx_id.mfg_num, + tiocx_btchar_get(cx_dev->cx_id.nasid)); +} + +static ssize_t store_cxdev_control(struct device *dev, const char *buf, + size_t count) +{ + int n; + struct cx_dev *cx_dev = to_cx_dev(dev); + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (count <= 0) + return 0; + + n = simple_strtoul(buf, NULL, 0); + + switch (n) { + case 1: + tiocx_reload(cx_dev); + break; + case 3: + tio_corelet_reset(cx_dev->cx_id.nasid, TIOCX_CORELET); + break; + default: + break; + } + + return count; +} + +DEVICE_ATTR(cxdev_control, 0644, show_cxdev_control, store_cxdev_control); + +static int __init tiocx_init(void) +{ + cnodeid_t cnodeid; + int found_tiocx_device = 0; + + bus_register(&tiocx_bus_type); + + for (cnodeid = 0; cnodeid < MAX_COMPACT_NODES; cnodeid++) { + nasid_t nasid; + + if ((nasid = cnodeid_to_nasid(cnodeid)) < 0) + break; /* No more nasids .. bail out of loop */ + + if ((nasid & 0x1) && is_fpga_brick(nasid)) { + struct hubdev_info *hubdev; + struct xwidget_info *widgetp; + + DBG("Found TIO at nasid 0x%x\n", nasid); + + hubdev = + (struct hubdev_info *)(NODEPDA(cnodeid)->pdinfo); + + widgetp = &hubdev->hdi_xwidget_info[TIOCX_CORELET]; + + /* The CE hangs off of the CX port but is not an FPGA */ + if (widgetp->xwi_hwid.part_num == TIO_CE_ASIC_PARTNUM) + continue; + + tio_corelet_reset(nasid, TIOCX_CORELET); + tio_conveyor_enable(nasid); + + if (cx_device_register + (nasid, widgetp->xwi_hwid.part_num, + widgetp->xwi_hwid.mfg_num, hubdev) < 0) + return -ENXIO; + else + found_tiocx_device++; + } + } + + /* It's ok if we find zero devices. */ + DBG("found_tiocx_device= %d\n", found_tiocx_device); + + return 0; +} + +static void __exit tiocx_exit(void) +{ + struct device *dev; + struct device *tdev; + + DBG("tiocx_exit\n"); + + /* + * Unregister devices. + */ + list_for_each_entry_safe(dev, tdev, &tiocx_bus_type.devices.list, + bus_list) { + if (dev) { + struct cx_dev *cx_dev = to_cx_dev(dev); + device_remove_file(dev, &dev_attr_cxdev_control); + cx_device_unregister(cx_dev); + } + } + + bus_unregister(&tiocx_bus_type); +} + +module_init(tiocx_init); +module_exit(tiocx_exit); + +/************************************************************************ + * Module licensing and description + ************************************************************************/ +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Bruce Losure <blosure@sgi.com>"); +MODULE_DESCRIPTION("TIOCX module"); +MODULE_SUPPORTED_DEVICE(DEVICE_NAME); diff --git a/arch/ia64/sn/kernel/xp_main.c b/arch/ia64/sn/kernel/xp_main.c new file mode 100644 index 00000000000..3be52a34c80 --- /dev/null +++ b/arch/ia64/sn/kernel/xp_main.c @@ -0,0 +1,289 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + */ + + +/* + * Cross Partition (XP) base. + * + * XP provides a base from which its users can interact + * with XPC, yet not be dependent on XPC. + * + */ + + +#include <linux/kernel.h> +#include <linux/interrupt.h> +#include <linux/module.h> +#include <asm/sn/intr.h> +#include <asm/sn/sn_sal.h> +#include <asm/sn/xp.h> + + +/* + * Target of nofault PIO read. + */ +u64 xp_nofault_PIOR_target; + + +/* + * xpc_registrations[] keeps track of xpc_connect()'s done by the kernel-level + * users of XPC. + */ +struct xpc_registration xpc_registrations[XPC_NCHANNELS]; + + +/* + * Initialize the XPC interface to indicate that XPC isn't loaded. + */ +static enum xpc_retval xpc_notloaded(void) { return xpcNotLoaded; } + +struct xpc_interface xpc_interface = { + (void (*)(int)) xpc_notloaded, + (void (*)(int)) xpc_notloaded, + (enum xpc_retval (*)(partid_t, int, u32, void **)) xpc_notloaded, + (enum xpc_retval (*)(partid_t, int, void *)) xpc_notloaded, + (enum xpc_retval (*)(partid_t, int, void *, xpc_notify_func, void *)) + xpc_notloaded, + (void (*)(partid_t, int, void *)) xpc_notloaded, + (enum xpc_retval (*)(partid_t, void *)) xpc_notloaded +}; + + +/* + * XPC calls this when it (the XPC module) has been loaded. + */ +void +xpc_set_interface(void (*connect)(int), + void (*disconnect)(int), + enum xpc_retval (*allocate)(partid_t, int, u32, void **), + enum xpc_retval (*send)(partid_t, int, void *), + enum xpc_retval (*send_notify)(partid_t, int, void *, + xpc_notify_func, void *), + void (*received)(partid_t, int, void *), + enum xpc_retval (*partid_to_nasids)(partid_t, void *)) +{ + xpc_interface.connect = connect; + xpc_interface.disconnect = disconnect; + xpc_interface.allocate = allocate; + xpc_interface.send = send; + xpc_interface.send_notify = send_notify; + xpc_interface.received = received; + xpc_interface.partid_to_nasids = partid_to_nasids; +} + + +/* + * XPC calls this when it (the XPC module) is being unloaded. + */ +void +xpc_clear_interface(void) +{ + xpc_interface.connect = (void (*)(int)) xpc_notloaded; + xpc_interface.disconnect = (void (*)(int)) xpc_notloaded; + xpc_interface.allocate = (enum xpc_retval (*)(partid_t, int, u32, + void **)) xpc_notloaded; + xpc_interface.send = (enum xpc_retval (*)(partid_t, int, void *)) + xpc_notloaded; + xpc_interface.send_notify = (enum xpc_retval (*)(partid_t, int, void *, + xpc_notify_func, void *)) xpc_notloaded; + xpc_interface.received = (void (*)(partid_t, int, void *)) + xpc_notloaded; + xpc_interface.partid_to_nasids = (enum xpc_retval (*)(partid_t, void *)) + xpc_notloaded; +} + + +/* + * Register for automatic establishment of a channel connection whenever + * a partition comes up. + * + * Arguments: + * + * ch_number - channel # to register for connection. + * func - function to call for asynchronous notification of channel + * state changes (i.e., connection, disconnection, error) and + * the arrival of incoming messages. + * key - pointer to optional user-defined value that gets passed back + * to the user on any callouts made to func. + * payload_size - size in bytes of the XPC message's payload area which + * contains a user-defined message. The user should make + * this large enough to hold their largest message. + * nentries - max #of XPC message entries a message queue can contain. + * The actual number, which is determined when a connection + * is established and may be less then requested, will be + * passed to the user via the xpcConnected callout. + * assigned_limit - max number of kthreads allowed to be processing + * messages (per connection) at any given instant. + * idle_limit - max number of kthreads allowed to be idle at any given + * instant. + */ +enum xpc_retval +xpc_connect(int ch_number, xpc_channel_func func, void *key, u16 payload_size, + u16 nentries, u32 assigned_limit, u32 idle_limit) +{ + struct xpc_registration *registration; + + + DBUG_ON(ch_number < 0 || ch_number >= XPC_NCHANNELS); + DBUG_ON(payload_size == 0 || nentries == 0); + DBUG_ON(func == NULL); + DBUG_ON(assigned_limit == 0 || idle_limit > assigned_limit); + + registration = &xpc_registrations[ch_number]; + + if (down_interruptible(®istration->sema) != 0) { + return xpcInterrupted; + } + + /* if XPC_CHANNEL_REGISTERED(ch_number) */ + if (registration->func != NULL) { + up(®istration->sema); + return xpcAlreadyRegistered; + } + + /* register the channel for connection */ + registration->msg_size = XPC_MSG_SIZE(payload_size); + registration->nentries = nentries; + registration->assigned_limit = assigned_limit; + registration->idle_limit = idle_limit; + registration->key = key; + registration->func = func; + + up(®istration->sema); + + xpc_interface.connect(ch_number); + + return xpcSuccess; +} + + +/* + * Remove the registration for automatic connection of the specified channel + * when a partition comes up. + * + * Before returning this xpc_disconnect() will wait for all connections on the + * specified channel have been closed/torndown. So the caller can be assured + * that they will not be receiving any more callouts from XPC to their + * function registered via xpc_connect(). + * + * Arguments: + * + * ch_number - channel # to unregister. + */ +void +xpc_disconnect(int ch_number) +{ + struct xpc_registration *registration; + + + DBUG_ON(ch_number < 0 || ch_number >= XPC_NCHANNELS); + + registration = &xpc_registrations[ch_number]; + + /* + * We've decided not to make this a down_interruptible(), since we + * figured XPC's users will just turn around and call xpc_disconnect() + * again anyways, so we might as well wait, if need be. + */ + down(®istration->sema); + + /* if !XPC_CHANNEL_REGISTERED(ch_number) */ + if (registration->func == NULL) { + up(®istration->sema); + return; + } + + /* remove the connection registration for the specified channel */ + registration->func = NULL; + registration->key = NULL; + registration->nentries = 0; + registration->msg_size = 0; + registration->assigned_limit = 0; + registration->idle_limit = 0; + + xpc_interface.disconnect(ch_number); + + up(®istration->sema); + + return; +} + + +int __init +xp_init(void) +{ + int ret, ch_number; + u64 func_addr = *(u64 *) xp_nofault_PIOR; + u64 err_func_addr = *(u64 *) xp_error_PIOR; + + + if (!ia64_platform_is("sn2")) { + return -ENODEV; + } + + /* + * Register a nofault code region which performs a cross-partition + * PIO read. If the PIO read times out, the MCA handler will consume + * the error and return to a kernel-provided instruction to indicate + * an error. This PIO read exists because it is guaranteed to timeout + * if the destination is down (AMO operations do not timeout on at + * least some CPUs on Shubs <= v1.2, which unfortunately we have to + * work around). + */ + if ((ret = sn_register_nofault_code(func_addr, err_func_addr, + err_func_addr, 1, 1)) != 0) { + printk(KERN_ERR "XP: can't register nofault code, error=%d\n", + ret); + } + /* + * Setup the nofault PIO read target. (There is no special reason why + * SH_IPI_ACCESS was selected.) + */ + if (is_shub2()) { + xp_nofault_PIOR_target = SH2_IPI_ACCESS0; + } else { + xp_nofault_PIOR_target = SH1_IPI_ACCESS; + } + + /* initialize the connection registration semaphores */ + for (ch_number = 0; ch_number < XPC_NCHANNELS; ch_number++) { + sema_init(&xpc_registrations[ch_number].sema, 1); /* mutex */ + } + + return 0; +} +module_init(xp_init); + + +void __exit +xp_exit(void) +{ + u64 func_addr = *(u64 *) xp_nofault_PIOR; + u64 err_func_addr = *(u64 *) xp_error_PIOR; + + + /* unregister the PIO read nofault code region */ + (void) sn_register_nofault_code(func_addr, err_func_addr, + err_func_addr, 1, 0); +} +module_exit(xp_exit); + + +MODULE_AUTHOR("Silicon Graphics, Inc."); +MODULE_DESCRIPTION("Cross Partition (XP) base"); +MODULE_LICENSE("GPL"); + +EXPORT_SYMBOL(xp_nofault_PIOR); +EXPORT_SYMBOL(xp_nofault_PIOR_target); +EXPORT_SYMBOL(xpc_registrations); +EXPORT_SYMBOL(xpc_interface); +EXPORT_SYMBOL(xpc_clear_interface); +EXPORT_SYMBOL(xpc_set_interface); +EXPORT_SYMBOL(xpc_connect); +EXPORT_SYMBOL(xpc_disconnect); + diff --git a/arch/ia64/sn/kernel/xp_nofault.S b/arch/ia64/sn/kernel/xp_nofault.S new file mode 100644 index 00000000000..b772543053c --- /dev/null +++ b/arch/ia64/sn/kernel/xp_nofault.S @@ -0,0 +1,31 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + */ + + +/* + * The xp_nofault_PIOR function takes a pointer to a remote PIO register + * and attempts to load and consume a value from it. This function + * will be registered as a nofault code block. In the event that the + * PIO read fails, the MCA handler will force the error to look + * corrected and vector to the xp_error_PIOR which will return an error. + * + * extern int xp_nofault_PIOR(void *remote_register); + */ + + .global xp_nofault_PIOR +xp_nofault_PIOR: + mov r8=r0 // Stage a success return value + ld8.acq r9=[r32];; // PIO Read the specified register + adds r9=1,r9 // Add to force a consume + br.ret.sptk.many b0;; // Return success + + .global xp_error_PIOR +xp_error_PIOR: + mov r8=1 // Return value of 1 + br.ret.sptk.many b0;; // Return failure + diff --git a/arch/ia64/sn/kernel/xpc.h b/arch/ia64/sn/kernel/xpc.h new file mode 100644 index 00000000000..1a0aed8490d --- /dev/null +++ b/arch/ia64/sn/kernel/xpc.h @@ -0,0 +1,991 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + */ + + +/* + * Cross Partition Communication (XPC) structures and macros. + */ + +#ifndef _IA64_SN_KERNEL_XPC_H +#define _IA64_SN_KERNEL_XPC_H + + +#include <linux/config.h> +#include <linux/interrupt.h> +#include <linux/sysctl.h> +#include <linux/device.h> +#include <asm/pgtable.h> +#include <asm/processor.h> +#include <asm/sn/bte.h> +#include <asm/sn/clksupport.h> +#include <asm/sn/addrs.h> +#include <asm/sn/mspec.h> +#include <asm/sn/shub_mmr.h> +#include <asm/sn/xp.h> + + +/* + * XPC Version numbers consist of a major and minor number. XPC can always + * talk to versions with same major #, and never talk to versions with a + * different major #. + */ +#define _XPC_VERSION(_maj, _min) (((_maj) << 4) | ((_min) & 0xf)) +#define XPC_VERSION_MAJOR(_v) ((_v) >> 4) +#define XPC_VERSION_MINOR(_v) ((_v) & 0xf) + + +/* + * The next macros define word or bit representations for given + * C-brick nasid in either the SAL provided bit array representing + * nasids in the partition/machine or the AMO_t array used for + * inter-partition initiation communications. + * + * For SN2 machines, C-Bricks are alway even numbered NASIDs. As + * such, some space will be saved by insisting that nasid information + * passed from SAL always be packed for C-Bricks and the + * cross-partition interrupts use the same packing scheme. + */ +#define XPC_NASID_W_INDEX(_n) (((_n) / 64) / 2) +#define XPC_NASID_B_INDEX(_n) (((_n) / 2) & (64 - 1)) +#define XPC_NASID_IN_ARRAY(_n, _p) ((_p)[XPC_NASID_W_INDEX(_n)] & \ + (1UL << XPC_NASID_B_INDEX(_n))) +#define XPC_NASID_FROM_W_B(_w, _b) (((_w) * 64 + (_b)) * 2) + +#define XPC_HB_DEFAULT_INTERVAL 5 /* incr HB every x secs */ +#define XPC_HB_CHECK_DEFAULT_TIMEOUT 20 /* check HB every x secs */ + +/* define the process name of HB checker and the CPU it is pinned to */ +#define XPC_HB_CHECK_THREAD_NAME "xpc_hb" +#define XPC_HB_CHECK_CPU 0 + +/* define the process name of the discovery thread */ +#define XPC_DISCOVERY_THREAD_NAME "xpc_discovery" + + +#define XPC_HB_ALLOWED(_p, _v) ((_v)->heartbeating_to_mask & (1UL << (_p))) +#define XPC_ALLOW_HB(_p, _v) (_v)->heartbeating_to_mask |= (1UL << (_p)) +#define XPC_DISALLOW_HB(_p, _v) (_v)->heartbeating_to_mask &= (~(1UL << (_p))) + + +/* + * Reserved Page provided by SAL. + * + * SAL provides one page per partition of reserved memory. When SAL + * initialization is complete, SAL_signature, SAL_version, partid, + * part_nasids, and mach_nasids are set. + * + * Note: Until vars_pa is set, the partition XPC code has not been initialized. + */ +struct xpc_rsvd_page { + u64 SAL_signature; /* SAL unique signature */ + u64 SAL_version; /* SAL specified version */ + u8 partid; /* partition ID from SAL */ + u8 version; + u8 pad[6]; /* pad to u64 align */ + u64 vars_pa; + u64 part_nasids[XP_NASID_MASK_WORDS] ____cacheline_aligned; + u64 mach_nasids[XP_NASID_MASK_WORDS] ____cacheline_aligned; +}; +#define XPC_RP_VERSION _XPC_VERSION(1,0) /* version 1.0 of the reserved page */ + +#define XPC_RSVD_PAGE_ALIGNED_SIZE \ + (L1_CACHE_ALIGN(sizeof(struct xpc_rsvd_page))) + + +/* + * Define the structures by which XPC variables can be exported to other + * partitions. (There are two: struct xpc_vars and struct xpc_vars_part) + */ + +/* + * The following structure describes the partition generic variables + * needed by other partitions in order to properly initialize. + * + * struct xpc_vars version number also applies to struct xpc_vars_part. + * Changes to either structure and/or related functionality should be + * reflected by incrementing either the major or minor version numbers + * of struct xpc_vars. + */ +struct xpc_vars { + u8 version; + u64 heartbeat; + u64 heartbeating_to_mask; + u64 kdb_status; /* 0 = machine running */ + int act_nasid; + int act_phys_cpuid; + u64 vars_part_pa; + u64 amos_page_pa; /* paddr of page of AMOs from MSPEC driver */ + AMO_t *amos_page; /* vaddr of page of AMOs from MSPEC driver */ + AMO_t *act_amos; /* pointer to the first activation AMO */ +}; +#define XPC_V_VERSION _XPC_VERSION(3,0) /* version 3.0 of the cross vars */ + +#define XPC_VARS_ALIGNED_SIZE (L1_CACHE_ALIGN(sizeof(struct xpc_vars))) + +/* + * The following structure describes the per partition specific variables. + * + * An array of these structures, one per partition, will be defined. As a + * partition becomes active XPC will copy the array entry corresponding to + * itself from that partition. It is desirable that the size of this + * structure evenly divide into a cacheline, such that none of the entries + * in this array crosses a cacheline boundary. As it is now, each entry + * occupies half a cacheline. + */ +struct xpc_vars_part { + u64 magic; + + u64 openclose_args_pa; /* physical address of open and close args */ + u64 GPs_pa; /* physical address of Get/Put values */ + + u64 IPI_amo_pa; /* physical address of IPI AMO_t structure */ + int IPI_nasid; /* nasid of where to send IPIs */ + int IPI_phys_cpuid; /* physical CPU ID of where to send IPIs */ + + u8 nchannels; /* #of defined channels supported */ + + u8 reserved[23]; /* pad to a full 64 bytes */ +}; + +/* + * The vars_part MAGIC numbers play a part in the first contact protocol. + * + * MAGIC1 indicates that the per partition specific variables for a remote + * partition have been initialized by this partition. + * + * MAGIC2 indicates that this partition has pulled the remote partititions + * per partition variables that pertain to this partition. + */ +#define XPC_VP_MAGIC1 0x0053524156435058L /* 'XPCVARS\0'L (little endian) */ +#define XPC_VP_MAGIC2 0x0073726176435058L /* 'XPCvars\0'L (little endian) */ + + + +/* + * Functions registered by add_timer() or called by kernel_thread() only + * allow for a single 64-bit argument. The following macros can be used to + * pack and unpack two (32-bit, 16-bit or 8-bit) arguments into or out from + * the passed argument. + */ +#define XPC_PACK_ARGS(_arg1, _arg2) \ + ((((u64) _arg1) & 0xffffffff) | \ + ((((u64) _arg2) & 0xffffffff) << 32)) + +#define XPC_UNPACK_ARG1(_args) (((u64) _args) & 0xffffffff) +#define XPC_UNPACK_ARG2(_args) ((((u64) _args) >> 32) & 0xffffffff) + + + +/* + * Define a Get/Put value pair (pointers) used with a message queue. + */ +struct xpc_gp { + s64 get; /* Get value */ + s64 put; /* Put value */ +}; + +#define XPC_GP_SIZE \ + L1_CACHE_ALIGN(sizeof(struct xpc_gp) * XPC_NCHANNELS) + + + +/* + * Define a structure that contains arguments associated with opening and + * closing a channel. + */ +struct xpc_openclose_args { + u16 reason; /* reason why channel is closing */ + u16 msg_size; /* sizeof each message entry */ + u16 remote_nentries; /* #of message entries in remote msg queue */ + u16 local_nentries; /* #of message entries in local msg queue */ + u64 local_msgqueue_pa; /* physical address of local message queue */ +}; + +#define XPC_OPENCLOSE_ARGS_SIZE \ + L1_CACHE_ALIGN(sizeof(struct xpc_openclose_args) * XPC_NCHANNELS) + + + +/* struct xpc_msg flags */ + +#define XPC_M_DONE 0x01 /* msg has been received/consumed */ +#define XPC_M_READY 0x02 /* msg is ready to be sent */ +#define XPC_M_INTERRUPT 0x04 /* send interrupt when msg consumed */ + + +#define XPC_MSG_ADDRESS(_payload) \ + ((struct xpc_msg *)((u8 *)(_payload) - XPC_MSG_PAYLOAD_OFFSET)) + + + +/* + * Defines notify entry. + * + * This is used to notify a message's sender that their message was received + * and consumed by the intended recipient. + */ +struct xpc_notify { + struct semaphore sema; /* notify semaphore */ + u8 type; /* type of notification */ + + /* the following two fields are only used if type == XPC_N_CALL */ + xpc_notify_func func; /* user's notify function */ + void *key; /* pointer to user's key */ +}; + +/* struct xpc_notify type of notification */ + +#define XPC_N_CALL 0x01 /* notify function provided by user */ + + + +/* + * Define the structure that manages all the stuff required by a channel. In + * particular, they are used to manage the messages sent across the channel. + * + * This structure is private to a partition, and is NOT shared across the + * partition boundary. + * + * There is an array of these structures for each remote partition. It is + * allocated at the time a partition becomes active. The array contains one + * of these structures for each potential channel connection to that partition. + * + * Each of these structures manages two message queues (circular buffers). + * They are allocated at the time a channel connection is made. One of + * these message queues (local_msgqueue) holds the locally created messages + * that are destined for the remote partition. The other of these message + * queues (remote_msgqueue) is a locally cached copy of the remote partition's + * own local_msgqueue. + * + * The following is a description of the Get/Put pointers used to manage these + * two message queues. Consider the local_msgqueue to be on one partition + * and the remote_msgqueue to be its cached copy on another partition. A + * description of what each of the lettered areas contains is included. + * + * + * local_msgqueue remote_msgqueue + * + * |/////////| |/////////| + * w_remote_GP.get --> +---------+ |/////////| + * | F | |/////////| + * remote_GP.get --> +---------+ +---------+ <-- local_GP->get + * | | | | + * | | | E | + * | | | | + * | | +---------+ <-- w_local_GP.get + * | B | |/////////| + * | | |////D////| + * | | |/////////| + * | | +---------+ <-- w_remote_GP.put + * | | |////C////| + * local_GP->put --> +---------+ +---------+ <-- remote_GP.put + * | | |/////////| + * | A | |/////////| + * | | |/////////| + * w_local_GP.put --> +---------+ |/////////| + * |/////////| |/////////| + * + * + * ( remote_GP.[get|put] are cached copies of the remote + * partition's local_GP->[get|put], and thus their values can + * lag behind their counterparts on the remote partition. ) + * + * + * A - Messages that have been allocated, but have not yet been sent to the + * remote partition. + * + * B - Messages that have been sent, but have not yet been acknowledged by the + * remote partition as having been received. + * + * C - Area that needs to be prepared for the copying of sent messages, by + * the clearing of the message flags of any previously received messages. + * + * D - Area into which sent messages are to be copied from the remote + * partition's local_msgqueue and then delivered to their intended + * recipients. [ To allow for a multi-message copy, another pointer + * (next_msg_to_pull) has been added to keep track of the next message + * number needing to be copied (pulled). It chases after w_remote_GP.put. + * Any messages lying between w_local_GP.get and next_msg_to_pull have + * been copied and are ready to be delivered. ] + * + * E - Messages that have been copied and delivered, but have not yet been + * acknowledged by the recipient as having been received. + * + * F - Messages that have been acknowledged, but XPC has not yet notified the + * sender that the message was received by its intended recipient. + * This is also an area that needs to be prepared for the allocating of + * new messages, by the clearing of the message flags of the acknowledged + * messages. + */ +struct xpc_channel { + partid_t partid; /* ID of remote partition connected */ + spinlock_t lock; /* lock for updating this structure */ + u32 flags; /* general flags */ + + enum xpc_retval reason; /* reason why channel is disconnect'g */ + int reason_line; /* line# disconnect initiated from */ + + u16 number; /* channel # */ + + u16 msg_size; /* sizeof each msg entry */ + u16 local_nentries; /* #of msg entries in local msg queue */ + u16 remote_nentries; /* #of msg entries in remote msg queue*/ + + void *local_msgqueue_base; /* base address of kmalloc'd space */ + struct xpc_msg *local_msgqueue; /* local message queue */ + void *remote_msgqueue_base; /* base address of kmalloc'd space */ + struct xpc_msg *remote_msgqueue;/* cached copy of remote partition's */ + /* local message queue */ + u64 remote_msgqueue_pa; /* phys addr of remote partition's */ + /* local message queue */ + + atomic_t references; /* #of external references to queues */ + + atomic_t n_on_msg_allocate_wq; /* #on msg allocation wait queue */ + wait_queue_head_t msg_allocate_wq; /* msg allocation wait queue */ + + /* queue of msg senders who want to be notified when msg received */ + + atomic_t n_to_notify; /* #of msg senders to notify */ + struct xpc_notify *notify_queue;/* notify queue for messages sent */ + + xpc_channel_func func; /* user's channel function */ + void *key; /* pointer to user's key */ + + struct semaphore msg_to_pull_sema; /* next msg to pull serialization */ + struct semaphore teardown_sema; /* wait for teardown completion */ + + struct xpc_openclose_args *local_openclose_args; /* args passed on */ + /* opening or closing of channel */ + + /* various flavors of local and remote Get/Put values */ + + struct xpc_gp *local_GP; /* local Get/Put values */ + struct xpc_gp remote_GP; /* remote Get/Put values */ + struct xpc_gp w_local_GP; /* working local Get/Put values */ + struct xpc_gp w_remote_GP; /* working remote Get/Put values */ + s64 next_msg_to_pull; /* Put value of next msg to pull */ + + /* kthread management related fields */ + +// >>> rethink having kthreads_assigned_limit and kthreads_idle_limit; perhaps +// >>> allow the assigned limit be unbounded and let the idle limit be dynamic +// >>> dependent on activity over the last interval of time + atomic_t kthreads_assigned; /* #of kthreads assigned to channel */ + u32 kthreads_assigned_limit; /* limit on #of kthreads assigned */ + atomic_t kthreads_idle; /* #of kthreads idle waiting for work */ + u32 kthreads_idle_limit; /* limit on #of kthreads idle */ + atomic_t kthreads_active; /* #of kthreads actively working */ + // >>> following field is temporary + u32 kthreads_created; /* total #of kthreads created */ + + wait_queue_head_t idle_wq; /* idle kthread wait queue */ + +} ____cacheline_aligned; + + +/* struct xpc_channel flags */ + +#define XPC_C_WASCONNECTED 0x00000001 /* channel was connected */ + +#define XPC_C_ROPENREPLY 0x00000002 /* remote open channel reply */ +#define XPC_C_OPENREPLY 0x00000004 /* local open channel reply */ +#define XPC_C_ROPENREQUEST 0x00000008 /* remote open channel request */ +#define XPC_C_OPENREQUEST 0x00000010 /* local open channel request */ + +#define XPC_C_SETUP 0x00000020 /* channel's msgqueues are alloc'd */ +#define XPC_C_CONNECTCALLOUT 0x00000040 /* channel connected callout made */ +#define XPC_C_CONNECTED 0x00000080 /* local channel is connected */ +#define XPC_C_CONNECTING 0x00000100 /* channel is being connected */ + +#define XPC_C_RCLOSEREPLY 0x00000200 /* remote close channel reply */ +#define XPC_C_CLOSEREPLY 0x00000400 /* local close channel reply */ +#define XPC_C_RCLOSEREQUEST 0x00000800 /* remote close channel request */ +#define XPC_C_CLOSEREQUEST 0x00001000 /* local close channel request */ + +#define XPC_C_DISCONNECTED 0x00002000 /* channel is disconnected */ +#define XPC_C_DISCONNECTING 0x00004000 /* channel is being disconnected */ + + + +/* + * Manages channels on a partition basis. There is one of these structures + * for each partition (a partition will never utilize the structure that + * represents itself). + */ +struct xpc_partition { + + /* XPC HB infrastructure */ + + u64 remote_rp_pa; /* phys addr of partition's rsvd pg */ + u64 remote_vars_pa; /* phys addr of partition's vars */ + u64 remote_vars_part_pa; /* phys addr of partition's vars part */ + u64 last_heartbeat; /* HB at last read */ + u64 remote_amos_page_pa; /* phys addr of partition's amos page */ + int remote_act_nasid; /* active part's act/deact nasid */ + int remote_act_phys_cpuid; /* active part's act/deact phys cpuid */ + u32 act_IRQ_rcvd; /* IRQs since activation */ + spinlock_t act_lock; /* protect updating of act_state */ + u8 act_state; /* from XPC HB viewpoint */ + enum xpc_retval reason; /* reason partition is deactivating */ + int reason_line; /* line# deactivation initiated from */ + int reactivate_nasid; /* nasid in partition to reactivate */ + + + /* XPC infrastructure referencing and teardown control */ + + u8 setup_state; /* infrastructure setup state */ + wait_queue_head_t teardown_wq; /* kthread waiting to teardown infra */ + atomic_t references; /* #of references to infrastructure */ + + + /* + * NONE OF THE PRECEDING FIELDS OF THIS STRUCTURE WILL BE CLEARED WHEN + * XPC SETS UP THE NECESSARY INFRASTRUCTURE TO SUPPORT CROSS PARTITION + * COMMUNICATION. ALL OF THE FOLLOWING FIELDS WILL BE CLEARED. (THE + * 'nchannels' FIELD MUST BE THE FIRST OF THE FIELDS TO BE CLEARED.) + */ + + + u8 nchannels; /* #of defined channels supported */ + atomic_t nchannels_active; /* #of channels that are not DISCONNECTED */ + struct xpc_channel *channels;/* array of channel structures */ + + void *local_GPs_base; /* base address of kmalloc'd space */ + struct xpc_gp *local_GPs; /* local Get/Put values */ + void *remote_GPs_base; /* base address of kmalloc'd space */ + struct xpc_gp *remote_GPs;/* copy of remote partition's local Get/Put */ + /* values */ + u64 remote_GPs_pa; /* phys address of remote partition's local */ + /* Get/Put values */ + + + /* fields used to pass args when opening or closing a channel */ + + void *local_openclose_args_base; /* base address of kmalloc'd space */ + struct xpc_openclose_args *local_openclose_args; /* local's args */ + void *remote_openclose_args_base; /* base address of kmalloc'd space */ + struct xpc_openclose_args *remote_openclose_args; /* copy of remote's */ + /* args */ + u64 remote_openclose_args_pa; /* phys addr of remote's args */ + + + /* IPI sending, receiving and handling related fields */ + + int remote_IPI_nasid; /* nasid of where to send IPIs */ + int remote_IPI_phys_cpuid; /* phys CPU ID of where to send IPIs */ + AMO_t *remote_IPI_amo_va; /* address of remote IPI AMO_t structure */ + + AMO_t *local_IPI_amo_va; /* address of IPI AMO_t structure */ + u64 local_IPI_amo; /* IPI amo flags yet to be handled */ + char IPI_owner[8]; /* IPI owner's name */ + struct timer_list dropped_IPI_timer; /* dropped IPI timer */ + + spinlock_t IPI_lock; /* IPI handler lock */ + + + /* channel manager related fields */ + + atomic_t channel_mgr_requests; /* #of requests to activate chan mgr */ + wait_queue_head_t channel_mgr_wq; /* channel mgr's wait queue */ + +} ____cacheline_aligned; + + +/* struct xpc_partition act_state values (for XPC HB) */ + +#define XPC_P_INACTIVE 0x00 /* partition is not active */ +#define XPC_P_ACTIVATION_REQ 0x01 /* created thread to activate */ +#define XPC_P_ACTIVATING 0x02 /* activation thread started */ +#define XPC_P_ACTIVE 0x03 /* xpc_partition_up() was called */ +#define XPC_P_DEACTIVATING 0x04 /* partition deactivation initiated */ + + +#define XPC_DEACTIVATE_PARTITION(_p, _reason) \ + xpc_deactivate_partition(__LINE__, (_p), (_reason)) + + +/* struct xpc_partition setup_state values */ + +#define XPC_P_UNSET 0x00 /* infrastructure was never setup */ +#define XPC_P_SETUP 0x01 /* infrastructure is setup */ +#define XPC_P_WTEARDOWN 0x02 /* waiting to teardown infrastructure */ +#define XPC_P_TORNDOWN 0x03 /* infrastructure is torndown */ + + +/* + * struct xpc_partition IPI_timer #of seconds to wait before checking for + * dropped IPIs. These occur whenever an IPI amo write doesn't complete until + * after the IPI was received. + */ +#define XPC_P_DROPPED_IPI_WAIT (0.25 * HZ) + + +#define XPC_PARTID(_p) ((partid_t) ((_p) - &xpc_partitions[0])) + + + +/* found in xp_main.c */ +extern struct xpc_registration xpc_registrations[]; + + +/* >>> found in xpc_main.c only */ +extern struct device *xpc_part; +extern struct device *xpc_chan; +extern irqreturn_t xpc_notify_IRQ_handler(int, void *, struct pt_regs *); +extern void xpc_dropped_IPI_check(struct xpc_partition *); +extern void xpc_activate_kthreads(struct xpc_channel *, int); +extern void xpc_create_kthreads(struct xpc_channel *, int); +extern void xpc_disconnect_wait(int); + + +/* found in xpc_main.c and efi-xpc.c */ +extern void xpc_activate_partition(struct xpc_partition *); + + +/* found in xpc_partition.c */ +extern int xpc_exiting; +extern int xpc_hb_interval; +extern int xpc_hb_check_interval; +extern struct xpc_vars *xpc_vars; +extern struct xpc_rsvd_page *xpc_rsvd_page; +extern struct xpc_vars_part *xpc_vars_part; +extern struct xpc_partition xpc_partitions[XP_MAX_PARTITIONS + 1]; +extern char xpc_remote_copy_buffer[]; +extern struct xpc_rsvd_page *xpc_rsvd_page_init(void); +extern void xpc_allow_IPI_ops(void); +extern void xpc_restrict_IPI_ops(void); +extern int xpc_identify_act_IRQ_sender(void); +extern enum xpc_retval xpc_mark_partition_active(struct xpc_partition *); +extern void xpc_mark_partition_inactive(struct xpc_partition *); +extern void xpc_discovery(void); +extern void xpc_check_remote_hb(void); +extern void xpc_deactivate_partition(const int, struct xpc_partition *, + enum xpc_retval); +extern enum xpc_retval xpc_initiate_partid_to_nasids(partid_t, void *); + + +/* found in xpc_channel.c */ +extern void xpc_initiate_connect(int); +extern void xpc_initiate_disconnect(int); +extern enum xpc_retval xpc_initiate_allocate(partid_t, int, u32, void **); +extern enum xpc_retval xpc_initiate_send(partid_t, int, void *); +extern enum xpc_retval xpc_initiate_send_notify(partid_t, int, void *, + xpc_notify_func, void *); +extern void xpc_initiate_received(partid_t, int, void *); +extern enum xpc_retval xpc_setup_infrastructure(struct xpc_partition *); +extern enum xpc_retval xpc_pull_remote_vars_part(struct xpc_partition *); +extern void xpc_process_channel_activity(struct xpc_partition *); +extern void xpc_connected_callout(struct xpc_channel *); +extern void xpc_deliver_msg(struct xpc_channel *); +extern void xpc_disconnect_channel(const int, struct xpc_channel *, + enum xpc_retval, unsigned long *); +extern void xpc_disconnected_callout(struct xpc_channel *); +extern void xpc_partition_down(struct xpc_partition *, enum xpc_retval); +extern void xpc_teardown_infrastructure(struct xpc_partition *); + + + +static inline void +xpc_wakeup_channel_mgr(struct xpc_partition *part) +{ + if (atomic_inc_return(&part->channel_mgr_requests) == 1) { + wake_up(&part->channel_mgr_wq); + } +} + + + +/* + * These next two inlines are used to keep us from tearing down a channel's + * msg queues while a thread may be referencing them. + */ +static inline void +xpc_msgqueue_ref(struct xpc_channel *ch) +{ + atomic_inc(&ch->references); +} + +static inline void +xpc_msgqueue_deref(struct xpc_channel *ch) +{ + s32 refs = atomic_dec_return(&ch->references); + + DBUG_ON(refs < 0); + if (refs == 0) { + xpc_wakeup_channel_mgr(&xpc_partitions[ch->partid]); + } +} + + + +#define XPC_DISCONNECT_CHANNEL(_ch, _reason, _irqflgs) \ + xpc_disconnect_channel(__LINE__, _ch, _reason, _irqflgs) + + +/* + * These two inlines are used to keep us from tearing down a partition's + * setup infrastructure while a thread may be referencing it. + */ +static inline void +xpc_part_deref(struct xpc_partition *part) +{ + s32 refs = atomic_dec_return(&part->references); + + + DBUG_ON(refs < 0); + if (refs == 0 && part->setup_state == XPC_P_WTEARDOWN) { + wake_up(&part->teardown_wq); + } +} + +static inline int +xpc_part_ref(struct xpc_partition *part) +{ + int setup; + + + atomic_inc(&part->references); + setup = (part->setup_state == XPC_P_SETUP); + if (!setup) { + xpc_part_deref(part); + } + return setup; +} + + + +/* + * The following macro is to be used for the setting of the reason and + * reason_line fields in both the struct xpc_channel and struct xpc_partition + * structures. + */ +#define XPC_SET_REASON(_p, _reason, _line) \ + { \ + (_p)->reason = _reason; \ + (_p)->reason_line = _line; \ + } + + + +/* + * The following set of macros and inlines are used for the sending and + * receiving of IPIs (also known as IRQs). There are two flavors of IPIs, + * one that is associated with partition activity (SGI_XPC_ACTIVATE) and + * the other that is associated with channel activity (SGI_XPC_NOTIFY). + */ + +static inline u64 +xpc_IPI_receive(AMO_t *amo) +{ + return FETCHOP_LOAD_OP(TO_AMO((u64) &amo->variable), FETCHOP_CLEAR); +} + + +static inline enum xpc_retval +xpc_IPI_send(AMO_t *amo, u64 flag, int nasid, int phys_cpuid, int vector) +{ + int ret = 0; + unsigned long irq_flags; + + + local_irq_save(irq_flags); + + FETCHOP_STORE_OP(TO_AMO((u64) &amo->variable), FETCHOP_OR, flag); + sn_send_IPI_phys(nasid, phys_cpuid, vector, 0); + + /* + * We must always use the nofault function regardless of whether we + * are on a Shub 1.1 system or a Shub 1.2 slice 0xc processor. If we + * didn't, we'd never know that the other partition is down and would + * keep sending IPIs and AMOs to it until the heartbeat times out. + */ + ret = xp_nofault_PIOR((u64 *) GLOBAL_MMR_ADDR(NASID_GET(&amo->variable), + xp_nofault_PIOR_target)); + + local_irq_restore(irq_flags); + + return ((ret == 0) ? xpcSuccess : xpcPioReadError); +} + + +/* + * IPIs associated with SGI_XPC_ACTIVATE IRQ. + */ + +/* + * Flag the appropriate AMO variable and send an IPI to the specified node. + */ +static inline void +xpc_activate_IRQ_send(u64 amos_page, int from_nasid, int to_nasid, + int to_phys_cpuid) +{ + int w_index = XPC_NASID_W_INDEX(from_nasid); + int b_index = XPC_NASID_B_INDEX(from_nasid); + AMO_t *amos = (AMO_t *) __va(amos_page + + (XP_MAX_PARTITIONS * sizeof(AMO_t))); + + + (void) xpc_IPI_send(&amos[w_index], (1UL << b_index), to_nasid, + to_phys_cpuid, SGI_XPC_ACTIVATE); +} + +static inline void +xpc_IPI_send_activate(struct xpc_vars *vars) +{ + xpc_activate_IRQ_send(vars->amos_page_pa, cnodeid_to_nasid(0), + vars->act_nasid, vars->act_phys_cpuid); +} + +static inline void +xpc_IPI_send_activated(struct xpc_partition *part) +{ + xpc_activate_IRQ_send(part->remote_amos_page_pa, cnodeid_to_nasid(0), + part->remote_act_nasid, part->remote_act_phys_cpuid); +} + +static inline void +xpc_IPI_send_reactivate(struct xpc_partition *part) +{ + xpc_activate_IRQ_send(xpc_vars->amos_page_pa, part->reactivate_nasid, + xpc_vars->act_nasid, xpc_vars->act_phys_cpuid); +} + + +/* + * IPIs associated with SGI_XPC_NOTIFY IRQ. + */ + +/* + * Send an IPI to the remote partition that is associated with the + * specified channel. + */ +#define XPC_NOTIFY_IRQ_SEND(_ch, _ipi_f, _irq_f) \ + xpc_notify_IRQ_send(_ch, _ipi_f, #_ipi_f, _irq_f) + +static inline void +xpc_notify_IRQ_send(struct xpc_channel *ch, u8 ipi_flag, char *ipi_flag_string, + unsigned long *irq_flags) +{ + struct xpc_partition *part = &xpc_partitions[ch->partid]; + enum xpc_retval ret; + + + if (likely(part->act_state != XPC_P_DEACTIVATING)) { + ret = xpc_IPI_send(part->remote_IPI_amo_va, + (u64) ipi_flag << (ch->number * 8), + part->remote_IPI_nasid, + part->remote_IPI_phys_cpuid, + SGI_XPC_NOTIFY); + dev_dbg(xpc_chan, "%s sent to partid=%d, channel=%d, ret=%d\n", + ipi_flag_string, ch->partid, ch->number, ret); + if (unlikely(ret != xpcSuccess)) { + if (irq_flags != NULL) { + spin_unlock_irqrestore(&ch->lock, *irq_flags); + } + XPC_DEACTIVATE_PARTITION(part, ret); + if (irq_flags != NULL) { + spin_lock_irqsave(&ch->lock, *irq_flags); + } + } + } +} + + +/* + * Make it look like the remote partition, which is associated with the + * specified channel, sent us an IPI. This faked IPI will be handled + * by xpc_dropped_IPI_check(). + */ +#define XPC_NOTIFY_IRQ_SEND_LOCAL(_ch, _ipi_f) \ + xpc_notify_IRQ_send_local(_ch, _ipi_f, #_ipi_f) + +static inline void +xpc_notify_IRQ_send_local(struct xpc_channel *ch, u8 ipi_flag, + char *ipi_flag_string) +{ + struct xpc_partition *part = &xpc_partitions[ch->partid]; + + + FETCHOP_STORE_OP(TO_AMO((u64) &part->local_IPI_amo_va->variable), + FETCHOP_OR, ((u64) ipi_flag << (ch->number * 8))); + dev_dbg(xpc_chan, "%s sent local from partid=%d, channel=%d\n", + ipi_flag_string, ch->partid, ch->number); +} + + +/* + * The sending and receiving of IPIs includes the setting of an AMO variable + * to indicate the reason the IPI was sent. The 64-bit variable is divided + * up into eight bytes, ordered from right to left. Byte zero pertains to + * channel 0, byte one to channel 1, and so on. Each byte is described by + * the following IPI flags. + */ + +#define XPC_IPI_CLOSEREQUEST 0x01 +#define XPC_IPI_CLOSEREPLY 0x02 +#define XPC_IPI_OPENREQUEST 0x04 +#define XPC_IPI_OPENREPLY 0x08 +#define XPC_IPI_MSGREQUEST 0x10 + + +/* given an AMO variable and a channel#, get its associated IPI flags */ +#define XPC_GET_IPI_FLAGS(_amo, _c) ((u8) (((_amo) >> ((_c) * 8)) & 0xff)) + +#define XPC_ANY_OPENCLOSE_IPI_FLAGS_SET(_amo) ((_amo) & 0x0f0f0f0f0f0f0f0f) +#define XPC_ANY_MSG_IPI_FLAGS_SET(_amo) ((_amo) & 0x1010101010101010) + + +static inline void +xpc_IPI_send_closerequest(struct xpc_channel *ch, unsigned long *irq_flags) +{ + struct xpc_openclose_args *args = ch->local_openclose_args; + + + args->reason = ch->reason; + + XPC_NOTIFY_IRQ_SEND(ch, XPC_IPI_CLOSEREQUEST, irq_flags); +} + +static inline void +xpc_IPI_send_closereply(struct xpc_channel *ch, unsigned long *irq_flags) +{ + XPC_NOTIFY_IRQ_SEND(ch, XPC_IPI_CLOSEREPLY, irq_flags); +} + +static inline void +xpc_IPI_send_openrequest(struct xpc_channel *ch, unsigned long *irq_flags) +{ + struct xpc_openclose_args *args = ch->local_openclose_args; + + + args->msg_size = ch->msg_size; + args->local_nentries = ch->local_nentries; + + XPC_NOTIFY_IRQ_SEND(ch, XPC_IPI_OPENREQUEST, irq_flags); +} + +static inline void +xpc_IPI_send_openreply(struct xpc_channel *ch, unsigned long *irq_flags) +{ + struct xpc_openclose_args *args = ch->local_openclose_args; + + + args->remote_nentries = ch->remote_nentries; + args->local_nentries = ch->local_nentries; + args->local_msgqueue_pa = __pa(ch->local_msgqueue); + + XPC_NOTIFY_IRQ_SEND(ch, XPC_IPI_OPENREPLY, irq_flags); +} + +static inline void +xpc_IPI_send_msgrequest(struct xpc_channel *ch) +{ + XPC_NOTIFY_IRQ_SEND(ch, XPC_IPI_MSGREQUEST, NULL); +} + +static inline void +xpc_IPI_send_local_msgrequest(struct xpc_channel *ch) +{ + XPC_NOTIFY_IRQ_SEND_LOCAL(ch, XPC_IPI_MSGREQUEST); +} + + +/* + * Memory for XPC's AMO variables is allocated by the MSPEC driver. These + * pages are located in the lowest granule. The lowest granule uses 4k pages + * for cached references and an alternate TLB handler to never provide a + * cacheable mapping for the entire region. This will prevent speculative + * reading of cached copies of our lines from being issued which will cause + * a PI FSB Protocol error to be generated by the SHUB. For XPC, we need 64 + * (XP_MAX_PARTITIONS) AMO variables for message notification (xpc_main.c) + * and an additional 16 AMO variables for partition activation (xpc_hb.c). + */ +static inline AMO_t * +xpc_IPI_init(partid_t partid) +{ + AMO_t *part_amo = xpc_vars->amos_page + partid; + + + xpc_IPI_receive(part_amo); + return part_amo; +} + + + +static inline enum xpc_retval +xpc_map_bte_errors(bte_result_t error) +{ + switch (error) { + case BTE_SUCCESS: return xpcSuccess; + case BTEFAIL_DIR: return xpcBteDirectoryError; + case BTEFAIL_POISON: return xpcBtePoisonError; + case BTEFAIL_WERR: return xpcBteWriteError; + case BTEFAIL_ACCESS: return xpcBteAccessError; + case BTEFAIL_PWERR: return xpcBtePWriteError; + case BTEFAIL_PRERR: return xpcBtePReadError; + case BTEFAIL_TOUT: return xpcBteTimeOutError; + case BTEFAIL_XTERR: return xpcBteXtalkError; + case BTEFAIL_NOTAVAIL: return xpcBteNotAvailable; + default: return xpcBteUnmappedError; + } +} + + + +static inline void * +xpc_kmalloc_cacheline_aligned(size_t size, int flags, void **base) +{ + /* see if kmalloc will give us cachline aligned memory by default */ + *base = kmalloc(size, flags); + if (*base == NULL) { + return NULL; + } + if ((u64) *base == L1_CACHE_ALIGN((u64) *base)) { + return *base; + } + kfree(*base); + + /* nope, we'll have to do it ourselves */ + *base = kmalloc(size + L1_CACHE_BYTES, flags); + if (*base == NULL) { + return NULL; + } + return (void *) L1_CACHE_ALIGN((u64) *base); +} + + +/* + * Check to see if there is any channel activity to/from the specified + * partition. + */ +static inline void +xpc_check_for_channel_activity(struct xpc_partition *part) +{ + u64 IPI_amo; + unsigned long irq_flags; + + + IPI_amo = xpc_IPI_receive(part->local_IPI_amo_va); + if (IPI_amo == 0) { + return; + } + + spin_lock_irqsave(&part->IPI_lock, irq_flags); + part->local_IPI_amo |= IPI_amo; + spin_unlock_irqrestore(&part->IPI_lock, irq_flags); + + dev_dbg(xpc_chan, "received IPI from partid=%d, IPI_amo=0x%lx\n", + XPC_PARTID(part), IPI_amo); + + xpc_wakeup_channel_mgr(part); +} + + +#endif /* _IA64_SN_KERNEL_XPC_H */ + diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c new file mode 100644 index 00000000000..0bf6fbcc46d --- /dev/null +++ b/arch/ia64/sn/kernel/xpc_channel.c @@ -0,0 +1,2297 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + */ + + +/* + * Cross Partition Communication (XPC) channel support. + * + * This is the part of XPC that manages the channels and + * sends/receives messages across them to/from other partitions. + * + */ + + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/sched.h> +#include <linux/cache.h> +#include <linux/interrupt.h> +#include <linux/slab.h> +#include <asm/sn/bte.h> +#include <asm/sn/sn_sal.h> +#include "xpc.h" + + +/* + * Set up the initial values for the XPartition Communication channels. + */ +static void +xpc_initialize_channels(struct xpc_partition *part, partid_t partid) +{ + int ch_number; + struct xpc_channel *ch; + + + for (ch_number = 0; ch_number < part->nchannels; ch_number++) { + ch = &part->channels[ch_number]; + + ch->partid = partid; + ch->number = ch_number; + ch->flags = XPC_C_DISCONNECTED; + + ch->local_GP = &part->local_GPs[ch_number]; + ch->local_openclose_args = + &part->local_openclose_args[ch_number]; + + atomic_set(&ch->kthreads_assigned, 0); + atomic_set(&ch->kthreads_idle, 0); + atomic_set(&ch->kthreads_active, 0); + + atomic_set(&ch->references, 0); + atomic_set(&ch->n_to_notify, 0); + + spin_lock_init(&ch->lock); + sema_init(&ch->msg_to_pull_sema, 1); /* mutex */ + + atomic_set(&ch->n_on_msg_allocate_wq, 0); + init_waitqueue_head(&ch->msg_allocate_wq); + init_waitqueue_head(&ch->idle_wq); + } +} + + +/* + * Setup the infrastructure necessary to support XPartition Communication + * between the specified remote partition and the local one. + */ +enum xpc_retval +xpc_setup_infrastructure(struct xpc_partition *part) +{ + int ret; + struct timer_list *timer; + partid_t partid = XPC_PARTID(part); + + + /* + * Zero out MOST of the entry for this partition. Only the fields + * starting with `nchannels' will be zeroed. The preceding fields must + * remain `viable' across partition ups and downs, since they may be + * referenced during this memset() operation. + */ + memset(&part->nchannels, 0, sizeof(struct xpc_partition) - + offsetof(struct xpc_partition, nchannels)); + + /* + * Allocate all of the channel structures as a contiguous chunk of + * memory. + */ + part->channels = kmalloc(sizeof(struct xpc_channel) * XPC_NCHANNELS, + GFP_KERNEL); + if (part->channels == NULL) { + dev_err(xpc_chan, "can't get memory for channels\n"); + return xpcNoMemory; + } + memset(part->channels, 0, sizeof(struct xpc_channel) * XPC_NCHANNELS); + + part->nchannels = XPC_NCHANNELS; + + + /* allocate all the required GET/PUT values */ + + part->local_GPs = xpc_kmalloc_cacheline_aligned(XPC_GP_SIZE, + GFP_KERNEL, &part->local_GPs_base); + if (part->local_GPs == NULL) { + kfree(part->channels); + part->channels = NULL; + dev_err(xpc_chan, "can't get memory for local get/put " + "values\n"); + return xpcNoMemory; + } + memset(part->local_GPs, 0, XPC_GP_SIZE); + + part->remote_GPs = xpc_kmalloc_cacheline_aligned(XPC_GP_SIZE, + GFP_KERNEL, &part->remote_GPs_base); + if (part->remote_GPs == NULL) { + kfree(part->channels); + part->channels = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + dev_err(xpc_chan, "can't get memory for remote get/put " + "values\n"); + return xpcNoMemory; + } + memset(part->remote_GPs, 0, XPC_GP_SIZE); + + + /* allocate all the required open and close args */ + + part->local_openclose_args = xpc_kmalloc_cacheline_aligned( + XPC_OPENCLOSE_ARGS_SIZE, GFP_KERNEL, + &part->local_openclose_args_base); + if (part->local_openclose_args == NULL) { + kfree(part->channels); + part->channels = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->remote_GPs_base); + part->remote_GPs = NULL; + dev_err(xpc_chan, "can't get memory for local connect args\n"); + return xpcNoMemory; + } + memset(part->local_openclose_args, 0, XPC_OPENCLOSE_ARGS_SIZE); + + part->remote_openclose_args = xpc_kmalloc_cacheline_aligned( + XPC_OPENCLOSE_ARGS_SIZE, GFP_KERNEL, + &part->remote_openclose_args_base); + if (part->remote_openclose_args == NULL) { + kfree(part->channels); + part->channels = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->remote_GPs_base); + part->remote_GPs = NULL; + kfree(part->local_openclose_args_base); + part->local_openclose_args = NULL; + dev_err(xpc_chan, "can't get memory for remote connect args\n"); + return xpcNoMemory; + } + memset(part->remote_openclose_args, 0, XPC_OPENCLOSE_ARGS_SIZE); + + + xpc_initialize_channels(part, partid); + + atomic_set(&part->nchannels_active, 0); + + + /* local_IPI_amo were set to 0 by an earlier memset() */ + + /* Initialize this partitions AMO_t structure */ + part->local_IPI_amo_va = xpc_IPI_init(partid); + + spin_lock_init(&part->IPI_lock); + + atomic_set(&part->channel_mgr_requests, 1); + init_waitqueue_head(&part->channel_mgr_wq); + + sprintf(part->IPI_owner, "xpc%02d", partid); + ret = request_irq(SGI_XPC_NOTIFY, xpc_notify_IRQ_handler, SA_SHIRQ, + part->IPI_owner, (void *) (u64) partid); + if (ret != 0) { + kfree(part->channels); + part->channels = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->remote_GPs_base); + part->remote_GPs = NULL; + kfree(part->local_openclose_args_base); + part->local_openclose_args = NULL; + kfree(part->remote_openclose_args_base); + part->remote_openclose_args = NULL; + dev_err(xpc_chan, "can't register NOTIFY IRQ handler, " + "errno=%d\n", -ret); + return xpcLackOfResources; + } + + /* Setup a timer to check for dropped IPIs */ + timer = &part->dropped_IPI_timer; + init_timer(timer); + timer->function = (void (*)(unsigned long)) xpc_dropped_IPI_check; + timer->data = (unsigned long) part; + timer->expires = jiffies + XPC_P_DROPPED_IPI_WAIT; + add_timer(timer); + + /* + * With the setting of the partition setup_state to XPC_P_SETUP, we're + * declaring that this partition is ready to go. + */ + (volatile u8) part->setup_state = XPC_P_SETUP; + + + /* + * Setup the per partition specific variables required by the + * remote partition to establish channel connections with us. + * + * The setting of the magic # indicates that these per partition + * specific variables are ready to be used. + */ + xpc_vars_part[partid].GPs_pa = __pa(part->local_GPs); + xpc_vars_part[partid].openclose_args_pa = + __pa(part->local_openclose_args); + xpc_vars_part[partid].IPI_amo_pa = __pa(part->local_IPI_amo_va); + xpc_vars_part[partid].IPI_nasid = cpuid_to_nasid(smp_processor_id()); + xpc_vars_part[partid].IPI_phys_cpuid = + cpu_physical_id(smp_processor_id()); + xpc_vars_part[partid].nchannels = part->nchannels; + (volatile u64) xpc_vars_part[partid].magic = XPC_VP_MAGIC1; + + return xpcSuccess; +} + + +/* + * Create a wrapper that hides the underlying mechanism for pulling a cacheline + * (or multiple cachelines) from a remote partition. + * + * src must be a cacheline aligned physical address on the remote partition. + * dst must be a cacheline aligned virtual address on this partition. + * cnt must be an cacheline sized + */ +static enum xpc_retval +xpc_pull_remote_cachelines(struct xpc_partition *part, void *dst, + const void *src, size_t cnt) +{ + bte_result_t bte_ret; + + + DBUG_ON((u64) src != L1_CACHE_ALIGN((u64) src)); + DBUG_ON((u64) dst != L1_CACHE_ALIGN((u64) dst)); + DBUG_ON(cnt != L1_CACHE_ALIGN(cnt)); + + if (part->act_state == XPC_P_DEACTIVATING) { + return part->reason; + } + + bte_ret = xp_bte_copy((u64) src, (u64) ia64_tpa((u64) dst), + (u64) cnt, (BTE_NORMAL | BTE_WACQUIRE), NULL); + if (bte_ret == BTE_SUCCESS) { + return xpcSuccess; + } + + dev_dbg(xpc_chan, "xp_bte_copy() from partition %d failed, ret=%d\n", + XPC_PARTID(part), bte_ret); + + return xpc_map_bte_errors(bte_ret); +} + + +/* + * Pull the remote per partititon specific variables from the specified + * partition. + */ +enum xpc_retval +xpc_pull_remote_vars_part(struct xpc_partition *part) +{ + u8 buffer[L1_CACHE_BYTES * 2]; + struct xpc_vars_part *pulled_entry_cacheline = + (struct xpc_vars_part *) L1_CACHE_ALIGN((u64) buffer); + struct xpc_vars_part *pulled_entry; + u64 remote_entry_cacheline_pa, remote_entry_pa; + partid_t partid = XPC_PARTID(part); + enum xpc_retval ret; + + + /* pull the cacheline that contains the variables we're interested in */ + + DBUG_ON(part->remote_vars_part_pa != + L1_CACHE_ALIGN(part->remote_vars_part_pa)); + DBUG_ON(sizeof(struct xpc_vars_part) != L1_CACHE_BYTES / 2); + + remote_entry_pa = part->remote_vars_part_pa + + sn_partition_id * sizeof(struct xpc_vars_part); + + remote_entry_cacheline_pa = (remote_entry_pa & ~(L1_CACHE_BYTES - 1)); + + pulled_entry = (struct xpc_vars_part *) ((u64) pulled_entry_cacheline + + (remote_entry_pa & (L1_CACHE_BYTES - 1))); + + ret = xpc_pull_remote_cachelines(part, pulled_entry_cacheline, + (void *) remote_entry_cacheline_pa, + L1_CACHE_BYTES); + if (ret != xpcSuccess) { + dev_dbg(xpc_chan, "failed to pull XPC vars_part from " + "partition %d, ret=%d\n", partid, ret); + return ret; + } + + + /* see if they've been set up yet */ + + if (pulled_entry->magic != XPC_VP_MAGIC1 && + pulled_entry->magic != XPC_VP_MAGIC2) { + + if (pulled_entry->magic != 0) { + dev_dbg(xpc_chan, "partition %d's XPC vars_part for " + "partition %d has bad magic value (=0x%lx)\n", + partid, sn_partition_id, pulled_entry->magic); + return xpcBadMagic; + } + + /* they've not been initialized yet */ + return xpcRetry; + } + + if (xpc_vars_part[partid].magic == XPC_VP_MAGIC1) { + + /* validate the variables */ + + if (pulled_entry->GPs_pa == 0 || + pulled_entry->openclose_args_pa == 0 || + pulled_entry->IPI_amo_pa == 0) { + + dev_err(xpc_chan, "partition %d's XPC vars_part for " + "partition %d are not valid\n", partid, + sn_partition_id); + return xpcInvalidAddress; + } + + /* the variables we imported look to be valid */ + + part->remote_GPs_pa = pulled_entry->GPs_pa; + part->remote_openclose_args_pa = + pulled_entry->openclose_args_pa; + part->remote_IPI_amo_va = + (AMO_t *) __va(pulled_entry->IPI_amo_pa); + part->remote_IPI_nasid = pulled_entry->IPI_nasid; + part->remote_IPI_phys_cpuid = pulled_entry->IPI_phys_cpuid; + + if (part->nchannels > pulled_entry->nchannels) { + part->nchannels = pulled_entry->nchannels; + } + + /* let the other side know that we've pulled their variables */ + + (volatile u64) xpc_vars_part[partid].magic = XPC_VP_MAGIC2; + } + + if (pulled_entry->magic == XPC_VP_MAGIC1) { + return xpcRetry; + } + + return xpcSuccess; +} + + +/* + * Get the IPI flags and pull the openclose args and/or remote GPs as needed. + */ +static u64 +xpc_get_IPI_flags(struct xpc_partition *part) +{ + unsigned long irq_flags; + u64 IPI_amo; + enum xpc_retval ret; + + + /* + * See if there are any IPI flags to be handled. + */ + + spin_lock_irqsave(&part->IPI_lock, irq_flags); + if ((IPI_amo = part->local_IPI_amo) != 0) { + part->local_IPI_amo = 0; + } + spin_unlock_irqrestore(&part->IPI_lock, irq_flags); + + + if (XPC_ANY_OPENCLOSE_IPI_FLAGS_SET(IPI_amo)) { + ret = xpc_pull_remote_cachelines(part, + part->remote_openclose_args, + (void *) part->remote_openclose_args_pa, + XPC_OPENCLOSE_ARGS_SIZE); + if (ret != xpcSuccess) { + XPC_DEACTIVATE_PARTITION(part, ret); + + dev_dbg(xpc_chan, "failed to pull openclose args from " + "partition %d, ret=%d\n", XPC_PARTID(part), + ret); + + /* don't bother processing IPIs anymore */ + IPI_amo = 0; + } + } + + if (XPC_ANY_MSG_IPI_FLAGS_SET(IPI_amo)) { + ret = xpc_pull_remote_cachelines(part, part->remote_GPs, + (void *) part->remote_GPs_pa, + XPC_GP_SIZE); + if (ret != xpcSuccess) { + XPC_DEACTIVATE_PARTITION(part, ret); + + dev_dbg(xpc_chan, "failed to pull GPs from partition " + "%d, ret=%d\n", XPC_PARTID(part), ret); + + /* don't bother processing IPIs anymore */ + IPI_amo = 0; + } + } + + return IPI_amo; +} + + +/* + * Allocate the local message queue and the notify queue. + */ +static enum xpc_retval +xpc_allocate_local_msgqueue(struct xpc_channel *ch) +{ + unsigned long irq_flags; + int nentries; + size_t nbytes; + + + // >>> may want to check for ch->flags & XPC_C_DISCONNECTING between + // >>> iterations of the for-loop, bail if set? + + // >>> should we impose a minumum #of entries? like 4 or 8? + for (nentries = ch->local_nentries; nentries > 0; nentries--) { + + nbytes = nentries * ch->msg_size; + ch->local_msgqueue = xpc_kmalloc_cacheline_aligned(nbytes, + (GFP_KERNEL | GFP_DMA), + &ch->local_msgqueue_base); + if (ch->local_msgqueue == NULL) { + continue; + } + memset(ch->local_msgqueue, 0, nbytes); + + nbytes = nentries * sizeof(struct xpc_notify); + ch->notify_queue = kmalloc(nbytes, (GFP_KERNEL | GFP_DMA)); + if (ch->notify_queue == NULL) { + kfree(ch->local_msgqueue_base); + ch->local_msgqueue = NULL; + continue; + } + memset(ch->notify_queue, 0, nbytes); + + spin_lock_irqsave(&ch->lock, irq_flags); + if (nentries < ch->local_nentries) { + dev_dbg(xpc_chan, "nentries=%d local_nentries=%d, " + "partid=%d, channel=%d\n", nentries, + ch->local_nentries, ch->partid, ch->number); + + ch->local_nentries = nentries; + } + spin_unlock_irqrestore(&ch->lock, irq_flags); + return xpcSuccess; + } + + dev_dbg(xpc_chan, "can't get memory for local message queue and notify " + "queue, partid=%d, channel=%d\n", ch->partid, ch->number); + return xpcNoMemory; +} + + +/* + * Allocate the cached remote message queue. + */ +static enum xpc_retval +xpc_allocate_remote_msgqueue(struct xpc_channel *ch) +{ + unsigned long irq_flags; + int nentries; + size_t nbytes; + + + DBUG_ON(ch->remote_nentries <= 0); + + // >>> may want to check for ch->flags & XPC_C_DISCONNECTING between + // >>> iterations of the for-loop, bail if set? + + // >>> should we impose a minumum #of entries? like 4 or 8? + for (nentries = ch->remote_nentries; nentries > 0; nentries--) { + + nbytes = nentries * ch->msg_size; + ch->remote_msgqueue = xpc_kmalloc_cacheline_aligned(nbytes, + (GFP_KERNEL | GFP_DMA), + &ch->remote_msgqueue_base); + if (ch->remote_msgqueue == NULL) { + continue; + } + memset(ch->remote_msgqueue, 0, nbytes); + + spin_lock_irqsave(&ch->lock, irq_flags); + if (nentries < ch->remote_nentries) { + dev_dbg(xpc_chan, "nentries=%d remote_nentries=%d, " + "partid=%d, channel=%d\n", nentries, + ch->remote_nentries, ch->partid, ch->number); + + ch->remote_nentries = nentries; + } + spin_unlock_irqrestore(&ch->lock, irq_flags); + return xpcSuccess; + } + + dev_dbg(xpc_chan, "can't get memory for cached remote message queue, " + "partid=%d, channel=%d\n", ch->partid, ch->number); + return xpcNoMemory; +} + + +/* + * Allocate message queues and other stuff associated with a channel. + * + * Note: Assumes all of the channel sizes are filled in. + */ +static enum xpc_retval +xpc_allocate_msgqueues(struct xpc_channel *ch) +{ + unsigned long irq_flags; + int i; + enum xpc_retval ret; + + + DBUG_ON(ch->flags & XPC_C_SETUP); + + if ((ret = xpc_allocate_local_msgqueue(ch)) != xpcSuccess) { + return ret; + } + + if ((ret = xpc_allocate_remote_msgqueue(ch)) != xpcSuccess) { + kfree(ch->local_msgqueue_base); + ch->local_msgqueue = NULL; + kfree(ch->notify_queue); + ch->notify_queue = NULL; + return ret; + } + + for (i = 0; i < ch->local_nentries; i++) { + /* use a semaphore as an event wait queue */ + sema_init(&ch->notify_queue[i].sema, 0); + } + + sema_init(&ch->teardown_sema, 0); /* event wait */ + + spin_lock_irqsave(&ch->lock, irq_flags); + ch->flags |= XPC_C_SETUP; + spin_unlock_irqrestore(&ch->lock, irq_flags); + + return xpcSuccess; +} + + +/* + * Process a connect message from a remote partition. + * + * Note: xpc_process_connect() is expecting to be called with the + * spin_lock_irqsave held and will leave it locked upon return. + */ +static void +xpc_process_connect(struct xpc_channel *ch, unsigned long *irq_flags) +{ + enum xpc_retval ret; + + + DBUG_ON(!spin_is_locked(&ch->lock)); + + if (!(ch->flags & XPC_C_OPENREQUEST) || + !(ch->flags & XPC_C_ROPENREQUEST)) { + /* nothing more to do for now */ + return; + } + DBUG_ON(!(ch->flags & XPC_C_CONNECTING)); + + if (!(ch->flags & XPC_C_SETUP)) { + spin_unlock_irqrestore(&ch->lock, *irq_flags); + ret = xpc_allocate_msgqueues(ch); + spin_lock_irqsave(&ch->lock, *irq_flags); + + if (ret != xpcSuccess) { + XPC_DISCONNECT_CHANNEL(ch, ret, irq_flags); + } + if (ch->flags & (XPC_C_CONNECTED | XPC_C_DISCONNECTING)) { + return; + } + + DBUG_ON(!(ch->flags & XPC_C_SETUP)); + DBUG_ON(ch->local_msgqueue == NULL); + DBUG_ON(ch->remote_msgqueue == NULL); + } + + if (!(ch->flags & XPC_C_OPENREPLY)) { + ch->flags |= XPC_C_OPENREPLY; + xpc_IPI_send_openreply(ch, irq_flags); + } + + if (!(ch->flags & XPC_C_ROPENREPLY)) { + return; + } + + DBUG_ON(ch->remote_msgqueue_pa == 0); + + ch->flags = (XPC_C_CONNECTED | XPC_C_SETUP); /* clear all else */ + + dev_info(xpc_chan, "channel %d to partition %d connected\n", + ch->number, ch->partid); + + spin_unlock_irqrestore(&ch->lock, *irq_flags); + xpc_create_kthreads(ch, 1); + spin_lock_irqsave(&ch->lock, *irq_flags); +} + + +/* + * Free up message queues and other stuff that were allocated for the specified + * channel. + * + * Note: ch->reason and ch->reason_line are left set for debugging purposes, + * they're cleared when XPC_C_DISCONNECTED is cleared. + */ +static void +xpc_free_msgqueues(struct xpc_channel *ch) +{ + DBUG_ON(!spin_is_locked(&ch->lock)); + DBUG_ON(atomic_read(&ch->n_to_notify) != 0); + + ch->remote_msgqueue_pa = 0; + ch->func = NULL; + ch->key = NULL; + ch->msg_size = 0; + ch->local_nentries = 0; + ch->remote_nentries = 0; + ch->kthreads_assigned_limit = 0; + ch->kthreads_idle_limit = 0; + + ch->local_GP->get = 0; + ch->local_GP->put = 0; + ch->remote_GP.get = 0; + ch->remote_GP.put = 0; + ch->w_local_GP.get = 0; + ch->w_local_GP.put = 0; + ch->w_remote_GP.get = 0; + ch->w_remote_GP.put = 0; + ch->next_msg_to_pull = 0; + + if (ch->flags & XPC_C_SETUP) { + ch->flags &= ~XPC_C_SETUP; + + dev_dbg(xpc_chan, "ch->flags=0x%x, partid=%d, channel=%d\n", + ch->flags, ch->partid, ch->number); + + kfree(ch->local_msgqueue_base); + ch->local_msgqueue = NULL; + kfree(ch->remote_msgqueue_base); + ch->remote_msgqueue = NULL; + kfree(ch->notify_queue); + ch->notify_queue = NULL; + + /* in case someone is waiting for the teardown to complete */ + up(&ch->teardown_sema); + } +} + + +/* + * spin_lock_irqsave() is expected to be held on entry. + */ +static void +xpc_process_disconnect(struct xpc_channel *ch, unsigned long *irq_flags) +{ + struct xpc_partition *part = &xpc_partitions[ch->partid]; + u32 ch_flags = ch->flags; + + + DBUG_ON(!spin_is_locked(&ch->lock)); + + if (!(ch->flags & XPC_C_DISCONNECTING)) { + return; + } + + DBUG_ON(!(ch->flags & XPC_C_CLOSEREQUEST)); + + /* make sure all activity has settled down first */ + + if (atomic_read(&ch->references) > 0) { + return; + } + DBUG_ON(atomic_read(&ch->kthreads_assigned) != 0); + + /* it's now safe to free the channel's message queues */ + + xpc_free_msgqueues(ch); + DBUG_ON(ch->flags & XPC_C_SETUP); + + if (part->act_state != XPC_P_DEACTIVATING) { + + /* as long as the other side is up do the full protocol */ + + if (!(ch->flags & XPC_C_RCLOSEREQUEST)) { + return; + } + + if (!(ch->flags & XPC_C_CLOSEREPLY)) { + ch->flags |= XPC_C_CLOSEREPLY; + xpc_IPI_send_closereply(ch, irq_flags); + } + + if (!(ch->flags & XPC_C_RCLOSEREPLY)) { + return; + } + } + + /* both sides are disconnected now */ + + ch->flags = XPC_C_DISCONNECTED; /* clear all flags, but this one */ + + atomic_dec(&part->nchannels_active); + + if (ch_flags & XPC_C_WASCONNECTED) { + dev_info(xpc_chan, "channel %d to partition %d disconnected, " + "reason=%d\n", ch->number, ch->partid, ch->reason); + } +} + + +/* + * Process a change in the channel's remote connection state. + */ +static void +xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, + u8 IPI_flags) +{ + unsigned long irq_flags; + struct xpc_openclose_args *args = + &part->remote_openclose_args[ch_number]; + struct xpc_channel *ch = &part->channels[ch_number]; + enum xpc_retval reason; + + + + spin_lock_irqsave(&ch->lock, irq_flags); + + + if (IPI_flags & XPC_IPI_CLOSEREQUEST) { + + dev_dbg(xpc_chan, "XPC_IPI_CLOSEREQUEST (reason=%d) received " + "from partid=%d, channel=%d\n", args->reason, + ch->partid, ch->number); + + /* + * If RCLOSEREQUEST is set, we're probably waiting for + * RCLOSEREPLY. We should find it and a ROPENREQUEST packed + * with this RCLOSEQREUQEST in the IPI_flags. + */ + + if (ch->flags & XPC_C_RCLOSEREQUEST) { + DBUG_ON(!(ch->flags & XPC_C_DISCONNECTING)); + DBUG_ON(!(ch->flags & XPC_C_CLOSEREQUEST)); + DBUG_ON(!(ch->flags & XPC_C_CLOSEREPLY)); + DBUG_ON(ch->flags & XPC_C_RCLOSEREPLY); + + DBUG_ON(!(IPI_flags & XPC_IPI_CLOSEREPLY)); + IPI_flags &= ~XPC_IPI_CLOSEREPLY; + ch->flags |= XPC_C_RCLOSEREPLY; + + /* both sides have finished disconnecting */ + xpc_process_disconnect(ch, &irq_flags); + } + + if (ch->flags & XPC_C_DISCONNECTED) { + // >>> explain this section + + if (!(IPI_flags & XPC_IPI_OPENREQUEST)) { + DBUG_ON(part->act_state != + XPC_P_DEACTIVATING); + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } + + XPC_SET_REASON(ch, 0, 0); + ch->flags &= ~XPC_C_DISCONNECTED; + + atomic_inc(&part->nchannels_active); + ch->flags |= (XPC_C_CONNECTING | XPC_C_ROPENREQUEST); + } + + IPI_flags &= ~(XPC_IPI_OPENREQUEST | XPC_IPI_OPENREPLY); + + /* + * The meaningful CLOSEREQUEST connection state fields are: + * reason = reason connection is to be closed + */ + + ch->flags |= XPC_C_RCLOSEREQUEST; + + if (!(ch->flags & XPC_C_DISCONNECTING)) { + reason = args->reason; + if (reason <= xpcSuccess || reason > xpcUnknownReason) { + reason = xpcUnknownReason; + } else if (reason == xpcUnregistering) { + reason = xpcOtherUnregistering; + } + + XPC_DISCONNECT_CHANNEL(ch, reason, &irq_flags); + } else { + xpc_process_disconnect(ch, &irq_flags); + } + } + + + if (IPI_flags & XPC_IPI_CLOSEREPLY) { + + dev_dbg(xpc_chan, "XPC_IPI_CLOSEREPLY received from partid=%d," + " channel=%d\n", ch->partid, ch->number); + + if (ch->flags & XPC_C_DISCONNECTED) { + DBUG_ON(part->act_state != XPC_P_DEACTIVATING); + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } + + DBUG_ON(!(ch->flags & XPC_C_CLOSEREQUEST)); + DBUG_ON(!(ch->flags & XPC_C_RCLOSEREQUEST)); + + ch->flags |= XPC_C_RCLOSEREPLY; + + if (ch->flags & XPC_C_CLOSEREPLY) { + /* both sides have finished disconnecting */ + xpc_process_disconnect(ch, &irq_flags); + } + } + + + if (IPI_flags & XPC_IPI_OPENREQUEST) { + + dev_dbg(xpc_chan, "XPC_IPI_OPENREQUEST (msg_size=%d, " + "local_nentries=%d) received from partid=%d, " + "channel=%d\n", args->msg_size, args->local_nentries, + ch->partid, ch->number); + + if ((ch->flags & XPC_C_DISCONNECTING) || + part->act_state == XPC_P_DEACTIVATING) { + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } + DBUG_ON(!(ch->flags & (XPC_C_DISCONNECTED | + XPC_C_OPENREQUEST))); + DBUG_ON(ch->flags & (XPC_C_ROPENREQUEST | XPC_C_ROPENREPLY | + XPC_C_OPENREPLY | XPC_C_CONNECTED)); + + /* + * The meaningful OPENREQUEST connection state fields are: + * msg_size = size of channel's messages in bytes + * local_nentries = remote partition's local_nentries + */ + DBUG_ON(args->msg_size == 0); + DBUG_ON(args->local_nentries == 0); + + ch->flags |= (XPC_C_ROPENREQUEST | XPC_C_CONNECTING); + ch->remote_nentries = args->local_nentries; + + + if (ch->flags & XPC_C_OPENREQUEST) { + if (args->msg_size != ch->msg_size) { + XPC_DISCONNECT_CHANNEL(ch, xpcUnequalMsgSizes, + &irq_flags); + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } + } else { + ch->msg_size = args->msg_size; + + XPC_SET_REASON(ch, 0, 0); + ch->flags &= ~XPC_C_DISCONNECTED; + + atomic_inc(&part->nchannels_active); + } + + xpc_process_connect(ch, &irq_flags); + } + + + if (IPI_flags & XPC_IPI_OPENREPLY) { + + dev_dbg(xpc_chan, "XPC_IPI_OPENREPLY (local_msgqueue_pa=0x%lx, " + "local_nentries=%d, remote_nentries=%d) received from " + "partid=%d, channel=%d\n", args->local_msgqueue_pa, + args->local_nentries, args->remote_nentries, + ch->partid, ch->number); + + if (ch->flags & (XPC_C_DISCONNECTING | XPC_C_DISCONNECTED)) { + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } + DBUG_ON(!(ch->flags & XPC_C_OPENREQUEST)); + DBUG_ON(!(ch->flags & XPC_C_ROPENREQUEST)); + DBUG_ON(ch->flags & XPC_C_CONNECTED); + + /* + * The meaningful OPENREPLY connection state fields are: + * local_msgqueue_pa = physical address of remote + * partition's local_msgqueue + * local_nentries = remote partition's local_nentries + * remote_nentries = remote partition's remote_nentries + */ + DBUG_ON(args->local_msgqueue_pa == 0); + DBUG_ON(args->local_nentries == 0); + DBUG_ON(args->remote_nentries == 0); + + ch->flags |= XPC_C_ROPENREPLY; + ch->remote_msgqueue_pa = args->local_msgqueue_pa; + + if (args->local_nentries < ch->remote_nentries) { + dev_dbg(xpc_chan, "XPC_IPI_OPENREPLY: new " + "remote_nentries=%d, old remote_nentries=%d, " + "partid=%d, channel=%d\n", + args->local_nentries, ch->remote_nentries, + ch->partid, ch->number); + + ch->remote_nentries = args->local_nentries; + } + if (args->remote_nentries < ch->local_nentries) { + dev_dbg(xpc_chan, "XPC_IPI_OPENREPLY: new " + "local_nentries=%d, old local_nentries=%d, " + "partid=%d, channel=%d\n", + args->remote_nentries, ch->local_nentries, + ch->partid, ch->number); + + ch->local_nentries = args->remote_nentries; + } + + xpc_process_connect(ch, &irq_flags); + } + + spin_unlock_irqrestore(&ch->lock, irq_flags); +} + + +/* + * Attempt to establish a channel connection to a remote partition. + */ +static enum xpc_retval +xpc_connect_channel(struct xpc_channel *ch) +{ + unsigned long irq_flags; + struct xpc_registration *registration = &xpc_registrations[ch->number]; + + + if (down_interruptible(®istration->sema) != 0) { + return xpcInterrupted; + } + + if (!XPC_CHANNEL_REGISTERED(ch->number)) { + up(®istration->sema); + return xpcUnregistered; + } + + spin_lock_irqsave(&ch->lock, irq_flags); + + DBUG_ON(ch->flags & XPC_C_CONNECTED); + DBUG_ON(ch->flags & XPC_C_OPENREQUEST); + + if (ch->flags & XPC_C_DISCONNECTING) { + spin_unlock_irqrestore(&ch->lock, irq_flags); + up(®istration->sema); + return ch->reason; + } + + + /* add info from the channel connect registration to the channel */ + + ch->kthreads_assigned_limit = registration->assigned_limit; + ch->kthreads_idle_limit = registration->idle_limit; + DBUG_ON(atomic_read(&ch->kthreads_assigned) != 0); + DBUG_ON(atomic_read(&ch->kthreads_idle) != 0); + DBUG_ON(atomic_read(&ch->kthreads_active) != 0); + + ch->func = registration->func; + DBUG_ON(registration->func == NULL); + ch->key = registration->key; + + ch->local_nentries = registration->nentries; + + if (ch->flags & XPC_C_ROPENREQUEST) { + if (registration->msg_size != ch->msg_size) { + /* the local and remote sides aren't the same */ + + /* + * Because XPC_DISCONNECT_CHANNEL() can block we're + * forced to up the registration sema before we unlock + * the channel lock. But that's okay here because we're + * done with the part that required the registration + * sema. XPC_DISCONNECT_CHANNEL() requires that the + * channel lock be locked and will unlock and relock + * the channel lock as needed. + */ + up(®istration->sema); + XPC_DISCONNECT_CHANNEL(ch, xpcUnequalMsgSizes, + &irq_flags); + spin_unlock_irqrestore(&ch->lock, irq_flags); + return xpcUnequalMsgSizes; + } + } else { + ch->msg_size = registration->msg_size; + + XPC_SET_REASON(ch, 0, 0); + ch->flags &= ~XPC_C_DISCONNECTED; + + atomic_inc(&xpc_partitions[ch->partid].nchannels_active); + } + + up(®istration->sema); + + + /* initiate the connection */ + + ch->flags |= (XPC_C_OPENREQUEST | XPC_C_CONNECTING); + xpc_IPI_send_openrequest(ch, &irq_flags); + + xpc_process_connect(ch, &irq_flags); + + spin_unlock_irqrestore(&ch->lock, irq_flags); + + return xpcSuccess; +} + + +/* + * Notify those who wanted to be notified upon delivery of their message. + */ +static void +xpc_notify_senders(struct xpc_channel *ch, enum xpc_retval reason, s64 put) +{ + struct xpc_notify *notify; + u8 notify_type; + s64 get = ch->w_remote_GP.get - 1; + + + while (++get < put && atomic_read(&ch->n_to_notify) > 0) { + + notify = &ch->notify_queue[get % ch->local_nentries]; + + /* + * See if the notify entry indicates it was associated with + * a message who's sender wants to be notified. It is possible + * that it is, but someone else is doing or has done the + * notification. + */ + notify_type = notify->type; + if (notify_type == 0 || + cmpxchg(¬ify->type, notify_type, 0) != + notify_type) { + continue; + } + + DBUG_ON(notify_type != XPC_N_CALL); + + atomic_dec(&ch->n_to_notify); + + if (notify->func != NULL) { + dev_dbg(xpc_chan, "notify->func() called, notify=0x%p, " + "msg_number=%ld, partid=%d, channel=%d\n", + (void *) notify, get, ch->partid, ch->number); + + notify->func(reason, ch->partid, ch->number, + notify->key); + + dev_dbg(xpc_chan, "notify->func() returned, " + "notify=0x%p, msg_number=%ld, partid=%d, " + "channel=%d\n", (void *) notify, get, + ch->partid, ch->number); + } + } +} + + +/* + * Clear some of the msg flags in the local message queue. + */ +static inline void +xpc_clear_local_msgqueue_flags(struct xpc_channel *ch) +{ + struct xpc_msg *msg; + s64 get; + + + get = ch->w_remote_GP.get; + do { + msg = (struct xpc_msg *) ((u64) ch->local_msgqueue + + (get % ch->local_nentries) * ch->msg_size); + msg->flags = 0; + } while (++get < (volatile s64) ch->remote_GP.get); +} + + +/* + * Clear some of the msg flags in the remote message queue. + */ +static inline void +xpc_clear_remote_msgqueue_flags(struct xpc_channel *ch) +{ + struct xpc_msg *msg; + s64 put; + + + put = ch->w_remote_GP.put; + do { + msg = (struct xpc_msg *) ((u64) ch->remote_msgqueue + + (put % ch->remote_nentries) * ch->msg_size); + msg->flags = 0; + } while (++put < (volatile s64) ch->remote_GP.put); +} + + +static void +xpc_process_msg_IPI(struct xpc_partition *part, int ch_number) +{ + struct xpc_channel *ch = &part->channels[ch_number]; + int nmsgs_sent; + + + ch->remote_GP = part->remote_GPs[ch_number]; + + + /* See what, if anything, has changed for each connected channel */ + + xpc_msgqueue_ref(ch); + + if (ch->w_remote_GP.get == ch->remote_GP.get && + ch->w_remote_GP.put == ch->remote_GP.put) { + /* nothing changed since GPs were last pulled */ + xpc_msgqueue_deref(ch); + return; + } + + if (!(ch->flags & XPC_C_CONNECTED)){ + xpc_msgqueue_deref(ch); + return; + } + + + /* + * First check to see if messages recently sent by us have been + * received by the other side. (The remote GET value will have + * changed since we last looked at it.) + */ + + if (ch->w_remote_GP.get != ch->remote_GP.get) { + + /* + * We need to notify any senders that want to be notified + * that their sent messages have been received by their + * intended recipients. We need to do this before updating + * w_remote_GP.get so that we don't allocate the same message + * queue entries prematurely (see xpc_allocate_msg()). + */ + if (atomic_read(&ch->n_to_notify) > 0) { + /* + * Notify senders that messages sent have been + * received and delivered by the other side. + */ + xpc_notify_senders(ch, xpcMsgDelivered, + ch->remote_GP.get); + } + + /* + * Clear msg->flags in previously sent messages, so that + * they're ready for xpc_allocate_msg(). + */ + xpc_clear_local_msgqueue_flags(ch); + + (volatile s64) ch->w_remote_GP.get = ch->remote_GP.get; + + dev_dbg(xpc_chan, "w_remote_GP.get changed to %ld, partid=%d, " + "channel=%d\n", ch->w_remote_GP.get, ch->partid, + ch->number); + + /* + * If anyone was waiting for message queue entries to become + * available, wake them up. + */ + if (atomic_read(&ch->n_on_msg_allocate_wq) > 0) { + wake_up(&ch->msg_allocate_wq); + } + } + + + /* + * Now check for newly sent messages by the other side. (The remote + * PUT value will have changed since we last looked at it.) + */ + + if (ch->w_remote_GP.put != ch->remote_GP.put) { + /* + * Clear msg->flags in previously received messages, so that + * they're ready for xpc_get_deliverable_msg(). + */ + xpc_clear_remote_msgqueue_flags(ch); + + (volatile s64) ch->w_remote_GP.put = ch->remote_GP.put; + + dev_dbg(xpc_chan, "w_remote_GP.put changed to %ld, partid=%d, " + "channel=%d\n", ch->w_remote_GP.put, ch->partid, + ch->number); + + nmsgs_sent = ch->w_remote_GP.put - ch->w_local_GP.get; + if (nmsgs_sent > 0) { + dev_dbg(xpc_chan, "msgs waiting to be copied and " + "delivered=%d, partid=%d, channel=%d\n", + nmsgs_sent, ch->partid, ch->number); + + if (ch->flags & XPC_C_CONNECTCALLOUT) { + xpc_activate_kthreads(ch, nmsgs_sent); + } + } + } + + xpc_msgqueue_deref(ch); +} + + +void +xpc_process_channel_activity(struct xpc_partition *part) +{ + unsigned long irq_flags; + u64 IPI_amo, IPI_flags; + struct xpc_channel *ch; + int ch_number; + + + IPI_amo = xpc_get_IPI_flags(part); + + /* + * Initiate channel connections for registered channels. + * + * For each connected channel that has pending messages activate idle + * kthreads and/or create new kthreads as needed. + */ + + for (ch_number = 0; ch_number < part->nchannels; ch_number++) { + ch = &part->channels[ch_number]; + + + /* + * Process any open or close related IPI flags, and then deal + * with connecting or disconnecting the channel as required. + */ + + IPI_flags = XPC_GET_IPI_FLAGS(IPI_amo, ch_number); + + if (XPC_ANY_OPENCLOSE_IPI_FLAGS_SET(IPI_flags)) { + xpc_process_openclose_IPI(part, ch_number, IPI_flags); + } + + + if (ch->flags & XPC_C_DISCONNECTING) { + spin_lock_irqsave(&ch->lock, irq_flags); + xpc_process_disconnect(ch, &irq_flags); + spin_unlock_irqrestore(&ch->lock, irq_flags); + continue; + } + + if (part->act_state == XPC_P_DEACTIVATING) { + continue; + } + + if (!(ch->flags & XPC_C_CONNECTED)) { + if (!(ch->flags & XPC_C_OPENREQUEST)) { + DBUG_ON(ch->flags & XPC_C_SETUP); + (void) xpc_connect_channel(ch); + } else { + spin_lock_irqsave(&ch->lock, irq_flags); + xpc_process_connect(ch, &irq_flags); + spin_unlock_irqrestore(&ch->lock, irq_flags); + } + continue; + } + + + /* + * Process any message related IPI flags, this may involve the + * activation of kthreads to deliver any pending messages sent + * from the other partition. + */ + + if (XPC_ANY_MSG_IPI_FLAGS_SET(IPI_flags)) { + xpc_process_msg_IPI(part, ch_number); + } + } +} + + +/* + * XPC's heartbeat code calls this function to inform XPC that a partition has + * gone down. XPC responds by tearing down the XPartition Communication + * infrastructure used for the just downed partition. + * + * XPC's heartbeat code will never call this function and xpc_partition_up() + * at the same time. Nor will it ever make multiple calls to either function + * at the same time. + */ +void +xpc_partition_down(struct xpc_partition *part, enum xpc_retval reason) +{ + unsigned long irq_flags; + int ch_number; + struct xpc_channel *ch; + + + dev_dbg(xpc_chan, "deactivating partition %d, reason=%d\n", + XPC_PARTID(part), reason); + + if (!xpc_part_ref(part)) { + /* infrastructure for this partition isn't currently set up */ + return; + } + + + /* disconnect all channels associated with the downed partition */ + + for (ch_number = 0; ch_number < part->nchannels; ch_number++) { + ch = &part->channels[ch_number]; + + + xpc_msgqueue_ref(ch); + spin_lock_irqsave(&ch->lock, irq_flags); + + XPC_DISCONNECT_CHANNEL(ch, reason, &irq_flags); + + spin_unlock_irqrestore(&ch->lock, irq_flags); + xpc_msgqueue_deref(ch); + } + + xpc_wakeup_channel_mgr(part); + + xpc_part_deref(part); +} + + +/* + * Teardown the infrastructure necessary to support XPartition Communication + * between the specified remote partition and the local one. + */ +void +xpc_teardown_infrastructure(struct xpc_partition *part) +{ + partid_t partid = XPC_PARTID(part); + + + /* + * We start off by making this partition inaccessible to local + * processes by marking it as no longer setup. Then we make it + * inaccessible to remote processes by clearing the XPC per partition + * specific variable's magic # (which indicates that these variables + * are no longer valid) and by ignoring all XPC notify IPIs sent to + * this partition. + */ + + DBUG_ON(atomic_read(&part->nchannels_active) != 0); + DBUG_ON(part->setup_state != XPC_P_SETUP); + part->setup_state = XPC_P_WTEARDOWN; + + xpc_vars_part[partid].magic = 0; + + + free_irq(SGI_XPC_NOTIFY, (void *) (u64) partid); + + + /* + * Before proceding with the teardown we have to wait until all + * existing references cease. + */ + wait_event(part->teardown_wq, (atomic_read(&part->references) == 0)); + + + /* now we can begin tearing down the infrastructure */ + + part->setup_state = XPC_P_TORNDOWN; + + /* in case we've still got outstanding timers registered... */ + del_timer_sync(&part->dropped_IPI_timer); + + kfree(part->remote_openclose_args_base); + part->remote_openclose_args = NULL; + kfree(part->local_openclose_args_base); + part->local_openclose_args = NULL; + kfree(part->remote_GPs_base); + part->remote_GPs = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; + part->local_IPI_amo_va = NULL; +} + + +/* + * Called by XP at the time of channel connection registration to cause + * XPC to establish connections to all currently active partitions. + */ +void +xpc_initiate_connect(int ch_number) +{ + partid_t partid; + struct xpc_partition *part; + struct xpc_channel *ch; + + + DBUG_ON(ch_number < 0 || ch_number >= XPC_NCHANNELS); + + for (partid = 1; partid < XP_MAX_PARTITIONS; partid++) { + part = &xpc_partitions[partid]; + + if (xpc_part_ref(part)) { + ch = &part->channels[ch_number]; + + if (!(ch->flags & XPC_C_DISCONNECTING)) { + DBUG_ON(ch->flags & XPC_C_OPENREQUEST); + DBUG_ON(ch->flags & XPC_C_CONNECTED); + DBUG_ON(ch->flags & XPC_C_SETUP); + + /* + * Initiate the establishment of a connection + * on the newly registered channel to the + * remote partition. + */ + xpc_wakeup_channel_mgr(part); + } + + xpc_part_deref(part); + } + } +} + + +void +xpc_connected_callout(struct xpc_channel *ch) +{ + unsigned long irq_flags; + + + /* let the registerer know that a connection has been established */ + + if (ch->func != NULL) { + dev_dbg(xpc_chan, "ch->func() called, reason=xpcConnected, " + "partid=%d, channel=%d\n", ch->partid, ch->number); + + ch->func(xpcConnected, ch->partid, ch->number, + (void *) (u64) ch->local_nentries, ch->key); + + dev_dbg(xpc_chan, "ch->func() returned, reason=xpcConnected, " + "partid=%d, channel=%d\n", ch->partid, ch->number); + } + + spin_lock_irqsave(&ch->lock, irq_flags); + ch->flags |= XPC_C_CONNECTCALLOUT; + spin_unlock_irqrestore(&ch->lock, irq_flags); +} + + +/* + * Called by XP at the time of channel connection unregistration to cause + * XPC to teardown all current connections for the specified channel. + * + * Before returning xpc_initiate_disconnect() will wait until all connections + * on the specified channel have been closed/torndown. So the caller can be + * assured that they will not be receiving any more callouts from XPC to the + * function they registered via xpc_connect(). + * + * Arguments: + * + * ch_number - channel # to unregister. + */ +void +xpc_initiate_disconnect(int ch_number) +{ + unsigned long irq_flags; + partid_t partid; + struct xpc_partition *part; + struct xpc_channel *ch; + + + DBUG_ON(ch_number < 0 || ch_number >= XPC_NCHANNELS); + + /* initiate the channel disconnect for every active partition */ + for (partid = 1; partid < XP_MAX_PARTITIONS; partid++) { + part = &xpc_partitions[partid]; + + if (xpc_part_ref(part)) { + ch = &part->channels[ch_number]; + xpc_msgqueue_ref(ch); + + spin_lock_irqsave(&ch->lock, irq_flags); + + XPC_DISCONNECT_CHANNEL(ch, xpcUnregistering, + &irq_flags); + + spin_unlock_irqrestore(&ch->lock, irq_flags); + + xpc_msgqueue_deref(ch); + xpc_part_deref(part); + } + } + + xpc_disconnect_wait(ch_number); +} + + +/* + * To disconnect a channel, and reflect it back to all who may be waiting. + * + * >>> An OPEN is not allowed until XPC_C_DISCONNECTING is cleared by + * >>> xpc_free_msgqueues(). + * + * THE CHANNEL IS TO BE LOCKED BY THE CALLER AND WILL REMAIN LOCKED UPON RETURN. + */ +void +xpc_disconnect_channel(const int line, struct xpc_channel *ch, + enum xpc_retval reason, unsigned long *irq_flags) +{ + u32 flags; + + + DBUG_ON(!spin_is_locked(&ch->lock)); + + if (ch->flags & (XPC_C_DISCONNECTING | XPC_C_DISCONNECTED)) { + return; + } + DBUG_ON(!(ch->flags & (XPC_C_CONNECTING | XPC_C_CONNECTED))); + + dev_dbg(xpc_chan, "reason=%d, line=%d, partid=%d, channel=%d\n", + reason, line, ch->partid, ch->number); + + XPC_SET_REASON(ch, reason, line); + + flags = ch->flags; + /* some of these may not have been set */ + ch->flags &= ~(XPC_C_OPENREQUEST | XPC_C_OPENREPLY | + XPC_C_ROPENREQUEST | XPC_C_ROPENREPLY | + XPC_C_CONNECTING | XPC_C_CONNECTED); + + ch->flags |= (XPC_C_CLOSEREQUEST | XPC_C_DISCONNECTING); + xpc_IPI_send_closerequest(ch, irq_flags); + + if (flags & XPC_C_CONNECTED) { + ch->flags |= XPC_C_WASCONNECTED; + } + + if (atomic_read(&ch->kthreads_idle) > 0) { + /* wake all idle kthreads so they can exit */ + wake_up_all(&ch->idle_wq); + } + + spin_unlock_irqrestore(&ch->lock, *irq_flags); + + + /* wake those waiting to allocate an entry from the local msg queue */ + + if (atomic_read(&ch->n_on_msg_allocate_wq) > 0) { + wake_up(&ch->msg_allocate_wq); + } + + /* wake those waiting for notify completion */ + + if (atomic_read(&ch->n_to_notify) > 0) { + xpc_notify_senders(ch, reason, ch->w_local_GP.put); + } + + spin_lock_irqsave(&ch->lock, *irq_flags); +} + + +void +xpc_disconnected_callout(struct xpc_channel *ch) +{ + /* + * Let the channel's registerer know that the channel is now + * disconnected. We don't want to do this if the registerer was never + * informed of a connection being made, unless the disconnect was for + * abnormal reasons. + */ + + if (ch->func != NULL) { + dev_dbg(xpc_chan, "ch->func() called, reason=%d, partid=%d, " + "channel=%d\n", ch->reason, ch->partid, ch->number); + + ch->func(ch->reason, ch->partid, ch->number, NULL, ch->key); + + dev_dbg(xpc_chan, "ch->func() returned, reason=%d, partid=%d, " + "channel=%d\n", ch->reason, ch->partid, ch->number); + } +} + + +/* + * Wait for a message entry to become available for the specified channel, + * but don't wait any longer than 1 jiffy. + */ +static enum xpc_retval +xpc_allocate_msg_wait(struct xpc_channel *ch) +{ + enum xpc_retval ret; + + + if (ch->flags & XPC_C_DISCONNECTING) { + DBUG_ON(ch->reason == xpcInterrupted); // >>> Is this true? + return ch->reason; + } + + atomic_inc(&ch->n_on_msg_allocate_wq); + ret = interruptible_sleep_on_timeout(&ch->msg_allocate_wq, 1); + atomic_dec(&ch->n_on_msg_allocate_wq); + + if (ch->flags & XPC_C_DISCONNECTING) { + ret = ch->reason; + DBUG_ON(ch->reason == xpcInterrupted); // >>> Is this true? + } else if (ret == 0) { + ret = xpcTimeout; + } else { + ret = xpcInterrupted; + } + + return ret; +} + + +/* + * Allocate an entry for a message from the message queue associated with the + * specified channel. + */ +static enum xpc_retval +xpc_allocate_msg(struct xpc_channel *ch, u32 flags, + struct xpc_msg **address_of_msg) +{ + struct xpc_msg *msg; + enum xpc_retval ret; + s64 put; + + + /* this reference will be dropped in xpc_send_msg() */ + xpc_msgqueue_ref(ch); + + if (ch->flags & XPC_C_DISCONNECTING) { + xpc_msgqueue_deref(ch); + return ch->reason; + } + if (!(ch->flags & XPC_C_CONNECTED)) { + xpc_msgqueue_deref(ch); + return xpcNotConnected; + } + + + /* + * Get the next available message entry from the local message queue. + * If none are available, we'll make sure that we grab the latest + * GP values. + */ + ret = xpcTimeout; + + while (1) { + + put = (volatile s64) ch->w_local_GP.put; + if (put - (volatile s64) ch->w_remote_GP.get < + ch->local_nentries) { + + /* There are available message entries. We need to try + * to secure one for ourselves. We'll do this by trying + * to increment w_local_GP.put as long as someone else + * doesn't beat us to it. If they do, we'll have to + * try again. + */ + if (cmpxchg(&ch->w_local_GP.put, put, put + 1) == + put) { + /* we got the entry referenced by put */ + break; + } + continue; /* try again */ + } + + + /* + * There aren't any available msg entries at this time. + * + * In waiting for a message entry to become available, + * we set a timeout in case the other side is not + * sending completion IPIs. This lets us fake an IPI + * that will cause the IPI handler to fetch the latest + * GP values as if an IPI was sent by the other side. + */ + if (ret == xpcTimeout) { + xpc_IPI_send_local_msgrequest(ch); + } + + if (flags & XPC_NOWAIT) { + xpc_msgqueue_deref(ch); + return xpcNoWait; + } + + ret = xpc_allocate_msg_wait(ch); + if (ret != xpcInterrupted && ret != xpcTimeout) { + xpc_msgqueue_deref(ch); + return ret; + } + } + + + /* get the message's address and initialize it */ + msg = (struct xpc_msg *) ((u64) ch->local_msgqueue + + (put % ch->local_nentries) * ch->msg_size); + + + DBUG_ON(msg->flags != 0); + msg->number = put; + + dev_dbg(xpc_chan, "w_local_GP.put changed to %ld; msg=0x%p, " + "msg_number=%ld, partid=%d, channel=%d\n", put + 1, + (void *) msg, msg->number, ch->partid, ch->number); + + *address_of_msg = msg; + + return xpcSuccess; +} + + +/* + * Allocate an entry for a message from the message queue associated with the + * specified channel. NOTE that this routine can sleep waiting for a message + * entry to become available. To not sleep, pass in the XPC_NOWAIT flag. + * + * Arguments: + * + * partid - ID of partition to which the channel is connected. + * ch_number - channel #. + * flags - see xpc.h for valid flags. + * payload - address of the allocated payload area pointer (filled in on + * return) in which the user-defined message is constructed. + */ +enum xpc_retval +xpc_initiate_allocate(partid_t partid, int ch_number, u32 flags, void **payload) +{ + struct xpc_partition *part = &xpc_partitions[partid]; + enum xpc_retval ret = xpcUnknownReason; + struct xpc_msg *msg; + + + DBUG_ON(partid <= 0 || partid >= XP_MAX_PARTITIONS); + DBUG_ON(ch_number < 0 || ch_number >= part->nchannels); + + *payload = NULL; + + if (xpc_part_ref(part)) { + ret = xpc_allocate_msg(&part->channels[ch_number], flags, &msg); + xpc_part_deref(part); + + if (msg != NULL) { + *payload = &msg->payload; + } + } + + return ret; +} + + +/* + * Now we actually send the messages that are ready to be sent by advancing + * the local message queue's Put value and then send an IPI to the recipient + * partition. + */ +static void +xpc_send_msgs(struct xpc_channel *ch, s64 initial_put) +{ + struct xpc_msg *msg; + s64 put = initial_put + 1; + int send_IPI = 0; + + + while (1) { + + while (1) { + if (put == (volatile s64) ch->w_local_GP.put) { + break; + } + + msg = (struct xpc_msg *) ((u64) ch->local_msgqueue + + (put % ch->local_nentries) * ch->msg_size); + + if (!(msg->flags & XPC_M_READY)) { + break; + } + + put++; + } + + if (put == initial_put) { + /* nothing's changed */ + break; + } + + if (cmpxchg_rel(&ch->local_GP->put, initial_put, put) != + initial_put) { + /* someone else beat us to it */ + DBUG_ON((volatile s64) ch->local_GP->put < initial_put); + break; + } + + /* we just set the new value of local_GP->put */ + + dev_dbg(xpc_chan, "local_GP->put changed to %ld, partid=%d, " + "channel=%d\n", put, ch->partid, ch->number); + + send_IPI = 1; + + /* + * We need to ensure that the message referenced by + * local_GP->put is not XPC_M_READY or that local_GP->put + * equals w_local_GP.put, so we'll go have a look. + */ + initial_put = put; + } + + if (send_IPI) { + xpc_IPI_send_msgrequest(ch); + } +} + + +/* + * Common code that does the actual sending of the message by advancing the + * local message queue's Put value and sends an IPI to the partition the + * message is being sent to. + */ +static enum xpc_retval +xpc_send_msg(struct xpc_channel *ch, struct xpc_msg *msg, u8 notify_type, + xpc_notify_func func, void *key) +{ + enum xpc_retval ret = xpcSuccess; + struct xpc_notify *notify = NULL; // >>> to keep the compiler happy!! + s64 put, msg_number = msg->number; + + + DBUG_ON(notify_type == XPC_N_CALL && func == NULL); + DBUG_ON((((u64) msg - (u64) ch->local_msgqueue) / ch->msg_size) != + msg_number % ch->local_nentries); + DBUG_ON(msg->flags & XPC_M_READY); + + if (ch->flags & XPC_C_DISCONNECTING) { + /* drop the reference grabbed in xpc_allocate_msg() */ + xpc_msgqueue_deref(ch); + return ch->reason; + } + + if (notify_type != 0) { + /* + * Tell the remote side to send an ACK interrupt when the + * message has been delivered. + */ + msg->flags |= XPC_M_INTERRUPT; + + atomic_inc(&ch->n_to_notify); + + notify = &ch->notify_queue[msg_number % ch->local_nentries]; + notify->func = func; + notify->key = key; + (volatile u8) notify->type = notify_type; + + // >>> is a mb() needed here? + + if (ch->flags & XPC_C_DISCONNECTING) { + /* + * An error occurred between our last error check and + * this one. We will try to clear the type field from + * the notify entry. If we succeed then + * xpc_disconnect_channel() didn't already process + * the notify entry. + */ + if (cmpxchg(¬ify->type, notify_type, 0) == + notify_type) { + atomic_dec(&ch->n_to_notify); + ret = ch->reason; + } + + /* drop the reference grabbed in xpc_allocate_msg() */ + xpc_msgqueue_deref(ch); + return ret; + } + } + + msg->flags |= XPC_M_READY; + + /* + * The preceding store of msg->flags must occur before the following + * load of ch->local_GP->put. + */ + mb(); + + /* see if the message is next in line to be sent, if so send it */ + + put = ch->local_GP->put; + if (put == msg_number) { + xpc_send_msgs(ch, put); + } + + /* drop the reference grabbed in xpc_allocate_msg() */ + xpc_msgqueue_deref(ch); + return ret; +} + + +/* + * Send a message previously allocated using xpc_initiate_allocate() on the + * specified channel connected to the specified partition. + * + * This routine will not wait for the message to be received, nor will + * notification be given when it does happen. Once this routine has returned + * the message entry allocated via xpc_initiate_allocate() is no longer + * accessable to the caller. + * + * This routine, although called by users, does not call xpc_part_ref() to + * ensure that the partition infrastructure is in place. It relies on the + * fact that we called xpc_msgqueue_ref() in xpc_allocate_msg(). + * + * Arguments: + * + * partid - ID of partition to which the channel is connected. + * ch_number - channel # to send message on. + * payload - pointer to the payload area allocated via + * xpc_initiate_allocate(). + */ +enum xpc_retval +xpc_initiate_send(partid_t partid, int ch_number, void *payload) +{ + struct xpc_partition *part = &xpc_partitions[partid]; + struct xpc_msg *msg = XPC_MSG_ADDRESS(payload); + enum xpc_retval ret; + + + dev_dbg(xpc_chan, "msg=0x%p, partid=%d, channel=%d\n", (void *) msg, + partid, ch_number); + + DBUG_ON(partid <= 0 || partid >= XP_MAX_PARTITIONS); + DBUG_ON(ch_number < 0 || ch_number >= part->nchannels); + DBUG_ON(msg == NULL); + + ret = xpc_send_msg(&part->channels[ch_number], msg, 0, NULL, NULL); + + return ret; +} + + +/* + * Send a message previously allocated using xpc_initiate_allocate on the + * specified channel connected to the specified partition. + * + * This routine will not wait for the message to be sent. Once this routine + * has returned the message entry allocated via xpc_initiate_allocate() is no + * longer accessable to the caller. + * + * Once the remote end of the channel has received the message, the function + * passed as an argument to xpc_initiate_send_notify() will be called. This + * allows the sender to free up or re-use any buffers referenced by the + * message, but does NOT mean the message has been processed at the remote + * end by a receiver. + * + * If this routine returns an error, the caller's function will NOT be called. + * + * This routine, although called by users, does not call xpc_part_ref() to + * ensure that the partition infrastructure is in place. It relies on the + * fact that we called xpc_msgqueue_ref() in xpc_allocate_msg(). + * + * Arguments: + * + * partid - ID of partition to which the channel is connected. + * ch_number - channel # to send message on. + * payload - pointer to the payload area allocated via + * xpc_initiate_allocate(). + * func - function to call with asynchronous notification of message + * receipt. THIS FUNCTION MUST BE NON-BLOCKING. + * key - user-defined key to be passed to the function when it's called. + */ +enum xpc_retval +xpc_initiate_send_notify(partid_t partid, int ch_number, void *payload, + xpc_notify_func func, void *key) +{ + struct xpc_partition *part = &xpc_partitions[partid]; + struct xpc_msg *msg = XPC_MSG_ADDRESS(payload); + enum xpc_retval ret; + + + dev_dbg(xpc_chan, "msg=0x%p, partid=%d, channel=%d\n", (void *) msg, + partid, ch_number); + + DBUG_ON(partid <= 0 || partid >= XP_MAX_PARTITIONS); + DBUG_ON(ch_number < 0 || ch_number >= part->nchannels); + DBUG_ON(msg == NULL); + DBUG_ON(func == NULL); + + ret = xpc_send_msg(&part->channels[ch_number], msg, XPC_N_CALL, + func, key); + return ret; +} + + +static struct xpc_msg * +xpc_pull_remote_msg(struct xpc_channel *ch, s64 get) +{ + struct xpc_partition *part = &xpc_partitions[ch->partid]; + struct xpc_msg *remote_msg, *msg; + u32 msg_index, nmsgs; + u64 msg_offset; + enum xpc_retval ret; + + + if (down_interruptible(&ch->msg_to_pull_sema) != 0) { + /* we were interrupted by a signal */ + return NULL; + } + + while (get >= ch->next_msg_to_pull) { + + /* pull as many messages as are ready and able to be pulled */ + + msg_index = ch->next_msg_to_pull % ch->remote_nentries; + + DBUG_ON(ch->next_msg_to_pull >= + (volatile s64) ch->w_remote_GP.put); + nmsgs = (volatile s64) ch->w_remote_GP.put - + ch->next_msg_to_pull; + if (msg_index + nmsgs > ch->remote_nentries) { + /* ignore the ones that wrap the msg queue for now */ + nmsgs = ch->remote_nentries - msg_index; + } + + msg_offset = msg_index * ch->msg_size; + msg = (struct xpc_msg *) ((u64) ch->remote_msgqueue + + msg_offset); + remote_msg = (struct xpc_msg *) (ch->remote_msgqueue_pa + + msg_offset); + + if ((ret = xpc_pull_remote_cachelines(part, msg, remote_msg, + nmsgs * ch->msg_size)) != xpcSuccess) { + + dev_dbg(xpc_chan, "failed to pull %d msgs starting with" + " msg %ld from partition %d, channel=%d, " + "ret=%d\n", nmsgs, ch->next_msg_to_pull, + ch->partid, ch->number, ret); + + XPC_DEACTIVATE_PARTITION(part, ret); + + up(&ch->msg_to_pull_sema); + return NULL; + } + + mb(); /* >>> this may not be needed, we're not sure */ + + ch->next_msg_to_pull += nmsgs; + } + + up(&ch->msg_to_pull_sema); + + /* return the message we were looking for */ + msg_offset = (get % ch->remote_nentries) * ch->msg_size; + msg = (struct xpc_msg *) ((u64) ch->remote_msgqueue + msg_offset); + + return msg; +} + + +/* + * Get a message to be delivered. + */ +static struct xpc_msg * +xpc_get_deliverable_msg(struct xpc_channel *ch) +{ + struct xpc_msg *msg = NULL; + s64 get; + + + do { + if ((volatile u32) ch->flags & XPC_C_DISCONNECTING) { + break; + } + + get = (volatile s64) ch->w_local_GP.get; + if (get == (volatile s64) ch->w_remote_GP.put) { + break; + } + + /* There are messages waiting to be pulled and delivered. + * We need to try to secure one for ourselves. We'll do this + * by trying to increment w_local_GP.get and hope that no one + * else beats us to it. If they do, we'll we'll simply have + * to try again for the next one. + */ + + if (cmpxchg(&ch->w_local_GP.get, get, get + 1) == get) { + /* we got the entry referenced by get */ + + dev_dbg(xpc_chan, "w_local_GP.get changed to %ld, " + "partid=%d, channel=%d\n", get + 1, + ch->partid, ch->number); + + /* pull the message from the remote partition */ + + msg = xpc_pull_remote_msg(ch, get); + + DBUG_ON(msg != NULL && msg->number != get); + DBUG_ON(msg != NULL && (msg->flags & XPC_M_DONE)); + DBUG_ON(msg != NULL && !(msg->flags & XPC_M_READY)); + + break; + } + + } while (1); + + return msg; +} + + +/* + * Deliver a message to its intended recipient. + */ +void +xpc_deliver_msg(struct xpc_channel *ch) +{ + struct xpc_msg *msg; + + + if ((msg = xpc_get_deliverable_msg(ch)) != NULL) { + + /* + * This ref is taken to protect the payload itself from being + * freed before the user is finished with it, which the user + * indicates by calling xpc_initiate_received(). + */ + xpc_msgqueue_ref(ch); + + atomic_inc(&ch->kthreads_active); + + if (ch->func != NULL) { + dev_dbg(xpc_chan, "ch->func() called, msg=0x%p, " + "msg_number=%ld, partid=%d, channel=%d\n", + (void *) msg, msg->number, ch->partid, + ch->number); + + /* deliver the message to its intended recipient */ + ch->func(xpcMsgReceived, ch->partid, ch->number, + &msg->payload, ch->key); + + dev_dbg(xpc_chan, "ch->func() returned, msg=0x%p, " + "msg_number=%ld, partid=%d, channel=%d\n", + (void *) msg, msg->number, ch->partid, + ch->number); + } + + atomic_dec(&ch->kthreads_active); + } +} + + +/* + * Now we actually acknowledge the messages that have been delivered and ack'd + * by advancing the cached remote message queue's Get value and if requested + * send an IPI to the message sender's partition. + */ +static void +xpc_acknowledge_msgs(struct xpc_channel *ch, s64 initial_get, u8 msg_flags) +{ + struct xpc_msg *msg; + s64 get = initial_get + 1; + int send_IPI = 0; + + + while (1) { + + while (1) { + if (get == (volatile s64) ch->w_local_GP.get) { + break; + } + + msg = (struct xpc_msg *) ((u64) ch->remote_msgqueue + + (get % ch->remote_nentries) * ch->msg_size); + + if (!(msg->flags & XPC_M_DONE)) { + break; + } + + msg_flags |= msg->flags; + get++; + } + + if (get == initial_get) { + /* nothing's changed */ + break; + } + + if (cmpxchg_rel(&ch->local_GP->get, initial_get, get) != + initial_get) { + /* someone else beat us to it */ + DBUG_ON((volatile s64) ch->local_GP->get <= + initial_get); + break; + } + + /* we just set the new value of local_GP->get */ + + dev_dbg(xpc_chan, "local_GP->get changed to %ld, partid=%d, " + "channel=%d\n", get, ch->partid, ch->number); + + send_IPI = (msg_flags & XPC_M_INTERRUPT); + + /* + * We need to ensure that the message referenced by + * local_GP->get is not XPC_M_DONE or that local_GP->get + * equals w_local_GP.get, so we'll go have a look. + */ + initial_get = get; + } + + if (send_IPI) { + xpc_IPI_send_msgrequest(ch); + } +} + + +/* + * Acknowledge receipt of a delivered message. + * + * If a message has XPC_M_INTERRUPT set, send an interrupt to the partition + * that sent the message. + * + * This function, although called by users, does not call xpc_part_ref() to + * ensure that the partition infrastructure is in place. It relies on the + * fact that we called xpc_msgqueue_ref() in xpc_deliver_msg(). + * + * Arguments: + * + * partid - ID of partition to which the channel is connected. + * ch_number - channel # message received on. + * payload - pointer to the payload area allocated via + * xpc_initiate_allocate(). + */ +void +xpc_initiate_received(partid_t partid, int ch_number, void *payload) +{ + struct xpc_partition *part = &xpc_partitions[partid]; + struct xpc_channel *ch; + struct xpc_msg *msg = XPC_MSG_ADDRESS(payload); + s64 get, msg_number = msg->number; + + + DBUG_ON(partid <= 0 || partid >= XP_MAX_PARTITIONS); + DBUG_ON(ch_number < 0 || ch_number >= part->nchannels); + + ch = &part->channels[ch_number]; + + dev_dbg(xpc_chan, "msg=0x%p, msg_number=%ld, partid=%d, channel=%d\n", + (void *) msg, msg_number, ch->partid, ch->number); + + DBUG_ON((((u64) msg - (u64) ch->remote_msgqueue) / ch->msg_size) != + msg_number % ch->remote_nentries); + DBUG_ON(msg->flags & XPC_M_DONE); + + msg->flags |= XPC_M_DONE; + + /* + * The preceding store of msg->flags must occur before the following + * load of ch->local_GP->get. + */ + mb(); + + /* + * See if this message is next in line to be acknowledged as having + * been delivered. + */ + get = ch->local_GP->get; + if (get == msg_number) { + xpc_acknowledge_msgs(ch, get, msg->flags); + } + + /* the call to xpc_msgqueue_ref() was done by xpc_deliver_msg() */ + xpc_msgqueue_deref(ch); +} + diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c new file mode 100644 index 00000000000..177ddb748eb --- /dev/null +++ b/arch/ia64/sn/kernel/xpc_main.c @@ -0,0 +1,1064 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + */ + + +/* + * Cross Partition Communication (XPC) support - standard version. + * + * XPC provides a message passing capability that crosses partition + * boundaries. This module is made up of two parts: + * + * partition This part detects the presence/absence of other + * partitions. It provides a heartbeat and monitors + * the heartbeats of other partitions. + * + * channel This part manages the channels and sends/receives + * messages across them to/from other partitions. + * + * There are a couple of additional functions residing in XP, which + * provide an interface to XPC for its users. + * + * + * Caveats: + * + * . We currently have no way to determine which nasid an IPI came + * from. Thus, xpc_IPI_send() does a remote AMO write followed by + * an IPI. The AMO indicates where data is to be pulled from, so + * after the IPI arrives, the remote partition checks the AMO word. + * The IPI can actually arrive before the AMO however, so other code + * must periodically check for this case. Also, remote AMO operations + * do not reliably time out. Thus we do a remote PIO read solely to + * know whether the remote partition is down and whether we should + * stop sending IPIs to it. This remote PIO read operation is set up + * in a special nofault region so SAL knows to ignore (and cleanup) + * any errors due to the remote AMO write, PIO read, and/or PIO + * write operations. + * + * If/when new hardware solves this IPI problem, we should abandon + * the current approach. + * + */ + + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/sched.h> +#include <linux/syscalls.h> +#include <linux/cache.h> +#include <linux/interrupt.h> +#include <linux/slab.h> +#include <asm/sn/intr.h> +#include <asm/sn/sn_sal.h> +#include <asm/uaccess.h> +#include "xpc.h" + + +/* define two XPC debug device structures to be used with dev_dbg() et al */ + +struct device_driver xpc_dbg_name = { + .name = "xpc" +}; + +struct device xpc_part_dbg_subname = { + .bus_id = {0}, /* set to "part" at xpc_init() time */ + .driver = &xpc_dbg_name +}; + +struct device xpc_chan_dbg_subname = { + .bus_id = {0}, /* set to "chan" at xpc_init() time */ + .driver = &xpc_dbg_name +}; + +struct device *xpc_part = &xpc_part_dbg_subname; +struct device *xpc_chan = &xpc_chan_dbg_subname; + + +/* systune related variables for /proc/sys directories */ + +static int xpc_hb_min = 1; +static int xpc_hb_max = 10; + +static int xpc_hb_check_min = 10; +static int xpc_hb_check_max = 120; + +static ctl_table xpc_sys_xpc_hb_dir[] = { + { + 1, + "hb_interval", + &xpc_hb_interval, + sizeof(int), + 0644, + NULL, + &proc_dointvec_minmax, + &sysctl_intvec, + NULL, + &xpc_hb_min, &xpc_hb_max + }, + { + 2, + "hb_check_interval", + &xpc_hb_check_interval, + sizeof(int), + 0644, + NULL, + &proc_dointvec_minmax, + &sysctl_intvec, + NULL, + &xpc_hb_check_min, &xpc_hb_check_max + }, + {0} +}; +static ctl_table xpc_sys_xpc_dir[] = { + { + 1, + "hb", + NULL, + 0, + 0555, + xpc_sys_xpc_hb_dir + }, + {0} +}; +static ctl_table xpc_sys_dir[] = { + { + 1, + "xpc", + NULL, + 0, + 0555, + xpc_sys_xpc_dir + }, + {0} +}; +static struct ctl_table_header *xpc_sysctl; + + +/* #of IRQs received */ +static atomic_t xpc_act_IRQ_rcvd; + +/* IRQ handler notifies this wait queue on receipt of an IRQ */ +static DECLARE_WAIT_QUEUE_HEAD(xpc_act_IRQ_wq); + +static unsigned long xpc_hb_check_timeout; + +/* xpc_hb_checker thread exited notification */ +static DECLARE_MUTEX_LOCKED(xpc_hb_checker_exited); + +/* xpc_discovery thread exited notification */ +static DECLARE_MUTEX_LOCKED(xpc_discovery_exited); + + +static struct timer_list xpc_hb_timer; + + +static void xpc_kthread_waitmsgs(struct xpc_partition *, struct xpc_channel *); + + +/* + * Notify the heartbeat check thread that an IRQ has been received. + */ +static irqreturn_t +xpc_act_IRQ_handler(int irq, void *dev_id, struct pt_regs *regs) +{ + atomic_inc(&xpc_act_IRQ_rcvd); + wake_up_interruptible(&xpc_act_IRQ_wq); + return IRQ_HANDLED; +} + + +/* + * Timer to produce the heartbeat. The timer structures function is + * already set when this is initially called. A tunable is used to + * specify when the next timeout should occur. + */ +static void +xpc_hb_beater(unsigned long dummy) +{ + xpc_vars->heartbeat++; + + if (jiffies >= xpc_hb_check_timeout) { + wake_up_interruptible(&xpc_act_IRQ_wq); + } + + xpc_hb_timer.expires = jiffies + (xpc_hb_interval * HZ); + add_timer(&xpc_hb_timer); +} + + +/* + * This thread is responsible for nearly all of the partition + * activation/deactivation. + */ +static int +xpc_hb_checker(void *ignore) +{ + int last_IRQ_count = 0; + int new_IRQ_count; + int force_IRQ=0; + + + /* this thread was marked active by xpc_hb_init() */ + + daemonize(XPC_HB_CHECK_THREAD_NAME); + + set_cpus_allowed(current, cpumask_of_cpu(XPC_HB_CHECK_CPU)); + + xpc_hb_check_timeout = jiffies + (xpc_hb_check_interval * HZ); + + while (!(volatile int) xpc_exiting) { + + /* wait for IRQ or timeout */ + (void) wait_event_interruptible(xpc_act_IRQ_wq, + (last_IRQ_count < atomic_read(&xpc_act_IRQ_rcvd) || + jiffies >= xpc_hb_check_timeout || + (volatile int) xpc_exiting)); + + dev_dbg(xpc_part, "woke up with %d ticks rem; %d IRQs have " + "been received\n", + (int) (xpc_hb_check_timeout - jiffies), + atomic_read(&xpc_act_IRQ_rcvd) - last_IRQ_count); + + + /* checking of remote heartbeats is skewed by IRQ handling */ + if (jiffies >= xpc_hb_check_timeout) { + dev_dbg(xpc_part, "checking remote heartbeats\n"); + xpc_check_remote_hb(); + + /* + * We need to periodically recheck to ensure no + * IPI/AMO pairs have been missed. That check + * must always reset xpc_hb_check_timeout. + */ + force_IRQ = 1; + } + + + new_IRQ_count = atomic_read(&xpc_act_IRQ_rcvd); + if (last_IRQ_count < new_IRQ_count || force_IRQ != 0) { + force_IRQ = 0; + + dev_dbg(xpc_part, "found an IRQ to process; will be " + "resetting xpc_hb_check_timeout\n"); + + last_IRQ_count += xpc_identify_act_IRQ_sender(); + if (last_IRQ_count < new_IRQ_count) { + /* retry once to help avoid missing AMO */ + (void) xpc_identify_act_IRQ_sender(); + } + last_IRQ_count = new_IRQ_count; + + xpc_hb_check_timeout = jiffies + + (xpc_hb_check_interval * HZ); + } + } + + dev_dbg(xpc_part, "heartbeat checker is exiting\n"); + + + /* mark this thread as inactive */ + up(&xpc_hb_checker_exited); + return 0; +} + + +/* + * This thread will attempt to discover other partitions to activate + * based on info provided by SAL. This new thread is short lived and + * will exit once discovery is complete. + */ +static int +xpc_initiate_discovery(void *ignore) +{ + daemonize(XPC_DISCOVERY_THREAD_NAME); + + xpc_discovery(); + + dev_dbg(xpc_part, "discovery thread is exiting\n"); + + /* mark this thread as inactive */ + up(&xpc_discovery_exited); + return 0; +} + + +/* + * Establish first contact with the remote partititon. This involves pulling + * the XPC per partition variables from the remote partition and waiting for + * the remote partition to pull ours. + */ +static enum xpc_retval +xpc_make_first_contact(struct xpc_partition *part) +{ + enum xpc_retval ret; + + + while ((ret = xpc_pull_remote_vars_part(part)) != xpcSuccess) { + if (ret != xpcRetry) { + XPC_DEACTIVATE_PARTITION(part, ret); + return ret; + } + + dev_dbg(xpc_chan, "waiting to make first contact with " + "partition %d\n", XPC_PARTID(part)); + + /* wait a 1/4 of a second or so */ + set_current_state(TASK_INTERRUPTIBLE); + (void) schedule_timeout(0.25 * HZ); + + if (part->act_state == XPC_P_DEACTIVATING) { + return part->reason; + } + } + + return xpc_mark_partition_active(part); +} + + +/* + * The first kthread assigned to a newly activated partition is the one + * created by XPC HB with which it calls xpc_partition_up(). XPC hangs on to + * that kthread until the partition is brought down, at which time that kthread + * returns back to XPC HB. (The return of that kthread will signify to XPC HB + * that XPC has dismantled all communication infrastructure for the associated + * partition.) This kthread becomes the channel manager for that partition. + * + * Each active partition has a channel manager, who, besides connecting and + * disconnecting channels, will ensure that each of the partition's connected + * channels has the required number of assigned kthreads to get the work done. + */ +static void +xpc_channel_mgr(struct xpc_partition *part) +{ + while (part->act_state != XPC_P_DEACTIVATING || + atomic_read(&part->nchannels_active) > 0) { + + xpc_process_channel_activity(part); + + + /* + * Wait until we've been requested to activate kthreads or + * all of the channel's message queues have been torn down or + * a signal is pending. + * + * The channel_mgr_requests is set to 1 after being awakened, + * This is done to prevent the channel mgr from making one pass + * through the loop for each request, since he will + * be servicing all the requests in one pass. The reason it's + * set to 1 instead of 0 is so that other kthreads will know + * that the channel mgr is running and won't bother trying to + * wake him up. + */ + atomic_dec(&part->channel_mgr_requests); + (void) wait_event_interruptible(part->channel_mgr_wq, + (atomic_read(&part->channel_mgr_requests) > 0 || + (volatile u64) part->local_IPI_amo != 0 || + ((volatile u8) part->act_state == + XPC_P_DEACTIVATING && + atomic_read(&part->nchannels_active) == 0))); + atomic_set(&part->channel_mgr_requests, 1); + + // >>> Does it need to wakeup periodically as well? In case we + // >>> miscalculated the #of kthreads to wakeup or create? + } +} + + +/* + * When XPC HB determines that a partition has come up, it will create a new + * kthread and that kthread will call this function to attempt to set up the + * basic infrastructure used for Cross Partition Communication with the newly + * upped partition. + * + * The kthread that was created by XPC HB and which setup the XPC + * infrastructure will remain assigned to the partition until the partition + * goes down. At which time the kthread will teardown the XPC infrastructure + * and then exit. + * + * XPC HB will put the remote partition's XPC per partition specific variables + * physical address into xpc_partitions[partid].remote_vars_part_pa prior to + * calling xpc_partition_up(). + */ +static void +xpc_partition_up(struct xpc_partition *part) +{ + DBUG_ON(part->channels != NULL); + + dev_dbg(xpc_chan, "activating partition %d\n", XPC_PARTID(part)); + + if (xpc_setup_infrastructure(part) != xpcSuccess) { + return; + } + + /* + * The kthread that XPC HB called us with will become the + * channel manager for this partition. It will not return + * back to XPC HB until the partition's XPC infrastructure + * has been dismantled. + */ + + (void) xpc_part_ref(part); /* this will always succeed */ + + if (xpc_make_first_contact(part) == xpcSuccess) { + xpc_channel_mgr(part); + } + + xpc_part_deref(part); + + xpc_teardown_infrastructure(part); +} + + +static int +xpc_activating(void *__partid) +{ + partid_t partid = (u64) __partid; + struct xpc_partition *part = &xpc_partitions[partid]; + unsigned long irq_flags; + struct sched_param param = { sched_priority: MAX_USER_RT_PRIO - 1 }; + int ret; + + + DBUG_ON(partid <= 0 || partid >= XP_MAX_PARTITIONS); + + spin_lock_irqsave(&part->act_lock, irq_flags); + + if (part->act_state == XPC_P_DEACTIVATING) { + part->act_state = XPC_P_INACTIVE; + spin_unlock_irqrestore(&part->act_lock, irq_flags); + part->remote_rp_pa = 0; + return 0; + } + + /* indicate the thread is activating */ + DBUG_ON(part->act_state != XPC_P_ACTIVATION_REQ); + part->act_state = XPC_P_ACTIVATING; + + XPC_SET_REASON(part, 0, 0); + spin_unlock_irqrestore(&part->act_lock, irq_flags); + + dev_dbg(xpc_part, "bringing partition %d up\n", partid); + + daemonize("xpc%02d", partid); + + /* + * This thread needs to run at a realtime priority to prevent a + * significant performance degradation. + */ + ret = sched_setscheduler(current, SCHED_FIFO, ¶m); + if (ret != 0) { + dev_warn(xpc_part, "unable to set pid %d to a realtime " + "priority, ret=%d\n", current->pid, ret); + } + + /* allow this thread and its children to run on any CPU */ + set_cpus_allowed(current, CPU_MASK_ALL); + + /* + * Register the remote partition's AMOs with SAL so it can handle + * and cleanup errors within that address range should the remote + * partition go down. We don't unregister this range because it is + * difficult to tell when outstanding writes to the remote partition + * are finished and thus when it is safe to unregister. This should + * not result in wasted space in the SAL xp_addr_region table because + * we should get the same page for remote_amos_page_pa after module + * reloads and system reboots. + */ + if (sn_register_xp_addr_region(part->remote_amos_page_pa, + PAGE_SIZE, 1) < 0) { + dev_warn(xpc_part, "xpc_partition_up(%d) failed to register " + "xp_addr region\n", partid); + + spin_lock_irqsave(&part->act_lock, irq_flags); + part->act_state = XPC_P_INACTIVE; + XPC_SET_REASON(part, xpcPhysAddrRegFailed, __LINE__); + spin_unlock_irqrestore(&part->act_lock, irq_flags); + part->remote_rp_pa = 0; + return 0; + } + + XPC_ALLOW_HB(partid, xpc_vars); + xpc_IPI_send_activated(part); + + + /* + * xpc_partition_up() holds this thread and marks this partition as + * XPC_P_ACTIVE by calling xpc_hb_mark_active(). + */ + (void) xpc_partition_up(part); + + xpc_mark_partition_inactive(part); + + if (part->reason == xpcReactivating) { + /* interrupting ourselves results in activating partition */ + xpc_IPI_send_reactivate(part); + } + + return 0; +} + + +void +xpc_activate_partition(struct xpc_partition *part) +{ + partid_t partid = XPC_PARTID(part); + unsigned long irq_flags; + pid_t pid; + + + spin_lock_irqsave(&part->act_lock, irq_flags); + + pid = kernel_thread(xpc_activating, (void *) ((u64) partid), 0); + + DBUG_ON(part->act_state != XPC_P_INACTIVE); + + if (pid > 0) { + part->act_state = XPC_P_ACTIVATION_REQ; + XPC_SET_REASON(part, xpcCloneKThread, __LINE__); + } else { + XPC_SET_REASON(part, xpcCloneKThreadFailed, __LINE__); + } + + spin_unlock_irqrestore(&part->act_lock, irq_flags); +} + + +/* + * Handle the receipt of a SGI_XPC_NOTIFY IRQ by seeing whether the specified + * partition actually sent it. Since SGI_XPC_NOTIFY IRQs may be shared by more + * than one partition, we use an AMO_t structure per partition to indicate + * whether a partition has sent an IPI or not. >>> If it has, then wake up the + * associated kthread to handle it. + * + * All SGI_XPC_NOTIFY IRQs received by XPC are the result of IPIs sent by XPC + * running on other partitions. + * + * Noteworthy Arguments: + * + * irq - Interrupt ReQuest number. NOT USED. + * + * dev_id - partid of IPI's potential sender. + * + * regs - processor's context before the processor entered + * interrupt code. NOT USED. + */ +irqreturn_t +xpc_notify_IRQ_handler(int irq, void *dev_id, struct pt_regs *regs) +{ + partid_t partid = (partid_t) (u64) dev_id; + struct xpc_partition *part = &xpc_partitions[partid]; + + + DBUG_ON(partid <= 0 || partid >= XP_MAX_PARTITIONS); + + if (xpc_part_ref(part)) { + xpc_check_for_channel_activity(part); + + xpc_part_deref(part); + } + return IRQ_HANDLED; +} + + +/* + * Check to see if xpc_notify_IRQ_handler() dropped any IPIs on the floor + * because the write to their associated IPI amo completed after the IRQ/IPI + * was received. + */ +void +xpc_dropped_IPI_check(struct xpc_partition *part) +{ + if (xpc_part_ref(part)) { + xpc_check_for_channel_activity(part); + + part->dropped_IPI_timer.expires = jiffies + + XPC_P_DROPPED_IPI_WAIT; + add_timer(&part->dropped_IPI_timer); + xpc_part_deref(part); + } +} + + +void +xpc_activate_kthreads(struct xpc_channel *ch, int needed) +{ + int idle = atomic_read(&ch->kthreads_idle); + int assigned = atomic_read(&ch->kthreads_assigned); + int wakeup; + + + DBUG_ON(needed <= 0); + + if (idle > 0) { + wakeup = (needed > idle) ? idle : needed; + needed -= wakeup; + + dev_dbg(xpc_chan, "wakeup %d idle kthreads, partid=%d, " + "channel=%d\n", wakeup, ch->partid, ch->number); + + /* only wakeup the requested number of kthreads */ + wake_up_nr(&ch->idle_wq, wakeup); + } + + if (needed <= 0) { + return; + } + + if (needed + assigned > ch->kthreads_assigned_limit) { + needed = ch->kthreads_assigned_limit - assigned; + // >>>should never be less than 0 + if (needed <= 0) { + return; + } + } + + dev_dbg(xpc_chan, "create %d new kthreads, partid=%d, channel=%d\n", + needed, ch->partid, ch->number); + + xpc_create_kthreads(ch, needed); +} + + +/* + * This function is where XPC's kthreads wait for messages to deliver. + */ +static void +xpc_kthread_waitmsgs(struct xpc_partition *part, struct xpc_channel *ch) +{ + do { + /* deliver messages to their intended recipients */ + + while ((volatile s64) ch->w_local_GP.get < + (volatile s64) ch->w_remote_GP.put && + !((volatile u32) ch->flags & + XPC_C_DISCONNECTING)) { + xpc_deliver_msg(ch); + } + + if (atomic_inc_return(&ch->kthreads_idle) > + ch->kthreads_idle_limit) { + /* too many idle kthreads on this channel */ + atomic_dec(&ch->kthreads_idle); + break; + } + + dev_dbg(xpc_chan, "idle kthread calling " + "wait_event_interruptible_exclusive()\n"); + + (void) wait_event_interruptible_exclusive(ch->idle_wq, + ((volatile s64) ch->w_local_GP.get < + (volatile s64) ch->w_remote_GP.put || + ((volatile u32) ch->flags & + XPC_C_DISCONNECTING))); + + atomic_dec(&ch->kthreads_idle); + + } while (!((volatile u32) ch->flags & XPC_C_DISCONNECTING)); +} + + +static int +xpc_daemonize_kthread(void *args) +{ + partid_t partid = XPC_UNPACK_ARG1(args); + u16 ch_number = XPC_UNPACK_ARG2(args); + struct xpc_partition *part = &xpc_partitions[partid]; + struct xpc_channel *ch; + int n_needed; + + + daemonize("xpc%02dc%d", partid, ch_number); + + dev_dbg(xpc_chan, "kthread starting, partid=%d, channel=%d\n", + partid, ch_number); + + ch = &part->channels[ch_number]; + + if (!(ch->flags & XPC_C_DISCONNECTING)) { + DBUG_ON(!(ch->flags & XPC_C_CONNECTED)); + + /* let registerer know that connection has been established */ + + if (atomic_read(&ch->kthreads_assigned) == 1) { + xpc_connected_callout(ch); + + /* + * It is possible that while the callout was being + * made that the remote partition sent some messages. + * If that is the case, we may need to activate + * additional kthreads to help deliver them. We only + * need one less than total #of messages to deliver. + */ + n_needed = ch->w_remote_GP.put - ch->w_local_GP.get - 1; + if (n_needed > 0 && + !(ch->flags & XPC_C_DISCONNECTING)) { + xpc_activate_kthreads(ch, n_needed); + } + } + + xpc_kthread_waitmsgs(part, ch); + } + + if (atomic_dec_return(&ch->kthreads_assigned) == 0 && + ((ch->flags & XPC_C_CONNECTCALLOUT) || + (ch->reason != xpcUnregistering && + ch->reason != xpcOtherUnregistering))) { + xpc_disconnected_callout(ch); + } + + + xpc_msgqueue_deref(ch); + + dev_dbg(xpc_chan, "kthread exiting, partid=%d, channel=%d\n", + partid, ch_number); + + xpc_part_deref(part); + return 0; +} + + +/* + * For each partition that XPC has established communications with, there is + * a minimum of one kernel thread assigned to perform any operation that + * may potentially sleep or block (basically the callouts to the asynchronous + * functions registered via xpc_connect()). + * + * Additional kthreads are created and destroyed by XPC as the workload + * demands. + * + * A kthread is assigned to one of the active channels that exists for a given + * partition. + */ +void +xpc_create_kthreads(struct xpc_channel *ch, int needed) +{ + unsigned long irq_flags; + pid_t pid; + u64 args = XPC_PACK_ARGS(ch->partid, ch->number); + + + while (needed-- > 0) { + pid = kernel_thread(xpc_daemonize_kthread, (void *) args, 0); + if (pid < 0) { + /* the fork failed */ + + if (atomic_read(&ch->kthreads_assigned) < + ch->kthreads_idle_limit) { + /* + * Flag this as an error only if we have an + * insufficient #of kthreads for the channel + * to function. + * + * No xpc_msgqueue_ref() is needed here since + * the channel mgr is doing this. + */ + spin_lock_irqsave(&ch->lock, irq_flags); + XPC_DISCONNECT_CHANNEL(ch, xpcLackOfResources, + &irq_flags); + spin_unlock_irqrestore(&ch->lock, irq_flags); + } + break; + } + + /* + * The following is done on behalf of the newly created + * kthread. That kthread is responsible for doing the + * counterpart to the following before it exits. + */ + (void) xpc_part_ref(&xpc_partitions[ch->partid]); + xpc_msgqueue_ref(ch); + atomic_inc(&ch->kthreads_assigned); + ch->kthreads_created++; // >>> temporary debug only!!! + } +} + + +void +xpc_disconnect_wait(int ch_number) +{ + partid_t partid; + struct xpc_partition *part; + struct xpc_channel *ch; + + + /* now wait for all callouts to the caller's function to cease */ + for (partid = 1; partid < XP_MAX_PARTITIONS; partid++) { + part = &xpc_partitions[partid]; + + if (xpc_part_ref(part)) { + ch = &part->channels[ch_number]; + +// >>> how do we keep from falling into the window between our check and going +// >>> down and coming back up where sema is re-inited? + if (ch->flags & XPC_C_SETUP) { + (void) down(&ch->teardown_sema); + } + + xpc_part_deref(part); + } + } +} + + +static void +xpc_do_exit(void) +{ + partid_t partid; + int active_part_count; + struct xpc_partition *part; + + + /* now it's time to eliminate our heartbeat */ + del_timer_sync(&xpc_hb_timer); + xpc_vars->heartbeating_to_mask = 0; + + /* indicate to others that our reserved page is uninitialized */ + xpc_rsvd_page->vars_pa = 0; + + /* + * Ignore all incoming interrupts. Without interupts the heartbeat + * checker won't activate any new partitions that may come up. + */ + free_irq(SGI_XPC_ACTIVATE, NULL); + + /* + * Cause the heartbeat checker and the discovery threads to exit. + * We don't want them attempting to activate new partitions as we + * try to deactivate the existing ones. + */ + xpc_exiting = 1; + wake_up_interruptible(&xpc_act_IRQ_wq); + + /* wait for the heartbeat checker thread to mark itself inactive */ + down(&xpc_hb_checker_exited); + + /* wait for the discovery thread to mark itself inactive */ + down(&xpc_discovery_exited); + + + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(0.3 * HZ); + set_current_state(TASK_RUNNING); + + + /* wait for all partitions to become inactive */ + + do { + active_part_count = 0; + + for (partid = 1; partid < XP_MAX_PARTITIONS; partid++) { + part = &xpc_partitions[partid]; + if (part->act_state != XPC_P_INACTIVE) { + active_part_count++; + + XPC_DEACTIVATE_PARTITION(part, xpcUnloading); + } + } + + if (active_part_count) { + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(0.3 * HZ); + set_current_state(TASK_RUNNING); + } + + } while (active_part_count > 0); + + + /* close down protections for IPI operations */ + xpc_restrict_IPI_ops(); + + + /* clear the interface to XPC's functions */ + xpc_clear_interface(); + + if (xpc_sysctl) { + unregister_sysctl_table(xpc_sysctl); + } +} + + +int __init +xpc_init(void) +{ + int ret; + partid_t partid; + struct xpc_partition *part; + pid_t pid; + + + /* + * xpc_remote_copy_buffer is used as a temporary buffer for bte_copy'ng + * both a partition's reserved page and its XPC variables. Its size was + * based on the size of a reserved page. So we need to ensure that the + * XPC variables will fit as well. + */ + if (XPC_VARS_ALIGNED_SIZE > XPC_RSVD_PAGE_ALIGNED_SIZE) { + dev_err(xpc_part, "xpc_remote_copy_buffer is not big enough\n"); + return -EPERM; + } + DBUG_ON((u64) xpc_remote_copy_buffer != + L1_CACHE_ALIGN((u64) xpc_remote_copy_buffer)); + + snprintf(xpc_part->bus_id, BUS_ID_SIZE, "part"); + snprintf(xpc_chan->bus_id, BUS_ID_SIZE, "chan"); + + xpc_sysctl = register_sysctl_table(xpc_sys_dir, 1); + + /* + * The first few fields of each entry of xpc_partitions[] need to + * be initialized now so that calls to xpc_connect() and + * xpc_disconnect() can be made prior to the activation of any remote + * partition. NOTE THAT NONE OF THE OTHER FIELDS BELONGING TO THESE + * ENTRIES ARE MEANINGFUL UNTIL AFTER AN ENTRY'S CORRESPONDING + * PARTITION HAS BEEN ACTIVATED. + */ + for (partid = 1; partid < XP_MAX_PARTITIONS; partid++) { + part = &xpc_partitions[partid]; + + DBUG_ON((u64) part != L1_CACHE_ALIGN((u64) part)); + + part->act_IRQ_rcvd = 0; + spin_lock_init(&part->act_lock); + part->act_state = XPC_P_INACTIVE; + XPC_SET_REASON(part, 0, 0); + part->setup_state = XPC_P_UNSET; + init_waitqueue_head(&part->teardown_wq); + atomic_set(&part->references, 0); + } + + /* + * Open up protections for IPI operations (and AMO operations on + * Shub 1.1 systems). + */ + xpc_allow_IPI_ops(); + + /* + * Interrupts being processed will increment this atomic variable and + * awaken the heartbeat thread which will process the interrupts. + */ + atomic_set(&xpc_act_IRQ_rcvd, 0); + + /* + * This is safe to do before the xpc_hb_checker thread has started + * because the handler releases a wait queue. If an interrupt is + * received before the thread is waiting, it will not go to sleep, + * but rather immediately process the interrupt. + */ + ret = request_irq(SGI_XPC_ACTIVATE, xpc_act_IRQ_handler, 0, + "xpc hb", NULL); + if (ret != 0) { + dev_err(xpc_part, "can't register ACTIVATE IRQ handler, " + "errno=%d\n", -ret); + + xpc_restrict_IPI_ops(); + + if (xpc_sysctl) { + unregister_sysctl_table(xpc_sysctl); + } + return -EBUSY; + } + + /* + * Fill the partition reserved page with the information needed by + * other partitions to discover we are alive and establish initial + * communications. + */ + xpc_rsvd_page = xpc_rsvd_page_init(); + if (xpc_rsvd_page == NULL) { + dev_err(xpc_part, "could not setup our reserved page\n"); + + free_irq(SGI_XPC_ACTIVATE, NULL); + xpc_restrict_IPI_ops(); + + if (xpc_sysctl) { + unregister_sysctl_table(xpc_sysctl); + } + return -EBUSY; + } + + + /* + * Set the beating to other partitions into motion. This is + * the last requirement for other partitions' discovery to + * initiate communications with us. + */ + init_timer(&xpc_hb_timer); + xpc_hb_timer.function = xpc_hb_beater; + xpc_hb_beater(0); + + + /* + * The real work-horse behind xpc. This processes incoming + * interrupts and monitors remote heartbeats. + */ + pid = kernel_thread(xpc_hb_checker, NULL, 0); + if (pid < 0) { + dev_err(xpc_part, "failed while forking hb check thread\n"); + + /* indicate to others that our reserved page is uninitialized */ + xpc_rsvd_page->vars_pa = 0; + + del_timer_sync(&xpc_hb_timer); + free_irq(SGI_XPC_ACTIVATE, NULL); + xpc_restrict_IPI_ops(); + + if (xpc_sysctl) { + unregister_sysctl_table(xpc_sysctl); + } + return -EBUSY; + } + + + /* + * Startup a thread that will attempt to discover other partitions to + * activate based on info provided by SAL. This new thread is short + * lived and will exit once discovery is complete. + */ + pid = kernel_thread(xpc_initiate_discovery, NULL, 0); + if (pid < 0) { + dev_err(xpc_part, "failed while forking discovery thread\n"); + + /* mark this new thread as a non-starter */ + up(&xpc_discovery_exited); + + xpc_do_exit(); + return -EBUSY; + } + + + /* set the interface to point at XPC's functions */ + xpc_set_interface(xpc_initiate_connect, xpc_initiate_disconnect, + xpc_initiate_allocate, xpc_initiate_send, + xpc_initiate_send_notify, xpc_initiate_received, + xpc_initiate_partid_to_nasids); + + return 0; +} +module_init(xpc_init); + + +void __exit +xpc_exit(void) +{ + xpc_do_exit(); +} +module_exit(xpc_exit); + + +MODULE_AUTHOR("Silicon Graphics, Inc."); +MODULE_DESCRIPTION("Cross Partition Communication (XPC) support"); +MODULE_LICENSE("GPL"); + +module_param(xpc_hb_interval, int, 0); +MODULE_PARM_DESC(xpc_hb_interval, "Number of seconds between " + "heartbeat increments."); + +module_param(xpc_hb_check_interval, int, 0); +MODULE_PARM_DESC(xpc_hb_check_interval, "Number of seconds between " + "heartbeat checks."); + diff --git a/arch/ia64/sn/kernel/xpc_partition.c b/arch/ia64/sn/kernel/xpc_partition.c new file mode 100644 index 00000000000..2c3c4a8af55 --- /dev/null +++ b/arch/ia64/sn/kernel/xpc_partition.c @@ -0,0 +1,984 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + */ + + +/* + * Cross Partition Communication (XPC) partition support. + * + * This is the part of XPC that detects the presence/absence of + * other partitions. It provides a heartbeat and monitors the + * heartbeats of other partitions. + * + */ + + +#include <linux/kernel.h> +#include <linux/sysctl.h> +#include <linux/cache.h> +#include <linux/mmzone.h> +#include <linux/nodemask.h> +#include <asm/sn/bte.h> +#include <asm/sn/intr.h> +#include <asm/sn/sn_sal.h> +#include <asm/sn/nodepda.h> +#include <asm/sn/addrs.h> +#include "xpc.h" + + +/* XPC is exiting flag */ +int xpc_exiting; + + +/* SH_IPI_ACCESS shub register value on startup */ +static u64 xpc_sh1_IPI_access; +static u64 xpc_sh2_IPI_access0; +static u64 xpc_sh2_IPI_access1; +static u64 xpc_sh2_IPI_access2; +static u64 xpc_sh2_IPI_access3; + + +/* original protection values for each node */ +u64 xpc_prot_vec[MAX_COMPACT_NODES]; + + +/* this partition's reserved page */ +struct xpc_rsvd_page *xpc_rsvd_page; + +/* this partition's XPC variables (within the reserved page) */ +struct xpc_vars *xpc_vars; +struct xpc_vars_part *xpc_vars_part; + + +/* + * For performance reasons, each entry of xpc_partitions[] is cacheline + * aligned. And xpc_partitions[] is padded with an additional entry at the + * end so that the last legitimate entry doesn't share its cacheline with + * another variable. + */ +struct xpc_partition xpc_partitions[XP_MAX_PARTITIONS + 1]; + + +/* + * Generic buffer used to store a local copy of the remote partitions + * reserved page or XPC variables. + * + * xpc_discovery runs only once and is a seperate thread that is + * very likely going to be processing in parallel with receiving + * interrupts. + */ +char ____cacheline_aligned + xpc_remote_copy_buffer[XPC_RSVD_PAGE_ALIGNED_SIZE]; + + +/* systune related variables */ +int xpc_hb_interval = XPC_HB_DEFAULT_INTERVAL; +int xpc_hb_check_interval = XPC_HB_CHECK_DEFAULT_TIMEOUT; + + +/* + * Given a nasid, get the physical address of the partition's reserved page + * for that nasid. This function returns 0 on any error. + */ +static u64 +xpc_get_rsvd_page_pa(int nasid, u64 buf, u64 buf_size) +{ + bte_result_t bte_res; + s64 status; + u64 cookie = 0; + u64 rp_pa = nasid; /* seed with nasid */ + u64 len = 0; + + + while (1) { + + status = sn_partition_reserved_page_pa(buf, &cookie, &rp_pa, + &len); + + dev_dbg(xpc_part, "SAL returned with status=%li, cookie=" + "0x%016lx, address=0x%016lx, len=0x%016lx\n", + status, cookie, rp_pa, len); + + if (status != SALRET_MORE_PASSES) { + break; + } + + if (len > buf_size) { + dev_err(xpc_part, "len (=0x%016lx) > buf_size\n", len); + status = SALRET_ERROR; + break; + } + + bte_res = xp_bte_copy(rp_pa, ia64_tpa(buf), buf_size, + (BTE_NOTIFY | BTE_WACQUIRE), NULL); + if (bte_res != BTE_SUCCESS) { + dev_dbg(xpc_part, "xp_bte_copy failed %i\n", bte_res); + status = SALRET_ERROR; + break; + } + } + + if (status != SALRET_OK) { + rp_pa = 0; + } + dev_dbg(xpc_part, "reserved page at phys address 0x%016lx\n", rp_pa); + return rp_pa; +} + + +/* + * Fill the partition reserved page with the information needed by + * other partitions to discover we are alive and establish initial + * communications. + */ +struct xpc_rsvd_page * +xpc_rsvd_page_init(void) +{ + struct xpc_rsvd_page *rp; + AMO_t *amos_page; + u64 rp_pa, next_cl, nasid_array = 0; + int i, ret; + + + /* get the local reserved page's address */ + + rp_pa = xpc_get_rsvd_page_pa(cnodeid_to_nasid(0), + (u64) xpc_remote_copy_buffer, + XPC_RSVD_PAGE_ALIGNED_SIZE); + if (rp_pa == 0) { + dev_err(xpc_part, "SAL failed to locate the reserved page\n"); + return NULL; + } + rp = (struct xpc_rsvd_page *) __va(rp_pa); + + if (rp->partid != sn_partition_id) { + dev_err(xpc_part, "the reserved page's partid of %d should be " + "%d\n", rp->partid, sn_partition_id); + return NULL; + } + + rp->version = XPC_RP_VERSION; + + /* + * Place the XPC variables on the cache line following the + * reserved page structure. + */ + next_cl = (u64) rp + XPC_RSVD_PAGE_ALIGNED_SIZE; + xpc_vars = (struct xpc_vars *) next_cl; + + /* + * Before clearing xpc_vars, see if a page of AMOs had been previously + * allocated. If not we'll need to allocate one and set permissions + * so that cross-partition AMOs are allowed. + * + * The allocated AMO page needs MCA reporting to remain disabled after + * XPC has unloaded. To make this work, we keep a copy of the pointer + * to this page (i.e., amos_page) in the struct xpc_vars structure, + * which is pointed to by the reserved page, and re-use that saved copy + * on subsequent loads of XPC. This AMO page is never freed, and its + * memory protections are never restricted. + */ + if ((amos_page = xpc_vars->amos_page) == NULL) { + amos_page = (AMO_t *) mspec_kalloc_page(0); + if (amos_page == NULL) { + dev_err(xpc_part, "can't allocate page of AMOs\n"); + return NULL; + } + + /* + * Open up AMO-R/W to cpu. This is done for Shub 1.1 systems + * when xpc_allow_IPI_ops() is called via xpc_hb_init(). + */ + if (!enable_shub_wars_1_1()) { + ret = sn_change_memprotect(ia64_tpa((u64) amos_page), + PAGE_SIZE, SN_MEMPROT_ACCESS_CLASS_1, + &nasid_array); + if (ret != 0) { + dev_err(xpc_part, "can't change memory " + "protections\n"); + mspec_kfree_page((unsigned long) amos_page); + return NULL; + } + } + } else if (!IS_AMO_ADDRESS((u64) amos_page)) { + /* + * EFI's XPBOOT can also set amos_page in the reserved page, + * but it happens to leave it as an uncached physical address + * and we need it to be an uncached virtual, so we'll have to + * convert it. + */ + if (!IS_AMO_PHYS_ADDRESS((u64) amos_page)) { + dev_err(xpc_part, "previously used amos_page address " + "is bad = 0x%p\n", (void *) amos_page); + return NULL; + } + amos_page = (AMO_t *) TO_AMO((u64) amos_page); + } + + memset(xpc_vars, 0, sizeof(struct xpc_vars)); + + /* + * Place the XPC per partition specific variables on the cache line + * following the XPC variables structure. + */ + next_cl += XPC_VARS_ALIGNED_SIZE; + memset((u64 *) next_cl, 0, sizeof(struct xpc_vars_part) * + XP_MAX_PARTITIONS); + xpc_vars_part = (struct xpc_vars_part *) next_cl; + xpc_vars->vars_part_pa = __pa(next_cl); + + xpc_vars->version = XPC_V_VERSION; + xpc_vars->act_nasid = cpuid_to_nasid(0); + xpc_vars->act_phys_cpuid = cpu_physical_id(0); + xpc_vars->amos_page = amos_page; /* save for next load of XPC */ + + + /* + * Initialize the activation related AMO variables. + */ + xpc_vars->act_amos = xpc_IPI_init(XP_MAX_PARTITIONS); + for (i = 1; i < XP_NASID_MASK_WORDS; i++) { + xpc_IPI_init(i + XP_MAX_PARTITIONS); + } + /* export AMO page's physical address to other partitions */ + xpc_vars->amos_page_pa = ia64_tpa((u64) xpc_vars->amos_page); + + /* + * This signifies to the remote partition that our reserved + * page is initialized. + */ + (volatile u64) rp->vars_pa = __pa(xpc_vars); + + return rp; +} + + +/* + * Change protections to allow IPI operations (and AMO operations on + * Shub 1.1 systems). + */ +void +xpc_allow_IPI_ops(void) +{ + int node; + int nasid; + + + // >>> Change SH_IPI_ACCESS code to use SAL call once it is available. + + if (is_shub2()) { + xpc_sh2_IPI_access0 = + (u64) HUB_L((u64 *) LOCAL_MMR_ADDR(SH2_IPI_ACCESS0)); + xpc_sh2_IPI_access1 = + (u64) HUB_L((u64 *) LOCAL_MMR_ADDR(SH2_IPI_ACCESS1)); + xpc_sh2_IPI_access2 = + (u64) HUB_L((u64 *) LOCAL_MMR_ADDR(SH2_IPI_ACCESS2)); + xpc_sh2_IPI_access3 = + (u64) HUB_L((u64 *) LOCAL_MMR_ADDR(SH2_IPI_ACCESS3)); + + for_each_online_node(node) { + nasid = cnodeid_to_nasid(node); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH2_IPI_ACCESS0), + -1UL); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH2_IPI_ACCESS1), + -1UL); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH2_IPI_ACCESS2), + -1UL); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH2_IPI_ACCESS3), + -1UL); + } + + } else { + xpc_sh1_IPI_access = + (u64) HUB_L((u64 *) LOCAL_MMR_ADDR(SH1_IPI_ACCESS)); + + for_each_online_node(node) { + nasid = cnodeid_to_nasid(node); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH1_IPI_ACCESS), + -1UL); + + /* + * Since the BIST collides with memory operations on + * SHUB 1.1 sn_change_memprotect() cannot be used. + */ + if (enable_shub_wars_1_1()) { + /* open up everything */ + xpc_prot_vec[node] = (u64) HUB_L((u64 *) + GLOBAL_MMR_ADDR(nasid, + SH1_MD_DQLP_MMR_DIR_PRIVEC0)); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, + SH1_MD_DQLP_MMR_DIR_PRIVEC0), + -1UL); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, + SH1_MD_DQRP_MMR_DIR_PRIVEC0), + -1UL); + } + } + } +} + + +/* + * Restrict protections to disallow IPI operations (and AMO operations on + * Shub 1.1 systems). + */ +void +xpc_restrict_IPI_ops(void) +{ + int node; + int nasid; + + + // >>> Change SH_IPI_ACCESS code to use SAL call once it is available. + + if (is_shub2()) { + + for_each_online_node(node) { + nasid = cnodeid_to_nasid(node); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH2_IPI_ACCESS0), + xpc_sh2_IPI_access0); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH2_IPI_ACCESS1), + xpc_sh2_IPI_access1); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH2_IPI_ACCESS2), + xpc_sh2_IPI_access2); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH2_IPI_ACCESS3), + xpc_sh2_IPI_access3); + } + + } else { + + for_each_online_node(node) { + nasid = cnodeid_to_nasid(node); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, SH1_IPI_ACCESS), + xpc_sh1_IPI_access); + + if (enable_shub_wars_1_1()) { + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, + SH1_MD_DQLP_MMR_DIR_PRIVEC0), + xpc_prot_vec[node]); + HUB_S((u64 *) GLOBAL_MMR_ADDR(nasid, + SH1_MD_DQRP_MMR_DIR_PRIVEC0), + xpc_prot_vec[node]); + } + } + } +} + + +/* + * At periodic intervals, scan through all active partitions and ensure + * their heartbeat is still active. If not, the partition is deactivated. + */ +void +xpc_check_remote_hb(void) +{ + struct xpc_vars *remote_vars; + struct xpc_partition *part; + partid_t partid; + bte_result_t bres; + + + remote_vars = (struct xpc_vars *) xpc_remote_copy_buffer; + + for (partid = 1; partid < XP_MAX_PARTITIONS; partid++) { + if (partid == sn_partition_id) { + continue; + } + + part = &xpc_partitions[partid]; + + if (part->act_state == XPC_P_INACTIVE || + part->act_state == XPC_P_DEACTIVATING) { + continue; + } + + /* pull the remote_hb cache line */ + bres = xp_bte_copy(part->remote_vars_pa, + ia64_tpa((u64) remote_vars), + XPC_VARS_ALIGNED_SIZE, + (BTE_NOTIFY | BTE_WACQUIRE), NULL); + if (bres != BTE_SUCCESS) { + XPC_DEACTIVATE_PARTITION(part, + xpc_map_bte_errors(bres)); + continue; + } + + dev_dbg(xpc_part, "partid = %d, heartbeat = %ld, last_heartbeat" + " = %ld, kdb_status = %ld, HB_mask = 0x%lx\n", partid, + remote_vars->heartbeat, part->last_heartbeat, + remote_vars->kdb_status, + remote_vars->heartbeating_to_mask); + + if (((remote_vars->heartbeat == part->last_heartbeat) && + (remote_vars->kdb_status == 0)) || + !XPC_HB_ALLOWED(sn_partition_id, remote_vars)) { + + XPC_DEACTIVATE_PARTITION(part, xpcNoHeartbeat); + continue; + } + + part->last_heartbeat = remote_vars->heartbeat; + } +} + + +/* + * Get a copy of the remote partition's rsvd page. + * + * remote_rp points to a buffer that is cacheline aligned for BTE copies and + * assumed to be of size XPC_RSVD_PAGE_ALIGNED_SIZE. + */ +static enum xpc_retval +xpc_get_remote_rp(int nasid, u64 *discovered_nasids, + struct xpc_rsvd_page *remote_rp, u64 *remote_rsvd_page_pa) +{ + int bres, i; + + + /* get the reserved page's physical address */ + + *remote_rsvd_page_pa = xpc_get_rsvd_page_pa(nasid, (u64) remote_rp, + XPC_RSVD_PAGE_ALIGNED_SIZE); + if (*remote_rsvd_page_pa == 0) { + return xpcNoRsvdPageAddr; + } + + + /* pull over the reserved page structure */ + + bres = xp_bte_copy(*remote_rsvd_page_pa, ia64_tpa((u64) remote_rp), + XPC_RSVD_PAGE_ALIGNED_SIZE, + (BTE_NOTIFY | BTE_WACQUIRE), NULL); + if (bres != BTE_SUCCESS) { + return xpc_map_bte_errors(bres); + } + + + if (discovered_nasids != NULL) { + for (i = 0; i < XP_NASID_MASK_WORDS; i++) { + discovered_nasids[i] |= remote_rp->part_nasids[i]; + } + } + + + /* check that the partid is for another partition */ + + if (remote_rp->partid < 1 || + remote_rp->partid > (XP_MAX_PARTITIONS - 1)) { + return xpcInvalidPartid; + } + + if (remote_rp->partid == sn_partition_id) { + return xpcLocalPartid; + } + + + if (XPC_VERSION_MAJOR(remote_rp->version) != + XPC_VERSION_MAJOR(XPC_RP_VERSION)) { + return xpcBadVersion; + } + + return xpcSuccess; +} + + +/* + * Get a copy of the remote partition's XPC variables. + * + * remote_vars points to a buffer that is cacheline aligned for BTE copies and + * assumed to be of size XPC_VARS_ALIGNED_SIZE. + */ +static enum xpc_retval +xpc_get_remote_vars(u64 remote_vars_pa, struct xpc_vars *remote_vars) +{ + int bres; + + + if (remote_vars_pa == 0) { + return xpcVarsNotSet; + } + + + /* pull over the cross partition variables */ + + bres = xp_bte_copy(remote_vars_pa, ia64_tpa((u64) remote_vars), + XPC_VARS_ALIGNED_SIZE, + (BTE_NOTIFY | BTE_WACQUIRE), NULL); + if (bres != BTE_SUCCESS) { + return xpc_map_bte_errors(bres); + } + + if (XPC_VERSION_MAJOR(remote_vars->version) != + XPC_VERSION_MAJOR(XPC_V_VERSION)) { + return xpcBadVersion; + } + + return xpcSuccess; +} + + +/* + * Prior code has determine the nasid which generated an IPI. Inspect + * that nasid to determine if its partition needs to be activated or + * deactivated. + * + * A partition is consider "awaiting activation" if our partition + * flags indicate it is not active and it has a heartbeat. A + * partition is considered "awaiting deactivation" if our partition + * flags indicate it is active but it has no heartbeat or it is not + * sending its heartbeat to us. + * + * To determine the heartbeat, the remote nasid must have a properly + * initialized reserved page. + */ +static void +xpc_identify_act_IRQ_req(int nasid) +{ + struct xpc_rsvd_page *remote_rp; + struct xpc_vars *remote_vars; + u64 remote_rsvd_page_pa; + u64 remote_vars_pa; + partid_t partid; + struct xpc_partition *part; + enum xpc_retval ret; + + + /* pull over the reserved page structure */ + + remote_rp = (struct xpc_rsvd_page *) xpc_remote_copy_buffer; + + ret = xpc_get_remote_rp(nasid, NULL, remote_rp, &remote_rsvd_page_pa); + if (ret != xpcSuccess) { + dev_warn(xpc_part, "unable to get reserved page from nasid %d, " + "which sent interrupt, reason=%d\n", nasid, ret); + return; + } + + remote_vars_pa = remote_rp->vars_pa; + partid = remote_rp->partid; + part = &xpc_partitions[partid]; + + + /* pull over the cross partition variables */ + + remote_vars = (struct xpc_vars *) xpc_remote_copy_buffer; + + ret = xpc_get_remote_vars(remote_vars_pa, remote_vars); + if (ret != xpcSuccess) { + + dev_warn(xpc_part, "unable to get XPC variables from nasid %d, " + "which sent interrupt, reason=%d\n", nasid, ret); + + XPC_DEACTIVATE_PARTITION(part, ret); + return; + } + + + part->act_IRQ_rcvd++; + + dev_dbg(xpc_part, "partid for nasid %d is %d; IRQs = %d; HB = " + "%ld:0x%lx\n", (int) nasid, (int) partid, part->act_IRQ_rcvd, + remote_vars->heartbeat, remote_vars->heartbeating_to_mask); + + + if (part->act_state == XPC_P_INACTIVE) { + + part->remote_rp_pa = remote_rsvd_page_pa; + dev_dbg(xpc_part, " remote_rp_pa = 0x%016lx\n", + part->remote_rp_pa); + + part->remote_vars_pa = remote_vars_pa; + dev_dbg(xpc_part, " remote_vars_pa = 0x%016lx\n", + part->remote_vars_pa); + + part->last_heartbeat = remote_vars->heartbeat; + dev_dbg(xpc_part, " last_heartbeat = 0x%016lx\n", + part->last_heartbeat); + + part->remote_vars_part_pa = remote_vars->vars_part_pa; + dev_dbg(xpc_part, " remote_vars_part_pa = 0x%016lx\n", + part->remote_vars_part_pa); + + part->remote_act_nasid = remote_vars->act_nasid; + dev_dbg(xpc_part, " remote_act_nasid = 0x%x\n", + part->remote_act_nasid); + + part->remote_act_phys_cpuid = remote_vars->act_phys_cpuid; + dev_dbg(xpc_part, " remote_act_phys_cpuid = 0x%x\n", + part->remote_act_phys_cpuid); + + part->remote_amos_page_pa = remote_vars->amos_page_pa; + dev_dbg(xpc_part, " remote_amos_page_pa = 0x%lx\n", + part->remote_amos_page_pa); + + xpc_activate_partition(part); + + } else if (part->remote_amos_page_pa != remote_vars->amos_page_pa || + !XPC_HB_ALLOWED(sn_partition_id, remote_vars)) { + + part->reactivate_nasid = nasid; + XPC_DEACTIVATE_PARTITION(part, xpcReactivating); + } +} + + +/* + * Loop through the activation AMO variables and process any bits + * which are set. Each bit indicates a nasid sending a partition + * activation or deactivation request. + * + * Return #of IRQs detected. + */ +int +xpc_identify_act_IRQ_sender(void) +{ + int word, bit; + u64 nasid_mask; + u64 nasid; /* remote nasid */ + int n_IRQs_detected = 0; + AMO_t *act_amos; + struct xpc_rsvd_page *rp = (struct xpc_rsvd_page *) xpc_rsvd_page; + + + act_amos = xpc_vars->act_amos; + + + /* scan through act AMO variable looking for non-zero entries */ + for (word = 0; word < XP_NASID_MASK_WORDS; word++) { + + nasid_mask = xpc_IPI_receive(&act_amos[word]); + if (nasid_mask == 0) { + /* no IRQs from nasids in this variable */ + continue; + } + + dev_dbg(xpc_part, "AMO[%d] gave back 0x%lx\n", word, + nasid_mask); + + + /* + * If this nasid has been added to the machine since + * our partition was reset, this will retain the + * remote nasid in our reserved pages machine mask. + * This is used in the event of module reload. + */ + rp->mach_nasids[word] |= nasid_mask; + + + /* locate the nasid(s) which sent interrupts */ + + for (bit = 0; bit < (8 * sizeof(u64)); bit++) { + if (nasid_mask & (1UL << bit)) { + n_IRQs_detected++; + nasid = XPC_NASID_FROM_W_B(word, bit); + dev_dbg(xpc_part, "interrupt from nasid %ld\n", + nasid); + xpc_identify_act_IRQ_req(nasid); + } + } + } + return n_IRQs_detected; +} + + +/* + * Mark specified partition as active. + */ +enum xpc_retval +xpc_mark_partition_active(struct xpc_partition *part) +{ + unsigned long irq_flags; + enum xpc_retval ret; + + + dev_dbg(xpc_part, "setting partition %d to ACTIVE\n", XPC_PARTID(part)); + + spin_lock_irqsave(&part->act_lock, irq_flags); + if (part->act_state == XPC_P_ACTIVATING) { + part->act_state = XPC_P_ACTIVE; + ret = xpcSuccess; + } else { + DBUG_ON(part->reason == xpcSuccess); + ret = part->reason; + } + spin_unlock_irqrestore(&part->act_lock, irq_flags); + + return ret; +} + + +/* + * Notify XPC that the partition is down. + */ +void +xpc_deactivate_partition(const int line, struct xpc_partition *part, + enum xpc_retval reason) +{ + unsigned long irq_flags; + partid_t partid = XPC_PARTID(part); + + + spin_lock_irqsave(&part->act_lock, irq_flags); + + if (part->act_state == XPC_P_INACTIVE) { + XPC_SET_REASON(part, reason, line); + spin_unlock_irqrestore(&part->act_lock, irq_flags); + if (reason == xpcReactivating) { + /* we interrupt ourselves to reactivate partition */ + xpc_IPI_send_reactivate(part); + } + return; + } + if (part->act_state == XPC_P_DEACTIVATING) { + if ((part->reason == xpcUnloading && reason != xpcUnloading) || + reason == xpcReactivating) { + XPC_SET_REASON(part, reason, line); + } + spin_unlock_irqrestore(&part->act_lock, irq_flags); + return; + } + + part->act_state = XPC_P_DEACTIVATING; + XPC_SET_REASON(part, reason, line); + + spin_unlock_irqrestore(&part->act_lock, irq_flags); + + XPC_DISALLOW_HB(partid, xpc_vars); + + dev_dbg(xpc_part, "bringing partition %d down, reason = %d\n", partid, + reason); + + xpc_partition_down(part, reason); +} + + +/* + * Mark specified partition as active. + */ +void +xpc_mark_partition_inactive(struct xpc_partition *part) +{ + unsigned long irq_flags; + + + dev_dbg(xpc_part, "setting partition %d to INACTIVE\n", + XPC_PARTID(part)); + + spin_lock_irqsave(&part->act_lock, irq_flags); + part->act_state = XPC_P_INACTIVE; + spin_unlock_irqrestore(&part->act_lock, irq_flags); + part->remote_rp_pa = 0; +} + + +/* + * SAL has provided a partition and machine mask. The partition mask + * contains a bit for each even nasid in our partition. The machine + * mask contains a bit for each even nasid in the entire machine. + * + * Using those two bit arrays, we can determine which nasids are + * known in the machine. Each should also have a reserved page + * initialized if they are available for partitioning. + */ +void +xpc_discovery(void) +{ + void *remote_rp_base; + struct xpc_rsvd_page *remote_rp; + struct xpc_vars *remote_vars; + u64 remote_rsvd_page_pa; + u64 remote_vars_pa; + int region; + int max_regions; + int nasid; + struct xpc_rsvd_page *rp; + partid_t partid; + struct xpc_partition *part; + u64 *discovered_nasids; + enum xpc_retval ret; + + + remote_rp = xpc_kmalloc_cacheline_aligned(XPC_RSVD_PAGE_ALIGNED_SIZE, + GFP_KERNEL, &remote_rp_base); + if (remote_rp == NULL) { + return; + } + remote_vars = (struct xpc_vars *) remote_rp; + + + discovered_nasids = kmalloc(sizeof(u64) * XP_NASID_MASK_WORDS, + GFP_KERNEL); + if (discovered_nasids == NULL) { + kfree(remote_rp_base); + return; + } + memset(discovered_nasids, 0, sizeof(u64) * XP_NASID_MASK_WORDS); + + rp = (struct xpc_rsvd_page *) xpc_rsvd_page; + + /* + * The term 'region' in this context refers to the minimum number of + * nodes that can comprise an access protection grouping. The access + * protection is in regards to memory, IOI and IPI. + */ +//>>> move the next two #defines into either include/asm-ia64/sn/arch.h or +//>>> include/asm-ia64/sn/addrs.h +#define SH1_MAX_REGIONS 64 +#define SH2_MAX_REGIONS 256 + max_regions = is_shub2() ? SH2_MAX_REGIONS : SH1_MAX_REGIONS; + + for (region = 0; region < max_regions; region++) { + + if ((volatile int) xpc_exiting) { + break; + } + + dev_dbg(xpc_part, "searching region %d\n", region); + + for (nasid = (region * sn_region_size * 2); + nasid < ((region + 1) * sn_region_size * 2); + nasid += 2) { + + if ((volatile int) xpc_exiting) { + break; + } + + dev_dbg(xpc_part, "checking nasid %d\n", nasid); + + + if (XPC_NASID_IN_ARRAY(nasid, rp->part_nasids)) { + dev_dbg(xpc_part, "PROM indicates Nasid %d is " + "part of the local partition; skipping " + "region\n", nasid); + break; + } + + if (!(XPC_NASID_IN_ARRAY(nasid, rp->mach_nasids))) { + dev_dbg(xpc_part, "PROM indicates Nasid %d was " + "not on Numa-Link network at reset\n", + nasid); + continue; + } + + if (XPC_NASID_IN_ARRAY(nasid, discovered_nasids)) { + dev_dbg(xpc_part, "Nasid %d is part of a " + "partition which was previously " + "discovered\n", nasid); + continue; + } + + + /* pull over the reserved page structure */ + + ret = xpc_get_remote_rp(nasid, discovered_nasids, + remote_rp, &remote_rsvd_page_pa); + if (ret != xpcSuccess) { + dev_dbg(xpc_part, "unable to get reserved page " + "from nasid %d, reason=%d\n", nasid, + ret); + + if (ret == xpcLocalPartid) { + break; + } + continue; + } + + remote_vars_pa = remote_rp->vars_pa; + + partid = remote_rp->partid; + part = &xpc_partitions[partid]; + + + /* pull over the cross partition variables */ + + ret = xpc_get_remote_vars(remote_vars_pa, remote_vars); + if (ret != xpcSuccess) { + dev_dbg(xpc_part, "unable to get XPC variables " + "from nasid %d, reason=%d\n", nasid, + ret); + + XPC_DEACTIVATE_PARTITION(part, ret); + continue; + } + + if (part->act_state != XPC_P_INACTIVE) { + dev_dbg(xpc_part, "partition %d on nasid %d is " + "already activating\n", partid, nasid); + break; + } + + /* + * Register the remote partition's AMOs with SAL so it + * can handle and cleanup errors within that address + * range should the remote partition go down. We don't + * unregister this range because it is difficult to + * tell when outstanding writes to the remote partition + * are finished and thus when it is thus safe to + * unregister. This should not result in wasted space + * in the SAL xp_addr_region table because we should + * get the same page for remote_act_amos_pa after + * module reloads and system reboots. + */ + if (sn_register_xp_addr_region( + remote_vars->amos_page_pa, + PAGE_SIZE, 1) < 0) { + dev_dbg(xpc_part, "partition %d failed to " + "register xp_addr region 0x%016lx\n", + partid, remote_vars->amos_page_pa); + + XPC_SET_REASON(part, xpcPhysAddrRegFailed, + __LINE__); + break; + } + + /* + * The remote nasid is valid and available. + * Send an interrupt to that nasid to notify + * it that we are ready to begin activation. + */ + dev_dbg(xpc_part, "sending an interrupt to AMO 0x%lx, " + "nasid %d, phys_cpuid 0x%x\n", + remote_vars->amos_page_pa, + remote_vars->act_nasid, + remote_vars->act_phys_cpuid); + + xpc_IPI_send_activate(remote_vars); + } + } + + kfree(discovered_nasids); + kfree(remote_rp_base); +} + + +/* + * Given a partid, get the nasids owned by that partition from the + * remote partition's reserved page. + */ +enum xpc_retval +xpc_initiate_partid_to_nasids(partid_t partid, void *nasid_mask) +{ + struct xpc_partition *part; + u64 part_nasid_pa; + int bte_res; + + + part = &xpc_partitions[partid]; + if (part->remote_rp_pa == 0) { + return xpcPartitionDown; + } + + part_nasid_pa = part->remote_rp_pa + + (u64) &((struct xpc_rsvd_page *) 0)->part_nasids; + + bte_res = xp_bte_copy(part_nasid_pa, ia64_tpa((u64) nasid_mask), + L1_CACHE_ALIGN(XP_NASID_MASK_BYTES), + (BTE_NOTIFY | BTE_WACQUIRE), NULL); + + return xpc_map_bte_errors(bte_res); +} + diff --git a/arch/ia64/sn/kernel/xpnet.c b/arch/ia64/sn/kernel/xpnet.c new file mode 100644 index 00000000000..78c13d676fa --- /dev/null +++ b/arch/ia64/sn/kernel/xpnet.c @@ -0,0 +1,715 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 1999,2001-2005 Silicon Graphics, Inc. All rights reserved. + */ + + +/* + * Cross Partition Network Interface (XPNET) support + * + * XPNET provides a virtual network layered on top of the Cross + * Partition communication layer. + * + * XPNET provides direct point-to-point and broadcast-like support + * for an ethernet-like device. The ethernet broadcast medium is + * replaced with a point-to-point message structure which passes + * pointers to a DMA-capable block that a remote partition should + * retrieve and pass to the upper level networking layer. + * + */ + + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/pci.h> +#include <linux/init.h> +#include <linux/ioport.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/delay.h> +#include <linux/ethtool.h> +#include <linux/mii.h> +#include <linux/smp.h> +#include <linux/string.h> +#include <asm/sn/bte.h> +#include <asm/sn/io.h> +#include <asm/sn/sn_sal.h> +#include <asm/types.h> +#include <asm/atomic.h> +#include <asm/sn/xp.h> + + +/* + * The message payload transferred by XPC. + * + * buf_pa is the physical address where the DMA should pull from. + * + * NOTE: for performance reasons, buf_pa should _ALWAYS_ begin on a + * cacheline boundary. To accomplish this, we record the number of + * bytes from the beginning of the first cacheline to the first useful + * byte of the skb (leadin_ignore) and the number of bytes from the + * last useful byte of the skb to the end of the last cacheline + * (tailout_ignore). + * + * size is the number of bytes to transfer which includes the skb->len + * (useful bytes of the senders skb) plus the leadin and tailout + */ +struct xpnet_message { + u16 version; /* Version for this message */ + u16 embedded_bytes; /* #of bytes embedded in XPC message */ + u32 magic; /* Special number indicating this is xpnet */ + u64 buf_pa; /* phys address of buffer to retrieve */ + u32 size; /* #of bytes in buffer */ + u8 leadin_ignore; /* #of bytes to ignore at the beginning */ + u8 tailout_ignore; /* #of bytes to ignore at the end */ + unsigned char data; /* body of small packets */ +}; + +/* + * Determine the size of our message, the cacheline aligned size, + * and then the number of message will request from XPC. + * + * XPC expects each message to exist in an individual cacheline. + */ +#define XPNET_MSG_SIZE (L1_CACHE_BYTES - XPC_MSG_PAYLOAD_OFFSET) +#define XPNET_MSG_DATA_MAX \ + (XPNET_MSG_SIZE - (u64)(&((struct xpnet_message *)0)->data)) +#define XPNET_MSG_ALIGNED_SIZE (L1_CACHE_ALIGN(XPNET_MSG_SIZE)) +#define XPNET_MSG_NENTRIES (PAGE_SIZE / XPNET_MSG_ALIGNED_SIZE) + + +#define XPNET_MAX_KTHREADS (XPNET_MSG_NENTRIES + 1) +#define XPNET_MAX_IDLE_KTHREADS (XPNET_MSG_NENTRIES + 1) + +/* + * Version number of XPNET implementation. XPNET can always talk to versions + * with same major #, and never talk to versions with a different version. + */ +#define _XPNET_VERSION(_major, _minor) (((_major) << 4) | (_minor)) +#define XPNET_VERSION_MAJOR(_v) ((_v) >> 4) +#define XPNET_VERSION_MINOR(_v) ((_v) & 0xf) + +#define XPNET_VERSION _XPNET_VERSION(1,0) /* version 1.0 */ +#define XPNET_VERSION_EMBED _XPNET_VERSION(1,1) /* version 1.1 */ +#define XPNET_MAGIC 0x88786984 /* "XNET" */ + +#define XPNET_VALID_MSG(_m) \ + ((XPNET_VERSION_MAJOR(_m->version) == XPNET_VERSION_MAJOR(XPNET_VERSION)) \ + && (msg->magic == XPNET_MAGIC)) + +#define XPNET_DEVICE_NAME "xp0" + + +/* + * When messages are queued with xpc_send_notify, a kmalloc'd buffer + * of the following type is passed as a notification cookie. When the + * notification function is called, we use the cookie to decide + * whether all outstanding message sends have completed. The skb can + * then be released. + */ +struct xpnet_pending_msg { + struct list_head free_list; + struct sk_buff *skb; + atomic_t use_count; +}; + +/* driver specific structure pointed to by the device structure */ +struct xpnet_dev_private { + struct net_device_stats stats; +}; + +struct net_device *xpnet_device; + +/* + * When we are notified of other partitions activating, we add them to + * our bitmask of partitions to which we broadcast. + */ +static u64 xpnet_broadcast_partitions; +/* protect above */ +static spinlock_t xpnet_broadcast_lock = SPIN_LOCK_UNLOCKED; + +/* + * Since the Block Transfer Engine (BTE) is being used for the transfer + * and it relies upon cache-line size transfers, we need to reserve at + * least one cache-line for head and tail alignment. The BTE is + * limited to 8MB transfers. + * + * Testing has shown that changing MTU to greater than 64KB has no effect + * on TCP as the two sides negotiate a Max Segment Size that is limited + * to 64K. Other protocols May use packets greater than this, but for + * now, the default is 64KB. + */ +#define XPNET_MAX_MTU (0x800000UL - L1_CACHE_BYTES) +/* 32KB has been determined to be the ideal */ +#define XPNET_DEF_MTU (0x8000UL) + + +/* + * The partition id is encapsulated in the MAC address. The following + * define locates the octet the partid is in. + */ +#define XPNET_PARTID_OCTET 1 +#define XPNET_LICENSE_OCTET 2 + + +/* + * Define the XPNET debug device structure that is to be used with dev_dbg(), + * dev_err(), dev_warn(), and dev_info(). + */ +struct device_driver xpnet_dbg_name = { + .name = "xpnet" +}; + +struct device xpnet_dbg_subname = { + .bus_id = {0}, /* set to "" */ + .driver = &xpnet_dbg_name +}; + +struct device *xpnet = &xpnet_dbg_subname; + +/* + * Packet was recevied by XPC and forwarded to us. + */ +static void +xpnet_receive(partid_t partid, int channel, struct xpnet_message *msg) +{ + struct sk_buff *skb; + bte_result_t bret; + struct xpnet_dev_private *priv = + (struct xpnet_dev_private *) xpnet_device->priv; + + + if (!XPNET_VALID_MSG(msg)) { + /* + * Packet with a different XPC version. Ignore. + */ + xpc_received(partid, channel, (void *) msg); + + priv->stats.rx_errors++; + + return; + } + dev_dbg(xpnet, "received 0x%lx, %d, %d, %d\n", msg->buf_pa, msg->size, + msg->leadin_ignore, msg->tailout_ignore); + + + /* reserve an extra cache line */ + skb = dev_alloc_skb(msg->size + L1_CACHE_BYTES); + if (!skb) { + dev_err(xpnet, "failed on dev_alloc_skb(%d)\n", + msg->size + L1_CACHE_BYTES); + + xpc_received(partid, channel, (void *) msg); + + priv->stats.rx_errors++; + + return; + } + + /* + * The allocated skb has some reserved space. + * In order to use bte_copy, we need to get the + * skb->data pointer moved forward. + */ + skb_reserve(skb, (L1_CACHE_BYTES - ((u64)skb->data & + (L1_CACHE_BYTES - 1)) + + msg->leadin_ignore)); + + /* + * Update the tail pointer to indicate data actually + * transferred. + */ + skb_put(skb, (msg->size - msg->leadin_ignore - msg->tailout_ignore)); + + /* + * Move the data over from the the other side. + */ + if ((XPNET_VERSION_MINOR(msg->version) == 1) && + (msg->embedded_bytes != 0)) { + dev_dbg(xpnet, "copying embedded message. memcpy(0x%p, 0x%p, " + "%lu)\n", skb->data, &msg->data, + (size_t) msg->embedded_bytes); + + memcpy(skb->data, &msg->data, (size_t) msg->embedded_bytes); + } else { + dev_dbg(xpnet, "transferring buffer to the skb->data area;\n\t" + "bte_copy(0x%p, 0x%p, %hu)\n", (void *)msg->buf_pa, + (void *)__pa((u64)skb->data & ~(L1_CACHE_BYTES - 1)), + msg->size); + + bret = bte_copy(msg->buf_pa, + __pa((u64)skb->data & ~(L1_CACHE_BYTES - 1)), + msg->size, (BTE_NOTIFY | BTE_WACQUIRE), NULL); + + if (bret != BTE_SUCCESS) { + // >>> Need better way of cleaning skb. Currently skb + // >>> appears in_use and we can't just call + // >>> dev_kfree_skb. + dev_err(xpnet, "bte_copy(0x%p, 0x%p, 0x%hx) returned " + "error=0x%x\n", (void *)msg->buf_pa, + (void *)__pa((u64)skb->data & + ~(L1_CACHE_BYTES - 1)), + msg->size, bret); + + xpc_received(partid, channel, (void *) msg); + + priv->stats.rx_errors++; + + return; + } + } + + dev_dbg(xpnet, "<skb->head=0x%p skb->data=0x%p skb->tail=0x%p " + "skb->end=0x%p skb->len=%d\n", (void *) skb->head, + (void *) skb->data, (void *) skb->tail, (void *) skb->end, + skb->len); + + skb->dev = xpnet_device; + skb->protocol = eth_type_trans(skb, xpnet_device); + skb->ip_summed = CHECKSUM_UNNECESSARY; + + dev_dbg(xpnet, "passing skb to network layer; \n\tskb->head=0x%p " + "skb->data=0x%p skb->tail=0x%p skb->end=0x%p skb->len=%d\n", + (void *) skb->head, (void *) skb->data, (void *) skb->tail, + (void *) skb->end, skb->len); + + + xpnet_device->last_rx = jiffies; + priv->stats.rx_packets++; + priv->stats.rx_bytes += skb->len + ETH_HLEN; + + netif_rx_ni(skb); + xpc_received(partid, channel, (void *) msg); +} + + +/* + * This is the handler which XPC calls during any sort of change in + * state or message reception on a connection. + */ +static void +xpnet_connection_activity(enum xpc_retval reason, partid_t partid, int channel, + void *data, void *key) +{ + long bp; + + + DBUG_ON(partid <= 0 || partid >= XP_MAX_PARTITIONS); + DBUG_ON(channel != XPC_NET_CHANNEL); + + switch(reason) { + case xpcMsgReceived: /* message received */ + DBUG_ON(data == NULL); + + xpnet_receive(partid, channel, (struct xpnet_message *) data); + break; + + case xpcConnected: /* connection completed to a partition */ + spin_lock_bh(&xpnet_broadcast_lock); + xpnet_broadcast_partitions |= 1UL << (partid -1 ); + bp = xpnet_broadcast_partitions; + spin_unlock_bh(&xpnet_broadcast_lock); + + netif_carrier_on(xpnet_device); + + dev_dbg(xpnet, "%s connection created to partition %d; " + "xpnet_broadcast_partitions=0x%lx\n", + xpnet_device->name, partid, bp); + break; + + default: + spin_lock_bh(&xpnet_broadcast_lock); + xpnet_broadcast_partitions &= ~(1UL << (partid -1 )); + bp = xpnet_broadcast_partitions; + spin_unlock_bh(&xpnet_broadcast_lock); + + if (bp == 0) { + netif_carrier_off(xpnet_device); + } + + dev_dbg(xpnet, "%s disconnected from partition %d; " + "xpnet_broadcast_partitions=0x%lx\n", + xpnet_device->name, partid, bp); + break; + + } +} + + +static int +xpnet_dev_open(struct net_device *dev) +{ + enum xpc_retval ret; + + + dev_dbg(xpnet, "calling xpc_connect(%d, 0x%p, NULL, %ld, %ld, %d, " + "%d)\n", XPC_NET_CHANNEL, xpnet_connection_activity, + XPNET_MSG_SIZE, XPNET_MSG_NENTRIES, XPNET_MAX_KTHREADS, + XPNET_MAX_IDLE_KTHREADS); + + ret = xpc_connect(XPC_NET_CHANNEL, xpnet_connection_activity, NULL, + XPNET_MSG_SIZE, XPNET_MSG_NENTRIES, + XPNET_MAX_KTHREADS, XPNET_MAX_IDLE_KTHREADS); + if (ret != xpcSuccess) { + dev_err(xpnet, "ifconfig up of %s failed on XPC connect, " + "ret=%d\n", dev->name, ret); + + return -ENOMEM; + } + + dev_dbg(xpnet, "ifconfig up of %s; XPC connected\n", dev->name); + + return 0; +} + + +static int +xpnet_dev_stop(struct net_device *dev) +{ + xpc_disconnect(XPC_NET_CHANNEL); + + dev_dbg(xpnet, "ifconfig down of %s; XPC disconnected\n", dev->name); + + return 0; +} + + +static int +xpnet_dev_change_mtu(struct net_device *dev, int new_mtu) +{ + /* 68 comes from min TCP+IP+MAC header */ + if ((new_mtu < 68) || (new_mtu > XPNET_MAX_MTU)) { + dev_err(xpnet, "ifconfig %s mtu %d failed; value must be " + "between 68 and %ld\n", dev->name, new_mtu, + XPNET_MAX_MTU); + return -EINVAL; + } + + dev->mtu = new_mtu; + dev_dbg(xpnet, "ifconfig %s mtu set to %d\n", dev->name, new_mtu); + return 0; +} + + +/* + * Required for the net_device structure. + */ +static int +xpnet_dev_set_config(struct net_device *dev, struct ifmap *new_map) +{ + return 0; +} + + +/* + * Return statistics to the caller. + */ +static struct net_device_stats * +xpnet_dev_get_stats(struct net_device *dev) +{ + struct xpnet_dev_private *priv; + + + priv = (struct xpnet_dev_private *) dev->priv; + + return &priv->stats; +} + + +/* + * Notification that the other end has received the message and + * DMA'd the skb information. At this point, they are done with + * our side. When all recipients are done processing, we + * release the skb and then release our pending message structure. + */ +static void +xpnet_send_completed(enum xpc_retval reason, partid_t partid, int channel, + void *__qm) +{ + struct xpnet_pending_msg *queued_msg = + (struct xpnet_pending_msg *) __qm; + + + DBUG_ON(queued_msg == NULL); + + dev_dbg(xpnet, "message to %d notified with reason %d\n", + partid, reason); + + if (atomic_dec_return(&queued_msg->use_count) == 0) { + dev_dbg(xpnet, "all acks for skb->head=-x%p\n", + (void *) queued_msg->skb->head); + + dev_kfree_skb_any(queued_msg->skb); + kfree(queued_msg); + } +} + + +/* + * Network layer has formatted a packet (skb) and is ready to place it + * "on the wire". Prepare and send an xpnet_message to all partitions + * which have connected with us and are targets of this packet. + * + * MAC-NOTE: For the XPNET driver, the MAC address contains the + * destination partition_id. If the destination partition id word + * is 0xff, this packet is to broadcast to all partitions. + */ +static int +xpnet_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct xpnet_pending_msg *queued_msg; + enum xpc_retval ret; + struct xpnet_message *msg; + u64 start_addr, end_addr; + long dp; + u8 second_mac_octet; + partid_t dest_partid; + struct xpnet_dev_private *priv; + u16 embedded_bytes; + + + priv = (struct xpnet_dev_private *) dev->priv; + + + dev_dbg(xpnet, ">skb->head=0x%p skb->data=0x%p skb->tail=0x%p " + "skb->end=0x%p skb->len=%d\n", (void *) skb->head, + (void *) skb->data, (void *) skb->tail, (void *) skb->end, + skb->len); + + + /* + * The xpnet_pending_msg tracks how many outstanding + * xpc_send_notifies are relying on this skb. When none + * remain, release the skb. + */ + queued_msg = kmalloc(sizeof(struct xpnet_pending_msg), GFP_ATOMIC); + if (queued_msg == NULL) { + dev_warn(xpnet, "failed to kmalloc %ld bytes; dropping " + "packet\n", sizeof(struct xpnet_pending_msg)); + + priv->stats.tx_errors++; + + return -ENOMEM; + } + + + /* get the beginning of the first cacheline and end of last */ + start_addr = ((u64) skb->data & ~(L1_CACHE_BYTES - 1)); + end_addr = L1_CACHE_ALIGN((u64) skb->tail); + + /* calculate how many bytes to embed in the XPC message */ + embedded_bytes = 0; + if (unlikely(skb->len <= XPNET_MSG_DATA_MAX)) { + /* skb->data does fit so embed */ + embedded_bytes = skb->len; + } + + + /* + * Since the send occurs asynchronously, we set the count to one + * and begin sending. Any sends that happen to complete before + * we are done sending will not free the skb. We will be left + * with that task during exit. This also handles the case of + * a packet destined for a partition which is no longer up. + */ + atomic_set(&queued_msg->use_count, 1); + queued_msg->skb = skb; + + + second_mac_octet = skb->data[XPNET_PARTID_OCTET]; + if (second_mac_octet == 0xff) { + /* we are being asked to broadcast to all partitions */ + dp = xpnet_broadcast_partitions; + } else if (second_mac_octet != 0) { + dp = xpnet_broadcast_partitions & + (1UL << (second_mac_octet - 1)); + } else { + /* 0 is an invalid partid. Ignore */ + dp = 0; + } + dev_dbg(xpnet, "destination Partitions mask (dp) = 0x%lx\n", dp); + + /* + * If we wanted to allow promiscous mode to work like an + * unswitched network, this would be a good point to OR in a + * mask of partitions which should be receiving all packets. + */ + + /* + * Main send loop. + */ + for (dest_partid = 1; dp && dest_partid < XP_MAX_PARTITIONS; + dest_partid++) { + + + if (!(dp & (1UL << (dest_partid - 1)))) { + /* not destined for this partition */ + continue; + } + + /* remove this partition from the destinations mask */ + dp &= ~(1UL << (dest_partid - 1)); + + + /* found a partition to send to */ + + ret = xpc_allocate(dest_partid, XPC_NET_CHANNEL, + XPC_NOWAIT, (void **)&msg); + if (unlikely(ret != xpcSuccess)) { + continue; + } + + msg->embedded_bytes = embedded_bytes; + if (unlikely(embedded_bytes != 0)) { + msg->version = XPNET_VERSION_EMBED; + dev_dbg(xpnet, "calling memcpy(0x%p, 0x%p, 0x%lx)\n", + &msg->data, skb->data, (size_t) embedded_bytes); + memcpy(&msg->data, skb->data, (size_t) embedded_bytes); + } else { + msg->version = XPNET_VERSION; + } + msg->magic = XPNET_MAGIC; + msg->size = end_addr - start_addr; + msg->leadin_ignore = (u64) skb->data - start_addr; + msg->tailout_ignore = end_addr - (u64) skb->tail; + msg->buf_pa = __pa(start_addr); + + dev_dbg(xpnet, "sending XPC message to %d:%d\nmsg->buf_pa=" + "0x%lx, msg->size=%u, msg->leadin_ignore=%u, " + "msg->tailout_ignore=%u\n", dest_partid, + XPC_NET_CHANNEL, msg->buf_pa, msg->size, + msg->leadin_ignore, msg->tailout_ignore); + + + atomic_inc(&queued_msg->use_count); + + ret = xpc_send_notify(dest_partid, XPC_NET_CHANNEL, msg, + xpnet_send_completed, queued_msg); + if (unlikely(ret != xpcSuccess)) { + atomic_dec(&queued_msg->use_count); + continue; + } + + } + + if (atomic_dec_return(&queued_msg->use_count) == 0) { + dev_dbg(xpnet, "no partitions to receive packet destined for " + "%d\n", dest_partid); + + + dev_kfree_skb(skb); + kfree(queued_msg); + } + + priv->stats.tx_packets++; + priv->stats.tx_bytes += skb->len; + + return 0; +} + + +/* + * Deal with transmit timeouts coming from the network layer. + */ +static void +xpnet_dev_tx_timeout (struct net_device *dev) +{ + struct xpnet_dev_private *priv; + + + priv = (struct xpnet_dev_private *) dev->priv; + + priv->stats.tx_errors++; + return; +} + + +static int __init +xpnet_init(void) +{ + int i; + u32 license_num; + int result = -ENOMEM; + + + dev_info(xpnet, "registering network device %s\n", XPNET_DEVICE_NAME); + + /* + * use ether_setup() to init the majority of our device + * structure and then override the necessary pieces. + */ + xpnet_device = alloc_netdev(sizeof(struct xpnet_dev_private), + XPNET_DEVICE_NAME, ether_setup); + if (xpnet_device == NULL) { + return -ENOMEM; + } + + netif_carrier_off(xpnet_device); + + xpnet_device->mtu = XPNET_DEF_MTU; + xpnet_device->change_mtu = xpnet_dev_change_mtu; + xpnet_device->open = xpnet_dev_open; + xpnet_device->get_stats = xpnet_dev_get_stats; + xpnet_device->stop = xpnet_dev_stop; + xpnet_device->hard_start_xmit = xpnet_dev_hard_start_xmit; + xpnet_device->tx_timeout = xpnet_dev_tx_timeout; + xpnet_device->set_config = xpnet_dev_set_config; + + /* + * Multicast assumes the LSB of the first octet is set for multicast + * MAC addresses. We chose the first octet of the MAC to be unlikely + * to collide with any vendor's officially issued MAC. + */ + xpnet_device->dev_addr[0] = 0xfe; + xpnet_device->dev_addr[XPNET_PARTID_OCTET] = sn_partition_id; + license_num = sn_partition_serial_number_val(); + for (i = 3; i >= 0; i--) { + xpnet_device->dev_addr[XPNET_LICENSE_OCTET + i] = + license_num & 0xff; + license_num = license_num >> 8; + } + + /* + * ether_setup() sets this to a multicast device. We are + * really not supporting multicast at this time. + */ + xpnet_device->flags &= ~IFF_MULTICAST; + + /* + * No need to checksum as it is a DMA transfer. The BTE will + * report an error if the data is not retrievable and the + * packet will be dropped. + */ + xpnet_device->features = NETIF_F_NO_CSUM; + + result = register_netdev(xpnet_device); + if (result != 0) { + free_netdev(xpnet_device); + } + + return result; +} +module_init(xpnet_init); + + +static void __exit +xpnet_exit(void) +{ + dev_info(xpnet, "unregistering network device %s\n", + xpnet_device[0].name); + + unregister_netdev(xpnet_device); + + free_netdev(xpnet_device); +} +module_exit(xpnet_exit); + + +MODULE_AUTHOR("Silicon Graphics, Inc."); +MODULE_DESCRIPTION("Cross Partition Network adapter (XPNET)"); +MODULE_LICENSE("GPL"); + diff --git a/arch/ia64/sn/pci/Makefile b/arch/ia64/sn/pci/Makefile index b5dca0097a8..2f915bce25f 100644 --- a/arch/ia64/sn/pci/Makefile +++ b/arch/ia64/sn/pci/Makefile @@ -7,4 +7,4 @@ # # Makefile for the sn pci general routines. -obj-y := pci_dma.o pcibr/ +obj-y := pci_dma.o tioca_provider.o pcibr/ diff --git a/arch/ia64/sn/pci/pci_dma.c b/arch/ia64/sn/pci/pci_dma.c index f680824f819..5da9bdbde7c 100644 --- a/arch/ia64/sn/pci/pci_dma.c +++ b/arch/ia64/sn/pci/pci_dma.c @@ -12,9 +12,8 @@ #include <linux/module.h> #include <asm/dma.h> #include <asm/sn/sn_sal.h> -#include "pci/pcibus_provider_defs.h" -#include "pci/pcidev.h" -#include "pci/pcibr_provider.h" +#include <asm/sn/pcibus_provider_defs.h> +#include <asm/sn/pcidev.h> #define SG_ENT_VIRT_ADDRESS(sg) (page_address((sg)->page) + (sg)->offset) #define SG_ENT_PHYS_ADDRESS(SG) virt_to_phys(SG_ENT_VIRT_ADDRESS(SG)) @@ -79,7 +78,8 @@ void *sn_dma_alloc_coherent(struct device *dev, size_t size, { void *cpuaddr; unsigned long phys_addr; - struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(to_pci_dev(dev)); + struct pci_dev *pdev = to_pci_dev(dev); + struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev); BUG_ON(dev->bus != &pci_bus_type); @@ -102,8 +102,7 @@ void *sn_dma_alloc_coherent(struct device *dev, size_t size, * resources. */ - *dma_handle = pcibr_dma_map(pcidev_info, phys_addr, size, - SN_PCIDMA_CONSISTENT); + *dma_handle = provider->dma_map_consistent(pdev, phys_addr, size); if (!*dma_handle) { printk(KERN_ERR "%s: out of ATEs\n", __FUNCTION__); free_pages((unsigned long)cpuaddr, get_order(size)); @@ -127,11 +126,12 @@ EXPORT_SYMBOL(sn_dma_alloc_coherent); void sn_dma_free_coherent(struct device *dev, size_t size, void *cpu_addr, dma_addr_t dma_handle) { - struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(to_pci_dev(dev)); + struct pci_dev *pdev = to_pci_dev(dev); + struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev); BUG_ON(dev->bus != &pci_bus_type); - pcibr_dma_unmap(pcidev_info, dma_handle, 0); + provider->dma_unmap(pdev, dma_handle, 0); free_pages((unsigned long)cpu_addr, get_order(size)); } EXPORT_SYMBOL(sn_dma_free_coherent); @@ -159,12 +159,13 @@ dma_addr_t sn_dma_map_single(struct device *dev, void *cpu_addr, size_t size, { dma_addr_t dma_addr; unsigned long phys_addr; - struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(to_pci_dev(dev)); + struct pci_dev *pdev = to_pci_dev(dev); + struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev); BUG_ON(dev->bus != &pci_bus_type); phys_addr = __pa(cpu_addr); - dma_addr = pcibr_dma_map(pcidev_info, phys_addr, size, 0); + dma_addr = provider->dma_map(pdev, phys_addr, size); if (!dma_addr) { printk(KERN_ERR "%s: out of ATEs\n", __FUNCTION__); return 0; @@ -187,10 +188,12 @@ EXPORT_SYMBOL(sn_dma_map_single); void sn_dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size, int direction) { - struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(to_pci_dev(dev)); + struct pci_dev *pdev = to_pci_dev(dev); + struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev); BUG_ON(dev->bus != &pci_bus_type); - pcibr_dma_unmap(pcidev_info, dma_addr, direction); + + provider->dma_unmap(pdev, dma_addr, direction); } EXPORT_SYMBOL(sn_dma_unmap_single); @@ -207,12 +210,13 @@ void sn_dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries, int direction) { int i; - struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(to_pci_dev(dev)); + struct pci_dev *pdev = to_pci_dev(dev); + struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev); BUG_ON(dev->bus != &pci_bus_type); for (i = 0; i < nhwentries; i++, sg++) { - pcibr_dma_unmap(pcidev_info, sg->dma_address, direction); + provider->dma_unmap(pdev, sg->dma_address, direction); sg->dma_address = (dma_addr_t) NULL; sg->dma_length = 0; } @@ -233,7 +237,8 @@ int sn_dma_map_sg(struct device *dev, struct scatterlist *sg, int nhwentries, { unsigned long phys_addr; struct scatterlist *saved_sg = sg; - struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(to_pci_dev(dev)); + struct pci_dev *pdev = to_pci_dev(dev); + struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev); int i; BUG_ON(dev->bus != &pci_bus_type); @@ -243,8 +248,8 @@ int sn_dma_map_sg(struct device *dev, struct scatterlist *sg, int nhwentries, */ for (i = 0; i < nhwentries; i++, sg++) { phys_addr = SG_ENT_PHYS_ADDRESS(sg); - sg->dma_address = pcibr_dma_map(pcidev_info, phys_addr, - sg->length, 0); + sg->dma_address = provider->dma_map(pdev, + phys_addr, sg->length); if (!sg->dma_address) { printk(KERN_ERR "%s: out of ATEs\n", __FUNCTION__); diff --git a/arch/ia64/sn/pci/pcibr/pcibr_ate.c b/arch/ia64/sn/pci/pcibr/pcibr_ate.c index 9d6854666f9..0e47bce85f2 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_ate.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_ate.c @@ -8,8 +8,8 @@ #include <linux/types.h> #include <asm/sn/sn_sal.h> -#include "pci/pcibus_provider_defs.h" -#include "pci/pcidev.h" +#include <asm/sn/pcibus_provider_defs.h> +#include <asm/sn/pcidev.h> #include "pci/pcibr_provider.h" int pcibr_invalidate_ate = 0; /* by default don't invalidate ATE on free */ diff --git a/arch/ia64/sn/pci/pcibr/pcibr_dma.c b/arch/ia64/sn/pci/pcibr/pcibr_dma.c index b1d66ac065c..64af2b2c178 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_dma.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_dma.c @@ -12,8 +12,8 @@ #include <asm/sn/geo.h> #include "xtalk/xwidgetdev.h" #include "xtalk/hubdev.h" -#include "pci/pcibus_provider_defs.h" -#include "pci/pcidev.h" +#include <asm/sn/pcibus_provider_defs.h> +#include <asm/sn/pcidev.h> #include "pci/tiocp.h" #include "pci/pic.h" #include "pci/pcibr_provider.h" @@ -40,7 +40,7 @@ extern int sn_ioif_inited; * we do not have to allocate entries in the PMU. */ -static uint64_t +static dma_addr_t pcibr_dmamap_ate32(struct pcidev_info *info, uint64_t paddr, size_t req_size, uint64_t flags) { @@ -109,7 +109,7 @@ pcibr_dmamap_ate32(struct pcidev_info *info, return pci_addr; } -static uint64_t +static dma_addr_t pcibr_dmatrans_direct64(struct pcidev_info * info, uint64_t paddr, uint64_t dma_attributes) { @@ -141,7 +141,7 @@ pcibr_dmatrans_direct64(struct pcidev_info * info, uint64_t paddr, } -static uint64_t +static dma_addr_t pcibr_dmatrans_direct32(struct pcidev_info * info, uint64_t paddr, size_t req_size, uint64_t flags) { @@ -180,11 +180,11 @@ pcibr_dmatrans_direct32(struct pcidev_info * info, * DMA mappings for Direct 64 and 32 do not have any DMA maps. */ void -pcibr_dma_unmap(struct pcidev_info *pcidev_info, dma_addr_t dma_handle, - int direction) +pcibr_dma_unmap(struct pci_dev *hwdev, dma_addr_t dma_handle, int direction) { - struct pcibus_info *pcibus_info = (struct pcibus_info *)pcidev_info-> - pdi_pcibus_info; + struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(hwdev); + struct pcibus_info *pcibus_info = + (struct pcibus_info *)pcidev_info->pdi_pcibus_info; if (IS_PCI32_MAPPED(dma_handle)) { int ate_index; @@ -301,7 +301,7 @@ void sn_dma_flush(uint64_t addr) spin_lock_irqsave(&((struct sn_flush_device_list *)p)-> sfdl_flush_lock, flags); - p->sfdl_flush_value = 0; + *p->sfdl_flush_addr = 0; /* force an interrupt. */ *(volatile uint32_t *)(p->sfdl_force_int_addr) = 1; @@ -316,64 +316,63 @@ void sn_dma_flush(uint64_t addr) } /* - * Wrapper DMA interface. Called from pci_dma.c routines. + * DMA interfaces. Called from pci_dma.c routines. */ -uint64_t -pcibr_dma_map(struct pcidev_info * pcidev_info, unsigned long phys_addr, - size_t size, unsigned int flags) +dma_addr_t +pcibr_dma_map(struct pci_dev * hwdev, unsigned long phys_addr, size_t size) { dma_addr_t dma_handle; - struct pci_dev *pcidev = pcidev_info->pdi_linux_pcidev; - - if (flags & SN_PCIDMA_CONSISTENT) { - /* sn_pci_alloc_consistent interfaces */ - if (pcidev->dev.coherent_dma_mask == ~0UL) { - dma_handle = - pcibr_dmatrans_direct64(pcidev_info, phys_addr, - PCI64_ATTR_BAR); - } else { - dma_handle = - (dma_addr_t) pcibr_dmamap_ate32(pcidev_info, - phys_addr, size, - PCI32_ATE_BAR); - } - } else { - /* map_sg/map_single interfaces */ + struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(hwdev); - /* SN cannot support DMA addresses smaller than 32 bits. */ - if (pcidev->dma_mask < 0x7fffffff) { - return 0; - } + /* SN cannot support DMA addresses smaller than 32 bits. */ + if (hwdev->dma_mask < 0x7fffffff) { + return 0; + } - if (pcidev->dma_mask == ~0UL) { + if (hwdev->dma_mask == ~0UL) { + /* + * Handle the most common case: 64 bit cards. This + * call should always succeed. + */ + + dma_handle = pcibr_dmatrans_direct64(pcidev_info, phys_addr, + PCI64_ATTR_PREF); + } else { + /* Handle 32-63 bit cards via direct mapping */ + dma_handle = pcibr_dmatrans_direct32(pcidev_info, phys_addr, + size, 0); + if (!dma_handle) { /* - * Handle the most common case: 64 bit cards. This - * call should always succeed. + * It is a 32 bit card and we cannot do direct mapping, + * so we use an ATE. */ - dma_handle = - pcibr_dmatrans_direct64(pcidev_info, phys_addr, - PCI64_ATTR_PREF); - } else { - /* Handle 32-63 bit cards via direct mapping */ - dma_handle = - pcibr_dmatrans_direct32(pcidev_info, phys_addr, - size, 0); - if (!dma_handle) { - /* - * It is a 32 bit card and we cannot do direct mapping, - * so we use an ATE. - */ - - dma_handle = - pcibr_dmamap_ate32(pcidev_info, phys_addr, - size, PCI32_ATE_PREF); - } + dma_handle = pcibr_dmamap_ate32(pcidev_info, phys_addr, + size, PCI32_ATE_PREF); } } return dma_handle; } +dma_addr_t +pcibr_dma_map_consistent(struct pci_dev * hwdev, unsigned long phys_addr, + size_t size) +{ + dma_addr_t dma_handle; + struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(hwdev); + + if (hwdev->dev.coherent_dma_mask == ~0UL) { + dma_handle = pcibr_dmatrans_direct64(pcidev_info, phys_addr, + PCI64_ATTR_BAR); + } else { + dma_handle = (dma_addr_t) pcibr_dmamap_ate32(pcidev_info, + phys_addr, size, + PCI32_ATE_BAR); + } + + return dma_handle; +} + EXPORT_SYMBOL(sn_dma_flush); diff --git a/arch/ia64/sn/pci/pcibr/pcibr_provider.c b/arch/ia64/sn/pci/pcibr/pcibr_provider.c index 92bd278cf7f..3893999d23d 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_provider.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_provider.c @@ -13,8 +13,8 @@ #include "xtalk/xwidgetdev.h" #include <asm/sn/geo.h> #include "xtalk/hubdev.h" -#include "pci/pcibus_provider_defs.h" -#include "pci/pcidev.h" +#include <asm/sn/pcibus_provider_defs.h> +#include <asm/sn/pcidev.h> #include "pci/pcibr_provider.h" #include <asm/sn/addrs.h> @@ -168,3 +168,23 @@ void pcibr_change_devices_irq(struct sn_irq_info *sn_irq_info) pcibr_force_interrupt(sn_irq_info); } } + +/* + * Provider entries for PIC/CP + */ + +struct sn_pcibus_provider pcibr_provider = { + .dma_map = pcibr_dma_map, + .dma_map_consistent = pcibr_dma_map_consistent, + .dma_unmap = pcibr_dma_unmap, + .bus_fixup = pcibr_bus_fixup, +}; + +int +pcibr_init_provider(void) +{ + sn_pci_provider[PCIIO_ASIC_TYPE_PIC] = &pcibr_provider; + sn_pci_provider[PCIIO_ASIC_TYPE_TIOCP] = &pcibr_provider; + + return 0; +} diff --git a/arch/ia64/sn/pci/pcibr/pcibr_reg.c b/arch/ia64/sn/pci/pcibr/pcibr_reg.c index 74a74a7d2a1..865c11c3b50 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_reg.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_reg.c @@ -8,8 +8,8 @@ #include <linux/types.h> #include <linux/interrupt.h> -#include "pci/pcibus_provider_defs.h" -#include "pci/pcidev.h" +#include <asm/sn/pcibus_provider_defs.h> +#include <asm/sn/pcidev.h> #include "pci/tiocp.h" #include "pci/pic.h" #include "pci/pcibr_provider.h" diff --git a/arch/ia64/sn/pci/tioca_provider.c b/arch/ia64/sn/pci/tioca_provider.c new file mode 100644 index 00000000000..8dae9eb4545 --- /dev/null +++ b/arch/ia64/sn/pci/tioca_provider.c @@ -0,0 +1,668 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2003-2005 Silicon Graphics, Inc. All Rights Reserved. + */ + +#include <linux/types.h> +#include <linux/interrupt.h> +#include <linux/pci.h> +#include <asm/sn/sn_sal.h> +#include <asm/sn/addrs.h> +#include <asm/sn/pcidev.h> +#include <asm/sn/pcibus_provider_defs.h> +#include <asm/sn/tioca_provider.h> + +uint32_t tioca_gart_found; +EXPORT_SYMBOL(tioca_gart_found); /* used by agp-sgi */ + +LIST_HEAD(tioca_list); +EXPORT_SYMBOL(tioca_list); /* used by agp-sgi */ + +static int tioca_gart_init(struct tioca_kernel *); + +/** + * tioca_gart_init - Initialize SGI TIOCA GART + * @tioca_common: ptr to common prom/kernel struct identifying the + * + * If the indicated tioca has devices present, initialize its associated + * GART MMR's and kernel memory. + */ +static int +tioca_gart_init(struct tioca_kernel *tioca_kern) +{ + uint64_t ap_reg; + uint64_t offset; + struct page *tmp; + struct tioca_common *tioca_common; + volatile struct tioca *ca_base; + + tioca_common = tioca_kern->ca_common; + ca_base = (struct tioca *)tioca_common->ca_common.bs_base; + + if (list_empty(tioca_kern->ca_devices)) + return 0; + + ap_reg = 0; + + /* + * Validate aperature size + */ + + switch (CA_APERATURE_SIZE >> 20) { + case 4: + ap_reg |= (0x3ff << CA_GART_AP_SIZE_SHFT); /* 4MB */ + break; + case 8: + ap_reg |= (0x3fe << CA_GART_AP_SIZE_SHFT); /* 8MB */ + break; + case 16: + ap_reg |= (0x3fc << CA_GART_AP_SIZE_SHFT); /* 16MB */ + break; + case 32: + ap_reg |= (0x3f8 << CA_GART_AP_SIZE_SHFT); /* 32 MB */ + break; + case 64: + ap_reg |= (0x3f0 << CA_GART_AP_SIZE_SHFT); /* 64 MB */ + break; + case 128: + ap_reg |= (0x3e0 << CA_GART_AP_SIZE_SHFT); /* 128 MB */ + break; + case 256: + ap_reg |= (0x3c0 << CA_GART_AP_SIZE_SHFT); /* 256 MB */ + break; + case 512: + ap_reg |= (0x380 << CA_GART_AP_SIZE_SHFT); /* 512 MB */ + break; + case 1024: + ap_reg |= (0x300 << CA_GART_AP_SIZE_SHFT); /* 1GB */ + break; + case 2048: + ap_reg |= (0x200 << CA_GART_AP_SIZE_SHFT); /* 2GB */ + break; + case 4096: + ap_reg |= (0x000 << CA_GART_AP_SIZE_SHFT); /* 4 GB */ + break; + default: + printk(KERN_ERR "%s: Invalid CA_APERATURE_SIZE " + "0x%lx\n", __FUNCTION__, (ulong) CA_APERATURE_SIZE); + return -1; + } + + /* + * Set up other aperature parameters + */ + + if (PAGE_SIZE >= 16384) { + tioca_kern->ca_ap_pagesize = 16384; + ap_reg |= CA_GART_PAGE_SIZE; + } else { + tioca_kern->ca_ap_pagesize = 4096; + } + + tioca_kern->ca_ap_size = CA_APERATURE_SIZE; + tioca_kern->ca_ap_bus_base = CA_APERATURE_BASE; + tioca_kern->ca_gart_entries = + tioca_kern->ca_ap_size / tioca_kern->ca_ap_pagesize; + + ap_reg |= (CA_GART_AP_ENB_AGP | CA_GART_AP_ENB_PCI); + ap_reg |= tioca_kern->ca_ap_bus_base; + + /* + * Allocate and set up the GART + */ + + tioca_kern->ca_gart_size = tioca_kern->ca_gart_entries * sizeof(u64); + tmp = + alloc_pages_node(tioca_kern->ca_closest_node, + GFP_KERNEL | __GFP_ZERO, + get_order(tioca_kern->ca_gart_size)); + + if (!tmp) { + printk(KERN_ERR "%s: Could not allocate " + "%lu bytes (order %d) for GART\n", + __FUNCTION__, + tioca_kern->ca_gart_size, + get_order(tioca_kern->ca_gart_size)); + return -ENOMEM; + } + + tioca_kern->ca_gart = page_address(tmp); + tioca_kern->ca_gart_coretalk_addr = + PHYS_TO_TIODMA(virt_to_phys(tioca_kern->ca_gart)); + + /* + * Compute PCI/AGP convenience fields + */ + + offset = CA_PCI32_MAPPED_BASE - CA_APERATURE_BASE; + tioca_kern->ca_pciap_base = CA_PCI32_MAPPED_BASE; + tioca_kern->ca_pciap_size = CA_PCI32_MAPPED_SIZE; + tioca_kern->ca_pcigart_start = offset / tioca_kern->ca_ap_pagesize; + tioca_kern->ca_pcigart_base = + tioca_kern->ca_gart_coretalk_addr + offset; + tioca_kern->ca_pcigart = + &tioca_kern->ca_gart[tioca_kern->ca_pcigart_start]; + tioca_kern->ca_pcigart_entries = + tioca_kern->ca_pciap_size / tioca_kern->ca_ap_pagesize; + tioca_kern->ca_pcigart_pagemap = + kcalloc(1, tioca_kern->ca_pcigart_entries / 8, GFP_KERNEL); + if (!tioca_kern->ca_pcigart_pagemap) { + free_pages((unsigned long)tioca_kern->ca_gart, + get_order(tioca_kern->ca_gart_size)); + return -1; + } + + offset = CA_AGP_MAPPED_BASE - CA_APERATURE_BASE; + tioca_kern->ca_gfxap_base = CA_AGP_MAPPED_BASE; + tioca_kern->ca_gfxap_size = CA_AGP_MAPPED_SIZE; + tioca_kern->ca_gfxgart_start = offset / tioca_kern->ca_ap_pagesize; + tioca_kern->ca_gfxgart_base = + tioca_kern->ca_gart_coretalk_addr + offset; + tioca_kern->ca_gfxgart = + &tioca_kern->ca_gart[tioca_kern->ca_gfxgart_start]; + tioca_kern->ca_gfxgart_entries = + tioca_kern->ca_gfxap_size / tioca_kern->ca_ap_pagesize; + + /* + * various control settings: + * use agp op-combining + * use GET semantics to fetch memory + * participate in coherency domain + * DISABLE GART PREFETCHING due to hw bug tracked in SGI PV930029 + */ + + ca_base->ca_control1 |= CA_AGPDMA_OP_ENB_COMBDELAY; /* PV895469 ? */ + ca_base->ca_control2 &= ~(CA_GART_MEM_PARAM); + ca_base->ca_control2 |= (0x2ull << CA_GART_MEM_PARAM_SHFT); + tioca_kern->ca_gart_iscoherent = 1; + ca_base->ca_control2 &= + ~(CA_GART_WR_PREFETCH_ENB | CA_GART_RD_PREFETCH_ENB); + + /* + * Unmask GART fetch error interrupts. Clear residual errors first. + */ + + ca_base->ca_int_status_alias = CA_GART_FETCH_ERR; + ca_base->ca_mult_error_alias = CA_GART_FETCH_ERR; + ca_base->ca_int_mask &= ~CA_GART_FETCH_ERR; + + /* + * Program the aperature and gart registers in TIOCA + */ + + ca_base->ca_gart_aperature = ap_reg; + ca_base->ca_gart_ptr_table = tioca_kern->ca_gart_coretalk_addr | 1; + + return 0; +} + +/** + * tioca_fastwrite_enable - enable AGP FW for a tioca and its functions + * @tioca_kernel: structure representing the CA + * + * Given a CA, scan all attached functions making sure they all support + * FastWrite. If so, enable FastWrite for all functions and the CA itself. + */ + +void +tioca_fastwrite_enable(struct tioca_kernel *tioca_kern) +{ + int cap_ptr; + uint64_t ca_control1; + uint32_t reg; + struct tioca *tioca_base; + struct pci_dev *pdev; + struct tioca_common *common; + + common = tioca_kern->ca_common; + + /* + * Scan all vga controllers on this bus making sure they all + * suport FW. If not, return. + */ + + list_for_each_entry(pdev, tioca_kern->ca_devices, bus_list) { + if (pdev->class != (PCI_CLASS_DISPLAY_VGA << 8)) + continue; + + cap_ptr = pci_find_capability(pdev, PCI_CAP_ID_AGP); + if (!cap_ptr) + return; /* no AGP CAP means no FW */ + + pci_read_config_dword(pdev, cap_ptr + PCI_AGP_STATUS, ®); + if (!(reg & PCI_AGP_STATUS_FW)) + return; /* function doesn't support FW */ + } + + /* + * Set fw for all vga fn's + */ + + list_for_each_entry(pdev, tioca_kern->ca_devices, bus_list) { + if (pdev->class != (PCI_CLASS_DISPLAY_VGA << 8)) + continue; + + cap_ptr = pci_find_capability(pdev, PCI_CAP_ID_AGP); + pci_read_config_dword(pdev, cap_ptr + PCI_AGP_COMMAND, ®); + reg |= PCI_AGP_COMMAND_FW; + pci_write_config_dword(pdev, cap_ptr + PCI_AGP_COMMAND, reg); + } + + /* + * Set ca's fw to match + */ + + tioca_base = (struct tioca *)common->ca_common.bs_base; + ca_control1 = tioca_base->ca_control1; + ca_control1 |= CA_AGP_FW_ENABLE; + tioca_base->ca_control1 = ca_control1; +} + +EXPORT_SYMBOL(tioca_fastwrite_enable); /* used by agp-sgi */ + +/** + * tioca_dma_d64 - create a DMA mapping using 64-bit direct mode + * @paddr: system physical address + * + * Map @paddr into 64-bit CA bus space. No device context is necessary. + * Bits 53:0 come from the coretalk address. We just need to mask in the + * following optional bits of the 64-bit pci address: + * + * 63:60 - Coretalk Packet Type - 0x1 for Mem Get/Put (coherent) + * 0x2 for PIO (non-coherent) + * We will always use 0x1 + * 55:55 - Swap bytes Currently unused + */ +static uint64_t +tioca_dma_d64(unsigned long paddr) +{ + dma_addr_t bus_addr; + + bus_addr = PHYS_TO_TIODMA(paddr); + + BUG_ON(!bus_addr); + BUG_ON(bus_addr >> 54); + + /* Set upper nibble to Cache Coherent Memory op */ + bus_addr |= (1UL << 60); + + return bus_addr; +} + +/** + * tioca_dma_d48 - create a DMA mapping using 48-bit direct mode + * @pdev: linux pci_dev representing the function + * @paddr: system physical address + * + * Map @paddr into 64-bit bus space of the CA associated with @pcidev_info. + * + * The CA agp 48 bit direct address falls out as follows: + * + * When direct mapping AGP addresses, the 48 bit AGP address is + * constructed as follows: + * + * [47:40] - Low 8 bits of the page Node ID extracted from coretalk + * address [47:40]. The upper 8 node bits are fixed + * and come from the xxx register bits [5:0] + * [39:38] - Chiplet ID extracted from coretalk address [39:38] + * [37:00] - node offset extracted from coretalk address [37:00] + * + * Since the node id in general will be non-zero, and the chiplet id + * will always be non-zero, it follows that the device must support + * a dma mask of at least 0xffffffffff (40 bits) to target node 0 + * and in general should be 0xffffffffffff (48 bits) to target nodes + * up to 255. Nodes above 255 need the support of the xxx register, + * and so a given CA can only directly target nodes in the range + * xxx - xxx+255. + */ +static uint64_t +tioca_dma_d48(struct pci_dev *pdev, uint64_t paddr) +{ + struct tioca_common *tioca_common; + struct tioca *ca_base; + uint64_t ct_addr; + dma_addr_t bus_addr; + uint32_t node_upper; + uint64_t agp_dma_extn; + struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(pdev); + + tioca_common = (struct tioca_common *)pcidev_info->pdi_pcibus_info; + ca_base = (struct tioca *)tioca_common->ca_common.bs_base; + + ct_addr = PHYS_TO_TIODMA(paddr); + if (!ct_addr) + return 0; + + bus_addr = (dma_addr_t) (ct_addr & 0xffffffffffff); + node_upper = ct_addr >> 48; + + if (node_upper > 64) { + printk(KERN_ERR "%s: coretalk addr 0x%p node id out " + "of range\n", __FUNCTION__, (void *)ct_addr); + return 0; + } + + agp_dma_extn = ca_base->ca_agp_dma_addr_extn; + if (node_upper != (agp_dma_extn >> CA_AGP_DMA_NODE_ID_SHFT)) { + printk(KERN_ERR "%s: coretalk upper node (%u) " + "mismatch with ca_agp_dma_addr_extn (%lu)\n", + __FUNCTION__, + node_upper, (agp_dma_extn >> CA_AGP_DMA_NODE_ID_SHFT)); + return 0; + } + + return bus_addr; +} + +/** + * tioca_dma_mapped - create a DMA mapping using a CA GART + * @pdev: linux pci_dev representing the function + * @paddr: host physical address to map + * @req_size: len (bytes) to map + * + * Map @paddr into CA address space using the GART mechanism. The mapped + * dma_addr_t is guarenteed to be contiguous in CA bus space. + */ +static dma_addr_t +tioca_dma_mapped(struct pci_dev *pdev, uint64_t paddr, size_t req_size) +{ + int i, ps, ps_shift, entry, entries, mapsize, last_entry; + uint64_t xio_addr, end_xio_addr; + struct tioca_common *tioca_common; + struct tioca_kernel *tioca_kern; + dma_addr_t bus_addr = 0; + struct tioca_dmamap *ca_dmamap; + void *map; + unsigned long flags; + struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(pdev);; + + tioca_common = (struct tioca_common *)pcidev_info->pdi_pcibus_info; + tioca_kern = (struct tioca_kernel *)tioca_common->ca_kernel_private; + + xio_addr = PHYS_TO_TIODMA(paddr); + if (!xio_addr) + return 0; + + spin_lock_irqsave(&tioca_kern->ca_lock, flags); + + /* + * allocate a map struct + */ + + ca_dmamap = kcalloc(1, sizeof(struct tioca_dmamap), GFP_ATOMIC); + if (!ca_dmamap) + goto map_return; + + /* + * Locate free entries that can hold req_size. Account for + * unaligned start/length when allocating. + */ + + ps = tioca_kern->ca_ap_pagesize; /* will be power of 2 */ + ps_shift = ffs(ps) - 1; + end_xio_addr = xio_addr + req_size - 1; + + entries = (end_xio_addr >> ps_shift) - (xio_addr >> ps_shift) + 1; + + map = tioca_kern->ca_pcigart_pagemap; + mapsize = tioca_kern->ca_pcigart_entries; + + entry = find_first_zero_bit(map, mapsize); + while (entry < mapsize) { + last_entry = find_next_bit(map, mapsize, entry); + + if (last_entry - entry >= entries) + break; + + entry = find_next_zero_bit(map, mapsize, last_entry); + } + + if (entry > mapsize) + goto map_return; + + for (i = 0; i < entries; i++) + set_bit(entry + i, map); + + bus_addr = tioca_kern->ca_pciap_base + (entry * ps); + + ca_dmamap->cad_dma_addr = bus_addr; + ca_dmamap->cad_gart_size = entries; + ca_dmamap->cad_gart_entry = entry; + list_add(&ca_dmamap->cad_list, &tioca_kern->ca_dmamaps); + + if (xio_addr % ps) { + tioca_kern->ca_pcigart[entry] = tioca_paddr_to_gart(xio_addr); + bus_addr += xio_addr & (ps - 1); + xio_addr &= ~(ps - 1); + xio_addr += ps; + entry++; + } + + while (xio_addr < end_xio_addr) { + tioca_kern->ca_pcigart[entry] = tioca_paddr_to_gart(xio_addr); + xio_addr += ps; + entry++; + } + + tioca_tlbflush(tioca_kern); + +map_return: + spin_unlock_irqrestore(&tioca_kern->ca_lock, flags); + return bus_addr; +} + +/** + * tioca_dma_unmap - release CA mapping resources + * @pdev: linux pci_dev representing the function + * @bus_addr: bus address returned by an earlier tioca_dma_map + * @dir: mapping direction (unused) + * + * Locate mapping resources associated with @bus_addr and release them. + * For mappings created using the direct modes (64 or 48) there are no + * resources to release. + */ +void +tioca_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir) +{ + int i, entry; + struct tioca_common *tioca_common; + struct tioca_kernel *tioca_kern; + struct tioca_dmamap *map; + struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(pdev); + unsigned long flags; + + tioca_common = (struct tioca_common *)pcidev_info->pdi_pcibus_info; + tioca_kern = (struct tioca_kernel *)tioca_common->ca_kernel_private; + + /* return straight away if this isn't be a mapped address */ + + if (bus_addr < tioca_kern->ca_pciap_base || + bus_addr >= (tioca_kern->ca_pciap_base + tioca_kern->ca_pciap_size)) + return; + + spin_lock_irqsave(&tioca_kern->ca_lock, flags); + + list_for_each_entry(map, &tioca_kern->ca_dmamaps, cad_list) + if (map->cad_dma_addr == bus_addr) + break; + + BUG_ON(map == NULL); + + entry = map->cad_gart_entry; + + for (i = 0; i < map->cad_gart_size; i++, entry++) { + clear_bit(entry, tioca_kern->ca_pcigart_pagemap); + tioca_kern->ca_pcigart[entry] = 0; + } + tioca_tlbflush(tioca_kern); + + list_del(&map->cad_list); + spin_unlock_irqrestore(&tioca_kern->ca_lock, flags); + kfree(map); +} + +/** + * tioca_dma_map - map pages for PCI DMA + * @pdev: linux pci_dev representing the function + * @paddr: host physical address to map + * @byte_count: bytes to map + * + * This is the main wrapper for mapping host physical pages to CA PCI space. + * The mapping mode used is based on the devices dma_mask. As a last resort + * use the GART mapped mode. + */ +uint64_t +tioca_dma_map(struct pci_dev *pdev, uint64_t paddr, size_t byte_count) +{ + uint64_t mapaddr; + + /* + * If card is 64 or 48 bit addresable, use a direct mapping. 32 + * bit direct is so restrictive w.r.t. where the memory resides that + * we don't use it even though CA has some support. + */ + + if (pdev->dma_mask == ~0UL) + mapaddr = tioca_dma_d64(paddr); + else if (pdev->dma_mask == 0xffffffffffffUL) + mapaddr = tioca_dma_d48(pdev, paddr); + else + mapaddr = 0; + + /* Last resort ... use PCI portion of CA GART */ + + if (mapaddr == 0) + mapaddr = tioca_dma_mapped(pdev, paddr, byte_count); + + return mapaddr; +} + +/** + * tioca_error_intr_handler - SGI TIO CA error interrupt handler + * @irq: unused + * @arg: pointer to tioca_common struct for the given CA + * @pt: unused + * + * Handle a CA error interrupt. Simply a wrapper around a SAL call which + * defers processing to the SGI prom. + */ +static irqreturn_t +tioca_error_intr_handler(int irq, void *arg, struct pt_regs *pt) +{ + struct tioca_common *soft = arg; + struct ia64_sal_retval ret_stuff; + uint64_t segment; + uint64_t busnum; + ret_stuff.status = 0; + ret_stuff.v0 = 0; + + segment = 0; + busnum = soft->ca_common.bs_persist_busnum; + + SAL_CALL_NOLOCK(ret_stuff, + (u64) SN_SAL_IOIF_ERROR_INTERRUPT, + segment, busnum, 0, 0, 0, 0, 0); + + return IRQ_HANDLED; +} + +/** + * tioca_bus_fixup - perform final PCI fixup for a TIO CA bus + * @prom_bussoft: Common prom/kernel struct representing the bus + * + * Replicates the tioca_common pointed to by @prom_bussoft in kernel + * space. Allocates and initializes a kernel-only area for a given CA, + * and sets up an irq for handling CA error interrupts. + * + * On successful setup, returns the kernel version of tioca_common back to + * the caller. + */ +void * +tioca_bus_fixup(struct pcibus_bussoft *prom_bussoft) +{ + struct tioca_common *tioca_common; + struct tioca_kernel *tioca_kern; + struct pci_bus *bus; + + /* sanity check prom rev */ + + if (sn_sal_rev_major() < 4 || + (sn_sal_rev_major() == 4 && sn_sal_rev_minor() < 6)) { + printk + (KERN_ERR "%s: SGI prom rev 4.06 or greater required " + "for tioca support\n", __FUNCTION__); + return NULL; + } + + /* + * Allocate kernel bus soft and copy from prom. + */ + + tioca_common = kcalloc(1, sizeof(struct tioca_common), GFP_KERNEL); + if (!tioca_common) + return NULL; + + memcpy(tioca_common, prom_bussoft, sizeof(struct tioca_common)); + tioca_common->ca_common.bs_base |= __IA64_UNCACHED_OFFSET; + + /* init kernel-private area */ + + tioca_kern = kcalloc(1, sizeof(struct tioca_kernel), GFP_KERNEL); + if (!tioca_kern) { + kfree(tioca_common); + return NULL; + } + + tioca_kern->ca_common = tioca_common; + spin_lock_init(&tioca_kern->ca_lock); + INIT_LIST_HEAD(&tioca_kern->ca_dmamaps); + tioca_kern->ca_closest_node = + nasid_to_cnodeid(tioca_common->ca_closest_nasid); + tioca_common->ca_kernel_private = (uint64_t) tioca_kern; + + bus = pci_find_bus(0, tioca_common->ca_common.bs_persist_busnum); + BUG_ON(!bus); + tioca_kern->ca_devices = &bus->devices; + + /* init GART */ + + if (tioca_gart_init(tioca_kern) < 0) { + kfree(tioca_kern); + kfree(tioca_common); + return NULL; + } + + tioca_gart_found++; + list_add(&tioca_kern->ca_list, &tioca_list); + + if (request_irq(SGI_TIOCA_ERROR, + tioca_error_intr_handler, + SA_SHIRQ, "TIOCA error", (void *)tioca_common)) + printk(KERN_WARNING + "%s: Unable to get irq %d. " + "Error interrupts won't be routed for TIOCA bus %d\n", + __FUNCTION__, SGI_TIOCA_ERROR, + (int)tioca_common->ca_common.bs_persist_busnum); + + return tioca_common; +} + +static struct sn_pcibus_provider tioca_pci_interfaces = { + .dma_map = tioca_dma_map, + .dma_map_consistent = tioca_dma_map, + .dma_unmap = tioca_dma_unmap, + .bus_fixup = tioca_bus_fixup, +}; + +/** + * tioca_init_provider - init SN PCI provider ops for TIO CA + */ +int +tioca_init_provider(void) +{ + sn_pci_provider[PCIIO_ASIC_TYPE_TIOCA] = &tioca_pci_interfaces; + return 0; +} diff --git a/arch/m32r/kernel/ptrace.c b/arch/m32r/kernel/ptrace.c index 8b40f362dd6..124f7c1b775 100644 --- a/arch/m32r/kernel/ptrace.c +++ b/arch/m32r/kernel/ptrace.c @@ -24,6 +24,7 @@ #include <linux/ptrace.h> #include <linux/user.h> #include <linux/string.h> +#include <linux/signal.h> #include <asm/cacheflush.h> #include <asm/io.h> @@ -665,7 +666,7 @@ do_ptrace(long request, struct task_struct *child, long addr, long data) case PTRACE_SYSCALL: case PTRACE_CONT: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -700,7 +701,7 @@ do_ptrace(long request, struct task_struct *child, long addr, long data) unsigned long pc, insn; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); if ((child->ptrace & PT_DTRACE) == 0) { diff --git a/arch/m68k/kernel/ptrace.c b/arch/m68k/kernel/ptrace.c index 0beb53333ba..f4e1e5eb8e1 100644 --- a/arch/m68k/kernel/ptrace.c +++ b/arch/m68k/kernel/ptrace.c @@ -19,6 +19,7 @@ #include <linux/ptrace.h> #include <linux/user.h> #include <linux/config.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -251,7 +252,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) long tmp; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { child->thread.work.syscall_trace = ~0; @@ -292,7 +293,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) long tmp; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; child->thread.work.syscall_trace = 0; tmp = get_reg(child, PT_SR) | (TRACE_BITS << 16); diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index fc4615b6d3a..e729bd28062 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -534,6 +534,11 @@ endchoice endmenu +config ISA_DMA_API + bool + depends on !M5272 + default y + menu "Bus options (PCI, PCMCIA, EISA, MCA, ISA)" config PCI diff --git a/arch/m68knommu/kernel/ptrace.c b/arch/m68knommu/kernel/ptrace.c index 15cf79080b1..9724e1cd82e 100644 --- a/arch/m68knommu/kernel/ptrace.c +++ b/arch/m68knommu/kernel/ptrace.c @@ -19,6 +19,7 @@ #include <linux/ptrace.h> #include <linux/user.h> #include <linux/config.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -240,7 +241,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) long tmp; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -278,7 +279,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) long tmp; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); tmp = get_reg(child, PT_SR) | (TRACE_BITS << 16); diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 5e666aad881..ab9944693f1 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -1656,3 +1656,7 @@ config GENERIC_HARDIRQS config GENERIC_IRQ_PROBE bool default y + +config ISA_DMA_API + bool + default y diff --git a/arch/mips/kernel/ptrace.c b/arch/mips/kernel/ptrace.c index 92f2c39afe2..92e70ca3bff 100644 --- a/arch/mips/kernel/ptrace.c +++ b/arch/mips/kernel/ptrace.c @@ -26,6 +26,7 @@ #include <linux/smp_lock.h> #include <linux/user.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/cpu.h> #include <asm/fpu.h> @@ -257,7 +258,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -300,25 +301,38 @@ out: return ret; } +static inline int audit_arch(void) +{ +#ifdef CONFIG_CPU_LITTLE_ENDIAN +#ifdef CONFIG_MIPS64 + if (!(current->thread.mflags & MF_32BIT_REGS)) + return AUDIT_ARCH_MIPSEL64; +#endif /* MIPS64 */ + return AUDIT_ARCH_MIPSEL; + +#else /* big endian... */ +#ifdef CONFIG_MIPS64 + if (!(current->thread.mflags & MF_32BIT_REGS)) + return AUDIT_ARCH_MIPS64; +#endif /* MIPS64 */ + return AUDIT_ARCH_MIPS; + +#endif /* endian */ +} + /* * Notification of system call entry/exit * - triggered by current->work.syscall_trace */ asmlinkage void do_syscall_trace(struct pt_regs *regs, int entryexit) { - if (unlikely(current->audit_context)) { - if (!entryexit) - audit_syscall_entry(current, regs->regs[2], - regs->regs[4], regs->regs[5], - regs->regs[6], regs->regs[7]); - else - audit_syscall_exit(current, regs->regs[2]); - } + if (unlikely(current->audit_context) && entryexit) + audit_syscall_exit(current, AUDITSC_RESULT(regs->regs[2]), regs->regs[2]); if (!test_thread_flag(TIF_SYSCALL_TRACE)) - return; + goto out; if (!(current->ptrace & PT_PTRACED)) - return; + goto out; /* The 0x80 provides a way for the tracing parent to distinguish between a syscall stop and SIGTRAP delivery */ @@ -334,4 +348,9 @@ asmlinkage void do_syscall_trace(struct pt_regs *regs, int entryexit) send_sig(current->exit_code, current, 1); current->exit_code = 0; } + out: + if (unlikely(current->audit_context) && !entryexit) + audit_syscall_entry(current, audit_arch(), regs->regs[2], + regs->regs[4], regs->regs[5], + regs->regs[6], regs->regs[7]); } diff --git a/arch/mips/kernel/ptrace32.c b/arch/mips/kernel/ptrace32.c index 611dee919d5..eee207969c2 100644 --- a/arch/mips/kernel/ptrace32.c +++ b/arch/mips/kernel/ptrace32.c @@ -24,6 +24,7 @@ #include <linux/smp_lock.h> #include <linux/user.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/cpu.h> #include <asm/fpu.h> @@ -241,7 +242,7 @@ asmlinkage int sys32_ptrace(int request, int pid, int addr, int data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned int) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c index 598bfe7426a..ae2a1312d4e 100644 --- a/arch/mips/kernel/syscall.c +++ b/arch/mips/kernel/syscall.c @@ -374,22 +374,6 @@ asmlinkage int sys_ipc (uint call, int first, int second, } /* - * Native ABI that is O32 or N64 version - */ -asmlinkage long sys_shmat(int shmid, char __user *shmaddr, - int shmflg, unsigned long *addr) -{ - unsigned long raddr; - int err; - - err = do_shmat(shmid, shmaddr, shmflg, &raddr); - if (err) - return err; - - return put_user(raddr, addr); -} - -/* * No implemented yet ... */ asmlinkage int sys_cachectl(char *addr, int nbytes, int op) diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index 5b5cd00d98c..e7e7c56fc21 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -45,6 +45,10 @@ config GENERIC_IRQ_PROBE config PM bool +config ISA_DMA_API + bool + default y + source "init/Kconfig" diff --git a/arch/parisc/kernel/ptrace.c b/arch/parisc/kernel/ptrace.c index 2937a923638..c07db9dff7c 100644 --- a/arch/parisc/kernel/ptrace.c +++ b/arch/parisc/kernel/ptrace.c @@ -17,6 +17,7 @@ #include <linux/personality.h> #include <linux/security.h> #include <linux/compat.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -285,7 +286,7 @@ long sys_ptrace(long request, pid_t pid, long addr, long data) ret = -EIO; DBG("sys_ptrace(%s)\n", request == PTRACE_SYSCALL ? "SYSCALL" : "CONT"); - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) goto out_tsk; child->ptrace &= ~(PT_SINGLESTEP|PT_BLOCKSTEP); if (request == PTRACE_SYSCALL) { @@ -311,7 +312,7 @@ long sys_ptrace(long request, pid_t pid, long addr, long data) case PTRACE_SINGLEBLOCK: DBG("sys_ptrace(SINGLEBLOCK)\n"); ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) goto out_tsk; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); child->ptrace &= ~PT_SINGLESTEP; @@ -328,7 +329,7 @@ long sys_ptrace(long request, pid_t pid, long addr, long data) case PTRACE_SINGLESTEP: DBG("sys_ptrace(SINGLESTEP)\n"); ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) goto out_tsk; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/parisc/kernel/sys_parisc.c b/arch/parisc/kernel/sys_parisc.c index 7958cd8c8bf..d15a1d53e10 100644 --- a/arch/parisc/kernel/sys_parisc.c +++ b/arch/parisc/kernel/sys_parisc.c @@ -161,17 +161,6 @@ asmlinkage unsigned long sys_mmap(unsigned long addr, unsigned long len, } } -long sys_shmat_wrapper(int shmid, char __user *shmaddr, int shmflag) -{ - unsigned long raddr; - int r; - - r = do_shmat(shmid, shmaddr, shmflag, &raddr); - if (r < 0) - return r; - return raddr; -} - /* Fucking broken ABI */ #ifdef CONFIG_64BIT diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S index 779b537100e..dcfa4d3d0e7 100644 --- a/arch/parisc/kernel/syscall_table.S +++ b/arch/parisc/kernel/syscall_table.S @@ -297,7 +297,7 @@ ENTRY_DIFF(msgrcv) ENTRY_SAME(msgget) /* 190 */ ENTRY_SAME(msgctl) - ENTRY_SAME(shmat_wrapper) + ENTRY_SAME(shmat) ENTRY_SAME(shmdt) ENTRY_SAME(shmget) ENTRY_SAME(shmctl) /* 195 */ diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig index 74aa1e92a39..d0d94e56b90 100644 --- a/arch/ppc/Kconfig +++ b/arch/ppc/Kconfig @@ -43,6 +43,10 @@ config GENERIC_NVRAM bool default y +config SCHED_NO_NO_OMIT_FRAME_POINTER + bool + default y + source "init/Kconfig" menu "Processor" @@ -53,6 +57,7 @@ choice config 6xx bool "6xx/7xx/74xx/52xx/82xx/83xx" + select PPC_FPU help There are four types of PowerPC chips supported. The more common types (601, 603, 604, 740, 750, 7400), the Motorola embedded @@ -86,6 +91,9 @@ config E500 endchoice +config PPC_FPU + bool + config BOOKE bool depends on E500 @@ -1075,6 +1083,10 @@ source kernel/power/Kconfig endmenu +config ISA_DMA_API + bool + default y + menu "Bus options" config ISA diff --git a/arch/ppc/Makefile b/arch/ppc/Makefile index 73cbdda5b59..0432a25b473 100644 --- a/arch/ppc/Makefile +++ b/arch/ppc/Makefile @@ -53,6 +53,7 @@ head-$(CONFIG_FSL_BOOKE) := arch/ppc/kernel/head_fsl_booke.o head-$(CONFIG_6xx) += arch/ppc/kernel/idle_6xx.o head-$(CONFIG_POWER4) += arch/ppc/kernel/idle_power4.o +head-$(CONFIG_PPC_FPU) += arch/ppc/kernel/fpu.o core-y += arch/ppc/kernel/ arch/ppc/platforms/ \ arch/ppc/mm/ arch/ppc/lib/ arch/ppc/syslib/ diff --git a/arch/ppc/boot/images/Makefile b/arch/ppc/boot/images/Makefile index 774de8e2387..f850fb0fb51 100644 --- a/arch/ppc/boot/images/Makefile +++ b/arch/ppc/boot/images/Makefile @@ -20,8 +20,9 @@ quiet_cmd_uimage = UIMAGE $@ targets += uImage $(obj)/uImage: $(obj)/vmlinux.gz + $(Q)rm -f $@ $(call if_changed,uimage) - @echo ' Image $@ is ready' + @echo ' Image: $@' $(if $(wildcard $@),'is ready','not made') # Files generated that shall be removed upon make clean clean-files := sImage vmapus vmlinux* miboot* zImage* uImage diff --git a/arch/ppc/kernel/Makefile b/arch/ppc/kernel/Makefile index 86bc878cb3e..b284451802c 100644 --- a/arch/ppc/kernel/Makefile +++ b/arch/ppc/kernel/Makefile @@ -9,6 +9,7 @@ extra-$(CONFIG_FSL_BOOKE) := head_fsl_booke.o extra-$(CONFIG_8xx) := head_8xx.o extra-$(CONFIG_6xx) += idle_6xx.o extra-$(CONFIG_POWER4) += idle_power4.o +extra-$(CONFIG_PPC_FPU) += fpu.o extra-y += vmlinux.lds obj-y := entry.o traps.o irq.o idle.o time.o misc.o \ diff --git a/arch/ppc/kernel/align.c b/arch/ppc/kernel/align.c index 79c92947503..ff81da9598d 100644 --- a/arch/ppc/kernel/align.c +++ b/arch/ppc/kernel/align.c @@ -290,6 +290,10 @@ fix_alignment(struct pt_regs *regs) /* lwm, stmw */ nb = (32 - reg) * 4; } + + if (!access_ok((flags & ST? VERIFY_WRITE: VERIFY_READ), addr, nb+nb0)) + return -EFAULT; /* bad address */ + rptr = (unsigned char *) ®s->gpr[reg]; if (flags & LD) { for (i = 0; i < nb; ++i) @@ -368,16 +372,24 @@ fix_alignment(struct pt_regs *regs) /* Single-precision FP load and store require conversions... */ case LD+F+S: +#ifdef CONFIG_PPC_FPU preempt_disable(); enable_kernel_fp(); cvt_fd(&data.f, &data.d, ¤t->thread.fpscr); preempt_enable(); +#else + return 0; +#endif break; case ST+F+S: +#ifdef CONFIG_PPC_FPU preempt_disable(); enable_kernel_fp(); cvt_df(&data.d, &data.f, ¤t->thread.fpscr); preempt_enable(); +#else + return 0; +#endif break; } diff --git a/arch/ppc/kernel/cpu_setup_6xx.S b/arch/ppc/kernel/cpu_setup_6xx.S index 74f781b486a..468721d9ebd 100644 --- a/arch/ppc/kernel/cpu_setup_6xx.S +++ b/arch/ppc/kernel/cpu_setup_6xx.S @@ -30,12 +30,14 @@ _GLOBAL(__setup_cpu_604) blr _GLOBAL(__setup_cpu_750) mflr r4 + bl __init_fpu_registers bl setup_common_caches bl setup_750_7400_hid0 mtlr r4 blr _GLOBAL(__setup_cpu_750cx) mflr r4 + bl __init_fpu_registers bl setup_common_caches bl setup_750_7400_hid0 bl setup_750cx @@ -43,6 +45,7 @@ _GLOBAL(__setup_cpu_750cx) blr _GLOBAL(__setup_cpu_750fx) mflr r4 + bl __init_fpu_registers bl setup_common_caches bl setup_750_7400_hid0 bl setup_750fx @@ -50,6 +53,7 @@ _GLOBAL(__setup_cpu_750fx) blr _GLOBAL(__setup_cpu_7400) mflr r4 + bl __init_fpu_registers bl setup_7400_workarounds bl setup_common_caches bl setup_750_7400_hid0 @@ -57,6 +61,7 @@ _GLOBAL(__setup_cpu_7400) blr _GLOBAL(__setup_cpu_7410) mflr r4 + bl __init_fpu_registers bl setup_7410_workarounds bl setup_common_caches bl setup_750_7400_hid0 @@ -80,7 +85,7 @@ setup_common_caches: bne 1f /* don't invalidate the D-cache */ ori r8,r8,HID0_DCI /* unless it wasn't enabled */ 1: sync - mtspr SPRN_HID0,r8 /* enable and invalidate caches */ + mtspr SPRN_HID0,r8 /* enable and invalidate caches */ sync mtspr SPRN_HID0,r11 /* enable caches */ sync @@ -152,9 +157,13 @@ setup_7410_workarounds: setup_750_7400_hid0: mfspr r11,SPRN_HID0 ori r11,r11,HID0_SGE | HID0_ABE | HID0_BHTE | HID0_BTIC + oris r11,r11,HID0_DPM@h BEGIN_FTR_SECTION - oris r11,r11,HID0_DPM@h /* enable dynamic power mgmt */ -END_FTR_SECTION_IFCLR(CPU_FTR_NO_DPM) + xori r11,r11,HID0_BTIC +END_FTR_SECTION_IFSET(CPU_FTR_NO_BTIC) +BEGIN_FTR_SECTION + xoris r11,r11,HID0_DPM@h /* disable dynamic power mgmt */ +END_FTR_SECTION_IFSET(CPU_FTR_NO_DPM) li r3,HID0_SPD andc r11,r11,r3 /* clear SPD: enable speculative */ li r3,0 @@ -218,13 +227,15 @@ setup_745x_specifics: /* All of the bits we have to set..... */ - ori r11,r11,HID0_SGE | HID0_FOLD | HID0_BHTE | HID0_LRSTK | HID0_BTIC + ori r11,r11,HID0_SGE | HID0_FOLD | HID0_BHTE + ori r11,r11,HID0_LRSTK | HID0_BTIC + oris r11,r11,HID0_DPM@h BEGIN_FTR_SECTION xori r11,r11,HID0_BTIC END_FTR_SECTION_IFSET(CPU_FTR_NO_BTIC) BEGIN_FTR_SECTION - oris r11,r11,HID0_DPM@h /* enable dynamic power mgmt */ -END_FTR_SECTION_IFCLR(CPU_FTR_NO_DPM) + xoris r11,r11,HID0_DPM@h /* disable dynamic power mgmt */ +END_FTR_SECTION_IFSET(CPU_FTR_NO_DPM) /* All of the bits we have to clear.... */ @@ -248,6 +259,25 @@ END_FTR_SECTION_IFCLR(CPU_FTR_NO_DPM) isync blr +/* + * Initialize the FPU registers. This is needed to work around an errata + * in some 750 cpus where using a not yet initialized FPU register after + * power on reset may hang the CPU + */ +_GLOBAL(__init_fpu_registers) + mfmsr r10 + ori r11,r10,MSR_FP + mtmsr r11 + isync + addis r9,r3,empty_zero_page@ha + addi r9,r9,empty_zero_page@l + REST_32FPRS(0,r9) + sync + mtmsr r10 + isync + blr + + /* Definitions for the table use to save CPU states */ #define CS_HID0 0 #define CS_HID1 4 diff --git a/arch/ppc/kernel/entry.S b/arch/ppc/kernel/entry.S index 035217d6c0f..5f075dbc4ee 100644 --- a/arch/ppc/kernel/entry.S +++ b/arch/ppc/kernel/entry.S @@ -563,6 +563,65 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) addi r1,r1,INT_FRAME_SIZE blr + .globl fast_exception_return +fast_exception_return: +#if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE)) + andi. r10,r9,MSR_RI /* check for recoverable interrupt */ + beq 1f /* if not, we've got problems */ +#endif + +2: REST_4GPRS(3, r11) + lwz r10,_CCR(r11) + REST_GPR(1, r11) + mtcr r10 + lwz r10,_LINK(r11) + mtlr r10 + REST_GPR(10, r11) + mtspr SPRN_SRR1,r9 + mtspr SPRN_SRR0,r12 + REST_GPR(9, r11) + REST_GPR(12, r11) + lwz r11,GPR11(r11) + SYNC + RFI + +#if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE)) +/* check if the exception happened in a restartable section */ +1: lis r3,exc_exit_restart_end@ha + addi r3,r3,exc_exit_restart_end@l + cmplw r12,r3 + bge 3f + lis r4,exc_exit_restart@ha + addi r4,r4,exc_exit_restart@l + cmplw r12,r4 + blt 3f + lis r3,fee_restarts@ha + tophys(r3,r3) + lwz r5,fee_restarts@l(r3) + addi r5,r5,1 + stw r5,fee_restarts@l(r3) + mr r12,r4 /* restart at exc_exit_restart */ + b 2b + + .comm fee_restarts,4 + +/* aargh, a nonrecoverable interrupt, panic */ +/* aargh, we don't know which trap this is */ +/* but the 601 doesn't implement the RI bit, so assume it's OK */ +3: +BEGIN_FTR_SECTION + b 2b +END_FTR_SECTION_IFSET(CPU_FTR_601) + li r10,-1 + stw r10,TRAP(r11) + addi r3,r1,STACK_FRAME_OVERHEAD + lis r10,MSR_KERNEL@h + ori r10,r10,MSR_KERNEL@l + bl transfer_to_handler_full + .long nonrecoverable_exception + .long ret_from_except +#endif + .globl sigreturn_exit sigreturn_exit: subi r1,r3,STACK_FRAME_OVERHEAD diff --git a/arch/ppc/kernel/fpu.S b/arch/ppc/kernel/fpu.S new file mode 100644 index 00000000000..6189b26f640 --- /dev/null +++ b/arch/ppc/kernel/fpu.S @@ -0,0 +1,133 @@ +/* + * FPU support code, moved here from head.S so that it can be used + * by chips which use other head-whatever.S files. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + */ + +#include <linux/config.h> +#include <asm/processor.h> +#include <asm/page.h> +#include <asm/mmu.h> +#include <asm/pgtable.h> +#include <asm/cputable.h> +#include <asm/cache.h> +#include <asm/thread_info.h> +#include <asm/ppc_asm.h> +#include <asm/offsets.h> + +/* + * This task wants to use the FPU now. + * On UP, disable FP for the task which had the FPU previously, + * and save its floating-point registers in its thread_struct. + * Load up this task's FP registers from its thread_struct, + * enable the FPU for the current task and return to the task. + */ + .globl load_up_fpu +load_up_fpu: + mfmsr r5 + ori r5,r5,MSR_FP +#ifdef CONFIG_PPC64BRIDGE + clrldi r5,r5,1 /* turn off 64-bit mode */ +#endif /* CONFIG_PPC64BRIDGE */ + SYNC + MTMSRD(r5) /* enable use of fpu now */ + isync +/* + * For SMP, we don't do lazy FPU switching because it just gets too + * horrendously complex, especially when a task switches from one CPU + * to another. Instead we call giveup_fpu in switch_to. + */ +#ifndef CONFIG_SMP + tophys(r6,0) /* get __pa constant */ + addis r3,r6,last_task_used_math@ha + lwz r4,last_task_used_math@l(r3) + cmpwi 0,r4,0 + beq 1f + add r4,r4,r6 + addi r4,r4,THREAD /* want last_task_used_math->thread */ + SAVE_32FPRS(0, r4) + mffs fr0 + stfd fr0,THREAD_FPSCR-4(r4) + lwz r5,PT_REGS(r4) + add r5,r5,r6 + lwz r4,_MSR-STACK_FRAME_OVERHEAD(r5) + li r10,MSR_FP|MSR_FE0|MSR_FE1 + andc r4,r4,r10 /* disable FP for previous task */ + stw r4,_MSR-STACK_FRAME_OVERHEAD(r5) +1: +#endif /* CONFIG_SMP */ + /* enable use of FP after return */ + mfspr r5,SPRN_SPRG3 /* current task's THREAD (phys) */ + lwz r4,THREAD_FPEXC_MODE(r5) + ori r9,r9,MSR_FP /* enable FP for current */ + or r9,r9,r4 + lfd fr0,THREAD_FPSCR-4(r5) + mtfsf 0xff,fr0 + REST_32FPRS(0, r5) +#ifndef CONFIG_SMP + subi r4,r5,THREAD + sub r4,r4,r6 + stw r4,last_task_used_math@l(r3) +#endif /* CONFIG_SMP */ + /* restore registers and return */ + /* we haven't used ctr or xer or lr */ + b fast_exception_return + +/* + * FP unavailable trap from kernel - print a message, but let + * the task use FP in the kernel until it returns to user mode. + */ + .globl KernelFP +KernelFP: + lwz r3,_MSR(r1) + ori r3,r3,MSR_FP + stw r3,_MSR(r1) /* enable use of FP after return */ + lis r3,86f@h + ori r3,r3,86f@l + mr r4,r2 /* current */ + lwz r5,_NIP(r1) + bl printk + b ret_from_except +86: .string "floating point used in kernel (task=%p, pc=%x)\n" + .align 4,0 + +/* + * giveup_fpu(tsk) + * Disable FP for the task given as the argument, + * and save the floating-point registers in its thread_struct. + * Enables the FPU for use in the kernel on return. + */ + .globl giveup_fpu +giveup_fpu: + mfmsr r5 + ori r5,r5,MSR_FP + SYNC_601 + ISYNC_601 + MTMSRD(r5) /* enable use of fpu now */ + SYNC_601 + isync + cmpwi 0,r3,0 + beqlr- /* if no previous owner, done */ + addi r3,r3,THREAD /* want THREAD of task */ + lwz r5,PT_REGS(r3) + cmpwi 0,r5,0 + SAVE_32FPRS(0, r3) + mffs fr0 + stfd fr0,THREAD_FPSCR-4(r3) + beq 1f + lwz r4,_MSR-STACK_FRAME_OVERHEAD(r5) + li r3,MSR_FP|MSR_FE0|MSR_FE1 + andc r4,r4,r3 /* disable FP for previous task */ + stw r4,_MSR-STACK_FRAME_OVERHEAD(r5) +1: +#ifndef CONFIG_SMP + li r5,0 + lis r4,last_task_used_math@ha + stw r5,last_task_used_math@l(r4) +#endif /* CONFIG_SMP */ + blr diff --git a/arch/ppc/kernel/head.S b/arch/ppc/kernel/head.S index 1a89a71e0ac..a931d773715 100644 --- a/arch/ppc/kernel/head.S +++ b/arch/ppc/kernel/head.S @@ -775,133 +775,6 @@ InstructionSegment: EXC_XFER_STD(0x480, UnknownException) #endif /* CONFIG_PPC64BRIDGE */ -/* - * This task wants to use the FPU now. - * On UP, disable FP for the task which had the FPU previously, - * and save its floating-point registers in its thread_struct. - * Load up this task's FP registers from its thread_struct, - * enable the FPU for the current task and return to the task. - */ -load_up_fpu: - mfmsr r5 - ori r5,r5,MSR_FP -#ifdef CONFIG_PPC64BRIDGE - clrldi r5,r5,1 /* turn off 64-bit mode */ -#endif /* CONFIG_PPC64BRIDGE */ - SYNC - MTMSRD(r5) /* enable use of fpu now */ - isync -/* - * For SMP, we don't do lazy FPU switching because it just gets too - * horrendously complex, especially when a task switches from one CPU - * to another. Instead we call giveup_fpu in switch_to. - */ -#ifndef CONFIG_SMP - tophys(r6,0) /* get __pa constant */ - addis r3,r6,last_task_used_math@ha - lwz r4,last_task_used_math@l(r3) - cmpwi 0,r4,0 - beq 1f - add r4,r4,r6 - addi r4,r4,THREAD /* want last_task_used_math->thread */ - SAVE_32FPRS(0, r4) - mffs fr0 - stfd fr0,THREAD_FPSCR-4(r4) - lwz r5,PT_REGS(r4) - add r5,r5,r6 - lwz r4,_MSR-STACK_FRAME_OVERHEAD(r5) - li r10,MSR_FP|MSR_FE0|MSR_FE1 - andc r4,r4,r10 /* disable FP for previous task */ - stw r4,_MSR-STACK_FRAME_OVERHEAD(r5) -1: -#endif /* CONFIG_SMP */ - /* enable use of FP after return */ - mfspr r5,SPRN_SPRG3 /* current task's THREAD (phys) */ - lwz r4,THREAD_FPEXC_MODE(r5) - ori r9,r9,MSR_FP /* enable FP for current */ - or r9,r9,r4 - lfd fr0,THREAD_FPSCR-4(r5) - mtfsf 0xff,fr0 - REST_32FPRS(0, r5) -#ifndef CONFIG_SMP - subi r4,r5,THREAD - sub r4,r4,r6 - stw r4,last_task_used_math@l(r3) -#endif /* CONFIG_SMP */ - /* restore registers and return */ - /* we haven't used ctr or xer or lr */ - /* fall through to fast_exception_return */ - - .globl fast_exception_return -fast_exception_return: - andi. r10,r9,MSR_RI /* check for recoverable interrupt */ - beq 1f /* if not, we've got problems */ -2: REST_4GPRS(3, r11) - lwz r10,_CCR(r11) - REST_GPR(1, r11) - mtcr r10 - lwz r10,_LINK(r11) - mtlr r10 - REST_GPR(10, r11) - mtspr SPRN_SRR1,r9 - mtspr SPRN_SRR0,r12 - REST_GPR(9, r11) - REST_GPR(12, r11) - lwz r11,GPR11(r11) - SYNC - RFI - -/* check if the exception happened in a restartable section */ -1: lis r3,exc_exit_restart_end@ha - addi r3,r3,exc_exit_restart_end@l - cmplw r12,r3 - bge 3f - lis r4,exc_exit_restart@ha - addi r4,r4,exc_exit_restart@l - cmplw r12,r4 - blt 3f - lis r3,fee_restarts@ha - tophys(r3,r3) - lwz r5,fee_restarts@l(r3) - addi r5,r5,1 - stw r5,fee_restarts@l(r3) - mr r12,r4 /* restart at exc_exit_restart */ - b 2b - - .comm fee_restarts,4 - -/* aargh, a nonrecoverable interrupt, panic */ -/* aargh, we don't know which trap this is */ -/* but the 601 doesn't implement the RI bit, so assume it's OK */ -3: -BEGIN_FTR_SECTION - b 2b -END_FTR_SECTION_IFSET(CPU_FTR_601) - li r10,-1 - stw r10,TRAP(r11) - addi r3,r1,STACK_FRAME_OVERHEAD - li r10,MSR_KERNEL - bl transfer_to_handler_full - .long nonrecoverable_exception - .long ret_from_except - -/* - * FP unavailable trap from kernel - print a message, but let - * the task use FP in the kernel until it returns to user mode. - */ -KernelFP: - lwz r3,_MSR(r1) - ori r3,r3,MSR_FP - stw r3,_MSR(r1) /* enable use of FP after return */ - lis r3,86f@h - ori r3,r3,86f@l - mr r4,r2 /* current */ - lwz r5,_NIP(r1) - bl printk - b ret_from_except -86: .string "floating point used in kernel (task=%p, pc=%x)\n" - .align 4,0 - #ifdef CONFIG_ALTIVEC /* Note that the AltiVec support is closely modeled after the FP * support. Changes to one are likely to be applicable to the @@ -1016,42 +889,6 @@ giveup_altivec: #endif /* CONFIG_ALTIVEC */ /* - * giveup_fpu(tsk) - * Disable FP for the task given as the argument, - * and save the floating-point registers in its thread_struct. - * Enables the FPU for use in the kernel on return. - */ - .globl giveup_fpu -giveup_fpu: - mfmsr r5 - ori r5,r5,MSR_FP - SYNC_601 - ISYNC_601 - MTMSRD(r5) /* enable use of fpu now */ - SYNC_601 - isync - cmpwi 0,r3,0 - beqlr- /* if no previous owner, done */ - addi r3,r3,THREAD /* want THREAD of task */ - lwz r5,PT_REGS(r3) - cmpwi 0,r5,0 - SAVE_32FPRS(0, r3) - mffs fr0 - stfd fr0,THREAD_FPSCR-4(r3) - beq 1f - lwz r4,_MSR-STACK_FRAME_OVERHEAD(r5) - li r3,MSR_FP|MSR_FE0|MSR_FE1 - andc r4,r4,r3 /* disable FP for previous task */ - stw r4,_MSR-STACK_FRAME_OVERHEAD(r5) -1: -#ifndef CONFIG_SMP - li r5,0 - lis r4,last_task_used_math@ha - stw r5,last_task_used_math@l(r4) -#endif /* CONFIG_SMP */ - blr - -/* * This code is jumped to from the startup code to copy * the kernel image to physical address 0. */ diff --git a/arch/ppc/kernel/head_44x.S b/arch/ppc/kernel/head_44x.S index 9ed8165a3d6..9b6a8e51365 100644 --- a/arch/ppc/kernel/head_44x.S +++ b/arch/ppc/kernel/head_44x.S @@ -426,7 +426,11 @@ interrupt_base: PROGRAM_EXCEPTION /* Floating Point Unavailable Interrupt */ +#ifdef CONFIG_PPC_FPU + FP_UNAVAILABLE_EXCEPTION +#else EXCEPTION(0x2010, FloatingPointUnavailable, UnknownException, EXC_XFER_EE) +#endif /* System Call Interrupt */ START_EXCEPTION(SystemCall) @@ -686,8 +690,10 @@ _GLOBAL(giveup_altivec) * * The 44x core does not have an FPU. */ +#ifndef CONFIG_PPC_FPU _GLOBAL(giveup_fpu) blr +#endif /* * extern void abort(void) diff --git a/arch/ppc/kernel/head_booke.h b/arch/ppc/kernel/head_booke.h index 884dac916bc..f213d12eec0 100644 --- a/arch/ppc/kernel/head_booke.h +++ b/arch/ppc/kernel/head_booke.h @@ -337,4 +337,11 @@ label: addi r3,r1,STACK_FRAME_OVERHEAD; \ EXC_XFER_LITE(0x0900, timer_interrupt) +#define FP_UNAVAILABLE_EXCEPTION \ + START_EXCEPTION(FloatingPointUnavailable) \ + NORMAL_EXCEPTION_PROLOG; \ + bne load_up_fpu; /* if from user, just load it up */ \ + addi r3,r1,STACK_FRAME_OVERHEAD; \ + EXC_XFER_EE_LITE(0x800, KernelFP) + #endif /* __HEAD_BOOKE_H__ */ diff --git a/arch/ppc/kernel/head_fsl_booke.S b/arch/ppc/kernel/head_fsl_booke.S index d64bf61d2b1..f22ddce3613 100644 --- a/arch/ppc/kernel/head_fsl_booke.S +++ b/arch/ppc/kernel/head_fsl_booke.S @@ -504,7 +504,11 @@ interrupt_base: PROGRAM_EXCEPTION /* Floating Point Unavailable Interrupt */ +#ifdef CONFIG_PPC_FPU + FP_UNAVAILABLE_EXCEPTION +#else EXCEPTION(0x0800, FloatingPointUnavailable, UnknownException, EXC_XFER_EE) +#endif /* System Call Interrupt */ START_EXCEPTION(SystemCall) @@ -916,10 +920,12 @@ _GLOBAL(giveup_spe) /* * extern void giveup_fpu(struct task_struct *prev) * - * The e500 core does not have an FPU. + * Not all FSL Book-E cores have an FPU */ +#ifndef CONFIG_PPC_FPU _GLOBAL(giveup_fpu) blr +#endif /* * extern void abort(void) diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S index 73f7c23b0dd..e4f1615ec13 100644 --- a/arch/ppc/kernel/misc.S +++ b/arch/ppc/kernel/misc.S @@ -1096,17 +1096,7 @@ _GLOBAL(_get_SP) * and exceptions as if the cpu had performed the load or store. */ -#if defined(CONFIG_4xx) || defined(CONFIG_E500) -_GLOBAL(cvt_fd) - lfs 0,0(r3) - stfd 0,0(r4) - blr - -_GLOBAL(cvt_df) - lfd 0,0(r3) - stfs 0,0(r4) - blr -#else +#ifdef CONFIG_PPC_FPU _GLOBAL(cvt_fd) lfd 0,-4(r5) /* load up fpscr value */ mtfsf 0xff,0 diff --git a/arch/ppc/kernel/pci.c b/arch/ppc/kernel/pci.c index 98f94b60204..47a15306823 100644 --- a/arch/ppc/kernel/pci.c +++ b/arch/ppc/kernel/pci.c @@ -1432,7 +1432,7 @@ pci_bus_to_hose(int bus) return NULL; } -void* +void __iomem * pci_bus_io_base(unsigned int bus) { struct pci_controller *hose; diff --git a/arch/ppc/kernel/ptrace.c b/arch/ppc/kernel/ptrace.c index 426b6f7d9de..59d59a8dc24 100644 --- a/arch/ppc/kernel/ptrace.c +++ b/arch/ppc/kernel/ptrace.c @@ -26,6 +26,7 @@ #include <linux/ptrace.h> #include <linux/user.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -356,7 +357,7 @@ int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -389,7 +390,7 @@ int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); set_single_step(child); diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c index e97ce635b99..5dfb42f1a15 100644 --- a/arch/ppc/kernel/setup.c +++ b/arch/ppc/kernel/setup.c @@ -221,27 +221,26 @@ int show_cpuinfo(struct seq_file *m, void *v) return err; } - switch (PVR_VER(pvr)) { - case 0x0020: /* 403 family */ - maj = PVR_MAJ(pvr) + 1; - min = PVR_MIN(pvr); - break; - case 0x1008: /* 740P/750P ?? */ - maj = ((pvr >> 8) & 0xFF) - 1; - min = pvr & 0xFF; - break; - case 0x8083: /* e300 */ - maj = PVR_MAJ(pvr); - min = PVR_MIN(pvr); - break; - case 0x8020: /* e500 */ + /* If we are a Freescale core do a simple check so + * we dont have to keep adding cases in the future */ + if ((PVR_VER(pvr) & 0x8000) == 0x8000) { maj = PVR_MAJ(pvr); min = PVR_MIN(pvr); - break; - default: - maj = (pvr >> 8) & 0xFF; - min = pvr & 0xFF; - break; + } else { + switch (PVR_VER(pvr)) { + case 0x0020: /* 403 family */ + maj = PVR_MAJ(pvr) + 1; + min = PVR_MIN(pvr); + break; + case 0x1008: /* 740P/750P ?? */ + maj = ((pvr >> 8) & 0xFF) - 1; + min = pvr & 0xFF; + break; + default: + maj = (pvr >> 8) & 0xFF; + min = pvr & 0xFF; + break; + } } seq_printf(m, "revision\t: %hd.%hd (pvr %04x %04x)\n", diff --git a/arch/ppc/kernel/signal.c b/arch/ppc/kernel/signal.c index 645eae19805..7c8437da09d 100644 --- a/arch/ppc/kernel/signal.c +++ b/arch/ppc/kernel/signal.c @@ -511,7 +511,7 @@ int sys_rt_sigreturn(int r3, int r4, int r5, int r6, int r7, int r8, } int sys_debug_setcontext(struct ucontext __user *ctx, - int ndbg, struct sig_dbg_op *dbg, + int ndbg, struct sig_dbg_op __user *dbg, int r6, int r7, int r8, struct pt_regs *regs) { @@ -632,7 +632,7 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, if (__put_user((unsigned long) ka->sa.sa_handler, &sc->handler) || __put_user(oldset->sig[0], &sc->oldmask) || __put_user(oldset->sig[1], &sc->_unused[3]) - || __put_user((struct pt_regs *)frame, &sc->regs) + || __put_user((struct pt_regs __user *)frame, &sc->regs) || __put_user(sig, &sc->signal)) goto badframe; diff --git a/arch/ppc/kernel/traps.c b/arch/ppc/kernel/traps.c index 002322a1f3c..f8e7e324a17 100644 --- a/arch/ppc/kernel/traps.c +++ b/arch/ppc/kernel/traps.c @@ -176,7 +176,7 @@ static inline int check_io_access(struct pt_regs *regs) #else #define get_mc_reason(regs) (mfspr(SPRN_MCSR)) #endif -#define REASON_FP 0 +#define REASON_FP ESR_FP #define REASON_ILLEGAL ESR_PIL #define REASON_PRIVILEGED ESR_PPR #define REASON_TRAP ESR_PTR @@ -403,7 +403,7 @@ static int emulate_string_inst(struct pt_regs *regs, u32 instword) u8 rA = (instword >> 16) & 0x1f; u8 NB_RB = (instword >> 11) & 0x1f; u32 num_bytes; - u32 EA; + unsigned long EA; int pos = 0; /* Early out if we are an invalid form of lswx */ diff --git a/arch/ppc/kernel/vmlinux.lds.S b/arch/ppc/kernel/vmlinux.lds.S index 0c0e714b84d..9353584fb71 100644 --- a/arch/ppc/kernel/vmlinux.lds.S +++ b/arch/ppc/kernel/vmlinux.lds.S @@ -145,6 +145,7 @@ SECTIONS __init_end = .; . = ALIGN(4096); + _sextratext = .; __pmac_begin = .; .pmac.text : { *(.pmac.text) } .pmac.data : { *(.pmac.data) } @@ -171,6 +172,7 @@ SECTIONS .openfirmware.data : { *(.openfirmware.data) } . = ALIGN(4096); __openfirmware_end = .; + _eextratext = .; __bss_start = .; .bss : diff --git a/arch/ppc/platforms/4xx/ebony.c b/arch/ppc/platforms/4xx/ebony.c index f63bca83e75..cd11734ef7c 100644 --- a/arch/ppc/platforms/4xx/ebony.c +++ b/arch/ppc/platforms/4xx/ebony.c @@ -149,7 +149,7 @@ ebony_map_irq(struct pci_dev *dev, unsigned char idsel, unsigned char pin) static void __init ebony_setup_pcix(void) { - void *pcix_reg_base; + void __iomem *pcix_reg_base; pcix_reg_base = ioremap64(PCIX0_REG_BASE, PCIX_REG_SIZE); @@ -210,9 +210,8 @@ ebony_setup_hose(void) hose->io_space.end = EBONY_PCI_UPPER_IO; hose->mem_space.start = EBONY_PCI_LOWER_MEM; hose->mem_space.end = EBONY_PCI_UPPER_MEM; - isa_io_base = - (unsigned long)ioremap64(EBONY_PCI_IO_BASE, EBONY_PCI_IO_SIZE); - hose->io_base_virt = (void *)isa_io_base; + hose->io_base_virt = ioremap64(EBONY_PCI_IO_BASE, EBONY_PCI_IO_SIZE); + isa_io_base = (unsigned long)hose->io_base_virt; setup_indirect_pci(hose, EBONY_PCI_CFGA_PLB32, diff --git a/arch/ppc/platforms/4xx/luan.c b/arch/ppc/platforms/4xx/luan.c index 1df2339f1f6..95359f748e7 100644 --- a/arch/ppc/platforms/4xx/luan.c +++ b/arch/ppc/platforms/4xx/luan.c @@ -223,9 +223,8 @@ luan_setup_hose(struct pci_controller *hose, hose->io_space.end = LUAN_PCIX_UPPER_IO; hose->mem_space.start = lower_mem; hose->mem_space.end = upper_mem; - isa_io_base = - (unsigned long)ioremap64(pcix_io_base, PCIX_IO_SIZE); - hose->io_base_virt = (void *)isa_io_base; + hose->io_base_virt = ioremap64(pcix_io_base, PCIX_IO_SIZE); + isa_io_base = (unsigned long) hose->io_base_virt; setup_indirect_pci(hose, cfga, cfgd); hose->set_cfg_type = 1; diff --git a/arch/ppc/platforms/4xx/ocotea.c b/arch/ppc/platforms/4xx/ocotea.c index 28de707434f..5f82a6bc704 100644 --- a/arch/ppc/platforms/4xx/ocotea.c +++ b/arch/ppc/platforms/4xx/ocotea.c @@ -227,9 +227,8 @@ ocotea_setup_hose(void) hose->io_space.end = OCOTEA_PCI_UPPER_IO; hose->mem_space.start = OCOTEA_PCI_LOWER_MEM; hose->mem_space.end = OCOTEA_PCI_UPPER_MEM; - isa_io_base = - (unsigned long)ioremap64(OCOTEA_PCI_IO_BASE, OCOTEA_PCI_IO_SIZE); - hose->io_base_virt = (void *)isa_io_base; + hose->io_base_virt = ioremap64(OCOTEA_PCI_IO_BASE, OCOTEA_PCI_IO_SIZE); + isa_io_base = (unsigned long) hose->io_base_virt; setup_indirect_pci(hose, OCOTEA_PCI_CFGA_PLB32, diff --git a/arch/ppc/platforms/chrp_pci.c b/arch/ppc/platforms/chrp_pci.c index 5bb6492ecf8..7d0ee308f66 100644 --- a/arch/ppc/platforms/chrp_pci.c +++ b/arch/ppc/platforms/chrp_pci.c @@ -129,7 +129,7 @@ static struct pci_ops rtas_pci_ops = rtas_write_config }; -volatile struct Hydra *Hydra = NULL; +volatile struct Hydra __iomem *Hydra = NULL; int __init hydra_init(void) @@ -175,13 +175,14 @@ chrp_pcibios_fixup(void) static void __init setup_python(struct pci_controller *hose, struct device_node *dev) { - u32 *reg, val; + u32 __iomem *reg; + u32 val; unsigned long addr = dev->addrs[0].address; setup_indirect_pci(hose, addr + 0xf8000, addr + 0xf8010); /* Clear the magic go-slow bit */ - reg = (u32 *) ioremap(dev->addrs[0].address + 0xf6000, 0x40); + reg = ioremap(dev->addrs[0].address + 0xf6000, 0x40); val = in_be32(®[12]); if (val & PRG_CL_RESET_VALID) { out_be32(®[12], val & ~PRG_CL_RESET_VALID); diff --git a/arch/ppc/platforms/chrp_setup.c b/arch/ppc/platforms/chrp_setup.c index f23c4f32076..57f29ab29bd 100644 --- a/arch/ppc/platforms/chrp_setup.c +++ b/arch/ppc/platforms/chrp_setup.c @@ -356,7 +356,7 @@ static void __init chrp_find_openpic(void) struct device_node *np; int len, i; unsigned int *iranges; - void *isu; + void __iomem *isu; np = find_type_devices("open-pic"); if (np == NULL || np->n_addrs == 0) diff --git a/arch/ppc/platforms/pmac_cache.S b/arch/ppc/platforms/pmac_cache.S index da34a9bc929..fb977de6b70 100644 --- a/arch/ppc/platforms/pmac_cache.S +++ b/arch/ppc/platforms/pmac_cache.S @@ -64,27 +64,39 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) mtspr SPRN_HID0,r4 /* Disable DPM */ sync - /* disp-flush L1 */ - li r4,0x4000 - mtctr r4 + /* Disp-flush L1. We have a weird problem here that I never + * totally figured out. On 750FX, using the ROM for the flush + * results in a non-working flush. We use that workaround for + * now until I finally understand what's going on. --BenH + */ + + /* ROM base by default */ lis r4,0xfff0 -1: lwzx r0,r0,r4 + mfpvr r3 + srwi r3,r3,16 + cmplwi cr0,r3,0x7000 + bne+ 1f + /* RAM base on 750FX */ + li r4,0 +1: li r4,0x4000 + mtctr r4 +1: lwz r0,0(r4) addi r4,r4,32 bdnz 1b sync isync - /* disable / invalidate / enable L1 data */ + /* Disable / invalidate / enable L1 data */ mfspr r3,SPRN_HID0 - rlwinm r0,r0,0,~HID0_DCE + rlwinm r3,r3,0,~(HID0_DCE | HID0_ICE) mtspr SPRN_HID0,r3 sync isync - ori r3,r3,HID0_DCE|HID0_DCI + ori r3,r3,(HID0_DCE|HID0_DCI|HID0_ICE|HID0_ICFI) sync isync mtspr SPRN_HID0,r3 - xori r3,r3,HID0_DCI + xori r3,r3,(HID0_DCI|HID0_ICFI) mtspr SPRN_HID0,r3 sync @@ -110,11 +122,20 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) lis r4,2 mtctr r4 lis r4,0xfff0 -1: lwzx r0,r0,r4 +1: lwz r0,0(r4) + addi r4,r4,32 + bdnz 1b + sync + isync + lis r4,2 + mtctr r4 + lis r4,0xfff0 +1: dcbf 0,r4 addi r4,r4,32 bdnz 1b sync isync + /* now disable L2 */ rlwinm r5,r5,0,~L2CR_L2E b 2f @@ -135,6 +156,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) mtspr SPRN_L2CR,r4 sync isync + + /* Wait for the invalidation to complete */ +1: mfspr r3,SPRN_L2CR + rlwinm. r0,r3,0,31,31 + bne 1b + + /* Clear L2I */ xoris r4,r4,L2CR_L2I@h sync mtspr SPRN_L2CR,r4 @@ -142,14 +170,16 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) /* now disable the L1 data cache */ mfspr r0,SPRN_HID0 - rlwinm r0,r0,0,~HID0_DCE + rlwinm r0,r0,0,~(HID0_DCE|HID0_ICE) mtspr SPRN_HID0,r0 sync isync /* Restore HID0[DPM] to whatever it was before */ sync - mtspr SPRN_HID0,r8 + mfspr r0,SPRN_HID0 + rlwimi r0,r8,0,11,11 /* Turn back HID0[DPM] */ + mtspr SPRN_HID0,r0 sync /* restore DR and EE */ @@ -201,7 +231,7 @@ flush_disable_745x: mtctr r4 li r4,0 1: - lwzx r0,r0,r4 + lwz r0,0(r4) addi r4,r4,32 /* Go to start of next cache line */ bdnz 1b isync diff --git a/arch/ppc/platforms/pmac_feature.c b/arch/ppc/platforms/pmac_feature.c index 46cbf36722d..867336ad5d3 100644 --- a/arch/ppc/platforms/pmac_feature.c +++ b/arch/ppc/platforms/pmac_feature.c @@ -1590,6 +1590,114 @@ intrepid_shutdown(struct macio_chip* macio, int sleep_mode) mdelay(10); } + +void __pmac pmac_tweak_clock_spreading(int enable) +{ + struct macio_chip* macio = &macio_chips[0]; + + /* Hack for doing clock spreading on some machines PowerBooks and + * iBooks. This implements the "platform-do-clockspreading" OF + * property as decoded manually on various models. For safety, we also + * check the product ID in the device-tree in cases we'll whack the i2c + * chip to make reasonably sure we won't set wrong values in there + * + * Of course, ultimately, we have to implement a real parser for + * the platform-do-* stuff... + */ + + if (macio->type == macio_intrepid) { + if (enable) + UN_OUT(UNI_N_CLOCK_SPREADING, 2); + else + UN_OUT(UNI_N_CLOCK_SPREADING, 0); + mdelay(40); + } + + while (machine_is_compatible("PowerBook5,2") || + machine_is_compatible("PowerBook5,3") || + machine_is_compatible("PowerBook6,2") || + machine_is_compatible("PowerBook6,3")) { + struct device_node *ui2c = of_find_node_by_type(NULL, "i2c"); + struct device_node *dt = of_find_node_by_name(NULL, "device-tree"); + u8 buffer[9]; + u32 *productID; + int i, rc, changed = 0; + + if (dt == NULL) + break; + productID = (u32 *)get_property(dt, "pid#", NULL); + if (productID == NULL) + break; + while(ui2c) { + struct device_node *p = of_get_parent(ui2c); + if (p && !strcmp(p->name, "uni-n")) + break; + ui2c = of_find_node_by_type(ui2c, "i2c"); + } + if (ui2c == NULL) + break; + DBG("Trying to bump clock speed for PID: %08x...\n", *productID); + rc = pmac_low_i2c_open(ui2c, 1); + if (rc != 0) + break; + pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_combined); + rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_read, 0x80, buffer, 9); + DBG("read result: %d,", rc); + if (rc != 0) { + pmac_low_i2c_close(ui2c); + break; + } + for (i=0; i<9; i++) + DBG(" %02x", buffer[i]); + DBG("\n"); + + switch(*productID) { + case 0x1182: /* AlBook 12" rev 2 */ + case 0x1183: /* iBook G4 12" */ + buffer[0] = (buffer[0] & 0x8f) | 0x70; + buffer[2] = (buffer[2] & 0x7f) | 0x00; + buffer[5] = (buffer[5] & 0x80) | 0x31; + buffer[6] = (buffer[6] & 0x40) | 0xb0; + buffer[7] = (buffer[7] & 0x00) | (enable ? 0xc0 : 0xba); + buffer[8] = (buffer[8] & 0x00) | 0x30; + changed = 1; + break; + case 0x3142: /* AlBook 15" (ATI M10) */ + case 0x3143: /* AlBook 17" (ATI M10) */ + buffer[0] = (buffer[0] & 0xaf) | 0x50; + buffer[2] = (buffer[2] & 0x7f) | 0x00; + buffer[5] = (buffer[5] & 0x80) | 0x31; + buffer[6] = (buffer[6] & 0x40) | 0xb0; + buffer[7] = (buffer[7] & 0x00) | (enable ? 0xd0 : 0xc0); + buffer[8] = (buffer[8] & 0x00) | 0x30; + changed = 1; + break; + default: + DBG("i2c-hwclock: Machine model not handled\n"); + break; + } + if (!changed) { + pmac_low_i2c_close(ui2c); + break; + } + pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_stdsub); + rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_write, 0x80, buffer, 9); + DBG("write result: %d,", rc); + pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_combined); + rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_read, 0x80, buffer, 9); + DBG("read result: %d,", rc); + if (rc != 0) { + pmac_low_i2c_close(ui2c); + break; + } + for (i=0; i<9; i++) + DBG(" %02x", buffer[i]); + pmac_low_i2c_close(ui2c); + break; + } +} + + static int __pmac core99_sleep(void) { @@ -1601,12 +1709,6 @@ core99_sleep(void) macio->type != macio_intrepid) return -ENODEV; - /* The device-tree contains that in the hwclock node */ - if (macio->type == macio_intrepid) { - UN_OUT(UNI_N_CLOCK_SPREADING, 0); - mdelay(40); - } - /* We power off the wireless slot in case it was not done * by the driver. We don't power it on automatically however */ @@ -1749,12 +1851,6 @@ core99_wake_up(void) UN_OUT(UNI_N_CLOCK_CNTL, save_unin_clock_ctl); udelay(100); - /* Restore clock spreading */ - if (macio->type == macio_intrepid) { - UN_OUT(UNI_N_CLOCK_SPREADING, 2); - mdelay(40); - } - return 0; } @@ -2149,7 +2245,7 @@ static struct pmac_mb_def pmac_mb_defs[] __pmacdata = { }, { "PowerBook1,1", "PowerBook 101 (Lombard)", PMAC_TYPE_101_PBOOK, paddington_features, - PMAC_MB_MAY_SLEEP | PMAC_MB_MOBILE + PMAC_MB_CAN_SLEEP | PMAC_MB_MOBILE }, { "PowerBook2,1", "iBook (first generation)", PMAC_TYPE_ORIG_IBOOK, core99_features, @@ -2718,97 +2814,11 @@ set_initial_features(void) MACIO_BIC(HEATHROW_FCR, HRW_SOUND_POWER_N); } - /* Hack for bumping clock speed on the new PowerBooks and the - * iBook G4. This implements the "platform-do-clockspreading" OF - * property. For safety, we also check the product ID in the - * device-tree to make reasonably sure we won't set wrong values - * in the clock chip. - * - * Of course, ultimately, we have to implement a real parser for - * the platform-do-* stuff... + /* Some machine models need the clock chip to be properly setup for + * clock spreading now. This should be a platform function but we + * don't do these at the moment */ - while (machine_is_compatible("PowerBook5,2") || - machine_is_compatible("PowerBook5,3") || - machine_is_compatible("PowerBook6,2") || - machine_is_compatible("PowerBook6,3")) { - struct device_node *ui2c = of_find_node_by_type(NULL, "i2c"); - struct device_node *dt = of_find_node_by_name(NULL, "device-tree"); - u8 buffer[9]; - u32 *productID; - int i, rc, changed = 0; - - if (dt == NULL) - break; - productID = (u32 *)get_property(dt, "pid#", NULL); - if (productID == NULL) - break; - while(ui2c) { - struct device_node *p = of_get_parent(ui2c); - if (p && !strcmp(p->name, "uni-n")) - break; - ui2c = of_find_node_by_type(ui2c, "i2c"); - } - if (ui2c == NULL) - break; - DBG("Trying to bump clock speed for PID: %08x...\n", *productID); - rc = pmac_low_i2c_open(ui2c, 1); - if (rc != 0) - break; - pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_combined); - rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_read, 0x80, buffer, 9); - DBG("read result: %d,", rc); - if (rc != 0) { - pmac_low_i2c_close(ui2c); - break; - } - for (i=0; i<9; i++) - DBG(" %02x", buffer[i]); - DBG("\n"); - - switch(*productID) { - case 0x1182: /* AlBook 12" rev 2 */ - case 0x1183: /* iBook G4 12" */ - buffer[0] = (buffer[0] & 0x8f) | 0x70; - buffer[2] = (buffer[2] & 0x7f) | 0x00; - buffer[5] = (buffer[5] & 0x80) | 0x31; - buffer[6] = (buffer[6] & 0x40) | 0xb0; - buffer[7] = (buffer[7] & 0x00) | 0xc0; - buffer[8] = (buffer[8] & 0x00) | 0x30; - changed = 1; - break; - case 0x3142: /* AlBook 15" (ATI M10) */ - case 0x3143: /* AlBook 17" (ATI M10) */ - buffer[0] = (buffer[0] & 0xaf) | 0x50; - buffer[2] = (buffer[2] & 0x7f) | 0x00; - buffer[5] = (buffer[5] & 0x80) | 0x31; - buffer[6] = (buffer[6] & 0x40) | 0xb0; - buffer[7] = (buffer[7] & 0x00) | 0xd0; - buffer[8] = (buffer[8] & 0x00) | 0x30; - changed = 1; - break; - default: - DBG("i2c-hwclock: Machine model not handled\n"); - break; - } - if (!changed) { - pmac_low_i2c_close(ui2c); - break; - } - pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_stdsub); - rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_write, 0x80, buffer, 9); - DBG("write result: %d,", rc); - pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_combined); - rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_read, 0x80, buffer, 9); - DBG("read result: %d,", rc); - if (rc != 0) { - pmac_low_i2c_close(ui2c); - break; - } - for (i=0; i<9; i++) - DBG(" %02x", buffer[i]); - pmac_low_i2c_close(ui2c); - break; - } + pmac_tweak_clock_spreading(1); #endif /* CONFIG_POWER4 */ diff --git a/arch/ppc/platforms/pmac_low_i2c.c b/arch/ppc/platforms/pmac_low_i2c.c index d07579f2b8b..08583fce169 100644 --- a/arch/ppc/platforms/pmac_low_i2c.c +++ b/arch/ppc/platforms/pmac_low_i2c.c @@ -54,7 +54,7 @@ struct low_i2c_host int mode; /* Current mode */ int channel; /* Current channel */ int num_channels; /* Number of channels */ - unsigned long base; /* For keywest-i2c, base address */ + void __iomem * base; /* For keywest-i2c, base address */ int bsteps; /* And register stepping */ int speed; /* And speed */ }; @@ -154,14 +154,12 @@ static const char *__kw_state_names[] = { static inline u8 __kw_read_reg(struct low_i2c_host *host, reg_t reg) { - return in_8(((volatile u8 *)host->base) - + (((unsigned)reg) << host->bsteps)); + return in_8(host->base + (((unsigned)reg) << host->bsteps)); } static inline void __kw_write_reg(struct low_i2c_host *host, reg_t reg, u8 val) { - out_8(((volatile u8 *)host->base) - + (((unsigned)reg) << host->bsteps), val); + out_8(host->base + (((unsigned)reg) << host->bsteps), val); (void)__kw_read_reg(host, reg_subaddr); } @@ -370,7 +368,7 @@ static void keywest_low_i2c_add(struct device_node *np) break; } host->mode = pmac_low_i2c_mode_std; - host->base = (unsigned long)ioremap(np->addrs[0].address + aoffset, + host->base = ioremap(np->addrs[0].address + aoffset, np->addrs[0].size); host->func = keywest_low_i2c_func; } diff --git a/arch/ppc/platforms/pmac_sleep.S b/arch/ppc/platforms/pmac_sleep.S index 3139b6766ad..f459ade1bd6 100644 --- a/arch/ppc/platforms/pmac_sleep.S +++ b/arch/ppc/platforms/pmac_sleep.S @@ -267,6 +267,10 @@ grackle_wake_up: /* Restore various CPU config stuffs */ bl __restore_cpu_setup + /* Make sure all FPRs have been initialized */ + bl reloc_offset + bl __init_fpu_registers + /* Invalidate & enable L1 cache, we don't care about * whatever the ROM may have tried to write to memory */ diff --git a/arch/ppc/platforms/pmac_smp.c b/arch/ppc/platforms/pmac_smp.c index 731841f9a5b..8e049dab4e6 100644 --- a/arch/ppc/platforms/pmac_smp.c +++ b/arch/ppc/platforms/pmac_smp.c @@ -91,11 +91,11 @@ extern void __secondary_start_psurge3(void); /* Temporary horrible hack */ #define PSURGE_QUAD_BIC(r, v) (PSURGE_QUAD_OUT((r), PSURGE_QUAD_IN(r) & ~(v))) /* virtual addresses for the above */ -static volatile u8 *hhead_base; -static volatile u8 *quad_base; -static volatile u32 *psurge_pri_intr; -static volatile u8 *psurge_sec_intr; -static volatile u32 *psurge_start; +static volatile u8 __iomem *hhead_base; +static volatile u8 __iomem *quad_base; +static volatile u32 __iomem *psurge_pri_intr; +static volatile u8 __iomem *psurge_sec_intr; +static volatile u32 __iomem *psurge_start; /* values for psurge_type */ #define PSURGE_NONE -1 @@ -322,10 +322,10 @@ static int __init smp_psurge_probe(void) /* All released cards using this HW design have 4 CPUs */ ncpus = 4; } else { - iounmap((void *) quad_base); + iounmap(quad_base); if ((in_8(hhead_base + HHEAD_CONFIG) & 0x02) == 0) { /* not a dual-cpu card */ - iounmap((void *) hhead_base); + iounmap(hhead_base); psurge_type = PSURGE_NONE; return 1; } diff --git a/arch/ppc/platforms/pmac_time.c b/arch/ppc/platforms/pmac_time.c index 09636546f44..de60ccc7db9 100644 --- a/arch/ppc/platforms/pmac_time.c +++ b/arch/ppc/platforms/pmac_time.c @@ -165,7 +165,7 @@ int __init via_calibrate_decr(void) { struct device_node *vias; - volatile unsigned char *via; + volatile unsigned char __iomem *via; int count = VIA_TIMER_FREQ_6 / 100; unsigned int dstart, dend; @@ -176,8 +176,7 @@ via_calibrate_decr(void) vias = find_devices("via"); if (vias == 0 || vias->n_addrs == 0) return 0; - via = (volatile unsigned char *) - ioremap(vias->addrs[0].address, vias->addrs[0].size); + via = ioremap(vias->addrs[0].address, vias->addrs[0].size); /* set timer 1 for continuous interrupts */ out_8(&via[ACR], (via[ACR] & ~T1MODE) | T1MODE_CONT); @@ -202,7 +201,7 @@ via_calibrate_decr(void) printk(KERN_INFO "via_calibrate_decr: ticks per jiffy = %u (%u ticks)\n", tb_ticks_per_jiffy, dstart - dend); - iounmap((void*)via); + iounmap(via); return 1; } diff --git a/arch/ppc/platforms/radstone_ppc7d.c b/arch/ppc/platforms/radstone_ppc7d.c index 2a99b43737a..c30607a972d 100644 --- a/arch/ppc/platforms/radstone_ppc7d.c +++ b/arch/ppc/platforms/radstone_ppc7d.c @@ -68,6 +68,7 @@ #define PPC7D_RST_PIN 17 /* GPP17 */ extern u32 mv64360_irq_base; +extern spinlock_t rtc_lock; static struct mv64x60_handle bh; static int ppc7d_has_alma; @@ -75,6 +76,11 @@ static int ppc7d_has_alma; extern void gen550_progress(char *, unsigned short); extern void gen550_init(int, struct uart_port *); +/* FIXME - move to h file */ +extern int ds1337_do_command(int id, int cmd, void *arg); +#define DS1337_GET_DATE 0 +#define DS1337_SET_DATE 1 + /* residual data */ unsigned char __res[sizeof(bd_t)]; @@ -253,6 +259,8 @@ static int ppc7d_show_cpuinfo(struct seq_file *m) u8 val1, val2; static int flash_sizes[4] = { 64, 32, 0, 16 }; static int flash_banks[4] = { 4, 3, 2, 1 }; + static int sdram_bank_sizes[4] = { 128, 256, 512, 1 }; + int sdram_num_banks = 2; static char *pci_modes[] = { "PCI33", "PCI66", "Unknown", "Unknown", "PCIX33", "PCIX66", @@ -279,13 +287,17 @@ static int ppc7d_show_cpuinfo(struct seq_file *m) (val1 == PPC7D_CPLD_MB_TYPE_PLL_100) ? 100 : (val1 == PPC7D_CPLD_MB_TYPE_PLL_64) ? 64 : 0); + val = inb(PPC7D_CPLD_MEM_CONFIG); + if (val & PPC7D_CPLD_SDRAM_BANK_NUM_MASK) sdram_num_banks--; + val = inb(PPC7D_CPLD_MEM_CONFIG_EXTEND); - val1 = val & PPC7D_CPLD_SDRAM_BANK_SIZE_MASK; - seq_printf(m, "SDRAM\t\t: %d%c", - (val1 == PPC7D_CPLD_SDRAM_BANK_SIZE_128M) ? 128 : - (val1 == PPC7D_CPLD_SDRAM_BANK_SIZE_256M) ? 256 : - (val1 == PPC7D_CPLD_SDRAM_BANK_SIZE_512M) ? 512 : 1, - (val1 == PPC7D_CPLD_SDRAM_BANK_SIZE_1G) ? 'G' : 'M'); + val1 = (val & PPC7D_CPLD_SDRAM_BANK_SIZE_MASK) >> 6; + seq_printf(m, "SDRAM\t\t: %d banks of %d%c, total %d%c", + sdram_num_banks, + sdram_bank_sizes[val1], + (sdram_bank_sizes[val1] < 128) ? 'G' : 'M', + sdram_num_banks * sdram_bank_sizes[val1], + (sdram_bank_sizes[val1] < 128) ? 'G' : 'M'); if (val2 & PPC7D_CPLD_MB_TYPE_ECC_FITTED_MASK) { seq_printf(m, " [ECC %sabled]", (val2 & PPC7D_CPLD_MB_TYPE_ECC_ENABLE_MASK) ? "en" : @@ -1236,6 +1248,38 @@ static void __init ppc7d_setup_arch(void) printk(KERN_INFO "Radstone Technology PPC7D\n"); if (ppc_md.progress) ppc_md.progress("ppc7d_setup_arch: exit", 0); + +} + +/* Real Time Clock support. + * PPC7D has a DS1337 accessed by I2C. + */ +static ulong ppc7d_get_rtc_time(void) +{ + struct rtc_time tm; + int result; + + spin_lock(&rtc_lock); + result = ds1337_do_command(0, DS1337_GET_DATE, &tm); + spin_unlock(&rtc_lock); + + if (result == 0) + result = mktime(tm.tm_year, tm.tm_mon, tm.tm_mday, tm.tm_hour, tm.tm_min, tm.tm_sec); + + return result; +} + +static int ppc7d_set_rtc_time(unsigned long nowtime) +{ + struct rtc_time tm; + int result; + + spin_lock(&rtc_lock); + to_tm(nowtime, &tm); + result = ds1337_do_command(0, DS1337_SET_DATE, &tm); + spin_unlock(&rtc_lock); + + return result; } /* This kernel command line parameter can be used to have the target @@ -1293,6 +1337,10 @@ static void ppc7d_init2(void) data8 |= 0x07; outb(data8, PPC7D_CPLD_LEDS); + /* Hook up RTC. We couldn't do this earlier because we need the I2C subsystem */ + ppc_md.set_rtc_time = ppc7d_set_rtc_time; + ppc_md.get_rtc_time = ppc7d_get_rtc_time; + pr_debug("%s: exit\n", __FUNCTION__); } diff --git a/arch/ppc/platforms/radstone_ppc7d.h b/arch/ppc/platforms/radstone_ppc7d.h index 4546fff2b0c..938375510be 100644 --- a/arch/ppc/platforms/radstone_ppc7d.h +++ b/arch/ppc/platforms/radstone_ppc7d.h @@ -240,6 +240,7 @@ #define PPC7D_CPLD_FLASH_CNTL 0x086E /* MEMORY_CONFIG_EXTEND */ +#define PPC7D_CPLD_SDRAM_BANK_NUM_MASK 0x02 #define PPC7D_CPLD_SDRAM_BANK_SIZE_MASK 0xc0 #define PPC7D_CPLD_SDRAM_BANK_SIZE_128M 0 #define PPC7D_CPLD_SDRAM_BANK_SIZE_256M 0x40 diff --git a/arch/ppc/syslib/cpm2_pic.c b/arch/ppc/syslib/cpm2_pic.c index 954b07fc1df..c867be6981c 100644 --- a/arch/ppc/syslib/cpm2_pic.c +++ b/arch/ppc/syslib/cpm2_pic.c @@ -107,6 +107,11 @@ static void cpm2_end_irq(unsigned int irq_nr) simr = &(cpm2_immr->im_intctl.ic_simrh); ppc_cached_irq_mask[word] |= 1 << bit; simr[word] = ppc_cached_irq_mask[word]; + /* + * Work around large numbers of spurious IRQs on PowerPC 82xx + * systems. + */ + mb(); } } diff --git a/arch/ppc/syslib/m8260_pci.c b/arch/ppc/syslib/m8260_pci.c index bd564fb35ab..057cc3f8ff3 100644 --- a/arch/ppc/syslib/m8260_pci.c +++ b/arch/ppc/syslib/m8260_pci.c @@ -171,10 +171,9 @@ void __init m8260_find_bridges(void) m8260_setup_pci(hose); hose->pci_mem_offset = MPC826x_PCI_MEM_OFFSET; - isa_io_base = - (unsigned long) ioremap(MPC826x_PCI_IO_BASE, + hose->io_base_virt = ioremap(MPC826x_PCI_IO_BASE, MPC826x_PCI_IO_SIZE); - hose->io_base_virt = (void *) isa_io_base; + isa_io_base = (unsigned long) hose->io_base_virt; /* setup resources */ pci_init_resource(&hose->mem_resources[0], diff --git a/arch/ppc/syslib/mpc52xx_pci.c b/arch/ppc/syslib/mpc52xx_pci.c index c723efd954a..59cf3e8bd1a 100644 --- a/arch/ppc/syslib/mpc52xx_pci.c +++ b/arch/ppc/syslib/mpc52xx_pci.c @@ -205,13 +205,11 @@ mpc52xx_find_bridges(void) hose->pci_mem_offset = MPC52xx_PCI_MEM_OFFSET; - isa_io_base = - (unsigned long) ioremap(MPC52xx_PCI_IO_BASE, - MPC52xx_PCI_IO_SIZE); - hose->io_base_virt = (void *) isa_io_base; + hose->io_base_virt = ioremap(MPC52xx_PCI_IO_BASE, MPC52xx_PCI_IO_SIZE); + isa_io_base = (unsigned long) hose->io_base_virt; hose->cfg_addr = &pci_regs->car; - hose->cfg_data = (void __iomem *) isa_io_base; + hose->cfg_data = hose->io_base_virt; /* Setup resources */ pci_init_resource(&hose->mem_resources[0], diff --git a/arch/ppc/syslib/ppc85xx_setup.c b/arch/ppc/syslib/ppc85xx_setup.c index 81f1968c326..152c3ef1312 100644 --- a/arch/ppc/syslib/ppc85xx_setup.c +++ b/arch/ppc/syslib/ppc85xx_setup.c @@ -280,16 +280,14 @@ mpc85xx_setup_hose(void) hose_a->io_space.end = MPC85XX_PCI1_UPPER_IO; hose_a->io_base_phys = MPC85XX_PCI1_IO_BASE; #ifdef CONFIG_85xx_PCI2 - isa_io_base = - (unsigned long) ioremap(MPC85XX_PCI1_IO_BASE, + hose_a->io_base_virt = ioremap(MPC85XX_PCI1_IO_BASE, MPC85XX_PCI1_IO_SIZE + MPC85XX_PCI2_IO_SIZE); #else - isa_io_base = - (unsigned long) ioremap(MPC85XX_PCI1_IO_BASE, + hose_a->io_base_virt = ioremap(MPC85XX_PCI1_IO_BASE, MPC85XX_PCI1_IO_SIZE); #endif - hose_a->io_base_virt = (void *) isa_io_base; + isa_io_base = (unsigned long)hose_a->io_base_virt; /* setup resources */ pci_init_resource(&hose_a->mem_resources[0], @@ -329,8 +327,8 @@ mpc85xx_setup_hose(void) hose_b->io_space.start = MPC85XX_PCI2_LOWER_IO; hose_b->io_space.end = MPC85XX_PCI2_UPPER_IO; hose_b->io_base_phys = MPC85XX_PCI2_IO_BASE; - hose_b->io_base_virt = (void *) isa_io_base + MPC85XX_PCI1_IO_SIZE; - + hose_b->io_base_virt = hose_a->io_base_virt + MPC85XX_PCI1_IO_SIZE; + /* setup resources */ pci_init_resource(&hose_b->mem_resources[0], MPC85XX_PCI2_LOWER_MEM, diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig index ef1f05e437c..257ff66d83c 100644 --- a/arch/ppc64/Kconfig +++ b/arch/ppc64/Kconfig @@ -40,6 +40,10 @@ config COMPAT bool default y +config SCHED_NO_NO_OMIT_FRAME_POINTER + bool + default y + # We optimistically allocate largepages from the VM, so make the limit # large enough (16MB). This badly named config option is actually # max order + 1 @@ -293,6 +297,9 @@ config SECCOMP endmenu +config ISA_DMA_API + bool + default y menu "General setup" diff --git a/arch/ppc64/Kconfig.debug b/arch/ppc64/Kconfig.debug index e341a129da8..46b1ce58da3 100644 --- a/arch/ppc64/Kconfig.debug +++ b/arch/ppc64/Kconfig.debug @@ -5,6 +5,9 @@ source "lib/Kconfig.debug" config DEBUG_STACKOVERFLOW bool "Check for stack overflows" depends on DEBUG_KERNEL + help + This option will cause messages to be printed if free stack space + drops below a certain limit. config KPROBES bool "Kprobes" diff --git a/arch/ppc64/Makefile b/arch/ppc64/Makefile index d33e20bcc52..691f3008e69 100644 --- a/arch/ppc64/Makefile +++ b/arch/ppc64/Makefile @@ -56,13 +56,20 @@ LDFLAGS_vmlinux := -Bstatic -e $(KERNELLOAD) -Ttext $(KERNELLOAD) CFLAGS += -msoft-float -pipe -mminimal-toc -mtraceback=none \ -mcall-aixdesc +GCC_VERSION := $(call cc-version) +GCC_BROKEN_VEC := $(shell if [ $(GCC_VERSION) -lt 0400 ] ; then echo "y"; fi ;) + ifeq ($(CONFIG_POWER4_ONLY),y) ifeq ($(CONFIG_ALTIVEC),y) +ifeq ($(GCC_BROKEN_VEC),y) CFLAGS += $(call cc-option,-mcpu=970) else CFLAGS += $(call cc-option,-mcpu=power4) endif else + CFLAGS += $(call cc-option,-mcpu=power4) +endif +else CFLAGS += $(call cc-option,-mtune=power4) endif diff --git a/arch/ppc64/boot/addnote.c b/arch/ppc64/boot/addnote.c index 66ff8103bf4..719663a694b 100644 --- a/arch/ppc64/boot/addnote.c +++ b/arch/ppc64/boot/addnote.c @@ -19,6 +19,7 @@ #include <unistd.h> #include <string.h> +/* CHRP note section */ char arch[] = "PowerPC"; #define N_DESCR 6 @@ -31,6 +32,29 @@ unsigned int descr[N_DESCR] = { 0x4000, /* load-base */ }; +/* RPA note section */ +char rpaname[] = "IBM,RPA-Client-Config"; + +/* + * Note: setting ignore_my_client_config *should* mean that OF ignores + * all the other fields, but there is a firmware bug which means that + * it looks at the splpar field at least. So these values need to be + * reasonable. + */ +#define N_RPA_DESCR 8 +unsigned int rpanote[N_RPA_DESCR] = { + 0, /* lparaffinity */ + 64, /* min_rmo_size */ + 0, /* min_rmo_percent */ + 40, /* max_pft_size */ + 1, /* splpar */ + -1, /* min_load */ + 0, /* new_mem_def */ + 1, /* ignore_my_client_config */ +}; + +#define ROUNDUP(len) (((len) + 3) & ~3) + unsigned char buf[512]; #define GET_16BE(off) ((buf[off] << 8) + (buf[(off)+1])) @@ -69,7 +93,7 @@ main(int ac, char **av) { int fd, n, i; int ph, ps, np; - int nnote, ns; + int nnote, nnote2, ns; if (ac != 2) { fprintf(stderr, "Usage: %s elf-file\n", av[0]); @@ -81,7 +105,8 @@ main(int ac, char **av) exit(1); } - nnote = strlen(arch) + 1 + (N_DESCR + 3) * 4; + nnote = 12 + ROUNDUP(strlen(arch) + 1) + sizeof(descr); + nnote2 = 12 + ROUNDUP(strlen(rpaname) + 1) + sizeof(rpanote); n = read(fd, buf, sizeof(buf)); if (n < 0) { @@ -104,7 +129,7 @@ main(int ac, char **av) np = GET_16BE(E_PHNUM); if (ph < E_HSIZE || ps < PH_HSIZE || np < 1) goto notelf; - if (ph + (np + 1) * ps + nnote > n) + if (ph + (np + 2) * ps + nnote + nnote2 > n) goto nospace; for (i = 0; i < np; ++i) { @@ -117,12 +142,12 @@ main(int ac, char **av) } /* XXX check that the area we want to use is all zeroes */ - for (i = 0; i < ps + nnote; ++i) + for (i = 0; i < 2 * ps + nnote + nnote2; ++i) if (buf[ph + i] != 0) goto nospace; /* fill in the program header entry */ - ns = ph + ps; + ns = ph + 2 * ps; PUT_32BE(ph + PH_TYPE, PT_NOTE); PUT_32BE(ph + PH_OFFSET, ns); PUT_32BE(ph + PH_FILESZ, nnote); @@ -134,11 +159,26 @@ main(int ac, char **av) PUT_32BE(ns + 8, 0x1275); strcpy(&buf[ns + 12], arch); ns += 12 + strlen(arch) + 1; - for (i = 0; i < N_DESCR; ++i) - PUT_32BE(ns + i * 4, descr[i]); + for (i = 0; i < N_DESCR; ++i, ns += 4) + PUT_32BE(ns, descr[i]); + + /* fill in the second program header entry and the RPA note area */ + ph += ps; + PUT_32BE(ph + PH_TYPE, PT_NOTE); + PUT_32BE(ph + PH_OFFSET, ns); + PUT_32BE(ph + PH_FILESZ, nnote2); + + /* fill in the note area we point to */ + PUT_32BE(ns, strlen(rpaname) + 1); + PUT_32BE(ns + 4, sizeof(rpanote)); + PUT_32BE(ns + 8, 0x12759999); + strcpy(&buf[ns + 12], rpaname); + ns += 12 + ROUNDUP(strlen(rpaname) + 1); + for (i = 0; i < N_RPA_DESCR; ++i, ns += 4) + PUT_32BE(ns, rpanote[i]); /* Update the number of program headers */ - PUT_16BE(E_PHNUM, np + 1); + PUT_16BE(E_PHNUM, np + 2); /* write back */ lseek(fd, (long) 0, SEEK_SET); @@ -155,11 +195,11 @@ main(int ac, char **av) exit(0); notelf: - fprintf(stderr, "%s does not appear to be an ELF file\n", av[0]); + fprintf(stderr, "%s does not appear to be an ELF file\n", av[1]); exit(1); nospace: fprintf(stderr, "sorry, I can't find space in %s to put the note\n", - av[0]); + av[1]); exit(1); } diff --git a/arch/ppc64/boot/main.c b/arch/ppc64/boot/main.c index b0fa86ad8b1..da12ea2ca46 100644 --- a/arch/ppc64/boot/main.c +++ b/arch/ppc64/boot/main.c @@ -14,7 +14,6 @@ #include <linux/string.h> #include <asm/processor.h> #include <asm/page.h> -#include <asm/bootinfo.h> extern void *finddevice(const char *); extern int getprop(void *, const char *, void *, int); diff --git a/arch/ppc64/boot/start.c b/arch/ppc64/boot/start.c deleted file mode 100644 index ea247e79b55..00000000000 --- a/arch/ppc64/boot/start.c +++ /dev/null @@ -1,654 +0,0 @@ -/* - * Copyright (C) Paul Mackerras 1997. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <stdarg.h> -#include <linux/types.h> -#include <linux/string.h> -#include <linux/ctype.h> - -#include <asm/div64.h> - -int (*prom)(void *); - -void *chosen_handle; -void *stdin; -void *stdout; -void *stderr; - -void exit(void); -void *finddevice(const char *name); -int getprop(void *phandle, const char *name, void *buf, int buflen); -void chrpboot(int a1, int a2, void *prom); /* in main.c */ - -void printk(char *fmt, ...); - -void -start(int a1, int a2, void *promptr) -{ - prom = (int (*)(void *)) promptr; - chosen_handle = finddevice("/chosen"); - if (chosen_handle == (void *) -1) - exit(); - if (getprop(chosen_handle, "stdout", &stdout, sizeof(stdout)) != 4) - exit(); - stderr = stdout; - if (getprop(chosen_handle, "stdin", &stdin, sizeof(stdin)) != 4) - exit(); - - chrpboot(a1, a2, promptr); - for (;;) - exit(); -} - -int -write(void *handle, void *ptr, int nb) -{ - struct prom_args { - char *service; - int nargs; - int nret; - void *ihandle; - void *addr; - int len; - int actual; - } args; - - args.service = "write"; - args.nargs = 3; - args.nret = 1; - args.ihandle = handle; - args.addr = ptr; - args.len = nb; - args.actual = -1; - (*prom)(&args); - return args.actual; -} - -int -read(void *handle, void *ptr, int nb) -{ - struct prom_args { - char *service; - int nargs; - int nret; - void *ihandle; - void *addr; - int len; - int actual; - } args; - - args.service = "read"; - args.nargs = 3; - args.nret = 1; - args.ihandle = handle; - args.addr = ptr; - args.len = nb; - args.actual = -1; - (*prom)(&args); - return args.actual; -} - -void -exit() -{ - struct prom_args { - char *service; - } args; - - for (;;) { - args.service = "exit"; - (*prom)(&args); - } -} - -void -pause(void) -{ - struct prom_args { - char *service; - } args; - - args.service = "enter"; - (*prom)(&args); -} - -void * -finddevice(const char *name) -{ - struct prom_args { - char *service; - int nargs; - int nret; - const char *devspec; - void *phandle; - } args; - - args.service = "finddevice"; - args.nargs = 1; - args.nret = 1; - args.devspec = name; - args.phandle = (void *) -1; - (*prom)(&args); - return args.phandle; -} - -void * -claim(unsigned long virt, unsigned long size, unsigned long align) -{ - struct prom_args { - char *service; - int nargs; - int nret; - unsigned int virt; - unsigned int size; - unsigned int align; - void *ret; - } args; - - args.service = "claim"; - args.nargs = 3; - args.nret = 1; - args.virt = virt; - args.size = size; - args.align = align; - (*prom)(&args); - return args.ret; -} - -int -getprop(void *phandle, const char *name, void *buf, int buflen) -{ - struct prom_args { - char *service; - int nargs; - int nret; - void *phandle; - const char *name; - void *buf; - int buflen; - int size; - } args; - - args.service = "getprop"; - args.nargs = 4; - args.nret = 1; - args.phandle = phandle; - args.name = name; - args.buf = buf; - args.buflen = buflen; - args.size = -1; - (*prom)(&args); - return args.size; -} - -int -putc(int c, void *f) -{ - char ch = c; - - if (c == '\n') - putc('\r', f); - return write(f, &ch, 1) == 1? c: -1; -} - -int -putchar(int c) -{ - return putc(c, stdout); -} - -int -fputs(char *str, void *f) -{ - int n = strlen(str); - - return write(f, str, n) == n? 0: -1; -} - -int -readchar(void) -{ - char ch; - - for (;;) { - switch (read(stdin, &ch, 1)) { - case 1: - return ch; - case -1: - printk("read(stdin) returned -1\r\n"); - return -1; - } - } -} - -static char line[256]; -static char *lineptr; -static int lineleft; - -int -getchar(void) -{ - int c; - - if (lineleft == 0) { - lineptr = line; - for (;;) { - c = readchar(); - if (c == -1 || c == 4) - break; - if (c == '\r' || c == '\n') { - *lineptr++ = '\n'; - putchar('\n'); - break; - } - switch (c) { - case 0177: - case '\b': - if (lineptr > line) { - putchar('\b'); - putchar(' '); - putchar('\b'); - --lineptr; - } - break; - case 'U' & 0x1F: - while (lineptr > line) { - putchar('\b'); - putchar(' '); - putchar('\b'); - --lineptr; - } - break; - default: - if (lineptr >= &line[sizeof(line) - 1]) - putchar('\a'); - else { - putchar(c); - *lineptr++ = c; - } - } - } - lineleft = lineptr - line; - lineptr = line; - } - if (lineleft == 0) - return -1; - --lineleft; - return *lineptr++; -} - - - -/* String functions lifted from lib/vsprintf.c and lib/ctype.c */ -unsigned char _ctype[] = { -_C,_C,_C,_C,_C,_C,_C,_C, /* 0-7 */ -_C,_C|_S,_C|_S,_C|_S,_C|_S,_C|_S,_C,_C, /* 8-15 */ -_C,_C,_C,_C,_C,_C,_C,_C, /* 16-23 */ -_C,_C,_C,_C,_C,_C,_C,_C, /* 24-31 */ -_S|_SP,_P,_P,_P,_P,_P,_P,_P, /* 32-39 */ -_P,_P,_P,_P,_P,_P,_P,_P, /* 40-47 */ -_D,_D,_D,_D,_D,_D,_D,_D, /* 48-55 */ -_D,_D,_P,_P,_P,_P,_P,_P, /* 56-63 */ -_P,_U|_X,_U|_X,_U|_X,_U|_X,_U|_X,_U|_X,_U, /* 64-71 */ -_U,_U,_U,_U,_U,_U,_U,_U, /* 72-79 */ -_U,_U,_U,_U,_U,_U,_U,_U, /* 80-87 */ -_U,_U,_U,_P,_P,_P,_P,_P, /* 88-95 */ -_P,_L|_X,_L|_X,_L|_X,_L|_X,_L|_X,_L|_X,_L, /* 96-103 */ -_L,_L,_L,_L,_L,_L,_L,_L, /* 104-111 */ -_L,_L,_L,_L,_L,_L,_L,_L, /* 112-119 */ -_L,_L,_L,_P,_P,_P,_P,_C, /* 120-127 */ -0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 128-143 */ -0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 144-159 */ -_S|_SP,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P, /* 160-175 */ -_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P, /* 176-191 */ -_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U, /* 192-207 */ -_U,_U,_U,_U,_U,_U,_U,_P,_U,_U,_U,_U,_U,_U,_U,_L, /* 208-223 */ -_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L, /* 224-239 */ -_L,_L,_L,_L,_L,_L,_L,_P,_L,_L,_L,_L,_L,_L,_L,_L}; /* 240-255 */ - -size_t strnlen(const char * s, size_t count) -{ - const char *sc; - - for (sc = s; count-- && *sc != '\0'; ++sc) - /* nothing */; - return sc - s; -} - -unsigned long simple_strtoul(const char *cp,char **endp,unsigned int base) -{ - unsigned long result = 0,value; - - if (!base) { - base = 10; - if (*cp == '0') { - base = 8; - cp++; - if ((*cp == 'x') && isxdigit(cp[1])) { - cp++; - base = 16; - } - } - } - while (isxdigit(*cp) && - (value = isdigit(*cp) ? *cp-'0' : toupper(*cp)-'A'+10) < base) { - result = result*base + value; - cp++; - } - if (endp) - *endp = (char *)cp; - return result; -} - -long simple_strtol(const char *cp,char **endp,unsigned int base) -{ - if(*cp=='-') - return -simple_strtoul(cp+1,endp,base); - return simple_strtoul(cp,endp,base); -} - -static int skip_atoi(const char **s) -{ - int i=0; - - while (isdigit(**s)) - i = i*10 + *((*s)++) - '0'; - return i; -} - -#define ZEROPAD 1 /* pad with zero */ -#define SIGN 2 /* unsigned/signed long */ -#define PLUS 4 /* show plus */ -#define SPACE 8 /* space if plus */ -#define LEFT 16 /* left justified */ -#define SPECIAL 32 /* 0x */ -#define LARGE 64 /* use 'ABCDEF' instead of 'abcdef' */ - -static char * number(char * str, long long num, int base, int size, int precision, int type) -{ - char c,sign,tmp[66]; - const char *digits="0123456789abcdefghijklmnopqrstuvwxyz"; - int i; - - if (type & LARGE) - digits = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"; - if (type & LEFT) - type &= ~ZEROPAD; - if (base < 2 || base > 36) - return 0; - c = (type & ZEROPAD) ? '0' : ' '; - sign = 0; - if (type & SIGN) { - if (num < 0) { - sign = '-'; - num = -num; - size--; - } else if (type & PLUS) { - sign = '+'; - size--; - } else if (type & SPACE) { - sign = ' '; - size--; - } - } - if (type & SPECIAL) { - if (base == 16) - size -= 2; - else if (base == 8) - size--; - } - i = 0; - if (num == 0) - tmp[i++]='0'; - else while (num != 0) - tmp[i++] = digits[do_div(num,base)]; - if (i > precision) - precision = i; - size -= precision; - if (!(type&(ZEROPAD+LEFT))) - while(size-->0) - *str++ = ' '; - if (sign) - *str++ = sign; - if (type & SPECIAL) { - if (base==8) - *str++ = '0'; - else if (base==16) { - *str++ = '0'; - *str++ = digits[33]; - } - } - if (!(type & LEFT)) - while (size-- > 0) - *str++ = c; - while (i < precision--) - *str++ = '0'; - while (i-- > 0) - *str++ = tmp[i]; - while (size-- > 0) - *str++ = ' '; - return str; -} - -/* Forward decl. needed for IP address printing stuff... */ -int sprintf(char * buf, const char *fmt, ...); - -int vsprintf(char *buf, const char *fmt, va_list args) -{ - int len; - unsigned long long num; - int i, base; - char * str; - const char *s; - - int flags; /* flags to number() */ - - int field_width; /* width of output field */ - int precision; /* min. # of digits for integers; max - number of chars for from string */ - int qualifier; /* 'h', 'l', or 'L' for integer fields */ - /* 'z' support added 23/7/1999 S.H. */ - /* 'z' changed to 'Z' --davidm 1/25/99 */ - - - for (str=buf ; *fmt ; ++fmt) { - if (*fmt != '%') { - *str++ = *fmt; - continue; - } - - /* process flags */ - flags = 0; - repeat: - ++fmt; /* this also skips first '%' */ - switch (*fmt) { - case '-': flags |= LEFT; goto repeat; - case '+': flags |= PLUS; goto repeat; - case ' ': flags |= SPACE; goto repeat; - case '#': flags |= SPECIAL; goto repeat; - case '0': flags |= ZEROPAD; goto repeat; - } - - /* get field width */ - field_width = -1; - if (isdigit(*fmt)) - field_width = skip_atoi(&fmt); - else if (*fmt == '*') { - ++fmt; - /* it's the next argument */ - field_width = va_arg(args, int); - if (field_width < 0) { - field_width = -field_width; - flags |= LEFT; - } - } - - /* get the precision */ - precision = -1; - if (*fmt == '.') { - ++fmt; - if (isdigit(*fmt)) - precision = skip_atoi(&fmt); - else if (*fmt == '*') { - ++fmt; - /* it's the next argument */ - precision = va_arg(args, int); - } - if (precision < 0) - precision = 0; - } - - /* get the conversion qualifier */ - qualifier = -1; - if (*fmt == 'h' || *fmt == 'l' || *fmt == 'L' || *fmt =='Z') { - qualifier = *fmt; - ++fmt; - } - - /* default base */ - base = 10; - - switch (*fmt) { - case 'c': - if (!(flags & LEFT)) - while (--field_width > 0) - *str++ = ' '; - *str++ = (unsigned char) va_arg(args, int); - while (--field_width > 0) - *str++ = ' '; - continue; - - case 's': - s = va_arg(args, char *); - if (!s) - s = "<NULL>"; - - len = strnlen(s, precision); - - if (!(flags & LEFT)) - while (len < field_width--) - *str++ = ' '; - for (i = 0; i < len; ++i) - *str++ = *s++; - while (len < field_width--) - *str++ = ' '; - continue; - - case 'p': - if (field_width == -1) { - field_width = 2*sizeof(void *); - flags |= ZEROPAD; - } - str = number(str, - (unsigned long) va_arg(args, void *), 16, - field_width, precision, flags); - continue; - - - case 'n': - if (qualifier == 'l') { - long * ip = va_arg(args, long *); - *ip = (str - buf); - } else if (qualifier == 'Z') { - size_t * ip = va_arg(args, size_t *); - *ip = (str - buf); - } else { - int * ip = va_arg(args, int *); - *ip = (str - buf); - } - continue; - - case '%': - *str++ = '%'; - continue; - - /* integer number formats - set up the flags and "break" */ - case 'o': - base = 8; - break; - - case 'X': - flags |= LARGE; - case 'x': - base = 16; - break; - - case 'd': - case 'i': - flags |= SIGN; - case 'u': - break; - - default: - *str++ = '%'; - if (*fmt) - *str++ = *fmt; - else - --fmt; - continue; - } - if (qualifier == 'L') - num = va_arg(args, long long); - else if (qualifier == 'l') { - num = va_arg(args, unsigned long); - if (flags & SIGN) - num = (signed long) num; - } else if (qualifier == 'Z') { - num = va_arg(args, size_t); - } else if (qualifier == 'h') { - num = (unsigned short) va_arg(args, int); - if (flags & SIGN) - num = (signed short) num; - } else { - num = va_arg(args, unsigned int); - if (flags & SIGN) - num = (signed int) num; - } - str = number(str, num, base, field_width, precision, flags); - } - *str = '\0'; - return str-buf; -} - -int sprintf(char * buf, const char *fmt, ...) -{ - va_list args; - int i; - - va_start(args, fmt); - i=vsprintf(buf,fmt,args); - va_end(args); - return i; -} - -static char sprint_buf[1024]; - -void -printk(char *fmt, ...) -{ - va_list args; - int n; - - va_start(args, fmt); - n = vsprintf(sprint_buf, fmt, args); - va_end(args); - write(stdout, sprint_buf, n); -} - -int -printf(char *fmt, ...) -{ - va_list args; - int n; - - va_start(args, fmt); - n = vsprintf(sprint_buf, fmt, args); - va_end(args); - write(stdout, sprint_buf, n); - return n; -} diff --git a/arch/ppc64/kernel/HvLpEvent.c b/arch/ppc64/kernel/HvLpEvent.c index 9802beefa21..f8f19637f73 100644 --- a/arch/ppc64/kernel/HvLpEvent.c +++ b/arch/ppc64/kernel/HvLpEvent.c @@ -45,7 +45,7 @@ int HvLpEvent_unregisterHandler( HvLpEvent_Type eventType ) /* We now sleep until all other CPUs have scheduled. This ensures that * the deletion is seen by all other CPUs, and that the deleted handler * isn't still running on another CPU when we return. */ - synchronize_kernel(); + synchronize_rcu(); } } return rc; diff --git a/arch/ppc64/kernel/misc.S b/arch/ppc64/kernel/misc.S index 90b41f48d21..b944717c1db 100644 --- a/arch/ppc64/kernel/misc.S +++ b/arch/ppc64/kernel/misc.S @@ -32,7 +32,7 @@ .text /* - * Returns (address we're running at) - (address we were linked at) + * Returns (address we were linked at) - (address we are running at) * for use before the text and data are mapped to KERNELBASE. */ diff --git a/arch/ppc64/kernel/nvram.c b/arch/ppc64/kernel/nvram.c index b9069c2d193..4e71781a441 100644 --- a/arch/ppc64/kernel/nvram.c +++ b/arch/ppc64/kernel/nvram.c @@ -339,9 +339,9 @@ static int nvram_remove_os_partition(void) static int nvram_create_os_partition(void) { struct list_head * p; - struct nvram_partition * part; - struct nvram_partition * new_part = NULL; - struct nvram_partition * free_part = NULL; + struct nvram_partition *part = NULL; + struct nvram_partition *new_part = NULL; + struct nvram_partition *free_part = NULL; int seq_init[2] = { 0, 0 }; loff_t tmp_index; long size = 0; @@ -364,13 +364,11 @@ static int nvram_create_os_partition(void) free_part = part; } } - if (!size) { + if (!size) return -ENOSPC; - } /* Create our OS partition */ - new_part = (struct nvram_partition *) - kmalloc(sizeof(struct nvram_partition), GFP_KERNEL); + new_part = kmalloc(sizeof(*new_part), GFP_KERNEL); if (!new_part) { printk(KERN_ERR "nvram_create_os_partition: kmalloc failed\n"); return -ENOMEM; @@ -379,7 +377,7 @@ static int nvram_create_os_partition(void) new_part->index = free_part->index; new_part->header.signature = NVRAM_SIG_OS; new_part->header.length = size; - sprintf(new_part->header.name, "ppc64,linux"); + strcpy(new_part->header.name, "ppc64,linux"); new_part->header.checksum = nvram_checksum(&new_part->header); rc = nvram_write_header(new_part); @@ -394,7 +392,8 @@ static int nvram_create_os_partition(void) tmp_index = new_part->index + NVRAM_HEADER_LEN; rc = ppc_md.nvram_write((char *)&seq_init, sizeof(seq_init), &tmp_index); if (rc <= 0) { - printk(KERN_ERR "nvram_create_os_partition: nvram_write failed (%d)\n", rc); + printk(KERN_ERR "nvram_create_os_partition: nvram_write " + "failed (%d)\n", rc); return rc; } diff --git a/arch/ppc64/kernel/pSeries_hvCall.S b/arch/ppc64/kernel/pSeries_hvCall.S index 0715d303801..176e8da7646 100644 --- a/arch/ppc64/kernel/pSeries_hvCall.S +++ b/arch/ppc64/kernel/pSeries_hvCall.S @@ -28,6 +28,8 @@ unsigned long *out3); R10 */ _GLOBAL(plpar_hcall) + HMT_MEDIUM + mfcr r0 std r8,STK_PARM(r8)(r1) /* Save out ptrs */ @@ -53,6 +55,8 @@ _GLOBAL(plpar_hcall) /* Simple interface with no output values (other than status) */ _GLOBAL(plpar_hcall_norets) + HMT_MEDIUM + mfcr r0 stw r0,8(r1) @@ -75,6 +79,8 @@ _GLOBAL(plpar_hcall_norets) unsigned long *out1); 120(R1) */ _GLOBAL(plpar_hcall_8arg_2ret) + HMT_MEDIUM + mfcr r0 ld r11,STK_PARM(r11)(r1) /* put arg8 in R11 */ stw r0,8(r1) @@ -99,6 +105,8 @@ _GLOBAL(plpar_hcall_8arg_2ret) unsigned long *out4); 112(R1) */ _GLOBAL(plpar_hcall_4out) + HMT_MEDIUM + mfcr r0 stw r0,8(r1) diff --git a/arch/ppc64/kernel/pSeries_smp.c b/arch/ppc64/kernel/pSeries_smp.c index c60d8cb2b84..fbad349ec58 100644 --- a/arch/ppc64/kernel/pSeries_smp.c +++ b/arch/ppc64/kernel/pSeries_smp.c @@ -326,13 +326,6 @@ static void __devinit smp_xics_setup_cpu(int cpu) cpu_clear(cpu, of_spin_map); - /* - * Put the calling processor into the GIQ. This is really only - * necessary from a secondary thread as the OF start-cpu interface - * performs this function for us on primary threads. - */ - rtas_set_indicator(GLOBAL_INTERRUPT_QUEUE, - (1UL << interrupt_server_size) - 1 - default_distrib_server, 1); } static DEFINE_SPINLOCK(timebase_lock); diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c index be3cc387c1e..d786d4b6af0 100644 --- a/arch/ppc64/kernel/pci.c +++ b/arch/ppc64/kernel/pci.c @@ -438,7 +438,7 @@ pgprot_t pci_phys_mem_access_prot(struct file *file, int i; if (page_is_ram(offset >> PAGE_SHIFT)) - return prot; + return __pgprot(prot); prot |= _PAGE_NO_CACHE | _PAGE_GUARDED; diff --git a/arch/ppc64/kernel/prom.c b/arch/ppc64/kernel/prom.c index 45a4ad08fbc..eb6538b5800 100644 --- a/arch/ppc64/kernel/prom.c +++ b/arch/ppc64/kernel/prom.c @@ -321,6 +321,10 @@ static int __devinit finish_node_interrupts(struct device_node *np, char *name = get_property(ic->parent, "name", NULL); if (name && !strcmp(name, "u3")) np->intrs[intrcount].line += 128; + else if (!(name && !strcmp(name, "mac-io"))) + /* ignore other cascaded controllers, such as + the k2-sata-root */ + break; } np->intrs[intrcount].sense = 1; if (n > 1) @@ -830,7 +834,7 @@ void __init unflatten_device_tree(void) { unsigned long start, mem, size; struct device_node **allnextp = &allnodes; - char *p; + char *p = NULL; int l = 0; DBG(" -> unflatten_device_tree()\n"); diff --git a/arch/ppc64/kernel/prom_init.c b/arch/ppc64/kernel/prom_init.c index 8dffa9ae262..35ec42de962 100644 --- a/arch/ppc64/kernel/prom_init.c +++ b/arch/ppc64/kernel/prom_init.c @@ -493,6 +493,113 @@ static void __init early_cmdline_parse(void) } /* + * To tell the firmware what our capabilities are, we have to pass + * it a fake 32-bit ELF header containing a couple of PT_NOTE sections + * that contain structures that contain the actual values. + */ +static struct fake_elf { + Elf32_Ehdr elfhdr; + Elf32_Phdr phdr[2]; + struct chrpnote { + u32 namesz; + u32 descsz; + u32 type; + char name[8]; /* "PowerPC" */ + struct chrpdesc { + u32 real_mode; + u32 real_base; + u32 real_size; + u32 virt_base; + u32 virt_size; + u32 load_base; + } chrpdesc; + } chrpnote; + struct rpanote { + u32 namesz; + u32 descsz; + u32 type; + char name[24]; /* "IBM,RPA-Client-Config" */ + struct rpadesc { + u32 lpar_affinity; + u32 min_rmo_size; + u32 min_rmo_percent; + u32 max_pft_size; + u32 splpar; + u32 min_load; + u32 new_mem_def; + u32 ignore_me; + } rpadesc; + } rpanote; +} fake_elf = { + .elfhdr = { + .e_ident = { 0x7f, 'E', 'L', 'F', + ELFCLASS32, ELFDATA2MSB, EV_CURRENT }, + .e_type = ET_EXEC, /* yeah right */ + .e_machine = EM_PPC, + .e_version = EV_CURRENT, + .e_phoff = offsetof(struct fake_elf, phdr), + .e_phentsize = sizeof(Elf32_Phdr), + .e_phnum = 2 + }, + .phdr = { + [0] = { + .p_type = PT_NOTE, + .p_offset = offsetof(struct fake_elf, chrpnote), + .p_filesz = sizeof(struct chrpnote) + }, [1] = { + .p_type = PT_NOTE, + .p_offset = offsetof(struct fake_elf, rpanote), + .p_filesz = sizeof(struct rpanote) + } + }, + .chrpnote = { + .namesz = sizeof("PowerPC"), + .descsz = sizeof(struct chrpdesc), + .type = 0x1275, + .name = "PowerPC", + .chrpdesc = { + .real_mode = ~0U, /* ~0 means "don't care" */ + .real_base = ~0U, + .real_size = ~0U, + .virt_base = ~0U, + .virt_size = ~0U, + .load_base = ~0U + }, + }, + .rpanote = { + .namesz = sizeof("IBM,RPA-Client-Config"), + .descsz = sizeof(struct rpadesc), + .type = 0x12759999, + .name = "IBM,RPA-Client-Config", + .rpadesc = { + .lpar_affinity = 0, + .min_rmo_size = 64, /* in megabytes */ + .min_rmo_percent = 0, + .max_pft_size = 48, /* 2^48 bytes max PFT size */ + .splpar = 1, + .min_load = ~0U, + .new_mem_def = 0 + } + } +}; + +static void __init prom_send_capabilities(void) +{ + unsigned long offset = reloc_offset(); + ihandle elfloader; + int ret; + + elfloader = call_prom("open", 1, 1, ADDR("/packages/elf-loader")); + if (elfloader == 0) { + prom_printf("couldn't open /packages/elf-loader\n"); + return; + } + ret = call_prom("call-method", 3, 1, ADDR("process-elf-header"), + elfloader, ADDR(&fake_elf)); + call_prom("close", 1, 0, elfloader); +} + +/* * Memory allocation strategy... our layout is normally: * * at 14Mb or more we vmlinux, then a gap and initrd. In some rare cases, initrd @@ -1448,6 +1555,12 @@ static void __init scan_dt_build_strings(phandle node, unsigned long *mem_start, } } +/* + * The Open Firmware 1275 specification states properties must be 31 bytes or + * less, however not all firmwares obey this. Make it 64 bytes to be safe. + */ +#define MAX_PROPERTY_NAME 64 + static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start, unsigned long *mem_end) { @@ -1457,7 +1570,7 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start, unsigned long soff; unsigned char *valp; unsigned long offset = reloc_offset(); - char pname[32]; + char pname[MAX_PROPERTY_NAME]; char *path; path = RELOC(prom_scratch); diff --git a/arch/ppc64/kernel/ptrace.c b/arch/ppc64/kernel/ptrace.c index 354a287c67e..9f8c6087ae5 100644 --- a/arch/ppc64/kernel/ptrace.c +++ b/arch/ppc64/kernel/ptrace.c @@ -28,6 +28,7 @@ #include <linux/security.h> #include <linux/audit.h> #include <linux/seccomp.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -162,7 +163,7 @@ int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -194,7 +195,7 @@ int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); set_single_step(child); @@ -304,14 +305,17 @@ static void do_syscall_trace(void) void do_syscall_trace_enter(struct pt_regs *regs) { + if (test_thread_flag(TIF_SYSCALL_TRACE) + && (current->ptrace & PT_PTRACED)) + do_syscall_trace(); + if (unlikely(current->audit_context)) - audit_syscall_entry(current, regs->gpr[0], + audit_syscall_entry(current, + test_thread_flag(TIF_32BIT)?AUDIT_ARCH_PPC:AUDIT_ARCH_PPC64, + regs->gpr[0], regs->gpr[3], regs->gpr[4], regs->gpr[5], regs->gpr[6]); - if (test_thread_flag(TIF_SYSCALL_TRACE) - && (current->ptrace & PT_PTRACED)) - do_syscall_trace(); } void do_syscall_trace_leave(struct pt_regs *regs) @@ -319,7 +323,9 @@ void do_syscall_trace_leave(struct pt_regs *regs) secure_computing(regs->gpr[0]); if (unlikely(current->audit_context)) - audit_syscall_exit(current, regs->result); + audit_syscall_exit(current, + (regs->ccr&0x1000)?AUDITSC_FAILURE:AUDITSC_SUCCESS, + regs->result); if ((test_thread_flag(TIF_SYSCALL_TRACE) || test_thread_flag(TIF_SINGLESTEP)) diff --git a/arch/ppc64/kernel/ptrace32.c b/arch/ppc64/kernel/ptrace32.c index ee81b1b776c..16436426c7e 100644 --- a/arch/ppc64/kernel/ptrace32.c +++ b/arch/ppc64/kernel/ptrace32.c @@ -26,6 +26,7 @@ #include <linux/ptrace.h> #include <linux/user.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -293,7 +294,7 @@ int sys32_ptrace(long request, long pid, unsigned long addr, unsigned long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -325,7 +326,7 @@ int sys32_ptrace(long request, long pid, unsigned long addr, unsigned long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); set_single_step(child); diff --git a/arch/ppc64/kernel/rtas_flash.c b/arch/ppc64/kernel/rtas_flash.c index 3213837282c..923e2e201a7 100644 --- a/arch/ppc64/kernel/rtas_flash.c +++ b/arch/ppc64/kernel/rtas_flash.c @@ -218,7 +218,7 @@ static void get_flash_status_msg(int status, char *buf) } /* Reading the proc file will show status (not the firmware contents) */ -static ssize_t rtas_flash_read(struct file *file, char *buf, +static ssize_t rtas_flash_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode); @@ -256,7 +256,7 @@ static ssize_t rtas_flash_read(struct file *file, char *buf, * count is. If the system is low on memory it will be just as well * that we fail.... */ -static ssize_t rtas_flash_write(struct file *file, const char *buffer, +static ssize_t rtas_flash_write(struct file *file, const char __user *buffer, size_t count, loff_t *off) { struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode); @@ -356,7 +356,7 @@ static void manage_flash(struct rtas_manage_flash_t *args_buf) args_buf->status = rc; } -static ssize_t manage_flash_read(struct file *file, char *buf, +static ssize_t manage_flash_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode); @@ -386,7 +386,7 @@ static ssize_t manage_flash_read(struct file *file, char *buf, return msglen; } -static ssize_t manage_flash_write(struct file *file, const char *buf, +static ssize_t manage_flash_write(struct file *file, const char __user *buf, size_t count, loff_t *off) { struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode); @@ -466,7 +466,7 @@ static int get_validate_flash_msg(struct rtas_validate_flash_t *args_buf, return n; } -static ssize_t validate_flash_read(struct file *file, char *buf, +static ssize_t validate_flash_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode); @@ -494,7 +494,7 @@ static ssize_t validate_flash_read(struct file *file, char *buf, return msglen; } -static ssize_t validate_flash_write(struct file *file, const char *buf, +static ssize_t validate_flash_write(struct file *file, const char __user *buf, size_t count, loff_t *off) { struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode); diff --git a/arch/ppc64/kernel/scanlog.c b/arch/ppc64/kernel/scanlog.c index 189b81a4198..4d70736619c 100644 --- a/arch/ppc64/kernel/scanlog.c +++ b/arch/ppc64/kernel/scanlog.c @@ -43,7 +43,7 @@ static int scanlog_debug; static unsigned int ibm_scan_log_dump; /* RTAS token */ static struct proc_dir_entry *proc_ppc64_scan_log_dump; /* The proc file */ -static ssize_t scanlog_read(struct file *file, char *buf, +static ssize_t scanlog_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { struct inode * inode = file->f_dentry->d_inode; @@ -129,7 +129,7 @@ static ssize_t scanlog_read(struct file *file, char *buf, /*NOTREACHED*/ } -static ssize_t scanlog_write(struct file * file, const char * buf, +static ssize_t scanlog_write(struct file * file, const char __user * buf, size_t count, loff_t *ppos) { char stkbuf[20]; diff --git a/arch/ppc64/kernel/signal.c b/arch/ppc64/kernel/signal.c index a95a2b49a1d..bf782276984 100644 --- a/arch/ppc64/kernel/signal.c +++ b/arch/ppc64/kernel/signal.c @@ -42,11 +42,7 @@ #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) -#ifndef MIN -#define MIN(a,b) (((a) < (b)) ? (a) : (b)) -#endif - -#define GP_REGS_SIZE MIN(sizeof(elf_gregset_t), sizeof(struct pt_regs)) +#define GP_REGS_SIZE min(sizeof(elf_gregset_t), sizeof(struct pt_regs)) #define FP_REGS_SIZE sizeof(elf_fpregset_t) #define TRAMP_TRACEBACK 3 diff --git a/arch/ppc64/kernel/signal32.c b/arch/ppc64/kernel/signal32.c index b0e167db6af..3c2fa5c284c 100644 --- a/arch/ppc64/kernel/signal32.c +++ b/arch/ppc64/kernel/signal32.c @@ -657,7 +657,7 @@ static int handle_rt_signal32(unsigned long sig, struct k_sigaction *ka, /* Save user registers on the stack */ frame = &rt_sf->uc.uc_mcontext; - if (put_user(regs->gpr[1], (unsigned long __user *)newsp)) + if (put_user(regs->gpr[1], (u32 __user *)newsp)) goto badframe; if (vdso32_rt_sigtramp && current->thread.vdso_base) { @@ -842,7 +842,7 @@ static int handle_signal32(unsigned long sig, struct k_sigaction *ka, regs->link = (unsigned long) frame->mctx.tramp; } - if (put_user(regs->gpr[1], (unsigned long __user *)newsp)) + if (put_user(regs->gpr[1], (u32 __user *)newsp)) goto badframe; regs->gpr[1] = (unsigned long) newsp; regs->gpr[3] = sig; diff --git a/arch/ppc64/kernel/smp.c b/arch/ppc64/kernel/smp.c index 1c92da3e452..3b906cd9403 100644 --- a/arch/ppc64/kernel/smp.c +++ b/arch/ppc64/kernel/smp.c @@ -125,7 +125,7 @@ void __devinit smp_generic_kick_cpu(int nr) * the processor will continue on to secondary_start */ paca[nr].cpu_start = 1; - mb(); + smp_mb(); } #endif /* CONFIG_PPC_MULTIPLATFORM */ @@ -256,7 +256,7 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic, } call_data = &data; - wmb(); + smp_wmb(); /* Send a message to all other CPUs and wait for them to respond */ smp_ops->message_pass(MSG_ALL_BUT_SELF, PPC_MSG_CALL_FUNCTION); @@ -431,7 +431,7 @@ int generic_cpu_enable(unsigned int cpu) /* get the target out of it's holding state */ per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; - wmb(); + smp_wmb(); while (!cpu_online(cpu)) cpu_relax(); @@ -447,7 +447,7 @@ void generic_cpu_die(unsigned int cpu) int i; for (i = 0; i < 100; i++) { - rmb(); + smp_rmb(); if (per_cpu(cpu_state, cpu) == CPU_DEAD) return; msleep(100); @@ -463,7 +463,7 @@ void generic_mach_cpu_die(void) cpu = smp_processor_id(); printk(KERN_DEBUG "CPU%d offline\n", cpu); __get_cpu_var(cpu_state) = CPU_DEAD; - wmb(); + smp_wmb(); while (__get_cpu_var(cpu_state) != CPU_UP_PREPARE) cpu_relax(); @@ -515,7 +515,7 @@ int __devinit __cpu_up(unsigned int cpu) * be written out to main store before we release * the processor. */ - mb(); + smp_mb(); /* wake up cpus */ DBG("smp: kicking cpu %d\n", cpu); diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c index 77ded5a363b..772a465b49f 100644 --- a/arch/ppc64/kernel/time.c +++ b/arch/ppc64/kernel/time.c @@ -221,15 +221,15 @@ static __inline__ void timer_recalc_offset(unsigned long cur_tb) temp_varp->tb_to_xs = do_gtod.varp->tb_to_xs; temp_varp->tb_orig_stamp = new_tb_orig_stamp; temp_varp->stamp_xsec = new_stamp_xsec; - mb(); + smp_mb(); do_gtod.varp = temp_varp; do_gtod.var_idx = temp_idx; ++(systemcfg->tb_update_count); - wmb(); + smp_wmb(); systemcfg->tb_orig_stamp = new_tb_orig_stamp; systemcfg->stamp_xsec = new_stamp_xsec; - wmb(); + smp_wmb(); ++(systemcfg->tb_update_count); } @@ -648,7 +648,7 @@ void ppc_adjtimex(void) temp_varp->tb_to_xs = new_tb_to_xs; temp_varp->stamp_xsec = new_stamp_xsec; temp_varp->tb_orig_stamp = do_gtod.varp->tb_orig_stamp; - mb(); + smp_mb(); do_gtod.varp = temp_varp; do_gtod.var_idx = temp_idx; @@ -662,10 +662,10 @@ void ppc_adjtimex(void) * loops back and reads them again until this criteria is met. */ ++(systemcfg->tb_update_count); - wmb(); + smp_wmb(); systemcfg->tb_to_xs = new_tb_to_xs; systemcfg->stamp_xsec = new_stamp_xsec; - wmb(); + smp_wmb(); ++(systemcfg->tb_update_count); write_sequnlock_irqrestore( &xtime_lock, flags ); diff --git a/arch/ppc64/kernel/vdso32/Makefile b/arch/ppc64/kernel/vdso32/Makefile index ede2f7e477c..0b1b0df973e 100644 --- a/arch/ppc64/kernel/vdso32/Makefile +++ b/arch/ppc64/kernel/vdso32/Makefile @@ -1,7 +1,7 @@ # List of files in the vdso, has to be asm only for now -obj-vdso32 = sigtramp.o gettimeofday.o datapage.o cacheflush.o +obj-vdso32 = sigtramp.o gettimeofday.o datapage.o cacheflush.o note.o # Build rules diff --git a/arch/ppc64/kernel/vdso32/cacheflush.S b/arch/ppc64/kernel/vdso32/cacheflush.S index c74fddb6afd..0ed7ea72171 100644 --- a/arch/ppc64/kernel/vdso32/cacheflush.S +++ b/arch/ppc64/kernel/vdso32/cacheflush.S @@ -47,6 +47,7 @@ V_FUNCTION_BEGIN(__kernel_sync_dicache) addi r6,r6,128 bdnz 1b isync + li r3,0 blr .cfi_endproc V_FUNCTION_END(__kernel_sync_dicache) @@ -59,6 +60,7 @@ V_FUNCTION_BEGIN(__kernel_sync_dicache_p5) .cfi_startproc sync isync + li r3,0 blr .cfi_endproc V_FUNCTION_END(__kernel_sync_dicache_p5) diff --git a/arch/ppc64/kernel/vdso32/gettimeofday.S b/arch/ppc64/kernel/vdso32/gettimeofday.S index ca7f415195c..2b48bf1fb10 100644 --- a/arch/ppc64/kernel/vdso32/gettimeofday.S +++ b/arch/ppc64/kernel/vdso32/gettimeofday.S @@ -58,6 +58,7 @@ V_FUNCTION_BEGIN(__kernel_gettimeofday) stw r5,TZONE_TZ_DSTTIME(r11) 1: mtlr r12 + li r3,0 blr 2: mr r3,r10 diff --git a/arch/ppc64/kernel/vdso32/note.S b/arch/ppc64/kernel/vdso32/note.S new file mode 100644 index 00000000000..d4b5be4f3d5 --- /dev/null +++ b/arch/ppc64/kernel/vdso32/note.S @@ -0,0 +1,25 @@ +/* + * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text. + * Here we can supply some information useful to userland. + */ + +#include <linux/uts.h> +#include <linux/version.h> + +#define ASM_ELF_NOTE_BEGIN(name, flags, vendor, type) \ + .section name, flags; \ + .balign 4; \ + .long 1f - 0f; /* name length */ \ + .long 3f - 2f; /* data length */ \ + .long type; /* note type */ \ +0: .asciz vendor; /* vendor name */ \ +1: .balign 4; \ +2: + +#define ASM_ELF_NOTE_END \ +3: .balign 4; /* pad out section */ \ + .previous + + ASM_ELF_NOTE_BEGIN(".note.kernel-version", "a", UTS_SYSNAME, 0) + .long LINUX_VERSION_CODE + ASM_ELF_NOTE_END diff --git a/arch/ppc64/kernel/vdso32/vdso32.lds.S b/arch/ppc64/kernel/vdso32/vdso32.lds.S index cca27bd03a5..11290c902ba 100644 --- a/arch/ppc64/kernel/vdso32/vdso32.lds.S +++ b/arch/ppc64/kernel/vdso32/vdso32.lds.S @@ -20,6 +20,8 @@ SECTIONS .gnu.version_d : { *(.gnu.version_d) } .gnu.version_r : { *(.gnu.version_r) } + .note : { *(.note.*) } :text :note + . = ALIGN (16); .text : { @@ -87,6 +89,7 @@ SECTIONS PHDRS { text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + note PT_NOTE FLAGS(4); /* PF_R */ dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */ } diff --git a/arch/ppc64/kernel/vdso64/Makefile b/arch/ppc64/kernel/vdso64/Makefile index bd3f70b1a38..ab39988452c 100644 --- a/arch/ppc64/kernel/vdso64/Makefile +++ b/arch/ppc64/kernel/vdso64/Makefile @@ -1,6 +1,6 @@ # List of files in the vdso, has to be asm only for now -obj-vdso64 = sigtramp.o gettimeofday.o datapage.o cacheflush.o +obj-vdso64 = sigtramp.o gettimeofday.o datapage.o cacheflush.o note.o # Build rules diff --git a/arch/ppc64/kernel/vdso64/cacheflush.S b/arch/ppc64/kernel/vdso64/cacheflush.S index d9696ffcf33..e0725b7b700 100644 --- a/arch/ppc64/kernel/vdso64/cacheflush.S +++ b/arch/ppc64/kernel/vdso64/cacheflush.S @@ -47,6 +47,7 @@ V_FUNCTION_BEGIN(__kernel_sync_dicache) addi r6,r6,128 bdnz 1b isync + li r3,0 blr .cfi_endproc V_FUNCTION_END(__kernel_sync_dicache) @@ -59,6 +60,7 @@ V_FUNCTION_BEGIN(__kernel_sync_dicache_p5) .cfi_startproc sync isync + li r3,0 blr .cfi_endproc V_FUNCTION_END(__kernel_sync_dicache_p5) diff --git a/arch/ppc64/kernel/vdso64/note.S b/arch/ppc64/kernel/vdso64/note.S new file mode 100644 index 00000000000..dc2a509f7e8 --- /dev/null +++ b/arch/ppc64/kernel/vdso64/note.S @@ -0,0 +1 @@ +#include "../vdso32/note.S" diff --git a/arch/ppc64/kernel/vdso64/vdso64.lds.S b/arch/ppc64/kernel/vdso64/vdso64.lds.S index 942c815c7bc..9cb28181da8 100644 --- a/arch/ppc64/kernel/vdso64/vdso64.lds.S +++ b/arch/ppc64/kernel/vdso64/vdso64.lds.S @@ -18,12 +18,14 @@ SECTIONS .gnu.version_d : { *(.gnu.version_d) } .gnu.version_r : { *(.gnu.version_r) } + .note : { *(.note.*) } :text :note + . = ALIGN (16); .text : { *(.text .stub .text.* .gnu.linkonce.t.*) *(.sfpr .glink) - } + } :text PROVIDE (__etext = .); PROVIDE (_etext = .); PROVIDE (etext = .); @@ -88,6 +90,7 @@ SECTIONS PHDRS { text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + note PT_NOTE FLAGS(4); /* PF_R */ dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */ } diff --git a/arch/ppc64/kernel/xics.c b/arch/ppc64/kernel/xics.c index eedd1d3c2a1..879f39b90a3 100644 --- a/arch/ppc64/kernel/xics.c +++ b/arch/ppc64/kernel/xics.c @@ -432,6 +432,7 @@ void xics_cause_IPI(int cpu) { ops->qirr_info(cpu, IPI_PRIORITY); } +#endif /* CONFIG_SMP */ void xics_setup_cpu(void) { @@ -439,9 +440,17 @@ void xics_setup_cpu(void) ops->cppr_info(cpu, 0xff); iosync(); -} -#endif /* CONFIG_SMP */ + /* + * Put the calling processor into the GIQ. This is really only + * necessary from a secondary thread as the OF start-cpu interface + * performs this function for us on primary threads. + * + * XXX: undo of teardown on kexec needs this too, as may hotplug + */ + rtas_set_indicator(GLOBAL_INTERRUPT_QUEUE, + (1UL << interrupt_server_size) - 1 - default_distrib_server, 1); +} void xics_init_IRQ(void) { @@ -563,8 +572,7 @@ nextnode: for (; i < NR_IRQS; ++i) get_irq_desc(i)->handler = &xics_pic; - ops->cppr_info(boot_cpuid, 0xff); - iosync(); + xics_setup_cpu(); ppc64_boot_msg(0x21, "XICS Done"); } diff --git a/arch/ppc64/mm/hash_low.S b/arch/ppc64/mm/hash_low.S index 8c0156a3700..c23d46956dd 100644 --- a/arch/ppc64/mm/hash_low.S +++ b/arch/ppc64/mm/hash_low.S @@ -85,7 +85,10 @@ _GLOBAL(__hash_page) bne- htab_wrong_access /* Check if PTE is busy */ andi. r0,r31,_PAGE_BUSY - bne- 1b + /* If so, just bail out and refault if needed. Someone else + * is changing this PTE anyway and might hash it. + */ + bne- bail_ok /* Prepare new PTE value (turn access RW into DIRTY, then * add BUSY,HASHPTE and ACCESSED) */ @@ -215,6 +218,10 @@ _GLOBAL(htab_call_hpte_remove) /* Try all again */ b htab_insert_pte +bail_ok: + li r3,0 + b bail + htab_pte_insert_ok: /* Insert slot number & secondary bit in PTE */ rldimi r30,r3,12,63-15 diff --git a/arch/ppc64/mm/hash_native.c b/arch/ppc64/mm/hash_native.c index 144657e0c3d..52b6b930534 100644 --- a/arch/ppc64/mm/hash_native.c +++ b/arch/ppc64/mm/hash_native.c @@ -320,8 +320,7 @@ static void native_flush_hash_range(unsigned long context, j = 0; for (i = 0; i < number; i++) { - if ((batch->addr[i] >= USER_START) && - (batch->addr[i] <= USER_END)) + if (batch->addr[i] < KERNELBASE) vsid = get_vsid(context, batch->addr[i]); else vsid = get_kernel_vsid(batch->addr[i]); diff --git a/arch/ppc64/mm/hash_utils.c b/arch/ppc64/mm/hash_utils.c index e48be12f518..0a0f97008d0 100644 --- a/arch/ppc64/mm/hash_utils.c +++ b/arch/ppc64/mm/hash_utils.c @@ -298,24 +298,23 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) int local = 0; cpumask_t tmp; + if ((ea & ~REGION_MASK) > EADDR_MASK) + return 1; + switch (REGION_ID(ea)) { case USER_REGION_ID: user_region = 1; mm = current->mm; - if ((ea > USER_END) || (! mm)) + if (! mm) return 1; vsid = get_vsid(mm->context.id, ea); break; case IO_REGION_ID: - if (ea > IMALLOC_END) - return 1; mm = &ioremap_mm; vsid = get_kernel_vsid(ea); break; case VMALLOC_REGION_ID: - if (ea > VMALLOC_END) - return 1; mm = &init_mm; vsid = get_kernel_vsid(ea); break; @@ -362,7 +361,7 @@ void flush_hash_page(unsigned long context, unsigned long ea, pte_t pte, unsigned long vsid, vpn, va, hash, secondary, slot; unsigned long huge = pte_huge(pte); - if ((ea >= USER_START) && (ea <= USER_END)) + if (ea < KERNELBASE) vsid = get_vsid(context, ea); else vsid = get_kernel_vsid(ea); diff --git a/arch/ppc64/mm/hugetlbpage.c b/arch/ppc64/mm/hugetlbpage.c index 390296efe3e..d3bf86a5c1a 100644 --- a/arch/ppc64/mm/hugetlbpage.c +++ b/arch/ppc64/mm/hugetlbpage.c @@ -42,7 +42,7 @@ static inline int hugepgd_index(unsigned long addr) return (addr & ~REGION_MASK) >> HUGEPGDIR_SHIFT; } -static pgd_t *hugepgd_offset(struct mm_struct *mm, unsigned long addr) +static pud_t *hugepgd_offset(struct mm_struct *mm, unsigned long addr) { int index; @@ -52,21 +52,21 @@ static pgd_t *hugepgd_offset(struct mm_struct *mm, unsigned long addr) index = hugepgd_index(addr); BUG_ON(index >= PTRS_PER_HUGEPGD); - return mm->context.huge_pgdir + index; + return (pud_t *)(mm->context.huge_pgdir + index); } -static inline pte_t *hugepte_offset(pgd_t *dir, unsigned long addr) +static inline pte_t *hugepte_offset(pud_t *dir, unsigned long addr) { int index; - if (pgd_none(*dir)) + if (pud_none(*dir)) return NULL; index = (addr >> HPAGE_SHIFT) % PTRS_PER_HUGEPTE; - return (pte_t *)pgd_page(*dir) + index; + return (pte_t *)pud_page(*dir) + index; } -static pgd_t *hugepgd_alloc(struct mm_struct *mm, unsigned long addr) +static pud_t *hugepgd_alloc(struct mm_struct *mm, unsigned long addr) { BUG_ON(! in_hugepage_area(mm->context, addr)); @@ -90,10 +90,9 @@ static pgd_t *hugepgd_alloc(struct mm_struct *mm, unsigned long addr) return hugepgd_offset(mm, addr); } -static pte_t *hugepte_alloc(struct mm_struct *mm, pgd_t *dir, - unsigned long addr) +static pte_t *hugepte_alloc(struct mm_struct *mm, pud_t *dir, unsigned long addr) { - if (! pgd_present(*dir)) { + if (! pud_present(*dir)) { pte_t *new; spin_unlock(&mm->page_table_lock); @@ -104,7 +103,7 @@ static pte_t *hugepte_alloc(struct mm_struct *mm, pgd_t *dir, * Because we dropped the lock, we should re-check the * entry, as somebody else could have populated it.. */ - if (pgd_present(*dir)) { + if (pud_present(*dir)) { if (new) kmem_cache_free(zero_cache, new); } else { @@ -115,7 +114,7 @@ static pte_t *hugepte_alloc(struct mm_struct *mm, pgd_t *dir, ptepage = virt_to_page(new); ptepage->mapping = (void *) mm; ptepage->index = addr & HUGEPGDIR_MASK; - pgd_populate(mm, dir, new); + pud_populate(mm, dir, new); } } @@ -124,28 +123,28 @@ static pte_t *hugepte_alloc(struct mm_struct *mm, pgd_t *dir, static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { - pgd_t *pgd; + pud_t *pud; BUG_ON(! in_hugepage_area(mm->context, addr)); - pgd = hugepgd_offset(mm, addr); - if (! pgd) + pud = hugepgd_offset(mm, addr); + if (! pud) return NULL; - return hugepte_offset(pgd, addr); + return hugepte_offset(pud, addr); } static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { - pgd_t *pgd; + pud_t *pud; BUG_ON(! in_hugepage_area(mm->context, addr)); - pgd = hugepgd_alloc(mm, addr); - if (! pgd) + pud = hugepgd_alloc(mm, addr); + if (! pud) return NULL; - return hugepte_alloc(mm, pgd, addr); + return hugepte_alloc(mm, pud, addr); } static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, @@ -709,10 +708,10 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm) /* cleanup any hugepte pages leftover */ for (i = 0; i < PTRS_PER_HUGEPGD; i++) { - pgd_t *pgd = pgdir + i; + pud_t *pud = (pud_t *)(pgdir + i); - if (! pgd_none(*pgd)) { - pte_t *pte = (pte_t *)pgd_page(*pgd); + if (! pud_none(*pud)) { + pte_t *pte = (pte_t *)pud_page(*pud); struct page *ptepage = virt_to_page(pte); ptepage->mapping = NULL; @@ -720,7 +719,7 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm) BUG_ON(memcmp(pte, empty_zero_page, PAGE_SIZE)); kmem_cache_free(zero_cache, pte); } - pgd_clear(pgd); + pud_clear(pud); } BUG_ON(memcmp(pgdir, empty_zero_page, PAGE_SIZE)); diff --git a/arch/ppc64/mm/imalloc.c b/arch/ppc64/mm/imalloc.c index 9d92b0d9cde..cb8727f3267 100644 --- a/arch/ppc64/mm/imalloc.c +++ b/arch/ppc64/mm/imalloc.c @@ -14,6 +14,7 @@ #include <asm/pgalloc.h> #include <asm/pgtable.h> #include <asm/semaphore.h> +#include <asm/imalloc.h> static DECLARE_MUTEX(imlist_sem); struct vm_struct * imlist = NULL; @@ -23,11 +24,11 @@ static int get_free_im_addr(unsigned long size, unsigned long *im_addr) unsigned long addr; struct vm_struct **p, *tmp; - addr = IMALLOC_START; + addr = ioremap_bot; for (p = &imlist; (tmp = *p) ; p = &tmp->next) { if (size + addr < (unsigned long) tmp->addr) break; - if ((unsigned long)tmp->addr >= IMALLOC_START) + if ((unsigned long)tmp->addr >= ioremap_bot) addr = tmp->size + (unsigned long) tmp->addr; if (addr > IMALLOC_END-size) return 1; diff --git a/arch/ppc64/mm/init.c b/arch/ppc64/mm/init.c index a7149b9fc35..4b42aff74d7 100644 --- a/arch/ppc64/mm/init.c +++ b/arch/ppc64/mm/init.c @@ -64,6 +64,7 @@ #include <asm/iommu.h> #include <asm/abs_addr.h> #include <asm/vdso.h> +#include <asm/imalloc.h> int mem_init_done; unsigned long ioremap_bot = IMALLOC_BASE; @@ -136,14 +137,78 @@ void iounmap(volatile void __iomem *addr) #else +static void unmap_im_area_pte(pmd_t *pmd, unsigned long addr, + unsigned long end) +{ + pte_t *pte; + + pte = pte_offset_kernel(pmd, addr); + do { + pte_t ptent = ptep_get_and_clear(&ioremap_mm, addr, pte); + WARN_ON(!pte_none(ptent) && !pte_present(ptent)); + } while (pte++, addr += PAGE_SIZE, addr != end); +} + +static inline void unmap_im_area_pmd(pud_t *pud, unsigned long addr, + unsigned long end) +{ + pmd_t *pmd; + unsigned long next; + + pmd = pmd_offset(pud, addr); + do { + next = pmd_addr_end(addr, end); + if (pmd_none_or_clear_bad(pmd)) + continue; + unmap_im_area_pte(pmd, addr, next); + } while (pmd++, addr = next, addr != end); +} + +static inline void unmap_im_area_pud(pgd_t *pgd, unsigned long addr, + unsigned long end) +{ + pud_t *pud; + unsigned long next; + + pud = pud_offset(pgd, addr); + do { + next = pud_addr_end(addr, end); + if (pud_none_or_clear_bad(pud)) + continue; + unmap_im_area_pmd(pud, addr, next); + } while (pud++, addr = next, addr != end); +} + +static void unmap_im_area(unsigned long addr, unsigned long end) +{ + struct mm_struct *mm = &ioremap_mm; + unsigned long next; + pgd_t *pgd; + + spin_lock(&mm->page_table_lock); + + pgd = pgd_offset_i(addr); + flush_cache_vunmap(addr, end); + do { + next = pgd_addr_end(addr, end); + if (pgd_none_or_clear_bad(pgd)) + continue; + unmap_im_area_pud(pgd, addr, next); + } while (pgd++, addr = next, addr != end); + flush_tlb_kernel_range(start, end); + + spin_unlock(&mm->page_table_lock); +} + /* * map_io_page currently only called by __ioremap * map_io_page adds an entry to the ioremap page table * and adds an entry to the HPT, possibly bolting it */ -static void map_io_page(unsigned long ea, unsigned long pa, int flags) +static int map_io_page(unsigned long ea, unsigned long pa, int flags) { pgd_t *pgdp; + pud_t *pudp; pmd_t *pmdp; pte_t *ptep; unsigned long vsid; @@ -151,9 +216,15 @@ static void map_io_page(unsigned long ea, unsigned long pa, int flags) if (mem_init_done) { spin_lock(&ioremap_mm.page_table_lock); pgdp = pgd_offset_i(ea); - pmdp = pmd_alloc(&ioremap_mm, pgdp, ea); + pudp = pud_alloc(&ioremap_mm, pgdp, ea); + if (!pudp) + return -ENOMEM; + pmdp = pmd_alloc(&ioremap_mm, pudp, ea); + if (!pmdp) + return -ENOMEM; ptep = pte_alloc_kernel(&ioremap_mm, pmdp, ea); - + if (!ptep) + return -ENOMEM; pa = abs_to_phys(pa); set_pte_at(&ioremap_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT, __pgprot(flags))); @@ -181,6 +252,7 @@ static void map_io_page(unsigned long ea, unsigned long pa, int flags) panic("map_io_page: could not insert mapping"); } } + return 0; } @@ -194,9 +266,14 @@ static void __iomem * __ioremap_com(unsigned long addr, unsigned long pa, flags |= pgprot_val(PAGE_KERNEL); for (i = 0; i < size; i += PAGE_SIZE) - map_io_page(ea+i, pa+i, flags); + if (map_io_page(ea+i, pa+i, flags)) + goto failure; return (void __iomem *) (ea + (addr & ~PAGE_MASK)); + failure: + if (mem_init_done) + unmap_im_area(ea, ea + size); + return NULL; } @@ -206,10 +283,11 @@ ioremap(unsigned long addr, unsigned long size) return __ioremap(addr, size, _PAGE_NO_CACHE | _PAGE_GUARDED); } -void __iomem * -__ioremap(unsigned long addr, unsigned long size, unsigned long flags) +void __iomem * __ioremap(unsigned long addr, unsigned long size, + unsigned long flags) { unsigned long pa, ea; + void __iomem *ret; /* * Choose an address to map it to. @@ -232,12 +310,16 @@ __ioremap(unsigned long addr, unsigned long size, unsigned long flags) if (area == NULL) return NULL; ea = (unsigned long)(area->addr); + ret = __ioremap_com(addr, pa, ea, size, flags); + if (!ret) + im_free(area->addr); } else { ea = ioremap_bot; - ioremap_bot += size; + ret = __ioremap_com(addr, pa, ea, size, flags); + if (ret) + ioremap_bot += size; } - - return __ioremap_com(addr, pa, ea, size, flags); + return ret; } #define IS_PAGE_ALIGNED(_val) ((_val) == ((_val) & PAGE_MASK)) @@ -246,6 +328,7 @@ int __ioremap_explicit(unsigned long pa, unsigned long ea, unsigned long size, unsigned long flags) { struct vm_struct *area; + void __iomem *ret; /* For now, require page-aligned values for pa, ea, and size */ if (!IS_PAGE_ALIGNED(pa) || !IS_PAGE_ALIGNED(ea) || @@ -276,7 +359,12 @@ int __ioremap_explicit(unsigned long pa, unsigned long ea, } } - if (__ioremap_com(pa, pa, ea, size, flags) != (void *) ea) { + ret = __ioremap_com(pa, pa, ea, size, flags); + if (ret == NULL) { + printk(KERN_ERR "ioremap_explicit() allocation failure !\n"); + return 1; + } + if (ret != (void *) ea) { printk(KERN_ERR "__ioremap_com() returned unexpected addr\n"); return 1; } @@ -284,69 +372,6 @@ int __ioremap_explicit(unsigned long pa, unsigned long ea, return 0; } -static void unmap_im_area_pte(pmd_t *pmd, unsigned long address, - unsigned long size) -{ - unsigned long base, end; - pte_t *pte; - - if (pmd_none(*pmd)) - return; - if (pmd_bad(*pmd)) { - pmd_ERROR(*pmd); - pmd_clear(pmd); - return; - } - - pte = pte_offset_kernel(pmd, address); - base = address & PMD_MASK; - address &= ~PMD_MASK; - end = address + size; - if (end > PMD_SIZE) - end = PMD_SIZE; - - do { - pte_t page; - page = ptep_get_and_clear(&ioremap_mm, base + address, pte); - address += PAGE_SIZE; - pte++; - if (pte_none(page)) - continue; - if (pte_present(page)) - continue; - printk(KERN_CRIT "Whee.. Swapped out page in kernel page" - " table\n"); - } while (address < end); -} - -static void unmap_im_area_pmd(pgd_t *dir, unsigned long address, - unsigned long size) -{ - unsigned long base, end; - pmd_t *pmd; - - if (pgd_none(*dir)) - return; - if (pgd_bad(*dir)) { - pgd_ERROR(*dir); - pgd_clear(dir); - return; - } - - pmd = pmd_offset(dir, address); - base = address & PGDIR_MASK; - address &= ~PGDIR_MASK; - end = address + size; - if (end > PGDIR_SIZE) - end = PGDIR_SIZE; - - do { - unmap_im_area_pte(pmd, base + address, end - address); - address = (address + PMD_SIZE) & PMD_MASK; - pmd++; - } while (address < end); -} - /* * Unmap an IO region and remove it from imalloc'd list. * Access to IO memory should be serialized by driver. @@ -356,39 +381,19 @@ static void unmap_im_area_pmd(pgd_t *dir, unsigned long address, */ void iounmap(volatile void __iomem *token) { - unsigned long address, start, end, size; - struct mm_struct *mm; - pgd_t *dir; + unsigned long address, size; void *addr; - if (!mem_init_done) { + if (!mem_init_done) return; - } addr = (void *) ((unsigned long __force) token & PAGE_MASK); - if ((size = im_free(addr)) == 0) { + if ((size = im_free(addr)) == 0) return; - } address = (unsigned long)addr; - start = address; - end = address + size; - - mm = &ioremap_mm; - spin_lock(&mm->page_table_lock); - - dir = pgd_offset_i(address); - flush_cache_vunmap(address, end); - do { - unmap_im_area_pmd(dir, address, end - address); - address = (address + PGDIR_SIZE) & PGDIR_MASK; - dir++; - } while (address && (address < end)); - flush_tlb_kernel_range(start, end); - - spin_unlock(&mm->page_table_lock); - return; + unmap_im_area(address, address + size); } static int iounmap_subset_regions(unsigned long addr, unsigned long size) @@ -664,7 +669,7 @@ void __init paging_init(void) zones_size[ZONE_DMA] = top_of_ram >> PAGE_SHIFT; zholes_size[ZONE_DMA] = (top_of_ram - total_ram) >> PAGE_SHIFT; - free_area_init_node(0, &contig_page_data, zones_size, + free_area_init_node(0, NODE_DATA(0), zones_size, __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size); } #endif /* CONFIG_DISCONTIGMEM */ diff --git a/arch/ppc64/mm/slb.c b/arch/ppc64/mm/slb.c index 6a20773f695..244150a0bc1 100644 --- a/arch/ppc64/mm/slb.c +++ b/arch/ppc64/mm/slb.c @@ -33,8 +33,8 @@ static inline unsigned long mk_vsid_data(unsigned long ea, unsigned long flags) return (get_kernel_vsid(ea) << SLB_VSID_SHIFT) | flags; } -static inline void create_slbe(unsigned long ea, unsigned long vsid, - unsigned long flags, unsigned long entry) +static inline void create_slbe(unsigned long ea, unsigned long flags, + unsigned long entry) { asm volatile("slbmte %0,%1" : : "r" (mk_vsid_data(ea, flags)), @@ -145,9 +145,8 @@ void slb_initialize(void) asm volatile("isync":::"memory"); asm volatile("slbmte %0,%0"::"r" (0) : "memory"); asm volatile("isync; slbia; isync":::"memory"); - create_slbe(KERNELBASE, get_kernel_vsid(KERNELBASE), flags, 0); - create_slbe(VMALLOCBASE, get_kernel_vsid(KERNELBASE), - SLB_VSID_KERNEL, 1); + create_slbe(KERNELBASE, flags, 0); + create_slbe(VMALLOCBASE, SLB_VSID_KERNEL, 1); /* We don't bolt the stack for the time being - we're in boot, * so the stack is in the bolted segment. By the time it goes * elsewhere, we'll call _switch() which will bolt in the new diff --git a/arch/ppc64/mm/stab.c b/arch/ppc64/mm/stab.c index 31491131d5e..df4bbe14153 100644 --- a/arch/ppc64/mm/stab.c +++ b/arch/ppc64/mm/stab.c @@ -19,6 +19,11 @@ #include <asm/paca.h> #include <asm/cputable.h> +struct stab_entry { + unsigned long esid_data; + unsigned long vsid_data; +}; + /* Both the segment table and SLB code uses the following cache */ #define NR_STAB_CACHE_ENTRIES 8 DEFINE_PER_CPU(long, stab_cache_ptr); diff --git a/arch/ppc64/xmon/ppc-opc.c b/arch/ppc64/xmon/ppc-opc.c index 1e4e7e31997..5ee8fc32f82 100644 --- a/arch/ppc64/xmon/ppc-opc.c +++ b/arch/ppc64/xmon/ppc-opc.c @@ -20,6 +20,7 @@ Software Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ +#include <linux/stddef.h> #include "nonstdio.h" #include "ppc.h" @@ -110,12 +111,12 @@ const struct powerpc_operand powerpc_operands[] = /* The zero index is used to indicate the end of the list of operands. */ #define UNUSED 0 - { 0, 0, 0, 0, 0 }, + { 0, 0, NULL, NULL, 0 }, /* The BA field in an XL form instruction. */ #define BA UNUSED + 1 #define BA_MASK (0x1f << 16) - { 5, 16, 0, 0, PPC_OPERAND_CR }, + { 5, 16, NULL, NULL, PPC_OPERAND_CR }, /* The BA field in an XL form instruction when it must be the same as the BT field in the same instruction. */ @@ -125,7 +126,7 @@ const struct powerpc_operand powerpc_operands[] = /* The BB field in an XL form instruction. */ #define BB BAT + 1 #define BB_MASK (0x1f << 11) - { 5, 11, 0, 0, PPC_OPERAND_CR }, + { 5, 11, NULL, NULL, PPC_OPERAND_CR }, /* The BB field in an XL form instruction when it must be the same as the BA field in the same instruction. */ @@ -168,21 +169,21 @@ const struct powerpc_operand powerpc_operands[] = /* The BF field in an X or XL form instruction. */ #define BF BDPA + 1 - { 3, 23, 0, 0, PPC_OPERAND_CR }, + { 3, 23, NULL, NULL, PPC_OPERAND_CR }, /* An optional BF field. This is used for comparison instructions, in which an omitted BF field is taken as zero. */ #define OBF BF + 1 - { 3, 23, 0, 0, PPC_OPERAND_CR | PPC_OPERAND_OPTIONAL }, + { 3, 23, NULL, NULL, PPC_OPERAND_CR | PPC_OPERAND_OPTIONAL }, /* The BFA field in an X or XL form instruction. */ #define BFA OBF + 1 - { 3, 18, 0, 0, PPC_OPERAND_CR }, + { 3, 18, NULL, NULL, PPC_OPERAND_CR }, /* The BI field in a B form or XL form instruction. */ #define BI BFA + 1 #define BI_MASK (0x1f << 16) - { 5, 16, 0, 0, PPC_OPERAND_CR }, + { 5, 16, NULL, NULL, PPC_OPERAND_CR }, /* The BO field in a B form instruction. Certain values are illegal. */ @@ -197,36 +198,36 @@ const struct powerpc_operand powerpc_operands[] = /* The BT field in an X or XL form instruction. */ #define BT BOE + 1 - { 5, 21, 0, 0, PPC_OPERAND_CR }, + { 5, 21, NULL, NULL, PPC_OPERAND_CR }, /* The condition register number portion of the BI field in a B form or XL form instruction. This is used for the extended conditional branch mnemonics, which set the lower two bits of the BI field. This field is optional. */ #define CR BT + 1 - { 3, 18, 0, 0, PPC_OPERAND_CR | PPC_OPERAND_OPTIONAL }, + { 3, 18, NULL, NULL, PPC_OPERAND_CR | PPC_OPERAND_OPTIONAL }, /* The CRB field in an X form instruction. */ #define CRB CR + 1 - { 5, 6, 0, 0, 0 }, + { 5, 6, NULL, NULL, 0 }, /* The CRFD field in an X form instruction. */ #define CRFD CRB + 1 - { 3, 23, 0, 0, PPC_OPERAND_CR }, + { 3, 23, NULL, NULL, PPC_OPERAND_CR }, /* The CRFS field in an X form instruction. */ #define CRFS CRFD + 1 - { 3, 0, 0, 0, PPC_OPERAND_CR }, + { 3, 0, NULL, NULL, PPC_OPERAND_CR }, /* The CT field in an X form instruction. */ #define CT CRFS + 1 - { 5, 21, 0, 0, PPC_OPERAND_OPTIONAL }, + { 5, 21, NULL, NULL, PPC_OPERAND_OPTIONAL }, /* The D field in a D form instruction. This is a displacement off a register, and implies that the next operand is a register in parentheses. */ #define D CT + 1 - { 16, 0, 0, 0, PPC_OPERAND_PARENS | PPC_OPERAND_SIGNED }, + { 16, 0, NULL, NULL, PPC_OPERAND_PARENS | PPC_OPERAND_SIGNED }, /* The DE field in a DE form instruction. This is like D, but is 12 bits only. */ @@ -252,40 +253,40 @@ const struct powerpc_operand powerpc_operands[] = /* The E field in a wrteei instruction. */ #define E DS + 1 - { 1, 15, 0, 0, 0 }, + { 1, 15, NULL, NULL, 0 }, /* The FL1 field in a POWER SC form instruction. */ #define FL1 E + 1 - { 4, 12, 0, 0, 0 }, + { 4, 12, NULL, NULL, 0 }, /* The FL2 field in a POWER SC form instruction. */ #define FL2 FL1 + 1 - { 3, 2, 0, 0, 0 }, + { 3, 2, NULL, NULL, 0 }, /* The FLM field in an XFL form instruction. */ #define FLM FL2 + 1 - { 8, 17, 0, 0, 0 }, + { 8, 17, NULL, NULL, 0 }, /* The FRA field in an X or A form instruction. */ #define FRA FLM + 1 #define FRA_MASK (0x1f << 16) - { 5, 16, 0, 0, PPC_OPERAND_FPR }, + { 5, 16, NULL, NULL, PPC_OPERAND_FPR }, /* The FRB field in an X or A form instruction. */ #define FRB FRA + 1 #define FRB_MASK (0x1f << 11) - { 5, 11, 0, 0, PPC_OPERAND_FPR }, + { 5, 11, NULL, NULL, PPC_OPERAND_FPR }, /* The FRC field in an A form instruction. */ #define FRC FRB + 1 #define FRC_MASK (0x1f << 6) - { 5, 6, 0, 0, PPC_OPERAND_FPR }, + { 5, 6, NULL, NULL, PPC_OPERAND_FPR }, /* The FRS field in an X form instruction or the FRT field in a D, X or A form instruction. */ #define FRS FRC + 1 #define FRT FRS - { 5, 21, 0, 0, PPC_OPERAND_FPR }, + { 5, 21, NULL, NULL, PPC_OPERAND_FPR }, /* The FXM field in an XFX instruction. */ #define FXM FRS + 1 @@ -298,11 +299,11 @@ const struct powerpc_operand powerpc_operands[] = /* The L field in a D or X form instruction. */ #define L FXM4 + 1 - { 1, 21, 0, 0, PPC_OPERAND_OPTIONAL }, + { 1, 21, NULL, NULL, PPC_OPERAND_OPTIONAL }, /* The LEV field in a POWER SC form instruction. */ #define LEV L + 1 - { 7, 5, 0, 0, 0 }, + { 7, 5, NULL, NULL, 0 }, /* The LI field in an I form instruction. The lower two bits are forced to zero. */ @@ -316,24 +317,24 @@ const struct powerpc_operand powerpc_operands[] = /* The LS field in an X (sync) form instruction. */ #define LS LIA + 1 - { 2, 21, 0, 0, PPC_OPERAND_OPTIONAL }, + { 2, 21, NULL, NULL, PPC_OPERAND_OPTIONAL }, /* The MB field in an M form instruction. */ #define MB LS + 1 #define MB_MASK (0x1f << 6) - { 5, 6, 0, 0, 0 }, + { 5, 6, NULL, NULL, 0 }, /* The ME field in an M form instruction. */ #define ME MB + 1 #define ME_MASK (0x1f << 1) - { 5, 1, 0, 0, 0 }, + { 5, 1, NULL, NULL, 0 }, /* The MB and ME fields in an M form instruction expressed a single operand which is a bitmask indicating which bits to select. This is a two operand form using PPC_OPERAND_NEXT. See the description in opcode/ppc.h for what this means. */ #define MBE ME + 1 - { 5, 6, 0, 0, PPC_OPERAND_OPTIONAL | PPC_OPERAND_NEXT }, + { 5, 6, NULL, NULL, PPC_OPERAND_OPTIONAL | PPC_OPERAND_NEXT }, { 32, 0, insert_mbe, extract_mbe, 0 }, /* The MB or ME field in an MD or MDS form instruction. The high @@ -345,7 +346,7 @@ const struct powerpc_operand powerpc_operands[] = /* The MO field in an mbar instruction. */ #define MO MB6 + 1 - { 5, 21, 0, 0, 0 }, + { 5, 21, NULL, NULL, 0 }, /* The NB field in an X form instruction. The value 32 is stored as 0. */ @@ -361,34 +362,34 @@ const struct powerpc_operand powerpc_operands[] = /* The RA field in an D, DS, DQ, X, XO, M, or MDS form instruction. */ #define RA NSI + 1 #define RA_MASK (0x1f << 16) - { 5, 16, 0, 0, PPC_OPERAND_GPR }, + { 5, 16, NULL, NULL, PPC_OPERAND_GPR }, /* The RA field in the DQ form lq instruction, which has special value restrictions. */ #define RAQ RA + 1 - { 5, 16, insert_raq, 0, PPC_OPERAND_GPR }, + { 5, 16, insert_raq, NULL, PPC_OPERAND_GPR }, /* The RA field in a D or X form instruction which is an updating load, which means that the RA field may not be zero and may not equal the RT field. */ #define RAL RAQ + 1 - { 5, 16, insert_ral, 0, PPC_OPERAND_GPR }, + { 5, 16, insert_ral, NULL, PPC_OPERAND_GPR }, /* The RA field in an lmw instruction, which has special value restrictions. */ #define RAM RAL + 1 - { 5, 16, insert_ram, 0, PPC_OPERAND_GPR }, + { 5, 16, insert_ram, NULL, PPC_OPERAND_GPR }, /* The RA field in a D or X form instruction which is an updating store or an updating floating point load, which means that the RA field may not be zero. */ #define RAS RAM + 1 - { 5, 16, insert_ras, 0, PPC_OPERAND_GPR }, + { 5, 16, insert_ras, NULL, PPC_OPERAND_GPR }, /* The RB field in an X, XO, M, or MDS form instruction. */ #define RB RAS + 1 #define RB_MASK (0x1f << 11) - { 5, 11, 0, 0, PPC_OPERAND_GPR }, + { 5, 11, NULL, NULL, PPC_OPERAND_GPR }, /* The RB field in an X form instruction when it must be the same as the RS field in the instruction. This is used for extended @@ -402,22 +403,22 @@ const struct powerpc_operand powerpc_operands[] = #define RS RBS + 1 #define RT RS #define RT_MASK (0x1f << 21) - { 5, 21, 0, 0, PPC_OPERAND_GPR }, + { 5, 21, NULL, NULL, PPC_OPERAND_GPR }, /* The RS field of the DS form stq instruction, which has special value restrictions. */ #define RSQ RS + 1 - { 5, 21, insert_rsq, 0, PPC_OPERAND_GPR }, + { 5, 21, insert_rsq, NULL, PPC_OPERAND_GPR }, /* The RT field of the DQ form lq instruction, which has special value restrictions. */ #define RTQ RSQ + 1 - { 5, 21, insert_rtq, 0, PPC_OPERAND_GPR }, + { 5, 21, insert_rtq, NULL, PPC_OPERAND_GPR }, /* The SH field in an X or M form instruction. */ #define SH RTQ + 1 #define SH_MASK (0x1f << 11) - { 5, 11, 0, 0, 0 }, + { 5, 11, NULL, NULL, 0 }, /* The SH field in an MD form instruction. This is split. */ #define SH6 SH + 1 @@ -426,12 +427,12 @@ const struct powerpc_operand powerpc_operands[] = /* The SI field in a D form instruction. */ #define SI SH6 + 1 - { 16, 0, 0, 0, PPC_OPERAND_SIGNED }, + { 16, 0, NULL, NULL, PPC_OPERAND_SIGNED }, /* The SI field in a D form instruction when we accept a wide range of positive values. */ #define SISIGNOPT SI + 1 - { 16, 0, 0, 0, PPC_OPERAND_SIGNED | PPC_OPERAND_SIGNOPT }, + { 16, 0, NULL, NULL, PPC_OPERAND_SIGNED | PPC_OPERAND_SIGNOPT }, /* The SPR field in an XFX form instruction. This is flipped--the lower 5 bits are stored in the upper 5 and vice- versa. */ @@ -443,25 +444,25 @@ const struct powerpc_operand powerpc_operands[] = /* The BAT index number in an XFX form m[ft]ibat[lu] instruction. */ #define SPRBAT SPR + 1 #define SPRBAT_MASK (0x3 << 17) - { 2, 17, 0, 0, 0 }, + { 2, 17, NULL, NULL, 0 }, /* The SPRG register number in an XFX form m[ft]sprg instruction. */ #define SPRG SPRBAT + 1 #define SPRG_MASK (0x3 << 16) - { 2, 16, 0, 0, 0 }, + { 2, 16, NULL, NULL, 0 }, /* The SR field in an X form instruction. */ #define SR SPRG + 1 - { 4, 16, 0, 0, 0 }, + { 4, 16, NULL, NULL, 0 }, /* The STRM field in an X AltiVec form instruction. */ #define STRM SR + 1 #define STRM_MASK (0x3 << 21) - { 2, 21, 0, 0, 0 }, + { 2, 21, NULL, NULL, 0 }, /* The SV field in a POWER SC form instruction. */ #define SV STRM + 1 - { 14, 2, 0, 0, 0 }, + { 14, 2, NULL, NULL, 0 }, /* The TBR field in an XFX form instruction. This is like the SPR field, but it is optional. */ @@ -471,52 +472,52 @@ const struct powerpc_operand powerpc_operands[] = /* The TO field in a D or X form instruction. */ #define TO TBR + 1 #define TO_MASK (0x1f << 21) - { 5, 21, 0, 0, 0 }, + { 5, 21, NULL, NULL, 0 }, /* The U field in an X form instruction. */ #define U TO + 1 - { 4, 12, 0, 0, 0 }, + { 4, 12, NULL, NULL, 0 }, /* The UI field in a D form instruction. */ #define UI U + 1 - { 16, 0, 0, 0, 0 }, + { 16, 0, NULL, NULL, 0 }, /* The VA field in a VA, VX or VXR form instruction. */ #define VA UI + 1 #define VA_MASK (0x1f << 16) - { 5, 16, 0, 0, PPC_OPERAND_VR }, + { 5, 16, NULL, NULL, PPC_OPERAND_VR }, /* The VB field in a VA, VX or VXR form instruction. */ #define VB VA + 1 #define VB_MASK (0x1f << 11) - { 5, 11, 0, 0, PPC_OPERAND_VR }, + { 5, 11, NULL, NULL, PPC_OPERAND_VR }, /* The VC field in a VA form instruction. */ #define VC VB + 1 #define VC_MASK (0x1f << 6) - { 5, 6, 0, 0, PPC_OPERAND_VR }, + { 5, 6, NULL, NULL, PPC_OPERAND_VR }, /* The VD or VS field in a VA, VX, VXR or X form instruction. */ #define VD VC + 1 #define VS VD #define VD_MASK (0x1f << 21) - { 5, 21, 0, 0, PPC_OPERAND_VR }, + { 5, 21, NULL, NULL, PPC_OPERAND_VR }, /* The SIMM field in a VX form instruction. */ #define SIMM VD + 1 - { 5, 16, 0, 0, PPC_OPERAND_SIGNED}, + { 5, 16, NULL, NULL, PPC_OPERAND_SIGNED}, /* The UIMM field in a VX form instruction. */ #define UIMM SIMM + 1 - { 5, 16, 0, 0, 0 }, + { 5, 16, NULL, NULL, 0 }, /* The SHB field in a VA form instruction. */ #define SHB UIMM + 1 - { 4, 6, 0, 0, 0 }, + { 4, 6, NULL, NULL, 0 }, /* The other UIMM field in a EVX form instruction. */ #define EVUIMM SHB + 1 - { 5, 11, 0, 0, 0 }, + { 5, 11, NULL, NULL, 0 }, /* The other UIMM field in a half word EVX form instruction. */ #define EVUIMM_2 EVUIMM + 1 @@ -533,11 +534,11 @@ const struct powerpc_operand powerpc_operands[] = /* The WS field. */ #define WS EVUIMM_8 + 1 #define WS_MASK (0x7 << 11) - { 3, 11, 0, 0, 0 }, + { 3, 11, NULL, NULL, 0 }, /* The L field in an mtmsrd instruction */ #define MTMSRD_L WS + 1 - { 1, 16, 0, 0, PPC_OPERAND_OPTIONAL }, + { 1, 16, NULL, NULL, PPC_OPERAND_OPTIONAL }, }; diff --git a/arch/s390/defconfig b/arch/s390/defconfig index 1358b420170..07fd0414a4b 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.11 -# Wed Mar 2 16:57:55 2005 +# Linux kernel version: 2.6.12-rc3 +# Fri Apr 22 15:30:58 2005 # CONFIG_MMU=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y @@ -15,6 +15,7 @@ CONFIG_UID16=y CONFIG_EXPERIMENTAL=y CONFIG_CLEAN_COMPILE=y CONFIG_LOCK_KERNEL=y +CONFIG_INIT_ENV_ARG_LIMIT=32 # # General setup @@ -26,24 +27,25 @@ CONFIG_SYSVIPC=y # CONFIG_BSD_PROCESS_ACCT is not set CONFIG_SYSCTL=y # CONFIG_AUDIT is not set -CONFIG_LOG_BUF_SHIFT=17 CONFIG_HOTPLUG=y CONFIG_KOBJECT_UEVENT=y CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y +# CONFIG_CPUSETS is not set # CONFIG_EMBEDDED is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set # CONFIG_KALLSYMS_EXTRA_PASS is not set +CONFIG_BASE_FULL=y CONFIG_FUTEX=y CONFIG_EPOLL=y -# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set CONFIG_SHMEM=y CONFIG_CC_ALIGN_FUNCTIONS=0 CONFIG_CC_ALIGN_LABELS=0 CONFIG_CC_ALIGN_LOOPS=0 CONFIG_CC_ALIGN_JUMPS=0 # CONFIG_TINY_SHMEM is not set +CONFIG_BASE_SMALL=0 # # Loadable module support @@ -261,7 +263,6 @@ CONFIG_NET=y # CONFIG_PACKET=y # CONFIG_PACKET_MMAP is not set -# CONFIG_NETLINK_DEV is not set CONFIG_UNIX=y CONFIG_NET_KEY=y CONFIG_INET=y @@ -329,6 +330,7 @@ CONFIG_NET_SCH_DSMARK=m CONFIG_NET_QOS=y CONFIG_NET_ESTIMATOR=y CONFIG_NET_CLS=y +# CONFIG_NET_CLS_BASIC is not set CONFIG_NET_CLS_TCINDEX=m CONFIG_NET_CLS_ROUTE4=m CONFIG_NET_CLS_ROUTE=y @@ -338,6 +340,7 @@ CONFIG_NET_CLS_U32=m # CONFIG_NET_CLS_IND is not set CONFIG_NET_CLS_RSVP=m CONFIG_NET_CLS_RSVP6=m +# CONFIG_NET_EMATCH is not set # CONFIG_NET_CLS_ACT is not set CONFIG_NET_CLS_POLICE=y @@ -393,6 +396,8 @@ CONFIG_CTC=m CONFIG_IUCV=m # CONFIG_NETIUCV is not set # CONFIG_SMSGIUCV is not set +# CONFIG_CLAW is not set +# CONFIG_MPC is not set CONFIG_QETH=y # @@ -532,10 +537,13 @@ CONFIG_MSDOS_PARTITION=y # # Kernel hacking # +# CONFIG_PRINTK_TIME is not set CONFIG_DEBUG_KERNEL=y CONFIG_MAGIC_SYSRQ=y +CONFIG_LOG_BUF_SHIFT=17 # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set +# CONFIG_DEBUG_SPINLOCK is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_KOBJECT is not set # CONFIG_DEBUG_INFO is not set @@ -560,6 +568,7 @@ CONFIG_CRYPTO=y # CONFIG_CRYPTO_SHA256 is not set # CONFIG_CRYPTO_SHA512 is not set # CONFIG_CRYPTO_WP512 is not set +# CONFIG_CRYPTO_TGR192 is not set # CONFIG_CRYPTO_DES is not set # CONFIG_CRYPTO_DES_Z990 is not set # CONFIG_CRYPTO_BLOWFISH is not set diff --git a/arch/s390/kernel/compat_ioctl.c b/arch/s390/kernel/compat_ioctl.c index 96571ff7115..03d03c6d3cb 100644 --- a/arch/s390/kernel/compat_ioctl.c +++ b/arch/s390/kernel/compat_ioctl.c @@ -16,6 +16,7 @@ #define CODE #include "../../../fs/compat_ioctl.c" #include <asm/dasd.h> +#include <asm/cmb.h> #include <asm/tape390.h> static int do_ioctl32_pointer(unsigned int fd, unsigned int cmd, @@ -58,7 +59,11 @@ COMPATIBLE_IOCTL(BIODASDPRRD) COMPATIBLE_IOCTL(BIODASDPSRD) COMPATIBLE_IOCTL(BIODASDGATTR) COMPATIBLE_IOCTL(BIODASDSATTR) - +#if defined(CONFIG_DASD_CMB) || defined(CONFIG_DASD_CMB_MODULE) +COMPATIBLE_IOCTL(BIODASDCMFENABLE) +COMPATIBLE_IOCTL(BIODASDCMFDISABLE) +COMPATIBLE_IOCTL(BIODASDREADALLCMB) +#endif #endif #if defined(CONFIG_S390_TAPE) || defined(CONFIG_S390_TAPE_MODULE) diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 647233c02fc..26889366929 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -32,6 +32,7 @@ #include <linux/user.h> #include <linux/security.h> #include <linux/audit.h> +#include <linux/signal.h> #include <asm/segment.h> #include <asm/page.h> @@ -609,7 +610,7 @@ do_ptrace(struct task_struct *child, long request, long addr, long data) /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ - if ((unsigned long) data >= _NSIG) + if (!valid_signal(data)) return -EIO; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -637,7 +638,7 @@ do_ptrace(struct task_struct *child, long request, long addr, long data) case PTRACE_SINGLESTEP: /* set the trap flag. */ - if ((unsigned long) data >= _NSIG) + if (!valid_signal(data)) return -EIO; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); child->exit_code = data; @@ -711,18 +712,13 @@ out: asmlinkage void syscall_trace(struct pt_regs *regs, int entryexit) { - if (unlikely(current->audit_context)) { - if (!entryexit) - audit_syscall_entry(current, regs->gprs[2], - regs->orig_gpr2, regs->gprs[3], - regs->gprs[4], regs->gprs[5]); - else - audit_syscall_exit(current, regs->gprs[2]); - } + if (unlikely(current->audit_context) && entryexit) + audit_syscall_exit(current, AUDITSC_RESULT(regs->gprs[2]), regs->gprs[2]); + if (!test_thread_flag(TIF_SYSCALL_TRACE)) - return; + goto out; if (!(current->ptrace & PT_PTRACED)) - return; + goto out; ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) ? 0x80 : 0)); @@ -735,4 +731,10 @@ syscall_trace(struct pt_regs *regs, int entryexit) send_sig(current->exit_code, current, 1); current->exit_code = 0; } + out: + if (unlikely(current->audit_context) && !entryexit) + audit_syscall_entry(current, + test_thread_flag(TIF_31BIT)?AUDIT_ARCH_S390:AUDIT_ARCH_S390X, + regs->gprs[2], regs->orig_gpr2, regs->gprs[3], + regs->gprs[4], regs->gprs[5]); } diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c index 11fd6d556d8..bee654abb6d 100644 --- a/arch/s390/kernel/s390_ksyms.c +++ b/arch/s390/kernel/s390_ksyms.c @@ -34,7 +34,6 @@ EXPORT_SYMBOL(__clear_user_asm); EXPORT_SYMBOL(__strncpy_from_user_asm); EXPORT_SYMBOL(__strnlen_user_asm); EXPORT_SYMBOL(diag10); -EXPORT_SYMBOL(default_storage_key); /* * semaphore ops diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index c879c40aa7a..df83215beac 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -44,6 +44,8 @@ #include <asm/cpcmd.h> #include <asm/lowcore.h> #include <asm/irq.h> +#include <asm/page.h> +#include <asm/ptrace.h> /* * Machine setup.. @@ -53,13 +55,14 @@ unsigned int console_devno = -1; unsigned int console_irq = -1; unsigned long memory_size = 0; unsigned long machine_flags = 0; -unsigned int default_storage_key = 0; struct { unsigned long addr, size, type; } memory_chunk[MEMORY_CHUNKS] = { { 0 } }; #define CHUNK_READ_WRITE 0 #define CHUNK_READ_ONLY 1 volatile int __cpu_logical_map[NR_CPUS]; /* logical cpu to cpu address */ +unsigned long __initdata zholes_size[MAX_NR_ZONES]; +static unsigned long __initdata memory_end; /* * Setup options @@ -78,11 +81,15 @@ static char command_line[COMMAND_LINE_SIZE] = { 0, }; static struct resource code_resource = { .name = "Kernel code", + .start = (unsigned long) &_text, + .end = (unsigned long) &_etext - 1, .flags = IORESOURCE_BUSY | IORESOURCE_MEM, }; static struct resource data_resource = { .name = "Kernel data", + .start = (unsigned long) &_etext, + .end = (unsigned long) &_edata - 1, .flags = IORESOURCE_BUSY | IORESOURCE_MEM, }; @@ -310,90 +317,50 @@ void machine_power_off(void) EXPORT_SYMBOL(machine_power_off); -/* - * Setup function called from init/main.c just after the banner - * was printed. - */ -extern char _pstart, _pend, _stext; +static void __init +add_memory_hole(unsigned long start, unsigned long end) +{ + unsigned long dma_pfn = MAX_DMA_ADDRESS >> PAGE_SHIFT; + + if (end <= dma_pfn) + zholes_size[ZONE_DMA] += end - start + 1; + else if (start > dma_pfn) + zholes_size[ZONE_NORMAL] += end - start + 1; + else { + zholes_size[ZONE_DMA] += dma_pfn - start + 1; + zholes_size[ZONE_NORMAL] += end - dma_pfn; + } +} -void __init setup_arch(char **cmdline_p) +static void __init +parse_cmdline_early(char **cmdline_p) { - unsigned long bootmap_size; - unsigned long memory_start, memory_end; - char c = ' ', cn, *to = command_line, *from = COMMAND_LINE; - unsigned long start_pfn, end_pfn; - static unsigned int smptrap=0; - unsigned long delay = 0; - struct _lowcore *lc; - int i; + char c = ' ', cn, *to = command_line, *from = COMMAND_LINE; + unsigned long delay = 0; - if (smptrap) - return; - smptrap=1; + /* Save unparsed command line copy for /proc/cmdline */ + memcpy(saved_command_line, COMMAND_LINE, COMMAND_LINE_SIZE); + saved_command_line[COMMAND_LINE_SIZE-1] = '\0'; - /* - * print what head.S has found out about the machine - */ -#ifndef CONFIG_ARCH_S390X - printk((MACHINE_IS_VM) ? - "We are running under VM (31 bit mode)\n" : - "We are running native (31 bit mode)\n"); - printk((MACHINE_HAS_IEEE) ? - "This machine has an IEEE fpu\n" : - "This machine has no IEEE fpu\n"); -#else /* CONFIG_ARCH_S390X */ - printk((MACHINE_IS_VM) ? - "We are running under VM (64 bit mode)\n" : - "We are running native (64 bit mode)\n"); -#endif /* CONFIG_ARCH_S390X */ - - ROOT_DEV = Root_RAM0; - memory_start = (unsigned long) &_end; /* fixit if use $CODELO etc*/ -#ifndef CONFIG_ARCH_S390X - memory_end = memory_size & ~0x400000UL; /* align memory end to 4MB */ - /* - * We need some free virtual space to be able to do vmalloc. - * On a machine with 2GB memory we make sure that we have at - * least 128 MB free space for vmalloc. - */ - if (memory_end > 1920*1024*1024) - memory_end = 1920*1024*1024; -#else /* CONFIG_ARCH_S390X */ - memory_end = memory_size & ~0x200000UL; /* detected in head.s */ -#endif /* CONFIG_ARCH_S390X */ - init_mm.start_code = PAGE_OFFSET; - init_mm.end_code = (unsigned long) &_etext; - init_mm.end_data = (unsigned long) &_edata; - init_mm.brk = (unsigned long) &_end; - - code_resource.start = (unsigned long) &_text; - code_resource.end = (unsigned long) &_etext - 1; - data_resource.start = (unsigned long) &_etext; - data_resource.end = (unsigned long) &_edata - 1; - - /* Save unparsed command line copy for /proc/cmdline */ - memcpy(saved_command_line, COMMAND_LINE, COMMAND_LINE_SIZE); - saved_command_line[COMMAND_LINE_SIZE-1] = '\0'; - - for (;;) { - /* - * "mem=XXX[kKmM]" sets memsize - */ - if (c == ' ' && strncmp(from, "mem=", 4) == 0) { - memory_end = simple_strtoul(from+4, &from, 0); - if ( *from == 'K' || *from == 'k' ) { - memory_end = memory_end << 10; - from++; - } else if ( *from == 'M' || *from == 'm' ) { - memory_end = memory_end << 20; - from++; - } - } - /* - * "ipldelay=XXX[sm]" sets ipl delay in seconds or minutes - */ - if (c == ' ' && strncmp(from, "ipldelay=", 9) == 0) { - delay = simple_strtoul(from+9, &from, 0); + for (;;) { + /* + * "mem=XXX[kKmM]" sets memsize + */ + if (c == ' ' && strncmp(from, "mem=", 4) == 0) { + memory_end = simple_strtoul(from+4, &from, 0); + if ( *from == 'K' || *from == 'k' ) { + memory_end = memory_end << 10; + from++; + } else if ( *from == 'M' || *from == 'm' ) { + memory_end = memory_end << 20; + from++; + } + } + /* + * "ipldelay=XXX[sm]" sets ipl delay in seconds or minutes + */ + if (c == ' ' && strncmp(from, "ipldelay=", 9) == 0) { + delay = simple_strtoul(from+9, &from, 0); if (*from == 's' || *from == 'S') { delay = delay*1000000; from++; @@ -403,24 +370,110 @@ void __init setup_arch(char **cmdline_p) } /* now wait for the requested amount of time */ udelay(delay); - } - cn = *(from++); - if (!cn) - break; - if (cn == '\n') - cn = ' '; /* replace newlines with space */ + } + cn = *(from++); + if (!cn) + break; + if (cn == '\n') + cn = ' '; /* replace newlines with space */ if (cn == 0x0d) cn = ' '; /* replace 0x0d with space */ - if (cn == ' ' && c == ' ') - continue; /* remove additional spaces */ - c = cn; - if (to - command_line >= COMMAND_LINE_SIZE) - break; - *(to++) = c; - } - if (c == ' ' && to > command_line) to--; - *to = '\0'; - *cmdline_p = command_line; + if (cn == ' ' && c == ' ') + continue; /* remove additional spaces */ + c = cn; + if (to - command_line >= COMMAND_LINE_SIZE) + break; + *(to++) = c; + } + if (c == ' ' && to > command_line) to--; + *to = '\0'; + *cmdline_p = command_line; +} + +static void __init +setup_lowcore(void) +{ + struct _lowcore *lc; + int lc_pages; + + /* + * Setup lowcore for boot cpu + */ + lc_pages = sizeof(void *) == 8 ? 2 : 1; + lc = (struct _lowcore *) + __alloc_bootmem(lc_pages * PAGE_SIZE, lc_pages * PAGE_SIZE, 0); + memset(lc, 0, lc_pages * PAGE_SIZE); + lc->restart_psw.mask = PSW_BASE_BITS | PSW_DEFAULT_KEY; + lc->restart_psw.addr = + PSW_ADDR_AMODE | (unsigned long) restart_int_handler; + lc->external_new_psw.mask = PSW_KERNEL_BITS; + lc->external_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) ext_int_handler; + lc->svc_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_IO | PSW_MASK_EXT; + lc->svc_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) system_call; + lc->program_new_psw.mask = PSW_KERNEL_BITS; + lc->program_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long)pgm_check_handler; + lc->mcck_new_psw.mask = PSW_KERNEL_BITS; + lc->mcck_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) mcck_int_handler; + lc->io_new_psw.mask = PSW_KERNEL_BITS; + lc->io_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) io_int_handler; + lc->ipl_device = S390_lowcore.ipl_device; + lc->jiffy_timer = -1LL; + lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE; + lc->async_stack = (unsigned long) + __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + ASYNC_SIZE; +#ifdef CONFIG_CHECK_STACK + lc->panic_stack = (unsigned long) + __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0) + PAGE_SIZE; +#endif + lc->current_task = (unsigned long) init_thread_union.thread_info.task; + lc->thread_info = (unsigned long) &init_thread_union; +#ifdef CONFIG_ARCH_S390X + if (MACHINE_HAS_DIAG44) + lc->diag44_opcode = 0x83000044; + else + lc->diag44_opcode = 0x07000700; +#endif /* CONFIG_ARCH_S390X */ + set_prefix((u32)(unsigned long) lc); +} + +static void __init +setup_resources(void) +{ + struct resource *res; + int i; + + for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) { + res = alloc_bootmem_low(sizeof(struct resource)); + res->flags = IORESOURCE_BUSY | IORESOURCE_MEM; + switch (memory_chunk[i].type) { + case CHUNK_READ_WRITE: + res->name = "System RAM"; + break; + case CHUNK_READ_ONLY: + res->name = "System ROM"; + res->flags |= IORESOURCE_READONLY; + break; + default: + res->name = "reserved"; + } + res->start = memory_chunk[i].addr; + res->end = memory_chunk[i].addr + memory_chunk[i].size - 1; + request_resource(&iomem_resource, res); + request_resource(res, &code_resource); + request_resource(res, &data_resource); + } +} + +static void __init +setup_memory(void) +{ + unsigned long bootmap_size; + unsigned long start_pfn, end_pfn, init_pfn; + unsigned long last_rw_end; + int i; /* * partially used pages are not usable - thus @@ -429,6 +482,10 @@ void __init setup_arch(char **cmdline_p) start_pfn = (__pa(&_end) + PAGE_SIZE - 1) >> PAGE_SHIFT; end_pfn = max_pfn = memory_end >> PAGE_SHIFT; + /* Initialize storage key for kernel pages */ + for (init_pfn = 0 ; init_pfn < start_pfn; init_pfn++) + page_set_storage_key(init_pfn << PAGE_SHIFT, PAGE_DEFAULT_KEY); + /* * Initialize the boot-time allocator (with low memory only): */ @@ -437,7 +494,9 @@ void __init setup_arch(char **cmdline_p) /* * Register RAM areas with the bootmem allocator. */ - for (i = 0; i < 16 && memory_chunk[i].size > 0; i++) { + last_rw_end = start_pfn; + + for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) { unsigned long start_chunk, end_chunk; if (memory_chunk[i].type != CHUNK_READ_WRITE) @@ -450,102 +509,98 @@ void __init setup_arch(char **cmdline_p) start_chunk = start_pfn; if (end_chunk > end_pfn) end_chunk = end_pfn; - if (start_chunk < end_chunk) + if (start_chunk < end_chunk) { + /* Initialize storage key for RAM pages */ + for (init_pfn = start_chunk ; init_pfn < end_chunk; + init_pfn++) + page_set_storage_key(init_pfn << PAGE_SHIFT, + PAGE_DEFAULT_KEY); free_bootmem(start_chunk << PAGE_SHIFT, (end_chunk - start_chunk) << PAGE_SHIFT); + if (last_rw_end < start_chunk) + add_memory_hole(last_rw_end, start_chunk - 1); + last_rw_end = end_chunk; + } } - /* - * Reserve the bootmem bitmap itself as well. We do this in two - * steps (first step was init_bootmem()) because this catches - * the (very unlikely) case of us accidentally initializing the - * bootmem allocator with an invalid RAM area. - */ - reserve_bootmem(start_pfn << PAGE_SHIFT, bootmap_size); + psw_set_key(PAGE_DEFAULT_KEY); + + if (last_rw_end < end_pfn - 1) + add_memory_hole(last_rw_end, end_pfn - 1); + + /* + * Reserve the bootmem bitmap itself as well. We do this in two + * steps (first step was init_bootmem()) because this catches + * the (very unlikely) case of us accidentally initializing the + * bootmem allocator with an invalid RAM area. + */ + reserve_bootmem(start_pfn << PAGE_SHIFT, bootmap_size); #ifdef CONFIG_BLK_DEV_INITRD - if (INITRD_START) { + if (INITRD_START) { if (INITRD_START + INITRD_SIZE <= memory_end) { reserve_bootmem(INITRD_START, INITRD_SIZE); initrd_start = INITRD_START; initrd_end = initrd_start + INITRD_SIZE; } else { - printk("initrd extends beyond end of memory " - "(0x%08lx > 0x%08lx)\ndisabling initrd\n", - initrd_start + INITRD_SIZE, memory_end); - initrd_start = initrd_end = 0; + printk("initrd extends beyond end of memory " + "(0x%08lx > 0x%08lx)\ndisabling initrd\n", + initrd_start + INITRD_SIZE, memory_end); + initrd_start = initrd_end = 0; } - } + } #endif +} - for (i = 0; i < 16 && memory_chunk[i].size > 0; i++) { - struct resource *res; - - res = alloc_bootmem_low(sizeof(struct resource)); - res->flags = IORESOURCE_BUSY | IORESOURCE_MEM; - - switch (memory_chunk[i].type) { - case CHUNK_READ_WRITE: - res->name = "System RAM"; - break; - case CHUNK_READ_ONLY: - res->name = "System ROM"; - res->flags |= IORESOURCE_READONLY; - break; - default: - res->name = "reserved"; - } - res->start = memory_chunk[i].addr; - res->end = memory_chunk[i].addr + memory_chunk[i].size - 1; - request_resource(&iomem_resource, res); - request_resource(res, &code_resource); - request_resource(res, &data_resource); - } +/* + * Setup function called from init/main.c just after the banner + * was printed. + */ +void __init +setup_arch(char **cmdline_p) +{ /* - * Setup lowcore for boot cpu + * print what head.S has found out about the machine */ #ifndef CONFIG_ARCH_S390X - lc = (struct _lowcore *) __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0); - memset(lc, 0, PAGE_SIZE); + printk((MACHINE_IS_VM) ? + "We are running under VM (31 bit mode)\n" : + "We are running native (31 bit mode)\n"); + printk((MACHINE_HAS_IEEE) ? + "This machine has an IEEE fpu\n" : + "This machine has no IEEE fpu\n"); #else /* CONFIG_ARCH_S390X */ - lc = (struct _lowcore *) __alloc_bootmem(2*PAGE_SIZE, 2*PAGE_SIZE, 0); - memset(lc, 0, 2*PAGE_SIZE); + printk((MACHINE_IS_VM) ? + "We are running under VM (64 bit mode)\n" : + "We are running native (64 bit mode)\n"); #endif /* CONFIG_ARCH_S390X */ - lc->restart_psw.mask = PSW_BASE_BITS; - lc->restart_psw.addr = - PSW_ADDR_AMODE | (unsigned long) restart_int_handler; - lc->external_new_psw.mask = PSW_KERNEL_BITS; - lc->external_new_psw.addr = - PSW_ADDR_AMODE | (unsigned long) ext_int_handler; - lc->svc_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_IO | PSW_MASK_EXT; - lc->svc_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) system_call; - lc->program_new_psw.mask = PSW_KERNEL_BITS; - lc->program_new_psw.addr = - PSW_ADDR_AMODE | (unsigned long)pgm_check_handler; - lc->mcck_new_psw.mask = PSW_KERNEL_BITS; - lc->mcck_new_psw.addr = - PSW_ADDR_AMODE | (unsigned long) mcck_int_handler; - lc->io_new_psw.mask = PSW_KERNEL_BITS; - lc->io_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) io_int_handler; - lc->ipl_device = S390_lowcore.ipl_device; - lc->jiffy_timer = -1LL; - lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE; - lc->async_stack = (unsigned long) - __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + ASYNC_SIZE; -#ifdef CONFIG_CHECK_STACK - lc->panic_stack = (unsigned long) - __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0) + PAGE_SIZE; -#endif - lc->current_task = (unsigned long) init_thread_union.thread_info.task; - lc->thread_info = (unsigned long) &init_thread_union; -#ifdef CONFIG_ARCH_S390X - if (MACHINE_HAS_DIAG44) - lc->diag44_opcode = 0x83000044; - else - lc->diag44_opcode = 0x07000700; + + ROOT_DEV = Root_RAM0; +#ifndef CONFIG_ARCH_S390X + memory_end = memory_size & ~0x400000UL; /* align memory end to 4MB */ + /* + * We need some free virtual space to be able to do vmalloc. + * On a machine with 2GB memory we make sure that we have at + * least 128 MB free space for vmalloc. + */ + if (memory_end > 1920*1024*1024) + memory_end = 1920*1024*1024; +#else /* CONFIG_ARCH_S390X */ + memory_end = memory_size & ~0x200000UL; /* detected in head.s */ #endif /* CONFIG_ARCH_S390X */ - set_prefix((u32)(unsigned long) lc); + + init_mm.start_code = PAGE_OFFSET; + init_mm.end_code = (unsigned long) &_etext; + init_mm.end_data = (unsigned long) &_edata; + init_mm.brk = (unsigned long) &_end; + + parse_cmdline_early(cmdline_p); + + setup_memory(); + setup_resources(); + setup_lowcore(); + cpu_init(); __cpu_logical_map[0] = S390_lowcore.cpu_data.cpu_addr; diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 061e81138dc..8ca48567678 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -244,7 +244,7 @@ int sysctl_hz_timer = 1; */ static inline void stop_hz_timer(void) { - __u64 timer; + __u64 timer, todval; if (sysctl_hz_timer != 0) return; @@ -265,8 +265,14 @@ static inline void stop_hz_timer(void) * for the next event. */ timer = (__u64) (next_timer_interrupt() - jiffies) + jiffies_64; - timer = jiffies_timer_cc + timer * CLK_TICKS_PER_JIFFY; - asm volatile ("SCKC %0" : : "m" (timer)); + todval = -1ULL; + /* Be careful about overflows. */ + if (timer < (-1ULL / CLK_TICKS_PER_JIFFY)) { + timer = jiffies_timer_cc + timer * CLK_TICKS_PER_JIFFY; + if (timer >= jiffies_timer_cc) + todval = timer; + } + asm volatile ("SCKC %0" : : "m" (todval)); } /* diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index bb6cf02418a..fa0726507b3 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -122,12 +122,17 @@ static void start_cpu_timer(void) struct vtimer_queue *vt_list; vt_list = &per_cpu(virt_cpu_timer, smp_processor_id()); - set_vtimer(vt_list->idle); + + /* CPU timer interrupt is pending, don't reprogramm it */ + if (vt_list->idle & 1LL<<63) + return; + + if (!list_empty(&vt_list->list)) + set_vtimer(vt_list->idle); } static void stop_cpu_timer(void) { - __u64 done; struct vtimer_queue *vt_list; vt_list = &per_cpu(virt_cpu_timer, smp_processor_id()); @@ -138,21 +143,17 @@ static void stop_cpu_timer(void) goto fire; } - /* store progress */ - asm volatile ("STPT %0" : "=m" (done)); + /* store the actual expire value */ + asm volatile ("STPT %0" : "=m" (vt_list->idle)); /* - * If done is negative we do not stop the CPU timer - * because we will get instantly an interrupt that - * will start the CPU timer again. + * If the CPU timer is negative we don't reprogramm + * it because we will get instantly an interrupt. */ - if (done & 1LL<<63) + if (vt_list->idle & 1LL<<63) return; - else - vt_list->offset += vt_list->to_expire - done; - /* save the actual expire value */ - vt_list->idle = done; + vt_list->offset += vt_list->to_expire - vt_list->idle; /* * We cannot halt the CPU timer, we just write a value that diff --git a/arch/s390/mm/cmm.c b/arch/s390/mm/cmm.c index d30cdb4248a..f5a5bc09b8f 100644 --- a/arch/s390/mm/cmm.c +++ b/arch/s390/mm/cmm.c @@ -20,6 +20,11 @@ #include <asm/pgalloc.h> #include <asm/uaccess.h> +static char *sender = "VMRMSVM"; +module_param(sender, charp, 0); +MODULE_PARM_DESC(sender, + "Guest name that may send SMSG messages (default VMRMSVM)"); + #include "../../../drivers/s390/net/smsgiucv.h" #define CMM_NR_PAGES ((PAGE_SIZE / sizeof(unsigned long)) - 2) @@ -367,10 +372,12 @@ static struct ctl_table cmm_dir_table[] = { #ifdef CONFIG_CMM_IUCV #define SMSG_PREFIX "CMM" static void -cmm_smsg_target(char *msg) +cmm_smsg_target(char *from, char *msg) { long pages, seconds; + if (strlen(sender) > 0 && strcmp(from, sender) != 0) + return; if (!cmm_skip_blanks(msg + strlen(SMSG_PREFIX), &msg)) return; if (strncmp(msg, "SHRINK", 6) == 0) { diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index 8e723bc7f79..6ec5cd981e7 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -101,6 +101,7 @@ extern unsigned long _end; extern unsigned long __init_begin; extern unsigned long __init_end; +extern unsigned long __initdata zholes_size[]; /* * paging_init() sets up the page tables */ @@ -163,10 +164,13 @@ void __init paging_init(void) local_flush_tlb(); { - unsigned long zones_size[MAX_NR_ZONES] = { 0, 0, 0}; + unsigned long zones_size[MAX_NR_ZONES]; + memset(zones_size, 0, sizeof(zones_size)); zones_size[ZONE_DMA] = max_low_pfn; - free_area_init(zones_size); + free_area_init_node(0, &contig_page_data, zones_size, + __pa(PAGE_OFFSET) >> PAGE_SHIFT, + zholes_size); } return; } @@ -184,9 +188,10 @@ void __init paging_init(void) _KERN_REGION_TABLE; static const int ssm_mask = 0x04000000L; - unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0}; + unsigned long zones_size[MAX_NR_ZONES]; unsigned long dma_pfn, high_pfn; + memset(zones_size, 0, sizeof(zones_size)); dma_pfn = MAX_DMA_ADDRESS >> PAGE_SHIFT; high_pfn = max_low_pfn; @@ -198,8 +203,8 @@ void __init paging_init(void) } /* Initialize mem_map[]. */ - free_area_init(zones_size); - + free_area_init_node(0, &contig_page_data, zones_size, + __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size); /* * map whole physical memory to virtual memory (identity mapping) diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index 722ea1d63c9..3468d512722 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -693,6 +693,10 @@ config RTC_9701JE endmenu +config ISA_DMA_API + bool + depends on MPC1211 + default y menu "Bus options (PCI, PCMCIA, EISA, MCA, ISA)" diff --git a/arch/sh/kernel/ptrace.c b/arch/sh/kernel/ptrace.c index 1b0dfb4d8ea..b28919b6568 100644 --- a/arch/sh/kernel/ptrace.c +++ b/arch/sh/kernel/ptrace.c @@ -20,6 +20,7 @@ #include <linux/user.h> #include <linux/slab.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/io.h> #include <asm/uaccess.h> @@ -197,7 +198,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -228,7 +229,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) struct pt_regs *dummy = NULL; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); if ((child->ptrace & PT_DTRACE) == 0) { diff --git a/arch/sh64/kernel/ptrace.c b/arch/sh64/kernel/ptrace.c index 800288c1562..fd2000956da 100644 --- a/arch/sh64/kernel/ptrace.c +++ b/arch/sh64/kernel/ptrace.c @@ -27,6 +27,7 @@ #include <linux/errno.h> #include <linux/ptrace.h> #include <linux/user.h> +#include <linux/signal.h> #include <asm/io.h> #include <asm/uaccess.h> @@ -255,7 +256,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -285,7 +286,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) struct pt_regs *regs; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); if ((child->ptrace & PT_DTRACE) == 0) { diff --git a/arch/sh64/kernel/sys_sh64.c b/arch/sh64/kernel/sys_sh64.c index 4546845b9ca..58ff7d522d8 100644 --- a/arch/sh64/kernel/sys_sh64.c +++ b/arch/sh64/kernel/sys_sh64.c @@ -283,18 +283,3 @@ asmlinkage int sys_uname(struct old_utsname * name) up_read(&uts_sem); return err?-EFAULT:0; } - -/* Copy from mips version */ -asmlinkage long sys_shmatcall(int shmid, char __user *shmaddr, - int shmflg) -{ - unsigned long raddr; - int err; - - err = do_shmat(shmid, shmaddr, shmflg, &raddr); - if (err) - return err; - - err = raddr; - return err; -} diff --git a/arch/sh64/kernel/syscalls.S b/arch/sh64/kernel/syscalls.S index 8ed417df3dc..6aabc63e451 100644 --- a/arch/sh64/kernel/syscalls.S +++ b/arch/sh64/kernel/syscalls.S @@ -268,7 +268,7 @@ sys_call_table: .long sys_msgrcv .long sys_msgget .long sys_msgctl - .long sys_shmatcall + .long sys_shmat .long sys_shmdt /* 245 */ .long sys_shmget .long sys_shmctl diff --git a/arch/sparc/kernel/process.c b/arch/sparc/kernel/process.c index 143fe2f3c1c..2c216ffeea9 100644 --- a/arch/sparc/kernel/process.c +++ b/arch/sparc/kernel/process.c @@ -83,9 +83,6 @@ void default_idle(void) */ void cpu_idle(void) { - if (current->pid != 0) - goto out; - /* endless idle loop with no priority at all */ for (;;) { if (ARCH_SUN4C_SUN4) { @@ -126,8 +123,6 @@ void cpu_idle(void) schedule(); check_pgt_cache(); } -out: - return; } #else @@ -333,6 +328,17 @@ void show_stack(struct task_struct *tsk, unsigned long *_ksp) printk("\n"); } +void dump_stack(void) +{ + unsigned long *ksp; + + __asm__ __volatile__("mov %%fp, %0" + : "=r" (ksp)); + show_stack(current, ksp); +} + +EXPORT_SYMBOL(dump_stack); + /* * Note: sparc64 has a pretty intricated thread_saved_pc, check it out. */ diff --git a/arch/sparc/kernel/ptrace.c b/arch/sparc/kernel/ptrace.c index c4f93bd2daf..475c4c13462 100644 --- a/arch/sparc/kernel/ptrace.c +++ b/arch/sparc/kernel/ptrace.c @@ -18,6 +18,7 @@ #include <linux/smp.h> #include <linux/smp_lock.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/pgtable.h> #include <asm/system.h> @@ -526,7 +527,7 @@ asmlinkage void do_ptrace(struct pt_regs *regs) addr = 1; case PTRACE_CONT: { /* restart after signal. */ - if (data > _NSIG) { + if (!valid_signal(data)) { pt_error_return(regs, EIO); goto out_tsk; } diff --git a/arch/sparc/kernel/sparc_ksyms.c b/arch/sparc/kernel/sparc_ksyms.c index f91b0e8d0dc..1bd430d0ca0 100644 --- a/arch/sparc/kernel/sparc_ksyms.c +++ b/arch/sparc/kernel/sparc_ksyms.c @@ -20,6 +20,7 @@ #include <linux/in6.h> #include <linux/spinlock.h> #include <linux/mm.h> +#include <linux/syscalls.h> #ifdef CONFIG_PCI #include <linux/pci.h> #endif @@ -89,6 +90,9 @@ extern void ___atomic24_sub(void); extern void ___set_bit(void); extern void ___clear_bit(void); extern void ___change_bit(void); +extern void ___rw_read_enter(void); +extern void ___rw_read_exit(void); +extern void ___rw_write_enter(void); /* Alias functions whose names begin with "." and export the aliases. * The module references will be fixed up by module_frob_arch_sections. @@ -121,9 +125,9 @@ EXPORT_SYMBOL(_do_write_unlock); #endif #else // XXX find what uses (or used) these. -// EXPORT_SYMBOL_PRIVATE(_rw_read_enter); -// EXPORT_SYMBOL_PRIVATE(_rw_read_exit); -// EXPORT_SYMBOL_PRIVATE(_rw_write_enter); +EXPORT_SYMBOL(___rw_read_enter); +EXPORT_SYMBOL(___rw_read_exit); +EXPORT_SYMBOL(___rw_write_enter); #endif /* semaphores */ EXPORT_SYMBOL(__up); @@ -144,6 +148,9 @@ EXPORT_SYMBOL(___set_bit); EXPORT_SYMBOL(___clear_bit); EXPORT_SYMBOL(___change_bit); +/* Per-CPU information table */ +EXPORT_PER_CPU_SYMBOL(__cpu_data); + #ifdef CONFIG_SMP /* IRQ implementation. */ EXPORT_SYMBOL(synchronize_irq); @@ -151,6 +158,10 @@ EXPORT_SYMBOL(synchronize_irq); /* Misc SMP information */ EXPORT_SYMBOL(__cpu_number_map); EXPORT_SYMBOL(__cpu_logical_map); + +/* CPU online map and active count. */ +EXPORT_SYMBOL(cpu_online_map); +EXPORT_SYMBOL(phys_cpu_present_map); #endif EXPORT_SYMBOL(__udelay); @@ -332,3 +343,6 @@ EXPORT_SYMBOL(do_BUG); /* Sun Power Management Idle Handler */ EXPORT_SYMBOL(pm_idle); + +/* Binfmt_misc needs this */ +EXPORT_SYMBOL(sys_close); diff --git a/arch/sparc/prom/memory.c b/arch/sparc/prom/memory.c index 46aa51afec1..c20e5309f8a 100644 --- a/arch/sparc/prom/memory.c +++ b/arch/sparc/prom/memory.c @@ -47,9 +47,9 @@ prom_sortmemlist(struct linux_mlist_v0 *thislist) char *tmpaddr; char *lowest; - for(i=0; thislist[i].theres_more != 0; i++) { + for(i=0; thislist[i].theres_more; i++) { lowest = thislist[i].start_adr; - for(mitr = i+1; thislist[mitr-1].theres_more != 0; mitr++) + for(mitr = i+1; thislist[mitr-1].theres_more; mitr++) if(thislist[mitr].start_adr < lowest) { lowest = thislist[mitr].start_adr; swapi = mitr; @@ -85,7 +85,7 @@ void __init prom_meminit(void) prom_phys_total[iter].num_bytes = mptr->num_bytes; prom_phys_total[iter].theres_more = &prom_phys_total[iter+1]; } - prom_phys_total[iter-1].theres_more = 0x0; + prom_phys_total[iter-1].theres_more = NULL; /* Second, the total prom taken descriptors. */ for(mptr = (*(romvec->pv_v0mem.v0_prommap)), iter=0; mptr; mptr=mptr->theres_more, iter++) { @@ -93,7 +93,7 @@ void __init prom_meminit(void) prom_prom_taken[iter].num_bytes = mptr->num_bytes; prom_prom_taken[iter].theres_more = &prom_prom_taken[iter+1]; } - prom_prom_taken[iter-1].theres_more = 0x0; + prom_prom_taken[iter-1].theres_more = NULL; /* Last, the available physical descriptors. */ for(mptr = (*(romvec->pv_v0mem.v0_available)), iter=0; mptr; mptr=mptr->theres_more, iter++) { @@ -101,7 +101,7 @@ void __init prom_meminit(void) prom_phys_avail[iter].num_bytes = mptr->num_bytes; prom_phys_avail[iter].theres_more = &prom_phys_avail[iter+1]; } - prom_phys_avail[iter-1].theres_more = 0x0; + prom_phys_avail[iter-1].theres_more = NULL; /* Sort all the lists. */ prom_sortmemlist(prom_phys_total); prom_sortmemlist(prom_prom_taken); @@ -124,7 +124,7 @@ void __init prom_meminit(void) prom_phys_avail[iter].theres_more = &prom_phys_avail[iter+1]; } - prom_phys_avail[iter-1].theres_more = 0x0; + prom_phys_avail[iter-1].theres_more = NULL; num_regs = prom_getproperty(node, "reg", (char *) prom_reg_memlist, @@ -138,7 +138,7 @@ void __init prom_meminit(void) prom_phys_total[iter].theres_more = &prom_phys_total[iter+1]; } - prom_phys_total[iter-1].theres_more = 0x0; + prom_phys_total[iter-1].theres_more = NULL; node = prom_getchild(prom_root_node); node = prom_searchsiblings(node, "virtual-memory"); @@ -158,7 +158,7 @@ void __init prom_meminit(void) prom_prom_taken[iter].theres_more = &prom_prom_taken[iter+1]; } - prom_prom_taken[iter-1].theres_more = 0x0; + prom_prom_taken[iter-1].theres_more = NULL; prom_sortmemlist(prom_prom_taken); @@ -182,15 +182,15 @@ void __init prom_meminit(void) case PROM_SUN4: #ifdef CONFIG_SUN4 /* how simple :) */ - prom_phys_total[0].start_adr = 0x0; + prom_phys_total[0].start_adr = NULL; prom_phys_total[0].num_bytes = *(sun4_romvec->memorysize); - prom_phys_total[0].theres_more = 0x0; - prom_prom_taken[0].start_adr = 0x0; + prom_phys_total[0].theres_more = NULL; + prom_prom_taken[0].start_adr = NULL; prom_prom_taken[0].num_bytes = 0x0; - prom_prom_taken[0].theres_more = 0x0; - prom_phys_avail[0].start_adr = 0x0; + prom_prom_taken[0].theres_more = NULL; + prom_phys_avail[0].start_adr = NULL; prom_phys_avail[0].num_bytes = *(sun4_romvec->memoryavail); - prom_phys_avail[0].theres_more = 0x0; + prom_phys_avail[0].theres_more = NULL; #endif break; diff --git a/arch/sparc/prom/sun4prom.c b/arch/sparc/prom/sun4prom.c index 69ca735f0d4..00390a2652a 100644 --- a/arch/sparc/prom/sun4prom.c +++ b/arch/sparc/prom/sun4prom.c @@ -151,7 +151,7 @@ struct linux_romvec * __init sun4_prom_init(void) * have more time, we can teach the penguin to say "By your * command" or "Activating turbo boost, Michael". :-) */ - sun4_romvec->setLEDs(0x0); + sun4_romvec->setLEDs(NULL); printk("PROMLIB: Old Sun4 boot PROM monitor %s, romvec version %d\n", sun4_romvec->monid, diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig index fb1189641c7..a72fd15d5ea 100644 --- a/arch/sparc64/Kconfig +++ b/arch/sparc64/Kconfig @@ -118,6 +118,7 @@ config VT_CONSOLE config HW_CONSOLE bool + depends on VT default y config SMP diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c index a38cb5036df..4dcb8af9409 100644 --- a/arch/sparc64/kernel/irq.c +++ b/arch/sparc64/kernel/irq.c @@ -756,7 +756,7 @@ void handler_irq(int irq, struct pt_regs *regs) clear_softint(clr_mask); } #else - int should_forward = 1; + int should_forward = 0; clear_softint(1 << irq); #endif @@ -1007,10 +1007,10 @@ static int retarget_one_irq(struct irqaction *p, int goal_cpu) } upa_writel(tid | IMAP_VALID, imap); - while (!cpu_online(goal_cpu)) { + do { if (++goal_cpu >= NR_CPUS) goal_cpu = 0; - } + } while (!cpu_online(goal_cpu)); return goal_cpu; } diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c index 26d3ec41da1..a0cd2b2494d 100644 --- a/arch/sparc64/kernel/process.c +++ b/arch/sparc64/kernel/process.c @@ -62,9 +62,6 @@ void default_idle(void) */ void cpu_idle(void) { - if (current->pid != 0) - return; - /* endless idle loop with no priority at all */ for (;;) { /* If current->work.need_resched is zero we should really @@ -80,7 +77,6 @@ void cpu_idle(void) schedule(); check_pgt_cache(); } - return; } #else diff --git a/arch/sparc64/kernel/ptrace.c b/arch/sparc64/kernel/ptrace.c index 5f080cf04b3..80a76e2ad73 100644 --- a/arch/sparc64/kernel/ptrace.c +++ b/arch/sparc64/kernel/ptrace.c @@ -19,6 +19,7 @@ #include <linux/smp.h> #include <linux/smp_lock.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/asi.h> #include <asm/pgtable.h> @@ -559,7 +560,7 @@ asmlinkage void do_ptrace(struct pt_regs *regs) addr = 1; case PTRACE_CONT: { /* restart after signal. */ - if (data > _NSIG) { + if (!valid_signal(data)) { pt_error_return(regs, EIO); goto out_tsk; } diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c index cad5a112280..e78cc53594f 100644 --- a/arch/sparc64/kernel/sparc64_ksyms.c +++ b/arch/sparc64/kernel/sparc64_ksyms.c @@ -278,7 +278,7 @@ EXPORT_SYMBOL(verify_compat_iovec); EXPORT_SYMBOL(dump_thread); EXPORT_SYMBOL(dump_fpu); -EXPORT_SYMBOL(__pte_alloc_one_kernel); +EXPORT_SYMBOL(pte_alloc_one_kernel); #ifndef CONFIG_SMP EXPORT_SYMBOL(pgt_quicklists); #endif diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c index 6a717d4d2bc..71b4e380769 100644 --- a/arch/sparc64/kernel/time.c +++ b/arch/sparc64/kernel/time.c @@ -48,7 +48,7 @@ DEFINE_SPINLOCK(mostek_lock); DEFINE_SPINLOCK(rtc_lock); -unsigned long mstk48t02_regs = 0UL; +void __iomem *mstk48t02_regs = NULL; #ifdef CONFIG_PCI unsigned long ds1287_regs = 0UL; #endif @@ -59,8 +59,8 @@ u64 jiffies_64 = INITIAL_JIFFIES; EXPORT_SYMBOL(jiffies_64); -static unsigned long mstk48t08_regs = 0UL; -static unsigned long mstk48t59_regs = 0UL; +static void __iomem *mstk48t08_regs; +static void __iomem *mstk48t59_regs; static int set_rtc_mmss(unsigned long); @@ -520,7 +520,7 @@ void timer_tick_interrupt(struct pt_regs *regs) /* Kick start a stopped clock (procedure from the Sun NVRAM/hostid FAQ). */ static void __init kick_start_clock(void) { - unsigned long regs = mstk48t02_regs; + void __iomem *regs = mstk48t02_regs; u8 sec, tmp; int i, count; @@ -604,7 +604,7 @@ static void __init kick_start_clock(void) /* Return nonzero if the clock chip battery is low. */ static int __init has_low_battery(void) { - unsigned long regs = mstk48t02_regs; + void __iomem *regs = mstk48t02_regs; u8 data1, data2; spin_lock_irq(&mostek_lock); @@ -623,7 +623,7 @@ static int __init has_low_battery(void) static void __init set_system_time(void) { unsigned int year, mon, day, hour, min, sec; - unsigned long mregs = mstk48t02_regs; + void __iomem *mregs = mstk48t02_regs; #ifdef CONFIG_PCI unsigned long dregs = ds1287_regs; #else @@ -843,7 +843,8 @@ void __init clock_probe(void) !strcmp(model, "m5823")) { ds1287_regs = edev->resource[0].start; } else { - mstk48t59_regs = edev->resource[0].start; + mstk48t59_regs = (void __iomem *) + edev->resource[0].start; mstk48t02_regs = mstk48t59_regs + MOSTEK_48T59_48T02; } break; @@ -865,7 +866,8 @@ try_isa_clock: !strcmp(model, "m5823")) { ds1287_regs = isadev->resource.start; } else { - mstk48t59_regs = isadev->resource.start; + mstk48t59_regs = (void __iomem *) + isadev->resource.start; mstk48t02_regs = mstk48t59_regs + MOSTEK_48T59_48T02; } break; @@ -893,21 +895,24 @@ try_isa_clock: } if(model[5] == '0' && model[6] == '2') { - mstk48t02_regs = (((u64)clk_reg[0].phys_addr) | - (((u64)clk_reg[0].which_io)<<32UL)); + mstk48t02_regs = (void __iomem *) + (((u64)clk_reg[0].phys_addr) | + (((u64)clk_reg[0].which_io)<<32UL)); } else if(model[5] == '0' && model[6] == '8') { - mstk48t08_regs = (((u64)clk_reg[0].phys_addr) | - (((u64)clk_reg[0].which_io)<<32UL)); + mstk48t08_regs = (void __iomem *) + (((u64)clk_reg[0].phys_addr) | + (((u64)clk_reg[0].which_io)<<32UL)); mstk48t02_regs = mstk48t08_regs + MOSTEK_48T08_48T02; } else { - mstk48t59_regs = (((u64)clk_reg[0].phys_addr) | - (((u64)clk_reg[0].which_io)<<32UL)); + mstk48t59_regs = (void __iomem *) + (((u64)clk_reg[0].phys_addr) | + (((u64)clk_reg[0].which_io)<<32UL)); mstk48t02_regs = mstk48t59_regs + MOSTEK_48T59_48T02; } break; } - if (mstk48t02_regs != 0UL) { + if (mstk48t02_regs != NULL) { /* Report a low battery voltage condition. */ if (has_low_battery()) prom_printf("NVRAM: Low battery voltage!\n"); @@ -1087,7 +1092,7 @@ unsigned long long sched_clock(void) static int set_rtc_mmss(unsigned long nowtime) { int real_seconds, real_minutes, chip_minutes; - unsigned long mregs = mstk48t02_regs; + void __iomem *mregs = mstk48t02_regs; #ifdef CONFIG_PCI unsigned long dregs = ds1287_regs; #else diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index db6fa77b4da..9c5222075da 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -1114,7 +1114,7 @@ struct pgtable_cache_struct pgt_quicklists; #else #define DC_ALIAS_SHIFT 0 #endif -pte_t *__pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) +pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { struct page *page; unsigned long color; diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 9a23df18212..c5292181a66 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -244,6 +244,7 @@ config KERNEL_HALF_GIGS config HIGHMEM bool "Highmem support" + depends on !64BIT config KERNEL_STACK_ORDER int "Kernel stack size order" diff --git a/arch/um/Kconfig_i386 b/arch/um/Kconfig_i386 index 203c242201b..e41f3748d30 100644 --- a/arch/um/Kconfig_i386 +++ b/arch/um/Kconfig_i386 @@ -1,4 +1,8 @@ -config 64_BIT +config UML_X86 + bool + default y + +config 64BIT bool default n diff --git a/arch/um/Kconfig_x86_64 b/arch/um/Kconfig_x86_64 index 768dc6626a8..fd8d7e8982b 100644 --- a/arch/um/Kconfig_x86_64 +++ b/arch/um/Kconfig_x86_64 @@ -1,4 +1,8 @@ -config 64_BIT +config UML_X86 + bool + default y + +config 64BIT bool default y diff --git a/arch/um/Makefile b/arch/um/Makefile index 97bca6b5ca9..f2a0c40a920 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -17,7 +17,7 @@ core-y += $(ARCH_DIR)/kernel/ \ # Have to precede the include because the included Makefiles reference them. SYMLINK_HEADERS := archparam.h system.h sigcontext.h processor.h ptrace.h \ - arch-signal.h module.h vm-flags.h + module.h vm-flags.h elf.h SYMLINK_HEADERS := $(foreach header,$(SYMLINK_HEADERS),include/asm-um/$(header)) # XXX: The "os" symlink is only used by arch/um/include/os.h, which includes @@ -44,6 +44,11 @@ ifneq ($(MAKEFILES-INCL),) endif ARCH_INCLUDE := -I$(ARCH_DIR)/include +ifneq ($(KBUILD_SRC),) +ARCH_INCLUDE += -I$(ARCH_DIR)/include2 +ARCH_INCLUDE += -I$(srctree)/$(ARCH_DIR)/include +MRPROPER_DIRS += $(ARCH_DIR)/include2 +endif SYS_DIR := $(ARCH_DIR)/include/sysdep-$(SUBARCH) include $(srctree)/$(ARCH_DIR)/Makefile-$(SUBARCH) @@ -94,17 +99,18 @@ define archhelp echo ' find in the kernel root.' endef +ifneq ($(KBUILD_SRC),) +$(shell mkdir -p $(ARCH_DIR) && ln -fsn $(srctree)/$(ARCH_DIR)/Kconfig_$(SUBARCH) $(ARCH_DIR)/Kconfig_arch) +CLEAN_FILES += $(ARCH_DIR)/Kconfig_arch +else $(shell cd $(ARCH_DIR) && ln -sf Kconfig_$(SUBARCH) Kconfig_arch) +endif -prepare: $(ARCH_SYMLINKS) $(SYS_HEADERS) $(GEN_HEADERS) \ - $(ARCH_DIR)/kernel/vmlinux.lds.S +prepare: $(ARCH_SYMLINKS) $(SYS_HEADERS) $(GEN_HEADERS) LINK-$(CONFIG_LD_SCRIPT_STATIC) += -static LINK-$(CONFIG_LD_SCRIPT_DYN) += -Wl,-rpath,/lib -LD_SCRIPT-$(CONFIG_LD_SCRIPT_STATIC) := uml.lds.S -LD_SCRIPT-$(CONFIG_LD_SCRIPT_DYN) := dyn.lds.S - CPP_MODE-$(CONFIG_MODE_TT) := -DMODE_TT CONFIG_KERNEL_STACK_ORDER ?= 2 STACK_SIZE := $(shell echo $$[ 4096 * (1 << $(CONFIG_KERNEL_STACK_ORDER)) ] ) @@ -126,7 +132,7 @@ define cmd_vmlinux__ $(CC) $(CFLAGS_vmlinux) -o $@ \ -Wl,-T,$(vmlinux-lds) $(vmlinux-init) \ -Wl,--start-group $(vmlinux-main) -Wl,--end-group \ - -L/usr/lib -lutil \ + -lutil \ $(filter-out $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) \ FORCE ,$^) ; rm -f linux endef @@ -145,31 +151,42 @@ archclean: @find . \( -name '*.bb' -o -name '*.bbg' -o -name '*.da' \ -o -name '*.gcov' \) -type f -print | xargs rm -f -#We need to re-preprocess this when the symlink dest changes. -#So we touch it when needed. -$(ARCH_DIR)/kernel/vmlinux.lds.S: FORCE - $(Q)if [ "$(shell readlink $@)" != "$(LD_SCRIPT-y)" ]; then \ - echo ' SYMLINK $@'; \ - ln -sf $(LD_SCRIPT-y) $@; \ - touch $@; \ - fi; - $(SYMLINK_HEADERS): @echo ' SYMLINK $@' +ifneq ($(KBUILD_SRC),) + ln -fsn $(srctree)/include/asm-um/$(basename $(notdir $@))-$(SUBARCH)$(suffix $@) $@ +else $(Q)cd $(TOPDIR)/$(dir $@) ; \ ln -sf $(basename $(notdir $@))-$(SUBARCH)$(suffix $@) $(notdir $@) +endif include/asm-um/arch: @echo ' SYMLINK $@' +ifneq ($(KBUILD_SRC),) + $(Q)mkdir -p include/asm-um + $(Q)ln -fsn $(srctree)/include/asm-$(SUBARCH) include/asm-um/arch +else $(Q)cd $(TOPDIR)/include/asm-um && ln -sf ../asm-$(SUBARCH) arch +endif $(ARCH_DIR)/include/sysdep: @echo ' SYMLINK $@' +ifneq ($(KBUILD_SRC),) + $(Q)mkdir -p $(ARCH_DIR)/include + $(Q)mkdir -p $(ARCH_DIR)/include2 + $(Q)ln -fsn sysdep-$(SUBARCH) $(ARCH_DIR)/include/sysdep + $(Q)ln -fsn $(srctree)/$(ARCH_DIR)/include/sysdep-$(SUBARCH) $(ARCH_DIR)/include2/sysdep +else $(Q)cd $(ARCH_DIR)/include && ln -sf sysdep-$(SUBARCH) sysdep +endif $(ARCH_DIR)/os: @echo ' SYMLINK $@' +ifneq ($(KBUILD_SRC),) + $(Q)ln -fsn $(srctree)/$(ARCH_DIR)/os-$(OS) $(ARCH_DIR)/os +else $(Q)cd $(ARCH_DIR) && ln -sf os-$(OS) os +endif # Generated files define filechk_umlconfig @@ -179,10 +196,31 @@ endef $(ARCH_DIR)/include/uml-config.h : include/linux/autoconf.h $(call filechk,umlconfig) +$(ARCH_DIR)/user-offsets.s: $(ARCH_DIR)/sys-$(SUBARCH)/user-offsets.c + $(CC) $(USER_CFLAGS) -S -o $@ $< + +$(ARCH_DIR)/user-offsets.h: $(ARCH_DIR)/user-offsets.s + $(call filechk,gen-asm-offsets) + +CLEAN_FILES += $(ARCH_DIR)/user-offsets.s $(ARCH_DIR)/user-offsets.h + +$(ARCH_DIR)/kernel-offsets.s: $(ARCH_DIR)/sys-$(SUBARCH)/kernel-offsets.c \ + $(ARCH_SYMLINKS) \ + $(SYS_DIR)/sc.h \ + include/asm include/linux/version.h \ + include/config/MARKER \ + $(ARCH_DIR)/include/user_constants.h + $(CC) $(CFLAGS) $(NOSTDINC_FLAGS) $(CPPFLAGS) -S -o $@ $< + +$(ARCH_DIR)/kernel-offsets.h: $(ARCH_DIR)/kernel-offsets.s + $(call filechk,gen-asm-offsets) + +CLEAN_FILES += $(ARCH_DIR)/kernel-offsets.s $(ARCH_DIR)/kernel-offsets.h + $(ARCH_DIR)/include/task.h: $(ARCH_DIR)/util/mk_task $(call filechk,gen_header) -$(ARCH_DIR)/include/user_constants.h: $(ARCH_DIR)/os/util/mk_user_constants +$(ARCH_DIR)/include/user_constants.h: $(ARCH_DIR)/os-$(OS)/util/mk_user_constants $(call filechk,gen_header) $(ARCH_DIR)/include/kern_constants.h: $(ARCH_DIR)/util/mk_constants @@ -191,20 +229,20 @@ $(ARCH_DIR)/include/kern_constants.h: $(ARCH_DIR)/util/mk_constants $(ARCH_DIR)/include/skas_ptregs.h: $(ARCH_DIR)/kernel/skas/util/mk_ptregs $(call filechk,gen_header) -$(ARCH_DIR)/os/util/mk_user_constants: $(ARCH_DIR)/os/util FORCE ; +$(ARCH_DIR)/os-$(OS)/util/mk_user_constants: $(ARCH_DIR)/os-$(OS)/util FORCE ; $(ARCH_DIR)/util/mk_task $(ARCH_DIR)/util/mk_constants: $(ARCH_DIR)/include/user_constants.h $(ARCH_DIR)/util \ FORCE ; $(ARCH_DIR)/kernel/skas/util/mk_ptregs: $(ARCH_DIR)/kernel/skas/util FORCE ; -$(ARCH_DIR)/util: scripts_basic $(SYS_DIR)/sc.h FORCE +$(ARCH_DIR)/util: scripts_basic $(SYS_DIR)/sc.h $(ARCH_DIR)/kernel-offsets.h FORCE $(Q)$(MAKE) $(build)=$@ -$(ARCH_DIR)/kernel/skas/util: scripts_basic FORCE +$(ARCH_DIR)/kernel/skas/util: scripts_basic $(ARCH_DIR)/user-offsets.h FORCE $(Q)$(MAKE) $(build)=$@ -$(ARCH_DIR)/os/util: scripts_basic FORCE +$(ARCH_DIR)/os-$(OS)/util: scripts_basic FORCE $(Q)$(MAKE) $(build)=$@ export SUBARCH USER_CFLAGS OS diff --git a/arch/um/Makefile-i386 b/arch/um/Makefile-i386 index 97b223bfa78..29e182d5a83 100644 --- a/arch/um/Makefile-i386 +++ b/arch/um/Makefile-i386 @@ -1,4 +1,4 @@ -SUBARCH_CORE := arch/um/sys-i386/ +SUBARCH_CORE := arch/um/sys-i386/ arch/i386/crypto/ TOP_ADDR := $(CONFIG_TOP_ADDR) @@ -32,10 +32,10 @@ $(SYS_DIR)/sc.h: $(SYS_UTIL_DIR)/mk_sc $(SYS_DIR)/thread.h: $(SYS_UTIL_DIR)/mk_thread $(call filechk,gen_header) -$(SYS_UTIL_DIR)/mk_sc: scripts_basic FORCE +$(SYS_UTIL_DIR)/mk_sc: scripts_basic $(ARCH_DIR)/user-offsets.h FORCE $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) $@ -$(SYS_UTIL_DIR)/mk_thread: scripts_basic $(ARCH_SYMLINKS) $(GEN_HEADERS) FORCE +$(SYS_UTIL_DIR)/mk_thread: scripts_basic $(ARCH_DIR)/kernel-offsets.h FORCE $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) $@ $(SYS_UTIL_DIR): scripts_basic include/asm FORCE diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64 index a77971133e9..32144562c27 100644 --- a/arch/um/Makefile-x86_64 +++ b/arch/um/Makefile-x86_64 @@ -23,10 +23,10 @@ $(SYS_DIR)/sc.h: $(SYS_UTIL_DIR)/mk_sc $(SYS_DIR)/thread.h: $(SYS_UTIL_DIR)/mk_thread $(call filechk,gen_header) -$(SYS_UTIL_DIR)/mk_sc: scripts_basic FORCE +$(SYS_UTIL_DIR)/mk_sc: scripts_basic $(ARCH_DIR)/user-offsets.h FORCE $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) $@ -$(SYS_UTIL_DIR)/mk_thread: scripts_basic $(ARCH_SYMLINKS) $(GEN_HEADERS) FORCE +$(SYS_UTIL_DIR)/mk_thread: scripts_basic $(GEN_HEADERS) $(ARCH_DIR)/kernel-offsets.h FORCE $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) $@ CLEAN_FILES += $(SYS_HEADERS) diff --git a/arch/um/defconfig b/arch/um/defconfig index fc3075c589d..4067c3aa5b6 100644 --- a/arch/um/defconfig +++ b/arch/um/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.12-rc1-bk1 -# Sun Mar 20 16:53:00 2005 +# Linux kernel version: 2.6.12-rc3-skas3-v9-pre2 +# Sun Apr 24 19:46:10 2005 # CONFIG_GENERIC_HARDIRQS=y CONFIG_UML=y @@ -15,7 +15,8 @@ CONFIG_GENERIC_CALIBRATE_DELAY=y # CONFIG_MODE_TT=y CONFIG_MODE_SKAS=y -# CONFIG_64_BIT is not set +CONFIG_UML_X86=y +# CONFIG_64BIT is not set CONFIG_TOP_ADDR=0xc0000000 # CONFIG_3_LEVEL_PGTABLES is not set CONFIG_ARCH_HAS_SC_SIGNALS=y @@ -41,6 +42,7 @@ CONFIG_UML_REAL_TIME_CLOCK=y CONFIG_EXPERIMENTAL=y CONFIG_CLEAN_COMPILE=y CONFIG_BROKEN_ON_SMP=y +CONFIG_INIT_ENV_ARG_LIMIT=32 # # General setup @@ -158,7 +160,6 @@ CONFIG_UML_NET_SLIRP=y # CONFIG_PACKET=y CONFIG_PACKET_MMAP=y -# CONFIG_NETLINK_DEV is not set CONFIG_UNIX=y # CONFIG_NET_KEY is not set CONFIG_INET=y @@ -412,6 +413,5 @@ CONFIG_DEBUG_INFO=y # CONFIG_DEBUG_FS is not set CONFIG_FRAME_POINTER=y CONFIG_PT_PROXY=y -# CONFIG_GPROF is not set # CONFIG_GCOV is not set # CONFIG_SYSCALL_DEBUG is not set diff --git a/arch/um/drivers/chan_kern.c b/arch/um/drivers/chan_kern.c index 1f77deb3fd2..0150038af79 100644 --- a/arch/um/drivers/chan_kern.c +++ b/arch/um/drivers/chan_kern.c @@ -22,7 +22,7 @@ #ifdef CONFIG_NOCONFIG_CHAN static void *not_configged_init(char *str, int device, struct chan_opts *opts) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(NULL); } @@ -30,27 +30,27 @@ static void *not_configged_init(char *str, int device, struct chan_opts *opts) static int not_configged_open(int input, int output, int primary, void *data, char **dev_out) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-ENODEV); } static void not_configged_close(int fd, void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); } static int not_configged_read(int fd, char *c_out, void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-EIO); } static int not_configged_write(int fd, const char *buf, int len, void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-EIO); } @@ -58,7 +58,7 @@ static int not_configged_write(int fd, const char *buf, int len, void *data) static int not_configged_console_write(int fd, const char *buf, int len, void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-EIO); } @@ -66,14 +66,14 @@ static int not_configged_console_write(int fd, const char *buf, int len, static int not_configged_window_size(int fd, void *data, unsigned short *rows, unsigned short *cols) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-ENODEV); } static void not_configged_free(void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); } diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 6924f273ced..025d3be8aca 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -39,19 +39,69 @@ static void line_timer_cb(void *arg) line_interrupt(line->driver->read_irq, arg, NULL); } -static int write_room(struct line *dev) +/* Returns the free space inside the ring buffer of this line. + * + * Should be called while holding line->lock (this does not modify datas). + */ +static int write_room(struct line *line) { int n; - if (dev->buffer == NULL) - return (LINE_BUFSIZE - 1); + if (line->buffer == NULL) + return LINE_BUFSIZE - 1; + + /* This is for the case where the buffer is wrapped! */ + n = line->head - line->tail; - n = dev->head - dev->tail; if (n <= 0) - n = LINE_BUFSIZE + n; - return (n - 1); + n = LINE_BUFSIZE + n; /* The other case */ + return n - 1; +} + +int line_write_room(struct tty_struct *tty) +{ + struct line *line = tty->driver_data; + unsigned long flags; + int room; + + if (tty->stopped) + return 0; + + spin_lock_irqsave(&line->lock, flags); + room = write_room(line); + spin_unlock_irqrestore(&line->lock, flags); + + /*XXX: Warning to remove */ + if (0 == room) + printk(KERN_DEBUG "%s: %s: no room left in buffer\n", + __FUNCTION__,tty->name); + return room; +} + +int line_chars_in_buffer(struct tty_struct *tty) +{ + struct line *line = tty->driver_data; + unsigned long flags; + int ret; + + spin_lock_irqsave(&line->lock, flags); + + /*write_room subtracts 1 for the needed NULL, so we readd it.*/ + ret = LINE_BUFSIZE - (write_room(line) + 1); + spin_unlock_irqrestore(&line->lock, flags); + + return ret; } +/* + * This copies the content of buf into the circular buffer associated with + * this line. + * The return value is the number of characters actually copied, i.e. the ones + * for which there was space: this function is not supposed to ever flush out + * the circular buffer. + * + * Must be called while holding line->lock! + */ static int buffer_data(struct line *line, const char *buf, int len) { int end, room; @@ -70,48 +120,95 @@ static int buffer_data(struct line *line, const char *buf, int len) len = (len > room) ? room : len; end = line->buffer + LINE_BUFSIZE - line->tail; - if(len < end){ + + if (len < end){ memcpy(line->tail, buf, len); line->tail += len; - } - else { + } else { + /* The circular buffer is wrapping */ memcpy(line->tail, buf, end); buf += end; memcpy(line->buffer, buf, len - end); line->tail = line->buffer + len - end; } - return(len); + return len; } +/* + * Flushes the ring buffer to the output channels. That is, write_chan is + * called, passing it line->head as buffer, and an appropriate count. + * + * On exit, returns 1 when the buffer is empty, + * 0 when the buffer is not empty on exit, + * and -errno when an error occurred. + * + * Must be called while holding line->lock!*/ static int flush_buffer(struct line *line) { int n, count; if ((line->buffer == NULL) || (line->head == line->tail)) - return(1); + return 1; if (line->tail < line->head) { + /* line->buffer + LINE_BUFSIZE is the end of the buffer! */ count = line->buffer + LINE_BUFSIZE - line->head; + n = write_chan(&line->chan_list, line->head, count, line->driver->write_irq); if (n < 0) - return(n); - if (n == count) + return n; + if (n == count) { + /* We have flushed from ->head to buffer end, now we + * must flush only from the beginning to ->tail.*/ line->head = line->buffer; - else { + } else { line->head += n; - return(0); + return 0; } } count = line->tail - line->head; n = write_chan(&line->chan_list, line->head, count, line->driver->write_irq); - if(n < 0) return(n); + + if(n < 0) + return n; line->head += n; - return(line->head == line->tail); + return line->head == line->tail; +} + +void line_flush_buffer(struct tty_struct *tty) +{ + struct line *line = tty->driver_data; + unsigned long flags; + int err; + + /*XXX: copied from line_write, verify if it is correct!*/ + if(tty->stopped) + return; + //return 0; + + spin_lock_irqsave(&line->lock, flags); + err = flush_buffer(line); + /*if (err == 1) + err = 0;*/ + spin_unlock_irqrestore(&line->lock, flags); + //return err; +} + +/* We map both ->flush_chars and ->put_char (which go in pair) onto ->flush_buffer + * and ->write. Hope it's not that bad.*/ +void line_flush_chars(struct tty_struct *tty) +{ + line_flush_buffer(tty); +} + +void line_put_char(struct tty_struct *tty, unsigned char ch) +{ + line_write(tty, &ch, sizeof(ch)); } int line_write(struct tty_struct *tty, const unsigned char *buf, int len) @@ -120,38 +217,31 @@ int line_write(struct tty_struct *tty, const unsigned char *buf, int len) unsigned long flags; int n, err, ret = 0; - if(tty->stopped) return 0; + if(tty->stopped) + return 0; - down(&line->sem); - if(line->head != line->tail){ - local_irq_save(flags); + spin_lock_irqsave(&line->lock, flags); + if (line->head != line->tail) { ret = buffer_data(line, buf, len); err = flush_buffer(line); - local_irq_restore(flags); - if(err <= 0 && (err != -EAGAIN || !ret)) + if (err <= 0 && (err != -EAGAIN || !ret)) ret = err; - } - else { + } else { n = write_chan(&line->chan_list, buf, len, line->driver->write_irq); - if(n < 0){ + if (n < 0) { ret = n; goto out_up; } len -= n; ret += n; - if(len > 0) + if (len > 0) ret += buffer_data(line, buf + n, len); } - out_up: - up(&line->sem); - return(ret); -} - -void line_put_char(struct tty_struct *tty, unsigned char ch) -{ - line_write(tty, &ch, sizeof(ch)); +out_up: + spin_unlock_irqrestore(&line->lock, flags); + return ret; } void line_set_termios(struct tty_struct *tty, struct termios * old) @@ -159,11 +249,6 @@ void line_set_termios(struct tty_struct *tty, struct termios * old) /* nothing */ } -int line_chars_in_buffer(struct tty_struct *tty) -{ - return 0; -} - static struct { int cmd; char *level; @@ -250,7 +335,7 @@ int line_ioctl(struct tty_struct *tty, struct file * file, ret = -ENOIOCTLCMD; break; } - return(ret); + return ret; } static irqreturn_t line_write_interrupt(int irq, void *data, @@ -260,18 +345,23 @@ static irqreturn_t line_write_interrupt(int irq, void *data, struct line *line = tty->driver_data; int err; + /* Interrupts are enabled here because we registered the interrupt with + * SA_INTERRUPT (see line_setup_irq).*/ + + spin_lock_irq(&line->lock); err = flush_buffer(line); - if(err == 0) - return(IRQ_NONE); - else if(err < 0){ + if (err == 0) { + return IRQ_NONE; + } else if(err < 0) { line->head = line->buffer; line->tail = line->buffer; } + spin_unlock_irq(&line->lock); if(tty == NULL) - return(IRQ_NONE); + return IRQ_NONE; - if(test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags) && + if (test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags) && (tty->ldisc.write_wakeup != NULL)) (tty->ldisc.write_wakeup)(tty); @@ -281,9 +371,9 @@ static irqreturn_t line_write_interrupt(int irq, void *data, * writes. */ - if(waitqueue_active(&tty->write_wait)) + if (waitqueue_active(&tty->write_wait)) wake_up_interruptible(&tty->write_wait); - return(IRQ_HANDLED); + return IRQ_HANDLED; } int line_setup_irq(int fd, int input, int output, struct tty_struct *tty) @@ -292,15 +382,18 @@ int line_setup_irq(int fd, int input, int output, struct tty_struct *tty) struct line_driver *driver = line->driver; int err = 0, flags = SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM; - if(input) err = um_request_irq(driver->read_irq, fd, IRQ_READ, + if (input) + err = um_request_irq(driver->read_irq, fd, IRQ_READ, line_interrupt, flags, driver->read_irq_name, tty); - if(err) return(err); - if(output) err = um_request_irq(driver->write_irq, fd, IRQ_WRITE, + if (err) + return err; + if (output) + err = um_request_irq(driver->write_irq, fd, IRQ_WRITE, line_write_interrupt, flags, driver->write_irq_name, tty); line->have_irq = 1; - return(err); + return err; } void line_disable(struct tty_struct *tty, int current_irq) @@ -336,7 +429,9 @@ int line_open(struct line *lines, struct tty_struct *tty, line = &lines[tty->index]; tty->driver_data = line; - down(&line->sem); + /* The IRQ which takes this lock is not yet enabled and won't be run + * before the end, so we don't need to use spin_lock_irq.*/ + spin_lock(&line->lock); if (tty->count == 1) { if (!line->valid) { err = -ENODEV; @@ -349,6 +444,7 @@ int line_open(struct line *lines, struct tty_struct *tty, err = open_chan(&line->chan_list); if(err) goto out; } + /* Here the interrupt is registered.*/ enable_chan(&line->chan_list, tty); INIT_WORK(&line->task, line_timer_cb, tty); } @@ -362,21 +458,36 @@ int line_open(struct line *lines, struct tty_struct *tty, line->count++; out: - up(&line->sem); - return(err); + spin_unlock(&line->lock); + return err; } +static void unregister_winch(struct tty_struct *tty); + void line_close(struct tty_struct *tty, struct file * filp) { struct line *line = tty->driver_data; - down(&line->sem); + /* XXX: I assume this should be called in process context, not with + * interrupts disabled! + */ + spin_lock_irq(&line->lock); + + /* We ignore the error anyway! */ + flush_buffer(line); + line->count--; if (tty->count == 1) { line_disable(tty, -1); tty->driver_data = NULL; } - up(&line->sem); + + if((line->count == 0) && line->sigio){ + unregister_winch(tty); + line->sigio = 0; + } + + spin_unlock_irq(&line->lock); } void close_lines(struct line *lines, int nlines) @@ -387,31 +498,41 @@ void close_lines(struct line *lines, int nlines) close_chan(&lines[i].chan_list); } -int line_setup(struct line *lines, int num, char *init, int all_allowed) +/* Common setup code for both startup command line and mconsole initialization. + * @lines contains the the array (of size @num) to modify; + * @init is the setup string; + * @all_allowed is a boolean saying if we can setup the whole @lines + * at once. For instance, it will be usually true for startup init. (where we + * can use con=xterm) and false for mconsole.*/ + +int line_setup(struct line *lines, unsigned int num, char *init, int all_allowed) { int i, n; char *end; - if(*init == '=') n = -1; - else { + if(*init == '=') { + /* We said con=/ssl= instead of con#=, so we are configuring all + * consoles at once.*/ + n = -1; + } else { n = simple_strtoul(init, &end, 0); if(*end != '='){ printk(KERN_ERR "line_setup failed to parse \"%s\"\n", init); - return(0); + return 0; } init = end; } init++; - if((n >= 0) && (n >= num)){ + + if (n >= (signed int) num) { printk("line_setup - %d out of range ((0 ... %d) allowed)\n", n, num - 1); - return(0); - } - else if (n >= 0){ + return 0; + } else if (n >= 0){ if (lines[n].count > 0) { printk("line_setup - device %d is open\n", n); - return(0); + return 0; } if (lines[n].init_pri <= INIT_ONE){ lines[n].init_pri = INIT_ONE; @@ -422,13 +543,11 @@ int line_setup(struct line *lines, int num, char *init, int all_allowed) lines[n].valid = 1; } } - } - else if(!all_allowed){ + } else if(!all_allowed){ printk("line_setup - can't configure all devices from " "mconsole\n"); - return(0); - } - else { + return 0; + } else { for(i = 0; i < num; i++){ if(lines[i].init_pri <= INIT_ALL){ lines[i].init_pri = INIT_ALL; @@ -440,21 +559,21 @@ int line_setup(struct line *lines, int num, char *init, int all_allowed) } } } - return(1); + return 1; } -int line_config(struct line *lines, int num, char *str) +int line_config(struct line *lines, unsigned int num, char *str) { char *new = uml_strdup(str); if(new == NULL){ printk("line_config - uml_strdup failed\n"); - return(-ENOMEM); + return -ENOMEM; } - return(!line_setup(lines, num, new, 0)); + return !line_setup(lines, num, new, 0); } -int line_get_config(char *name, struct line *lines, int num, char *str, +int line_get_config(char *name, struct line *lines, unsigned int num, char *str, int size, char **error_out) { struct line *line; @@ -464,47 +583,33 @@ int line_get_config(char *name, struct line *lines, int num, char *str, dev = simple_strtoul(name, &end, 0); if((*end != '\0') || (end == name)){ *error_out = "line_get_config failed to parse device number"; - return(0); + return 0; } if((dev < 0) || (dev >= num)){ - *error_out = "device number of of range"; - return(0); + *error_out = "device number out of range"; + return 0; } line = &lines[dev]; - down(&line->sem); + spin_lock(&line->lock); if(!line->valid) CONFIG_CHUNK(str, size, n, "none", 1); else if(line->count == 0) CONFIG_CHUNK(str, size, n, line->init_str, 1); else n = chan_config_string(&line->chan_list, str, size, error_out); - up(&line->sem); + spin_unlock(&line->lock); - return(n); + return n; } -int line_remove(struct line *lines, int num, char *str) +int line_remove(struct line *lines, unsigned int num, char *str) { char config[sizeof("conxxxx=none\0")]; sprintf(config, "%s=none", str); - return(!line_setup(lines, num, config, 0)); -} - -int line_write_room(struct tty_struct *tty) -{ - struct line *dev = tty->driver_data; - int room; - - if (tty->stopped) - return 0; - room = write_room(dev); - if (0 == room) - printk(KERN_DEBUG "%s: %s: no room left in buffer\n", - __FUNCTION__,tty->name); - return room; + return !line_setup(lines, num, config, 0); } struct tty_driver *line_register_devfs(struct lines *set, @@ -553,7 +658,7 @@ void lines_init(struct line *lines, int nlines) for(i = 0; i < nlines; i++){ line = &lines[i]; INIT_LIST_HEAD(&line->chan_list); - sema_init(&line->sem, 1); + spin_lock_init(&line->lock); if(line->init_str != NULL){ line->init_str = uml_strdup(line->init_str); if(line->init_str == NULL) @@ -587,7 +692,7 @@ irqreturn_t winch_interrupt(int irq, void *data, struct pt_regs *unused) "errno = %d\n", -err); printk("fd %d is losing SIGWINCH support\n", winch->tty_fd); - return(IRQ_HANDLED); + return IRQ_HANDLED; } goto out; } @@ -603,7 +708,7 @@ irqreturn_t winch_interrupt(int irq, void *data, struct pt_regs *unused) out: if(winch->fd != -1) reactivate_fd(winch->fd, WINCH_IRQ); - return(IRQ_HANDLED); + return IRQ_HANDLED; } DECLARE_MUTEX(winch_handler_sem); @@ -625,7 +730,7 @@ void register_winch_irq(int fd, int tty_fd, int pid, struct tty_struct *tty) .pid = pid, .tty = tty }); list_add(&winch->list, &winch_handlers); - if(um_request_irq(WINCH_IRQ, fd, IRQ_READ, winch_interrupt, + if(um_request_irq(WINCH_IRQ, fd, IRQ_READ, winch_interrupt, SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM, "winch", winch) < 0) printk("register_winch_irq - failed to register IRQ\n"); @@ -633,6 +738,34 @@ void register_winch_irq(int fd, int tty_fd, int pid, struct tty_struct *tty) up(&winch_handler_sem); } +static void unregister_winch(struct tty_struct *tty) +{ + struct list_head *ele; + struct winch *winch, *found = NULL; + + down(&winch_handler_sem); + list_for_each(ele, &winch_handlers){ + winch = list_entry(ele, struct winch, list); + if(winch->tty == tty){ + found = winch; + break; + } + } + + if(found == NULL) + goto out; + + if(winch->pid != -1) + os_kill_process(winch->pid, 1); + + free_irq_by_irq_and_dev(WINCH_IRQ, winch); + free_irq(WINCH_IRQ, winch); + list_del(&winch->list); + kfree(winch); + out: + up(&winch_handler_sem); +} + static void winch_cleanup(void) { struct list_head *ele; @@ -656,26 +789,16 @@ char *add_xterm_umid(char *base) int len; umid = get_umid(1); - if(umid == NULL) return(base); + if(umid == NULL) + return base; len = strlen(base) + strlen(" ()") + strlen(umid) + 1; title = kmalloc(len, GFP_KERNEL); if(title == NULL){ printk("Failed to allocate buffer for xterm title\n"); - return(base); + return base; } snprintf(title, len, "%s (%s)", base, umid); - return(title); + return title; } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/ssl.c b/arch/um/drivers/ssl.c index c5839c3141f..a2bac429f3d 100644 --- a/arch/um/drivers/ssl.c +++ b/arch/um/drivers/ssl.c @@ -107,11 +107,6 @@ int ssl_open(struct tty_struct *tty, struct file *filp) } #if 0 -static int ssl_chars_in_buffer(struct tty_struct *tty) -{ - return(0); -} - static void ssl_flush_buffer(struct tty_struct *tty) { return; @@ -149,11 +144,11 @@ static struct tty_operations ssl_ops = { .put_char = line_put_char, .write_room = line_write_room, .chars_in_buffer = line_chars_in_buffer, + .flush_buffer = line_flush_buffer, + .flush_chars = line_flush_chars, .set_termios = line_set_termios, .ioctl = line_ioctl, #if 0 - .flush_chars = ssl_flush_chars, - .flush_buffer = ssl_flush_buffer, .throttle = ssl_throttle, .unthrottle = ssl_unthrottle, .stop = ssl_stop, @@ -171,10 +166,11 @@ static void ssl_console_write(struct console *c, const char *string, unsigned len) { struct line *line = &serial_lines[c->index]; + unsigned long flags; - down(&line->sem); + spin_lock_irqsave(&line->lock, flags); console_write_chan(&line->chan_list, string, len); - up(&line->sem); + spin_unlock_irqrestore(&line->lock, flags); } static struct tty_driver *ssl_console_device(struct console *c, int *index) @@ -238,14 +234,3 @@ static int ssl_chan_setup(char *str) __setup("ssl", ssl_chan_setup); __channel_help(ssl_chan_setup, "ssl"); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/stdio_console.c b/arch/um/drivers/stdio_console.c index e604d7c8769..361d0be342b 100644 --- a/arch/um/drivers/stdio_console.c +++ b/arch/um/drivers/stdio_console.c @@ -116,8 +116,11 @@ static struct tty_operations console_ops = { .open = con_open, .close = line_close, .write = line_write, + .put_char = line_put_char, .write_room = line_write_room, .chars_in_buffer = line_chars_in_buffer, + .flush_buffer = line_flush_buffer, + .flush_chars = line_flush_chars, .set_termios = line_set_termios, .ioctl = line_ioctl, }; @@ -126,10 +129,11 @@ static void uml_console_write(struct console *console, const char *string, unsigned len) { struct line *line = &vts[console->index]; + unsigned long flags; - down(&line->sem); + spin_lock_irqsave(&line->lock, flags); console_write_chan(&line->chan_list, string, len); - up(&line->sem); + spin_unlock_irqrestore(&line->lock, flags); } static struct tty_driver *uml_console_device(struct console *c, int *index) @@ -192,14 +196,3 @@ static int console_chan_setup(char *str) } __setup("con", console_chan_setup); __channel_help(console_chan_setup, "con"); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 4d8b165bfa4..9a56ff94308 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -156,6 +156,7 @@ static struct gendisk *fake_gendisk[MAX_DEV]; static struct openflags global_openflags = OPEN_FLAGS; struct cow { + /* This is the backing file, actually */ char *file; int fd; unsigned long *bitmap; @@ -927,10 +928,14 @@ static int ubd_open(struct inode *inode, struct file *filp) } } dev->count++; - if((filp->f_mode & FMODE_WRITE) && !dev->openflags.w){ + set_disk_ro(disk, !dev->openflags.w); + + /* This should no more be needed. And it didn't work anyway to exclude + * read-write remounting of filesystems.*/ + /*if((filp->f_mode & FMODE_WRITE) && !dev->openflags.w){ if(--dev->count == 0) ubd_close(dev); err = -EROFS; - } + }*/ out: return(err); } @@ -1096,6 +1101,7 @@ static int prepare_request(struct request *req, struct io_thread_req *io_req) if(req->rq_status == RQ_INACTIVE) return(1); + /* This should be impossible now */ if((rq_data_dir(req) == WRITE) && !dev->openflags.w){ printk("Write attempted on readonly ubd device %s\n", disk->disk_name); @@ -1243,6 +1249,7 @@ static int ubd_check_remapped(int fd, unsigned long address, int is_write, /* It's a write to a ubd device */ + /* This should be impossible now */ if(!dev->openflags.w){ /* It's a write access on a read-only device - probably * shouldn't happen. If the kernel is trying to change @@ -1605,8 +1612,7 @@ void do_io(struct io_thread_req *req) } } while((n < len) && (n != 0)); if (n < len) memset(&buf[n], 0, len - n); - } - else { + } else { n = os_write_file(req->fds[bit], buf, len); if(n != len){ printk("do_io - write failed err = %d " diff --git a/arch/um/drivers/xterm_kern.c b/arch/um/drivers/xterm_kern.c index 7917b9d1cec..a4fdf3584ad 100644 --- a/arch/um/drivers/xterm_kern.c +++ b/arch/um/drivers/xterm_kern.c @@ -7,7 +7,6 @@ #include "linux/slab.h" #include "linux/signal.h" #include "linux/interrupt.h" -#include "asm/semaphore.h" #include "asm/irq.h" #include "irq_user.h" #include "irq_kern.h" diff --git a/arch/um/include/common-offsets.h b/arch/um/include/common-offsets.h new file mode 100644 index 00000000000..d705daa2d85 --- /dev/null +++ b/arch/um/include/common-offsets.h @@ -0,0 +1,14 @@ +/* for use by sys-$SUBARCH/kernel-offsets.c */ + +OFFSET(TASK_REGS, task_struct, thread.regs); +OFFSET(TASK_PID, task_struct, pid); +DEFINE(UM_KERN_PAGE_SIZE, PAGE_SIZE); +DEFINE(UM_NSEC_PER_SEC, NSEC_PER_SEC); +DEFINE_STR(UM_KERN_EMERG, KERN_EMERG); +DEFINE_STR(UM_KERN_ALERT, KERN_ALERT); +DEFINE_STR(UM_KERN_CRIT, KERN_CRIT); +DEFINE_STR(UM_KERN_ERR, KERN_ERR); +DEFINE_STR(UM_KERN_WARNING, KERN_WARNING); +DEFINE_STR(UM_KERN_NOTICE, KERN_NOTICE); +DEFINE_STR(UM_KERN_INFO, KERN_INFO); +DEFINE_STR(UM_KERN_DEBUG, KERN_DEBUG); diff --git a/arch/um/include/kern_util.h b/arch/um/include/kern_util.h index 15389c886b4..e5fec557019 100644 --- a/arch/um/include/kern_util.h +++ b/arch/um/include/kern_util.h @@ -8,6 +8,7 @@ #include "linux/threads.h" #include "sysdep/ptrace.h" +#include "sysdep/faultinfo.h" extern int ncpus; extern char *linux_prog; @@ -31,8 +32,8 @@ extern int current_pid(void); extern unsigned long alloc_stack(int order, int atomic); extern int do_signal(void); extern int is_stack_fault(unsigned long sp); -extern unsigned long segv(unsigned long address, unsigned long ip, - int is_write, int is_user, void *sc); +extern unsigned long segv(struct faultinfo fi, unsigned long ip, + int is_user, void *sc); extern int handle_page_fault(unsigned long address, unsigned long ip, int is_write, int is_user, int *code_out); extern void syscall_ready(void); @@ -82,7 +83,7 @@ extern void timer_irq(union uml_pt_regs *regs); extern void unprotect_stack(unsigned long stack); extern void do_uml_exitcalls(void); extern int attach_debugger(int idle_pid, int pid, int stop); -extern void bad_segv(unsigned long address, unsigned long ip, int is_write); +extern void bad_segv(struct faultinfo fi, unsigned long ip); extern int config_gdb(char *str); extern int remove_gdb(void); extern char *uml_strdup(char *string); diff --git a/arch/um/include/line.h b/arch/um/include/line.h index 6d81ecc17be..4c5e92c04cc 100644 --- a/arch/um/include/line.h +++ b/arch/um/include/line.h @@ -10,7 +10,7 @@ #include "linux/workqueue.h" #include "linux/tty.h" #include "linux/interrupt.h" -#include "asm/semaphore.h" +#include "linux/spinlock.h" #include "chan_user.h" #include "mconsole_kern.h" @@ -37,10 +37,18 @@ struct line { struct list_head chan_list; int valid; int count; - struct semaphore sem; + /*This lock is actually, mostly, local to*/ + spinlock_t lock; + + /* Yes, this is a real circular buffer. + * XXX: And this should become a struct kfifo! + * + * buffer points to a buffer allocated on demand, of length + * LINE_BUFSIZE, head to the start of the ring, tail to the end.*/ char *buffer; char *head; char *tail; + int sigio; struct work_struct task; struct line_driver *driver; @@ -52,7 +60,6 @@ struct line { init_pri : INIT_STATIC, \ chan_list : { }, \ valid : 1, \ - sem : { }, \ buffer : NULL, \ head : NULL, \ tail : NULL, \ @@ -69,15 +76,18 @@ struct lines { extern void line_close(struct tty_struct *tty, struct file * filp); extern int line_open(struct line *lines, struct tty_struct *tty, struct chan_opts *opts); -extern int line_setup(struct line *lines, int num, char *init, +extern int line_setup(struct line *lines, unsigned int sizeof_lines, char *init, int all_allowed); extern int line_write(struct tty_struct *tty, const unsigned char *buf, int len); extern void line_put_char(struct tty_struct *tty, unsigned char ch); extern void line_set_termios(struct tty_struct *tty, struct termios * old); extern int line_chars_in_buffer(struct tty_struct *tty); +extern void line_flush_buffer(struct tty_struct *tty); +extern void line_flush_chars(struct tty_struct *tty); extern int line_write_room(struct tty_struct *tty); extern int line_ioctl(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg); + extern char *add_xterm_umid(char *base); extern int line_setup_irq(int fd, int input, int output, struct tty_struct *tty); extern void line_close_chan(struct line *line); @@ -89,20 +99,10 @@ extern struct tty_driver * line_register_devfs(struct lines *set, int nlines); extern void lines_init(struct line *lines, int nlines); extern void close_lines(struct line *lines, int nlines); -extern int line_config(struct line *lines, int num, char *str); -extern int line_remove(struct line *lines, int num, char *str); -extern int line_get_config(char *dev, struct line *lines, int num, char *str, + +extern int line_config(struct line *lines, unsigned int sizeof_lines, char *str); +extern int line_remove(struct line *lines, unsigned int sizeof_lines, char *str); +extern int line_get_config(char *dev, struct line *lines, unsigned int sizeof_lines, char *str, int size, char **error_out); #endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/include/os.h b/arch/um/include/os.h index 07340c8cf20..d246d5a2460 100644 --- a/arch/um/include/os.h +++ b/arch/um/include/os.h @@ -160,6 +160,7 @@ extern void os_kill_process(int pid, int reap_child); extern void os_kill_ptraced_process(int pid, int reap_child); extern void os_usr1_process(int pid); extern int os_getpid(void); +extern int os_getpgrp(void); extern int os_map_memory(void *virt, int fd, unsigned long long off, unsigned long len, int r, int w, int x); diff --git a/arch/um/include/skas_ptrace.h b/arch/um/include/skas_ptrace.h index cfb5fb4f5b9..cd2327d09c8 100644 --- a/arch/um/include/skas_ptrace.h +++ b/arch/um/include/skas_ptrace.h @@ -6,22 +6,11 @@ #ifndef __SKAS_PTRACE_H #define __SKAS_PTRACE_H -struct ptrace_faultinfo { - int is_write; - unsigned long addr; -}; - -struct ptrace_ldt { - int func; - void *ptr; - unsigned long bytecount; -}; - #define PTRACE_FAULTINFO 52 -#define PTRACE_SIGPENDING 53 -#define PTRACE_LDT 54 #define PTRACE_SWITCH_MM 55 +#include "sysdep/skas_ptrace.h" + #endif /* diff --git a/arch/um/include/sysdep-i386/checksum.h b/arch/um/include/sysdep-i386/checksum.h index 3a2a45811aa..764ba4db478 100644 --- a/arch/um/include/sysdep-i386/checksum.h +++ b/arch/um/include/sysdep-i386/checksum.h @@ -24,19 +24,6 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); /* - * the same as csum_partial, but copies from src while it - * checksums, and handles user-space pointer exceptions correctly, when needed. - * - * here even more important to align src and dst on a 32-bit (or even - * better 64-bit) boundary - */ - -unsigned int csum_partial_copy_to(const unsigned char *src, unsigned char *dst, - int len, int sum, int *err_ptr); -unsigned int csum_partial_copy_from(const unsigned char *src, unsigned char *dst, - int len, int sum, int *err_ptr); - -/* * Note: when you get a NULL pointer exception here this means someone * passed in an incorrect kernel address to one of these functions. * @@ -52,11 +39,24 @@ unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned char * return(csum_partial(dst, len, sum)); } +/* + * the same as csum_partial, but copies from src while it + * checksums, and handles user-space pointer exceptions correctly, when needed. + * + * here even more important to align src and dst on a 32-bit (or even + * better 64-bit) boundary + */ + static __inline__ unsigned int csum_partial_copy_from_user(const unsigned char *src, unsigned char *dst, int len, int sum, int *err_ptr) { - return csum_partial_copy_from(src, dst, len, sum, err_ptr); + if(copy_from_user(dst, src, len)){ + *err_ptr = -EFAULT; + return(-1); + } + + return csum_partial(dst, len, sum); } /* @@ -67,7 +67,6 @@ unsigned int csum_partial_copy_from_user(const unsigned char *src, unsigned char */ #define csum_partial_copy_fromuser csum_partial_copy_from_user -unsigned int csum_partial_copy(const unsigned char *src, unsigned char *dst, int len, int sum); /* * This is a version of ip_compute_csum() optimized for IP headers, @@ -196,8 +195,14 @@ static __inline__ unsigned int csum_and_copy_to_user(const unsigned char *src, unsigned char *dst, int len, int sum, int *err_ptr) { - if (access_ok(VERIFY_WRITE, dst, len)) - return(csum_partial_copy_to(src, dst, len, sum, err_ptr)); + if (access_ok(VERIFY_WRITE, dst, len)){ + if(copy_to_user(dst, src, len)){ + *err_ptr = -EFAULT; + return(-1); + } + + return csum_partial(src, len, sum); + } if (len) *err_ptr = -EFAULT; diff --git a/arch/um/include/sysdep-i386/faultinfo.h b/arch/um/include/sysdep-i386/faultinfo.h new file mode 100644 index 00000000000..db437cc373b --- /dev/null +++ b/arch/um/include/sysdep-i386/faultinfo.h @@ -0,0 +1,29 @@ +/* + * Copyright (C) 2004 Fujitsu Siemens Computers GmbH + * Author: Bodo Stroesser <bstroesser@fujitsu-siemens.com> + * Licensed under the GPL + */ + +#ifndef __FAULTINFO_I386_H +#define __FAULTINFO_I386_H + +/* this structure contains the full arch-specific faultinfo + * from the traps. + * On i386, ptrace_faultinfo unfortunately doesn't provide + * all the info, since trap_no is missing. + * All common elements are defined at the same position in + * both structures, thus making it easy to copy the + * contents without knowledge about the structure elements. + */ +struct faultinfo { + int error_code; /* in ptrace_faultinfo misleadingly called is_write */ + unsigned long cr2; /* in ptrace_faultinfo called addr */ + int trap_no; /* missing in ptrace_faultinfo */ +}; + +#define FAULT_WRITE(fi) ((fi).error_code & 2) +#define FAULT_ADDRESS(fi) ((fi).cr2) + +#define PTRACE_FULL_FAULTINFO 0 + +#endif diff --git a/arch/um/include/sysdep-i386/ptrace.h b/arch/um/include/sysdep-i386/ptrace.h index 661d495e204..84ec7ff5cf8 100644 --- a/arch/um/include/sysdep-i386/ptrace.h +++ b/arch/um/include/sysdep-i386/ptrace.h @@ -31,6 +31,7 @@ extern int sysemu_supported; #ifdef UML_CONFIG_MODE_SKAS #include "skas_ptregs.h" +#include "sysdep/faultinfo.h" #define REGS_IP(r) ((r)[HOST_IP]) #define REGS_SP(r) ((r)[HOST_SP]) @@ -53,12 +54,6 @@ extern int sysemu_supported; #define REGS_RESTART_SYSCALL(r) IP_RESTART_SYSCALL(REGS_IP(r)) -#define REGS_SEGV_IS_FIXABLE(r) SEGV_IS_FIXABLE((r)->trap_type) - -#define REGS_FAULT_ADDR(r) ((r)->fault_addr) - -#define REGS_FAULT_WRITE(r) FAULT_WRITE((r)->fault_type) - #endif #ifndef PTRACE_SYSEMU_SINGLESTEP #define PTRACE_SYSEMU_SINGLESTEP 32 @@ -71,6 +66,7 @@ union uml_pt_regs { struct tt_regs { long syscall; void *sc; + struct faultinfo faultinfo; } tt; #endif #ifdef UML_CONFIG_MODE_SKAS @@ -78,9 +74,7 @@ union uml_pt_regs { unsigned long regs[HOST_FRAME_SIZE]; unsigned long fp[HOST_FP_SIZE]; unsigned long xfp[HOST_XFP_SIZE]; - unsigned long fault_addr; - unsigned long fault_type; - unsigned long trap_type; + struct faultinfo faultinfo; long syscall; int is_user; } skas; @@ -217,15 +211,8 @@ struct syscall_args { #define UPT_SYSCALL_NR(r) UPT_ORIG_EAX(r) #define UPT_SYSCALL_RET(r) UPT_EAX(r) -#define UPT_SEGV_IS_FIXABLE(r) \ - CHOOSE_MODE(SC_SEGV_IS_FIXABLE(UPT_SC(r)), \ - REGS_SEGV_IS_FIXABLE(&r->skas)) - -#define UPT_FAULT_ADDR(r) \ - __CHOOSE_MODE(SC_FAULT_ADDR(UPT_SC(r)), REGS_FAULT_ADDR(&r->skas)) - -#define UPT_FAULT_WRITE(r) \ - CHOOSE_MODE(SC_FAULT_WRITE(UPT_SC(r)), REGS_FAULT_WRITE(&r->skas)) +#define UPT_FAULTINFO(r) \ + CHOOSE_MODE((&(r)->tt.faultinfo), (&(r)->skas.faultinfo)) #endif diff --git a/arch/um/include/sysdep-i386/sigcontext.h b/arch/um/include/sysdep-i386/sigcontext.h index dfee589de36..1fe72926516 100644 --- a/arch/um/include/sysdep-i386/sigcontext.h +++ b/arch/um/include/sysdep-i386/sigcontext.h @@ -13,15 +13,12 @@ #define SC_RESTART_SYSCALL(sc) IP_RESTART_SYSCALL(SC_IP(sc)) #define SC_SET_SYSCALL_RETURN(sc, result) SC_EAX(sc) = (result) -#define SC_FAULT_ADDR(sc) SC_CR2(sc) -#define SC_FAULT_TYPE(sc) SC_ERR(sc) - -#define FAULT_WRITE(err) (err & 2) -#define TO_SC_ERR(is_write) ((is_write) ? 2 : 0) - -#define SC_FAULT_WRITE(sc) (FAULT_WRITE(SC_ERR(sc))) - -#define SC_TRAP_TYPE(sc) SC_TRAPNO(sc) +#define GET_FAULTINFO_FROM_SC(fi,sc) \ + { \ + (fi).cr2 = SC_CR2(sc); \ + (fi).error_code = SC_ERR(sc); \ + (fi).trap_no = SC_TRAPNO(sc); \ + } /* ptrace expects that, at the start of a system call, %eax contains * -ENOSYS, so this makes it so. @@ -29,9 +26,7 @@ #define SC_START_SYSCALL(sc) do SC_EAX(sc) = -ENOSYS; while(0) /* This is Page Fault */ -#define SEGV_IS_FIXABLE(trap) (trap == 14) - -#define SC_SEGV_IS_FIXABLE(sc) (SEGV_IS_FIXABLE(SC_TRAPNO(sc))) +#define SEGV_IS_FIXABLE(fi) ((fi)->trap_no == 14) extern unsigned long *sc_sigmask(void *sc_ptr); extern int sc_get_fpregs(unsigned long buf, void *sc_ptr); diff --git a/arch/um/include/sysdep-i386/signal.h b/arch/um/include/sysdep-i386/signal.h index b1e1f7a7749..07518b16213 100644 --- a/arch/um/include/sysdep-i386/signal.h +++ b/arch/um/include/sysdep-i386/signal.h @@ -8,6 +8,8 @@ #include <signal.h> +#define ARCH_SIGHDLR_PARAM int sig + #define ARCH_GET_SIGCONTEXT(sc, sig) \ do sc = (struct sigcontext *) (&sig + 1); while(0) diff --git a/arch/um/include/sysdep-i386/skas_ptrace.h b/arch/um/include/sysdep-i386/skas_ptrace.h new file mode 100644 index 00000000000..e27b8a79177 --- /dev/null +++ b/arch/um/include/sysdep-i386/skas_ptrace.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_I386_SKAS_PTRACE_H +#define __SYSDEP_I386_SKAS_PTRACE_H + +struct ptrace_faultinfo { + int is_write; + unsigned long addr; +}; + +struct ptrace_ldt { + int func; + void *ptr; + unsigned long bytecount; +}; + +#define PTRACE_LDT 54 + +#endif diff --git a/arch/um/include/sysdep-i386/syscalls.h b/arch/um/include/sysdep-i386/syscalls.h index 5db81ec9087..be0a3e3469e 100644 --- a/arch/um/include/sysdep-i386/syscalls.h +++ b/arch/um/include/sysdep-i386/syscalls.h @@ -22,102 +22,3 @@ extern syscall_handler_t old_mmap_i386; extern long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff); - -/* On i386 they choose a meaningless naming.*/ -#define __NR_kexec_load __NR_sys_kexec_load - -#define ARCH_SYSCALLS \ - [ __NR_waitpid ] = (syscall_handler_t *) sys_waitpid, \ - [ __NR_break ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_oldstat ] = (syscall_handler_t *) sys_stat, \ - [ __NR_umount ] = (syscall_handler_t *) sys_oldumount, \ - [ __NR_stime ] = um_stime, \ - [ __NR_oldfstat ] = (syscall_handler_t *) sys_fstat, \ - [ __NR_stty ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_gtty ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_nice ] = (syscall_handler_t *) sys_nice, \ - [ __NR_ftime ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_prof ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_signal ] = (syscall_handler_t *) sys_signal, \ - [ __NR_lock ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_mpx ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_ulimit ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_oldolduname ] = (syscall_handler_t *) sys_olduname, \ - [ __NR_sigaction ] = (syscall_handler_t *) sys_sigaction, \ - [ __NR_sgetmask ] = (syscall_handler_t *) sys_sgetmask, \ - [ __NR_ssetmask ] = (syscall_handler_t *) sys_ssetmask, \ - [ __NR_sigsuspend ] = (syscall_handler_t *) sys_sigsuspend, \ - [ __NR_sigpending ] = (syscall_handler_t *) sys_sigpending, \ - [ __NR_oldlstat ] = (syscall_handler_t *) sys_lstat, \ - [ __NR_readdir ] = old_readdir, \ - [ __NR_profil ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_socketcall ] = (syscall_handler_t *) sys_socketcall, \ - [ __NR_olduname ] = (syscall_handler_t *) sys_uname, \ - [ __NR_iopl ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_idle ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_ipc ] = (syscall_handler_t *) sys_ipc, \ - [ __NR_sigreturn ] = (syscall_handler_t *) sys_sigreturn, \ - [ __NR_sigprocmask ] = (syscall_handler_t *) sys_sigprocmask, \ - [ __NR_bdflush ] = (syscall_handler_t *) sys_bdflush, \ - [ __NR__llseek ] = (syscall_handler_t *) sys_llseek, \ - [ __NR__newselect ] = (syscall_handler_t *) sys_select, \ - [ __NR_vm86 ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_mmap ] = (syscall_handler_t *) old_mmap_i386, \ - [ __NR_ugetrlimit ] = (syscall_handler_t *) sys_getrlimit, \ - [ __NR_mmap2 ] = (syscall_handler_t *) sys_mmap2, \ - [ __NR_truncate64 ] = (syscall_handler_t *) sys_truncate64, \ - [ __NR_ftruncate64 ] = (syscall_handler_t *) sys_ftruncate64, \ - [ __NR_stat64 ] = (syscall_handler_t *) sys_stat64, \ - [ __NR_lstat64 ] = (syscall_handler_t *) sys_lstat64, \ - [ __NR_fstat64 ] = (syscall_handler_t *) sys_fstat64, \ - [ __NR_fcntl64 ] = (syscall_handler_t *) sys_fcntl64, \ - [ __NR_sendfile64 ] = (syscall_handler_t *) sys_sendfile64, \ - [ __NR_statfs64 ] = (syscall_handler_t *) sys_statfs64, \ - [ __NR_fstatfs64 ] = (syscall_handler_t *) sys_fstatfs64, \ - [ __NR_fadvise64_64 ] = (syscall_handler_t *) sys_fadvise64_64, \ - [ __NR_select ] = (syscall_handler_t *) old_select, \ - [ __NR_vm86old ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_modify_ldt ] = (syscall_handler_t *) sys_modify_ldt, \ - [ __NR_lchown32 ] = (syscall_handler_t *) sys_lchown, \ - [ __NR_getuid32 ] = (syscall_handler_t *) sys_getuid, \ - [ __NR_getgid32 ] = (syscall_handler_t *) sys_getgid, \ - [ __NR_geteuid32 ] = (syscall_handler_t *) sys_geteuid, \ - [ __NR_getegid32 ] = (syscall_handler_t *) sys_getegid, \ - [ __NR_setreuid32 ] = (syscall_handler_t *) sys_setreuid, \ - [ __NR_setregid32 ] = (syscall_handler_t *) sys_setregid, \ - [ __NR_getgroups32 ] = (syscall_handler_t *) sys_getgroups, \ - [ __NR_setgroups32 ] = (syscall_handler_t *) sys_setgroups, \ - [ __NR_fchown32 ] = (syscall_handler_t *) sys_fchown, \ - [ __NR_setresuid32 ] = (syscall_handler_t *) sys_setresuid, \ - [ __NR_getresuid32 ] = (syscall_handler_t *) sys_getresuid, \ - [ __NR_setresgid32 ] = (syscall_handler_t *) sys_setresgid, \ - [ __NR_getresgid32 ] = (syscall_handler_t *) sys_getresgid, \ - [ __NR_chown32 ] = (syscall_handler_t *) sys_chown, \ - [ __NR_setuid32 ] = (syscall_handler_t *) sys_setuid, \ - [ __NR_setgid32 ] = (syscall_handler_t *) sys_setgid, \ - [ __NR_setfsuid32 ] = (syscall_handler_t *) sys_setfsuid, \ - [ __NR_setfsgid32 ] = (syscall_handler_t *) sys_setfsgid, \ - [ __NR_pivot_root ] = (syscall_handler_t *) sys_pivot_root, \ - [ __NR_mincore ] = (syscall_handler_t *) sys_mincore, \ - [ __NR_madvise ] = (syscall_handler_t *) sys_madvise, \ - [ 222 ] = (syscall_handler_t *) sys_ni_syscall, \ - [ 223 ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_set_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_get_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ - [ 251 ] = (syscall_handler_t *) sys_ni_syscall, \ - [ 285 ] = (syscall_handler_t *) sys_ni_syscall, - -/* 222 doesn't yet have a name in include/asm-i386/unistd.h */ - -#define LAST_ARCH_SYSCALL 285 - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/include/sysdep-ia64/skas_ptrace.h b/arch/um/include/sysdep-ia64/skas_ptrace.h new file mode 100644 index 00000000000..25a38e71570 --- /dev/null +++ b/arch/um/include/sysdep-ia64/skas_ptrace.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_IA64_SKAS_PTRACE_H +#define __SYSDEP_IA64_SKAS_PTRACE_H + +struct ptrace_faultinfo { + int is_write; + unsigned long addr; +}; + +struct ptrace_ldt { + int func; + void *ptr; + unsigned long bytecount; +}; + +#define PTRACE_LDT 54 + +#endif diff --git a/arch/um/include/sysdep-ppc/skas_ptrace.h b/arch/um/include/sysdep-ppc/skas_ptrace.h new file mode 100644 index 00000000000..d9fbbac10de --- /dev/null +++ b/arch/um/include/sysdep-ppc/skas_ptrace.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_PPC_SKAS_PTRACE_H +#define __SYSDEP_PPC_SKAS_PTRACE_H + +struct ptrace_faultinfo { + int is_write; + unsigned long addr; +}; + +struct ptrace_ldt { + int func; + void *ptr; + unsigned long bytecount; +}; + +#define PTRACE_LDT 54 + +#endif diff --git a/arch/um/include/sysdep-x86_64/faultinfo.h b/arch/um/include/sysdep-x86_64/faultinfo.h new file mode 100644 index 00000000000..cb917b0d566 --- /dev/null +++ b/arch/um/include/sysdep-x86_64/faultinfo.h @@ -0,0 +1,29 @@ +/* + * Copyright (C) 2004 Fujitsu Siemens Computers GmbH + * Author: Bodo Stroesser <bstroesser@fujitsu-siemens.com> + * Licensed under the GPL + */ + +#ifndef __FAULTINFO_X86_64_H +#define __FAULTINFO_X86_64_H + +/* this structure contains the full arch-specific faultinfo + * from the traps. + * On i386, ptrace_faultinfo unfortunately doesn't provide + * all the info, since trap_no is missing. + * All common elements are defined at the same position in + * both structures, thus making it easy to copy the + * contents without knowledge about the structure elements. + */ +struct faultinfo { + int error_code; /* in ptrace_faultinfo misleadingly called is_write */ + unsigned long cr2; /* in ptrace_faultinfo called addr */ + int trap_no; /* missing in ptrace_faultinfo */ +}; + +#define FAULT_WRITE(fi) ((fi).error_code & 2) +#define FAULT_ADDRESS(fi) ((fi).cr2) + +#define PTRACE_FULL_FAULTINFO 1 + +#endif diff --git a/arch/um/include/sysdep-x86_64/ptrace.h b/arch/um/include/sysdep-x86_64/ptrace.h index 915c82daffb..348e8fcd513 100644 --- a/arch/um/include/sysdep-x86_64/ptrace.h +++ b/arch/um/include/sysdep-x86_64/ptrace.h @@ -9,6 +9,7 @@ #include "uml-config.h" #include "user_constants.h" +#include "sysdep/faultinfo.h" #define MAX_REG_OFFSET (UM_FRAME_SIZE) #define MAX_REG_NR ((MAX_REG_OFFSET) / sizeof(unsigned long)) @@ -83,6 +84,7 @@ union uml_pt_regs { long syscall; unsigned long orig_rax; void *sc; + struct faultinfo faultinfo; } tt; #endif #ifdef UML_CONFIG_MODE_SKAS @@ -90,9 +92,7 @@ union uml_pt_regs { /* XXX */ unsigned long regs[27]; unsigned long fp[65]; - unsigned long fault_addr; - unsigned long fault_type; - unsigned long trap_type; + struct faultinfo faultinfo; long syscall; int is_user; } skas; @@ -241,14 +241,8 @@ struct syscall_args { CHOOSE_MODE(SC_SEGV_IS_FIXABLE(UPT_SC(r)), \ REGS_SEGV_IS_FIXABLE(&r->skas)) -#define UPT_FAULT_ADDR(r) \ - __CHOOSE_MODE(SC_FAULT_ADDR(UPT_SC(r)), REGS_FAULT_ADDR(&r->skas)) - -#define UPT_FAULT_WRITE(r) \ - CHOOSE_MODE(SC_FAULT_WRITE(UPT_SC(r)), REGS_FAULT_WRITE(&r->skas)) - -#define UPT_TRAP(r) __CHOOSE_MODE(SC_TRAP_TYPE(UPT_SC(r)), REGS_TRAP(&r->skas)) -#define UPT_ERR(r) __CHOOSE_MODE(SC_FAULT_TYPE(UPT_SC(r)), REGS_ERR(&r->skas)) +#define UPT_FAULTINFO(r) \ + CHOOSE_MODE((&(r)->tt.faultinfo), (&(r)->skas.faultinfo)) #endif diff --git a/arch/um/include/sysdep-x86_64/sigcontext.h b/arch/um/include/sysdep-x86_64/sigcontext.h index 1e38a54ff4c..2a78260d15a 100644 --- a/arch/um/include/sysdep-x86_64/sigcontext.h +++ b/arch/um/include/sysdep-x86_64/sigcontext.h @@ -17,11 +17,12 @@ #define SC_FAULT_ADDR(sc) SC_CR2(sc) #define SC_FAULT_TYPE(sc) SC_ERR(sc) -#define FAULT_WRITE(err) ((err) & 2) - -#define SC_FAULT_WRITE(sc) FAULT_WRITE(SC_FAULT_TYPE(sc)) - -#define SC_TRAP_TYPE(sc) SC_TRAPNO(sc) +#define GET_FAULTINFO_FROM_SC(fi,sc) \ + { \ + (fi).cr2 = SC_CR2(sc); \ + (fi).error_code = SC_ERR(sc); \ + (fi).trap_no = SC_TRAPNO(sc); \ + } /* ptrace expects that, at the start of a system call, %eax contains * -ENOSYS, so this makes it so. @@ -29,8 +30,8 @@ #define SC_START_SYSCALL(sc) do SC_RAX(sc) = -ENOSYS; while(0) -#define SEGV_IS_FIXABLE(trap) ((trap) == 14) -#define SC_SEGV_IS_FIXABLE(sc) SEGV_IS_FIXABLE(SC_TRAP_TYPE(sc)) +/* This is Page Fault */ +#define SEGV_IS_FIXABLE(fi) ((fi)->trap_no == 14) extern unsigned long *sc_sigmask(void *sc_ptr); diff --git a/arch/um/include/sysdep-x86_64/signal.h b/arch/um/include/sysdep-x86_64/signal.h index e5e52756fab..6142897af3d 100644 --- a/arch/um/include/sysdep-x86_64/signal.h +++ b/arch/um/include/sysdep-x86_64/signal.h @@ -6,6 +6,8 @@ #ifndef __X86_64_SIGNAL_H_ #define __X86_64_SIGNAL_H_ +#define ARCH_SIGHDLR_PARAM int sig + #define ARCH_GET_SIGCONTEXT(sc, sig_addr) \ do { \ struct ucontext *__uc; \ diff --git a/arch/um/include/sysdep-x86_64/skas_ptrace.h b/arch/um/include/sysdep-x86_64/skas_ptrace.h new file mode 100644 index 00000000000..95db4be786e --- /dev/null +++ b/arch/um/include/sysdep-x86_64/skas_ptrace.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_X86_64_SKAS_PTRACE_H +#define __SYSDEP_X86_64_SKAS_PTRACE_H + +struct ptrace_faultinfo { + int is_write; + unsigned long addr; +}; + +struct ptrace_ldt { + int func; + void *ptr; + unsigned long bytecount; +}; + +#define PTRACE_LDT 54 + +#endif diff --git a/arch/um/include/sysdep-x86_64/syscalls.h b/arch/um/include/sysdep-x86_64/syscalls.h index b187a4157ff..67923cca569 100644 --- a/arch/um/include/sysdep-x86_64/syscalls.h +++ b/arch/um/include/sysdep-x86_64/syscalls.h @@ -26,66 +26,9 @@ extern syscall_handler_t *ia32_sys_call_table[]; extern long old_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff); -extern syscall_handler_t wrap_sys_shmat; extern syscall_handler_t sys_modify_ldt; extern syscall_handler_t sys_arch_prctl; -#define ARCH_SYSCALLS \ - [ __NR_mmap ] = (syscall_handler_t *) old_mmap, \ - [ __NR_select ] = (syscall_handler_t *) sys_select, \ - [ __NR_mincore ] = (syscall_handler_t *) sys_mincore, \ - [ __NR_madvise ] = (syscall_handler_t *) sys_madvise, \ - [ __NR_shmget ] = (syscall_handler_t *) sys_shmget, \ - [ __NR_shmat ] = (syscall_handler_t *) wrap_sys_shmat, \ - [ __NR_shmctl ] = (syscall_handler_t *) sys_shmctl, \ - [ __NR_semop ] = (syscall_handler_t *) sys_semop, \ - [ __NR_semget ] = (syscall_handler_t *) sys_semget, \ - [ __NR_semctl ] = (syscall_handler_t *) sys_semctl, \ - [ __NR_shmdt ] = (syscall_handler_t *) sys_shmdt, \ - [ __NR_msgget ] = (syscall_handler_t *) sys_msgget, \ - [ __NR_msgsnd ] = (syscall_handler_t *) sys_msgsnd, \ - [ __NR_msgrcv ] = (syscall_handler_t *) sys_msgrcv, \ - [ __NR_msgctl ] = (syscall_handler_t *) sys_msgctl, \ - [ __NR_pivot_root ] = (syscall_handler_t *) sys_pivot_root, \ - [ __NR_tuxcall ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_security ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_epoll_ctl_old ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_epoll_wait_old ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_modify_ldt ] = (syscall_handler_t *) sys_modify_ldt, \ - [ __NR_arch_prctl ] = (syscall_handler_t *) sys_arch_prctl, \ - [ __NR_socket ] = (syscall_handler_t *) sys_socket, \ - [ __NR_connect ] = (syscall_handler_t *) sys_connect, \ - [ __NR_accept ] = (syscall_handler_t *) sys_accept, \ - [ __NR_recvfrom ] = (syscall_handler_t *) sys_recvfrom, \ - [ __NR_recvmsg ] = (syscall_handler_t *) sys_recvmsg, \ - [ __NR_sendmsg ] = (syscall_handler_t *) sys_sendmsg, \ - [ __NR_bind ] = (syscall_handler_t *) sys_bind, \ - [ __NR_listen ] = (syscall_handler_t *) sys_listen, \ - [ __NR_getsockname ] = (syscall_handler_t *) sys_getsockname, \ - [ __NR_getpeername ] = (syscall_handler_t *) sys_getpeername, \ - [ __NR_socketpair ] = (syscall_handler_t *) sys_socketpair, \ - [ __NR_sendto ] = (syscall_handler_t *) sys_sendto, \ - [ __NR_shutdown ] = (syscall_handler_t *) sys_shutdown, \ - [ __NR_setsockopt ] = (syscall_handler_t *) sys_setsockopt, \ - [ __NR_getsockopt ] = (syscall_handler_t *) sys_getsockopt, \ - [ __NR_iopl ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_set_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_get_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_semtimedop ] = (syscall_handler_t *) sys_semtimedop, \ - [ 251 ] = (syscall_handler_t *) sys_ni_syscall, - -#define LAST_ARCH_SYSCALL 251 -#define NR_syscalls 1024 +#define NR_syscalls (__NR_syscall_max + 1) #endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index dc796c1bf39..dca65426385 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -4,13 +4,13 @@ # extra-y := vmlinux.lds -clean-files := vmlinux.lds.S config.tmp +clean-files := -obj-y = checksum.o config.o exec_kern.o exitcode.o \ +obj-y = config.o exec_kern.o exitcode.o \ helper.o init_task.o irq.o irq_user.o ksyms.o main.o mem.o mem_user.o \ physmem.o process.o process_kern.o ptrace.o reboot.o resource.o \ sigio_user.o sigio_kern.o signal_kern.o signal_user.o smp.o \ - syscall_kern.o sysrq.o sys_call_table.o tempfile.o time.o time_kern.o \ + syscall_kern.o sysrq.o tempfile.o time.o time_kern.o \ tlb.o trap_kern.o trap_user.o uaccess_user.o um_arch.o umid.o \ user_util.o @@ -34,7 +34,7 @@ USER_OBJS := $(user-objs-y) config.o helper.o main.o process.o tempfile.o \ include arch/um/scripts/Makefile.rules -targets += config.c +targets := config.c config.tmp # Be careful with the below Sed code - sed is pitfall-rich! # We use sed to lower build requirements, for "embedded" builders for instance. @@ -53,6 +53,7 @@ quiet_cmd_quote2 = QUOTE $@ cmd_quote2 = sed -e '/CONFIG/{' \ -e 's/"CONFIG"\;/""/' \ -e 'r $(obj)/config.tmp' \ - -e 'a""\;' \ + -e 'a \' \ + -e '""\;' \ -e '}' \ $< > $@ diff --git a/arch/um/kernel/checksum.c b/arch/um/kernel/checksum.c index e69b2be951d..e69de29bb2d 100644 --- a/arch/um/kernel/checksum.c +++ b/arch/um/kernel/checksum.c @@ -1,36 +0,0 @@ -#include "asm/uaccess.h" -#include "linux/errno.h" -#include "linux/module.h" - -unsigned int arch_csum_partial(const unsigned char *buff, int len, int sum); - -unsigned int csum_partial(unsigned char *buff, int len, int sum) -{ - return arch_csum_partial(buff, len, sum); -} - -EXPORT_SYMBOL(csum_partial); - -unsigned int csum_partial_copy_to(const unsigned char *src, - unsigned char __user *dst, int len, int sum, - int *err_ptr) -{ - if(copy_to_user(dst, src, len)){ - *err_ptr = -EFAULT; - return(-1); - } - - return(arch_csum_partial(src, len, sum)); -} - -unsigned int csum_partial_copy_from(const unsigned char __user *src, - unsigned char *dst, int len, int sum, - int *err_ptr) -{ - if(copy_from_user(dst, src, len)){ - *err_ptr = -EFAULT; - return(-1); - } - - return arch_csum_partial(dst, len, sum); -} diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c index d71e8f00810..d44fb528254 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -163,7 +163,6 @@ void __init init_IRQ(void) irq_desc[i].handler = &SIGIO_irq_type; enable_irq(i); } - init_irq_signals(0); } /* diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c index b41d3397d07..78d69dc74b2 100644 --- a/arch/um/kernel/ksyms.c +++ b/arch/um/kernel/ksyms.c @@ -10,7 +10,6 @@ #include "linux/spinlock.h" #include "linux/highmem.h" #include "asm/current.h" -#include "asm/delay.h" #include "asm/processor.h" #include "asm/unistd.h" #include "asm/pgalloc.h" @@ -28,8 +27,6 @@ EXPORT_SYMBOL(uml_physmem); EXPORT_SYMBOL(set_signals); EXPORT_SYMBOL(get_signals); EXPORT_SYMBOL(kernel_thread); -EXPORT_SYMBOL(__const_udelay); -EXPORT_SYMBOL(__udelay); EXPORT_SYMBOL(sys_waitpid); EXPORT_SYMBOL(task_size); EXPORT_SYMBOL(flush_tlb_range); diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index f76a2692adc..51f8e5a8ac6 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -65,8 +65,6 @@ void init_new_thread_signals(int altstack) SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); set_handler(SIGBUS, (__sighandler_t) sig_handler, flags, SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); - set_handler(SIGWINCH, (__sighandler_t) sig_handler, flags, - SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); set_handler(SIGUSR2, (__sighandler_t) sig_handler, flags, SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); signal(SIGHUP, SIG_IGN); diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c index 1d719d5b4bb..7a943696f95 100644 --- a/arch/um/kernel/process_kern.c +++ b/arch/um/kernel/process_kern.c @@ -161,10 +161,6 @@ void *get_current(void) return(current); } -void prepare_to_copy(struct task_struct *tsk) -{ -} - int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, unsigned long stack_top, struct task_struct * p, struct pt_regs *regs) diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c index 3a99ee6d94e..3efa59a941a 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -98,12 +98,14 @@ long sys_ptrace(long request, long pid, long addr, long data) if(addr < MAX_REG_OFFSET){ tmp = getreg(child, addr); } +#if defined(CONFIG_UML_X86) && !defined(CONFIG_64BIT) else if((addr >= offsetof(struct user, u_debugreg[0])) && (addr <= offsetof(struct user, u_debugreg[7]))){ addr -= offsetof(struct user, u_debugreg[0]); addr = addr >> 2; tmp = child->thread.arch.debugregs[addr]; } +#endif ret = put_user(tmp, (unsigned long __user *) data); break; } @@ -127,7 +129,7 @@ long sys_ptrace(long request, long pid, long addr, long data) ret = putreg(child, addr, data); break; } -#if 0 /* XXX x86_64 */ +#if defined(CONFIG_UML_X86) && !defined(CONFIG_64BIT) else if((addr >= offsetof(struct user, u_debugreg[0])) && (addr <= offsetof(struct user, u_debugreg[7]))){ addr -= offsetof(struct user, u_debugreg[0]); @@ -143,7 +145,7 @@ long sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; child->ptrace &= ~PT_DTRACE; @@ -179,7 +181,7 @@ long sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); child->ptrace |= PT_DTRACE; @@ -250,23 +252,19 @@ long sys_ptrace(long request, long pid, long addr, long data) break; #endif case PTRACE_FAULTINFO: { - struct ptrace_faultinfo fault; - - fault = ((struct ptrace_faultinfo) - { .is_write = child->thread.err, - .addr = child->thread.cr2 }); - ret = copy_to_user((unsigned long __user *) data, &fault, - sizeof(fault)); + /* Take the info from thread->arch->faultinfo, + * but transfer max. sizeof(struct ptrace_faultinfo). + * On i386, ptrace_faultinfo is smaller! + */ + ret = copy_to_user((unsigned long __user *) data, + &child->thread.arch.faultinfo, + sizeof(struct ptrace_faultinfo)); if(ret) break; break; } - case PTRACE_SIGPENDING: - ret = copy_to_user((unsigned long __user *) data, - &child->pending.signal, - sizeof(child->pending.signal)); - break; +#ifdef PTRACE_LDT case PTRACE_LDT: { struct ptrace_ldt ldt; @@ -282,6 +280,7 @@ long sys_ptrace(long request, long pid, long addr, long data) ret = -EIO; break; } +#endif #ifdef CONFIG_PROC_MM case PTRACE_SWITCH_MM: { struct mm_struct *old = child->mm; @@ -337,15 +336,18 @@ void syscall_trace(union uml_pt_regs *regs, int entryexit) if (unlikely(current->audit_context)) { if (!entryexit) - audit_syscall_entry(current, - UPT_SYSCALL_NR(®s->regs), - UPT_SYSCALL_ARG1(®s->regs), - UPT_SYSCALL_ARG2(®s->regs), - UPT_SYSCALL_ARG3(®s->regs), - UPT_SYSCALL_ARG4(®s->regs)); - else - audit_syscall_exit(current, - UPT_SYSCALL_RET(®s->regs)); + audit_syscall_entry(current, + HOST_AUDIT_ARCH, + UPT_SYSCALL_NR(regs), + UPT_SYSCALL_ARG1(regs), + UPT_SYSCALL_ARG2(regs), + UPT_SYSCALL_ARG3(regs), + UPT_SYSCALL_ARG4(regs)); + else { + int res = UPT_SYSCALL_RET(regs); + audit_syscall_exit(current, AUDITSC_RESULT(res), + res); + } } /* Fake a debug trap */ diff --git a/arch/um/kernel/sigio_user.c b/arch/um/kernel/sigio_user.c index 668df13d8c9..e89218958f3 100644 --- a/arch/um/kernel/sigio_user.c +++ b/arch/um/kernel/sigio_user.c @@ -182,6 +182,7 @@ static int write_sigio_thread(void *unused) int i, n, respond_fd; char c; + signal(SIGWINCH, SIG_IGN); fds = ¤t_poll; while(1){ n = poll(fds->poll, fds->used, -1); diff --git a/arch/um/kernel/skas/include/skas.h b/arch/um/kernel/skas/include/skas.h index f0702c2c720..96b51dba347 100644 --- a/arch/um/kernel/skas/include/skas.h +++ b/arch/um/kernel/skas/include/skas.h @@ -27,9 +27,10 @@ extern void map(int fd, unsigned long virt, unsigned long len, int r, int w, extern int unmap(int fd, void *addr, unsigned long len); extern int protect(int fd, unsigned long addr, unsigned long len, int r, int w, int x); -extern void user_signal(int sig, union uml_pt_regs *regs); +extern void user_signal(int sig, union uml_pt_regs *regs, int pid); extern int new_mm(int from); extern void start_userspace(int cpu); +extern void get_skas_faultinfo(int pid, struct faultinfo * fi); extern long execute_syscall_skas(void *r); #endif diff --git a/arch/um/kernel/skas/include/uaccess-skas.h b/arch/um/kernel/skas/include/uaccess-skas.h index 11986c9b9dd..c35620385da 100644 --- a/arch/um/kernel/skas/include/uaccess-skas.h +++ b/arch/um/kernel/skas/include/uaccess-skas.h @@ -18,7 +18,7 @@ ((unsigned long) (addr) + (size) <= FIXADDR_USER_END) && \ ((unsigned long) (addr) + (size) >= (unsigned long)(addr)))) -static inline int verify_area_skas(int type, const void * addr, +static inline int __deprecated verify_area_skas(int type, const void * addr, unsigned long size) { return(access_ok_skas(type, addr, size) ? 0 : -EFAULT); diff --git a/arch/um/kernel/skas/process.c b/arch/um/kernel/skas/process.c index b4ffaaa8124..4dc13bc8cfd 100644 --- a/arch/um/kernel/skas/process.c +++ b/arch/um/kernel/skas/process.c @@ -4,6 +4,7 @@ */ #include <stdlib.h> +#include <string.h> #include <unistd.h> #include <errno.h> #include <signal.h> @@ -27,27 +28,37 @@ #include "chan_user.h" #include "signal_user.h" #include "registers.h" +#include "process.h" int is_skas_winch(int pid, int fd, void *data) { - if(pid != os_getpid()) + if(pid != os_getpgrp()) return(0); register_winch_irq(-1, fd, -1, data); return(1); } -static void handle_segv(int pid) +void get_skas_faultinfo(int pid, struct faultinfo * fi) { - struct ptrace_faultinfo fault; int err; - err = ptrace(PTRACE_FAULTINFO, pid, 0, &fault); + err = ptrace(PTRACE_FAULTINFO, pid, 0, fi); if(err) - panic("handle_segv - PTRACE_FAULTINFO failed, errno = %d\n", - errno); + panic("get_skas_faultinfo - PTRACE_FAULTINFO failed, " + "errno = %d\n", errno); + + /* Special handling for i386, which has different structs */ + if (sizeof(struct ptrace_faultinfo) < sizeof(struct faultinfo)) + memset((char *)fi + sizeof(struct ptrace_faultinfo), 0, + sizeof(struct faultinfo) - + sizeof(struct ptrace_faultinfo)); +} - segv(fault.addr, 0, FAULT_WRITE(fault.is_write), 1, NULL); +static void handle_segv(int pid, union uml_pt_regs * regs) +{ + get_skas_faultinfo(pid, ®s->skas.faultinfo); + segv(regs->skas.faultinfo, 0, 1, NULL); } /*To use the same value of using_sysemu as the caller, ask it that value (in local_using_sysemu)*/ @@ -163,7 +174,7 @@ void userspace(union uml_pt_regs *regs) if(WIFSTOPPED(status)){ switch(WSTOPSIG(status)){ case SIGSEGV: - handle_segv(pid); + handle_segv(pid, regs); break; case SIGTRAP + 0x80: handle_trap(pid, regs, local_using_sysemu); @@ -177,7 +188,7 @@ void userspace(union uml_pt_regs *regs) case SIGBUS: case SIGFPE: case SIGWINCH: - user_signal(WSTOPSIG(status), regs); + user_signal(WSTOPSIG(status), regs, pid); break; default: printk("userspace - child stopped with signal " @@ -249,6 +260,10 @@ int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr) sigjmp_buf **switch_buf = switch_buf_ptr; int n; + set_handler(SIGWINCH, (__sighandler_t) sig_handler, + SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGALRM, + SIGVTALRM, -1); + *fork_buf_ptr = &initial_jmpbuf; n = sigsetjmp(initial_jmpbuf, 1); if(n == 0) diff --git a/arch/um/kernel/skas/trap_user.c b/arch/um/kernel/skas/trap_user.c index 8e9b46d4702..0dee1d95c80 100644 --- a/arch/um/kernel/skas/trap_user.c +++ b/arch/um/kernel/skas/trap_user.c @@ -5,12 +5,15 @@ #include <signal.h> #include <errno.h> -#include "sysdep/ptrace.h" #include "signal_user.h" #include "user_util.h" #include "kern_util.h" #include "task.h" #include "sigcontext.h" +#include "skas.h" +#include "ptrace_user.h" +#include "sysdep/ptrace.h" +#include "sysdep/ptrace_user.h" void sig_handler_common_skas(int sig, void *sc_ptr) { @@ -31,9 +34,11 @@ void sig_handler_common_skas(int sig, void *sc_ptr) r = &TASK_REGS(get_current())->skas; save_user = r->is_user; r->is_user = 0; - r->fault_addr = SC_FAULT_ADDR(sc); - r->fault_type = SC_FAULT_TYPE(sc); - r->trap_type = SC_TRAP_TYPE(sc); + if ( sig == SIGFPE || sig == SIGSEGV || + sig == SIGBUS || sig == SIGILL || + sig == SIGTRAP ) { + GET_FAULTINFO_FROM_SC(r->faultinfo, sc); + } change_sig(SIGUSR1, 1); info = &sig_info[sig]; @@ -45,14 +50,17 @@ void sig_handler_common_skas(int sig, void *sc_ptr) r->is_user = save_user; } -void user_signal(int sig, union uml_pt_regs *regs) +extern int ptrace_faultinfo; + +void user_signal(int sig, union uml_pt_regs *regs, int pid) { struct signal_info *info; + int segv = ((sig == SIGFPE) || (sig == SIGSEGV) || (sig == SIGBUS) || + (sig == SIGILL) || (sig == SIGTRAP)); regs->skas.is_user = 1; - regs->skas.fault_addr = 0; - regs->skas.fault_type = 0; - regs->skas.trap_type = 0; + if (segv) + get_skas_faultinfo(pid, ®s->skas.faultinfo); info = &sig_info[sig]; (*info->handler)(sig, regs); diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c index 7575ec489b6..f7da9d02767 100644 --- a/arch/um/kernel/skas/uaccess.c +++ b/arch/um/kernel/skas/uaccess.c @@ -3,6 +3,7 @@ * Licensed under the GPL */ +#include "linux/compiler.h" #include "linux/stddef.h" #include "linux/kernel.h" #include "linux/string.h" @@ -61,8 +62,7 @@ static void do_buffer_op(void *jmpbuf, void *arg_ptr) void *arg; int *res; - /* Some old gccs recognize __va_copy, but not va_copy */ - __va_copy(args, *(va_list *)arg_ptr); + va_copy(args, *(va_list *)arg_ptr); addr = va_arg(args, unsigned long); len = va_arg(args, int); is_write = va_arg(args, int); diff --git a/arch/um/kernel/skas/util/Makefile b/arch/um/kernel/skas/util/Makefile index 17f5909d60f..f7b7eba8334 100644 --- a/arch/um/kernel/skas/util/Makefile +++ b/arch/um/kernel/skas/util/Makefile @@ -2,3 +2,4 @@ hostprogs-y := mk_ptregs always := $(hostprogs-y) mk_ptregs-objs := mk_ptregs-$(SUBARCH).o +HOSTCFLAGS_mk_ptregs-$(SUBARCH).o := -I$(objtree)/arch/um diff --git a/arch/um/kernel/skas/util/mk_ptregs-i386.c b/arch/um/kernel/skas/util/mk_ptregs-i386.c index 0788dd05bca..1f96e1eeb8a 100644 --- a/arch/um/kernel/skas/util/mk_ptregs-i386.c +++ b/arch/um/kernel/skas/util/mk_ptregs-i386.c @@ -1,8 +1,7 @@ #include <stdio.h> -#include <asm/ptrace.h> -#include <asm/user.h> +#include <user-offsets.h> -#define PRINT_REG(name, val) printf("#define HOST_%s %d\n", (name), (val)) +#define SHOW(name) printf("#define %s %d\n", #name, name) int main(int argc, char **argv) { @@ -12,28 +11,27 @@ int main(int argc, char **argv) printf("#ifndef __SKAS_PT_REGS_\n"); printf("#define __SKAS_PT_REGS_\n"); printf("\n"); - printf("#define HOST_FRAME_SIZE %d\n", FRAME_SIZE); - printf("#define HOST_FP_SIZE %d\n", - sizeof(struct user_i387_struct) / sizeof(unsigned long)); - printf("#define HOST_XFP_SIZE %d\n", - sizeof(struct user_fxsr_struct) / sizeof(unsigned long)); + SHOW(HOST_FRAME_SIZE); + SHOW(HOST_FP_SIZE); + SHOW(HOST_XFP_SIZE); + + SHOW(HOST_IP); + SHOW(HOST_SP); + SHOW(HOST_EFLAGS); + SHOW(HOST_EAX); + SHOW(HOST_EBX); + SHOW(HOST_ECX); + SHOW(HOST_EDX); + SHOW(HOST_ESI); + SHOW(HOST_EDI); + SHOW(HOST_EBP); + SHOW(HOST_CS); + SHOW(HOST_SS); + SHOW(HOST_DS); + SHOW(HOST_FS); + SHOW(HOST_ES); + SHOW(HOST_GS); - PRINT_REG("IP", EIP); - PRINT_REG("SP", UESP); - PRINT_REG("EFLAGS", EFL); - PRINT_REG("EAX", EAX); - PRINT_REG("EBX", EBX); - PRINT_REG("ECX", ECX); - PRINT_REG("EDX", EDX); - PRINT_REG("ESI", ESI); - PRINT_REG("EDI", EDI); - PRINT_REG("EBP", EBP); - PRINT_REG("CS", CS); - PRINT_REG("SS", SS); - PRINT_REG("DS", DS); - PRINT_REG("FS", FS); - PRINT_REG("ES", ES); - PRINT_REG("GS", GS); printf("\n"); printf("#endif\n"); return(0); diff --git a/arch/um/kernel/skas/util/mk_ptregs-x86_64.c b/arch/um/kernel/skas/util/mk_ptregs-x86_64.c index 67aee92a70e..5fccbfe35f7 100644 --- a/arch/um/kernel/skas/util/mk_ptregs-x86_64.c +++ b/arch/um/kernel/skas/util/mk_ptregs-x86_64.c @@ -5,11 +5,10 @@ */ #include <stdio.h> -#define __FRAME_OFFSETS -#include <asm/ptrace.h> +#include <user-offsets.h> -#define PRINT_REG(name, val) \ - printf("#define HOST_%s (%d / sizeof(unsigned long))\n", (name), (val)) +#define SHOW(name) \ + printf("#define %s (%d / sizeof(unsigned long))\n", #name, name) int main(int argc, char **argv) { @@ -18,36 +17,35 @@ int main(int argc, char **argv) printf("\n"); printf("#ifndef __SKAS_PT_REGS_\n"); printf("#define __SKAS_PT_REGS_\n"); - printf("#define HOST_FRAME_SIZE (%d / sizeof(unsigned long))\n", - FRAME_SIZE); - PRINT_REG("RBX", RBX); - PRINT_REG("RCX", RCX); - PRINT_REG("RDI", RDI); - PRINT_REG("RSI", RSI); - PRINT_REG("RDX", RDX); - PRINT_REG("RBP", RBP); - PRINT_REG("RAX", RAX); - PRINT_REG("R8", R8); - PRINT_REG("R9", R9); - PRINT_REG("R10", R10); - PRINT_REG("R11", R11); - PRINT_REG("R12", R12); - PRINT_REG("R13", R13); - PRINT_REG("R14", R14); - PRINT_REG("R15", R15); - PRINT_REG("ORIG_RAX", ORIG_RAX); - PRINT_REG("CS", CS); - PRINT_REG("SS", SS); - PRINT_REG("EFLAGS", EFLAGS); + SHOW(HOST_FRAME_SIZE); + SHOW(HOST_RBX); + SHOW(HOST_RCX); + SHOW(HOST_RDI); + SHOW(HOST_RSI); + SHOW(HOST_RDX); + SHOW(HOST_RBP); + SHOW(HOST_RAX); + SHOW(HOST_R8); + SHOW(HOST_R9); + SHOW(HOST_R10); + SHOW(HOST_R11); + SHOW(HOST_R12); + SHOW(HOST_R13); + SHOW(HOST_R14); + SHOW(HOST_R15); + SHOW(HOST_ORIG_RAX); + SHOW(HOST_CS); + SHOW(HOST_SS); + SHOW(HOST_EFLAGS); #if 0 - PRINT_REG("FS", FS); - PRINT_REG("GS", GS); - PRINT_REG("DS", DS); - PRINT_REG("ES", ES); + SHOW(HOST_FS); + SHOW(HOST_GS); + SHOW(HOST_DS); + SHOW(HOST_ES); #endif - PRINT_REG("IP", RIP); - PRINT_REG("SP", RSP); + SHOW(HOST_IP); + SHOW(HOST_SP); printf("#define HOST_FP_SIZE 0\n"); printf("#define HOST_XFP_SIZE 0\n"); printf("\n"); diff --git a/arch/um/kernel/sys_call_table.c b/arch/um/kernel/sys_call_table.c deleted file mode 100644 index 7fc06c85b29..00000000000 --- a/arch/um/kernel/sys_call_table.c +++ /dev/null @@ -1,276 +0,0 @@ -/* - * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) - * Copyright 2003 PathScale, Inc. - * Licensed under the GPL - */ - -#include "linux/config.h" -#include "linux/unistd.h" -#include "linux/sys.h" -#include "linux/swap.h" -#include "linux/syscalls.h" -#include "linux/sysctl.h" -#include "asm/signal.h" -#include "sysdep/syscalls.h" -#include "kern_util.h" - -#ifdef CONFIG_NFSD -#define NFSSERVCTL sys_nfsservctl -#else -#define NFSSERVCTL sys_ni_syscall -#endif - -#define LAST_GENERIC_SYSCALL __NR_keyctl - -#if LAST_GENERIC_SYSCALL > LAST_ARCH_SYSCALL -#define LAST_SYSCALL LAST_GENERIC_SYSCALL -#else -#define LAST_SYSCALL LAST_ARCH_SYSCALL -#endif - -extern syscall_handler_t sys_fork; -extern syscall_handler_t sys_execve; -extern syscall_handler_t um_time; -extern syscall_handler_t um_stime; -extern syscall_handler_t sys_pipe; -extern syscall_handler_t sys_olduname; -extern syscall_handler_t sys_sigaction; -extern syscall_handler_t sys_sigsuspend; -extern syscall_handler_t old_readdir; -extern syscall_handler_t sys_uname; -extern syscall_handler_t sys_ipc; -extern syscall_handler_t sys_sigreturn; -extern syscall_handler_t sys_clone; -extern syscall_handler_t sys_rt_sigreturn; -extern syscall_handler_t sys_sigaltstack; -extern syscall_handler_t sys_vfork; -extern syscall_handler_t old_select; -extern syscall_handler_t sys_modify_ldt; -extern syscall_handler_t sys_rt_sigsuspend; -extern syscall_handler_t sys_mbind; -extern syscall_handler_t sys_get_mempolicy; -extern syscall_handler_t sys_set_mempolicy; -extern syscall_handler_t sys_sys_setaltroot; - -syscall_handler_t *sys_call_table[] = { - [ __NR_restart_syscall ] = (syscall_handler_t *) sys_restart_syscall, - [ __NR_exit ] = (syscall_handler_t *) sys_exit, - [ __NR_fork ] = (syscall_handler_t *) sys_fork, - [ __NR_read ] = (syscall_handler_t *) sys_read, - [ __NR_write ] = (syscall_handler_t *) sys_write, - - /* These three are declared differently in asm/unistd.h */ - [ __NR_open ] = (syscall_handler_t *) sys_open, - [ __NR_close ] = (syscall_handler_t *) sys_close, - [ __NR_creat ] = (syscall_handler_t *) sys_creat, - [ __NR_link ] = (syscall_handler_t *) sys_link, - [ __NR_unlink ] = (syscall_handler_t *) sys_unlink, - [ __NR_execve ] = (syscall_handler_t *) sys_execve, - - /* declared differently in kern_util.h */ - [ __NR_chdir ] = (syscall_handler_t *) sys_chdir, - [ __NR_time ] = um_time, - [ __NR_mknod ] = (syscall_handler_t *) sys_mknod, - [ __NR_chmod ] = (syscall_handler_t *) sys_chmod, - [ __NR_lchown ] = (syscall_handler_t *) sys_lchown16, - [ __NR_lseek ] = (syscall_handler_t *) sys_lseek, - [ __NR_getpid ] = (syscall_handler_t *) sys_getpid, - [ __NR_mount ] = (syscall_handler_t *) sys_mount, - [ __NR_setuid ] = (syscall_handler_t *) sys_setuid16, - [ __NR_getuid ] = (syscall_handler_t *) sys_getuid16, - [ __NR_ptrace ] = (syscall_handler_t *) sys_ptrace, - [ __NR_alarm ] = (syscall_handler_t *) sys_alarm, - [ __NR_pause ] = (syscall_handler_t *) sys_pause, - [ __NR_utime ] = (syscall_handler_t *) sys_utime, - [ __NR_access ] = (syscall_handler_t *) sys_access, - [ __NR_sync ] = (syscall_handler_t *) sys_sync, - [ __NR_kill ] = (syscall_handler_t *) sys_kill, - [ __NR_rename ] = (syscall_handler_t *) sys_rename, - [ __NR_mkdir ] = (syscall_handler_t *) sys_mkdir, - [ __NR_rmdir ] = (syscall_handler_t *) sys_rmdir, - - /* Declared differently in asm/unistd.h */ - [ __NR_dup ] = (syscall_handler_t *) sys_dup, - [ __NR_pipe ] = (syscall_handler_t *) sys_pipe, - [ __NR_times ] = (syscall_handler_t *) sys_times, - [ __NR_brk ] = (syscall_handler_t *) sys_brk, - [ __NR_setgid ] = (syscall_handler_t *) sys_setgid16, - [ __NR_getgid ] = (syscall_handler_t *) sys_getgid16, - [ __NR_geteuid ] = (syscall_handler_t *) sys_geteuid16, - [ __NR_getegid ] = (syscall_handler_t *) sys_getegid16, - [ __NR_acct ] = (syscall_handler_t *) sys_acct, - [ __NR_umount2 ] = (syscall_handler_t *) sys_umount, - [ __NR_ioctl ] = (syscall_handler_t *) sys_ioctl, - [ __NR_fcntl ] = (syscall_handler_t *) sys_fcntl, - [ __NR_setpgid ] = (syscall_handler_t *) sys_setpgid, - [ __NR_umask ] = (syscall_handler_t *) sys_umask, - [ __NR_chroot ] = (syscall_handler_t *) sys_chroot, - [ __NR_ustat ] = (syscall_handler_t *) sys_ustat, - [ __NR_dup2 ] = (syscall_handler_t *) sys_dup2, - [ __NR_getppid ] = (syscall_handler_t *) sys_getppid, - [ __NR_getpgrp ] = (syscall_handler_t *) sys_getpgrp, - [ __NR_setsid ] = (syscall_handler_t *) sys_setsid, - [ __NR_setreuid ] = (syscall_handler_t *) sys_setreuid16, - [ __NR_setregid ] = (syscall_handler_t *) sys_setregid16, - [ __NR_sethostname ] = (syscall_handler_t *) sys_sethostname, - [ __NR_setrlimit ] = (syscall_handler_t *) sys_setrlimit, - [ __NR_getrlimit ] = (syscall_handler_t *) sys_old_getrlimit, - [ __NR_getrusage ] = (syscall_handler_t *) sys_getrusage, - [ __NR_gettimeofday ] = (syscall_handler_t *) sys_gettimeofday, - [ __NR_settimeofday ] = (syscall_handler_t *) sys_settimeofday, - [ __NR_getgroups ] = (syscall_handler_t *) sys_getgroups16, - [ __NR_setgroups ] = (syscall_handler_t *) sys_setgroups16, - [ __NR_symlink ] = (syscall_handler_t *) sys_symlink, - [ __NR_readlink ] = (syscall_handler_t *) sys_readlink, - [ __NR_uselib ] = (syscall_handler_t *) sys_uselib, - [ __NR_swapon ] = (syscall_handler_t *) sys_swapon, - [ __NR_reboot ] = (syscall_handler_t *) sys_reboot, - [ __NR_munmap ] = (syscall_handler_t *) sys_munmap, - [ __NR_truncate ] = (syscall_handler_t *) sys_truncate, - [ __NR_ftruncate ] = (syscall_handler_t *) sys_ftruncate, - [ __NR_fchmod ] = (syscall_handler_t *) sys_fchmod, - [ __NR_fchown ] = (syscall_handler_t *) sys_fchown16, - [ __NR_getpriority ] = (syscall_handler_t *) sys_getpriority, - [ __NR_setpriority ] = (syscall_handler_t *) sys_setpriority, - [ __NR_statfs ] = (syscall_handler_t *) sys_statfs, - [ __NR_fstatfs ] = (syscall_handler_t *) sys_fstatfs, - [ __NR_ioperm ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_syslog ] = (syscall_handler_t *) sys_syslog, - [ __NR_setitimer ] = (syscall_handler_t *) sys_setitimer, - [ __NR_getitimer ] = (syscall_handler_t *) sys_getitimer, - [ __NR_stat ] = (syscall_handler_t *) sys_newstat, - [ __NR_lstat ] = (syscall_handler_t *) sys_newlstat, - [ __NR_fstat ] = (syscall_handler_t *) sys_newfstat, - [ __NR_vhangup ] = (syscall_handler_t *) sys_vhangup, - [ __NR_wait4 ] = (syscall_handler_t *) sys_wait4, - [ __NR_swapoff ] = (syscall_handler_t *) sys_swapoff, - [ __NR_sysinfo ] = (syscall_handler_t *) sys_sysinfo, - [ __NR_fsync ] = (syscall_handler_t *) sys_fsync, - [ __NR_clone ] = (syscall_handler_t *) sys_clone, - [ __NR_setdomainname ] = (syscall_handler_t *) sys_setdomainname, - [ __NR_uname ] = (syscall_handler_t *) sys_newuname, - [ __NR_adjtimex ] = (syscall_handler_t *) sys_adjtimex, - [ __NR_mprotect ] = (syscall_handler_t *) sys_mprotect, - [ __NR_create_module ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_init_module ] = (syscall_handler_t *) sys_init_module, - [ __NR_delete_module ] = (syscall_handler_t *) sys_delete_module, - [ __NR_get_kernel_syms ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_quotactl ] = (syscall_handler_t *) sys_quotactl, - [ __NR_getpgid ] = (syscall_handler_t *) sys_getpgid, - [ __NR_fchdir ] = (syscall_handler_t *) sys_fchdir, - [ __NR_sysfs ] = (syscall_handler_t *) sys_sysfs, - [ __NR_personality ] = (syscall_handler_t *) sys_personality, - [ __NR_afs_syscall ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_setfsuid ] = (syscall_handler_t *) sys_setfsuid16, - [ __NR_setfsgid ] = (syscall_handler_t *) sys_setfsgid16, - [ __NR_getdents ] = (syscall_handler_t *) sys_getdents, - [ __NR_flock ] = (syscall_handler_t *) sys_flock, - [ __NR_msync ] = (syscall_handler_t *) sys_msync, - [ __NR_readv ] = (syscall_handler_t *) sys_readv, - [ __NR_writev ] = (syscall_handler_t *) sys_writev, - [ __NR_getsid ] = (syscall_handler_t *) sys_getsid, - [ __NR_fdatasync ] = (syscall_handler_t *) sys_fdatasync, - [ __NR__sysctl ] = (syscall_handler_t *) sys_sysctl, - [ __NR_mlock ] = (syscall_handler_t *) sys_mlock, - [ __NR_munlock ] = (syscall_handler_t *) sys_munlock, - [ __NR_mlockall ] = (syscall_handler_t *) sys_mlockall, - [ __NR_munlockall ] = (syscall_handler_t *) sys_munlockall, - [ __NR_sched_setparam ] = (syscall_handler_t *) sys_sched_setparam, - [ __NR_sched_getparam ] = (syscall_handler_t *) sys_sched_getparam, - [ __NR_sched_setscheduler ] = (syscall_handler_t *) sys_sched_setscheduler, - [ __NR_sched_getscheduler ] = (syscall_handler_t *) sys_sched_getscheduler, - [ __NR_sched_yield ] = (syscall_handler_t *) yield, - [ __NR_sched_get_priority_max ] = (syscall_handler_t *) sys_sched_get_priority_max, - [ __NR_sched_get_priority_min ] = (syscall_handler_t *) sys_sched_get_priority_min, - [ __NR_sched_rr_get_interval ] = (syscall_handler_t *) sys_sched_rr_get_interval, - [ __NR_nanosleep ] = (syscall_handler_t *) sys_nanosleep, - [ __NR_mremap ] = (syscall_handler_t *) sys_mremap, - [ __NR_setresuid ] = (syscall_handler_t *) sys_setresuid16, - [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid16, - [ __NR_query_module ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_poll ] = (syscall_handler_t *) sys_poll, - [ __NR_nfsservctl ] = (syscall_handler_t *) NFSSERVCTL, - [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid16, - [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid16, - [ __NR_prctl ] = (syscall_handler_t *) sys_prctl, - [ __NR_rt_sigreturn ] = (syscall_handler_t *) sys_rt_sigreturn, - [ __NR_rt_sigaction ] = (syscall_handler_t *) sys_rt_sigaction, - [ __NR_rt_sigprocmask ] = (syscall_handler_t *) sys_rt_sigprocmask, - [ __NR_rt_sigpending ] = (syscall_handler_t *) sys_rt_sigpending, - [ __NR_rt_sigtimedwait ] = (syscall_handler_t *) sys_rt_sigtimedwait, - [ __NR_rt_sigqueueinfo ] = (syscall_handler_t *) sys_rt_sigqueueinfo, - [ __NR_rt_sigsuspend ] = (syscall_handler_t *) sys_rt_sigsuspend, - [ __NR_pread64 ] = (syscall_handler_t *) sys_pread64, - [ __NR_pwrite64 ] = (syscall_handler_t *) sys_pwrite64, - [ __NR_chown ] = (syscall_handler_t *) sys_chown16, - [ __NR_getcwd ] = (syscall_handler_t *) sys_getcwd, - [ __NR_capget ] = (syscall_handler_t *) sys_capget, - [ __NR_capset ] = (syscall_handler_t *) sys_capset, - [ __NR_sigaltstack ] = (syscall_handler_t *) sys_sigaltstack, - [ __NR_sendfile ] = (syscall_handler_t *) sys_sendfile, - [ __NR_getpmsg ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_putpmsg ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_vfork ] = (syscall_handler_t *) sys_vfork, - [ __NR_getdents64 ] = (syscall_handler_t *) sys_getdents64, - [ __NR_gettid ] = (syscall_handler_t *) sys_gettid, - [ __NR_readahead ] = (syscall_handler_t *) sys_readahead, - [ __NR_setxattr ] = (syscall_handler_t *) sys_setxattr, - [ __NR_lsetxattr ] = (syscall_handler_t *) sys_lsetxattr, - [ __NR_fsetxattr ] = (syscall_handler_t *) sys_fsetxattr, - [ __NR_getxattr ] = (syscall_handler_t *) sys_getxattr, - [ __NR_lgetxattr ] = (syscall_handler_t *) sys_lgetxattr, - [ __NR_fgetxattr ] = (syscall_handler_t *) sys_fgetxattr, - [ __NR_listxattr ] = (syscall_handler_t *) sys_listxattr, - [ __NR_llistxattr ] = (syscall_handler_t *) sys_llistxattr, - [ __NR_flistxattr ] = (syscall_handler_t *) sys_flistxattr, - [ __NR_removexattr ] = (syscall_handler_t *) sys_removexattr, - [ __NR_lremovexattr ] = (syscall_handler_t *) sys_lremovexattr, - [ __NR_fremovexattr ] = (syscall_handler_t *) sys_fremovexattr, - [ __NR_tkill ] = (syscall_handler_t *) sys_tkill, - [ __NR_futex ] = (syscall_handler_t *) sys_futex, - [ __NR_sched_setaffinity ] = (syscall_handler_t *) sys_sched_setaffinity, - [ __NR_sched_getaffinity ] = (syscall_handler_t *) sys_sched_getaffinity, - [ __NR_io_setup ] = (syscall_handler_t *) sys_io_setup, - [ __NR_io_destroy ] = (syscall_handler_t *) sys_io_destroy, - [ __NR_io_getevents ] = (syscall_handler_t *) sys_io_getevents, - [ __NR_io_submit ] = (syscall_handler_t *) sys_io_submit, - [ __NR_io_cancel ] = (syscall_handler_t *) sys_io_cancel, - [ __NR_exit_group ] = (syscall_handler_t *) sys_exit_group, - [ __NR_lookup_dcookie ] = (syscall_handler_t *) sys_lookup_dcookie, - [ __NR_epoll_create ] = (syscall_handler_t *) sys_epoll_create, - [ __NR_epoll_ctl ] = (syscall_handler_t *) sys_epoll_ctl, - [ __NR_epoll_wait ] = (syscall_handler_t *) sys_epoll_wait, - [ __NR_remap_file_pages ] = (syscall_handler_t *) sys_remap_file_pages, - [ __NR_set_tid_address ] = (syscall_handler_t *) sys_set_tid_address, - [ __NR_timer_create ] = (syscall_handler_t *) sys_timer_create, - [ __NR_timer_settime ] = (syscall_handler_t *) sys_timer_settime, - [ __NR_timer_gettime ] = (syscall_handler_t *) sys_timer_gettime, - [ __NR_timer_getoverrun ] = (syscall_handler_t *) sys_timer_getoverrun, - [ __NR_timer_delete ] = (syscall_handler_t *) sys_timer_delete, - [ __NR_clock_settime ] = (syscall_handler_t *) sys_clock_settime, - [ __NR_clock_gettime ] = (syscall_handler_t *) sys_clock_gettime, - [ __NR_clock_getres ] = (syscall_handler_t *) sys_clock_getres, - [ __NR_clock_nanosleep ] = (syscall_handler_t *) sys_clock_nanosleep, - [ __NR_tgkill ] = (syscall_handler_t *) sys_tgkill, - [ __NR_utimes ] = (syscall_handler_t *) sys_utimes, - [ __NR_fadvise64 ] = (syscall_handler_t *) sys_fadvise64, - [ __NR_vserver ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_mbind ] = (syscall_handler_t *) sys_mbind, - [ __NR_get_mempolicy ] = (syscall_handler_t *) sys_get_mempolicy, - [ __NR_set_mempolicy ] = (syscall_handler_t *) sys_set_mempolicy, - [ __NR_mq_open ] = (syscall_handler_t *) sys_mq_open, - [ __NR_mq_unlink ] = (syscall_handler_t *) sys_mq_unlink, - [ __NR_mq_timedsend ] = (syscall_handler_t *) sys_mq_timedsend, - [ __NR_mq_timedreceive ] = (syscall_handler_t *) sys_mq_timedreceive, - [ __NR_mq_notify ] = (syscall_handler_t *) sys_mq_notify, - [ __NR_mq_getsetattr ] = (syscall_handler_t *) sys_mq_getsetattr, - [ __NR_kexec_load ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_waitid ] = (syscall_handler_t *) sys_waitid, - [ __NR_add_key ] = (syscall_handler_t *) sys_add_key, - [ __NR_request_key ] = (syscall_handler_t *) sys_request_key, - [ __NR_keyctl ] = (syscall_handler_t *) sys_keyctl, - - ARCH_SYSCALLS - [ LAST_SYSCALL + 1 ... NR_syscalls ] = - (syscall_handler_t *) sys_ni_syscall -}; diff --git a/arch/um/kernel/time_kern.c b/arch/um/kernel/time_kern.c index 2461cd73ca8..6516fc52afe 100644 --- a/arch/um/kernel/time_kern.c +++ b/arch/um/kernel/time_kern.c @@ -48,8 +48,6 @@ static unsigned long long prev_usecs; static long long delta; /* Deviation per interval */ #endif -#define MILLION 1000000 - void timer_irq(union uml_pt_regs *regs) { unsigned long long ticks = 0; @@ -136,22 +134,6 @@ long um_stime(int __user *tptr) return 0; } -void __udelay(unsigned long usecs) -{ - int i, n; - - n = (loops_per_jiffy * HZ * usecs) / MILLION; - for(i=0;i<n;i++) ; -} - -void __const_udelay(unsigned long usecs) -{ - int i, n; - - n = (loops_per_jiffy * HZ * usecs) / MILLION; - for(i=0;i<n;i++) ; -} - void timer_handler(int sig, union uml_pt_regs *regs) { local_irq_disable(); diff --git a/arch/um/kernel/trap_kern.c b/arch/um/kernel/trap_kern.c index 47e766e6ba1..54e2ec33a43 100644 --- a/arch/um/kernel/trap_kern.c +++ b/arch/um/kernel/trap_kern.c @@ -133,12 +133,19 @@ static int check_remapped_addr(unsigned long address, int is_write) return(0); } -unsigned long segv(unsigned long address, unsigned long ip, int is_write, - int is_user, void *sc) +/* + * We give a *copy* of the faultinfo in the regs to segv. + * This must be done, since nesting SEGVs could overwrite + * the info in the regs. A pointer to the info then would + * give us bad data! + */ +unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, void *sc) { struct siginfo si; void *catcher; int err; + int is_write = FAULT_WRITE(fi); + unsigned long address = FAULT_ADDRESS(fi); if(!is_user && (address >= start_vm) && (address < end_vm)){ flush_tlb_kernel_vm(); @@ -159,7 +166,7 @@ unsigned long segv(unsigned long address, unsigned long ip, int is_write, } else if(current->thread.fault_addr != NULL) panic("fault_addr set but no fault catcher"); - else if(arch_fixup(ip, sc)) + else if(!is_user && arch_fixup(ip, sc)) return(0); if(!is_user) @@ -171,6 +178,7 @@ unsigned long segv(unsigned long address, unsigned long ip, int is_write, si.si_errno = 0; si.si_code = BUS_ADRERR; si.si_addr = (void *)address; + current->thread.arch.faultinfo = fi; force_sig_info(SIGBUS, &si, current); } else if(err == -ENOMEM){ @@ -180,22 +188,20 @@ unsigned long segv(unsigned long address, unsigned long ip, int is_write, else { si.si_signo = SIGSEGV; si.si_addr = (void *) address; - current->thread.cr2 = address; - current->thread.err = is_write; + current->thread.arch.faultinfo = fi; force_sig_info(SIGSEGV, &si, current); } return(0); } -void bad_segv(unsigned long address, unsigned long ip, int is_write) +void bad_segv(struct faultinfo fi, unsigned long ip) { struct siginfo si; si.si_signo = SIGSEGV; si.si_code = SEGV_ACCERR; - si.si_addr = (void *) address; - current->thread.cr2 = address; - current->thread.err = is_write; + si.si_addr = (void *) FAULT_ADDRESS(fi); + current->thread.arch.faultinfo = fi; force_sig_info(SIGSEGV, &si, current); } @@ -204,6 +210,7 @@ void relay_signal(int sig, union uml_pt_regs *regs) if(arch_handle_signal(sig, regs)) return; if(!UPT_IS_USER(regs)) panic("Kernel mode signal %d", sig); + current->thread.arch.faultinfo = *UPT_FAULTINFO(regs); force_sig(sig, current); } diff --git a/arch/um/kernel/trap_user.c b/arch/um/kernel/trap_user.c index 50a4042a509..f825a6eda3f 100644 --- a/arch/um/kernel/trap_user.c +++ b/arch/um/kernel/trap_user.c @@ -54,23 +54,22 @@ struct { void segv_handler(int sig, union uml_pt_regs *regs) { int index, max; + struct faultinfo * fi = UPT_FAULTINFO(regs); - if(UPT_IS_USER(regs) && !UPT_SEGV_IS_FIXABLE(regs)){ - bad_segv(UPT_FAULT_ADDR(regs), UPT_IP(regs), - UPT_FAULT_WRITE(regs)); + if(UPT_IS_USER(regs) && !SEGV_IS_FIXABLE(fi)){ + bad_segv(*fi, UPT_IP(regs)); return; } max = sizeof(segfault_record)/sizeof(segfault_record[0]); index = next_trap_index(max); nsegfaults++; - segfault_record[index].address = UPT_FAULT_ADDR(regs); + segfault_record[index].address = FAULT_ADDRESS(*fi); segfault_record[index].pid = os_getpid(); - segfault_record[index].is_write = UPT_FAULT_WRITE(regs); + segfault_record[index].is_write = FAULT_WRITE(*fi); segfault_record[index].sp = UPT_SP(regs); segfault_record[index].is_user = UPT_IS_USER(regs); - segv(UPT_FAULT_ADDR(regs), UPT_IP(regs), UPT_FAULT_WRITE(regs), - UPT_IS_USER(regs), regs); + segv(*fi, UPT_IP(regs), UPT_IS_USER(regs), regs); } void usr2_handler(int sig, union uml_pt_regs *regs) diff --git a/arch/um/kernel/tt/Makefile b/arch/um/kernel/tt/Makefile index 3d5177df350..c3faea21a99 100644 --- a/arch/um/kernel/tt/Makefile +++ b/arch/um/kernel/tt/Makefile @@ -4,6 +4,7 @@ # extra-y := unmap_fin.o +targets := unmap.o clean-files := unmap_tmp.o obj-y = exec_kern.o exec_user.o gdb.o ksyms.o mem.o mem_user.o process_kern.o \ diff --git a/arch/um/kernel/tt/include/uaccess-tt.h b/arch/um/kernel/tt/include/uaccess-tt.h index f0bad010ceb..bb69d6b7d02 100644 --- a/arch/um/kernel/tt/include/uaccess-tt.h +++ b/arch/um/kernel/tt/include/uaccess-tt.h @@ -33,7 +33,7 @@ extern unsigned long uml_physmem; (((unsigned long) (addr) <= ((unsigned long) (addr) + (size))) && \ (under_task_size(addr, size) || is_stack(addr, size)))) -static inline int verify_area_tt(int type, const void * addr, +static inline int __deprecated verify_area_tt(int type, const void * addr, unsigned long size) { return(access_ok_tt(type, addr, size) ? 0 : -EFAULT); diff --git a/arch/um/kernel/tt/tracer.c b/arch/um/kernel/tt/tracer.c index 7b5d937e595..d11e7399d7a 100644 --- a/arch/um/kernel/tt/tracer.c +++ b/arch/um/kernel/tt/tracer.c @@ -26,6 +26,7 @@ #include "kern_util.h" #include "chan_user.h" #include "ptrace_user.h" +#include "irq_user.h" #include "mode.h" #include "tt.h" @@ -33,7 +34,7 @@ static int tracer_winch[2]; int is_tracer_winch(int pid, int fd, void *data) { - if(pid != tracing_pid) + if(pid != os_getpgrp()) return(0); register_winch_irq(tracer_winch[0], fd, -1, data); @@ -89,8 +90,10 @@ void tracer_panic(char *format, ...) static void tracer_segv(int sig, struct sigcontext sc) { + struct faultinfo fi; + GET_FAULTINFO_FROM_SC(fi, &sc); printf("Tracing thread segfault at address 0x%lx, ip 0x%lx\n", - SC_FAULT_ADDR(&sc), SC_IP(&sc)); + FAULT_ADDRESS(fi), SC_IP(&sc)); while(1) pause(); } @@ -117,6 +120,7 @@ static int signal_tramp(void *arg) signal(SIGSEGV, (__sighandler_t) sig_handler); set_cmdline("(idle thread)"); set_init_pid(os_getpid()); + init_irq_signals(0); proc = arg; return((*proc)(NULL)); } diff --git a/arch/um/kernel/tt/trap_user.c b/arch/um/kernel/tt/trap_user.c index 92a3820ca54..fc108615bea 100644 --- a/arch/um/kernel/tt/trap_user.c +++ b/arch/um/kernel/tt/trap_user.c @@ -7,6 +7,7 @@ #include <errno.h> #include <signal.h> #include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" #include "signal_user.h" #include "user_util.h" #include "kern_util.h" @@ -28,6 +29,11 @@ void sig_handler_common_tt(int sig, void *sc_ptr) change_sig(SIGSEGV, 1); r = &TASK_REGS(get_current())->tt; + if ( sig == SIGFPE || sig == SIGSEGV || + sig == SIGBUS || sig == SIGILL || + sig == SIGTRAP ) { + GET_FAULTINFO_FROM_SC(r->faultinfo, sc); + } save_regs = *r; is_user = user_context(SC_SP(sc)); r->sc = sc; diff --git a/arch/um/kernel/vmlinux.lds.S b/arch/um/kernel/vmlinux.lds.S new file mode 100644 index 00000000000..1660a769674 --- /dev/null +++ b/arch/um/kernel/vmlinux.lds.S @@ -0,0 +1,6 @@ +#include <linux/config.h> +#ifdef CONFIG_LD_SCRIPT_STATIC +#include "uml.lds.S" +#else +#include "dyn.lds.S" +#endif diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c index ba9ca1cc790..1e126bfd31a 100644 --- a/arch/um/os-Linux/process.c +++ b/arch/um/os-Linux/process.c @@ -123,6 +123,11 @@ int os_getpid(void) return(getpid()); } +int os_getpgrp(void) +{ + return getpgrp(); +} + int os_map_memory(void *virt, int fd, unsigned long long off, unsigned long len, int r, int w, int x) { diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c index 7eac1baf597..c7bfd5ee392 100644 --- a/arch/um/os-Linux/signal.c +++ b/arch/um/os-Linux/signal.c @@ -8,7 +8,7 @@ #include "mode.h" #include "sysdep/signal.h" -void sig_handler(int sig) +void sig_handler(ARCH_SIGHDLR_PARAM) { struct sigcontext *sc; @@ -19,7 +19,7 @@ void sig_handler(int sig) extern int timer_irq_inited; -void alarm_handler(int sig) +void alarm_handler(ARCH_SIGHDLR_PARAM) { struct sigcontext *sc; diff --git a/arch/um/os-Linux/sys-i386/registers.c b/arch/um/os-Linux/sys-i386/registers.c index 148645b1448..9a0ad094d92 100644 --- a/arch/um/os-Linux/sys-i386/registers.c +++ b/arch/um/os-Linux/sys-i386/registers.c @@ -105,14 +105,15 @@ void init_registers(int pid) panic("check_ptrace : PTRACE_GETREGS failed, errno = %d", err); + errno = 0; err = ptrace(PTRACE_GETFPXREGS, pid, 0, exec_fpx_regs); if(!err) return; + if(errno != EIO) + panic("check_ptrace : PTRACE_GETFPXREGS failed, errno = %d", + errno); have_fpx_regs = 0; - if(err != EIO) - panic("check_ptrace : PTRACE_GETFPXREGS failed, errno = %d", - err); err = ptrace(PTRACE_GETFPREGS, pid, 0, exec_fp_regs); if(err) diff --git a/arch/um/os-Linux/util/Makefile b/arch/um/os-Linux/util/Makefile index fb00ddf969b..9778aed0c31 100644 --- a/arch/um/os-Linux/util/Makefile +++ b/arch/um/os-Linux/util/Makefile @@ -1,4 +1,4 @@ hostprogs-y := mk_user_constants always := $(hostprogs-y) -mk_user_constants-objs := mk_user_constants.o +HOSTCFLAGS_mk_user_constants.o := -I$(objtree)/arch/um diff --git a/arch/um/os-Linux/util/mk_user_constants.c b/arch/um/os-Linux/util/mk_user_constants.c index 0933518aa8b..4838f30eecf 100644 --- a/arch/um/os-Linux/util/mk_user_constants.c +++ b/arch/um/os-Linux/util/mk_user_constants.c @@ -1,11 +1,5 @@ #include <stdio.h> -#include <asm/types.h> -/* For some reason, x86_64 nowhere defines u64 and u32, even though they're - * used throughout the headers. - */ -typedef __u64 u64; -typedef __u32 u32; -#include <asm/user.h> +#include <user-offsets.h> int main(int argc, char **argv) { @@ -20,7 +14,7 @@ int main(int argc, char **argv) * x86_64 (216 vs 168 bytes). user_regs_struct is the correct size on * both x86_64 and i386. */ - printf("#define UM_FRAME_SIZE %d\n", (int) sizeof(struct user_regs_struct)); + printf("#define UM_FRAME_SIZE %d\n", __UM_FRAME_SIZE); printf("\n"); printf("#endif\n"); diff --git a/arch/um/scripts/Makefile.rules b/arch/um/scripts/Makefile.rules index 143f6fea076..0b2491883d9 100644 --- a/arch/um/scripts/Makefile.rules +++ b/arch/um/scripts/Makefile.rules @@ -2,12 +2,27 @@ # arch/um: Generic definitions # =========================================================================== -USER_SINGLE_OBJS = $(foreach f,$(patsubst %.o,%,$(obj-y) $(obj-m)),$($(f)-objs)) -USER_OBJS += $(filter %_user.o,$(obj-y) $(obj-m) $(USER_SINGLE_OBJS)) - +USER_SINGLE_OBJS := \ + $(foreach f,$(patsubst %.o,%,$(obj-y) $(obj-m)),$($(f)-objs)) +USER_OBJS += $(filter %_user.o,$(obj-y) $(obj-m) $(USER_SINGLE_OBJS)) USER_OBJS := $(foreach file,$(USER_OBJS),$(obj)/$(file)) -$(USER_OBJS): c_flags = -Wp,-MD,$(depfile) $(USER_CFLAGS) $(CFLAGS_$(notdir $@)) +$(USER_OBJS) : c_flags = -Wp,-MD,$(depfile) $(USER_CFLAGS) \ + $(CFLAGS_$(notdir $@)) quiet_cmd_make_link = SYMLINK $@ -cmd_make_link = rm -f $@; ln -sf $(srctree)/arch/$(SUBARCH)/$($(notdir $@)-dir)/$(notdir $@) $@ +cmd_make_link = ln -sf $(srctree)/arch/$(SUBARCH)/$($(notdir $@)-dir)/$(notdir $@) $@ + +# this needs to be before the foreach, because targets does not accept +# complete paths like $(obj)/$(f). To make sure this works, use a := assignment, +# or we will get $(obj)/$(f) in the "targets" value. +# Also, this forces you to use the := syntax when assigning to targets. +# Otherwise the line below will cause an infinite loop (if you don't know why, +# just do it). + +targets := $(targets) $(SYMLINKS) + +SYMLINKS := $(foreach f,$(SYMLINKS),$(obj)/$(f)) + +$(SYMLINKS): FORCE + $(call if_changed,make_link) diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile index 71b47e61860..fcd67c3414e 100644 --- a/arch/um/sys-i386/Makefile +++ b/arch/um/sys-i386/Makefile @@ -1,29 +1,19 @@ obj-y = bitops.o bugs.o checksum.o delay.o fault.o ksyms.o ldt.o ptrace.o \ - ptrace_user.o semaphore.o signal.o sigcontext.o syscalls.o sysrq.o + ptrace_user.o semaphore.o signal.o sigcontext.o syscalls.o sysrq.o \ + sys_call_table.o obj-$(CONFIG_HIGHMEM) += highmem.o obj-$(CONFIG_MODULES) += module.o USER_OBJS := bugs.o ptrace_user.o sigcontext.o fault.o -include arch/um/scripts/Makefile.rules - SYMLINKS = bitops.c semaphore.c highmem.c module.c -# this needs to be before the foreach, because clean-files does not accept -# complete paths like $(src)/$f. -clean-files := $(SYMLINKS) - -targets += $(SYMLINKS) - -SYMLINKS := $(foreach f,$(SYMLINKS),$(obj)/$f) - bitops.c-dir = lib semaphore.c-dir = kernel highmem.c-dir = mm module.c-dir = kernel -$(SYMLINKS): FORCE - $(call if_changed,make_link) - subdir- := util + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/sys-i386/checksum.S b/arch/um/sys-i386/checksum.S index a11171fb622..d98b2fff3d0 100644 --- a/arch/um/sys-i386/checksum.S +++ b/arch/um/sys-i386/checksum.S @@ -38,7 +38,7 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) .text .align 4 -.globl arch_csum_partial +.globl csum_partial #ifndef CONFIG_X86_USE_PPRO_CHECKSUM @@ -49,7 +49,7 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) * Fortunately, it is easy to convert 2-byte alignment to 4-byte * alignment for the unrolled loop. */ -arch_csum_partial: +csum_partial: pushl %esi pushl %ebx movl 20(%esp),%eax # Function arg: unsigned int sum @@ -119,7 +119,7 @@ arch_csum_partial: /* Version for PentiumII/PPro */ -arch_csum_partial: +csum_partial: pushl %esi pushl %ebx movl 20(%esp),%eax # Function arg: unsigned int sum diff --git a/arch/um/sys-i386/delay.c b/arch/um/sys-i386/delay.c index 20d37dbbaf0..e9892eef51c 100644 --- a/arch/um/sys-i386/delay.c +++ b/arch/um/sys-i386/delay.c @@ -1,3 +1,6 @@ +#include "linux/delay.h" +#include "asm/param.h" + void __delay(unsigned long time) { /* Stolen from the i386 __loop_delay */ @@ -12,3 +15,18 @@ void __delay(unsigned long time) :"0" (time)); } +void __udelay(unsigned long usecs) +{ + int i, n; + + n = (loops_per_jiffy * HZ * usecs) / MILLION; + for(i=0;i<n;i++) ; +} + +void __const_udelay(unsigned long usecs) +{ + int i, n; + + n = (loops_per_jiffy * HZ * usecs) / MILLION; + for(i=0;i<n;i++) ; +} diff --git a/arch/um/sys-i386/kernel-offsets.c b/arch/um/sys-i386/kernel-offsets.c new file mode 100644 index 00000000000..9f8ecd1fdd9 --- /dev/null +++ b/arch/um/sys-i386/kernel-offsets.c @@ -0,0 +1,25 @@ +#include <linux/config.h> +#include <linux/stddef.h> +#include <linux/sched.h> +#include <linux/time.h> +#include <asm/page.h> + +#define DEFINE(sym, val) \ + asm volatile("\n->" #sym " %0 " #val : : "i" (val)) + +#define STR(x) #x +#define DEFINE_STR(sym, val) asm volatile("\n->" #sym " " STR(val) " " #val: : ) + +#define BLANK() asm volatile("\n->" : : ) + +#define OFFSET(sym, str, mem) \ + DEFINE(sym, offsetof(struct str, mem)); + +void foo(void) +{ + OFFSET(TASK_DEBUGREGS, task_struct, thread.arch.debugregs); +#ifdef CONFIG_MODE_TT + OFFSET(TASK_EXTERN_PID, task_struct, thread.mode.tt.extern_pid); +#endif +#include <common-offsets.h> +} diff --git a/arch/um/sys-i386/ksyms.c b/arch/um/sys-i386/ksyms.c index 74f70a12045..db524ab3f74 100644 --- a/arch/um/sys-i386/ksyms.c +++ b/arch/um/sys-i386/ksyms.c @@ -2,6 +2,7 @@ #include "linux/in6.h" #include "linux/rwsem.h" #include "asm/byteorder.h" +#include "asm/delay.h" #include "asm/semaphore.h" #include "asm/uaccess.h" #include "asm/checksum.h" @@ -13,5 +14,8 @@ EXPORT_SYMBOL(__down_failed_trylock); EXPORT_SYMBOL(__up_wakeup); /* Networking helper routines. */ -EXPORT_SYMBOL(csum_partial_copy_from); -EXPORT_SYMBOL(csum_partial_copy_to); +EXPORT_SYMBOL(csum_partial); + +/* delay core functions */ +EXPORT_SYMBOL(__const_udelay); +EXPORT_SYMBOL(__udelay); diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c index 31bcb2f997d..dc755b0b9db 100644 --- a/arch/um/sys-i386/ldt.c +++ b/arch/um/sys-i386/ldt.c @@ -25,7 +25,7 @@ int sys_modify_ldt_tt(int func, void __user *ptr, unsigned long bytecount) #endif #ifdef CONFIG_MODE_SKAS -extern int userspace_pid; +extern int userspace_pid[]; #include "skas_ptrace.h" @@ -56,7 +56,8 @@ int sys_modify_ldt_skas(int func, void __user *ptr, unsigned long bytecount) ldt = ((struct ptrace_ldt) { .func = func, .ptr = buf, .bytecount = bytecount }); - res = ptrace(PTRACE_LDT, userspace_pid, 0, (unsigned long) &ldt); +#warning Need to look up userspace_pid by cpu + res = ptrace(PTRACE_LDT, userspace_pid[0], 0, (unsigned long) &ldt); if(res < 0) goto out; diff --git a/arch/um/sys-i386/signal.c b/arch/um/sys-i386/signal.c index 76ba87254b2..03913ca5d25 100644 --- a/arch/um/sys-i386/signal.c +++ b/arch/um/sys-i386/signal.c @@ -47,9 +47,6 @@ static int copy_sc_from_user_skas(struct pt_regs *regs, REGS_CS(regs->regs.skas.regs) = sc.cs; REGS_EFLAGS(regs->regs.skas.regs) = sc.eflags; REGS_SS(regs->regs.skas.regs) = sc.ss; - regs->regs.skas.fault_addr = sc.cr2; - regs->regs.skas.fault_type = FAULT_WRITE(sc.err); - regs->regs.skas.trap_type = sc.trapno; err = restore_fp_registers(userspace_pid[0], fpregs); if(err < 0){ @@ -62,11 +59,11 @@ static int copy_sc_from_user_skas(struct pt_regs *regs, } int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, - struct pt_regs *regs, unsigned long fault_addr, - int fault_type) + struct pt_regs *regs) { struct sigcontext sc; unsigned long fpregs[HOST_FP_SIZE]; + struct faultinfo * fi = ¤t->thread.arch.faultinfo; int err; sc.gs = REGS_GS(regs->regs.skas.regs); @@ -86,9 +83,9 @@ int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, sc.eflags = REGS_EFLAGS(regs->regs.skas.regs); sc.esp_at_signal = regs->regs.skas.regs[UESP]; sc.ss = regs->regs.skas.regs[SS]; - sc.cr2 = fault_addr; - sc.err = TO_SC_ERR(fault_type); - sc.trapno = regs->regs.skas.trap_type; + sc.cr2 = fi->cr2; + sc.err = fi->error_code; + sc.trapno = fi->trap_no; err = save_fp_registers(userspace_pid[0], fpregs); if(err < 0){ @@ -167,9 +164,7 @@ static int copy_sc_to_user(struct sigcontext *to, struct _fpstate *fp, { return(CHOOSE_MODE(copy_sc_to_user_tt(to, fp, UPT_SC(&from->regs), sizeof(*fp)), - copy_sc_to_user_skas(to, fp, from, - current->thread.cr2, - current->thread.err))); + copy_sc_to_user_skas(to, fp, from))); } static int copy_ucontext_to_user(struct ucontext *uc, struct _fpstate *fp, diff --git a/arch/um/sys-i386/sys_call_table.S b/arch/um/sys-i386/sys_call_table.S new file mode 100644 index 00000000000..ad75c27afe3 --- /dev/null +++ b/arch/um/sys-i386/sys_call_table.S @@ -0,0 +1,16 @@ +#include <linux/linkage.h> +/* Steal i386 syscall table for our purposes, but with some slight changes.*/ + +#define sys_iopl sys_ni_syscall +#define sys_ioperm sys_ni_syscall + +#define sys_vm86old sys_ni_syscall +#define sys_vm86 sys_ni_syscall +#define sys_set_thread_area sys_ni_syscall +#define sys_get_thread_area sys_ni_syscall + +#define sys_stime um_stime +#define sys_time um_time +#define old_mmap old_mmap_i386 + +#include "../../i386/kernel/syscall_table.S" diff --git a/arch/um/sys-i386/user-offsets.c b/arch/um/sys-i386/user-offsets.c new file mode 100644 index 00000000000..3ceaabceb3d --- /dev/null +++ b/arch/um/sys-i386/user-offsets.c @@ -0,0 +1,69 @@ +#include <stdio.h> +#include <signal.h> +#include <asm/ptrace.h> +#include <asm/user.h> +#include <linux/stddef.h> + +#define DEFINE(sym, val) \ + asm volatile("\n->" #sym " %0 " #val : : "i" (val)) + +#define OFFSET(sym, str, mem) \ + DEFINE(sym, offsetof(struct str, mem)); + +void foo(void) +{ + OFFSET(SC_IP, sigcontext, eip); + OFFSET(SC_SP, sigcontext, esp); + OFFSET(SC_FS, sigcontext, fs); + OFFSET(SC_GS, sigcontext, gs); + OFFSET(SC_DS, sigcontext, ds); + OFFSET(SC_ES, sigcontext, es); + OFFSET(SC_SS, sigcontext, ss); + OFFSET(SC_CS, sigcontext, cs); + OFFSET(SC_EFLAGS, sigcontext, eflags); + OFFSET(SC_EAX, sigcontext, eax); + OFFSET(SC_EBX, sigcontext, ebx); + OFFSET(SC_ECX, sigcontext, ecx); + OFFSET(SC_EDX, sigcontext, edx); + OFFSET(SC_EDI, sigcontext, edi); + OFFSET(SC_ESI, sigcontext, esi); + OFFSET(SC_EBP, sigcontext, ebp); + OFFSET(SC_TRAPNO, sigcontext, trapno); + OFFSET(SC_ERR, sigcontext, err); + OFFSET(SC_CR2, sigcontext, cr2); + OFFSET(SC_FPSTATE, sigcontext, fpstate); + OFFSET(SC_SIGMASK, sigcontext, oldmask); + OFFSET(SC_FP_CW, _fpstate, cw); + OFFSET(SC_FP_SW, _fpstate, sw); + OFFSET(SC_FP_TAG, _fpstate, tag); + OFFSET(SC_FP_IPOFF, _fpstate, ipoff); + OFFSET(SC_FP_CSSEL, _fpstate, cssel); + OFFSET(SC_FP_DATAOFF, _fpstate, dataoff); + OFFSET(SC_FP_DATASEL, _fpstate, datasel); + OFFSET(SC_FP_ST, _fpstate, _st); + OFFSET(SC_FXSR_ENV, _fpstate, _fxsr_env); + + DEFINE(HOST_FRAME_SIZE, FRAME_SIZE); + DEFINE(HOST_FP_SIZE, + sizeof(struct user_i387_struct) / sizeof(unsigned long)); + DEFINE(HOST_XFP_SIZE, + sizeof(struct user_fxsr_struct) / sizeof(unsigned long)); + + DEFINE(HOST_IP, EIP); + DEFINE(HOST_SP, UESP); + DEFINE(HOST_EFLAGS, EFL); + DEFINE(HOST_EAX, EAX); + DEFINE(HOST_EBX, EBX); + DEFINE(HOST_ECX, ECX); + DEFINE(HOST_EDX, EDX); + DEFINE(HOST_ESI, ESI); + DEFINE(HOST_EDI, EDI); + DEFINE(HOST_EBP, EBP); + DEFINE(HOST_CS, CS); + DEFINE(HOST_SS, SS); + DEFINE(HOST_DS, DS); + DEFINE(HOST_FS, FS); + DEFINE(HOST_ES, ES); + DEFINE(HOST_GS, GS); + DEFINE(__UM_FRAME_SIZE, sizeof(struct user_regs_struct)); +} diff --git a/arch/um/sys-i386/util/Makefile b/arch/um/sys-i386/util/Makefile index 34860f9ca7b..bf61afd0b04 100644 --- a/arch/um/sys-i386/util/Makefile +++ b/arch/um/sys-i386/util/Makefile @@ -1,8 +1,5 @@ - hostprogs-y := mk_sc mk_thread always := $(hostprogs-y) -mk_thread-objs := mk_thread_kern.o mk_thread_user.o - -HOSTCFLAGS_mk_thread_kern.o := $(CFLAGS) $(CPPFLAGS) -HOSTCFLAGS_mk_thread_user.o := $(USER_CFLAGS) +HOSTCFLAGS_mk_sc.o := -I$(objtree)/arch/um +HOSTCFLAGS_mk_thread.o := -I$(objtree)/arch/um diff --git a/arch/um/sys-i386/util/mk_sc.c b/arch/um/sys-i386/util/mk_sc.c index 85cbd30396f..04c0d73433a 100644 --- a/arch/um/sys-i386/util/mk_sc.c +++ b/arch/um/sys-i386/util/mk_sc.c @@ -1,52 +1,51 @@ #include <stdio.h> -#include <signal.h> -#include <linux/stddef.h> +#include <user-offsets.h> #define SC_OFFSET(name, field) \ - printf("#define " name "(sc) *((unsigned long *) &(((char *) (sc))[%d]))\n",\ - offsetof(struct sigcontext, field)) + printf("#define " #name "(sc) *((unsigned long *) &(((char *) (sc))[%d]))\n",\ + name) #define SC_FP_OFFSET(name, field) \ - printf("#define " name \ + printf("#define " #name \ "(sc) *((unsigned long *) &(((char *) (SC_FPSTATE(sc)))[%d]))\n",\ - offsetof(struct _fpstate, field)) + name) #define SC_FP_OFFSET_PTR(name, field, type) \ - printf("#define " name \ + printf("#define " #name \ "(sc) ((" type " *) &(((char *) (SC_FPSTATE(sc)))[%d]))\n",\ - offsetof(struct _fpstate, field)) + name) int main(int argc, char **argv) { - SC_OFFSET("SC_IP", eip); - SC_OFFSET("SC_SP", esp); - SC_OFFSET("SC_FS", fs); - SC_OFFSET("SC_GS", gs); - SC_OFFSET("SC_DS", ds); - SC_OFFSET("SC_ES", es); - SC_OFFSET("SC_SS", ss); - SC_OFFSET("SC_CS", cs); - SC_OFFSET("SC_EFLAGS", eflags); - SC_OFFSET("SC_EAX", eax); - SC_OFFSET("SC_EBX", ebx); - SC_OFFSET("SC_ECX", ecx); - SC_OFFSET("SC_EDX", edx); - SC_OFFSET("SC_EDI", edi); - SC_OFFSET("SC_ESI", esi); - SC_OFFSET("SC_EBP", ebp); - SC_OFFSET("SC_TRAPNO", trapno); - SC_OFFSET("SC_ERR", err); - SC_OFFSET("SC_CR2", cr2); - SC_OFFSET("SC_FPSTATE", fpstate); - SC_OFFSET("SC_SIGMASK", oldmask); - SC_FP_OFFSET("SC_FP_CW", cw); - SC_FP_OFFSET("SC_FP_SW", sw); - SC_FP_OFFSET("SC_FP_TAG", tag); - SC_FP_OFFSET("SC_FP_IPOFF", ipoff); - SC_FP_OFFSET("SC_FP_CSSEL", cssel); - SC_FP_OFFSET("SC_FP_DATAOFF", dataoff); - SC_FP_OFFSET("SC_FP_DATASEL", datasel); - SC_FP_OFFSET_PTR("SC_FP_ST", _st, "struct _fpstate"); - SC_FP_OFFSET_PTR("SC_FXSR_ENV", _fxsr_env, "void"); + SC_OFFSET(SC_IP, eip); + SC_OFFSET(SC_SP, esp); + SC_OFFSET(SC_FS, fs); + SC_OFFSET(SC_GS, gs); + SC_OFFSET(SC_DS, ds); + SC_OFFSET(SC_ES, es); + SC_OFFSET(SC_SS, ss); + SC_OFFSET(SC_CS, cs); + SC_OFFSET(SC_EFLAGS, eflags); + SC_OFFSET(SC_EAX, eax); + SC_OFFSET(SC_EBX, ebx); + SC_OFFSET(SC_ECX, ecx); + SC_OFFSET(SC_EDX, edx); + SC_OFFSET(SC_EDI, edi); + SC_OFFSET(SC_ESI, esi); + SC_OFFSET(SC_EBP, ebp); + SC_OFFSET(SC_TRAPNO, trapno); + SC_OFFSET(SC_ERR, err); + SC_OFFSET(SC_CR2, cr2); + SC_OFFSET(SC_FPSTATE, fpstate); + SC_OFFSET(SC_SIGMASK, oldmask); + SC_FP_OFFSET(SC_FP_CW, cw); + SC_FP_OFFSET(SC_FP_SW, sw); + SC_FP_OFFSET(SC_FP_TAG, tag); + SC_FP_OFFSET(SC_FP_IPOFF, ipoff); + SC_FP_OFFSET(SC_FP_CSSEL, cssel); + SC_FP_OFFSET(SC_FP_DATAOFF, dataoff); + SC_FP_OFFSET(SC_FP_DATASEL, datasel); + SC_FP_OFFSET_PTR(SC_FP_ST, _st, "struct _fpstate"); + SC_FP_OFFSET_PTR(SC_FXSR_ENV, _fxsr_env, "void"); return(0); } diff --git a/arch/um/sys-i386/util/mk_thread.c b/arch/um/sys-i386/util/mk_thread.c new file mode 100644 index 00000000000..7470d0dda67 --- /dev/null +++ b/arch/um/sys-i386/util/mk_thread.c @@ -0,0 +1,22 @@ +#include <stdio.h> +#include <kernel-offsets.h> + +int main(int argc, char **argv) +{ + printf("/*\n"); + printf(" * Generated by mk_thread\n"); + printf(" */\n"); + printf("\n"); + printf("#ifndef __UM_THREAD_H\n"); + printf("#define __UM_THREAD_H\n"); + printf("\n"); + printf("#define TASK_DEBUGREGS(task) ((unsigned long *) " + "&(((char *) (task))[%d]))\n", TASK_DEBUGREGS); +#ifdef TASK_EXTERN_PID + printf("#define TASK_EXTERN_PID(task) *((int *) &(((char *) (task))[%d]))\n", + TASK_EXTERN_PID); +#endif + printf("\n"); + printf("#endif\n"); + return(0); +} diff --git a/arch/um/sys-i386/util/mk_thread_kern.c b/arch/um/sys-i386/util/mk_thread_kern.c deleted file mode 100644 index 948b1ce8523..00000000000 --- a/arch/um/sys-i386/util/mk_thread_kern.c +++ /dev/null @@ -1,22 +0,0 @@ -#include "linux/config.h" -#include "linux/stddef.h" -#include "linux/sched.h" - -extern void print_head(void); -extern void print_constant_ptr(char *name, int value); -extern void print_constant(char *name, char *type, int value); -extern void print_tail(void); - -#define THREAD_OFFSET(field) offsetof(struct task_struct, thread.field) - -int main(int argc, char **argv) -{ - print_head(); - print_constant_ptr("TASK_DEBUGREGS", THREAD_OFFSET(arch.debugregs)); -#ifdef CONFIG_MODE_TT - print_constant("TASK_EXTERN_PID", "int", THREAD_OFFSET(mode.tt.extern_pid)); -#endif - print_tail(); - return(0); -} - diff --git a/arch/um/sys-i386/util/mk_thread_user.c b/arch/um/sys-i386/util/mk_thread_user.c deleted file mode 100644 index 2620cd6aa1f..00000000000 --- a/arch/um/sys-i386/util/mk_thread_user.c +++ /dev/null @@ -1,30 +0,0 @@ -#include <stdio.h> - -void print_head(void) -{ - printf("/*\n"); - printf(" * Generated by mk_thread\n"); - printf(" */\n"); - printf("\n"); - printf("#ifndef __UM_THREAD_H\n"); - printf("#define __UM_THREAD_H\n"); - printf("\n"); -} - -void print_constant_ptr(char *name, int value) -{ - printf("#define %s(task) ((unsigned long *) " - "&(((char *) (task))[%d]))\n", name, value); -} - -void print_constant(char *name, char *type, int value) -{ - printf("#define %s(task) *((%s *) &(((char *) (task))[%d]))\n", name, type, - value); -} - -void print_tail(void) -{ - printf("\n"); - printf("#endif\n"); -} diff --git a/arch/um/sys-x86_64/Makefile b/arch/um/sys-x86_64/Makefile index 2129e314355..3d7da911cc8 100644 --- a/arch/um/sys-x86_64/Makefile +++ b/arch/um/sys-x86_64/Makefile @@ -4,24 +4,20 @@ # Licensed under the GPL # +#XXX: why into lib-y? lib-y = bitops.o bugs.o csum-partial.o delay.o fault.o mem.o memcpy.o \ ptrace.o ptrace_user.o semaphore.o sigcontext.o signal.o \ - syscalls.o sysrq.o thunk.o + syscalls.o sysrq.o thunk.o syscall_table.o + +obj-y := ksyms.o +obj-$(CONFIG_MODULES) += module.o um_module.o USER_OBJS := ptrace_user.o sigcontext.o include arch/um/scripts/Makefile.rules SYMLINKS = bitops.c csum-copy.S csum-partial.c csum-wrappers.c memcpy.S \ - semaphore.c thunk.S - -# this needs to be before the foreach, because clean-files does not accept -# complete paths like $(src)/$f. -clean-files := $(SYMLINKS) - -targets += $(SYMLINKS) - -SYMLINKS := $(foreach f,$(SYMLINKS),$(obj)/$f) + semaphore.c thunk.S module.c bitops.c-dir = lib csum-copy.S-dir = lib @@ -30,8 +26,8 @@ csum-wrappers.c-dir = lib memcpy.S-dir = lib semaphore.c-dir = kernel thunk.S-dir = lib - -$(SYMLINKS): FORCE - $(call if_changed,make_link) +module.c-dir = kernel CFLAGS_csum-partial.o := -Dcsum_partial=arch_csum_partial + +subdir- := util diff --git a/arch/um/sys-x86_64/delay.c b/arch/um/sys-x86_64/delay.c index f3b5187942b..651332aeec2 100644 --- a/arch/um/sys-x86_64/delay.c +++ b/arch/um/sys-x86_64/delay.c @@ -5,7 +5,9 @@ * Licensed under the GPL */ +#include "linux/delay.h" #include "asm/processor.h" +#include "asm/param.h" void __delay(unsigned long loops) { @@ -14,6 +16,22 @@ void __delay(unsigned long loops) for(i = 0; i < loops; i++) ; } +void __udelay(unsigned long usecs) +{ + int i, n; + + n = (loops_per_jiffy * HZ * usecs) / MILLION; + for(i=0;i<n;i++) ; +} + +void __const_udelay(unsigned long usecs) +{ + int i, n; + + n = (loops_per_jiffy * HZ * usecs) / MILLION; + for(i=0;i<n;i++) ; +} + /* * Overrides for Emacs so that we follow Linus's tabbing style. * Emacs will notice this stuff at the end of the file and automatically diff --git a/arch/um/sys-x86_64/kernel-offsets.c b/arch/um/sys-x86_64/kernel-offsets.c new file mode 100644 index 00000000000..220e875cbe2 --- /dev/null +++ b/arch/um/sys-x86_64/kernel-offsets.c @@ -0,0 +1,24 @@ +#include <linux/config.h> +#include <linux/stddef.h> +#include <linux/sched.h> +#include <linux/time.h> +#include <asm/page.h> + +#define DEFINE(sym, val) \ + asm volatile("\n->" #sym " %0 " #val : : "i" (val)) + +#define DEFINE_STR1(x) #x +#define DEFINE_STR(sym, val) asm volatile("\n->" #sym " " DEFINE_STR1(val) " " #val: : ) + +#define BLANK() asm volatile("\n->" : : ) + +#define OFFSET(sym, str, mem) \ + DEFINE(sym, offsetof(struct str, mem)); + +void foo(void) +{ +#ifdef CONFIG_MODE_TT + OFFSET(TASK_EXTERN_PID, task_struct, thread.mode.tt.extern_pid); +#endif +#include <common-offsets.h> +} diff --git a/arch/um/sys-x86_64/ksyms.c b/arch/um/sys-x86_64/ksyms.c new file mode 100644 index 00000000000..a27f0ee6a4f --- /dev/null +++ b/arch/um/sys-x86_64/ksyms.c @@ -0,0 +1,20 @@ +#include "linux/module.h" +#include "linux/in6.h" +#include "linux/rwsem.h" +#include "asm/byteorder.h" +#include "asm/semaphore.h" +#include "asm/uaccess.h" +#include "asm/checksum.h" +#include "asm/errno.h" + +EXPORT_SYMBOL(__down_failed); +EXPORT_SYMBOL(__down_failed_interruptible); +EXPORT_SYMBOL(__down_failed_trylock); +EXPORT_SYMBOL(__up_wakeup); + +/*XXX: we need them because they would be exported by x86_64 */ +EXPORT_SYMBOL(__memcpy); + +/* Networking helper routines. */ +/*EXPORT_SYMBOL(csum_partial_copy_from); +EXPORT_SYMBOL(csum_partial_copy_to);*/ diff --git a/arch/um/sys-x86_64/signal.c b/arch/um/sys-x86_64/signal.c index 5bc5a0d796e..b740177066a 100644 --- a/arch/um/sys-x86_64/signal.c +++ b/arch/um/sys-x86_64/signal.c @@ -57,7 +57,7 @@ static int copy_sc_from_user_skas(struct pt_regs *regs, int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, struct pt_regs *regs, unsigned long mask) { - unsigned long eflags; + struct faultinfo * fi = ¤t->thread.arch.faultinfo; int err = 0; err |= __put_user(0, &to->gs); @@ -84,14 +84,16 @@ int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, err |= PUTREG(regs, R14, to, r14); err |= PUTREG(regs, R15, to, r15); err |= PUTREG(regs, CS, to, cs); /* XXX x86_64 doesn't do this */ - err |= __put_user(current->thread.err, &to->err); - err |= __put_user(current->thread.trap_no, &to->trapno); + + err |= __put_user(fi->cr2, &to->cr2); + err |= __put_user(fi->error_code, &to->err); + err |= __put_user(fi->trap_no, &to->trapno); + err |= PUTREG(regs, RIP, to, rip); err |= PUTREG(regs, EFLAGS, to, eflags); #undef PUTREG err |= __put_user(mask, &to->oldmask); - err |= __put_user(current->thread.cr2, &to->cr2); return(err); } diff --git a/arch/um/sys-x86_64/syscall_table.c b/arch/um/sys-x86_64/syscall_table.c new file mode 100644 index 00000000000..34b2e842864 --- /dev/null +++ b/arch/um/sys-x86_64/syscall_table.c @@ -0,0 +1,59 @@ +/* System call table for UML/x86-64, copied from arch/x86_64/kernel/syscall.c + * with some changes for UML. */ + +#include <linux/linkage.h> +#include <linux/sys.h> +#include <linux/cache.h> +#include <linux/config.h> + +#define __NO_STUBS + +/* Below you can see, in terms of #define's, the differences between the x86-64 + * and the UML syscall table. */ + +/* Not going to be implemented by UML, since we have no hardware. */ +#define stub_iopl sys_ni_syscall +#define sys_ioperm sys_ni_syscall + +/* The UML TLS problem. Note that x86_64 does not implement this, so the below + * is needed only for the ia32 compatibility. */ +/*#define sys_set_thread_area sys_ni_syscall +#define sys_get_thread_area sys_ni_syscall*/ + +/* For __NR_time. The x86-64 name hopefully will change from sys_time64 to + * sys_time (since the current situation is bogus). I've sent a patch to cleanup + * this. Remove below the obsoleted line. */ +#define sys_time64 um_time +#define sys_time um_time + +/* On UML we call it this way ("old" means it's not mmap2) */ +#define sys_mmap old_mmap +/* On x86-64 sys_uname is actually sys_newuname plus a compatibility trick. + * See arch/x86_64/kernel/sys_x86_64.c */ +#define sys_uname sys_uname64 + +#define stub_clone sys_clone +#define stub_fork sys_fork +#define stub_vfork sys_vfork +#define stub_execve sys_execve +#define stub_rt_sigsuspend sys_rt_sigsuspend +#define stub_sigaltstack sys_sigaltstack +#define stub_rt_sigreturn sys_rt_sigreturn + +#define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ; +#undef _ASM_X86_64_UNISTD_H_ +#include <asm-x86_64/unistd.h> + +#undef __SYSCALL +#define __SYSCALL(nr, sym) [ nr ] = sym, +#undef _ASM_X86_64_UNISTD_H_ + +typedef void (*sys_call_ptr_t)(void); + +extern void sys_ni_syscall(void); + +sys_call_ptr_t sys_call_table[__NR_syscall_max+1] __cacheline_aligned = { + /* Smells like a like a compiler bug -- it doesn't work when the & below is removed. */ + [0 ... __NR_syscall_max] = &sys_ni_syscall, +#include <asm-x86_64/unistd.h> +}; diff --git a/arch/um/sys-x86_64/syscalls.c b/arch/um/sys-x86_64/syscalls.c index 68205a03364..2a575ef52bb 100644 --- a/arch/um/sys-x86_64/syscalls.c +++ b/arch/um/sys-x86_64/syscalls.c @@ -7,6 +7,8 @@ #include "linux/linkage.h" #include "linux/slab.h" #include "linux/shm.h" +#include "linux/utsname.h" +#include "linux/personality.h" #include "asm/uaccess.h" #define __FRAME_OFFSETS #include "asm/ptrace.h" @@ -14,11 +16,15 @@ #include "asm/prctl.h" /* XXX This should get the constants from libc */ #include "choose-mode.h" -asmlinkage long wrap_sys_shmat(int shmid, char __user *shmaddr, int shmflg) +asmlinkage long sys_uname64(struct new_utsname __user * name) { - unsigned long raddr; - - return do_shmat(shmid, shmaddr, shmflg, &raddr) ?: (long) raddr; + int err; + down_read(&uts_sem); + err = copy_to_user(name, &system_utsname, sizeof (*name)); + up_read(&uts_sem); + if (personality(current->personality) == PER_LINUX32) + err |= copy_to_user(&name->machine, "i686", 5); + return err ? -EFAULT : 0; } #ifdef CONFIG_MODE_TT diff --git a/arch/um/sys-x86_64/um_module.c b/arch/um/sys-x86_64/um_module.c new file mode 100644 index 00000000000..8b8eff1bd97 --- /dev/null +++ b/arch/um/sys-x86_64/um_module.c @@ -0,0 +1,19 @@ +#include <linux/vmalloc.h> +#include <linux/moduleloader.h> + +/*Copied from i386 arch/i386/kernel/module.c */ +void *module_alloc(unsigned long size) +{ + if (size == 0) + return NULL; + return vmalloc_exec(size); +} + +/* Free memory returned from module_alloc */ +void module_free(struct module *mod, void *module_region) +{ + vfree(module_region); + /* FIXME: If module_region == mod->init_region, trim exception + table entries. */ +} + diff --git a/arch/um/sys-x86_64/user-offsets.c b/arch/um/sys-x86_64/user-offsets.c new file mode 100644 index 00000000000..5e14792e483 --- /dev/null +++ b/arch/um/sys-x86_64/user-offsets.c @@ -0,0 +1,78 @@ +#include <stdio.h> +#include <stddef.h> +#include <signal.h> +#define __FRAME_OFFSETS +#include <asm/ptrace.h> +#include <asm/user.h> + +#define DEFINE(sym, val) \ + asm volatile("\n->" #sym " %0 " #val : : "i" (val)) + +#define OFFSET(sym, str, mem) \ + DEFINE(sym, offsetof(struct str, mem)); + +void foo(void) +{ + OFFSET(SC_RBX, sigcontext, rbx); + OFFSET(SC_RCX, sigcontext, rcx); + OFFSET(SC_RDX, sigcontext, rdx); + OFFSET(SC_RSI, sigcontext, rsi); + OFFSET(SC_RDI, sigcontext, rdi); + OFFSET(SC_RBP, sigcontext, rbp); + OFFSET(SC_RAX, sigcontext, rax); + OFFSET(SC_R8, sigcontext, r8); + OFFSET(SC_R9, sigcontext, r9); + OFFSET(SC_R10, sigcontext, r10); + OFFSET(SC_R11, sigcontext, r11); + OFFSET(SC_R12, sigcontext, r12); + OFFSET(SC_R13, sigcontext, r13); + OFFSET(SC_R14, sigcontext, r14); + OFFSET(SC_R15, sigcontext, r15); + OFFSET(SC_IP, sigcontext, rip); + OFFSET(SC_SP, sigcontext, rsp); + OFFSET(SC_CR2, sigcontext, cr2); + OFFSET(SC_ERR, sigcontext, err); + OFFSET(SC_TRAPNO, sigcontext, trapno); + OFFSET(SC_CS, sigcontext, cs); + OFFSET(SC_FS, sigcontext, fs); + OFFSET(SC_GS, sigcontext, gs); + OFFSET(SC_EFLAGS, sigcontext, eflags); + OFFSET(SC_SIGMASK, sigcontext, oldmask); +#if 0 + OFFSET(SC_ORIG_RAX, sigcontext, orig_rax); + OFFSET(SC_DS, sigcontext, ds); + OFFSET(SC_ES, sigcontext, es); + OFFSET(SC_SS, sigcontext, ss); +#endif + + DEFINE(HOST_FRAME_SIZE, FRAME_SIZE); + DEFINE(HOST_RBX, RBX); + DEFINE(HOST_RCX, RCX); + DEFINE(HOST_RDI, RDI); + DEFINE(HOST_RSI, RSI); + DEFINE(HOST_RDX, RDX); + DEFINE(HOST_RBP, RBP); + DEFINE(HOST_RAX, RAX); + DEFINE(HOST_R8, R8); + DEFINE(HOST_R9, R9); + DEFINE(HOST_R10, R10); + DEFINE(HOST_R11, R11); + DEFINE(HOST_R12, R12); + DEFINE(HOST_R13, R13); + DEFINE(HOST_R14, R14); + DEFINE(HOST_R15, R15); + DEFINE(HOST_ORIG_RAX, ORIG_RAX); + DEFINE(HOST_CS, CS); + DEFINE(HOST_SS, SS); + DEFINE(HOST_EFLAGS, EFLAGS); +#if 0 + DEFINE(HOST_FS, FS); + DEFINE(HOST_GS, GS); + DEFINE(HOST_DS, DS); + DEFINE(HOST_ES, ES); +#endif + + DEFINE(HOST_IP, RIP); + DEFINE(HOST_SP, RSP); + DEFINE(__UM_FRAME_SIZE, sizeof(struct user_regs_struct)); +} diff --git a/arch/um/sys-x86_64/util/Makefile b/arch/um/sys-x86_64/util/Makefile index 00260798086..75b052cfc20 100644 --- a/arch/um/sys-x86_64/util/Makefile +++ b/arch/um/sys-x86_64/util/Makefile @@ -4,7 +4,5 @@ hostprogs-y := mk_sc mk_thread always := $(hostprogs-y) -mk_thread-objs := mk_thread_kern.o mk_thread_user.o - -HOSTCFLAGS_mk_thread_kern.o := $(CFLAGS) $(CPPFLAGS) -HOSTCFLAGS_mk_thread_user.o := $(USER_CFLAGS) +HOSTCFLAGS_mk_sc.o := -I$(objtree)/arch/um +HOSTCFLAGS_mk_thread.o := -I$(objtree)/arch/um diff --git a/arch/um/sys-x86_64/util/mk_sc.c b/arch/um/sys-x86_64/util/mk_sc.c index c236e213918..7619bc377c1 100644 --- a/arch/um/sys-x86_64/util/mk_sc.c +++ b/arch/um/sys-x86_64/util/mk_sc.c @@ -3,56 +3,45 @@ */ #include <stdio.h> -#include <signal.h> -#include <linux/stddef.h> +#include <user-offsets.h> -#define SC_OFFSET(name, field) \ - printf("#define " name \ - "(sc) *((unsigned long *) &(((char *) (sc))[%ld]))\n",\ - offsetof(struct sigcontext, field)) - -#define SC_FP_OFFSET(name, field) \ - printf("#define " name \ - "(sc) *((unsigned long *) &(((char *) (SC_FPSTATE(sc)))[%ld]))\n",\ - offsetof(struct _fpstate, field)) - -#define SC_FP_OFFSET_PTR(name, field, type) \ - printf("#define " name \ - "(sc) ((" type " *) &(((char *) (SC_FPSTATE(sc)))[%d]))\n",\ - offsetof(struct _fpstate, field)) +#define SC_OFFSET(name) \ + printf("#define " #name \ + "(sc) *((unsigned long *) &(((char *) (sc))[%d]))\n",\ + name) int main(int argc, char **argv) { - SC_OFFSET("SC_RBX", rbx); - SC_OFFSET("SC_RCX", rcx); - SC_OFFSET("SC_RDX", rdx); - SC_OFFSET("SC_RSI", rsi); - SC_OFFSET("SC_RDI", rdi); - SC_OFFSET("SC_RBP", rbp); - SC_OFFSET("SC_RAX", rax); - SC_OFFSET("SC_R8", r8); - SC_OFFSET("SC_R9", r9); - SC_OFFSET("SC_R10", r10); - SC_OFFSET("SC_R11", r11); - SC_OFFSET("SC_R12", r12); - SC_OFFSET("SC_R13", r13); - SC_OFFSET("SC_R14", r14); - SC_OFFSET("SC_R15", r15); - SC_OFFSET("SC_IP", rip); - SC_OFFSET("SC_SP", rsp); - SC_OFFSET("SC_CR2", cr2); - SC_OFFSET("SC_ERR", err); - SC_OFFSET("SC_TRAPNO", trapno); - SC_OFFSET("SC_CS", cs); - SC_OFFSET("SC_FS", fs); - SC_OFFSET("SC_GS", gs); - SC_OFFSET("SC_EFLAGS", eflags); - SC_OFFSET("SC_SIGMASK", oldmask); + SC_OFFSET(SC_RBX); + SC_OFFSET(SC_RCX); + SC_OFFSET(SC_RDX); + SC_OFFSET(SC_RSI); + SC_OFFSET(SC_RDI); + SC_OFFSET(SC_RBP); + SC_OFFSET(SC_RAX); + SC_OFFSET(SC_R8); + SC_OFFSET(SC_R9); + SC_OFFSET(SC_R10); + SC_OFFSET(SC_R11); + SC_OFFSET(SC_R12); + SC_OFFSET(SC_R13); + SC_OFFSET(SC_R14); + SC_OFFSET(SC_R15); + SC_OFFSET(SC_IP); + SC_OFFSET(SC_SP); + SC_OFFSET(SC_CR2); + SC_OFFSET(SC_ERR); + SC_OFFSET(SC_TRAPNO); + SC_OFFSET(SC_CS); + SC_OFFSET(SC_FS); + SC_OFFSET(SC_GS); + SC_OFFSET(SC_EFLAGS); + SC_OFFSET(SC_SIGMASK); #if 0 - SC_OFFSET("SC_ORIG_RAX", orig_rax); - SC_OFFSET("SC_DS", ds); - SC_OFFSET("SC_ES", es); - SC_OFFSET("SC_SS", ss); + SC_OFFSET(SC_ORIG_RAX); + SC_OFFSET(SC_DS); + SC_OFFSET(SC_ES); + SC_OFFSET(SC_SS); #endif return(0); } diff --git a/arch/um/sys-x86_64/util/mk_thread.c b/arch/um/sys-x86_64/util/mk_thread.c new file mode 100644 index 00000000000..15517396e9c --- /dev/null +++ b/arch/um/sys-x86_64/util/mk_thread.c @@ -0,0 +1,20 @@ +#include <stdio.h> +#include <kernel-offsets.h> + +int main(int argc, char **argv) +{ + printf("/*\n"); + printf(" * Generated by mk_thread\n"); + printf(" */\n"); + printf("\n"); + printf("#ifndef __UM_THREAD_H\n"); + printf("#define __UM_THREAD_H\n"); + printf("\n"); +#ifdef TASK_EXTERN_PID + printf("#define TASK_EXTERN_PID(task) *((int *) &(((char *) (task))[%d]))\n", + TASK_EXTERN_PID); +#endif + printf("\n"); + printf("#endif\n"); + return(0); +} diff --git a/arch/um/sys-x86_64/util/mk_thread_kern.c b/arch/um/sys-x86_64/util/mk_thread_kern.c deleted file mode 100644 index a281673f02b..00000000000 --- a/arch/um/sys-x86_64/util/mk_thread_kern.c +++ /dev/null @@ -1,21 +0,0 @@ -#include "linux/config.h" -#include "linux/stddef.h" -#include "linux/sched.h" - -extern void print_head(void); -extern void print_constant_ptr(char *name, int value); -extern void print_constant(char *name, char *type, int value); -extern void print_tail(void); - -#define THREAD_OFFSET(field) offsetof(struct task_struct, thread.field) - -int main(int argc, char **argv) -{ - print_head(); -#ifdef CONFIG_MODE_TT - print_constant("TASK_EXTERN_PID", "int", THREAD_OFFSET(mode.tt.extern_pid)); -#endif - print_tail(); - return(0); -} - diff --git a/arch/um/sys-x86_64/util/mk_thread_user.c b/arch/um/sys-x86_64/util/mk_thread_user.c deleted file mode 100644 index 7989725568b..00000000000 --- a/arch/um/sys-x86_64/util/mk_thread_user.c +++ /dev/null @@ -1,30 +0,0 @@ -#include <stdio.h> - -void print_head(void) -{ - printf("/*\n"); - printf(" * Generated by mk_thread\n"); - printf(" */\n"); - printf("\n"); - printf("#ifndef __UM_THREAD_H\n"); - printf("#define __UM_THREAD_H\n"); - printf("\n"); -} - -void print_constant_ptr(char *name, int value) -{ - printf("#define %s(task) ((unsigned long *) " - "&(((char *) (task))[%d]))\n", name, value); -} - -void print_constant(char *name, char *type, int value) -{ - printf("#define %s(task) *((%s *) &(((char *) (task))[%d]))\n", name, type, - value); -} - -void print_tail(void) -{ - printf("\n"); - printf("#endif\n"); -} diff --git a/arch/um/util/Makefile b/arch/um/util/Makefile index e2ab71209f3..4c7551c2803 100644 --- a/arch/um/util/Makefile +++ b/arch/um/util/Makefile @@ -1,8 +1,5 @@ hostprogs-y := mk_task mk_constants always := $(hostprogs-y) -mk_task-objs := mk_task_user.o mk_task_kern.o -mk_constants-objs := mk_constants_user.o mk_constants_kern.o - -HOSTCFLAGS_mk_task_kern.o := $(CFLAGS) $(CPPFLAGS) -HOSTCFLAGS_mk_constants_kern.o := $(CFLAGS) $(CPPFLAGS) +HOSTCFLAGS_mk_task.o := -I$(objtree)/arch/um +HOSTCFLAGS_mk_constants.o := -I$(objtree)/arch/um diff --git a/arch/um/util/mk_constants.c b/arch/um/util/mk_constants.c new file mode 100644 index 00000000000..ab217becc36 --- /dev/null +++ b/arch/um/util/mk_constants.c @@ -0,0 +1,32 @@ +#include <stdio.h> +#include <kernel-offsets.h> + +#define SHOW_INT(sym) printf("#define %s %d\n", #sym, sym) +#define SHOW_STR(sym) printf("#define %s %s\n", #sym, sym) + +int main(int argc, char **argv) +{ + printf("/*\n"); + printf(" * Generated by mk_constants\n"); + printf(" */\n"); + printf("\n"); + printf("#ifndef __UM_CONSTANTS_H\n"); + printf("#define __UM_CONSTANTS_H\n"); + printf("\n"); + + SHOW_INT(UM_KERN_PAGE_SIZE); + + SHOW_STR(UM_KERN_EMERG); + SHOW_STR(UM_KERN_ALERT); + SHOW_STR(UM_KERN_CRIT); + SHOW_STR(UM_KERN_ERR); + SHOW_STR(UM_KERN_WARNING); + SHOW_STR(UM_KERN_NOTICE); + SHOW_STR(UM_KERN_INFO); + SHOW_STR(UM_KERN_DEBUG); + + SHOW_INT(UM_NSEC_PER_SEC); + printf("\n"); + printf("#endif\n"); + return(0); +} diff --git a/arch/um/util/mk_constants_kern.c b/arch/um/util/mk_constants_kern.c deleted file mode 100644 index cdcb1232a1e..00000000000 --- a/arch/um/util/mk_constants_kern.c +++ /dev/null @@ -1,28 +0,0 @@ -#include "linux/kernel.h" -#include "linux/stringify.h" -#include "linux/time.h" -#include "asm/page.h" - -extern void print_head(void); -extern void print_constant_str(char *name, char *value); -extern void print_constant_int(char *name, int value); -extern void print_tail(void); - -int main(int argc, char **argv) -{ - print_head(); - print_constant_int("UM_KERN_PAGE_SIZE", PAGE_SIZE); - - print_constant_str("UM_KERN_EMERG", KERN_EMERG); - print_constant_str("UM_KERN_ALERT", KERN_ALERT); - print_constant_str("UM_KERN_CRIT", KERN_CRIT); - print_constant_str("UM_KERN_ERR", KERN_ERR); - print_constant_str("UM_KERN_WARNING", KERN_WARNING); - print_constant_str("UM_KERN_NOTICE", KERN_NOTICE); - print_constant_str("UM_KERN_INFO", KERN_INFO); - print_constant_str("UM_KERN_DEBUG", KERN_DEBUG); - - print_constant_int("UM_NSEC_PER_SEC", NSEC_PER_SEC); - print_tail(); - return(0); -} diff --git a/arch/um/util/mk_constants_user.c b/arch/um/util/mk_constants_user.c deleted file mode 100644 index 8f4d7e50be7..00000000000 --- a/arch/um/util/mk_constants_user.c +++ /dev/null @@ -1,28 +0,0 @@ -#include <stdio.h> - -void print_head(void) -{ - printf("/*\n"); - printf(" * Generated by mk_constants\n"); - printf(" */\n"); - printf("\n"); - printf("#ifndef __UM_CONSTANTS_H\n"); - printf("#define __UM_CONSTANTS_H\n"); - printf("\n"); -} - -void print_constant_str(char *name, char *value) -{ - printf("#define %s \"%s\"\n", name, value); -} - -void print_constant_int(char *name, int value) -{ - printf("#define %s %d\n", name, value); -} - -void print_tail(void) -{ - printf("\n"); - printf("#endif\n"); -} diff --git a/arch/um/util/mk_task_user.c b/arch/um/util/mk_task.c index 9db849f3f3a..36c9606505e 100644 --- a/arch/um/util/mk_task_user.c +++ b/arch/um/util/mk_task.c @@ -1,18 +1,19 @@ #include <stdio.h> +#include <kernel-offsets.h> -void print(char *name, char *type, int offset) +void print_ptr(char *name, char *type, int offset) { - printf("#define %s(task) *((%s *) &(((char *) (task))[%d]))\n", name, type, + printf("#define %s(task) ((%s *) &(((char *) (task))[%d]))\n", name, type, offset); } -void print_ptr(char *name, char *type, int offset) +void print(char *name, char *type, int offset) { - printf("#define %s(task) ((%s *) &(((char *) (task))[%d]))\n", name, type, + printf("#define %s(task) *((%s *) &(((char *) (task))[%d]))\n", name, type, offset); } -void print_head(void) +int main(int argc, char **argv) { printf("/*\n"); printf(" * Generated by mk_task\n"); @@ -21,10 +22,9 @@ void print_head(void) printf("#ifndef __TASK_H\n"); printf("#define __TASK_H\n"); printf("\n"); -} - -void print_tail(void) -{ + print_ptr("TASK_REGS", "union uml_pt_regs", TASK_REGS); + print("TASK_PID", "int", TASK_PID); printf("\n"); printf("#endif\n"); + return(0); } diff --git a/arch/um/util/mk_task_kern.c b/arch/um/util/mk_task_kern.c deleted file mode 100644 index c218103315e..00000000000 --- a/arch/um/util/mk_task_kern.c +++ /dev/null @@ -1,17 +0,0 @@ -#include "linux/sched.h" -#include "linux/stddef.h" - -extern void print(char *name, char *type, int offset); -extern void print_ptr(char *name, char *type, int offset); -extern void print_head(void); -extern void print_tail(void); - -int main(int argc, char **argv) -{ - print_head(); - print_ptr("TASK_REGS", "union uml_pt_regs", - offsetof(struct task_struct, thread.regs)); - print("TASK_PID", "int", offsetof(struct task_struct, pid)); - print_tail(); - return(0); -} diff --git a/arch/v850/kernel/ptrace.c b/arch/v850/kernel/ptrace.c index 8fa780757dc..4726b87f5e5 100644 --- a/arch/v850/kernel/ptrace.c +++ b/arch/v850/kernel/ptrace.c @@ -23,6 +23,7 @@ #include <linux/sched.h> #include <linux/smp_lock.h> #include <linux/ptrace.h> +#include <linux/signal.h> #include <asm/errno.h> #include <asm/ptrace.h> @@ -208,7 +209,7 @@ int sys_ptrace(long request, long pid, long addr, long data) /* Execute a single instruction. */ case PTRACE_SINGLESTEP: rval = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; /* Turn CHILD's single-step flag on or off. */ diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 80c38c5d71f..44ee7f6acf7 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -379,6 +379,11 @@ config GENERIC_IRQ_PROBE bool default y +# we have no ISA slots, but we do have ISA-style DMA. +config ISA_DMA_API + bool + default y + menu "Power management options" source kernel/power/Kconfig diff --git a/arch/x86_64/boot/bootsect.S b/arch/x86_64/boot/bootsect.S index bb15d406ee9..011b7a4993d 100644 --- a/arch/x86_64/boot/bootsect.S +++ b/arch/x86_64/boot/bootsect.S @@ -63,7 +63,7 @@ msg_loop: jz die movb $0xe, %ah movw $7, %bx - int $0x10 + int $0x10 jmp msg_loop die: @@ -71,7 +71,7 @@ die: xorw %ax, %ax int $0x16 int $0x19 - + # int 0x19 should never return. In case it does anyway, # invoke the BIOS reset code... ljmp $0xf000,$0xfff0 diff --git a/arch/x86_64/boot/setup.S b/arch/x86_64/boot/setup.S index 3e838be9dbe..75d4d2ad93b 100644 --- a/arch/x86_64/boot/setup.S +++ b/arch/x86_64/boot/setup.S @@ -160,7 +160,7 @@ ramdisk_max: .long 0xffffffff trampoline: call start_of_setup .align 16 # The offset at this point is 0x240 - .space (0x7ff-0x240+1) # E820 & EDD space (ending at 0x7ff) + .space (0xeff-0x240+1) # E820 & EDD space (ending at 0xeff) # End of setup header ##################################################### start_of_setup: @@ -412,9 +412,9 @@ jmpe820: # sizeof(e820rec). # good820: - movb (E820NR), %al # up to 32 entries + movb (E820NR), %al # up to 128 entries cmpb $E820MAX, %al - jnl bail820 + jae bail820 incb (E820NR) movw %di, %ax diff --git a/arch/x86_64/ia32/vsyscall.lds b/arch/x86_64/ia32/vsyscall.lds index fa4b4dd4a9f..f2e75ed4c6c 100644 --- a/arch/x86_64/ia32/vsyscall.lds +++ b/arch/x86_64/ia32/vsyscall.lds @@ -36,6 +36,7 @@ SECTIONS .text.rtsigreturn : { *(.text.rtsigreturn) } :text =0x90909090 + .note : { *(.note.*) } :text :note .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr .eh_frame : { KEEP (*(.eh_frame)) } :text .dynamic : { *(.dynamic) } :text :dynamic @@ -55,6 +56,7 @@ PHDRS { text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + note PT_NOTE FLAGS(4); /* PF_R */ eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */ } diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index 56516ac92e5..7c154dfff64 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -2,6 +2,12 @@ * Handle the memory map. * The functions here do the job until bootmem takes over. * $Id: e820.c,v 1.4 2002/09/19 19:25:32 ak Exp $ + * + * Getting sanitize_e820_map() in sync with i386 version by applying change: + * - Provisions for empty E820 memory regions (reported by certain BIOSes). + * Alex Achenbach <xela@slit.de>, December 2002. + * Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> + * */ #include <linux/config.h> #include <linux/kernel.h> @@ -277,7 +283,7 @@ static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) int chgidx, still_changing; int overlap_entries; int new_bios_entry; - int old_nr, new_nr; + int old_nr, new_nr, chg_nr; int i; /* @@ -331,20 +337,24 @@ static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) for (i=0; i < 2*old_nr; i++) change_point[i] = &change_point_list[i]; - /* record all known change-points (starting and ending addresses) */ + /* record all known change-points (starting and ending addresses), + omitting those that are for empty memory regions */ chgidx = 0; for (i=0; i < old_nr; i++) { - change_point[chgidx]->addr = biosmap[i].addr; - change_point[chgidx++]->pbios = &biosmap[i]; - change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size; - change_point[chgidx++]->pbios = &biosmap[i]; + if (biosmap[i].size != 0) { + change_point[chgidx]->addr = biosmap[i].addr; + change_point[chgidx++]->pbios = &biosmap[i]; + change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size; + change_point[chgidx++]->pbios = &biosmap[i]; + } } + chg_nr = chgidx; /* sort change-point list by memory addresses (low -> high) */ still_changing = 1; while (still_changing) { still_changing = 0; - for (i=1; i < 2*old_nr; i++) { + for (i=1; i < chg_nr; i++) { /* if <current_addr> > <last_addr>, swap */ /* or, if current=<start_addr> & last=<end_addr>, swap */ if ((change_point[i]->addr < change_point[i-1]->addr) || @@ -367,7 +377,7 @@ static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) last_type = 0; /* start with undefined memory type */ last_addr = 0; /* start with 0 as last starting address */ /* loop through change-points, determining affect on the new bios map */ - for (chgidx=0; chgidx < 2*old_nr; chgidx++) + for (chgidx=0; chgidx < chg_nr; chgidx++) { /* keep track of all overlapping bios entries */ if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr) diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c index 750bcd0655d..e3a19e8ebbf 100644 --- a/arch/x86_64/kernel/early_printk.c +++ b/arch/x86_64/kernel/early_printk.c @@ -60,7 +60,7 @@ static struct console early_vga_console = { /* Serial functions loosely based on a similar package from Klaus P. Gerlicher */ -int early_serial_base = 0x3f8; /* ttyS0 */ +static int early_serial_base = 0x3f8; /* ttyS0 */ #define XMTRDY 0x20 diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 3233a15cc4e..1086b5fcac2 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -296,6 +296,7 @@ int_very_careful: call syscall_trace_leave popq %rdi andl $~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edi + cli jmp int_restore_rest int_signal: @@ -307,6 +308,7 @@ int_signal: 1: movl $_TIF_NEED_RESCHED,%edi int_restore_rest: RESTORE_REST + cli jmp int_with_check CFI_ENDPROC @@ -490,7 +492,8 @@ retint_signal: call do_notify_resume RESTORE_REST cli - GET_THREAD_INFO(%rcx) + GET_THREAD_INFO(%rcx) + movl $_TIF_WORK_MASK,%edi jmp retint_check #ifdef CONFIG_PREEMPT diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c index 6cad46c98a2..0f8c78dcd38 100644 --- a/arch/x86_64/kernel/head64.c +++ b/arch/x86_64/kernel/head64.c @@ -29,8 +29,6 @@ static void __init clear_bss(void) (unsigned long) __bss_end - (unsigned long) __bss_start); } -extern char x86_boot_params[2048]; - #define NEW_CL_POINTER 0x228 /* Relative to real mode data */ #define OLD_CL_MAGIC_ADDR 0x90020 #define OLD_CL_MAGIC 0xA33F @@ -44,7 +42,7 @@ static void __init copy_bootdata(char *real_mode_data) int new_data; char * command_line; - memcpy(x86_boot_params, real_mode_data, 2048); + memcpy(x86_boot_params, real_mode_data, BOOT_PARAM_SIZE); new_data = *(int *) (x86_boot_params + NEW_CL_POINTER); if (!new_data) { if (OLD_CL_MAGIC != * (u16 *) OLD_CL_MAGIC_ADDR) { @@ -93,9 +91,6 @@ void __init x86_64_start_kernel(char * real_mode_data) #ifdef CONFIG_SMP cpu_set(0, cpu_online_map); #endif - /* default console: */ - if (!strstr(saved_command_line, "console=")) - strcat(saved_command_line, " console=tty0"); s = strstr(saved_command_line, "earlyprintk="); if (s != NULL) setup_early_printk(s); diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 29a25729548..60be58617eb 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1607,7 +1607,6 @@ static inline void check_timer(void) disable_8259A_irq(0); setup_nmi(); enable_8259A_irq(0); - check_nmi_watchdog(); } return; } @@ -1627,7 +1626,6 @@ static inline void check_timer(void) nmi_watchdog_default(); if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); - check_nmi_watchdog(); } return; } diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c index 4f2a852299b..f77f8a0ff18 100644 --- a/arch/x86_64/kernel/kprobes.c +++ b/arch/x86_64/kernel/kprobes.c @@ -355,6 +355,13 @@ static void resume_execution(struct kprobe *p, struct pt_regs *regs) *tos &= ~(TF_MASK | IF_MASK); *tos |= kprobe_old_rflags; break; + case 0xc3: /* ret/lret */ + case 0xcb: + case 0xc2: + case 0xca: + regs->eflags &= ~TF_MASK; + /* rip is already adjusted, no more changes required*/ + return; case 0xe8: /* call relative - Fix return addr */ *tos = orig_rip + (*tos - copy_rip); break; diff --git a/arch/x86_64/kernel/module.c b/arch/x86_64/kernel/module.c index c2ffea8845e..bac195c74bc 100644 --- a/arch/x86_64/kernel/module.c +++ b/arch/x86_64/kernel/module.c @@ -30,9 +30,12 @@ #define DEBUGP(fmt...) +#ifndef CONFIG_UML void module_free(struct module *mod, void *module_region) { vfree(module_region); + /* FIXME: If module_region == mod->init_region, trim exception + table entries. */ } void *module_alloc(unsigned long size) @@ -51,6 +54,7 @@ void *module_alloc(unsigned long size) return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL_EXEC); } +#endif /* We don't need anything special. */ int module_frob_arch_sections(Elf_Ehdr *hdr, diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index e00d4adec36..61de0b34a01 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -112,17 +112,20 @@ static __init int cpu_has_lapic(void) } } -int __init check_nmi_watchdog (void) +static int __init check_nmi_watchdog (void) { int counts[NR_CPUS]; int cpu; + if (nmi_watchdog == NMI_NONE) + return 0; + if (nmi_watchdog == NMI_LOCAL_APIC && !cpu_has_lapic()) { nmi_watchdog = NMI_NONE; return -1; } - printk(KERN_INFO "testing NMI watchdog ... "); + printk(KERN_INFO "Testing NMI watchdog ... "); for (cpu = 0; cpu < NR_CPUS; cpu++) counts[cpu] = cpu_pda[cpu].__nmi_count; @@ -148,6 +151,8 @@ int __init check_nmi_watchdog (void) return 0; } +/* Have this called later during boot so counters are updating */ +late_initcall(check_nmi_watchdog); int __init setup_nmi_watchdog(char *str) { diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 9922d2ba24a..761b6d35e33 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -402,10 +402,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp, p->thread.fs = me->thread.fs; p->thread.gs = me->thread.gs; - asm("movl %%gs,%0" : "=m" (p->thread.gsindex)); - asm("movl %%fs,%0" : "=m" (p->thread.fsindex)); - asm("movl %%es,%0" : "=m" (p->thread.es)); - asm("movl %%ds,%0" : "=m" (p->thread.ds)); + asm("mov %%gs,%0" : "=m" (p->thread.gsindex)); + asm("mov %%fs,%0" : "=m" (p->thread.fsindex)); + asm("mov %%es,%0" : "=m" (p->thread.es)); + asm("mov %%ds,%0" : "=m" (p->thread.ds)); if (unlikely(me->thread.io_bitmap_ptr != NULL)) { p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL); @@ -468,11 +468,11 @@ struct task_struct *__switch_to(struct task_struct *prev_p, struct task_struct * * Switch DS and ES. * This won't pick up thread selector changes, but I guess that is ok. */ - asm volatile("movl %%es,%0" : "=m" (prev->es)); + asm volatile("mov %%es,%0" : "=m" (prev->es)); if (unlikely(next->es | prev->es)) loadsegment(es, next->es); - asm volatile ("movl %%ds,%0" : "=m" (prev->ds)); + asm volatile ("mov %%ds,%0" : "=m" (prev->ds)); if (unlikely(next->ds | prev->ds)) loadsegment(ds, next->ds); diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c index c7011675007..e26e86bb56f 100644 --- a/arch/x86_64/kernel/ptrace.c +++ b/arch/x86_64/kernel/ptrace.c @@ -18,6 +18,7 @@ #include <linux/security.h> #include <linux/audit.h> #include <linux/seccomp.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -467,7 +468,7 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child,TIF_SYSCALL_TRACE); @@ -529,7 +530,7 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data case PTRACE_SINGLESTEP: /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child,TIF_SYSCALL_TRACE); set_singlestep(child); @@ -634,20 +635,29 @@ asmlinkage void syscall_trace_enter(struct pt_regs *regs) /* do the secure computing check first */ secure_computing(regs->orig_rax); - if (unlikely(current->audit_context)) - audit_syscall_entry(current, regs->orig_rax, - regs->rdi, regs->rsi, - regs->rdx, regs->r10); - if (test_thread_flag(TIF_SYSCALL_TRACE) && (current->ptrace & PT_PTRACED)) syscall_trace(regs); + + if (unlikely(current->audit_context)) { + if (test_thread_flag(TIF_IA32)) { + audit_syscall_entry(current, AUDIT_ARCH_I386, + regs->orig_rax, + regs->rbx, regs->rcx, + regs->rdx, regs->rsi); + } else { + audit_syscall_entry(current, AUDIT_ARCH_X86_64, + regs->orig_rax, + regs->rdi, regs->rsi, + regs->rdx, regs->r10); + } + } } asmlinkage void syscall_trace_leave(struct pt_regs *regs) { if (unlikely(current->audit_context)) - audit_syscall_exit(current, regs->rax); + audit_syscall_exit(current, AUDITSC_RESULT(regs->rax), regs->rax); if ((test_thread_flag(TIF_SYSCALL_TRACE) || test_thread_flag(TIF_SINGLESTEP)) diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index b18c114c764..2129cf9ba6b 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -977,7 +977,7 @@ void __init identify_cpu(struct cpuinfo_x86 *c) if ((xlvl & 0xffff0000) == 0x80000000) { if (xlvl >= 0x80000001) { c->x86_capability[1] = cpuid_edx(0x80000001); - c->x86_capability[5] = cpuid_ecx(0x80000001); + c->x86_capability[6] = cpuid_ecx(0x80000001); } if (xlvl >= 0x80000004) get_model_name(c); /* Default name */ @@ -1076,7 +1076,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", NULL, /* AMD-defined */ - "pni", NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, "nx", NULL, "mmxext", NULL, NULL, "fxsr_opt", NULL, NULL, NULL, "lm", "3dnowext", "3dnow", @@ -1100,11 +1100,17 @@ static int show_cpuinfo(struct seq_file *m, void *v) NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + /* VIA/Cyrix/Centaur-defined */ + NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en", + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + /* AMD-defined (#2) */ "lahf_lm", "cmp_legacy", NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, }; static char *x86_power_flags[] = { "ts", /* temperature sensor */ diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c index b5305b04bc4..678b7ac33b8 100644 --- a/arch/x86_64/kernel/setup64.c +++ b/arch/x86_64/kernel/setup64.c @@ -12,6 +12,7 @@ #include <linux/string.h> #include <linux/bootmem.h> #include <linux/bitops.h> +#include <asm/bootsetup.h> #include <asm/pda.h> #include <asm/pgtable.h> #include <asm/processor.h> @@ -26,7 +27,7 @@ #include <asm/mman.h> #include <asm/numa.h> -char x86_boot_params[2048] __initdata = {0,}; +char x86_boot_params[BOOT_PARAM_SIZE] __initdata = {0,}; cpumask_t cpu_initialized __initdata = CPU_MASK_NONE; diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c index 7760224cdfe..d439ced150c 100644 --- a/arch/x86_64/kernel/signal.c +++ b/arch/x86_64/kernel/signal.c @@ -83,7 +83,7 @@ sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, struct rt_sigframe { - char *pretcode; + char __user *pretcode; struct ucontext uc; struct siginfo info; }; diff --git a/arch/x86_64/kernel/sys_x86_64.c b/arch/x86_64/kernel/sys_x86_64.c index 477d8be57d6..dbebd5ccba6 100644 --- a/arch/x86_64/kernel/sys_x86_64.c +++ b/arch/x86_64/kernel/sys_x86_64.c @@ -152,12 +152,6 @@ asmlinkage long sys_uname(struct new_utsname __user * name) return err ? -EFAULT : 0; } -asmlinkage long wrap_sys_shmat(int shmid, char __user *shmaddr, int shmflg) -{ - unsigned long raddr; - return do_shmat(shmid,shmaddr,shmflg,&raddr) ?: (long)raddr; -} - asmlinkage long sys_time64(long __user * tloc) { struct timeval now; diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c index 88626e62688..a43dedb58fa 100644 --- a/arch/x86_64/kernel/x8664_ksyms.c +++ b/arch/x86_64/kernel/x8664_ksyms.c @@ -139,35 +139,23 @@ EXPORT_SYMBOL_GPL(unset_nmi_callback); #undef memmove #undef memchr #undef strlen -#undef strcpy #undef strncmp #undef strncpy #undef strchr -#undef strcmp -#undef strcpy -#undef strcat -#undef memcmp extern void * memset(void *,int,__kernel_size_t); extern size_t strlen(const char *); extern void * memmove(void * dest,const void *src,size_t count); -extern char * strcpy(char * dest,const char *src); -extern int strcmp(const char * cs,const char * ct); extern void *memchr(const void *s, int c, size_t n); extern void * memcpy(void *,const void *,__kernel_size_t); extern void * __memcpy(void *,const void *,__kernel_size_t); -extern char * strcat(char *, const char *); -extern int memcmp(const void * cs,const void * ct,size_t count); EXPORT_SYMBOL(memset); EXPORT_SYMBOL(strlen); EXPORT_SYMBOL(memmove); -EXPORT_SYMBOL(strcpy); EXPORT_SYMBOL(strncmp); EXPORT_SYMBOL(strncpy); EXPORT_SYMBOL(strchr); -EXPORT_SYMBOL(strcmp); -EXPORT_SYMBOL(strcat); EXPORT_SYMBOL(strncat); EXPORT_SYMBOL(memchr); EXPORT_SYMBOL(strrchr); @@ -175,7 +163,6 @@ EXPORT_SYMBOL(strnlen); EXPORT_SYMBOL(memscan); EXPORT_SYMBOL(memcpy); EXPORT_SYMBOL(__memcpy); -EXPORT_SYMBOL(memcmp); #ifdef CONFIG_RWSEM_XCHGADD_ALGORITHM /* prototypes are wrong, these are assembly with custom calling functions */ diff --git a/arch/x86_64/lib/putuser.S b/arch/x86_64/lib/putuser.S index d9d3e5ec9ad..5828b819166 100644 --- a/arch/x86_64/lib/putuser.S +++ b/arch/x86_64/lib/putuser.S @@ -49,8 +49,8 @@ __put_user_2: jc 20f cmpq threadinfo_addr_limit(%r8),%rcx jae 20f -2: decq %rcx - movw %dx,(%rcx) + decq %rcx +2: movw %dx,(%rcx) xorl %eax,%eax ret 20: decq %rcx @@ -64,8 +64,8 @@ __put_user_4: jc 30f cmpq threadinfo_addr_limit(%r8),%rcx jae 30f -3: subq $3,%rcx - movl %edx,(%rcx) + subq $3,%rcx +3: movl %edx,(%rcx) xorl %eax,%eax ret 30: subq $3,%rcx @@ -79,8 +79,8 @@ __put_user_8: jc 40f cmpq threadinfo_addr_limit(%r8),%rcx jae 40f -4: subq $7,%rcx - movq %rdx,(%rcx) + subq $7,%rcx +4: movq %rdx,(%rcx) xorl %eax,%eax ret 40: subq $7,%rcx diff --git a/crypto/Kconfig b/crypto/Kconfig index 536754faf4d..90d6089d60e 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -146,7 +146,7 @@ config CRYPTO_SERPENT config CRYPTO_AES tristate "AES cipher algorithms" - depends on CRYPTO && !(X86 && !X86_64) + depends on CRYPTO && !((X86 || UML_X86) && !64BIT) help AES cipher algorithms (FIPS-197). AES uses the Rijndael algorithm. @@ -166,7 +166,7 @@ config CRYPTO_AES config CRYPTO_AES_586 tristate "AES cipher algorithms (i586)" - depends on CRYPTO && (X86 && !X86_64) + depends on CRYPTO && ((X86 || UML_X86) && !64BIT) help AES cipher algorithms (FIPS-197). AES uses the Rijndael algorithm. diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 05a17812d52..ff64d333e95 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -838,7 +838,7 @@ int acpi_processor_cst_has_changed (struct acpi_processor *pr) /* Fall back to the default idle loop */ pm_idle = pm_idle_save; - synchronize_kernel(); + synchronize_sched(); /* Relies on interrupts forcing exit from idle. */ pr->flags.power = 0; result = acpi_processor_get_power_info(pr); diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index e7ca0662656..119c94093a1 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -379,8 +379,8 @@ ACPI_DEVICE_ATTR(eject, 0200, NULL, acpi_eject_store); /** * setup_sys_fs_device_files - sets up the device files under device namespace - * @@dev: acpi_device object - * @@func: function pointer to create or destroy the device file + * @dev: acpi_device object + * @func: function pointer to create or destroy the device file */ static void setup_sys_fs_device_files ( diff --git a/drivers/atm/eni.c b/drivers/atm/eni.c index 78e34ee79df..10da3693476 100644 --- a/drivers/atm/eni.c +++ b/drivers/atm/eni.c @@ -59,7 +59,6 @@ * - doesn't support OAM cells * - eni_put_free may hang if not putting memory fragments that _complete_ * 2^n block (never happens in real life, though) - * - keeps IRQ even if initialization fails */ @@ -1802,22 +1801,22 @@ static int __devinit eni_start(struct atm_dev *dev) if (request_irq(eni_dev->irq,&eni_int,SA_SHIRQ,DEV_LABEL,dev)) { printk(KERN_ERR DEV_LABEL "(itf %d): IRQ%d is already in use\n", dev->number,eni_dev->irq); - return -EAGAIN; + error = -EAGAIN; + goto out; } - /* @@@ should release IRQ on error */ pci_set_master(eni_dev->pci_dev); if ((error = pci_write_config_word(eni_dev->pci_dev,PCI_COMMAND, PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER | (eni_dev->asic ? PCI_COMMAND_PARITY | PCI_COMMAND_SERR : 0)))) { printk(KERN_ERR DEV_LABEL "(itf %d): can't enable memory+" "master (0x%02x)\n",dev->number,error); - return error; + goto free_irq; } if ((error = pci_write_config_byte(eni_dev->pci_dev,PCI_TONGA_CTRL, END_SWAP_DMA))) { printk(KERN_ERR DEV_LABEL "(itf %d): can't set endian swap " "(0x%02x)\n",dev->number,error); - return error; + goto free_irq; } /* determine addresses of internal tables */ eni_dev->vci = eni_dev->ram; @@ -1839,7 +1838,8 @@ static int __devinit eni_start(struct atm_dev *dev) if (!eni_dev->free_list) { printk(KERN_ERR DEV_LABEL "(itf %d): couldn't get free page\n", dev->number); - return -ENOMEM; + error = -ENOMEM; + goto free_irq; } eni_dev->free_len = 0; eni_put_free(eni_dev,buf,buffer_mem); @@ -1855,17 +1855,26 @@ static int __devinit eni_start(struct atm_dev *dev) */ eni_out(0xffffffff,MID_IE); error = start_tx(dev); - if (error) return error; + if (error) goto free_list; error = start_rx(dev); - if (error) return error; + if (error) goto free_list; error = dev->phy->start(dev); - if (error) return error; + if (error) goto free_list; eni_out(eni_in(MID_MC_S) | (1 << MID_INT_SEL_SHIFT) | MID_TX_LOCK_MODE | MID_DMA_ENABLE | MID_TX_ENABLE | MID_RX_ENABLE, MID_MC_S); /* Tonga uses SBus INTReq1 */ (void) eni_in(MID_ISA); /* clear Midway interrupts */ return 0; + +free_list: + kfree(eni_dev->free_list); + +free_irq: + free_irq(eni_dev->irq, eni_dev); + +out: + return error; } diff --git a/drivers/atm/fore200e.c b/drivers/atm/fore200e.c index 196b3364462..9e65bfb85ba 100644 --- a/drivers/atm/fore200e.c +++ b/drivers/atm/fore200e.c @@ -2792,8 +2792,6 @@ static void __devexit fore200e_pca_remove_one(struct pci_dev *pci_dev) fore200e = pci_get_drvdata(pci_dev); - list_del(&fore200e->entry); - fore200e_shutdown(fore200e); kfree(fore200e); pci_disable_device(pci_dev); @@ -2850,7 +2848,7 @@ fore200e_module_init(void) } #ifdef CONFIG_ATM_FORE200E_PCA - if (!pci_module_init(&fore200e_pca_driver)) + if (!pci_register_driver(&fore200e_pca_driver)) return 0; #endif diff --git a/drivers/atm/he.c b/drivers/atm/he.c index c2c31a5f451..3022c548a13 100644 --- a/drivers/atm/he.c +++ b/drivers/atm/he.c @@ -70,6 +70,7 @@ #include <linux/sched.h> #include <linux/timer.h> #include <linux/interrupt.h> +#include <linux/dma-mapping.h> #include <asm/io.h> #include <asm/byteorder.h> #include <asm/uaccess.h> @@ -371,7 +372,7 @@ he_init_one(struct pci_dev *pci_dev, const struct pci_device_id *pci_ent) if (pci_enable_device(pci_dev)) return -EIO; - if (pci_set_dma_mask(pci_dev, HE_DMA_MASK) != 0) { + if (pci_set_dma_mask(pci_dev, DMA_32BIT_MASK) != 0) { printk(KERN_WARNING "he: no suitable dma available\n"); err = -EIO; goto init_one_failure; diff --git a/drivers/atm/he.h b/drivers/atm/he.h index 1a903859343..1dc277547a7 100644 --- a/drivers/atm/he.h +++ b/drivers/atm/he.h @@ -380,8 +380,6 @@ struct he_vcc #define PCI_VENDOR_ID_FORE 0x1127 #define PCI_DEVICE_ID_FORE_HE 0x400 -#define HE_DMA_MASK 0xffffffff - #define GEN_CNTL_0 0x40 #define INT_PROC_ENBL (1<<25) #define SLAVE_ENDIAN_MODE (1<<16) diff --git a/drivers/base/bus.c b/drivers/base/bus.c index f4fa27315fb..2b3902c867d 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -405,9 +405,8 @@ void device_release_driver(struct device * dev) static void driver_detach(struct device_driver * drv) { - struct list_head * entry, * next; - list_for_each_safe(entry, next, &drv->devices) { - struct device * dev = container_of(entry, struct device, driver_list); + while (!list_empty(&drv->devices)) { + struct device * dev = container_of(drv->devices.next, struct device, driver_list); device_release_driver(dev); } } diff --git a/drivers/base/core.c b/drivers/base/core.c index a7cedd8cefe..268a9c8d168 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -139,7 +139,7 @@ static int dev_hotplug(struct kset *kset, struct kobject *kobj, char **envp, buffer = &buffer[length]; buffer_size -= length; - if (dev->bus->hotplug) { + if (dev->bus && dev->bus->hotplug) { /* have the bus specific function add its stuff */ retval = dev->bus->hotplug (dev, envp, num_envp, buffer, buffer_size); if (retval) { diff --git a/drivers/base/platform.c b/drivers/base/platform.c index cd6453905a9..3a5f4c99179 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -115,7 +115,7 @@ int platform_add_devices(struct platform_device **devs, int num) /** * platform_device_register - add a platform-level device - * @dev: platform device we're adding + * @pdev: platform device we're adding * */ int platform_device_register(struct platform_device * pdev) @@ -174,7 +174,7 @@ int platform_device_register(struct platform_device * pdev) /** * platform_device_unregister - remove a platform-level device - * @dev: platform device we're removing + * @pdev: platform device we're removing * * Note that this function will also release all memory- and port-based * resources owned by the device (@dev->resource). diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c index 423bbf2000d..3760edfdc65 100644 --- a/drivers/block/DAC960.c +++ b/drivers/block/DAC960.c @@ -3,6 +3,7 @@ Linux Driver for Mylex DAC960/AcceleRAID/eXtremeRAID PCI RAID Controllers Copyright 1998-2001 by Leonard N. Zubkoff <lnz@dandelion.com> + Portions Copyright 2002 by Mylex (An IBM Business Unit) This program is free software; you may redistribute and/or modify it under the terms of the GNU General Public License Version 2 as published by the @@ -532,6 +533,34 @@ static void DAC960_WaitForCommand(DAC960_Controller_T *Controller) spin_lock_irq(&Controller->queue_lock); } +/* + DAC960_GEM_QueueCommand queues Command for DAC960 GEM Series Controllers. +*/ + +static void DAC960_GEM_QueueCommand(DAC960_Command_T *Command) +{ + DAC960_Controller_T *Controller = Command->Controller; + void __iomem *ControllerBaseAddress = Controller->BaseAddress; + DAC960_V2_CommandMailbox_T *CommandMailbox = &Command->V2.CommandMailbox; + DAC960_V2_CommandMailbox_T *NextCommandMailbox = + Controller->V2.NextCommandMailbox; + + CommandMailbox->Common.CommandIdentifier = Command->CommandIdentifier; + DAC960_GEM_WriteCommandMailbox(NextCommandMailbox, CommandMailbox); + + if (Controller->V2.PreviousCommandMailbox1->Words[0] == 0 || + Controller->V2.PreviousCommandMailbox2->Words[0] == 0) + DAC960_GEM_MemoryMailboxNewCommand(ControllerBaseAddress); + + Controller->V2.PreviousCommandMailbox2 = + Controller->V2.PreviousCommandMailbox1; + Controller->V2.PreviousCommandMailbox1 = NextCommandMailbox; + + if (++NextCommandMailbox > Controller->V2.LastCommandMailbox) + NextCommandMailbox = Controller->V2.FirstCommandMailbox; + + Controller->V2.NextCommandMailbox = NextCommandMailbox; +} /* DAC960_BA_QueueCommand queues Command for DAC960 BA Series Controllers. @@ -1464,6 +1493,17 @@ static boolean DAC960_V2_EnableMemoryMailboxInterface(DAC960_Controller_T Controller->V2.FirstStatusMailboxDMA; switch (Controller->HardwareType) { + case DAC960_GEM_Controller: + while (DAC960_GEM_HardwareMailboxFullP(ControllerBaseAddress)) + udelay(1); + DAC960_GEM_WriteHardwareMailbox(ControllerBaseAddress, CommandMailboxDMA); + DAC960_GEM_HardwareMailboxNewCommand(ControllerBaseAddress); + while (!DAC960_GEM_HardwareMailboxStatusAvailableP(ControllerBaseAddress)) + udelay(1); + CommandStatus = DAC960_GEM_ReadCommandStatus(ControllerBaseAddress); + DAC960_GEM_AcknowledgeHardwareMailboxInterrupt(ControllerBaseAddress); + DAC960_GEM_AcknowledgeHardwareMailboxStatus(ControllerBaseAddress); + break; case DAC960_BA_Controller: while (DAC960_BA_HardwareMailboxFullP(ControllerBaseAddress)) udelay(1); @@ -2627,6 +2667,9 @@ static void DAC960_DetectCleanup(DAC960_Controller_T *Controller) if (Controller->MemoryMappedAddress) { switch(Controller->HardwareType) { + case DAC960_GEM_Controller: + DAC960_GEM_DisableInterrupts(Controller->BaseAddress); + break; case DAC960_BA_Controller: DAC960_BA_DisableInterrupts(Controller->BaseAddress); break; @@ -2705,6 +2748,9 @@ DAC960_DetectController(struct pci_dev *PCI_Device, switch (Controller->HardwareType) { + case DAC960_GEM_Controller: + Controller->PCI_Address = pci_resource_start(PCI_Device, 0); + break; case DAC960_BA_Controller: Controller->PCI_Address = pci_resource_start(PCI_Device, 0); break; @@ -2756,6 +2802,36 @@ DAC960_DetectController(struct pci_dev *PCI_Device, BaseAddress = Controller->BaseAddress; switch (Controller->HardwareType) { + case DAC960_GEM_Controller: + DAC960_GEM_DisableInterrupts(BaseAddress); + DAC960_GEM_AcknowledgeHardwareMailboxStatus(BaseAddress); + udelay(1000); + while (DAC960_GEM_InitializationInProgressP(BaseAddress)) + { + if (DAC960_GEM_ReadErrorStatus(BaseAddress, &ErrorStatus, + &Parameter0, &Parameter1) && + DAC960_ReportErrorStatus(Controller, ErrorStatus, + Parameter0, Parameter1)) + goto Failure; + udelay(10); + } + if (!DAC960_V2_EnableMemoryMailboxInterface(Controller)) + { + DAC960_Error("Unable to Enable Memory Mailbox Interface " + "for Controller at\n", Controller); + goto Failure; + } + DAC960_GEM_EnableInterrupts(BaseAddress); + Controller->QueueCommand = DAC960_GEM_QueueCommand; + Controller->ReadControllerConfiguration = + DAC960_V2_ReadControllerConfiguration; + Controller->ReadDeviceConfiguration = + DAC960_V2_ReadDeviceConfiguration; + Controller->ReportDeviceConfiguration = + DAC960_V2_ReportDeviceConfiguration; + Controller->QueueReadWriteCommand = + DAC960_V2_QueueReadWriteCommand; + break; case DAC960_BA_Controller: DAC960_BA_DisableInterrupts(BaseAddress); DAC960_BA_AcknowledgeHardwareMailboxStatus(BaseAddress); @@ -5189,6 +5265,47 @@ static void DAC960_V2_ProcessCompletedCommand(DAC960_Command_T *Command) wake_up(&Controller->CommandWaitQueue); } +/* + DAC960_GEM_InterruptHandler handles hardware interrupts from DAC960 GEM Series + Controllers. +*/ + +static irqreturn_t DAC960_GEM_InterruptHandler(int IRQ_Channel, + void *DeviceIdentifier, + struct pt_regs *InterruptRegisters) +{ + DAC960_Controller_T *Controller = (DAC960_Controller_T *) DeviceIdentifier; + void __iomem *ControllerBaseAddress = Controller->BaseAddress; + DAC960_V2_StatusMailbox_T *NextStatusMailbox; + unsigned long flags; + + spin_lock_irqsave(&Controller->queue_lock, flags); + DAC960_GEM_AcknowledgeInterrupt(ControllerBaseAddress); + NextStatusMailbox = Controller->V2.NextStatusMailbox; + while (NextStatusMailbox->Fields.CommandIdentifier > 0) + { + DAC960_V2_CommandIdentifier_T CommandIdentifier = + NextStatusMailbox->Fields.CommandIdentifier; + DAC960_Command_T *Command = Controller->Commands[CommandIdentifier-1]; + Command->V2.CommandStatus = NextStatusMailbox->Fields.CommandStatus; + Command->V2.RequestSenseLength = + NextStatusMailbox->Fields.RequestSenseLength; + Command->V2.DataTransferResidue = + NextStatusMailbox->Fields.DataTransferResidue; + NextStatusMailbox->Words[0] = 0; + if (++NextStatusMailbox > Controller->V2.LastStatusMailbox) + NextStatusMailbox = Controller->V2.FirstStatusMailbox; + DAC960_V2_ProcessCompletedCommand(Command); + } + Controller->V2.NextStatusMailbox = NextStatusMailbox; + /* + Attempt to remove additional I/O Requests from the Controller's + I/O Request Queue and queue them to the Controller. + */ + DAC960_ProcessRequest(Controller); + spin_unlock_irqrestore(&Controller->queue_lock, flags); + return IRQ_HANDLED; +} /* DAC960_BA_InterruptHandler handles hardware interrupts from DAC960 BA Series @@ -6962,6 +7079,14 @@ static void DAC960_gam_cleanup(void) #endif /* DAC960_GAM_MINOR */ +static struct DAC960_privdata DAC960_GEM_privdata = { + .HardwareType = DAC960_GEM_Controller, + .FirmwareType = DAC960_V2_Controller, + .InterruptHandler = DAC960_GEM_InterruptHandler, + .MemoryWindowSize = DAC960_GEM_RegisterWindowSize, +}; + + static struct DAC960_privdata DAC960_BA_privdata = { .HardwareType = DAC960_BA_Controller, .FirmwareType = DAC960_V2_Controller, @@ -7007,6 +7132,13 @@ static struct DAC960_privdata DAC960_P_privdata = { static struct pci_device_id DAC960_id_table[] = { { .vendor = PCI_VENDOR_ID_MYLEX, + .device = PCI_DEVICE_ID_MYLEX_DAC960_GEM, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + .driver_data = (unsigned long) &DAC960_GEM_privdata, + }, + { + .vendor = PCI_VENDOR_ID_MYLEX, .device = PCI_DEVICE_ID_MYLEX_DAC960_BA, .subvendor = PCI_ANY_ID, .subdevice = PCI_ANY_ID, diff --git a/drivers/block/DAC960.h b/drivers/block/DAC960.h index d5e8e7190c9..a82f37f749a 100644 --- a/drivers/block/DAC960.h +++ b/drivers/block/DAC960.h @@ -2114,7 +2114,8 @@ typedef enum DAC960_LA_Controller = 3, /* DAC1164P */ DAC960_PG_Controller = 4, /* DAC960PTL/PJ/PG */ DAC960_PD_Controller = 5, /* DAC960PU/PD/PL/P */ - DAC960_P_Controller = 6 /* DAC960PU/PD/PL/P */ + DAC960_P_Controller = 6, /* DAC960PU/PD/PL/P */ + DAC960_GEM_Controller = 7, /* AcceleRAID 4/5/600 */ } DAC960_HardwareType_T; @@ -2541,6 +2542,320 @@ void dma_addr_writeql(dma_addr_t addr, void __iomem *write_address) } /* + Define the DAC960 GEM Series Controller Interface Register Offsets. + */ + +#define DAC960_GEM_RegisterWindowSize 0x600 + +typedef enum +{ + DAC960_GEM_InboundDoorBellRegisterReadSetOffset = 0x214, + DAC960_GEM_InboundDoorBellRegisterClearOffset = 0x218, + DAC960_GEM_OutboundDoorBellRegisterReadSetOffset = 0x224, + DAC960_GEM_OutboundDoorBellRegisterClearOffset = 0x228, + DAC960_GEM_InterruptStatusRegisterOffset = 0x208, + DAC960_GEM_InterruptMaskRegisterReadSetOffset = 0x22C, + DAC960_GEM_InterruptMaskRegisterClearOffset = 0x230, + DAC960_GEM_CommandMailboxBusAddressOffset = 0x510, + DAC960_GEM_CommandStatusOffset = 0x518, + DAC960_GEM_ErrorStatusRegisterReadSetOffset = 0x224, + DAC960_GEM_ErrorStatusRegisterClearOffset = 0x228, +} +DAC960_GEM_RegisterOffsets_T; + +/* + Define the structure of the DAC960 GEM Series Inbound Door Bell + */ + +typedef union DAC960_GEM_InboundDoorBellRegister +{ + unsigned int All; + struct { + unsigned int :24; + boolean HardwareMailboxNewCommand:1; + boolean AcknowledgeHardwareMailboxStatus:1; + boolean GenerateInterrupt:1; + boolean ControllerReset:1; + boolean MemoryMailboxNewCommand:1; + unsigned int :3; + } Write; + struct { + unsigned int :24; + boolean HardwareMailboxFull:1; + boolean InitializationInProgress:1; + unsigned int :6; + } Read; +} +DAC960_GEM_InboundDoorBellRegister_T; + +/* + Define the structure of the DAC960 GEM Series Outbound Door Bell Register. + */ +typedef union DAC960_GEM_OutboundDoorBellRegister +{ + unsigned int All; + struct { + unsigned int :24; + boolean AcknowledgeHardwareMailboxInterrupt:1; + boolean AcknowledgeMemoryMailboxInterrupt:1; + unsigned int :6; + } Write; + struct { + unsigned int :24; + boolean HardwareMailboxStatusAvailable:1; + boolean MemoryMailboxStatusAvailable:1; + unsigned int :6; + } Read; +} +DAC960_GEM_OutboundDoorBellRegister_T; + +/* + Define the structure of the DAC960 GEM Series Interrupt Mask Register. + */ +typedef union DAC960_GEM_InterruptMaskRegister +{ + unsigned int All; + struct { + unsigned int :16; + unsigned int :8; + unsigned int HardwareMailboxInterrupt:1; + unsigned int MemoryMailboxInterrupt:1; + unsigned int :6; + } Bits; +} +DAC960_GEM_InterruptMaskRegister_T; + +/* + Define the structure of the DAC960 GEM Series Error Status Register. + */ + +typedef union DAC960_GEM_ErrorStatusRegister +{ + unsigned int All; + struct { + unsigned int :24; + unsigned int :5; + boolean ErrorStatusPending:1; + unsigned int :2; + } Bits; +} +DAC960_GEM_ErrorStatusRegister_T; + +/* + Define inline functions to provide an abstraction for reading and writing the + DAC960 GEM Series Controller Interface Registers. +*/ + +static inline +void DAC960_GEM_HardwareMailboxNewCommand(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InboundDoorBellRegister_T InboundDoorBellRegister; + InboundDoorBellRegister.All = 0; + InboundDoorBellRegister.Write.HardwareMailboxNewCommand = true; + writel(InboundDoorBellRegister.All, + ControllerBaseAddress + DAC960_GEM_InboundDoorBellRegisterReadSetOffset); +} + +static inline +void DAC960_GEM_AcknowledgeHardwareMailboxStatus(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InboundDoorBellRegister_T InboundDoorBellRegister; + InboundDoorBellRegister.All = 0; + InboundDoorBellRegister.Write.AcknowledgeHardwareMailboxStatus = true; + writel(InboundDoorBellRegister.All, + ControllerBaseAddress + DAC960_GEM_InboundDoorBellRegisterClearOffset); +} + +static inline +void DAC960_GEM_GenerateInterrupt(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InboundDoorBellRegister_T InboundDoorBellRegister; + InboundDoorBellRegister.All = 0; + InboundDoorBellRegister.Write.GenerateInterrupt = true; + writel(InboundDoorBellRegister.All, + ControllerBaseAddress + DAC960_GEM_InboundDoorBellRegisterReadSetOffset); +} + +static inline +void DAC960_GEM_ControllerReset(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InboundDoorBellRegister_T InboundDoorBellRegister; + InboundDoorBellRegister.All = 0; + InboundDoorBellRegister.Write.ControllerReset = true; + writel(InboundDoorBellRegister.All, + ControllerBaseAddress + DAC960_GEM_InboundDoorBellRegisterReadSetOffset); +} + +static inline +void DAC960_GEM_MemoryMailboxNewCommand(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InboundDoorBellRegister_T InboundDoorBellRegister; + InboundDoorBellRegister.All = 0; + InboundDoorBellRegister.Write.MemoryMailboxNewCommand = true; + writel(InboundDoorBellRegister.All, + ControllerBaseAddress + DAC960_GEM_InboundDoorBellRegisterReadSetOffset); +} + +static inline +boolean DAC960_GEM_HardwareMailboxFullP(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InboundDoorBellRegister_T InboundDoorBellRegister; + InboundDoorBellRegister.All = + readl(ControllerBaseAddress + + DAC960_GEM_InboundDoorBellRegisterReadSetOffset); + return InboundDoorBellRegister.Read.HardwareMailboxFull; +} + +static inline +boolean DAC960_GEM_InitializationInProgressP(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InboundDoorBellRegister_T InboundDoorBellRegister; + InboundDoorBellRegister.All = + readl(ControllerBaseAddress + + DAC960_GEM_InboundDoorBellRegisterReadSetOffset); + return InboundDoorBellRegister.Read.InitializationInProgress; +} + +static inline +void DAC960_GEM_AcknowledgeHardwareMailboxInterrupt(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_OutboundDoorBellRegister_T OutboundDoorBellRegister; + OutboundDoorBellRegister.All = 0; + OutboundDoorBellRegister.Write.AcknowledgeHardwareMailboxInterrupt = true; + writel(OutboundDoorBellRegister.All, + ControllerBaseAddress + DAC960_GEM_OutboundDoorBellRegisterClearOffset); +} + +static inline +void DAC960_GEM_AcknowledgeMemoryMailboxInterrupt(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_OutboundDoorBellRegister_T OutboundDoorBellRegister; + OutboundDoorBellRegister.All = 0; + OutboundDoorBellRegister.Write.AcknowledgeMemoryMailboxInterrupt = true; + writel(OutboundDoorBellRegister.All, + ControllerBaseAddress + DAC960_GEM_OutboundDoorBellRegisterClearOffset); +} + +static inline +void DAC960_GEM_AcknowledgeInterrupt(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_OutboundDoorBellRegister_T OutboundDoorBellRegister; + OutboundDoorBellRegister.All = 0; + OutboundDoorBellRegister.Write.AcknowledgeHardwareMailboxInterrupt = true; + OutboundDoorBellRegister.Write.AcknowledgeMemoryMailboxInterrupt = true; + writel(OutboundDoorBellRegister.All, + ControllerBaseAddress + DAC960_GEM_OutboundDoorBellRegisterClearOffset); +} + +static inline +boolean DAC960_GEM_HardwareMailboxStatusAvailableP(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_OutboundDoorBellRegister_T OutboundDoorBellRegister; + OutboundDoorBellRegister.All = + readl(ControllerBaseAddress + + DAC960_GEM_OutboundDoorBellRegisterReadSetOffset); + return OutboundDoorBellRegister.Read.HardwareMailboxStatusAvailable; +} + +static inline +boolean DAC960_GEM_MemoryMailboxStatusAvailableP(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_OutboundDoorBellRegister_T OutboundDoorBellRegister; + OutboundDoorBellRegister.All = + readl(ControllerBaseAddress + + DAC960_GEM_OutboundDoorBellRegisterReadSetOffset); + return OutboundDoorBellRegister.Read.MemoryMailboxStatusAvailable; +} + +static inline +void DAC960_GEM_EnableInterrupts(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InterruptMaskRegister_T InterruptMaskRegister; + InterruptMaskRegister.All = 0; + InterruptMaskRegister.Bits.HardwareMailboxInterrupt = true; + InterruptMaskRegister.Bits.MemoryMailboxInterrupt = true; + writel(InterruptMaskRegister.All, + ControllerBaseAddress + DAC960_GEM_InterruptMaskRegisterClearOffset); +} + +static inline +void DAC960_GEM_DisableInterrupts(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InterruptMaskRegister_T InterruptMaskRegister; + InterruptMaskRegister.All = 0; + InterruptMaskRegister.Bits.HardwareMailboxInterrupt = true; + InterruptMaskRegister.Bits.MemoryMailboxInterrupt = true; + writel(InterruptMaskRegister.All, + ControllerBaseAddress + DAC960_GEM_InterruptMaskRegisterReadSetOffset); +} + +static inline +boolean DAC960_GEM_InterruptsEnabledP(void __iomem *ControllerBaseAddress) +{ + DAC960_GEM_InterruptMaskRegister_T InterruptMaskRegister; + InterruptMaskRegister.All = + readl(ControllerBaseAddress + + DAC960_GEM_InterruptMaskRegisterReadSetOffset); + return !(InterruptMaskRegister.Bits.HardwareMailboxInterrupt || + InterruptMaskRegister.Bits.MemoryMailboxInterrupt); +} + +static inline +void DAC960_GEM_WriteCommandMailbox(DAC960_V2_CommandMailbox_T + *MemoryCommandMailbox, + DAC960_V2_CommandMailbox_T + *CommandMailbox) +{ + memcpy(&MemoryCommandMailbox->Words[1], &CommandMailbox->Words[1], + sizeof(DAC960_V2_CommandMailbox_T) - sizeof(unsigned int)); + wmb(); + MemoryCommandMailbox->Words[0] = CommandMailbox->Words[0]; + mb(); +} + +static inline +void DAC960_GEM_WriteHardwareMailbox(void __iomem *ControllerBaseAddress, + dma_addr_t CommandMailboxDMA) +{ + dma_addr_writeql(CommandMailboxDMA, + ControllerBaseAddress + + DAC960_GEM_CommandMailboxBusAddressOffset); +} + +static inline DAC960_V2_CommandIdentifier_T +DAC960_GEM_ReadCommandIdentifier(void __iomem *ControllerBaseAddress) +{ + return readw(ControllerBaseAddress + DAC960_GEM_CommandStatusOffset); +} + +static inline DAC960_V2_CommandStatus_T +DAC960_GEM_ReadCommandStatus(void __iomem *ControllerBaseAddress) +{ + return readw(ControllerBaseAddress + DAC960_GEM_CommandStatusOffset + 2); +} + +static inline boolean +DAC960_GEM_ReadErrorStatus(void __iomem *ControllerBaseAddress, + unsigned char *ErrorStatus, + unsigned char *Parameter0, + unsigned char *Parameter1) +{ + DAC960_GEM_ErrorStatusRegister_T ErrorStatusRegister; + ErrorStatusRegister.All = + readl(ControllerBaseAddress + DAC960_GEM_ErrorStatusRegisterReadSetOffset); + if (!ErrorStatusRegister.Bits.ErrorStatusPending) return false; + ErrorStatusRegister.Bits.ErrorStatusPending = false; + *ErrorStatus = ErrorStatusRegister.All; + *Parameter0 = + readb(ControllerBaseAddress + DAC960_GEM_CommandMailboxBusAddressOffset + 0); + *Parameter1 = + readb(ControllerBaseAddress + DAC960_GEM_CommandMailboxBusAddressOffset + 1); + writel(0x03000000, ControllerBaseAddress + + DAC960_GEM_ErrorStatusRegisterClearOffset); + return true; +} + +/* Define the DAC960 BA Series Controller Interface Register Offsets. */ diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig index e830be1a3ae..b594768b024 100644 --- a/drivers/block/Kconfig +++ b/drivers/block/Kconfig @@ -6,7 +6,7 @@ menu "Block devices" config BLK_DEV_FD tristate "Normal floppy disk support" - depends on (!ARCH_S390 && !M68K && !IA64 && !UML) || Q40 || (SUN3X && BROKEN) || ARCH_RPC || ARCH_EBSA285 + depends on (!ARCH_S390 && !M68K && !IA64 && !UML && !ARM) || Q40 || (SUN3X && BROKEN) || ARCH_RPC || ARCH_EBSA285 ---help--- If you want to use the floppy disk drive(s) of your PC under Linux, say Y. Information about this driver, especially important for IBM @@ -105,7 +105,7 @@ config ATARI_SLM config BLK_DEV_XD tristate "XT hard disk support" - depends on ISA + depends on ISA && ISA_DMA_API help Very old 8 bit hard disk controllers used in the IBM XT computer will be supported if you say Y here. diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index aa8b547ffaf..721ba808604 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -1,5 +1,5 @@ /* Copyright (c) 2004 Coraid, Inc. See COPYING for GPL terms. */ -#define VERSION "6" +#define VERSION "10" #define AOE_MAJOR 152 #define DEVICE_NAME "aoe" diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index 4780f7926d4..0e97fcb9f3a 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -37,6 +37,13 @@ static ssize_t aoedisk_show_netif(struct gendisk * disk, char *page) return snprintf(page, PAGE_SIZE, "%s\n", d->ifp->name); } +/* firmware version */ +static ssize_t aoedisk_show_fwver(struct gendisk * disk, char *page) +{ + struct aoedev *d = disk->private_data; + + return snprintf(page, PAGE_SIZE, "0x%04x\n", (unsigned int) d->fw_ver); +} static struct disk_attribute disk_attr_state = { .attr = {.name = "state", .mode = S_IRUGO }, @@ -50,6 +57,10 @@ static struct disk_attribute disk_attr_netif = { .attr = {.name = "netif", .mode = S_IRUGO }, .show = aoedisk_show_netif }; +static struct disk_attribute disk_attr_fwver = { + .attr = {.name = "firmware-version", .mode = S_IRUGO }, + .show = aoedisk_show_fwver +}; static void aoedisk_add_sysfs(struct aoedev *d) @@ -57,6 +68,7 @@ aoedisk_add_sysfs(struct aoedev *d) sysfs_create_file(&d->gd->kobj, &disk_attr_state.attr); sysfs_create_file(&d->gd->kobj, &disk_attr_mac.attr); sysfs_create_file(&d->gd->kobj, &disk_attr_netif.attr); + sysfs_create_file(&d->gd->kobj, &disk_attr_fwver.attr); } void aoedisk_rm_sysfs(struct aoedev *d) @@ -64,6 +76,7 @@ aoedisk_rm_sysfs(struct aoedev *d) sysfs_remove_link(&d->gd->kobj, "state"); sysfs_remove_link(&d->gd->kobj, "mac"); sysfs_remove_link(&d->gd->kobj, "netif"); + sysfs_remove_link(&d->gd->kobj, "firmware-version"); } static int diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index ec16c64dd11..6e231c5a119 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -109,25 +109,22 @@ aoedev_set(ulong sysminor, unsigned char *addr, struct net_device *ifp, ulong bu spin_lock_irqsave(&devlist_lock, flags); for (d=devlist; d; d=d->next) - if (d->sysminor == sysminor - || memcmp(d->addr, addr, sizeof d->addr) == 0) + if (d->sysminor == sysminor) break; if (d == NULL && (d = aoedev_newdev(bufcnt)) == NULL) { spin_unlock_irqrestore(&devlist_lock, flags); printk(KERN_INFO "aoe: aoedev_set: aoedev_newdev failure.\n"); return NULL; - } + } /* if newdev, (d->flags & DEVFL_UP) == 0 for below */ spin_unlock_irqrestore(&devlist_lock, flags); spin_lock_irqsave(&d->lock, flags); d->ifp = ifp; - - if (d->sysminor != sysminor - || (d->flags & DEVFL_UP) == 0) { + memcpy(d->addr, addr, sizeof d->addr); + if ((d->flags & DEVFL_UP) == 0) { aoedev_downdev(d); /* flushes outstanding frames */ - memcpy(d->addr, addr, sizeof d->addr); d->sysminor = sysminor; d->aoemajor = AOEMAJOR(sysminor); d->aoeminor = AOEMINOR(sysminor); diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index bc92aacb6da..9e6f51c528b 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -7,6 +7,7 @@ #include <linux/hdreg.h> #include <linux/blkdev.h> #include <linux/netdevice.h> +#include <linux/moduleparam.h> #include "aoe.h" #define NECODES 5 @@ -26,6 +27,19 @@ enum { }; static char aoe_iflist[IFLISTSZ]; +module_param_string(aoe_iflist, aoe_iflist, IFLISTSZ, 0600); +MODULE_PARM_DESC(aoe_iflist, "aoe_iflist=\"dev1 [dev2 ...]\"\n"); + +#ifndef MODULE +static int __init aoe_iflist_setup(char *str) +{ + strncpy(aoe_iflist, str, IFLISTSZ); + aoe_iflist[IFLISTSZ - 1] = '\0'; + return 1; +} + +__setup("aoe_iflist=", aoe_iflist_setup); +#endif int is_aoe_netif(struct net_device *ifp) @@ -36,7 +50,8 @@ is_aoe_netif(struct net_device *ifp) if (aoe_iflist[0] == '\0') return 1; - for (p = aoe_iflist; *p; p = q + strspn(q, WHITESPACE)) { + p = aoe_iflist + strspn(aoe_iflist, WHITESPACE); + for (; *p; p = q + strspn(q, WHITESPACE)) { q = p + strcspn(p, WHITESPACE); if (q != p) len = q - p; diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 42dfa281a88..f0c1084b840 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -3345,7 +3345,7 @@ static inline int set_geometry(unsigned int cmd, struct floppy_struct *g, struct block_device *bdev = opened_bdev[cnt]; if (!bdev || ITYPE(drive_state[cnt].fd_device) != type) continue; - __invalidate_device(bdev, 0); + __invalidate_device(bdev); } up(&open_lock); } else { diff --git a/drivers/block/genhd.c b/drivers/block/genhd.c index ab4db71375e..8bbe01d4b48 100644 --- a/drivers/block/genhd.c +++ b/drivers/block/genhd.c @@ -14,6 +14,7 @@ #include <linux/slab.h> #include <linux/kmod.h> #include <linux/kobj_map.h> +#include <linux/buffer_head.h> #define MAX_PROBE_HASH 255 /* random */ @@ -676,7 +677,8 @@ int invalidate_partition(struct gendisk *disk, int index) int res = 0; struct block_device *bdev = bdget_disk(disk, index); if (bdev) { - res = __invalidate_device(bdev, 1); + fsync_bdev(bdev); + res = __invalidate_device(bdev); bdput(bdev); } return res; diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index efdf04450bf..9e268ddedfb 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -78,6 +78,7 @@ #define DBG_RX 0x0200 #define DBG_TX 0x0400 static unsigned int debugflags; +static unsigned int nbds_max = 16; #endif /* NDEBUG */ static struct nbd_device nbd_dev[MAX_NBD]; @@ -647,7 +648,13 @@ static int __init nbd_init(void) return -EIO; } - for (i = 0; i < MAX_NBD; i++) { + if (nbds_max > MAX_NBD) { + printk(KERN_CRIT "nbd: cannot allocate more than %u nbds; %u requested.\n", MAX_NBD, + nbds_max); + return -EINVAL; + } + + for (i = 0; i < nbds_max; i++) { struct gendisk *disk = alloc_disk(1); if (!disk) goto out; @@ -673,7 +680,7 @@ static int __init nbd_init(void) dprintk(DBG_INIT, "nbd: debugflags=0x%x\n", debugflags); devfs_mk_dir("nbd"); - for (i = 0; i < MAX_NBD; i++) { + for (i = 0; i < nbds_max; i++) { struct gendisk *disk = nbd_dev[i].disk; nbd_dev[i].file = NULL; nbd_dev[i].magic = LO_MAGIC; @@ -706,8 +713,9 @@ out: static void __exit nbd_cleanup(void) { int i; - for (i = 0; i < MAX_NBD; i++) { + for (i = 0; i < nbds_max; i++) { struct gendisk *disk = nbd_dev[i].disk; + nbd_dev[i].magic = 0; if (disk) { del_gendisk(disk); blk_cleanup_queue(disk->queue); @@ -725,6 +733,8 @@ module_exit(nbd_cleanup); MODULE_DESCRIPTION("Network Block Device"); MODULE_LICENSE("GPL"); +module_param(nbds_max, int, 0444); +MODULE_PARM_DESC(nbds_max, "How many network block devices to initialize."); #ifndef NDEBUG module_param(debugflags, int, 0644); MODULE_PARM_DESC(debugflags, "flags for controlling debug output"); diff --git a/drivers/block/noop-iosched.c b/drivers/block/noop-iosched.c index 888c477e02b..b1730b62c37 100644 --- a/drivers/block/noop-iosched.c +++ b/drivers/block/noop-iosched.c @@ -13,34 +13,13 @@ static int elevator_noop_merge(request_queue_t *q, struct request **req, struct bio *bio) { - struct list_head *entry = &q->queue_head; - struct request *__rq; int ret; - if ((ret = elv_try_last_merge(q, bio))) { + ret = elv_try_last_merge(q, bio); + if (ret != ELEVATOR_NO_MERGE) *req = q->last_merge; - return ret; - } - while ((entry = entry->prev) != &q->queue_head) { - __rq = list_entry_rq(entry); - - if (__rq->flags & (REQ_SOFTBARRIER | REQ_HARDBARRIER)) - break; - else if (__rq->flags & REQ_STARTED) - break; - - if (!blk_fs_request(__rq)) - continue; - - if ((ret = elv_try_merge(__rq, bio))) { - *req = __rq; - q->last_merge = __rq; - return ret; - } - } - - return ELEVATOR_NO_MERGE; + return ret; } static void elevator_noop_merge_requests(request_queue_t *q, struct request *req, diff --git a/drivers/block/rd.c b/drivers/block/rd.c index 145c1fbffe0..f8a8b68dced 100644 --- a/drivers/block/rd.c +++ b/drivers/block/rd.c @@ -74,7 +74,7 @@ static struct request_queue *rd_queue[CONFIG_BLK_DEV_RAM_COUNT]; * architecture-specific setup routine (from the stored boot sector * information). */ -int rd_size = CONFIG_BLK_DEV_RAM_SIZE; /* Size of the RAM disks */ +static int rd_size = CONFIG_BLK_DEV_RAM_SIZE; /* Size of the RAM disks */ /* * It would be very desirable to have a soft-blocksize (that in the case * of the ramdisk driver is also the hardblocksize ;) of PAGE_SIZE because diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c index 9deca49c71f..beaa561f2ed 100644 --- a/drivers/cdrom/cdrom.c +++ b/drivers/cdrom/cdrom.c @@ -645,7 +645,7 @@ static int cdrom_mrw_exit(struct cdrom_device_info *cdi) ret = cdrom_mrw_bgformat_susp(cdi, 0); } - if (!ret) + if (!ret && cdi->media_written) ret = cdrom_flush_cache(cdi); return ret; diff --git a/drivers/cdrom/cdu31a.c b/drivers/cdrom/cdu31a.c index 647a71b12a2..ac96de15d83 100644 --- a/drivers/cdrom/cdu31a.c +++ b/drivers/cdrom/cdu31a.c @@ -292,7 +292,7 @@ module_param(cdu31a_irq, int, 0); /* The interrupt handler will wake this queue up when it gets an interrupts. */ -DECLARE_WAIT_QUEUE_HEAD(cdu31a_irq_wait); +static DECLARE_WAIT_QUEUE_HEAD(cdu31a_irq_wait); static int irq_flag = 0; static int curr_control_reg = 0; /* Current value of the control register */ @@ -2947,7 +2947,7 @@ static int scd_block_media_changed(struct gendisk *disk) return cdrom_media_changed(&scd_info); } -struct block_device_operations scd_bdops = +static struct block_device_operations scd_bdops = { .owner = THIS_MODULE, .open = scd_block_open, @@ -3216,7 +3216,7 @@ errout3: } -void __exit cdu31a_exit(void) +static void __exit cdu31a_exit(void) { del_gendisk(scd_gendisk); put_disk(scd_gendisk); diff --git a/drivers/cdrom/mcdx.c b/drivers/cdrom/mcdx.c index ccde7ab491d..07bbd24e3c1 100644 --- a/drivers/cdrom/mcdx.c +++ b/drivers/cdrom/mcdx.c @@ -107,20 +107,20 @@ static const char *mcdx_c_version The _direct_ size is the number of sectors we're allowed to skip directly (performing a read instead of requesting the new sector needed */ -const int REQUEST_SIZE = 800; /* should be less then 255 * 4 */ -const int DIRECT_SIZE = 400; /* should be less then REQUEST_SIZE */ +static const int REQUEST_SIZE = 800; /* should be less then 255 * 4 */ +static const int DIRECT_SIZE = 400; /* should be less then REQUEST_SIZE */ enum drivemodes { TOC, DATA, RAW, COOKED }; enum datamodes { MODE0, MODE1, MODE2 }; enum resetmodes { SOFT, HARD }; -const int SINGLE = 0x01; /* single speed drive (FX001S, LU) */ -const int DOUBLE = 0x02; /* double speed drive (FX001D, ..? */ -const int DOOR = 0x04; /* door locking capability */ -const int MULTI = 0x08; /* multi session capability */ +static const int SINGLE = 0x01; /* single speed drive (FX001S, LU) */ +static const int DOUBLE = 0x02; /* double speed drive (FX001D, ..? */ +static const int DOOR = 0x04; /* door locking capability */ +static const int MULTI = 0x08; /* multi session capability */ -const unsigned char READ1X = 0xc0; -const unsigned char READ2X = 0xc1; +static const unsigned char READ1X = 0xc0; +static const unsigned char READ2X = 0xc1; /* DECLARATIONS ****************************************************/ @@ -210,9 +210,7 @@ struct s_drive_stuff { repeated here to show what's going on. And to sense, if they're changed elsewhere. */ -/* declared in blk.h */ -int mcdx_init(void); -void do_mcdx_request(request_queue_t * q); +static int mcdx_init(void); static int mcdx_block_open(struct inode *inode, struct file *file) { @@ -569,7 +567,7 @@ static int mcdx_audio_ioctl(struct cdrom_device_info *cdi, } } -void do_mcdx_request(request_queue_t * q) +static void do_mcdx_request(request_queue_t * q) { struct s_drive_stuff *stuffp; struct request *req; @@ -1028,7 +1026,7 @@ int __mcdx_init(void) return 0; } -void __exit mcdx_exit(void) +static void __exit mcdx_exit(void) { int i; @@ -1075,7 +1073,7 @@ module_exit(mcdx_exit); /* Support functions ************************************************/ -int __init mcdx_init_drive(int drive) +static int __init mcdx_init_drive(int drive) { struct s_version version; struct gendisk *disk; @@ -1261,7 +1259,7 @@ int __init mcdx_init_drive(int drive) return 0; } -int __init mcdx_init(void) +static int __init mcdx_init(void) { int drive; xwarn("Version 2.14(hs) \n"); diff --git a/drivers/cdrom/sbpcd.c b/drivers/cdrom/sbpcd.c index fc2c433f6a2..452d3467515 100644 --- a/drivers/cdrom/sbpcd.c +++ b/drivers/cdrom/sbpcd.c @@ -5895,7 +5895,7 @@ int __init sbpcd_init(void) } /*==========================================================================*/ #ifdef MODULE -void sbpcd_exit(void) +static void sbpcd_exit(void) { int j; diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index 096a1202ea0..5ed6515ae01 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -153,7 +153,7 @@ config DIGIEPCA config ESPSERIAL tristate "Hayes ESP serial port support" - depends on SERIAL_NONSTANDARD && ISA && BROKEN_ON_SMP + depends on SERIAL_NONSTANDARD && ISA && BROKEN_ON_SMP && ISA_DMA_API help This is a driver which supports Hayes ESP serial ports. Both single port cards and multiport cards are supported. Make sure to read @@ -195,7 +195,7 @@ config ISI config SYNCLINK tristate "Microgate SyncLink card support" - depends on SERIAL_NONSTANDARD && PCI + depends on SERIAL_NONSTANDARD && PCI && ISA_DMA_API help Provides support for the SyncLink ISA and PCI multiprotocol serial adapters. These adapters support asynchronous and HDLC bit @@ -399,6 +399,20 @@ config SGI_SNSC controller communication from user space (you want this!), say Y. Otherwise, say N. +config SGI_TIOCX + bool "SGI TIO CX driver support" + depends on (IA64_SGI_SN2 || IA64_GENERIC) + help + If you have an SGI Altix and you have fpga devices attached + to your TIO, say Y here, otherwise say N. + +config SGI_MBCS + tristate "SGI FPGA Core Services driver support" + depends on SGI_TIOCX + help + If you have an SGI Altix with an attached SABrick + say Y or M here, otherwise say N. + source "drivers/serial/Kconfig" config UNIX98_PTYS @@ -968,7 +982,7 @@ config MAX_RAW_DEVS config HANGCHECK_TIMER tristate "Hangcheck timer" - depends on X86_64 || X86 + depends on X86_64 || X86 || IA64 || PPC64 || ARCH_S390 help The hangcheck-timer module detects when the system has gone out to lunch past a certain margin. It can reboot the system diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 54ed76af1a4..e3f5c32aac5 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -42,11 +42,12 @@ obj-$(CONFIG_SX) += sx.o generic_serial.o obj-$(CONFIG_RIO) += rio/ generic_serial.o obj-$(CONFIG_HVC_CONSOLE) += hvc_console.o hvsi.o obj-$(CONFIG_RAW_DRIVER) += raw.o -obj-$(CONFIG_SGI_SNSC) += snsc.o +obj-$(CONFIG_SGI_SNSC) += snsc.o snsc_event.o obj-$(CONFIG_MMTIMER) += mmtimer.o obj-$(CONFIG_VIOCONS) += viocons.o obj-$(CONFIG_VIOTAPE) += viotape.o obj-$(CONFIG_HVCS) += hvcs.o +obj-$(CONFIG_SGI_MBCS) += mbcs.o obj-$(CONFIG_PRINTER) += lp.o obj-$(CONFIG_TIPAR) += tipar.o diff --git a/drivers/char/agp/ali-agp.c b/drivers/char/agp/ali-agp.c index c86a22c5499..0212febda65 100644 --- a/drivers/char/agp/ali-agp.c +++ b/drivers/char/agp/ali-agp.c @@ -192,7 +192,7 @@ static struct aper_size_info_32 ali_generic_sizes[7] = {4, 1024, 0, 3} }; -struct agp_bridge_driver ali_generic_bridge = { +static struct agp_bridge_driver ali_generic_bridge = { .owner = THIS_MODULE, .aperture_sizes = ali_generic_sizes, .size_type = U32_APER_SIZE, @@ -215,7 +215,7 @@ struct agp_bridge_driver ali_generic_bridge = { .agp_destroy_page = ali_destroy_page, }; -struct agp_bridge_driver ali_m1541_bridge = { +static struct agp_bridge_driver ali_m1541_bridge = { .owner = THIS_MODULE, .aperture_sizes = ali_generic_sizes, .size_type = U32_APER_SIZE, diff --git a/drivers/char/agp/amd-k7-agp.c b/drivers/char/agp/amd-k7-agp.c index f1ea87ea6b6..e62a3c2c44a 100644 --- a/drivers/char/agp/amd-k7-agp.c +++ b/drivers/char/agp/amd-k7-agp.c @@ -358,7 +358,7 @@ static struct gatt_mask amd_irongate_masks[] = {.mask = 1, .type = 0} }; -struct agp_bridge_driver amd_irongate_driver = { +static struct agp_bridge_driver amd_irongate_driver = { .owner = THIS_MODULE, .aperture_sizes = amd_irongate_sizes, .size_type = LVL2_APER_SIZE, diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c index 905f0629c44..399c042f68f 100644 --- a/drivers/char/agp/amd64-agp.c +++ b/drivers/char/agp/amd64-agp.c @@ -243,7 +243,7 @@ static void amd64_cleanup(void) } -struct agp_bridge_driver amd_8151_driver = { +static struct agp_bridge_driver amd_8151_driver = { .owner = THIS_MODULE, .aperture_sizes = amd_8151_sizes, .size_type = U32_APER_SIZE, diff --git a/drivers/char/agp/ati-agp.c b/drivers/char/agp/ati-agp.c index 757dde006fc..a65f8827c28 100644 --- a/drivers/char/agp/ati-agp.c +++ b/drivers/char/agp/ati-agp.c @@ -393,7 +393,7 @@ static int ati_free_gatt_table(struct agp_bridge_data *bridge) return 0; } -struct agp_bridge_driver ati_generic_bridge = { +static struct agp_bridge_driver ati_generic_bridge = { .owner = THIS_MODULE, .aperture_sizes = ati_generic_sizes, .size_type = LVL2_APER_SIZE, diff --git a/drivers/char/agp/backend.c b/drivers/char/agp/backend.c index c3442f3c648..2f3dfb63bdc 100644 --- a/drivers/char/agp/backend.c +++ b/drivers/char/agp/backend.c @@ -97,7 +97,7 @@ void agp_backend_release(struct agp_bridge_data *bridge) EXPORT_SYMBOL(agp_backend_release); -struct { int mem, agp; } maxes_table[] = { +static struct { int mem, agp; } maxes_table[] = { {0, 0}, {32, 4}, {64, 28}, @@ -322,7 +322,7 @@ static int __init agp_init(void) return 0; } -void __exit agp_exit(void) +static void __exit agp_exit(void) { } diff --git a/drivers/char/agp/efficeon-agp.c b/drivers/char/agp/efficeon-agp.c index 2a87cecdc91..1383c3165ea 100644 --- a/drivers/char/agp/efficeon-agp.c +++ b/drivers/char/agp/efficeon-agp.c @@ -303,7 +303,7 @@ static int efficeon_remove_memory(struct agp_memory * mem, off_t pg_start, int t } -struct agp_bridge_driver efficeon_driver = { +static struct agp_bridge_driver efficeon_driver = { .owner = THIS_MODULE, .aperture_sizes = efficeon_generic_sizes, .size_type = LVL2_APER_SIZE, diff --git a/drivers/char/agp/frontend.c b/drivers/char/agp/frontend.c index f633623ac80..3dfb6648547 100644 --- a/drivers/char/agp/frontend.c +++ b/drivers/char/agp/frontend.c @@ -235,7 +235,7 @@ static void agp_insert_into_pool(struct agp_memory * temp) /* File private list routines */ -struct agp_file_private *agp_find_private(pid_t pid) +static struct agp_file_private *agp_find_private(pid_t pid) { struct agp_file_private *curr; @@ -250,7 +250,7 @@ struct agp_file_private *agp_find_private(pid_t pid) return NULL; } -void agp_insert_file_private(struct agp_file_private * priv) +static void agp_insert_file_private(struct agp_file_private * priv) { struct agp_file_private *prev; @@ -262,7 +262,7 @@ void agp_insert_file_private(struct agp_file_private * priv) agp_fe.file_priv_list = priv; } -void agp_remove_file_private(struct agp_file_private * priv) +static void agp_remove_file_private(struct agp_file_private * priv) { struct agp_file_private *next; struct agp_file_private *prev; diff --git a/drivers/char/agp/nvidia-agp.c b/drivers/char/agp/nvidia-agp.c index 4f7a3e8bc91..80dafa3030b 100644 --- a/drivers/char/agp/nvidia-agp.c +++ b/drivers/char/agp/nvidia-agp.c @@ -288,7 +288,7 @@ static struct gatt_mask nvidia_generic_masks[] = }; -struct agp_bridge_driver nvidia_driver = { +static struct agp_bridge_driver nvidia_driver = { .owner = THIS_MODULE, .aperture_sizes = nvidia_generic_sizes, .size_type = U8_APER_SIZE, diff --git a/drivers/char/agp/sis-agp.c b/drivers/char/agp/sis-agp.c index cfccacb2a64..ebc05554045 100644 --- a/drivers/char/agp/sis-agp.c +++ b/drivers/char/agp/sis-agp.c @@ -119,7 +119,7 @@ static struct aper_size_info_8 sis_generic_sizes[7] = {4, 1024, 0, 3} }; -struct agp_bridge_driver sis_driver = { +static struct agp_bridge_driver sis_driver = { .owner = THIS_MODULE, .aperture_sizes = sis_generic_sizes, .size_type = U8_APER_SIZE, diff --git a/drivers/char/agp/sworks-agp.c b/drivers/char/agp/sworks-agp.c index bb338d9134e..10c23302dd8 100644 --- a/drivers/char/agp/sworks-agp.c +++ b/drivers/char/agp/sworks-agp.c @@ -409,7 +409,7 @@ static void serverworks_agp_enable(struct agp_bridge_data *bridge, u32 mode) agp_device_command(command, 0); } -struct agp_bridge_driver sworks_driver = { +static struct agp_bridge_driver sworks_driver = { .owner = THIS_MODULE, .aperture_sizes = serverworks_sizes, .size_type = LVL2_APER_SIZE, diff --git a/drivers/char/agp/via-agp.c b/drivers/char/agp/via-agp.c index e1451dd9b6a..c847df575cf 100644 --- a/drivers/char/agp/via-agp.c +++ b/drivers/char/agp/via-agp.c @@ -170,7 +170,7 @@ static void via_tlbflush_agp3(struct agp_memory *mem) } -struct agp_bridge_driver via_agp3_driver = { +static struct agp_bridge_driver via_agp3_driver = { .owner = THIS_MODULE, .aperture_sizes = agp3_generic_sizes, .size_type = U8_APER_SIZE, @@ -193,7 +193,7 @@ struct agp_bridge_driver via_agp3_driver = { .agp_destroy_page = agp_generic_destroy_page, }; -struct agp_bridge_driver via_driver = { +static struct agp_bridge_driver via_driver = { .owner = THIS_MODULE, .aperture_sizes = via_generic_sizes, .size_type = U8_APER_SIZE, diff --git a/drivers/char/dtlk.c b/drivers/char/dtlk.c index 903e4c3cc20..a229915ce1b 100644 --- a/drivers/char/dtlk.c +++ b/drivers/char/dtlk.c @@ -52,7 +52,7 @@ #define KERNEL #include <linux/types.h> #include <linux/fs.h> -#include <linux/mm.h> /* for verify_area */ +#include <linux/mm.h> #include <linux/errno.h> /* for -EBUSY */ #include <linux/ioport.h> /* for request_region */ #include <linux/delay.h> /* for loops_per_jiffy */ diff --git a/drivers/char/hangcheck-timer.c b/drivers/char/hangcheck-timer.c index 83d6b37b36c..78e650fc5b4 100644 --- a/drivers/char/hangcheck-timer.c +++ b/drivers/char/hangcheck-timer.c @@ -3,7 +3,7 @@ * * Driver for a little io fencing timer. * - * Copyright (C) 2002 Oracle Corporation. All rights reserved. + * Copyright (C) 2002, 2003 Oracle. All rights reserved. * * Author: Joel Becker <joel.becker@oracle.com> * @@ -44,11 +44,14 @@ #include <linux/fs.h> #include <linux/mm.h> #include <linux/reboot.h> +#include <linux/smp_lock.h> #include <linux/init.h> +#include <linux/delay.h> #include <asm/uaccess.h> +#include <linux/sysrq.h> -#define VERSION_STR "0.5.0" +#define VERSION_STR "0.9.0" #define DEFAULT_IOFENCE_MARGIN 60 /* Default fudge factor, in seconds */ #define DEFAULT_IOFENCE_TICK 180 /* Default timer timeout, in seconds */ @@ -56,18 +59,89 @@ static int hangcheck_tick = DEFAULT_IOFENCE_TICK; static int hangcheck_margin = DEFAULT_IOFENCE_MARGIN; static int hangcheck_reboot; /* Defaults to not reboot */ +static int hangcheck_dump_tasks; /* Defaults to not dumping SysRQ T */ -/* Driver options */ +/* options - modular */ module_param(hangcheck_tick, int, 0); MODULE_PARM_DESC(hangcheck_tick, "Timer delay."); module_param(hangcheck_margin, int, 0); MODULE_PARM_DESC(hangcheck_margin, "If the hangcheck timer has been delayed more than hangcheck_margin seconds, the driver will fire."); module_param(hangcheck_reboot, int, 0); MODULE_PARM_DESC(hangcheck_reboot, "If nonzero, the machine will reboot when the timer margin is exceeded."); +module_param(hangcheck_dump_tasks, int, 0); +MODULE_PARM_DESC(hangcheck_dump_tasks, "If nonzero, the machine will dump the system task state when the timer margin is exceeded."); -MODULE_AUTHOR("Joel Becker"); +MODULE_AUTHOR("Oracle"); MODULE_DESCRIPTION("Hangcheck-timer detects when the system has gone out to lunch past a certain margin."); MODULE_LICENSE("GPL"); +MODULE_VERSION(VERSION_STR); + +/* options - nonmodular */ +#ifndef MODULE + +static int __init hangcheck_parse_tick(char *str) +{ + int par; + if (get_option(&str,&par)) + hangcheck_tick = par; + return 1; +} + +static int __init hangcheck_parse_margin(char *str) +{ + int par; + if (get_option(&str,&par)) + hangcheck_margin = par; + return 1; +} + +static int __init hangcheck_parse_reboot(char *str) +{ + int par; + if (get_option(&str,&par)) + hangcheck_reboot = par; + return 1; +} + +static int __init hangcheck_parse_dump_tasks(char *str) +{ + int par; + if (get_option(&str,&par)) + hangcheck_dump_tasks = par; + return 1; +} + +__setup("hcheck_tick", hangcheck_parse_tick); +__setup("hcheck_margin", hangcheck_parse_margin); +__setup("hcheck_reboot", hangcheck_parse_reboot); +__setup("hcheck_dump_tasks", hangcheck_parse_dump_tasks); +#endif /* not MODULE */ + +#if defined(CONFIG_X86) || defined(CONFIG_X86_64) +# define HAVE_MONOTONIC +# define TIMER_FREQ 1000000000ULL +#elif defined(CONFIG_ARCH_S390) +/* FA240000 is 1 Second in the IBM time universe (Page 4-38 Principles of Op for zSeries */ +# define TIMER_FREQ 0xFA240000ULL +#elif defined(CONFIG_IA64) +# define TIMER_FREQ ((unsigned long long)local_cpu_data->itc_freq) +#elif defined(CONFIG_PPC64) +# define TIMER_FREQ (HZ*loops_per_jiffy) +#endif + +#ifdef HAVE_MONOTONIC +extern unsigned long long monotonic_clock(void); +#else +static inline unsigned long long monotonic_clock(void) +{ +# ifdef __s390__ + /* returns the TOD. see 4-38 Principles of Op of zSeries */ + return get_clock(); +# else + return get_cycles(); +# endif /* __s390__ */ +} +#endif /* HAVE_MONOTONIC */ /* Last time scheduled */ @@ -78,7 +152,6 @@ static void hangcheck_fire(unsigned long); static struct timer_list hangcheck_ticktock = TIMER_INITIALIZER(hangcheck_fire, 0, 0); -extern unsigned long long monotonic_clock(void); static void hangcheck_fire(unsigned long data) { @@ -92,6 +165,12 @@ static void hangcheck_fire(unsigned long data) tsc_diff = (cur_tsc + (~0ULL - hangcheck_tsc)); /* or something */ if (tsc_diff > hangcheck_tsc_margin) { + if (hangcheck_dump_tasks) { + printk(KERN_CRIT "Hangcheck: Task state:\n"); +#ifdef CONFIG_MAGIC_SYSRQ + handle_sysrq('t', NULL, NULL); +#endif /* CONFIG_MAGIC_SYSRQ */ + } if (hangcheck_reboot) { printk(KERN_CRIT "Hangcheck: hangcheck is restarting the machine.\n"); machine_restart(NULL); @@ -108,10 +187,16 @@ static int __init hangcheck_init(void) { printk("Hangcheck: starting hangcheck timer %s (tick is %d seconds, margin is %d seconds).\n", VERSION_STR, hangcheck_tick, hangcheck_margin); - - hangcheck_tsc_margin = hangcheck_margin + hangcheck_tick; - hangcheck_tsc_margin *= 1000000000; - +#if defined (HAVE_MONOTONIC) + printk("Hangcheck: Using monotonic_clock().\n"); +#elif defined(__s390__) + printk("Hangcheck: Using TOD.\n"); +#else + printk("Hangcheck: Using get_cycles().\n"); +#endif /* HAVE_MONOTONIC */ + hangcheck_tsc_margin = + (unsigned long long)(hangcheck_margin + hangcheck_tick); + hangcheck_tsc_margin *= (unsigned long long)TIMER_FREQ; hangcheck_tsc = monotonic_clock(); mod_timer(&hangcheck_ticktock, jiffies + (hangcheck_tick*HZ)); @@ -123,6 +208,7 @@ static int __init hangcheck_init(void) static void __exit hangcheck_exit(void) { del_timer_sync(&hangcheck_ticktock); + printk("Hangcheck: Stopped hangcheck timer.\n"); } module_init(hangcheck_init); diff --git a/drivers/char/ipmi/ipmi_bt_sm.c b/drivers/char/ipmi/ipmi_bt_sm.c index 225b330115b..5ce9c626903 100644 --- a/drivers/char/ipmi/ipmi_bt_sm.c +++ b/drivers/char/ipmi/ipmi_bt_sm.c @@ -235,7 +235,6 @@ static void reset_flags(struct si_sm_data *bt) if (BT_STATUS & BT_B_BUSY) BT_CONTROL(BT_B_BUSY); BT_CONTROL(BT_CLR_WR_PTR); BT_CONTROL(BT_SMS_ATN); - BT_INTMASK_W(BT_BMC_HWRST); #ifdef DEVELOPMENT_ONLY_NOT_FOR_PRODUCTION if (BT_STATUS & BT_B2H_ATN) { int i; diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index a6606a1aced..d7fb452af7f 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -2588,28 +2588,20 @@ handle_msg_timeout(struct ipmi_recv_msg *msg) deliver_response(msg); } -static void -send_from_recv_msg(ipmi_smi_t intf, struct ipmi_recv_msg *recv_msg, - struct ipmi_smi_msg *smi_msg, - unsigned char seq, long seqid) +static struct ipmi_smi_msg * +smi_from_recv_msg(ipmi_smi_t intf, struct ipmi_recv_msg *recv_msg, + unsigned char seq, long seqid) { - if (!smi_msg) - smi_msg = ipmi_alloc_smi_msg(); + struct ipmi_smi_msg *smi_msg = ipmi_alloc_smi_msg(); if (!smi_msg) /* If we can't allocate the message, then just return, we get 4 retries, so this should be ok. */ - return; + return NULL; memcpy(smi_msg->data, recv_msg->msg.data, recv_msg->msg.data_len); smi_msg->data_size = recv_msg->msg.data_len; smi_msg->msgid = STORE_SEQ_IN_MSGID(seq, seqid); - /* Send the new message. We send with a zero priority. It - timed out, I doubt time is that critical now, and high - priority messages are really only for messages to the local - MC, which don't get resent. */ - intf->handlers->sender(intf->send_info, smi_msg, 0); - #ifdef DEBUG_MSGING { int m; @@ -2619,6 +2611,7 @@ send_from_recv_msg(ipmi_smi_t intf, struct ipmi_recv_msg *recv_msg, printk("\n"); } #endif + return smi_msg; } static void @@ -2683,14 +2676,13 @@ ipmi_timeout_handler(long timeout_period) intf->timed_out_ipmb_commands++; spin_unlock(&intf->counter_lock); } else { + struct ipmi_smi_msg *smi_msg; /* More retries, send again. */ /* Start with the max timer, set to normal timer after the message is sent. */ ent->timeout = MAX_MSG_TIMEOUT; ent->retries_left--; - send_from_recv_msg(intf, ent->recv_msg, NULL, - j, ent->seqid); spin_lock(&intf->counter_lock); if (ent->recv_msg->addr.addr_type == IPMI_LAN_ADDR_TYPE) @@ -2698,6 +2690,20 @@ ipmi_timeout_handler(long timeout_period) else intf->retransmitted_ipmb_commands++; spin_unlock(&intf->counter_lock); + smi_msg = smi_from_recv_msg(intf, + ent->recv_msg, j, ent->seqid); + if(!smi_msg) + continue; + + spin_unlock_irqrestore(&(intf->seq_lock),flags); + /* Send the new message. We send with a zero + * priority. It timed out, I doubt time is + * that critical now, and high priority + * messages are really only for messages to the + * local MC, which don't get resent. */ + intf->handlers->sender(intf->send_info, + smi_msg, 0); + spin_lock_irqsave(&(intf->seq_lock), flags); } } spin_unlock_irqrestore(&(intf->seq_lock), flags); diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 29de259a981..298574e1606 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -100,6 +100,11 @@ enum si_intf_state { /* FIXME - add watchdog stuff. */ }; +/* Some BT-specific defines we need here. */ +#define IPMI_BT_INTMASK_REG 2 +#define IPMI_BT_INTMASK_CLEAR_IRQ_BIT 2 +#define IPMI_BT_INTMASK_ENABLE_IRQ_BIT 1 + enum si_type { SI_KCS, SI_SMIC, SI_BT }; @@ -875,6 +880,17 @@ static irqreturn_t si_irq_handler(int irq, void *data, struct pt_regs *regs) return IRQ_HANDLED; } +static irqreturn_t si_bt_irq_handler(int irq, void *data, struct pt_regs *regs) +{ + struct smi_info *smi_info = data; + /* We need to clear the IRQ flag for the BT interface. */ + smi_info->io.outputb(&smi_info->io, IPMI_BT_INTMASK_REG, + IPMI_BT_INTMASK_CLEAR_IRQ_BIT + | IPMI_BT_INTMASK_ENABLE_IRQ_BIT); + return si_irq_handler(irq, data, regs); +} + + static struct ipmi_smi_handlers handlers = { .owner = THIS_MODULE, @@ -1001,11 +1017,22 @@ static int std_irq_setup(struct smi_info *info) if (!info->irq) return 0; - rv = request_irq(info->irq, - si_irq_handler, - SA_INTERRUPT, - DEVICE_NAME, - info); + if (info->si_type == SI_BT) { + rv = request_irq(info->irq, + si_bt_irq_handler, + SA_INTERRUPT, + DEVICE_NAME, + info); + if (!rv) + /* Enable the interrupt in the BT interface. */ + info->io.outputb(&info->io, IPMI_BT_INTMASK_REG, + IPMI_BT_INTMASK_ENABLE_IRQ_BIT); + } else + rv = request_irq(info->irq, + si_irq_handler, + SA_INTERRUPT, + DEVICE_NAME, + info); if (rv) { printk(KERN_WARNING "ipmi_si: %s unable to claim interrupt %d," @@ -1024,6 +1051,9 @@ static void std_irq_cleanup(struct smi_info *info) if (!info->irq) return; + if (info->si_type == SI_BT) + /* Disable the interrupt in the BT interface. */ + info->io.outputb(&info->io, IPMI_BT_INTMASK_REG, 0); free_irq(info->irq, info); } @@ -1526,8 +1556,17 @@ static int try_init_acpi(int intf_num, struct smi_info **new_info) info->irq_setup = NULL; } - regspacings[intf_num] = spmi->addr.register_bit_width / 8; - info->io.regspacing = spmi->addr.register_bit_width / 8; + if (spmi->addr.register_bit_width) { + /* A (hopefully) properly formed register bit width. */ + regspacings[intf_num] = spmi->addr.register_bit_width / 8; + info->io.regspacing = spmi->addr.register_bit_width / 8; + } else { + /* Some broken systems get this wrong and set the value + * to zero. Assume it is the default spacing. If that + * is wrong, too bad, the vendor should fix the tables. */ + regspacings[intf_num] = DEFAULT_REGSPACING; + info->io.regspacing = DEFAULT_REGSPACING; + } regsizes[intf_num] = regspacings[intf_num]; info->io.regsize = regsizes[intf_num]; regshifts[intf_num] = spmi->addr.register_bit_offset; @@ -1578,15 +1617,15 @@ typedef struct dmi_header u16 handle; } dmi_header_t; -static int decode_dmi(dmi_header_t *dm, int intf_num) +static int decode_dmi(dmi_header_t __iomem *dm, int intf_num) { - u8 *data = (u8 *)dm; + u8 __iomem *data = (u8 __iomem *)dm; unsigned long base_addr; u8 reg_spacing; - u8 len = dm->length; + u8 len = readb(&dm->length); dmi_ipmi_data_t *ipmi_data = dmi_data+intf_num; - ipmi_data->type = data[4]; + ipmi_data->type = readb(&data[4]); memcpy(&base_addr, data+8, sizeof(unsigned long)); if (len >= 0x11) { @@ -1601,12 +1640,12 @@ static int decode_dmi(dmi_header_t *dm, int intf_num) } /* If bit 4 of byte 0x10 is set, then the lsb for the address is odd. */ - ipmi_data->base_addr = base_addr | ((data[0x10] & 0x10) >> 4); + ipmi_data->base_addr = base_addr | ((readb(&data[0x10]) & 0x10) >> 4); - ipmi_data->irq = data[0x11]; + ipmi_data->irq = readb(&data[0x11]); /* The top two bits of byte 0x10 hold the register spacing. */ - reg_spacing = (data[0x10] & 0xC0) >> 6; + reg_spacing = (readb(&data[0x10]) & 0xC0) >> 6; switch(reg_spacing){ case 0x00: /* Byte boundaries */ ipmi_data->offset = 1; @@ -1623,12 +1662,18 @@ static int decode_dmi(dmi_header_t *dm, int intf_num) } } else { /* Old DMI spec. */ - ipmi_data->base_addr = base_addr; + /* Note that technically, the lower bit of the base + * address should be 1 if the address is I/O and 0 if + * the address is in memory. So many systems get that + * wrong (and all that I have seen are I/O) so we just + * ignore that bit and assume I/O. Systems that use + * memory should use the newer spec, anyway. */ + ipmi_data->base_addr = base_addr & 0xfffe; ipmi_data->addr_space = IPMI_IO_ADDR_SPACE; ipmi_data->offset = 1; } - ipmi_data->slave_addr = data[6]; + ipmi_data->slave_addr = readb(&data[6]); if (is_new_interface(-1, ipmi_data->addr_space,ipmi_data->base_addr)) { dmi_data_entries++; @@ -1642,9 +1687,9 @@ static int decode_dmi(dmi_header_t *dm, int intf_num) static int dmi_table(u32 base, int len, int num) { - u8 *buf; - struct dmi_header *dm; - u8 *data; + u8 __iomem *buf; + struct dmi_header __iomem *dm; + u8 __iomem *data; int i=1; int status=-1; int intf_num = 0; @@ -1657,12 +1702,12 @@ static int dmi_table(u32 base, int len, int num) while(i<num && (data - buf) < len) { - dm=(dmi_header_t *)data; + dm=(dmi_header_t __iomem *)data; - if((data-buf+dm->length) >= len) + if((data-buf+readb(&dm->length)) >= len) break; - if (dm->type == 38) { + if (readb(&dm->type) == 38) { if (decode_dmi(dm, intf_num) == 0) { intf_num++; if (intf_num >= SI_MAX_DRIVERS) @@ -1670,8 +1715,8 @@ static int dmi_table(u32 base, int len, int num) } } - data+=dm->length; - while((data-buf) < len && (*data || data[1])) + data+=readb(&dm->length); + while((data-buf) < len && (readb(data)||readb(data+1))) data++; data+=2; i++; @@ -2199,7 +2244,7 @@ static int init_one_smi(int intf_num, struct smi_info **smi) /* Wait until we know that we are out of any interrupt handlers might have been running before we freed the interrupt. */ - synchronize_kernel(); + synchronize_sched(); if (new_smi->si_sm) { if (new_smi->handlers) @@ -2312,7 +2357,7 @@ static void __exit cleanup_one_si(struct smi_info *to_clean) /* Wait until we know that we are out of any interrupt handlers might have been running before we freed the interrupt. */ - synchronize_kernel(); + synchronize_sched(); /* Wait for the timer to stop. This avoids problems with race conditions removing the timer here. */ diff --git a/drivers/char/ipmi/ipmi_si_sm.h b/drivers/char/ipmi/ipmi_si_sm.h index a0212b00401..62791dd4298 100644 --- a/drivers/char/ipmi/ipmi_si_sm.h +++ b/drivers/char/ipmi/ipmi_si_sm.h @@ -51,7 +51,7 @@ struct si_sm_io /* Generic info used by the actual handling routines, the state machine shouldn't touch these. */ void *info; - void *addr; + void __iomem *addr; int regspacing; int regsize; int regshift; diff --git a/drivers/char/ipmi/ipmi_watchdog.c b/drivers/char/ipmi/ipmi_watchdog.c index fd7093879c6..fcd1c02a32c 100644 --- a/drivers/char/ipmi/ipmi_watchdog.c +++ b/drivers/char/ipmi/ipmi_watchdog.c @@ -709,11 +709,11 @@ static int ipmi_close(struct inode *ino, struct file *filep) if (expect_close == 42) { ipmi_watchdog_state = WDOG_TIMEOUT_NONE; ipmi_set_timeout(IPMI_SET_TIMEOUT_NO_HB); - clear_bit(0, &ipmi_wdog_open); } else { printk(KERN_CRIT PFX "Unexpected close, not stopping watchdog!\n"); ipmi_heartbeat(); } + clear_bit(0, &ipmi_wdog_open); } ipmi_fasync (-1, filep, 0); diff --git a/drivers/char/keyboard.c b/drivers/char/keyboard.c index 3ce51c6a1b1..7b19e02f112 100644 --- a/drivers/char/keyboard.c +++ b/drivers/char/keyboard.c @@ -1026,7 +1026,8 @@ static void kbd_rawcode(unsigned char data) put_queue(vc, data); } -void kbd_keycode(unsigned int keycode, int down, int hw_raw, struct pt_regs *regs) +static void kbd_keycode(unsigned int keycode, int down, + int hw_raw, struct pt_regs *regs) { struct vc_data *vc = vc_cons[fg_console].d; unsigned short keysym, *key_map; diff --git a/drivers/char/mbcs.c b/drivers/char/mbcs.c new file mode 100644 index 00000000000..ac9cfa9701e --- /dev/null +++ b/drivers/char/mbcs.c @@ -0,0 +1,849 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2005 Silicon Graphics, Inc. All rights reserved. + */ + +/* + * MOATB Core Services driver. + */ + +#include <linux/config.h> +#include <linux/interrupt.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/types.h> +#include <linux/ioport.h> +#include <linux/notifier.h> +#include <linux/reboot.h> +#include <linux/init.h> +#include <linux/fs.h> +#include <linux/delay.h> +#include <linux/device.h> +#include <linux/mm.h> +#include <linux/uio.h> +#include <asm/io.h> +#include <asm/uaccess.h> +#include <asm/system.h> +#include <asm/pgtable.h> +#include <asm/sn/addrs.h> +#include <asm/sn/intr.h> +#include <asm/sn/tiocx.h> +#include "mbcs.h" + +#define MBCS_DEBUG 0 +#if MBCS_DEBUG +#define DBG(fmt...) printk(KERN_ALERT fmt) +#else +#define DBG(fmt...) +#endif +int mbcs_major; + +LIST_HEAD(soft_list); + +/* + * file operations + */ +struct file_operations mbcs_ops = { + .open = mbcs_open, + .llseek = mbcs_sram_llseek, + .read = mbcs_sram_read, + .write = mbcs_sram_write, + .mmap = mbcs_gscr_mmap, +}; + +struct mbcs_callback_arg { + int minor; + struct cx_dev *cx_dev; +}; + +static inline void mbcs_getdma_init(struct getdma *gdma) +{ + memset(gdma, 0, sizeof(struct getdma)); + gdma->DoneIntEnable = 1; +} + +static inline void mbcs_putdma_init(struct putdma *pdma) +{ + memset(pdma, 0, sizeof(struct putdma)); + pdma->DoneIntEnable = 1; +} + +static inline void mbcs_algo_init(struct algoblock *algo_soft) +{ + memset(algo_soft, 0, sizeof(struct algoblock)); +} + +static inline void mbcs_getdma_set(void *mmr, + uint64_t hostAddr, + uint64_t localAddr, + uint64_t localRamSel, + uint64_t numPkts, + uint64_t amoEnable, + uint64_t intrEnable, + uint64_t peerIO, + uint64_t amoHostDest, + uint64_t amoModType, uint64_t intrHostDest, + uint64_t intrVector) +{ + union dma_control rdma_control; + union dma_amo_dest amo_dest; + union intr_dest intr_dest; + union dma_localaddr local_addr; + union dma_hostaddr host_addr; + + rdma_control.dma_control_reg = 0; + amo_dest.dma_amo_dest_reg = 0; + intr_dest.intr_dest_reg = 0; + local_addr.dma_localaddr_reg = 0; + host_addr.dma_hostaddr_reg = 0; + + host_addr.dma_sys_addr = hostAddr; + MBCS_MMR_SET(mmr, MBCS_RD_DMA_SYS_ADDR, host_addr.dma_hostaddr_reg); + + local_addr.dma_ram_addr = localAddr; + local_addr.dma_ram_sel = localRamSel; + MBCS_MMR_SET(mmr, MBCS_RD_DMA_LOC_ADDR, local_addr.dma_localaddr_reg); + + rdma_control.dma_op_length = numPkts; + rdma_control.done_amo_en = amoEnable; + rdma_control.done_int_en = intrEnable; + rdma_control.pio_mem_n = peerIO; + MBCS_MMR_SET(mmr, MBCS_RD_DMA_CTRL, rdma_control.dma_control_reg); + + amo_dest.dma_amo_sys_addr = amoHostDest; + amo_dest.dma_amo_mod_type = amoModType; + MBCS_MMR_SET(mmr, MBCS_RD_DMA_AMO_DEST, amo_dest.dma_amo_dest_reg); + + intr_dest.address = intrHostDest; + intr_dest.int_vector = intrVector; + MBCS_MMR_SET(mmr, MBCS_RD_DMA_INT_DEST, intr_dest.intr_dest_reg); + +} + +static inline void mbcs_putdma_set(void *mmr, + uint64_t hostAddr, + uint64_t localAddr, + uint64_t localRamSel, + uint64_t numPkts, + uint64_t amoEnable, + uint64_t intrEnable, + uint64_t peerIO, + uint64_t amoHostDest, + uint64_t amoModType, + uint64_t intrHostDest, uint64_t intrVector) +{ + union dma_control wdma_control; + union dma_amo_dest amo_dest; + union intr_dest intr_dest; + union dma_localaddr local_addr; + union dma_hostaddr host_addr; + + wdma_control.dma_control_reg = 0; + amo_dest.dma_amo_dest_reg = 0; + intr_dest.intr_dest_reg = 0; + local_addr.dma_localaddr_reg = 0; + host_addr.dma_hostaddr_reg = 0; + + host_addr.dma_sys_addr = hostAddr; + MBCS_MMR_SET(mmr, MBCS_WR_DMA_SYS_ADDR, host_addr.dma_hostaddr_reg); + + local_addr.dma_ram_addr = localAddr; + local_addr.dma_ram_sel = localRamSel; + MBCS_MMR_SET(mmr, MBCS_WR_DMA_LOC_ADDR, local_addr.dma_localaddr_reg); + + wdma_control.dma_op_length = numPkts; + wdma_control.done_amo_en = amoEnable; + wdma_control.done_int_en = intrEnable; + wdma_control.pio_mem_n = peerIO; + MBCS_MMR_SET(mmr, MBCS_WR_DMA_CTRL, wdma_control.dma_control_reg); + + amo_dest.dma_amo_sys_addr = amoHostDest; + amo_dest.dma_amo_mod_type = amoModType; + MBCS_MMR_SET(mmr, MBCS_WR_DMA_AMO_DEST, amo_dest.dma_amo_dest_reg); + + intr_dest.address = intrHostDest; + intr_dest.int_vector = intrVector; + MBCS_MMR_SET(mmr, MBCS_WR_DMA_INT_DEST, intr_dest.intr_dest_reg); + +} + +static inline void mbcs_algo_set(void *mmr, + uint64_t amoHostDest, + uint64_t amoModType, + uint64_t intrHostDest, + uint64_t intrVector, uint64_t algoStepCount) +{ + union dma_amo_dest amo_dest; + union intr_dest intr_dest; + union algo_step step; + + step.algo_step_reg = 0; + intr_dest.intr_dest_reg = 0; + amo_dest.dma_amo_dest_reg = 0; + + amo_dest.dma_amo_sys_addr = amoHostDest; + amo_dest.dma_amo_mod_type = amoModType; + MBCS_MMR_SET(mmr, MBCS_ALG_AMO_DEST, amo_dest.dma_amo_dest_reg); + + intr_dest.address = intrHostDest; + intr_dest.int_vector = intrVector; + MBCS_MMR_SET(mmr, MBCS_ALG_INT_DEST, intr_dest.intr_dest_reg); + + step.alg_step_cnt = algoStepCount; + MBCS_MMR_SET(mmr, MBCS_ALG_STEP, step.algo_step_reg); +} + +static inline int mbcs_getdma_start(struct mbcs_soft *soft) +{ + void *mmr_base; + struct getdma *gdma; + uint64_t numPkts; + union cm_control cm_control; + + mmr_base = soft->mmr_base; + gdma = &soft->getdma; + + /* check that host address got setup */ + if (!gdma->hostAddr) + return -1; + + numPkts = + (gdma->bytes + (MBCS_CACHELINE_SIZE - 1)) / MBCS_CACHELINE_SIZE; + + /* program engine */ + mbcs_getdma_set(mmr_base, tiocx_dma_addr(gdma->hostAddr), + gdma->localAddr, + (gdma->localAddr < MB2) ? 0 : + (gdma->localAddr < MB4) ? 1 : + (gdma->localAddr < MB6) ? 2 : 3, + numPkts, + gdma->DoneAmoEnable, + gdma->DoneIntEnable, + gdma->peerIO, + gdma->amoHostDest, + gdma->amoModType, + gdma->intrHostDest, gdma->intrVector); + + /* start engine */ + cm_control.cm_control_reg = MBCS_MMR_GET(mmr_base, MBCS_CM_CONTROL); + cm_control.rd_dma_go = 1; + MBCS_MMR_SET(mmr_base, MBCS_CM_CONTROL, cm_control.cm_control_reg); + + return 0; + +} + +static inline int mbcs_putdma_start(struct mbcs_soft *soft) +{ + void *mmr_base; + struct putdma *pdma; + uint64_t numPkts; + union cm_control cm_control; + + mmr_base = soft->mmr_base; + pdma = &soft->putdma; + + /* check that host address got setup */ + if (!pdma->hostAddr) + return -1; + + numPkts = + (pdma->bytes + (MBCS_CACHELINE_SIZE - 1)) / MBCS_CACHELINE_SIZE; + + /* program engine */ + mbcs_putdma_set(mmr_base, tiocx_dma_addr(pdma->hostAddr), + pdma->localAddr, + (pdma->localAddr < MB2) ? 0 : + (pdma->localAddr < MB4) ? 1 : + (pdma->localAddr < MB6) ? 2 : 3, + numPkts, + pdma->DoneAmoEnable, + pdma->DoneIntEnable, + pdma->peerIO, + pdma->amoHostDest, + pdma->amoModType, + pdma->intrHostDest, pdma->intrVector); + + /* start engine */ + cm_control.cm_control_reg = MBCS_MMR_GET(mmr_base, MBCS_CM_CONTROL); + cm_control.wr_dma_go = 1; + MBCS_MMR_SET(mmr_base, MBCS_CM_CONTROL, cm_control.cm_control_reg); + + return 0; + +} + +static inline int mbcs_algo_start(struct mbcs_soft *soft) +{ + struct algoblock *algo_soft = &soft->algo; + void *mmr_base = soft->mmr_base; + union cm_control cm_control; + + if (down_interruptible(&soft->algolock)) + return -ERESTARTSYS; + + atomic_set(&soft->algo_done, 0); + + mbcs_algo_set(mmr_base, + algo_soft->amoHostDest, + algo_soft->amoModType, + algo_soft->intrHostDest, + algo_soft->intrVector, algo_soft->algoStepCount); + + /* start algorithm */ + cm_control.cm_control_reg = MBCS_MMR_GET(mmr_base, MBCS_CM_CONTROL); + cm_control.alg_done_int_en = 1; + cm_control.alg_go = 1; + MBCS_MMR_SET(mmr_base, MBCS_CM_CONTROL, cm_control.cm_control_reg); + + up(&soft->algolock); + + return 0; +} + +static inline ssize_t +do_mbcs_sram_dmawrite(struct mbcs_soft *soft, uint64_t hostAddr, + size_t len, loff_t * off) +{ + int rv = 0; + + if (down_interruptible(&soft->dmawritelock)) + return -ERESTARTSYS; + + atomic_set(&soft->dmawrite_done, 0); + + soft->putdma.hostAddr = hostAddr; + soft->putdma.localAddr = *off; + soft->putdma.bytes = len; + + if (mbcs_putdma_start(soft) < 0) { + DBG(KERN_ALERT "do_mbcs_sram_dmawrite: " + "mbcs_putdma_start failed\n"); + rv = -EAGAIN; + goto dmawrite_exit; + } + + if (wait_event_interruptible(soft->dmawrite_queue, + atomic_read(&soft->dmawrite_done))) { + rv = -ERESTARTSYS; + goto dmawrite_exit; + } + + rv = len; + *off += len; + +dmawrite_exit: + up(&soft->dmawritelock); + + return rv; +} + +static inline ssize_t +do_mbcs_sram_dmaread(struct mbcs_soft *soft, uint64_t hostAddr, + size_t len, loff_t * off) +{ + int rv = 0; + + if (down_interruptible(&soft->dmareadlock)) + return -ERESTARTSYS; + + atomic_set(&soft->dmawrite_done, 0); + + soft->getdma.hostAddr = hostAddr; + soft->getdma.localAddr = *off; + soft->getdma.bytes = len; + + if (mbcs_getdma_start(soft) < 0) { + DBG(KERN_ALERT "mbcs_strategy: mbcs_getdma_start failed\n"); + rv = -EAGAIN; + goto dmaread_exit; + } + + if (wait_event_interruptible(soft->dmaread_queue, + atomic_read(&soft->dmaread_done))) { + rv = -ERESTARTSYS; + goto dmaread_exit; + } + + rv = len; + *off += len; + +dmaread_exit: + up(&soft->dmareadlock); + + return rv; +} + +int mbcs_open(struct inode *ip, struct file *fp) +{ + struct mbcs_soft *soft; + int minor; + + minor = iminor(ip); + + list_for_each_entry(soft, &soft_list, list) { + if (soft->nasid == minor) { + fp->private_data = soft->cxdev; + return 0; + } + } + + return -ENODEV; +} + +ssize_t mbcs_sram_read(struct file * fp, char __user *buf, size_t len, loff_t * off) +{ + struct cx_dev *cx_dev = fp->private_data; + struct mbcs_soft *soft = cx_dev->soft; + uint64_t hostAddr; + int rv = 0; + + hostAddr = __get_dma_pages(GFP_KERNEL, get_order(len)); + if (hostAddr == 0) + return -ENOMEM; + + rv = do_mbcs_sram_dmawrite(soft, hostAddr, len, off); + if (rv < 0) + goto exit; + + if (copy_to_user(buf, (void *)hostAddr, len)) + rv = -EFAULT; + + exit: + free_pages(hostAddr, get_order(len)); + + return rv; +} + +ssize_t +mbcs_sram_write(struct file * fp, const char __user *buf, size_t len, loff_t * off) +{ + struct cx_dev *cx_dev = fp->private_data; + struct mbcs_soft *soft = cx_dev->soft; + uint64_t hostAddr; + int rv = 0; + + hostAddr = __get_dma_pages(GFP_KERNEL, get_order(len)); + if (hostAddr == 0) + return -ENOMEM; + + if (copy_from_user((void *)hostAddr, buf, len)) { + rv = -EFAULT; + goto exit; + } + + rv = do_mbcs_sram_dmaread(soft, hostAddr, len, off); + + exit: + free_pages(hostAddr, get_order(len)); + + return rv; +} + +loff_t mbcs_sram_llseek(struct file * filp, loff_t off, int whence) +{ + loff_t newpos; + + switch (whence) { + case 0: /* SEEK_SET */ + newpos = off; + break; + + case 1: /* SEEK_CUR */ + newpos = filp->f_pos + off; + break; + + case 2: /* SEEK_END */ + newpos = MBCS_SRAM_SIZE + off; + break; + + default: /* can't happen */ + return -EINVAL; + } + + if (newpos < 0) + return -EINVAL; + + filp->f_pos = newpos; + + return newpos; +} + +static uint64_t mbcs_pioaddr(struct mbcs_soft *soft, uint64_t offset) +{ + uint64_t mmr_base; + + mmr_base = (uint64_t) (soft->mmr_base + offset); + + return mmr_base; +} + +static void mbcs_debug_pioaddr_set(struct mbcs_soft *soft) +{ + soft->debug_addr = mbcs_pioaddr(soft, MBCS_DEBUG_START); +} + +static void mbcs_gscr_pioaddr_set(struct mbcs_soft *soft) +{ + soft->gscr_addr = mbcs_pioaddr(soft, MBCS_GSCR_START); +} + +int mbcs_gscr_mmap(struct file *fp, struct vm_area_struct *vma) +{ + struct cx_dev *cx_dev = fp->private_data; + struct mbcs_soft *soft = cx_dev->soft; + + if (vma->vm_pgoff != 0) + return -EINVAL; + + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + + /* Remap-pfn-range will mark the range VM_IO and VM_RESERVED */ + if (remap_pfn_range(vma, + vma->vm_start, + __pa(soft->gscr_addr) >> PAGE_SHIFT, + PAGE_SIZE, + vma->vm_page_prot)) + return -EAGAIN; + + return 0; +} + +/** + * mbcs_completion_intr_handler - Primary completion handler. + * @irq: irq + * @arg: soft struct for device + * @ep: regs + * + */ +static irqreturn_t +mbcs_completion_intr_handler(int irq, void *arg, struct pt_regs *ep) +{ + struct mbcs_soft *soft = (struct mbcs_soft *)arg; + void *mmr_base; + union cm_status cm_status; + union cm_control cm_control; + + mmr_base = soft->mmr_base; + cm_status.cm_status_reg = MBCS_MMR_GET(mmr_base, MBCS_CM_STATUS); + + if (cm_status.rd_dma_done) { + /* stop dma-read engine, clear status */ + cm_control.cm_control_reg = + MBCS_MMR_GET(mmr_base, MBCS_CM_CONTROL); + cm_control.rd_dma_clr = 1; + MBCS_MMR_SET(mmr_base, MBCS_CM_CONTROL, + cm_control.cm_control_reg); + atomic_set(&soft->dmaread_done, 1); + wake_up(&soft->dmaread_queue); + } + if (cm_status.wr_dma_done) { + /* stop dma-write engine, clear status */ + cm_control.cm_control_reg = + MBCS_MMR_GET(mmr_base, MBCS_CM_CONTROL); + cm_control.wr_dma_clr = 1; + MBCS_MMR_SET(mmr_base, MBCS_CM_CONTROL, + cm_control.cm_control_reg); + atomic_set(&soft->dmawrite_done, 1); + wake_up(&soft->dmawrite_queue); + } + if (cm_status.alg_done) { + /* clear status */ + cm_control.cm_control_reg = + MBCS_MMR_GET(mmr_base, MBCS_CM_CONTROL); + cm_control.alg_done_clr = 1; + MBCS_MMR_SET(mmr_base, MBCS_CM_CONTROL, + cm_control.cm_control_reg); + atomic_set(&soft->algo_done, 1); + wake_up(&soft->algo_queue); + } + + return IRQ_HANDLED; +} + +/** + * mbcs_intr_alloc - Allocate interrupts. + * @dev: device pointer + * + */ +static int mbcs_intr_alloc(struct cx_dev *dev) +{ + struct sn_irq_info *sn_irq; + struct mbcs_soft *soft; + struct getdma *getdma; + struct putdma *putdma; + struct algoblock *algo; + + soft = dev->soft; + getdma = &soft->getdma; + putdma = &soft->putdma; + algo = &soft->algo; + + soft->get_sn_irq = NULL; + soft->put_sn_irq = NULL; + soft->algo_sn_irq = NULL; + + sn_irq = tiocx_irq_alloc(dev->cx_id.nasid, TIOCX_CORELET, -1, -1, -1); + if (sn_irq == NULL) + return -EAGAIN; + soft->get_sn_irq = sn_irq; + getdma->intrHostDest = sn_irq->irq_xtalkaddr; + getdma->intrVector = sn_irq->irq_irq; + if (request_irq(sn_irq->irq_irq, + (void *)mbcs_completion_intr_handler, SA_SHIRQ, + "MBCS get intr", (void *)soft)) { + tiocx_irq_free(soft->get_sn_irq); + return -EAGAIN; + } + + sn_irq = tiocx_irq_alloc(dev->cx_id.nasid, TIOCX_CORELET, -1, -1, -1); + if (sn_irq == NULL) { + free_irq(soft->get_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->get_sn_irq); + return -EAGAIN; + } + soft->put_sn_irq = sn_irq; + putdma->intrHostDest = sn_irq->irq_xtalkaddr; + putdma->intrVector = sn_irq->irq_irq; + if (request_irq(sn_irq->irq_irq, + (void *)mbcs_completion_intr_handler, SA_SHIRQ, + "MBCS put intr", (void *)soft)) { + tiocx_irq_free(soft->put_sn_irq); + free_irq(soft->get_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->get_sn_irq); + return -EAGAIN; + } + + sn_irq = tiocx_irq_alloc(dev->cx_id.nasid, TIOCX_CORELET, -1, -1, -1); + if (sn_irq == NULL) { + free_irq(soft->put_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->put_sn_irq); + free_irq(soft->get_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->get_sn_irq); + return -EAGAIN; + } + soft->algo_sn_irq = sn_irq; + algo->intrHostDest = sn_irq->irq_xtalkaddr; + algo->intrVector = sn_irq->irq_irq; + if (request_irq(sn_irq->irq_irq, + (void *)mbcs_completion_intr_handler, SA_SHIRQ, + "MBCS algo intr", (void *)soft)) { + tiocx_irq_free(soft->algo_sn_irq); + free_irq(soft->put_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->put_sn_irq); + free_irq(soft->get_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->get_sn_irq); + return -EAGAIN; + } + + return 0; +} + +/** + * mbcs_intr_dealloc - Remove interrupts. + * @dev: device pointer + * + */ +static void mbcs_intr_dealloc(struct cx_dev *dev) +{ + struct mbcs_soft *soft; + + soft = dev->soft; + + free_irq(soft->get_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->get_sn_irq); + free_irq(soft->put_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->put_sn_irq); + free_irq(soft->algo_sn_irq->irq_irq, soft); + tiocx_irq_free(soft->algo_sn_irq); +} + +static inline int mbcs_hw_init(struct mbcs_soft *soft) +{ + void *mmr_base = soft->mmr_base; + union cm_control cm_control; + union cm_req_timeout cm_req_timeout; + uint64_t err_stat; + + cm_req_timeout.cm_req_timeout_reg = + MBCS_MMR_GET(mmr_base, MBCS_CM_REQ_TOUT); + + cm_req_timeout.time_out = MBCS_CM_CONTROL_REQ_TOUT_MASK; + MBCS_MMR_SET(mmr_base, MBCS_CM_REQ_TOUT, + cm_req_timeout.cm_req_timeout_reg); + + mbcs_gscr_pioaddr_set(soft); + mbcs_debug_pioaddr_set(soft); + + /* clear errors */ + err_stat = MBCS_MMR_GET(mmr_base, MBCS_CM_ERR_STAT); + MBCS_MMR_SET(mmr_base, MBCS_CM_CLR_ERR_STAT, err_stat); + MBCS_MMR_ZERO(mmr_base, MBCS_CM_ERROR_DETAIL1); + + /* enable interrupts */ + /* turn off 2^23 (INT_EN_PIO_REQ_ADDR_INV) */ + MBCS_MMR_SET(mmr_base, MBCS_CM_ERR_INT_EN, 0x3ffffff7e00ffUL); + + /* arm status regs and clear engines */ + cm_control.cm_control_reg = MBCS_MMR_GET(mmr_base, MBCS_CM_CONTROL); + cm_control.rearm_stat_regs = 1; + cm_control.alg_clr = 1; + cm_control.wr_dma_clr = 1; + cm_control.rd_dma_clr = 1; + + MBCS_MMR_SET(mmr_base, MBCS_CM_CONTROL, cm_control.cm_control_reg); + + return 0; +} + +static ssize_t show_algo(struct device *dev, char *buf) +{ + struct cx_dev *cx_dev = to_cx_dev(dev); + struct mbcs_soft *soft = cx_dev->soft; + uint64_t debug0; + + /* + * By convention, the first debug register contains the + * algorithm number and revision. + */ + debug0 = *(uint64_t *) soft->debug_addr; + + return sprintf(buf, "0x%lx 0x%lx\n", + (debug0 >> 32), (debug0 & 0xffffffff)); +} + +static ssize_t store_algo(struct device *dev, const char *buf, size_t count) +{ + int n; + struct cx_dev *cx_dev = to_cx_dev(dev); + struct mbcs_soft *soft = cx_dev->soft; + + if (count <= 0) + return 0; + + n = simple_strtoul(buf, NULL, 0); + + if (n == 1) { + mbcs_algo_start(soft); + if (wait_event_interruptible(soft->algo_queue, + atomic_read(&soft->algo_done))) + return -ERESTARTSYS; + } + + return count; +} + +DEVICE_ATTR(algo, 0644, show_algo, store_algo); + +/** + * mbcs_probe - Initialize for device + * @dev: device pointer + * @device_id: id table pointer + * + */ +static int mbcs_probe(struct cx_dev *dev, const struct cx_device_id *id) +{ + struct mbcs_soft *soft; + + dev->soft = NULL; + + soft = kcalloc(1, sizeof(struct mbcs_soft), GFP_KERNEL); + if (soft == NULL) + return -ENOMEM; + + soft->nasid = dev->cx_id.nasid; + list_add(&soft->list, &soft_list); + soft->mmr_base = (void *)tiocx_swin_base(dev->cx_id.nasid); + dev->soft = soft; + soft->cxdev = dev; + + init_waitqueue_head(&soft->dmawrite_queue); + init_waitqueue_head(&soft->dmaread_queue); + init_waitqueue_head(&soft->algo_queue); + + init_MUTEX(&soft->dmawritelock); + init_MUTEX(&soft->dmareadlock); + init_MUTEX(&soft->algolock); + + mbcs_getdma_init(&soft->getdma); + mbcs_putdma_init(&soft->putdma); + mbcs_algo_init(&soft->algo); + + mbcs_hw_init(soft); + + /* Allocate interrupts */ + mbcs_intr_alloc(dev); + + device_create_file(&dev->dev, &dev_attr_algo); + + return 0; +} + +static int mbcs_remove(struct cx_dev *dev) +{ + if (dev->soft) { + mbcs_intr_dealloc(dev); + kfree(dev->soft); + } + + device_remove_file(&dev->dev, &dev_attr_algo); + + return 0; +} + +const struct cx_device_id __devinitdata mbcs_id_table[] = { + { + .part_num = MBCS_PART_NUM, + .mfg_num = MBCS_MFG_NUM, + }, + { + .part_num = MBCS_PART_NUM_ALG0, + .mfg_num = MBCS_MFG_NUM, + }, + {0, 0} +}; + +MODULE_DEVICE_TABLE(cx, mbcs_id_table); + +struct cx_drv mbcs_driver = { + .name = DEVICE_NAME, + .id_table = mbcs_id_table, + .probe = mbcs_probe, + .remove = mbcs_remove, +}; + +static void __exit mbcs_exit(void) +{ + int rv; + + rv = unregister_chrdev(mbcs_major, DEVICE_NAME); + if (rv < 0) + DBG(KERN_ALERT "Error in unregister_chrdev: %d\n", rv); + + cx_driver_unregister(&mbcs_driver); +} + +static int __init mbcs_init(void) +{ + int rv; + + // Put driver into chrdevs[]. Get major number. + rv = register_chrdev(mbcs_major, DEVICE_NAME, &mbcs_ops); + if (rv < 0) { + DBG(KERN_ALERT "mbcs_init: can't get major number. %d\n", rv); + return rv; + } + mbcs_major = rv; + + return cx_driver_register(&mbcs_driver); +} + +module_init(mbcs_init); +module_exit(mbcs_exit); + +MODULE_AUTHOR("Bruce Losure <blosure@sgi.com>"); +MODULE_DESCRIPTION("Driver for MOATB Core Services"); +MODULE_LICENSE("GPL"); diff --git a/drivers/char/mbcs.h b/drivers/char/mbcs.h new file mode 100644 index 00000000000..e7fd47e4325 --- /dev/null +++ b/drivers/char/mbcs.h @@ -0,0 +1,553 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2005 Silicon Graphics, Inc. All rights reserved. + */ + +#ifndef __MBCS_H__ +#define __MBCS_H__ + +/* + * General macros + */ +#define MB (1024*1024) +#define MB2 (2*MB) +#define MB4 (4*MB) +#define MB6 (6*MB) + +/* + * Offsets and masks + */ +#define MBCS_CM_ID 0x0000 /* Identification */ +#define MBCS_CM_STATUS 0x0008 /* Status */ +#define MBCS_CM_ERROR_DETAIL1 0x0010 /* Error Detail1 */ +#define MBCS_CM_ERROR_DETAIL2 0x0018 /* Error Detail2 */ +#define MBCS_CM_CONTROL 0x0020 /* Control */ +#define MBCS_CM_REQ_TOUT 0x0028 /* Request Time-out */ +#define MBCS_CM_ERR_INT_DEST 0x0038 /* Error Interrupt Destination */ +#define MBCS_CM_TARG_FL 0x0050 /* Target Flush */ +#define MBCS_CM_ERR_STAT 0x0060 /* Error Status */ +#define MBCS_CM_CLR_ERR_STAT 0x0068 /* Clear Error Status */ +#define MBCS_CM_ERR_INT_EN 0x0070 /* Error Interrupt Enable */ +#define MBCS_RD_DMA_SYS_ADDR 0x0100 /* Read DMA System Address */ +#define MBCS_RD_DMA_LOC_ADDR 0x0108 /* Read DMA Local Address */ +#define MBCS_RD_DMA_CTRL 0x0110 /* Read DMA Control */ +#define MBCS_RD_DMA_AMO_DEST 0x0118 /* Read DMA AMO Destination */ +#define MBCS_RD_DMA_INT_DEST 0x0120 /* Read DMA Interrupt Destination */ +#define MBCS_RD_DMA_AUX_STAT 0x0130 /* Read DMA Auxillary Status */ +#define MBCS_WR_DMA_SYS_ADDR 0x0200 /* Write DMA System Address */ +#define MBCS_WR_DMA_LOC_ADDR 0x0208 /* Write DMA Local Address */ +#define MBCS_WR_DMA_CTRL 0x0210 /* Write DMA Control */ +#define MBCS_WR_DMA_AMO_DEST 0x0218 /* Write DMA AMO Destination */ +#define MBCS_WR_DMA_INT_DEST 0x0220 /* Write DMA Interrupt Destination */ +#define MBCS_WR_DMA_AUX_STAT 0x0230 /* Write DMA Auxillary Status */ +#define MBCS_ALG_AMO_DEST 0x0300 /* Algorithm AMO Destination */ +#define MBCS_ALG_INT_DEST 0x0308 /* Algorithm Interrupt Destination */ +#define MBCS_ALG_OFFSETS 0x0310 +#define MBCS_ALG_STEP 0x0318 /* Algorithm Step */ + +#define MBCS_GSCR_START 0x0000000 +#define MBCS_DEBUG_START 0x0100000 +#define MBCS_RAM0_START 0x0200000 +#define MBCS_RAM1_START 0x0400000 +#define MBCS_RAM2_START 0x0600000 + +#define MBCS_CM_CONTROL_REQ_TOUT_MASK 0x0000000000ffffffUL +//#define PIO_BASE_ADDR_BASE_OFFSET_MASK 0x00fffffffff00000UL + +#define MBCS_SRAM_SIZE (1024*1024) +#define MBCS_CACHELINE_SIZE 128 + +/* + * MMR get's and put's + */ +#define MBCS_MMR_ADDR(mmr_base, offset)((uint64_t *)(mmr_base + offset)) +#define MBCS_MMR_SET(mmr_base, offset, value) { \ + uint64_t *mbcs_mmr_set_u64p, readback; \ + mbcs_mmr_set_u64p = (uint64_t *)(mmr_base + offset); \ + *mbcs_mmr_set_u64p = value; \ + readback = *mbcs_mmr_set_u64p; \ +} +#define MBCS_MMR_GET(mmr_base, offset) *(uint64_t *)(mmr_base + offset) +#define MBCS_MMR_ZERO(mmr_base, offset) MBCS_MMR_SET(mmr_base, offset, 0) + +/* + * MBCS mmr structures + */ +union cm_id { + uint64_t cm_id_reg; + struct { + uint64_t always_one:1, // 0 + mfg_id:11, // 11:1 + part_num:16, // 27:12 + bitstream_rev:8, // 35:28 + :28; // 63:36 + }; +}; + +union cm_status { + uint64_t cm_status_reg; + struct { + uint64_t pending_reads:8, // 7:0 + pending_writes:8, // 15:8 + ice_rsp_credits:8, // 23:16 + ice_req_credits:8, // 31:24 + cm_req_credits:8, // 39:32 + :1, // 40 + rd_dma_in_progress:1, // 41 + rd_dma_done:1, // 42 + :1, // 43 + wr_dma_in_progress:1, // 44 + wr_dma_done:1, // 45 + alg_waiting:1, // 46 + alg_pipe_running:1, // 47 + alg_done:1, // 48 + :3, // 51:49 + pending_int_reqs:8, // 59:52 + :3, // 62:60 + alg_half_speed_sel:1; // 63 + }; +}; + +union cm_error_detail1 { + uint64_t cm_error_detail1_reg; + struct { + uint64_t packet_type:4, // 3:0 + source_id:2, // 5:4 + data_size:2, // 7:6 + tnum:8, // 15:8 + byte_enable:8, // 23:16 + gfx_cred:8, // 31:24 + read_type:2, // 33:32 + pio_or_memory:1, // 34 + head_cw_error:1, // 35 + :12, // 47:36 + head_error_bit:1, // 48 + data_error_bit:1, // 49 + :13, // 62:50 + valid:1; // 63 + }; +}; + +union cm_error_detail2 { + uint64_t cm_error_detail2_reg; + struct { + uint64_t address:56, // 55:0 + :8; // 63:56 + }; +}; + +union cm_control { + uint64_t cm_control_reg; + struct { + uint64_t cm_id:2, // 1:0 + :2, // 3:2 + max_trans:5, // 8:4 + :3, // 11:9 + address_mode:1, // 12 + :7, // 19:13 + credit_limit:8, // 27:20 + :5, // 32:28 + rearm_stat_regs:1, // 33 + prescalar_byp:1, // 34 + force_gap_war:1, // 35 + rd_dma_go:1, // 36 + wr_dma_go:1, // 37 + alg_go:1, // 38 + rd_dma_clr:1, // 39 + wr_dma_clr:1, // 40 + alg_clr:1, // 41 + :2, // 43:42 + alg_wait_step:1, // 44 + alg_done_amo_en:1, // 45 + alg_done_int_en:1, // 46 + :1, // 47 + alg_sram0_locked:1, // 48 + alg_sram1_locked:1, // 49 + alg_sram2_locked:1, // 50 + alg_done_clr:1, // 51 + :12; // 63:52 + }; +}; + +union cm_req_timeout { + uint64_t cm_req_timeout_reg; + struct { + uint64_t time_out:24, // 23:0 + :40; // 63:24 + }; +}; + +union intr_dest { + uint64_t intr_dest_reg; + struct { + uint64_t address:56, // 55:0 + int_vector:8; // 63:56 + }; +}; + +union cm_error_status { + uint64_t cm_error_status_reg; + struct { + uint64_t ecc_sbe:1, // 0 + ecc_mbe:1, // 1 + unsupported_req:1, // 2 + unexpected_rsp:1, // 3 + bad_length:1, // 4 + bad_datavalid:1, // 5 + buffer_overflow:1, // 6 + request_timeout:1, // 7 + :8, // 15:8 + head_inv_data_size:1, // 16 + rsp_pactype_inv:1, // 17 + head_sb_err:1, // 18 + missing_head:1, // 19 + head_inv_rd_type:1, // 20 + head_cmd_err_bit:1, // 21 + req_addr_align_inv:1, // 22 + pio_req_addr_inv:1, // 23 + req_range_dsize_inv:1, // 24 + early_term:1, // 25 + early_tail:1, // 26 + missing_tail:1, // 27 + data_flit_sb_err:1, // 28 + cm2hcm_req_cred_of:1, // 29 + cm2hcm_rsp_cred_of:1, // 30 + rx_bad_didn:1, // 31 + rd_dma_err_rsp:1, // 32 + rd_dma_tnum_tout:1, // 33 + rd_dma_multi_tnum_tou:1, // 34 + wr_dma_err_rsp:1, // 35 + wr_dma_tnum_tout:1, // 36 + wr_dma_multi_tnum_tou:1, // 37 + alg_data_overflow:1, // 38 + alg_data_underflow:1, // 39 + ram0_access_conflict:1, // 40 + ram1_access_conflict:1, // 41 + ram2_access_conflict:1, // 42 + ram0_perr:1, // 43 + ram1_perr:1, // 44 + ram2_perr:1, // 45 + int_gen_rsp_err:1, // 46 + int_gen_tnum_tout:1, // 47 + rd_dma_prog_err:1, // 48 + wr_dma_prog_err:1, // 49 + :14; // 63:50 + }; +}; + +union cm_clr_error_status { + uint64_t cm_clr_error_status_reg; + struct { + uint64_t clr_ecc_sbe:1, // 0 + clr_ecc_mbe:1, // 1 + clr_unsupported_req:1, // 2 + clr_unexpected_rsp:1, // 3 + clr_bad_length:1, // 4 + clr_bad_datavalid:1, // 5 + clr_buffer_overflow:1, // 6 + clr_request_timeout:1, // 7 + :8, // 15:8 + clr_head_inv_data_siz:1, // 16 + clr_rsp_pactype_inv:1, // 17 + clr_head_sb_err:1, // 18 + clr_missing_head:1, // 19 + clr_head_inv_rd_type:1, // 20 + clr_head_cmd_err_bit:1, // 21 + clr_req_addr_align_in:1, // 22 + clr_pio_req_addr_inv:1, // 23 + clr_req_range_dsize_i:1, // 24 + clr_early_term:1, // 25 + clr_early_tail:1, // 26 + clr_missing_tail:1, // 27 + clr_data_flit_sb_err:1, // 28 + clr_cm2hcm_req_cred_o:1, // 29 + clr_cm2hcm_rsp_cred_o:1, // 30 + clr_rx_bad_didn:1, // 31 + clr_rd_dma_err_rsp:1, // 32 + clr_rd_dma_tnum_tout:1, // 33 + clr_rd_dma_multi_tnum:1, // 34 + clr_wr_dma_err_rsp:1, // 35 + clr_wr_dma_tnum_tout:1, // 36 + clr_wr_dma_multi_tnum:1, // 37 + clr_alg_data_overflow:1, // 38 + clr_alg_data_underflo:1, // 39 + clr_ram0_access_confl:1, // 40 + clr_ram1_access_confl:1, // 41 + clr_ram2_access_confl:1, // 42 + clr_ram0_perr:1, // 43 + clr_ram1_perr:1, // 44 + clr_ram2_perr:1, // 45 + clr_int_gen_rsp_err:1, // 46 + clr_int_gen_tnum_tout:1, // 47 + clr_rd_dma_prog_err:1, // 48 + clr_wr_dma_prog_err:1, // 49 + :14; // 63:50 + }; +}; + +union cm_error_intr_enable { + uint64_t cm_error_intr_enable_reg; + struct { + uint64_t int_en_ecc_sbe:1, // 0 + int_en_ecc_mbe:1, // 1 + int_en_unsupported_re:1, // 2 + int_en_unexpected_rsp:1, // 3 + int_en_bad_length:1, // 4 + int_en_bad_datavalid:1, // 5 + int_en_buffer_overflo:1, // 6 + int_en_request_timeou:1, // 7 + :8, // 15:8 + int_en_head_inv_data_:1, // 16 + int_en_rsp_pactype_in:1, // 17 + int_en_head_sb_err:1, // 18 + int_en_missing_head:1, // 19 + int_en_head_inv_rd_ty:1, // 20 + int_en_head_cmd_err_b:1, // 21 + int_en_req_addr_align:1, // 22 + int_en_pio_req_addr_i:1, // 23 + int_en_req_range_dsiz:1, // 24 + int_en_early_term:1, // 25 + int_en_early_tail:1, // 26 + int_en_missing_tail:1, // 27 + int_en_data_flit_sb_e:1, // 28 + int_en_cm2hcm_req_cre:1, // 29 + int_en_cm2hcm_rsp_cre:1, // 30 + int_en_rx_bad_didn:1, // 31 + int_en_rd_dma_err_rsp:1, // 32 + int_en_rd_dma_tnum_to:1, // 33 + int_en_rd_dma_multi_t:1, // 34 + int_en_wr_dma_err_rsp:1, // 35 + int_en_wr_dma_tnum_to:1, // 36 + int_en_wr_dma_multi_t:1, // 37 + int_en_alg_data_overf:1, // 38 + int_en_alg_data_under:1, // 39 + int_en_ram0_access_co:1, // 40 + int_en_ram1_access_co:1, // 41 + int_en_ram2_access_co:1, // 42 + int_en_ram0_perr:1, // 43 + int_en_ram1_perr:1, // 44 + int_en_ram2_perr:1, // 45 + int_en_int_gen_rsp_er:1, // 46 + int_en_int_gen_tnum_t:1, // 47 + int_en_rd_dma_prog_er:1, // 48 + int_en_wr_dma_prog_er:1, // 49 + :14; // 63:50 + }; +}; + +struct cm_mmr { + union cm_id id; + union cm_status status; + union cm_error_detail1 err_detail1; + union cm_error_detail2 err_detail2; + union cm_control control; + union cm_req_timeout req_timeout; + uint64_t reserved1[1]; + union intr_dest int_dest; + uint64_t reserved2[2]; + uint64_t targ_flush; + uint64_t reserved3[1]; + union cm_error_status err_status; + union cm_clr_error_status clr_err_status; + union cm_error_intr_enable int_enable; +}; + +union dma_hostaddr { + uint64_t dma_hostaddr_reg; + struct { + uint64_t dma_sys_addr:56, // 55:0 + :8; // 63:56 + }; +}; + +union dma_localaddr { + uint64_t dma_localaddr_reg; + struct { + uint64_t dma_ram_addr:21, // 20:0 + dma_ram_sel:2, // 22:21 + :41; // 63:23 + }; +}; + +union dma_control { + uint64_t dma_control_reg; + struct { + uint64_t dma_op_length:16, // 15:0 + :18, // 33:16 + done_amo_en:1, // 34 + done_int_en:1, // 35 + :1, // 36 + pio_mem_n:1, // 37 + :26; // 63:38 + }; +}; + +union dma_amo_dest { + uint64_t dma_amo_dest_reg; + struct { + uint64_t dma_amo_sys_addr:56, // 55:0 + dma_amo_mod_type:3, // 58:56 + :5; // 63:59 + }; +}; + +union rdma_aux_status { + uint64_t rdma_aux_status_reg; + struct { + uint64_t op_num_pacs_left:17, // 16:0 + :5, // 21:17 + lrsp_buff_empty:1, // 22 + :17, // 39:23 + pending_reqs_left:6, // 45:40 + :18; // 63:46 + }; +}; + +struct rdma_mmr { + union dma_hostaddr host_addr; + union dma_localaddr local_addr; + union dma_control control; + union dma_amo_dest amo_dest; + union intr_dest intr_dest; + union rdma_aux_status aux_status; +}; + +union wdma_aux_status { + uint64_t wdma_aux_status_reg; + struct { + uint64_t op_num_pacs_left:17, // 16:0 + :4, // 20:17 + lreq_buff_empty:1, // 21 + :18, // 39:22 + pending_reqs_left:6, // 45:40 + :18; // 63:46 + }; +}; + +struct wdma_mmr { + union dma_hostaddr host_addr; + union dma_localaddr local_addr; + union dma_control control; + union dma_amo_dest amo_dest; + union intr_dest intr_dest; + union wdma_aux_status aux_status; +}; + +union algo_step { + uint64_t algo_step_reg; + struct { + uint64_t alg_step_cnt:16, // 15:0 + :48; // 63:16 + }; +}; + +struct algo_mmr { + union dma_amo_dest amo_dest; + union intr_dest intr_dest; + union { + uint64_t algo_offset_reg; + struct { + uint64_t sram0_offset:7, // 6:0 + reserved0:1, // 7 + sram1_offset:7, // 14:8 + reserved1:1, // 15 + sram2_offset:7, // 22:16 + reserved2:14; // 63:23 + }; + } sram_offset; + union algo_step step; +}; + +struct mbcs_mmr { + struct cm_mmr cm; + uint64_t reserved1[17]; + struct rdma_mmr rdDma; + uint64_t reserved2[25]; + struct wdma_mmr wrDma; + uint64_t reserved3[25]; + struct algo_mmr algo; + uint64_t reserved4[156]; +}; + +/* + * defines + */ +#define DEVICE_NAME "mbcs" +#define MBCS_PART_NUM 0xfff0 +#define MBCS_PART_NUM_ALG0 0xf001 +#define MBCS_MFG_NUM 0x1 + +struct algoblock { + uint64_t amoHostDest; + uint64_t amoModType; + uint64_t intrHostDest; + uint64_t intrVector; + uint64_t algoStepCount; +}; + +struct getdma { + uint64_t hostAddr; + uint64_t localAddr; + uint64_t bytes; + uint64_t DoneAmoEnable; + uint64_t DoneIntEnable; + uint64_t peerIO; + uint64_t amoHostDest; + uint64_t amoModType; + uint64_t intrHostDest; + uint64_t intrVector; +}; + +struct putdma { + uint64_t hostAddr; + uint64_t localAddr; + uint64_t bytes; + uint64_t DoneAmoEnable; + uint64_t DoneIntEnable; + uint64_t peerIO; + uint64_t amoHostDest; + uint64_t amoModType; + uint64_t intrHostDest; + uint64_t intrVector; +}; + +struct mbcs_soft { + struct list_head list; + struct cx_dev *cxdev; + int major; + int nasid; + void *mmr_base; + wait_queue_head_t dmawrite_queue; + wait_queue_head_t dmaread_queue; + wait_queue_head_t algo_queue; + struct sn_irq_info *get_sn_irq; + struct sn_irq_info *put_sn_irq; + struct sn_irq_info *algo_sn_irq; + struct getdma getdma; + struct putdma putdma; + struct algoblock algo; + uint64_t gscr_addr; // pio addr + uint64_t ram0_addr; // pio addr + uint64_t ram1_addr; // pio addr + uint64_t ram2_addr; // pio addr + uint64_t debug_addr; // pio addr + atomic_t dmawrite_done; + atomic_t dmaread_done; + atomic_t algo_done; + struct semaphore dmawritelock; + struct semaphore dmareadlock; + struct semaphore algolock; +}; + +extern int mbcs_open(struct inode *ip, struct file *fp); +extern ssize_t mbcs_sram_read(struct file *fp, char __user *buf, size_t len, + loff_t * off); +extern ssize_t mbcs_sram_write(struct file *fp, const char __user *buf, size_t len, + loff_t * off); +extern loff_t mbcs_sram_llseek(struct file *filp, loff_t off, int whence); +extern int mbcs_gscr_mmap(struct file *fp, struct vm_area_struct *vma); + +#endif // __MBCS_H__ diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c index a91ae271cf0..763893e289b 100644 --- a/drivers/char/rio/rio_linux.c +++ b/drivers/char/rio/rio_linux.c @@ -221,7 +221,7 @@ static int rio_probe_addrs[]= {0xc0000, 0xd0000, 0xe0000}; /* Set the mask to all-ones. This alas, only supports 32 interrupts. Some architectures may need more. -- Changed to LONG to support up to 64 bits on 64bit architectures. -- REW 20/06/99 */ -long rio_irqmask = -1; +static long rio_irqmask = -1; MODULE_AUTHOR("Rogier Wolff <R.E.Wolff@bitwizard.nl>, Patrick van de Lageweg <patrick@bitwizard.nl>"); MODULE_DESCRIPTION("RIO driver"); diff --git a/drivers/char/s3c2410-rtc.c b/drivers/char/s3c2410-rtc.c index 8e61be34a1d..ed867db550a 100644 --- a/drivers/char/s3c2410-rtc.c +++ b/drivers/char/s3c2410-rtc.c @@ -116,7 +116,7 @@ static void s3c2410_rtc_setfreq(int freq) /* Time read/write */ -static void s3c2410_rtc_gettime(struct rtc_time *rtc_tm) +static int s3c2410_rtc_gettime(struct rtc_time *rtc_tm) { unsigned int have_retried = 0; @@ -151,6 +151,8 @@ static void s3c2410_rtc_gettime(struct rtc_time *rtc_tm) rtc_tm->tm_year += 100; rtc_tm->tm_mon -= 1; + + return 0; } @@ -171,7 +173,7 @@ static int s3c2410_rtc_settime(struct rtc_time *tm) return 0; } -static void s3c2410_rtc_getalarm(struct rtc_wkalrm *alrm) +static int s3c2410_rtc_getalarm(struct rtc_wkalrm *alrm) { struct rtc_time *alm_tm = &alrm->time; unsigned int alm_en; @@ -231,6 +233,8 @@ static void s3c2410_rtc_getalarm(struct rtc_wkalrm *alrm) } /* todo - set alrm->enabled ? */ + + return 0; } static int s3c2410_rtc_setalarm(struct rtc_wkalrm *alrm) diff --git a/drivers/char/snsc.c b/drivers/char/snsc.c index ffb9143376b..e3c0b52d943 100644 --- a/drivers/char/snsc.c +++ b/drivers/char/snsc.c @@ -374,6 +374,7 @@ scdrv_init(void) void *salbuf; struct class_simple *snsc_class; dev_t first_dev, dev; + nasid_t event_nasid = ia64_sn_get_console_nasid(); if (alloc_chrdev_region(&first_dev, 0, numionodes, SYSCTL_BASENAME) < 0) { @@ -441,6 +442,13 @@ scdrv_init(void) ia64_sn_irtr_intr_enable(scd->scd_nasid, 0 /*ignored */ , SAL_IROUTER_INTR_RECV); + + /* on the console nasid, prepare to receive + * system controller environmental events + */ + if(scd->scd_nasid == event_nasid) { + scdrv_event_init(scd); + } } return 0; } diff --git a/drivers/char/snsc.h b/drivers/char/snsc.h index c22c6c55e25..a9efc13cc85 100644 --- a/drivers/char/snsc.h +++ b/drivers/char/snsc.h @@ -47,4 +47,44 @@ struct sysctl_data_s { nasid_t scd_nasid; /* Node on which subchannels are opened. */ }; + +/* argument types */ +#define IR_ARG_INT 0x00 /* 4-byte integer (big-endian) */ +#define IR_ARG_ASCII 0x01 /* null-terminated ASCII string */ +#define IR_ARG_UNKNOWN 0x80 /* unknown data type. The low + * 7 bits will contain the data + * length. */ +#define IR_ARG_UNKNOWN_LENGTH_MASK 0x7f + + +/* system controller event codes */ +#define EV_CLASS_MASK 0xf000ul +#define EV_SEVERITY_MASK 0x0f00ul +#define EV_COMPONENT_MASK 0x00fful + +#define EV_CLASS_POWER 0x1000ul +#define EV_CLASS_FAN 0x2000ul +#define EV_CLASS_TEMP 0x3000ul +#define EV_CLASS_ENV 0x4000ul +#define EV_CLASS_TEST_FAULT 0x5000ul +#define EV_CLASS_TEST_WARNING 0x6000ul +#define EV_CLASS_PWRD_NOTIFY 0x8000ul + +#define EV_SEVERITY_POWER_STABLE 0x0000ul +#define EV_SEVERITY_POWER_LOW_WARNING 0x0100ul +#define EV_SEVERITY_POWER_HIGH_WARNING 0x0200ul +#define EV_SEVERITY_POWER_HIGH_FAULT 0x0300ul +#define EV_SEVERITY_POWER_LOW_FAULT 0x0400ul + +#define EV_SEVERITY_FAN_STABLE 0x0000ul +#define EV_SEVERITY_FAN_WARNING 0x0100ul +#define EV_SEVERITY_FAN_FAULT 0x0200ul + +#define EV_SEVERITY_TEMP_STABLE 0x0000ul +#define EV_SEVERITY_TEMP_ADVISORY 0x0100ul +#define EV_SEVERITY_TEMP_CRITICAL 0x0200ul +#define EV_SEVERITY_TEMP_FAULT 0x0300ul + +void scdrv_event_init(struct sysctl_data_s *); + #endif /* _SN_SYSCTL_H_ */ diff --git a/drivers/char/snsc_event.c b/drivers/char/snsc_event.c new file mode 100644 index 00000000000..d692af57213 --- /dev/null +++ b/drivers/char/snsc_event.c @@ -0,0 +1,304 @@ +/* + * SN Platform system controller communication support + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2004 Silicon Graphics, Inc. All rights reserved. + */ + +/* + * System controller event handler + * + * These routines deal with environmental events arriving from the + * system controllers. + */ + +#include <linux/interrupt.h> +#include <linux/sched.h> +#include <linux/byteorder/generic.h> +#include <asm/sn/sn_sal.h> +#include "snsc.h" + +static struct subch_data_s *event_sd; + +void scdrv_event(unsigned long); +DECLARE_TASKLET(sn_sysctl_event, scdrv_event, 0); + +/* + * scdrv_event_interrupt + * + * Pull incoming environmental events off the physical link to the + * system controller and put them in a temporary holding area in SAL. + * Schedule scdrv_event() to move them along to their ultimate + * destination. + */ +static irqreturn_t +scdrv_event_interrupt(int irq, void *subch_data, struct pt_regs *regs) +{ + struct subch_data_s *sd = subch_data; + unsigned long flags; + int status; + + spin_lock_irqsave(&sd->sd_rlock, flags); + status = ia64_sn_irtr_intr(sd->sd_nasid, sd->sd_subch); + + if ((status > 0) && (status & SAL_IROUTER_INTR_RECV)) { + tasklet_schedule(&sn_sysctl_event); + } + spin_unlock_irqrestore(&sd->sd_rlock, flags); + return IRQ_HANDLED; +} + + +/* + * scdrv_parse_event + * + * Break an event (as read from SAL) into useful pieces so we can decide + * what to do with it. + */ +static int +scdrv_parse_event(char *event, int *src, int *code, int *esp_code, char *desc) +{ + char *desc_end; + + /* record event source address */ + *src = be32_to_cpup((__be32 *)event); + event += 4; /* move on to event code */ + + /* record the system controller's event code */ + *code = be32_to_cpup((__be32 *)event); + event += 4; /* move on to event arguments */ + + /* how many arguments are in the packet? */ + if (*event++ != 2) { + /* if not 2, give up */ + return -1; + } + + /* parse out the ESP code */ + if (*event++ != IR_ARG_INT) { + /* not an integer argument, so give up */ + return -1; + } + *esp_code = be32_to_cpup((__be32 *)event); + event += 4; + + /* parse out the event description */ + if (*event++ != IR_ARG_ASCII) { + /* not an ASCII string, so give up */ + return -1; + } + event[CHUNKSIZE-1] = '\0'; /* ensure this string ends! */ + event += 2; /* skip leading CR/LF */ + desc_end = desc + sprintf(desc, "%s", event); + + /* strip trailing CR/LF (if any) */ + for (desc_end--; + (desc_end != desc) && ((*desc_end == 0xd) || (*desc_end == 0xa)); + desc_end--) { + *desc_end = '\0'; + } + + return 0; +} + + +/* + * scdrv_event_severity + * + * Figure out how urgent a message we should write to the console/syslog + * via printk. + */ +static char * +scdrv_event_severity(int code) +{ + int ev_class = (code & EV_CLASS_MASK); + int ev_severity = (code & EV_SEVERITY_MASK); + char *pk_severity = KERN_NOTICE; + + switch (ev_class) { + case EV_CLASS_POWER: + switch (ev_severity) { + case EV_SEVERITY_POWER_LOW_WARNING: + case EV_SEVERITY_POWER_HIGH_WARNING: + pk_severity = KERN_WARNING; + break; + case EV_SEVERITY_POWER_HIGH_FAULT: + case EV_SEVERITY_POWER_LOW_FAULT: + pk_severity = KERN_ALERT; + break; + } + break; + case EV_CLASS_FAN: + switch (ev_severity) { + case EV_SEVERITY_FAN_WARNING: + pk_severity = KERN_WARNING; + break; + case EV_SEVERITY_FAN_FAULT: + pk_severity = KERN_CRIT; + break; + } + break; + case EV_CLASS_TEMP: + switch (ev_severity) { + case EV_SEVERITY_TEMP_ADVISORY: + pk_severity = KERN_WARNING; + break; + case EV_SEVERITY_TEMP_CRITICAL: + pk_severity = KERN_CRIT; + break; + case EV_SEVERITY_TEMP_FAULT: + pk_severity = KERN_ALERT; + break; + } + break; + case EV_CLASS_ENV: + pk_severity = KERN_ALERT; + break; + case EV_CLASS_TEST_FAULT: + pk_severity = KERN_ALERT; + break; + case EV_CLASS_TEST_WARNING: + pk_severity = KERN_WARNING; + break; + case EV_CLASS_PWRD_NOTIFY: + pk_severity = KERN_ALERT; + break; + } + + return pk_severity; +} + + +/* + * scdrv_dispatch_event + * + * Do the right thing with an incoming event. That's often nothing + * more than printing it to the system log. For power-down notifications + * we start a graceful shutdown. + */ +static void +scdrv_dispatch_event(char *event, int len) +{ + int code, esp_code, src; + char desc[CHUNKSIZE]; + char *severity; + + if (scdrv_parse_event(event, &src, &code, &esp_code, desc) < 0) { + /* ignore uninterpretible event */ + return; + } + + /* how urgent is the message? */ + severity = scdrv_event_severity(code); + + if ((code & EV_CLASS_MASK) == EV_CLASS_PWRD_NOTIFY) { + struct task_struct *p; + + /* give a SIGPWR signal to init proc */ + + /* first find init's task */ + read_lock(&tasklist_lock); + for_each_process(p) { + if (p->pid == 1) + break; + } + if (p) { /* we found init's task */ + printk(KERN_EMERG "Power off indication received. Initiating power fail sequence...\n"); + force_sig(SIGPWR, p); + } else { /* failed to find init's task - just give message(s) */ + printk(KERN_WARNING "Failed to find init proc to handle power off!\n"); + printk("%s|$(0x%x)%s\n", severity, esp_code, desc); + } + read_unlock(&tasklist_lock); + } else { + /* print to system log */ + printk("%s|$(0x%x)%s\n", severity, esp_code, desc); + } +} + + +/* + * scdrv_event + * + * Called as a tasklet when an event arrives from the L1. Read the event + * from where it's temporarily stored in SAL and call scdrv_dispatch_event() + * to send it on its way. Keep trying to read events until SAL indicates + * that there are no more immediately available. + */ +void +scdrv_event(unsigned long dummy) +{ + int status; + int len; + unsigned long flags; + struct subch_data_s *sd = event_sd; + + /* anything to read? */ + len = CHUNKSIZE; + spin_lock_irqsave(&sd->sd_rlock, flags); + status = ia64_sn_irtr_recv(sd->sd_nasid, sd->sd_subch, + sd->sd_rb, &len); + + while (!(status < 0)) { + spin_unlock_irqrestore(&sd->sd_rlock, flags); + scdrv_dispatch_event(sd->sd_rb, len); + len = CHUNKSIZE; + spin_lock_irqsave(&sd->sd_rlock, flags); + status = ia64_sn_irtr_recv(sd->sd_nasid, sd->sd_subch, + sd->sd_rb, &len); + } + spin_unlock_irqrestore(&sd->sd_rlock, flags); +} + + +/* + * scdrv_event_init + * + * Sets up a system controller subchannel to begin receiving event + * messages. This is sort of a specialized version of scdrv_open() + * in drivers/char/sn_sysctl.c. + */ +void +scdrv_event_init(struct sysctl_data_s *scd) +{ + int rv; + + event_sd = kmalloc(sizeof (struct subch_data_s), GFP_KERNEL); + if (event_sd == NULL) { + printk(KERN_WARNING "%s: couldn't allocate subchannel info" + " for event monitoring\n", __FUNCTION__); + return; + } + + /* initialize subch_data_s fields */ + memset(event_sd, 0, sizeof (struct subch_data_s)); + event_sd->sd_nasid = scd->scd_nasid; + spin_lock_init(&event_sd->sd_rlock); + + /* ask the system controllers to send events to this node */ + event_sd->sd_subch = ia64_sn_sysctl_event_init(scd->scd_nasid); + + if (event_sd->sd_subch < 0) { + kfree(event_sd); + printk(KERN_WARNING "%s: couldn't open event subchannel\n", + __FUNCTION__); + return; + } + + /* hook event subchannel up to the system controller interrupt */ + rv = request_irq(SGI_UART_VECTOR, scdrv_event_interrupt, + SA_SHIRQ | SA_INTERRUPT, + "system controller events", event_sd); + if (rv) { + printk(KERN_WARNING "%s: irq request failed (%d)\n", + __FUNCTION__, rv); + ia64_sn_irtr_close(event_sd->sd_nasid, event_sd->sd_subch); + kfree(event_sd); + return; + } +} + + diff --git a/drivers/char/sonypi.c b/drivers/char/sonypi.c index c812191417c..fd042060809 100644 --- a/drivers/char/sonypi.c +++ b/drivers/char/sonypi.c @@ -1021,11 +1021,11 @@ static int sonypi_misc_ioctl(struct inode *ip, struct file *fp, ret = -EIO; break; } - if (copy_to_user((u8 *)arg, &val8, sizeof(val8))) + if (copy_to_user(argp, &val8, sizeof(val8))) ret = -EFAULT; break; case SONYPI_IOCSFAN: - if (copy_from_user(&val8, (u8 *)arg, sizeof(val8))) { + if (copy_from_user(&val8, argp, sizeof(val8))) { ret = -EFAULT; break; } @@ -1038,7 +1038,7 @@ static int sonypi_misc_ioctl(struct inode *ip, struct file *fp, ret = -EIO; break; } - if (copy_to_user((u8 *)arg, &val8, sizeof(val8))) + if (copy_to_user(argp, &val8, sizeof(val8))) ret = -EFAULT; break; default: diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c index c789d5ceac7..50e0b612a8a 100644 --- a/drivers/char/specialix.c +++ b/drivers/char/specialix.c @@ -1987,10 +1987,9 @@ static inline int sx_set_serial_info(struct specialix_port * port, func_enter(); /* - error = verify_area(VERIFY_READ, (void *) newinfo, sizeof(tmp)); - if (error) { + if (!access_ok(VERIFY_READ, (void *) newinfo, sizeof(tmp))) { func_exit(); - return error; + return -EFAULT; } */ if (copy_from_user(&tmp, newinfo, sizeof(tmp))) { @@ -2046,14 +2045,12 @@ static inline int sx_get_serial_info(struct specialix_port * port, { struct serial_struct tmp; struct specialix_board *bp = port_Board(port); - // int error; func_enter(); /* - error = verify_area(VERIFY_WRITE, (void *) retinfo, sizeof(tmp)); - if (error) - return error; + if (!access_ok(VERIFY_WRITE, (void *) retinfo, sizeof(tmp))) + return -EFAULT; */ memset(&tmp, 0, sizeof(tmp)); diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index de166608c59..b8899f560b5 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -466,7 +466,7 @@ static int stl_parsebrd(stlconf_t *confp, char **argp); static unsigned long stl_atol(char *str); -int stl_init(void); +static int stl_init(void); static int stl_open(struct tty_struct *tty, struct file *filp); static void stl_close(struct tty_struct *tty, struct file *filp); static int stl_write(struct tty_struct *tty, const unsigned char *buf, int count); @@ -3063,7 +3063,7 @@ static struct tty_operations stl_ops = { /*****************************************************************************/ -int __init stl_init(void) +static int __init stl_init(void) { int i; printk(KERN_INFO "%s: version %s\n", stl_drvtitle, stl_drvversion); diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c index ca36087d4f8..87235330fdb 100644 --- a/drivers/char/tpm/tpm.c +++ b/drivers/char/tpm/tpm.c @@ -143,7 +143,7 @@ static ssize_t tpm_transmit(struct tpm_chip *chip, const char *buf, return -ENODATA; if (count > bufsiz) { dev_err(&chip->pci_dev->dev, - "invalid count value %x %x \n", count, bufsiz); + "invalid count value %x %zx \n", count, bufsiz); return -E2BIG; } @@ -151,7 +151,7 @@ static ssize_t tpm_transmit(struct tpm_chip *chip, const char *buf, if ((len = chip->vendor->send(chip, (u8 *) buf, count)) < 0) { dev_err(&chip->pci_dev->dev, - "tpm_transmit: tpm_send: error %d\n", len); + "tpm_transmit: tpm_send: error %zd\n", len); return len; } @@ -188,7 +188,7 @@ out_recv: len = chip->vendor->recv(chip, (u8 *) buf, bufsiz); if (len < 0) dev_err(&chip->pci_dev->dev, - "tpm_transmit: tpm_recv: error %d\n", len); + "tpm_transmit: tpm_recv: error %zd\n", len); up(&chip->tpm_mutex); return len; } diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 06e5a3f1836..26e5e19ed85 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -185,7 +185,7 @@ char *tty_name(struct tty_struct *tty, char *buf) EXPORT_SYMBOL(tty_name); -inline int tty_paranoia_check(struct tty_struct *tty, struct inode *inode, +int tty_paranoia_check(struct tty_struct *tty, struct inode *inode, const char *routine) { #ifdef TTY_PARANOIA_CHECK diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c index 5d386f4bea4..8971484b956 100644 --- a/drivers/char/vt_ioctl.c +++ b/drivers/char/vt_ioctl.c @@ -24,6 +24,7 @@ #include <linux/major.h> #include <linux/fs.h> #include <linux/console.h> +#include <linux/signal.h> #include <asm/io.h> #include <asm/uaccess.h> @@ -641,7 +642,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, extern int spawnpid, spawnsig; if (!perm || !capable(CAP_KILL)) return -EPERM; - if (arg < 1 || arg > _NSIG || arg == SIGKILL) + if (!valid_signal(arg) || arg < 1 || arg == SIGKILL) return -EINVAL; spawnpid = current->pid; spawnsig = arg; diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index b30001f3161..8e561313d09 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -223,7 +223,7 @@ static inline void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) } if ((val == CPUFREQ_PRECHANGE && ci->old < ci->new) || (val == CPUFREQ_POSTCHANGE && ci->old > ci->new) || - (val == CPUFREQ_RESUMECHANGE)) { + (val == CPUFREQ_RESUMECHANGE || val == CPUFREQ_SUSPENDCHANGE)) { loops_per_jiffy = cpufreq_scale(l_p_j_ref, l_p_j_ref_freq, ci->new); dprintk("scaling loops_per_jiffy to %lu for frequency %u kHz\n", loops_per_jiffy, ci->new); } @@ -866,11 +866,90 @@ EXPORT_SYMBOL(cpufreq_get); /** + * cpufreq_suspend - let the low level driver prepare for suspend + */ + +static int cpufreq_suspend(struct sys_device * sysdev, u32 state) +{ + int cpu = sysdev->id; + unsigned int ret = 0; + unsigned int cur_freq = 0; + struct cpufreq_policy *cpu_policy; + + dprintk("resuming cpu %u\n", cpu); + + if (!cpu_online(cpu)) + return 0; + + /* we may be lax here as interrupts are off. Nonetheless + * we need to grab the correct cpu policy, as to check + * whether we really run on this CPU. + */ + + cpu_policy = cpufreq_cpu_get(cpu); + if (!cpu_policy) + return -EINVAL; + + /* only handle each CPU group once */ + if (unlikely(cpu_policy->cpu != cpu)) { + cpufreq_cpu_put(cpu_policy); + return 0; + } + + if (cpufreq_driver->suspend) { + ret = cpufreq_driver->suspend(cpu_policy, state); + if (ret) { + printk(KERN_ERR "cpufreq: suspend failed in ->suspend " + "step on CPU %u\n", cpu_policy->cpu); + cpufreq_cpu_put(cpu_policy); + return ret; + } + } + + + if (cpufreq_driver->flags & CPUFREQ_CONST_LOOPS) + goto out; + + if (cpufreq_driver->get) + cur_freq = cpufreq_driver->get(cpu_policy->cpu); + + if (!cur_freq || !cpu_policy->cur) { + printk(KERN_ERR "cpufreq: suspend failed to assert current " + "frequency is what timing core thinks it is.\n"); + goto out; + } + + if (unlikely(cur_freq != cpu_policy->cur)) { + struct cpufreq_freqs freqs; + + if (!(cpufreq_driver->flags & CPUFREQ_PM_NO_WARN)) + printk(KERN_DEBUG "Warning: CPU frequency is %u, " + "cpufreq assumed %u kHz.\n", + cur_freq, cpu_policy->cur); + + freqs.cpu = cpu; + freqs.old = cpu_policy->cur; + freqs.new = cur_freq; + + notifier_call_chain(&cpufreq_transition_notifier_list, + CPUFREQ_SUSPENDCHANGE, &freqs); + adjust_jiffies(CPUFREQ_SUSPENDCHANGE, &freqs); + + cpu_policy->cur = cur_freq; + } + + out: + cpufreq_cpu_put(cpu_policy); + return 0; +} + +/** * cpufreq_resume - restore proper CPU frequency handling after resume * * 1.) resume CPUfreq hardware support (cpufreq_driver->resume()) * 2.) if ->target and !CPUFREQ_CONST_LOOPS: verify we're in sync - * 3.) schedule call cpufreq_update_policy() ASAP as interrupts are restored. + * 3.) schedule call cpufreq_update_policy() ASAP as interrupts are + * restored. */ static int cpufreq_resume(struct sys_device * sysdev) { @@ -915,21 +994,26 @@ static int cpufreq_resume(struct sys_device * sysdev) cur_freq = cpufreq_driver->get(cpu_policy->cpu); if (!cur_freq || !cpu_policy->cur) { - printk(KERN_ERR "cpufreq: resume failed to assert current frequency is what timing core thinks it is.\n"); + printk(KERN_ERR "cpufreq: resume failed to assert " + "current frequency is what timing core " + "thinks it is.\n"); goto out; } if (unlikely(cur_freq != cpu_policy->cur)) { struct cpufreq_freqs freqs; - printk(KERN_WARNING "Warning: CPU frequency is %u, " - "cpufreq assumed %u kHz.\n", cur_freq, cpu_policy->cur); + if (!(cpufreq_driver->flags & CPUFREQ_PM_NO_WARN)) + printk(KERN_WARNING "Warning: CPU frequency" + "is %u, cpufreq assumed %u kHz.\n", + cur_freq, cpu_policy->cur); freqs.cpu = cpu; freqs.old = cpu_policy->cur; freqs.new = cur_freq; - notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_RESUMECHANGE, &freqs); + notifier_call_chain(&cpufreq_transition_notifier_list, + CPUFREQ_RESUMECHANGE, &freqs); adjust_jiffies(CPUFREQ_RESUMECHANGE, &freqs); cpu_policy->cur = cur_freq; @@ -945,6 +1029,7 @@ out: static struct sysdev_driver cpufreq_sysdev_driver = { .add = cpufreq_add_dev, .remove = cpufreq_remove_dev, + .suspend = cpufreq_suspend, .resume = cpufreq_resume, }; diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index a2827e43f0f..a0018de3bef 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -237,7 +237,7 @@ config I2C_KEYWEST config I2C_MPC tristate "MPC107/824x/85xx/52xx" - depends on I2C && PPC + depends on I2C && PPC32 help If you say yes to this option, support will be included for the built-in I2C interface on the MPC107/Tsi107/MPC8240/MPC8245 and diff --git a/drivers/i2c/busses/i2c-mpc.c b/drivers/i2c/busses/i2c-mpc.c index 75b8d867dae..6f33496d31c 100644 --- a/drivers/i2c/busses/i2c-mpc.c +++ b/drivers/i2c/busses/i2c-mpc.c @@ -55,7 +55,7 @@ #define CSR_RXAK 0x01 struct mpc_i2c { - char *base; + void __iomem *base; u32 interrupt; wait_queue_head_t queue; struct i2c_adapter adap; @@ -444,7 +444,7 @@ static int fsl_i2c_probe(struct device *device) fail_add: if (i2c->irq != 0) - free_irq(i2c->irq, 0); + free_irq(i2c->irq, NULL); fail_irq: iounmap(i2c->base); fail_map: diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c index 9d70ba5ea59..16b3e2d8bfb 100644 --- a/drivers/ide/pci/sis5513.c +++ b/drivers/ide/pci/sis5513.c @@ -726,7 +726,7 @@ static int sis5513_config_xfer_rate (ide_drive_t *drive) */ /* Chip detection and general config */ -static unsigned int __init init_chipset_sis5513 (struct pci_dev *dev, const char *name) +static unsigned int __devinit init_chipset_sis5513 (struct pci_dev *dev, const char *name) { struct pci_dev *host; int i = 0; @@ -879,7 +879,7 @@ static unsigned int __init init_chipset_sis5513 (struct pci_dev *dev, const char return 0; } -static unsigned int __init ata66_sis5513 (ide_hwif_t *hwif) +static unsigned int __devinit ata66_sis5513 (ide_hwif_t *hwif) { u8 ata66 = 0; @@ -897,7 +897,7 @@ static unsigned int __init ata66_sis5513 (ide_hwif_t *hwif) return ata66; } -static void __init init_hwif_sis5513 (ide_hwif_t *hwif) +static void __devinit init_hwif_sis5513 (ide_hwif_t *hwif) { hwif->autodma = 0; diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c index 6dc273a8132..569f1676744 100644 --- a/drivers/ide/ppc/pmac.c +++ b/drivers/ide/ppc/pmac.c @@ -1204,6 +1204,8 @@ pmac_ide_do_suspend(ide_hwif_t *hwif) } #endif /* CONFIG_BLK_DEV_IDE_PMAC_BLINK */ + disable_irq(pmif->irq); + /* The media bay will handle itself just fine */ if (pmif->mediabay) return 0; @@ -1236,7 +1238,6 @@ pmac_ide_do_resume(ide_hwif_t *hwif) ppc_md.feature_call(PMAC_FTR_IDE_ENABLE, pmif->node, pmif->aapl_bus_id, 1); msleep(10); ppc_md.feature_call(PMAC_FTR_IDE_RESET, pmif->node, pmif->aapl_bus_id, 0); - msleep(jiffies_to_msecs(IDE_WAKEUP_DELAY)); /* Kauai has it different */ if (pmif->kauai_fcr) { @@ -1244,11 +1245,15 @@ pmac_ide_do_resume(ide_hwif_t *hwif) fcr |= KAUAI_FCR_UATA_RESET_N | KAUAI_FCR_UATA_ENABLE; writel(fcr, pmif->kauai_fcr); } + + msleep(jiffies_to_msecs(IDE_WAKEUP_DELAY)); } /* Sanitize drive timings */ sanitize_timings(pmif); + enable_irq(pmif->irq); + return 0; } diff --git a/drivers/ieee1394/ohci1394.c b/drivers/ieee1394/ohci1394.c index 97ff364c043..6cb0b586c29 100644 --- a/drivers/ieee1394/ohci1394.c +++ b/drivers/ieee1394/ohci1394.c @@ -539,10 +539,8 @@ static void ohci_initialize(struct ti_ohci *ohci) initialize_dma_trm_ctx(&ohci->at_req_context); initialize_dma_trm_ctx(&ohci->at_resp_context); - /* Initialize IR Legacy DMA */ + /* Initialize IR Legacy DMA channel mask */ ohci->ir_legacy_channels = 0; - initialize_dma_rcv_ctx(&ohci->ir_legacy_context, 1); - DBGMSG("ISO receive legacy context activated"); /* * Accept AT requests from all nodes. This probably @@ -1032,6 +1030,8 @@ static int ohci_devctl(struct hpsb_host *host, enum devctl_cmd cmd, int arg) case ISO_LISTEN_CHANNEL: { u64 mask; + struct dma_rcv_ctx *d = &ohci->ir_legacy_context; + int ir_legacy_active; if (arg<0 || arg>63) { PRINT(KERN_ERR, @@ -1052,9 +1052,37 @@ static int ohci_devctl(struct hpsb_host *host, enum devctl_cmd cmd, int arg) return -EFAULT; } + ir_legacy_active = ohci->ir_legacy_channels; + ohci->ISO_channel_usage |= mask; ohci->ir_legacy_channels |= mask; + spin_unlock_irqrestore(&ohci->IR_channel_lock, flags); + + if (!ir_legacy_active) { + if (ohci1394_register_iso_tasklet(ohci, + &ohci->ir_legacy_tasklet) < 0) { + PRINT(KERN_ERR, "No IR DMA context available"); + return -EBUSY; + } + + /* the IR context can be assigned to any DMA context + * by ohci1394_register_iso_tasklet */ + d->ctx = ohci->ir_legacy_tasklet.context; + d->ctrlSet = OHCI1394_IsoRcvContextControlSet + + 32*d->ctx; + d->ctrlClear = OHCI1394_IsoRcvContextControlClear + + 32*d->ctx; + d->cmdPtr = OHCI1394_IsoRcvCommandPtr + 32*d->ctx; + d->ctxtMatch = OHCI1394_IsoRcvContextMatch + 32*d->ctx; + + initialize_dma_rcv_ctx(&ohci->ir_legacy_context, 1); + + PRINT(KERN_ERR, "IR legacy activated"); + } + + spin_lock_irqsave(&ohci->IR_channel_lock, flags); + if (arg>31) reg_write(ohci, OHCI1394_IRMultiChanMaskHiSet, 1<<(arg-32)); @@ -1101,6 +1129,12 @@ static int ohci_devctl(struct hpsb_host *host, enum devctl_cmd cmd, int arg) spin_unlock_irqrestore(&ohci->IR_channel_lock, flags); DBGMSG("Listening disabled on channel %d", arg); + + if (ohci->ir_legacy_channels == 0) { + stop_dma_rcv_ctx(&ohci->ir_legacy_context); + DBGMSG("ISO legacy receive context stopped"); + } + break; } default: @@ -1270,8 +1304,10 @@ static int ohci_iso_recv_init(struct hpsb_iso *iso) OHCI_ISO_RECEIVE, ohci_iso_recv_task, (unsigned long) iso); - if (ohci1394_register_iso_tasklet(recv->ohci, &recv->task) < 0) + if (ohci1394_register_iso_tasklet(recv->ohci, &recv->task) < 0) { + ret = -EBUSY; goto err; + } recv->task_active = 1; @@ -1896,8 +1932,10 @@ static int ohci_iso_xmit_init(struct hpsb_iso *iso) ohci1394_init_iso_tasklet(&xmit->task, OHCI_ISO_TRANSMIT, ohci_iso_xmit_task, (unsigned long) iso); - if (ohci1394_register_iso_tasklet(xmit->ohci, &xmit->task) < 0) + if (ohci1394_register_iso_tasklet(xmit->ohci, &xmit->task) < 0) { + ret = -EBUSY; goto err; + } xmit->task_active = 1; @@ -2691,7 +2729,7 @@ static void dma_rcv_tasklet (unsigned long data) (cond_le32_to_cpu(d->spb[length/4-1], ohci->no_swap_incoming)>>16)&0x1f, (cond_le32_to_cpu(d->spb[length/4-1], ohci->no_swap_incoming)>>21)&0x3, tcode, length, d->ctx, - (cond_le32_to_cpu(d->spb[length/4-1], ohci->no_swap_incoming)>>10)&0x3f); + (cond_le32_to_cpu(d->spb[0], ohci->no_swap_incoming)>>10)&0x3f); ack = (((cond_le32_to_cpu(d->spb[length/4-1], ohci->no_swap_incoming)>>16)&0x1f) == 0x11) ? 1 : 0; @@ -2754,7 +2792,7 @@ static void dma_trm_tasklet (unsigned long data) d->ctx); else DBGMSG("Packet sent to node %d tcode=0x%X tLabel=" - "0x%02X ack=0x%X spd=%d dataLength=%d ctx=%d", + "%d ack=0x%X spd=%d dataLength=%d ctx=%d", (le32_to_cpu(d->prg_cpu[d->sent_ind]->data[1])>>16)&0x3f, (le32_to_cpu(d->prg_cpu[d->sent_ind]->data[0])>>4)&0xf, (le32_to_cpu(d->prg_cpu[d->sent_ind]->data[0])>>10)&0x3f, @@ -2763,7 +2801,7 @@ static void dma_trm_tasklet (unsigned long data) d->ctx); else DBGMSG("Packet sent to node %d tcode=0x%X tLabel=" - "0x%02X ack=0x%X spd=%d data=0x%08X ctx=%d", + "%d ack=0x%X spd=%d data=0x%08X ctx=%d", (le32_to_cpu(d->prg_cpu[d->sent_ind]->data[1]) >>16)&0x3f, (le32_to_cpu(d->prg_cpu[d->sent_ind]->data[0]) @@ -2999,20 +3037,6 @@ alloc_dma_rcv_ctx(struct ti_ohci *ohci, struct dma_rcv_ctx *d, ohci1394_init_iso_tasklet(&ohci->ir_legacy_tasklet, OHCI_ISO_MULTICHANNEL_RECEIVE, dma_rcv_tasklet, (unsigned long) d); - if (ohci1394_register_iso_tasklet(ohci, - &ohci->ir_legacy_tasklet) < 0) { - PRINT(KERN_ERR, "No IR DMA context available"); - free_dma_rcv_ctx(d); - return -EBUSY; - } - - /* the IR context can be assigned to any DMA context - * by ohci1394_register_iso_tasklet */ - d->ctx = ohci->ir_legacy_tasklet.context; - d->ctrlSet = OHCI1394_IsoRcvContextControlSet + 32*d->ctx; - d->ctrlClear = OHCI1394_IsoRcvContextControlClear + 32*d->ctx; - d->cmdPtr = OHCI1394_IsoRcvCommandPtr + 32*d->ctx; - d->ctxtMatch = OHCI1394_IsoRcvContextMatch + 32*d->ctx; } else { d->ctrlSet = context_base + OHCI1394_ContextControlSet; d->ctrlClear = context_base + OHCI1394_ContextControlClear; @@ -3413,7 +3437,6 @@ static void ohci1394_pci_remove(struct pci_dev *pdev) switch (ohci->init_state) { case OHCI_INIT_DONE: - stop_dma_rcv_ctx(&ohci->ir_legacy_context); hpsb_remove_host(ohci->host); /* Clear out BUS Options */ diff --git a/drivers/infiniband/hw/mthca/mthca_memfree.c b/drivers/infiniband/hw/mthca/mthca_memfree.c index 5824b6d3769..637b30e3559 100644 --- a/drivers/infiniband/hw/mthca/mthca_memfree.c +++ b/drivers/infiniband/hw/mthca/mthca_memfree.c @@ -32,6 +32,8 @@ * $Id$ */ +#include <linux/mm.h> + #include "mthca_memfree.h" #include "mthca_dev.h" #include "mthca_cmd.h" diff --git a/drivers/input/joystick/spaceorb.c b/drivers/input/joystick/spaceorb.c index c76cf8ff29c..874367bfab0 100644 --- a/drivers/input/joystick/spaceorb.c +++ b/drivers/input/joystick/spaceorb.c @@ -116,7 +116,7 @@ static void spaceorb_process_packet(struct spaceorb *spaceorb, struct pt_regs *r case 'K': /* Button data */ if (spaceorb->idx != 5) return; - for (i = 0; i < 7; i++) + for (i = 0; i < 6; i++) input_report_key(dev, spaceorb_buttons[i], (data[2] >> i) & 1); break; diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c index f7304f0ce54..ff66ed4ee2c 100644 --- a/drivers/input/keyboard/atkbd.c +++ b/drivers/input/keyboard/atkbd.c @@ -678,7 +678,7 @@ static void atkbd_disconnect(struct serio *serio) atkbd_disable(atkbd); /* make sure we don't have a command in flight */ - synchronize_kernel(); + synchronize_sched(); /* Allow atkbd_interrupt()s to complete. */ flush_scheduled_work(); device_remove_file(&serio->dev, &atkbd_attr_extra); diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c index 06163538bb2..12dee8e9fbb 100644 --- a/drivers/isdn/capi/capi.c +++ b/drivers/isdn/capi/capi.c @@ -60,12 +60,12 @@ MODULE_LICENSE("GPL"); static struct class_simple *capi_class; -int capi_major = 68; /* allocated */ +static int capi_major = 68; /* allocated */ #ifdef CONFIG_ISDN_CAPI_MIDDLEWARE #define CAPINC_NR_PORTS 32 #define CAPINC_MAX_PORTS 256 -int capi_ttymajor = 191; -int capi_ttyminors = CAPINC_NR_PORTS; +static int capi_ttymajor = 191; +static int capi_ttyminors = CAPINC_NR_PORTS; #endif /* CONFIG_ISDN_CAPI_MIDDLEWARE */ module_param_named(major, capi_major, uint, 0); @@ -268,7 +268,7 @@ static void capiminor_free(struct capiminor *mp) kfree(mp); } -struct capiminor *capiminor_find(unsigned int minor) +static struct capiminor *capiminor_find(unsigned int minor) { struct list_head *l; struct capiminor *p = NULL; @@ -1166,7 +1166,7 @@ static int capinc_tty_write_room(struct tty_struct *tty) return room; } -int capinc_tty_chars_in_buffer(struct tty_struct *tty) +static int capinc_tty_chars_in_buffer(struct tty_struct *tty) { struct capiminor *mp = (struct capiminor *)tty->driver_data; if (!mp || !mp->nccip) { diff --git a/drivers/isdn/capi/kcapi_proc.c b/drivers/isdn/capi/kcapi_proc.c index 16dc5418ff4..2cc8b27e4c3 100644 --- a/drivers/isdn/capi/kcapi_proc.c +++ b/drivers/isdn/capi/kcapi_proc.c @@ -89,14 +89,14 @@ static int contrstats_show(struct seq_file *seq, void *v) return 0; } -struct seq_operations seq_controller_ops = { +static struct seq_operations seq_controller_ops = { .start = controller_start, .next = controller_next, .stop = controller_stop, .show = controller_show, }; -struct seq_operations seq_contrstats_ops = { +static struct seq_operations seq_contrstats_ops = { .start = controller_start, .next = controller_next, .stop = controller_stop, @@ -192,14 +192,14 @@ applstats_show(struct seq_file *seq, void *v) return 0; } -struct seq_operations seq_applications_ops = { +static struct seq_operations seq_applications_ops = { .start = applications_start, .next = applications_next, .stop = applications_stop, .show = applications_show, }; -struct seq_operations seq_applstats_ops = { +static struct seq_operations seq_applstats_ops = { .start = applications_start, .next = applications_next, .stop = applications_stop, @@ -287,7 +287,7 @@ static int capi_driver_show(struct seq_file *seq, void *v) return 0; } -struct seq_operations seq_capi_driver_ops = { +static struct seq_operations seq_capi_driver_ops = { .start = capi_driver_start, .next = capi_driver_next, .stop = capi_driver_stop, diff --git a/drivers/isdn/divert/isdn_divert.c b/drivers/isdn/divert/isdn_divert.c index 1eb112213f0..0bfd698726a 100644 --- a/drivers/isdn/divert/isdn_divert.c +++ b/drivers/isdn/divert/isdn_divert.c @@ -383,7 +383,7 @@ divert_rule *getruleptr(int idx) /*************************************************/ /* called from common module on an incoming call */ /*************************************************/ -int isdn_divert_icall(isdn_ctrl *ic) +static int isdn_divert_icall(isdn_ctrl *ic) { int retval = 0; unsigned long flags; struct call_struc *cs = NULL; @@ -552,7 +552,7 @@ void deleteprocs(void) /****************************************************/ /* put a address including address type into buffer */ /****************************************************/ -int put_address(char *st, u_char *p, int len) +static int put_address(char *st, u_char *p, int len) { u_char retval = 0; u_char adr_typ = 0; /* network standard */ @@ -595,7 +595,7 @@ int put_address(char *st, u_char *p, int len) /*************************************/ /* report a succesfull interrogation */ /*************************************/ -int interrogate_success(isdn_ctrl *ic, struct call_struc *cs) +static int interrogate_success(isdn_ctrl *ic, struct call_struc *cs) { char *src = ic->parm.dss1_io.data; int restlen = ic->parm.dss1_io.datalen; int cnt = 1; @@ -689,7 +689,7 @@ int interrogate_success(isdn_ctrl *ic, struct call_struc *cs) /*********************************************/ /* callback for protocol specific extensions */ /*********************************************/ -int prot_stat_callback(isdn_ctrl *ic) +static int prot_stat_callback(isdn_ctrl *ic) { struct call_struc *cs, *cs1; int i; unsigned long flags; @@ -781,7 +781,7 @@ int prot_stat_callback(isdn_ctrl *ic) /***************************/ /* status callback from HL */ /***************************/ -int isdn_divert_stat_callback(isdn_ctrl *ic) +static int isdn_divert_stat_callback(isdn_ctrl *ic) { struct call_struc *cs, *cs1; unsigned long flags; int retval; diff --git a/drivers/isdn/hisax/hfc4s8s_l1.c b/drivers/isdn/hisax/hfc4s8s_l1.c index 1ac46c26b93..ba1d028343e 100644 --- a/drivers/isdn/hisax/hfc4s8s_l1.c +++ b/drivers/isdn/hisax/hfc4s8s_l1.c @@ -29,6 +29,7 @@ #include <linux/timer.h> #include <linux/skbuff.h> #include <linux/wait.h> +#include <asm/io.h> #include "hisax_if.h" #include "hfc4s8s_l1.h" diff --git a/drivers/macintosh/adbhid.c b/drivers/macintosh/adbhid.c index 8f93d01d892..db654e8bd67 100644 --- a/drivers/macintosh/adbhid.c +++ b/drivers/macintosh/adbhid.c @@ -555,6 +555,42 @@ adbhid_buttons_input(unsigned char *data, int nb, struct pt_regs *regs, int auto #endif /* CONFIG_PMAC_BACKLIGHT */ input_report_key(&adbhid[id]->input, KEY_BRIGHTNESSUP, down); break; + + case 0xc: /* videomode switch */ + input_report_key(&adbhid[id]->input, KEY_SWITCHVIDEOMODE, down); + break; + + case 0xd: /* keyboard illumination toggle */ + input_report_key(&adbhid[id]->input, KEY_KBDILLUMTOGGLE, down); + break; + + case 0xe: /* keyboard illumination decrease */ + input_report_key(&adbhid[id]->input, KEY_KBDILLUMDOWN, down); + break; + + case 0xf: + switch (data[1]) { + case 0x8f: + case 0x0f: + /* keyboard illumination increase */ + input_report_key(&adbhid[id]->input, KEY_KBDILLUMUP, down); + break; + + case 0x7f: + case 0xff: + /* keypad overlay toogle */ + break; + + default: + printk(KERN_INFO "Unhandled ADB_MISC event %02x, %02x, %02x, %02x\n", + data[0], data[1], data[2], data[3]); + break; + } + break; + default: + printk(KERN_INFO "Unhandled ADB_MISC event %02x, %02x, %02x, %02x\n", + data[0], data[1], data[2], data[3]); + break; } } break; @@ -775,6 +811,10 @@ adbhid_input_register(int id, int default_id, int original_handler_id, set_bit(KEY_BRIGHTNESSUP, adbhid[id]->input.keybit); set_bit(KEY_BRIGHTNESSDOWN, adbhid[id]->input.keybit); set_bit(KEY_EJECTCD, adbhid[id]->input.keybit); + set_bit(KEY_SWITCHVIDEOMODE, adbhid[id]->input.keybit); + set_bit(KEY_KBDILLUMTOGGLE, adbhid[id]->input.keybit); + set_bit(KEY_KBDILLUMDOWN, adbhid[id]->input.keybit); + set_bit(KEY_KBDILLUMUP, adbhid[id]->input.keybit); break; } if (adbhid[id]->name[0]) diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c index fdea1a3a631..e654aa5eecd 100644 --- a/drivers/macintosh/via-pmu.c +++ b/drivers/macintosh/via-pmu.c @@ -2351,6 +2351,10 @@ pmac_suspend_devices(void) return -EBUSY; } + /* Disable clock spreading on some machines */ + pmac_tweak_clock_spreading(0); + + /* Stop preemption */ preempt_disable(); /* Make sure the decrementer won't interrupt us */ @@ -2417,11 +2421,12 @@ pmac_wakeup_devices(void) /* Re-enable local CPU interrupts */ local_irq_enable(); - mdelay(100); - preempt_enable(); + /* Re-enable clock spreading on some machines */ + pmac_tweak_clock_spreading(1); + /* Resume devices */ device_resume(); diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 77619a56e2b..0dd6c2b5391 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -331,25 +331,19 @@ crypt_alloc_buffer(struct crypt_config *cc, unsigned int size, struct bio *bio; unsigned int nr_iovecs = (size + PAGE_SIZE - 1) >> PAGE_SHIFT; int gfp_mask = GFP_NOIO | __GFP_HIGHMEM; - unsigned long flags = current->flags; unsigned int i; /* - * Tell VM to act less aggressively and fail earlier. - * This is not necessary but increases throughput. + * Use __GFP_NOMEMALLOC to tell the VM to act less aggressively and + * to fail earlier. This is not necessary but increases throughput. * FIXME: Is this really intelligent? */ - current->flags &= ~PF_MEMALLOC; - if (base_bio) - bio = bio_clone(base_bio, GFP_NOIO); + bio = bio_clone(base_bio, GFP_NOIO|__GFP_NOMEMALLOC); else - bio = bio_alloc(GFP_NOIO, nr_iovecs); - if (!bio) { - if (flags & PF_MEMALLOC) - current->flags |= PF_MEMALLOC; + bio = bio_alloc(GFP_NOIO|__GFP_NOMEMALLOC, nr_iovecs); + if (!bio) return NULL; - } /* if the last bio was not complete, continue where that one ended */ bio->bi_idx = *bio_vec_idx; @@ -386,9 +380,6 @@ crypt_alloc_buffer(struct crypt_config *cc, unsigned int size, size -= bv->bv_len; } - if (flags & PF_MEMALLOC) - current->flags |= PF_MEMALLOC; - if (!bio->bi_size) { bio_put(bio); return NULL; diff --git a/drivers/md/dm-emc.c b/drivers/md/dm-emc.c index 70065866459..c7067674dcb 100644 --- a/drivers/md/dm-emc.c +++ b/drivers/md/dm-emc.c @@ -223,8 +223,10 @@ static struct emc_handler *alloc_emc_handler(void) { struct emc_handler *h = kmalloc(sizeof(*h), GFP_KERNEL); - if (h) + if (h) { + memset(h, 0, sizeof(*h)); spin_lock_init(&h->lock); + } return h; } @@ -259,8 +261,6 @@ static int emc_create(struct hw_handler *hwh, unsigned argc, char **argv) if (!h) return -ENOMEM; - memset(h, 0, sizeof(*h)); - hwh->context = h; if ((h->short_trespass = short_trespass)) diff --git a/drivers/md/dm-hw-handler.c b/drivers/md/dm-hw-handler.c index ae63772e44c..4cc0010e015 100644 --- a/drivers/md/dm-hw-handler.c +++ b/drivers/md/dm-hw-handler.c @@ -23,7 +23,7 @@ struct hwh_internal { static LIST_HEAD(_hw_handlers); static DECLARE_RWSEM(_hwh_lock); -struct hwh_internal *__find_hw_handler_type(const char *name) +static struct hwh_internal *__find_hw_handler_type(const char *name) { struct hwh_internal *hwhi; diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 43763a0bd09..1e97b3c12bd 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -101,6 +101,7 @@ typedef int (*action_fn) (struct pgpath *pgpath); static kmem_cache_t *_mpio_cache; +struct workqueue_struct *kmultipathd; static void process_queued_ios(void *data); static void trigger_event(void *data); @@ -308,7 +309,7 @@ static int map_io(struct multipath *m, struct bio *bio, struct mpath_io *mpio, bio_list_add(&m->queued_ios, bio); m->queue_size++; if (m->pg_init_required || !m->queue_io) - schedule_work(&m->process_queued_ios); + queue_work(kmultipathd, &m->process_queued_ios); pgpath = NULL; r = 0; } else if (!pgpath) @@ -334,7 +335,7 @@ static int queue_if_no_path(struct multipath *m, unsigned queue_if_no_path) m->queue_if_no_path = queue_if_no_path; if (!m->queue_if_no_path) - schedule_work(&m->process_queued_ios); + queue_work(kmultipathd, &m->process_queued_ios); spin_unlock_irqrestore(&m->lock, flags); @@ -800,7 +801,7 @@ static int fail_path(struct pgpath *pgpath) if (pgpath == m->current_pgpath) m->current_pgpath = NULL; - schedule_work(&m->trigger_event); + queue_work(kmultipathd, &m->trigger_event); out: spin_unlock_irqrestore(&m->lock, flags); @@ -837,9 +838,9 @@ static int reinstate_path(struct pgpath *pgpath) m->current_pgpath = NULL; if (!m->nr_valid_paths++) - schedule_work(&m->process_queued_ios); + queue_work(kmultipathd, &m->process_queued_ios); - schedule_work(&m->trigger_event); + queue_work(kmultipathd, &m->trigger_event); out: spin_unlock_irqrestore(&m->lock, flags); @@ -883,7 +884,7 @@ static void bypass_pg(struct multipath *m, struct priority_group *pg, spin_unlock_irqrestore(&m->lock, flags); - schedule_work(&m->trigger_event); + queue_work(kmultipathd, &m->trigger_event); } /* @@ -913,7 +914,7 @@ static int switch_pg_num(struct multipath *m, const char *pgstr) } spin_unlock_irqrestore(&m->lock, flags); - schedule_work(&m->trigger_event); + queue_work(kmultipathd, &m->trigger_event); return 0; } @@ -968,7 +969,7 @@ void dm_pg_init_complete(struct path *path, unsigned err_flags) m->current_pgpath = NULL; m->current_pg = NULL; } - schedule_work(&m->process_queued_ios); + queue_work(kmultipathd, &m->process_queued_ios); spin_unlock_irqrestore(&m->lock, flags); } @@ -1018,7 +1019,7 @@ static int do_end_io(struct multipath *m, struct bio *bio, bio_list_add(&m->queued_ios, bio); m->queue_size++; if (!m->queue_io) - schedule_work(&m->process_queued_ios); + queue_work(kmultipathd, &m->process_queued_ios); spin_unlock(&m->lock); return 1; /* io not complete */ @@ -1057,7 +1058,7 @@ static void multipath_presuspend(struct dm_target *ti) spin_lock_irqsave(&m->lock, flags); m->suspended = 1; if (m->queue_if_no_path) - schedule_work(&m->process_queued_ios); + queue_work(kmultipathd, &m->process_queued_ios); spin_unlock_irqrestore(&m->lock, flags); } @@ -1274,6 +1275,15 @@ static int __init dm_multipath_init(void) return -EINVAL; } + kmultipathd = create_workqueue("kmpathd"); + if (!kmultipathd) { + DMERR("%s: failed to create workqueue kmpathd", + multipath_target.name); + dm_unregister_target(&multipath_target); + kmem_cache_destroy(_mpio_cache); + return -ENOMEM; + } + DMINFO("dm-multipath version %u.%u.%u loaded", multipath_target.version[0], multipath_target.version[1], multipath_target.version[2]); @@ -1285,6 +1295,8 @@ static void __exit dm_multipath_exit(void) { int r; + destroy_workqueue(kmultipathd); + r = dm_unregister_target(&multipath_target); if (r < 0) DMERR("%s: target unregister failed %d", diff --git a/drivers/md/dm-path-selector.c b/drivers/md/dm-path-selector.c index ac5c4bbec6c..a28c1c2b4ef 100644 --- a/drivers/md/dm-path-selector.c +++ b/drivers/md/dm-path-selector.c @@ -26,7 +26,7 @@ struct ps_internal { static LIST_HEAD(_path_selectors); static DECLARE_RWSEM(_ps_lock); -struct ps_internal *__find_path_selector_type(const char *name) +static struct ps_internal *__find_path_selector_type(const char *name) { struct ps_internal *psi; diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index ee175d4906c..18e9b9953fc 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -242,7 +242,7 @@ static void free_devices(struct list_head *devices) } } -void table_destroy(struct dm_table *t) +static void table_destroy(struct dm_table *t) { unsigned int i; diff --git a/drivers/md/dm-zero.c b/drivers/md/dm-zero.c index 7febc2cac73..51c0639b248 100644 --- a/drivers/md/dm-zero.c +++ b/drivers/md/dm-zero.c @@ -55,7 +55,7 @@ static struct target_type zero_target = { .map = zero_map, }; -int __init dm_zero_init(void) +static int __init dm_zero_init(void) { int r = dm_register_target(&zero_target); @@ -65,7 +65,7 @@ int __init dm_zero_init(void) return r; } -void __exit dm_zero_exit(void) +static void __exit dm_zero_exit(void) { int r = dm_unregister_target(&zero_target); diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 243ff6884e8..f6b03957efc 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -97,6 +97,7 @@ struct mapped_device { * freeze/thaw support require holding onto a super block */ struct super_block *frozen_sb; + struct block_device *frozen_bdev; }; #define MIN_IOS 256 @@ -990,44 +991,50 @@ int dm_swap_table(struct mapped_device *md, struct dm_table *table) */ static int __lock_fs(struct mapped_device *md) { - struct block_device *bdev; + int error = -ENOMEM; if (test_and_set_bit(DMF_FS_LOCKED, &md->flags)) return 0; - bdev = bdget_disk(md->disk, 0); - if (!bdev) { + md->frozen_bdev = bdget_disk(md->disk, 0); + if (!md->frozen_bdev) { DMWARN("bdget failed in __lock_fs"); - return -ENOMEM; + goto out; } WARN_ON(md->frozen_sb); - md->frozen_sb = freeze_bdev(bdev); + + md->frozen_sb = freeze_bdev(md->frozen_bdev); + if (IS_ERR(md->frozen_sb)) { + error = PTR_ERR(md->frozen_sb); + goto out_bdput; + } + /* don't bdput right now, we don't want the bdev * to go away while it is locked. We'll bdput * in __unlock_fs */ return 0; + +out_bdput: + bdput(md->frozen_bdev); + md->frozen_sb = NULL; + md->frozen_bdev = NULL; +out: + clear_bit(DMF_FS_LOCKED, &md->flags); + return error; } -static int __unlock_fs(struct mapped_device *md) +static void __unlock_fs(struct mapped_device *md) { - struct block_device *bdev; - if (!test_and_clear_bit(DMF_FS_LOCKED, &md->flags)) - return 0; + return; - bdev = bdget_disk(md->disk, 0); - if (!bdev) { - DMWARN("bdget failed in __unlock_fs"); - return -ENOMEM; - } + thaw_bdev(md->frozen_bdev, md->frozen_sb); + bdput(md->frozen_bdev); - thaw_bdev(bdev, md->frozen_sb); md->frozen_sb = NULL; - bdput(bdev); - bdput(bdev); - return 0; + md->frozen_bdev = NULL; } /* @@ -1041,37 +1048,37 @@ int dm_suspend(struct mapped_device *md) { struct dm_table *map; DECLARE_WAITQUEUE(wait, current); + int error = -EINVAL; /* Flush I/O to the device. */ down_read(&md->lock); - if (test_bit(DMF_BLOCK_IO, &md->flags)) { - up_read(&md->lock); - return -EINVAL; - } + if (test_bit(DMF_BLOCK_IO, &md->flags)) + goto out_read_unlock; + + error = __lock_fs(md); + if (error) + goto out_read_unlock; map = dm_get_table(md); if (map) dm_table_presuspend_targets(map); - __lock_fs(md); up_read(&md->lock); /* - * First we set the BLOCK_IO flag so no more ios will be - * mapped. + * First we set the BLOCK_IO flag so no more ios will be mapped. + * + * If the flag is already set we know another thread is trying to + * suspend as well, so we leave the fs locked for this thread. */ + error = -EINVAL; down_write(&md->lock); - if (test_bit(DMF_BLOCK_IO, &md->flags)) { - /* - * If we get here we know another thread is - * trying to suspend as well, so we leave the fs - * locked for this thread. - */ - up_write(&md->lock); - return -EINVAL; + if (test_and_set_bit(DMF_BLOCK_IO, &md->flags)) { + if (map) + dm_table_put(map); + goto out_write_unlock; } - set_bit(DMF_BLOCK_IO, &md->flags); add_wait_queue(&md->wait, &wait); up_write(&md->lock); @@ -1099,12 +1106,9 @@ int dm_suspend(struct mapped_device *md) remove_wait_queue(&md->wait, &wait); /* were we interrupted ? */ - if (atomic_read(&md->pending)) { - __unlock_fs(md); - clear_bit(DMF_BLOCK_IO, &md->flags); - up_write(&md->lock); - return -EINTR; - } + error = -EINTR; + if (atomic_read(&md->pending)) + goto out_unfreeze; set_bit(DMF_SUSPENDED, &md->flags); @@ -1115,6 +1119,18 @@ int dm_suspend(struct mapped_device *md) up_write(&md->lock); return 0; + +out_unfreeze: + /* FIXME Undo dm_table_presuspend_targets */ + __unlock_fs(md); + clear_bit(DMF_BLOCK_IO, &md->flags); +out_write_unlock: + up_write(&md->lock); + return error; + +out_read_unlock: + up_read(&md->lock); + return error; } int dm_resume(struct mapped_device *md) diff --git a/drivers/md/md.c b/drivers/md/md.c index 97af857d8a8..d899204d374 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -957,7 +957,7 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev) } -struct super_type super_types[] = { +static struct super_type super_types[] = { [0] = { .name = "0.90.0", .owner = THIS_MODULE, @@ -2740,7 +2740,7 @@ static struct block_device_operations md_fops = .revalidate_disk= md_revalidate, }; -int md_thread(void * arg) +static int md_thread(void * arg) { mdk_thread_t *thread = arg; @@ -3232,7 +3232,7 @@ void md_handle_safemode(mddev_t *mddev) } -DECLARE_WAIT_QUEUE_HEAD(resync_wait); +static DECLARE_WAIT_QUEUE_HEAD(resync_wait); #define SYNC_MARKS 10 #define SYNC_MARK_STEP (3*HZ) @@ -3575,8 +3575,8 @@ void md_check_recovery(mddev_t *mddev) } } -int md_notify_reboot(struct notifier_block *this, - unsigned long code, void *x) +static int md_notify_reboot(struct notifier_block *this, + unsigned long code, void *x) { struct list_head *tmp; mddev_t *mddev; @@ -3599,7 +3599,7 @@ int md_notify_reboot(struct notifier_block *this, return NOTIFY_DONE; } -struct notifier_block md_notifier = { +static struct notifier_block md_notifier = { .notifier_call = md_notify_reboot, .next = NULL, .priority = INT_MAX, /* before any real devices */ @@ -3616,7 +3616,7 @@ static void md_geninit(void) p->proc_fops = &md_seq_fops; } -int __init md_init(void) +static int __init md_init(void) { int minor; diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index c9b134cd153..4e4bfde3db5 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -103,7 +103,8 @@ static void multipath_end_bh_io (struct multipath_bh *mp_bh, int err) mempool_free(mp_bh, conf->pool); } -int multipath_end_request(struct bio *bio, unsigned int bytes_done, int error) +static int multipath_end_request(struct bio *bio, unsigned int bytes_done, + int error) { int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); struct multipath_bh * mp_bh = (struct multipath_bh *)(bio->bi_private); @@ -355,7 +356,7 @@ static int multipath_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index a389394b52f..83380b5d659 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -797,7 +797,7 @@ static int raid1_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index b100bfe4fdc..e9dc2876a62 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -977,7 +977,7 @@ static int raid10_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 52c3a81c4aa..e96e2a10a9c 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -1873,7 +1873,7 @@ static int raid5_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/md/raid6main.c b/drivers/md/raid6main.c index 7e30ab29691..8a33f351e09 100644 --- a/drivers/md/raid6main.c +++ b/drivers/md/raid6main.c @@ -2038,7 +2038,7 @@ static int raid6_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/media/common/ir-common.c b/drivers/media/common/ir-common.c index 8c842e2f59a..84a49d2ec91 100644 --- a/drivers/media/common/ir-common.c +++ b/drivers/media/common/ir-common.c @@ -131,10 +131,10 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = { [ 18 ] = KEY_KP0, [ 0 ] = KEY_POWER, -// [ 27 ] = MTS button + [ 27 ] = KEY_LANGUAGE, //MTS button [ 2 ] = KEY_TUNER, // TV/FM [ 30 ] = KEY_VIDEO, -// [ 22 ] = display button + [ 22 ] = KEY_INFO, //display button [ 4 ] = KEY_VOLUMEUP, [ 8 ] = KEY_VOLUMEDOWN, [ 12 ] = KEY_CHANNELUP, @@ -142,7 +142,7 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = { [ 3 ] = KEY_ZOOM, // fullscreen [ 31 ] = KEY_SUBTITLE, // closed caption/teletext [ 32 ] = KEY_SLEEP, -// [ 41 ] = boss key + [ 41 ] = KEY_SEARCH, //boss key [ 20 ] = KEY_MUTE, [ 43 ] = KEY_RED, [ 44 ] = KEY_GREEN, @@ -150,17 +150,17 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = { [ 46 ] = KEY_BLUE, [ 24 ] = KEY_KPPLUS, //fine tune + [ 25 ] = KEY_KPMINUS, //fine tune - -// [ 42 ] = picture in picture - [ 33 ] = KEY_KPDOT, + [ 42 ] = KEY_ANGLE, //picture in picture + [ 33 ] = KEY_KPDOT, [ 19 ] = KEY_KPENTER, -// [ 17 ] = recall + [ 17 ] = KEY_AGAIN, //recall [ 34 ] = KEY_BACK, [ 35 ] = KEY_PLAYPAUSE, [ 36 ] = KEY_NEXT, -// [ 37 ] = time shifting + [ 37 ] = KEY_T, //time shifting [ 38 ] = KEY_STOP, - [ 39 ] = KEY_RECORD -// [ 40 ] = snapshot + [ 39 ] = KEY_RECORD, + [ 40 ] = KEY_SHUFFLE //snapshot }; EXPORT_SYMBOL_GPL(ir_codes_winfast); diff --git a/drivers/media/dvb/frontends/cx22702.c b/drivers/media/dvb/frontends/cx22702.c index 1930b513eef..011860ce36c 100644 --- a/drivers/media/dvb/frontends/cx22702.c +++ b/drivers/media/dvb/frontends/cx22702.c @@ -32,6 +32,7 @@ #include <linux/slab.h> #include <linux/delay.h> #include "dvb_frontend.h" +#include "dvb-pll.h" #include "cx22702.h" @@ -203,7 +204,19 @@ static int cx22702_set_tps (struct dvb_frontend* fe, struct dvb_frontend_paramet /* set PLL */ cx22702_writereg (state, 0x0D, cx22702_readreg(state,0x0D) &0xfe); - state->config->pll_set(fe, p); + if (state->config->pll_set) { + state->config->pll_set(fe, p); + } else if (state->config->pll_desc) { + u8 pllbuf[4]; + struct i2c_msg msg = { .addr = state->config->pll_address, + .buf = pllbuf, .len = 4 }; + dvb_pll_configure(state->config->pll_desc, pllbuf, + p->frequency, + p->u.ofdm.bandwidth); + i2c_transfer(state->i2c, &msg, 1); + } else { + BUG(); + } cx22702_writereg (state, 0x0D, cx22702_readreg(state,0x0D) | 1); /* set inversion */ diff --git a/drivers/media/dvb/frontends/cx22702.h b/drivers/media/dvb/frontends/cx22702.h index 6e34f997aba..559fdb90666 100644 --- a/drivers/media/dvb/frontends/cx22702.h +++ b/drivers/media/dvb/frontends/cx22702.h @@ -36,6 +36,9 @@ struct cx22702_config u8 demod_address; /* PLL maintenance */ + u8 pll_address; + struct dvb_pll_desc *pll_desc; + int (*pll_init)(struct dvb_frontend* fe); int (*pll_set)(struct dvb_frontend* fe, struct dvb_frontend_parameters* params); }; diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig index c1b3542dad8..d3dd4228b72 100644 --- a/drivers/media/video/Kconfig +++ b/drivers/media/video/Kconfig @@ -252,6 +252,7 @@ config VIDEO_SAA7134_DVB depends on VIDEO_SAA7134 && DVB_CORE select VIDEO_BUF_DVB select DVB_MT352 + select DVB_CX22702 ---help--- This adds support for DVB cards based on the Philips saa7134 chip. diff --git a/drivers/media/video/bttv-driver.c b/drivers/media/video/bttv-driver.c index c13f222fe6b..033cc5498f2 100644 --- a/drivers/media/video/bttv-driver.c +++ b/drivers/media/video/bttv-driver.c @@ -3169,7 +3169,7 @@ static struct video_device radio_template = /* ----------------------------------------------------------------------- */ /* some debug code */ -int bttv_risc_decode(u32 risc) +static int bttv_risc_decode(u32 risc) { static char *instr[16] = { [ BT848_RISC_WRITE >> 28 ] = "write", @@ -3206,8 +3206,8 @@ int bttv_risc_decode(u32 risc) return incr[risc >> 28] ? incr[risc >> 28] : 1; } -void bttv_risc_disasm(struct bttv *btv, - struct btcx_riscmem *risc) +static void bttv_risc_disasm(struct bttv *btv, + struct btcx_riscmem *risc) { unsigned int i,j,n; diff --git a/drivers/media/video/bttv-i2c.c b/drivers/media/video/bttv-i2c.c index e42f1ec13f3..e3f477dff82 100644 --- a/drivers/media/video/bttv-i2c.c +++ b/drivers/media/video/bttv-i2c.c @@ -29,6 +29,7 @@ #include <linux/moduleparam.h> #include <linux/init.h> #include <linux/delay.h> +#include <linux/jiffies.h> #include <asm/io.h> #include "bttvp.h" @@ -130,17 +131,14 @@ static u32 functionality(struct i2c_adapter *adap) static int bttv_i2c_wait_done(struct bttv *btv) { - DECLARE_WAITQUEUE(wait, current); int rc = 0; - add_wait_queue(&btv->i2c_queue, &wait); - if (0 == btv->i2c_done) - msleep_interruptible(20); - remove_wait_queue(&btv->i2c_queue, &wait); + /* timeout */ + if (wait_event_interruptible_timeout(btv->i2c_queue, + btv->i2c_done, msecs_to_jiffies(85)) == -ERESTARTSYS) + + rc = -EIO; - if (0 == btv->i2c_done) - /* timeout */ - rc = -EIO; if (btv->i2c_done & BT848_INT_RACK) rc = 1; btv->i2c_done = 0; diff --git a/drivers/media/video/cx88/cx88-core.c b/drivers/media/video/cx88/cx88-core.c index 26a6138015c..1ff79b5a883 100644 --- a/drivers/media/video/cx88/cx88-core.c +++ b/drivers/media/video/cx88/cx88-core.c @@ -429,7 +429,7 @@ int cx88_sram_channel_setup(struct cx88_core *core, /* ------------------------------------------------------------------ */ /* debug helper code */ -int cx88_risc_decode(u32 risc) +static int cx88_risc_decode(u32 risc) { static char *instr[16] = { [ RISC_SYNC >> 28 ] = "sync", @@ -542,7 +542,7 @@ void cx88_sram_channel_dump(struct cx88_core *core, core->name,cx_read(ch->cnt2_reg)); } -char *cx88_pci_irqs[32] = { +static char *cx88_pci_irqs[32] = { "vid", "aud", "ts", "vip", "hst", "5", "6", "tm1", "src_dma", "dst_dma", "risc_rd_err", "risc_wr_err", "brdg_err", "src_dma_err", "dst_dma_err", "ipb_dma_err", @@ -1206,7 +1206,6 @@ void cx88_core_put(struct cx88_core *core, struct pci_dev *pci) /* ------------------------------------------------------------------ */ EXPORT_SYMBOL(cx88_print_ioctl); -EXPORT_SYMBOL(cx88_pci_irqs); EXPORT_SYMBOL(cx88_vid_irqs); EXPORT_SYMBOL(cx88_mpeg_irqs); EXPORT_SYMBOL(cx88_print_irqbits); diff --git a/drivers/media/video/cx88/cx88-dvb.c b/drivers/media/video/cx88/cx88-dvb.c index bc6f18c4535..9d15d3d5a2b 100644 --- a/drivers/media/video/cx88/cx88-dvb.c +++ b/drivers/media/video/cx88/cx88-dvb.c @@ -31,7 +31,7 @@ #include <linux/suspend.h> /* those two frontends need merging via linuxtv cvs ... */ -#define HAVE_CX22702 0 +#define HAVE_CX22702 1 #define HAVE_OR51132 1 #include "cx88.h" @@ -91,7 +91,7 @@ static void dvb_buf_release(struct videobuf_queue *q, struct videobuf_buffer *vb cx88_free_buffer(dev->pci, (struct cx88_buffer*)vb); } -struct videobuf_queue_ops dvb_qops = { +static struct videobuf_queue_ops dvb_qops = { .buf_setup = dvb_buf_setup, .buf_prepare = dvb_buf_prepare, .buf_queue = dvb_buf_queue, @@ -191,7 +191,7 @@ static int or51132_set_ts_param(struct dvb_frontend* fe, return 0; } -struct or51132_config pchdtv_hd3000 = { +static struct or51132_config pchdtv_hd3000 = { .demod_address = 0x15, .pll_address = 0x61, .pll_desc = &dvb_pll_thomson_dtt7610, @@ -243,10 +243,8 @@ static int dvb_register(struct cx8802_dev *dev) break; #endif default: - printk("%s: The frontend of your DVB/ATSC card isn't supported yet\n" - "%s: you might want to look out for patches here:\n" - "%s: http://dl.bytesex.org/patches/\n", - dev->core->name, dev->core->name, dev->core->name); + printk("%s: The frontend of your DVB/ATSC card isn't supported yet\n", + dev->core->name); break; } if (NULL == dev->dvb.frontend) { @@ -308,9 +306,11 @@ static int __devinit dvb_probe(struct pci_dev *pci_dev, dev); err = dvb_register(dev); if (0 != err) - goto fail_free; + goto fail_fini; return 0; + fail_fini: + cx8802_fini_common(dev); fail_free: kfree(dev); fail_core: diff --git a/drivers/media/video/cx88/cx88-i2c.c b/drivers/media/video/cx88/cx88-i2c.c index 60800172c02..0725b1288f4 100644 --- a/drivers/media/video/cx88/cx88-i2c.c +++ b/drivers/media/video/cx88/cx88-i2c.c @@ -45,7 +45,7 @@ MODULE_PARM_DESC(i2c_scan,"scan i2c bus at insmod time"); /* ----------------------------------------------------------------------- */ -void cx8800_bit_setscl(void *data, int state) +static void cx8800_bit_setscl(void *data, int state) { struct cx88_core *core = data; @@ -57,7 +57,7 @@ void cx8800_bit_setscl(void *data, int state) cx_read(MO_I2C); } -void cx8800_bit_setsda(void *data, int state) +static void cx8800_bit_setsda(void *data, int state) { struct cx88_core *core = data; diff --git a/drivers/media/video/cx88/cx88-vbi.c b/drivers/media/video/cx88/cx88-vbi.c index 471e508b074..0584ff47638 100644 --- a/drivers/media/video/cx88/cx88-vbi.c +++ b/drivers/media/video/cx88/cx88-vbi.c @@ -46,9 +46,9 @@ void cx8800_vbi_fmt(struct cx8800_dev *dev, struct v4l2_format *f) } } -int cx8800_start_vbi_dma(struct cx8800_dev *dev, - struct cx88_dmaqueue *q, - struct cx88_buffer *buf) +static int cx8800_start_vbi_dma(struct cx8800_dev *dev, + struct cx88_dmaqueue *q, + struct cx88_buffer *buf) { struct cx88_core *core = dev->core; diff --git a/drivers/media/video/cx88/cx88-video.c b/drivers/media/video/cx88/cx88-video.c index 701f594e181..d1f5c92f0ce 100644 --- a/drivers/media/video/cx88/cx88-video.c +++ b/drivers/media/video/cx88/cx88-video.c @@ -325,7 +325,7 @@ static struct cx88_ctrl cx8800_ctls[] = { .shift = 0, } }; -const int CX8800_CTLS = ARRAY_SIZE(cx8800_ctls); +static const int CX8800_CTLS = ARRAY_SIZE(cx8800_ctls); /* ------------------------------------------------------------------- */ /* resource management */ @@ -665,7 +665,7 @@ static void buffer_release(struct videobuf_queue *q, struct videobuf_buffer *vb) cx88_free_buffer(fh->dev->pci,buf); } -struct videobuf_queue_ops cx8800_video_qops = { +static struct videobuf_queue_ops cx8800_video_qops = { .buf_setup = buffer_setup, .buf_prepare = buffer_prepare, .buf_queue = buffer_queue, @@ -1924,7 +1924,7 @@ static struct file_operations video_fops = .llseek = no_llseek, }; -struct video_device cx8800_video_template = +static struct video_device cx8800_video_template = { .name = "cx8800-video", .type = VID_TYPE_CAPTURE|VID_TYPE_TUNER|VID_TYPE_SCALES, @@ -1933,7 +1933,7 @@ struct video_device cx8800_video_template = .minor = -1, }; -struct video_device cx8800_vbi_template = +static struct video_device cx8800_vbi_template = { .name = "cx8800-vbi", .type = VID_TYPE_TELETEXT|VID_TYPE_TUNER, @@ -1951,7 +1951,7 @@ static struct file_operations radio_fops = .llseek = no_llseek, }; -struct video_device cx8800_radio_template = +static struct video_device cx8800_radio_template = { .name = "cx8800-radio", .type = VID_TYPE_TUNER, @@ -2226,7 +2226,7 @@ static int cx8800_resume(struct pci_dev *pci_dev) /* ----------------------------------------------------------- */ -struct pci_device_id cx8800_pci_tbl[] = { +static struct pci_device_id cx8800_pci_tbl[] = { { .vendor = 0x14f1, .device = 0x8800, diff --git a/drivers/media/video/cx88/cx88.h b/drivers/media/video/cx88/cx88.h index b351d9eae61..88eaaaba5ad 100644 --- a/drivers/media/video/cx88/cx88.h +++ b/drivers/media/video/cx88/cx88.h @@ -420,7 +420,6 @@ struct cx8802_dev { /* ----------------------------------------------------------- */ /* cx88-core.c */ -extern char *cx88_pci_irqs[32]; extern char *cx88_vid_irqs[32]; extern char *cx88_mpeg_irqs[32]; extern void cx88_print_irqbits(char *name, char *tag, char **strings, @@ -472,9 +471,6 @@ extern void cx88_core_put(struct cx88_core *core, /* cx88-vbi.c */ void cx8800_vbi_fmt(struct cx8800_dev *dev, struct v4l2_format *f); -int cx8800_start_vbi_dma(struct cx8800_dev *dev, - struct cx88_dmaqueue *q, - struct cx88_buffer *buf); int cx8800_stop_vbi_dma(struct cx8800_dev *dev); int cx8800_restart_vbi_queue(struct cx8800_dev *dev, struct cx88_dmaqueue *q); diff --git a/drivers/media/video/msp3400.c b/drivers/media/video/msp3400.c index c97df705df5..7fbb8581a87 100644 --- a/drivers/media/video/msp3400.c +++ b/drivers/media/video/msp3400.c @@ -380,7 +380,9 @@ static void msp3400c_setvolume(struct i2c_client *client, int val = 0, bal = 0; if (!muted) { - val = (volume * 0x7F / 65535) << 8; + /* 0x7f instead if 0x73 here has sound quality issues, + * probably due to overmodulation + clipping ... */ + val = (volume * 0x73 / 65535) << 8; } if (val) { bal = (balance / 256) - 128; @@ -997,7 +999,13 @@ static int msp34xx_modus(int norm) { switch (norm) { case VIDEO_MODE_PAL: +#if 1 + /* experimental: not sure this works with all chip versions */ + return 0x7003; +#else + /* previous value, try this if it breaks ... */ return 0x1003; +#endif case VIDEO_MODE_NTSC: /* BTSC */ return 0x2003; case VIDEO_MODE_SECAM: @@ -1264,6 +1272,7 @@ static int msp34xxg_thread(void *data) int val, std, i; printk("msp34xxg: daemon started\n"); + msp->source = 1; /* default */ for (;;) { d2printk(KERN_DEBUG "msp34xxg: thread: sleep\n"); msp34xx_sleep(msp,-1); @@ -1334,8 +1343,9 @@ static void msp34xxg_set_source(struct i2c_client *client, int source) /* fix matrix mode to stereo and let the msp choose what * to output according to 'source', as recommended + * for MONO (source==0) downmixing set bit[7:0] to 0x30 */ - int value = (source&0x07)<<8|(source==0 ? 0x00:0x20); + int value = (source&0x07)<<8|(source==0 ? 0x30:0x20); dprintk("msp34xxg: set source to %d (0x%x)\n", source, value); msp3400c_write(client, I2C_MSP3400C_DFP, @@ -1359,7 +1369,7 @@ static void msp34xxg_set_source(struct i2c_client *client, int source) msp3400c_write(client, I2C_MSP3400C_DEM, 0x22, /* a2 threshold for stereo/bilingual */ - source==0 ? 0x7f0:stereo_threshold); + stereo_threshold); msp->source=source; } @@ -1394,7 +1404,7 @@ static void msp34xxg_detect_stereo(struct i2c_client *client) static void msp34xxg_set_audmode(struct i2c_client *client, int audmode) { struct msp3400c *msp = i2c_get_clientdata(client); - int source = 0; + int source; switch (audmode) { case V4L2_TUNER_MODE_MONO: @@ -1410,9 +1420,10 @@ static void msp34xxg_set_audmode(struct i2c_client *client, int audmode) case V4L2_TUNER_MODE_LANG2: source=4; /* stereo or B */ break; - default: /* doing nothing: a safe, sane default */ + default: audmode = 0; - return; + source = 1; + break; } msp->audmode = audmode; msp34xxg_set_source(client, source); @@ -1514,12 +1525,9 @@ static int msp_attach(struct i2c_adapter *adap, int addr, int kind) msp->opmode = opmode; if (OPMODE_AUTO == msp->opmode) { -#if 0 /* seems to work for ivtv only, disable by default for now ... */ if (HAVE_SIMPLER(msp)) msp->opmode = OPMODE_SIMPLER; - else -#endif - if (HAVE_SIMPLE(msp)) + else if (HAVE_SIMPLE(msp)) msp->opmode = OPMODE_SIMPLE; else msp->opmode = OPMODE_MANUAL; diff --git a/drivers/media/video/saa7134/saa6752hs.c b/drivers/media/video/saa7134/saa6752hs.c index cee13584c9c..fe6abe34168 100644 --- a/drivers/media/video/saa7134/saa6752hs.c +++ b/drivers/media/video/saa7134/saa6752hs.c @@ -32,9 +32,32 @@ MODULE_LICENSE("GPL"); static struct i2c_driver driver; static struct i2c_client client_template; +enum saa6752hs_videoformat { + SAA6752HS_VF_D1 = 0, /* standard D1 video format: 720x576 */ + SAA6752HS_VF_2_3_D1 = 1,/* 2/3D1 video format: 480x576 */ + SAA6752HS_VF_1_2_D1 = 2,/* 1/2D1 video format: 352x576 */ + SAA6752HS_VF_SIF = 3, /* SIF video format: 352x288 */ + SAA6752HS_VF_UNKNOWN, +}; + +static const struct v4l2_format v4l2_format_table[] = +{ + [SAA6752HS_VF_D1] = { + .fmt.pix.width = 720, .fmt.pix.height = 576 }, + [SAA6752HS_VF_2_3_D1] = { + .fmt.pix.width = 480, .fmt.pix.height = 576 }, + [SAA6752HS_VF_1_2_D1] = { + .fmt.pix.width = 352, .fmt.pix.height = 576 }, + [SAA6752HS_VF_SIF] = { + .fmt.pix.width = 352, .fmt.pix.height = 288 }, + [SAA6752HS_VF_UNKNOWN] = { + .fmt.pix.width = 0, .fmt.pix.height = 0}, +}; + struct saa6752hs_state { struct i2c_client client; struct v4l2_mpeg_compression params; + enum saa6752hs_videoformat video_format; }; enum saa6752hs_command { @@ -256,6 +279,51 @@ static int saa6752hs_set_bitrate(struct i2c_client* client, return 0; } +static void saa6752hs_set_subsampling(struct i2c_client* client, + struct v4l2_format* f) +{ + struct saa6752hs_state *h = i2c_get_clientdata(client); + int dist_352, dist_480, dist_720; + + /* + FIXME: translate and round width/height into EMPRESS + subsample type: + + type | PAL | NTSC + --------------------------- + SIF | 352x288 | 352x240 + 1/2 D1 | 352x576 | 352x480 + 2/3 D1 | 480x576 | 480x480 + D1 | 720x576 | 720x480 + */ + + dist_352 = abs(f->fmt.pix.width - 352); + dist_480 = abs(f->fmt.pix.width - 480); + dist_720 = abs(f->fmt.pix.width - 720); + if (dist_720 < dist_480) { + f->fmt.pix.width = 720; + f->fmt.pix.height = 576; + h->video_format = SAA6752HS_VF_D1; + } + else if (dist_480 < dist_352) { + f->fmt.pix.width = 480; + f->fmt.pix.height = 576; + h->video_format = SAA6752HS_VF_2_3_D1; + } + else { + f->fmt.pix.width = 352; + if (abs(f->fmt.pix.height - 576) < + abs(f->fmt.pix.height - 288)) { + f->fmt.pix.height = 576; + h->video_format = SAA6752HS_VF_1_2_D1; + } + else { + f->fmt.pix.height = 288; + h->video_format = SAA6752HS_VF_SIF; + } + } +} + static void saa6752hs_set_params(struct i2c_client* client, struct v4l2_mpeg_compression* params) @@ -315,7 +383,7 @@ static int saa6752hs_init(struct i2c_client* client) // Set video format - must be done first as it resets other settings buf[0] = 0x41; - buf[1] = 0 /* MPEG_VIDEO_FORMAT_D1 */; + buf[1] = h->video_format; i2c_master_send(client, buf, 2); // set bitrate @@ -494,6 +562,25 @@ saa6752hs_command(struct i2c_client *client, unsigned int cmd, void *arg) case VIDIOC_G_MPEGCOMP: *params = h->params; break; + case VIDIOC_G_FMT: + { + struct v4l2_format *f = arg; + + if (h->video_format == SAA6752HS_VF_UNKNOWN) + h->video_format = SAA6752HS_VF_D1; + f->fmt.pix.width = + v4l2_format_table[h->video_format].fmt.pix.width; + f->fmt.pix.height = + v4l2_format_table[h->video_format].fmt.pix.height; + break ; + } + case VIDIOC_S_FMT: + { + struct v4l2_format *f = arg; + + saa6752hs_set_subsampling(client, f); + break; + } default: /* nothing */ break; diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index 180d3175ea5..c51eb7f078d 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -183,12 +183,12 @@ struct saa7134_board saa7134_boards[] = { .name = "LifeView FlyTV Platinum FM", .audio_clock = 0x00200000, .tuner_type = TUNER_PHILIPS_TDA8290, -// .gpiomask = 0xe000, + .gpiomask = 0x1E000, /* Set GP16 and unused 15,14,13 to Output */ .inputs = {{ .name = name_tv, .vmux = 1, .amux = TV, -// .gpio = 0x0000, + .gpio = 0x10000, /* GP16=1 selects TV input */ .tv = 1, },{ /* .name = name_tv_mono, @@ -212,12 +212,12 @@ struct saa7134_board saa7134_boards[] = { .amux = LINE2, // .gpio = 0x4000, }}, -/* .radio = { + .radio = { .name = name_radio, - .amux = LINE2, - .gpio = 0x2000, + .amux = TV, + .gpio = 0x00000, /* GP16=0 selects FM radio antenna */ }, -*/ }, + }, [SAA7134_BOARD_EMPRESS] = { /* "Gert Vervoort" <gert.vervoort@philips.com> */ .name = "EMPRESS", @@ -1628,11 +1628,17 @@ struct pci_device_id saa7134_pci_tbl[] = { },{ .vendor = PCI_VENDOR_ID_PHILIPS, .device = PCI_DEVICE_ID_PHILIPS_SAA7133, - .subvendor = 0x5168, + .subvendor = 0x5168, /* Animation Technologies (LifeView) */ .subdevice = 0x0214, /* Standard PCI, LR214WF */ .driver_data = SAA7134_BOARD_FLYTVPLATINUM_FM, },{ .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x1489, /* KYE */ + .subdevice = 0x0214, /* Genius VideoWonder ProTV */ + .driver_data = SAA7134_BOARD_FLYTVPLATINUM_FM, /* is an LR214WF actually */ + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, .device = PCI_DEVICE_ID_PHILIPS_SAA7134, .subvendor = 0x16be, .subdevice = 0x0003, @@ -1948,6 +1954,7 @@ int saa7134_board_init1(struct saa7134_dev *dev) dev->has_remote = 1; board_flyvideo(dev); break; + case SAA7134_BOARD_FLYTVPLATINUM_FM: case SAA7134_BOARD_CINERGY400: case SAA7134_BOARD_CINERGY600: case SAA7134_BOARD_CINERGY600_MK3: diff --git a/drivers/media/video/saa7134/saa7134-dvb.c b/drivers/media/video/saa7134/saa7134-dvb.c index dd4a6c8ee65..c2873ae029f 100644 --- a/drivers/media/video/saa7134/saa7134-dvb.c +++ b/drivers/media/video/saa7134/saa7134-dvb.c @@ -172,7 +172,7 @@ static int fe_request_firmware(struct dvb_frontend* fe, return request_firmware(fw, name, &dev->pci->dev); } -struct tda1004x_config medion_cardbus = { +static struct tda1004x_config medion_cardbus = { .demod_address = 0x08, /* not sure this is correct */ .invert = 0, .invert_oclk = 0, diff --git a/drivers/media/video/saa7134/saa7134-empress.c b/drivers/media/video/saa7134/saa7134-empress.c index 2021e099e35..fa135733690 100644 --- a/drivers/media/video/saa7134/saa7134-empress.c +++ b/drivers/media/video/saa7134/saa7134-empress.c @@ -233,10 +233,7 @@ static int ts_do_ioctl(struct inode *inode, struct file *file, memset(f,0,sizeof(*f)); f->type = V4L2_BUF_TYPE_VIDEO_CAPTURE; - /* FIXME: translate subsampling type EMPRESS into - * width/height: */ - f->fmt.pix.width = 720; /* D1 */ - f->fmt.pix.height = 576; + saa7134_i2c_call_clients(dev, cmd, arg); f->fmt.pix.pixelformat = V4L2_PIX_FMT_MPEG; f->fmt.pix.sizeimage = TS_PACKET_SIZE * dev->ts.nr_packets; return 0; @@ -249,20 +246,7 @@ static int ts_do_ioctl(struct inode *inode, struct file *file, if (f->type != V4L2_BUF_TYPE_VIDEO_CAPTURE) return -EINVAL; - /* - FIXME: translate and round width/height into EMPRESS - subsample type: - - type | PAL | NTSC - --------------------------- - SIF | 352x288 | 352x240 - 1/2 D1 | 352x576 | 352x480 - 2/3 D1 | 480x576 | 480x480 - D1 | 720x576 | 720x480 - */ - - f->fmt.pix.width = 720; /* D1 */ - f->fmt.pix.height = 576; + saa7134_i2c_call_clients(dev, cmd, arg); f->fmt.pix.pixelformat = V4L2_PIX_FMT_MPEG; f->fmt.pix.sizeimage = TS_PACKET_SIZE* dev->ts.nr_packets; return 0; diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c index 727d437e07d..ca50cf531f2 100644 --- a/drivers/media/video/saa7134/saa7134-input.c +++ b/drivers/media/video/saa7134/saa7134-input.c @@ -379,6 +379,7 @@ int saa7134_input_init1(struct saa7134_dev *dev) switch (dev->board) { case SAA7134_BOARD_FLYVIDEO2000: case SAA7134_BOARD_FLYVIDEO3000: + case SAA7134_BOARD_FLYTVPLATINUM_FM: ir_codes = flyvideo_codes; mask_keycode = 0xEC00000; mask_keydown = 0x0040000; diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c index 065eb4007b1..80dc34f18c2 100644 --- a/drivers/media/video/tvaudio.c +++ b/drivers/media/video/tvaudio.c @@ -991,7 +991,7 @@ static int tda9874a_initialize(struct CHIPSTATE *chip) { if (tda9874a_SIF > 2) tda9874a_SIF = 1; - if (tda9874a_STD >= 8) + if (tda9874a_STD > 8) tda9874a_STD = 0; if(tda9874a_AMSEL > 1) tda9874a_AMSEL = 0; diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig index 72f2b466b81..2e70d74fbde 100644 --- a/drivers/mmc/Kconfig +++ b/drivers/mmc/Kconfig @@ -51,7 +51,7 @@ config MMC_PXA config MMC_WBSD tristate "Winbond W83L51xD SD/MMC Card Interface support" - depends on MMC && ISA + depends on MMC && ISA && ISA_DMA_API help This selects the Winbond(R) W83L51xD Secure digital and Multimedia card Interface. diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c index 43e2ac532f8..b5e07604343 100644 --- a/drivers/net/3c59x.c +++ b/drivers/net/3c59x.c @@ -1581,7 +1581,8 @@ vortex_up(struct net_device *dev) if (VORTEX_PCI(vp)) { pci_set_power_state(VORTEX_PCI(vp), PCI_D0); /* Go active */ - pci_restore_state(VORTEX_PCI(vp)); + if (vp->pm_state_valid) + pci_restore_state(VORTEX_PCI(vp)); pci_enable_device(VORTEX_PCI(vp)); } @@ -2741,6 +2742,7 @@ vortex_down(struct net_device *dev, int final_down) outl(0, ioaddr + DownListPtr); if (final_down && VORTEX_PCI(vp)) { + vp->pm_state_valid = 1; pci_save_state(VORTEX_PCI(vp)); acpi_set_WOL(dev); } @@ -3243,9 +3245,10 @@ static void acpi_set_WOL(struct net_device *dev) outw(RxEnable, ioaddr + EL3_CMD); pci_enable_wake(VORTEX_PCI(vp), 0, 1); + + /* Change the power state to D3; RxEnable doesn't take effect. */ + pci_set_power_state(VORTEX_PCI(vp), PCI_D3hot); } - /* Change the power state to D3; RxEnable doesn't take effect. */ - pci_set_power_state(VORTEX_PCI(vp), PCI_D3hot); } diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 68242bda4b9..3a0a55b62aa 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -589,7 +589,7 @@ config EL2 config ELPLUS tristate "3c505 \"EtherLink Plus\" support" - depends on NET_VENDOR_3COM && ISA + depends on NET_VENDOR_3COM && ISA && ISA_DMA_API ---help--- Information about this network (Ethernet) card can be found in <file:Documentation/networking/3c505.txt>. If you have a card of @@ -630,7 +630,7 @@ config EL3 config 3C515 tristate "3c515 ISA \"Fast EtherLink\"" - depends on NET_VENDOR_3COM && (ISA || EISA) + depends on NET_VENDOR_3COM && (ISA || EISA) && ISA_DMA_API help If you have a 3Com ISA EtherLink XL "Corkscrew" 3c515 Fast Ethernet network card, say Y and read the Ethernet-HOWTO, available from @@ -708,7 +708,7 @@ config TYPHOON config LANCE tristate "AMD LANCE and PCnet (AT1500 and NE2100) support" - depends on NET_ETHERNET && ISA + depends on NET_ETHERNET && ISA && ISA_DMA_API help If you have a network (Ethernet) card of this type, say Y and read the Ethernet-HOWTO, available from @@ -864,7 +864,7 @@ config NI52 config NI65 tristate "NI6510 support" - depends on NET_VENDOR_RACAL && ISA + depends on NET_VENDOR_RACAL && ISA && ISA_DMA_API help If you have a network (Ethernet) card of this type, say Y and read the Ethernet-HOWTO, available from @@ -1072,7 +1072,7 @@ config NE2000 config ZNET tristate "Zenith Z-Note support (EXPERIMENTAL)" - depends on NET_ISA && EXPERIMENTAL + depends on NET_ISA && EXPERIMENTAL && ISA_DMA_API help The Zenith Z-Note notebook computer has a built-in network (Ethernet) card, and this is the Linux driver for it. Note that the diff --git a/drivers/net/Space.c b/drivers/net/Space.c index fc519377b5a..fb433325aa2 100644 --- a/drivers/net/Space.c +++ b/drivers/net/Space.c @@ -7,7 +7,7 @@ * * Version: @(#)Space.c 1.0.7 08/12/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Donald J. Becker, <becker@scyld.com> * diff --git a/drivers/net/appletalk/Kconfig b/drivers/net/appletalk/Kconfig index 60b19679ca5..69c488d933a 100644 --- a/drivers/net/appletalk/Kconfig +++ b/drivers/net/appletalk/Kconfig @@ -13,7 +13,7 @@ config DEV_APPLETALK config LTPC tristate "Apple/Farallon LocalTalk PC support" - depends on DEV_APPLETALK && (ISA || EISA) + depends on DEV_APPLETALK && (ISA || EISA) && ISA_DMA_API help This allows you to use the AppleTalk PC card to connect to LocalTalk networks. The card is also known as the Farallon PhoneNet PC card. diff --git a/drivers/net/appletalk/cops.c b/drivers/net/appletalk/cops.c index 2161c2d585f..9edaa183227 100644 --- a/drivers/net/appletalk/cops.c +++ b/drivers/net/appletalk/cops.c @@ -65,7 +65,7 @@ static const char *version = #include <linux/etherdevice.h> #include <linux/skbuff.h> #include <linux/if_arp.h> -#include <linux/if_ltalk.h> /* For ltalk_setup() */ +#include <linux/if_ltalk.h> #include <linux/delay.h> /* For udelay() */ #include <linux/atalk.h> #include <linux/spinlock.h> @@ -223,7 +223,7 @@ struct net_device * __init cops_probe(int unit) int base_addr; int err = 0; - dev = alloc_netdev(sizeof(struct cops_local), "lt%d", ltalk_setup); + dev = alloc_ltalkdev(sizeof(struct cops_local)); if (!dev) return ERR_PTR(-ENOMEM); diff --git a/drivers/net/appletalk/cops_ffdrv.h b/drivers/net/appletalk/cops_ffdrv.h index 4131b4a7a65..31cf8c9c947 100644 --- a/drivers/net/appletalk/cops_ffdrv.h +++ b/drivers/net/appletalk/cops_ffdrv.h @@ -28,7 +28,7 @@ #ifdef CONFIG_COPS_DAYNA -unsigned char ffdrv_code[] = { +static const unsigned char ffdrv_code[] = { 58,3,0,50,228,149,33,255,255,34,226,149, 249,17,40,152,33,202,154,183,237,82,77,68, 11,107,98,19,54,0,237,176,175,50,80,0, diff --git a/drivers/net/appletalk/cops_ltdrv.h b/drivers/net/appletalk/cops_ltdrv.h index 05de66dd920..4afb8e18ba6 100644 --- a/drivers/net/appletalk/cops_ltdrv.h +++ b/drivers/net/appletalk/cops_ltdrv.h @@ -27,7 +27,7 @@ #ifdef CONFIG_COPS_TANGENT -unsigned char ltdrv_code[] = { +static const unsigned char ltdrv_code[] = { 58,3,0,50,148,10,33,143,15,62,85,119, 190,32,9,62,170,119,190,32,3,35,24,241, 34,146,10,249,17,150,10,33,143,15,183,237, diff --git a/drivers/net/appletalk/ltpc.c b/drivers/net/appletalk/ltpc.c index ad8e943231a..db4f369637b 100644 --- a/drivers/net/appletalk/ltpc.c +++ b/drivers/net/appletalk/ltpc.c @@ -1039,7 +1039,7 @@ struct net_device * __init ltpc_probe(void) unsigned long f; unsigned long timeout; - dev = alloc_netdev(sizeof(struct ltpc_private), "lt%d", ltalk_setup); + dev = alloc_ltalkdev(sizeof(struct ltpc_private)); if (!dev) goto out; diff --git a/drivers/net/arcnet/capmode.c b/drivers/net/arcnet/capmode.c index 16e155b0412..66485585ab3 100644 --- a/drivers/net/arcnet/capmode.c +++ b/drivers/net/arcnet/capmode.c @@ -48,7 +48,7 @@ static int prepare_tx(struct net_device *dev, struct archdr *pkt, int length, static int ack_tx(struct net_device *dev, int acked); -struct ArcProto capmode_proto = +static struct ArcProto capmode_proto = { 'r', XMTU, diff --git a/drivers/net/gt96100eth.h b/drivers/net/gt96100eth.h index 2f4bfd4dacb..395869c5ed3 100644 --- a/drivers/net/gt96100eth.h +++ b/drivers/net/gt96100eth.h @@ -214,7 +214,7 @@ typedef struct { u32 cmdstat; u32 next; u32 buff_ptr; -} gt96100_td_t __attribute__ ((packed)); +} __attribute__ ((packed)) gt96100_td_t; typedef struct { #ifdef DESC_BE @@ -227,7 +227,7 @@ typedef struct { u32 cmdstat; u32 next; u32 buff_ptr; -} gt96100_rd_t __attribute__ ((packed)); +} __attribute__ ((packed)) gt96100_rd_t; /* Values for the Tx command-status descriptor entry. */ diff --git a/drivers/net/hamradio/6pack.c b/drivers/net/hamradio/6pack.c index 067b353e1cb..89454915b85 100644 --- a/drivers/net/hamradio/6pack.c +++ b/drivers/net/hamradio/6pack.c @@ -394,13 +394,11 @@ static void sp_bump(struct sixpack *sp, char cmd) if ((skb = dev_alloc_skb(count)) == NULL) goto out_mem; - skb->dev = sp->dev; ptr = skb_put(skb, count); *ptr++ = cmd; /* KISS command */ memcpy(ptr, sp->cooked_buf + 1, count); - skb->mac.raw = skb->data; - skb->protocol = htons(ETH_P_AX25); + skb->protocol = ax25_type_trans(skb, sp->dev); netif_rx(skb); sp->dev->last_rx = jiffies; sp->stats.rx_packets++; diff --git a/drivers/net/hamradio/Kconfig b/drivers/net/hamradio/Kconfig index 34068f81d45..7cdebe1a0b6 100644 --- a/drivers/net/hamradio/Kconfig +++ b/drivers/net/hamradio/Kconfig @@ -45,7 +45,7 @@ config BPQETHER config DMASCC tristate "High-speed (DMA) SCC driver for AX.25" - depends on ISA && AX25 && BROKEN_ON_SMP + depends on ISA && AX25 && BROKEN_ON_SMP && ISA_DMA_API ---help--- This is a driver for high-speed SCC boards, i.e. those supporting DMA on one port. You usually use those boards to connect your @@ -78,7 +78,7 @@ config DMASCC config SCC tristate "Z8530 SCC driver" - depends on ISA && AX25 + depends on ISA && AX25 && ISA_DMA_API ---help--- These cards are used to connect your Linux box to an amateur radio in order to communicate with other computers. If you want to use diff --git a/drivers/net/hamradio/baycom_epp.c b/drivers/net/hamradio/baycom_epp.c index e8cb87d906f..1c563f905a5 100644 --- a/drivers/net/hamradio/baycom_epp.c +++ b/drivers/net/hamradio/baycom_epp.c @@ -601,12 +601,10 @@ static void do_rxpacket(struct net_device *dev) bc->stats.rx_dropped++; return; } - skb->dev = dev; cp = skb_put(skb, pktlen); *cp++ = 0; /* KISS kludge */ memcpy(cp, bc->hdlcrx.buf, pktlen - 1); - skb->protocol = htons(ETH_P_AX25); - skb->mac.raw = skb->data; + skb->protocol = ax25_type_trans(skb, dev); netif_rx(skb); dev->last_rx = jiffies; bc->stats.rx_packets++; diff --git a/drivers/net/hamradio/bpqether.c b/drivers/net/hamradio/bpqether.c index ef1a359e227..ba9f0580e1f 100644 --- a/drivers/net/hamradio/bpqether.c +++ b/drivers/net/hamradio/bpqether.c @@ -211,11 +211,7 @@ static int bpq_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_ty ptr = skb_push(skb, 1); *ptr = 0; - skb->dev = dev; - skb->protocol = htons(ETH_P_AX25); - skb->mac.raw = skb->data; - skb->pkt_type = PACKET_HOST; - + skb->protocol = ax25_type_trans(skb, dev); netif_rx(skb); dev->last_rx = jiffies; unlock: @@ -272,8 +268,6 @@ static int bpq_xmit(struct sk_buff *skb, struct net_device *dev) skb = newskb; } - skb->protocol = htons(ETH_P_AX25); - ptr = skb_push(skb, 2); *ptr++ = (size + 5) % 256; @@ -287,7 +281,7 @@ static int bpq_xmit(struct sk_buff *skb, struct net_device *dev) return -ENODEV; } - skb->dev = dev; + skb->protocol = ax25_type_trans(skb, dev); skb->nh.raw = skb->data; dev->hard_header(skb, dev, ETH_P_BPQ, bpq->dest_addr, NULL, 0); bpq->stats.tx_packets++; diff --git a/drivers/net/hamradio/dmascc.c b/drivers/net/hamradio/dmascc.c index f3269b70a8c..f515245a3fd 100644 --- a/drivers/net/hamradio/dmascc.c +++ b/drivers/net/hamradio/dmascc.c @@ -1306,9 +1306,7 @@ static void rx_bh(void *arg) data = skb_put(skb, cb + 1); data[0] = 0; memcpy(&data[1], priv->rx_buf[i], cb); - skb->dev = priv->dev; - skb->protocol = ntohs(ETH_P_AX25); - skb->mac.raw = skb->data; + skb->protocol = ax25_type_trans(skb, priv->dev); netif_rx(skb); priv->dev->last_rx = jiffies; priv->stats.rx_packets++; diff --git a/drivers/net/hamradio/hdlcdrv.c b/drivers/net/hamradio/hdlcdrv.c index b89959a596d..b4c836e4fe8 100644 --- a/drivers/net/hamradio/hdlcdrv.c +++ b/drivers/net/hamradio/hdlcdrv.c @@ -174,12 +174,10 @@ static void hdlc_rx_flag(struct net_device *dev, struct hdlcdrv_state *s) s->stats.rx_dropped++; return; } - skb->dev = dev; cp = skb_put(skb, pkt_len); *cp++ = 0; /* KISS kludge */ memcpy(cp, s->hdlcrx.buffer, pkt_len - 1); - skb->protocol = htons(ETH_P_AX25); - skb->mac.raw = skb->data; + skb->protocol = ax25_type_trans(skb, dev); netif_rx(skb); dev->last_rx = jiffies; s->stats.rx_packets++; diff --git a/drivers/net/hamradio/mkiss.c b/drivers/net/hamradio/mkiss.c index d9ea080aea0..62790511098 100644 --- a/drivers/net/hamradio/mkiss.c +++ b/drivers/net/hamradio/mkiss.c @@ -332,12 +332,10 @@ static void ax_bump(struct ax_disp *ax) return; } - skb->dev = ax->dev; spin_lock_bh(&ax->buflock); memcpy(skb_put(skb,count), ax->rbuff, count); spin_unlock_bh(&ax->buflock); - skb->mac.raw = skb->data; - skb->protocol = htons(ETH_P_AX25); + skb->protocol = ax25_type_trans(skb, ax->dev); netif_rx(skb); ax->dev->last_rx = jiffies; ax->rx_packets++; diff --git a/drivers/net/hamradio/scc.c b/drivers/net/hamradio/scc.c index ce9e7af020d..ece1b1a1318 100644 --- a/drivers/net/hamradio/scc.c +++ b/drivers/net/hamradio/scc.c @@ -1630,10 +1630,7 @@ static void scc_net_rx(struct scc_channel *scc, struct sk_buff *skb) scc->dev_stat.rx_packets++; scc->dev_stat.rx_bytes += skb->len; - skb->dev = scc->dev; - skb->protocol = htons(ETH_P_AX25); - skb->mac.raw = skb->data; - skb->pkt_type = PACKET_HOST; + skb->protocol = ax25_type_trans(skb, scc->dev); netif_rx(skb); scc->dev->last_rx = jiffies; diff --git a/drivers/net/hamradio/yam.c b/drivers/net/hamradio/yam.c index fd7b00fe38e..41213ef602d 100644 --- a/drivers/net/hamradio/yam.c +++ b/drivers/net/hamradio/yam.c @@ -522,12 +522,10 @@ static inline void yam_rx_flag(struct net_device *dev, struct yam_port *yp) ++yp->stats.rx_dropped; } else { unsigned char *cp; - skb->dev = dev; cp = skb_put(skb, pkt_len); *cp++ = 0; /* KISS kludge */ memcpy(cp, yp->rx_buf, pkt_len - 1); - skb->protocol = htons(ETH_P_AX25); - skb->mac.raw = skb->data; + skb->protocol = ax25_type_trans(skb, dev); netif_rx(skb); dev->last_rx = jiffies; ++yp->stats.rx_packets; diff --git a/drivers/net/hp100.c b/drivers/net/hp100.c index acb170152bb..b3a898c5a58 100644 --- a/drivers/net/hp100.c +++ b/drivers/net/hp100.c @@ -13,8 +13,8 @@ ** This driver has only been tested with ** -- HP J2585B 10/100 Mbit/s PCI Busmaster ** -- HP J2585A 10/100 Mbit/s PCI -** -- HP J2970 10 Mbit/s PCI Combo 10base-T/BNC -** -- HP J2973 10 Mbit/s PCI 10base-T +** -- HP J2970A 10 Mbit/s PCI Combo 10base-T/BNC +** -- HP J2973A 10 Mbit/s PCI 10base-T ** -- HP J2573 10/100 ISA ** -- Compex ReadyLink ENET100-VG4 10/100 Mbit/s PCI / EISA ** -- Compex FreedomLine 100/VG 10/100 Mbit/s ISA / EISA / PCI diff --git a/drivers/net/irda/Kconfig b/drivers/net/irda/Kconfig index 6bf76a444d4..1c553d7efdd 100644 --- a/drivers/net/irda/Kconfig +++ b/drivers/net/irda/Kconfig @@ -310,7 +310,7 @@ config SIGMATEL_FIR config NSC_FIR tristate "NSC PC87108/PC87338" - depends on IRDA + depends on IRDA && ISA_DMA_API help Say Y here if you want to build support for the NSC PC87108 and PC87338 IrDA chipsets. This driver supports SIR, @@ -321,7 +321,7 @@ config NSC_FIR config WINBOND_FIR tristate "Winbond W83977AF (IR)" - depends on IRDA + depends on IRDA && ISA_DMA_API help Say Y here if you want to build IrDA support for the Winbond W83977AF super-io chipset. This driver should be used for the IrDA @@ -347,7 +347,7 @@ config AU1000_FIR config SMC_IRCC_FIR tristate "SMSC IrCC (EXPERIMENTAL)" - depends on EXPERIMENTAL && IRDA + depends on EXPERIMENTAL && IRDA && ISA_DMA_API help Say Y here if you want to build support for the SMC Infrared Communications Controller. It is used in a wide variety of @@ -357,7 +357,7 @@ config SMC_IRCC_FIR config ALI_FIR tristate "ALi M5123 FIR (EXPERIMENTAL)" - depends on EXPERIMENTAL && IRDA + depends on EXPERIMENTAL && IRDA && ISA_DMA_API help Say Y here if you want to build support for the ALi M5123 FIR Controller. The ALi M5123 FIR Controller is embedded in ALi M1543C, @@ -385,7 +385,7 @@ config SA1100_FIR config VIA_FIR tristate "VIA VT8231/VT1211 SIR/MIR/FIR" - depends on IRDA + depends on IRDA && ISA_DMA_API help Say Y here if you want to build support for the VIA VT8231 and VIA VT1211 IrDA controllers, found on the motherboards using diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index 2ffc31708d5..b33111e2131 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -7,7 +7,7 @@ * * Version: @(#)loopback.c 1.0.4b 08/16/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Donald Becker, <becker@scyld.com> * diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index 7e94d455533..0405e1f0d3d 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -99,7 +99,7 @@ static spinlock_t mv643xx_eth_phy_lock = SPIN_LOCK_UNLOCKED; static inline u32 mv_read(int offset) { - void *__iomem reg_base; + void __iomem *reg_base; reg_base = mv643xx_eth_shared_base - MV643XX_ETH_SHARED_REGS; @@ -108,7 +108,7 @@ static inline u32 mv_read(int offset) static inline void mv_write(int offset, u32 data) { - void * __iomem reg_base; + void __iomem *reg_base; reg_base = mv643xx_eth_shared_base - MV643XX_ETH_SHARED_REGS; writel(data, reg_base + offset); diff --git a/drivers/net/ppp_deflate.c b/drivers/net/ppp_deflate.c index 507d6328d4e..3872088fdd1 100644 --- a/drivers/net/ppp_deflate.c +++ b/drivers/net/ppp_deflate.c @@ -87,8 +87,7 @@ static void z_comp_free(void *arg) if (state) { zlib_deflateEnd(&state->strm); - if (state->strm.workspace) - vfree(state->strm.workspace); + vfree(state->strm.workspace); kfree(state); } } @@ -308,8 +307,7 @@ static void z_decomp_free(void *arg) if (state) { zlib_inflateEnd(&state->strm); - if (state->strm.workspace) - kfree(state->strm.workspace); + kfree(state->strm.workspace); kfree(state); } } diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c index c456dc81b87..3b377f6cd4a 100644 --- a/drivers/net/ppp_generic.c +++ b/drivers/net/ppp_generic.c @@ -2467,14 +2467,10 @@ static void ppp_destroy_interface(struct ppp *ppp) skb_queue_purge(&ppp->mrq); #endif /* CONFIG_PPP_MULTILINK */ #ifdef CONFIG_PPP_FILTER - if (ppp->pass_filter) { - kfree(ppp->pass_filter); - ppp->pass_filter = NULL; - } - if (ppp->active_filter) { - kfree(ppp->active_filter); - ppp->active_filter = NULL; - } + kfree(ppp->pass_filter); + ppp->pass_filter = NULL; + kfree(ppp->active_filter); + ppp->active_filter = NULL; #endif /* CONFIG_PPP_FILTER */ kfree(ppp); diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index 07e2df09491..c59507f8a76 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -2385,7 +2385,7 @@ core_down: } /* Give a racing hard_start_xmit a few cycles to complete. */ - synchronize_kernel(); + synchronize_sched(); /* FIXME: should this be synchronize_irq()? */ /* * And now for the 50k$ question: are IRQ disabled or not ? diff --git a/drivers/net/slip.c b/drivers/net/slip.c index 4ce52f5f241..8f7841c0374 100644 --- a/drivers/net/slip.c +++ b/drivers/net/slip.c @@ -185,15 +185,12 @@ sl_alloc_bufs(struct slip *sl, int mtu) /* Cleanup */ err_exit: #ifdef SL_INCLUDE_CSLIP - if (cbuff) - kfree(cbuff); + kfree(cbuff); if (slcomp) slhc_free(slcomp); #endif - if (xbuff) - kfree(xbuff); - if (rbuff) - kfree(rbuff); + kfree(xbuff); + kfree(rbuff); return err; } @@ -204,13 +201,13 @@ sl_free_bufs(struct slip *sl) void * tmp; /* Free all SLIP frame buffers. */ - if ((tmp = xchg(&sl->rbuff, NULL)) != NULL) - kfree(tmp); - if ((tmp = xchg(&sl->xbuff, NULL)) != NULL) - kfree(tmp); + tmp = xchg(&sl->rbuff, NULL); + kfree(tmp); + tmp = xchg(&sl->xbuff, NULL); + kfree(tmp); #ifdef SL_INCLUDE_CSLIP - if ((tmp = xchg(&sl->cbuff, NULL)) != NULL) - kfree(tmp); + tmp = xchg(&sl->cbuff, NULL); + kfree(tmp); if ((tmp = xchg(&sl->slcomp, NULL)) != NULL) slhc_free(tmp); #endif @@ -297,13 +294,10 @@ done_on_bh: spin_unlock_bh(&sl->lock); done: - if (xbuff) - kfree(xbuff); - if (rbuff) - kfree(rbuff); + kfree(xbuff); + kfree(rbuff); #ifdef SL_INCLUDE_CSLIP - if (cbuff) - kfree(cbuff); + kfree(cbuff); #endif return err; } diff --git a/drivers/net/sunbmac.c b/drivers/net/sunbmac.c index 025dcd867ea..f88f5e32b71 100644 --- a/drivers/net/sunbmac.c +++ b/drivers/net/sunbmac.c @@ -37,8 +37,18 @@ #include "sunbmac.h" +#define DRV_NAME "sunbmac" +#define DRV_VERSION "2.0" +#define DRV_RELDATE "11/24/03" +#define DRV_AUTHOR "David S. Miller (davem@redhat.com)" + static char version[] __initdata = - "sunbmac.c:v2.0 24/Nov/03 David S. Miller (davem@redhat.com)\n"; + DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n"; + +MODULE_VERSION(DRV_VERSION); +MODULE_AUTHOR(DRV_AUTHOR); +MODULE_DESCRIPTION("Sun BigMAC 100baseT ethernet driver"); +MODULE_LICENSE("GPL"); #undef DEBUG_PROBE #undef DEBUG_TX @@ -1321,4 +1331,3 @@ static void __exit bigmac_cleanup(void) module_init(bigmac_probe); module_exit(bigmac_cleanup); -MODULE_LICENSE("GPL"); diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c index d837b3c3572..f02fe4119b2 100644 --- a/drivers/net/sunhme.c +++ b/drivers/net/sunhme.c @@ -13,9 +13,6 @@ * argument : macaddr=0x00,0x10,0x20,0x30,0x40,0x50 */ -static char version[] = - "sunhme.c:v2.02 24/Aug/2003 David S. Miller (davem@redhat.com)\n"; - #include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> @@ -67,15 +64,24 @@ static char version[] = #include "sunhme.h" +#define DRV_NAME "sunhme" +#define DRV_VERSION "2.02" +#define DRV_RELDATE "8/24/03" +#define DRV_AUTHOR "David S. Miller (davem@redhat.com)" + +static char version[] = + DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n"; -#define DRV_NAME "sunhme" +MODULE_VERSION(DRV_VERSION); +MODULE_AUTHOR(DRV_AUTHOR); +MODULE_DESCRIPTION("Sun HappyMealEthernet(HME) 10/100baseT ethernet driver"); +MODULE_LICENSE("GPL"); static int macaddr[6]; /* accept MAC address of the form macaddr=0x08,0x00,0x20,0x30,0x40,0x50 */ module_param_array(macaddr, int, NULL, 0); MODULE_PARM_DESC(macaddr, "Happy Meal MAC address to set"); -MODULE_LICENSE("GPL"); static struct happy_meal *root_happy_dev; diff --git a/drivers/net/sunlance.c b/drivers/net/sunlance.c index 62d464c7ef5..b7d87d4690b 100644 --- a/drivers/net/sunlance.c +++ b/drivers/net/sunlance.c @@ -69,9 +69,6 @@ #undef DEBUG_DRIVER -static char version[] = - "sunlance.c:v2.02 24/Aug/03 Miguel de Icaza (miguel@nuclecu.unam.mx)\n"; - static char lancestr[] = "LANCE"; #include <linux/config.h> @@ -108,6 +105,19 @@ static char lancestr[] = "LANCE"; #include <asm/auxio.h> /* For tpe-link-test? setting */ #include <asm/irq.h> +#define DRV_NAME "sunlance" +#define DRV_VERSION "2.02" +#define DRV_RELDATE "8/24/03" +#define DRV_AUTHOR "Miguel de Icaza (miguel@nuclecu.unam.mx)" + +static char version[] = + DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n"; + +MODULE_VERSION(DRV_VERSION); +MODULE_AUTHOR(DRV_AUTHOR); +MODULE_DESCRIPTION("Sun Lance ethernet driver"); +MODULE_LICENSE("GPL"); + /* Define: 2^4 Tx buffers and 2^4 Rx buffers */ #ifndef LANCE_LOG_TX_BUFFERS #define LANCE_LOG_TX_BUFFERS 4 @@ -1611,4 +1621,3 @@ static void __exit sparc_lance_cleanup(void) module_init(sparc_lance_probe); module_exit(sparc_lance_cleanup); -MODULE_LICENSE("GPL"); diff --git a/drivers/net/sunqe.c b/drivers/net/sunqe.c index 37ef1b82a6c..1f2323be60d 100644 --- a/drivers/net/sunqe.c +++ b/drivers/net/sunqe.c @@ -7,9 +7,6 @@ * Copyright (C) 1996, 1999, 2003 David S. Miller (davem@redhat.com) */ -static char version[] = - "sunqe.c:v3.0 8/24/03 David S. Miller (davem@redhat.com)\n"; - #include <linux/module.h> #include <linux/kernel.h> #include <linux/types.h> @@ -43,6 +40,19 @@ static char version[] = #include "sunqe.h" +#define DRV_NAME "sunqe" +#define DRV_VERSION "3.0" +#define DRV_RELDATE "8/24/03" +#define DRV_AUTHOR "David S. Miller (davem@redhat.com)" + +static char version[] = + DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n"; + +MODULE_VERSION(DRV_VERSION); +MODULE_AUTHOR(DRV_AUTHOR); +MODULE_DESCRIPTION("Sun QuadEthernet 10baseT SBUS card driver"); +MODULE_LICENSE("GPL"); + static struct sunqec *root_qec_dev; static void qe_set_multicast(struct net_device *dev); @@ -1040,4 +1050,3 @@ static void __exit qec_cleanup(void) module_init(qec_probe); module_exit(qec_cleanup); -MODULE_LICENSE("GPL"); diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 12de80884b1..f79b02e80e7 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -61,8 +61,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.25" -#define DRV_MODULE_RELDATE "March 24, 2005" +#define DRV_MODULE_VERSION "3.27" +#define DRV_MODULE_RELDATE "May 5, 2005" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 @@ -85,8 +85,7 @@ /* hardware minimum and maximum for a single frame's data payload */ #define TG3_MIN_MTU 60 #define TG3_MAX_MTU(tp) \ - ((GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5705 && \ - GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5750) ? 9000 : 1500) + (!(tp->tg3_flags2 & TG3_FLG2_5705_PLUS) ? 9000 : 1500) /* These numbers seem to be hard coded in the NIC firmware somehow. * You can't change the ring sizes, but you can change where you place @@ -205,6 +204,8 @@ static struct pci_device_id tg3_pci_tbl[] = { PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5751F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5752, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5753, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5753M, @@ -425,9 +426,30 @@ static void tg3_enable_ints(struct tg3 *tp) tg3_cond_int(tp); } +static inline unsigned int tg3_has_work(struct tg3 *tp) +{ + struct tg3_hw_status *sblk = tp->hw_status; + unsigned int work_exists = 0; + + /* check for phy events */ + if (!(tp->tg3_flags & + (TG3_FLAG_USE_LINKCHG_REG | + TG3_FLAG_POLL_SERDES))) { + if (sblk->status & SD_STATUS_LINK_CHG) + work_exists = 1; + } + /* check for RX/TX work to do */ + if (sblk->idx[0].tx_consumer != tp->tx_cons || + sblk->idx[0].rx_producer != tp->rx_rcb_ptr) + work_exists = 1; + + return work_exists; +} + /* tg3_restart_ints - * similar to tg3_enable_ints, but it can return without flushing the - * PIO write which reenables interrupts + * similar to tg3_enable_ints, but it accurately determines whether there + * is new work pending and can return without flushing the PIO write + * which reenables interrupts */ static void tg3_restart_ints(struct tg3 *tp) { @@ -436,7 +458,9 @@ static void tg3_restart_ints(struct tg3 *tp) tw32_mailbox(MAILBOX_INTERRUPT_0 + TG3_64BIT_REG_LOW, 0x00000000); mmiowb(); - tg3_cond_int(tp); + if (tg3_has_work(tp)) + tw32(HOSTCC_MODE, tp->coalesce_mode | + (HOSTCC_MODE_ENABLE | HOSTCC_MODE_NOW)); } static inline void tg3_netif_stop(struct tg3 *tp) @@ -860,8 +884,7 @@ out: if ((tp->phy_id & PHY_ID_MASK) == PHY_ID_BCM5401) { /* Cannot do read-modify-write on 5401 */ tg3_writephy(tp, MII_TG3_AUX_CTRL, 0x4c20); - } else if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5705 && - GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5750) { + } else if (!(tp->tg3_flags2 & TG3_FLG2_5705_PLUS)) { u32 phy_reg; /* Set bit 14 with read-modify-write to preserve other bits */ @@ -873,8 +896,7 @@ out: /* Set phy register 0x10 bit 0 to high fifo elasticity to support * jumbo frames transmission. */ - if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5705 && - GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5750) { + if (!(tp->tg3_flags2 & TG3_FLG2_5705_PLUS)) { u32 phy_reg; if (!tg3_readphy(tp, MII_TG3_EXT_CTRL, &phy_reg)) @@ -1006,8 +1028,13 @@ static int tg3_set_power_state(struct tg3 *tp, int state) pci_write_config_word(tp->pdev, pm + PCI_PM_CTRL, power_control); - tw32_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl); - udelay(100); + udelay(100); /* Delay after power state change */ + + /* Switch out of Vaux if it is not a LOM */ + if (!(tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT)) { + tw32_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl); + udelay(100); + } return 0; @@ -1068,7 +1095,7 @@ static int tg3_set_power_state(struct tg3 *tp, int state) mac_mode = MAC_MODE_PORT_MODE_TBI; } - if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5750) + if (!(tp->tg3_flags2 & TG3_FLG2_5750_PLUS)) tw32(MAC_LED_CTRL, tp->led_ctrl); if (((power_caps & PCI_PM_CAP_PME_D3cold) && @@ -1095,7 +1122,7 @@ static int tg3_set_power_state(struct tg3 *tp, int state) CLOCK_CTRL_ALTCLK | CLOCK_CTRL_PWRDOWN_PLL133); udelay(40); - } else if (!((GET_ASIC_REV(tp->pci_chip_rev_id) == 5750) && + } else if (!((tp->tg3_flags2 & TG3_FLG2_5750_PLUS) && (tp->tg3_flags & TG3_FLAG_ENABLE_ASF))) { u32 newbits1, newbits2; @@ -1152,6 +1179,7 @@ static int tg3_set_power_state(struct tg3 *tp, int state) /* Finally, set the new power state. */ pci_write_config_word(tp->pdev, pm + PCI_PM_CTRL, power_control); + udelay(100); /* Delay after power state change */ tg3_write_sig_post_reset(tp, RESET_KIND_SHUTDOWN); @@ -2681,8 +2709,8 @@ static int tg3_vlan_rx(struct tg3 *tp, struct sk_buff *skb, u16 vlan_tag) static int tg3_rx(struct tg3 *tp, int budget) { u32 work_mask; - u32 rx_rcb_ptr = tp->rx_rcb_ptr; - u16 hw_idx, sw_idx; + u32 sw_idx = tp->rx_rcb_ptr; + u16 hw_idx; int received; hw_idx = tp->hw_status->idx[0].rx_producer; @@ -2691,7 +2719,6 @@ static int tg3_rx(struct tg3 *tp, int budget) * the opaque cookie. */ rmb(); - sw_idx = rx_rcb_ptr % TG3_RX_RCB_RING_SIZE(tp); work_mask = 0; received = 0; while (sw_idx != hw_idx && budget > 0) { @@ -2796,14 +2823,19 @@ static int tg3_rx(struct tg3 *tp, int budget) next_pkt: (*post_ptr)++; next_pkt_nopost: - rx_rcb_ptr++; - sw_idx = rx_rcb_ptr % TG3_RX_RCB_RING_SIZE(tp); + sw_idx++; + sw_idx %= TG3_RX_RCB_RING_SIZE(tp); + + /* Refresh hw_idx to see if there is new work */ + if (sw_idx == hw_idx) { + hw_idx = tp->hw_status->idx[0].rx_producer; + rmb(); + } } /* ACK the status ring. */ - tp->rx_rcb_ptr = rx_rcb_ptr; - tw32_rx_mbox(MAILBOX_RCVRET_CON_IDX_0 + TG3_64BIT_REG_LOW, - (rx_rcb_ptr % TG3_RX_RCB_RING_SIZE(tp))); + tp->rx_rcb_ptr = sw_idx; + tw32_rx_mbox(MAILBOX_RCVRET_CON_IDX_0 + TG3_64BIT_REG_LOW, sw_idx); /* Refill RX ring(s). */ if (work_mask & RXD_OPAQUE_RING_STD) { @@ -2882,24 +2914,41 @@ static int tg3_poll(struct net_device *netdev, int *budget) return (done ? 0 : 1); } -static inline unsigned int tg3_has_work(struct net_device *dev, struct tg3 *tp) +/* MSI ISR - No need to check for interrupt sharing and no need to + * flush status block and interrupt mailbox. PCI ordering rules + * guarantee that MSI will arrive after the status block. + */ +static irqreturn_t tg3_msi(int irq, void *dev_id, struct pt_regs *regs) { + struct net_device *dev = dev_id; + struct tg3 *tp = netdev_priv(dev); struct tg3_hw_status *sblk = tp->hw_status; - unsigned int work_exists = 0; + unsigned long flags; - /* check for phy events */ - if (!(tp->tg3_flags & - (TG3_FLAG_USE_LINKCHG_REG | - TG3_FLAG_POLL_SERDES))) { - if (sblk->status & SD_STATUS_LINK_CHG) - work_exists = 1; + spin_lock_irqsave(&tp->lock, flags); + + /* + * writing any value to intr-mbox-0 clears PCI INTA# and + * chip-internal interrupt pending events. + * writing non-zero to intr-mbox-0 additional tells the + * NIC to stop sending us irqs, engaging "in-intr-handler" + * event coalescing. + */ + tw32_mailbox(MAILBOX_INTERRUPT_0 + TG3_64BIT_REG_LOW, 0x00000001); + sblk->status &= ~SD_STATUS_UPDATED; + + if (likely(tg3_has_work(tp))) + netif_rx_schedule(dev); /* schedule NAPI poll */ + else { + /* no work, re-enable interrupts + */ + tw32_mailbox(MAILBOX_INTERRUPT_0 + TG3_64BIT_REG_LOW, + 0x00000000); } - /* check for RX/TX work to do */ - if (sblk->idx[0].tx_consumer != tp->tx_cons || - sblk->idx[0].rx_producer != tp->rx_rcb_ptr) - work_exists = 1; - return work_exists; + spin_unlock_irqrestore(&tp->lock, flags); + + return IRQ_RETVAL(1); } static irqreturn_t tg3_interrupt(int irq, void *dev_id, struct pt_regs *regs) @@ -2935,7 +2984,7 @@ static irqreturn_t tg3_interrupt(int irq, void *dev_id, struct pt_regs *regs) tr32(MAILBOX_INTERRUPT_0 + TG3_64BIT_REG_LOW); sblk->status &= ~SD_STATUS_UPDATED; - if (likely(tg3_has_work(dev, tp))) + if (likely(tg3_has_work(tp))) netif_rx_schedule(dev); /* schedule NAPI poll */ else { /* no work, shared interrupt perhaps? re-enable @@ -2954,13 +3003,31 @@ static irqreturn_t tg3_interrupt(int irq, void *dev_id, struct pt_regs *regs) return IRQ_RETVAL(handled); } +/* ISR for interrupt test */ +static irqreturn_t tg3_test_isr(int irq, void *dev_id, + struct pt_regs *regs) +{ + struct net_device *dev = dev_id; + struct tg3 *tp = netdev_priv(dev); + struct tg3_hw_status *sblk = tp->hw_status; + + if (sblk->status & SD_STATUS_UPDATED) { + tw32_mailbox(MAILBOX_INTERRUPT_0 + TG3_64BIT_REG_LOW, + 0x00000001); + return IRQ_RETVAL(1); + } + return IRQ_RETVAL(0); +} + static int tg3_init_hw(struct tg3 *); -static int tg3_halt(struct tg3 *); +static int tg3_halt(struct tg3 *, int); #ifdef CONFIG_NET_POLL_CONTROLLER static void tg3_poll_controller(struct net_device *dev) { - tg3_interrupt(dev->irq, dev, NULL); + struct tg3 *tp = netdev_priv(dev); + + tg3_interrupt(tp->pdev->irq, dev, NULL); } #endif @@ -2977,7 +3044,7 @@ static void tg3_reset_task(void *_data) restart_timer = tp->tg3_flags2 & TG3_FLG2_RESTART_TIMER; tp->tg3_flags2 &= ~TG3_FLG2_RESTART_TIMER; - tg3_halt(tp); + tg3_halt(tp, 0); tg3_init_hw(tp); tg3_netif_start(tp); @@ -3323,7 +3390,7 @@ static int tg3_change_mtu(struct net_device *dev, int new_mtu) spin_lock_irq(&tp->lock); spin_lock(&tp->tx_lock); - tg3_halt(tp); + tg3_halt(tp, 1); tg3_set_mtu(dev, tp, new_mtu); @@ -3590,7 +3657,7 @@ err_out: /* To stop a block, clear the enable bit and poll till it * clears. tp->lock is held. */ -static int tg3_stop_block(struct tg3 *tp, unsigned long ofs, u32 enable_bit) +static int tg3_stop_block(struct tg3 *tp, unsigned long ofs, u32 enable_bit, int silent) { unsigned int i; u32 val; @@ -3623,7 +3690,7 @@ static int tg3_stop_block(struct tg3 *tp, unsigned long ofs, u32 enable_bit) break; } - if (i == MAX_WAIT_CNT) { + if (i == MAX_WAIT_CNT && !silent) { printk(KERN_ERR PFX "tg3_stop_block timed out, " "ofs=%lx enable_bit=%x\n", ofs, enable_bit); @@ -3634,7 +3701,7 @@ static int tg3_stop_block(struct tg3 *tp, unsigned long ofs, u32 enable_bit) } /* tp->lock is held. */ -static int tg3_abort_hw(struct tg3 *tp) +static int tg3_abort_hw(struct tg3 *tp, int silent) { int i, err; @@ -3644,22 +3711,20 @@ static int tg3_abort_hw(struct tg3 *tp) tw32_f(MAC_RX_MODE, tp->rx_mode); udelay(10); - err = tg3_stop_block(tp, RCVBDI_MODE, RCVBDI_MODE_ENABLE); - err |= tg3_stop_block(tp, RCVLPC_MODE, RCVLPC_MODE_ENABLE); - err |= tg3_stop_block(tp, RCVLSC_MODE, RCVLSC_MODE_ENABLE); - err |= tg3_stop_block(tp, RCVDBDI_MODE, RCVDBDI_MODE_ENABLE); - err |= tg3_stop_block(tp, RCVDCC_MODE, RCVDCC_MODE_ENABLE); - err |= tg3_stop_block(tp, RCVCC_MODE, RCVCC_MODE_ENABLE); - - err |= tg3_stop_block(tp, SNDBDS_MODE, SNDBDS_MODE_ENABLE); - err |= tg3_stop_block(tp, SNDBDI_MODE, SNDBDI_MODE_ENABLE); - err |= tg3_stop_block(tp, SNDDATAI_MODE, SNDDATAI_MODE_ENABLE); - err |= tg3_stop_block(tp, RDMAC_MODE, RDMAC_MODE_ENABLE); - err |= tg3_stop_block(tp, SNDDATAC_MODE, SNDDATAC_MODE_ENABLE); - err |= tg3_stop_block(tp, DMAC_MODE, DMAC_MODE_ENABLE); - err |= tg3_stop_block(tp, SNDBDC_MODE, SNDBDC_MODE_ENABLE); - if (err) - goto out; + err = tg3_stop_block(tp, RCVBDI_MODE, RCVBDI_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, RCVLPC_MODE, RCVLPC_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, RCVLSC_MODE, RCVLSC_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, RCVDBDI_MODE, RCVDBDI_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, RCVDCC_MODE, RCVDCC_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, RCVCC_MODE, RCVCC_MODE_ENABLE, silent); + + err |= tg3_stop_block(tp, SNDBDS_MODE, SNDBDS_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, SNDBDI_MODE, SNDBDI_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, SNDDATAI_MODE, SNDDATAI_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, RDMAC_MODE, RDMAC_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, SNDDATAC_MODE, SNDDATAC_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, DMAC_MODE, DMAC_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, SNDBDC_MODE, SNDBDC_MODE_ENABLE, silent); tp->mac_mode &= ~MAC_MODE_TDE_ENABLE; tw32_f(MAC_MODE, tp->mac_mode); @@ -3677,27 +3742,24 @@ static int tg3_abort_hw(struct tg3 *tp) printk(KERN_ERR PFX "tg3_abort_hw timed out for %s, " "TX_MODE_ENABLE will not clear MAC_TX_MODE=%08x\n", tp->dev->name, tr32(MAC_TX_MODE)); - return -ENODEV; + err |= -ENODEV; } - err = tg3_stop_block(tp, HOSTCC_MODE, HOSTCC_MODE_ENABLE); - err |= tg3_stop_block(tp, WDMAC_MODE, WDMAC_MODE_ENABLE); - err |= tg3_stop_block(tp, MBFREE_MODE, MBFREE_MODE_ENABLE); + err |= tg3_stop_block(tp, HOSTCC_MODE, HOSTCC_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, WDMAC_MODE, WDMAC_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, MBFREE_MODE, MBFREE_MODE_ENABLE, silent); tw32(FTQ_RESET, 0xffffffff); tw32(FTQ_RESET, 0x00000000); - err |= tg3_stop_block(tp, BUFMGR_MODE, BUFMGR_MODE_ENABLE); - err |= tg3_stop_block(tp, MEMARB_MODE, MEMARB_MODE_ENABLE); - if (err) - goto out; + err |= tg3_stop_block(tp, BUFMGR_MODE, BUFMGR_MODE_ENABLE, silent); + err |= tg3_stop_block(tp, MEMARB_MODE, MEMARB_MODE_ENABLE, silent); if (tp->hw_status) memset(tp->hw_status, 0, TG3_HW_STATUS_SIZE); if (tp->hw_stats) memset(tp->hw_stats, 0, sizeof(struct tg3_hw_stats)); -out: return err; } @@ -3727,6 +3789,28 @@ static void tg3_nvram_unlock(struct tg3 *tp) } /* tp->lock is held. */ +static void tg3_enable_nvram_access(struct tg3 *tp) +{ + if ((tp->tg3_flags2 & TG3_FLG2_5750_PLUS) && + !(tp->tg3_flags2 & TG3_FLG2_PROTECTED_NVRAM)) { + u32 nvaccess = tr32(NVRAM_ACCESS); + + tw32(NVRAM_ACCESS, nvaccess | ACCESS_ENABLE); + } +} + +/* tp->lock is held. */ +static void tg3_disable_nvram_access(struct tg3 *tp) +{ + if ((tp->tg3_flags2 & TG3_FLG2_5750_PLUS) && + !(tp->tg3_flags2 & TG3_FLG2_PROTECTED_NVRAM)) { + u32 nvaccess = tr32(NVRAM_ACCESS); + + tw32(NVRAM_ACCESS, nvaccess & ~ACCESS_ENABLE); + } +} + +/* tp->lock is held. */ static void tg3_write_sig_pre_reset(struct tg3 *tp, int kind) { if (!(tp->tg3_flags2 & TG3_FLG2_SUN_570X)) @@ -3967,7 +4051,7 @@ static int tg3_chip_reset(struct tg3 *tp) tg3_read_mem(tp, NIC_SRAM_DATA_CFG, &nic_cfg); if (nic_cfg & NIC_SRAM_DATA_CFG_ASF_ENABLE) { tp->tg3_flags |= TG3_FLAG_ENABLE_ASF; - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) + if (tp->tg3_flags2 & TG3_FLG2_5750_PLUS) tp->tg3_flags2 |= TG3_FLG2_ASF_NEW_HANDSHAKE; } } @@ -3997,7 +4081,7 @@ static void tg3_stop_fw(struct tg3 *tp) } /* tp->lock is held. */ -static int tg3_halt(struct tg3 *tp) +static int tg3_halt(struct tg3 *tp, int silent) { int err; @@ -4005,7 +4089,7 @@ static int tg3_halt(struct tg3 *tp) tg3_write_sig_pre_reset(tp, RESET_KIND_SHUTDOWN); - tg3_abort_hw(tp); + tg3_abort_hw(tp, silent); err = tg3_chip_reset(tp); tg3_write_sig_legacy(tp, RESET_KIND_SHUTDOWN); @@ -4974,9 +5058,7 @@ static int tg3_reset_hw(struct tg3 *tp) tg3_write_sig_pre_reset(tp, RESET_KIND_INIT); if (tp->tg3_flags & TG3_FLAG_INIT_COMPLETE) { - err = tg3_abort_hw(tp); - if (err) - return err; + tg3_abort_hw(tp, 1); } err = tg3_chip_reset(tp); @@ -5041,7 +5123,7 @@ static int tg3_reset_hw(struct tg3 *tp) tw32(GRC_MISC_CFG, val); /* Initialize MBUF/DESC pool. */ - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { + if (tp->tg3_flags2 & TG3_FLG2_5750_PLUS) { /* Do nothing. */ } else if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5705) { tw32(BUFMGR_MB_POOL_ADDR, NIC_SRAM_MBUF_POOL_BASE); @@ -5238,6 +5320,8 @@ static int tg3_reset_hw(struct tg3 *tp) RDMAC_MODE_LNGREAD_ENAB); if (tp->tg3_flags & TG3_FLAG_SPLIT_MODE) rdmac_mode |= RDMAC_MODE_SPLIT_ENABLE; + + /* If statement applies to 5705 and 5750 PCI devices only */ if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5705 && tp->pci_chip_rev_id != CHIPREV_ID_5705_A0) || (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750)) { @@ -5251,6 +5335,9 @@ static int tg3_reset_hw(struct tg3 *tp) } } + if (tp->tg3_flags2 & TG3_FLG2_PCI_EXPRESS) + rdmac_mode |= RDMAC_MODE_FIFO_LONG_BURST; + #if TG3_TSO_SUPPORT != 0 if (tp->tg3_flags2 & TG3_FLG2_HW_TSO) rdmac_mode |= (1 << 27); @@ -5332,10 +5419,28 @@ static int tg3_reset_hw(struct tg3 *tp) tw32_f(MAC_MODE, tp->mac_mode | MAC_MODE_RXSTAT_CLEAR | MAC_MODE_TXSTAT_CLEAR); udelay(40); - tp->grc_local_ctrl = GRC_LCLCTRL_INT_ON_ATTN | GRC_LCLCTRL_AUTO_SEEPROM; - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5700) + /* tp->grc_local_ctrl is partially set up during tg3_get_invariants(). + * If TG3_FLAG_EEPROM_WRITE_PROT is set, we should read the + * register to preserve the GPIO settings for LOMs. The GPIOs, + * whether used as inputs or outputs, are set by boot code after + * reset. + */ + if (tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT) { + u32 gpio_mask; + + gpio_mask = GRC_LCLCTRL_GPIO_OE0 | GRC_LCLCTRL_GPIO_OE2 | + GRC_LCLCTRL_GPIO_OUTPUT0 | GRC_LCLCTRL_GPIO_OUTPUT2; + + if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5752) + gpio_mask |= GRC_LCLCTRL_GPIO_OE3 | + GRC_LCLCTRL_GPIO_OUTPUT3; + + tp->grc_local_ctrl |= tr32(GRC_LOCAL_CTRL) & gpio_mask; + + /* GPIO1 must be driven high for eeprom write protect */ tp->grc_local_ctrl |= (GRC_LCLCTRL_GPIO_OE1 | GRC_LCLCTRL_GPIO_OUTPUT1); + } tw32_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl); udelay(100); @@ -5353,6 +5458,7 @@ static int tg3_reset_hw(struct tg3 *tp) WDMAC_MODE_FIFOURUN_ENAB | WDMAC_MODE_FIFOOREAD_ENAB | WDMAC_MODE_LNGREAD_ENAB); + /* If statement applies to 5705 and 5750 PCI devices only */ if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5705 && tp->pci_chip_rev_id != CHIPREV_ID_5705_A0) || GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { @@ -5706,6 +5812,118 @@ static void tg3_timer(unsigned long __opaque) add_timer(&tp->timer); } +static int tg3_test_interrupt(struct tg3 *tp) +{ + struct net_device *dev = tp->dev; + int err, i; + u32 int_mbox = 0; + + tg3_disable_ints(tp); + + free_irq(tp->pdev->irq, dev); + + err = request_irq(tp->pdev->irq, tg3_test_isr, + SA_SHIRQ | SA_SAMPLE_RANDOM, dev->name, dev); + if (err) + return err; + + tg3_enable_ints(tp); + + tw32_f(HOSTCC_MODE, tp->coalesce_mode | HOSTCC_MODE_ENABLE | + HOSTCC_MODE_NOW); + + for (i = 0; i < 5; i++) { + int_mbox = tr32(MAILBOX_INTERRUPT_0 + TG3_64BIT_REG_LOW); + if (int_mbox != 0) + break; + msleep(10); + } + + tg3_disable_ints(tp); + + free_irq(tp->pdev->irq, dev); + + if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) + err = request_irq(tp->pdev->irq, tg3_msi, + SA_SAMPLE_RANDOM, dev->name, dev); + else + err = request_irq(tp->pdev->irq, tg3_interrupt, + SA_SHIRQ | SA_SAMPLE_RANDOM, dev->name, dev); + + if (err) + return err; + + if (int_mbox != 0) + return 0; + + return -EIO; +} + +/* Returns 0 if MSI test succeeds or MSI test fails and INTx mode is + * successfully restored + */ +static int tg3_test_msi(struct tg3 *tp) +{ + struct net_device *dev = tp->dev; + int err; + u16 pci_cmd; + + if (!(tp->tg3_flags2 & TG3_FLG2_USING_MSI)) + return 0; + + /* Turn off SERR reporting in case MSI terminates with Master + * Abort. + */ + pci_read_config_word(tp->pdev, PCI_COMMAND, &pci_cmd); + pci_write_config_word(tp->pdev, PCI_COMMAND, + pci_cmd & ~PCI_COMMAND_SERR); + + err = tg3_test_interrupt(tp); + + pci_write_config_word(tp->pdev, PCI_COMMAND, pci_cmd); + + if (!err) + return 0; + + /* other failures */ + if (err != -EIO) + return err; + + /* MSI test failed, go back to INTx mode */ + printk(KERN_WARNING PFX "%s: No interrupt was generated using MSI, " + "switching to INTx mode. Please report this failure to " + "the PCI maintainer and include system chipset information.\n", + tp->dev->name); + + free_irq(tp->pdev->irq, dev); + pci_disable_msi(tp->pdev); + + tp->tg3_flags2 &= ~TG3_FLG2_USING_MSI; + + err = request_irq(tp->pdev->irq, tg3_interrupt, + SA_SHIRQ | SA_SAMPLE_RANDOM, dev->name, dev); + + if (err) + return err; + + /* Need to reset the chip because the MSI cycle may have terminated + * with Master Abort. + */ + spin_lock_irq(&tp->lock); + spin_lock(&tp->tx_lock); + + tg3_halt(tp, 1); + err = tg3_init_hw(tp); + + spin_unlock(&tp->tx_lock); + spin_unlock_irq(&tp->lock); + + if (err) + free_irq(tp->pdev->irq, dev); + + return err; +} + static int tg3_open(struct net_device *dev) { struct tg3 *tp = netdev_priv(dev); @@ -5727,10 +5945,29 @@ static int tg3_open(struct net_device *dev) if (err) return err; - err = request_irq(dev->irq, tg3_interrupt, - SA_SHIRQ, dev->name, dev); + if ((tp->tg3_flags2 & TG3_FLG2_5750_PLUS) && + (GET_CHIP_REV(tp->pci_chip_rev_id) != CHIPREV_5750_AX) && + (GET_CHIP_REV(tp->pci_chip_rev_id) != CHIPREV_5750_BX)) { + if (pci_enable_msi(tp->pdev) == 0) { + u32 msi_mode; + + msi_mode = tr32(MSGINT_MODE); + tw32(MSGINT_MODE, msi_mode | MSGINT_MODE_ENABLE); + tp->tg3_flags2 |= TG3_FLG2_USING_MSI; + } + } + if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) + err = request_irq(tp->pdev->irq, tg3_msi, + SA_SAMPLE_RANDOM, dev->name, dev); + else + err = request_irq(tp->pdev->irq, tg3_interrupt, + SA_SHIRQ | SA_SAMPLE_RANDOM, dev->name, dev); if (err) { + if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) { + pci_disable_msi(tp->pdev); + tp->tg3_flags2 &= ~TG3_FLG2_USING_MSI; + } tg3_free_consistent(tp); return err; } @@ -5740,7 +5977,7 @@ static int tg3_open(struct net_device *dev) err = tg3_init_hw(tp); if (err) { - tg3_halt(tp); + tg3_halt(tp, 1); tg3_free_rings(tp); } else { tp->timer_offset = HZ / 10; @@ -5751,23 +5988,47 @@ static int tg3_open(struct net_device *dev) tp->timer.expires = jiffies + tp->timer_offset; tp->timer.data = (unsigned long) tp; tp->timer.function = tg3_timer; - add_timer(&tp->timer); - - tp->tg3_flags |= TG3_FLAG_INIT_COMPLETE; } spin_unlock(&tp->tx_lock); spin_unlock_irq(&tp->lock); if (err) { - free_irq(dev->irq, dev); + free_irq(tp->pdev->irq, dev); + if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) { + pci_disable_msi(tp->pdev); + tp->tg3_flags2 &= ~TG3_FLG2_USING_MSI; + } tg3_free_consistent(tp); return err; } + if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) { + err = tg3_test_msi(tp); + if (err) { + spin_lock_irq(&tp->lock); + spin_lock(&tp->tx_lock); + + if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) { + pci_disable_msi(tp->pdev); + tp->tg3_flags2 &= ~TG3_FLG2_USING_MSI; + } + tg3_halt(tp, 1); + tg3_free_rings(tp); + tg3_free_consistent(tp); + + spin_unlock(&tp->tx_lock); + spin_unlock_irq(&tp->lock); + + return err; + } + } + spin_lock_irq(&tp->lock); spin_lock(&tp->tx_lock); + add_timer(&tp->timer); + tp->tg3_flags |= TG3_FLAG_INIT_COMPLETE; tg3_enable_ints(tp); spin_unlock(&tp->tx_lock); @@ -6025,7 +6286,7 @@ static int tg3_close(struct net_device *dev) tg3_disable_ints(tp); - tg3_halt(tp); + tg3_halt(tp, 1); tg3_free_rings(tp); tp->tg3_flags &= ~(TG3_FLAG_INIT_COMPLETE | @@ -6035,7 +6296,11 @@ static int tg3_close(struct net_device *dev) spin_unlock(&tp->tx_lock); spin_unlock_irq(&tp->lock); - free_irq(dev->irq, dev); + free_irq(tp->pdev->irq, dev); + if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) { + pci_disable_msi(tp->pdev); + tp->tg3_flags2 &= ~TG3_FLG2_USING_MSI; + } memcpy(&tp->net_stats_prev, tg3_get_stats(tp->dev), sizeof(tp->net_stats_prev)); @@ -6509,10 +6774,12 @@ static int tg3_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, start = cpu_to_le32(start); len += b_offset; offset &= ~3; + if (len < 4) + len = 4; } odd_len = 0; - if ((len & 3) && ((len > 4) || (b_offset == 0))) { + if (len & 3) { /* adjustments to end on required 4 byte boundary */ odd_len = 1; len = (len + 3) & ~3; @@ -6739,7 +7006,7 @@ static int tg3_set_ringparam(struct net_device *dev, struct ethtool_ringparam *e tp->tx_pending = ering->tx_pending; if (netif_running(dev)) { - tg3_halt(tp); + tg3_halt(tp, 1); tg3_init_hw(tp); tg3_netif_start(tp); } @@ -6782,7 +7049,7 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam tp->tg3_flags &= ~TG3_FLAG_TX_PAUSE; if (netif_running(dev)) { - tg3_halt(tp); + tg3_halt(tp, 1); tg3_init_hw(tp); tg3_netif_start(tp); } @@ -7067,6 +7334,67 @@ static void __devinit tg3_get_nvram_info(struct tg3 *tp) } } +static void __devinit tg3_get_5752_nvram_info(struct tg3 *tp) +{ + u32 nvcfg1; + + nvcfg1 = tr32(NVRAM_CFG1); + + /* NVRAM protection for TPM */ + if (nvcfg1 & (1 << 27)) + tp->tg3_flags2 |= TG3_FLG2_PROTECTED_NVRAM; + + switch (nvcfg1 & NVRAM_CFG1_5752VENDOR_MASK) { + case FLASH_5752VENDOR_ATMEL_EEPROM_64KHZ: + case FLASH_5752VENDOR_ATMEL_EEPROM_376KHZ: + tp->nvram_jedecnum = JEDEC_ATMEL; + tp->tg3_flags |= TG3_FLAG_NVRAM_BUFFERED; + break; + case FLASH_5752VENDOR_ATMEL_FLASH_BUFFERED: + tp->nvram_jedecnum = JEDEC_ATMEL; + tp->tg3_flags |= TG3_FLAG_NVRAM_BUFFERED; + tp->tg3_flags2 |= TG3_FLG2_FLASH; + break; + case FLASH_5752VENDOR_ST_M45PE10: + case FLASH_5752VENDOR_ST_M45PE20: + case FLASH_5752VENDOR_ST_M45PE40: + tp->nvram_jedecnum = JEDEC_ST; + tp->tg3_flags |= TG3_FLAG_NVRAM_BUFFERED; + tp->tg3_flags2 |= TG3_FLG2_FLASH; + break; + } + + if (tp->tg3_flags2 & TG3_FLG2_FLASH) { + switch (nvcfg1 & NVRAM_CFG1_5752PAGE_SIZE_MASK) { + case FLASH_5752PAGE_SIZE_256: + tp->nvram_pagesize = 256; + break; + case FLASH_5752PAGE_SIZE_512: + tp->nvram_pagesize = 512; + break; + case FLASH_5752PAGE_SIZE_1K: + tp->nvram_pagesize = 1024; + break; + case FLASH_5752PAGE_SIZE_2K: + tp->nvram_pagesize = 2048; + break; + case FLASH_5752PAGE_SIZE_4K: + tp->nvram_pagesize = 4096; + break; + case FLASH_5752PAGE_SIZE_264: + tp->nvram_pagesize = 264; + break; + } + } + else { + /* For eeprom, set pagesize to maximum eeprom size */ + tp->nvram_pagesize = ATMEL_AT24C512_CHIP_SIZE; + + nvcfg1 &= ~NVRAM_CFG1_COMPAT_BYPASS; + tw32(NVRAM_CFG1, nvcfg1); + } +} + /* Chips other than 5700/5701 use the NVRAM for fetching info. */ static void __devinit tg3_nvram_init(struct tg3 *tp) { @@ -7093,20 +7421,16 @@ static void __devinit tg3_nvram_init(struct tg3 *tp) GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5701) { tp->tg3_flags |= TG3_FLAG_NVRAM; - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { - u32 nvaccess = tr32(NVRAM_ACCESS); + tg3_enable_nvram_access(tp); - tw32(NVRAM_ACCESS, nvaccess | ACCESS_ENABLE); - } + if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5752) + tg3_get_5752_nvram_info(tp); + else + tg3_get_nvram_info(tp); - tg3_get_nvram_info(tp); tg3_get_nvram_size(tp); - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { - u32 nvaccess = tr32(NVRAM_ACCESS); - - tw32(NVRAM_ACCESS, nvaccess & ~ACCESS_ENABLE); - } + tg3_disable_nvram_access(tp); } else { tp->tg3_flags &= ~(TG3_FLAG_NVRAM | TG3_FLAG_NVRAM_BUFFERED); @@ -7195,11 +7519,7 @@ static int tg3_nvram_read(struct tg3 *tp, u32 offset, u32 *val) tg3_nvram_lock(tp); - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { - u32 nvaccess = tr32(NVRAM_ACCESS); - - tw32_f(NVRAM_ACCESS, nvaccess | ACCESS_ENABLE); - } + tg3_enable_nvram_access(tp); tw32(NVRAM_ADDR, offset); ret = tg3_nvram_exec_cmd(tp, NVRAM_CMD_RD | NVRAM_CMD_GO | @@ -7210,11 +7530,7 @@ static int tg3_nvram_read(struct tg3 *tp, u32 offset, u32 *val) tg3_nvram_unlock(tp); - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { - u32 nvaccess = tr32(NVRAM_ACCESS); - - tw32_f(NVRAM_ACCESS, nvaccess & ~ACCESS_ENABLE); - } + tg3_disable_nvram_access(tp); return ret; } @@ -7277,7 +7593,7 @@ static int tg3_nvram_write_block_unbuffered(struct tg3 *tp, u32 offset, u32 len, while (len) { int j; - u32 phy_addr, page_off, size, nvaccess; + u32 phy_addr, page_off, size; phy_addr = offset & ~pagemask; @@ -7300,8 +7616,7 @@ static int tg3_nvram_write_block_unbuffered(struct tg3 *tp, u32 offset, u32 len, offset = offset + (pagesize - page_off); - nvaccess = tr32(NVRAM_ACCESS); - tw32_f(NVRAM_ACCESS, nvaccess | ACCESS_ENABLE); + tg3_enable_nvram_access(tp); /* * Before we can erase the flash page, we need @@ -7425,8 +7740,8 @@ static int tg3_nvram_write_block(struct tg3 *tp, u32 offset, u32 len, u8 *buf) } if (tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT) { - tw32_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl | - GRC_LCLCTRL_GPIO_OE1); + tw32_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl & + ~GRC_LCLCTRL_GPIO_OUTPUT1); udelay(40); } @@ -7438,13 +7753,10 @@ static int tg3_nvram_write_block(struct tg3 *tp, u32 offset, u32 len, u8 *buf) tg3_nvram_lock(tp); - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { - u32 nvaccess = tr32(NVRAM_ACCESS); - - tw32(NVRAM_ACCESS, nvaccess | ACCESS_ENABLE); - + tg3_enable_nvram_access(tp); + if ((tp->tg3_flags2 & TG3_FLG2_5750_PLUS) && + !(tp->tg3_flags2 & TG3_FLG2_PROTECTED_NVRAM)) tw32(NVRAM_WRITE1, 0x406); - } grc_mode = tr32(GRC_MODE); tw32(GRC_MODE, grc_mode | GRC_MODE_NVRAM_WR_ENABLE); @@ -7463,17 +7775,12 @@ static int tg3_nvram_write_block(struct tg3 *tp, u32 offset, u32 len, u8 *buf) grc_mode = tr32(GRC_MODE); tw32(GRC_MODE, grc_mode & ~GRC_MODE_NVRAM_WR_ENABLE); - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { - u32 nvaccess = tr32(NVRAM_ACCESS); - - tw32(NVRAM_ACCESS, nvaccess & ~ACCESS_ENABLE); - } + tg3_disable_nvram_access(tp); tg3_nvram_unlock(tp); } if (tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT) { - tw32_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl | - GRC_LCLCTRL_GPIO_OE1 | GRC_LCLCTRL_GPIO_OUTPUT1); + tw32_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl); udelay(40); } @@ -7537,21 +7844,27 @@ static inline struct subsys_tbl_ent *lookup_by_subsys(struct tg3 *tp) return NULL; } -static int __devinit tg3_phy_probe(struct tg3 *tp) +/* Since this function may be called in D3-hot power state during + * tg3_init_one(), only config cycles are allowed. + */ +static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp) { - u32 eeprom_phy_id, hw_phy_id_1, hw_phy_id_2; - u32 hw_phy_id, hw_phy_id_masked; u32 val; - int eeprom_signature_found, eeprom_phy_serdes, err; + + /* Make sure register accesses (indirect or otherwise) + * will function correctly. + */ + pci_write_config_dword(tp->pdev, TG3PCI_MISC_HOST_CTRL, + tp->misc_host_ctrl); tp->phy_id = PHY_ID_INVALID; - eeprom_phy_id = PHY_ID_INVALID; - eeprom_phy_serdes = 0; - eeprom_signature_found = 0; + tp->led_ctrl = LED_CTRL_MODE_PHY_1; + tg3_read_mem(tp, NIC_SRAM_DATA_SIG, &val); if (val == NIC_SRAM_DATA_SIG_MAGIC) { u32 nic_cfg, led_cfg; - u32 nic_phy_id, ver, cfg2 = 0; + u32 nic_phy_id, ver, cfg2 = 0, eeprom_phy_id; + int eeprom_phy_serdes = 0; tg3_read_mem(tp, NIC_SRAM_DATA_CFG, &nic_cfg); tp->nic_sram_data_cfg = nic_cfg; @@ -7564,8 +7877,6 @@ static int __devinit tg3_phy_probe(struct tg3 *tp) (ver > 0) && (ver < 0x100)) tg3_read_mem(tp, NIC_SRAM_DATA_CFG_2, &cfg2); - eeprom_signature_found = 1; - if ((nic_cfg & NIC_SRAM_DATA_CFG_PHY_TYPE_MASK) == NIC_SRAM_DATA_CFG_PHY_TYPE_FIBER) eeprom_phy_serdes = 1; @@ -7581,10 +7892,14 @@ static int __devinit tg3_phy_probe(struct tg3 *tp) } else eeprom_phy_id = 0; - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) { + tp->phy_id = eeprom_phy_id; + if (eeprom_phy_serdes) + tp->tg3_flags2 |= TG3_FLG2_PHY_SERDES; + + if (tp->tg3_flags2 & TG3_FLG2_5750_PLUS) led_cfg = cfg2 & (NIC_SRAM_DATA_CFG_LED_MODE_MASK | SHASTA_EXT_LED_MODE_MASK); - } else + else led_cfg = nic_cfg & NIC_SRAM_DATA_CFG_LED_MODE_MASK; switch (led_cfg) { @@ -7634,7 +7949,7 @@ static int __devinit tg3_phy_probe(struct tg3 *tp) if (nic_cfg & NIC_SRAM_DATA_CFG_ASF_ENABLE) { tp->tg3_flags |= TG3_FLAG_ENABLE_ASF; - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) + if (tp->tg3_flags2 & TG3_FLG2_5750_PLUS) tp->tg3_flags2 |= TG3_FLG2_ASF_NEW_HANDSHAKE; } if (nic_cfg & NIC_SRAM_DATA_CFG_FIBER_WOL) @@ -7648,6 +7963,13 @@ static int __devinit tg3_phy_probe(struct tg3 *tp) if (cfg2 & (1 << 18)) tp->tg3_flags2 |= TG3_FLG2_SERDES_PREEMPHASIS; } +} + +static int __devinit tg3_phy_probe(struct tg3 *tp) +{ + u32 hw_phy_id_1, hw_phy_id_2; + u32 hw_phy_id, hw_phy_id_masked; + int err; /* Reading the PHY ID register can conflict with ASF * firwmare access to the PHY hardware. @@ -7676,10 +7998,10 @@ static int __devinit tg3_phy_probe(struct tg3 *tp) if (hw_phy_id_masked == PHY_ID_BCM8002) tp->tg3_flags2 |= TG3_FLG2_PHY_SERDES; } else { - if (eeprom_signature_found) { - tp->phy_id = eeprom_phy_id; - if (eeprom_phy_serdes) - tp->tg3_flags2 |= TG3_FLG2_PHY_SERDES; + if (tp->phy_id != PHY_ID_INVALID) { + /* Do nothing, phy ID already set up in + * tg3_get_eeprom_hw_cfg(). + */ } else { struct subsys_tbl_ent *p; @@ -7750,9 +8072,6 @@ skip_phy_reset: err = tg3_init_5401phy_dsp(tp); } - if (!eeprom_signature_found) - tp->led_ctrl = LED_CTRL_MODE_PHY_1; - if (tp->tg3_flags2 & TG3_FLG2_PHY_SERDES) tp->link_config.advertising = (ADVERTISED_1000baseT_Half | @@ -7917,6 +8236,12 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) tp->pci_chip_rev_id = (misc_ctrl_reg >> MISC_HOST_CTRL_CHIPREV_SHIFT); + /* Wrong chip ID in 5752 A0. This code can be removed later + * as A0 is not in production. + */ + if (tp->pci_chip_rev_id == CHIPREV_ID_5752_A0_HW) + tp->pci_chip_rev_id = CHIPREV_ID_5752_A0; + /* Initialize misc host control in PCI block. */ tp->misc_host_ctrl |= (misc_ctrl_reg & MISC_HOST_CTRL_CHIPREV); @@ -7931,11 +8256,15 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) tp->pci_hdr_type = (cacheline_sz_reg >> 16) & 0xff; tp->pci_bist = (cacheline_sz_reg >> 24) & 0xff; + if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750 || + GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5752) + tp->tg3_flags2 |= TG3_FLG2_5750_PLUS; + if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5705) || - (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750)) + (tp->tg3_flags2 & TG3_FLG2_5750_PLUS)) tp->tg3_flags2 |= TG3_FLG2_5705_PLUS; - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) + if (tp->tg3_flags2 & TG3_FLG2_5750_PLUS) tp->tg3_flags2 |= TG3_FLG2_HW_TSO; if (pci_find_capability(tp->pdev, PCI_CAP_ID_EXP) != 0) @@ -8013,6 +8342,31 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) pci_write_config_dword(tp->pdev, TG3PCI_PCISTATE, pci_state_reg); } + /* Get eeprom hw config before calling tg3_set_power_state(). + * In particular, the TG3_FLAG_EEPROM_WRITE_PROT flag must be + * determined before calling tg3_set_power_state() so that + * we know whether or not to switch out of Vaux power. + * When the flag is set, it means that GPIO1 is used for eeprom + * write protect and also implies that it is a LOM where GPIOs + * are not used to switch power. + */ + tg3_get_eeprom_hw_cfg(tp); + + /* Set up tp->grc_local_ctrl before calling tg3_set_power_state(). + * GPIO1 driven high will bring 5700's external PHY out of reset. + * It is also used as eeprom write protect on LOMs. + */ + tp->grc_local_ctrl = GRC_LCLCTRL_INT_ON_ATTN | GRC_LCLCTRL_AUTO_SEEPROM; + if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5700) || + (tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT)) + tp->grc_local_ctrl |= (GRC_LCLCTRL_GPIO_OE1 | + GRC_LCLCTRL_GPIO_OUTPUT1); + /* Unused GPIO3 must be driven as output on 5752 because there + * are no pull-up resistors on unused GPIO pins. + */ + else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5752) + tp->grc_local_ctrl |= GRC_LCLCTRL_GPIO_OE3; + /* Force the chip into D0. */ err = tg3_set_power_state(tp, 0); if (err) { @@ -8065,8 +8419,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) if (tp->pci_chip_rev_id == CHIPREV_ID_5704_A0) tp->tg3_flags2 |= TG3_FLG2_PHY_5704_A0_BUG; - if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5705 || - GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750) + if (tp->tg3_flags2 & TG3_FLG2_5705_PLUS) tp->tg3_flags2 |= TG3_FLG2_PHY_BER_BUG; /* Only 5701 and later support tagged irq status mode. @@ -8628,6 +8981,7 @@ static char * __devinit tg3_phy_string(struct tg3 *tp) case PHY_ID_BCM5704: return "5704"; case PHY_ID_BCM5705: return "5705"; case PHY_ID_BCM5750: return "5750"; + case PHY_ID_BCM5752: return "5752"; case PHY_ID_BCM8002: return "8002/serdes"; case 0: return "serdes"; default: return "unknown"; @@ -8878,7 +9232,7 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, (tr32(WDMAC_MODE) & WDMAC_MODE_ENABLE)) { pci_save_state(tp->pdev); tw32(MEMARB_MODE, MEMARB_MODE_ENABLE); - tg3_halt(tp); + tg3_halt(tp, 1); } err = tg3_test_dma(tp); @@ -9001,7 +9355,7 @@ static int tg3_suspend(struct pci_dev *pdev, pm_message_t state) spin_lock_irq(&tp->lock); spin_lock(&tp->tx_lock); - tg3_halt(tp); + tg3_halt(tp, 1); spin_unlock(&tp->tx_lock); spin_unlock_irq(&tp->lock); diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h index d48887d9032..8de6f21037b 100644 --- a/drivers/net/tg3.h +++ b/drivers/net/tg3.h @@ -125,6 +125,9 @@ #define CHIPREV_ID_5750_A0 0x4000 #define CHIPREV_ID_5750_A1 0x4001 #define CHIPREV_ID_5750_A3 0x4003 +#define CHIPREV_ID_5752_A0_HW 0x5000 +#define CHIPREV_ID_5752_A0 0x6000 +#define CHIPREV_ID_5752_A1 0x6001 #define GET_ASIC_REV(CHIP_REV_ID) ((CHIP_REV_ID) >> 12) #define ASIC_REV_5700 0x07 #define ASIC_REV_5701 0x00 @@ -132,6 +135,7 @@ #define ASIC_REV_5704 0x02 #define ASIC_REV_5705 0x03 #define ASIC_REV_5750 0x04 +#define ASIC_REV_5752 0x06 #define GET_CHIP_REV(CHIP_REV_ID) ((CHIP_REV_ID) >> 8) #define CHIPREV_5700_AX 0x70 #define CHIPREV_5700_BX 0x71 @@ -1307,6 +1311,9 @@ #define GRC_LCLCTRL_CLEARINT 0x00000002 #define GRC_LCLCTRL_SETINT 0x00000004 #define GRC_LCLCTRL_INT_ON_ATTN 0x00000008 +#define GRC_LCLCTRL_GPIO_INPUT3 0x00000020 +#define GRC_LCLCTRL_GPIO_OE3 0x00000040 +#define GRC_LCLCTRL_GPIO_OUTPUT3 0x00000080 #define GRC_LCLCTRL_GPIO_INPUT0 0x00000100 #define GRC_LCLCTRL_GPIO_INPUT1 0x00000200 #define GRC_LCLCTRL_GPIO_INPUT2 0x00000400 @@ -1392,6 +1399,20 @@ #define FLASH_VENDOR_SAIFUN 0x01000003 #define FLASH_VENDOR_SST_SMALL 0x00000001 #define FLASH_VENDOR_SST_LARGE 0x02000001 +#define NVRAM_CFG1_5752VENDOR_MASK 0x03c00003 +#define FLASH_5752VENDOR_ATMEL_EEPROM_64KHZ 0x00000000 +#define FLASH_5752VENDOR_ATMEL_EEPROM_376KHZ 0x02000000 +#define FLASH_5752VENDOR_ATMEL_FLASH_BUFFERED 0x02000003 +#define FLASH_5752VENDOR_ST_M45PE10 0x02400000 +#define FLASH_5752VENDOR_ST_M45PE20 0x02400002 +#define FLASH_5752VENDOR_ST_M45PE40 0x02400001 +#define NVRAM_CFG1_5752PAGE_SIZE_MASK 0x70000000 +#define FLASH_5752PAGE_SIZE_256 0x00000000 +#define FLASH_5752PAGE_SIZE_512 0x10000000 +#define FLASH_5752PAGE_SIZE_1K 0x20000000 +#define FLASH_5752PAGE_SIZE_2K 0x30000000 +#define FLASH_5752PAGE_SIZE_4K 0x40000000 +#define FLASH_5752PAGE_SIZE_264 0x50000000 #define NVRAM_CFG2 0x00007018 #define NVRAM_CFG3 0x0000701c #define NVRAM_SWARB 0x00007020 @@ -2100,6 +2121,9 @@ struct tg3 { #define TG3_FLG2_HW_TSO 0x00010000 #define TG3_FLG2_SERDES_PREEMPHASIS 0x00020000 #define TG3_FLG2_5705_PLUS 0x00040000 +#define TG3_FLG2_5750_PLUS 0x00080000 +#define TG3_FLG2_PROTECTED_NVRAM 0x00100000 +#define TG3_FLG2_USING_MSI 0x00200000 u32 split_mode_max_reqs; #define SPLIT_MODE_5704_MAX_REQ 3 @@ -2145,6 +2169,7 @@ struct tg3 { #define PHY_ID_BCM5704 0x60008190 #define PHY_ID_BCM5705 0x600081a0 #define PHY_ID_BCM5750 0x60008180 +#define PHY_ID_BCM5752 0x60008100 #define PHY_ID_BCM8002 0x60010140 #define PHY_ID_INVALID 0xffffffff #define PHY_ID_REV_MASK 0x0000000f diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig index 35791934a60..66b94668ddd 100644 --- a/drivers/net/wan/Kconfig +++ b/drivers/net/wan/Kconfig @@ -26,7 +26,7 @@ config WAN # There is no way to detect a comtrol sv11 - force it modular for now. config HOSTESS_SV11 tristate "Comtrol Hostess SV-11 support" - depends on WAN && ISA && m + depends on WAN && ISA && m && ISA_DMA_API help Driver for Comtrol Hostess SV-11 network card which operates on low speed synchronous serial links at up to @@ -38,7 +38,7 @@ config HOSTESS_SV11 # The COSA/SRP driver has not been tested as non-modular yet. config COSA tristate "COSA/SRP sync serial boards support" - depends on WAN && ISA && m + depends on WAN && ISA && m && ISA_DMA_API ---help--- Driver for COSA and SRP synchronous serial boards. @@ -127,7 +127,7 @@ config LANMEDIA # There is no way to detect a Sealevel board. Force it modular config SEALEVEL_4021 tristate "Sealevel Systems 4021 support" - depends on WAN && ISA && m + depends on WAN && ISA && m && ISA_DMA_API help This is a driver for the Sealevel Systems ACB 56 serial I/O adapter. diff --git a/drivers/net/wan/cycx_x25.c b/drivers/net/wan/cycx_x25.c index 5b48cd8568f..02d57c0b424 100644 --- a/drivers/net/wan/cycx_x25.c +++ b/drivers/net/wan/cycx_x25.c @@ -436,9 +436,7 @@ static int cycx_wan_new_if(struct wan_device *wandev, struct net_device *dev, } if (err) { - if (chan->local_addr) - kfree(chan->local_addr); - + kfree(chan->local_addr); kfree(chan); return err; } @@ -458,9 +456,7 @@ static int cycx_wan_del_if(struct wan_device *wandev, struct net_device *dev) struct cycx_x25_channel *chan = dev->priv; if (chan->svc) { - if (chan->local_addr) - kfree(chan->local_addr); - + kfree(chan->local_addr); if (chan->state == WAN_CONNECTED) del_timer(&chan->timer); } diff --git a/drivers/net/wan/pc300_drv.c b/drivers/net/wan/pc300_drv.c index d67be2587d4..3e7753b1071 100644 --- a/drivers/net/wan/pc300_drv.c +++ b/drivers/net/wan/pc300_drv.c @@ -3427,7 +3427,7 @@ cpc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { static int first_time = 1; ucchar cpc_rev_id; - int err = 0, eeprom_outdated = 0; + int err, eeprom_outdated = 0; ucshort device_id; pc300_t *card; @@ -3439,15 +3439,21 @@ cpc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) #endif } + if ((err = pci_enable_device(pdev)) < 0) + return err; + card = (pc300_t *) kmalloc(sizeof(pc300_t), GFP_KERNEL); if (card == NULL) { printk("PC300 found at RAM 0x%08lx, " "but could not allocate card structure.\n", pci_resource_start(pdev, 3)); - return -ENOMEM; + err = -ENOMEM; + goto err_disable_dev; } memset(card, 0, sizeof(pc300_t)); + err = -ENODEV; + /* read PCI configuration area */ device_id = ent->device; card->hw.irq = pdev->irq; @@ -3507,7 +3513,6 @@ cpc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) printk("PC300 found at RAM 0x%08x, " "but could not allocate PLX mem region.\n", card->hw.ramphys); - err = -ENODEV; goto err_release_io; } if (!request_mem_region(card->hw.ramphys, card->hw.alloc_ramsize, @@ -3515,7 +3520,6 @@ cpc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) printk("PC300 found at RAM 0x%08x, " "but could not allocate RAM mem region.\n", card->hw.ramphys); - err = -ENODEV; goto err_release_plx; } if (!request_mem_region(card->hw.scaphys, card->hw.scasize, @@ -3523,13 +3527,9 @@ cpc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) printk("PC300 found at RAM 0x%08x, " "but could not allocate SCA mem region.\n", card->hw.ramphys); - err = -ENODEV; goto err_release_ram; } - if ((err = pci_enable_device(pdev)) != 0) - goto err_release_sca; - card->hw.plxbase = ioremap(card->hw.plxphys, card->hw.plxsize); card->hw.rambase = ioremap(card->hw.ramphys, card->hw.alloc_ramsize); card->hw.scabase = ioremap(card->hw.scaphys, card->hw.scasize); @@ -3619,7 +3619,6 @@ err_io_unmap: iounmap(card->hw.falcbase); release_mem_region(card->hw.falcphys, card->hw.falcsize); } -err_release_sca: release_mem_region(card->hw.scaphys, card->hw.scasize); err_release_ram: release_mem_region(card->hw.ramphys, card->hw.alloc_ramsize); @@ -3628,7 +3627,9 @@ err_release_plx: err_release_io: release_region(card->hw.iophys, card->hw.iosize); kfree(card); - return -ENODEV; +err_disable_dev: + pci_disable_device(pdev); + return err; } static void __devexit cpc_remove_one(struct pci_dev *pdev) @@ -3662,6 +3663,7 @@ static void __devexit cpc_remove_one(struct pci_dev *pdev) if (card->hw.irq) free_irq(card->hw.irq, card); kfree(card); + pci_disable_device(pdev); } } diff --git a/drivers/net/wan/pc300_tty.c b/drivers/net/wan/pc300_tty.c index 29f84ad0873..8454bf6caaa 100644 --- a/drivers/net/wan/pc300_tty.c +++ b/drivers/net/wan/pc300_tty.c @@ -400,10 +400,8 @@ static void cpc_tty_close(struct tty_struct *tty, struct file *flip) cpc_tty->buf_rx.last = NULL; } - if (cpc_tty->buf_tx) { - kfree(cpc_tty->buf_tx); - cpc_tty->buf_tx = NULL; - } + kfree(cpc_tty->buf_tx); + cpc_tty->buf_tx = NULL; CPC_TTY_DBG("%s: TTY closed\n",cpc_tty->name); @@ -666,7 +664,7 @@ static void cpc_tty_rx_work(void * data) unsigned long port; int i, j; st_cpc_tty_area *cpc_tty; - volatile st_cpc_rx_buf * buf; + volatile st_cpc_rx_buf *buf; char flags=0,flg_rx=1; struct tty_ldisc *ld; @@ -680,9 +678,9 @@ static void cpc_tty_rx_work(void * data) cpc_tty = &cpc_tty_area[port]; if ((buf=cpc_tty->buf_rx.first) != 0) { - if(cpc_tty->tty) { + if (cpc_tty->tty) { ld = tty_ldisc_ref(cpc_tty->tty); - if(ld) { + if (ld) { if (ld->receive_buf) { CPC_TTY_DBG("%s: call line disc. receive_buf\n",cpc_tty->name); ld->receive_buf(cpc_tty->tty, (char *)(buf->data), &flags, buf->size); @@ -691,7 +689,7 @@ static void cpc_tty_rx_work(void * data) } } cpc_tty->buf_rx.first = cpc_tty->buf_rx.first->next; - kfree((unsigned char *)buf); + kfree(buf); buf = cpc_tty->buf_rx.first; flg_rx = 1; } @@ -733,7 +731,7 @@ static void cpc_tty_rx_disc_frame(pc300ch_t *pc300chan) void cpc_tty_receive(pc300dev_t *pc300dev) { - st_cpc_tty_area *cpc_tty; + st_cpc_tty_area *cpc_tty; pc300ch_t *pc300chan = (pc300ch_t *)pc300dev->chan; pc300_t *card = (pc300_t *)pc300chan->card; int ch = pc300chan->channel; @@ -742,7 +740,7 @@ void cpc_tty_receive(pc300dev_t *pc300dev) int rx_len, rx_aux; volatile unsigned char status; unsigned short first_bd = pc300chan->rx_first_bd; - st_cpc_rx_buf *new=NULL; + st_cpc_rx_buf *new = NULL; unsigned char dsr_rx; if (pc300dev->cpc_tty == NULL) { @@ -762,7 +760,7 @@ void cpc_tty_receive(pc300dev_t *pc300dev) if (status & DST_EOM) { break; } - ptdescr=(pcsca_bd_t __iomem *)(card->hw.rambase+cpc_readl(&ptdescr->next)); + ptdescr = (pcsca_bd_t __iomem *)(card->hw.rambase+cpc_readl(&ptdescr->next)); } if (!rx_len) { @@ -771,10 +769,7 @@ void cpc_tty_receive(pc300dev_t *pc300dev) cpc_writel(card->hw.scabase + DRX_REG(EDAL, ch), RX_BD_ADDR(ch, pc300chan->rx_last_bd)); } - if (new) { - kfree(new); - new = NULL; - } + kfree(new); return; } @@ -787,7 +782,7 @@ void cpc_tty_receive(pc300dev_t *pc300dev) continue; } - new = (st_cpc_rx_buf *) kmalloc(rx_len + sizeof(st_cpc_rx_buf), GFP_ATOMIC); + new = (st_cpc_rx_buf *)kmalloc(rx_len + sizeof(st_cpc_rx_buf), GFP_ATOMIC); if (new == 0) { cpc_tty_rx_disc_frame(pc300chan); continue; diff --git a/drivers/net/wan/sdla_chdlc.c b/drivers/net/wan/sdla_chdlc.c index afbe0024e3e..496d29237e9 100644 --- a/drivers/net/wan/sdla_chdlc.c +++ b/drivers/net/wan/sdla_chdlc.c @@ -3664,15 +3664,10 @@ static void wanpipe_tty_close(struct tty_struct *tty, struct file * filp) chdlc_disable_comm_shutdown(card); unlock_adapter_irq(&card->wandev.lock,&smp_flags); - if (card->tty_buf){ - kfree(card->tty_buf); - card->tty_buf=NULL; - } - - if (card->tty_rx){ - kfree(card->tty_rx); - card->tty_rx=NULL; - } + kfree(card->tty_buf); + card->tty_buf = NULL; + kfree(card->tty_rx); + card->tty_rx = NULL; } return; } diff --git a/drivers/net/wan/x25_asy.c b/drivers/net/wan/x25_asy.c index 8c5cfcb5582..1c540d82555 100644 --- a/drivers/net/wan/x25_asy.c +++ b/drivers/net/wan/x25_asy.c @@ -107,13 +107,9 @@ static struct x25_asy *x25_asy_alloc(void) static void x25_asy_free(struct x25_asy *sl) { /* Free all X.25 frame buffers. */ - if (sl->rbuff) { - kfree(sl->rbuff); - } + kfree(sl->rbuff); sl->rbuff = NULL; - if (sl->xbuff) { - kfree(sl->xbuff); - } + kfree(sl->xbuff); sl->xbuff = NULL; if (!test_and_clear_bit(SLF_INUSE, &sl->flags)) { @@ -134,10 +130,8 @@ static int x25_asy_change_mtu(struct net_device *dev, int newmtu) { printk("%s: unable to grow X.25 buffers, MTU change cancelled.\n", dev->name); - if (xbuff != NULL) - kfree(xbuff); - if (rbuff != NULL) - kfree(rbuff); + kfree(xbuff); + kfree(rbuff); return -ENOMEM; } @@ -169,10 +163,8 @@ static int x25_asy_change_mtu(struct net_device *dev, int newmtu) spin_unlock_bh(&sl->lock); - if (xbuff != NULL) - kfree(xbuff); - if (rbuff != NULL) - kfree(rbuff); + kfree(xbuff); + kfree(rbuff); return 0; } diff --git a/drivers/oprofile/oprofile_files.c b/drivers/oprofile/oprofile_files.c index 9abedeaa567..a72006c08f2 100644 --- a/drivers/oprofile/oprofile_files.c +++ b/drivers/oprofile/oprofile_files.c @@ -18,13 +18,13 @@ unsigned long fs_buffer_size = 131072; unsigned long fs_cpu_buffer_size = 8192; unsigned long fs_buffer_watershed = 32768; /* FIXME: tune */ -static ssize_t depth_read(struct file * file, char * buf, size_t count, loff_t * offset) +static ssize_t depth_read(struct file * file, char __user * buf, size_t count, loff_t * offset) { return oprofilefs_ulong_to_user(backtrace_depth, buf, count, offset); } -static ssize_t depth_write(struct file * file, char const * buf, size_t count, loff_t * offset) +static ssize_t depth_write(struct file * file, char const __user * buf, size_t count, loff_t * offset) { unsigned long val; int retval; diff --git a/drivers/parport/Kconfig b/drivers/parport/Kconfig index 731010e0e6f..16a2e6ae37f 100644 --- a/drivers/parport/Kconfig +++ b/drivers/parport/Kconfig @@ -34,7 +34,7 @@ config PARPORT config PARPORT_PC tristate "PC-style hardware" - depends on PARPORT && (!SPARC64 || PCI) && (!SPARC32 || BROKEN) + depends on PARPORT && (!SPARC64 || PCI) && !SPARC32 ---help--- You should say Y here if you have a PC-style parallel port. All IBM PC compatible computers and some Alphas have PC-style diff --git a/drivers/parport/parport_pc.c b/drivers/parport/parport_pc.c index c5774e7855d..e7f3bcb7900 100644 --- a/drivers/parport/parport_pc.c +++ b/drivers/parport/parport_pc.c @@ -67,6 +67,10 @@ #define PARPORT_PC_MAX_PORTS PARPORT_MAX +#ifdef CONFIG_ISA_DMA_API +#define HAS_DMA +#endif + /* ECR modes */ #define ECR_SPP 00 #define ECR_PS2 01 @@ -610,6 +614,7 @@ dump_parport_state ("leave fifo_write_block_pio", port); return length - left; } +#ifdef HAS_DMA static size_t parport_pc_fifo_write_block_dma (struct parport *port, const void *buf, size_t length) { @@ -732,6 +737,17 @@ dump_parport_state ("enter fifo_write_block_dma", port); dump_parport_state ("leave fifo_write_block_dma", port); return length - left; } +#endif + +static inline size_t parport_pc_fifo_write_block(struct parport *port, + const void *buf, size_t length) +{ +#ifdef HAS_DMA + if (port->dma != PARPORT_DMA_NONE) + return parport_pc_fifo_write_block_dma (port, buf, length); +#endif + return parport_pc_fifo_write_block_pio (port, buf, length); +} /* Parallel Port FIFO mode (ECP chipsets) */ static size_t parport_pc_compat_write_block_pio (struct parport *port, @@ -758,10 +774,7 @@ static size_t parport_pc_compat_write_block_pio (struct parport *port, port->physport->ieee1284.phase = IEEE1284_PH_FWD_DATA; /* Write the data to the FIFO. */ - if (port->dma != PARPORT_DMA_NONE) - written = parport_pc_fifo_write_block_dma (port, buf, length); - else - written = parport_pc_fifo_write_block_pio (port, buf, length); + written = parport_pc_fifo_write_block(port, buf, length); /* Finish up. */ /* For some hardware we don't want to touch the mode until @@ -856,10 +869,7 @@ static size_t parport_pc_ecp_write_block_pio (struct parport *port, port->physport->ieee1284.phase = IEEE1284_PH_FWD_DATA; /* Write the data to the FIFO. */ - if (port->dma != PARPORT_DMA_NONE) - written = parport_pc_fifo_write_block_dma (port, buf, length); - else - written = parport_pc_fifo_write_block_pio (port, buf, length); + written = parport_pc_fifo_write_block(port, buf, length); /* Finish up. */ /* For some hardware we don't want to touch the mode until @@ -2285,6 +2295,7 @@ struct parport *parport_pc_probe_port (unsigned long int base, } #ifdef CONFIG_PARPORT_PC_FIFO +#ifdef HAS_DMA if (p->dma != PARPORT_DMA_NONE) { if (request_dma (p->dma, p->name)) { printk (KERN_WARNING "%s: dma %d in use, " @@ -2306,7 +2317,8 @@ struct parport *parport_pc_probe_port (unsigned long int base, } } } -#endif /* CONFIG_PARPORT_PC_FIFO */ +#endif +#endif } /* Done probing. Now put the port into a sensible start-up state. */ @@ -2367,11 +2379,13 @@ void parport_pc_unregister_port (struct parport *p) if (p->modes & PARPORT_MODE_ECP) release_region(p->base_hi, 3); #ifdef CONFIG_PARPORT_PC_FIFO +#ifdef HAS_DMA if (priv->dma_buf) pci_free_consistent(priv->dev, PAGE_SIZE, priv->dma_buf, priv->dma_handle); -#endif /* CONFIG_PARPORT_PC_FIFO */ +#endif +#endif kfree (p->private_data); parport_put_port(p); kfree (ops); /* hope no-one cached it */ diff --git a/drivers/pci/hotplug.c b/drivers/pci/hotplug.c index d471b3ea5d1..021d0f76bc4 100644 --- a/drivers/pci/hotplug.c +++ b/drivers/pci/hotplug.c @@ -120,6 +120,10 @@ static int pci_visit_bridge (struct pci_visit * fn, /** * pci_visit_dev - scans the pci buses. + * @fn: callback functions that are called while visiting + * @wrapped_dev: the device to scan + * @wrapped_parent: the bus where @wrapped_dev is connected to + * * Every bus and every function is presented to a custom * function that can act upon it. */ diff --git a/drivers/pci/hotplug/ibmphp.h b/drivers/pci/hotplug/ibmphp.h index 5bc039da647..c22e0284d7b 100644 --- a/drivers/pci/hotplug/ibmphp.h +++ b/drivers/pci/hotplug/ibmphp.h @@ -196,7 +196,7 @@ struct ebda_hpc_bus { /******************************************************************** -* THREE TYPE OF HOT PLUG CONTROLER * +* THREE TYPE OF HOT PLUG CONTROLLER * ********************************************************************/ struct isa_ctlr_access { diff --git a/drivers/pci/hotplug/ibmphp_hpc.c b/drivers/pci/hotplug/ibmphp_hpc.c index 6894b548c8c..1a3eb8d3d4c 100644 --- a/drivers/pci/hotplug/ibmphp_hpc.c +++ b/drivers/pci/hotplug/ibmphp_hpc.c @@ -64,7 +64,7 @@ static int to_debug = FALSE; #define WPG_I2C_OR 0x2000 // I2C OR operation //---------------------------------------------------------------------------- -// Command set for I2C Master Operation Setup Regisetr +// Command set for I2C Master Operation Setup Register //---------------------------------------------------------------------------- #define WPG_READATADDR_MASK 0x00010000 // read,bytes,I2C shifted,index #define WPG_WRITEATADDR_MASK 0x40010000 // write,bytes,I2C shifted,index @@ -835,7 +835,7 @@ static void poll_hpc (void) if (ibmphp_shutdown) break; - /* try to get the lock to do some kind of harware access */ + /* try to get the lock to do some kind of hardware access */ down (&semOperations); switch (poll_state) { @@ -906,7 +906,7 @@ static void poll_hpc (void) poll_state = POLL_LATCH_REGISTER; break; } - /* give up the harware semaphore */ + /* give up the hardware semaphore */ up (&semOperations); /* sleep for a short time just for good measure */ msleep(100); diff --git a/drivers/pci/hotplug/ibmphp_pci.c b/drivers/pci/hotplug/ibmphp_pci.c index 2335fac65fb..8122fe734aa 100644 --- a/drivers/pci/hotplug/ibmphp_pci.c +++ b/drivers/pci/hotplug/ibmphp_pci.c @@ -1308,10 +1308,10 @@ static int unconfigure_boot_device (u8 busno, u8 device, u8 function) /* ????????? DO WE NEED TO WRITE ANYTHING INTO THE PCI CONFIG SPACE BACK ?????????? */ } else { /* This is Memory */ - start_address &= PCI_BASE_ADDRESS_MEM_MASK; if (start_address & PCI_BASE_ADDRESS_MEM_PREFETCH) { /* pfmem */ debug ("start address of pfmem is %x\n", start_address); + start_address &= PCI_BASE_ADDRESS_MEM_MASK; if (ibmphp_find_resource (bus, start_address, &pfmem, PFMEM) < 0) { err ("cannot find corresponding PFMEM resource to remove\n"); @@ -1325,6 +1325,8 @@ static int unconfigure_boot_device (u8 busno, u8 device, u8 function) } else { /* regular memory */ debug ("start address of mem is %x\n", start_address); + start_address &= PCI_BASE_ADDRESS_MEM_MASK; + if (ibmphp_find_resource (bus, start_address, &mem, MEM) < 0) { err ("cannot find corresponding MEM resource to remove\n"); return -EIO; @@ -1422,9 +1424,9 @@ static int unconfigure_boot_bridge (u8 busno, u8 device, u8 function) /* ????????? DO WE NEED TO WRITE ANYTHING INTO THE PCI CONFIG SPACE BACK ?????????? */ } else { /* This is Memory */ - start_address &= PCI_BASE_ADDRESS_MEM_MASK; if (start_address & PCI_BASE_ADDRESS_MEM_PREFETCH) { /* pfmem */ + start_address &= PCI_BASE_ADDRESS_MEM_MASK; if (ibmphp_find_resource (bus, start_address, &pfmem, PFMEM) < 0) { err ("cannot find corresponding PFMEM resource to remove\n"); return -EINVAL; @@ -1436,6 +1438,7 @@ static int unconfigure_boot_bridge (u8 busno, u8 device, u8 function) } } else { /* regular memory */ + start_address &= PCI_BASE_ADDRESS_MEM_MASK; if (ibmphp_find_resource (bus, start_address, &mem, MEM) < 0) { err ("cannot find corresponding MEM resource to remove\n"); return -EINVAL; diff --git a/drivers/pci/hotplug/pci_hotplug.h b/drivers/pci/hotplug/pci_hotplug.h index 57ace325168..88d44f7fef2 100644 --- a/drivers/pci/hotplug/pci_hotplug.h +++ b/drivers/pci/hotplug/pci_hotplug.h @@ -150,7 +150,7 @@ struct hotplug_slot_info { * @name: the name of the slot being registered. This string must * be unique amoung slots registered on this system. * @ops: pointer to the &struct hotplug_slot_ops to be used for this slot - * @info: pointer to the &struct hotplug_slot_info for the inital values for + * @info: pointer to the &struct hotplug_slot_info for the initial values for * this slot. * @release: called during pci_hp_deregister to free memory allocated in a * hotplug_slot structure. diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index 72baf749e65..ed1fd8d6178 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -90,6 +90,22 @@ static struct hotplug_slot_ops pciehp_hotplug_slot_ops = { .get_cur_bus_speed = get_cur_bus_speed, }; +/** + * release_slot - free up the memory used by a slot + * @hotplug_slot: slot to free + */ +static void release_slot(struct hotplug_slot *hotplug_slot) +{ + struct slot *slot = hotplug_slot->private; + + dbg("%s - physical_slot = %s\n", __FUNCTION__, hotplug_slot->name); + + kfree(slot->hotplug_slot->info); + kfree(slot->hotplug_slot->name); + kfree(slot->hotplug_slot); + kfree(slot); +} + static int init_slots(struct controller *ctrl) { struct slot *new_slot; @@ -139,7 +155,8 @@ static int init_slots(struct controller *ctrl) /* register this slot with the hotplug pci core */ new_slot->hotplug_slot->private = new_slot; - make_slot_name (new_slot->hotplug_slot->name, SLOT_NAME_SIZE, new_slot); + new_slot->hotplug_slot->release = &release_slot; + make_slot_name(new_slot->hotplug_slot->name, SLOT_NAME_SIZE, new_slot); new_slot->hotplug_slot->ops = &pciehp_hotplug_slot_ops; new_slot->hpc_ops->get_power_status(new_slot, &(new_slot->hotplug_slot->info->power_status)); @@ -188,10 +205,6 @@ static int cleanup_slots (struct controller * ctrl) while (old_slot) { next_slot = old_slot->next; pci_hp_deregister (old_slot->hotplug_slot); - kfree(old_slot->hotplug_slot->info); - kfree(old_slot->hotplug_slot->name); - kfree(old_slot->hotplug_slot); - kfree(old_slot); old_slot = next_slot; } diff --git a/drivers/pci/hotplug/pcihp_skeleton.c b/drivers/pci/hotplug/pcihp_skeleton.c index 6605d6bda52..3194d51c6ec 100644 --- a/drivers/pci/hotplug/pcihp_skeleton.c +++ b/drivers/pci/hotplug/pcihp_skeleton.c @@ -297,7 +297,7 @@ static int __init init_slots(void) hotplug_slot->ops = &skel_hotplug_slot_ops; /* - * Initilize the slot info structure with some known + * Initialize the slot info structure with some known * good values. */ info->power_status = get_power_status(slot); diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 22ecd3b058b..30206ac43c4 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -522,7 +522,7 @@ void pci_scan_msi_device(struct pci_dev *dev) * msi_capability_init - configure device's MSI capability structure * @dev: pointer to the pci_dev data structure of MSI device function * - * Setup the MSI capability structure of device funtion with a single + * Setup the MSI capability structure of device function with a single * MSI vector, regardless of device function is capable of handling * multiple messages. A return of zero indicates the successful setup * of an entry zero with the new MSI vector or non-zero for otherwise. @@ -599,7 +599,7 @@ static int msi_capability_init(struct pci_dev *dev) * msix_capability_init - configure device's MSI-X capability * @dev: pointer to the pci_dev data structure of MSI-X device function * - * Setup the MSI-X capability structure of device funtion with a + * Setup the MSI-X capability structure of device function with a * single MSI-X vector. A return of zero indicates the successful setup of * requested MSI-X entries with allocated vectors or non-zero for otherwise. **/ @@ -1074,7 +1074,7 @@ void pci_disable_msix(struct pci_dev* dev) * msi_remove_pci_irq_vectors - reclaim MSI(X) vectors to unused state * @dev: pointer to the pci_dev data structure of MSI(X) device function * - * Being called during hotplug remove, from which the device funciton + * Being called during hotplug remove, from which the device function * is hot-removed. All previous assigned MSI/MSI-X vectors, if * allocated for this device function, are reclaimed to unused state, * which may be used later on. diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c index 968eb32f292..bc01d34e263 100644 --- a/drivers/pci/pci-acpi.c +++ b/drivers/pci/pci-acpi.c @@ -19,7 +19,7 @@ static u32 ctrlset_buf[3] = {0, 0, 0}; static u32 global_ctrlsets = 0; -u8 OSC_UUID[16] = {0x5B, 0x4D, 0xDB, 0x33, 0xF7, 0x1F, 0x1C, 0x40, 0x96, 0x57, 0x74, 0x41, 0xC0, 0x3D, 0xD7, 0x66}; +static u8 OSC_UUID[16] = {0x5B, 0x4D, 0xDB, 0x33, 0xF7, 0x1F, 0x1C, 0x40, 0x96, 0x57, 0x74, 0x41, 0xC0, 0x3D, 0xD7, 0x66}; static acpi_status acpi_query_osc ( diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 37b7961efc4..fe98553c978 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -318,6 +318,14 @@ static int pci_device_resume(struct device * dev) return 0; } +static void pci_device_shutdown(struct device *dev) +{ + struct pci_dev *pci_dev = to_pci_dev(dev); + struct pci_driver *drv = pci_dev->driver; + + if (drv && drv->shutdown) + drv->shutdown(pci_dev); +} #define kobj_to_pci_driver(obj) container_of(obj, struct device_driver, kobj) #define attr_to_driver_attribute(obj) container_of(obj, struct driver_attribute, attr) @@ -373,7 +381,7 @@ pci_populate_driver_dir(struct pci_driver *drv) * * Adds the driver structure to the list of registered drivers. * Returns a negative value on error, otherwise 0. - * If no error occured, the driver remains registered even if + * If no error occurred, the driver remains registered even if * no device was claimed during registration. */ int pci_register_driver(struct pci_driver *drv) @@ -385,6 +393,7 @@ int pci_register_driver(struct pci_driver *drv) drv->driver.bus = &pci_bus_type; drv->driver.probe = pci_device_probe; drv->driver.remove = pci_device_remove; + drv->driver.shutdown = pci_device_shutdown, drv->driver.owner = drv->owner; drv->driver.kobj.ktype = &pci_driver_kobj_type; pci_init_dynids(&drv->dynids); diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index d57ae71d32b..8568b207f18 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -91,6 +91,7 @@ pci_read_config(struct kobject *kobj, char *buf, loff_t off, size_t count) struct pci_dev *dev = to_pci_dev(container_of(kobj,struct device,kobj)); unsigned int size = 64; loff_t init_off = off; + u8 *data = (u8*) buf; /* Several chips lock up trying to read undefined config space */ if (capable(CAP_SYS_ADMIN)) { @@ -108,30 +109,47 @@ pci_read_config(struct kobject *kobj, char *buf, loff_t off, size_t count) size = count; } - while (off & 3) { - unsigned char val; + if ((off & 1) && size) { + u8 val; pci_read_config_byte(dev, off, &val); - buf[off - init_off] = val; + data[off - init_off] = val; off++; - if (--size == 0) - break; + size--; + } + + if ((off & 3) && size > 2) { + u16 val; + pci_read_config_word(dev, off, &val); + data[off - init_off] = val & 0xff; + data[off - init_off + 1] = (val >> 8) & 0xff; + off += 2; + size -= 2; } while (size > 3) { - unsigned int val; + u32 val; pci_read_config_dword(dev, off, &val); - buf[off - init_off] = val & 0xff; - buf[off - init_off + 1] = (val >> 8) & 0xff; - buf[off - init_off + 2] = (val >> 16) & 0xff; - buf[off - init_off + 3] = (val >> 24) & 0xff; + data[off - init_off] = val & 0xff; + data[off - init_off + 1] = (val >> 8) & 0xff; + data[off - init_off + 2] = (val >> 16) & 0xff; + data[off - init_off + 3] = (val >> 24) & 0xff; off += 4; size -= 4; } - while (size > 0) { - unsigned char val; + if (size >= 2) { + u16 val; + pci_read_config_word(dev, off, &val); + data[off - init_off] = val & 0xff; + data[off - init_off + 1] = (val >> 8) & 0xff; + off += 2; + size -= 2; + } + + if (size > 0) { + u8 val; pci_read_config_byte(dev, off, &val); - buf[off - init_off] = val; + data[off - init_off] = val; off++; --size; } @@ -145,6 +163,7 @@ pci_write_config(struct kobject *kobj, char *buf, loff_t off, size_t count) struct pci_dev *dev = to_pci_dev(container_of(kobj,struct device,kobj)); unsigned int size = count; loff_t init_off = off; + u8 *data = (u8*) buf; if (off > dev->cfg_size) return 0; @@ -152,26 +171,41 @@ pci_write_config(struct kobject *kobj, char *buf, loff_t off, size_t count) size = dev->cfg_size - off; count = size; } - - while (off & 3) { - pci_write_config_byte(dev, off, buf[off - init_off]); + + if ((off & 1) && size) { + pci_write_config_byte(dev, off, data[off - init_off]); off++; - if (--size == 0) - break; + size--; } + + if ((off & 3) && size > 2) { + u16 val = data[off - init_off]; + val |= (u16) data[off - init_off + 1] << 8; + pci_write_config_word(dev, off, val); + off += 2; + size -= 2; + } while (size > 3) { - unsigned int val = buf[off - init_off]; - val |= (unsigned int) buf[off - init_off + 1] << 8; - val |= (unsigned int) buf[off - init_off + 2] << 16; - val |= (unsigned int) buf[off - init_off + 3] << 24; + u32 val = data[off - init_off]; + val |= (u32) data[off - init_off + 1] << 8; + val |= (u32) data[off - init_off + 2] << 16; + val |= (u32) data[off - init_off + 3] << 24; pci_write_config_dword(dev, off, val); off += 4; size -= 4; } + + if (size >= 2) { + u16 val = data[off - init_off]; + val |= (u16) data[off - init_off + 1] << 8; + pci_write_config_word(dev, off, val); + off += 2; + size -= 2; + } - while (size > 0) { - pci_write_config_byte(dev, off, buf[off - init_off]); + if (size) { + pci_write_config_byte(dev, off, data[off - init_off]); off++; --size; } diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index bfbff833526..f04b9ffe415 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -16,6 +16,7 @@ #include <linux/module.h> #include <linux/spinlock.h> #include <asm/dma.h> /* isa_dma_bridge_buggy */ +#include "pci.h" /** @@ -398,10 +399,10 @@ pci_enable_device(struct pci_dev *dev) { int err; - dev->is_enabled = 1; if ((err = pci_enable_device_bars(dev, (1 << PCI_NUM_RESOURCES) - 1))) return err; pci_fixup_device(pci_fixup_enable, dev); + dev->is_enabled = 1; return 0; } @@ -427,16 +428,15 @@ pci_disable_device(struct pci_dev *dev) { u16 pci_command; - dev->is_enabled = 0; - dev->is_busmaster = 0; - pci_read_config_word(dev, PCI_COMMAND, &pci_command); if (pci_command & PCI_COMMAND_MASTER) { pci_command &= ~PCI_COMMAND_MASTER; pci_write_config_word(dev, PCI_COMMAND, pci_command); } + dev->is_busmaster = 0; pcibios_disable_device(dev); + dev->is_enabled = 0; } /** @@ -749,17 +749,6 @@ pci_set_dma_mask(struct pci_dev *dev, u64 mask) } int -pci_dac_set_dma_mask(struct pci_dev *dev, u64 mask) -{ - if (!pci_dac_dma_supported(dev, mask)) - return -EIO; - - dev->dma_mask = mask; - - return 0; -} - -int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask) { if (!pci_dma_supported(dev, mask)) @@ -821,7 +810,6 @@ EXPORT_SYMBOL(pci_set_master); EXPORT_SYMBOL(pci_set_mwi); EXPORT_SYMBOL(pci_clear_mwi); EXPORT_SYMBOL(pci_set_dma_mask); -EXPORT_SYMBOL(pci_dac_set_dma_mask); EXPORT_SYMBOL(pci_set_consistent_dma_mask); EXPORT_SYMBOL(pci_assign_resource); EXPORT_SYMBOL(pci_find_parent_resource); diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 6f0edadd132..b7ae87823c6 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -9,6 +9,7 @@ #include <linux/slab.h> #include <linux/module.h> #include <linux/cpumask.h> +#include "pci.h" #define CARDBUS_LATENCY_TIMER 176 /* secondary latency timer */ #define CARDBUS_RESERVE_BUSNR 3 diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c index 84cc4f620d8..e68bbfb1e7c 100644 --- a/drivers/pci/proc.c +++ b/drivers/pci/proc.c @@ -15,6 +15,7 @@ #include <asm/uaccess.h> #include <asm/byteorder.h> +#include "pci.h" static int proc_initialized; /* = 0 */ diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 15a39805168..026aa04669a 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -18,6 +18,7 @@ #include <linux/pci.h> #include <linux/init.h> #include <linux/delay.h> +#include "pci.h" /* Deal with broken BIOS'es that neglect to enable passive release, which can cause problems in combination with the 82441FX/PPro MTRRs */ @@ -328,6 +329,7 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_12, DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801DB_0, quirk_ich4_lpc_acpi ); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801DB_12, quirk_ich4_lpc_acpi ); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801EB_0, quirk_ich4_lpc_acpi ); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ESB_1, quirk_ich4_lpc_acpi ); /* * VIA ACPI: One IO region pointed to by longword at diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index 3e64ff64b38..838575e3fac 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -14,7 +14,7 @@ /** * pci_enable_rom - enable ROM decoding for a PCI device - * @dev: PCI device to enable + * @pdev: PCI device to enable * * Enable ROM decoding on @dev. This involves simply turning on the last * bit of the PCI ROM BAR. Note that some cards may share address decoders @@ -32,7 +32,7 @@ static void pci_enable_rom(struct pci_dev *pdev) /** * pci_disable_rom - disable ROM decoding for a PCI device - * @dev: PCI device to disable + * @pdev: PCI device to disable * * Disable ROM decoding on a PCI device by turning off the last bit in the * ROM BAR. @@ -47,7 +47,7 @@ static void pci_disable_rom(struct pci_dev *pdev) /** * pci_map_rom - map a PCI ROM to kernel space - * @dev: pointer to pci device struct + * @pdev: pointer to pci device struct * @size: pointer to receive size of pci window over ROM * @return: kernel virtual pointer to image of ROM * @@ -132,7 +132,7 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) /** * pci_map_rom_copy - map a PCI ROM to kernel space, create a copy - * @dev: pointer to pci device struct + * @pdev: pointer to pci device struct * @size: pointer to receive size of pci window over ROM * @return: kernel virtual pointer to image of ROM * @@ -166,7 +166,7 @@ void __iomem *pci_map_rom_copy(struct pci_dev *pdev, size_t *size) /** * pci_unmap_rom - unmap the ROM from kernel space - * @dev: pointer to pci device struct + * @pdev: pointer to pci device struct * @rom: virtual address of the previous mapping * * Remove a mapping of a previously mapped ROM @@ -187,7 +187,7 @@ void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom) /** * pci_remove_rom - disable the ROM and remove its sysfs attribute - * @dev: pointer to pci device struct + * @pdev: pointer to pci device struct * * Remove the rom file in sysfs and disable ROM decoding. */ @@ -206,7 +206,7 @@ void pci_remove_rom(struct pci_dev *pdev) /** * pci_cleanup_rom - internal routine for freeing the ROM copy created * by pci_map_rom_copy called from remove.c - * @dev: pointer to pci device struct + * @pdev: pointer to pci device struct * * Free the copied ROM if we allocated one. */ diff --git a/drivers/pcmcia/pd6729.c b/drivers/pcmcia/pd6729.c index 3f4364341d8..20642f0e7bf 100644 --- a/drivers/pcmcia/pd6729.c +++ b/drivers/pcmcia/pd6729.c @@ -744,7 +744,7 @@ static int __devinit pd6729_pci_probe(struct pci_dev *dev, for (i = 0; i < MAX_SOCKETS; i++) { socket[i].io_base = pci_resource_start(dev, 0); - socket[i].socket.features |= SS_CAP_PCCARD; + socket[i].socket.features |= SS_CAP_PAGE_REGS | SS_CAP_PCCARD; socket[i].socket.map_size = 0x1000; socket[i].socket.irq_mask = mask; socket[i].socket.pci_irq = dev->irq; diff --git a/drivers/pcmcia/ti113x.h b/drivers/pcmcia/ti113x.h index 52c073a9d7e..a8a1d104524 100644 --- a/drivers/pcmcia/ti113x.h +++ b/drivers/pcmcia/ti113x.h @@ -442,6 +442,25 @@ out: } +/* changes the irq of func1 to match that of func0 */ +static int ti12xx_align_irqs(struct yenta_socket *socket, int *old_irq) +{ + struct pci_dev *func0; + + /* find func0 device */ + func0 = pci_get_slot(socket->dev->bus, socket->dev->devfn & ~0x07); + if (!func0) + return 0; + + if (old_irq) + *old_irq = socket->cb_irq; + socket->cb_irq = socket->dev->irq = func0->irq; + + pci_dev_put(func0); + + return 1; +} + /* * ties INTA and INTB together. also changes the devices irq to that of * the function 0 device. call from func1 only. @@ -449,26 +468,22 @@ out: */ static int ti12xx_tie_interrupts(struct yenta_socket *socket, int *old_irq) { - struct pci_dev *func0; u32 sysctl; + int ret; sysctl = config_readl(socket, TI113X_SYSTEM_CONTROL); if (sysctl & TI122X_SCR_INTRTIE) return 0; - /* find func0 device */ - func0 = pci_get_slot(socket->dev->bus, socket->dev->devfn & ~0x07); - if (!func0) + /* align */ + ret = ti12xx_align_irqs(socket, old_irq); + if (!ret) return 0; - /* change the interrupt to match func0, tie 'em up */ - *old_irq = socket->cb_irq; - socket->cb_irq = socket->dev->irq = func0->irq; + /* tie */ sysctl |= TI122X_SCR_INTRTIE; config_writel(socket, TI113X_SYSTEM_CONTROL, sysctl); - pci_dev_put(func0); - return 1; } @@ -489,7 +504,7 @@ static void ti12xx_untie_interrupts(struct yenta_socket *socket, int old_irq) */ static void ti12xx_irqroute_func1(struct yenta_socket *socket) { - u32 mfunc, mfunc_old, devctl; + u32 mfunc, mfunc_old, devctl, sysctl; int pci_irq_status; mfunc = mfunc_old = config_readl(socket, TI122X_MFUNC); @@ -497,6 +512,11 @@ static void ti12xx_irqroute_func1(struct yenta_socket *socket) printk(KERN_INFO "Yenta TI: socket %s, mfunc 0x%08x, devctl 0x%02x\n", pci_name(socket->dev), mfunc, devctl); + /* if IRQs are configured as tied, align irq of func1 with func0 */ + sysctl = config_readl(socket, TI113X_SYSTEM_CONTROL); + if (sysctl & TI122X_SCR_INTRTIE) + ti12xx_align_irqs(socket, NULL); + /* make sure PCI interrupts are enabled before probing */ ti_init(socket); diff --git a/drivers/pnp/manager.c b/drivers/pnp/manager.c index 639e0425348..65ecef73853 100644 --- a/drivers/pnp/manager.c +++ b/drivers/pnp/manager.c @@ -253,7 +253,7 @@ void pnp_init_resource_table(struct pnp_resource_table *table) /** * pnp_clean_resources - clears resources that were not manually set - * @res - the resources to clean + * @res: the resources to clean * */ static void pnp_clean_resource_table(struct pnp_resource_table * res) diff --git a/drivers/pnp/pnpacpi/rsparser.c b/drivers/pnp/pnpacpi/rsparser.c index c0ddb1eb8c4..dd61e09029b 100644 --- a/drivers/pnp/pnpacpi/rsparser.c +++ b/drivers/pnp/pnpacpi/rsparser.c @@ -94,8 +94,8 @@ static void pnpacpi_parse_allocated_dmaresource(struct pnp_resource_table * res, int dma) { int i = 0; - while (!(res->dma_resource[i].flags & IORESOURCE_UNSET) && - i < PNP_MAX_DMA) + while (i < PNP_MAX_DMA && + !(res->dma_resource[i].flags & IORESOURCE_UNSET)) i++; if (i < PNP_MAX_DMA) { res->dma_resource[i].flags = IORESOURCE_DMA; // Also clears _UNSET flag diff --git a/drivers/pnp/pnpbios/rsparser.c b/drivers/pnp/pnpbios/rsparser.c index 618ac15a9e9..79bce7b7574 100644 --- a/drivers/pnp/pnpbios/rsparser.c +++ b/drivers/pnp/pnpbios/rsparser.c @@ -72,7 +72,9 @@ static void pnpbios_parse_allocated_dmaresource(struct pnp_resource_table * res, int dma) { int i = 0; - while (!(res->dma_resource[i].flags & IORESOURCE_UNSET) && i < PNP_MAX_DMA) i++; + while (i < PNP_MAX_DMA && + !(res->dma_resource[i].flags & IORESOURCE_UNSET)) + i++; if (i < PNP_MAX_DMA) { res->dma_resource[i].flags = IORESOURCE_DMA; // Also clears _UNSET flag if (dma == -1) { diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index b755bac6ccb..02cfe244e06 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -7,7 +7,7 @@ * Bugreports.to..: <Linux390@de.ibm.com> * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999-2001 * - * $Revision: 1.158 $ + * $Revision: 1.161 $ */ #include <linux/config.h> @@ -1131,13 +1131,17 @@ __dasd_process_blk_queue(struct dasd_device * device) request_queue_t *queue; struct request *req; struct dasd_ccw_req *cqr; - int nr_queued; + int nr_queued, feature_ro; queue = device->request_queue; /* No queue ? Then there is nothing to do. */ if (queue == NULL) return; + feature_ro = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature_ro < 0) /* no devmap */ + return; + /* * We requeue request from the block device queue to the ccw * queue only in two states. In state DASD_STATE_READY the @@ -1157,8 +1161,8 @@ __dasd_process_blk_queue(struct dasd_device * device) elv_next_request(queue) && nr_queued < DASD_CHANQ_MAX_SIZE) { req = elv_next_request(queue); - if (test_bit(DASD_FLAG_RO, &device->flags) && - rq_data_dir(req) == WRITE) { + + if (feature_ro && rq_data_dir(req) == WRITE) { DBF_DEV_EVENT(DBF_ERR, device, "Rejecting write request %p", req); @@ -1631,6 +1635,7 @@ dasd_setup_queue(struct dasd_device * device) blk_queue_max_hw_segments(device->request_queue, -1L); blk_queue_max_segment_size(device->request_queue, -1L); blk_queue_segment_boundary(device->request_queue, -1L); + blk_queue_ordered(device->request_queue, 1); } /* @@ -1803,13 +1808,17 @@ dasd_generic_set_online (struct ccw_device *cdev, { struct dasd_device *device; - int rc; + int feature_diag, rc; + + feature_diag = dasd_get_feature(cdev, DASD_FEATURE_USEDIAG); + if (feature_diag < 0) + return feature_diag; device = dasd_create_device(cdev); if (IS_ERR(device)) return PTR_ERR(device); - if (test_bit(DASD_FLAG_USE_DIAG, &device->flags)) { + if (feature_diag) { if (!dasd_diag_discipline_pointer) { printk (KERN_WARNING "dasd_generic couldn't online device %s " diff --git a/drivers/s390/block/dasd_cmb.c b/drivers/s390/block/dasd_cmb.c index ed1ab474c0c..4f365bff275 100644 --- a/drivers/s390/block/dasd_cmb.c +++ b/drivers/s390/block/dasd_cmb.c @@ -1,5 +1,5 @@ /* - * linux/drivers/s390/block/dasd_cmb.c ($Revision: 1.6 $) + * linux/drivers/s390/block/dasd_cmb.c ($Revision: 1.9 $) * * Linux on zSeries Channel Measurement Facility support * (dasd device driver interface) @@ -23,7 +23,6 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #include <linux/init.h> -#include <linux/ioctl32.h> #include <linux/module.h> #include <asm/ccwdev.h> #include <asm/cmb.h> @@ -84,27 +83,13 @@ dasd_ioctl_readall_cmb(struct block_device *bdev, int no, long args) static inline int ioctl_reg(unsigned int no, dasd_ioctl_fn_t handler) { - int ret; - ret = dasd_ioctl_no_register(THIS_MODULE, no, handler); -#ifdef CONFIG_COMPAT - if (ret) - return ret; - - ret = register_ioctl32_conversion(no, NULL); - if (ret) - dasd_ioctl_no_unregister(THIS_MODULE, no, handler); -#endif - return ret; + return dasd_ioctl_no_register(THIS_MODULE, no, handler); } static inline void ioctl_unreg(unsigned int no, dasd_ioctl_fn_t handler) { dasd_ioctl_no_unregister(THIS_MODULE, no, handler); -#ifdef CONFIG_COMPAT - unregister_ioctl32_conversion(no); -#endif - } static void diff --git a/drivers/s390/block/dasd_devmap.c b/drivers/s390/block/dasd_devmap.c index ad1841a96c8..1aedc48e5f8 100644 --- a/drivers/s390/block/dasd_devmap.c +++ b/drivers/s390/block/dasd_devmap.c @@ -11,7 +11,7 @@ * functions may not be called from interrupt context. In particular * dasd_get_device is a no-no from interrupt context. * - * $Revision: 1.37 $ + * $Revision: 1.40 $ */ #include <linux/config.h> @@ -513,14 +513,6 @@ dasd_create_device(struct ccw_device *cdev) if (!devmap->device) { devmap->device = device; device->devindex = devmap->devindex; - if (devmap->features & DASD_FEATURE_READONLY) - set_bit(DASD_FLAG_RO, &device->flags); - else - clear_bit(DASD_FLAG_RO, &device->flags); - if (devmap->features & DASD_FEATURE_USEDIAG) - set_bit(DASD_FLAG_USE_DIAG, &device->flags); - else - clear_bit(DASD_FLAG_USE_DIAG, &device->flags); get_device(&cdev->dev); device->cdev = cdev; rc = 0; @@ -651,14 +643,8 @@ dasd_ro_store(struct device *dev, const char *buf, size_t count) devmap->features |= DASD_FEATURE_READONLY; else devmap->features &= ~DASD_FEATURE_READONLY; - if (devmap->device) { - if (devmap->device->gdp) - set_disk_ro(devmap->device->gdp, ro_flag); - if (ro_flag) - set_bit(DASD_FLAG_RO, &devmap->device->flags); - else - clear_bit(DASD_FLAG_RO, &devmap->device->flags); - } + if (devmap->device && devmap->device->gdp) + set_disk_ro(devmap->device->gdp, ro_flag); spin_unlock(&dasd_devmap_lock); return count; } @@ -739,6 +725,45 @@ static struct attribute_group dasd_attr_group = { .attrs = dasd_attrs, }; +/* + * Return value of the specified feature. + */ +int +dasd_get_feature(struct ccw_device *cdev, int feature) +{ + struct dasd_devmap *devmap; + + devmap = dasd_find_busid(cdev->dev.bus_id); + if (IS_ERR(devmap)) + return (int) PTR_ERR(devmap); + + return ((devmap->features & feature) != 0); +} + +/* + * Set / reset given feature. + * Flag indicates wether to set (!=0) or the reset (=0) the feature. + */ +int +dasd_set_feature(struct ccw_device *cdev, int feature, int flag) +{ + struct dasd_devmap *devmap; + + devmap = dasd_find_busid(cdev->dev.bus_id); + if (IS_ERR(devmap)) + return (int) PTR_ERR(devmap); + + spin_lock(&dasd_devmap_lock); + if (flag) + devmap->features |= feature; + else + devmap->features &= ~feature; + + spin_unlock(&dasd_devmap_lock); + return 0; +} + + int dasd_add_sysfs_files(struct ccw_device *cdev) { diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c index 838aedf78a5..811060e10c0 100644 --- a/drivers/s390/block/dasd_eckd.c +++ b/drivers/s390/block/dasd_eckd.c @@ -7,7 +7,7 @@ * Bugreports.to..: <Linux390@de.ibm.com> * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000 * - * $Revision: 1.69 $ + * $Revision: 1.71 $ */ #include <linux/config.h> @@ -1101,7 +1101,8 @@ dasd_eckd_build_cp(struct dasd_device * device, struct request *req) if (dasd_eckd_cdl_special(blk_per_trk, recid)){ rcmd |= 0x8; count = dasd_eckd_cdl_reclen(recid); - if (count < blksize) + if (count < blksize && + rq_data_dir(req) == READ) memset(dst + count, 0xe5, blksize - count); } diff --git a/drivers/s390/block/dasd_genhd.c b/drivers/s390/block/dasd_genhd.c index 1d52db406b2..96c49349701 100644 --- a/drivers/s390/block/dasd_genhd.c +++ b/drivers/s390/block/dasd_genhd.c @@ -9,7 +9,7 @@ * * gendisk related functions for the dasd driver. * - * $Revision: 1.48 $ + * $Revision: 1.50 $ */ #include <linux/config.h> @@ -31,12 +31,16 @@ int dasd_gendisk_alloc(struct dasd_device *device) { struct gendisk *gdp; - int len; + int len, feature_ro; /* Make sure the minor for this device exists. */ if (device->devindex >= DASD_PER_MAJOR) return -EBUSY; + feature_ro = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature_ro < 0) + return feature_ro; + gdp = alloc_disk(1 << DASD_PARTN_BITS); if (!gdp) return -ENOMEM; @@ -71,7 +75,7 @@ dasd_gendisk_alloc(struct dasd_device *device) sprintf(gdp->devfs_name, "dasd/%s", device->cdev->dev.bus_id); - if (test_bit(DASD_FLAG_RO, &device->flags)) + if (feature_ro) set_disk_ro(gdp, 1); gdp->private_data = device; gdp->queue = device->request_queue; diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index 4586e0ecc52..a9f38b23598 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -6,7 +6,7 @@ * Bugreports.to..: <Linux390@de.ibm.com> * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000 * - * $Revision: 1.63 $ + * $Revision: 1.64 $ */ #ifndef DASD_INT_H @@ -329,8 +329,6 @@ struct dasd_device { #define DASD_STOPPED_DC_EIO 16 /* disconnected, return -EIO */ /* per device flags */ -#define DASD_FLAG_RO 0 /* device is read-only */ -#define DASD_FLAG_USE_DIAG 1 /* use diag disciplnie */ #define DASD_FLAG_DSC_ERROR 2 /* return -EIO when disconnected */ #define DASD_FLAG_OFFLINE 3 /* device is in offline processing */ @@ -501,6 +499,9 @@ void dasd_devmap_exit(void); struct dasd_device *dasd_create_device(struct ccw_device *); void dasd_delete_device(struct dasd_device *); +int dasd_get_feature(struct ccw_device *, int); +int dasd_set_feature(struct ccw_device *, int, int); + int dasd_add_sysfs_files(struct ccw_device *); void dasd_remove_sysfs_files(struct ccw_device *); diff --git a/drivers/s390/block/dasd_ioctl.c b/drivers/s390/block/dasd_ioctl.c index f1892baa3b1..980c555aa53 100644 --- a/drivers/s390/block/dasd_ioctl.c +++ b/drivers/s390/block/dasd_ioctl.c @@ -7,6 +7,8 @@ * Bugreports.to..: <Linux390@de.ibm.com> * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999-2001 * + * $Revision: 1.45 $ + * * i/o controls for the dasd driver. */ #include <linux/config.h> @@ -294,6 +296,7 @@ dasd_ioctl_format(struct block_device *bdev, int no, long args) { struct dasd_device *device; struct format_data_t fdata; + int feature_ro; if (!capable(CAP_SYS_ADMIN)) return -EACCES; @@ -304,7 +307,11 @@ dasd_ioctl_format(struct block_device *bdev, int no, long args) if (device == NULL) return -ENODEV; - if (test_bit(DASD_FLAG_RO, &device->flags)) + + feature_ro = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature_ro < 0) + return feature_ro; + if (feature_ro) return -EROFS; if (copy_from_user(&fdata, (void __user *) args, sizeof (struct format_data_t))) @@ -377,7 +384,7 @@ dasd_ioctl_information(struct block_device *bdev, int no, long args) struct dasd_device *device; struct dasd_information2_t *dasd_info; unsigned long flags; - int rc; + int rc, feature_ro; struct ccw_device *cdev; device = bdev->bd_disk->private_data; @@ -387,6 +394,10 @@ dasd_ioctl_information(struct block_device *bdev, int no, long args) if (!device->discipline->fill_info) return -EINVAL; + feature_ro = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature_ro < 0) + return feature_ro; + dasd_info = kmalloc(sizeof(struct dasd_information2_t), GFP_KERNEL); if (dasd_info == NULL) return -ENOMEM; @@ -415,9 +426,8 @@ dasd_ioctl_information(struct block_device *bdev, int no, long args) if ((device->state < DASD_STATE_READY) || (dasd_check_blocksize(device->bp_block))) dasd_info->format = DASD_FORMAT_NONE; - - dasd_info->features |= test_bit(DASD_FLAG_RO, &device->flags) ? - DASD_FEATURE_READONLY : DASD_FEATURE_DEFAULT; + + dasd_info->features |= feature_ro; if (device->discipline) memcpy(dasd_info->type, device->discipline->name, 4); @@ -460,7 +470,7 @@ static int dasd_ioctl_set_ro(struct block_device *bdev, int no, long args) { struct dasd_device *device; - int intval; + int intval, rc; if (!capable(CAP_SYS_ADMIN)) return -EACCES; @@ -472,12 +482,11 @@ dasd_ioctl_set_ro(struct block_device *bdev, int no, long args) device = bdev->bd_disk->private_data; if (device == NULL) return -ENODEV; + set_disk_ro(bdev->bd_disk, intval); - if (intval) - set_bit(DASD_FLAG_RO, &device->flags); - else - clear_bit(DASD_FLAG_RO, &device->flags); - return 0; + rc = dasd_set_feature(device->cdev, DASD_FEATURE_READONLY, intval); + + return rc; } /* diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c index 353d41118c6..d7f19745911 100644 --- a/drivers/s390/block/dasd_proc.c +++ b/drivers/s390/block/dasd_proc.c @@ -9,7 +9,7 @@ * * /proc interface for the dasd driver. * - * $Revision: 1.30 $ + * $Revision: 1.31 $ */ #include <linux/config.h> @@ -54,6 +54,7 @@ dasd_devices_show(struct seq_file *m, void *v) { struct dasd_device *device; char *substr; + int feature; device = dasd_device_from_devindex((unsigned long) v - 1); if (IS_ERR(device)) @@ -77,7 +78,10 @@ dasd_devices_show(struct seq_file *m, void *v) else seq_printf(m, " is ????????"); /* Print devices features. */ - substr = test_bit(DASD_FLAG_RO, &device->flags) ? "(ro)" : " "; + feature = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature < 0) + return 0; + substr = feature ? "(ro)" : " "; seq_printf(m, "%4s: ", substr); /* Print device status information. */ switch ((device != NULL) ? device->state : -1) { diff --git a/drivers/s390/cio/airq.c b/drivers/s390/cio/airq.c index 3720e77b465..83e6a060668 100644 --- a/drivers/s390/cio/airq.c +++ b/drivers/s390/cio/airq.c @@ -45,7 +45,7 @@ s390_register_adapter_interrupt (adapter_int_handler_t handler) else ret = (cmpxchg(&adapter_handler, NULL, handler) ? -EBUSY : 0); if (!ret) - synchronize_kernel(); + synchronize_sched(); /* Allow interrupts to complete. */ sprintf (dbf_txt, "ret:%d", ret); CIO_TRACE_EVENT (4, dbf_txt); @@ -65,7 +65,7 @@ s390_unregister_adapter_interrupt (adapter_int_handler_t handler) ret = -EINVAL; else { adapter_handler = NULL; - synchronize_kernel(); + synchronize_sched(); /* Allow interrupts to complete. */ ret = 0; } sprintf (dbf_txt, "ret:%d", ret); diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index 99ce5a56798..1d9b3f18d8d 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -1,7 +1,7 @@ /* * drivers/s390/cio/cio.c * S/390 common I/O routines -- low level i/o calls - * $Revision: 1.131 $ + * $Revision: 1.133 $ * * Copyright (C) 1999-2002 IBM Deutschland Entwicklung GmbH, * IBM Corporation @@ -228,7 +228,7 @@ cio_start_key (struct subchannel *sch, /* subchannel structure */ int cio_start (struct subchannel *sch, struct ccw1 *cpa, __u8 lpm) { - return cio_start_key(sch, cpa, lpm, default_storage_key); + return cio_start_key(sch, cpa, lpm, PAGE_DEFAULT_KEY); } /* diff --git a/drivers/s390/cio/device_ops.c b/drivers/s390/cio/device_ops.c index 11e260e0b9c..02d01a0de16 100644 --- a/drivers/s390/cio/device_ops.c +++ b/drivers/s390/cio/device_ops.c @@ -1,7 +1,7 @@ /* * drivers/s390/cio/device_ops.c * - * $Revision: 1.55 $ + * $Revision: 1.56 $ * * Copyright (C) 2002 IBM Deutschland Entwicklung GmbH, * IBM Corporation @@ -128,7 +128,7 @@ ccw_device_start(struct ccw_device *cdev, struct ccw1 *cpa, unsigned long intparm, __u8 lpm, unsigned long flags) { return ccw_device_start_key(cdev, cpa, intparm, lpm, - default_storage_key, flags); + PAGE_DEFAULT_KEY, flags); } int @@ -137,7 +137,7 @@ ccw_device_start_timeout(struct ccw_device *cdev, struct ccw1 *cpa, int expires) { return ccw_device_start_timeout_key(cdev, cpa, intparm, lpm, - default_storage_key, flags, + PAGE_DEFAULT_KEY, flags, expires); } diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h index 9ad14db2414..b6daadac4e8 100644 --- a/drivers/s390/cio/qdio.h +++ b/drivers/s390/cio/qdio.h @@ -1,7 +1,9 @@ #ifndef _CIO_QDIO_H #define _CIO_QDIO_H -#define VERSION_CIO_QDIO_H "$Revision: 1.26 $" +#include <asm/page.h> + +#define VERSION_CIO_QDIO_H "$Revision: 1.32 $" #ifdef CONFIG_QDIO_DEBUG #define QDIO_VERBOSE_LEVEL 9 @@ -42,7 +44,7 @@ #define QDIO_Q_LAPS 5 -#define QDIO_STORAGE_KEY 0 +#define QDIO_STORAGE_KEY PAGE_DEFAULT_KEY #define L2_CACHELINE_SIZE 256 #define INDICATORS_PER_CACHELINE (L2_CACHELINE_SIZE/sizeof(__u32)) diff --git a/drivers/s390/crypto/z90main.c b/drivers/s390/crypto/z90main.c index a98c00c0255..9ec29bb41b2 100644 --- a/drivers/s390/crypto/z90main.c +++ b/drivers/s390/crypto/z90main.c @@ -385,8 +385,8 @@ static int z90crypt_release(struct inode *, struct file *); static ssize_t z90crypt_read(struct file *, char __user *, size_t, loff_t *); static ssize_t z90crypt_write(struct file *, const char __user *, size_t, loff_t *); -static int z90crypt_ioctl(struct inode *, struct file *, - unsigned int, unsigned long); +static long z90crypt_unlocked_ioctl(struct file *, unsigned int, unsigned long); +static long z90crypt_compat_ioctl(struct file *, unsigned int, unsigned long); static void z90crypt_reader_task(unsigned long); static void z90crypt_schedule_reader_task(unsigned long); @@ -433,12 +433,15 @@ static atomic_t total_open; static atomic_t z90crypt_step; static struct file_operations z90crypt_fops = { - .owner = THIS_MODULE, - .read = z90crypt_read, - .write = z90crypt_write, - .ioctl = z90crypt_ioctl, - .open = z90crypt_open, - .release = z90crypt_release + .owner = THIS_MODULE, + .read = z90crypt_read, + .write = z90crypt_write, + .unlocked_ioctl = z90crypt_unlocked_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = z90crypt_compat_ioctl, +#endif + .open = z90crypt_open, + .release = z90crypt_release }; #ifndef Z90CRYPT_USE_HOTPLUG @@ -474,14 +477,13 @@ struct ica_rsa_modexpo_32 { // For 32-bit callers compat_uptr_t n_modulus; }; -static int -trans_modexpo32(unsigned int fd, unsigned int cmd, unsigned long arg, - struct file *file) +static long +trans_modexpo32(struct file *filp, unsigned int cmd, unsigned long arg) { struct ica_rsa_modexpo_32 __user *mex32u = compat_ptr(arg); struct ica_rsa_modexpo_32 mex32k; struct ica_rsa_modexpo __user *mex64; - int ret = 0; + long ret = 0; unsigned int i; if (!access_ok(VERIFY_WRITE, mex32u, sizeof(struct ica_rsa_modexpo_32))) @@ -498,7 +500,7 @@ trans_modexpo32(unsigned int fd, unsigned int cmd, unsigned long arg, __put_user(compat_ptr(mex32k.b_key), &mex64->b_key) || __put_user(compat_ptr(mex32k.n_modulus), &mex64->n_modulus)) return -EFAULT; - ret = sys_ioctl(fd, cmd, (unsigned long)mex64); + ret = z90crypt_unlocked_ioctl(filp, cmd, (unsigned long)mex64); if (!ret) if (__get_user(i, &mex64->outputdatalength) || __put_user(i, &mex32u->outputdatalength)) @@ -518,14 +520,13 @@ struct ica_rsa_modexpo_crt_32 { // For 32-bit callers compat_uptr_t u_mult_inv; }; -static int -trans_modexpo_crt32(unsigned int fd, unsigned int cmd, unsigned long arg, - struct file *file) +static long +trans_modexpo_crt32(struct file *filp, unsigned int cmd, unsigned long arg) { struct ica_rsa_modexpo_crt_32 __user *crt32u = compat_ptr(arg); struct ica_rsa_modexpo_crt_32 crt32k; struct ica_rsa_modexpo_crt __user *crt64; - int ret = 0; + long ret = 0; unsigned int i; if (!access_ok(VERIFY_WRITE, crt32u, @@ -546,9 +547,8 @@ trans_modexpo_crt32(unsigned int fd, unsigned int cmd, unsigned long arg, __put_user(compat_ptr(crt32k.np_prime), &crt64->np_prime) || __put_user(compat_ptr(crt32k.nq_prime), &crt64->nq_prime) || __put_user(compat_ptr(crt32k.u_mult_inv), &crt64->u_mult_inv)) - ret = -EFAULT; - if (!ret) - ret = sys_ioctl(fd, cmd, (unsigned long)crt64); + return -EFAULT; + ret = z90crypt_unlocked_ioctl(filp, cmd, (unsigned long)crt64); if (!ret) if (__get_user(i, &crt64->outputdatalength) || __put_user(i, &crt32u->outputdatalength)) @@ -556,66 +556,34 @@ trans_modexpo_crt32(unsigned int fd, unsigned int cmd, unsigned long arg, return ret; } -static int compatible_ioctls[] = { - ICAZ90STATUS, Z90QUIESCE, Z90STAT_TOTALCOUNT, Z90STAT_PCICACOUNT, - Z90STAT_PCICCCOUNT, Z90STAT_PCIXCCCOUNT, Z90STAT_PCIXCCMCL2COUNT, - Z90STAT_PCIXCCMCL3COUNT, Z90STAT_CEX2CCOUNT, Z90STAT_REQUESTQ_COUNT, - Z90STAT_PENDINGQ_COUNT, Z90STAT_TOTALOPEN_COUNT, Z90STAT_DOMAIN_INDEX, - Z90STAT_STATUS_MASK, Z90STAT_QDEPTH_MASK, Z90STAT_PERDEV_REQCNT, -}; - -static void z90_unregister_ioctl32s(void) -{ - int i; - - unregister_ioctl32_conversion(ICARSAMODEXPO); - unregister_ioctl32_conversion(ICARSACRT); - - for(i = 0; i < ARRAY_SIZE(compatible_ioctls); i++) - unregister_ioctl32_conversion(compatible_ioctls[i]); -} - -static int z90_register_ioctl32s(void) -{ - int result, i; - - result = register_ioctl32_conversion(ICARSAMODEXPO, trans_modexpo32); - if (result == -EBUSY) { - unregister_ioctl32_conversion(ICARSAMODEXPO); - result = register_ioctl32_conversion(ICARSAMODEXPO, - trans_modexpo32); - } - if (result) - return result; - result = register_ioctl32_conversion(ICARSACRT, trans_modexpo_crt32); - if (result == -EBUSY) { - unregister_ioctl32_conversion(ICARSACRT); - result = register_ioctl32_conversion(ICARSACRT, - trans_modexpo_crt32); - } - if (result) - return result; - - for(i = 0; i < ARRAY_SIZE(compatible_ioctls); i++) { - result = register_ioctl32_conversion(compatible_ioctls[i], 0); - if (result == -EBUSY) { - unregister_ioctl32_conversion(compatible_ioctls[i]); - result = register_ioctl32_conversion( - compatible_ioctls[i], 0); - } - if (result) - return result; - } - return 0; -} -#else // !CONFIG_COMPAT -static inline void z90_unregister_ioctl32s(void) -{ -} - -static inline int z90_register_ioctl32s(void) +static long +z90crypt_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { - return 0; + switch (cmd) { + case ICAZ90STATUS: + case Z90QUIESCE: + case Z90STAT_TOTALCOUNT: + case Z90STAT_PCICACOUNT: + case Z90STAT_PCICCCOUNT: + case Z90STAT_PCIXCCCOUNT: + case Z90STAT_PCIXCCMCL2COUNT: + case Z90STAT_PCIXCCMCL3COUNT: + case Z90STAT_CEX2CCOUNT: + case Z90STAT_REQUESTQ_COUNT: + case Z90STAT_PENDINGQ_COUNT: + case Z90STAT_TOTALOPEN_COUNT: + case Z90STAT_DOMAIN_INDEX: + case Z90STAT_STATUS_MASK: + case Z90STAT_QDEPTH_MASK: + case Z90STAT_PERDEV_REQCNT: + return z90crypt_unlocked_ioctl(filp, cmd, arg); + case ICARSAMODEXPO: + return trans_modexpo32(filp, cmd, arg); + case ICARSACRT: + return trans_modexpo_crt32(filp, cmd, arg); + default: + return -ENOIOCTLCMD; + } } #endif @@ -730,14 +698,9 @@ z90crypt_init_module(void) reader_timer.expires = jiffies + (READERTIME * HZ / 1000); add_timer(&reader_timer); - if ((result = z90_register_ioctl32s())) - goto init_module_cleanup; - return 0; // success init_module_cleanup: - z90_unregister_ioctl32s(); - #ifndef Z90CRYPT_USE_HOTPLUG if ((nresult = misc_deregister(&z90crypt_misc_device))) PRINTK("misc_deregister failed with %d.\n", nresult); @@ -763,8 +726,6 @@ z90crypt_cleanup_module(void) PDEBUG("PID %d\n", PID()); - z90_unregister_ioctl32s(); - remove_proc_entry("driver/z90crypt", 0); #ifndef Z90CRYPT_USE_HOTPLUG @@ -800,7 +761,7 @@ z90crypt_cleanup_module(void) * z90crypt_release * z90crypt_read * z90crypt_write - * z90crypt_ioctl + * z90crypt_unlocked_ioctl * z90crypt_status * z90crypt_status_write * disable_card @@ -1804,9 +1765,8 @@ z90crypt_rsa(struct priv_data *private_data_p, pid_t pid, * This function is a little long, but it's really just one large switch * statement. */ -static int -z90crypt_ioctl(struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg) +static long +z90crypt_unlocked_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { struct priv_data *private_data_p = filp->private_data; unsigned char *status; diff --git a/drivers/s390/net/smsgiucv.c b/drivers/s390/net/smsgiucv.c index a3d28585956..1e3f7f3c662 100644 --- a/drivers/s390/net/smsgiucv.c +++ b/drivers/s390/net/smsgiucv.c @@ -32,7 +32,7 @@ struct smsg_callback { struct list_head list; char *prefix; int len; - void (*callback)(char *str); + void (*callback)(char *from, char *str); }; MODULE_AUTHOR @@ -55,8 +55,9 @@ smsg_message_pending(iucv_MessagePending *eib, void *pgm_data) { struct smsg_callback *cb; unsigned char *msg; + unsigned char sender[9]; unsigned short len; - int rc; + int rc, i; len = eib->ln1msg2.ipbfln1f; msg = kmalloc(len + 1, GFP_ATOMIC|GFP_DMA); @@ -69,10 +70,18 @@ smsg_message_pending(iucv_MessagePending *eib, void *pgm_data) if (rc == 0) { msg[len] = 0; EBCASC(msg, len); + memcpy(sender, msg, 8); + sender[8] = 0; + /* Remove trailing whitespace from the sender name. */ + for (i = 7; i >= 0; i--) { + if (sender[i] != ' ' && sender[i] != '\t') + break; + sender[i] = 0; + } spin_lock(&smsg_list_lock); list_for_each_entry(cb, &smsg_list, list) if (strncmp(msg + 8, cb->prefix, cb->len) == 0) { - cb->callback(msg + 8); + cb->callback(sender, msg + 8); break; } spin_unlock(&smsg_list_lock); @@ -91,7 +100,7 @@ static struct device_driver smsg_driver = { }; int -smsg_register_callback(char *prefix, void (*callback)(char *str)) +smsg_register_callback(char *prefix, void (*callback)(char *from, char *str)) { struct smsg_callback *cb; @@ -108,7 +117,7 @@ smsg_register_callback(char *prefix, void (*callback)(char *str)) } void -smsg_unregister_callback(char *prefix, void (*callback)(char *str)) +smsg_unregister_callback(char *prefix, void (*callback)(char *from, char *str)) { struct smsg_callback *cb, *tmp; diff --git a/drivers/s390/net/smsgiucv.h b/drivers/s390/net/smsgiucv.h index 04cd8715296..67f5d4f8378 100644 --- a/drivers/s390/net/smsgiucv.h +++ b/drivers/s390/net/smsgiucv.h @@ -5,6 +5,6 @@ * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) */ -int smsg_register_callback(char *, void (*)(char *)); -void smsg_unregister_callback(char *, void (*)(char *)); +int smsg_register_callback(char *, void (*)(char *, char *)); +void smsg_unregister_callback(char *, void (*)(char *, char *)); diff --git a/drivers/sbus/char/rtc.c b/drivers/sbus/char/rtc.c index bf3273eb1c8..9b988baf0b5 100644 --- a/drivers/sbus/char/rtc.c +++ b/drivers/sbus/char/rtc.c @@ -28,10 +28,46 @@ static int rtc_busy = 0; +/* This is the structure layout used by drivers/char/rtc.c, we + * support that driver's ioctls so that things are less messy in + * userspace. + */ +struct rtc_time_generic { + int tm_sec; + int tm_min; + int tm_hour; + int tm_mday; + int tm_mon; + int tm_year; + int tm_wday; + int tm_yday; + int tm_isdst; +}; +#define RTC_AIE_ON _IO('p', 0x01) /* Alarm int. enable on */ +#define RTC_AIE_OFF _IO('p', 0x02) /* ... off */ +#define RTC_UIE_ON _IO('p', 0x03) /* Update int. enable on */ +#define RTC_UIE_OFF _IO('p', 0x04) /* ... off */ +#define RTC_PIE_ON _IO('p', 0x05) /* Periodic int. enable on */ +#define RTC_PIE_OFF _IO('p', 0x06) /* ... off */ +#define RTC_WIE_ON _IO('p', 0x0f) /* Watchdog int. enable on */ +#define RTC_WIE_OFF _IO('p', 0x10) /* ... off */ +#define RTC_RD_TIME _IOR('p', 0x09, struct rtc_time_generic) /* Read RTC time */ +#define RTC_SET_TIME _IOW('p', 0x0a, struct rtc_time_generic) /* Set RTC time */ +#define RTC_ALM_SET _IOW('p', 0x07, struct rtc_time) /* Set alarm time */ +#define RTC_ALM_READ _IOR('p', 0x08, struct rtc_time) /* Read alarm time */ +#define RTC_IRQP_READ _IOR('p', 0x0b, unsigned long) /* Read IRQ rate */ +#define RTC_IRQP_SET _IOW('p', 0x0c, unsigned long) /* Set IRQ rate */ +#define RTC_EPOCH_READ _IOR('p', 0x0d, unsigned long) /* Read epoch */ +#define RTC_EPOCH_SET _IOW('p', 0x0e, unsigned long) /* Set epoch */ +#define RTC_WKALM_SET _IOW('p', 0x0f, struct rtc_wkalrm)/* Set wakeup alarm*/ +#define RTC_WKALM_RD _IOR('p', 0x10, struct rtc_wkalrm)/* Get wakeup alarm*/ +#define RTC_PLL_GET _IOR('p', 0x11, struct rtc_pll_info) /* Get PLL correction */ +#define RTC_PLL_SET _IOW('p', 0x12, struct rtc_pll_info) /* Set PLL correction */ + /* Retrieve the current date and time from the real time clock. */ static void get_rtc_time(struct rtc_time *t) { - void * __iomem regs = mstk48t02_regs; + void __iomem *regs = mstk48t02_regs; u8 tmp; spin_lock_irq(&mostek_lock); @@ -58,7 +94,7 @@ static void get_rtc_time(struct rtc_time *t) /* Set the current date and time inthe real time clock. */ void set_rtc_time(struct rtc_time *t) { - void * __iomem regs = mstk48t02_regs; + void __iomem *regs = mstk48t02_regs; u8 tmp; spin_lock_irq(&mostek_lock); @@ -82,29 +118,87 @@ void set_rtc_time(struct rtc_time *t) spin_unlock_irq(&mostek_lock); } +static int put_rtc_time_generic(void __user *argp, struct rtc_time *tm) +{ + struct rtc_time_generic __user *utm = argp; + + if (__put_user(tm->sec, &utm->tm_sec) || + __put_user(tm->min, &utm->tm_min) || + __put_user(tm->hour, &utm->tm_hour) || + __put_user(tm->dom, &utm->tm_mday) || + __put_user(tm->month, &utm->tm_mon) || + __put_user(tm->year, &utm->tm_year) || + __put_user(tm->dow, &utm->tm_wday) || + __put_user(0, &utm->tm_yday) || + __put_user(0, &utm->tm_isdst)) + return -EFAULT; + + return 0; +} + +static int get_rtc_time_generic(struct rtc_time *tm, void __user *argp) +{ + struct rtc_time_generic __user *utm = argp; + + if (__get_user(tm->sec, &utm->tm_sec) || + __get_user(tm->min, &utm->tm_min) || + __get_user(tm->hour, &utm->tm_hour) || + __get_user(tm->dom, &utm->tm_mday) || + __get_user(tm->month, &utm->tm_mon) || + __get_user(tm->year, &utm->tm_year) || + __get_user(tm->dow, &utm->tm_wday)) + return -EFAULT; + + return 0; +} + static int rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg) { struct rtc_time rtc_tm; void __user *argp = (void __user *)arg; - switch (cmd) - { + switch (cmd) { + /* No interrupt support, return an error + * compatible with drivers/char/rtc.c + */ + case RTC_AIE_OFF: + case RTC_AIE_ON: + case RTC_PIE_OFF: + case RTC_PIE_ON: + case RTC_UIE_OFF: + case RTC_UIE_ON: + case RTC_IRQP_READ: + case RTC_IRQP_SET: + case RTC_EPOCH_SET: + case RTC_EPOCH_READ: + return -EINVAL; + case RTCGET: + case RTC_RD_TIME: memset(&rtc_tm, 0, sizeof(struct rtc_time)); get_rtc_time(&rtc_tm); - if (copy_to_user(argp, &rtc_tm, sizeof(struct rtc_time))) + if (cmd == RTCGET) { + if (copy_to_user(argp, &rtc_tm, + sizeof(struct rtc_time))) + return -EFAULT; + } else if (put_rtc_time_generic(argp, &rtc_tm)) return -EFAULT; return 0; case RTCSET: + case RTC_SET_TIME: if (!capable(CAP_SYS_TIME)) return -EPERM; - if (copy_from_user(&rtc_tm, argp, sizeof(struct rtc_time))) + if (cmd == RTCSET) { + if (copy_from_user(&rtc_tm, argp, + sizeof(struct rtc_time))) + return -EFAULT; + } else if (get_rtc_time_generic(&rtc_tm, argp)) return -EFAULT; set_rtc_time(&rtc_tm); @@ -156,7 +250,7 @@ static int __init rtc_sun_init(void) /* It is possible we are being driven by some other RTC chip * and thus another RTC driver is handling things. */ - if (mstk48t02_regs == 0) + if (!mstk48t02_regs) return -ENODEV; error = misc_register(&rtc_dev); @@ -164,6 +258,7 @@ static int __init rtc_sun_init(void) printk(KERN_ERR "rtc: unable to get misc minor for Mostek\n"); return error; } + printk("rtc_sun_init: Registered Mostek RTC driver.\n"); return 0; } diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 750b11cefd9..1811cb24031 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -260,7 +260,7 @@ config SCSI_3W_9XXX config SCSI_7000FASST tristate "7000FASST SCSI support" - depends on ISA && SCSI + depends on ISA && SCSI && ISA_DMA_API help This driver supports the Western Digital 7000 SCSI host adapter family. Some information is in the source: @@ -295,7 +295,7 @@ config SCSI_AHA152X config SCSI_AHA1542 tristate "Adaptec AHA1542 support" - depends on ISA && SCSI + depends on ISA && SCSI && ISA_DMA_API ---help--- This is support for a SCSI host adapter. It is explained in section 3.4 of the SCSI-HOWTO, available from @@ -515,7 +515,7 @@ config SCSI_SATA_VITESSE config SCSI_BUSLOGIC tristate "BusLogic SCSI support" - depends on (PCI || ISA || MCA) && SCSI && (BROKEN || !SPARC64) + depends on (PCI || ISA || MCA) && SCSI && ISA_DMA_API ---help--- This is support for BusLogic MultiMaster and FlashPoint SCSI Host Adapters. Consult the SCSI-HOWTO, available from @@ -571,7 +571,7 @@ config SCSI_DTC3280 config SCSI_EATA tristate "EATA ISA/EISA/PCI (DPT and generic EATA/DMA-compliant boards) support" - depends on (ISA || EISA || PCI) && SCSI && (BROKEN || !SPARC64) + depends on (ISA || EISA || PCI) && SCSI && ISA_DMA_API ---help--- This driver supports all EATA/DMA-compliant SCSI host adapters. DPT ISA and all EISA I/O addresses are probed looking for the "EATA" @@ -665,7 +665,7 @@ config SCSI_FD_MCS config SCSI_GDTH tristate "Intel/ICP (former GDT SCSI Disk Array) RAID Controller support" - depends on (ISA || EISA || PCI) && SCSI && (BROKEN || !SPARC64) + depends on (ISA || EISA || PCI) && SCSI && ISA_DMA_API ---help--- Formerly called GDT SCSI Disk Array Controller Support. @@ -1416,7 +1416,7 @@ config SCSI_T128 config SCSI_U14_34F tristate "UltraStor 14F/34F support" - depends on ISA && SCSI + depends on ISA && SCSI && ISA_DMA_API ---help--- This is support for the UltraStor 14F and 34F SCSI-2 host adapters. The source at <file:drivers/scsi/u14-34f.c> contains some diff --git a/drivers/scsi/NCR53C9x.c b/drivers/scsi/NCR53C9x.c index 3c86655a5f3..74b93564a25 100644 --- a/drivers/scsi/NCR53C9x.c +++ b/drivers/scsi/NCR53C9x.c @@ -1824,7 +1824,10 @@ static int esp_do_data(struct NCR_ESP *esp, struct ESP_regs *eregs) /* loop */ while (hmuch) { int j, fifo_stuck = 0, newphase; - unsigned long flags, timeout; + unsigned long timeout; +#if 0 + unsigned long flags; +#endif #if 0 if ( i % 10 ) ESPDATA(("\r")); diff --git a/drivers/scsi/esp.c b/drivers/scsi/esp.c index d8ab73b6803..f6900538be9 100644 --- a/drivers/scsi/esp.c +++ b/drivers/scsi/esp.c @@ -49,6 +49,8 @@ #include <scsi/scsi_host.h> #include <scsi/scsi_tcq.h> +#define DRV_VERSION "1.101" + #define DEBUG_ESP /* #define DEBUG_ESP_HME */ /* #define DEBUG_ESP_DATA */ @@ -1145,7 +1147,7 @@ static int __init esp_detect(struct scsi_host_template *tpnt) static struct sbus_dev esp_dev; int esps_in_use = 0; - espchain = 0; + espchain = NULL; if (sun4_esp_physaddr) { memset (&esp_dev, 0, sizeof(esp_dev)); @@ -2511,7 +2513,7 @@ static inline void esp_reconnect(struct esp *esp, struct scsi_cmnd *sp) ESPLOG(("esp%d: Weird, being reselected but disconnected " "command queue is empty.\n", esp->esp_id)); esp->snip = 0; - esp->current_SC = 0; + esp->current_SC = NULL; sp->SCp.phase = not_issued; append_SC(&esp->issue_SC, sp); } @@ -4146,7 +4148,7 @@ static int esp_work_bus(struct esp *esp) } static espfunc_t isvc_vector[] = { - 0, + NULL, esp_do_phase_determine, esp_do_resetbus, esp_finish_reset, @@ -4398,5 +4400,8 @@ static struct scsi_host_template driver_template = { #include "scsi_module.c" +MODULE_DESCRIPTION("EnhancedScsiProcessor Sun SCSI driver"); +MODULE_AUTHOR("David S. Miller (davem@redhat.com)"); MODULE_LICENSE("GPL"); +MODULE_VERSION(DRV_VERSION); diff --git a/drivers/scsi/pas16.c b/drivers/scsi/pas16.c index e70dedb0d0a..7976947c032 100644 --- a/drivers/scsi/pas16.c +++ b/drivers/scsi/pas16.c @@ -137,7 +137,7 @@ static unsigned short pas16_addr = 0; static int pas16_irq = 0; -int scsi_irq_translate[] = +static const int scsi_irq_translate[] = { 0, 0, 1, 2, 3, 4, 5, 6, 0, 0, 7, 8, 9, 0, 10, 11 }; /* The default_irqs array contains values used to set the irq into the @@ -145,7 +145,7 @@ int scsi_irq_translate[] = * irq jumpers on the board). The first value in the array will be * assigned to logical board 0, the next to board 1, etc. */ -int default_irqs[] __initdata = +static int default_irqs[] __initdata = { PAS16_DEFAULT_BOARD_1_IRQ, PAS16_DEFAULT_BOARD_2_IRQ, PAS16_DEFAULT_BOARD_3_IRQ, @@ -177,7 +177,7 @@ static struct base { #define NO_BASES (sizeof (bases) / sizeof (struct base)) -unsigned short pas16_offset[ 8 ] = +static const unsigned short pas16_offset[ 8 ] = { 0x1c00, /* OUTPUT_DATA_REG */ 0x1c01, /* INITIATOR_COMMAND_REG */ diff --git a/drivers/scsi/pluto.c b/drivers/scsi/pluto.c index 7bb0a2e5674..c01b7191fcf 100644 --- a/drivers/scsi/pluto.c +++ b/drivers/scsi/pluto.c @@ -45,7 +45,7 @@ static struct ctrl_inquiry { Scsi_Cmnd cmd; char inquiry[256]; fc_channel *fc; -} *fcs __initdata = { 0 }; +} *fcs __initdata; static int fcscount __initdata = 0; static atomic_t fcss __initdata = ATOMIC_INIT(0); DECLARE_MUTEX_LOCKED(fc_sem); diff --git a/drivers/scsi/qla2xxx/qla_sup.c b/drivers/scsi/qla2xxx/qla_sup.c index 0e75fbb77b6..32583bbb487 100644 --- a/drivers/scsi/qla2xxx/qla_sup.c +++ b/drivers/scsi/qla2xxx/qla_sup.c @@ -85,7 +85,7 @@ qla2x00_unlock_nvram_access(scsi_qla_host_t *ha) void qla2x00_release_nvram_protection(scsi_qla_host_t *ha) { - device_reg_t *reg; + device_reg_t __iomem *reg; uint32_t word; reg = ha->iobase; diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 1a135f38e78..2bf1ee2b47b 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -584,7 +584,7 @@ static int scsi_request_sense(struct scsi_cmnd *scmd) memcpy(scmd->cmnd, generic_sense, sizeof(generic_sense)); - scsi_result = kmalloc(252, GFP_ATOMIC | (scmd->device->host->hostt->unchecked_isa_dma) ? __GFP_DMA : 0); + scsi_result = kmalloc(252, GFP_ATOMIC | ((scmd->device->host->hostt->unchecked_isa_dma) ? __GFP_DMA : 0)); if (unlikely(!scsi_result)) { diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 287d197a7c1..cca772624ae 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -801,7 +801,7 @@ static int scsi_probe_and_add_lun(struct scsi_target *starget, if (!sreq) goto out_free_sdev; result = kmalloc(256, GFP_ATOMIC | - (shost->unchecked_isa_dma) ? __GFP_DMA : 0); + ((shost->unchecked_isa_dma) ? __GFP_DMA : 0)); if (!result) goto out_free_sreq; diff --git a/drivers/scsi/sym53c416.c b/drivers/scsi/sym53c416.c index f26c3a29e63..ebfddd40ce6 100644 --- a/drivers/scsi/sym53c416.c +++ b/drivers/scsi/sym53c416.c @@ -809,7 +809,7 @@ static int sym53c416_host_reset(Scsi_Cmnd *SCpnt) /* printk("sym53c416_reset\n"); */ base = SCpnt->device->host->io_port; /* search scsi_id - fixme, we shouldnt need to iterate for this! */ - for(i = 0; i < host_index && scsi_id != -1; i++) + for(i = 0; i < host_index && scsi_id == -1; i++) if(hosts[i].base == base) scsi_id = hosts[i].scsi_id; outb(RESET_CHIP, base + COMMAND_REG); diff --git a/drivers/scsi/ultrastor.c b/drivers/scsi/ultrastor.c index a00095cc74c..97f4d9112b4 100644 --- a/drivers/scsi/ultrastor.c +++ b/drivers/scsi/ultrastor.c @@ -945,7 +945,7 @@ static int ultrastor_abort(Scsi_Cmnd *SCpnt) config.mscp[mscp_index].SCint, SCpnt); #endif if (config.mscp[mscp_index].SCint == 0) - return FAILURE; + return FAILED; if (config.mscp[mscp_index].SCint != SCpnt) panic("Bad abort"); config.mscp[mscp_index].SCint = NULL; diff --git a/drivers/serial/21285.c b/drivers/serial/21285.c index f8504b0adeb..33fbda79f35 100644 --- a/drivers/serial/21285.c +++ b/drivers/serial/21285.c @@ -110,7 +110,7 @@ static irqreturn_t serial21285_rx_chars(int irq, void *dev_id, struct pt_regs *r port->icount.rx++; rxs = *CSR_RXSTAT | RXSTAT_DUMMY_READ; - if (rxs & RXSTAT_ANYERR) { + if (unlikely(rxs & RXSTAT_ANYERR)) { if (rxs & RXSTAT_PARITY) port->icount.parity++; else if (rxs & RXSTAT_FRAME) diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index 218b69372c0..0d9358608fd 100644 --- a/drivers/serial/8250.c +++ b/drivers/serial/8250.c @@ -51,7 +51,7 @@ * share_irqs - whether we pass SA_SHIRQ to request_irq(). This option * is unsafe when used on edge-triggered interrupts. */ -unsigned int share_irqs = SERIAL8250_SHARE_IRQS; +static unsigned int share_irqs = SERIAL8250_SHARE_IRQS; /* * Debugging. diff --git a/drivers/serial/8250_hp300.c b/drivers/serial/8250_hp300.c index b8d51eb56bf..4315afe9c08 100644 --- a/drivers/serial/8250_hp300.c +++ b/drivers/serial/8250_hp300.c @@ -9,15 +9,15 @@ #include <linux/init.h> #include <linux/string.h> #include <linux/kernel.h> -#include <linux/tty.h> #include <linux/serial.h> -#include <linux/serialP.h> #include <linux/serial_core.h> #include <linux/delay.h> #include <linux/dio.h> #include <linux/console.h> #include <asm/io.h> +#include "8250.h" + #if !defined(CONFIG_HPDCA) && !defined(CONFIG_HPAPCI) #warning CONFIG_8250 defined but neither CONFIG_HPDCA nor CONFIG_HPAPCI defined, are you sure? #endif @@ -163,7 +163,7 @@ int __init hp300_setup_serial_console(void) static int __devinit hpdca_init_one(struct dio_dev *d, const struct dio_device_id *ent) { - struct serial_struct serial_req; + struct uart_port port; int line; #ifdef CONFIG_SERIAL_8250_CONSOLE @@ -172,21 +172,22 @@ static int __devinit hpdca_init_one(struct dio_dev *d, return 0; } #endif - memset(&serial_req, 0, sizeof(struct serial_struct)); + memset(&port, 0, sizeof(struct uart_port)); /* Memory mapped I/O */ - serial_req.io_type = SERIAL_IO_MEM; - serial_req.flags = UPF_SKIP_TEST | UPF_SHARE_IRQ | UPF_BOOT_AUTOCONF; - serial_req.irq = d->ipl; - serial_req.baud_base = HPDCA_BAUD_BASE; - serial_req.iomap_base = (d->resource.start + UART_OFFSET); - serial_req.iomem_base = (char *)(serial_req.iomap_base + DIO_VIRADDRBASE); - serial_req.iomem_reg_shift = 1; - line = register_serial(&serial_req); + port.iotype = UPIO_MEM; + port.flags = UPF_SKIP_TEST | UPF_SHARE_IRQ | UPF_BOOT_AUTOCONF; + port.irq = d->ipl; + port.uartclk = HPDCA_BAUD_BASE * 16; + port.mapbase = (d->resource.start + UART_OFFSET); + port.membase = (char *)(port.mapbase + DIO_VIRADDRBASE); + port.regshift = 1; + port.dev = &d->dev; + line = serial8250_register_port(&port); if (line < 0) { printk(KERN_NOTICE "8250_hp300: register_serial() DCA scode %d" - " irq %d failed\n", d->scode, serial_req.irq); + " irq %d failed\n", d->scode, port.irq); return -ENOMEM; } @@ -209,7 +210,7 @@ static int __init hp300_8250_init(void) #ifdef CONFIG_HPAPCI int line; unsigned long base; - struct serial_struct serial_req; + struct uart_port uport; struct hp300_port *port; int i; #endif @@ -251,25 +252,25 @@ static int __init hp300_8250_init(void) if (!port) return -ENOMEM; - memset(&serial_req, 0, sizeof(struct serial_struct)); + memset(&uport, 0, sizeof(struct uart_port)); base = (FRODO_BASE + FRODO_APCI_OFFSET(i)); /* Memory mapped I/O */ - serial_req.io_type = SERIAL_IO_MEM; - serial_req.flags = UPF_SKIP_TEST | UPF_SHARE_IRQ | UPF_BOOT_AUTOCONF; + uport.iotype = UPIO_MEM; + uport.flags = UPF_SKIP_TEST | UPF_SHARE_IRQ | UPF_BOOT_AUTOCONF; /* XXX - no interrupt support yet */ - serial_req.irq = 0; - serial_req.baud_base = HPAPCI_BAUD_BASE; - serial_req.iomap_base = base; - serial_req.iomem_base = (char *)(serial_req.iomap_base + DIO_VIRADDRBASE); - serial_req.iomem_reg_shift = 2; + uport.irq = 0; + uport.uartclk = HPAPCI_BAUD_BASE * 16; + uport.mapbase = base; + uport.membase = (char *)(base + DIO_VIRADDRBASE); + uport.regshift = 2; - line = register_serial(&serial_req); + line = serial8250_register_port(&uport); if (line < 0) { printk(KERN_NOTICE "8250_hp300: register_serial() APCI %d" - " irq %d failed\n", i, serial_req.irq); + " irq %d failed\n", i, uport.irq); kfree(port); continue; } @@ -299,7 +300,7 @@ static void __devexit hpdca_remove_one(struct dio_dev *d) /* Disable board-interrupts */ out_8(d->resource.start + DIO_VIRADDRBASE + DCA_IC, 0); } - unregister_serial(line); + serial8250_unregister_port(line); } #endif @@ -309,7 +310,7 @@ static void __exit hp300_8250_exit(void) struct hp300_port *port, *to_free; for (port = hp300_ports; port; ) { - unregister_serial(port->line); + serial8250_unregister_port(port->line); to_free = port; port = port->next; kfree(to_free); diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c index f8d90d0ecfe..de54bdc5398 100644 --- a/drivers/serial/8250_pci.c +++ b/drivers/serial/8250_pci.c @@ -1009,6 +1009,8 @@ get_pci_irq(struct pci_dev *dev, struct pci_board *board, int idx) * n = number of serial ports * baud = baud rate * + * This table is sorted by (in order): baud, bt, bn, n. + * * Please note: in theory if n = 1, _bt infix should make no difference. * ie, pbn_b0_1_115200 is the same as pbn_b0_bt_1_115200 */ diff --git a/drivers/serial/amba-pl010.c b/drivers/serial/amba-pl010.c index 484f6fb900b..f2a5e2933c4 100644 --- a/drivers/serial/amba-pl010.c +++ b/drivers/serial/amba-pl010.c @@ -172,7 +172,7 @@ pl010_rx_chars(struct uart_port *port) * out of the main execution path */ rsr = UART_GET_RSR(port) | UART_DUMMY_RSR_RX; - if (rsr & UART01x_RSR_ANY) { + if (unlikely(rsr & UART01x_RSR_ANY)) { if (rsr & UART01x_RSR_BE) { rsr &= ~(UART01x_RSR_FE | UART01x_RSR_PE); port->icount.brk++; diff --git a/drivers/serial/amba-pl011.c b/drivers/serial/amba-pl011.c index ff658a830f3..d5cbef3fe8b 100644 --- a/drivers/serial/amba-pl011.c +++ b/drivers/serial/amba-pl011.c @@ -137,7 +137,7 @@ pl011_rx_chars(struct uart_amba_port *uap) * out of the main execution path */ rsr = readw(uap->port.membase + UART01x_RSR) | UART_DUMMY_RSR_RX; - if (rsr & UART01x_RSR_ANY) { + if (unlikely(rsr & UART01x_RSR_ANY)) { if (rsr & UART01x_RSR_BE) { rsr &= ~(UART01x_RSR_FE | UART01x_RSR_PE); uap->port.icount.brk++; diff --git a/drivers/serial/clps711x.c b/drivers/serial/clps711x.c index 16592fae47f..6242f3090a9 100644 --- a/drivers/serial/clps711x.c +++ b/drivers/serial/clps711x.c @@ -116,54 +116,43 @@ static irqreturn_t clps711xuart_int_rx(int irq, void *dev_id, struct pt_regs *re * Note that the error handling code is * out of the main execution path */ - if (ch & UART_ANY_ERR) - goto handle_error; + if (unlikely(ch & UART_ANY_ERR)) { + if (ch & UARTDR_PARERR) + port->icount.parity++; + else if (ch & UARTDR_FRMERR) + port->icount.frame++; + if (ch & UARTDR_OVERR) + port->icount.overrun++; - if (uart_handle_sysrq_char(port, ch, regs)) - goto ignore_char; + ch &= port->read_status_mask; - error_return: - tty_insert_flip_char(tty, ch, flg); - ignore_char: - status = clps_readl(SYSFLG(port)); - } - out: - tty_flip_buffer_push(tty); - return IRQ_HANDLED; + if (ch & UARTDR_PARERR) + flg = TTY_PARITY; + else if (ch & UARTDR_FRMERR) + flg = TTY_FRAME; - handle_error: - if (ch & UARTDR_PARERR) - port->icount.parity++; - else if (ch & UARTDR_FRMERR) - port->icount.frame++; - if (ch & UARTDR_OVERR) - port->icount.overrun++; - - if (ch & port->ignore_status_mask) { - if (++ignored > 100) - goto out; - goto ignore_char; - } - ch &= port->read_status_mask; +#ifdef SUPPORT_SYSRQ + port->sysrq = 0; +#endif + } - if (ch & UARTDR_PARERR) - flg = TTY_PARITY; - else if (ch & UARTDR_FRMERR) - flg = TTY_FRAME; + if (uart_handle_sysrq_char(port, ch, regs)) + goto ignore_char; - if (ch & UARTDR_OVERR) { /* * CHECK: does overrun affect the current character? * ASSUMPTION: it does not. */ - tty_insert_flip_char(tty, ch, flg); - ch = 0; - flg = TTY_OVERRUN; + if ((ch & port->ignore_status_mask & ~RXSTAT_OVERRUN) == 0) + tty_insert_flip_char(tty, ch, flg); + if ((ch & ~port->ignore_status_mask & RXSTAT_OVERRUN) == 0) + tty_insert_flip_char(tty, 0, TTY_OVERRUN); + + ignore_char: + status = clps_readl(SYSFLG(port)); } -#ifdef SUPPORT_SYSRQ - port->sysrq = 0; -#endif - goto error_return; + tty_flip_buffer_push(tty); + return IRQ_HANDLED; } static irqreturn_t clps711xuart_int_tx(int irq, void *dev_id, struct pt_regs *regs) diff --git a/drivers/serial/imx.c b/drivers/serial/imx.c index c682c6308cd..01a8726a3f9 100644 --- a/drivers/serial/imx.c +++ b/drivers/serial/imx.c @@ -321,18 +321,39 @@ static void imx_break_ctl(struct uart_port *port, int break_state) #define TXTL 2 /* reset default */ #define RXTL 1 /* reset default */ +static int imx_setup_ufcr(struct imx_port *sport, unsigned int mode) +{ + unsigned int val; + unsigned int ufcr_rfdiv; + + /* set receiver / transmitter trigger level. + * RFDIV is set such way to satisfy requested uartclk value + */ + val = TXTL<<10 | RXTL; + ufcr_rfdiv = (imx_get_perclk1() + sport->port.uartclk / 2) / sport->port.uartclk; + + if(!ufcr_rfdiv) + ufcr_rfdiv = 1; + + if(ufcr_rfdiv >= 7) + ufcr_rfdiv = 6; + else + ufcr_rfdiv = 6 - ufcr_rfdiv; + + val |= UFCR_RFDIV & (ufcr_rfdiv << 7); + + UFCR((u32)sport->port.membase) = val; + + return 0; +} + static int imx_startup(struct uart_port *port) { struct imx_port *sport = (struct imx_port *)port; int retval; - unsigned int val; unsigned long flags; - /* set receiver / transmitter trigger level. We assume - * that RFDIV has been set by the arch setup or by the bootloader. - */ - val = (UFCR((u32)sport->port.membase) & UFCR_RFDIV) | TXTL<<10 | RXTL; - UFCR((u32)sport->port.membase) = val; + imx_setup_ufcr(sport, 0); /* disable the DREN bit (Data Ready interrupt enable) before * requesting IRQs @@ -737,9 +758,12 @@ static void __init imx_console_get_options(struct imx_port *sport, int *baud, int *parity, int *bits) { + if ( UCR1((u32)sport->port.membase) | UCR1_UARTEN ) { /* ok, the port was enabled */ unsigned int ucr2, ubir,ubmr, uartclk; + unsigned int baud_raw; + unsigned int ucfr_rfdiv; ucr2 = UCR2((u32)sport->port.membase); @@ -758,9 +782,35 @@ imx_console_get_options(struct imx_port *sport, int *baud, ubir = UBIR((u32)sport->port.membase) & 0xffff; ubmr = UBMR((u32)sport->port.membase) & 0xffff; - uartclk = sport->port.uartclk; - *baud = ((uartclk/16) * (ubir + 1)) / (ubmr + 1); + + ucfr_rfdiv = (UFCR((u32)sport->port.membase) & UFCR_RFDIV) >> 7; + if (ucfr_rfdiv == 6) + ucfr_rfdiv = 7; + else + ucfr_rfdiv = 6 - ucfr_rfdiv; + + uartclk = imx_get_perclk1(); + uartclk /= ucfr_rfdiv; + + { /* + * The next code provides exact computation of + * baud_raw = round(((uartclk/16) * (ubir + 1)) / (ubmr + 1)) + * without need of float support or long long division, + * which would be required to prevent 32bit arithmetic overflow + */ + unsigned int mul = ubir + 1; + unsigned int div = 16 * (ubmr + 1); + unsigned int rem = uartclk % div; + + baud_raw = (uartclk / div) * mul; + baud_raw += (rem * mul + div / 2) / div; + *baud = (baud_raw + 50) / 100 * 100; + } + + if(*baud != baud_raw) + printk(KERN_INFO "Serial: Console IMX rounded baud rate from %d to %d\n", + baud_raw, *baud); } } @@ -787,6 +837,8 @@ imx_console_setup(struct console *co, char *options) else imx_console_get_options(sport, &baud, &parity, &bits); + imx_setup_ufcr(sport, 0); + return uart_set_options(&sport->port, co, baud, parity, bits, flow); } diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index d054f126570..ba4e13a22a5 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -838,7 +838,7 @@ static int inline port_init(struct ioc4_port *port) port->ip_tx_prod = readl(&port->ip_serial_regs->stcir) & PROD_CONS_MASK; writel(port->ip_tx_prod, &port->ip_serial_regs->stpir); port->ip_rx_cons = readl(&port->ip_serial_regs->srpir) & PROD_CONS_MASK; - writel(port->ip_rx_cons, &port->ip_serial_regs->srcir); + writel(port->ip_rx_cons | IOC4_SRCIR_ARM, &port->ip_serial_regs->srcir); /* Disable interrupts for this 16550 */ uart = port->ip_uart_regs; @@ -1272,8 +1272,9 @@ static inline int set_rx_timeout(struct ioc4_port *port, int timeout) * and set the rx threshold to that amount. There are 4 chars * per ring entry, so we'll divide the number of chars that will * arrive in timeout by 4. + * So .... timeout * baud / 10 / HZ / 4, with HZ = 100. */ - threshold = timeout * port->ip_baud / 10 / HZ / 4; + threshold = timeout * port->ip_baud / 4000; if (threshold == 0) threshold = 1; /* otherwise we'll intr all the time! */ @@ -1285,8 +1286,10 @@ static inline int set_rx_timeout(struct ioc4_port *port, int timeout) writel(port->ip_sscr, &port->ip_serial_regs->sscr); - /* Now set the rx timeout to the given value */ - timeout = timeout * IOC4_SRTR_HZ / HZ; + /* Now set the rx timeout to the given value + * again timeout * IOC4_SRTR_HZ / HZ + */ + timeout = timeout * IOC4_SRTR_HZ / 100; if (timeout > IOC4_SRTR_CNT) timeout = IOC4_SRTR_CNT; @@ -1380,7 +1383,7 @@ config_port(struct ioc4_port *port, if (port->ip_tx_lowat == 0) port->ip_tx_lowat = 1; - set_rx_timeout(port, port->ip_rx_timeout); + set_rx_timeout(port, 2); return 0; } @@ -1685,8 +1688,8 @@ ioc4_change_speed(struct uart_port *the_port, { struct ioc4_port *port = get_ioc4_port(the_port); int baud, bits; - unsigned cflag, cval; - int new_parity = 0, new_parity_enable = 0, new_stop = 1, new_data = 8; + unsigned cflag; + int new_parity = 0, new_parity_enable = 0, new_stop = 0, new_data = 8; struct uart_info *info = the_port->info; cflag = new_termios->c_cflag; @@ -1694,48 +1697,35 @@ ioc4_change_speed(struct uart_port *the_port, switch (cflag & CSIZE) { case CS5: new_data = 5; - cval = 0x00; bits = 7; break; case CS6: new_data = 6; - cval = 0x01; bits = 8; break; case CS7: new_data = 7; - cval = 0x02; bits = 9; break; case CS8: new_data = 8; - cval = 0x03; bits = 10; break; default: /* cuz we always need a default ... */ new_data = 5; - cval = 0x00; bits = 7; break; } if (cflag & CSTOPB) { - cval |= 0x04; bits++; new_stop = 1; } if (cflag & PARENB) { - cval |= UART_LCR_PARITY; bits++; new_parity_enable = 1; - } - if (cflag & PARODD) { - cval |= UART_LCR_EPAR; - new_parity = 1; - } - if (cflag & IGNPAR) { - cval &= ~UART_LCR_PARITY; - new_parity_enable = 0; + if (cflag & PARODD) + new_parity = 1; } baud = uart_get_baud_rate(the_port, new_termios, old_termios, MIN_BAUD_SUPPORTED, MAX_BAUD_SUPPORTED); @@ -1765,10 +1755,15 @@ ioc4_change_speed(struct uart_port *the_port, the_port->ignore_status_mask &= ~N_DATA_READY; } - if (cflag & CRTSCTS) + if (cflag & CRTSCTS) { info->flags |= ASYNC_CTS_FLOW; - else + port->ip_sscr |= IOC4_SSCR_HFC_EN; + } + else { info->flags &= ~ASYNC_CTS_FLOW; + port->ip_sscr &= ~IOC4_SSCR_HFC_EN; + } + writel(port->ip_sscr, &port->ip_serial_regs->sscr); /* Set the configuration and proper notification call */ DPRINT_CONFIG(("%s : port 0x%p cflag 0%o " @@ -1825,12 +1820,6 @@ static inline int ic4_startup_local(struct uart_port *the_port) /* set the speed of the serial port */ ioc4_change_speed(the_port, info->tty->termios, (struct termios *)0); - /* enable hardware flow control - after ioc4_change_speed because - * ASYNC_CTS_FLOW is set there */ - if (info->flags & ASYNC_CTS_FLOW) { - port->ip_sscr |= IOC4_SSCR_HFC_EN; - writel(port->ip_sscr, &port->ip_serial_regs->sscr); - } info->flags |= UIF_INITIALIZED; return 0; } @@ -1847,7 +1836,6 @@ static void ioc4_cb_output_lowat(struct ioc4_port *port) } } - /** * handle_intr - service any interrupts for the given port - 2nd level * called via sd_intr diff --git a/drivers/serial/jsm/jsm.h b/drivers/serial/jsm/jsm.h index e0717611c94..777829fa330 100644 --- a/drivers/serial/jsm/jsm.h +++ b/drivers/serial/jsm/jsm.h @@ -393,7 +393,6 @@ int jsm_tty_init(struct jsm_board *); int jsm_uart_port_init(struct jsm_board *); int jsm_remove_uart_port(struct jsm_board *); void jsm_input(struct jsm_channel *ch); -void jsm_carrier(struct jsm_channel *ch); void jsm_check_queue_flow_control(struct jsm_channel *ch); #endif diff --git a/drivers/serial/jsm/jsm_neo.c b/drivers/serial/jsm/jsm_neo.c index 9b79c1ff6c7..3a11a69feb4 100644 --- a/drivers/serial/jsm/jsm_neo.c +++ b/drivers/serial/jsm/jsm_neo.c @@ -688,7 +688,7 @@ static void neo_flush_uart_read(struct jsm_channel *ch) /* * No locks are assumed to be held when calling this function. */ -void neo_clear_break(struct jsm_channel *ch, int force) +static void neo_clear_break(struct jsm_channel *ch, int force) { unsigned long lock_flags; diff --git a/drivers/serial/jsm/jsm_tty.c b/drivers/serial/jsm/jsm_tty.c index 24fe76c2883..98de2258fd0 100644 --- a/drivers/serial/jsm/jsm_tty.c +++ b/drivers/serial/jsm/jsm_tty.c @@ -31,6 +31,8 @@ #include "jsm.h" +static void jsm_carrier(struct jsm_channel *ch); + static inline int jsm_get_mstat(struct jsm_channel *ch) { unsigned char mstat; @@ -755,7 +757,7 @@ void jsm_input(struct jsm_channel *ch) jsm_printk(IOCTL, INFO, &ch->ch_bd->pci_dev, "finish\n"); } -void jsm_carrier(struct jsm_channel *ch) +static void jsm_carrier(struct jsm_channel *ch) { struct jsm_board *bd; diff --git a/drivers/serial/mpsc.c b/drivers/serial/mpsc.c index d0dfc3cf924..a8314aee2ab 100644 --- a/drivers/serial/mpsc.c +++ b/drivers/serial/mpsc.c @@ -329,8 +329,8 @@ mpsc_sdma_stop(struct mpsc_port_info *pi) mpsc_sdma_cmd(pi, SDMA_SDCM_AR | SDMA_SDCM_AT); /* Clear the SDMA current and first TX and RX pointers */ - mpsc_sdma_set_tx_ring(pi, 0); - mpsc_sdma_set_rx_ring(pi, 0); + mpsc_sdma_set_tx_ring(pi, NULL); + mpsc_sdma_set_rx_ring(pi, NULL); /* Disable interrupts */ mpsc_sdma_intr_mask(pi, 0xf); @@ -1540,8 +1540,8 @@ mpsc_shared_unmap_regs(void) MPSC_SDMA_INTR_REG_BLOCK_SIZE); } - mpsc_shared_regs.mpsc_routing_base = 0; - mpsc_shared_regs.sdma_intr_base = 0; + mpsc_shared_regs.mpsc_routing_base = NULL; + mpsc_shared_regs.sdma_intr_base = NULL; mpsc_shared_regs.mpsc_routing_base_p = 0; mpsc_shared_regs.sdma_intr_base_p = 0; @@ -1678,9 +1678,9 @@ mpsc_drv_unmap_regs(struct mpsc_port_info *pi) release_mem_region(pi->brg_base_p, MPSC_BRG_REG_BLOCK_SIZE); } - pi->mpsc_base = 0; - pi->sdma_base = 0; - pi->brg_base = 0; + pi->mpsc_base = NULL; + pi->sdma_base = NULL; + pi->brg_base = NULL; pi->mpsc_base_p = 0; pi->sdma_base_p = 0; diff --git a/drivers/serial/mpsc.h b/drivers/serial/mpsc.h index 1f7294b7095..678dbcf06c8 100644 --- a/drivers/serial/mpsc.h +++ b/drivers/serial/mpsc.h @@ -83,8 +83,8 @@ struct mpsc_shared_regs { phys_addr_t mpsc_routing_base_p; phys_addr_t sdma_intr_base_p; - void *mpsc_routing_base; - void *sdma_intr_base; + void __iomem *mpsc_routing_base; + void __iomem *sdma_intr_base; u32 MPSC_MRR_m; u32 MPSC_RCRR_m; @@ -120,9 +120,9 @@ struct mpsc_port_info { phys_addr_t brg_base_p; /* Virtual addresses of various blocks of registers (from platform) */ - void *mpsc_base; - void *sdma_base; - void *brg_base; + void __iomem *mpsc_base; + void __iomem *sdma_base; + void __iomem *brg_base; /* Descriptor ring and buffer allocations */ void *dma_region; diff --git a/drivers/serial/s3c2410.c b/drivers/serial/s3c2410.c index bd6782aeb83..435750d40a4 100644 --- a/drivers/serial/s3c2410.c +++ b/drivers/serial/s3c2410.c @@ -364,7 +364,7 @@ s3c24xx_serial_rx_chars(int irq, void *dev_id, struct pt_regs *regs) flag = TTY_NORMAL; port->icount.rx++; - if (uerstat & S3C2410_UERSTAT_ANY) { + if (unlikely(uerstat & S3C2410_UERSTAT_ANY)) { dbg("rxerr: port ch=0x%02x, rxs=0x%08x\n", ch, uerstat); diff --git a/drivers/serial/sa1100.c b/drivers/serial/sa1100.c index 086065210d1..157218bc6c6 100644 --- a/drivers/serial/sa1100.c +++ b/drivers/serial/sa1100.c @@ -214,56 +214,39 @@ sa1100_rx_chars(struct sa1100_port *sport, struct pt_regs *regs) * note that the error handling code is * out of the main execution path */ - if (status & UTSR1_TO_SM(UTSR1_PRE | UTSR1_FRE | UTSR1_ROR)) - goto handle_error; + if (status & UTSR1_TO_SM(UTSR1_PRE | UTSR1_FRE | UTSR1_ROR)) { + if (status & UTSR1_TO_SM(UTSR1_PRE)) + sport->port.icount.parity++; + else if (status & UTSR1_TO_SM(UTSR1_FRE)) + sport->port.icount.frame++; + if (status & UTSR1_TO_SM(UTSR1_ROR)) + sport->port.icount.overrun++; + + status &= sport->port.read_status_mask; + + if (status & UTSR1_TO_SM(UTSR1_PRE)) + flg = TTY_PARITY; + else if (status & UTSR1_TO_SM(UTSR1_FRE)) + flg = TTY_FRAME; + +#ifdef SUPPORT_SYSRQ + sport->port.sysrq = 0; +#endif + } if (uart_handle_sysrq_char(&sport->port, ch, regs)) goto ignore_char; - error_return: - tty_insert_flip_char(tty, ch, flg); + if ((status & port->ignore_status_mask & ~UTSR1_TO_SM(UTSR1_ROR)) == 0) + tty_insert_flip_char(tty, ch, flg); + if (status & ~port->ignore_status_mask & UTSR1_TO_SM(UTSR1_ROR)) + tty_insert_flip_char(tty, 0, TTY_OVERRUN); + ignore_char: status = UTSR1_TO_SM(UART_GET_UTSR1(sport)) | UTSR0_TO_SM(UART_GET_UTSR0(sport)); } - out: tty_flip_buffer_push(tty); - return; - - handle_error: - if (status & UTSR1_TO_SM(UTSR1_PRE)) - sport->port.icount.parity++; - else if (status & UTSR1_TO_SM(UTSR1_FRE)) - sport->port.icount.frame++; - if (status & UTSR1_TO_SM(UTSR1_ROR)) - sport->port.icount.overrun++; - - if (status & sport->port.ignore_status_mask) { - if (++ignored > 100) - goto out; - goto ignore_char; - } - - status &= sport->port.read_status_mask; - - if (status & UTSR1_TO_SM(UTSR1_PRE)) - flg = TTY_PARITY; - else if (status & UTSR1_TO_SM(UTSR1_FRE)) - flg = TTY_FRAME; - - if (status & UTSR1_TO_SM(UTSR1_ROR)) { - /* - * overrun does *not* affect the character - * we read from the FIFO - */ - tty_insert_flip_char(tty, ch, flg); - ch = 0; - flg = TTY_OVERRUN; - } -#ifdef SUPPORT_SYSRQ - sport->port.sysrq = 0; -#endif - goto error_return; } static void sa1100_tx_chars(struct sa1100_port *sport) diff --git a/drivers/serial/serial_cs.c b/drivers/serial/serial_cs.c index 9034f9ad37c..6eeb48f6a48 100644 --- a/drivers/serial/serial_cs.c +++ b/drivers/serial/serial_cs.c @@ -107,6 +107,13 @@ struct serial_info { int line[4]; }; +struct serial_cfg_mem { + tuple_t tuple; + cisparse_t parse; + u_char buf[256]; +}; + + static void serial_config(dev_link_t * link); static int serial_event(event_t event, int priority, event_callback_args_t * args); @@ -357,14 +364,24 @@ static int simple_config(dev_link_t *link) static int size_table[2] = { 8, 16 }; client_handle_t handle = link->handle; struct serial_info *info = link->priv; - tuple_t tuple; - u_char buf[256]; - cisparse_t parse; - cistpl_cftable_entry_t *cf = &parse.cftable_entry; + struct serial_cfg_mem *cfg_mem; + tuple_t *tuple; + u_char *buf; + cisparse_t *parse; + cistpl_cftable_entry_t *cf; config_info_t config; int i, j, try; int s; + cfg_mem = kmalloc(sizeof(struct serial_cfg_mem), GFP_KERNEL); + if (!cfg_mem) + return -1; + + tuple = &cfg_mem->tuple; + parse = &cfg_mem->parse; + cf = &parse->cftable_entry; + buf = cfg_mem->buf; + /* If the card is already configured, look up the port and irq */ i = pcmcia_get_configuration_info(handle, &config); if ((i == CS_SUCCESS) && (config.Attributes & CONF_VALID_CLIENT)) { @@ -377,21 +394,23 @@ static int simple_config(dev_link_t *link) port = config.BasePort1 + 0x28; info->slave = 1; } - if (info->slave) + if (info->slave) { + kfree(cfg_mem); return setup_serial(handle, info, port, config.AssignedIRQ); + } } link->conf.Vcc = config.Vcc; /* First pass: look for a config entry that looks normal. */ - tuple.TupleData = (cisdata_t *) buf; - tuple.TupleOffset = 0; - tuple.TupleDataMax = 255; - tuple.Attributes = 0; - tuple.DesiredTuple = CISTPL_CFTABLE_ENTRY; + tuple->TupleData = (cisdata_t *) buf; + tuple->TupleOffset = 0; + tuple->TupleDataMax = 255; + tuple->Attributes = 0; + tuple->DesiredTuple = CISTPL_CFTABLE_ENTRY; /* Two tries: without IO aliases, then with aliases */ for (s = 0; s < 2; s++) { for (try = 0; try < 2; try++) { - i = first_tuple(handle, &tuple, &parse); + i = first_tuple(handle, tuple, parse); while (i != CS_NO_MORE_ITEMS) { if (i != CS_SUCCESS) goto next_entry; @@ -409,14 +428,14 @@ static int simple_config(dev_link_t *link) goto found_port; } next_entry: - i = next_tuple(handle, &tuple, &parse); + i = next_tuple(handle, tuple, parse); } } } /* Second pass: try to find an entry that isn't picky about its base address, then try to grab any standard serial port address, and finally try to get any free port. */ - i = first_tuple(handle, &tuple, &parse); + i = first_tuple(handle, tuple, parse); while (i != CS_NO_MORE_ITEMS) { if ((i == CS_SUCCESS) && (cf->io.nwin > 0) && ((cf->io.flags & CISTPL_IO_LINES_MASK) <= 3)) { @@ -429,7 +448,7 @@ next_entry: goto found_port; } } - i = next_tuple(handle, &tuple, &parse); + i = next_tuple(handle, tuple, parse); } found_port: @@ -437,6 +456,7 @@ next_entry: printk(KERN_NOTICE "serial_cs: no usable port range found, giving up\n"); cs_error(link->handle, RequestIO, i); + kfree(cfg_mem); return -1; } @@ -450,9 +470,10 @@ next_entry: i = pcmcia_request_configuration(link->handle, &link->conf); if (i != CS_SUCCESS) { cs_error(link->handle, RequestConfiguration, i); + kfree(cfg_mem); return -1; } - + kfree(cfg_mem); return setup_serial(handle, info, link->io.BasePort1, link->irq.AssignedIRQ); } @@ -460,29 +481,39 @@ static int multi_config(dev_link_t * link) { client_handle_t handle = link->handle; struct serial_info *info = link->priv; - tuple_t tuple; - u_char buf[256]; - cisparse_t parse; - cistpl_cftable_entry_t *cf = &parse.cftable_entry; + struct serial_cfg_mem *cfg_mem; + tuple_t *tuple; + u_char *buf; + cisparse_t *parse; + cistpl_cftable_entry_t *cf; config_info_t config; - int i, base2 = 0; + int i, rc, base2 = 0; + + cfg_mem = kmalloc(sizeof(struct serial_cfg_mem), GFP_KERNEL); + if (!cfg_mem) + return -1; + tuple = &cfg_mem->tuple; + parse = &cfg_mem->parse; + cf = &parse->cftable_entry; + buf = cfg_mem->buf; i = pcmcia_get_configuration_info(handle, &config); if (i != CS_SUCCESS) { cs_error(handle, GetConfigurationInfo, i); - return -1; + rc = -1; + goto free_cfg_mem; } link->conf.Vcc = config.Vcc; - tuple.TupleData = (cisdata_t *) buf; - tuple.TupleOffset = 0; - tuple.TupleDataMax = 255; - tuple.Attributes = 0; - tuple.DesiredTuple = CISTPL_CFTABLE_ENTRY; + tuple->TupleData = (cisdata_t *) buf; + tuple->TupleOffset = 0; + tuple->TupleDataMax = 255; + tuple->Attributes = 0; + tuple->DesiredTuple = CISTPL_CFTABLE_ENTRY; /* First, look for a generic full-sized window */ link->io.NumPorts1 = info->multi * 8; - i = first_tuple(handle, &tuple, &parse); + i = first_tuple(handle, tuple, parse); while (i != CS_NO_MORE_ITEMS) { /* The quad port cards have bad CIS's, so just look for a window larger than 8 ports and assume it will be right */ @@ -497,14 +528,14 @@ static int multi_config(dev_link_t * link) if (i == CS_SUCCESS) break; } - i = next_tuple(handle, &tuple, &parse); + i = next_tuple(handle, tuple, parse); } /* If that didn't work, look for two windows */ if (i != CS_SUCCESS) { link->io.NumPorts1 = link->io.NumPorts2 = 8; info->multi = 2; - i = first_tuple(handle, &tuple, &parse); + i = first_tuple(handle, tuple, parse); while (i != CS_NO_MORE_ITEMS) { if ((i == CS_SUCCESS) && (cf->io.nwin == 2)) { link->conf.ConfigIndex = cf->index; @@ -517,13 +548,14 @@ static int multi_config(dev_link_t * link) if (i == CS_SUCCESS) break; } - i = next_tuple(handle, &tuple, &parse); + i = next_tuple(handle, tuple, parse); } } if (i != CS_SUCCESS) { cs_error(link->handle, RequestIO, i); - return -1; + rc = -1; + goto free_cfg_mem; } i = pcmcia_request_irq(link->handle, &link->irq); @@ -541,7 +573,8 @@ static int multi_config(dev_link_t * link) i = pcmcia_request_configuration(link->handle, &link->conf); if (i != CS_SUCCESS) { cs_error(link->handle, RequestConfiguration, i); - return -1; + rc = -1; + goto free_cfg_mem; } /* The Oxford Semiconductor OXCF950 cards are in fact single-port: @@ -554,17 +587,23 @@ static int multi_config(dev_link_t * link) setup_serial(handle, info, link->io.BasePort1, link->irq.AssignedIRQ); outb(12, base2 + 1); } - return 0; + rc = 0; + goto free_cfg_mem; } setup_serial(handle, info, link->io.BasePort1, link->irq.AssignedIRQ); /* The Nokia cards are not really multiport cards */ - if (info->manfid == MANFID_NOKIA) - return 0; + if (info->manfid == MANFID_NOKIA) { + rc = 0; + goto free_cfg_mem; + } for (i = 0; i < info->multi - 1; i++) - setup_serial(handle, info, base2 + (8 * i), link->irq.AssignedIRQ); - - return 0; + setup_serial(handle, info, base2 + (8 * i), + link->irq.AssignedIRQ); + rc = 0; +free_cfg_mem: + kfree(cfg_mem); + return rc; } /*====================================================================== @@ -579,39 +618,49 @@ void serial_config(dev_link_t * link) { client_handle_t handle = link->handle; struct serial_info *info = link->priv; - tuple_t tuple; - u_short buf[128]; - cisparse_t parse; - cistpl_cftable_entry_t *cf = &parse.cftable_entry; + struct serial_cfg_mem *cfg_mem; + tuple_t *tuple; + u_char *buf; + cisparse_t *parse; + cistpl_cftable_entry_t *cf; int i, last_ret, last_fn; DEBUG(0, "serial_config(0x%p)\n", link); - tuple.TupleData = (cisdata_t *) buf; - tuple.TupleOffset = 0; - tuple.TupleDataMax = 255; - tuple.Attributes = 0; + cfg_mem = kmalloc(sizeof(struct serial_cfg_mem), GFP_KERNEL); + if (!cfg_mem) + goto failed; + + tuple = &cfg_mem->tuple; + parse = &cfg_mem->parse; + cf = &parse->cftable_entry; + buf = cfg_mem->buf; + + tuple->TupleData = (cisdata_t *) buf; + tuple->TupleOffset = 0; + tuple->TupleDataMax = 255; + tuple->Attributes = 0; /* Get configuration register information */ - tuple.DesiredTuple = CISTPL_CONFIG; - last_ret = first_tuple(handle, &tuple, &parse); + tuple->DesiredTuple = CISTPL_CONFIG; + last_ret = first_tuple(handle, tuple, parse); if (last_ret != CS_SUCCESS) { last_fn = ParseTuple; goto cs_failed; } - link->conf.ConfigBase = parse.config.base; - link->conf.Present = parse.config.rmask[0]; + link->conf.ConfigBase = parse->config.base; + link->conf.Present = parse->config.rmask[0]; /* Configure card */ link->state |= DEV_CONFIG; /* Is this a compliant multifunction card? */ - tuple.DesiredTuple = CISTPL_LONGLINK_MFC; - tuple.Attributes = TUPLE_RETURN_COMMON | TUPLE_RETURN_LINK; - info->multi = (first_tuple(handle, &tuple, &parse) == CS_SUCCESS); + tuple->DesiredTuple = CISTPL_LONGLINK_MFC; + tuple->Attributes = TUPLE_RETURN_COMMON | TUPLE_RETURN_LINK; + info->multi = (first_tuple(handle, tuple, parse) == CS_SUCCESS); /* Is this a multiport card? */ - tuple.DesiredTuple = CISTPL_MANFID; - if (first_tuple(handle, &tuple, &parse) == CS_SUCCESS) { + tuple->DesiredTuple = CISTPL_MANFID; + if (first_tuple(handle, tuple, parse) == CS_SUCCESS) { info->manfid = le16_to_cpu(buf[0]); for (i = 0; i < MULTI_COUNT; i++) if ((info->manfid == multi_id[i].manfid) && @@ -623,13 +672,13 @@ void serial_config(dev_link_t * link) /* Another check for dual-serial cards: look for either serial or multifunction cards that ask for appropriate IO port ranges */ - tuple.DesiredTuple = CISTPL_FUNCID; + tuple->DesiredTuple = CISTPL_FUNCID; if ((info->multi == 0) && - ((first_tuple(handle, &tuple, &parse) != CS_SUCCESS) || - (parse.funcid.func == CISTPL_FUNCID_MULTI) || - (parse.funcid.func == CISTPL_FUNCID_SERIAL))) { - tuple.DesiredTuple = CISTPL_CFTABLE_ENTRY; - if (first_tuple(handle, &tuple, &parse) == CS_SUCCESS) { + ((first_tuple(handle, tuple, parse) != CS_SUCCESS) || + (parse->funcid.func == CISTPL_FUNCID_MULTI) || + (parse->funcid.func == CISTPL_FUNCID_SERIAL))) { + tuple->DesiredTuple = CISTPL_CFTABLE_ENTRY; + if (first_tuple(handle, tuple, parse) == CS_SUCCESS) { if ((cf->io.nwin == 1) && (cf->io.win[0].len % 8 == 0)) info->multi = cf->io.win[0].len >> 3; if ((cf->io.nwin == 2) && (cf->io.win[0].len == 8) && @@ -664,6 +713,7 @@ void serial_config(dev_link_t * link) link->dev = &info->node[0]; link->state &= ~DEV_CONFIG_PENDING; + kfree(cfg_mem); return; cs_failed: @@ -671,6 +721,7 @@ void serial_config(dev_link_t * link) failed: serial_remove(link); link->state &= ~DEV_CONFIG_PENDING; + kfree(cfg_mem); } /*====================================================================== diff --git a/drivers/serial/serial_lh7a40x.c b/drivers/serial/serial_lh7a40x.c index 4ce3a41f161..85cfa08d3ba 100644 --- a/drivers/serial/serial_lh7a40x.c +++ b/drivers/serial/serial_lh7a40x.c @@ -162,7 +162,7 @@ lh7a40xuart_rx_chars (struct uart_port* port) flag = TTY_NORMAL; ++port->icount.rx; - if (data & RxError) { /* Quick check, short-circuit */ + if (unlikely(data & RxError)) { /* Quick check, short-circuit */ if (data & RxBreak) { data &= ~(RxFramingError | RxParityError); ++port->icount.brk; diff --git a/drivers/serial/sn_console.c b/drivers/serial/sn_console.c index ffaab9b90fd..fee6418e84c 100644 --- a/drivers/serial/sn_console.c +++ b/drivers/serial/sn_console.c @@ -787,7 +787,7 @@ static void __init sn_sal_switch_to_interrupts(struct sn_cons_port *port) static void sn_sal_console_write(struct console *, const char *, unsigned); static int __init sn_sal_console_setup(struct console *, char *); -extern struct uart_driver sal_console_uart; +static struct uart_driver sal_console_uart; extern struct tty_driver *uart_console_device(struct console *, int *); static struct console sal_console = { @@ -801,7 +801,7 @@ static struct console sal_console = { #define SAL_CONSOLE &sal_console -struct uart_driver sal_console_uart = { +static struct uart_driver sal_console_uart = { .owner = THIS_MODULE, .driver_name = "sn_console", .dev_name = DEVICE_NAME, diff --git a/drivers/serial/sunsab.c b/drivers/serial/sunsab.c index 8caaf2e5e47..39b788d95e3 100644 --- a/drivers/serial/sunsab.c +++ b/drivers/serial/sunsab.c @@ -682,7 +682,8 @@ static void calc_ebrg(int baud, int *n_ret, int *m_ret) /* Internal routine, port->lock is held and local interrupts are disabled. */ static void sunsab_convert_to_sab(struct uart_sunsab_port *up, unsigned int cflag, - unsigned int iflag, int baud) + unsigned int iflag, unsigned int baud, + unsigned int quot) { unsigned int ebrg; unsigned char dafo; @@ -766,6 +767,9 @@ static void sunsab_convert_to_sab(struct uart_sunsab_port *up, unsigned int cfla up->port.ignore_status_mask |= (SAB82532_ISR0_RPF | SAB82532_ISR0_TCD); + uart_update_timeout(&up->port, cflag, + (up->port.uartclk / (16 * quot))); + /* Now bang the new settings into the chip. */ sunsab_cec_wait(up); sunsab_tec_wait(up); @@ -784,10 +788,11 @@ static void sunsab_set_termios(struct uart_port *port, struct termios *termios, { struct uart_sunsab_port *up = (struct uart_sunsab_port *) port; unsigned long flags; - int baud = uart_get_baud_rate(port, termios, old, 0, 4000000); + unsigned int baud = uart_get_baud_rate(port, termios, old, 0, 4000000); + unsigned int quot = uart_get_divisor(port, baud); spin_lock_irqsave(&up->port.lock, flags); - sunsab_convert_to_sab(up, termios->c_cflag, termios->c_iflag, baud); + sunsab_convert_to_sab(up, termios->c_cflag, termios->c_iflag, baud, quot); spin_unlock_irqrestore(&up->port.lock, flags); } @@ -880,7 +885,7 @@ static int sunsab_console_setup(struct console *con, char *options) { struct uart_sunsab_port *up = &sunsab_ports[con->index]; unsigned long flags; - int baud; + unsigned int baud, quot; printk("Console: ttyS%d (SAB82532)\n", (sunsab_reg.minor - 64) + con->index); @@ -926,7 +931,8 @@ static int sunsab_console_setup(struct console *con, char *options) SAB82532_IMR1_XPR; writeb(up->interrupt_mask1, &up->regs->w.imr1); - sunsab_convert_to_sab(up, con->cflag, 0, baud); + quot = uart_get_divisor(&up->port, baud); + sunsab_convert_to_sab(up, con->cflag, 0, baud, quot); sunsab_set_mctrl(&up->port, TIOCM_DTR | TIOCM_RTS); spin_unlock_irqrestore(&up->port.lock, flags); diff --git a/drivers/serial/sunsu.c b/drivers/serial/sunsu.c index 23d19d39432..ddc97c905e1 100644 --- a/drivers/serial/sunsu.c +++ b/drivers/serial/sunsu.c @@ -1285,6 +1285,7 @@ static struct uart_driver sunsu_reg = { static int __init sunsu_kbd_ms_init(struct uart_sunsu_port *up, int channel) { + int quot, baud; #ifdef CONFIG_SERIO struct serio *serio; #endif @@ -1293,10 +1294,14 @@ static int __init sunsu_kbd_ms_init(struct uart_sunsu_port *up, int channel) up->port.type = PORT_UNKNOWN; up->port.uartclk = (SU_BASE_BAUD * 16); - if (up->su_type == SU_PORT_KBD) + if (up->su_type == SU_PORT_KBD) { up->cflag = B1200 | CS8 | CLOCAL | CREAD; - else + baud = 1200; + } else { up->cflag = B4800 | CS8 | CLOCAL | CREAD; + baud = 4800; + } + quot = up->port.uartclk / (16 * baud); sunsu_autoconfig(up); if (up->port.type == PORT_UNKNOWN) @@ -1337,6 +1342,8 @@ static int __init sunsu_kbd_ms_init(struct uart_sunsu_port *up, int channel) } #endif + sunsu_change_speed(&up->port, up->cflag, 0, quot); + sunsu_startup(&up->port); return 0; } diff --git a/drivers/telephony/ixj.c b/drivers/telephony/ixj.c index 5538756f13b..d5863b8b56e 100644 --- a/drivers/telephony/ixj.c +++ b/drivers/telephony/ixj.c @@ -41,9 +41,6 @@ * ***************************************************************************/ -static char ixj_c_rcsid[] = "$Id: ixj.c,v 4.7 2001/08/13 06:19:33 craigs Exp $"; -static char ixj_c_revision[] = "$Revision: 4.7 $"; - /* * $Log: ixj.c,v $ * @@ -6172,8 +6169,14 @@ static int ixj_ioctl(struct inode *inode, struct file *file_p, unsigned int cmd, retval = j->serial; break; case IXJCTL_VERSION: - if (copy_to_user(argp, ixj_c_revision, strlen(ixj_c_revision))) - retval = -EFAULT; + { + char arg_str[100]; + snprintf(arg_str, sizeof(arg_str), + "\nDriver version %i.%i.%i", IXJ_VER_MAJOR, + IXJ_VER_MINOR, IXJ_BLD_VER); + if (copy_to_user(argp, arg_str, strlen(arg_str))) + retval = -EFAULT; + } break; case PHONE_RING_CADENCE: j->ring_cadence = arg; @@ -7168,9 +7171,6 @@ static int ixj_get_status_proc(char *buf) int cnt; IXJ *j; len = 0; - len += sprintf(buf + len, "%s", ixj_c_rcsid); - len += sprintf(buf + len, "\n%s", ixj_h_rcsid); - len += sprintf(buf + len, "\n%s", ixjuser_h_rcsid); len += sprintf(buf + len, "\nDriver version %i.%i.%i", IXJ_VER_MAJOR, IXJ_VER_MINOR, IXJ_BLD_VER); len += sprintf(buf + len, "\nsizeof IXJ struct %Zd bytes", sizeof(IXJ)); len += sprintf(buf + len, "\nsizeof DAA struct %Zd bytes", sizeof(DAA_REGS)); @@ -7790,7 +7790,7 @@ static int __init ixj_init(void) if ((probe = ixj_probe_pci(&cnt)) < 0) { return probe; } - printk("%s\n", ixj_c_rcsid); + printk(KERN_INFO "ixj driver initialized.\n"); create_proc_read_entry ("ixj", 0, NULL, ixj_read_proc, NULL); return probe; } diff --git a/drivers/telephony/ixj.h b/drivers/telephony/ixj.h index 143818a5612..51e3f7f6597 100644 --- a/drivers/telephony/ixj.h +++ b/drivers/telephony/ixj.h @@ -38,8 +38,6 @@ * TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS. * *****************************************************************************/ -static char ixj_h_rcsid[] = "$Id: ixj.h,v 4.1 2001/08/04 14:49:27 craigs Exp $"; - #define IXJ_VERSION 3031 #include <linux/version.h> diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index e12c5be1e0a..f50aaf25c98 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -431,7 +431,7 @@ nomem: * (2) error, where io->status is a negative errno value. The number * of io->bytes transferred before the error is usually less * than requested, and can be nonzero. - * (3) cancelation, a type of error with status -ECONNRESET that + * (3) cancellation, a type of error with status -ECONNRESET that * is initiated by usb_sg_cancel(). * * When this function returns, all memory allocated through usb_sg_init() or @@ -1282,7 +1282,7 @@ static void release_interface(struct device *dev) * bus rwsem; usb device driver probe() methods cannot use this routine. * * Returns zero on success, or else the status code returned by the - * underlying call that failed. On succesful completion, each interface + * underlying call that failed. On successful completion, each interface * in the original device configuration has been destroyed, and each one * in the new configuration has been probed by all relevant usb device * drivers currently known to the kernel. diff --git a/drivers/usb/core/urb.c b/drivers/usb/core/urb.c index 16972159a57..0faf18d511d 100644 --- a/drivers/usb/core/urb.c +++ b/drivers/usb/core/urb.c @@ -121,7 +121,7 @@ struct urb * usb_get_urb(struct urb *urb) * describing that request to the USB subsystem. Request completion will * be indicated later, asynchronously, by calling the completion handler. * The three types of completion are success, error, and unlink - * (a software-induced fault, also called "request cancelation"). + * (a software-induced fault, also called "request cancellation"). * * URBs may be submitted in interrupt context. * @@ -170,7 +170,7 @@ struct urb * usb_get_urb(struct urb *urb) * As of Linux 2.6, all USB endpoint transfer queues support depths greater * than one. This was previously a HCD-specific behavior, except for ISO * transfers. Non-isochronous endpoint queues are inactive during cleanup - * after faults (transfer errors or cancelation). + * after faults (transfer errors or cancellation). * * Reserved Bandwidth Transfers: * @@ -395,7 +395,7 @@ int usb_submit_urb(struct urb *urb, int mem_flags) * * This routine cancels an in-progress request. URBs complete only * once per submission, and may be canceled only once per submission. - * Successful cancelation means the requests's completion handler will + * Successful cancellation means the requests's completion handler will * be called with a status code indicating that the request has been * canceled (rather than any other code) and will quickly be removed * from host controller data structures. diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c index c231b4bef31..25cf7e9eccf 100644 --- a/drivers/usb/core/usb.c +++ b/drivers/usb/core/usb.c @@ -611,11 +611,10 @@ static int usb_hotplug (struct device *dev, char **envp, int num_envp, if (add_hotplug_env_var(envp, num_envp, &i, buffer, buffer_size, &length, - "MODALIAS=usb:v%04Xp%04Xdl%04Xdh%04Xdc%02Xdsc%02Xdp%02Xic%02Xisc%02Xip%02X", + "MODALIAS=usb:v%04Xp%04Xd%04Xdc%02Xdsc%02Xdp%02Xic%02Xisc%02Xip%02X", le16_to_cpu(usb_dev->descriptor.idVendor), le16_to_cpu(usb_dev->descriptor.idProduct), le16_to_cpu(usb_dev->descriptor.bcdDevice), - le16_to_cpu(usb_dev->descriptor.bcdDevice), usb_dev->descriptor.bDeviceClass, usb_dev->descriptor.bDeviceSubClass, usb_dev->descriptor.bDeviceProtocol, @@ -626,11 +625,10 @@ static int usb_hotplug (struct device *dev, char **envp, int num_envp, } else { if (add_hotplug_env_var(envp, num_envp, &i, buffer, buffer_size, &length, - "MODALIAS=usb:v%04Xp%04Xdl%04Xdh%04Xdc%02Xdsc%02Xdp%02Xic*isc*ip*", + "MODALIAS=usb:v%04Xp%04Xd%04Xdc%02Xdsc%02Xdp%02Xic*isc*ip*", le16_to_cpu(usb_dev->descriptor.idVendor), le16_to_cpu(usb_dev->descriptor.idProduct), le16_to_cpu(usb_dev->descriptor.bcdDevice), - le16_to_cpu(usb_dev->descriptor.bcdDevice), usb_dev->descriptor.bDeviceClass, usb_dev->descriptor.bDeviceSubClass, usb_dev->descriptor.bDeviceProtocol)) diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c index 3993156c2e8..3f783cbdc7c 100644 --- a/drivers/usb/gadget/ether.c +++ b/drivers/usb/gadget/ether.c @@ -569,7 +569,7 @@ static const struct usb_cdc_ether_desc ether_desc = { /* include the status endpoint if we can, even where it's optional. * use wMaxPacketSize big enough to fit CDC_NOTIFY_SPEED_CHANGE in one - * packet, to simplify cancelation; and a big transfer interval, to + * packet, to simplify cancellation; and a big transfer interval, to * waste less bandwidth. * * some drivers (like Linux 2.4 cdc-ether!) "need" it to exist even diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index 2cff67ccce4..1e5e6ddef78 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -275,7 +275,7 @@ static const char *CHIP; * * After opening, configure non-control endpoints. Then use normal * stream read() and write() requests; and maybe ioctl() to get more - * precise FIFO status when recovering from cancelation. + * precise FIFO status when recovering from cancellation. */ static void epio_complete (struct usb_ep *ep, struct usb_request *req) diff --git a/drivers/usb/gadget/lh7a40x_udc.c b/drivers/usb/gadget/lh7a40x_udc.c index 0def9f70e88..df75ab65a5e 100644 --- a/drivers/usb/gadget/lh7a40x_udc.c +++ b/drivers/usb/gadget/lh7a40x_udc.c @@ -705,7 +705,7 @@ void nuke(struct lh7a40x_ep *ep, int status) done(ep, req, status); } - /* Disable IRQ if EP is enabled (has decriptor) */ + /* Disable IRQ if EP is enabled (has descriptor) */ if (ep->desc) pio_irq_disable(ep_index(ep)); } diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c index f1762ed6db6..4d591c764e3 100644 --- a/drivers/usb/gadget/serial.c +++ b/drivers/usb/gadget/serial.c @@ -240,7 +240,7 @@ struct gs_dev { struct usb_ep *dev_notify_ep; /* address of notify endpoint */ struct usb_ep *dev_in_ep; /* address of in endpoint */ struct usb_ep *dev_out_ep; /* address of out endpoint */ - struct usb_endpoint_descriptor /* desciptor of notify ep */ + struct usb_endpoint_descriptor /* descriptor of notify ep */ *dev_notify_ep_desc; struct usb_endpoint_descriptor /* descriptor of in endpoint */ *dev_in_ep_desc; diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c index 84d2b93aca3..bc69bd7aceb 100644 --- a/drivers/usb/host/ehci-hcd.c +++ b/drivers/usb/host/ehci-hcd.c @@ -346,6 +346,22 @@ ehci_reboot (struct notifier_block *self, unsigned long code, void *null) return 0; } +static void ehci_port_power (struct ehci_hcd *ehci, int is_on) +{ + unsigned port; + + if (!HCS_PPC (ehci->hcs_params)) + return; + + ehci_dbg (ehci, "...power%s ports...\n", is_on ? "up" : "down"); + for (port = HCS_N_PORTS (ehci->hcs_params); port > 0; ) + (void) ehci_hub_control(ehci_to_hcd(ehci), + is_on ? SetPortFeature : ClearPortFeature, + USB_PORT_FEAT_POWER, + port--, NULL, 0); + msleep(20); +} + /* called by khubd or root hub init threads */ @@ -362,8 +378,10 @@ static int ehci_hc_reset (struct usb_hcd *hcd) dbg_hcs_params (ehci, "reset"); dbg_hcc_params (ehci, "reset"); + /* cache this readonly data; minimize chip reads */ + ehci->hcs_params = readl (&ehci->caps->hcs_params); + #ifdef CONFIG_PCI - /* EHCI 0.96 and later may have "extended capabilities" */ if (hcd->self.controller->bus == &pci_bus_type) { struct pci_dev *pdev = to_pci_dev(hcd->self.controller); @@ -383,9 +401,30 @@ static int ehci_hc_reset (struct usb_hcd *hcd) break; } + /* optional debug port, normally in the first BAR */ + temp = pci_find_capability (pdev, 0x0a); + if (temp) { + pci_read_config_dword(pdev, temp, &temp); + temp >>= 16; + if ((temp & (3 << 13)) == (1 << 13)) { + temp &= 0x1fff; + ehci->debug = hcd->regs + temp; + temp = readl (&ehci->debug->control); + ehci_info (ehci, "debug port %d%s\n", + HCS_DEBUG_PORT(ehci->hcs_params), + (temp & DBGP_ENABLED) + ? " IN USE" + : ""); + if (!(temp & DBGP_ENABLED)) + ehci->debug = NULL; + } + } + temp = HCC_EXT_CAPS (readl (&ehci->caps->hcc_params)); } else temp = 0; + + /* EHCI 0.96 and later may have "extended capabilities" */ while (temp && count--) { u32 cap; @@ -414,8 +453,7 @@ static int ehci_hc_reset (struct usb_hcd *hcd) ehci_reset (ehci); #endif - /* cache this readonly data; minimize PCI reads */ - ehci->hcs_params = readl (&ehci->caps->hcs_params); + ehci_port_power (ehci, 0); /* at least the Genesys GL880S needs fixup here */ temp = HCS_N_CC(ehci->hcs_params) * HCS_N_PCC(ehci->hcs_params); @@ -657,16 +695,11 @@ done2: static void ehci_stop (struct usb_hcd *hcd) { struct ehci_hcd *ehci = hcd_to_ehci (hcd); - u8 rh_ports, port; ehci_dbg (ehci, "stop\n"); /* Turn off port power on all root hub ports. */ - rh_ports = HCS_N_PORTS (ehci->hcs_params); - for (port = 1; port <= rh_ports; port++) - (void) ehci_hub_control(hcd, - ClearPortFeature, USB_PORT_FEAT_POWER, - port, NULL, 0); + ehci_port_power (ehci, 0); /* no more interrupts ... */ del_timer_sync (&ehci->watchdog); @@ -748,7 +781,6 @@ static int ehci_resume (struct usb_hcd *hcd) unsigned port; struct usb_device *root = hcd->self.root_hub; int retval = -EINVAL; - int powerup = 0; // maybe restore (PCI) FLADJ @@ -766,8 +798,6 @@ static int ehci_resume (struct usb_hcd *hcd) up (&hcd->self.root_hub->serialize); break; } - if ((status & PORT_POWER) == 0) - powerup = 1; if (!root->children [port]) continue; dbg_port (ehci, __FUNCTION__, port + 1, status); @@ -794,16 +824,9 @@ static int ehci_resume (struct usb_hcd *hcd) retval = ehci_start (hcd); /* here we "know" root ports should always stay powered; - * but some controllers may lost all power. + * but some controllers may lose all power. */ - if (powerup) { - ehci_dbg (ehci, "...powerup ports...\n"); - for (port = HCS_N_PORTS (ehci->hcs_params); port > 0; ) - (void) ehci_hub_control(hcd, - SetPortFeature, USB_PORT_FEAT_POWER, - port--, NULL, 0); - msleep(20); - } + ehci_port_power (ehci, 1); } return retval; diff --git a/drivers/usb/host/ehci-hub.c b/drivers/usb/host/ehci-hub.c index 2373537fabe..02fefab3501 100644 --- a/drivers/usb/host/ehci-hub.c +++ b/drivers/usb/host/ehci-hub.c @@ -281,6 +281,8 @@ ehci_hub_descriptor ( temp = 0x0008; /* per-port overcurrent reporting */ if (HCS_PPC (ehci->hcs_params)) temp |= 0x0001; /* per-port power control */ + else + temp |= 0x0002; /* no power switching */ #if 0 // re-enable when we support USB_PORT_FEAT_INDICATOR below. if (HCS_INDICATOR (ehci->hcs_params)) diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h index e763a8399a7..4df49823175 100644 --- a/drivers/usb/host/ehci.h +++ b/drivers/usb/host/ehci.h @@ -47,6 +47,12 @@ struct ehci_stats { #define EHCI_MAX_ROOT_PORTS 15 /* see HCS_N_PORTS */ struct ehci_hcd { /* one per controller */ + /* glue to PCI and HCD framework */ + struct ehci_caps __iomem *caps; + struct ehci_regs __iomem *regs; + struct ehci_dbg_port __iomem *debug; + + __u32 hcs_params; /* cached register copy */ spinlock_t lock; /* async schedule support */ @@ -84,11 +90,6 @@ struct ehci_hcd { /* one per controller */ unsigned is_tdi_rh_tt:1; /* TDI roothub with TT */ - /* glue to PCI and HCD framework */ - struct ehci_caps __iomem *caps; - struct ehci_regs __iomem *regs; - __u32 hcs_params; /* cached register copy */ - /* irq statistics */ #ifdef EHCI_STATS struct ehci_stats stats; @@ -165,7 +166,7 @@ struct ehci_caps { /* these fields are specified as 8 and 16 bit registers, * but some hosts can't perform 8 or 16 bit PCI accesses. */ - u32 hc_capbase; + u32 hc_capbase; #define HC_LENGTH(p) (((p)>>00)&0x00ff) /* bits 7:0 */ #define HC_VERSION(p) (((p)>>16)&0xffff) /* bits 31:16 */ u32 hcs_params; /* HCSPARAMS - offset 0x4 */ @@ -273,7 +274,7 @@ struct ehci_dbg_port { #define DBGP_ENABLED (1<<28) #define DBGP_DONE (1<<16) #define DBGP_INUSE (1<<10) -#define DBGP_ERRCODE(x) (((x)>>7)&0x0f) +#define DBGP_ERRCODE(x) (((x)>>7)&0x07) # define DBGP_ERR_BAD 1 # define DBGP_ERR_SIGNAL 2 #define DBGP_ERROR (1<<6) @@ -282,11 +283,11 @@ struct ehci_dbg_port { #define DBGP_LEN(x) (((x)>>0)&0x0f) u32 pids; #define DBGP_PID_GET(x) (((x)>>16)&0xff) -#define DBGP_PID_SET(data,tok) (((data)<<8)|(tok)); +#define DBGP_PID_SET(data,tok) (((data)<<8)|(tok)) u32 data03; u32 data47; u32 address; -#define DBGP_EPADDR(dev,ep) (((dev)<<8)|(ep)); +#define DBGP_EPADDR(dev,ep) (((dev)<<8)|(ep)) } __attribute__ ((packed)); /*-------------------------------------------------------------------------*/ diff --git a/drivers/usb/host/hc_crisv10.c b/drivers/usb/host/hc_crisv10.c index 376f8a034f6..d9883d774d3 100644 --- a/drivers/usb/host/hc_crisv10.c +++ b/drivers/usb/host/hc_crisv10.c @@ -4329,7 +4329,7 @@ static int __init etrax_usb_hc_init(void) bus->bus_name="ETRAX 100LX"; bus->hcpriv = hc; - /* Initalize RH to the default address. + /* Initialize RH to the default address. And make sure that we have no status change indication */ hc->rh.numports = 2; /* The RH has two ports */ hc->rh.devnum = 1; diff --git a/drivers/usb/host/sl811-hcd.c b/drivers/usb/host/sl811-hcd.c index d309e292198..a374b769207 100644 --- a/drivers/usb/host/sl811-hcd.c +++ b/drivers/usb/host/sl811-hcd.c @@ -134,7 +134,7 @@ static void port_power(struct sl811 *sl811, int is_on) /* This is a PIO-only HCD. Queueing appends URBs to the endpoint's queue, * and may start I/O. Endpoint queues are scanned during completion irq - * handlers (one per packet: ACK, NAK, faults, etc) and urb cancelation. + * handlers (one per packet: ACK, NAK, faults, etc) and urb cancellation. * * Using an external DMA engine to copy a packet at a time could work, * though setup/teardown costs may be too big to make it worthwhile. @@ -738,7 +738,7 @@ retry: } #endif - /* port status seems wierd until after reset, so + /* port status seems weird until after reset, so * force the reset and make khubd clean up later. */ sl811->port1 |= (1 << USB_PORT_FEAT_C_CONNECTION) diff --git a/drivers/usb/image/mdc800.c b/drivers/usb/image/mdc800.c index 5791723e608..a330a4b50e1 100644 --- a/drivers/usb/image/mdc800.c +++ b/drivers/usb/image/mdc800.c @@ -23,7 +23,7 @@ * * * The driver brings the USB functions of the MDC800 to Linux. - * To use the Camera you must support the USB Protocoll of the camera + * To use the Camera you must support the USB Protocol of the camera * to the Kernel Node. * The Driver uses a misc device Node. Create it with : * mknod /dev/mustek c 180 32 diff --git a/drivers/usb/image/microtek.c b/drivers/usb/image/microtek.c index cab89a970c7..7d21a4f5c42 100644 --- a/drivers/usb/image/microtek.c +++ b/drivers/usb/image/microtek.c @@ -335,7 +335,7 @@ static int mts_scsi_abort (Scsi_Cmnd *srb) mts_urb_abort(desc); - return FAILURE; + return FAILED; } static int mts_scsi_host_reset (Scsi_Cmnd *srb) diff --git a/drivers/usb/input/aiptek.c b/drivers/usb/input/aiptek.c index 2d76be62f4e..94ce2a9ad50 100644 --- a/drivers/usb/input/aiptek.c +++ b/drivers/usb/input/aiptek.c @@ -386,7 +386,7 @@ static int aiptek_convert_from_2s_complement(unsigned char c) * convention above.) I therefore have taken over REL_MISC and ABS_MISC * (for relative and absolute reports, respectively) for communicating * Proximity. Why two events? I thought it interesting to know if the - * Proximity event occured while the tablet was in absolute or relative + * Proximity event occurred while the tablet was in absolute or relative * mode. * * Other tablets use the notion of a certain minimum stylus pressure diff --git a/drivers/usb/input/ati_remote.c b/drivers/usb/input/ati_remote.c index 355add5c29f..860df26323b 100644 --- a/drivers/usb/input/ati_remote.c +++ b/drivers/usb/input/ati_remote.c @@ -619,7 +619,7 @@ static void ati_remote_delete(struct ati_remote *ati_remote) if (ati_remote->outbuf) usb_buffer_free(ati_remote->udev, DATA_BUFSIZE, - ati_remote->inbuf, ati_remote->outbuf_dma); + ati_remote->outbuf, ati_remote->outbuf_dma); if (ati_remote->irq_urb) usb_free_urb(ati_remote->irq_urb); diff --git a/drivers/usb/input/mtouchusb.c b/drivers/usb/input/mtouchusb.c index 6b45a66d58c..ab1a2a30ce7 100644 --- a/drivers/usb/input/mtouchusb.c +++ b/drivers/usb/input/mtouchusb.c @@ -32,7 +32,7 @@ * Changed reset from standard USB dev reset to vendor reset * Changed data sent to host from compensated to raw coordinates * Eliminated vendor/product module params - * Performed multiple successfull tests with an EXII-5010UC + * Performed multiple successful tests with an EXII-5010UC * * 1.5 02/27/2005 ddstreet@ieee.org * Added module parameter to select raw or hw-calibrated coordinate reporting diff --git a/drivers/usb/input/usbkbd.c b/drivers/usb/input/usbkbd.c index 01514b0551b..7038fb9d1ce 100644 --- a/drivers/usb/input/usbkbd.c +++ b/drivers/usb/input/usbkbd.c @@ -133,7 +133,8 @@ resubmit: kbd->usbdev->devpath, i); } -int usb_kbd_event(struct input_dev *dev, unsigned int type, unsigned int code, int value) +static int usb_kbd_event(struct input_dev *dev, unsigned int type, + unsigned int code, int value) { struct usb_kbd *kbd = dev->private; diff --git a/drivers/usb/media/ov511.c b/drivers/usb/media/ov511.c index d6051822416..036c485d1d1 100644 --- a/drivers/usb/media/ov511.c +++ b/drivers/usb/media/ov511.c @@ -5041,7 +5041,7 @@ ov6xx0_configure(struct usb_ov511 *ov) { OV511_I2C_BUS, 0x2a, 0x04 }, /* Disable framerate adjust */ // { OV511_I2C_BUS, 0x2b, 0xac }, /* Framerate; Set 2a[7] first */ { OV511_I2C_BUS, 0x2d, 0x99 }, - { OV511_I2C_BUS, 0x33, 0xa0 }, /* Color Procesing Parameter */ + { OV511_I2C_BUS, 0x33, 0xa0 }, /* Color Processing Parameter */ { OV511_I2C_BUS, 0x34, 0xd2 }, /* Max A/D range */ { OV511_I2C_BUS, 0x38, 0x8b }, { OV511_I2C_BUS, 0x39, 0x40 }, diff --git a/drivers/usb/media/pwc/pwc-ctrl.c b/drivers/usb/media/pwc/pwc-ctrl.c index 26aa914bc54..42ec468d52d 100644 --- a/drivers/usb/media/pwc/pwc-ctrl.c +++ b/drivers/usb/media/pwc/pwc-ctrl.c @@ -418,6 +418,44 @@ static inline int set_video_mode_Kiara(struct pwc_device *pdev, int size, int fr +static void pwc_set_image_buffer_size(struct pwc_device *pdev) +{ + int i, factor = 0, filler = 0; + + /* for PALETTE_YUV420P */ + switch(pdev->vpalette) + { + case VIDEO_PALETTE_YUV420P: + factor = 6; + filler = 128; + break; + case VIDEO_PALETTE_RAW: + factor = 6; /* can be uncompressed YUV420P */ + filler = 0; + break; + } + + /* Set sizes in bytes */ + pdev->image.size = pdev->image.x * pdev->image.y * factor / 4; + pdev->view.size = pdev->view.x * pdev->view.y * factor / 4; + + /* Align offset, or you'll get some very weird results in + YUV420 mode... x must be multiple of 4 (to get the Y's in + place), and y even (or you'll mixup U & V). This is less of a + problem for YUV420P. + */ + pdev->offset.x = ((pdev->view.x - pdev->image.x) / 2) & 0xFFFC; + pdev->offset.y = ((pdev->view.y - pdev->image.y) / 2) & 0xFFFE; + + /* Fill buffers with gray or black */ + for (i = 0; i < MAX_IMAGES; i++) { + if (pdev->image_ptr[i] != NULL) + memset(pdev->image_ptr[i], filler, pdev->view.size); + } +} + + + /** @pdev: device structure @width: viewport width @@ -475,44 +513,6 @@ int pwc_set_video_mode(struct pwc_device *pdev, int width, int height, int frame } -void pwc_set_image_buffer_size(struct pwc_device *pdev) -{ - int i, factor = 0, filler = 0; - - /* for PALETTE_YUV420P */ - switch(pdev->vpalette) - { - case VIDEO_PALETTE_YUV420P: - factor = 6; - filler = 128; - break; - case VIDEO_PALETTE_RAW: - factor = 6; /* can be uncompressed YUV420P */ - filler = 0; - break; - } - - /* Set sizes in bytes */ - pdev->image.size = pdev->image.x * pdev->image.y * factor / 4; - pdev->view.size = pdev->view.x * pdev->view.y * factor / 4; - - /* Align offset, or you'll get some very weird results in - YUV420 mode... x must be multiple of 4 (to get the Y's in - place), and y even (or you'll mixup U & V). This is less of a - problem for YUV420P. - */ - pdev->offset.x = ((pdev->view.x - pdev->image.x) / 2) & 0xFFFC; - pdev->offset.y = ((pdev->view.y - pdev->image.y) / 2) & 0xFFFE; - - /* Fill buffers with gray or black */ - for (i = 0; i < MAX_IMAGES; i++) { - if (pdev->image_ptr[i] != NULL) - memset(pdev->image_ptr[i], filler, pdev->view.size); - } -} - - - /* BRIGHTNESS */ int pwc_get_brightness(struct pwc_device *pdev) @@ -949,7 +949,7 @@ int pwc_set_leds(struct pwc_device *pdev, int on_value, int off_value) return SendControlMsg(SET_STATUS_CTL, LED_FORMATTER, 2); } -int pwc_get_leds(struct pwc_device *pdev, int *on_value, int *off_value) +static int pwc_get_leds(struct pwc_device *pdev, int *on_value, int *off_value) { unsigned char buf[2]; int ret; @@ -1100,7 +1100,7 @@ static inline int pwc_mpt_set_angle(struct pwc_device *pdev, int pan, int tilt) unsigned char buf[4]; /* set new relative angle; angles are expressed in degrees * 100, - but cam as .5 degree resolution, hence devide by 200. Also + but cam as .5 degree resolution, hence divide by 200. Also the angle must be multiplied by 64 before it's send to the cam (??) */ diff --git a/drivers/usb/media/pwc/pwc-if.c b/drivers/usb/media/pwc/pwc-if.c index 100a5a4f03a..cca47f480a8 100644 --- a/drivers/usb/media/pwc/pwc-if.c +++ b/drivers/usb/media/pwc/pwc-if.c @@ -129,7 +129,7 @@ static int default_mbufs = 2; /* Default number of mmap() buffers */ int pwc_trace = TRACE_MODULE | TRACE_FLOW | TRACE_PWCX; static int power_save = 0; static int led_on = 100, led_off = 0; /* defaults to LED that is on while in use */ - int pwc_preferred_compression = 2; /* 0..3 = uncompressed..high */ +static int pwc_preferred_compression = 2; /* 0..3 = uncompressed..high */ static struct { int type; char serial_number[30]; @@ -272,7 +272,7 @@ static int pwc_allocate_buffers(struct pwc_device *pdev) return -ENXIO; } #endif - /* Allocate Isochronuous pipe buffers */ + /* Allocate Isochronous pipe buffers */ for (i = 0; i < MAX_ISO_BUFS; i++) { if (pdev->sbuf[i].data == NULL) { kbuf = kmalloc(ISO_BUFFER_SIZE, GFP_KERNEL); @@ -322,7 +322,7 @@ static int pwc_allocate_buffers(struct pwc_device *pdev) case 730: case 740: case 750: - Trace(TRACE_MEMORY,"private_data(%Zd)\n",sizeof(struct pwc_dec23_private)); + Trace(TRACE_MEMORY,"private_data(%zu)\n",sizeof(struct pwc_dec23_private)); kbuf = kmalloc(sizeof(struct pwc_dec23_private), GFP_KERNEL); /* Timon & Kiara */ break; case 645: @@ -850,7 +850,7 @@ static int pwc_isoc_init(struct pwc_device *pdev) if (pdev->vmax_packet_size < 0 || pdev->vmax_packet_size > ISO_MAX_FRAME_SIZE) { Err("Failed to find packet size for video endpoint in current alternate setting.\n"); - return -ENFILE; /* Odd error, that should be noticable */ + return -ENFILE; /* Odd error, that should be noticeable */ } /* Set alternate interface */ @@ -1179,7 +1179,7 @@ static ssize_t pwc_video_read(struct file *file, char __user * buf, DECLARE_WAITQUEUE(wait, current); int bytes_to_read; - Trace(TRACE_READ, "video_read(0x%p, %p, %Zd) called.\n", vdev, buf, count); + Trace(TRACE_READ, "video_read(0x%p, %p, %zu) called.\n", vdev, buf, count); if (vdev == NULL) return -EFAULT; pdev = vdev->priv; @@ -2128,7 +2128,7 @@ static int __init usb_pwc_init(void) if (leds[1] >= 0) led_off = leds[1]; - /* Big device node whoopla. Basicly, it allows you to assign a + /* Big device node whoopla. Basically, it allows you to assign a device node (/dev/videoX) to a camera, based on its type & serial number. The format is [type[.serialnumber]:]node. diff --git a/drivers/usb/media/pwc/pwc-ioctl.h b/drivers/usb/media/pwc/pwc-ioctl.h index 65805eaa9a1..5f9cb08bc02 100644 --- a/drivers/usb/media/pwc/pwc-ioctl.h +++ b/drivers/usb/media/pwc/pwc-ioctl.h @@ -75,7 +75,7 @@ #define PWC_FPS_SNAPSHOT 0x00400000 -/* structure for transfering x & y coordinates */ +/* structure for transferring x & y coordinates */ struct pwc_coord { int x, y; /* guess what */ diff --git a/drivers/usb/media/pwc/pwc.h b/drivers/usb/media/pwc/pwc.h index 53b516d29cf..267869dab18 100644 --- a/drivers/usb/media/pwc/pwc.h +++ b/drivers/usb/media/pwc/pwc.h @@ -226,9 +226,8 @@ struct pwc_device extern "C" { #endif -/* Global variables */ +/* Global variable */ extern int pwc_trace; -extern int pwc_preferred_compression; /** functions in pwc-if.c */ int pwc_try_video_mode(struct pwc_device *pdev, int width, int height, int new_fps, int new_compression, int new_snapshot); @@ -243,8 +242,6 @@ void pwc_construct(struct pwc_device *pdev); /** Functions in pwc-ctrl.c */ /* Request a certain video mode. Returns < 0 if not possible */ extern int pwc_set_video_mode(struct pwc_device *pdev, int width, int height, int frames, int compression, int snapshot); -/* Calculate the number of bytes per image (not frame) */ -extern void pwc_set_image_buffer_size(struct pwc_device *pdev); /* Various controls; should be obvious. Value 0..65535, or < 0 on error */ extern int pwc_get_brightness(struct pwc_device *pdev); @@ -256,7 +253,6 @@ extern int pwc_set_gamma(struct pwc_device *pdev, int value); extern int pwc_get_saturation(struct pwc_device *pdev); extern int pwc_set_saturation(struct pwc_device *pdev, int value); extern int pwc_set_leds(struct pwc_device *pdev, int on_value, int off_value); -extern int pwc_get_leds(struct pwc_device *pdev, int *on_value, int *off_value); extern int pwc_get_cmos_sensor(struct pwc_device *pdev, int *sensor); /* Power down or up the camera; not supported by all models */ diff --git a/drivers/usb/media/sn9c102_core.c b/drivers/usb/media/sn9c102_core.c index 898401cf7dc..31d57400d5b 100644 --- a/drivers/usb/media/sn9c102_core.c +++ b/drivers/usb/media/sn9c102_core.c @@ -429,7 +429,7 @@ sn9c102_i2c_try_read(struct sn9c102_device* cam, } -int +static int sn9c102_i2c_try_write(struct sn9c102_device* cam, struct sn9c102_sensor* sensor, u8 address, u8 value) { @@ -785,7 +785,7 @@ static int sn9c102_stop_transfer(struct sn9c102_device* cam) } -int sn9c102_stream_interrupt(struct sn9c102_device* cam) +static int sn9c102_stream_interrupt(struct sn9c102_device* cam) { int err = 0; diff --git a/drivers/usb/media/sn9c102_sensor.h b/drivers/usb/media/sn9c102_sensor.h index 16f7483559f..6a7adebcb4b 100644 --- a/drivers/usb/media/sn9c102_sensor.h +++ b/drivers/usb/media/sn9c102_sensor.h @@ -145,8 +145,6 @@ static const struct usb_device_id sn9c102_id_table[] = { \ */ /* The "try" I2C I/O versions are used when probing the sensor */ -extern int sn9c102_i2c_try_write(struct sn9c102_device*,struct sn9c102_sensor*, - u8 address, u8 value); extern int sn9c102_i2c_try_read(struct sn9c102_device*,struct sn9c102_sensor*, u8 address); diff --git a/drivers/usb/misc/legousbtower.c b/drivers/usb/misc/legousbtower.c index dd4580cb57e..7d06105763d 100644 --- a/drivers/usb/misc/legousbtower.c +++ b/drivers/usb/misc/legousbtower.c @@ -859,7 +859,7 @@ static int tower_probe (struct usb_interface *interface, const struct usb_device info ("udev is NULL."); } - /* allocate memory for our device state and intialize it */ + /* allocate memory for our device state and initialize it */ dev = kmalloc (sizeof(struct lego_usb_tower), GFP_KERNEL); diff --git a/drivers/usb/misc/sisusbvga/sisusb.c b/drivers/usb/misc/sisusbvga/sisusb.c index 57b82d53a94..2fd12264fd5 100644 --- a/drivers/usb/misc/sisusbvga/sisusb.c +++ b/drivers/usb/misc/sisusbvga/sisusb.c @@ -983,7 +983,7 @@ static int sisusb_write_mem_bulk(struct sisusb_usb_data *sisusb, u32 addr, msgcount++; if (msgcount < 500) printk(KERN_ERR - "sisusbvga[%d]: Wrote %Zd of " + "sisusbvga[%d]: Wrote %zd of " "%d bytes, error %d\n", sisusb->minor, *bytes_written, length, ret); @@ -3105,6 +3105,7 @@ static void sisusb_disconnect(struct usb_interface *intf) static struct usb_device_id sisusb_table [] = { { USB_DEVICE(0x0711, 0x0900) }, { USB_DEVICE(0x182d, 0x021c) }, + { USB_DEVICE(0x182d, 0x0269) }, { } }; diff --git a/drivers/usb/net/pegasus.c b/drivers/usb/net/pegasus.c index a02be795d63..d976790312a 100644 --- a/drivers/usb/net/pegasus.c +++ b/drivers/usb/net/pegasus.c @@ -1388,11 +1388,11 @@ static int pegasus_resume (struct usb_interface *intf) if (netif_running(pegasus->net)) { pegasus->rx_urb->status = 0; pegasus->rx_urb->actual_length = 0; - read_bulk_callback(pegasus->rx_urb, 0); + read_bulk_callback(pegasus->rx_urb, NULL); pegasus->intr_urb->status = 0; pegasus->intr_urb->actual_length = 0; - intr_callback(pegasus->intr_urb, 0); + intr_callback(pegasus->intr_urb, NULL); queue_delayed_work(pegasus_workqueue, &pegasus->carrier_check, CARRIER_CHECK_DELAY); diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index bbaef047d53..f6bc6b3b333 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -1,6 +1,6 @@ /* * USB Networking Links - * Copyright (C) 2000-2003 by David Brownell <dbrownell@users.sourceforge.net> + * Copyright (C) 2000-2005 by David Brownell <dbrownell@users.sourceforge.net> * Copyright (C) 2002 Pavel Machek <pavel@ucw.cz> * Copyright (C) 2003-2005 David Hollis <dhollis@davehollis.com> * Copyright (C) 2005 Phil Chang <pchang23@sbcglobal.net> @@ -210,6 +210,7 @@ struct usbnet { # define EVENT_RX_HALT 1 # define EVENT_RX_MEMORY 2 # define EVENT_STS_SPLIT 3 +# define EVENT_LINK_RESET 4 }; // device-specific info used by the driver @@ -243,6 +244,9 @@ struct driver_info { /* for status polling */ void (*status)(struct usbnet *, struct urb *); + /* link reset handling, called from defer_kevent */ + int (*link_reset)(struct usbnet *); + /* fixup rx packet (strip framing) */ int (*rx_fixup)(struct usbnet *dev, struct sk_buff *skb); @@ -304,6 +308,7 @@ static void usbnet_get_drvinfo (struct net_device *, struct ethtool_drvinfo *); static u32 usbnet_get_link (struct net_device *); static u32 usbnet_get_msglevel (struct net_device *); static void usbnet_set_msglevel (struct net_device *, u32); +static void defer_kevent (struct usbnet *, int); /* mostly for PDA style devices, which are always connected if present */ static int always_connected (struct usbnet *dev) @@ -426,7 +431,7 @@ static void skb_return (struct usbnet *dev, struct sk_buff *skb) dev->stats.rx_bytes += skb->len; if (netif_msg_rx_status (dev)) - devdbg (dev, "< rx, len %zd, type 0x%x", + devdbg (dev, "< rx, len %zu, type 0x%x", skb->len + sizeof (struct ethhdr), skb->protocol); memset (skb->cb, 0, sizeof (struct skb_data)); status = netif_rx (skb); @@ -501,6 +506,7 @@ static const struct driver_info an2720_info = { #define AX_CMD_WRITE_MULTI_FILTER 0x16 #define AX_CMD_READ_NODE_ID 0x17 #define AX_CMD_READ_PHY_ID 0x19 +#define AX_CMD_READ_MEDIUM_STATUS 0x1a #define AX_CMD_WRITE_MEDIUM_MODE 0x1b #define AX_CMD_READ_MONITOR_MODE 0x1c #define AX_CMD_WRITE_MONITOR_MODE 0x1d @@ -515,11 +521,14 @@ static const struct driver_info an2720_info = { #define AX_MONITOR_MAGIC 0x04 #define AX_MONITOR_HSFS 0x10 +/* AX88172 Medium Status Register values */ +#define AX_MEDIUM_FULL_DUPLEX 0x02 +#define AX_MEDIUM_TX_ABORT_ALLOW 0x04 +#define AX_MEDIUM_FLOW_CONTROL_EN 0x10 + #define AX_MCAST_FILTER_SIZE 8 #define AX_MAX_MCAST 64 -#define AX_INTERRUPT_BUFSIZE 8 - #define AX_EEPROM_LEN 0x40 #define AX_SWRESET_CLEAR 0x00 @@ -535,15 +544,33 @@ static const struct driver_info an2720_info = { #define AX88772_IPG1_DEFAULT 0x0c #define AX88772_IPG2_DEFAULT 0x12 +#define AX88772_MEDIUM_FULL_DUPLEX 0x0002 +#define AX88772_MEDIUM_RESERVED 0x0004 +#define AX88772_MEDIUM_RX_FC_ENABLE 0x0010 +#define AX88772_MEDIUM_TX_FC_ENABLE 0x0020 +#define AX88772_MEDIUM_PAUSE_FORMAT 0x0080 +#define AX88772_MEDIUM_RX_ENABLE 0x0100 +#define AX88772_MEDIUM_100MB 0x0200 +#define AX88772_MEDIUM_DEFAULT \ + (AX88772_MEDIUM_FULL_DUPLEX | AX88772_MEDIUM_RX_FC_ENABLE | \ + AX88772_MEDIUM_TX_FC_ENABLE | AX88772_MEDIUM_100MB | \ + AX88772_MEDIUM_RESERVED | AX88772_MEDIUM_RX_ENABLE ) + #define AX_EEPROM_MAGIC 0xdeadbeef /* This structure cannot exceed sizeof(unsigned long [5]) AKA 20 bytes */ struct ax8817x_data { u8 multi_filter[AX_MCAST_FILTER_SIZE]; - struct urb *int_urb; - u8 *int_buf; }; +struct ax88172_int_data { + u16 res1; + u8 link; + u16 res2; + u8 status; + u16 res3; +} __attribute__ ((packed)); + static int ax8817x_read_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index, u16 size, void *data) { @@ -586,25 +613,23 @@ static void ax8817x_async_cmd_callback(struct urb *urb, struct pt_regs *regs) usb_free_urb(urb); } -static void ax8817x_interrupt_complete(struct urb *urb, struct pt_regs *regs) +static void ax8817x_status(struct usbnet *dev, struct urb *urb) { - struct usbnet *dev = (struct usbnet *)urb->context; - struct ax8817x_data *data = (struct ax8817x_data *)&dev->data; + struct ax88172_int_data *event; int link; - if (urb->status < 0) { - devdbg(dev,"ax8817x_interrupt_complete() failed with %d", - urb->status); - } else { - link = data->int_buf[2] & 0x01; - if (netif_carrier_ok(dev->net) != link) { - if (link) - netif_carrier_on(dev->net); - else - netif_carrier_off(dev->net); - devdbg(dev, "ax8817x - Link Status is: %d", link); - } - usb_submit_urb(data->int_urb, GFP_ATOMIC); + if (urb->actual_length < 8) + return; + + event = urb->transfer_buffer; + link = event->link & 0x01; + if (netif_carrier_ok(dev->net) != link) { + if (link) { + netif_carrier_on(dev->net); + defer_kevent (dev, EVENT_LINK_RESET ); + } else + netif_carrier_off(dev->net); + devdbg(dev, "ax8817x - Link Status is: %d", link); } } @@ -711,6 +736,20 @@ static void ax8817x_mdio_write(struct net_device *netdev, int phy_id, int loc, i ax8817x_write_cmd(dev, AX_CMD_SET_HW_MII, 0, 0, 0, &buf); } +static int ax88172_link_reset(struct usbnet *dev) +{ + u16 lpa; + u8 mode; + + mode = AX_MEDIUM_TX_ABORT_ALLOW | AX_MEDIUM_FLOW_CONTROL_EN; + lpa = ax8817x_mdio_read(dev->net, dev->mii.phy_id, MII_LPA); + if (lpa & LPA_DUPLEX) + mode |= AX_MEDIUM_FULL_DUPLEX; + ax8817x_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE, mode, 0, 0, NULL); + + return 0; +} + static void ax8817x_get_wol(struct net_device *net, struct ethtool_wolinfo *wolinfo) { struct usbnet *dev = netdev_priv(net); @@ -824,35 +863,13 @@ static int ax8817x_bind(struct usbnet *dev, struct usb_interface *intf) void *buf; int i; unsigned long gpio_bits = dev->driver_info->data; - struct ax8817x_data *data = (struct ax8817x_data *)dev->data; get_endpoints(dev,intf); - if ((data->int_urb = usb_alloc_urb (0, GFP_KERNEL)) == NULL) { - dbg ("%s: cannot allocate interrupt URB", - dev->net->name); - ret = -ENOMEM; - goto out1; - } - - if ((data->int_buf = kmalloc(AX_INTERRUPT_BUFSIZE, GFP_KERNEL)) == NULL) { - dbg ("%s: cannot allocate memory for interrupt buffer", - dev->net->name); - ret = -ENOMEM; - goto out1; - } - memset(data->int_buf, 0, AX_INTERRUPT_BUFSIZE); - - usb_fill_int_urb (data->int_urb, dev->udev, - usb_rcvintpipe (dev->udev, 1), - data->int_buf, AX_INTERRUPT_BUFSIZE, - ax8817x_interrupt_complete, dev, - dev->udev->speed == USB_SPEED_HIGH ? 8 : 100); - buf = kmalloc(ETH_ALEN, GFP_KERNEL); if(!buf) { ret = -ENOMEM; - goto out2; + goto out1; } /* Toggle the GPIOs in a manufacturer/model specific way */ @@ -860,32 +877,32 @@ static int ax8817x_bind(struct usbnet *dev, struct usb_interface *intf) if ((ret = ax8817x_write_cmd(dev, AX_CMD_WRITE_GPIOS, (gpio_bits >> (i * 8)) & 0xff, 0, 0, buf)) < 0) - goto out3; + goto out2; msleep(5); } if ((ret = ax8817x_write_cmd(dev, AX_CMD_WRITE_RX_CTL, 0x80, 0, 0, buf)) < 0) { dbg("send AX_CMD_WRITE_RX_CTL failed: %d", ret); - goto out3; + goto out2; } /* Get the MAC address */ memset(buf, 0, ETH_ALEN); if ((ret = ax8817x_read_cmd(dev, AX_CMD_READ_NODE_ID, 0, 0, 6, buf)) < 0) { dbg("read AX_CMD_READ_NODE_ID failed: %d", ret); - goto out3; + goto out2; } memcpy(dev->net->dev_addr, buf, ETH_ALEN); /* Get the PHY id */ if ((ret = ax8817x_read_cmd(dev, AX_CMD_READ_PHY_ID, 0, 0, 2, buf)) < 0) { dbg("error on read AX_CMD_READ_PHY_ID: %02x", ret); - goto out3; + goto out2; } else if (ret < 2) { /* this should always return 2 bytes */ dbg("AX_CMD_READ_PHY_ID returned less than 2 bytes: ret=%02x", ret); ret = -EIO; - goto out3; + goto out2; } /* Initialize MII structure */ @@ -899,36 +916,18 @@ static int ax8817x_bind(struct usbnet *dev, struct usb_interface *intf) dev->net->set_multicast_list = ax8817x_set_multicast; dev->net->ethtool_ops = &ax8817x_ethtool_ops; - ax8817x_mdio_write(dev->net, dev->mii.phy_id, MII_BMCR, - cpu_to_le16(BMCR_RESET)); + ax8817x_mdio_write(dev->net, dev->mii.phy_id, MII_BMCR, BMCR_RESET); ax8817x_mdio_write(dev->net, dev->mii.phy_id, MII_ADVERTISE, - cpu_to_le16(ADVERTISE_ALL | ADVERTISE_CSMA | 0x0400)); + ADVERTISE_ALL | ADVERTISE_CSMA | ADVERTISE_PAUSE_CAP); mii_nway_restart(&dev->mii); - if((ret = usb_submit_urb(data->int_urb, GFP_KERNEL)) < 0) { - dbg("Failed to submit interrupt URB: %02x", ret); - goto out2; - } - return 0; -out3: - kfree(buf); out2: - kfree(data->int_buf); + kfree(buf); out1: - usb_free_urb(data->int_urb); return ret; } -static void ax8817x_unbind(struct usbnet *dev, struct usb_interface *intf) -{ - struct ax8817x_data *data = (struct ax8817x_data *)dev->data; - - usb_kill_urb(data->int_urb); - usb_free_urb(data->int_urb); - kfree(data->int_buf); -} - static struct ethtool_ops ax88772_ethtool_ops = { .get_drvinfo = ax8817x_get_drvinfo, .get_link = ethtool_op_get_link, @@ -946,64 +945,44 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf) { int ret; void *buf; - struct ax8817x_data *data = (struct ax8817x_data *)dev->data; get_endpoints(dev,intf); - if ((data->int_urb = usb_alloc_urb (0, GFP_KERNEL)) == 0) { - dbg ("Cannot allocate interrupt URB"); - ret = -ENOMEM; - goto out1; - } - - if ((data->int_buf = kmalloc(AX_INTERRUPT_BUFSIZE, GFP_KERNEL)) == NULL) { - dbg ("Cannot allocate memory for interrupt buffer"); - ret = -ENOMEM; - goto out1; - } - memset(data->int_buf, 0, AX_INTERRUPT_BUFSIZE); - - usb_fill_int_urb (data->int_urb, dev->udev, - usb_rcvintpipe (dev->udev, 1), - data->int_buf, AX_INTERRUPT_BUFSIZE, - ax8817x_interrupt_complete, dev, - dev->udev->speed == USB_SPEED_HIGH ? 8 : 100); - buf = kmalloc(6, GFP_KERNEL); if(!buf) { dbg ("Cannot allocate memory for buffer"); ret = -ENOMEM; - goto out2; + goto out1; } if ((ret = ax8817x_write_cmd(dev, AX_CMD_WRITE_GPIOS, 0x00B0, 0, 0, buf)) < 0) - goto out3; + goto out2; msleep(5); if ((ret = ax8817x_write_cmd(dev, AX_CMD_SW_PHY_SELECT, 0x0001, 0, 0, buf)) < 0) { dbg("Select PHY #1 failed: %d", ret); - goto out3; + goto out2; } if ((ret = ax8817x_write_cmd(dev, AX_CMD_SW_RESET, AX_SWRESET_IPPD, 0, 0, buf)) < 0) { dbg("Failed to power down internal PHY: %d", ret); - goto out3; + goto out2; } msleep(150); if ((ret = ax8817x_write_cmd(dev, AX_CMD_SW_RESET, AX_SWRESET_CLEAR, 0, 0, buf)) < 0) { dbg("Failed to perform software reset: %d", ret); - goto out3; + goto out2; } msleep(150); if ((ret = ax8817x_write_cmd(dev, AX_CMD_SW_RESET, AX_SWRESET_IPRL | AX_SWRESET_PRL, 0, 0, buf)) < 0) { dbg("Failed to set Internal/External PHY reset control: %d", ret); - goto out3; + goto out2; } msleep(150); @@ -1011,27 +990,27 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf) ax8817x_write_cmd(dev, AX_CMD_WRITE_RX_CTL, 0x0000, 0, 0, buf)) < 0) { dbg("Failed to reset RX_CTL: %d", ret); - goto out3; + goto out2; } /* Get the MAC address */ memset(buf, 0, ETH_ALEN); if ((ret = ax8817x_read_cmd(dev, AX88772_CMD_READ_NODE_ID, 0, 0, ETH_ALEN, buf)) < 0) { dbg("Failed to read MAC address: %d", ret); - goto out3; + goto out2; } memcpy(dev->net->dev_addr, buf, ETH_ALEN); if ((ret = ax8817x_write_cmd(dev, AX_CMD_SET_SW_MII, 0, 0, 0, buf)) < 0) { dbg("Enabling software MII failed: %d", ret); - goto out3; + goto out2; } if (((ret = ax8817x_read_cmd(dev, AX_CMD_READ_MII_REG, 0x0010, 2, 2, buf)) < 0) || (*((u16 *)buf) != 0x003b)) { dbg("Read PHY register 2 must be 0x3b00: %d", ret); - goto out3; + goto out2; } /* Initialize MII structure */ @@ -1044,26 +1023,26 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf) /* Get the PHY id */ if ((ret = ax8817x_read_cmd(dev, AX_CMD_READ_PHY_ID, 0, 0, 2, buf)) < 0) { dbg("Error reading PHY ID: %02x", ret); - goto out3; + goto out2; } else if (ret < 2) { /* this should always return 2 bytes */ dbg("AX_CMD_READ_PHY_ID returned less than 2 bytes: ret=%02x", ret); ret = -EIO; - goto out3; + goto out2; } dev->mii.phy_id = *((u8 *)buf + 1); if ((ret = ax8817x_write_cmd(dev, AX_CMD_SW_RESET, AX_SWRESET_PRL, 0, 0, buf)) < 0) { dbg("Set external PHY reset pin level: %d", ret); - goto out3; + goto out2; } msleep(150); if ((ret = ax8817x_write_cmd(dev, AX_CMD_SW_RESET, AX_SWRESET_IPRL | AX_SWRESET_PRL, 0, 0, buf)) < 0) { dbg("Set Internal/External PHY reset control: %d", ret); - goto out3; + goto out2; } msleep(150); @@ -1071,25 +1050,24 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf) dev->net->set_multicast_list = ax8817x_set_multicast; dev->net->ethtool_ops = &ax88772_ethtool_ops; - ax8817x_mdio_write(dev->net, dev->mii.phy_id, MII_BMCR, - cpu_to_le16(BMCR_RESET)); + ax8817x_mdio_write(dev->net, dev->mii.phy_id, MII_BMCR, BMCR_RESET); ax8817x_mdio_write(dev->net, dev->mii.phy_id, MII_ADVERTISE, - cpu_to_le16(ADVERTISE_ALL | ADVERTISE_CSMA)); + ADVERTISE_ALL | ADVERTISE_CSMA); mii_nway_restart(&dev->mii); - if ((ret = ax8817x_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE, 0x0336, 0, 0, buf)) < 0) { + if ((ret = ax8817x_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE, AX88772_MEDIUM_DEFAULT, 0, 0, buf)) < 0) { dbg("Write medium mode register: %d", ret); - goto out3; + goto out2; } if ((ret = ax8817x_write_cmd(dev, AX_CMD_WRITE_IPG0, AX88772_IPG0_DEFAULT | AX88772_IPG1_DEFAULT,AX88772_IPG2_DEFAULT, 0, buf)) < 0) { dbg("Write IPG,IPG1,IPG2 failed: %d", ret); - goto out3; + goto out2; } if ((ret = ax8817x_write_cmd(dev, AX_CMD_SET_HW_MII, 0, 0, 0, &buf)) < 0) { dbg("Failed to set hardware MII: %02x", ret); - goto out3; + goto out2; } /* Set RX_CTL to default values with 2k buffer, and enable cactus */ @@ -1097,25 +1075,16 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf) ax8817x_write_cmd(dev, AX_CMD_WRITE_RX_CTL, 0x0088, 0, 0, buf)) < 0) { dbg("Reset RX_CTL failed: %d", ret); - goto out3; - } - - if((ret = usb_submit_urb(data->int_urb, GFP_KERNEL)) < 0) { - dbg("Failed to submit interrupt URB: %02x", ret); - goto out3; + goto out2; } kfree(buf); return 0; -out3: - kfree(buf); out2: - kfree(data->int_buf); + kfree(buf); out1: - usb_free_urb(data->int_urb); - return ret; } @@ -1213,10 +1182,29 @@ static struct sk_buff *ax88772_tx_fixup(struct usbnet *dev, struct sk_buff *skb, return skb; } +static int ax88772_link_reset(struct usbnet *dev) +{ + u16 lpa; + u16 mode; + + mode = AX88772_MEDIUM_DEFAULT; + lpa = ax8817x_mdio_read(dev->net, dev->mii.phy_id, MII_LPA); + + if ((lpa & LPA_DUPLEX) == 0) + mode &= ~AX88772_MEDIUM_FULL_DUPLEX; + if ((lpa & LPA_100) == 0) + mode &= ~AX88772_MEDIUM_100MB; + ax8817x_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE, mode, 0, 0, NULL); + + return 0; +} + static const struct driver_info ax8817x_info = { .description = "ASIX AX8817x USB 2.0 Ethernet", .bind = ax8817x_bind, - .unbind = ax8817x_unbind, + .status = ax8817x_status, + .link_reset = ax88172_link_reset, + .reset = ax88172_link_reset, .flags = FLAG_ETHER, .data = 0x00130103, }; @@ -1224,7 +1212,9 @@ static const struct driver_info ax8817x_info = { static const struct driver_info dlink_dub_e100_info = { .description = "DLink DUB-E100 USB Ethernet", .bind = ax8817x_bind, - .unbind = ax8817x_unbind, + .status = ax8817x_status, + .link_reset = ax88172_link_reset, + .reset = ax88172_link_reset, .flags = FLAG_ETHER, .data = 0x009f9d9f, }; @@ -1232,7 +1222,9 @@ static const struct driver_info dlink_dub_e100_info = { static const struct driver_info netgear_fa120_info = { .description = "Netgear FA-120 USB Ethernet", .bind = ax8817x_bind, - .unbind = ax8817x_unbind, + .status = ax8817x_status, + .link_reset = ax88172_link_reset, + .reset = ax88172_link_reset, .flags = FLAG_ETHER, .data = 0x00130103, }; @@ -1240,7 +1232,9 @@ static const struct driver_info netgear_fa120_info = { static const struct driver_info hawking_uf200_info = { .description = "Hawking UF200 USB Ethernet", .bind = ax8817x_bind, - .unbind = ax8817x_unbind, + .status = ax8817x_status, + .link_reset = ax88172_link_reset, + .reset = ax88172_link_reset, .flags = FLAG_ETHER, .data = 0x001f1d1f, }; @@ -1248,7 +1242,9 @@ static const struct driver_info hawking_uf200_info = { static const struct driver_info ax88772_info = { .description = "ASIX AX88772 USB 2.0 Ethernet", .bind = ax88772_bind, - .unbind = ax8817x_unbind, + .status = ax8817x_status, + .link_reset = ax88772_link_reset, + .reset = ax88772_link_reset, .flags = FLAG_ETHER | FLAG_FRAMING_AX, .rx_fixup = ax88772_rx_fixup, .tx_fixup = ax88772_tx_fixup, @@ -2661,7 +2657,7 @@ static const struct driver_info blob_info = { * All known Zaurii lie about their standards conformance. Most lie by * saying they support CDC Ethernet. Some lie and say they support CDC * MDLM (as if for access to cell phone modems). Someone, please beat - * on Sharp for a while with a cluestick. + * on Sharp (and other such vendors) for a while with a cluestick. * *-------------------------------------------------------------------------*/ @@ -2714,13 +2710,6 @@ static const struct driver_info zaurus_pxa_info = { }; #define ZAURUS_PXA_INFO ((unsigned long)&zaurus_pxa_info) -static const struct driver_info zaurus_pxa_mdlm_info = { - .description = "Sharp Zaurus, PXA-255 based", - .flags = FLAG_FRAMING_Z, - .check_connect = always_connected, - .tx_fixup = zaurus_tx_fixup, -}; - static const struct driver_info olympus_mxl_info = { .description = "Olympus R1000", .flags = FLAG_FRAMING_Z, @@ -2731,6 +2720,133 @@ static const struct driver_info olympus_mxl_info = { }; #define OLYMPUS_MXL_INFO ((unsigned long)&olympus_mxl_info) + +/* Some more recent products using Lineo/Belcarra code will wrongly claim + * CDC MDLM conformance. They aren't conformant: data endpoints live + * in the control interface, there's no data interface, and it's not used + * to talk to a cell phone radio. But at least we can detect these two + * pseudo-classes, rather than growing this product list with entries for + * each new nonconformant product (sigh). + */ +static const u8 safe_guid[16] = { + 0x5d, 0x34, 0xcf, 0x66, 0x11, 0x18, 0x11, 0xd6, + 0xa2, 0x1a, 0x00, 0x01, 0x02, 0xca, 0x9a, 0x7f, +}; +static const u8 blan_guid[16] = { + 0x74, 0xf0, 0x3d, 0xbd, 0x1e, 0xc1, 0x44, 0x70, + 0xa3, 0x67, 0x71, 0x34, 0xc9, 0xf5, 0x54, 0x37, +}; + +static int blan_mdlm_bind (struct usbnet *dev, struct usb_interface *intf) +{ + u8 *buf = intf->cur_altsetting->extra; + int len = intf->cur_altsetting->extralen; + struct usb_cdc_mdlm_desc *desc = NULL; + struct usb_cdc_mdlm_detail_desc *detail = NULL; + + while (len > 3) { + if (buf [1] != USB_DT_CS_INTERFACE) + goto next_desc; + + /* use bDescriptorSubType, and just verify that we get a + * "BLAN" (or "SAFE") descriptor. + */ + switch (buf [2]) { + case USB_CDC_MDLM_TYPE: + if (desc) { + dev_dbg (&intf->dev, "extra MDLM\n"); + goto bad_desc; + } + desc = (void *) buf; + if (desc->bLength != sizeof *desc) { + dev_dbg (&intf->dev, "MDLM len %u\n", + desc->bLength); + goto bad_desc; + } + /* expect bcdVersion 1.0, ignore */ + if (memcmp(&desc->bGUID, blan_guid, 16) + || memcmp(&desc->bGUID, blan_guid, 16) ) { + /* hey, this one might _really_ be MDLM! */ + dev_dbg (&intf->dev, "MDLM guid\n"); + goto bad_desc; + } + break; + case USB_CDC_MDLM_DETAIL_TYPE: + if (detail) { + dev_dbg (&intf->dev, "extra MDLM detail\n"); + goto bad_desc; + } + detail = (void *) buf; + switch (detail->bGuidDescriptorType) { + case 0: /* "SAFE" */ + if (detail->bLength != (sizeof *detail + 2)) + goto bad_detail; + break; + case 1: /* "BLAN" */ + if (detail->bLength != (sizeof *detail + 3)) + goto bad_detail; + break; + default: + goto bad_detail; + } + + /* assuming we either noticed BLAN already, or will + * find it soon, there are some data bytes here: + * - bmNetworkCapabilities (unused) + * - bmDataCapabilities (bits, see below) + * - bPad (ignored, for PADAFTER -- BLAN-only) + * bits are: + * - 0x01 -- Zaurus framing (add CRC) + * - 0x02 -- PADBEFORE + * - 0x04 -- PADAFTER + * - 0x08 -- "fermat" packet mangling (for hw bugs) + */ + if (detail->bDetailData[1] != 0x01) { + /* bmDataCapabilites == 0 would be fine too, + * but framing is minidriver-coupled for now. + */ +bad_detail: + dev_dbg (&intf->dev, + "bad MDLM detail, %d %d %d\n", + detail->bLength, + detail->bDetailData[0], + detail->bDetailData[2]); + goto bad_desc; + } + break; + } +next_desc: + len -= buf [0]; /* bLength */ + buf += buf [0]; + } + + if (!desc || !detail) { + dev_dbg (&intf->dev, "missing cdc mdlm %s%sdescriptor\n", + desc ? "" : "func ", + detail ? "" : "detail "); + goto bad_desc; + } + + /* There's probably a CDC Ethernet descriptor there, but we can't + * rely on the Ethernet address it provides since not all vendors + * bother to make it unique. Likewise there's no point in tracking + * of the CDC event notifications. + */ + return get_endpoints (dev, intf); + +bad_desc: + dev_info (&dev->udev->dev, "unsupported MDLM descriptors\n"); + return -ENODEV; +} + +static const struct driver_info bogus_mdlm_info = { + .description = "pseudo-MDLM (BLAN) device", + .flags = FLAG_FRAMING_Z, + .check_connect = always_connected, + .tx_fixup = zaurus_tx_fixup, + .bind = blan_mdlm_bind, +}; + #else /* blacklist all those devices */ @@ -3307,6 +3423,19 @@ kevent (void *data) } } + if (test_bit (EVENT_LINK_RESET, &dev->flags)) { + struct driver_info *info = dev->driver_info; + int retval = 0; + + clear_bit (EVENT_LINK_RESET, &dev->flags); + if(info->link_reset && (retval = info->link_reset(dev)) < 0) { + devinfo(dev, "link reset failed (%d) usbnet usb-%s-%s, %s", + retval, + dev->udev->bus->bus_name, dev->udev->devpath, + info->description); + } + } + if (dev->flags) devdbg (dev, "kevent done, flags = 0x%lx", dev->flags); @@ -3942,6 +4071,9 @@ static const struct usb_device_id products [] = { USB_DEVICE (0x8086, 0x07d3), // "blob" bootloader .driver_info = (unsigned long) &blob_info, }, { + USB_DEVICE (0x22b8, 0x600c), // USBNET Motorola E680 + .driver_info = (unsigned long) &linuxdev_info, +}, { // Linux Ethernet/RNDIS gadget on pxa210/25x/26x // e.g. Gumstix, current OpenZaurus, ... USB_DEVICE_VER (0x0525, 0xa4a2, 0x0203, 0x0203), @@ -4020,30 +4152,14 @@ static const struct usb_device_id products [] = { }, #ifdef CONFIG_USB_ZAURUS - /* at least some (reports vary) PXA units have very different - * lies about their standards support: they claim to be cell - * phones giving direct radio access (which they aren't). + /* At least some (reports vary) PXA units have very different lies + * about their standards support: they claim to be cell phones with + * direct access to their radios. (They don't conform to CDC MDLM.) */ { - .match_flags = USB_DEVICE_ID_MATCH_INT_INFO - | USB_DEVICE_ID_MATCH_DEVICE, - .idVendor = 0x04DD, - /* Sharp ROM v1.32 */ - .idProduct = 0x8006, /* SL-5600 */ - .bInterfaceClass = USB_CLASS_COMM, - .bInterfaceSubClass = USB_CDC_SUBCLASS_MDLM, - .bInterfaceProtocol = USB_CDC_PROTO_NONE, - .driver_info = (unsigned long) &zaurus_pxa_mdlm_info, -}, { - .match_flags = USB_DEVICE_ID_MATCH_INT_INFO - | USB_DEVICE_ID_MATCH_DEVICE, - .idVendor = 0x04DD, - /* reported with some C860 units */ - .idProduct = 0x9031, /* C-860 */ - .bInterfaceClass = USB_CLASS_COMM, - .bInterfaceSubClass = USB_CDC_SUBCLASS_MDLM, - .bInterfaceProtocol = USB_CDC_PROTO_NONE, - .driver_info = (unsigned long) &zaurus_pxa_mdlm_info, + USB_INTERFACE_INFO (USB_CLASS_COMM, USB_CDC_SUBCLASS_MDLM, + USB_CDC_PROTO_NONE), + .driver_info = (unsigned long) &bogus_mdlm_info, }, #endif diff --git a/drivers/usb/net/zd1201.c b/drivers/usb/net/zd1201.c index c81cd0a619b..341ae5f732d 100644 --- a/drivers/usb/net/zd1201.c +++ b/drivers/usb/net/zd1201.c @@ -45,7 +45,7 @@ MODULE_PARM_DESC(ap, "If non-zero Access Point firmware will be loaded"); MODULE_DEVICE_TABLE(usb, zd1201_table); -int zd1201_fw_upload(struct usb_device *dev, int apfw) +static int zd1201_fw_upload(struct usb_device *dev, int apfw) { const struct firmware *fw_entry; char* data; @@ -111,7 +111,7 @@ exit: return err; } -void zd1201_usbfree(struct urb *urb, struct pt_regs *regs) +static void zd1201_usbfree(struct urb *urb, struct pt_regs *regs) { struct zd1201 *zd = urb->context; @@ -142,7 +142,8 @@ void zd1201_usbfree(struct urb *urb, struct pt_regs *regs) total: 4 + 2 + 2 + 2 + 2 + 4 = 16 */ -int zd1201_docmd(struct zd1201 *zd, int cmd, int parm0, int parm1, int parm2) +static int zd1201_docmd(struct zd1201 *zd, int cmd, int parm0, + int parm1, int parm2) { unsigned char *command; int ret; @@ -175,15 +176,15 @@ int zd1201_docmd(struct zd1201 *zd, int cmd, int parm0, int parm1, int parm2) } /* Callback after sending out a packet */ -void zd1201_usbtx(struct urb *urb, struct pt_regs *regs) +static void zd1201_usbtx(struct urb *urb, struct pt_regs *regs) { struct zd1201 *zd = urb->context; netif_wake_queue(zd->dev); return; } -/* Incomming data */ -void zd1201_usbrx(struct urb *urb, struct pt_regs *regs) +/* Incoming data */ +static void zd1201_usbrx(struct urb *urb, struct pt_regs *regs) { struct zd1201 *zd = urb->context; int free = 0; @@ -613,7 +614,7 @@ static inline int zd1201_setconfig16(struct zd1201 *zd, int rid, short val) return (zd1201_setconfig(zd, rid, &zdval, sizeof(__le16), 1)); } -int zd1201_drvr_start(struct zd1201 *zd) +static int zd1201_drvr_start(struct zd1201 *zd) { int err, i; short max; @@ -771,7 +772,7 @@ static int zd1201_net_stop(struct net_device *dev) /* RFC 1042 encapsulates Ethernet frames in 802.11 frames by prefixing them with 0xaa, 0xaa, 0x03) followed by a SNAP OID of 0 - (0x00, 0x00, 0x00). Zd requires an additionnal padding, copy + (0x00, 0x00, 0x00). Zd requires an additional padding, copy of ethernet addresses, length of the standard RFC 1042 packet and a command byte (which is nul for tx). @@ -1097,7 +1098,7 @@ static int zd1201_get_range(struct net_device *dev, /* Little bit of magic here: we only get the quality if we poll * for it, and we never get an actual request to trigger such - * a poll. Therefore we 'asume' that the user will soon ask for + * a poll. Therefore we 'assume' that the user will soon ask for * the stats after asking the bssid. */ static int zd1201_get_wap(struct net_device *dev, @@ -1107,7 +1108,7 @@ static int zd1201_get_wap(struct net_device *dev, unsigned char buffer[6]; if (!zd1201_getconfig(zd, ZD1201_RID_COMMSQUALITY, buffer, 6)) { - /* Unfortunatly the quality and noise reported is useless. + /* Unfortunately the quality and noise reported is useless. they seem to be accumulators that increase until you read them, unless we poll on a fixed interval we can't use them @@ -1739,7 +1740,8 @@ static const struct iw_handler_def zd1201_iw_handlers = { .private_args = (struct iw_priv_args *) zd1201_private_args, }; -int zd1201_probe(struct usb_interface *interface, const struct usb_device_id *id) +static int zd1201_probe(struct usb_interface *interface, + const struct usb_device_id *id) { struct zd1201 *zd; struct usb_device *usb; @@ -1851,7 +1853,7 @@ err_zd: return err; } -void zd1201_disconnect(struct usb_interface *interface) +static void zd1201_disconnect(struct usb_interface *interface) { struct zd1201 *zd=(struct zd1201 *)usb_get_intfdata(interface); struct hlist_node *node, *node2; @@ -1882,7 +1884,7 @@ void zd1201_disconnect(struct usb_interface *interface) kfree(zd); } -struct usb_driver zd1201_usb = { +static struct usb_driver zd1201_usb = { .owner = THIS_MODULE, .name = "zd1201", .probe = zd1201_probe, diff --git a/drivers/usb/serial/Kconfig b/drivers/usb/serial/Kconfig index b869076d9c7..bc798edf035 100644 --- a/drivers/usb/serial/Kconfig +++ b/drivers/usb/serial/Kconfig @@ -53,6 +53,15 @@ config USB_SERIAL_GENERIC support" be compiled as a module for this driver to be used properly. +config USB_SERIAL_AIRPRIME + tristate "USB AirPrime CDMA Wireless Driver" + depends on USB_SERIAL + help + Say Y here if you want to use a AirPrime CDMA Wireless PC card. + + To compile this driver as a module, choose M here: the + module will be called airprime. + config USB_SERIAL_BELKIN tristate "USB Belkin and Peracom Single Port Serial Driver" depends on USB_SERIAL @@ -395,6 +404,15 @@ config USB_SERIAL_PL2303 To compile this driver as a module, choose M here: the module will be called pl2303. +config USB_SERIAL_HP4X + tristate "USB HP4x Calculators support" + depends on USB_SERIAL + help + Say Y here if you want to use an Hewlett-Packard 4x Calculator. + + To compile this driver as a module, choose M here: the + module will be called hp4x. + config USB_SERIAL_SAFE tristate "USB Safe Serial (Encapsulated) Driver (EXPERIMENTAL)" depends on USB_SERIAL && EXPERIMENTAL diff --git a/drivers/usb/serial/Makefile b/drivers/usb/serial/Makefile index 351b81855b1..d56ff6d86cc 100644 --- a/drivers/usb/serial/Makefile +++ b/drivers/usb/serial/Makefile @@ -11,6 +11,7 @@ usbserial-obj-$(CONFIG_USB_EZUSB) += ezusb.o usbserial-objs := usb-serial.o generic.o bus.o $(usbserial-obj-y) +obj-$(CONFIG_USB_SERIAL_AIRPRIME) += airprime.o obj-$(CONFIG_USB_SERIAL_BELKIN) += belkin_sa.o obj-$(CONFIG_USB_SERIAL_CP2101) += cp2101.o obj-$(CONFIG_USB_SERIAL_CYBERJACK) += cyberjack.o @@ -21,6 +22,7 @@ obj-$(CONFIG_USB_SERIAL_EDGEPORT_TI) += io_ti.o obj-$(CONFIG_USB_SERIAL_EMPEG) += empeg.o obj-$(CONFIG_USB_SERIAL_FTDI_SIO) += ftdi_sio.o obj-$(CONFIG_USB_SERIAL_GARMIN) += garmin_gps.o +obj-$(CONFIG_USB_SERIAL_HP4X) += hp4x.o obj-$(CONFIG_USB_SERIAL_IPAQ) += ipaq.o obj-$(CONFIG_USB_SERIAL_IPW) += ipw.o obj-$(CONFIG_USB_SERIAL_IR) += ir-usb.o diff --git a/drivers/usb/serial/airprime.c b/drivers/usb/serial/airprime.c new file mode 100644 index 00000000000..a4ce0008d69 --- /dev/null +++ b/drivers/usb/serial/airprime.c @@ -0,0 +1,63 @@ +/* + * AirPrime CDMA Wireless Serial USB driver + * + * Copyright (C) 2005 Greg Kroah-Hartman <gregkh@suse.de> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/tty.h> +#include <linux/module.h> +#include <linux/usb.h> +#include "usb-serial.h" + +static struct usb_device_id id_table [] = { + { USB_DEVICE(0xf3d, 0x0112) }, + { }, +}; +MODULE_DEVICE_TABLE(usb, id_table); + +static struct usb_driver airprime_driver = { + .owner = THIS_MODULE, + .name = "airprime", + .probe = usb_serial_probe, + .disconnect = usb_serial_disconnect, + .id_table = id_table, +}; + +static struct usb_serial_device_type airprime_device = { + .owner = THIS_MODULE, + .name = "airprime", + .id_table = id_table, + .num_interrupt_in = NUM_DONT_CARE, + .num_bulk_in = NUM_DONT_CARE, + .num_bulk_out = NUM_DONT_CARE, + .num_ports = 1, +}; + +static int __init airprime_init(void) +{ + int retval; + + retval = usb_serial_register(&airprime_device); + if (retval) + return retval; + retval = usb_register(&airprime_driver); + if (retval) + usb_serial_deregister(&airprime_device); + return retval; +} + +static void __exit airprime_exit(void) +{ + usb_deregister(&airprime_driver); + usb_serial_deregister(&airprime_device); +} + +module_init(airprime_init); +module_exit(airprime_exit); +MODULE_LICENSE("GPL"); diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c index d165f42d560..f34a9bb6a21 100644 --- a/drivers/usb/serial/cypress_m8.c +++ b/drivers/usb/serial/cypress_m8.c @@ -16,6 +16,14 @@ * See http://geocities.com/i0xox0i for information on this driver and the * earthmate usb device. * + * Lonnie Mendez <dignome@gmail.com> + * 4-29-2005 + * Fixed problem where setting or retreiving the serial config would fail with + * EPIPE. Removed CRTS toggling so the driver behaves more like other usbserial + * adapters. Issued new interval of 1ms instead of the default 10ms. As a + * result, transfer speed has been substantially increased. From avg. 850bps to + * avg. 3300bps. initial termios has also been modified. Cleaned up code and + * formatting issues so it is more readable. Replaced the C++ style comments. * * Lonnie Mendez <dignome@gmail.com> * 12-15-2004 @@ -32,12 +40,6 @@ * 10-2003 * Driver first released. * - * - * Long Term TODO: - * Improve transfer speeds - both read/write are somewhat slow - * at this point. - * Improve debugging. Show modem line status with debug output and - * implement filtering for certain data as a module parameter. */ /* Thanks to Neil Whelchel for writing the first cypress m8 implementation for linux. */ @@ -72,11 +74,12 @@ static int debug; #endif static int stats; +static int interval; /* * Version Information */ -#define DRIVER_VERSION "v1.08" +#define DRIVER_VERSION "v1.09" #define DRIVER_AUTHOR "Lonnie Mendez <dignome@gmail.com>, Neil Whelchel <koyama@firstlight.net>" #define DRIVER_DESC "Cypress USB to Serial Driver" @@ -130,7 +133,6 @@ struct cypress_private { char prev_status, diff_status; /* used for TIOCMIWAIT */ /* we pass a pointer to this as the arguement sent to cypress_set_termios old_termios */ struct termios tmp_termios; /* stores the old termios settings */ - char calledfromopen; /* used when issuing lines on open - fixes rts drop bug */ }; /* write buffer structure */ @@ -168,10 +170,8 @@ static void cypress_buf_free(struct cypress_buf *cb); static void cypress_buf_clear(struct cypress_buf *cb); static unsigned int cypress_buf_data_avail(struct cypress_buf *cb); static unsigned int cypress_buf_space_avail(struct cypress_buf *cb); -static unsigned int cypress_buf_put(struct cypress_buf *cb, const char *buf, - unsigned int count); -static unsigned int cypress_buf_get(struct cypress_buf *cb, char *buf, - unsigned int count); +static unsigned int cypress_buf_put(struct cypress_buf *cb, const char *buf, unsigned int count); +static unsigned int cypress_buf_get(struct cypress_buf *cb, char *buf, unsigned int count); static struct usb_serial_device_type cypress_earthmate_device = { @@ -234,14 +234,13 @@ static struct usb_serial_device_type cypress_hidcom_device = { *****************************************************************************/ -/* This function can either set or retreive the current serial line settings */ +/* This function can either set or retrieve the current serial line settings */ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_mask, int data_bits, int stop_bits, int parity_enable, int parity_type, int reset, int cypress_request_type) { - int i, n_baud_rate = 0, retval = 0; + int new_baudrate = 0, retval = 0, tries = 0; struct cypress_private *priv; - __u8 feature_buffer[5]; - __u8 config; + __u8 feature_buffer[8]; unsigned long flags; dbg("%s", __FUNCTION__); @@ -256,7 +255,8 @@ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_m * of 57600bps (I have no idea whether DeLorme chose to use the general purpose * firmware or not), if you need to modify this speed setting for your own * project please add your own chiptype and modify the code likewise. The - * Cypress HID->COM device will work successfully up to 115200bps. + * Cypress HID->COM device will work successfully up to 115200bps (but the + * actual throughput is around 3kBps). */ if (baud_mask != priv->cbr_mask) { dbg("%s - baud rate is changing", __FUNCTION__); @@ -265,109 +265,114 @@ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_m * but are not used with NMEA and SiRF protocols */ if ( (baud_mask == B300) || (baud_mask == B600) ) { - err("%s - failed setting baud rate, unsupported speed (default to 4800)", + err("%s - failed setting baud rate, unsupported speed", __FUNCTION__); - n_baud_rate = 4800; - } else if ( (n_baud_rate = mask_to_rate(baud_mask)) == -1) { - err("%s - failed setting baud rate, unsupported speed (default to 4800)", + new_baudrate = priv->baud_rate; + } else if ( (new_baudrate = mask_to_rate(baud_mask)) == -1) { + err("%s - failed setting baud rate, unsupported speed", __FUNCTION__); - n_baud_rate = 4800; + new_baudrate = priv->baud_rate; } } else if (priv->chiptype == CT_CYPHIDCOM) { - if ( (n_baud_rate = mask_to_rate(baud_mask)) == -1) { - err("%s - failed setting baud rate, unsupported speed (default to 4800)", + if ( (new_baudrate = mask_to_rate(baud_mask)) == -1) { + err("%s - failed setting baud rate, unsupported speed", __FUNCTION__); - n_baud_rate = 4800; + new_baudrate = priv->baud_rate; } } else if (priv->chiptype == CT_GENERIC) { - if ( (n_baud_rate = mask_to_rate(baud_mask)) == -1) { - err("%s - failed setting baud rate, unsupported speed (default to 4800)", + if ( (new_baudrate = mask_to_rate(baud_mask)) == -1) { + err("%s - failed setting baud rate, unsupported speed", __FUNCTION__); - n_baud_rate = 4800; + new_baudrate = priv->baud_rate; } } else { - info("%s - please define your chiptype, using 4800bps default", __FUNCTION__); - n_baud_rate = 4800; + info("%s - please define your chiptype", __FUNCTION__); + new_baudrate = priv->baud_rate; } } else { /* baud rate not changing, keep the old */ - n_baud_rate = priv->baud_rate; + new_baudrate = priv->baud_rate; } - dbg("%s - baud rate is being sent as %d", __FUNCTION__, n_baud_rate); - + dbg("%s - baud rate is being sent as %d", __FUNCTION__, new_baudrate); - /* - * This algorithm accredited to Jiang Jay Zhang... thanks for all the help! - */ - for (i = 0; i < 4; ++i) { - feature_buffer[i] = ( n_baud_rate >> (i*8) & 0xFF ); - } + memset(feature_buffer, 0, 8); + /* fill the feature_buffer with new configuration */ + *((u_int32_t *)feature_buffer) = new_baudrate; - config = 0; // reset config byte - config |= data_bits; // assign data bits in 2 bit space ( max 3 ) + feature_buffer[4] |= data_bits; /* assign data bits in 2 bit space ( max 3 ) */ /* 1 bit gap */ - config |= (stop_bits << 3); // assign stop bits in 1 bit space - config |= (parity_enable << 4); // assign parity flag in 1 bit space - config |= (parity_type << 5); // assign parity type in 1 bit space + feature_buffer[4] |= (stop_bits << 3); /* assign stop bits in 1 bit space */ + feature_buffer[4] |= (parity_enable << 4); /* assign parity flag in 1 bit space */ + feature_buffer[4] |= (parity_type << 5); /* assign parity type in 1 bit space */ /* 1 bit gap */ - config |= (reset << 7); // assign reset at end of byte, 1 bit space - - feature_buffer[4] = config; + feature_buffer[4] |= (reset << 7); /* assign reset at end of byte, 1 bit space */ dbg("%s - device is being sent this feature report:", __FUNCTION__); dbg("%s - %02X - %02X - %02X - %02X - %02X", __FUNCTION__, feature_buffer[0], feature_buffer[1], feature_buffer[2], feature_buffer[3], feature_buffer[4]); + do { retval = usb_control_msg (port->serial->dev, usb_sndctrlpipe(port->serial->dev, 0), HID_REQ_SET_REPORT, USB_DIR_OUT | USB_RECIP_INTERFACE | USB_TYPE_CLASS, - 0x0300, 0, feature_buffer, 5, 500); + 0x0300, 0, feature_buffer, 8, 500); + + if (tries++ >= 3) + break; - if (retval != 5) + if (retval == EPIPE) + usb_clear_halt(port->serial->dev, 0x00); + } while (retval != 8 && retval != ENODEV); + + if (retval != 8) err("%s - failed sending serial line settings - %d", __FUNCTION__, retval); else { spin_lock_irqsave(&priv->lock, flags); - priv->baud_rate = n_baud_rate; + priv->baud_rate = new_baudrate; priv->cbr_mask = baud_mask; - priv->current_config = config; - ++priv->cmd_count; + priv->current_config = feature_buffer[4]; spin_unlock_irqrestore(&priv->lock, flags); } break; case CYPRESS_GET_CONFIG: dbg("%s - retreiving serial line settings", __FUNCTION__); - /* reset values in feature buffer */ - memset(feature_buffer, 0, 5); + /* set initial values in feature buffer */ + memset(feature_buffer, 0, 8); + do { retval = usb_control_msg (port->serial->dev, usb_rcvctrlpipe(port->serial->dev, 0), HID_REQ_GET_REPORT, USB_DIR_IN | USB_RECIP_INTERFACE | USB_TYPE_CLASS, - 0x0300, 0, feature_buffer, 5, 500); + 0x0300, 0, feature_buffer, 8, 500); + + if (tries++ >= 3) + break; + + if (retval == EPIPE) + usb_clear_halt(port->serial->dev, 0x00); + } while (retval != 5 && retval != ENODEV); + if (retval != 5) { err("%s - failed to retreive serial line settings - %d", __FUNCTION__, retval); return retval; } else { spin_lock_irqsave(&priv->lock, flags); + /* store the config in one byte, and later use bit masks to check values */ priv->current_config = feature_buffer[4]; - /* reverse the process above to get the baud_mask value */ - n_baud_rate = 0; // reset bits - for (i = 0; i < 4; ++i) { - n_baud_rate |= ( feature_buffer[i] << (i*8) ); - } + priv->baud_rate = *((u_int32_t *)feature_buffer); - priv->baud_rate = n_baud_rate; - if ( (priv->cbr_mask = rate_to_mask(n_baud_rate)) == 0x40) + if ( (priv->cbr_mask = rate_to_mask(priv->baud_rate)) == 0x40) dbg("%s - failed setting the baud mask (not defined)", __FUNCTION__); - ++priv->cmd_count; spin_unlock_irqrestore(&priv->lock, flags); } - break; - default: - err("%s - unsupported serial control command issued", __FUNCTION__); } + spin_lock_irqsave(&priv->lock, flags); + ++priv->cmd_count; + spin_unlock_irqrestore(&priv->lock, flags); + return retval; } /* cypress_serial_control */ -/* given a baud mask, it will return speed on success */ +/* given a baud mask, it will return integer baud on success */ static int mask_to_rate (unsigned mask) { int rate; @@ -438,11 +443,12 @@ static int generic_startup (struct usb_serial *serial) usb_reset_configuration (serial->dev); + interval = 1; priv->cmd_ctrl = 0; priv->line_control = 0; priv->termios_initialized = 0; - priv->calledfromopen = 0; priv->rx_flags = 0; + priv->cbr_mask = B300; usb_set_serial_port_data(serial->port[0], priv); return (0); @@ -513,7 +519,6 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp) dbg("%s - port %d", __FUNCTION__, port->number); /* clear halts before open */ - usb_clear_halt(serial->dev, 0x00); usb_clear_halt(serial->dev, 0x81); usb_clear_halt(serial->dev, 0x02); @@ -531,7 +536,6 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp) /* raise both lines and set termios */ spin_lock_irqsave(&priv->lock, flags); priv->line_control = CONTROL_DTR | CONTROL_RTS; - priv->calledfromopen = 1; priv->cmd_ctrl = 1; spin_unlock_irqrestore(&priv->lock, flags); result = cypress_write(port, NULL, 0); @@ -553,7 +557,7 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp) usb_fill_int_urb(port->interrupt_in_urb, serial->dev, usb_rcvintpipe(serial->dev, port->interrupt_in_endpointAddress), port->interrupt_in_urb->transfer_buffer, port->interrupt_in_urb->transfer_buffer_length, - cypress_read_int_callback, port, port->interrupt_in_urb->interval); + cypress_read_int_callback, port, interval); result = usb_submit_urb(port->interrupt_in_urb, GFP_KERNEL); if (result){ @@ -680,12 +684,12 @@ static void cypress_send(struct usb_serial_port *port) spin_lock_irqsave(&priv->lock, flags); switch (port->interrupt_out_size) { case 32: - // this is for the CY7C64013... + /* this is for the CY7C64013... */ offset = 2; port->interrupt_out_buffer[0] = priv->line_control; break; case 8: - // this is for the CY7C63743... + /* this is for the CY7C63743... */ offset = 1; port->interrupt_out_buffer[0] = priv->line_control; break; @@ -738,6 +742,7 @@ send: port->interrupt_out_urb->transfer_buffer_length = actual_size; port->interrupt_out_urb->dev = port->serial->dev; + port->interrupt_out_urb->interval = interval; result = usb_submit_urb (port->interrupt_out_urb, GFP_ATOMIC); if (result) { dev_err(&port->dev, "%s - failed submitting write urb, error %d\n", __FUNCTION__, @@ -910,7 +915,7 @@ static void cypress_set_termios (struct usb_serial_port *port, struct termios *o unsigned cflag, iflag, baud_mask; unsigned long flags; __u8 oldlines; - int linechange; + int linechange = 0; dbg("%s - port %d", __FUNCTION__, port->number); @@ -996,15 +1001,7 @@ static void cypress_set_termios (struct usb_serial_port *port, struct termios *o case B115200: dbg("%s - setting baud 115200bps", __FUNCTION__); break; default: dbg("%s - unknown masked baud rate", __FUNCTION__); } - priv->line_control |= CONTROL_DTR; - - /* toggle CRTSCTS? - don't do this if being called from cypress_open */ - if (!priv->calledfromopen) { - if (cflag & CRTSCTS) - priv->line_control |= CONTROL_RTS; - else - priv->line_control &= ~CONTROL_RTS; - } + priv->line_control = (CONTROL_DTR | CONTROL_RTS); } spin_unlock_irqrestore(&priv->lock, flags); @@ -1014,8 +1011,6 @@ static void cypress_set_termios (struct usb_serial_port *port, struct termios *o cypress_serial_control(port, baud_mask, data_bits, stop_bits, parity_enable, parity_type, 0, CYPRESS_SET_CONFIG); - msleep(50); /* give some time between change and read (50ms) */ - /* we perform a CYPRESS_GET_CONFIG so that the current settings are filled into the private structure * this should confirm that all is working if it returns what we just set */ cypress_serial_control(port, 0, 0, 0, 0, 0, 0, CYPRESS_GET_CONFIG); @@ -1031,7 +1026,6 @@ static void cypress_set_termios (struct usb_serial_port *port, struct termios *o dbg("Using custom termios settings for a baud rate of 4800bps."); /* define custom termios settings for NMEA protocol */ - tty->termios->c_iflag /* input modes - */ &= ~(IGNBRK /* disable ignore break */ | BRKINT /* disable break causes interrupt */ @@ -1052,23 +1046,16 @@ static void cypress_set_termios (struct usb_serial_port *port, struct termios *o | ISIG /* disable interrupt, quit, and suspend special characters */ | IEXTEN); /* disable non-POSIX special characters */ - } else if (priv->chiptype == CT_CYPHIDCOM) { - - // Software app handling it for device... + } /* CT_CYPHIDCOM: Application should handle this for device */ - } linechange = (priv->line_control != oldlines); spin_unlock_irqrestore(&priv->lock, flags); /* if necessary, set lines */ - if (!priv->calledfromopen && linechange) { + if (linechange) { priv->cmd_ctrl = 1; cypress_write(port, NULL, 0); } - - if (priv->calledfromopen) - priv->calledfromopen = 0; - } /* cypress_set_termios */ @@ -1164,7 +1151,7 @@ static void cypress_read_int_callback(struct urb *urb, struct pt_regs *regs) spin_lock_irqsave(&priv->lock, flags); switch(urb->actual_length) { case 32: - // This is for the CY7C64013... + /* This is for the CY7C64013... */ priv->current_status = data[0] & 0xF8; bytes = data[1]+2; i=2; @@ -1172,7 +1159,7 @@ static void cypress_read_int_callback(struct urb *urb, struct pt_regs *regs) havedata = 1; break; case 8: - // This is for the CY7C63743... + /* This is for the CY7C63743... */ priv->current_status = data[0] & 0xF8; bytes = (data[0] & 0x07)+1; i=1; @@ -1245,7 +1232,7 @@ continue_read: port->interrupt_in_urb->transfer_buffer, port->interrupt_in_urb->transfer_buffer_length, cypress_read_int_callback, port, - port->interrupt_in_urb->interval); + interval); result = usb_submit_urb(port->interrupt_in_urb, GFP_ATOMIC); if (result) dev_err(&urb->dev->dev, "%s - failed resubmitting read urb, error %d\n", __FUNCTION__, result); @@ -1274,6 +1261,8 @@ static void cypress_write_int_callback(struct urb *urb, struct pt_regs *regs) dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status); priv->write_urb_in_use = 0; return; + case -EPIPE: /* no break needed */ + usb_clear_halt(port->serial->dev, 0x02); default: /* error in the urb, so we have to resubmit it */ dbg("%s - Overflow in write", __FUNCTION__); @@ -1535,3 +1524,5 @@ module_param(debug, bool, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(debug, "Debug enabled or not"); module_param(stats, bool, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(stats, "Enable statistics or not"); +module_param(interval, int, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(interval, "Overrides interrupt interval"); diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index 4c788c767a9..52394f08a94 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -76,7 +76,7 @@ * Defererence pointers after any paranoid checks, not before. * * (21/Jun/2003) Erik Nygren - * Added support for Home Electronics Tira-1 IR tranceiver using FT232BM chip. + * Added support for Home Electronics Tira-1 IR transceiver using FT232BM chip. * See <http://www.home-electro.com/tira1.htm>. Only operates properly * at 100000 and RTS-CTS, so set custom divisor mode on startup. * Also force the Tira-1 and USB-UIRT to only use their custom baud rates. @@ -91,7 +91,7 @@ * Minor whitespace and comment changes. * * (12/Jun/2003) David Norwood - * Added support for USB-UIRT IR tranceiver using 8U232AM chip. + * Added support for USB-UIRT IR transceiver using 8U232AM chip. * See <http://home.earthlink.net/~jrhees/USBUIRT/index.htm>. Only * operates properly at 312500, so set custom divisor mode on startup. * @@ -272,6 +272,7 @@ static int debug; static struct usb_device_id id_table_sio [] = { { USB_DEVICE(FTDI_VID, FTDI_SIO_PID) }, + { USB_DEVICE(MOBILITY_VID, MOBILITY_USB_SERIAL_PID) }, { } /* Terminating entry */ }; @@ -296,7 +297,6 @@ static struct usb_device_id id_table_8U232AM [] = { { USB_DEVICE_VER(FTDI_VID, FTDI_IRTRANS_PID, 0, 0x3ff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_PID, 0, 0x3ff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_ALT_PID, 0, 0x3ff) }, - { USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_ALT_ALT_PID, 0, 0x3ff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_RELAIS_PID, 0, 0x3ff) }, { USB_DEVICE(INTERBIOMETRICS_VID, INTERBIOMETRICS_IOBOARD_PID) }, { USB_DEVICE(INTERBIOMETRICS_VID, INTERBIOMETRICS_MINI_IOBOARD_PID) }, @@ -369,11 +369,14 @@ static struct usb_device_id id_table_8U232AM [] = { { USB_DEVICE_VER(INTREPID_VID, INTREPID_NEOVI_PID, 0, 0x3ff) }, { USB_DEVICE_VER(FALCOM_VID, FALCOM_TWIST_PID, 0, 0x3ff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_SUUNTO_SPORTS_PID, 0, 0x3ff) }, - { USB_DEVICE_VER(FTDI_RM_VID, FTDI_RMCANVIEW_PID, 0, 0x3ff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_RM_CANVIEW_PID, 0, 0x3ff) }, { USB_DEVICE_VER(BANDB_VID, BANDB_USOTL4_PID, 0, 0x3ff) }, { USB_DEVICE_VER(BANDB_VID, BANDB_USTL4_PID, 0, 0x3ff) }, { USB_DEVICE_VER(BANDB_VID, BANDB_USO9ML2_PID, 0, 0x3ff) }, { USB_DEVICE_VER(FTDI_VID, EVER_ECO_PRO_CDS, 0, 0x3ff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_4N_GALAXY_DE_0_PID, 0, 0x3ff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_4N_GALAXY_DE_1_PID, 0, 0x3ff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_4N_GALAXY_DE_2_PID, 0, 0x3ff) }, { } /* Terminating entry */ }; @@ -382,7 +385,6 @@ static struct usb_device_id id_table_FT232BM [] = { { USB_DEVICE_VER(FTDI_VID, FTDI_IRTRANS_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_ALT_PID, 0x400, 0xffff) }, - { USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_ALT_ALT_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_RELAIS_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FTDI_NF_RIC_VID, FTDI_NF_RIC_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_XF_632_PID, 0x400, 0xffff) }, @@ -485,11 +487,15 @@ static struct usb_device_id id_table_FT232BM [] = { { USB_DEVICE_VER(INTREPID_VID, INTREPID_NEOVI_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FALCOM_VID, FALCOM_TWIST_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FTDI_VID, FTDI_SUUNTO_SPORTS_PID, 0x400, 0xffff) }, - { USB_DEVICE_VER(FTDI_RM_VID, FTDI_RMCANVIEW_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_RM_CANVIEW_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(BANDB_VID, BANDB_USOTL4_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(BANDB_VID, BANDB_USTL4_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(BANDB_VID, BANDB_USO9ML2_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FTDI_VID, EVER_ECO_PRO_CDS, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_4N_GALAXY_DE_0_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_4N_GALAXY_DE_1_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_4N_GALAXY_DE_2_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_ACTIVE_ROBOTS_PID, 0x400, 0xffff) }, { } /* Terminating entry */ }; @@ -517,7 +523,6 @@ static struct usb_device_id id_table_combined [] = { { USB_DEVICE(FTDI_VID, FTDI_SIO_PID) }, { USB_DEVICE(FTDI_VID, FTDI_8U232AM_PID) }, { USB_DEVICE(FTDI_VID, FTDI_8U232AM_ALT_PID) }, - { USB_DEVICE(FTDI_VID, FTDI_8U232AM_ALT_ALT_PID) }, { USB_DEVICE(FTDI_VID, FTDI_8U2232C_PID) }, { USB_DEVICE(FTDI_VID, FTDI_RELAIS_PID) }, { USB_DEVICE(INTERBIOMETRICS_VID, INTERBIOMETRICS_IOBOARD_PID) }, @@ -596,6 +601,22 @@ static struct usb_device_id id_table_combined [] = { { USB_DEVICE(FTDI_VID, PROTEGO_R2X0) }, { USB_DEVICE(FTDI_VID, PROTEGO_SPECIAL_3) }, { USB_DEVICE(FTDI_VID, PROTEGO_SPECIAL_4) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E808_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E809_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E80A_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E80B_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E80C_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E80D_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E80E_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E80F_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E888_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E889_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E88A_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E88B_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E88C_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E88D_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E88E_PID, 0x400, 0xffff) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_GUDEADS_E88F_PID, 0x400, 0xffff) }, { USB_DEVICE(FTDI_VID, FTDI_ELV_UO100_PID) }, { USB_DEVICE_VER(FTDI_VID, LINX_SDMUSBQSS_PID, 0x400, 0xffff) }, { USB_DEVICE_VER(FTDI_VID, LINX_MASTERDEVEL2_PID, 0x400, 0xffff) }, @@ -609,11 +630,16 @@ static struct usb_device_id id_table_combined [] = { { USB_DEVICE(INTREPID_VID, INTREPID_NEOVI_PID) }, { USB_DEVICE(FALCOM_VID, FALCOM_TWIST_PID) }, { USB_DEVICE(FTDI_VID, FTDI_SUUNTO_SPORTS_PID) }, - { USB_DEVICE(FTDI_RM_VID, FTDI_RMCANVIEW_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_RM_CANVIEW_PID) }, { USB_DEVICE(BANDB_VID, BANDB_USOTL4_PID) }, { USB_DEVICE(BANDB_VID, BANDB_USTL4_PID) }, { USB_DEVICE(BANDB_VID, BANDB_USO9ML2_PID) }, { USB_DEVICE(FTDI_VID, EVER_ECO_PRO_CDS) }, + { USB_DEVICE(FTDI_VID, FTDI_4N_GALAXY_DE_0_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_4N_GALAXY_DE_1_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_4N_GALAXY_DE_2_PID) }, + { USB_DEVICE(MOBILITY_VID, MOBILITY_USB_SERIAL_PID) }, + { USB_DEVICE_VER(FTDI_VID, FTDI_ACTIVE_ROBOTS_PID, 0x400, 0xffff) }, { } /* Terminating entry */ }; @@ -1457,10 +1483,10 @@ static int ftdi_FT2232C_startup (struct usb_serial *serial) inter = serial->interface->altsetting->desc.bInterfaceNumber; if (inter) { - priv->interface = INTERFACE_B; + priv->interface = PIT_SIOB; } else { - priv->interface = INTERFACE_A; + priv->interface = PIT_SIOA; } priv->baud_base = 48000000 / 2; /* Would be / 16, but FT2232C supports multiple of 0.125 divisor fractions! */ diff --git a/drivers/usb/serial/ftdi_sio.h b/drivers/usb/serial/ftdi_sio.h index be5d60bf90b..a52bb13a9ce 100644 --- a/drivers/usb/serial/ftdi_sio.h +++ b/drivers/usb/serial/ftdi_sio.h @@ -26,7 +26,6 @@ #define FTDI_SIO_PID 0x8372 /* Product Id SIO application of 8U100AX */ #define FTDI_8U232AM_PID 0x6001 /* Similar device to SIO above */ #define FTDI_8U232AM_ALT_PID 0x6006 /* FTDI's alternate PID for above */ -#define FTDI_8U232AM_ALT_ALT_PID 0xf3c0 /* FTDI's second alternate PID for above */ #define FTDI_8U2232C_PID 0x6010 /* Dual channel device */ #define FTDI_RELAIS_PID 0xFA10 /* Relais device from Rudolf Gugler */ #define FTDI_NF_RIC_VID 0x0DCD /* Vendor Id */ @@ -137,7 +136,7 @@ /* * Home Electronics (www.home-electro.com) USB gadgets */ -#define FTDI_HE_TIRA1_PID 0xFA78 /* Tira-1 IR tranceiver */ +#define FTDI_HE_TIRA1_PID 0xFA78 /* Tira-1 IR transceiver */ /* USB-UIRT - An infrared receiver and transmitter using the 8U232AM chip */ /* http://home.earthlink.net/~jrhees/USBUIRT/index.htm */ @@ -157,7 +156,8 @@ */ #define OCT_VID 0x0B39 /* OCT vendor ID */ /* Note: OCT US101 is also rebadged as Dick Smith Electronics (NZ) XH6381 */ -/* Also rebadged as SIIG Inc. model US2308 */ +/* Also rebadged as Dick Smith Electronics (Aus) XH6451 */ +/* Also rebadged as SIIG Inc. model US2308 hardware version 1 */ #define OCT_US101_PID 0x0421 /* OCT US101 USB to RS-232 */ /* an infrared receiver for user access control with IR tags */ @@ -236,10 +236,10 @@ /* * RM Michaelides CANview USB (http://www.rmcan.com) - * CAN filedbus interface adapter, addad by port GmbH www.port.de) + * CAN fieldbus interface adapter, added by port GmbH www.port.de) + * Ian Abbott changed the macro names for consistency. */ -#define FTDI_RM_VID 0x0403 /* Vendor Id */ -#define FTDI_RMCANVIEW_PID 0xfd60 /* Product Id */ +#define FTDI_RM_CANVIEW_PID 0xfd60 /* Product Id */ /* * EVER Eco Pro UPS (http://www.ever.com.pl/) @@ -247,6 +247,26 @@ #define EVER_ECO_PRO_CDS 0xe520 /* RS-232 converter */ +/* + * 4N-GALAXY.DE PIDs for CAN-USB, USB-RS232, USB-RS422, USB-RS485, + * USB-TTY activ, USB-TTY passiv. Some PIDs are used by several devices + * and I'm not entirely sure which are used by which. + */ +#define FTDI_4N_GALAXY_DE_0_PID 0x8372 +#define FTDI_4N_GALAXY_DE_1_PID 0xF3C0 +#define FTDI_4N_GALAXY_DE_2_PID 0xF3C1 + +/* + * Mobility Electronics products. + */ +#define MOBILITY_VID 0x1342 +#define MOBILITY_USB_SERIAL_PID 0x0202 /* EasiDock USB 200 serial */ + +/* + * Active Robots product ids. + */ +#define FTDI_ACTIVE_ROBOTS_PID 0xE548 /* USB comms board */ + /* Commands */ #define FTDI_SIO_RESET 0 /* Reset the port */ #define FTDI_SIO_MODEM_CTRL 1 /* Set the modem control register */ @@ -259,10 +279,6 @@ #define FTDI_SIO_SET_LATENCY_TIMER 9 /* Set the latency timer */ #define FTDI_SIO_GET_LATENCY_TIMER 10 /* Get the latency timer */ -/* Port interface code for FT2232C */ -#define INTERFACE_A 1 -#define INTERFACE_B 2 - /* * BmRequestType: 1100 0000b diff --git a/drivers/usb/serial/hp4x.c b/drivers/usb/serial/hp4x.c new file mode 100644 index 00000000000..64d55fbd206 --- /dev/null +++ b/drivers/usb/serial/hp4x.c @@ -0,0 +1,85 @@ +/* + * HP4x Calculators Serial USB driver + * + * Copyright (C) 2005 Arthur Huillet (ahuillet@users.sf.net) + * Copyright (C) 2001-2005 Greg Kroah-Hartman (greg@kroah.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * See Documentation/usb/usb-serial.txt for more information on using this driver + */ + +#include <linux/config.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/tty.h> +#include <linux/module.h> +#include <linux/usb.h> +#include "usb-serial.h" + +/* + * Version Information + */ +#define DRIVER_VERSION "v1.00" +#define DRIVER_DESC "HP4x (48/49) Generic Serial driver" + +#define HP_VENDOR_ID 0x03f0 +#define HP49GP_PRODUCT_ID 0x0121 + +static struct usb_device_id id_table [] = { + { USB_DEVICE(HP_VENDOR_ID, HP49GP_PRODUCT_ID) }, + { } /* Terminating entry */ +}; + +MODULE_DEVICE_TABLE(usb, id_table); + +static struct usb_driver hp49gp_driver = { + .owner = THIS_MODULE, + .name = "HP4X", + .probe = usb_serial_probe, + .disconnect = usb_serial_disconnect, + .id_table = id_table, +}; + +static struct usb_serial_device_type hp49gp_device = { + .owner = THIS_MODULE, + .name = "HP4X", + .id_table = id_table, + .num_interrupt_in = NUM_DONT_CARE, + .num_bulk_in = NUM_DONT_CARE, + .num_bulk_out = NUM_DONT_CARE, + .num_ports = 1, +}; + +static int __init hp49gp_init(void) +{ + int retval; + retval = usb_serial_register(&hp49gp_device); + if (retval) + goto failed_usb_serial_register; + retval = usb_register(&hp49gp_driver); + if (retval) + goto failed_usb_register; + info(DRIVER_DESC " " DRIVER_VERSION); + return 0; +failed_usb_register: + usb_serial_deregister(&hp49gp_device); +failed_usb_serial_register: + return retval; +} + +static void __exit hp49gp_exit(void) +{ + usb_deregister(&hp49gp_driver); + usb_serial_deregister(&hp49gp_device); +} + +module_init(hp49gp_init); +module_exit(hp49gp_exit); + +MODULE_DESCRIPTION(DRIVER_DESC); +MODULE_VERSION(DRIVER_VERSION); +MODULE_LICENSE("GPL"); diff --git a/drivers/usb/serial/io_usbvend.h b/drivers/usb/serial/io_usbvend.h index 8c1fa5e722b..f1804fd5a3d 100644 --- a/drivers/usb/serial/io_usbvend.h +++ b/drivers/usb/serial/io_usbvend.h @@ -289,7 +289,7 @@ // // -// Edgeport Compatiblity Descriptor +// Edgeport Compatibility Descriptor // // This descriptor is only returned by Edgeport-compatible devices // supporting the EPiC spec. True ION devices do not return this diff --git a/drivers/usb/serial/keyspan_usa90msg.h b/drivers/usb/serial/keyspan_usa90msg.h index dd935b62c1a..86708ecd873 100644 --- a/drivers/usb/serial/keyspan_usa90msg.h +++ b/drivers/usb/serial/keyspan_usa90msg.h @@ -19,7 +19,7 @@ This file is available under a BSD-style copyright - 2. The name of InnoSys Incorprated may not be used to endorse or promote + 2. The name of InnoSys Incorporated may not be used to endorse or promote products derived from this software without specific prior written permission. diff --git a/drivers/usb/storage/debug.c b/drivers/usb/storage/debug.c index d76483706bc..5a9321705a7 100644 --- a/drivers/usb/storage/debug.c +++ b/drivers/usb/storage/debug.c @@ -47,6 +47,7 @@ #include <linux/cdrom.h> #include <scsi/scsi.h> #include <scsi/scsi_cmnd.h> +#include <scsi/scsi_dbg.h> #include "debug.h" #include "scsi.h" diff --git a/drivers/usb/storage/shuttle_usbat.c b/drivers/usb/storage/shuttle_usbat.c index 7eff03d9b04..f3b60288696 100644 --- a/drivers/usb/storage/shuttle_usbat.c +++ b/drivers/usb/storage/shuttle_usbat.c @@ -786,7 +786,7 @@ static int usbat_flash_check_media(struct us_data *us, if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; - // Check for media existance + // Check for media existence rc = usbat_flash_check_media_present(uio); if (rc == USBAT_FLASH_MEDIA_NONE) { info->sense_key = 0x02; diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index fa68dea6bc6..d2891f47579 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1,5 +1,5 @@ /* Driver for USB Mass Storage compliant devices - * Ununsual Devices File + * Unusual Devices File * * $Id: unusual_devs.h,v 1.32 2002/02/25 02:41:24 mdharm Exp $ * @@ -48,6 +48,14 @@ * USB development list <linux-usb-devel@lists.sourceforge.net>. */ +/* patch submitted by Vivian Bregier <Vivian.Bregier@imag.fr> + */ +UNUSUAL_DEV( 0x03eb, 0x2002, 0x0100, 0x0100, + "ATMEL", + "SND1 Storage", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_IGNORE_RESIDUE), + UNUSUAL_DEV( 0x03ee, 0x6901, 0x0000, 0x0100, "Mitsumi", "USB FDD", @@ -517,14 +525,32 @@ UNUSUAL_DEV( 0x05ab, 0x5701, 0x0100, 0x0110, 0 ), #endif +/* Submitted by Sven Anderson <sven-linux@anderson.de> + * There are at least four ProductIDs used for iPods, so I added 0x1202 and + * 0x1204. They just need the US_FL_FIX_CAPACITY. As the bcdDevice appears + * to change with firmware updates, I changed the range to maximum for all + * iPod entries. + */ +UNUSUAL_DEV( 0x05ac, 0x1202, 0x0000, 0x9999, + "Apple", + "iPod", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_FIX_CAPACITY ), + /* Reported by Avi Kivity <avi@argo.co.il> */ -UNUSUAL_DEV( 0x05ac, 0x1203, 0x0001, 0x0001, +UNUSUAL_DEV( 0x05ac, 0x1203, 0x0000, 0x9999, "Apple", "iPod", US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_CAPACITY ), -UNUSUAL_DEV( 0x05ac, 0x1205, 0x0001, 0x0001, +UNUSUAL_DEV( 0x05ac, 0x1204, 0x0000, 0x9999, + "Apple", + "iPod", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_FIX_CAPACITY ), + +UNUSUAL_DEV( 0x05ac, 0x1205, 0x0000, 0x9999, "Apple", "iPod", US_SC_DEVICE, US_PR_DEVICE, NULL, @@ -976,6 +1002,13 @@ UNUSUAL_DEV( 0x1019, 0x0c55, 0x0000, 0x9999, US_SC_DEVICE, US_PR_DEVICE, usb_stor_ucr61s2b_init, 0 ), +/* Reported by Vilius Bilinkevicius <vilisas AT xxx DOT lt) */ +UNUSUAL_DEV( 0x132b, 0x000b, 0x0001, 0x0001, + "Minolta", + "Dimage Z10", + US_SC_DEVICE, US_PR_DEVICE, NULL, + 0 ), + /* Reported by Kotrla Vitezslav <kotrla@ceb.cz> */ UNUSUAL_DEV( 0x1370, 0x6828, 0x0110, 0x0110, "SWISSBIT", diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index 2a1c5965de2..6be8fbec0a0 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -198,6 +198,14 @@ config FB_SA1100 If you plan to use the LCD display with your SA-1100 system, say Y here. +config FB_IMX + tristate "Motorola i.MX LCD support" + depends on FB && ARM && ARCH_IMX + select FB_CFB_FILLRECT + select FB_CFB_COPYAREA + select FB_CFB_IMAGEBLIT + select FB_SOFT_CURSOR + config FB_CYBER2000 tristate "CyberPro 2000/2010/5000 support" depends on FB && PCI && (BROKEN || !SPARC64) diff --git a/drivers/video/Makefile b/drivers/video/Makefile index 92265b741dc..bd8dc0ffe72 100644 --- a/drivers/video/Makefile +++ b/drivers/video/Makefile @@ -90,6 +90,7 @@ obj-$(CONFIG_FB_PMAGB_B) += pmagb-b-fb.o obj-$(CONFIG_FB_MAXINE) += maxinefb.o obj-$(CONFIG_FB_TX3912) += tx3912fb.o obj-$(CONFIG_FB_S1D13XXX) += s1d13xxxfb.o +obj-$(CONFIG_FB_IMX) += imxfb.o # Platform or fallback drivers go here obj-$(CONFIG_FB_VESA) += vesafb.o diff --git a/drivers/video/amba-clcd.c b/drivers/video/amba-clcd.c index acdba0c67fb..321dbe91dc1 100644 --- a/drivers/video/amba-clcd.c +++ b/drivers/video/amba-clcd.c @@ -125,28 +125,28 @@ clcdfb_set_bitfields(struct clcd_fb *fb, struct fb_var_screeninfo *var) case 2: case 4: case 8: - var->red.length = 8; + var->red.length = var->bits_per_pixel; var->red.offset = 0; - var->green.length = 8; + var->green.length = var->bits_per_pixel; var->green.offset = 0; - var->blue.length = 8; + var->blue.length = var->bits_per_pixel; var->blue.offset = 0; break; case 16: var->red.length = 5; - var->green.length = 5; + var->green.length = 6; var->blue.length = 5; if (fb->panel->cntl & CNTL_BGR) { - var->red.offset = 10; + var->red.offset = 11; var->green.offset = 5; var->blue.offset = 0; } else { var->red.offset = 0; var->green.offset = 5; - var->blue.offset = 10; + var->blue.offset = 11; } break; - case 24: + case 32: if (fb->panel->cntl & CNTL_LCDTFT) { var->red.length = 8; var->green.length = 8; @@ -178,6 +178,12 @@ static int clcdfb_check_var(struct fb_var_screeninfo *var, struct fb_info *info) if (fb->board->check) ret = fb->board->check(fb, var); + + if (ret == 0 && + var->xres_virtual * var->bits_per_pixel / 8 * + var->yres_virtual > fb->fb.fix.smem_len) + ret = -EINVAL; + if (ret == 0) ret = clcdfb_set_bitfields(fb, var); @@ -250,7 +256,7 @@ clcdfb_setcolreg(unsigned int regno, unsigned int red, unsigned int green, convert_bitfield(green, &fb->fb.var.green) | convert_bitfield(red, &fb->fb.var.red); - if (fb->fb.var.bits_per_pixel == 8 && regno < 256) { + if (fb->fb.fix.visual == FB_VISUAL_PSEUDOCOLOR && regno < 256) { int hw_reg = CLCD_PALETTE + ((regno * 2) & ~3); u32 val, mask, newval; diff --git a/drivers/video/aty/radeon_base.c b/drivers/video/aty/radeon_base.c index e8eb124754b..ee25b9e8db6 100644 --- a/drivers/video/aty/radeon_base.c +++ b/drivers/video/aty/radeon_base.c @@ -1057,13 +1057,14 @@ static int radeonfb_blank (int blank, struct fb_info *info) return radeon_screen_blank(rinfo, blank, 0); } -static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, - unsigned blue, unsigned transp, struct fb_info *info) +static int radeon_setcolreg (unsigned regno, unsigned red, unsigned green, + unsigned blue, unsigned transp, + struct radeonfb_info *rinfo) { - struct radeonfb_info *rinfo = info->par; u32 pindex; unsigned int i; - + + if (regno > 255) return 1; @@ -1078,20 +1079,7 @@ static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, pindex = regno; if (!rinfo->asleep) { - u32 dac_cntl2, vclk_cntl = 0; - radeon_fifo_wait(9); - if (rinfo->is_mobility) { - vclk_cntl = INPLL(VCLK_ECP_CNTL); - OUTPLL(VCLK_ECP_CNTL, vclk_cntl & ~PIXCLK_DAC_ALWAYS_ONb); - } - - /* Make sure we are on first palette */ - if (rinfo->has_CRTC2) { - dac_cntl2 = INREG(DAC_CNTL2); - dac_cntl2 &= ~DAC2_PALETTE_ACCESS_CNTL; - OUTREG(DAC_CNTL2, dac_cntl2); - } if (rinfo->bpp == 16) { pindex = regno * 8; @@ -1101,24 +1089,27 @@ static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, if (rinfo->depth == 15 && regno > 31) return 1; - /* For 565, the green component is mixed one order below */ + /* For 565, the green component is mixed one order + * below + */ if (rinfo->depth == 16) { OUTREG(PALETTE_INDEX, pindex>>1); - OUTREG(PALETTE_DATA, (rinfo->palette[regno>>1].red << 16) | - (green << 8) | (rinfo->palette[regno>>1].blue)); + OUTREG(PALETTE_DATA, + (rinfo->palette[regno>>1].red << 16) | + (green << 8) | + (rinfo->palette[regno>>1].blue)); green = rinfo->palette[regno<<1].green; } } if (rinfo->depth != 16 || regno < 32) { OUTREG(PALETTE_INDEX, pindex); - OUTREG(PALETTE_DATA, (red << 16) | (green << 8) | blue); + OUTREG(PALETTE_DATA, (red << 16) | + (green << 8) | blue); } - if (rinfo->is_mobility) - OUTPLL(VCLK_ECP_CNTL, vclk_cntl); } if (regno < 16) { - u32 *pal = info->pseudo_palette; + u32 *pal = rinfo->info->pseudo_palette; switch (rinfo->depth) { case 15: pal[regno] = (regno << 10) | (regno << 5) | regno; @@ -1138,6 +1129,84 @@ static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, return 0; } +static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, + unsigned blue, unsigned transp, + struct fb_info *info) +{ + struct radeonfb_info *rinfo = info->par; + u32 dac_cntl2, vclk_cntl = 0; + int rc; + + if (!rinfo->asleep) { + if (rinfo->is_mobility) { + vclk_cntl = INPLL(VCLK_ECP_CNTL); + OUTPLL(VCLK_ECP_CNTL, + vclk_cntl & ~PIXCLK_DAC_ALWAYS_ONb); + } + + /* Make sure we are on first palette */ + if (rinfo->has_CRTC2) { + dac_cntl2 = INREG(DAC_CNTL2); + dac_cntl2 &= ~DAC2_PALETTE_ACCESS_CNTL; + OUTREG(DAC_CNTL2, dac_cntl2); + } + } + + rc = radeon_setcolreg (regno, red, green, blue, transp, rinfo); + + if (!rinfo->asleep && rinfo->is_mobility) + OUTPLL(VCLK_ECP_CNTL, vclk_cntl); + + return rc; +} + +static int radeonfb_setcmap(struct fb_cmap *cmap, struct fb_info *info) +{ + struct radeonfb_info *rinfo = info->par; + u16 *red, *green, *blue, *transp; + u32 dac_cntl2, vclk_cntl = 0; + int i, start, rc = 0; + + if (!rinfo->asleep) { + if (rinfo->is_mobility) { + vclk_cntl = INPLL(VCLK_ECP_CNTL); + OUTPLL(VCLK_ECP_CNTL, + vclk_cntl & ~PIXCLK_DAC_ALWAYS_ONb); + } + + /* Make sure we are on first palette */ + if (rinfo->has_CRTC2) { + dac_cntl2 = INREG(DAC_CNTL2); + dac_cntl2 &= ~DAC2_PALETTE_ACCESS_CNTL; + OUTREG(DAC_CNTL2, dac_cntl2); + } + } + + red = cmap->red; + green = cmap->green; + blue = cmap->blue; + transp = cmap->transp; + start = cmap->start; + + for (i = 0; i < cmap->len; i++) { + u_int hred, hgreen, hblue, htransp = 0xffff; + + hred = *red++; + hgreen = *green++; + hblue = *blue++; + if (transp) + htransp = *transp++; + rc = radeon_setcolreg (start++, hred, hgreen, hblue, htransp, + rinfo); + if (rc) + break; + } + + if (!rinfo->asleep && rinfo->is_mobility) + OUTPLL(VCLK_ECP_CNTL, vclk_cntl); + + return rc; +} static void radeon_save_state (struct radeonfb_info *rinfo, struct radeon_regs *save) @@ -1796,6 +1865,7 @@ static struct fb_ops radeonfb_ops = { .fb_check_var = radeonfb_check_var, .fb_set_par = radeonfb_set_par, .fb_setcolreg = radeonfb_setcolreg, + .fb_setcmap = radeonfb_setcmap, .fb_pan_display = radeonfb_pan_display, .fb_blank = radeonfb_blank, .fb_ioctl = radeonfb_ioctl, diff --git a/drivers/video/console/fbcon.c b/drivers/video/console/fbcon.c index 59e3b4b4e7e..b209adbd508 100644 --- a/drivers/video/console/fbcon.c +++ b/drivers/video/console/fbcon.c @@ -906,10 +906,13 @@ static void fbcon_init(struct vc_data *vc, int init) struct vc_data *svc = *default_mode; struct display *t, *p = &fb_display[vc->vc_num]; int logo = 1, new_rows, new_cols, rows, cols, charcnt = 256; - int cap = info->flags; + int cap; if (info_idx == -1 || info == NULL) return; + + cap = info->flags; + if (vc != svc || logo_shown == FBCON_LOGO_DONTSHOW || (info->fix.type == FB_TYPE_TEXT)) logo = 0; diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c index 7d1ae06667c..bcf59b28a14 100644 --- a/drivers/video/console/vgacon.c +++ b/drivers/video/console/vgacon.c @@ -337,6 +337,8 @@ static void vgacon_init(struct vc_data *c, int init) c->vc_scan_lines = vga_scan_lines; c->vc_font.height = vga_video_font_height; c->vc_complement_mask = 0x7700; + if (vga_512_chars) + c->vc_hi_font_mask = 0x0800; p = *c->vc_uni_pagedir_loc; if (c->vc_uni_pagedir_loc == &c->vc_uni_pagedir || !--c->vc_uni_pagedir_loc[1]) diff --git a/drivers/video/cyber2000fb.c b/drivers/video/cyber2000fb.c index 8b1b7c687a9..3894b2a501d 100644 --- a/drivers/video/cyber2000fb.c +++ b/drivers/video/cyber2000fb.c @@ -90,6 +90,8 @@ struct cfb_info { */ u_char ramdac_ctrl; u_char ramdac_powerdown; + + u32 pseudo_palette[16]; }; static char *default_font = "Acorn8x8"; @@ -1223,9 +1225,7 @@ cyberpro_alloc_fb_info(unsigned int id, char *name) { struct cfb_info *cfb; - cfb = kmalloc(sizeof(struct cfb_info) + - sizeof(u32) * 16, GFP_KERNEL); - + cfb = kmalloc(sizeof(struct cfb_info), GFP_KERNEL); if (!cfb) return NULL; @@ -1281,7 +1281,7 @@ cyberpro_alloc_fb_info(unsigned int id, char *name) cfb->fb.fbops = &cyber2000fb_ops; cfb->fb.flags = FBINFO_DEFAULT | FBINFO_HWACCEL_YPAN; - cfb->fb.pseudo_palette = (void *)(cfb + 1); + cfb->fb.pseudo_palette = cfb->pseudo_palette; fb_alloc_cmap(&cfb->fb.cmap, NR_PALETTE, 0); diff --git a/drivers/video/fbcmap.c b/drivers/video/fbcmap.c index c51f8fb5c1d..4e5ce8f7d65 100644 --- a/drivers/video/fbcmap.c +++ b/drivers/video/fbcmap.c @@ -222,8 +222,11 @@ int fb_set_cmap(struct fb_cmap *cmap, struct fb_info *info) transp = cmap->transp; start = cmap->start; - if (start < 0 || !info->fbops->fb_setcolreg) + if (start < 0 || (!info->fbops->fb_setcolreg && + !info->fbops->fb_setcmap)) return -EINVAL; + if (info->fbops->fb_setcmap) + return info->fbops->fb_setcmap(cmap, info); for (i = 0; i < cmap->len; i++) { hred = *red++; hgreen = *green++; @@ -250,8 +253,33 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info) transp = cmap->transp; start = cmap->start; - if (start < 0 || !info->fbops->fb_setcolreg) + if (start < 0 || (!info->fbops->fb_setcolreg && + !info->fbops->fb_setcmap)) return -EINVAL; + + /* If we can batch, do it */ + if (info->fbops->fb_setcmap && cmap->len > 1) { + struct fb_cmap umap; + int size = cmap->len * sizeof(u16); + int rc; + + memset(&umap, 0, sizeof(struct fb_cmap)); + rc = fb_alloc_cmap(&umap, cmap->len, transp != NULL); + if (rc) + return rc; + if (copy_from_user(umap.red, red, size) || + copy_from_user(umap.green, green, size) || + copy_from_user(umap.blue, blue, size) || + (transp && copy_from_user(umap.transp, transp, size))) { + rc = -EFAULT; + } + umap.start = start; + if (rc == 0) + rc = info->fbops->fb_setcmap(&umap, info); + fb_dealloc_cmap(&umap); + return rc; + } + for (i = 0; i < cmap->len; i++, red++, blue++, green++) { if (get_user(hred, red) || get_user(hgreen, green) || diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 25f460ca0da..7705070191d 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1257,6 +1257,8 @@ int fb_new_modelist(struct fb_info *info) static char *video_options[FB_MAX]; static int ofonly; +extern const char *global_mode_option; + /** * fb_get_options - get kernel boot parameters * @name: framebuffer name as it would appear in @@ -1297,9 +1299,6 @@ int fb_get_options(char *name, char **option) return retval; } - -extern const char *global_mode_option; - /** * video_setup - process command line options * @options: string of options @@ -1313,7 +1312,7 @@ extern const char *global_mode_option; * Returns zero. * */ -int __init video_setup(char *options) +static int __init video_setup(char *options) { int i, global = 0; diff --git a/drivers/video/fbmon.c b/drivers/video/fbmon.c index 978def01358..6cd1976548d 100644 --- a/drivers/video/fbmon.c +++ b/drivers/video/fbmon.c @@ -34,7 +34,6 @@ #include <asm/prom.h> #include <asm/pci-bridge.h> #endif -#include <video/edid.h> #include "edid.h" /* diff --git a/drivers/video/fbsysfs.c b/drivers/video/fbsysfs.c index 2bdda4010b8..c78a2c5961d 100644 --- a/drivers/video/fbsysfs.c +++ b/drivers/video/fbsysfs.c @@ -354,7 +354,7 @@ static ssize_t show_pan(struct class_device *class_device, char *buf) fb_info->var.xoffset); } -struct class_device_attribute class_device_attrs[] = { +static struct class_device_attribute class_device_attrs[] = { __ATTR(bits_per_pixel, S_IRUGO|S_IWUSR, show_bpp, store_bpp), __ATTR(blank, S_IRUGO|S_IWUSR, show_blank, store_blank), __ATTR(color_map, S_IRUGO|S_IWUSR, show_cmap, store_cmap), diff --git a/drivers/video/i810/i810_main.c b/drivers/video/i810/i810_main.c index 9ec8781794c..a9a618f2aa6 100644 --- a/drivers/video/i810/i810_main.c +++ b/drivers/video/i810/i810_main.c @@ -999,8 +999,16 @@ static int i810_check_params(struct fb_var_screeninfo *var, info->monspecs.dclkmin = 15000000; if (fb_validate_mode(var, info)) { - if (fb_get_mode(FB_MAXTIMINGS, 0, var, info)) + if (fb_get_mode(FB_MAXTIMINGS, 0, var, info)) { + int default_sync = (info->monspecs.hfmin-HFMIN) + |(info->monspecs.hfmax-HFMAX) + |(info->monspecs.vfmin-VFMIN) + |(info->monspecs.vfmax-VFMAX); + printk("i810fb: invalid video mode%s\n", + default_sync ? "" : + ". Specifying vsyncN/hsyncN parameters may help"); return -EINVAL; + } } var->xres = xres; @@ -2023,10 +2031,10 @@ MODULE_PARM_DESC(vyres, "Virtual vertical resolution in scanlines" " (default = 480)"); module_param(hsync1, int, 0); MODULE_PARM_DESC(hsync1, "Minimum horizontal frequency of monitor in KHz" - " (default = 31)"); + " (default = 29)"); module_param(hsync2, int, 0); MODULE_PARM_DESC(hsync2, "Maximum horizontal frequency of monitor in KHz" - " (default = 31)"); + " (default = 30)"); module_param(vsync1, int, 0); MODULE_PARM_DESC(vsync1, "Minimum vertical frequency of monitor in Hz" " (default = 50)"); diff --git a/drivers/video/imsttfb.c b/drivers/video/imsttfb.c index 5a72ca3c013..7b9bf45ab6f 100644 --- a/drivers/video/imsttfb.c +++ b/drivers/video/imsttfb.c @@ -1287,12 +1287,12 @@ imsttfb_ioctl(struct inode *inode, struct file *file, u_int cmd, case FBIMSTT_SETCMAPREG: if (copy_from_user(reg, argp, 8) || reg[0] > (0x1000 - sizeof(reg[0])) / sizeof(reg[0])) return -EFAULT; - write_reg_le32(((u_int *)par->cmap_regs), reg[0], reg[1]); + write_reg_le32(((u_int __iomem *)par->cmap_regs), reg[0], reg[1]); return 0; case FBIMSTT_GETCMAPREG: if (copy_from_user(reg, argp, 4) || reg[0] > (0x1000 - sizeof(reg[0])) / sizeof(reg[0])) return -EFAULT; - reg[1] = read_reg_le32(((u_int *)par->cmap_regs), reg[0]); + reg[1] = read_reg_le32(((u_int __iomem *)par->cmap_regs), reg[0]); if (copy_to_user((void __user *)(arg + 4), ®[1], 4)) return -EFAULT; return 0; diff --git a/drivers/video/imxfb.c b/drivers/video/imxfb.c new file mode 100644 index 00000000000..8fe1c12a17b --- /dev/null +++ b/drivers/video/imxfb.c @@ -0,0 +1,695 @@ +/* + * linux/drivers/video/imxfb.c + * + * Freescale i.MX Frame Buffer device driver + * + * Copyright (C) 2004 Sascha Hauer, Pengutronix + * Based on acornfb.c Copyright (C) Russell King. + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file COPYING in the main directory of this archive for + * more details. + * + * Please direct your questions and comments on this driver to the following + * email address: + * + * linux-arm-kernel@lists.arm.linux.org.uk + */ + +//#define DEBUG 1 + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/errno.h> +#include <linux/string.h> +#include <linux/interrupt.h> +#include <linux/slab.h> +#include <linux/fb.h> +#include <linux/delay.h> +#include <linux/init.h> +#include <linux/ioport.h> +#include <linux/cpufreq.h> +#include <linux/device.h> +#include <linux/dma-mapping.h> + +#include <asm/hardware.h> +#include <asm/io.h> +#include <asm/mach-types.h> +#include <asm/uaccess.h> +#include <asm/arch/imxfb.h> + +/* + * Complain if VAR is out of range. + */ +#define DEBUG_VAR 1 + +#include "imxfb.h" + +static struct imxfb_rgb def_rgb_16 = { + .red = { .offset = 8, .length = 4, }, + .green = { .offset = 4, .length = 4, }, + .blue = { .offset = 0, .length = 4, }, + .transp = { .offset = 0, .length = 0, }, +}; + +static struct imxfb_rgb def_rgb_8 = { + .red = { .offset = 0, .length = 8, }, + .green = { .offset = 0, .length = 8, }, + .blue = { .offset = 0, .length = 8, }, + .transp = { .offset = 0, .length = 0, }, +}; + +static int imxfb_activate_var(struct fb_var_screeninfo *var, struct fb_info *info); + +static inline u_int chan_to_field(u_int chan, struct fb_bitfield *bf) +{ + chan &= 0xffff; + chan >>= 16 - bf->length; + return chan << bf->offset; +} + +#define LCDC_PALETTE(x) __REG2(IMX_LCDC_BASE+0x800, (x)<<2) +static int +imxfb_setpalettereg(u_int regno, u_int red, u_int green, u_int blue, + u_int trans, struct fb_info *info) +{ + struct imxfb_info *fbi = info->par; + u_int val, ret = 1; + +#define CNVT_TOHW(val,width) ((((val)<<(width))+0x7FFF-(val))>>16) + if (regno < fbi->palette_size) { + val = (CNVT_TOHW(red, 4) << 8) | + (CNVT_TOHW(green,4) << 4) | + CNVT_TOHW(blue, 4); + + LCDC_PALETTE(regno) = val; + ret = 0; + } + return ret; +} + +static int +imxfb_setcolreg(u_int regno, u_int red, u_int green, u_int blue, + u_int trans, struct fb_info *info) +{ + struct imxfb_info *fbi = info->par; + unsigned int val; + int ret = 1; + + /* + * If inverse mode was selected, invert all the colours + * rather than the register number. The register number + * is what you poke into the framebuffer to produce the + * colour you requested. + */ + if (fbi->cmap_inverse) { + red = 0xffff - red; + green = 0xffff - green; + blue = 0xffff - blue; + } + + /* + * If greyscale is true, then we convert the RGB value + * to greyscale no mater what visual we are using. + */ + if (info->var.grayscale) + red = green = blue = (19595 * red + 38470 * green + + 7471 * blue) >> 16; + + switch (info->fix.visual) { + case FB_VISUAL_TRUECOLOR: + /* + * 12 or 16-bit True Colour. We encode the RGB value + * according to the RGB bitfield information. + */ + if (regno < 16) { + u32 *pal = info->pseudo_palette; + + val = chan_to_field(red, &info->var.red); + val |= chan_to_field(green, &info->var.green); + val |= chan_to_field(blue, &info->var.blue); + + pal[regno] = val; + ret = 0; + } + break; + + case FB_VISUAL_STATIC_PSEUDOCOLOR: + case FB_VISUAL_PSEUDOCOLOR: + ret = imxfb_setpalettereg(regno, red, green, blue, trans, info); + break; + } + + return ret; +} + +/* + * imxfb_check_var(): + * Round up in the following order: bits_per_pixel, xres, + * yres, xres_virtual, yres_virtual, xoffset, yoffset, grayscale, + * bitfields, horizontal timing, vertical timing. + */ +static int +imxfb_check_var(struct fb_var_screeninfo *var, struct fb_info *info) +{ + struct imxfb_info *fbi = info->par; + int rgbidx; + + if (var->xres < MIN_XRES) + var->xres = MIN_XRES; + if (var->yres < MIN_YRES) + var->yres = MIN_YRES; + if (var->xres > fbi->max_xres) + var->xres = fbi->max_xres; + if (var->yres > fbi->max_yres) + var->yres = fbi->max_yres; + var->xres_virtual = max(var->xres_virtual, var->xres); + var->yres_virtual = max(var->yres_virtual, var->yres); + + pr_debug("var->bits_per_pixel=%d\n", var->bits_per_pixel); + switch (var->bits_per_pixel) { + case 16: + rgbidx = RGB_16; + break; + case 8: + rgbidx = RGB_8; + break; + default: + rgbidx = RGB_16; + } + + /* + * Copy the RGB parameters for this display + * from the machine specific parameters. + */ + var->red = fbi->rgb[rgbidx]->red; + var->green = fbi->rgb[rgbidx]->green; + var->blue = fbi->rgb[rgbidx]->blue; + var->transp = fbi->rgb[rgbidx]->transp; + + pr_debug("RGBT length = %d:%d:%d:%d\n", + var->red.length, var->green.length, var->blue.length, + var->transp.length); + + pr_debug("RGBT offset = %d:%d:%d:%d\n", + var->red.offset, var->green.offset, var->blue.offset, + var->transp.offset); + + return 0; +} + +/* + * imxfb_set_par(): + * Set the user defined part of the display for the specified console + */ +static int imxfb_set_par(struct fb_info *info) +{ + struct imxfb_info *fbi = info->par; + struct fb_var_screeninfo *var = &info->var; + + pr_debug("set_par\n"); + + if (var->bits_per_pixel == 16) + info->fix.visual = FB_VISUAL_TRUECOLOR; + else if (!fbi->cmap_static) + info->fix.visual = FB_VISUAL_PSEUDOCOLOR; + else { + /* + * Some people have weird ideas about wanting static + * pseudocolor maps. I suspect their user space + * applications are broken. + */ + info->fix.visual = FB_VISUAL_STATIC_PSEUDOCOLOR; + } + + info->fix.line_length = var->xres_virtual * + var->bits_per_pixel / 8; + fbi->palette_size = var->bits_per_pixel == 8 ? 256 : 16; + + imxfb_activate_var(var, info); + + return 0; +} + +static void imxfb_enable_controller(struct imxfb_info *fbi) +{ + pr_debug("Enabling LCD controller\n"); + + /* initialize LCDC */ + LCDC_RMCR &= ~RMCR_LCDC_EN; /* just to be safe... */ + + LCDC_SSA = fbi->screen_dma; + /* physical screen start address */ + LCDC_VPW = VPW_VPW(fbi->max_xres * fbi->max_bpp / 8 / 4); + + LCDC_POS = 0x00000000; /* panning offset 0 (0 pixel offset) */ + + /* disable hardware cursor */ + LCDC_CPOS &= ~(CPOS_CC0 | CPOS_CC1); + + /* fixed burst length (see erratum 11) */ + LCDC_DMACR = DMACR_BURST | DMACR_HM(8) | DMACR_TM(2); + + LCDC_RMCR = RMCR_LCDC_EN; + + if(fbi->backlight_power) + fbi->backlight_power(1); + if(fbi->lcd_power) + fbi->lcd_power(1); +} + +static void imxfb_disable_controller(struct imxfb_info *fbi) +{ + pr_debug("Disabling LCD controller\n"); + + if(fbi->backlight_power) + fbi->backlight_power(0); + if(fbi->lcd_power) + fbi->lcd_power(0); + + LCDC_RMCR = 0; +} + +static int imxfb_blank(int blank, struct fb_info *info) +{ + struct imxfb_info *fbi = info->par; + + pr_debug("imxfb_blank: blank=%d\n", blank); + + switch (blank) { + case FB_BLANK_POWERDOWN: + case FB_BLANK_VSYNC_SUSPEND: + case FB_BLANK_HSYNC_SUSPEND: + case FB_BLANK_NORMAL: + imxfb_disable_controller(fbi); + break; + + case FB_BLANK_UNBLANK: + imxfb_enable_controller(fbi); + break; + } + return 0; +} + +static struct fb_ops imxfb_ops = { + .owner = THIS_MODULE, + .fb_check_var = imxfb_check_var, + .fb_set_par = imxfb_set_par, + .fb_setcolreg = imxfb_setcolreg, + .fb_fillrect = cfb_fillrect, + .fb_copyarea = cfb_copyarea, + .fb_imageblit = cfb_imageblit, + .fb_blank = imxfb_blank, + .fb_cursor = soft_cursor, /* FIXME: i.MX can do hardware cursor */ +}; + +/* + * imxfb_activate_var(): + * Configures LCD Controller based on entries in var parameter. Settings are + * only written to the controller if changes were made. + */ +static int imxfb_activate_var(struct fb_var_screeninfo *var, struct fb_info *info) +{ + struct imxfb_info *fbi = info->par; + pr_debug("var: xres=%d hslen=%d lm=%d rm=%d\n", + var->xres, var->hsync_len, + var->left_margin, var->right_margin); + pr_debug("var: yres=%d vslen=%d um=%d bm=%d\n", + var->yres, var->vsync_len, + var->upper_margin, var->lower_margin); + +#if DEBUG_VAR + if (var->xres < 16 || var->xres > 1024) + printk(KERN_ERR "%s: invalid xres %d\n", + info->fix.id, var->xres); + if (var->hsync_len < 1 || var->hsync_len > 64) + printk(KERN_ERR "%s: invalid hsync_len %d\n", + info->fix.id, var->hsync_len); + if (var->left_margin > 255) + printk(KERN_ERR "%s: invalid left_margin %d\n", + info->fix.id, var->left_margin); + if (var->right_margin > 255) + printk(KERN_ERR "%s: invalid right_margin %d\n", + info->fix.id, var->right_margin); + if (var->yres < 1 || var->yres > 511) + printk(KERN_ERR "%s: invalid yres %d\n", + info->fix.id, var->yres); + if (var->vsync_len > 100) + printk(KERN_ERR "%s: invalid vsync_len %d\n", + info->fix.id, var->vsync_len); + if (var->upper_margin > 63) + printk(KERN_ERR "%s: invalid upper_margin %d\n", + info->fix.id, var->upper_margin); + if (var->lower_margin > 255) + printk(KERN_ERR "%s: invalid lower_margin %d\n", + info->fix.id, var->lower_margin); +#endif + + LCDC_HCR = HCR_H_WIDTH(var->hsync_len) | + HCR_H_WAIT_1(var->left_margin) | + HCR_H_WAIT_2(var->right_margin); + + LCDC_VCR = VCR_V_WIDTH(var->vsync_len) | + VCR_V_WAIT_1(var->upper_margin) | + VCR_V_WAIT_2(var->lower_margin); + + LCDC_SIZE = SIZE_XMAX(var->xres) | SIZE_YMAX(var->yres); + LCDC_PCR = fbi->pcr; + LCDC_PWMR = fbi->pwmr; + LCDC_LSCR1 = fbi->lscr1; + + return 0; +} + +static void imxfb_setup_gpio(struct imxfb_info *fbi) +{ + int width; + + LCDC_RMCR &= ~(RMCR_LCDC_EN | RMCR_SELF_REF); + + if( fbi->pcr & PCR_TFT ) + width = 16; + else + width = 1 << ((fbi->pcr >> 28) & 0x3); + + switch(width) { + case 16: + imx_gpio_mode(PD30_PF_LD15); + imx_gpio_mode(PD29_PF_LD14); + imx_gpio_mode(PD28_PF_LD13); + imx_gpio_mode(PD27_PF_LD12); + imx_gpio_mode(PD26_PF_LD11); + imx_gpio_mode(PD25_PF_LD10); + imx_gpio_mode(PD24_PF_LD9); + imx_gpio_mode(PD23_PF_LD8); + case 8: + imx_gpio_mode(PD22_PF_LD7); + imx_gpio_mode(PD21_PF_LD6); + imx_gpio_mode(PD20_PF_LD5); + imx_gpio_mode(PD19_PF_LD4); + case 4: + imx_gpio_mode(PD18_PF_LD3); + imx_gpio_mode(PD17_PF_LD2); + case 2: + imx_gpio_mode(PD16_PF_LD1); + case 1: + imx_gpio_mode(PD15_PF_LD0); + } + + /* initialize GPIOs */ + imx_gpio_mode(PD6_PF_LSCLK); + imx_gpio_mode(PD10_PF_SPL_SPR); + imx_gpio_mode(PD11_PF_CONTRAST); + imx_gpio_mode(PD14_PF_FLM_VSYNC); + imx_gpio_mode(PD13_PF_LP_HSYNC); + imx_gpio_mode(PD7_PF_REV); + imx_gpio_mode(PD8_PF_CLS); + +#ifndef CONFIG_MACH_PIMX1 + /* on PiMX1 used as buffers enable signal + */ + imx_gpio_mode(PD9_PF_PS); +#endif + +#ifndef CONFIG_MACH_MX1FS2 + /* on mx1fs2 this pin is used to (de)activate the display, so we need + * it as a normal gpio + */ + imx_gpio_mode(PD12_PF_ACD_OE); +#endif + +} + +#ifdef CONFIG_PM +/* + * Power management hooks. Note that we won't be called from IRQ context, + * unlike the blank functions above, so we may sleep. + */ +static int imxfb_suspend(struct device *dev, u32 state, u32 level) +{ + struct imxfb_info *fbi = dev_get_drvdata(dev); + pr_debug("%s\n",__FUNCTION__); + + if (level == SUSPEND_DISABLE || level == SUSPEND_POWER_DOWN) + imxfb_disable_controller(fbi); + return 0; +} + +static int imxfb_resume(struct device *dev, u32 level) +{ + struct imxfb_info *fbi = dev_get_drvdata(dev); + pr_debug("%s\n",__FUNCTION__); + + if (level == RESUME_ENABLE) + imxfb_enable_controller(fbi); + return 0; +} +#else +#define imxfb_suspend NULL +#define imxfb_resume NULL +#endif + +static int __init imxfb_init_fbinfo(struct device *dev) +{ + struct imxfb_mach_info *inf = dev->platform_data; + struct fb_info *info = dev_get_drvdata(dev); + struct imxfb_info *fbi = info->par; + + pr_debug("%s\n",__FUNCTION__); + + info->pseudo_palette = kmalloc( sizeof(u32) * 16, GFP_KERNEL); + if (!info->pseudo_palette) + return -ENOMEM; + + memset(fbi, 0, sizeof(struct imxfb_info)); + fbi->dev = dev; + + strlcpy(info->fix.id, IMX_NAME, sizeof(info->fix.id)); + + info->fix.type = FB_TYPE_PACKED_PIXELS; + info->fix.type_aux = 0; + info->fix.xpanstep = 0; + info->fix.ypanstep = 0; + info->fix.ywrapstep = 0; + info->fix.accel = FB_ACCEL_NONE; + + info->var.nonstd = 0; + info->var.activate = FB_ACTIVATE_NOW; + info->var.height = -1; + info->var.width = -1; + info->var.accel_flags = 0; + info->var.vmode = FB_VMODE_NONINTERLACED; + + info->fbops = &imxfb_ops; + info->flags = FBINFO_FLAG_DEFAULT; + info->pseudo_palette = (fbi + 1); + + fbi->rgb[RGB_16] = &def_rgb_16; + fbi->rgb[RGB_8] = &def_rgb_8; + + fbi->max_xres = inf->xres; + info->var.xres = inf->xres; + info->var.xres_virtual = inf->xres; + fbi->max_yres = inf->yres; + info->var.yres = inf->yres; + info->var.yres_virtual = inf->yres; + fbi->max_bpp = inf->bpp; + info->var.bits_per_pixel = inf->bpp; + info->var.pixclock = inf->pixclock; + info->var.hsync_len = inf->hsync_len; + info->var.left_margin = inf->left_margin; + info->var.right_margin = inf->right_margin; + info->var.vsync_len = inf->vsync_len; + info->var.upper_margin = inf->upper_margin; + info->var.lower_margin = inf->lower_margin; + info->var.sync = inf->sync; + info->var.grayscale = inf->cmap_greyscale; + fbi->cmap_inverse = inf->cmap_inverse; + fbi->pcr = inf->pcr; + fbi->lscr1 = inf->lscr1; + fbi->pwmr = inf->pwmr; + fbi->lcd_power = inf->lcd_power; + fbi->backlight_power = inf->backlight_power; + info->fix.smem_len = fbi->max_xres * fbi->max_yres * + fbi->max_bpp / 8; + + return 0; +} + +/* + * Allocates the DRAM memory for the frame buffer. This buffer is + * remapped into a non-cached, non-buffered, memory region to + * allow pixel writes to occur without flushing the cache. + * Once this area is remapped, all virtual memory access to the + * video memory should occur at the new region. + */ +static int __init imxfb_map_video_memory(struct fb_info *info) +{ + struct imxfb_info *fbi = info->par; + + fbi->map_size = PAGE_ALIGN(info->fix.smem_len); + fbi->map_cpu = dma_alloc_writecombine(fbi->dev, fbi->map_size, + &fbi->map_dma,GFP_KERNEL); + + if (fbi->map_cpu) { + info->screen_base = fbi->map_cpu; + fbi->screen_cpu = fbi->map_cpu; + fbi->screen_dma = fbi->map_dma; + info->fix.smem_start = fbi->screen_dma; + } + + return fbi->map_cpu ? 0 : -ENOMEM; +} + +static int __init imxfb_probe(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct imxfb_info *fbi; + struct fb_info *info; + struct imxfb_mach_info *inf; + struct resource *res; + int ret; + + printk("i.MX Framebuffer driver\n"); + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if(!res) + return -ENODEV; + + inf = dev->platform_data; + if(!inf) { + dev_err(dev,"No platform_data available\n"); + return -ENOMEM; + } + + info = framebuffer_alloc(sizeof(struct imxfb_info), dev); + if(!info) + return -ENOMEM; + + fbi = info->par; + + dev_set_drvdata(dev, info); + + ret = imxfb_init_fbinfo(dev); + if( ret < 0 ) + goto failed_init; + + res = request_mem_region(res->start, res->end - res->start + 1, "IMXFB"); + if (!res) { + ret = -EBUSY; + goto failed_regs; + } + + if (!inf->fixed_screen_cpu) { + ret = imxfb_map_video_memory(info); + if (ret) { + dev_err(dev, "Failed to allocate video RAM: %d\n", ret); + ret = -ENOMEM; + goto failed_map; + } + } else { + /* Fixed framebuffer mapping enables location of the screen in eSRAM */ + fbi->map_cpu = inf->fixed_screen_cpu; + fbi->map_dma = inf->fixed_screen_dma; + info->screen_base = fbi->map_cpu; + fbi->screen_cpu = fbi->map_cpu; + fbi->screen_dma = fbi->map_dma; + info->fix.smem_start = fbi->screen_dma; + } + + /* + * This makes sure that our colour bitfield + * descriptors are correctly initialised. + */ + imxfb_check_var(&info->var, info); + + ret = fb_alloc_cmap(&info->cmap, 1<<info->var.bits_per_pixel, 0); + if (ret < 0) + goto failed_cmap; + + imxfb_setup_gpio(fbi); + + imxfb_set_par(info); + ret = register_framebuffer(info); + if (ret < 0) { + dev_err(dev, "failed to register framebuffer\n"); + goto failed_register; + } + + imxfb_enable_controller(fbi); + + return 0; + +failed_register: + fb_dealloc_cmap(&info->cmap); +failed_cmap: + if (!inf->fixed_screen_cpu) + dma_free_writecombine(dev,fbi->map_size,fbi->map_cpu, + fbi->map_dma); +failed_map: + kfree(info->pseudo_palette); +failed_regs: + release_mem_region(res->start, res->end - res->start); +failed_init: + dev_set_drvdata(dev, NULL); + framebuffer_release(info); + return ret; +} + +static int imxfb_remove(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct fb_info *info = dev_get_drvdata(dev); + struct resource *res; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + + /* disable LCD controller */ + LCDC_RMCR &= ~RMCR_LCDC_EN; + + unregister_framebuffer(info); + + fb_dealloc_cmap(&info->cmap); + kfree(info->pseudo_palette); + framebuffer_release(info); + + release_mem_region(res->start, res->end - res->start + 1); + dev_set_drvdata(dev, NULL); + + return 0; +} + +void imxfb_shutdown(struct device * dev) +{ + /* disable LCD Controller */ + LCDC_RMCR &= ~RMCR_LCDC_EN; +} + +static struct device_driver imxfb_driver = { + .name = "imx-fb", + .bus = &platform_bus_type, + .probe = imxfb_probe, + .suspend = imxfb_suspend, + .resume = imxfb_resume, + .remove = imxfb_remove, + .shutdown = imxfb_shutdown, +}; + +int __init imxfb_init(void) +{ + return driver_register(&imxfb_driver); +} + +static void __exit imxfb_cleanup(void) +{ + driver_unregister(&imxfb_driver); +} + +module_init(imxfb_init); +module_exit(imxfb_cleanup); + +MODULE_DESCRIPTION("Motorola i.MX framebuffer driver"); +MODULE_AUTHOR("Sascha Hauer, Pengutronix"); +MODULE_LICENSE("GPL"); diff --git a/drivers/video/imxfb.h b/drivers/video/imxfb.h new file mode 100644 index 00000000000..128c3ee515c --- /dev/null +++ b/drivers/video/imxfb.h @@ -0,0 +1,72 @@ +/* + * linux/drivers/video/imxfb.h + * + * Freescale i.MX Frame Buffer device driver + * + * Copyright (C) 2004 S.Hauer, Pengutronix + * + * Copyright (C) 1999 Eric A. Thomas + * Based on acornfb.c Copyright (C) Russell King. + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file COPYING in the main directory of this archive + * for more details. + */ + +/* + * These are the bitfields for each + * display depth that we support. + */ +struct imxfb_rgb { + struct fb_bitfield red; + struct fb_bitfield green; + struct fb_bitfield blue; + struct fb_bitfield transp; +}; + +#define RGB_16 (0) +#define RGB_8 (1) +#define NR_RGB 2 + +struct imxfb_info { + struct device *dev; + struct imxfb_rgb *rgb[NR_RGB]; + + u_int max_bpp; + u_int max_xres; + u_int max_yres; + + /* + * These are the addresses we mapped + * the framebuffer memory region to. + */ + dma_addr_t map_dma; + u_char * map_cpu; + u_int map_size; + + u_char * screen_cpu; + dma_addr_t screen_dma; + u_int palette_size; + + dma_addr_t dbar1; + dma_addr_t dbar2; + + u_int pcr; + u_int pwmr; + u_int lscr1; + u_int cmap_inverse:1, + cmap_static:1, + unused:30; + + void (*lcd_power)(int); + void (*backlight_power)(int); +}; + +#define IMX_NAME "IMX" + +/* + * Minimum X and Y resolutions + */ +#define MIN_XRES 64 +#define MIN_YRES 64 + diff --git a/drivers/video/intelfb/intelfbdrv.c b/drivers/video/intelfb/intelfbdrv.c index 6a05b700083..549e2293926 100644 --- a/drivers/video/intelfb/intelfbdrv.c +++ b/drivers/video/intelfb/intelfbdrv.c @@ -135,9 +135,45 @@ #endif #include "intelfb.h" -#include "intelfbdrv.h" #include "intelfbhw.h" +static void __devinit get_initial_mode(struct intelfb_info *dinfo); +static void update_dinfo(struct intelfb_info *dinfo, + struct fb_var_screeninfo *var); +static int intelfb_get_fix(struct fb_fix_screeninfo *fix, + struct fb_info *info); + +static int intelfb_check_var(struct fb_var_screeninfo *var, + struct fb_info *info); +static int intelfb_set_par(struct fb_info *info); +static int intelfb_setcolreg(unsigned regno, unsigned red, unsigned green, + unsigned blue, unsigned transp, + struct fb_info *info); + +static int intelfb_blank(int blank, struct fb_info *info); +static int intelfb_pan_display(struct fb_var_screeninfo *var, + struct fb_info *info); + +static void intelfb_fillrect(struct fb_info *info, + const struct fb_fillrect *rect); +static void intelfb_copyarea(struct fb_info *info, + const struct fb_copyarea *region); +static void intelfb_imageblit(struct fb_info *info, + const struct fb_image *image); +static int intelfb_cursor(struct fb_info *info, + struct fb_cursor *cursor); + +static int intelfb_sync(struct fb_info *info); + +static int intelfb_ioctl(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long arg, + struct fb_info *info); + +static int __devinit intelfb_pci_register(struct pci_dev *pdev, + const struct pci_device_id *ent); +static void __devexit intelfb_pci_unregister(struct pci_dev *pdev); +static int __devinit intelfb_set_fbinfo(struct intelfb_info *dinfo); + /* * Limiting the class to PCI_CLASS_DISPLAY_VGA prevents function 1 of the * mobile chipsets from being registered. diff --git a/drivers/video/intelfb/intelfbdrv.h b/drivers/video/intelfb/intelfbdrv.h deleted file mode 100644 index cc305812888..00000000000 --- a/drivers/video/intelfb/intelfbdrv.h +++ /dev/null @@ -1,68 +0,0 @@ -#ifndef _INTELFBDRV_H -#define _INTELFBDRV_H - -/* - ****************************************************************************** - * intelfb - * - * Linux framebuffer driver for Intel(R) 830M/845G/852GM/855GM/865G/915G - * integrated graphics chips. - * - * Copyright © 2004 Sylvain Meyer - * - * Author: Sylvain Meyer - * - ****************************************************************************** - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. -*/ - -static void __devinit get_initial_mode(struct intelfb_info *dinfo); -static void update_dinfo(struct intelfb_info *dinfo, - struct fb_var_screeninfo *var); -static int intelfb_get_fix(struct fb_fix_screeninfo *fix, - struct fb_info *info); - -static int intelfb_check_var(struct fb_var_screeninfo *var, - struct fb_info *info); -static int intelfb_set_par(struct fb_info *info); -static int intelfb_setcolreg(unsigned regno, unsigned red, unsigned green, - unsigned blue, unsigned transp, - struct fb_info *info); - -static int intelfb_blank(int blank, struct fb_info *info); -static int intelfb_pan_display(struct fb_var_screeninfo *var, - struct fb_info *info); - -static void intelfb_fillrect(struct fb_info *info, - const struct fb_fillrect *rect); -static void intelfb_copyarea(struct fb_info *info, - const struct fb_copyarea *region); -static void intelfb_imageblit(struct fb_info *info, - const struct fb_image *image); -static int intelfb_cursor(struct fb_info *info, - struct fb_cursor *cursor); - -static int intelfb_sync(struct fb_info *info); - -static int intelfb_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg, - struct fb_info *info); - -static int __devinit intelfb_pci_register(struct pci_dev *pdev, - const struct pci_device_id *ent); -static void __devexit intelfb_pci_unregister(struct pci_dev *pdev); -static int __devinit intelfb_set_fbinfo(struct intelfb_info *dinfo); - -#endif diff --git a/drivers/video/logo/Kconfig b/drivers/video/logo/Kconfig index 849b47b210e..6ba10e3acef 100644 --- a/drivers/video/logo/Kconfig +++ b/drivers/video/logo/Kconfig @@ -45,7 +45,7 @@ config LOGO_SGI_CLUT224 config LOGO_SUN_CLUT224 bool "224-color Sun Linux logo" - depends on LOGO && (SPARC || SPARC64) + depends on LOGO && (SPARC32 || SPARC64) default y config LOGO_SUPERH_MONO diff --git a/drivers/video/nvidia/nvidia.c b/drivers/video/nvidia/nvidia.c index 3a6555a8aaa..47733f58153 100644 --- a/drivers/video/nvidia/nvidia.c +++ b/drivers/video/nvidia/nvidia.c @@ -408,6 +408,7 @@ static int hwcur __devinitdata = 0; static int noaccel __devinitdata = 0; static int noscale __devinitdata = 0; static int paneltweak __devinitdata = 0; +static int vram __devinitdata = 0; #ifdef CONFIG_MTRR static int nomtrr __devinitdata = 0; #endif @@ -1180,7 +1181,7 @@ static int nvidiafb_check_var(struct fb_var_screeninfo *var, var->xres_virtual = (var->xres_virtual + 63) & ~63; - vramlen = info->fix.smem_len; + vramlen = info->screen_size; pitch = ((var->xres_virtual * var->bits_per_pixel) + 7) / 8; memlen = pitch * var->yres_virtual; @@ -1343,7 +1344,7 @@ static int __devinit nvidia_set_fbinfo(struct fb_info *info) /* maximize virtual vertical length */ lpitch = info->var.xres_virtual * ((info->var.bits_per_pixel + 7) >> 3); - info->var.yres_virtual = info->fix.smem_len / lpitch; + info->var.yres_virtual = info->screen_size / lpitch; info->pixmap.scan_align = 4; info->pixmap.buf_align = 4; @@ -1507,12 +1508,20 @@ static int __devinit nvidiafb_probe(struct pci_dev *pd, par->FbAddress = nvidiafb_fix.smem_start; par->FbMapSize = par->RamAmountKBytes * 1024; + if (vram && vram * 1024 * 1024 < par->FbMapSize) + par->FbMapSize = vram * 1024 * 1024; + + /* Limit amount of vram to 64 MB */ + if (par->FbMapSize > 64 * 1024 * 1024) + par->FbMapSize = 64 * 1024 * 1024; + par->FbUsableSize = par->FbMapSize - (128 * 1024); par->ScratchBufferSize = (par->Architecture < NV_ARCH_10) ? 8 * 1024 : 16 * 1024; par->ScratchBufferStart = par->FbUsableSize - par->ScratchBufferSize; info->screen_base = ioremap(nvidiafb_fix.smem_start, par->FbMapSize); - nvidiafb_fix.smem_len = par->FbUsableSize; + info->screen_size = par->FbUsableSize; + nvidiafb_fix.smem_len = par->RamAmountKBytes * 1024; if (!info->screen_base) { printk(KERN_ERR PFX "cannot ioremap FB base\n"); @@ -1524,7 +1533,8 @@ static int __devinit nvidiafb_probe(struct pci_dev *pd, #ifdef CONFIG_MTRR if (!nomtrr) { par->mtrr.vram = mtrr_add(nvidiafb_fix.smem_start, - par->FbMapSize, MTRR_TYPE_WRCOMB, 1); + par->RamAmountKBytes * 1024, + MTRR_TYPE_WRCOMB, 1); if (par->mtrr.vram < 0) { printk(KERN_ERR PFX "unable to setup MTRR\n"); } else { @@ -1566,9 +1576,9 @@ static int __devinit nvidiafb_probe(struct pci_dev *pd, err_out_iounmap_fb: iounmap(info->screen_base); + err_out_free_base1: fb_destroy_modedb(info->monspecs.modedb); nvidia_delete_i2c_busses(par); - err_out_free_base1: iounmap(par->REGS); err_out_free_base0: pci_release_regions(pd); @@ -1645,6 +1655,8 @@ static int __devinit nvidiafb_setup(char *options) noscale = 1; } else if (!strncmp(this_opt, "paneltweak:", 11)) { paneltweak = simple_strtoul(this_opt+11, NULL, 0); + } else if (!strncmp(this_opt, "vram:", 5)) { + vram = simple_strtoul(this_opt+5, NULL, 0); #ifdef CONFIG_MTRR } else if (!strncmp(this_opt, "nomtrr", 6)) { nomtrr = 1; @@ -1716,6 +1728,10 @@ module_param(forceCRTC, int, 0); MODULE_PARM_DESC(forceCRTC, "Forces usage of a particular CRTC in case autodetection " "fails. (0 or 1) (default=autodetect)"); +module_param(vram, int, 0); +MODULE_PARM_DESC(vram, + "amount of framebuffer memory to remap in MiB" + "(default=0 - remap entire memory)"); #ifdef CONFIG_MTRR module_param(nomtrr, bool, 0); MODULE_PARM_DESC(nomtrr, "Disables MTRR support (0 or 1=disabled) " diff --git a/drivers/video/radeonfb.c b/drivers/video/radeonfb.c index d9a084e77a6..c46387024b1 100644 --- a/drivers/video/radeonfb.c +++ b/drivers/video/radeonfb.c @@ -2107,7 +2107,7 @@ static void radeon_write_mode (struct radeonfb_info *rinfo, if (rinfo->arch == RADEON_M6) { - for (i=0; i<8; i++) + for (i=0; i<7; i++) OUTREG(common_regs_m6[i].reg, common_regs_m6[i].val); } else { for (i=0; i<9; i++) diff --git a/drivers/video/savage/savagefb_driver.c b/drivers/video/savage/savagefb_driver.c index 5bb8d60f35c..03d74e8ee06 100644 --- a/drivers/video/savage/savagefb_driver.c +++ b/drivers/video/savage/savagefb_driver.c @@ -1498,7 +1498,7 @@ static int __devinit savage_map_mmio (struct fb_info *info) info->fix.mmio_start = par->mmio.pbase; info->fix.mmio_len = par->mmio.len; - par->bci_base = (u32*)(par->mmio.vbase + BCI_BUFFER_OFFSET); + par->bci_base = (u32 __iomem *)(par->mmio.vbase + BCI_BUFFER_OFFSET); par->bci_ptr = 0; savage_enable_mmio (par); @@ -1514,7 +1514,7 @@ static void __devinit savage_unmap_mmio (struct fb_info *info) savage_disable_mmio(par); if (par->mmio.vbase) { - iounmap ((void *)par->mmio.vbase); + iounmap(par->mmio.vbase); par->mmio.vbase = NULL; } } @@ -1553,7 +1553,7 @@ static int __devinit savage_map_video (struct fb_info *info, #endif /* Clear framebuffer, it's all white in memory after boot */ - memset (par->video.vbase, 0, par->video.len); + memset_io (par->video.vbase, 0, par->video.len); return 0; } diff --git a/drivers/video/sis/init.c b/drivers/video/sis/init.c index 1994054d45f..ecfd72178db 100644 --- a/drivers/video/sis/init.c +++ b/drivers/video/sis/init.c @@ -1384,7 +1384,7 @@ SiSInitPCIetc(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo) /* HELPER: SetLVDSetc */ /*********************************************/ -void +static void SiSSetLVDSetc(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo) { USHORT temp; @@ -1625,7 +1625,7 @@ SiS_ResetSegmentRegisters(SiS_Private *SiS_Pr,PSIS_HW_INFO HwInfo) /* HELPER: GetVBType */ /*********************************************/ -void +static void SiS_GetVBType(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo) { USHORT flag=0, rev=0, nolcd=0, p4_0f, p4_25, p4_27; diff --git a/drivers/video/sis/init.h b/drivers/video/sis/init.h index 35030d30043..7e36b7ac147 100644 --- a/drivers/video/sis/init.h +++ b/drivers/video/sis/init.h @@ -2394,11 +2394,9 @@ void SiS_SetRegOR(SISIOADDRESS Port,USHORT Index, USHORT DataOR); void SiS_DisplayOn(SiS_Private *SiS_Pr); void SiS_DisplayOff(SiS_Private *SiS_Pr); void SiSRegInit(SiS_Private *SiS_Pr, SISIOADDRESS BaseAddr); -void SiSSetLVDSetc(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); BOOLEAN SiSDetermineROMLayout661(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); void SiS_SetEnableDstn(SiS_Private *SiS_Pr, int enable); void SiS_SetEnableFstn(SiS_Private *SiS_Pr, int enable); -void SiS_GetVBType(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); BOOLEAN SiS_SearchModeID(SiS_Private *SiS_Pr, USHORT *ModeNo, USHORT *ModeIdIndex); UCHAR SiS_GetModePtr(SiS_Private *SiS_Pr, USHORT ModeNo, USHORT ModeIdIndex); USHORT SiS_GetColorDepth(SiS_Private *SiS_Pr, USHORT ModeNo, USHORT ModeIdIndex); @@ -2444,7 +2442,6 @@ extern void SiS_GetLCDResInfo(SiS_Private *SiS_Pr, USHORT ModeNo, USHORT Mod extern void SiS_SetYPbPr(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); extern void SiS_SetTVMode(SiS_Private *SiS_Pr, USHORT ModeNo, USHORT ModeIdIndex, PSIS_HW_INFO HwInfo); extern void SiS_UnLockCRT2(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); -extern void SiS_LockCRT2(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); extern void SiS_DisableBridge(SiS_Private *, PSIS_HW_INFO); extern BOOLEAN SiS_SetCRT2Group(SiS_Private *, PSIS_HW_INFO, USHORT); extern USHORT SiS_GetRatePtr(SiS_Private *SiS_Pr, USHORT ModeNo, USHORT ModeIdIndex, diff --git a/drivers/video/sis/init301.c b/drivers/video/sis/init301.c index 2bc5b809791..274dacd54bb 100644 --- a/drivers/video/sis/init301.c +++ b/drivers/video/sis/init301.c @@ -86,6 +86,7 @@ #define SiS_I2CDELAYSHORT 150 static USHORT SiS_GetBIOSLCDResInfo(SiS_Private *SiS_Pr); +static void SiS_SetCH70xx(SiS_Private *SiS_Pr, USHORT tempbx); /*********************************************/ /* HELPER: Lock/Unlock CRT2 */ @@ -100,7 +101,7 @@ SiS_UnLockCRT2(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo) SiS_SetRegOR(SiS_Pr->SiS_Part1Port,0x24,0x01); } -void +static void SiS_LockCRT2(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo) { if(HwInfo->jChipType >= SIS_315H) @@ -4236,7 +4237,7 @@ SiS_DisableBridge(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo) * from outside the context of a mode switch! * MUST call getVBType before calling this */ -void +static void SiS_EnableBridge(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo) { USHORT temp=0,tempah; @@ -9219,7 +9220,7 @@ SiS_SetCH701x(SiS_Private *SiS_Pr, USHORT tempbx) SiS_SetChReg(SiS_Pr, tempbx, 0); } -void +static void SiS_SetCH70xx(SiS_Private *SiS_Pr, USHORT tempbx) { if(SiS_Pr->SiS_IF_DEF_CH70xx == 1) @@ -9323,7 +9324,7 @@ SiS_GetCH701x(SiS_Private *SiS_Pr, USHORT tempbx) /* Read from Chrontel 70xx */ /* Parameter is [Register no (S7-S0)] */ -USHORT +static USHORT SiS_GetCH70xx(SiS_Private *SiS_Pr, USHORT tempbx) { if(SiS_Pr->SiS_IF_DEF_CH70xx == 1) diff --git a/drivers/video/sis/init301.h b/drivers/video/sis/init301.h index f05aebc994b..f84eb54164a 100644 --- a/drivers/video/sis/init301.h +++ b/drivers/video/sis/init301.h @@ -293,7 +293,6 @@ static UCHAR SiS300_TrumpionData[7][80] = { #endif void SiS_UnLockCRT2(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); -void SiS_LockCRT2(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); void SiS_EnableCRT2(SiS_Private *SiS_Pr); USHORT SiS_GetRatePtr(SiS_Private *SiS_Pr, USHORT ModeNo, USHORT ModeIdIndex, PSIS_HW_INFO HwInfo); void SiS_WaitRetrace1(SiS_Private *SiS_Pr); @@ -310,7 +309,6 @@ USHORT SiS_GetVCLK2Ptr(SiS_Private *SiS_Pr, USHORT ModeNo, USHORT ModeIdIndex, USHORT RefreshRateTableIndex, PSIS_HW_INFO HwInfo); USHORT SiS_GetResInfo(SiS_Private *SiS_Pr,USHORT ModeNo,USHORT ModeIdIndex); void SiS_DisableBridge(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); -void SiS_EnableBridge(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); BOOLEAN SiS_SetCRT2Group(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo, USHORT ModeNo); void SiS_SiS30xBLOn(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); void SiS_SiS30xBLOff(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); @@ -319,8 +317,6 @@ void SiS_SetCH700x(SiS_Private *SiS_Pr, USHORT tempax); USHORT SiS_GetCH700x(SiS_Private *SiS_Pr, USHORT tempax); void SiS_SetCH701x(SiS_Private *SiS_Pr, USHORT tempax); USHORT SiS_GetCH701x(SiS_Private *SiS_Pr, USHORT tempax); -void SiS_SetCH70xx(SiS_Private *SiS_Pr, USHORT tempax); -USHORT SiS_GetCH70xx(SiS_Private *SiS_Pr, USHORT tempax); void SiS_SetCH70xxANDOR(SiS_Private *SiS_Pr, USHORT tempax,USHORT tempbh); #ifdef SIS315H static void SiS_Chrontel701xOn(SiS_Private *SiS_Pr, PSIS_HW_INFO HwInfo); diff --git a/drivers/video/sis/sis_main.c b/drivers/video/sis/sis_main.c index b773c98f651..69826603681 100644 --- a/drivers/video/sis/sis_main.c +++ b/drivers/video/sis/sis_main.c @@ -4762,7 +4762,8 @@ static void __devinit sisfb_post_sis315330(struct pci_dev *pdev) #endif -int __devinit sisfb_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +static int __devinit sisfb_probe(struct pci_dev *pdev, + const struct pci_device_id *ent) { struct sisfb_chip_info *chipinfo = &sisfb_chip_info[ent->driver_data]; struct sis_video_info *ivideo = NULL; @@ -5940,7 +5941,7 @@ MODULE_PARM_DESC(videoram, #endif #endif -int __devinit sisfb_init_module(void) +static int __devinit sisfb_init_module(void) { sisfb_setdefaultparms(); diff --git a/drivers/video/tcx.c b/drivers/video/tcx.c index e2fa9e1ddc3..1986a8b3833 100644 --- a/drivers/video/tcx.c +++ b/drivers/video/tcx.c @@ -36,6 +36,7 @@ static int tcx_blank(int, struct fb_info *); static int tcx_mmap(struct fb_info *, struct file *, struct vm_area_struct *); static int tcx_ioctl(struct inode *, struct file *, unsigned int, unsigned long, struct fb_info *); +static int tcx_pan_display(struct fb_var_screeninfo *, struct fb_info *); /* * Frame buffer operations @@ -45,6 +46,7 @@ static struct fb_ops tcx_ops = { .owner = THIS_MODULE, .fb_setcolreg = tcx_setcolreg, .fb_blank = tcx_blank, + .fb_pan_display = tcx_pan_display, .fb_fillrect = cfb_fillrect, .fb_copyarea = cfb_copyarea, .fb_imageblit = cfb_imageblit, @@ -153,6 +155,12 @@ static void tcx_reset (struct fb_info *info) spin_unlock_irqrestore(&par->lock, flags); } +static int tcx_pan_display(struct fb_var_screeninfo *var, struct fb_info *info) +{ + tcx_reset(info); + return 0; +} + /** * tcx_setcolreg - Optional function. Sets a color register. * @regno: boolean, 0 copy local, 1 get_user() function @@ -366,6 +374,9 @@ static void tcx_init_one(struct sbus_dev *sdev) all->par.lowdepth = prom_getbool(sdev->prom_node, "tcx-8-bit"); sbusfb_fill_var(&all->info.var, sdev->prom_node, 8); + all->info.var.red.length = 8; + all->info.var.green.length = 8; + all->info.var.blue.length = 8; linebytes = prom_getintdefault(sdev->prom_node, "linebytes", all->info.var.xres); @@ -439,6 +450,7 @@ static void tcx_init_one(struct sbus_dev *sdev) return; } + fb_set_cmap(&all->info.cmap, &all->info); tcx_init_fix(&all->info, linebytes); if (register_framebuffer(&all->info) < 0) { @@ -466,7 +478,7 @@ int __init tcx_init(void) return -ENODEV; for_all_sbusdev(sdev, sbus) { - if (!strcmp(sdev->prom_name, "tcx")) + if (!strcmp(sdev->prom_name, "SUNW,tcx")) tcx_init_one(sdev); } diff --git a/drivers/video/tdfxfb.c b/drivers/video/tdfxfb.c index c34ba39b6f7..7044226c5d4 100644 --- a/drivers/video/tdfxfb.c +++ b/drivers/video/tdfxfb.c @@ -317,30 +317,49 @@ static inline void do_setpalentry(struct tdfx_par *par, unsigned regno, u32 c) static u32 do_calc_pll(int freq, int* freq_out) { - int m, n, k, best_m, best_n, best_k, f_cur, best_error; + int m, n, k, best_m, best_n, best_k, best_error; int fref = 14318; - /* this really could be done with more intelligence -- - 255*63*4 = 64260 iterations is silly */ best_error = freq; best_n = best_m = best_k = 0; - for (n = 1; n < 256; n++) { - for (m = 1; m < 64; m++) { - for (k = 0; k < 4; k++) { - f_cur = fref*(n + 2)/(m + 2)/(1 << k); - if (abs(f_cur - freq) < best_error) { - best_error = abs(f_cur-freq); - best_n = n; - best_m = m; - best_k = k; + + for (k = 3; k >= 0; k--) { + for (m = 63; m >= 0; m--) { + /* + * Estimate value of n that produces target frequency + * with current m and k + */ + int n_estimated = (freq * (m + 2) * (1 << k) / fref) - 2; + + /* Search neighborhood of estimated n */ + for (n = max(0, n_estimated - 1); + n <= min(255, n_estimated + 1); n++) { + /* + * Calculate PLL freqency with current m, k and + * estimated n + */ + int f = fref * (n + 2) / (m + 2) / (1 << k); + int error = abs (f - freq); + + /* + * If this is the closest we've come to the + * target frequency then remember n, m and k + */ + if (error < best_error) { + best_error = error; + best_n = n; + best_m = m; + best_k = k; } } } } + n = best_n; m = best_m; k = best_k; *freq_out = fref*(n + 2)/(m + 2)/(1 << k); + return (n << 8) | (m << 2) | k; } @@ -411,36 +430,35 @@ static void do_write_regs(struct fb_info *info, struct banshee_reg* reg) static unsigned long do_lfb_size(struct tdfx_par *par, unsigned short dev_id) { - u32 draminit0 = 0; - u32 draminit1 = 0; - u32 miscinit1 = 0; - u32 lfbsize = 0; - int sgram_p = 0; + u32 draminit0; + u32 draminit1; + u32 miscinit1; + + int num_chips; + int chip_size; /* in MB */ + u32 lfbsize; + int has_sgram; draminit0 = tdfx_inl(par, DRAMINIT0); draminit1 = tdfx_inl(par, DRAMINIT1); + + num_chips = (draminit0 & DRAMINIT0_SGRAM_NUM) ? 8 : 4; - if ((dev_id == PCI_DEVICE_ID_3DFX_BANSHEE) || - (dev_id == PCI_DEVICE_ID_3DFX_VOODOO3)) { - sgram_p = (draminit1 & DRAMINIT1_MEM_SDRAM) ? 0 : 1; - - lfbsize = sgram_p ? - (((draminit0 & DRAMINIT0_SGRAM_NUM) ? 2 : 1) * - ((draminit0 & DRAMINIT0_SGRAM_TYPE) ? 8 : 4) * 1024 * 1024) : - 16 * 1024 * 1024; + if (dev_id < PCI_DEVICE_ID_3DFX_VOODOO5) { + /* Banshee/Voodoo3 */ + has_sgram = draminit1 & DRAMINIT1_MEM_SDRAM; + chip_size = has_sgram ? ((draminit0 & DRAMINIT0_SGRAM_TYPE) ? 2 : 1) + : 2; } else { /* Voodoo4/5 */ - u32 chips, psize, banks; - - chips = ((draminit0 & (1 << 26)) == 0) ? 4 : 8; - psize = 1 << ((draminit0 & 0x38000000) >> 28); - banks = ((draminit0 & (1 << 30)) == 0) ? 2 : 4; - lfbsize = chips * psize * banks; - lfbsize <<= 20; - } - /* disable block writes for SDRAM (why?) */ + has_sgram = 0; + chip_size = 1 << ((draminit0 & DRAMINIT0_SGRAM_TYPE_MASK) >> DRAMINIT0_SGRAM_TYPE_SHIFT); + } + lfbsize = num_chips * chip_size * 1024 * 1024; + + /* disable block writes for SDRAM */ miscinit1 = tdfx_inl(par, MISCINIT1); - miscinit1 |= sgram_p ? 0 : MISCINIT1_2DBLOCK_DIS; + miscinit1 |= has_sgram ? 0 : MISCINIT1_2DBLOCK_DIS; miscinit1 |= MISCINIT1_CLUT_INV; banshee_make_room(par, 1); diff --git a/drivers/video/tgafb.c b/drivers/video/tgafb.c index 3099630d0c3..9d9d2009ad8 100644 --- a/drivers/video/tgafb.c +++ b/drivers/video/tgafb.c @@ -45,9 +45,7 @@ static void tgafb_fillrect(struct fb_info *, const struct fb_fillrect *); static void tgafb_copyarea(struct fb_info *, const struct fb_copyarea *); static int tgafb_pci_register(struct pci_dev *, const struct pci_device_id *); -#ifdef MODULE static void tgafb_pci_unregister(struct pci_dev *); -#endif static const char *mode_option = "640x480@60"; @@ -1484,7 +1482,6 @@ tgafb_pci_register(struct pci_dev *pdev, const struct pci_device_id *ent) return ret; } -#ifdef MODULE static void __exit tgafb_pci_unregister(struct pci_dev *pdev) { @@ -1500,6 +1497,7 @@ tgafb_pci_unregister(struct pci_dev *pdev) kfree(info); } +#ifdef MODULE static void __exit tgafb_exit(void) { diff --git a/drivers/video/vesafb.c b/drivers/video/vesafb.c index 8fc1278d7fb..3027841f9c2 100644 --- a/drivers/video/vesafb.c +++ b/drivers/video/vesafb.c @@ -19,9 +19,6 @@ #include <linux/fb.h> #include <linux/ioport.h> #include <linux/init.h> -#ifdef __i386__ -#include <video/edid.h> -#endif #include <asm/io.h> #include <asm/mtrr.h> diff --git a/fs/afs/file.c b/fs/afs/file.c index 6b6bb7c8abf..23c12512802 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -131,8 +131,7 @@ static int afs_file_readpage(struct file *file, struct page *page) vnode = AFS_FS_I(inode); - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); ret = -ESTALE; if (vnode->flags & AFS_VNODE_DELETED) @@ -40,9 +40,6 @@ #define dprintk(x...) do { ; } while (0) #endif -long aio_run = 0; /* for testing only */ -long aio_wakeups = 0; /* for testing only */ - /*------ sysctl variables----*/ atomic_t aio_nr = ATOMIC_INIT(0); /* current system wide number of aio requests */ unsigned aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ @@ -58,7 +55,7 @@ static void aio_fput_routine(void *); static DECLARE_WORK(fput_work, aio_fput_routine, NULL); static DEFINE_SPINLOCK(fput_lock); -LIST_HEAD(fput_head); +static LIST_HEAD(fput_head); static void aio_kick_handler(void *); @@ -290,7 +287,7 @@ static void aio_cancel_all(struct kioctx *ctx) spin_unlock_irq(&ctx->ctx_lock); } -void wait_for_all_aios(struct kioctx *ctx) +static void wait_for_all_aios(struct kioctx *ctx) { struct task_struct *tsk = current; DECLARE_WAITQUEUE(wait, tsk); @@ -405,7 +402,6 @@ static struct kiocb fastcall *__aio_get_req(struct kioctx *ctx) req->ki_ctx = ctx; req->ki_cancel = NULL; req->ki_retry = NULL; - req->ki_obj.user = NULL; req->ki_dtor = NULL; req->private = NULL; INIT_LIST_HEAD(&req->ki_run_list); @@ -451,11 +447,6 @@ static inline void really_put_req(struct kioctx *ctx, struct kiocb *req) { if (req->ki_dtor) req->ki_dtor(req); - req->ki_ctx = NULL; - req->ki_filp = NULL; - req->ki_obj.user = NULL; - req->ki_dtor = NULL; - req->private = NULL; kmem_cache_free(kiocb_cachep, req); ctx->reqs_active--; @@ -592,7 +583,7 @@ static void use_mm(struct mm_struct *mm) * Comments: Called with ctx->ctx_lock held. This nests * task_lock instead ctx_lock. */ -void unuse_mm(struct mm_struct *mm) +static void unuse_mm(struct mm_struct *mm) { struct task_struct *tsk = current; @@ -623,7 +614,6 @@ static inline int __queue_kicked_iocb(struct kiocb *iocb) if (list_empty(&iocb->ki_run_list)) { list_add_tail(&iocb->ki_run_list, &ctx->run_list); - iocb->ki_queued++; return 1; } return 0; @@ -664,10 +654,8 @@ static ssize_t aio_run_iocb(struct kiocb *iocb) } if (!(iocb->ki_retried & 0xff)) { - pr_debug("%ld retry: %d of %d (kick %ld, Q %ld run %ld, wake %ld)\n", - iocb->ki_retried, - iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes, - iocb->ki_kicked, iocb->ki_queued, aio_run, aio_wakeups); + pr_debug("%ld retry: %d of %d\n", iocb->ki_retried, + iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes); } if (!(retry = iocb->ki_retry)) { @@ -774,7 +762,6 @@ out: static int __aio_run_iocbs(struct kioctx *ctx) { struct kiocb *iocb; - int count = 0; LIST_HEAD(run_list); list_splice_init(&ctx->run_list, &run_list); @@ -789,9 +776,7 @@ static int __aio_run_iocbs(struct kioctx *ctx) aio_run_iocb(iocb); if (__aio_put_req(ctx, iocb)) /* drop extra ref */ put_ioctx(ctx); - count++; } - aio_run++; if (!list_empty(&ctx->run_list)) return 1; return 0; @@ -879,7 +864,7 @@ static void aio_kick_handler(void *data) * and if required activate the aio work queue to process * it */ -void queue_kicked_iocb(struct kiocb *iocb) +static void queue_kicked_iocb(struct kiocb *iocb) { struct kioctx *ctx = iocb->ki_ctx; unsigned long flags; @@ -890,10 +875,8 @@ void queue_kicked_iocb(struct kiocb *iocb) spin_lock_irqsave(&ctx->ctx_lock, flags); run = __queue_kicked_iocb(iocb); spin_unlock_irqrestore(&ctx->ctx_lock, flags); - if (run) { + if (run) aio_queue_work(ctx); - aio_wakeups++; - } } /* @@ -913,7 +896,6 @@ void fastcall kick_iocb(struct kiocb *iocb) return; } - iocb->ki_kicked++; /* If its already kicked we shouldn't queue it again */ if (!kiocbTryKick(iocb)) { queue_kicked_iocb(iocb); @@ -984,7 +966,8 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) tail = info->tail; event = aio_ring_event(info, tail, KM_IRQ0); - tail = (tail + 1) % info->nr; + if (++tail >= info->nr) + tail = 0; event->obj = (u64)(unsigned long)iocb->ki_obj.user; event->data = iocb->ki_user_data; @@ -1008,10 +991,8 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) pr_debug("added to ring %p at [%lu]\n", iocb, tail); - pr_debug("%ld retries: %d of %d (kicked %ld, Q %ld run %ld wake %ld)\n", - iocb->ki_retried, - iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes, - iocb->ki_kicked, iocb->ki_queued, aio_run, aio_wakeups); + pr_debug("%ld retries: %d of %d\n", iocb->ki_retried, + iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes); put_rq: /* everything turned out well, dispose of the aiocb. */ ret = __aio_put_req(ctx, iocb); @@ -1119,7 +1100,6 @@ static int read_events(struct kioctx *ctx, int i = 0; struct io_event ent; struct aio_timeout to; - int event_loop = 0; /* testing only */ int retry = 0; /* needed to zero any padding within an entry (there shouldn't be @@ -1186,7 +1166,6 @@ retry: if (to.timed_out) /* Only check after read evt */ break; schedule(); - event_loop++; if (signal_pending(tsk)) { ret = -EINTR; break; @@ -1214,9 +1193,6 @@ retry: if (timeout) clear_timeout(&to); out: - pr_debug("event loop executed %d times\n", event_loop); - pr_debug("aio_run %ld\n", aio_run); - pr_debug("aio_wakeups %ld\n", aio_wakeups); return i ? i : ret; } @@ -1401,7 +1377,7 @@ static ssize_t aio_fsync(struct kiocb *iocb) * Performs the initial checks and aio retry method * setup for the kiocb at the time of io submission. */ -ssize_t aio_setup_iocb(struct kiocb *kiocb) +static ssize_t aio_setup_iocb(struct kiocb *kiocb) { struct file *file = kiocb->ki_filp; ssize_t ret = 0; @@ -1470,7 +1446,8 @@ ssize_t aio_setup_iocb(struct kiocb *kiocb) * because this callback isn't used for wait queues which * are nested inside ioctx lock (i.e. ctx->wait) */ -int aio_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) +static int aio_wake_function(wait_queue_t *wait, unsigned mode, + int sync, void *key) { struct kiocb *iocb = container_of(wait, struct kiocb, ki_wait); @@ -1514,8 +1491,7 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, } req->ki_filp = file; - iocb->aio_key = req->ki_key; - ret = put_user(iocb->aio_key, &user_iocb->aio_key); + ret = put_user(req->ki_key, &user_iocb->aio_key); if (unlikely(ret)) { dprintk("EFAULT: aio_key\n"); goto out_put_req; @@ -1530,13 +1506,7 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, req->ki_opcode = iocb->aio_lio_opcode; init_waitqueue_func_entry(&req->ki_wait, aio_wake_function); INIT_LIST_HEAD(&req->ki_wait.task_list); - req->ki_run_list.next = req->ki_run_list.prev = NULL; - req->ki_retry = NULL; req->ki_retried = 0; - req->ki_kicked = 0; - req->ki_queued = 0; - aio_run = 0; - aio_wakeups = 0; ret = aio_setup_iocb(req); @@ -1544,10 +1514,14 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, goto out_put_req; spin_lock_irq(&ctx->ctx_lock); - list_add_tail(&req->ki_run_list, &ctx->run_list); - /* drain the run list */ - while (__aio_run_iocbs(ctx)) - ; + if (likely(list_empty(&ctx->run_list))) { + aio_run_iocb(req); + } else { + list_add_tail(&req->ki_run_list, &ctx->run_list); + /* drain the run list */ + while (__aio_run_iocbs(ctx)) + ; + } spin_unlock_irq(&ctx->ctx_lock); aio_put_req(req); /* drop extra ref to req */ return 0; @@ -1620,7 +1594,8 @@ asmlinkage long sys_io_submit(aio_context_t ctx_id, long nr, * Finds a given iocb for cancellation. * MUST be called with ctx->ctx_lock held. */ -struct kiocb *lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb, u32 key) +static struct kiocb *lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb, + u32 key) { struct list_head *pos; /* TODO: use a hash or array, this sucks. */ diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index f5a52c87172..c7b2b889018 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -84,6 +84,7 @@ struct autofs_wait_queue { char *name; /* This is for status reporting upon return */ int status; + atomic_t notified; atomic_t wait_ctr; }; @@ -101,6 +102,7 @@ struct autofs_sb_info { int needs_reghost; struct super_block *sb; struct semaphore wq_sem; + spinlock_t fs_lock; struct autofs_wait_queue *queues; /* Wait queue pointer */ }; @@ -126,9 +128,18 @@ static inline int autofs4_oz_mode(struct autofs_sb_info *sbi) { static inline int autofs4_ispending(struct dentry *dentry) { struct autofs_info *inf = autofs4_dentry_ino(dentry); + int pending = 0; - return (dentry->d_flags & DCACHE_AUTOFS_PENDING) || - (inf != NULL && inf->flags & AUTOFS_INF_EXPIRING); + if (dentry->d_flags & DCACHE_AUTOFS_PENDING) + return 1; + + if (inf) { + spin_lock(&inf->sbi->fs_lock); + pending = inf->flags & AUTOFS_INF_EXPIRING; + spin_unlock(&inf->sbi->fs_lock); + } + + return pending; } static inline void autofs4_copy_atime(struct file *src, struct file *dst) diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index 31540a6404d..500425e24fb 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -99,6 +99,10 @@ static int autofs4_check_tree(struct vfsmount *mnt, if (!autofs4_can_expire(top, timeout, do_now)) return 0; + /* Is someone visiting anywhere in the tree ? */ + if (may_umount_tree(mnt)) + return 0; + spin_lock(&dcache_lock); repeat: next = this_parent->d_subdirs.next; @@ -270,10 +274,18 @@ static struct dentry *autofs4_expire(struct super_block *sb, /* Case 2: tree mount, expire iff entire tree is not busy */ if (!exp_leaves) { + /* Lock the tree as we must expire as a whole */ + spin_lock(&sbi->fs_lock); if (autofs4_check_tree(mnt, dentry, timeout, do_now)) { - expired = dentry; - break; + struct autofs_info *inf = autofs4_dentry_ino(dentry); + + /* Set this flag early to catch sys_chdir and the like */ + inf->flags |= AUTOFS_INF_EXPIRING; + spin_unlock(&sbi->fs_lock); + expired = dentry; + break; } + spin_unlock(&sbi->fs_lock); /* Case 3: direct mount, expire individual leaves */ } else { expired = autofs4_check_leaves(mnt, dentry, timeout, do_now); diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index a5256074662..4bb14cc6804 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -206,6 +206,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) sbi->version = 0; sbi->sub_version = 0; init_MUTEX(&sbi->wq_sem); + spin_lock_init(&sbi->fs_lock); sbi->queues = NULL; s->s_blocksize = 1024; s->s_blocksize_bits = 10; diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 1ab24a662e0..5a40d36e5a5 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -210,17 +210,8 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, wq->len = len; wq->status = -EINTR; /* Status return if interrupted */ atomic_set(&wq->wait_ctr, 2); + atomic_set(&wq->notified, 1); up(&sbi->wq_sem); - - DPRINTK("new wait id = 0x%08lx, name = %.*s, nfy=%d", - (unsigned long) wq->wait_queue_token, wq->len, wq->name, notify); - /* autofs4_notify_daemon() may block */ - if (notify != NFY_NONE) { - autofs4_notify_daemon(sbi,wq, - notify == NFY_MOUNT ? - autofs_ptype_missing : - autofs_ptype_expire_multi); - } } else { atomic_inc(&wq->wait_ctr); up(&sbi->wq_sem); @@ -229,6 +220,17 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, (unsigned long) wq->wait_queue_token, wq->len, wq->name, notify); } + if (notify != NFY_NONE && atomic_dec_and_test(&wq->notified)) { + int type = (notify == NFY_MOUNT ? + autofs_ptype_missing : autofs_ptype_expire_multi); + + DPRINTK(("new wait id = 0x%08lx, name = %.*s, nfy=%d\n", + (unsigned long) wq->wait_queue_token, wq->len, wq->name, notify)); + + /* autofs4_notify_daemon() may block */ + autofs4_notify_daemon(sbi, wq, type); + } + /* wq->name is NULL if and only if the lock is already released */ if ( sbi->catatonic ) { diff --git a/fs/bad_inode.c b/fs/bad_inode.c index 672a31924f3..e172180a1d8 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -47,7 +47,7 @@ static struct file_operations bad_file_ops = .get_unmapped_area = EIO_ERROR, }; -struct inode_operations bad_inode_ops = +static struct inode_operations bad_inode_ops = { .create = EIO_ERROR, .lookup = EIO_ERROR, diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 6ae62cbf7c2..ce9423bb2de 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -945,7 +945,7 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs) retval = arch_setup_additional_pages(bprm, executable_stack); if (retval < 0) { send_sig(SIGKILL, current, 0); - goto out_free_dentry; + goto out; } #endif /* ARCH_HAS_SETUP_ADDITIONAL_PAGES */ @@ -140,6 +140,7 @@ inline void bio_init(struct bio *bio) * bio_alloc_bioset - allocate a bio for I/O * @gfp_mask: the GFP_ mask given to the slab allocator * @nr_iovecs: number of iovecs to pre-allocate + * @bs: the bio_set to allocate from * * Description: * bio_alloc_bioset will first try it's on mempool to satisfy the allocation. @@ -629,6 +630,7 @@ out: /** * bio_map_user - map user address into bio + * @q: the request_queue_t for the bio * @bdev: destination block device * @uaddr: start of user address * @len: length in bytes diff --git a/fs/block_dev.c b/fs/block_dev.c index d19d07c49ad..c0cbd1bc1a0 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -530,7 +530,7 @@ int check_disk_change(struct block_device *bdev) if (!bdops->media_changed(bdev->bd_disk)) return 0; - if (__invalidate_device(bdev, 0)) + if (__invalidate_device(bdev)) printk("VFS: busy inodes on changed media.\n"); if (bdops->revalidate_disk) diff --git a/fs/buffer.c b/fs/buffer.c index 3b12cf947ab..6f88dcc6d00 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -218,7 +218,7 @@ struct super_block *freeze_bdev(struct block_device *bdev) sb = get_super(bdev); if (sb && !(sb->s_flags & MS_RDONLY)) { sb->s_frozen = SB_FREEZE_WRITE; - wmb(); + smp_wmb(); sync_inodes_sb(sb, 0); DQUOT_SYNC(sb); @@ -235,7 +235,7 @@ struct super_block *freeze_bdev(struct block_device *bdev) sync_inodes_sb(sb, 1); sb->s_frozen = SB_FREEZE_TRANS; - wmb(); + smp_wmb(); sync_blockdev(sb->s_bdev); @@ -263,7 +263,7 @@ void thaw_bdev(struct block_device *bdev, struct super_block *sb) if (sb->s_op->unlockfs) sb->s_op->unlockfs(sb); sb->s_frozen = SB_UNFROZEN; - wmb(); + smp_wmb(); wake_up(&sb->s_wait_unfrozen); drop_super(sb); } @@ -774,15 +774,14 @@ repeat: /** * sync_mapping_buffers - write out and wait upon a mapping's "associated" * buffers - * @buffer_mapping - the mapping which backs the buffers' data - * @mapping - the mapping which wants those buffers written + * @mapping: the mapping which wants those buffers written * * Starts I/O against the buffers at mapping->private_list, and waits upon * that I/O. * - * Basically, this is a convenience function for fsync(). @buffer_mapping is - * the blockdev which "owns" the buffers and @mapping is a file or directory - * which needs those buffers to be written for a successful fsync(). + * Basically, this is a convenience function for fsync(). + * @mapping is a file or directory which needs those buffers to be written for + * a successful fsync(). */ int sync_mapping_buffers(struct address_space *mapping) { @@ -1211,7 +1210,7 @@ grow_buffers(struct block_device *bdev, sector_t block, int size) return 1; } -struct buffer_head * +static struct buffer_head * __getblk_slow(struct block_device *bdev, sector_t block, int size) { /* Size must be multiple of hard sectorsize */ @@ -1263,6 +1262,7 @@ __getblk_slow(struct block_device *bdev, sector_t block, int size) /** * mark_buffer_dirty - mark a buffer_head as needing writeout + * @bh: the buffer_head to mark dirty * * mark_buffer_dirty() will set the dirty bit against the buffer, then set its * backing page dirty, then tag the page as dirty in its address_space's radix @@ -1501,6 +1501,7 @@ EXPORT_SYMBOL(__breadahead); /** * __bread() - reads a specified block and returns the bh + * @bdev: the block_device to read from * @block: number of block * @size: size (in bytes) to read * @@ -1808,7 +1809,6 @@ static int __block_write_full_page(struct inode *inode, struct page *page, } while (bh != head); do { - get_bh(bh); if (!buffer_mapped(bh)) continue; /* @@ -1837,7 +1837,6 @@ static int __block_write_full_page(struct inode *inode, struct page *page, */ BUG_ON(PageWriteback(page)); set_page_writeback(page); - unlock_page(page); do { struct buffer_head *next = bh->b_this_page; @@ -1845,9 +1844,9 @@ static int __block_write_full_page(struct inode *inode, struct page *page, submit_bh(WRITE, bh); nr_underway++; } - put_bh(bh); bh = next; } while (bh != head); + unlock_page(page); err = 0; done: @@ -1886,7 +1885,6 @@ recover: bh = head; /* Recovery: lock and submit the mapped buffers */ do { - get_bh(bh); if (buffer_mapped(bh) && buffer_dirty(bh)) { lock_buffer(bh); mark_buffer_async_write(bh); @@ -1909,7 +1907,6 @@ recover: submit_bh(WRITE, bh); nr_underway++; } - put_bh(bh); bh = next; } while (bh != head); goto done; @@ -1952,7 +1949,7 @@ static int __block_prepare_write(struct inode *inode, struct page *page, if (!buffer_mapped(bh)) { err = get_block(inode, block, bh, 1); if (err) - goto out; + break; if (buffer_new(bh)) { clear_buffer_new(bh); unmap_underlying_metadata(bh->b_bdev, @@ -1994,10 +1991,12 @@ static int __block_prepare_write(struct inode *inode, struct page *page, while(wait_bh > wait) { wait_on_buffer(*--wait_bh); if (!buffer_uptodate(*wait_bh)) - return -EIO; + err = -EIO; } - return 0; -out: + if (!err) + return err; + + /* Error case: */ /* * Zero out any newly allocated blocks to avoid exposing stale * data. If BH_New is set, we know that the block was newly @@ -2078,8 +2077,7 @@ int block_read_full_page(struct page *page, get_block_t *get_block) int nr, i; int fully_mapped = 1; - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); blocksize = 1 << inode->i_blkbits; if (!page_has_buffers(page)) create_empty_buffers(page, blocksize, 0); @@ -2917,7 +2915,7 @@ drop_buffers(struct page *page, struct buffer_head **buffers_to_free) bh = head; do { - if (buffer_write_io_error(bh)) + if (buffer_write_io_error(bh) && page->mapping) set_bit(AS_EIO, &page->mapping->flags); if (buffer_busy(bh)) goto failed; @@ -3115,7 +3113,7 @@ void __init buffer_init(void) bh_cachep = kmem_cache_create("buffer_head", sizeof(struct buffer_head), 0, - SLAB_PANIC, init_buffer_head, NULL); + SLAB_RECLAIM_ACCOUNT|SLAB_PANIC, init_buffer_head, NULL); /* * Limit the bh occupancy to 10% of ZONE_NORMAL diff --git a/fs/char_dev.c b/fs/char_dev.c index a745b1d9e54..c1e3537909f 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -328,7 +328,7 @@ void cd_forget(struct inode *inode) spin_unlock(&cdev_lock); } -void cdev_purge(struct cdev *cdev) +static void cdev_purge(struct cdev *cdev) { spin_lock(&cdev_lock); while (!list_empty(&cdev->list)) { diff --git a/fs/cifs/AUTHORS b/fs/cifs/AUTHORS index acce36e25d2..72fdc10dfdd 100644 --- a/fs/cifs/AUTHORS +++ b/fs/cifs/AUTHORS @@ -4,13 +4,16 @@ Steve French (sfrench@samba.org) The author wishes to express his appreciation and thanks to: Andrew Tridgell (Samba team) for his early suggestions about smb/cifs VFS -improvements. Thanks to IBM for allowing me the time and test resources to pursue -this project. Jim McDonough from IBM (and the Samba Team) for his help. -The IBM Linux JFS team for explaining many esoteric Linux filesystem features. +improvements. Thanks to IBM for allowing me time and test resources to pursue +this project, to Jim McDonough from IBM (and the Samba Team) for his help, to +the IBM Linux JFS team for explaining many esoteric Linux filesystem features. +Jeremy Allison of the Samba team has done invaluable work in adding the server +side of the original CIFS Unix extensions and reviewing and implementing +portions of the newer CIFS POSIX extensions into the Samba 3 file server. Thank Dave Boutcher of IBM Rochester (author of the OS/400 smb/cifs filesystem client) -for proving years ago that a very good smb/cifs client could be done on a Unix like -operating system. Volker Lendecke, Andrew Tridgell, Urban Widmark, John Newbigin -and others for their work on the Linux smbfs module over the years. Thanks to +for proving years ago that very good smb/cifs clients could be done on Unix-like +operating systems. Volker Lendecke, Andrew Tridgell, Urban Widmark, John +Newbigin and others for their work on the Linux smbfs module. Thanks to the other members of the Storage Network Industry Association CIFS Technical Workgroup for their work specifying this highly complex protocol and finally thanks to the Samba team for their technical advice and encouragement. @@ -24,9 +27,11 @@ Shobhit Dayal Sergey Vlasov Richard Hughes Yury Umanets -Mark Hamzy +Mark Hamzy (for some of the early cifs IPv6 work) Domen Puncer -Jesper Juhl +Jesper Juhl (in particular for lots of whitespace/formatting cleanup) +Vince Negri and Dave Stahl (for finding an important caching bug) +Adrian Bunk (kcalloc cleanups) Test case and Bug Report contributors ------------------------------------- @@ -36,7 +41,8 @@ Rene Scharfe, Martin Josefsson, Alexander Wild, Anthony Liguori, Lars Muller, Urban Widmark, Massimiliano Ferrero, Howard Owen, Olaf Kirch, Kieron Briggs, Nick Millington and others. Also special mention to the Stanford Checker (SWAT) which pointed out many minor -bugs in error paths. +bugs in error paths. Valuable suggestions also have come from Al Viro +and Dave Miller. And thanks to the IBM LTC and Power test teams and SuSE testers for finding multiple bugs during excellent stress test runs. diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 5316c8dd6bf..95483baab70 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,7 +1,44 @@ -Version 1.31 +Version 1.34 +------------ +Fix error mapping of the TOO_MANY_LINKS (hardlinks) case. +Do not oops if root user kills cifs oplock kernel thread or +kills the cifsd thread (NB: killing the cifs kernel threads is not +recommended, unmount and rmmod cifs will kill them when they are +no longer needed). Fix readdir to ASCII servers (ie older servers +which do not support Unicode) and also require asterik. + + +Version 1.33 +------------ +Fix caching problem, in which readdir of directory containing a file +which was cached could cause the file's time stamp to be updated +without invalidating the readahead data (so we could get stale +file data on the client for that file even as the server copy changed). +Cleanup response processing so cifsd can not loop when abnormally +terminated. + + +Version 1.32 ------------ Fix oops in ls when Transact2 FindFirst (or FindNext) returns more than one transact response for an SMB request and search entry split across two frames. +Add support for lsattr (getting ext2/ext3/reiserfs attr flags from the server) +as new protocol extensions. Do not send Get/Set calls for POSIX ACLs +unless server explicitly claims to support them in CIFS Unix extensions +POSIX ACL capability bit. Fix packet signing when multiuser mounting with +different users from the same client to the same server. Fix oops in +cifs_close. Add mount option for remapping reserved characters in +filenames (also allow recognizing files with created by SFU which have any +of these seven reserved characters, except backslash, to be recognized). +Fix invalid transact2 message (we were sometimes trying to interpret +oplock breaks as SMB responses). Add ioctl for checking that the +current uid matches the uid of the mounter (needed by umount.cifs). +Reduce the number of large buffer allocations in cifs response processing +(significantly reduces memory pressure under heavy stress with multiple +processes accessing the same server at the same time). + +Version 1.31 +------------ Fix updates of DOS attributes and time fields so that files on NT4 servers do not get marked delete on close. Display sizes of cifs buffer pools in cifs stats. Fix oops in unmount when cifsd thread being killed by diff --git a/fs/cifs/README b/fs/cifs/README index 0f20edc935b..e74df0c7325 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -32,9 +32,9 @@ the cifs download to your kernel build directory e.g. 6) make modules (or "make" if CIFS VFS not to be built as a module) For Linux 2.6: -1) Download the kernel (e.g. from http://www.kernel.org or from bitkeeper -at bk://linux.bkbits.net/linux-2.5) and change directory into the top -of the kernel directory tree (e.g. /usr/src/linux-2.5.73) +1) Download the kernel (e.g. from http://www.kernel.org) +and change directory into the top of the kernel directory tree +(e.g. /usr/src/linux-2.5.73) 2) make menuconfig (or make xconfig) 3) select cifs from within the network filesystem choices 4) save and exit @@ -75,7 +75,7 @@ Allowing User Mounts ==================== To permit users to mount and unmount over directories they own is possible with the cifs vfs. A way to enable such mounting is to mark the mount.cifs -utility as suid (e.g. "chmod +s /sbin/mount/cifs). To enable users to +utility as suid (e.g. "chmod +s /sbin/mount.cifs). To enable users to umount shares they mount requires 1) mount.cifs version 1.4 or later 2) an entry for the share in /etc/fstab indicating that a user may @@ -97,6 +97,26 @@ mount.cifs with the following flag: There is a corresponding manual page for cifs mounting in the Samba 3.0 and later source tree in docs/manpages/mount.cifs.8 +Allowing User Unmounts +====================== +To permit users to ummount directories that they have user mounted (see above), +the utility umount.cifs may be used. It may be invoked directly, or if +umount.cifs is placed in /sbin, umount can invoke the cifs umount helper +(at least for most versions of the umount utility) for umount of cifs +mounts, unless umount is invoked with -i (which will avoid invoking a umount +helper). As with mount.cifs, to enable user unmounts umount.cifs must be marked +as suid (e.g. "chmod +s /sbin/umount.cifs") or equivalent (some distributions +allow adding entries to a file to the /etc/permissions file to achieve the +equivalent suid effect). For this utility to succeed the target path +must be a cifs mount, and the uid of the current user must match the uid +of the user who mounted the resource. + +Also note that the customary way of allowing user mounts and unmounts is +(instead of using mount.cifs and unmount.cifs as suid) to add a line +to the file /etc/fstab for each //server/share you wish to mount, but +this can become unwieldy when potential mount targets include many +or unpredictable UNC names. + Samba Considerations ==================== To get the maximum benefit from the CIFS VFS, we recommend using a server that @@ -376,6 +396,19 @@ A partial list of the supported mount options follows: attributes) to the server (default) e.g. via setfattr and getfattr utilities. nouser_xattr Do not allow getfattr/setfattr to get/set xattrs + mapchars Translate six of the seven reserved characters (not backslash) + *?<>|: + to the remap range (above 0xF000), which also + allows the CIFS client to recognize files created with + such characters by Windows's POSIX emulation. This can + also be useful when mounting to most versions of Samba + (which also forbids creating and opening files + whose names contain any of these seven characters). + This has no effect if the server does not support + Unicode on the wire. + nomapchars Do not translate any of these seven characters (default). + remount remount the share (often used to change from ro to rw mounts + or vice versa) The mount.cifs mount helper also accepts a few mount options before -o including: @@ -392,7 +425,7 @@ Misc /proc/fs/cifs Flags and Debug Info ======================================= Informational pseudo-files: DebugData Displays information about active CIFS sessions - and shares. + and shares, as well as the cifs.ko version. Stats Lists summary resource usage information as well as per share statistics, if CONFIG_CIFS_STATS in enabled in the kernel configuration. @@ -449,7 +482,7 @@ and for more extensive tracing including the start of smb requests and responses Two other experimental features are under development and to test require enabling CONFIG_CIFS_EXPERIMENTAL - More efficient write operations and SMB buffer handling + More efficient write operations DNOTIFY fcntl: needed for support of directory change notification and perhaps later for file leases) @@ -467,8 +500,8 @@ returned success. Also note that "cat /proc/fs/cifs/DebugData" will display information about the active sessions and the shares that are mounted. Note: NTLMv2 enablement -will not work since they its implementation is not quite complete yet. -Do not alter these configuration values unless you are doing specific testing. +will not work since its implementation is not quite complete yet. Do not alter +the ExtendedSecurity configuration value unless you are doing specific testing. Enabling extended security works to Windows 2000 Workstations and XP but not to Windows 2000 server or Samba since it does not usually send "raw NTLMSSP" (instead it sends NTLMSSP encapsulated in SPNEGO/GSSAPI, which support is not diff --git a/fs/cifs/TODO b/fs/cifs/TODO index f4e3e1f67ee..8cc881694e2 100644 --- a/fs/cifs/TODO +++ b/fs/cifs/TODO @@ -1,4 +1,4 @@ -version 1.22 July 30, 2004 +version 1.34 April 29, 2005 A Partial List of Missing Features ================================== @@ -14,7 +14,7 @@ b) Better pam/winbind integration (e.g. to handle uid mapping better) c) multi-user mounts - multiplexed sessionsetups over single vc -(ie tcp session) - prettying up needed, and more testing needed +(ie tcp session) - more testing needed d) Kerberos/SPNEGO session setup support - (started) @@ -67,12 +67,26 @@ q) implement support for security and trusted categories of xattrs r) Implement O_DIRECT flag on open (already supported on mount) -KNOWN BUGS (updated December 10, 2004) +s) Allow remapping of last remaining character (\) to +0xF000 which +(this character is valid for POSIX but not for Windows) + +t) Create UID mapping facility so server UIDs can be mapped on a per +mount or a per server basis to client UIDs or nobody if no mapping +exists. This is helpful when Unix extensions are negotiated to +allow better permission checking when UIDs differ on the server +and client. Add new protocol request to the CIFS protocol +standard for asking the server for the corresponding name of a +particular uid. + +KNOWN BUGS (updated April 29, 2005) ==================================== +See http://bugzilla.samba.org - search on product "CifsVFS" for +current bug list. + 1) existing symbolic links (Windows reparse points) are recognized but can not be created remotely. They are implemented for Samba and those that -support the CIFS Unix extensions but Samba has a bug currently handling -symlink text beginning with slash +support the CIFS Unix extensions, although earlier versions of Samba +overly restrict the pathnames. 2) follow_link and readdir code does not follow dfs junctions but recognizes them 3) create of new files to FAT partitions on Windows servers can @@ -98,7 +112,5 @@ there are some easy changes that can be done to parallelize sequential writes, and when signing is disabled to request larger read sizes (larger than negotiated size) and send larger write sizes to modern servers. -4) More exhaustively test the recently added NT4 support against various -NT4 service pack levels, and fix cifs_setattr for setting file times and -size to fall back to level 1 when error invalid level returned. - +4) More exhaustively test against less common servers. More testing +against Windows 9x, Windows ME servers. diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index db28b561cd4..4061e43471c 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -1,7 +1,7 @@ /* * fs/cifs_debug.c * - * Copyright (C) International Business Machines Corp., 2000,2003 + * Copyright (C) International Business Machines Corp., 2000,2005 * * Modified by Steve French (sfrench@us.ibm.com) * @@ -29,6 +29,7 @@ #include "cifsglob.h" #include "cifsproto.h" #include "cifs_debug.h" +#include "cifsfs.h" void cifs_dump_mem(char *label, void *data, int length) @@ -78,8 +79,9 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, "Display Internal CIFS Data Structures for Debugging\n" "---------------------------------------------------\n"); buf += length; - - length = sprintf(buf, "Servers:\n"); + length = sprintf(buf,"CIFS Version %s\n",CIFS_VERSION); + buf += length; + length = sprintf(buf, "Servers:"); buf += length; i = 0; @@ -87,12 +89,21 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, list_for_each(tmp, &GlobalSMBSessionList) { i++; ses = list_entry(tmp, struct cifsSesInfo, cifsSessionList); - length = - sprintf(buf, - "\n%d) Name: %s Domain: %s Mounts: %d ServerOS: %s \n\tServerNOS: %s\tCapabilities: 0x%x\n\tSMB session status: %d\t", - i, ses->serverName, ses->serverDomain, atomic_read(&ses->inUse), - ses->serverOS, ses->serverNOS, ses->capabilities,ses->status); - buf += length; + if((ses->serverDomain == NULL) || (ses->serverOS == NULL) || + (ses->serverNOS == NULL)) { + buf += sprintf("\nentry for %s not fully displayed\n\t", + ses->serverName); + + } else { + length = + sprintf(buf, + "\n%d) Name: %s Domain: %s Mounts: %d ServerOS: %s \n\tServerNOS: %s\tCapabilities: 0x%x\n\tSMB session status: %d\t", + i, ses->serverName, ses->serverDomain, + atomic_read(&ses->inUse), + ses->serverOS, ses->serverNOS, + ses->capabilities,ses->status); + buf += length; + } if(ses->server) { buf += sprintf(buf, "TCP status: %d\n\tLocal Users To Server: %d SecMode: 0x%x Req Active: %d", ses->server->tcpStatus, @@ -100,7 +111,7 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, ses->server->secMode, atomic_read(&ses->server->inFlight)); - length = sprintf(buf, "\nMIDs: \n"); + length = sprintf(buf, "\nMIDs:\n"); buf += length; spin_lock(&GlobalMid_Lock); @@ -109,7 +120,12 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, mid_q_entry, qhead); if(mid_entry) { - length = sprintf(buf,"State: %d com: %d pid: %d tsk: %p mid %d\n",mid_entry->midState,mid_entry->command,mid_entry->pid,mid_entry->tsk,mid_entry->mid); + length = sprintf(buf,"State: %d com: %d pid: %d tsk: %p mid %d\n", + mid_entry->midState, + (int)mid_entry->command, + mid_entry->pid, + mid_entry->tsk, + mid_entry->mid); buf += length; } } @@ -121,7 +137,7 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, sprintf(buf, "\n"); buf++; - length = sprintf(buf, "\nShares:\n"); + length = sprintf(buf, "Shares:"); buf += length; i = 0; @@ -200,7 +216,8 @@ cifs_stats_read(char *buf, char **beginBuffer, off_t offset, buf += item_length; item_length = sprintf(buf,"SMB Request/Response Buffer: %d Pool size: %d\n", - bufAllocCount.counter,cifs_min_rcv + tcpSesAllocCount.counter); + bufAllocCount.counter, + cifs_min_rcv + tcpSesAllocCount.counter); length += item_length; buf += item_length; item_length = diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index 77da902d8f3..ec00d61d530 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -23,6 +23,7 @@ #define CIFS_MOUNT_SERVER_INUM 4 /* inode numbers from uniqueid from server */ #define CIFS_MOUNT_DIRECT_IO 8 /* do not write nor read through page cache */ #define CIFS_MOUNT_NO_XATTR 0x10 /* if set - disable xattr support */ +#define CIFS_MOUNT_MAP_SPECIAL_CHR 0x20 /* remap illegal chars in filenames */ struct cifs_sb_info { struct cifsTconInfo *tcon; /* primary mount */ diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c index a17adf4cb9b..99a096d3f84 100644 --- a/fs/cifs/cifs_unicode.c +++ b/fs/cifs/cifs_unicode.c @@ -76,8 +76,8 @@ cifs_strtoUCS(wchar_t * to, const char *from, int len, charlen)); to[i] = cpu_to_le16(0x003f); /* a question mark */ charlen = 1; - } - to[i] = cpu_to_le16(to[i]); + } else + to[i] = cpu_to_le16(to[i]); } diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index 78829e7d8cd..1959c7c4b18 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -50,7 +50,7 @@ static int cifs_calculate_signature(const struct smb_hdr * cifs_pdu, const char return 0; } -int cifs_sign_smb(struct smb_hdr * cifs_pdu, struct cifsSesInfo * ses, +int cifs_sign_smb(struct smb_hdr * cifs_pdu, struct TCP_Server_Info * server, __u32 * pexpected_response_sequence_number) { int rc = 0; @@ -59,21 +59,21 @@ int cifs_sign_smb(struct smb_hdr * cifs_pdu, struct cifsSesInfo * ses, /* BB remember to initialize sequence number elsewhere and initialize mac_signing key elsewhere BB */ /* BB remember to add code to save expected sequence number in midQ entry BB */ - if((cifs_pdu == NULL) || (ses == NULL)) + if((cifs_pdu == NULL) || (server == NULL)) return -EINVAL; if((cifs_pdu->Flags2 & SMBFLG2_SECURITY_SIGNATURE) == 0) return rc; spin_lock(&GlobalMid_Lock); - cifs_pdu->Signature.Sequence.SequenceNumber = cpu_to_le32(ses->sequence_number); + cifs_pdu->Signature.Sequence.SequenceNumber = cpu_to_le32(server->sequence_number); cifs_pdu->Signature.Sequence.Reserved = 0; - *pexpected_response_sequence_number = ses->sequence_number++; - ses->sequence_number++; + *pexpected_response_sequence_number = server->sequence_number++; + server->sequence_number++; spin_unlock(&GlobalMid_Lock); - rc = cifs_calculate_signature(cifs_pdu, ses->mac_signing_key,smb_signature); + rc = cifs_calculate_signature(cifs_pdu, server->mac_signing_key,smb_signature); if(rc) memset(cifs_pdu->Signature.SecuritySignature, 0, 8); else @@ -190,7 +190,7 @@ int CalcNTLMv2_partial_mac_key(struct cifsSesInfo * ses, struct nls_table * nls_ hmac_md5_update((const unsigned char *) unicode_buf, (user_name_len+dom_name_len)*2,&ctx); - hmac_md5_final(ses->mac_signing_key,&ctx); + hmac_md5_final(ses->server->mac_signing_key,&ctx); kfree(ucase_buf); kfree(unicode_buf); return 0; @@ -200,7 +200,7 @@ void CalcNTLMv2_response(const struct cifsSesInfo * ses,char * v2_session_respon struct HMACMD5Context context; memcpy(v2_session_response + 8, ses->server->cryptKey,8); /* gen_blob(v2_session_response + 16); */ - hmac_md5_init_limK_to_64(ses->mac_signing_key, 16, &context); + hmac_md5_init_limK_to_64(ses->server->mac_signing_key, 16, &context); hmac_md5_update(ses->server->cryptKey,8,&context); /* hmac_md5_update(v2_session_response+16)client thing,8,&context); */ /* BB fix */ diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 5082fce3c56..8cc23e7d0d5 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -169,7 +169,8 @@ cifs_put_super(struct super_block *sb) static int cifs_statfs(struct super_block *sb, struct kstatfs *buf) { - int xid, rc = -EOPNOTSUPP; + int xid; + int rc = -EOPNOTSUPP; struct cifs_sb_info *cifs_sb; struct cifsTconInfo *pTcon; @@ -181,31 +182,34 @@ cifs_statfs(struct super_block *sb, struct kstatfs *buf) buf->f_type = CIFS_MAGIC_NUMBER; /* instead could get the real value via SMB_QUERY_FS_ATTRIBUTE_INFO */ - buf->f_namelen = PATH_MAX; /* PATH_MAX may be too long - it would presumably - be length of total path, note that some servers may be - able to support more than this, but best to be safe - since Win2k and others can not handle very long filenames */ + buf->f_namelen = PATH_MAX; /* PATH_MAX may be too long - it would + presumably be total path, but note + that some servers (includinng Samba 3) + have a shorter maximum path */ buf->f_files = 0; /* undefined */ buf->f_ffree = 0; /* unlimited */ #ifdef CONFIG_CIFS_EXPERIMENTAL /* BB we could add a second check for a QFS Unix capability bit */ - if (pTcon->ses->capabilities & CAP_UNIX) - rc = CIFSSMBQFSPosixInfo(xid, pTcon, buf, cifs_sb->local_nls); +/* BB FIXME check CIFS_POSIX_EXTENSIONS Unix cap first FIXME BB */ + if ((pTcon->ses->capabilities & CAP_UNIX) && (CIFS_POSIX_EXTENSIONS & + le64_to_cpu(pTcon->fsUnixInfo.Capability))) + rc = CIFSSMBQFSPosixInfo(xid, pTcon, buf); /* Only need to call the old QFSInfo if failed on newer one */ if(rc) #endif /* CIFS_EXPERIMENTAL */ - rc = CIFSSMBQFSInfo(xid, pTcon, buf, cifs_sb->local_nls); + rc = CIFSSMBQFSInfo(xid, pTcon, buf); /* int f_type; __fsid_t f_fsid; int f_namelen; */ - /* BB get from info put in tcon struct at mount time with call to QFSAttrInfo */ + /* BB get from info in tcon struct at mount time call to QFSAttrInfo */ FreeXid(xid); - return 0; /* always return success? what if volume is no longer available? */ + return 0; /* always return success? what if volume is no + longer available? */ } static int cifs_permission(struct inode * inode, int mask, struct nameidata *nd) @@ -559,6 +563,10 @@ struct file_operations cifs_file_ops = { .flush = cifs_flush, .mmap = cifs_file_mmap, .sendfile = generic_file_sendfile, +#ifdef CONFIG_CIFS_POSIX + .ioctl = cifs_ioctl, +#endif /* CONFIG_CIFS_POSIX */ + #ifdef CONFIG_CIFS_EXPERIMENTAL .readv = generic_file_readv, .writev = generic_file_writev, @@ -579,6 +587,10 @@ struct file_operations cifs_file_direct_ops = { .fsync = cifs_fsync, .flush = cifs_flush, .sendfile = generic_file_sendfile, /* BB removeme BB */ +#ifdef CONFIG_CIFS_POSIX + .ioctl = cifs_ioctl, +#endif /* CONFIG_CIFS_POSIX */ + #ifdef CONFIG_CIFS_EXPERIMENTAL .dir_notify = cifs_dir_notify, #endif /* CONFIG_CIFS_EXPERIMENTAL */ @@ -591,6 +603,7 @@ struct file_operations cifs_dir_ops = { #ifdef CONFIG_CIFS_EXPERIMENTAL .dir_notify = cifs_dir_notify, #endif /* CONFIG_CIFS_EXPERIMENTAL */ + .ioctl = cifs_ioctl, }; static void @@ -822,6 +835,7 @@ static int cifs_oplock_thread(void * dummyarg) } } while(!signal_pending(current)); complete_and_exit (&cifs_oplock_exited, 0); + oplockThread = NULL; } static int __init diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 451f18af320..d00b3bfe1a5 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -1,7 +1,7 @@ /* * fs/cifs/cifsfs.h * - * Copyright (c) International Business Machines Corp., 2002 + * Copyright (c) International Business Machines Corp., 2002, 2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -91,8 +91,10 @@ extern int cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname); extern int cifs_removexattr(struct dentry *, const char *); extern int cifs_setxattr(struct dentry *, const char *, const void *, - size_t, int); + size_t, int); extern ssize_t cifs_getxattr(struct dentry *, const char *, void *, size_t); extern ssize_t cifs_listxattr(struct dentry *, char *, size_t); -#define CIFS_VERSION "1.31" +extern int cifs_ioctl (struct inode * inode, struct file * filep, + unsigned int command, unsigned long arg); +#define CIFS_VERSION "1.34" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 69aff1a7da9..81babab265e 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -1,7 +1,7 @@ /* * fs/cifs/cifsglob.h * - * Copyright (C) International Business Machines Corp., 2002,2003 + * Copyright (C) International Business Machines Corp., 2002,2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -149,6 +149,8 @@ struct TCP_Server_Info { __u16 timeZone; char cryptKey[CIFS_CRYPTO_KEY_SIZE]; char workstation_RFC1001_name[16]; /* 16th byte is always zero */ + __u32 sequence_number; /* needed for CIFS PDU signature */ + char mac_signing_key[CIFS_SESSION_KEY_SIZE + 16]; }; /* @@ -174,17 +176,16 @@ struct cifsSesInfo { struct TCP_Server_Info *server; /* pointer to server info */ atomic_t inUse; /* # of mounts (tree connections) on this ses */ enum statusEnum status; - __u32 sequence_number; /* needed for CIFS PDU signature */ __u16 ipc_tid; /* special tid for connection to IPC share */ __u16 flags; - char mac_signing_key[CIFS_SESSION_KEY_SIZE + 16]; - char *serverOS; /* name of operating system underlying the server */ - char *serverNOS; /* name of network operating system that the server is running */ + char *serverOS; /* name of operating system underlying server */ + char *serverNOS; /* name of network operating system of server */ char *serverDomain; /* security realm of server */ int Suid; /* remote smb uid */ uid_t linux_uid; /* local Linux uid */ int capabilities; - char serverName[SERVER_NAME_LEN_WITH_NULL * 2]; /* BB make bigger for tcp names - will ipv6 and sctp addresses fit here?? */ + char serverName[SERVER_NAME_LEN_WITH_NULL * 2]; /* BB make bigger for + TCP names - will ipv6 and sctp addresses fit? */ char userName[MAX_USERNAME_SIZE + 1]; char domainName[MAX_USERNAME_SIZE + 1]; char * password; @@ -312,12 +313,15 @@ struct mid_q_entry { __u16 mid; /* multiplex id */ __u16 pid; /* process id */ __u32 sequence_number; /* for CIFS signing */ - __u16 command; /* smb command code */ struct timeval when_sent; /* time when smb sent */ struct cifsSesInfo *ses; /* smb was sent to this server */ struct task_struct *tsk; /* task waiting for response */ struct smb_hdr *resp_buf; /* response buffer */ int midState; /* wish this were enum but can not pass to wait_event */ + __u8 command; /* smb command code */ + unsigned multiPart:1; /* multiple responses to one SMB request */ + unsigned largeBuf:1; /* if valid response, is pointer to large buf */ + unsigned multiResp:1; /* multiple trans2 responses for one request */ }; struct oplock_q_entry { diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index bcd4a6136f0..aede6a81316 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -330,7 +330,7 @@ struct smb_hdr { }; /* given a pointer to an smb_hdr retrieve the value of byte count */ #define BCC(smb_var) ( *(__u16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2* smb_var->WordCount) ) ) - +#define BCC_LE(smb_var) ( *(__le16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2* smb_var->WordCount) ) ) /* given a pointer to an smb_hdr retrieve the pointer to the byte area */ #define pByteArea(smb_var) ((unsigned char *)smb_var + sizeof(struct smb_hdr) + (2* smb_var->WordCount) + 2 ) @@ -762,6 +762,16 @@ typedef struct smb_com_lock_req { LOCKING_ANDX_RANGE Locks[1]; } LOCK_REQ; + +typedef struct cifs_posix_lock { + __le16 lock_type; /* 0 = Read, 1 = Write, 2 = Unlock */ + __le16 lock_flags; /* 1 = Wait (only valid for setlock) */ + __le32 pid; + __le64 start; + __le64 length; + /* BB what about additional owner info to identify network client */ +} CIFS_POSIX_LOCK; + typedef struct smb_com_lock_rsp { struct smb_hdr hdr; /* wct = 2 */ __u8 AndXCommand; @@ -1098,6 +1108,8 @@ struct smb_t2_rsp { #define SMB_QUERY_POSIX_ACL 0x204 #define SMB_QUERY_XATTR 0x205 #define SMB_QUERY_ATTR_FLAGS 0x206 /* append,immutable etc. */ +#define SMB_QUERY_POSIX_PERMISSION 0x207 +#define SMB_QUERY_POSIX_LOCK 0x208 #define SMB_QUERY_FILE_INTERNAL_INFO 0x3ee #define SMB_QUERY_FILE_ACCESS_INFO 0x3f0 #define SMB_QUERY_FILE_NAME_INFO2 0x3f1 /* 0x30 bytes */ @@ -1116,6 +1128,7 @@ struct smb_t2_rsp { #define SMB_SET_POSIX_ACL 0x204 #define SMB_SET_XATTR 0x205 #define SMB_SET_ATTR_FLAGS 0x206 /* append, immutable etc. */ +#define SMB_SET_POSIX_LOCK 0x208 #define SMB_SET_FILE_BASIC_INFO2 0x3ec #define SMB_SET_FILE_RENAME_INFORMATION 0x3f2 /* BB check if qpathinfo level too */ #define SMB_FILE_ALL_INFO2 0x3fa @@ -1237,9 +1250,25 @@ struct smb_com_transaction2_sfi_rsp { struct smb_hdr hdr; /* wct = 10 + SetupCount */ struct trans2_resp t2; __u16 ByteCount; - __u16 Reserved2; /* parameter word reserved - present for infolevels > 100 */ + __u16 Reserved2; /* parameter word reserved - + present for infolevels > 100 */ +}; + +struct smb_t2_qfi_req { + struct smb_hdr hdr; + struct trans2_req t2; + __u8 Pad; + __u16 Fid; + __le16 InformationLevel; }; +struct smb_t2_qfi_rsp { + struct smb_hdr hdr; /* wct = 10 + SetupCount */ + struct trans2_resp t2; + __u16 ByteCount; + __u16 Reserved2; /* parameter word reserved - + present for infolevels > 100 */ +}; /* * Flags on T2 FINDFIRST and FINDNEXT @@ -1524,9 +1553,10 @@ typedef struct { } FILE_SYSTEM_UNIX_INFO; /* Unix extensions info, level 0x200 */ /* Linux/Unix extensions capability flags */ #define CIFS_UNIX_FCNTL_CAP 0x00000001 /* support for fcntl locks */ -#define CIFS_UNIX_POSIX_ACL_CAP 0x00000002 -#define CIFS_UNIX_XATTR_CAP 0x00000004 /*support for new namespace*/ - +#define CIFS_UNIX_POSIX_ACL_CAP 0x00000002 /* support getfacl/setfacl */ +#define CIFS_UNIX_XATTR_CAP 0x00000004 /* support new namespace */ +#define CIFS_UNIX_EXTATTR_CAP 0x00000008 /* support chattr/chflag */ +#define CIFS_POSIX_EXTENSIONS 0x00000010 /* support for new QFSInfo */ typedef struct { /* For undefined recommended transfer size return -1 in that field */ __le32 OptimalTransferSize; /* bsize on some os, iosize on other os */ @@ -1971,14 +2001,39 @@ struct xsymlink { char path[1024]; }; -typedef struct { +typedef struct file_xattr_info { /* BB do we need another field for flags? BB */ __u32 xattr_name_len; __u32 xattr_value_len; char xattr_name[0]; /* followed by xattr_value[xattr_value_len], no pad */ -} FILE_XATTR_INFO; /* extended attribute, info level 205 */ - +} FILE_XATTR_INFO; /* extended attribute, info level 0x205 */ + + +/* flags for chattr command */ +#define EXT_SECURE_DELETE 0x00000001 /* EXT3_SECRM_FL */ +#define EXT_ENABLE_UNDELETE 0x00000002 /* EXT3_UNRM_FL */ +/* Reserved for compress file 0x4 */ +#define EXT_SYNCHRONOUS 0x00000008 /* EXT3_SYNC_FL */ +#define EXT_IMMUTABLE_FL 0x00000010 /* EXT3_IMMUTABLE_FL */ +#define EXT_OPEN_APPEND_ONLY 0x00000020 /* EXT3_APPEND_FL */ +#define EXT_DO_NOT_BACKUP 0x00000040 /* EXT3_NODUMP_FL */ +#define EXT_NO_UPDATE_ATIME 0x00000080 /* EXT3_NOATIME_FL */ +/* 0x100 through 0x800 reserved for compression flags and are GET-ONLY */ +#define EXT_HASH_TREE_INDEXED_DIR 0x00001000 /* GET-ONLY EXT3_INDEX_FL */ +/* 0x2000 reserved for IMAGIC_FL */ +#define EXT_JOURNAL_THIS_FILE 0x00004000 /* GET-ONLY EXT3_JOURNAL_DATA_FL */ +/* 0x8000 reserved for EXT3_NOTAIL_FL */ +#define EXT_SYNCHRONOUS_DIR 0x00010000 /* EXT3_DIRSYNC_FL */ +#define EXT_TOPDIR 0x00020000 /* EXT3_TOPDIR_FL */ + +#define EXT_SET_MASK 0x000300FF +#define EXT_GET_MASK 0x0003DFFF + +typedef struct file_chattr_info { + __le64 mask; /* list of all possible attribute bits */ + __le64 mode; /* list of actual attribute bits on this inode */ +} FILE_CHATTR_INFO; /* ext attributes (chattr, chflags) level 0x206 */ #endif diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 787eef4d86d..0010511083f 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -57,10 +57,11 @@ extern int decode_negTokenInit(unsigned char *security_blob, int length, extern int cifs_inet_pton(int, char * source, void *dst); extern int map_smb_to_linux_error(struct smb_hdr *smb); extern void header_assemble(struct smb_hdr *, char /* command */ , - const struct cifsTconInfo *, int - /* length of fixed section (word count) in two byte units */ + const struct cifsTconInfo *, int /* specifies length + of fixed section (word count) in two byte units */ ); -extern struct oplock_q_entry * AllocOplockQEntry(struct inode *, u16, struct cifsTconInfo *); +extern struct oplock_q_entry * AllocOplockQEntry(struct inode *, u16, + struct cifsTconInfo *); extern void DeleteOplockQEntry(struct oplock_q_entry *); extern struct timespec cifs_NTtimeToUnix(u64 /* utc nanoseconds since 1601 */ ); extern u64 cifs_UnixTimeToNT(struct timespec); @@ -88,7 +89,7 @@ extern int CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, extern int CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, const char *searchName, const struct nls_table *nls_codepage, - __u16 *searchHandle, struct cifs_search_info * psrch_inf); + __u16 *searchHandle, struct cifs_search_info * psrch_inf, int map); extern int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, __u16 searchHandle, struct cifs_search_info * psrch_inf); @@ -99,42 +100,42 @@ extern int CIFSFindClose(const int, struct cifsTconInfo *tcon, extern int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_ALL_INFO * findData, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, int remap); extern int CIFSSMBUnixQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_UNIX_BASIC_INFO * pFindData, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, int remap); extern int CIFSGetDFSRefer(const int xid, struct cifsSesInfo *ses, const unsigned char *searchName, unsigned char **targetUNCs, unsigned int *number_of_UNC_in_array, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, int remap); extern int connect_to_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, int remap); extern int get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, - const char *old_path, const struct nls_table *nls_codepage, - unsigned int *pnum_referrals, unsigned char ** preferrals); + const char *old_path, + const struct nls_table *nls_codepage, + unsigned int *pnum_referrals, + unsigned char ** preferrals, + int remap); extern int CIFSSMBQFSInfo(const int xid, struct cifsTconInfo *tcon, - struct kstatfs *FSData, - const struct nls_table *nls_codepage); + struct kstatfs *FSData); extern int CIFSSMBQFSAttributeInfo(const int xid, - struct cifsTconInfo *tcon, - const struct nls_table *nls_codepage); -extern int CIFSSMBQFSDeviceInfo(const int xid, struct cifsTconInfo *tcon, - const struct nls_table *nls_codepage); -extern int CIFSSMBQFSUnixInfo(const int xid, struct cifsTconInfo *tcon, - const struct nls_table *nls_codepage); + struct cifsTconInfo *tcon); +extern int CIFSSMBQFSDeviceInfo(const int xid, struct cifsTconInfo *tcon); +extern int CIFSSMBQFSUnixInfo(const int xid, struct cifsTconInfo *tcon); extern int CIFSSMBQFSPosixInfo(const int xid, struct cifsTconInfo *tcon, - struct kstatfs *FSData, const struct nls_table *nls_codepage); + struct kstatfs *FSData); extern int CIFSSMBSetTimes(const int xid, struct cifsTconInfo *tcon, const char *fileName, const FILE_BASIC_INFO * data, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, const FILE_BASIC_INFO * data, __u16 fid); #if 0 @@ -143,36 +144,49 @@ extern int CIFSSMBSetAttrLegacy(int xid, struct cifsTconInfo *tcon, const struct nls_table *nls_codepage); #endif /* possibly unneeded function */ extern int CIFSSMBSetEOF(const int xid, struct cifsTconInfo *tcon, - const char *fileName, __u64 size,int setAllocationSizeFlag, - const struct nls_table *nls_codepage); + const char *fileName, __u64 size, + int setAllocationSizeFlag, + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, - __u64 size, __u16 fileHandle,__u32 opener_pid, int AllocSizeFlag); + __u64 size, __u16 fileHandle,__u32 opener_pid, + int AllocSizeFlag); extern int CIFSSMBUnixSetPerms(const int xid, struct cifsTconInfo *pTcon, char *full_path, __u64 mode, __u64 uid, - __u64 gid, dev_t dev, const struct nls_table *nls_codepage); + __u64 gid, dev_t dev, + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSSMBMkDir(const int xid, struct cifsTconInfo *tcon, const char *newName, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSSMBRmDir(const int xid, struct cifsTconInfo *tcon, - const char *name, const struct nls_table *nls_codepage); + const char *name, const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSSMBDelFile(const int xid, struct cifsTconInfo *tcon, const char *name, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSSMBRename(const int xid, struct cifsTconInfo *tcon, const char *fromName, const char *toName, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSSMBRenameOpenFile(const int xid,struct cifsTconInfo *pTcon, - int netfid, char * target_name, const struct nls_table *nls_codepage); + int netfid, char * target_name, + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSCreateHardLink(const int xid, struct cifsTconInfo *tcon, const char *fromName, const char *toName, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSUnixCreateHardLink(const int xid, struct cifsTconInfo *tcon, const char *fromName, const char *toName, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSUnixCreateSymLink(const int xid, struct cifsTconInfo *tcon, const char *fromName, const char *toName, @@ -192,7 +206,7 @@ extern int CIFSSMBOpen(const int xid, struct cifsTconInfo *tcon, const char *fileName, const int disposition, const int access_flags, const int omode, __u16 * netfid, int *pOplock, FILE_ALL_INFO *, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, int remap); extern int CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, const int smb_file_id); @@ -211,8 +225,13 @@ extern int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, const char __user *buf,const int long_op); extern int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, __u64 * inode_number, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, + int remap_special_chars); #endif /* CONFIG_CIFS_EXPERIMENTAL */ +extern int cifs_convertUCSpath(char *target, const __u16 *source, int maxlen, + const struct nls_table * codepage); +extern int cifsConvertToUCS(__le16 * target, const char *source, int maxlen, + const struct nls_table * cp, int mapChars); extern int CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, const __u16 netfid, const __u64 len, @@ -230,7 +249,7 @@ extern void tconInfoFree(struct cifsTconInfo *); extern int cifs_reconnect(struct TCP_Server_Info *server); -extern int cifs_sign_smb(struct smb_hdr *, struct cifsSesInfo *,__u32 *); +extern int cifs_sign_smb(struct smb_hdr *, struct TCP_Server_Info *,__u32 *); extern int cifs_verify_signature(struct smb_hdr *, const char * mac_key, __u32 expected_sequence_number); extern int cifs_calculate_mac_key(char * key,const char * rn,const char * pass); @@ -241,29 +260,31 @@ extern int CIFSSMBCopy(int xid, const char *fromName, const __u16 target_tid, const char *toName, const int flags, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, + int remap_special_chars); extern int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, - const int notify_subdirs,const __u16 netfid,__u32 filter, - const struct nls_table *nls_codepage); + const int notify_subdirs,const __u16 netfid, + __u32 filter, const struct nls_table *nls_codepage); extern ssize_t CIFSSMBQAllEAs(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, char * EAData, - size_t bufsize, const struct nls_table *nls_codepage); + size_t bufsize, const struct nls_table *nls_codepage, + int remap_special_chars); extern ssize_t CIFSSMBQueryEA(const int xid,struct cifsTconInfo * tcon, const unsigned char * searchName,const unsigned char * ea_name, unsigned char * ea_value, size_t buf_size, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSSMBSetEA(const int xid, struct cifsTconInfo *tcon, const char *fileName, const char * ea_name, const void * ea_value, const __u16 ea_value_len, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSSMBGetPosixACL(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, char *acl_inf, const int buflen,const int acl_type, - const struct nls_table *nls_codepage); + const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSSMBSetPosixACL(const int xid, struct cifsTconInfo *tcon, const unsigned char *fileName, const char *local_acl, const int buflen, const int acl_type, - const struct nls_table *nls_codepage); -int cifs_ioctl (struct inode * inode, struct file * filep, - unsigned int command, unsigned long arg); + const struct nls_table *nls_codepage, int remap_special_chars); +extern int CIFSGetExtAttr(const int xid, struct cifsTconInfo *tcon, + const int netfid, __u64 * pExtAttrBits, __u64 *pMask); #endif /* _CIFSPROTO_H */ diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index df6a619a682..741ff0c69f3 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -75,7 +75,8 @@ static void mark_open_files_invalid(struct cifsTconInfo * pTcon) } } write_unlock(&GlobalSMBSeslock); - /* BB Add call to invalidate_inodes(sb) for all superblocks mounted to this tcon */ + /* BB Add call to invalidate_inodes(sb) for all superblocks mounted + to this tcon */ } /* If the return code is zero, this function must fill in request_buf pointer */ @@ -89,11 +90,12 @@ small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, check for tcp and smb session status done differently for those three - in the calling routine */ if(tcon) { - if((tcon->ses) && (tcon->ses->server)){ + if((tcon->ses) && (tcon->ses->status != CifsExiting) && + (tcon->ses->server)){ struct nls_table *nls_codepage; /* Give Demultiplex thread up to 10 seconds to - reconnect, should be greater than cifs socket - timeout which is 7 seconds */ + reconnect, should be greater than cifs socket + timeout which is 7 seconds */ while(tcon->ses->server->tcpStatus == CifsNeedReconnect) { wait_event_interruptible_timeout(tcon->ses->server->response_q, (tcon->ses->server->tcpStatus == CifsGood), 10 * HZ); @@ -103,8 +105,9 @@ small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, (tcon->ses->status == CifsExiting)) { cFYI(1,("gave up waiting on reconnect in smb_init")); return -EHOSTDOWN; - } /* else "hard" mount - keep retrying until - process is killed or server comes back up */ + } /* else "hard" mount - keep retrying + until process is killed or server + comes back on-line */ } else /* TCP session is reestablished now */ break; @@ -115,23 +118,24 @@ small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, simultaneously reconnect the same SMB session */ down(&tcon->ses->sesSem); if(tcon->ses->status == CifsNeedReconnect) - rc = cifs_setup_session(0, tcon->ses, nls_codepage); + rc = cifs_setup_session(0, tcon->ses, + nls_codepage); if(!rc && (tcon->tidStatus == CifsNeedReconnect)) { mark_open_files_invalid(tcon); - rc = CIFSTCon(0, tcon->ses, tcon->treeName, tcon, - nls_codepage); + rc = CIFSTCon(0, tcon->ses, tcon->treeName, tcon + , nls_codepage); up(&tcon->ses->sesSem); if(rc == 0) atomic_inc(&tconInfoReconnectCount); cFYI(1, ("reconnect tcon rc = %d", rc)); /* Removed call to reopen open files here - - it is safer (and faster) to reopen files - one at a time as needed in read and write */ + it is safer (and faster) to reopen files + one at a time as needed in read and write */ /* Check if handle based operation so we - know whether we can continue or not without - returning to caller to reset file handle */ + know whether we can continue or not without + returning to caller to reset file handle */ switch(smb_command) { case SMB_COM_READ_ANDX: case SMB_COM_WRITE_ANDX: @@ -182,22 +186,25 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, check for tcp and smb session status done differently for those three - in the calling routine */ if(tcon) { - if((tcon->ses) && (tcon->ses->server)){ + if((tcon->ses) && (tcon->ses->status != CifsExiting) && + (tcon->ses->server)){ struct nls_table *nls_codepage; - /* Give Demultiplex thread up to 10 seconds to - reconnect, should be greater than cifs socket - timeout which is 7 seconds */ + /* Give Demultiplex thread up to 10 seconds to + reconnect, should be greater than cifs socket + timeout which is 7 seconds */ while(tcon->ses->server->tcpStatus == CifsNeedReconnect) { wait_event_interruptible_timeout(tcon->ses->server->response_q, (tcon->ses->server->tcpStatus == CifsGood), 10 * HZ); - if(tcon->ses->server->tcpStatus == CifsNeedReconnect) { + if(tcon->ses->server->tcpStatus == + CifsNeedReconnect) { /* on "soft" mounts we wait once */ if((tcon->retry == FALSE) || (tcon->ses->status == CifsExiting)) { cFYI(1,("gave up waiting on reconnect in smb_init")); return -EHOSTDOWN; - } /* else "hard" mount - keep retrying until - process is killed or server comes back up */ + } /* else "hard" mount - keep retrying + until process is killed or server + comes on-line */ } else /* TCP session is reestablished now */ break; @@ -208,23 +215,24 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, simultaneously reconnect the same SMB session */ down(&tcon->ses->sesSem); if(tcon->ses->status == CifsNeedReconnect) - rc = cifs_setup_session(0, tcon->ses, nls_codepage); + rc = cifs_setup_session(0, tcon->ses, + nls_codepage); if(!rc && (tcon->tidStatus == CifsNeedReconnect)) { mark_open_files_invalid(tcon); - rc = CIFSTCon(0, tcon->ses, tcon->treeName, tcon, - nls_codepage); + rc = CIFSTCon(0, tcon->ses, tcon->treeName, + tcon, nls_codepage); up(&tcon->ses->sesSem); if(rc == 0) atomic_inc(&tconInfoReconnectCount); cFYI(1, ("reconnect tcon rc = %d", rc)); /* Removed call to reopen open files here - - it is safer (and faster) to reopen files - one at a time as needed in read and write */ + it is safer (and faster) to reopen files + one at a time as needed in read and write */ /* Check if handle based operation so we - know whether we can continue or not without - returning to caller to reset file handle */ + know whether we can continue or not without + returning to caller to reset file handle */ switch(smb_command) { case SMB_COM_READ_ANDX: case SMB_COM_WRITE_ANDX: @@ -286,7 +294,8 @@ static int validate_t2(struct smb_t2_rsp * pSMB) if(total_size < 512) { total_size+=le16_to_cpu(pSMB->t2_rsp.DataCount); /* BCC le converted in SendReceive */ - pBCC = (pSMB->hdr.WordCount * 2) + sizeof(struct smb_hdr) + + pBCC = (pSMB->hdr.WordCount * 2) + + sizeof(struct smb_hdr) + (char *)pSMB; if((total_size <= (*(u16 *)pBCC)) && (total_size < @@ -337,8 +346,10 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) (struct smb_hdr *) pSMBr, &bytes_returned, 0); if (rc == 0) { server->secMode = pSMBr->SecurityMode; - server->secType = NTLM; /* BB override default for NTLMv2 or krb*/ - /* one byte - no need to convert this or EncryptionKeyLen from le,*/ + server->secType = NTLM; /* BB override default for + NTLMv2 or kerberos v5 */ + /* one byte - no need to convert this or EncryptionKeyLen + from little endian */ server->maxReq = le16_to_cpu(pSMBr->MaxMpxCount); /* probably no need to store and check maxvcs */ server->maxBuf = @@ -374,7 +385,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) pSMBr->u.extended_response. GUID, 16) != 0) { cFYI(1, - ("UID of server does not match previous connection to same ip address")); + ("UID of server does not match previous connection to same ip address")); memcpy(server-> server_GUID, pSMBr->u. @@ -454,17 +465,18 @@ CIFSSMBTDis(const int xid, struct cifsTconInfo *tcon) up(&tcon->tconSem); return -EIO; } - rc = small_smb_init(SMB_COM_TREE_DISCONNECT, 0, tcon, (void **)&smb_buffer); + rc = small_smb_init(SMB_COM_TREE_DISCONNECT, 0, tcon, + (void **)&smb_buffer); if (rc) { up(&tcon->tconSem); return rc; } else { smb_buffer_response = smb_buffer; /* BB removeme BB */ - } + } rc = SendReceive(xid, tcon->ses, smb_buffer, smb_buffer_response, &length, 0); if (rc) - cFYI(1, (" Tree disconnect failed %d", rc)); + cFYI(1, ("Tree disconnect failed %d", rc)); if (smb_buffer) cifs_small_buf_release(smb_buffer); @@ -538,8 +550,8 @@ CIFSSMBLogoff(const int xid, struct cifsSesInfo *ses) } int -CIFSSMBDelFile(const int xid, struct cifsTconInfo *tcon, - const char *fileName, const struct nls_table *nls_codepage) +CIFSSMBDelFile(const int xid, struct cifsTconInfo *tcon, const char *fileName, + const struct nls_table *nls_codepage, int remap) { DELETE_FILE_REQ *pSMB = NULL; DELETE_FILE_RSP *pSMBr = NULL; @@ -555,12 +567,11 @@ DelFileRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->fileName, fileName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->fileName, fileName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; - } else { /* BB improve the check for buffer overruns BB */ + } else { /* BB improve check for buffer overruns BB */ name_len = strnlen(fileName, PATH_MAX); name_len++; /* trailing null */ strncpy(pSMB->fileName, fileName, name_len); @@ -589,8 +600,8 @@ DelFileRetry: } int -CIFSSMBRmDir(const int xid, struct cifsTconInfo *tcon, - const char *dirName, const struct nls_table *nls_codepage) +CIFSSMBRmDir(const int xid, struct cifsTconInfo *tcon, const char *dirName, + const struct nls_table *nls_codepage, int remap) { DELETE_DIRECTORY_REQ *pSMB = NULL; DELETE_DIRECTORY_RSP *pSMBr = NULL; @@ -606,12 +617,11 @@ RmDirRetry: return rc; if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { - name_len = cifs_strtoUCS((wchar_t *) pSMB->DirName, dirName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + name_len = cifsConvertToUCS((__le16 *) pSMB->DirName, dirName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; - } else { /* BB improve the check for buffer overruns BB */ + } else { /* BB improve check for buffer overruns BB */ name_len = strnlen(dirName, PATH_MAX); name_len++; /* trailing null */ strncpy(pSMB->DirName, dirName, name_len); @@ -639,7 +649,7 @@ RmDirRetry: int CIFSSMBMkDir(const int xid, struct cifsTconInfo *tcon, - const char *name, const struct nls_table *nls_codepage) + const char *name, const struct nls_table *nls_codepage, int remap) { int rc = 0; CREATE_DIRECTORY_REQ *pSMB = NULL; @@ -655,12 +665,11 @@ MkDirRetry: return rc; if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { - name_len = cifs_strtoUCS((wchar_t *) pSMB->DirName, name, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + name_len = cifsConvertToUCS((__u16 *) pSMB->DirName, name, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; - } else { /* BB improve the check for buffer overruns BB */ + } else { /* BB improve check for buffer overruns BB */ name_len = strnlen(name, PATH_MAX); name_len++; /* trailing null */ strncpy(pSMB->DirName, name, name_len); @@ -690,7 +699,7 @@ CIFSSMBOpen(const int xid, struct cifsTconInfo *tcon, const char *fileName, const int openDisposition, const int access_flags, const int create_options, __u16 * netfid, int *pOplock, FILE_ALL_INFO * pfile_info, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { int rc = -EACCES; OPEN_REQ *pSMB = NULL; @@ -710,14 +719,12 @@ openRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { count = 1; /* account for one byte pad to word boundary */ name_len = - cifs_strtoUCS((wchar_t *) (pSMB->fileName + 1), - fileName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) (pSMB->fileName + 1), + fileName, PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; pSMB->NameLength = cpu_to_le16(name_len); - } else { /* BB improve the check for buffer overruns BB */ + } else { /* BB improve check for buffer overruns BB */ count = 0; /* no pad */ name_len = strnlen(fileName, PATH_MAX); name_len++; /* trailing null */ @@ -746,7 +753,8 @@ openRetry: pSMB->ShareAccess = cpu_to_le32(FILE_SHARE_ALL); pSMB->CreateDisposition = cpu_to_le32(openDisposition); pSMB->CreateOptions = cpu_to_le32(create_options); - pSMB->ImpersonationLevel = cpu_to_le32(SECURITY_IMPERSONATION); /* BB ??*/ + /* BB Expirement with various impersonation levels and verify */ + pSMB->ImpersonationLevel = cpu_to_le32(SECURITY_IMPERSONATION); pSMB->SecurityFlags = SECURITY_CONTEXT_TRACKING | SECURITY_EFFECTIVE_ONLY; @@ -760,7 +768,7 @@ openRetry: if (rc) { cFYI(1, ("Error in Open = %d", rc)); } else { - *pOplock = pSMBr->OplockLevel; /* one byte no need to le_to_cpu */ + *pOplock = pSMBr->OplockLevel; /* 1 byte no need to le_to_cpu */ *netfid = pSMBr->Fid; /* cifs fid stays in le */ /* Let caller know file was created so we can set the mode. */ /* Do we care about the CreateAction in any other cases? */ @@ -1017,11 +1025,13 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, __u16 count; cFYI(1, ("In CIFSSMBLock - timeout %d numLock %d",waitFlag,numLock)); - rc = smb_init(SMB_COM_LOCKING_ANDX, 8, tcon, (void **) &pSMB, - (void **) &pSMBr); + rc = small_smb_init(SMB_COM_LOCKING_ANDX, 8, tcon, (void **) &pSMB); + if (rc) return rc; + pSMBr = (LOCK_RSP *)pSMB; /* BB removeme BB */ + if(lockType == LOCKING_ANDX_OPLOCK_RELEASE) { timeout = -1; /* no response expected */ pSMB->Timeout = 0; @@ -1059,7 +1069,7 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, if (rc) { cFYI(1, ("Send error in Lock = %d", rc)); } - cifs_buf_release(pSMB); + cifs_small_buf_release(pSMB); /* Note: On -EAGAIN error only caller can retry on handle based calls since file handle passed in no longer valid */ @@ -1108,7 +1118,7 @@ CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, int smb_file_id) int CIFSSMBRename(const int xid, struct cifsTconInfo *tcon, const char *fromName, const char *toName, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { int rc = 0; RENAME_REQ *pSMB = NULL; @@ -1131,18 +1141,16 @@ renameRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->OldFileName, fromName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->OldFileName, fromName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; pSMB->OldFileName[name_len] = 0x04; /* pad */ /* protocol requires ASCII signature byte on Unicode string */ pSMB->OldFileName[name_len + 1] = 0x00; name_len2 = - cifs_strtoUCS((wchar_t *) & pSMB-> - OldFileName[name_len + 2], toName, PATH_MAX, - nls_codepage); + cifsConvertToUCS((__u16 *) &pSMB->OldFileName[name_len + 2], + toName, PATH_MAX, nls_codepage, remap); name_len2 += 1 /* trailing null */ + 1 /* Signature word */ ; name_len2 *= 2; /* convert to bytes */ } else { /* BB improve the check for buffer overruns BB */ @@ -1182,7 +1190,8 @@ renameRetry: } int CIFSSMBRenameOpenFile(const int xid,struct cifsTconInfo *pTcon, - int netfid, char * target_name, const struct nls_table * nls_codepage) + int netfid, char * target_name, + const struct nls_table * nls_codepage, int remap) { struct smb_com_transaction2_sfi_req *pSMB = NULL; struct smb_com_transaction2_sfi_rsp *pSMBr = NULL; @@ -1227,9 +1236,11 @@ int CIFSSMBRenameOpenFile(const int xid,struct cifsTconInfo *pTcon, /* unicode only call */ if(target_name == NULL) { sprintf(dummy_string,"cifs%x",pSMB->hdr.Mid); - len_of_str = cifs_strtoUCS((wchar_t *) rename_info->target_name, dummy_string, 24, nls_codepage); + len_of_str = cifsConvertToUCS((__u16 *)rename_info->target_name, + dummy_string, 24, nls_codepage, remap); } else { - len_of_str = cifs_strtoUCS((wchar_t *) rename_info->target_name, target_name, PATH_MAX, nls_codepage); + len_of_str = cifsConvertToUCS((__u16 *)rename_info->target_name, + target_name, PATH_MAX, nls_codepage, remap); } rename_info->target_name_len = cpu_to_le32(2 * len_of_str); count = 12 /* sizeof(struct set_file_rename) */ + (2 * len_of_str) + 2; @@ -1263,7 +1274,7 @@ int CIFSSMBRenameOpenFile(const int xid,struct cifsTconInfo *pTcon, int CIFSSMBCopy(const int xid, struct cifsTconInfo *tcon, const char * fromName, const __u16 target_tid, const char *toName, const int flags, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { int rc = 0; COPY_REQ *pSMB = NULL; @@ -1285,18 +1296,16 @@ copyRetry: pSMB->Flags = cpu_to_le16(flags & COPY_TREE); if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { - name_len = cifs_strtoUCS((wchar_t *) pSMB->OldFileName, - fromName, - PATH_MAX /* find define for this maxpathcomponent */, - nls_codepage); + name_len = cifsConvertToUCS((__u16 *) pSMB->OldFileName, + fromName, PATH_MAX, nls_codepage, + remap); name_len++; /* trailing null */ name_len *= 2; pSMB->OldFileName[name_len] = 0x04; /* pad */ /* protocol requires ASCII signature byte on Unicode string */ pSMB->OldFileName[name_len + 1] = 0x00; - name_len2 = cifs_strtoUCS((wchar_t *) & pSMB-> - OldFileName[name_len + 2], toName, PATH_MAX, - nls_codepage); + name_len2 = cifsConvertToUCS((__u16 *)&pSMB->OldFileName[name_len + 2], + toName, PATH_MAX, nls_codepage, remap); name_len2 += 1 /* trailing null */ + 1 /* Signature word */ ; name_len2 *= 2; /* convert to bytes */ } else { /* BB improve the check for buffer overruns BB */ @@ -1425,7 +1434,7 @@ createSymLinkRetry: int CIFSUnixCreateHardLink(const int xid, struct cifsTconInfo *tcon, const char *fromName, const char *toName, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { TRANSACTION2_SPI_REQ *pSMB = NULL; TRANSACTION2_SPI_RSP *pSMBr = NULL; @@ -1444,9 +1453,8 @@ createHardLinkRetry: return rc; if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { - name_len = cifs_strtoUCS((wchar_t *) pSMB->FileName, toName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + name_len = cifsConvertToUCS((__u16 *) pSMB->FileName, toName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; @@ -1468,9 +1476,8 @@ createHardLinkRetry: data_offset = (char *) (&pSMB->hdr.Protocol) + offset; if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len_target = - cifs_strtoUCS((wchar_t *) data_offset, fromName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) data_offset, fromName, PATH_MAX, + nls_codepage, remap); name_len_target++; /* trailing null */ name_len_target *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -1512,7 +1519,7 @@ createHardLinkRetry: int CIFSCreateHardLink(const int xid, struct cifsTconInfo *tcon, const char *fromName, const char *toName, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { int rc = 0; NT_RENAME_REQ *pSMB = NULL; @@ -1539,17 +1546,15 @@ winCreateHardLinkRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->OldFileName, fromName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->OldFileName, fromName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; pSMB->OldFileName[name_len] = 0; /* pad */ pSMB->OldFileName[name_len + 1] = 0x04; name_len2 = - cifs_strtoUCS((wchar_t *) & pSMB-> - OldFileName[name_len + 2], toName, PATH_MAX, - nls_codepage); + cifsConvertToUCS((__u16 *)&pSMB->OldFileName[name_len + 2], + toName, PATH_MAX, nls_codepage, remap); name_len2 += 1 /* trailing null */ + 1 /* Signature word */ ; name_len2 *= 2; /* convert to bytes */ } else { /* BB improve the check for buffer overruns BB */ @@ -1659,6 +1664,7 @@ querySymLinkRetry: name_len = UniStrnlen((wchar_t *) ((char *) &pSMBr->hdr.Protocol +data_offset), min_t(const int, buflen,count) / 2); + /* BB FIXME investigate remapping reserved chars here */ cifs_strfromUCS_le(symlinkinfo, (wchar_t *) ((char *)&pSMBr->hdr.Protocol + data_offset), @@ -1793,7 +1799,8 @@ static void cifs_convert_ace(posix_acl_xattr_entry * ace, struct cifs_posix_ace } /* Convert ACL from CIFS POSIX wire format to local Linux POSIX ACL xattr */ -static int cifs_copy_posix_acl(char * trgt,char * src, const int buflen,const int acl_type,const int size_of_data_area) +static int cifs_copy_posix_acl(char * trgt,char * src, const int buflen, + const int acl_type,const int size_of_data_area) { int size = 0; int i; @@ -1912,7 +1919,7 @@ int CIFSSMBGetPosixACL(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, char *acl_inf, const int buflen, const int acl_type, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { /* SMB_QUERY_POSIX_ACL */ TRANSACTION2_QPI_REQ *pSMB = NULL; @@ -1932,8 +1939,8 @@ queryAclRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, searchName, PATH_MAX - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, searchName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; pSMB->FileName[name_len] = 0; @@ -1997,8 +2004,9 @@ queryAclRetry: int CIFSSMBSetPosixACL(const int xid, struct cifsTconInfo *tcon, const unsigned char *fileName, - const char *local_acl, const int buflen, const int acl_type, - const struct nls_table *nls_codepage) + const char *local_acl, const int buflen, + const int acl_type, + const struct nls_table *nls_codepage, int remap) { struct smb_com_transaction2_spi_req *pSMB = NULL; struct smb_com_transaction2_spi_rsp *pSMBr = NULL; @@ -2016,8 +2024,8 @@ setAclRetry: return rc; if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, fileName, PATH_MAX - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, fileName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -2072,13 +2080,96 @@ setACLerrorExit: return rc; } -#endif +/* BB fix tabs in this function FIXME BB */ +int +CIFSGetExtAttr(const int xid, struct cifsTconInfo *tcon, + const int netfid, __u64 * pExtAttrBits, __u64 *pMask) +{ + int rc = 0; + struct smb_t2_qfi_req *pSMB = NULL; + struct smb_t2_qfi_rsp *pSMBr = NULL; + int bytes_returned; + __u16 params, byte_count; + + cFYI(1,("In GetExtAttr")); + if(tcon == NULL) + return -ENODEV; + +GetExtAttrRetry: + rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB, + (void **) &pSMBr); + if (rc) + return rc; + + params = 2 /* level */ +2 /* fid */; + pSMB->t2.TotalDataCount = 0; + pSMB->t2.MaxParameterCount = cpu_to_le16(4); + /* BB find exact max data count below from sess structure BB */ + pSMB->t2.MaxDataCount = cpu_to_le16(4000); + pSMB->t2.MaxSetupCount = 0; + pSMB->t2.Reserved = 0; + pSMB->t2.Flags = 0; + pSMB->t2.Timeout = 0; + pSMB->t2.Reserved2 = 0; + pSMB->t2.ParameterOffset = cpu_to_le16(offsetof(struct smb_t2_qfi_req, + Fid) - 4); + pSMB->t2.DataCount = 0; + pSMB->t2.DataOffset = 0; + pSMB->t2.SetupCount = 1; + pSMB->t2.Reserved3 = 0; + pSMB->t2.SubCommand = cpu_to_le16(TRANS2_QUERY_FILE_INFORMATION); + byte_count = params + 1 /* pad */ ; + pSMB->t2.TotalParameterCount = cpu_to_le16(params); + pSMB->t2.ParameterCount = pSMB->t2.TotalParameterCount; + pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_ATTR_FLAGS); + pSMB->Pad = 0; + pSMB->Fid = netfid; + pSMB->hdr.smb_buf_length += byte_count; + pSMB->t2.ByteCount = cpu_to_le16(byte_count); + + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned, 0); + if (rc) { + cFYI(1, ("error %d in GetExtAttr", rc)); + } else { + /* decode response */ + rc = validate_t2((struct smb_t2_rsp *)pSMBr); + if (rc || (pSMBr->ByteCount < 2)) + /* BB also check enough total bytes returned */ + /* If rc should we check for EOPNOSUPP and + disable the srvino flag? or in caller? */ + rc = -EIO; /* bad smb */ + else { + __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); + __u16 count = le16_to_cpu(pSMBr->t2.DataCount); + struct file_chattr_info * pfinfo; + /* BB Do we need a cast or hash here ? */ + if(count != 16) { + cFYI(1, ("Illegal size ret in GetExtAttr")); + rc = -EIO; + goto GetExtAttrOut; + } + pfinfo = (struct file_chattr_info *) + (data_offset + (char *) &pSMBr->hdr.Protocol); + *pExtAttrBits = le64_to_cpu(pfinfo->mode); + *pMask = le64_to_cpu(pfinfo->mask); + } + } +GetExtAttrOut: + cifs_buf_release(pSMB); + if (rc == -EAGAIN) + goto GetExtAttrRetry; + return rc; +} + + +#endif /* CONFIG_POSIX */ int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_ALL_INFO * pFindData, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { /* level 263 SMB_QUERY_FILE_ALL_INFO */ TRANSACTION2_QPI_REQ *pSMB = NULL; @@ -2097,9 +2188,8 @@ QPathInfoRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, searchName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, searchName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -2160,7 +2250,7 @@ int CIFSSMBUnixQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_UNIX_BASIC_INFO * pFindData, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { /* SMB_QUERY_FILE_UNIX_BASIC */ TRANSACTION2_QPI_REQ *pSMB = NULL; @@ -2179,9 +2269,8 @@ UnixQPathInfoRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, searchName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, searchName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -2261,7 +2350,7 @@ findUniqueRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, searchName, PATH_MAX + cifsConvertToUCS((wchar_t *) pSMB->FileName, searchName, PATH_MAX /* find define for this maxpathcomponent */ , nls_codepage); name_len++; /* trailing null */ @@ -2325,7 +2414,7 @@ CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, const char *searchName, const struct nls_table *nls_codepage, __u16 * pnetfid, - struct cifs_search_info * psrch_inf) + struct cifs_search_info * psrch_inf, int remap) { /* level 257 SMB_ */ TRANSACTION2_FFIRST_REQ *pSMB = NULL; @@ -2336,7 +2425,7 @@ CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, int name_len; __u16 params, byte_count; - cFYI(1, ("In FindFirst")); + cFYI(1, ("In FindFirst for %s",searchName)); findFirstRetry: rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB, @@ -2346,20 +2435,30 @@ findFirstRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName,searchName, - PATH_MAX, nls_codepage); - name_len++; /* trailing null */ + cifsConvertToUCS((__u16 *) pSMB->FileName,searchName, + PATH_MAX, nls_codepage, remap); + /* We can not add the asterik earlier in case + it got remapped to 0xF03A as if it were part of the + directory name instead of a wildcard */ name_len *= 2; + pSMB->FileName[name_len] = '\\'; + pSMB->FileName[name_len+1] = 0; + pSMB->FileName[name_len+2] = '*'; + pSMB->FileName[name_len+3] = 0; + name_len += 4; /* now the trailing null */ pSMB->FileName[name_len] = 0; /* null terminate just in case */ pSMB->FileName[name_len+1] = 0; + name_len += 2; } else { /* BB add check for overrun of SMB buf BB */ name_len = strnlen(searchName, PATH_MAX); - name_len++; /* trailing null */ /* BB fix here and in unicode clause above ie if(name_len > buffersize-header) free buffer exit; BB */ strncpy(pSMB->FileName, searchName, name_len); - pSMB->FileName[name_len] = 0; /* just in case */ + pSMB->FileName[name_len] = '\\'; + pSMB->FileName[name_len+1] = '*'; + pSMB->FileName[name_len+2] = 0; + name_len += 3; } params = 12 + name_len /* includes null */ ; @@ -2422,7 +2521,6 @@ findFirstRetry: psrch_inf->srch_entries_start = (char *) &pSMBr->hdr.Protocol + le16_to_cpu(pSMBr->t2.DataOffset); - parms = (T2_FFIRST_RSP_PARMS *)((char *) &pSMBr->hdr.Protocol + le16_to_cpu(pSMBr->t2.ParameterOffset)); @@ -2434,7 +2532,6 @@ findFirstRetry: psrch_inf->entries_in_buffer = le16_to_cpu(parms->SearchCount); psrch_inf->index_of_last_entry = psrch_inf->entries_in_buffer; -/*cFYI(1,("entries in buf %d index_of_last %d",psrch_inf->entries_in_buffer,psrch_inf->index_of_last_entry)); */ *pnetfid = parms->SearchHandle; } else { cifs_buf_release(pSMB); @@ -2608,7 +2705,7 @@ int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, __u64 * inode_number, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { int rc = 0; TRANSACTION2_QPI_REQ *pSMB = NULL; @@ -2629,8 +2726,8 @@ GetInodeNumberRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, searchName, - PATH_MAX,nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, searchName, + PATH_MAX,nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -2704,7 +2801,7 @@ CIFSGetDFSRefer(const int xid, struct cifsSesInfo *ses, const unsigned char *searchName, unsigned char **targetUNCs, unsigned int *number_of_UNC_in_array, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { /* TRANS2_GET_DFS_REFERRAL */ TRANSACTION2_GET_DFS_REFER_REQ *pSMB = NULL; @@ -2740,10 +2837,8 @@ getDFSRetry: if (ses->capabilities & CAP_UNICODE) { pSMB->hdr.Flags2 |= SMBFLG2_UNICODE; name_len = - cifs_strtoUCS((wchar_t *) pSMB->RequestFileName, - searchName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->RequestFileName, + searchName, PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -2871,8 +2966,7 @@ GetDFSRefExit: } int -CIFSSMBQFSInfo(const int xid, struct cifsTconInfo *tcon, - struct kstatfs *FSData, const struct nls_table *nls_codepage) +CIFSSMBQFSInfo(const int xid, struct cifsTconInfo *tcon, struct kstatfs *FSData) { /* level 0x103 SMB_QUERY_FILE_SYSTEM_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -2955,8 +3049,7 @@ QFSInfoRetry: } int -CIFSSMBQFSAttributeInfo(const int xid, struct cifsTconInfo *tcon, - const struct nls_table *nls_codepage) +CIFSSMBQFSAttributeInfo(const int xid, struct cifsTconInfo *tcon) { /* level 0x105 SMB_QUERY_FILE_SYSTEM_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -3024,8 +3117,7 @@ QFSAttributeRetry: } int -CIFSSMBQFSDeviceInfo(const int xid, struct cifsTconInfo *tcon, - const struct nls_table *nls_codepage) +CIFSSMBQFSDeviceInfo(const int xid, struct cifsTconInfo *tcon) { /* level 0x104 SMB_QUERY_FILE_SYSTEM_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -3078,8 +3170,8 @@ QFSDeviceRetry: else { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); response_data = - (FILE_SYSTEM_DEVICE_INFO - *) (((char *) &pSMBr->hdr.Protocol) + + (FILE_SYSTEM_DEVICE_INFO *) + (((char *) &pSMBr->hdr.Protocol) + data_offset); memcpy(&tcon->fsDevInfo, response_data, sizeof (FILE_SYSTEM_DEVICE_INFO)); @@ -3094,8 +3186,7 @@ QFSDeviceRetry: } int -CIFSSMBQFSUnixInfo(const int xid, struct cifsTconInfo *tcon, - const struct nls_table *nls_codepage) +CIFSSMBQFSUnixInfo(const int xid, struct cifsTconInfo *tcon) { /* level 0x200 SMB_QUERY_CIFS_UNIX_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -3166,7 +3257,7 @@ QFSUnixRetry: int CIFSSMBQFSPosixInfo(const int xid, struct cifsTconInfo *tcon, - struct kstatfs *FSData, const struct nls_table *nls_codepage) + struct kstatfs *FSData) { /* level 0x201 SMB_QUERY_CIFS_POSIX_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -3258,7 +3349,8 @@ QFSPosixRetry: int CIFSSMBSetEOF(const int xid, struct cifsTconInfo *tcon, const char *fileName, - __u64 size, int SetAllocation, const struct nls_table *nls_codepage) + __u64 size, int SetAllocation, + const struct nls_table *nls_codepage, int remap) { struct smb_com_transaction2_spi_req *pSMB = NULL; struct smb_com_transaction2_spi_rsp *pSMBr = NULL; @@ -3277,9 +3369,8 @@ SetEOFRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, fileName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, fileName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -3360,11 +3451,13 @@ CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, __u64 size, cFYI(1, ("SetFileSize (via SetFileInfo) %lld", (long long)size)); - rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB, - (void **) &pSMBr); + rc = small_smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB); + if (rc) return rc; + pSMBr = (struct smb_com_transaction2_sfi_rsp *)pSMB; + pSMB->hdr.Pid = cpu_to_le16((__u16)pid_of_opener); pSMB->hdr.PidHigh = cpu_to_le16((__u16)(pid_of_opener >> 16)); @@ -3424,7 +3517,7 @@ CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, __u64 size, } if (pSMB) - cifs_buf_release(pSMB); + cifs_small_buf_release(pSMB); /* Note: On -EAGAIN error only caller can retry on handle based calls since file handle passed in no longer valid */ @@ -3450,11 +3543,13 @@ CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, const FILE_BASIC_I __u16 params, param_offset, offset, byte_count, count; cFYI(1, ("Set Times (via SetFileInfo)")); - rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB, - (void **) &pSMBr); + rc = small_smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB); + if (rc) return rc; + pSMBr = (struct smb_com_transaction2_sfi_rsp *)pSMB; + /* At this point there is no need to override the current pid with the pid of the opener, but that could change if we someday use an existing handle (rather than opening one on the fly) */ @@ -3500,7 +3595,7 @@ CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, const FILE_BASIC_I cFYI(1,("Send error in Set Time (SetFileInfo) = %d",rc)); } - cifs_buf_release(pSMB); + cifs_small_buf_release(pSMB); /* Note: On -EAGAIN error only caller can retry on handle based calls since file handle passed in no longer valid */ @@ -3512,7 +3607,7 @@ CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, const FILE_BASIC_I int CIFSSMBSetTimes(const int xid, struct cifsTconInfo *tcon, const char *fileName, const FILE_BASIC_INFO * data, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { TRANSACTION2_SPI_REQ *pSMB = NULL; TRANSACTION2_SPI_RSP *pSMBr = NULL; @@ -3532,9 +3627,8 @@ SetTimesRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, fileName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, fileName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -3614,7 +3708,7 @@ SetAttrLgcyRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->fileName, fileName, + ConvertToUCS((wchar_t *) pSMB->fileName, fileName, PATH_MAX, nls_codepage); name_len++; /* trailing null */ name_len *= 2; @@ -3644,8 +3738,9 @@ SetAttrLgcyRetry: int CIFSSMBUnixSetPerms(const int xid, struct cifsTconInfo *tcon, - char *fileName, __u64 mode, __u64 uid, __u64 gid, - dev_t device, const struct nls_table *nls_codepage) + char *fileName, __u64 mode, __u64 uid, __u64 gid, + dev_t device, const struct nls_table *nls_codepage, + int remap) { TRANSACTION2_SPI_REQ *pSMB = NULL; TRANSACTION2_SPI_RSP *pSMBr = NULL; @@ -3664,9 +3759,8 @@ setPermsRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, fileName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, fileName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -3789,7 +3883,7 @@ ssize_t CIFSSMBQAllEAs(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, char * EAData, size_t buf_size, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { /* BB assumes one setup word */ TRANSACTION2_QPI_REQ *pSMB = NULL; @@ -3810,9 +3904,8 @@ QAllEAsRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, searchName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((wchar_t *) pSMB->FileName, searchName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -3934,7 +4027,7 @@ QAllEAsRetry: ssize_t CIFSSMBQueryEA(const int xid,struct cifsTconInfo * tcon, const unsigned char * searchName,const unsigned char * ea_name, unsigned char * ea_value, size_t buf_size, - const struct nls_table *nls_codepage) + const struct nls_table *nls_codepage, int remap) { TRANSACTION2_QPI_REQ *pSMB = NULL; TRANSACTION2_QPI_RSP *pSMBr = NULL; @@ -3954,9 +4047,8 @@ QEARetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, searchName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, searchName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ @@ -4082,7 +4174,8 @@ QEARetry: int CIFSSMBSetEA(const int xid, struct cifsTconInfo *tcon, const char *fileName, const char * ea_name, const void * ea_value, - const __u16 ea_value_len, const struct nls_table *nls_codepage) + const __u16 ea_value_len, const struct nls_table *nls_codepage, + int remap) { struct smb_com_transaction2_spi_req *pSMB = NULL; struct smb_com_transaction2_spi_rsp *pSMBr = NULL; @@ -4101,9 +4194,8 @@ SetEARetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, fileName, PATH_MAX - /* find define for this maxpathcomponent */ - , nls_codepage); + cifsConvertToUCS((__u16 *) pSMB->FileName, fileName, + PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; } else { /* BB improve the check for buffer overruns BB */ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 40470b9d547..e568cc47a7f 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1,7 +1,7 @@ /* * fs/cifs/connect.c * - * Copyright (C) International Business Machines Corp., 2002,2004 + * Copyright (C) International Business Machines Corp., 2002,2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -28,6 +28,7 @@ #include <linux/ctype.h> #include <linux/utsname.h> #include <linux/mempool.h> +#include <linux/delay.h> #include <asm/uaccess.h> #include <asm/processor.h> #include "cifspdu.h" @@ -72,6 +73,7 @@ struct smb_vol { unsigned no_xattr:1; /* set if xattr (EA) support should be disabled*/ unsigned server_ino:1; /* use inode numbers from server ie UniqueId */ unsigned direct_io:1; + unsigned remap:1; /* set to remap seven reserved chars in filenames */ unsigned int rsize; unsigned int wsize; unsigned int sockopt; @@ -114,7 +116,7 @@ cifs_reconnect(struct TCP_Server_Info *server) spin_unlock(&GlobalMid_Lock); server->maxBuf = 0; - cFYI(1, ("Reconnecting tcp session ")); + cFYI(1, ("Reconnecting tcp session")); /* before reconnecting the tcp session, mark the smb session (uid) and the tid bad so they are not used until reconnected */ @@ -155,9 +157,10 @@ cifs_reconnect(struct TCP_Server_Info *server) qhead); if(mid_entry) { if(mid_entry->midState == MID_REQUEST_SUBMITTED) { - /* Mark other intransit requests as needing retry so - we do not immediately mark the session bad again - (ie after we reconnect below) as they timeout too */ + /* Mark other intransit requests as needing + retry so we do not immediately mark the + session bad again (ie after we reconnect + below) as they timeout too */ mid_entry->midState = MID_RETRY_NEEDED; } } @@ -175,14 +178,14 @@ cifs_reconnect(struct TCP_Server_Info *server) server->workstation_RFC1001_name); } if(rc) { - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(3 * HZ); + msleep(3000); } else { atomic_inc(&tcpSesReconnectCount); spin_lock(&GlobalMid_Lock); if(server->tcpStatus != CifsExiting) server->tcpStatus = CifsGood; - spin_unlock(&GlobalMid_Lock); + server->sequence_number = 0; + spin_unlock(&GlobalMid_Lock); /* atomic_set(&server->inFlight,0);*/ wake_up(&server->response_q); } @@ -190,12 +193,129 @@ cifs_reconnect(struct TCP_Server_Info *server) return rc; } +/* + return codes: + 0 not a transact2, or all data present + >0 transact2 with that much data missing + -EINVAL = invalid transact2 + + */ +static int check2ndT2(struct smb_hdr * pSMB, unsigned int maxBufSize) +{ + struct smb_t2_rsp * pSMBt; + int total_data_size; + int data_in_this_rsp; + int remaining; + + if(pSMB->Command != SMB_COM_TRANSACTION2) + return 0; + + /* check for plausible wct, bcc and t2 data and parm sizes */ + /* check for parm and data offset going beyond end of smb */ + if(pSMB->WordCount != 10) { /* coalesce_t2 depends on this */ + cFYI(1,("invalid transact2 word count")); + return -EINVAL; + } + + pSMBt = (struct smb_t2_rsp *)pSMB; + + total_data_size = le16_to_cpu(pSMBt->t2_rsp.TotalDataCount); + data_in_this_rsp = le16_to_cpu(pSMBt->t2_rsp.DataCount); + + remaining = total_data_size - data_in_this_rsp; + + if(remaining == 0) + return 0; + else if(remaining < 0) { + cFYI(1,("total data %d smaller than data in frame %d", + total_data_size, data_in_this_rsp)); + return -EINVAL; + } else { + cFYI(1,("missing %d bytes from transact2, check next response", + remaining)); + if(total_data_size > maxBufSize) { + cERROR(1,("TotalDataSize %d is over maximum buffer %d", + total_data_size,maxBufSize)); + return -EINVAL; + } + return remaining; + } +} + +static int coalesce_t2(struct smb_hdr * psecond, struct smb_hdr *pTargetSMB) +{ + struct smb_t2_rsp *pSMB2 = (struct smb_t2_rsp *)psecond; + struct smb_t2_rsp *pSMBt = (struct smb_t2_rsp *)pTargetSMB; + int total_data_size; + int total_in_buf; + int remaining; + int total_in_buf2; + char * data_area_of_target; + char * data_area_of_buf2; + __u16 byte_count; + + total_data_size = le16_to_cpu(pSMBt->t2_rsp.TotalDataCount); + + if(total_data_size != le16_to_cpu(pSMB2->t2_rsp.TotalDataCount)) { + cFYI(1,("total data sizes of primary and secondary t2 differ")); + } + + total_in_buf = le16_to_cpu(pSMBt->t2_rsp.DataCount); + + remaining = total_data_size - total_in_buf; + + if(remaining < 0) + return -EINVAL; + + if(remaining == 0) /* nothing to do, ignore */ + return 0; + + total_in_buf2 = le16_to_cpu(pSMB2->t2_rsp.DataCount); + if(remaining < total_in_buf2) { + cFYI(1,("transact2 2nd response contains too much data")); + } + + /* find end of first SMB data area */ + data_area_of_target = (char *)&pSMBt->hdr.Protocol + + le16_to_cpu(pSMBt->t2_rsp.DataOffset); + /* validate target area */ + + data_area_of_buf2 = (char *) &pSMB2->hdr.Protocol + + le16_to_cpu(pSMB2->t2_rsp.DataOffset); + + data_area_of_target += total_in_buf; + + /* copy second buffer into end of first buffer */ + memcpy(data_area_of_target,data_area_of_buf2,total_in_buf2); + total_in_buf += total_in_buf2; + pSMBt->t2_rsp.DataCount = cpu_to_le16(total_in_buf); + byte_count = le16_to_cpu(BCC_LE(pTargetSMB)); + byte_count += total_in_buf2; + BCC_LE(pTargetSMB) = cpu_to_le16(byte_count); + + byte_count = be32_to_cpu(pTargetSMB->smb_buf_length); + byte_count += total_in_buf2; + + /* BB also add check that we are not beyond maximum buffer size */ + + pTargetSMB->smb_buf_length = cpu_to_be32(byte_count); + + if(remaining == total_in_buf2) { + cFYI(1,("found the last secondary response")); + return 0; /* we are done */ + } else /* more responses to go */ + return 1; + +} + static int cifs_demultiplex_thread(struct TCP_Server_Info *server) { int length; unsigned int pdu_length, total_read; struct smb_hdr *smb_buffer = NULL; + struct smb_hdr *bigbuf = NULL; + struct smb_hdr *smallbuf = NULL; struct msghdr smb_msg; struct kvec iov; struct socket *csocket = server->ssocket; @@ -204,6 +324,9 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) struct task_struct *task_to_wake = NULL; struct mid_q_entry *mid_entry; char *temp; + int isLargeBuf = FALSE; + int isMultiRsp; + int reconnect; daemonize("cifsd"); allow_signal(SIGKILL); @@ -221,17 +344,34 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) } while (server->tcpStatus != CifsExiting) { - if (smb_buffer == NULL) - smb_buffer = cifs_buf_get(); - else - memset(smb_buffer, 0, sizeof (struct smb_hdr)); - - if (smb_buffer == NULL) { - cERROR(1,("Can not get memory for SMB response")); - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(HZ * 3); /* give system time to free memory */ - continue; + if (bigbuf == NULL) { + bigbuf = cifs_buf_get(); + if(bigbuf == NULL) { + cERROR(1,("No memory for large SMB response")); + msleep(3000); + /* retry will check if exiting */ + continue; + } + } else if(isLargeBuf) { + /* we are reusing a dirtry large buf, clear its start */ + memset(bigbuf, 0, sizeof (struct smb_hdr)); } + + if (smallbuf == NULL) { + smallbuf = cifs_small_buf_get(); + if(smallbuf == NULL) { + cERROR(1,("No memory for SMB response")); + msleep(1000); + /* retry will check if exiting */ + continue; + } + /* beginning of smb buffer is cleared in our buf_get */ + } else /* if existing small buf clear beginning */ + memset(smallbuf, 0, sizeof (struct smb_hdr)); + + isLargeBuf = FALSE; + isMultiRsp = FALSE; + smb_buffer = smallbuf; iov.iov_base = smb_buffer; iov.iov_len = 4; smb_msg.msg_control = NULL; @@ -243,176 +383,257 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) if(server->tcpStatus == CifsExiting) { break; } else if (server->tcpStatus == CifsNeedReconnect) { - cFYI(1,("Reconnecting after server stopped responding")); + cFYI(1,("Reconnect after server stopped responding")); cifs_reconnect(server); cFYI(1,("call to reconnect done")); csocket = server->ssocket; continue; } else if ((length == -ERESTARTSYS) || (length == -EAGAIN)) { - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(1); /* minimum sleep to prevent looping + msleep(1); /* minimum sleep to prevent looping allowing socket to clear and app threads to set tcpStatus CifsNeedReconnect if server hung */ continue; } else if (length <= 0) { if(server->tcpStatus == CifsNew) { - cFYI(1,("tcp session abended prematurely (after SMBnegprot)")); - /* some servers kill tcp session rather than returning - smb negprot error in which case reconnecting here is - not going to help - return error to mount */ + cFYI(1,("tcp session abend after SMBnegprot")); + /* some servers kill the TCP session rather than + returning an SMB negprot error, in which + case reconnecting here is not going to help, + and so simply return error to mount */ break; } if(length == -EINTR) { cFYI(1,("cifsd thread killed")); break; } - cFYI(1,("Reconnecting after unexpected peek error %d",length)); + cFYI(1,("Reconnect after unexpected peek error %d", + length)); cifs_reconnect(server); csocket = server->ssocket; wake_up(&server->response_q); continue; - } else if (length > 3) { - pdu_length = ntohl(smb_buffer->smb_buf_length); - /* Only read pdu_length after below checks for too short (due - to e.g. int overflow) and too long ie beyond end of buf */ - cFYI(1,("rfc1002 length(big endian)0x%x)", pdu_length+4)); - - temp = (char *) smb_buffer; - if (temp[0] == (char) RFC1002_SESSION_KEEP_ALIVE) { - cFYI(0,("Received 4 byte keep alive packet")); - } else if (temp[0] == (char) RFC1002_POSITIVE_SESSION_RESPONSE) { - cFYI(1,("Good RFC 1002 session rsp")); - } else if (temp[0] == (char)RFC1002_NEGATIVE_SESSION_RESPONSE) { - /* we get this from Windows 98 instead of error on SMB negprot response */ - cFYI(1,("Negative RFC 1002 Session Response Error 0x%x)",temp[4])); - if(server->tcpStatus == CifsNew) { - /* if nack on negprot (rather than - ret of smb negprot error) reconnecting - not going to help, ret error to mount */ - break; - } else { - /* give server a second to - clean up before reconnect attempt */ - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(HZ); - /* always try 445 first on reconnect - since we get NACK on some if we ever - connected to port 139 (the NACK is - since we do not begin with RFC1001 - session initialize frame) */ - server->addr.sockAddr.sin_port = htons(CIFS_PORT); - cifs_reconnect(server); - csocket = server->ssocket; - wake_up(&server->response_q); - continue; - } - } else if (temp[0] != (char) 0) { - cERROR(1,("Unknown RFC 1002 frame")); - cifs_dump_mem(" Received Data: ", temp, length); - cifs_reconnect(server); - csocket = server->ssocket; - continue; - } else { - if((pdu_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) - || (pdu_length < sizeof (struct smb_hdr) - 1 - 4)) { - cERROR(1, - ("Invalid size SMB length %d and pdu_length %d", - length, pdu_length+4)); - cifs_reconnect(server); - csocket = server->ssocket; - wake_up(&server->response_q); - continue; - } else { /* length ok */ - length = 0; - iov.iov_base = 4 + (char *)smb_buffer; - iov.iov_len = pdu_length; - for (total_read = 0; - total_read < pdu_length; - total_read += length) { - length = kernel_recvmsg(csocket, &smb_msg, - &iov, 1, - pdu_length - total_read, 0); - if (length == 0) { - cERROR(1, - ("Zero length receive when expecting %d ", - pdu_length - total_read)); - cifs_reconnect(server); - csocket = server->ssocket; - wake_up(&server->response_q); - continue; - } - } - length += 4; /* account for rfc1002 hdr */ - } + } else if (length < 4) { + cFYI(1, + ("Frame under four bytes received (%d bytes long)", + length)); + cifs_reconnect(server); + csocket = server->ssocket; + wake_up(&server->response_q); + continue; + } - dump_smb(smb_buffer, length); - if (checkSMB - (smb_buffer, smb_buffer->Mid, total_read+4)) { - cERROR(1, ("Bad SMB Received ")); - continue; - } + /* the right amount was read from socket - 4 bytes */ - task_to_wake = NULL; - spin_lock(&GlobalMid_Lock); - list_for_each(tmp, &server->pending_mid_q) { - mid_entry = list_entry(tmp, struct - mid_q_entry, - qhead); + pdu_length = ntohl(smb_buffer->smb_buf_length); + cFYI(1,("rfc1002 length(big endian)0x%x)", pdu_length+4)); - if ((mid_entry->mid == smb_buffer->Mid) && (mid_entry->midState == MID_REQUEST_SUBMITTED)) { - cFYI(1, - (" Mid 0x%x matched - waking up ",mid_entry->mid)); - task_to_wake = mid_entry->tsk; - mid_entry->resp_buf = - smb_buffer; - mid_entry->midState = - MID_RESPONSE_RECEIVED; - } - } - spin_unlock(&GlobalMid_Lock); - if (task_to_wake) { - smb_buffer = NULL; /* will be freed by users thread after he is done */ - wake_up_process(task_to_wake); - } else if (is_valid_oplock_break(smb_buffer) == FALSE) { - cERROR(1, ("No task to wake, unknown frame rcvd!")); - cifs_dump_mem("Received Data is: ",temp,sizeof(struct smb_hdr)); - } + temp = (char *) smb_buffer; + if (temp[0] == (char) RFC1002_SESSION_KEEP_ALIVE) { + continue; + } else if (temp[0] == (char)RFC1002_POSITIVE_SESSION_RESPONSE) { + cFYI(1,("Good RFC 1002 session rsp")); + continue; + } else if (temp[0] == (char)RFC1002_NEGATIVE_SESSION_RESPONSE) { + /* we get this from Windows 98 instead of + an error on SMB negprot response */ + cFYI(1,("Negative RFC1002 Session Response Error 0x%x)", + temp[4])); + if(server->tcpStatus == CifsNew) { + /* if nack on negprot (rather than + ret of smb negprot error) reconnecting + not going to help, ret error to mount */ + break; + } else { + /* give server a second to + clean up before reconnect attempt */ + msleep(1000); + /* always try 445 first on reconnect + since we get NACK on some if we ever + connected to port 139 (the NACK is + since we do not begin with RFC1001 + session initialize frame) */ + server->addr.sockAddr.sin_port = + htons(CIFS_PORT); + cifs_reconnect(server); + csocket = server->ssocket; + wake_up(&server->response_q); + continue; } - } else { - cFYI(1, - ("Frame less than four bytes received %d bytes long.", - length)); + } else if (temp[0] != (char) 0) { + cERROR(1,("Unknown RFC 1002 frame")); + cifs_dump_mem(" Received Data: ", temp, length); + cifs_reconnect(server); + csocket = server->ssocket; + continue; + } + + /* else we have an SMB response */ + if((pdu_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) || + (pdu_length < sizeof (struct smb_hdr) - 1 - 4)) { + cERROR(1, ("Invalid size SMB length %d pdu_length %d", + length, pdu_length+4)); cifs_reconnect(server); csocket = server->ssocket; wake_up(&server->response_q); continue; + } + + /* else length ok */ + reconnect = 0; + + if(pdu_length > MAX_CIFS_HDR_SIZE - 4) { + isLargeBuf = TRUE; + memcpy(bigbuf, smallbuf, 4); + smb_buffer = bigbuf; } - } + length = 0; + iov.iov_base = 4 + (char *)smb_buffer; + iov.iov_len = pdu_length; + for (total_read = 0; total_read < pdu_length; + total_read += length) { + length = kernel_recvmsg(csocket, &smb_msg, &iov, 1, + pdu_length - total_read, 0); + if((server->tcpStatus == CifsExiting) || + (length == -EINTR)) { + /* then will exit */ + reconnect = 2; + break; + } else if (server->tcpStatus == CifsNeedReconnect) { + cifs_reconnect(server); + csocket = server->ssocket; + /* Reconnect wakes up rspns q */ + /* Now we will reread sock */ + reconnect = 1; + break; + } else if ((length == -ERESTARTSYS) || + (length == -EAGAIN)) { + msleep(1); /* minimum sleep to prevent looping, + allowing socket to clear and app + threads to set tcpStatus + CifsNeedReconnect if server hung*/ + continue; + } else if (length <= 0) { + cERROR(1,("Received no data, expecting %d", + pdu_length - total_read)); + cifs_reconnect(server); + csocket = server->ssocket; + reconnect = 1; + break; + } + } + if(reconnect == 2) + break; + else if(reconnect == 1) + continue; + + length += 4; /* account for rfc1002 hdr */ + + + dump_smb(smb_buffer, length); + if (checkSMB (smb_buffer, smb_buffer->Mid, total_read+4)) { + cERROR(1, ("Bad SMB Received ")); + continue; + } + + + task_to_wake = NULL; + spin_lock(&GlobalMid_Lock); + list_for_each(tmp, &server->pending_mid_q) { + mid_entry = list_entry(tmp, struct mid_q_entry, qhead); + + if ((mid_entry->mid == smb_buffer->Mid) && + (mid_entry->midState == MID_REQUEST_SUBMITTED) && + (mid_entry->command == smb_buffer->Command)) { + if(check2ndT2(smb_buffer,server->maxBuf) > 0) { + /* We have a multipart transact2 resp */ + isMultiRsp = TRUE; + if(mid_entry->resp_buf) { + /* merge response - fix up 1st*/ + if(coalesce_t2(smb_buffer, + mid_entry->resp_buf)) { + break; + } else { + /* all parts received */ + goto multi_t2_fnd; + } + } else { + if(!isLargeBuf) { + cERROR(1,("1st trans2 resp needs bigbuf")); + /* BB maybe we can fix this up, switch + to already allocated large buffer? */ + } else { + /* Have first buffer */ + mid_entry->resp_buf = + smb_buffer; + mid_entry->largeBuf = 1; + bigbuf = NULL; + } + } + break; + } + mid_entry->resp_buf = smb_buffer; + if(isLargeBuf) + mid_entry->largeBuf = 1; + else + mid_entry->largeBuf = 0; +multi_t2_fnd: + task_to_wake = mid_entry->tsk; + mid_entry->midState = MID_RESPONSE_RECEIVED; + break; + } + } + spin_unlock(&GlobalMid_Lock); + if (task_to_wake) { + /* Was previous buf put in mpx struct for multi-rsp? */ + if(!isMultiRsp) { + /* smb buffer will be freed by user thread */ + if(isLargeBuf) { + bigbuf = NULL; + } else + smallbuf = NULL; + } + wake_up_process(task_to_wake); + } else if ((is_valid_oplock_break(smb_buffer) == FALSE) + && (isMultiRsp == FALSE)) { + cERROR(1, ("No task to wake, unknown frame rcvd!")); + cifs_dump_mem("Received Data is: ",temp,sizeof(struct smb_hdr)); + } + } /* end while !EXITING */ + spin_lock(&GlobalMid_Lock); server->tcpStatus = CifsExiting; server->tsk = NULL; - atomic_set(&server->inFlight, 0); + /* check if we have blocked requests that need to free */ + /* Note that cifs_max_pending is normally 50, but + can be set at module install time to as little as two */ + if(atomic_read(&server->inFlight) >= cifs_max_pending) + atomic_set(&server->inFlight, cifs_max_pending - 1); + /* We do not want to set the max_pending too low or we + could end up with the counter going negative */ spin_unlock(&GlobalMid_Lock); /* Although there should not be any requests blocked on this queue it can not hurt to be paranoid and try to wake up requests - that may haven been blocked when more than 50 at time were on the wire + that may haven been blocked when more than 50 at time were on the wire to the same server - they now will see the session is in exit state and get out of SendReceive. */ wake_up_all(&server->request_q); /* give those requests time to exit */ - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(HZ/8); - + msleep(125); + if(server->ssocket) { sock_release(csocket); server->ssocket = NULL; } - if (smb_buffer) /* buffer usually freed in free_mid - need to free it on error or exit */ - cifs_buf_release(smb_buffer); + /* buffer usuallly freed in free_mid - need to free it here on exit */ + if (bigbuf != NULL) + cifs_buf_release(bigbuf); + if (smallbuf != NULL) + cifs_small_buf_release(smallbuf); read_lock(&GlobalSMBSeslock); if (list_empty(&server->pending_mid_q)) { - /* loop through server session structures attached to this and mark them dead */ + /* loop through server session structures attached to this and + mark them dead */ list_for_each(tmp, &GlobalSMBSessionList) { ses = list_entry(tmp, struct cifsSesInfo, @@ -424,12 +645,23 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) } read_unlock(&GlobalSMBSeslock); } else { + /* although we can not zero the server struct pointer yet, + since there are active requests which may depnd on them, + mark the corresponding SMB sessions as exiting too */ + list_for_each(tmp, &GlobalSMBSessionList) { + ses = list_entry(tmp, struct cifsSesInfo, + cifsSessionList); + if (ses->server == server) { + ses->status = CifsExiting; + } + } + spin_lock(&GlobalMid_Lock); list_for_each(tmp, &server->pending_mid_q) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); if (mid_entry->midState == MID_REQUEST_SUBMITTED) { cFYI(1, - (" Clearing Mid 0x%x - waking up ",mid_entry->mid)); + ("Clearing Mid 0x%x - waking up ",mid_entry->mid)); task_to_wake = mid_entry->tsk; if(task_to_wake) { wake_up_process(task_to_wake); @@ -438,47 +670,51 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) } spin_unlock(&GlobalMid_Lock); read_unlock(&GlobalSMBSeslock); - set_current_state(TASK_INTERRUPTIBLE); /* 1/8th of sec is more than enough time for them to exit */ - schedule_timeout(HZ/8); + msleep(125); } if (list_empty(&server->pending_mid_q)) { /* mpx threads have not exited yet give them at least the smb send timeout time for long ops */ + /* due to delays on oplock break requests, we need + to wait at least 45 seconds before giving up + on a request getting a response and going ahead + and killing cifsd */ cFYI(1, ("Wait for exit from demultiplex thread")); - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(46 * HZ); + msleep(46000); /* if threads still have not exited they are probably never coming home not much else we can do but free the memory */ } - kfree(server); write_lock(&GlobalSMBSeslock); atomic_dec(&tcpSesAllocCount); length = tcpSesAllocCount.counter; + + /* last chance to mark ses pointers invalid + if there are any pointing to this (e.g + if a crazy root user tried to kill cifsd + kernel thread explicitly this might happen) */ + list_for_each(tmp, &GlobalSMBSessionList) { + ses = list_entry(tmp, struct cifsSesInfo, + cifsSessionList); + if (ses->server == server) { + ses->server = NULL; + } + } write_unlock(&GlobalSMBSeslock); + + kfree(server); if(length > 0) { mempool_resize(cifs_req_poolp, length + cifs_min_rcv, GFP_KERNEL); } - - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(HZ/4); + + msleep(250); return 0; } -static void * -cifs_kcalloc(size_t size, unsigned int __nocast type) -{ - void *addr; - addr = kmalloc(size, type); - if (addr) - memset(addr, 0, size); - return addr; -} - static int cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) { @@ -495,7 +731,8 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) /* does not have to be a perfect mapping since the field is informational, only used for servers that do not support port 445 and it can be overridden at mount time */ - vol->source_rfc1001_name[i] = toupper(system_utsname.nodename[i]); + vol->source_rfc1001_name[i] = + toupper(system_utsname.nodename[i]); } vol->source_rfc1001_name[15] = 0; @@ -570,14 +807,17 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) /* NB: password legally can have multiple commas and the only illegal character in a password is null */ - if ((value[temp_len] == 0) && (value[temp_len+1] == separator[0])) { + if ((value[temp_len] == 0) && + (value[temp_len+1] == separator[0])) { /* reinsert comma */ value[temp_len] = separator[0]; temp_len+=2; /* move after the second comma */ while(value[temp_len] != 0) { if (value[temp_len] == separator[0]) { - if (value[temp_len+1] == separator[0]) { - temp_len++; /* skip second comma */ + if (value[temp_len+1] == + separator[0]) { + /* skip second comma */ + temp_len++; } else { /* single comma indicating start of next parm */ @@ -596,17 +836,26 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) /* go from value to value + temp_len condensing double commas to singles. Note that this ends up allocating a few bytes too many, which is ok */ - vol->password = cifs_kcalloc(temp_len, GFP_KERNEL); + vol->password = kcalloc(1, temp_len, GFP_KERNEL); + if(vol->password == NULL) { + printk("CIFS: no memory for pass\n"); + return 1; + } for(i=0,j=0;i<temp_len;i++,j++) { vol->password[j] = value[i]; - if(value[i] == separator[0] && value[i+1] == separator[0]) { + if(value[i] == separator[0] + && value[i+1] == separator[0]) { /* skip second comma */ i++; } } vol->password[j] = 0; } else { - vol->password = cifs_kcalloc(temp_len + 1, GFP_KERNEL); + vol->password = kcalloc(1, temp_len+1, GFP_KERNEL); + if(vol->password == NULL) { + printk("CIFS: no memory for pass\n"); + return 1; + } strcpy(vol->password, value); } } else if (strnicmp(data, "ip", 2) == 0) { @@ -770,6 +1019,10 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) vol->noperm = 0; } else if (strnicmp(data, "noperm", 6) == 0) { vol->noperm = 1; + } else if (strnicmp(data, "mapchars", 8) == 0) { + vol->remap = 1; + } else if (strnicmp(data, "nomapchars", 10) == 0) { + vol->remap = 0; } else if (strnicmp(data, "setuids", 7) == 0) { vol->setuids = 1; } else if (strnicmp(data, "nosetuids", 9) == 0) { @@ -918,14 +1171,15 @@ find_unc(__be32 new_target_ip_addr, char *uncName, char *userName) int connect_to_dfs_path(int xid, struct cifsSesInfo *pSesInfo, - const char *old_path, const struct nls_table *nls_codepage) + const char *old_path, const struct nls_table *nls_codepage, + int remap) { unsigned char *referrals = NULL; unsigned int num_referrals; int rc = 0; rc = get_dfs_path(xid, pSesInfo,old_path, nls_codepage, - &num_referrals, &referrals); + &num_referrals, &referrals, remap); /* BB Add in code to: if valid refrl, if not ip address contact the helper that resolves tcp names, mount to it, try to @@ -940,7 +1194,8 @@ connect_to_dfs_path(int xid, struct cifsSesInfo *pSesInfo, int get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, const struct nls_table *nls_codepage, - unsigned int *pnum_referrals, unsigned char ** preferrals) + unsigned int *pnum_referrals, + unsigned char ** preferrals, int remap) { char *temp_unc; int rc = 0; @@ -965,7 +1220,7 @@ get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, } if (rc == 0) rc = CIFSGetDFSRefer(xid, pSesInfo, old_path, preferrals, - pnum_referrals, nls_codepage); + pnum_referrals, nls_codepage, remap); return rc; } @@ -1062,7 +1317,7 @@ ipv4_connect(struct sockaddr_in *psin_server, struct socket **csocket, sessinit is sent but no second negprot */ struct rfc1002_session_packet * ses_init_buf; struct smb_hdr * smb_buf; - ses_init_buf = cifs_kcalloc(sizeof(struct rfc1002_session_packet), GFP_KERNEL); + ses_init_buf = kcalloc(1, sizeof(struct rfc1002_session_packet), GFP_KERNEL); if(ses_init_buf) { ses_init_buf->trailer.session_req.called_len = 32; rfc1002mangle(ses_init_buf->trailer.session_req.called_name, @@ -1352,6 +1607,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, } else rc = 0; memcpy(srvTcp->workstation_RFC1001_name, volume_info.source_rfc1001_name,16); + srvTcp->sequence_number = 0; } } @@ -1419,6 +1675,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_SET_UID; if(volume_info.server_ino) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_SERVER_INUM; + if(volume_info.remap) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_MAP_SPECIAL_CHR; if(volume_info.no_xattr) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_XATTR; if(volume_info.direct_io) { @@ -1447,11 +1705,10 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if ((strchr(volume_info.UNC + 3, '\\') == NULL) && (strchr(volume_info.UNC + 3, '/') == NULL)) { - rc = connect_to_dfs_path(xid, - pSesInfo, - "", - cifs_sb-> - local_nls); + rc = connect_to_dfs_path(xid, pSesInfo, + "", cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if(volume_info.UNC) kfree(volume_info.UNC); FreeXid(xid); @@ -1514,10 +1771,10 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, tcon->ses = pSesInfo; /* do not care if following two calls succeed - informational only */ - CIFSSMBQFSDeviceInfo(xid, tcon, cifs_sb->local_nls); - CIFSSMBQFSAttributeInfo(xid, tcon, cifs_sb->local_nls); + CIFSSMBQFSDeviceInfo(xid, tcon); + CIFSSMBQFSAttributeInfo(xid, tcon); if (tcon->ses->capabilities & CAP_UNIX) { - if(!CIFSSMBQFSUnixInfo(xid, tcon, cifs_sb->local_nls)) { + if(!CIFSSMBQFSUnixInfo(xid, tcon)) { if(!volume_info.no_psx_acl) { if(CIFS_UNIX_POSIX_ACL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) @@ -1707,7 +1964,9 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, /* We look for obvious messed up bcc or strings in response so we do not go off the end since (at least) WIN2K and Windows XP have a major bug in not null terminating last Unicode string in response */ - ses->serverOS = cifs_kcalloc(2 * (len + 1), GFP_KERNEL); + ses->serverOS = kcalloc(1, 2 * (len + 1), GFP_KERNEL); + if(ses->serverOS == NULL) + goto sesssetup_nomem; cifs_strfromUCS_le(ses->serverOS, (wchar_t *)bcc_ptr, len,nls_codepage); bcc_ptr += 2 * (len + 1); @@ -1717,7 +1976,9 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, if (remaining_words > 0) { len = UniStrnlen((wchar_t *)bcc_ptr, remaining_words-1); - ses->serverNOS =cifs_kcalloc(2 * (len + 1),GFP_KERNEL); + ses->serverNOS = kcalloc(1, 2 * (len + 1),GFP_KERNEL); + if(ses->serverNOS == NULL) + goto sesssetup_nomem; cifs_strfromUCS_le(ses->serverNOS, (wchar_t *)bcc_ptr,len,nls_codepage); bcc_ptr += 2 * (len + 1); @@ -1730,10 +1991,12 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, } remaining_words -= len + 1; if (remaining_words > 0) { - len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); + len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); /* last string is not always null terminated (for e.g. for Windows XP & 2000) */ ses->serverDomain = - cifs_kcalloc(2*(len+1),GFP_KERNEL); + kcalloc(1, 2*(len+1),GFP_KERNEL); + if(ses->serverDomain == NULL) + goto sesssetup_nomem; cifs_strfromUCS_le(ses->serverDomain, (wchar_t *)bcc_ptr,len,nls_codepage); bcc_ptr += 2 * (len + 1); @@ -1741,21 +2004,25 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, ses->serverDomain[1+(2*len)] = 0; } /* else no more room so create dummy domain string */ else - ses->serverDomain = - cifs_kcalloc(2, - GFP_KERNEL); + ses->serverDomain = + kcalloc(1, 2, GFP_KERNEL); } else { /* no room so create dummy domain and NOS string */ + /* if these kcallocs fail not much we + can do, but better to not fail the + sesssetup itself */ ses->serverDomain = - cifs_kcalloc(2, GFP_KERNEL); + kcalloc(1, 2, GFP_KERNEL); ses->serverNOS = - cifs_kcalloc(2, GFP_KERNEL); + kcalloc(1, 2, GFP_KERNEL); } } else { /* ASCII */ len = strnlen(bcc_ptr, 1024); if (((long) bcc_ptr + len) - (long) pByteArea(smb_buffer_response) <= BCC(smb_buffer_response)) { - ses->serverOS = cifs_kcalloc(len + 1,GFP_KERNEL); + ses->serverOS = kcalloc(1, len + 1,GFP_KERNEL); + if(ses->serverOS == NULL) + goto sesssetup_nomem; strncpy(ses->serverOS,bcc_ptr, len); bcc_ptr += len; @@ -1763,14 +2030,18 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr++; len = strnlen(bcc_ptr, 1024); - ses->serverNOS = cifs_kcalloc(len + 1,GFP_KERNEL); + ses->serverNOS = kcalloc(1, len + 1,GFP_KERNEL); + if(ses->serverNOS == NULL) + goto sesssetup_nomem; strncpy(ses->serverNOS, bcc_ptr, len); bcc_ptr += len; bcc_ptr[0] = 0; bcc_ptr++; len = strnlen(bcc_ptr, 1024); - ses->serverDomain = cifs_kcalloc(len + 1,GFP_KERNEL); + ses->serverDomain = kcalloc(1, len + 1,GFP_KERNEL); + if(ses->serverDomain == NULL) + goto sesssetup_nomem; strncpy(ses->serverDomain, bcc_ptr, len); bcc_ptr += len; bcc_ptr[0] = 0; @@ -1790,7 +2061,9 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, smb_buffer_response->WordCount)); rc = -EIO; } - +sesssetup_nomem: /* do not return an error on nomem for the info strings, + since that could make reconnection harder, and + reconnection might be needed to free memory */ if (smb_buffer) cifs_buf_release(smb_buffer); @@ -1967,7 +2240,7 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, the end since (at least) WIN2K and Windows XP have a major bug in not null terminating last Unicode string in response */ ses->serverOS = - cifs_kcalloc(2 * (len + 1), GFP_KERNEL); + kcalloc(1, 2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverOS, (wchar_t *) bcc_ptr, len, @@ -1981,7 +2254,7 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, remaining_words - 1); ses->serverNOS = - cifs_kcalloc(2 * (len + 1), + kcalloc(1, 2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverNOS, (wchar_t *)bcc_ptr, @@ -1994,7 +2267,7 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, if (remaining_words > 0) { len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); /* last string is not always null terminated (for e.g. for Windows XP & 2000) */ - ses->serverDomain = cifs_kcalloc(2*(len+1),GFP_KERNEL); + ses->serverDomain = kcalloc(1, 2*(len+1),GFP_KERNEL); cifs_strfromUCS_le(ses->serverDomain, (wchar_t *)bcc_ptr, len, @@ -2005,10 +2278,10 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, } /* else no more room so create dummy domain string */ else ses->serverDomain = - cifs_kcalloc(2,GFP_KERNEL); + kcalloc(1, 2,GFP_KERNEL); } else { /* no room so create dummy domain and NOS string */ - ses->serverDomain = cifs_kcalloc(2, GFP_KERNEL); - ses->serverNOS = cifs_kcalloc(2, GFP_KERNEL); + ses->serverDomain = kcalloc(1, 2, GFP_KERNEL); + ses->serverNOS = kcalloc(1, 2, GFP_KERNEL); } } else { /* ASCII */ @@ -2016,7 +2289,7 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, if (((long) bcc_ptr + len) - (long) pByteArea(smb_buffer_response) <= BCC(smb_buffer_response)) { - ses->serverOS = cifs_kcalloc(len + 1, GFP_KERNEL); + ses->serverOS = kcalloc(1, len + 1, GFP_KERNEL); strncpy(ses->serverOS, bcc_ptr, len); bcc_ptr += len; @@ -2024,14 +2297,14 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr++; len = strnlen(bcc_ptr, 1024); - ses->serverNOS = cifs_kcalloc(len + 1,GFP_KERNEL); + ses->serverNOS = kcalloc(1, len + 1,GFP_KERNEL); strncpy(ses->serverNOS, bcc_ptr, len); bcc_ptr += len; bcc_ptr[0] = 0; bcc_ptr++; len = strnlen(bcc_ptr, 1024); - ses->serverDomain = cifs_kcalloc(len + 1, GFP_KERNEL); + ses->serverDomain = kcalloc(1, len + 1, GFP_KERNEL); strncpy(ses->serverDomain, bcc_ptr, len); bcc_ptr += len; bcc_ptr[0] = 0; @@ -2281,7 +2554,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, the end since (at least) WIN2K and Windows XP have a major bug in not null terminating last Unicode string in response */ ses->serverOS = - cifs_kcalloc(2 * (len + 1), GFP_KERNEL); + kcalloc(1, 2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverOS, (wchar_t *) bcc_ptr, len, @@ -2296,7 +2569,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, remaining_words - 1); ses->serverNOS = - cifs_kcalloc(2 * (len + 1), + kcalloc(1, 2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses-> serverNOS, @@ -2313,7 +2586,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); /* last string is not always null terminated (for e.g. for Windows XP & 2000) */ ses->serverDomain = - cifs_kcalloc(2 * + kcalloc(1, 2 * (len + 1), GFP_KERNEL); @@ -2339,13 +2612,13 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, } /* else no more room so create dummy domain string */ else ses->serverDomain = - cifs_kcalloc(2, + kcalloc(1, 2, GFP_KERNEL); } else { /* no room so create dummy domain and NOS string */ ses->serverDomain = - cifs_kcalloc(2, GFP_KERNEL); + kcalloc(1, 2, GFP_KERNEL); ses->serverNOS = - cifs_kcalloc(2, GFP_KERNEL); + kcalloc(1, 2, GFP_KERNEL); } } else { /* ASCII */ len = strnlen(bcc_ptr, 1024); @@ -2353,7 +2626,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, pByteArea(smb_buffer_response) <= BCC(smb_buffer_response)) { ses->serverOS = - cifs_kcalloc(len + 1, + kcalloc(1, len + 1, GFP_KERNEL); strncpy(ses->serverOS, bcc_ptr, len); @@ -2364,7 +2637,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, len = strnlen(bcc_ptr, 1024); ses->serverNOS = - cifs_kcalloc(len + 1, + kcalloc(1, len + 1, GFP_KERNEL); strncpy(ses->serverNOS, bcc_ptr, len); bcc_ptr += len; @@ -2373,7 +2646,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, len = strnlen(bcc_ptr, 1024); ses->serverDomain = - cifs_kcalloc(len + 1, + kcalloc(1, len + 1, GFP_KERNEL); strncpy(ses->serverDomain, bcc_ptr, len); bcc_ptr += len; @@ -2675,7 +2948,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, the end since (at least) WIN2K and Windows XP have a major bug in not null terminating last Unicode string in response */ ses->serverOS = - cifs_kcalloc(2 * (len + 1), GFP_KERNEL); + kcalloc(1, 2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverOS, (wchar_t *) bcc_ptr, len, @@ -2690,7 +2963,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, remaining_words - 1); ses->serverNOS = - cifs_kcalloc(2 * (len + 1), + kcalloc(1, 2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses-> serverNOS, @@ -2706,7 +2979,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); /* last string not always null terminated (e.g. for Windows XP & 2000) */ ses->serverDomain = - cifs_kcalloc(2 * + kcalloc(1, 2 * (len + 1), GFP_KERNEL); @@ -2731,17 +3004,17 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, = 0; } /* else no more room so create dummy domain string */ else - ses->serverDomain = cifs_kcalloc(2,GFP_KERNEL); + ses->serverDomain = kcalloc(1, 2,GFP_KERNEL); } else { /* no room so create dummy domain and NOS string */ - ses->serverDomain = cifs_kcalloc(2, GFP_KERNEL); - ses->serverNOS = cifs_kcalloc(2, GFP_KERNEL); + ses->serverDomain = kcalloc(1, 2, GFP_KERNEL); + ses->serverNOS = kcalloc(1, 2, GFP_KERNEL); } } else { /* ASCII */ len = strnlen(bcc_ptr, 1024); if (((long) bcc_ptr + len) - (long) pByteArea(smb_buffer_response) <= BCC(smb_buffer_response)) { - ses->serverOS = cifs_kcalloc(len + 1,GFP_KERNEL); + ses->serverOS = kcalloc(1, len + 1,GFP_KERNEL); strncpy(ses->serverOS,bcc_ptr, len); bcc_ptr += len; @@ -2749,14 +3022,14 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr++; len = strnlen(bcc_ptr, 1024); - ses->serverNOS = cifs_kcalloc(len+1,GFP_KERNEL); + ses->serverNOS = kcalloc(1, len+1,GFP_KERNEL); strncpy(ses->serverNOS, bcc_ptr, len); bcc_ptr += len; bcc_ptr[0] = 0; bcc_ptr++; len = strnlen(bcc_ptr, 1024); - ses->serverDomain = cifs_kcalloc(len+1,GFP_KERNEL); + ses->serverDomain = kcalloc(1, len+1,GFP_KERNEL); strncpy(ses->serverDomain, bcc_ptr, len); bcc_ptr += len; bcc_ptr[0] = 0; @@ -2868,7 +3141,7 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, if(tcon->nativeFileSystem) kfree(tcon->nativeFileSystem); tcon->nativeFileSystem = - cifs_kcalloc(length + 2, GFP_KERNEL); + kcalloc(1, length + 2, GFP_KERNEL); cifs_strfromUCS_le(tcon->nativeFileSystem, (wchar_t *) bcc_ptr, length, nls_codepage); @@ -2886,7 +3159,7 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, if(tcon->nativeFileSystem) kfree(tcon->nativeFileSystem); tcon->nativeFileSystem = - cifs_kcalloc(length + 1, GFP_KERNEL); + kcalloc(1, length + 1, GFP_KERNEL); strncpy(tcon->nativeFileSystem, bcc_ptr, length); } @@ -2959,6 +3232,7 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, int rc = 0; char ntlm_session_key[CIFS_SESSION_KEY_SIZE]; int ntlmv2_flag = FALSE; + int first_time = 0; /* what if server changes its buffer size after dropping the session? */ if(pSesInfo->server->maxBuf == 0) /* no need to send on reconnect */ { @@ -2977,12 +3251,13 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, spin_unlock(&GlobalMid_Lock); } + first_time = 1; } if (!rc) { pSesInfo->capabilities = pSesInfo->server->capabilities; if(linuxExtEnabled == 0) pSesInfo->capabilities &= (~CAP_UNIX); - pSesInfo->sequence_number = 0; + /* pSesInfo->sequence_number = 0;*/ cFYI(1,("Security Mode: 0x%x Capabilities: 0x%x Time Zone: %d", pSesInfo->server->secMode, pSesInfo->server->capabilities, @@ -3015,7 +3290,10 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, v2_response = kmalloc(16 + 64 /* blob */, GFP_KERNEL); if(v2_response) { CalcNTLMv2_response(pSesInfo,v2_response); -/* cifs_calculate_ntlmv2_mac_key(pSesInfo->mac_signing_key, response, ntlm_session_key, */ + /* if(first_time) + cifs_calculate_ntlmv2_mac_key( + pSesInfo->server->mac_signing_key, + response, ntlm_session_key, */ kfree(v2_response); /* BB Put dummy sig in SessSetup PDU? */ } else { @@ -3028,9 +3306,11 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, pSesInfo->server->cryptKey, ntlm_session_key); - cifs_calculate_mac_key(pSesInfo->mac_signing_key, - ntlm_session_key, - pSesInfo->password); + if(first_time) + cifs_calculate_mac_key( + pSesInfo->server->mac_signing_key, + ntlm_session_key, + pSesInfo->password); } /* for better security the weaker lanman hash not sent in AuthSessSetup so we no longer calculate it */ @@ -3046,8 +3326,11 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, pSesInfo->server->cryptKey, ntlm_session_key); - cifs_calculate_mac_key(pSesInfo->mac_signing_key, - ntlm_session_key, pSesInfo->password); + if(first_time) + cifs_calculate_mac_key( + pSesInfo->server->mac_signing_key, + ntlm_session_key, pSesInfo->password); + rc = CIFSSessSetup(xid, pSesInfo, ntlm_session_key, nls_info); } diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index f54e1866f0f..e3137aa48cd 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -101,68 +101,15 @@ cifs_bp_rename_retry: return full_path; } -/* Note: caller must free return buffer */ -char * -build_wildcard_path_from_dentry(struct dentry *direntry) +/* char * build_wildcard_path_from_dentry(struct dentry *direntry) { - struct dentry *temp; - int namelen = 0; - char *full_path; - - if(direntry == NULL) - return NULL; /* not much we can do if dentry is freed and - we need to reopen the file after it was closed implicitly - when the server crashed */ - -cifs_bwp_rename_retry: - for (temp = direntry; !IS_ROOT(temp);) { - namelen += (1 + temp->d_name.len); - temp = temp->d_parent; - if(temp == NULL) { - cERROR(1,("corrupt dentry")); - return NULL; - } - } - - full_path = kmalloc(namelen+3, GFP_KERNEL); if(full_path == NULL) return full_path; full_path[namelen] = '\\'; full_path[namelen+1] = '*'; - full_path[namelen+2] = 0; /* trailing null */ - - for (temp = direntry; !IS_ROOT(temp);) { - namelen -= 1 + temp->d_name.len; - if (namelen < 0) { - break; - } else { - full_path[namelen] = '\\'; - strncpy(full_path + namelen + 1, temp->d_name.name, - temp->d_name.len); - cFYI(0, (" name: %s ", full_path + namelen)); - } - temp = temp->d_parent; - if(temp == NULL) { - cERROR(1,("corrupt dentry")); - kfree(full_path); - return NULL; - } - } - if (namelen != 0) { - cERROR(1, - ("We did not end path lookup where we expected namelen is %d", - namelen)); - /* presumably this is only possible if we were racing with a rename - of one of the parent directories (we can not lock the dentries - above us to prevent this, but retrying should be harmless) */ - kfree(full_path); - namelen = 0; - goto cifs_bwp_rename_retry; - } - - return full_path; -} + full_path[namelen+2] = 0; +BB remove above eight lines BB */ /* Inode operations in similar order to how they appear in the Linux file fs.h */ @@ -235,7 +182,8 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, rc = CIFSSMBOpen(xid, pTcon, full_path, disposition, desiredAccess, CREATE_NOT_DIR, - &fileHandle, &oplock, buf, cifs_sb->local_nls); + &fileHandle, &oplock, buf, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc) { cFYI(1, ("cifs_create returned 0x%x ", rc)); } else { @@ -248,13 +196,17 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, (__u64)current->euid, (__u64)current->egid, 0 /* dev */, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, (__u64)-1, (__u64)-1, 0 /* dev */, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { /* BB implement via Windows security descriptors */ @@ -284,51 +236,48 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, /* mknod case - do not leave file open */ CIFSSMBClose(xid, pTcon, fileHandle); } else if(newinode) { - pCifsFile = (struct cifsFileInfo *) + pCifsFile = kmalloc(sizeof (struct cifsFileInfo), GFP_KERNEL); - - if (pCifsFile) { - memset((char *)pCifsFile, 0, - sizeof (struct cifsFileInfo)); - pCifsFile->netfid = fileHandle; - pCifsFile->pid = current->tgid; - pCifsFile->pInode = newinode; - pCifsFile->invalidHandle = FALSE; - pCifsFile->closePend = FALSE; - init_MUTEX(&pCifsFile->fh_sem); - /* put the following in at open now */ - /* pCifsFile->pfile = file; */ - write_lock(&GlobalSMBSeslock); - list_add(&pCifsFile->tlist,&pTcon->openFileList); - pCifsInode = CIFS_I(newinode); - if(pCifsInode) { + + if(pCifsFile == NULL) + goto cifs_create_out; + memset((char *)pCifsFile, 0, + sizeof (struct cifsFileInfo)); + pCifsFile->netfid = fileHandle; + pCifsFile->pid = current->tgid; + pCifsFile->pInode = newinode; + pCifsFile->invalidHandle = FALSE; + pCifsFile->closePend = FALSE; + init_MUTEX(&pCifsFile->fh_sem); + /* set the following in open now + pCifsFile->pfile = file; */ + write_lock(&GlobalSMBSeslock); + list_add(&pCifsFile->tlist,&pTcon->openFileList); + pCifsInode = CIFS_I(newinode); + if(pCifsInode) { /* if readable file instance put first in list*/ - if (write_only == TRUE) { - list_add_tail(&pCifsFile->flist, - &pCifsInode->openFileList); - } else { - list_add(&pCifsFile->flist, - &pCifsInode->openFileList); - } - if((oplock & 0xF) == OPLOCK_EXCLUSIVE) { - pCifsInode->clientCanCacheAll = TRUE; - pCifsInode->clientCanCacheRead = TRUE; - cFYI(1,("Exclusive Oplock granted on inode %p", - newinode)); - } else if((oplock & 0xF) == OPLOCK_READ) - pCifsInode->clientCanCacheRead = TRUE; + if (write_only == TRUE) { + list_add_tail(&pCifsFile->flist, + &pCifsInode->openFileList); + } else { + list_add(&pCifsFile->flist, + &pCifsInode->openFileList); } - write_unlock(&GlobalSMBSeslock); + if((oplock & 0xF) == OPLOCK_EXCLUSIVE) { + pCifsInode->clientCanCacheAll = TRUE; + pCifsInode->clientCanCacheRead = TRUE; + cFYI(1,("Exclusive Oplock for inode %p", + newinode)); + } else if((oplock & 0xF) == OPLOCK_READ) + pCifsInode->clientCanCacheRead = TRUE; } + write_unlock(&GlobalSMBSeslock); } } - - if (buf) - kfree(buf); - if (full_path) - kfree(full_path); +cifs_create_out: + kfree(buf); + kfree(full_path); FreeXid(xid); - return rc; } @@ -359,11 +308,15 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, dev_t dev if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { rc = CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode,(__u64)current->euid,(__u64)current->egid, - device_number, cifs_sb->local_nls); + device_number, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { rc = CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, (__u64)-1, (__u64)-1, - device_number, cifs_sb->local_nls); + device_number, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } if(!rc) { @@ -375,10 +328,8 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, dev_t dev } } - if (full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); - return rc; } @@ -447,43 +398,11 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, struct name if file exists or not but no access BB */ } - if (full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); return ERR_PTR(rc); } -int -cifs_dir_open(struct inode *inode, struct file *file) -{ /* NB: currently unused since searches are opened in readdir */ - int rc = 0; - int xid; - struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; - char *full_path = NULL; - - xid = GetXid(); - - cifs_sb = CIFS_SB(inode->i_sb); - pTcon = cifs_sb->tcon; - - if(file->f_dentry) { - down(&file->f_dentry->d_sb->s_vfs_rename_sem); - full_path = build_wildcard_path_from_dentry(file->f_dentry); - up(&file->f_dentry->d_sb->s_vfs_rename_sem); - } else { - FreeXid(xid); - return -EIO; - } - - cFYI(1, ("inode = 0x%p and full path is %s", inode, full_path)); - - if (full_path) - kfree(full_path); - FreeXid(xid); - return rc; -} - static int cifs_d_revalidate(struct dentry *direntry, struct nameidata *nd) { diff --git a/fs/cifs/fcntl.c b/fs/cifs/fcntl.c index 9d24c40f196..7d2a9202c39 100644 --- a/fs/cifs/fcntl.c +++ b/fs/cifs/fcntl.c @@ -92,7 +92,8 @@ int cifs_dir_notify(struct file * file, unsigned long arg) cERROR(1,("cifs dir notify on file %s with arg 0x%lx",full_path,arg)); /* BB removeme BB */ rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, GENERIC_READ | SYNCHRONIZE, 0 /* create options */, - &netfid, &oplock,NULL, cifs_sb->local_nls); + &netfid, &oplock,NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); /* BB fixme - add this handle to a notify handle list */ if(rc) { cERROR(1,("Could not open directory for notify")); /* BB remove BB */ diff --git a/fs/cifs/file.c b/fs/cifs/file.c index dcab7cf1b53..dde2d251fc3 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -254,7 +254,8 @@ int cifs_open(struct inode *inode, struct file *file) } rc = CIFSSMBOpen(xid, pTcon, full_path, disposition, desiredAccess, CREATE_NOT_DIR, &netfid, &oplock, buf, - cifs_sb->local_nls); + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags + & CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc) { cFYI(1, ("cifs_open returned 0x%x ", rc)); goto out; @@ -287,7 +288,9 @@ int cifs_open(struct inode *inode, struct file *file) CIFSSMBUnixSetPerms(xid, pTcon, full_path, inode->i_mode, (__u64)-1, (__u64)-1, 0 /* dev */, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { /* BB implement via Windows security descriptors eg CIFSSMBWinSetPerms(xid, pTcon, full_path, mode, @@ -387,7 +390,8 @@ static int cifs_reopen_file(struct inode *inode, struct file *file, } */ rc = CIFSSMBOpen(xid, pTcon, full_path, disposition, desiredAccess, CREATE_NOT_DIR, &netfid, &oplock, NULL, - cifs_sb->local_nls); + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc) { up(&pCifsFile->fh_sem); cFYI(1, ("cifs_open returned 0x%x ", rc)); @@ -465,8 +469,10 @@ int cifs_close(struct inode *inode, struct file *file) write_lock(&file->f_owner.lock); } } + write_lock(&GlobalSMBSeslock); list_del(&pSMBFile->flist); list_del(&pSMBFile->tlist); + write_unlock(&GlobalSMBSeslock); write_unlock(&file->f_owner.lock); kfree(pSMBFile->search_resume_name); kfree(file->private_data); @@ -506,7 +512,8 @@ int cifs_closedir(struct inode *inode, struct file *file) pTcon = cifs_sb->tcon; cFYI(1, ("Freeing private data in close dir")); - if (pCFileStruct->srch_inf.endOfSearch == FALSE) { + if ((pCFileStruct->srch_inf.endOfSearch == FALSE) && + (pCFileStruct->invalidHandle == FALSE)) { pCFileStruct->invalidHandle = TRUE; rc = CIFSFindClose(xid, pTcon, pCFileStruct->netfid); cFYI(1, ("Closing uncompleted readdir with rc %d", diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index d73b0aa8677..67094728826 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -44,7 +44,8 @@ int cifs_get_inode_info_unix(struct inode **pinode, cFYI(1, (" Getting info on %s ", search_path)); /* could have done a find first instead but this returns more info */ rc = CIFSSMBUnixQPathInfo(xid, pTcon, search_path, &findData, - cifs_sb->local_nls); + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); /* dump_mem("\nUnixQPathInfo return data", &findData, sizeof(findData)); */ if (rc) { @@ -63,7 +64,9 @@ int cifs_get_inode_info_unix(struct inode **pinode, strncat(tmp_path, search_path, MAX_PATHCONF); rc = connect_to_dfs_path(xid, pTcon->ses, /* treename + */ tmp_path, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); kfree(tmp_path); /* BB fix up inode etc. */ @@ -210,7 +213,8 @@ int cifs_get_inode_info(struct inode **pinode, pfindData = (FILE_ALL_INFO *)buf; /* could do find first instead but this returns more info */ rc = CIFSSMBQPathInfo(xid, pTcon, search_path, pfindData, - cifs_sb->local_nls); + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } /* dump_mem("\nQPathInfo return data",&findData, sizeof(findData)); */ if (rc) { @@ -230,7 +234,9 @@ int cifs_get_inode_info(struct inode **pinode, strncat(tmp_path, search_path, MAX_PATHCONF); rc = connect_to_dfs_path(xid, pTcon->ses, /* treename + */ tmp_path, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); kfree(tmp_path); /* BB fix up inode etc. */ } else if (rc) { @@ -268,7 +274,9 @@ int cifs_get_inode_info(struct inode **pinode, rc1 = CIFSGetSrvInodeNumber(xid, pTcon, search_path, &inode_num, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if(rc1) { cFYI(1,("GetSrvInodeNum rc %d", rc1)); /* BB EOPNOSUPP disable SERVER_INUM? */ @@ -410,7 +418,8 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) FreeXid(xid); return -ENOMEM; } - rc = CIFSSMBDelFile(xid, pTcon, full_path, cifs_sb->local_nls); + rc = CIFSSMBDelFile(xid, pTcon, full_path, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if (!rc) { direntry->d_inode->i_nlink--; @@ -422,10 +431,14 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, DELETE, CREATE_NOT_DIR | CREATE_DELETE_ON_CLOSE, - &netfid, &oplock, NULL, cifs_sb->local_nls); + &netfid, &oplock, NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc==0) { CIFSSMBRenameOpenFile(xid, pTcon, netfid, NULL, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); CIFSSMBClose(xid, pTcon, netfid); direntry->d_inode->i_nlink--; } @@ -439,7 +452,9 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) if (!(pTcon->ses->flags & CIFS_SES_NT4)) rc = CIFSSMBSetTimes(xid, pTcon, full_path, pinfo_buf, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); else rc = -EOPNOTSUPP; @@ -461,7 +476,9 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) FILE_OPEN, SYNCHRONIZE | FILE_WRITE_ATTRIBUTES, 0, &netfid, &oplock, NULL, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc==0) { rc = CIFSSMBSetFileTimes(xid, pTcon, pinfo_buf, @@ -472,8 +489,10 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) kfree(pinfo_buf); } if (rc==0) { - rc = CIFSSMBDelFile(xid, pTcon, full_path, - cifs_sb->local_nls); + rc = CIFSSMBDelFile(xid, pTcon, full_path, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (!rc) { direntry->d_inode->i_nlink--; } else if (rc == -ETXTBSY) { @@ -485,11 +504,15 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) CREATE_NOT_DIR | CREATE_DELETE_ON_CLOSE, &netfid, &oplock, NULL, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc==0) { CIFSSMBRenameOpenFile(xid, pTcon, netfid, NULL, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); CIFSSMBClose(xid, pTcon, netfid); direntry->d_inode->i_nlink--; } @@ -534,7 +557,8 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) return -ENOMEM; } /* BB add setting the equivalent of mode via CreateX w/ACLs */ - rc = CIFSSMBMkDir(xid, pTcon, full_path, cifs_sb->local_nls); + rc = CIFSSMBMkDir(xid, pTcon, full_path, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc) { cFYI(1, ("cifs_mkdir returned 0x%x ", rc)); d_drop(direntry); @@ -558,12 +582,16 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) (__u64)current->euid, (__u64)current->egid, 0 /* dev_t */, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, (__u64)-1, (__u64)-1, 0 /* dev_t */, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { /* BB to be implemented via Windows secrty descriptors @@ -600,7 +628,8 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry) return -ENOMEM; } - rc = CIFSSMBRmDir(xid, pTcon, full_path, cifs_sb->local_nls); + rc = CIFSSMBRmDir(xid, pTcon, full_path, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if (!rc) { inode->i_nlink--; @@ -653,7 +682,9 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, } rc = CIFSSMBRename(xid, pTcon, fromName, toName, - cifs_sb_source->local_nls); + cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc == -EEXIST) { /* check if they are the same file because rename of hardlinked files is a noop */ @@ -665,11 +696,16 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, if (info_buf_source != NULL) { info_buf_target = info_buf_source + 1; rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, - info_buf_source, cifs_sb_source->local_nls); + info_buf_source, cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc == 0) { rc = CIFSSMBUnixQPathInfo(xid, pTcon, toName, info_buf_target, - cifs_sb_target->local_nls); + cifs_sb_target->local_nls, + /* remap based on source sb */ + cifs_sb_source->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } if ((rc == 0) && (info_buf_source->UniqueId == @@ -685,7 +721,9 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, cifs_unlink(target_inode, target_direntry); rc = CIFSSMBRename(xid, pTcon, fromName, toName, - cifs_sb_source->local_nls); + cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags + & CIFS_MOUNT_MAP_SPECIAL_CHR); } kfree(info_buf_source); } /* if we can not get memory just leave rc as EEXIST */ @@ -705,10 +743,14 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, might not right be right access to request */ rc = CIFSSMBOpen(xid, pTcon, fromName, FILE_OPEN, GENERIC_READ, CREATE_NOT_DIR, &netfid, &oplock, NULL, - cifs_sb_source->local_nls); + cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc==0) { CIFSSMBRenameOpenFile(xid, pTcon, netfid, toName, - cifs_sb_source->local_nls); + cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); CIFSSMBClose(xid, pTcon, netfid); } } @@ -962,7 +1004,9 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) it by handle */ rc = CIFSSMBSetEOF(xid, pTcon, full_path, attrs->ia_size, FALSE, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); cFYI(1, (" SetEOF by path (setattrs) rc = %d", rc)); } @@ -999,7 +1043,9 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) if ((cifs_sb->tcon->ses->capabilities & CAP_UNIX) && (attrs->ia_valid & (ATTR_MODE | ATTR_GID | ATTR_UID))) rc = CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, uid, gid, - 0 /* dev_t */, cifs_sb->local_nls); + 0 /* dev_t */, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); else if (attrs->ia_valid & ATTR_MODE) { if ((mode & S_IWUGO) == 0) /* not writeable */ { if ((cifsInode->cifsAttrs & ATTR_READONLY) == 0) @@ -1048,7 +1094,9 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) via Handle (SetFileInfo) instead of by path */ if (!(pTcon->ses->flags & CIFS_SES_NT4)) rc = CIFSSMBSetTimes(xid, pTcon, full_path, &time_buf, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); else rc = -EOPNOTSUPP; @@ -1063,7 +1111,9 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, SYNCHRONIZE | FILE_WRITE_ATTRIBUTES, CREATE_NOT_DIR, &netfid, &oplock, - NULL, cifs_sb->local_nls); + NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc==0) { rc = CIFSSMBSetFileTimes(xid, pTcon, &time_buf, netfid); diff --git a/fs/cifs/ioctl.c b/fs/cifs/ioctl.c index b4b8e201d42..b0ea6687ab5 100644 --- a/fs/cifs/ioctl.c +++ b/fs/cifs/ioctl.c @@ -20,30 +20,93 @@ * along with this library; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ + #include <linux/fs.h> #include <linux/ext2_fs.h> #include "cifspdu.h" #include "cifsglob.h" #include "cifsproto.h" #include "cifs_debug.h" +#include "cifsfs.h" + +#define CIFS_IOC_CHECKUMOUNT _IO(0xCF, 2) int cifs_ioctl (struct inode * inode, struct file * filep, unsigned int command, unsigned long arg) { int rc = -ENOTTY; /* strange error - but the precedent */ + int xid; + struct cifs_sb_info *cifs_sb; +#ifdef CONFIG_CIFS_POSIX + __u64 ExtAttrBits = 0; + __u64 ExtAttrMask = 0; + __u64 caps; + struct cifsTconInfo *tcon; + struct cifsFileInfo *pSMBFile = + (struct cifsFileInfo *)filep->private_data; +#endif /* CONFIG_CIFS_POSIX */ + + xid = GetXid(); + + cFYI(1,("ioctl file %p cmd %u arg %lu",filep,command,arg)); + + cifs_sb = CIFS_SB(inode->i_sb); + #ifdef CONFIG_CIFS_POSIX - cFYI(1,("ioctl file %p cmd %u arg %lu",filep,command,arg)); + tcon = cifs_sb->tcon; + if(tcon) + caps = le64_to_cpu(tcon->fsUnixInfo.Capability); + else { + rc = -EIO; + FreeXid(xid); + return -EIO; + } +#endif /* CONFIG_CIFS_POSIX */ + switch(command) { + case CIFS_IOC_CHECKUMOUNT: + cFYI(1,("User unmount attempted")); + if(cifs_sb->mnt_uid == current->uid) + rc = 0; + else { + rc = -EACCES; + cFYI(1,("uids do not match")); + } + break; +#ifdef CONFIG_CIFS_POSIX case EXT2_IOC_GETFLAGS: - cFYI(1,("get flags not implemented yet")); - return -EOPNOTSUPP; + if(CIFS_UNIX_EXTATTR_CAP & caps) { + if (pSMBFile == NULL) + break; + rc = CIFSGetExtAttr(xid, tcon, pSMBFile->netfid, + &ExtAttrBits, &ExtAttrMask); + if(rc == 0) + rc = put_user(ExtAttrBits & + EXT2_FL_USER_VISIBLE, + (int __user *)arg); + } + break; + case EXT2_IOC_SETFLAGS: + if(CIFS_UNIX_EXTATTR_CAP & caps) { + if(get_user(ExtAttrBits,(int __user *)arg)) { + rc = -EFAULT; + break; + } + if (pSMBFile == NULL) + break; + /* rc= CIFSGetExtAttr(xid,tcon,pSMBFile->netfid, + extAttrBits, &ExtAttrMask);*/ + + } cFYI(1,("set flags not implemented yet")); - return -EOPNOTSUPP; + break; +#endif /* CONFIG_CIFS_POSIX */ default: cFYI(1,("unsupported ioctl")); - return rc; + break; } -#endif /* CONFIG_CIFS_POSIX */ + + FreeXid(xid); return rc; } diff --git a/fs/cifs/link.c b/fs/cifs/link.c index 1455810ba1c..bde0fabfece 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -59,10 +59,14 @@ cifs_hardlink(struct dentry *old_file, struct inode *inode, if (cifs_sb_target->tcon->ses->capabilities & CAP_UNIX) rc = CIFSUnixCreateHardLink(xid, pTcon, fromName, toName, - cifs_sb_target->local_nls); + cifs_sb_target->local_nls, + cifs_sb_target->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); else { rc = CIFSCreateHardLink(xid, pTcon, fromName, toName, - cifs_sb_target->local_nls); + cifs_sb_target->local_nls, + cifs_sb_target->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if(rc == -EIO) rc = -EOPNOTSUPP; } @@ -260,7 +264,10 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) cifs_sb->local_nls); else { rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, GENERIC_READ, - OPEN_REPARSE_POINT,&fid, &oplock, NULL, cifs_sb->local_nls); + OPEN_REPARSE_POINT,&fid, &oplock, NULL, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if(!rc) { rc = CIFSSMBQueryReparseLinkInfo(xid, pTcon, full_path, tmpbuffer, @@ -279,7 +286,10 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) strncpy(tmp_path, pTcon->treeName, MAX_TREE_SIZE); strncat(tmp_path, full_path, MAX_PATHCONF); rc = get_dfs_path(xid, pTcon->ses, tmp_path, - cifs_sb->local_nls, &num_referrals, &referrals); + cifs_sb->local_nls, + &num_referrals, &referrals, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); cFYI(1,("Get DFS for %s rc = %d ",tmp_path, rc)); if((num_referrals == 0) && (rc == 0)) rc = -EACCES; diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 7b38d3059a8..db14b503d89 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -28,6 +28,7 @@ #include "cifs_debug.h" #include "smberr.h" #include "nterr.h" +#include "cifs_unicode.h" extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; @@ -451,25 +452,30 @@ is_valid_oplock_break(struct smb_hdr *buf) atomic_inc(&tcon->num_oplock_brks); #endif list_for_each(tmp1,&tcon->openFileList){ - netfile = list_entry(tmp1,struct cifsFileInfo,tlist); + netfile = list_entry(tmp1,struct cifsFileInfo, + tlist); if(pSMB->Fid == netfile->netfid) { struct cifsInodeInfo *pCifsInode; read_unlock(&GlobalSMBSeslock); - cFYI(1,("Matching file id, processing oplock break")); + cFYI(1,("file id match, oplock break")); pCifsInode = CIFS_I(netfile->pInode); pCifsInode->clientCanCacheAll = FALSE; if(pSMB->OplockLevel == 0) - pCifsInode->clientCanCacheRead = FALSE; + pCifsInode->clientCanCacheRead + = FALSE; pCifsInode->oplockPending = TRUE; - AllocOplockQEntry(netfile->pInode, netfile->netfid, tcon); + AllocOplockQEntry(netfile->pInode, + netfile->netfid, + tcon); cFYI(1,("about to wake up oplock thd")); - wake_up_process(oplockThread); + if(oplockThread) + wake_up_process(oplockThread); return TRUE; } } read_unlock(&GlobalSMBSeslock); - cFYI(1,("No matching file for oplock break on connection")); + cFYI(1,("No matching file for oplock break")); return TRUE; } } @@ -490,7 +496,7 @@ dump_smb(struct smb_hdr *smb_buf, int smb_buf_length) buffer = (unsigned char *) smb_buf; for (i = 0, j = 0; i < smb_buf_length; i++, j++) { - if (i % 8 == 0) { /* we have reached the beginning of line */ + if (i % 8 == 0) { /* have reached the beginning of line */ printk(KERN_DEBUG "| "); j = 0; } @@ -501,7 +507,7 @@ dump_smb(struct smb_hdr *smb_buf, int smb_buf_length) else debug_line[1 + (2 * j)] = '_'; - if (i % 8 == 7) { /* we have reached end of line, time to print ascii */ + if (i % 8 == 7) { /* reached end of line, time to print ascii */ debug_line[16] = 0; printk(" | %s\n", debug_line); } @@ -514,3 +520,141 @@ dump_smb(struct smb_hdr *smb_buf, int smb_buf_length) printk( " | %s\n", debug_line); return; } + +/* Windows maps these to the user defined 16 bit Unicode range since they are + reserved symbols (along with \ and /), otherwise illegal to store + in filenames in NTFS */ +#define UNI_ASTERIK (__u16) ('*' + 0xF000) +#define UNI_QUESTION (__u16) ('?' + 0xF000) +#define UNI_COLON (__u16) (':' + 0xF000) +#define UNI_GRTRTHAN (__u16) ('>' + 0xF000) +#define UNI_LESSTHAN (__u16) ('<' + 0xF000) +#define UNI_PIPE (__u16) ('|' + 0xF000) +#define UNI_SLASH (__u16) ('\\' + 0xF000) + +/* Convert 16 bit Unicode pathname from wire format to string in current code + page. Conversion may involve remapping up the seven characters that are + only legal in POSIX-like OS (if they are present in the string). Path + names are little endian 16 bit Unicode on the wire */ +int +cifs_convertUCSpath(char *target, const __le16 * source, int maxlen, + const struct nls_table * cp) +{ + int i,j,len; + __u16 src_char; + + for(i = 0, j = 0; i < maxlen; i++) { + src_char = le16_to_cpu(source[i]); + switch (src_char) { + case 0: + goto cUCS_out; /* BB check this BB */ + case UNI_COLON: + target[j] = ':'; + break; + case UNI_ASTERIK: + target[j] = '*'; + break; + case UNI_QUESTION: + target[j] = '?'; + break; + /* BB We can not handle remapping slash until + all the calls to build_path_from_dentry + are modified, as they use slash as separator BB */ + /* case UNI_SLASH: + target[j] = '\\'; + break;*/ + case UNI_PIPE: + target[j] = '|'; + break; + case UNI_GRTRTHAN: + target[j] = '>'; + break; + case UNI_LESSTHAN: + target[j] = '<'; + default: + len = cp->uni2char(src_char, &target[j], + NLS_MAX_CHARSET_SIZE); + if(len > 0) { + j += len; + continue; + } else { + target[j] = '?'; + } + } + j++; + /* make sure we do not overrun callers allocated temp buffer */ + if(j >= (2 * NAME_MAX)) + break; + } +cUCS_out: + target[j] = 0; + return j; +} + +/* Convert 16 bit Unicode pathname to wire format from string in current code + page. Conversion may involve remapping up the seven characters that are + only legal in POSIX-like OS (if they are present in the string). Path + names are little endian 16 bit Unicode on the wire */ +int +cifsConvertToUCS(__le16 * target, const char *source, int maxlen, + const struct nls_table * cp, int mapChars) +{ + int i,j,charlen; + int len_remaining = maxlen; + char src_char; + + if(!mapChars) + return cifs_strtoUCS((wchar_t *) target, source, PATH_MAX, cp); + + for(i = 0, j = 0; i < maxlen; j++) { + src_char = source[i]; + switch (src_char) { + case 0: + goto ctoUCS_out; + case ':': + target[j] = cpu_to_le16(UNI_COLON); + break; + case '*': + target[j] = cpu_to_le16(UNI_ASTERIK); + break; + case '?': + target[j] = cpu_to_le16(UNI_QUESTION); + break; + case '<': + target[j] = cpu_to_le16(UNI_LESSTHAN); + break; + case '>': + target[j] = cpu_to_le16(UNI_GRTRTHAN); + break; + case '|': + target[j] = cpu_to_le16(UNI_PIPE); + break; + /* BB We can not handle remapping slash until + all the calls to build_path_from_dentry + are modified, as they use slash as separator BB */ + /* case '\\': + target[j] = cpu_to_le16(UNI_SLASH); + break;*/ + default: + charlen = cp->char2uni(source+i, + len_remaining, target+j); + /* if no match, use question mark, which + at least in some cases servers as wild card */ + if(charlen < 1) { + target[j] = cpu_to_le16(0x003f); + charlen = 1; + } + len_remaining -= charlen; + /* character may take more than one byte in the + the source string, but will take exactly two + bytes in the target string */ + i+= charlen; + continue; + } + i++; /* move to next char in source string */ + len_remaining--; + } + +ctoUCS_out: + return i; +} diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index 4e34c89cec5..a92af41d441 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -78,6 +78,7 @@ static const struct smb_to_posix_error mapping_table_ERRDOS[] = { {ErrQuota, -EDQUOT}, {ErrNotALink, -ENOLINK}, {ERRnetlogonNotStarted,-ENOPROTOOPT}, + {ErrTooManyLinks,-EMLINK}, {0, 0} }; @@ -206,7 +207,7 @@ static const struct { { ERRDOS, ERRgeneral, NT_STATUS_UNSUCCESSFUL}, { ERRDOS, ERRbadfunc, NT_STATUS_NOT_IMPLEMENTED}, { - ERRDOS, 87, NT_STATUS_INVALID_INFO_CLASS}, { + ERRDOS, ERRinvlevel, NT_STATUS_INVALID_INFO_CLASS}, { ERRDOS, 24, NT_STATUS_INFO_LENGTH_MISMATCH}, { ERRHRD, ERRgeneral, NT_STATUS_ACCESS_VIOLATION}, { ERRHRD, ERRgeneral, NT_STATUS_IN_PAGE_ERROR}, { @@ -742,7 +743,7 @@ static const struct { ERRDOS, 182, NT_STATUS_DRIVER_ORDINAL_NOT_FOUND}, { ERRDOS, 127, NT_STATUS_DRIVER_ENTRYPOINT_NOT_FOUND}, { ERRDOS, 288, NT_STATUS_RESOURCE_NOT_OWNED}, { - ERRHRD, ERRgeneral, NT_STATUS_TOO_MANY_LINKS}, { + ERRDOS, ErrTooManyLinks, NT_STATUS_TOO_MANY_LINKS}, { ERRHRD, ERRgeneral, NT_STATUS_QUOTA_LIST_INCONSISTENT}, { ERRHRD, ERRgeneral, NT_STATUS_FILE_IS_OFFLINE}, { ERRDOS, 21, 0xc000026e}, { diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index f8bea395ec9..22557716f9a 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -3,7 +3,7 @@ * * Directory search handling * - * Copyright (C) International Business Machines Corp., 2004 + * Copyright (C) International Business Machines Corp., 2004, 2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -65,14 +65,14 @@ static int construct_dentry(struct qstr *qstring, struct file *file, struct cifsTconInfo *pTcon; int rc = 0; - cFYI(1, ("For %s ", qstring->name)); + cFYI(1, ("For %s", qstring->name)); cifs_sb = CIFS_SB(file->f_dentry->d_sb); pTcon = cifs_sb->tcon; qstring->hash = full_name_hash(qstring->name, qstring->len); tmp_dentry = d_lookup(file->f_dentry, qstring); if (tmp_dentry) { - cFYI(0, (" existing dentry with inode 0x%p", tmp_dentry->d_inode)); + cFYI(0, ("existing dentry with inode 0x%p", tmp_dentry->d_inode)); *ptmp_inode = tmp_dentry->d_inode; /* BB overwrite old name? i.e. tmp_dentry->d_name and tmp_dentry->d_name.len??*/ if(*ptmp_inode == NULL) { @@ -105,8 +105,11 @@ static int construct_dentry(struct qstr *qstring, struct file *file, } static void fill_in_inode(struct inode *tmp_inode, - FILE_DIRECTORY_INFO *pfindData, int *pobject_type) + FILE_DIRECTORY_INFO *pfindData, int *pobject_type, int isNewInode) { + loff_t local_size; + struct timespec local_mtime; + struct cifsInodeInfo *cifsInfo = CIFS_I(tmp_inode); struct cifs_sb_info *cifs_sb = CIFS_SB(tmp_inode->i_sb); __u32 attr = le32_to_cpu(pfindData->ExtFileAttributes); @@ -116,6 +119,10 @@ static void fill_in_inode(struct inode *tmp_inode, cifsInfo->cifsAttrs = attr; cifsInfo->time = jiffies; + /* save mtime and size */ + local_mtime = tmp_inode->i_mtime; + local_size = tmp_inode->i_size; + /* Linux can not store file creation time unfortunately so ignore it */ tmp_inode->i_atime = cifs_NTtimeToUnix(le64_to_cpu(pfindData->LastAccessTime)); @@ -134,7 +141,6 @@ static void fill_in_inode(struct inode *tmp_inode, tmp_inode->i_mode = cifs_sb->mnt_file_mode; } - cFYI(0,("CIFS FFIRST: Attributes came in as 0x%x",attr)); if (attr & ATTR_DIRECTORY) { *pobject_type = DT_DIR; /* override default perms since we do not lock dirs */ @@ -175,30 +181,46 @@ static void fill_in_inode(struct inode *tmp_inode, (unsigned long)tmp_inode->i_size, tmp_inode->i_blocks, tmp_inode->i_blksize)); if (S_ISREG(tmp_inode->i_mode)) { - cFYI(1, (" File inode ")); + cFYI(1, ("File inode")); tmp_inode->i_op = &cifs_file_inode_ops; if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) tmp_inode->i_fop = &cifs_file_direct_ops; else tmp_inode->i_fop = &cifs_file_ops; tmp_inode->i_data.a_ops = &cifs_addr_ops; + + if(isNewInode) + return; /* No sense invalidating pages for new inode since we + have not started caching readahead file data yet */ + + if (timespec_equal(&tmp_inode->i_mtime, &local_mtime) && + (local_size == tmp_inode->i_size)) { + cFYI(1, ("inode exists but unchanged")); + } else { + /* file may have changed on server */ + cFYI(1, ("invalidate inode, readdir detected change")); + invalidate_remote_inode(tmp_inode); + } } else if (S_ISDIR(tmp_inode->i_mode)) { - cFYI(1, (" Directory inode")); + cFYI(1, ("Directory inode")); tmp_inode->i_op = &cifs_dir_inode_ops; tmp_inode->i_fop = &cifs_dir_ops; } else if (S_ISLNK(tmp_inode->i_mode)) { - cFYI(1, (" Symbolic Link inode ")); + cFYI(1, ("Symbolic Link inode")); tmp_inode->i_op = &cifs_symlink_inode_ops; } else { - cFYI(1, (" Init special inode ")); + cFYI(1, ("Init special inode")); init_special_inode(tmp_inode, tmp_inode->i_mode, tmp_inode->i_rdev); } } static void unix_fill_in_inode(struct inode *tmp_inode, - FILE_UNIX_INFO *pfindData, int *pobject_type) + FILE_UNIX_INFO *pfindData, int *pobject_type, int isNewInode) { + loff_t local_size; + struct timespec local_mtime; + struct cifsInodeInfo *cifsInfo = CIFS_I(tmp_inode); struct cifs_sb_info *cifs_sb = CIFS_SB(tmp_inode->i_sb); @@ -208,6 +230,10 @@ static void unix_fill_in_inode(struct inode *tmp_inode, cifsInfo->time = jiffies; atomic_inc(&cifsInfo->inUse); + /* save mtime and size */ + local_mtime = tmp_inode->i_mtime; + local_size = tmp_inode->i_size; + tmp_inode->i_atime = cifs_NTtimeToUnix(le64_to_cpu(pfindData->LastAccessTime)); tmp_inode->i_mtime = @@ -265,6 +291,19 @@ static void unix_fill_in_inode(struct inode *tmp_inode, else tmp_inode->i_fop = &cifs_file_ops; tmp_inode->i_data.a_ops = &cifs_addr_ops; + + if(isNewInode) + return; /* No sense invalidating pages for new inode since we + have not started caching readahead file data yet */ + + if (timespec_equal(&tmp_inode->i_mtime, &local_mtime) && + (local_size == tmp_inode->i_size)) { + cFYI(1, ("inode exists but unchanged")); + } else { + /* file may have changed on server */ + cFYI(1, ("invalidate inode, readdir detected change")); + invalidate_remote_inode(tmp_inode); + } } else if (S_ISDIR(tmp_inode->i_mode)) { cFYI(1, ("Directory inode")); tmp_inode->i_op = &cifs_dir_inode_ops; @@ -314,15 +353,16 @@ static int initiate_cifs_search(const int xid, struct file *file) return -EINVAL; down(&file->f_dentry->d_sb->s_vfs_rename_sem); - full_path = build_wildcard_path_from_dentry(file->f_dentry); + full_path = build_path_from_dentry(file->f_dentry); up(&file->f_dentry->d_sb->s_vfs_rename_sem); if(full_path == NULL) { return -ENOMEM; } - cFYI(1, ("Full path: %s start at: %lld ", full_path, file->f_pos)); + cFYI(1, ("Full path: %s start at: %lld", full_path, file->f_pos)); +ffirst_retry: /* test for Unix extensions */ if (pTcon->ses->capabilities & CAP_UNIX) { cifsFile->srch_inf.info_level = SMB_FIND_FILE_UNIX; @@ -332,10 +372,16 @@ static int initiate_cifs_search(const int xid, struct file *file) cifsFile->srch_inf.info_level = SMB_FIND_FILE_DIRECTORY_INFO; } - rc = CIFSFindFirst(xid, pTcon,full_path,cifs_sb->local_nls, - &cifsFile->netfid, &cifsFile->srch_inf); + rc = CIFSFindFirst(xid, pTcon,full_path,cifs_sb->local_nls, + &cifsFile->netfid, &cifsFile->srch_inf, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if(rc == 0) cifsFile->invalidHandle = FALSE; + if((rc == -EOPNOTSUPP) && + (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) { + cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SERVER_INUM; + goto ffirst_retry; + } kfree(full_path); return rc; } @@ -363,10 +409,15 @@ static char *nxt_dir_entry(char *old_entry, char *end_of_smb) cFYI(1,("new entry %p old entry %p",new_entry,old_entry)); /* validate that new_entry is not past end of SMB */ if(new_entry >= end_of_smb) { - cFYI(1,("search entry %p began after end of SMB %p old entry %p", - new_entry,end_of_smb,old_entry)); + cERROR(1, + ("search entry %p began after end of SMB %p old entry %p", + new_entry, end_of_smb, old_entry)); return NULL; - } else + } else if (new_entry + sizeof(FILE_DIRECTORY_INFO) > end_of_smb) { + cERROR(1,("search entry %p extends after end of SMB %p", + new_entry, end_of_smb)); + return NULL; + } else return new_entry; } @@ -594,7 +645,12 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, if(unicode) { /* BB fixme - test with long names */ /* Note converted filename can be longer than in unicode */ - pqst->len = cifs_strfromUCS_le((char *)pqst->name,(wchar_t *)filename,len/2,nlt); + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR) + pqst->len = cifs_convertUCSpath((char *)pqst->name, + (__le16 *)filename, len/2, nlt); + else + pqst->len = cifs_strfromUCS_le((char *)pqst->name, + (wchar_t *)filename,len/2,nlt); } else { pqst->name = filename; pqst->len = len; @@ -654,10 +710,15 @@ static int cifs_filldir(char *pfindEntry, struct file *file, insert_inode_hash(tmp_inode); } + /* we pass in rc below, indicating whether it is a new inode, + so we can figure out whether to invalidate the inode cached + data if the file has changed */ if(pCifsF->srch_inf.info_level == SMB_FIND_FILE_UNIX) { - unix_fill_in_inode(tmp_inode,(FILE_UNIX_INFO *)pfindEntry,&obj_type); + unix_fill_in_inode(tmp_inode, + (FILE_UNIX_INFO *)pfindEntry,&obj_type, rc); } else { - fill_in_inode(tmp_inode,(FILE_DIRECTORY_INFO *)pfindEntry,&obj_type); + fill_in_inode(tmp_inode, + (FILE_DIRECTORY_INFO *)pfindEntry,&obj_type, rc); } rc = filldir(direntry,qstring.name,qstring.len,file->f_pos,tmp_inode->i_ino,obj_type); @@ -823,7 +884,11 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) end_of_smb = cifsFile->srch_inf.ntwrk_buf_start + smbCalcSize((struct smb_hdr *) cifsFile->srch_inf.ntwrk_buf_start); - tmp_buf = kmalloc(NAME_MAX+1,GFP_KERNEL); + /* To be safe - for UCS to UTF-8 with strings loaded + with the rare long characters alloc more to account for + such multibyte target UTF-8 characters. cifs_unicode.c, + which actually does the conversion, has the same limit */ + tmp_buf = kmalloc((2 * NAME_MAX) + 4, GFP_KERNEL); for(i=0;(i<num_to_fill) && (rc == 0);i++) { if(current_entry == NULL) { /* evaluate whether this case is an error */ @@ -832,19 +897,6 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) break; } - /* BB FIXME - need to enable the below code BB */ - - /* if((!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) || - (cifsFile->srch_inf.info_level != - something that supports server inodes)) { - create dentry - create inode - fill in inode new_inode (getting local i_ino) - } - also create local inode for performance reasons (so we - have a cache of inode metadata) unless this new mount - parm says otherwise */ - rc = cifs_filldir(current_entry, file, filldir, direntry,tmp_buf); file->f_pos++; diff --git a/fs/cifs/smberr.h b/fs/cifs/smberr.h index e21f1384661..cd41c67ff8d 100644 --- a/fs/cifs/smberr.h +++ b/fs/cifs/smberr.h @@ -22,94 +22,159 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#define SUCCESS 0 /* The request was successful. */ -#define ERRDOS 0x01 /* Error is from the core DOS operating system set */ -#define ERRSRV 0x02 /* Error is generated by the file server daemon */ -#define ERRHRD 0x03 /* Error is a hardware error. */ -#define ERRCMD 0xFF /* Command was not in the "SMB" format. */ +#define SUCCESS 0x00 /* The request was successful. */ +#define ERRDOS 0x01 /* Error is from the core DOS operating system set */ +#define ERRSRV 0x02 /* Error is generated by the file server daemon */ +#define ERRHRD 0x03 /* Error is a hardware error. */ +#define ERRCMD 0xFF /* Command was not in the "SMB" format. */ /* The following error codes may be generated with the SUCCESS error class.*/ -#define SUCCESS 0 /* The request was successful. */ +/*#define SUCCESS 0 The request was successful. */ /* The following error codes may be generated with the ERRDOS error class.*/ -#define ERRbadfunc 1 /* Invalid function. The server did not recognize or could not perform a system call generated by the server, e.g., set the DIRECTORY attribute on a data file, invalid seek mode. */ -#define ERRbadfile 2 /*File not found. The last component of a file's pathname could not be found. */ -#define ERRbadpath 3 /* Directory invalid. A directory component in a pathname could not be found. */ -#define ERRnofids 4 /* Too many open files. The server has no file handles available. */ -#define ERRnoaccess 5 /* Access denied, the client's context does not permit the requested function. This includes the following conditions: invalid rename command, write to Fid open for read only, read on Fid open for write only, attempt to delete a non-empty directory */ -#define ERRbadfid 6 /* Invalid file handle. The file handle specified was not recognized by the server. */ -#define ERRbadmcb 7 /* Memory control blocks destroyed. */ -#define ERRnomem 8 /* Insufficient server memory to perform the requested function. */ -#define ERRbadmem 9 /* Invalid memory block address. */ -#define ERRbadenv 10 /* Invalid environment. */ -#define ERRbadformat 11 /* Invalid format. */ -#define ERRbadaccess 12 /* Invalid open mode. */ -#define ERRbaddata 13 /* Invalid data (generated only by IOCTL calls within the server). */ -#define ERRbaddrive 15 /* Invalid drive specified. */ -#define ERRremcd 16 /* A Delete Directory request attempted to remove the server's current directory. */ -#define ERRdiffdevice 17 /* Not same device (e.g., a cross volume rename was attempted */ -#define ERRnofiles 18 /* A File Search command can find no more files matching the specified criteria. */ -#define ERRgeneral 31 -#define ERRbadshare 32 /* The sharing mode specified for an Open conflicts with existing FIDs on the file. */ -#define ERRlock 33 /* A Lock request conflicted with an existing lock or specified an invalid mode, or an Unlock requested attempted to remove a lock held by another process. */ -#define ERRunsup 50 -#define ERRnosuchshare 67 -#define ERRfilexists 80 /* The file named in the request already exists. */ -#define ERRinvparm 87 -#define ERRdiskfull 112 -#define ERRinvname 123 -#define ERRinvlevel 124 -#define ERRdirnotempty 145 -#define ERRnotlocked 158 -#define ERRalreadyexists 183 -#define ERRbadpipe 230 -#define ERRpipebusy 231 -#define ERRpipeclosing 232 -#define ERRnotconnected 233 -#define ERRmoredata 234 -#define ERReasnotsupported 282 -#define ErrQuota 0x200 /* The operation would cause a quota limit to be exceeded. */ -#define ErrNotALink 0x201 /* A link operation was performed on a pathname that - was not a link. */ +#define ERRbadfunc 1 /* Invalid function. The server did not + recognize or could not perform a + system call generated by the server, + e.g., set the DIRECTORY attribute on + a data file, invalid seek mode. */ +#define ERRbadfile 2 /* File not found. The last component + of a file's pathname could not be + found. */ +#define ERRbadpath 3 /* Directory invalid. A directory + component in a pathname could not be + found. */ +#define ERRnofids 4 /* Too many open files. The server has + no file handles available. */ +#define ERRnoaccess 5 /* Access denied, the client's context + does not permit the requested + function. This includes the + following conditions: invalid rename + command, write to Fid open for read + only, read on Fid open for write + only, attempt to delete a non-empty + directory */ +#define ERRbadfid 6 /* Invalid file handle. The file handle + specified was not recognized by the + server. */ +#define ERRbadmcb 7 /* Memory control blocks destroyed. */ +#define ERRnomem 8 /* Insufficient server memory to + perform the requested function. */ +#define ERRbadmem 9 /* Invalid memory block address. */ +#define ERRbadenv 10 /* Invalid environment. */ +#define ERRbadformat 11 /* Invalid format. */ +#define ERRbadaccess 12 /* Invalid open mode. */ +#define ERRbaddata 13 /* Invalid data (generated only by + IOCTL calls within the server). */ +#define ERRbaddrive 15 /* Invalid drive specified. */ +#define ERRremcd 16 /* A Delete Directory request attempted + to remove the server's current + directory. */ +#define ERRdiffdevice 17 /* Not same device (e.g., a cross + volume rename was attempted */ +#define ERRnofiles 18 /* A File Search command can find no + more files matching the specified + criteria. */ +#define ERRgeneral 31 +#define ERRbadshare 32 /* The sharing mode specified for an + Open conflicts with existing FIDs on + the file. */ +#define ERRlock 33 /* A Lock request conflicted with an + existing lock or specified an + invalid mode, or an Unlock requested + attempted to remove a lock held by + another process. */ +#define ERRunsup 50 +#define ERRnosuchshare 67 +#define ERRfilexists 80 /* The file named in the request + already exists. */ +#define ERRinvparm 87 +#define ERRdiskfull 112 +#define ERRinvname 123 +#define ERRinvlevel 124 +#define ERRdirnotempty 145 +#define ERRnotlocked 158 +#define ERRalreadyexists 183 +#define ERRbadpipe 230 +#define ERRpipebusy 231 +#define ERRpipeclosing 232 +#define ERRnotconnected 233 +#define ERRmoredata 234 +#define ERReasnotsupported 282 +#define ErrQuota 0x200 /* The operation would cause a quota + limit to be exceeded. */ +#define ErrNotALink 0x201 /* A link operation was performed on a + pathname that was not a link. */ -/* Following error codes may be generated with the ERRSRV error -class.*/ +/* Below errors are used internally (do not come over the wire) for passthrough + from STATUS codes to POSIX only */ +#define ErrTooManyLinks 0xFFFE -#define ERRerror 1 /* Non-specific error code. It is returned under the following conditions: resource other than disk space exhausted (e.g. TIDs), first SMB command was not negotiate, multiple negotiates attempted, and internal server error. */ -#define ERRbadpw 2 /* Bad password - name/password pair in a TreeConnect or Session Setup are invalid. */ -#define ERRbadtype 3 /* used for indicating DFS referral needed */ -#define ERRaccess 4 /* The client does not have the necessary access rights within the specified context for requested function. */ -#define ERRinvtid 5 /* The Tid specified in a command was invalid. */ -#define ERRinvnetname 6 /* Invalid network name in tree connect. */ -#define ERRinvdevice 7 /* Invalid device - printer request made to non-printer connection or non-printer request made to printer connection. */ -#define ERRqfull 49 /* Print queue full (files) -- returned by open print file. */ -#define ERRqtoobig 50 /* Print queue full -- no space. */ -#define ERRqeof 51 /* EOF on print queue dump */ -#define ERRinvpfid 52 /* Invalid print file FID. */ -#define ERRsmbcmd 64 /* The server did not recognize the command received. */ -#define ERRsrverror 65 /* The server encountered an internal error, e.g., system file unavailable. */ -#define ERRbadBID 66 /* (obsolete) */ -#define ERRfilespecs 67 /* The Fid and pathname parameters contained an invalid combination of values. */ -#define ERRbadLink 68 /* (obsolete) */ -#define ERRbadpermits 69 /* The access permissions specified for a file or directory are not a valid combination. */ -#define ERRbadPID 70 -#define ERRsetattrmode 71 /* attribute (mode) is invalid */ -#define ERRpaused 81 /* Server is paused */ -#define ERRmsgoff 82 /* reserved - messaging off */ -#define ERRnoroom 83 /* reserved - no room for message */ -#define ERRrmuns 87 /* reserved - too many remote names */ -#define ERRtimeout 88 /* operation timed out */ -#define ERRnoresource 89 /* No resources available for request */ -#define ERRtoomanyuids 90 /* Too many UIDs active on this session */ -#define ERRbaduid 91 /* The UID is not known as a valid user */ -#define ERRusempx 250 /* temporarily unable to use raw */ -#define ERRusestd 251 /* temporarily unable to use either raw or mpx */ -#define ERR_NOTIFY_ENUM_DIR 1024 -#define ERRaccountexpired 2239 -#define ERRbadclient 2240 -#define ERRbadLogonTime 2241 -#define ERRpasswordExpired 2242 -#define ERRnetlogonNotStarted 2455 -#define ERRnosupport 0xFFFF +/* Following error codes may be generated with the ERRSRV error class.*/ + +#define ERRerror 1 /* Non-specific error code. It is + returned under the following + conditions: resource other than disk + space exhausted (e.g. TIDs), first + SMB command was not negotiate, + multiple negotiates attempted, and + internal server error. */ +#define ERRbadpw 2 /* Bad password - name/password pair in + a TreeConnect or Session Setup are + invalid. */ +#define ERRbadtype 3 /* used for indicating DFS referral + needed */ +#define ERRaccess 4 /* The client does not have the + necessary access rights within the + specified context for requested + function. */ +#define ERRinvtid 5 /* The Tid specified in a command was + invalid. */ +#define ERRinvnetname 6 /* Invalid network name in tree + connect. */ +#define ERRinvdevice 7 /* Invalid device - printer request + made to non-printer connection or + non-printer request made to printer + connection. */ +#define ERRqfull 49 /* Print queue full (files) -- returned + by open print file. */ +#define ERRqtoobig 50 /* Print queue full -- no space. */ +#define ERRqeof 51 /* EOF on print queue dump */ +#define ERRinvpfid 52 /* Invalid print file FID. */ +#define ERRsmbcmd 64 /* The server did not recognize the + command received. */ +#define ERRsrverror 65 /* The server encountered an internal + error, e.g., system file + unavailable. */ +#define ERRbadBID 66 /* (obsolete) */ +#define ERRfilespecs 67 /* The Fid and pathname parameters + contained an invalid combination of + values. */ +#define ERRbadLink 68 /* (obsolete) */ +#define ERRbadpermits 69 /* The access permissions specified for + a file or directory are not a valid + combination. */ +#define ERRbadPID 70 +#define ERRsetattrmode 71 /* attribute (mode) is invalid */ +#define ERRpaused 81 /* Server is paused */ +#define ERRmsgoff 82 /* reserved - messaging off */ +#define ERRnoroom 83 /* reserved - no room for message */ +#define ERRrmuns 87 /* reserved - too many remote names */ +#define ERRtimeout 88 /* operation timed out */ +#define ERRnoresource 89 /* No resources available for request + */ +#define ERRtoomanyuids 90 /* Too many UIDs active on this session + */ +#define ERRbaduid 91 /* The UID is not known as a valid user + */ +#define ERRusempx 250 /* temporarily unable to use raw */ +#define ERRusestd 251 /* temporarily unable to use either raw + or mpx */ +#define ERR_NOTIFY_ENUM_DIR 1024 +#define ERRaccountexpired 2239 +#define ERRbadclient 2240 +#define ERRbadLogonTime 2241 +#define ERRpasswordExpired 2242 +#define ERRnetlogonNotStarted 2455 +#define ERRnosupport 0xFFFF diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index af13e526b15..0046c219833 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -1,7 +1,7 @@ /* * fs/cifs/transport.c * - * Copyright (C) International Business Machines Corp., 2002,2004 + * Copyright (C) International Business Machines Corp., 2002,2005 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -41,7 +41,7 @@ AllocMidQEntry(struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) struct mid_q_entry *temp; if (ses == NULL) { - cERROR(1, ("Null session passed in to AllocMidQEntry ")); + cERROR(1, ("Null session passed in to AllocMidQEntry")); return NULL; } if (ses->server == NULL) { @@ -79,7 +79,10 @@ DeleteMidQEntry(struct mid_q_entry *midEntry) list_del(&midEntry->qhead); atomic_dec(&midCount); spin_unlock(&GlobalMid_Lock); - cifs_buf_release(midEntry->resp_buf); + if(midEntry->largeBuf) + cifs_buf_release(midEntry->resp_buf); + else + cifs_small_buf_release(midEntry->resp_buf); mempool_free(midEntry, cifs_mid_poolp); } @@ -182,14 +185,14 @@ smb_send(struct socket *ssocket, struct smb_hdr *smb_buffer, int smb_sendv(struct socket *ssocket, struct smb_hdr *smb_buffer, - unsigned int smb_buf_length, struct kvec * write_vector /* page list */, struct sockaddr *sin) + unsigned int smb_buf_length, struct kvec * write_vector + /* page list */, struct sockaddr *sin) { int rc = 0; int i = 0; struct msghdr smb_msg; number_of_pages += 1; /* account for SMB header */ struct kvec * piov = kmalloc(number_of_pages * sizeof(struct kvec)); - if(i=0;i<num_pages-1;i++ unsigned len = smb_buf_length + 4; if(ssocket == NULL) @@ -213,7 +216,8 @@ smb_sendv(struct socket *ssocket, struct smb_hdr *smb_buffer, dump_smb(smb_buffer, len); while (len > 0) { - rc = kernel_sendmsg(ssocket, &smb_msg, &iov, number_of_pages, len?); + rc = kernel_sendmsg(ssocket, &smb_msg, &iov, number_of_pages, + len); if ((rc == -ENOSPC) || (rc == -EAGAIN)) { i++; if(i > 60) { @@ -266,6 +270,9 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, + if(ses->server->tcpStatus == CIFS_EXITING) + return -ENOENT; + /* Ensure that we do not send more than 50 overlapping requests to the same server. We may make this configurable later or use ses->maxReq */ @@ -346,11 +353,12 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, } /* BB can we sign efficiently in this path? */ - rc = cifs_sign_smb(in_buf, ses, &midQ->sequence_number); + rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); midQ->midState = MID_REQUEST_SUBMITTED; -/* rc = smb_sendv(ses->server->ssocket, in_buf, in_buf->smb_buf_length, piovec, - (struct sockaddr *) &(ses->server->addr.sockAddr));*/ +/* rc = smb_sendv(ses->server->ssocket, in_buf, in_buf->smb_buf_length, + piovec, + (struct sockaddr *) &(ses->server->addr.sockAddr));*/ if(rc < 0) { DeleteMidQEntry(midQ); up(&ses->server->tcpSem); @@ -396,6 +404,9 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, return -EIO; } + if(ses->server->tcpStatus == CifsExiting) + return -ENOENT; + /* Ensure that we do not send more than 50 overlapping requests to the same server. We may make this configurable later or use ses->maxReq */ @@ -405,7 +416,8 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, } else { spin_lock(&GlobalMid_Lock); while(1) { - if(atomic_read(&ses->server->inFlight) >= cifs_max_pending){ + if(atomic_read(&ses->server->inFlight) >= + cifs_max_pending){ spin_unlock(&GlobalMid_Lock); wait_event(ses->server->request_q, atomic_read(&ses->server->inFlight) @@ -475,7 +487,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, return -EIO; } - rc = cifs_sign_smb(in_buf, ses, &midQ->sequence_number); + rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); midQ->midState = MID_REQUEST_SUBMITTED; rc = smb_send(ses->server->ssocket, in_buf, in_buf->smb_buf_length, @@ -493,7 +505,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, up(&ses->server->tcpSem); if (long_op == -1) goto cifs_no_response_exit; - else if (long_op == 2) /* writes past end of file can take looooong time */ + else if (long_op == 2) /* writes past end of file can take loong time */ timeout = 300 * HZ; else if (long_op == 1) timeout = 45 * HZ; /* should be greater than @@ -559,8 +571,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, } if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) { - cERROR(1, - ("Frame too large received. Length: %d Xid: %d", + cERROR(1, ("Frame too large received. Length: %d Xid: %d", receive_len, xid)); rc = -EIO; } else { /* rcvd frame is ok */ @@ -575,15 +586,20 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, dump_smb(out_buf, 92); /* convert the length into a more usable form */ if((receive_len > 24) && - (ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED))) { - rc = cifs_verify_signature(out_buf, ses->mac_signing_key,midQ->sequence_number); /* BB fix BB */ - if(rc) - cFYI(1,("Unexpected signature received from server")); + (ses->server->secMode & (SECMODE_SIGN_REQUIRED | + SECMODE_SIGN_ENABLED))) { + rc = cifs_verify_signature(out_buf, + ses->server->mac_signing_key, + midQ->sequence_number+1); + if(rc) { + cERROR(1,("Unexpected SMB signature")); + /* BB FIXME add code to kill session */ + } } *pbytes_returned = out_buf->smb_buf_length; - /* BB special case reconnect tid and reconnect uid here? */ + /* BB special case reconnect tid and uid here? */ rc = map_smb_to_linux_error(out_buf); /* convert ByteCount if necessary */ diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index 549afa184fd..c1e02eff1d2 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -83,7 +83,8 @@ int cifs_removexattr(struct dentry * direntry, const char * ea_name) ea_name+=5; /* skip past user. prefix */ rc = CIFSSMBSetEA(xid,pTcon,full_path,ea_name,NULL, - (__u16)0, cifs_sb->local_nls); + (__u16)0, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); } remove_ea_exit: if (full_path) @@ -147,32 +148,40 @@ int cifs_setxattr(struct dentry * direntry, const char * ea_name, } ea_name += 5; /* skip past user. prefix */ rc = CIFSSMBSetEA(xid,pTcon,full_path,ea_name,ea_value, - (__u16)value_size, cifs_sb->local_nls); + (__u16)value_size, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); } else if(strncmp(ea_name, CIFS_XATTR_OS2_PREFIX,4) == 0) { if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) goto set_ea_exit; ea_name += 4; /* skip past os2. prefix */ rc = CIFSSMBSetEA(xid,pTcon,full_path,ea_name,ea_value, - (__u16)value_size, cifs_sb->local_nls); + (__u16)value_size, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); } else { int temp; temp = strncmp(ea_name,POSIX_ACL_XATTR_ACCESS, strlen(POSIX_ACL_XATTR_ACCESS)); if (temp == 0) { #ifdef CONFIG_CIFS_POSIX - rc = CIFSSMBSetPosixACL(xid, pTcon,full_path,ea_value, - (const int)value_size, ACL_TYPE_ACCESS, - cifs_sb->local_nls); + if(sb->s_flags & MS_POSIXACL) + rc = CIFSSMBSetPosixACL(xid, pTcon,full_path, + ea_value, (const int)value_size, + ACL_TYPE_ACCESS,cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); cFYI(1,("set POSIX ACL rc %d",rc)); #else cFYI(1,("set POSIX ACL not supported")); #endif } else if(strncmp(ea_name,POSIX_ACL_XATTR_DEFAULT,strlen(POSIX_ACL_XATTR_DEFAULT)) == 0) { #ifdef CONFIG_CIFS_POSIX - rc = CIFSSMBSetPosixACL(xid, pTcon,full_path,ea_value, - (const int)value_size, ACL_TYPE_DEFAULT, - cifs_sb->local_nls); + if(sb->s_flags & MS_POSIXACL) + rc = CIFSSMBSetPosixACL(xid, pTcon,full_path, + ea_value, (const int)value_size, + ACL_TYPE_DEFAULT, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); cFYI(1,("set POSIX default ACL rc %d",rc)); #else cFYI(1,("set default POSIX ACL not supported")); @@ -238,27 +247,35 @@ ssize_t cifs_getxattr(struct dentry * direntry, const char * ea_name, } /* BB add else when above is implemented */ ea_name += 5; /* skip past user. prefix */ rc = CIFSSMBQueryEA(xid,pTcon,full_path,ea_name,ea_value, - buf_size, cifs_sb->local_nls); + buf_size, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); } else if(strncmp(ea_name, CIFS_XATTR_OS2_PREFIX,4) == 0) { if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) goto get_ea_exit; ea_name += 4; /* skip past os2. prefix */ rc = CIFSSMBQueryEA(xid,pTcon,full_path,ea_name,ea_value, - buf_size, cifs_sb->local_nls); + buf_size, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); } else if(strncmp(ea_name,POSIX_ACL_XATTR_ACCESS,strlen(POSIX_ACL_XATTR_ACCESS)) == 0) { #ifdef CONFIG_CIFS_POSIX - rc = CIFSSMBGetPosixACL(xid, pTcon, full_path, + if(sb->s_flags & MS_POSIXACL) + rc = CIFSSMBGetPosixACL(xid, pTcon, full_path, ea_value, buf_size, ACL_TYPE_ACCESS, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); #else cFYI(1,("query POSIX ACL not supported yet")); #endif /* CONFIG_CIFS_POSIX */ } else if(strncmp(ea_name,POSIX_ACL_XATTR_DEFAULT,strlen(POSIX_ACL_XATTR_DEFAULT)) == 0) { #ifdef CONFIG_CIFS_POSIX - rc = CIFSSMBGetPosixACL(xid, pTcon, full_path, + if(sb->s_flags & MS_POSIXACL) + rc = CIFSSMBGetPosixACL(xid, pTcon, full_path, ea_value, buf_size, ACL_TYPE_DEFAULT, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); #else cFYI(1,("query POSIX default ACL not supported yet")); #endif @@ -324,7 +341,9 @@ ssize_t cifs_listxattr(struct dentry * direntry, char * data, size_t buf_size) search server for EAs or streams to returns as xattrs */ rc = CIFSSMBQAllEAs(xid,pTcon,full_path,data,buf_size, - cifs_sb->local_nls); + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (full_path) kfree(full_path); diff --git a/fs/compat.c b/fs/compat.c index 67c0b94d114..728cd836538 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -809,7 +809,7 @@ static void *do_smb_super_data_conv(void *raw_data) struct compat_nfs_string { compat_uint_t len; - compat_uptr_t __user data; + compat_uptr_t data; }; static inline void compat_nfs_string(struct nfs_string *dst, @@ -834,10 +834,10 @@ struct compat_nfs4_mount_data_v1 { struct compat_nfs_string mnt_path; struct compat_nfs_string hostname; compat_uint_t host_addrlen; - compat_uptr_t __user host_addr; + compat_uptr_t host_addr; compat_int_t proto; compat_int_t auth_flavourlen; - compat_uptr_t __user auth_flavours; + compat_uptr_t auth_flavours; }; static int do_nfs4_super_data_conv(void *raw_data) diff --git a/fs/dcache.c b/fs/dcache.c index 496a4e08369..3aa8a7e980d 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -39,7 +39,7 @@ int sysctl_vfs_cache_pressure = 100; EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure); __cacheline_aligned_in_smp DEFINE_SPINLOCK(dcache_lock); -seqlock_t rename_lock __cacheline_aligned_in_smp = SEQLOCK_UNLOCKED; +static seqlock_t rename_lock __cacheline_aligned_in_smp = SEQLOCK_UNLOCKED; EXPORT_SYMBOL(dcache_lock); diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 05b966cd6f7..9900e333655 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -320,7 +320,7 @@ static struct super_block *eventpollfs_get_sb(struct file_system_type *fs_type, /* * This semaphore is used to serialize ep_free() and eventpoll_release_file(). */ -struct semaphore epsem; +static struct semaphore epsem; /* Safe wake up implementation */ static struct poll_safewake psw; diff --git a/fs/exec.c b/fs/exec.c index a8394499926..e56ee243702 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -197,7 +197,8 @@ static int count(char __user * __user * argv, int max) * memory to free pages in kernel mem. These are in a format ready * to be put directly into the top of new user memory. */ -int copy_strings(int argc,char __user * __user * argv, struct linux_binprm *bprm) +static int copy_strings(int argc, char __user * __user * argv, + struct linux_binprm *bprm) { struct page *kmapped_page = NULL; char *kaddr = NULL; @@ -868,9 +869,11 @@ int flush_old_exec(struct linux_binprm * bprm) if (current->euid == current->uid && current->egid == current->gid) current->mm->dumpable = 1; name = bprm->filename; + + /* Copies the binary name from after last slash */ for (i=0; (ch = *(name++)) != '\0';) { if (ch == '/') - i = 0; + i = 0; /* overwrite what we wrote */ else if (i < (sizeof(tcomm) - 1)) tcomm[i++] = ch; diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 040eb288bb1..0d5fa73b18d 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -455,12 +455,11 @@ static unsigned long ext3_find_near(struct inode *inode, Indirect *ind) * @goal: place to store the result. * * Normally this function find the prefered place for block allocation, - * stores it in *@goal and returns zero. If the branch had been changed - * under us we return -EAGAIN. + * stores it in *@goal and returns zero. */ -static int ext3_find_goal(struct inode *inode, long block, Indirect chain[4], - Indirect *partial, unsigned long *goal) +static unsigned long ext3_find_goal(struct inode *inode, long block, + Indirect chain[4], Indirect *partial) { struct ext3_block_alloc_info *block_i = EXT3_I(inode)->i_block_alloc_info; @@ -470,15 +469,10 @@ static int ext3_find_goal(struct inode *inode, long block, Indirect chain[4], */ if (block_i && (block == block_i->last_alloc_logical_block + 1) && (block_i->last_alloc_physical_block != 0)) { - *goal = block_i->last_alloc_physical_block + 1; - return 0; + return block_i->last_alloc_physical_block + 1; } - if (verify_chain(chain, partial)) { - *goal = ext3_find_near(inode, partial); - return 0; - } - return -EAGAIN; + return ext3_find_near(inode, partial); } /** @@ -582,12 +576,9 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, * @where: location of missing link * @num: number of blocks we are adding * - * This function verifies that chain (up to the missing link) had not - * changed, fills the missing link and does all housekeeping needed in + * This function fills the missing link and does all housekeeping needed in * inode (->i_blocks, etc.). In case of success we end up with the full - * chain to new block and return 0. Otherwise (== chain had been changed) - * we free the new blocks (forgetting their buffer_heads, indeed) and - * return -EAGAIN. + * chain to new block and return 0. */ static int ext3_splice_branch(handle_t *handle, struct inode *inode, long block, @@ -608,12 +599,6 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, long block, if (err) goto err_out; } - /* Verify that place we are splicing to is still there and vacant */ - - if (!verify_chain(chain, where-1) || *where->p) - /* Writer: end */ - goto changed; - /* That's it */ *where->p = where->key; @@ -657,26 +642,11 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, long block, } return err; -changed: - /* - * AKPM: if where[i].bh isn't part of the current updating - * transaction then we explode nastily. Test this code path. - */ - jbd_debug(1, "the chain changed: try again\n"); - err = -EAGAIN; - err_out: for (i = 1; i < num; i++) { BUFFER_TRACE(where[i].bh, "call journal_forget"); ext3_journal_forget(handle, where[i].bh); } - /* For the normal collision cleanup case, we free up the blocks. - * On genuine filesystem errors we don't even think about doing - * that. */ - if (err == -EAGAIN) - for (i = 0; i < num; i++) - ext3_free_blocks(handle, inode, - le32_to_cpu(where[i].key), 1); return err; } @@ -708,7 +678,7 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock, unsigned long goal; int left; int boundary = 0; - int depth = ext3_block_to_path(inode, iblock, offsets, &boundary); + const int depth = ext3_block_to_path(inode, iblock, offsets, &boundary); struct ext3_inode_info *ei = EXT3_I(inode); J_ASSERT(handle != NULL || create == 0); @@ -716,54 +686,55 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock, if (depth == 0) goto out; -reread: partial = ext3_get_branch(inode, depth, offsets, chain, &err); /* Simplest case - block found, no allocation needed */ if (!partial) { clear_buffer_new(bh_result); -got_it: - map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); - if (boundary) - set_buffer_boundary(bh_result); - /* Clean up and exit */ - partial = chain+depth-1; /* the whole chain */ - goto cleanup; + goto got_it; } /* Next simple case - plain lookup or failed read of indirect block */ - if (!create || err == -EIO) { -cleanup: + if (!create || err == -EIO) + goto cleanup; + + down(&ei->truncate_sem); + + /* + * If the indirect block is missing while we are reading + * the chain(ext3_get_branch() returns -EAGAIN err), or + * if the chain has been changed after we grab the semaphore, + * (either because another process truncated this branch, or + * another get_block allocated this branch) re-grab the chain to see if + * the request block has been allocated or not. + * + * Since we already block the truncate/other get_block + * at this point, we will have the current copy of the chain when we + * splice the branch into the tree. + */ + if (err == -EAGAIN || !verify_chain(chain, partial)) { while (partial > chain) { - BUFFER_TRACE(partial->bh, "call brelse"); brelse(partial->bh); partial--; } - BUFFER_TRACE(bh_result, "returned"); -out: - return err; + partial = ext3_get_branch(inode, depth, offsets, chain, &err); + if (!partial) { + up(&ei->truncate_sem); + if (err) + goto cleanup; + clear_buffer_new(bh_result); + goto got_it; + } } /* - * Indirect block might be removed by truncate while we were - * reading it. Handling of that case (forget what we've got and - * reread) is taken out of the main path. - */ - if (err == -EAGAIN) - goto changed; - - goal = 0; - down(&ei->truncate_sem); - - /* lazy initialize the block allocation info here if necessary */ - if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) { + * Okay, we need to do block allocation. Lazily initialize the block + * allocation info here if necessary + */ + if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) ext3_init_block_alloc_info(inode); - } - if (ext3_find_goal(inode, iblock, chain, partial, &goal) < 0) { - up(&ei->truncate_sem); - goto changed; - } + goal = ext3_find_goal(inode, iblock, chain, partial); left = (chain + depth) - partial; @@ -771,38 +742,45 @@ out: * Block out ext3_truncate while we alter the tree */ err = ext3_alloc_branch(handle, inode, left, goal, - offsets+(partial-chain), partial); + offsets + (partial - chain), partial); - /* The ext3_splice_branch call will free and forget any buffers + /* + * The ext3_splice_branch call will free and forget any buffers * on the new chain if there is a failure, but that risks using * up transaction credits, especially for bitmaps where the * credits cannot be returned. Can we handle this somehow? We - * may need to return -EAGAIN upwards in the worst case. --sct */ + * may need to return -EAGAIN upwards in the worst case. --sct + */ if (!err) err = ext3_splice_branch(handle, inode, iblock, chain, partial, left); - /* i_disksize growing is protected by truncate_sem - * don't forget to protect it if you're about to implement - * concurrent ext3_get_block() -bzzz */ + /* + * i_disksize growing is protected by truncate_sem. Don't forget to + * protect it if you're about to implement concurrent + * ext3_get_block() -bzzz + */ if (!err && extend_disksize && inode->i_size > ei->i_disksize) ei->i_disksize = inode->i_size; up(&ei->truncate_sem); - if (err == -EAGAIN) - goto changed; if (err) goto cleanup; set_buffer_new(bh_result); - goto got_it; - -changed: +got_it: + map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); + if (boundary) + set_buffer_boundary(bh_result); + /* Clean up and exit */ + partial = chain + depth - 1; /* the whole chain */ +cleanup: while (partial > chain) { - jbd_debug(1, "buffer chain changed, retrying\n"); - BUFFER_TRACE(partial->bh, "brelsing"); + BUFFER_TRACE(partial->bh, "call brelse"); brelse(partial->bh); partial--; } - goto reread; + BUFFER_TRACE(bh_result, "returned"); +out: + return err; } static int ext3_get_block(struct inode *inode, sector_t iblock, @@ -866,12 +844,6 @@ get_block: return ret; } -static int ext3_writepages_get_block(struct inode *inode, sector_t iblock, - struct buffer_head *bh, int create) -{ - return ext3_direct_io_get_blocks(inode, iblock, 1, bh, create); -} - /* * `handle' can be NULL if create is zero */ @@ -1345,45 +1317,6 @@ out_fail: return ret; } -static int -ext3_writeback_writepage_helper(struct page *page, - struct writeback_control *wbc) -{ - return block_write_full_page(page, ext3_get_block, wbc); -} - -static int -ext3_writeback_writepages(struct address_space *mapping, - struct writeback_control *wbc) -{ - struct inode *inode = mapping->host; - handle_t *handle = NULL; - int err, ret = 0; - - if (!mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) - return ret; - - handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - return ret; - } - - ret = __mpage_writepages(mapping, wbc, ext3_writepages_get_block, - ext3_writeback_writepage_helper); - - /* - * Need to reaquire the handle since ext3_writepages_get_block() - * can restart the handle - */ - handle = journal_current_handle(); - - err = ext3_journal_stop(handle); - if (!ret) - ret = err; - return ret; -} - static int ext3_writeback_writepage(struct page *page, struct writeback_control *wbc) { @@ -1621,7 +1554,6 @@ static struct address_space_operations ext3_writeback_aops = { .readpage = ext3_readpage, .readpages = ext3_readpages, .writepage = ext3_writeback_writepage, - .writepages = ext3_writeback_writepages, .sync_page = block_sync_page, .prepare_write = ext3_prepare_write, .commit_write = ext3_writeback_commit_write, diff --git a/fs/fcntl.c b/fs/fcntl.c index 3e7ab16ed15..286a9f8f3d4 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -15,6 +15,7 @@ #include <linux/module.h> #include <linux/security.h> #include <linux/ptrace.h> +#include <linux/signal.h> #include <asm/poll.h> #include <asm/siginfo.h> @@ -308,7 +309,7 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, break; case F_SETSIG: /* arg == 0 restores default behaviour. */ - if (arg < 0 || arg > _NSIG) { + if (!valid_signal(arg)) { break; } err = 0; diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index d6efb36cab2..8e050fa5821 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -512,7 +512,8 @@ restart: } /** - * sync_inodes + * sync_inodes - writes all inodes to disk + * @wait: wait for completion * * sync_inodes() goes through each super block's dirty inode list, writes the * inodes out, waits on the writeout and puts the inodes back on the normal @@ -604,6 +605,7 @@ EXPORT_SYMBOL(sync_inode); /** * generic_osync_inode - flush all dirty data for a given inode to disk * @inode: inode to write + * @mapping: the address_space that should be flushed * @what: what to write and wait upon * * This can be called by file_write functions for files which have the diff --git a/fs/hfs/mdb.c b/fs/hfs/mdb.c index 4efb640c4d0..217e32f37e0 100644 --- a/fs/hfs/mdb.c +++ b/fs/hfs/mdb.c @@ -333,6 +333,8 @@ void hfs_mdb_close(struct super_block *sb) * Release the resources associated with the in-core MDB. */ void hfs_mdb_put(struct super_block *sb) { + if (!HFS_SB(sb)) + return; /* free the B-trees */ hfs_btree_close(HFS_SB(sb)->ext_tree); hfs_btree_close(HFS_SB(sb)->cat_tree); @@ -340,4 +342,7 @@ void hfs_mdb_put(struct super_block *sb) /* free the buffers holding the primary and alternate MDBs */ brelse(HFS_SB(sb)->mdb_bh); brelse(HFS_SB(sb)->alt_mdb_bh); + + kfree(HFS_SB(sb)); + sb->s_fs_info = NULL; } diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 1e2c193134c..ab783f6afa3 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -297,7 +297,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) res = -EINVAL; if (!parse_options((char *)data, sbi)) { hfs_warn("hfs_fs: unable to parse mount options.\n"); - goto bail3; + goto bail; } sb->s_op = &hfs_super_operations; @@ -310,7 +310,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) hfs_warn("VFS: Can't find a HFS filesystem on dev %s.\n", hfs_mdb_name(sb)); res = -EINVAL; - goto bail2; + goto bail; } /* try to get the root inode */ @@ -340,10 +340,8 @@ bail_iput: iput(root_inode); bail_no_root: hfs_warn("hfs_fs: get root inode failed.\n"); +bail: hfs_mdb_put(sb); -bail2: -bail3: - kfree(sbi); return res; } diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 5f8044664a3..d55ad67b8e4 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -208,7 +208,9 @@ static void hfsplus_write_super(struct super_block *sb) static void hfsplus_put_super(struct super_block *sb) { dprint(DBG_SUPER, "hfsplus_put_super\n"); - if (!(sb->s_flags & MS_RDONLY)) { + if (!sb->s_fs_info) + return; + if (!(sb->s_flags & MS_RDONLY) && HFSPLUS_SB(sb).s_vhdr) { struct hfsplus_vh *vhdr = HFSPLUS_SB(sb).s_vhdr; vhdr->modify_date = hfsp_now2mt(); @@ -226,6 +228,8 @@ static void hfsplus_put_super(struct super_block *sb) brelse(HFSPLUS_SB(sb).s_vhbh); if (HFSPLUS_SB(sb).nls) unload_nls(HFSPLUS_SB(sb).nls); + kfree(sb->s_fs_info); + sb->s_fs_info = NULL; } static int hfsplus_statfs(struct super_block *sb, struct kstatfs *buf) diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index a88ad292485..14a0d339d03 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -521,7 +521,7 @@ int hostfs_commit_write(struct file *file, struct page *page, unsigned from, static struct address_space_operations hostfs_aops = { .writepage = hostfs_writepage, .readpage = hostfs_readpage, -/* .set_page_dirty = __set_page_dirty_nobuffers, */ + .set_page_dirty = __set_page_dirty_nobuffers, .prepare_write = hostfs_prepare_write, .commit_write = hostfs_commit_write }; @@ -991,13 +991,17 @@ static int hostfs_fill_sb_common(struct super_block *sb, void *d, int silent) goto out_put; err = read_inode(root_inode); - if(err) - goto out_put; + if(err){ + /* No iput in this case because the dput does that for us */ + dput(sb->s_root); + sb->s_root = NULL; + goto out_free; + } return(0); out_put: - iput(root_inode); + iput(root_inode); out_free: kfree(name); out: diff --git a/fs/inode.c b/fs/inode.c index af8fd78d209..801fe7f3628 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -26,7 +26,6 @@ * This is needed for the following functions: * - inode_has_buffers * - invalidate_inode_buffers - * - fsync_bdev * - invalidate_bdev * * FIXME: remove all knowledge of the buffer layer from this file @@ -332,14 +331,6 @@ static int invalidate_list(struct list_head *head, struct list_head *dispose) return busy; } -/* - * This is a two-stage process. First we collect all - * offending inodes onto the throw-away list, and in - * the second stage we actually dispose of them. This - * is because we don't want to sleep while messing - * with the global lists.. - */ - /** * invalidate_inodes - discard the inodes on a device * @sb: superblock @@ -366,16 +357,11 @@ int invalidate_inodes(struct super_block * sb) EXPORT_SYMBOL(invalidate_inodes); -int __invalidate_device(struct block_device *bdev, int do_sync) +int __invalidate_device(struct block_device *bdev) { - struct super_block *sb; - int res; + struct super_block *sb = get_super(bdev); + int res = 0; - if (do_sync) - fsync_bdev(bdev); - - res = 0; - sb = get_super(bdev); if (sb) { /* * no need to lock the super, get_super holds the @@ -390,7 +376,6 @@ int __invalidate_device(struct block_device *bdev, int do_sync) invalidate_bdev(bdev, 0); return res; } - EXPORT_SYMBOL(__invalidate_device); static int can_unuse(struct inode *inode) @@ -1336,7 +1321,7 @@ void __init inode_init(unsigned long mempages) /* inode slab cache */ inode_cachep = kmem_cache_create("inode_cache", sizeof(struct inode), - 0, SLAB_PANIC, init_once, NULL); + 0, SLAB_RECLAIM_ACCOUNT|SLAB_PANIC, init_once, NULL); set_shrinker(DEFAULT_SEEKS, shrink_icache_memory); /* Hash may have been set up in inode_init_early */ diff --git a/fs/isofs/compress.c b/fs/isofs/compress.c index fb42c3f3bf0..34a44e45168 100644 --- a/fs/isofs/compress.c +++ b/fs/isofs/compress.c @@ -18,29 +18,12 @@ #include <linux/config.h> #include <linux/module.h> - -#include <linux/stat.h> -#include <linux/time.h> -#include <linux/iso_fs.h> -#include <linux/kernel.h> -#include <linux/major.h> -#include <linux/mm.h> -#include <linux/string.h> -#include <linux/slab.h> -#include <linux/errno.h> #include <linux/init.h> -#include <linux/nls.h> -#include <linux/ctype.h> -#include <linux/smp_lock.h> -#include <linux/blkdev.h> + #include <linux/vmalloc.h> #include <linux/zlib.h> -#include <linux/buffer_head.h> - -#include <asm/system.h> -#include <asm/uaccess.h> -#include <asm/semaphore.h> +#include "isofs.h" #include "zisofs.h" /* This should probably be global. */ diff --git a/fs/isofs/dir.c b/fs/isofs/dir.c index 14d86de6637..6030956b894 100644 --- a/fs/isofs/dir.c +++ b/fs/isofs/dir.c @@ -10,20 +10,9 @@ * * isofs directory handling functions */ -#include <linux/errno.h> -#include <linux/fs.h> -#include <linux/iso_fs.h> -#include <linux/kernel.h> -#include <linux/stat.h> -#include <linux/string.h> -#include <linux/mm.h> -#include <linux/slab.h> -#include <linux/time.h> #include <linux/config.h> #include <linux/smp_lock.h> -#include <linux/buffer_head.h> - -#include <asm/uaccess.h> +#include "isofs.h" static int isofs_readdir(struct file *, void *, filldir_t); diff --git a/fs/isofs/export.c b/fs/isofs/export.c index e4252c96087..4af856a7fda 100644 --- a/fs/isofs/export.c +++ b/fs/isofs/export.c @@ -13,11 +13,7 @@ * fs/exportfs/expfs.c. */ -#include <linux/buffer_head.h> -#include <linux/errno.h> -#include <linux/fs.h> -#include <linux/iso_fs.h> -#include <linux/kernel.h> +#include "isofs.h" static struct dentry * isofs_export_iget(struct super_block *sb, diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index b9256e65e14..abd7b12eeca 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -12,29 +12,18 @@ */ #include <linux/config.h> +#include <linux/init.h> #include <linux/module.h> -#include <linux/stat.h> -#include <linux/time.h> -#include <linux/iso_fs.h> -#include <linux/kernel.h> -#include <linux/major.h> -#include <linux/mm.h> -#include <linux/string.h> #include <linux/slab.h> -#include <linux/errno.h> -#include <linux/cdrom.h> -#include <linux/init.h> #include <linux/nls.h> #include <linux/ctype.h> #include <linux/smp_lock.h> -#include <linux/blkdev.h> -#include <linux/buffer_head.h> -#include <linux/vfs.h> +#include <linux/statfs.h> +#include <linux/cdrom.h> #include <linux/parser.h> -#include <asm/system.h> -#include <asm/uaccess.h> +#include "isofs.h" #include "zisofs.h" #define BEQUIET diff --git a/fs/isofs/isofs.h b/fs/isofs/isofs.h new file mode 100644 index 00000000000..9ce7b51fb61 --- /dev/null +++ b/fs/isofs/isofs.h @@ -0,0 +1,190 @@ +#include <linux/fs.h> +#include <linux/buffer_head.h> +#include <linux/iso_fs.h> +#include <asm/unaligned.h> + +enum isofs_file_format { + isofs_file_normal = 0, + isofs_file_sparse = 1, + isofs_file_compressed = 2, +}; + +/* + * iso fs inode data in memory + */ +struct iso_inode_info { + unsigned long i_iget5_block; + unsigned long i_iget5_offset; + unsigned int i_first_extent; + unsigned char i_file_format; + unsigned char i_format_parm[3]; + unsigned long i_next_section_block; + unsigned long i_next_section_offset; + off_t i_section_size; + struct inode vfs_inode; +}; + +/* + * iso9660 super-block data in memory + */ +struct isofs_sb_info { + unsigned long s_ninodes; + unsigned long s_nzones; + unsigned long s_firstdatazone; + unsigned long s_log_zone_size; + unsigned long s_max_size; + + unsigned char s_high_sierra; /* A simple flag */ + unsigned char s_mapping; + int s_rock_offset; /* offset of SUSP fields within SU area */ + unsigned char s_rock; + unsigned char s_joliet_level; + unsigned char s_utf8; + unsigned char s_cruft; /* Broken disks with high + byte of length containing + junk */ + unsigned char s_unhide; + unsigned char s_nosuid; + unsigned char s_nodev; + unsigned char s_nocompress; + + mode_t s_mode; + gid_t s_gid; + uid_t s_uid; + struct nls_table *s_nls_iocharset; /* Native language support table */ +}; + +static inline struct isofs_sb_info *ISOFS_SB(struct super_block *sb) +{ + return sb->s_fs_info; +} + +static inline struct iso_inode_info *ISOFS_I(struct inode *inode) +{ + return container_of(inode, struct iso_inode_info, vfs_inode); +} + +static inline int isonum_711(char *p) +{ + return *(u8 *)p; +} +static inline int isonum_712(char *p) +{ + return *(s8 *)p; +} +static inline unsigned int isonum_721(char *p) +{ + return le16_to_cpu(get_unaligned((__le16 *)p)); +} +static inline unsigned int isonum_722(char *p) +{ + return be16_to_cpu(get_unaligned((__le16 *)p)); +} +static inline unsigned int isonum_723(char *p) +{ + /* Ignore bigendian datum due to broken mastering programs */ + return le16_to_cpu(get_unaligned((__le16 *)p)); +} +static inline unsigned int isonum_731(char *p) +{ + return le32_to_cpu(get_unaligned((__le32 *)p)); +} +static inline unsigned int isonum_732(char *p) +{ + return be32_to_cpu(get_unaligned((__le32 *)p)); +} +static inline unsigned int isonum_733(char *p) +{ + /* Ignore bigendian datum due to broken mastering programs */ + return le32_to_cpu(get_unaligned((__le32 *)p)); +} +extern int iso_date(char *, int); + +struct inode; /* To make gcc happy */ + +extern int parse_rock_ridge_inode(struct iso_directory_record *, struct inode *); +extern int get_rock_ridge_filename(struct iso_directory_record *, char *, struct inode *); +extern int isofs_name_translate(struct iso_directory_record *, char *, struct inode *); + +int get_joliet_filename(struct iso_directory_record *, unsigned char *, struct inode *); +int get_acorn_filename(struct iso_directory_record *, char *, struct inode *); + +extern struct dentry *isofs_lookup(struct inode *, struct dentry *, struct nameidata *); +extern struct buffer_head *isofs_bread(struct inode *, sector_t); +extern int isofs_get_blocks(struct inode *, sector_t, struct buffer_head **, unsigned long); + +extern struct inode *isofs_iget(struct super_block *sb, + unsigned long block, + unsigned long offset); + +/* Because the inode number is no longer relevant to finding the + * underlying meta-data for an inode, we are free to choose a more + * convenient 32-bit number as the inode number. The inode numbering + * scheme was recommended by Sergey Vlasov and Eric Lammerts. */ +static inline unsigned long isofs_get_ino(unsigned long block, + unsigned long offset, + unsigned long bufbits) +{ + return (block << (bufbits - 5)) | (offset >> 5); +} + +/* Every directory can have many redundant directory entries scattered + * throughout the directory tree. First there is the directory entry + * with the name of the directory stored in the parent directory. + * Then, there is the "." directory entry stored in the directory + * itself. Finally, there are possibly many ".." directory entries + * stored in all the subdirectories. + * + * In order for the NFS get_parent() method to work and for the + * general consistency of the dcache, we need to make sure the + * "i_iget5_block" and "i_iget5_offset" all point to exactly one of + * the many redundant entries for each directory. We normalize the + * block and offset by always making them point to the "." directory. + * + * Notice that we do not use the entry for the directory with the name + * that is located in the parent directory. Even though choosing this + * first directory is more natural, it is much easier to find the "." + * entry in the NFS get_parent() method because it is implicitly + * encoded in the "extent + ext_attr_length" fields of _all_ the + * redundant entries for the directory. Thus, it can always be + * reached regardless of which directory entry you have in hand. + * + * This works because the "." entry is simply the first directory + * record when you start reading the file that holds all the directory + * records, and this file starts at "extent + ext_attr_length" blocks. + * Because the "." entry is always the first entry listed in the + * directories file, the normalized "offset" value is always 0. + * + * You should pass the directory entry in "de". On return, "block" + * and "offset" will hold normalized values. Only directories are + * affected making it safe to call even for non-directory file + * types. */ +static inline void +isofs_normalize_block_and_offset(struct iso_directory_record* de, + unsigned long *block, + unsigned long *offset) +{ + /* Only directories are normalized. */ + if (de->flags[0] & 2) { + *offset = 0; + *block = (unsigned long)isonum_733(de->extent) + + (unsigned long)isonum_711(de->ext_attr_length); + } +} + +extern struct inode_operations isofs_dir_inode_operations; +extern struct file_operations isofs_dir_operations; +extern struct address_space_operations isofs_symlink_aops; +extern struct export_operations isofs_export_ops; + +/* The following macros are used to check for memory leaks. */ +#ifdef LEAK_CHECK +#define free_s leak_check_free_s +#define malloc leak_check_malloc +#define sb_bread leak_check_bread +#define brelse leak_check_brelse +extern void * leak_check_malloc(unsigned int size); +extern void leak_check_free_s(void * obj, int size); +extern struct buffer_head * leak_check_bread(struct super_block *sb, int block); +extern void leak_check_brelse(struct buffer_head * bh); +#endif /* LEAK_CHECK */ diff --git a/fs/isofs/joliet.c b/fs/isofs/joliet.c index 86c50e22fc8..2931de7f1a6 100644 --- a/fs/isofs/joliet.c +++ b/fs/isofs/joliet.c @@ -6,11 +6,9 @@ * Joliet: Microsoft's Unicode extensions to iso9660 */ -#include <linux/string.h> +#include <linux/types.h> #include <linux/nls.h> -#include <linux/mm.h> -#include <linux/iso_fs.h> -#include <asm/unaligned.h> +#include "isofs.h" /* * Convert Unicode 16 to UTF8 or ASCII. diff --git a/fs/isofs/namei.c b/fs/isofs/namei.c index 9569fc44102..690edf37173 100644 --- a/fs/isofs/namei.c +++ b/fs/isofs/namei.c @@ -6,20 +6,9 @@ * (C) 1991 Linus Torvalds - minix filesystem */ -#include <linux/time.h> -#include <linux/iso_fs.h> -#include <linux/kernel.h> -#include <linux/string.h> -#include <linux/stat.h> -#include <linux/fcntl.h> -#include <linux/mm.h> -#include <linux/errno.h> #include <linux/config.h> /* Joliet? */ #include <linux/smp_lock.h> -#include <linux/buffer_head.h> -#include <linux/dcache.h> - -#include <asm/uaccess.h> +#include "isofs.h" /* * ok, we cannot use strncmp, as the name is not in our data space. diff --git a/fs/isofs/rock.c b/fs/isofs/rock.c index 8bdd3e40954..089e79c6558 100644 --- a/fs/isofs/rock.c +++ b/fs/isofs/rock.c @@ -6,17 +6,11 @@ * Rock Ridge Extensions to iso9660 */ -#include <linux/stat.h> -#include <linux/time.h> -#include <linux/iso_fs.h> -#include <linux/string.h> -#include <linux/mm.h> #include <linux/slab.h> #include <linux/pagemap.h> #include <linux/smp_lock.h> -#include <linux/buffer_head.h> -#include <asm/page.h> +#include "isofs.h" #include "rock.h" /* These functions are designed to read the system areas of a directory record diff --git a/fs/isofs/util.c b/fs/isofs/util.c index 3f6d9c1ac95..01e1ee7a998 100644 --- a/fs/isofs/util.c +++ b/fs/isofs/util.c @@ -2,9 +2,7 @@ * linux/fs/isofs/util.c */ -#include <linux/time.h> -#include <linux/fs.h> -#include <linux/iso_fs.h> +#include "isofs.h" /* * We have to convert from a MM/DD/YY format to the Unix ctime format. @@ -80,4 +78,3 @@ int iso_date(char * p, int flag) } return crtime; } - diff --git a/fs/jffs2/compr_rubin.c b/fs/jffs2/compr_rubin.c index 450d6624181..09422388fb9 100644 --- a/fs/jffs2/compr_rubin.c +++ b/fs/jffs2/compr_rubin.c @@ -228,8 +228,10 @@ int jffs2_rubinmips_compress(unsigned char *data_in, unsigned char *cpage_out, return rubin_do_compress(BIT_DIVIDER_MIPS, bits_mips, data_in, cpage_out, sourcelen, dstlen); } #endif -int jffs2_dynrubin_compress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t *sourcelen, uint32_t *dstlen, void *model) +static int jffs2_dynrubin_compress(unsigned char *data_in, + unsigned char *cpage_out, + uint32_t *sourcelen, uint32_t *dstlen, + void *model) { int bits[8]; unsigned char histo[256]; @@ -306,15 +308,19 @@ static void rubin_do_decompress(int bit_divider, int *bits, unsigned char *cdata } -int jffs2_rubinmips_decompress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t sourcelen, uint32_t dstlen, void *model) +static int jffs2_rubinmips_decompress(unsigned char *data_in, + unsigned char *cpage_out, + uint32_t sourcelen, uint32_t dstlen, + void *model) { rubin_do_decompress(BIT_DIVIDER_MIPS, bits_mips, data_in, cpage_out, sourcelen, dstlen); return 0; } -int jffs2_dynrubin_decompress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t sourcelen, uint32_t dstlen, void *model) +static int jffs2_dynrubin_decompress(unsigned char *data_in, + unsigned char *cpage_out, + uint32_t sourcelen, uint32_t dstlen, + void *model) { int bits[8]; int c; diff --git a/fs/jffs2/compr_zlib.c b/fs/jffs2/compr_zlib.c index 9f9932c22ad..078a30e406b 100644 --- a/fs/jffs2/compr_zlib.c +++ b/fs/jffs2/compr_zlib.c @@ -69,8 +69,10 @@ static void free_workspaces(void) #define free_workspaces() do { } while(0) #endif /* __KERNEL__ */ -int jffs2_zlib_compress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t *sourcelen, uint32_t *dstlen, void *model) +static int jffs2_zlib_compress(unsigned char *data_in, + unsigned char *cpage_out, + uint32_t *sourcelen, uint32_t *dstlen, + void *model) { int ret; @@ -135,8 +137,10 @@ int jffs2_zlib_compress(unsigned char *data_in, unsigned char *cpage_out, return ret; } -int jffs2_zlib_decompress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t srclen, uint32_t destlen, void *model) +static int jffs2_zlib_decompress(unsigned char *data_in, + unsigned char *cpage_out, + uint32_t srclen, uint32_t destlen, + void *model) { int ret; int wbits = MAX_WBITS; diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index 0c607c1388f..771a554701d 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -79,8 +79,7 @@ static int jffs2_do_readpage_nolock (struct inode *inode, struct page *pg) D2(printk(KERN_DEBUG "jffs2_do_readpage_nolock(): ino #%lu, page at offset 0x%lx\n", inode->i_ino, pg->index << PAGE_CACHE_SHIFT)); - if (!PageLocked(pg)) - PAGE_BUG(pg); + BUG_ON(!PageLocked(pg)); pg_buf = kmap(pg); /* FIXME: Can kmap fail? */ diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c index 7bc906677b0..24a689179af 100644 --- a/fs/jfs/inode.c +++ b/fs/jfs/inode.c @@ -175,31 +175,22 @@ jfs_get_blocks(struct inode *ip, sector_t lblock, unsigned long max_blocks, { s64 lblock64 = lblock; int rc = 0; - int take_locks; xad_t xad; s64 xaddr; int xflag; - s32 xlen; - - /* - * If this is a special inode (imap, dmap) - * the lock should already be taken - */ - take_locks = (JFS_IP(ip)->fileset != AGGREGATE_I); + s32 xlen = max_blocks; /* * Take appropriate lock on inode */ - if (take_locks) { - if (create) - IWRITE_LOCK(ip); - else - IREAD_LOCK(ip); - } + if (create) + IWRITE_LOCK(ip); + else + IREAD_LOCK(ip); if (((lblock64 << ip->i_sb->s_blocksize_bits) < ip->i_size) && - (xtLookup(ip, lblock64, max_blocks, &xflag, &xaddr, &xlen, 0) - == 0) && xlen) { + (!xtLookup(ip, lblock64, max_blocks, &xflag, &xaddr, &xlen, 0)) && + xaddr) { if (xflag & XAD_NOTRECORDED) { if (!create) /* @@ -238,7 +229,7 @@ jfs_get_blocks(struct inode *ip, sector_t lblock, unsigned long max_blocks, #ifdef _JFS_4K if ((rc = extHint(ip, lblock64 << ip->i_sb->s_blocksize_bits, &xad))) goto unlock; - rc = extAlloc(ip, max_blocks, lblock64, &xad, FALSE); + rc = extAlloc(ip, xlen, lblock64, &xad, FALSE); if (rc) goto unlock; @@ -258,12 +249,10 @@ jfs_get_blocks(struct inode *ip, sector_t lblock, unsigned long max_blocks, /* * Release lock on inode */ - if (take_locks) { - if (create) - IWRITE_UNLOCK(ip); - else - IREAD_UNLOCK(ip); - } + if (create) + IWRITE_UNLOCK(ip); + else + IREAD_UNLOCK(ip); return rc; } diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c index d86e467c6e4..69007fd546e 100644 --- a/fs/jfs/jfs_dmap.c +++ b/fs/jfs/jfs_dmap.c @@ -471,6 +471,7 @@ dbUpdatePMap(struct inode *ipbmap, struct metapage *mp; struct jfs_log *log; int lsn, difft, diffp; + unsigned long flags; /* the blocks better be within the mapsize. */ if (blkno + nblocks > bmp->db_mapsize) { @@ -504,6 +505,7 @@ dbUpdatePMap(struct inode *ipbmap, 0); if (mp == NULL) return -EIO; + metapage_wait_for_io(mp); } dp = (struct dmap *) mp->data; @@ -578,34 +580,32 @@ dbUpdatePMap(struct inode *ipbmap, if (mp->lsn != 0) { /* inherit older/smaller lsn */ logdiff(diffp, mp->lsn, log); + LOGSYNC_LOCK(log, flags); if (difft < diffp) { mp->lsn = lsn; /* move bp after tblock in logsync list */ - LOGSYNC_LOCK(log); list_move(&mp->synclist, &tblk->synclist); - LOGSYNC_UNLOCK(log); } /* inherit younger/larger clsn */ - LOGSYNC_LOCK(log); logdiff(difft, tblk->clsn, log); logdiff(diffp, mp->clsn, log); if (difft > diffp) mp->clsn = tblk->clsn; - LOGSYNC_UNLOCK(log); + LOGSYNC_UNLOCK(log, flags); } else { mp->log = log; mp->lsn = lsn; /* insert bp after tblock in logsync list */ - LOGSYNC_LOCK(log); + LOGSYNC_LOCK(log, flags); log->count++; list_add(&mp->synclist, &tblk->synclist); mp->clsn = tblk->clsn; - LOGSYNC_UNLOCK(log); + LOGSYNC_UNLOCK(log, flags); } } diff --git a/fs/jfs/jfs_dtree.c b/fs/jfs/jfs_dtree.c index e357890adfb..ac41f72d6d5 100644 --- a/fs/jfs/jfs_dtree.c +++ b/fs/jfs/jfs_dtree.c @@ -212,7 +212,7 @@ static struct metapage *read_index_page(struct inode *inode, s64 blkno) s32 xlen; rc = xtLookup(inode, blkno, 1, &xflag, &xaddr, &xlen, 1); - if (rc || (xlen == 0)) + if (rc || (xaddr == 0)) return NULL; return read_metapage(inode, xaddr, PSIZE, 1); @@ -231,7 +231,7 @@ static struct metapage *get_index_page(struct inode *inode, s64 blkno) s32 xlen; rc = xtLookup(inode, blkno, 1, &xflag, &xaddr, &xlen, 1); - if (rc || (xlen == 0)) + if (rc || (xaddr == 0)) return NULL; return get_metapage(inode, xaddr, PSIZE, 1); @@ -3181,7 +3181,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) d = (struct ldtentry *) & p->slot[stbl[i]]; if (((long) jfs_dirent + d->namlen + 1) > - (dirent_buf + PSIZE)) { + (dirent_buf + PAGE_SIZE)) { /* DBCS codepages could overrun dirent_buf */ index = i; overflow = 1; diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index 78383130162..7acff2ce3c8 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -502,7 +502,7 @@ struct inode *diReadSpecial(struct super_block *sb, ino_t inum, int secondary) } - ip->i_mapping->a_ops = &jfs_aops; + ip->i_mapping->a_ops = &jfs_metapage_aops; mapping_set_gfp_mask(ip->i_mapping, GFP_NOFS); /* Allocations to metadata inodes should not affect quotas */ @@ -2573,9 +2573,18 @@ diNewIAG(struct inomap * imap, int *iagnop, int agno, struct metapage ** mpp) goto out; } - /* assign a buffer for the page */ - mp = get_metapage(ipimap, xaddr, PSIZE, 1); - if (!mp) { + /* + * start transaction of update of the inode map + * addressing structure pointing to the new iag page; + */ + tid = txBegin(sb, COMMIT_FORCE); + down(&JFS_IP(ipimap)->commit_sem); + + /* update the inode map addressing structure to point to it */ + if ((rc = + xtInsert(tid, ipimap, 0, blkno, xlen, &xaddr, 0))) { + txEnd(tid); + up(&JFS_IP(ipimap)->commit_sem); /* Free the blocks allocated for the iag since it was * not successfully added to the inode map */ @@ -2584,6 +2593,29 @@ diNewIAG(struct inomap * imap, int *iagnop, int agno, struct metapage ** mpp) /* release the inode map lock */ IWRITE_UNLOCK(ipimap); + goto out; + } + + /* update the inode map's inode to reflect the extension */ + ipimap->i_size += PSIZE; + inode_add_bytes(ipimap, PSIZE); + + /* assign a buffer for the page */ + mp = get_metapage(ipimap, blkno, PSIZE, 0); + if (!mp) { + /* + * This is very unlikely since we just created the + * extent, but let's try to handle it correctly + */ + xtTruncate(tid, ipimap, ipimap->i_size - PSIZE, + COMMIT_PWMAP); + + txAbort(tid, 0); + txEnd(tid); + + /* release the inode map lock */ + IWRITE_UNLOCK(ipimap); + rc = -EIO; goto out; } @@ -2605,41 +2637,11 @@ diNewIAG(struct inomap * imap, int *iagnop, int agno, struct metapage ** mpp) iagp->inosmap[i] = cpu_to_le32(ONES); /* - * Invalidate the page after writing and syncing it. - * After it's initialized, we access it in a different - * address space + * Write and sync the metapage */ - set_bit(META_discard, &mp->flag); flush_metapage(mp); /* - * start tyransaction of update of the inode map - * addressing structure pointing to the new iag page; - */ - tid = txBegin(sb, COMMIT_FORCE); - down(&JFS_IP(ipimap)->commit_sem); - - /* update the inode map addressing structure to point to it */ - if ((rc = - xtInsert(tid, ipimap, 0, blkno, xlen, &xaddr, 0))) { - txEnd(tid); - up(&JFS_IP(ipimap)->commit_sem); - /* Free the blocks allocated for the iag since it was - * not successfully added to the inode map - */ - dbFree(ipimap, xaddr, (s64) xlen); - - /* release the inode map lock */ - IWRITE_UNLOCK(ipimap); - - goto out; - } - - /* update the inode map's inode to reflect the extension */ - ipimap->i_size += PSIZE; - inode_add_bytes(ipimap, PSIZE); - - /* * txCommit(COMMIT_FORCE) will synchronously write address * index pages and inode after commit in careful update order * of address index pages (right to left, bottom up); @@ -2789,6 +2791,7 @@ diUpdatePMap(struct inode *ipimap, u32 mask; struct jfs_log *log; int lsn, difft, diffp; + unsigned long flags; imap = JFS_IP(ipimap)->i_imap; /* get the iag number containing the inode */ @@ -2805,6 +2808,7 @@ diUpdatePMap(struct inode *ipimap, IREAD_UNLOCK(ipimap); if (rc) return (rc); + metapage_wait_for_io(mp); iagp = (struct iag *) mp->data; /* get the inode number and extent number of the inode within * the iag and the inode number within the extent. @@ -2868,30 +2872,28 @@ diUpdatePMap(struct inode *ipimap, /* inherit older/smaller lsn */ logdiff(difft, lsn, log); logdiff(diffp, mp->lsn, log); + LOGSYNC_LOCK(log, flags); if (difft < diffp) { mp->lsn = lsn; /* move mp after tblock in logsync list */ - LOGSYNC_LOCK(log); list_move(&mp->synclist, &tblk->synclist); - LOGSYNC_UNLOCK(log); } /* inherit younger/larger clsn */ - LOGSYNC_LOCK(log); assert(mp->clsn); logdiff(difft, tblk->clsn, log); logdiff(diffp, mp->clsn, log); if (difft > diffp) mp->clsn = tblk->clsn; - LOGSYNC_UNLOCK(log); + LOGSYNC_UNLOCK(log, flags); } else { mp->log = log; mp->lsn = lsn; /* insert mp after tblock in logsync list */ - LOGSYNC_LOCK(log); + LOGSYNC_LOCK(log, flags); log->count++; list_add(&mp->synclist, &tblk->synclist); mp->clsn = tblk->clsn; - LOGSYNC_UNLOCK(log); + LOGSYNC_UNLOCK(log, flags); } write_metapage(mp); return (0); diff --git a/fs/jfs/jfs_incore.h b/fs/jfs/jfs_incore.h index ebd77c1bed6..c0fd7b3eadc 100644 --- a/fs/jfs/jfs_incore.h +++ b/fs/jfs/jfs_incore.h @@ -165,6 +165,7 @@ struct jfs_sb_info { /* Formerly in ipbmap */ struct bmap *bmap; /* incore bmap descriptor */ struct nls_table *nls_tab; /* current codepage */ + struct inode *direct_inode; /* metadata inode */ uint state; /* mount/recovery state */ unsigned long flag; /* mount time flags */ uint p_state; /* state prior to going no integrity */ diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index b6a6869ebb4..dfa1200daa6 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -234,6 +234,7 @@ int lmLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, int lsn; int diffp, difft; struct metapage *mp = NULL; + unsigned long flags; jfs_info("lmLog: log:0x%p tblk:0x%p, lrd:0x%p tlck:0x%p", log, tblk, lrd, tlck); @@ -254,7 +255,7 @@ int lmLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, */ lsn = log->lsn; - LOGSYNC_LOCK(log); + LOGSYNC_LOCK(log, flags); /* * initialize page lsn if first log write of the page @@ -310,7 +311,7 @@ int lmLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, } } - LOGSYNC_UNLOCK(log); + LOGSYNC_UNLOCK(log, flags); /* * write the log record @@ -334,7 +335,6 @@ int lmLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, return lsn; } - /* * NAME: lmWriteRecord() * @@ -927,9 +927,8 @@ static void lmPostGC(struct lbuf * bp) * calculate new value of i_nextsync which determines when * this code is called again. * - * this is called only from lmLog(). - * - * PARAMETER: ip - pointer to logs inode. + * PARAMETERS: log - log structure + * nosyncwait - 1 if called asynchronously * * RETURN: 0 * @@ -945,6 +944,15 @@ static int lmLogSync(struct jfs_log * log, int nosyncwait) struct lrd lrd; int lsn; struct logsyncblk *lp; + struct jfs_sb_info *sbi; + unsigned long flags; + + /* push dirty metapages out to disk */ + list_for_each_entry(sbi, &log->sb_list, log_list) { + filemap_flush(sbi->ipbmap->i_mapping); + filemap_flush(sbi->ipimap->i_mapping); + filemap_flush(sbi->direct_inode->i_mapping); + } /* * forward syncpt @@ -954,10 +962,7 @@ static int lmLogSync(struct jfs_log * log, int nosyncwait) */ if (log->sync == log->syncpt) { - LOGSYNC_LOCK(log); - /* ToDo: push dirty metapages out to disk */ -// bmLogSync(log); - + LOGSYNC_LOCK(log, flags); if (list_empty(&log->synclist)) log->sync = log->lsn; else { @@ -965,7 +970,7 @@ static int lmLogSync(struct jfs_log * log, int nosyncwait) struct logsyncblk, synclist); log->sync = lp->lsn; } - LOGSYNC_UNLOCK(log); + LOGSYNC_UNLOCK(log, flags); } @@ -974,27 +979,6 @@ static int lmLogSync(struct jfs_log * log, int nosyncwait) * reset syncpt = sync */ if (log->sync != log->syncpt) { - struct jfs_sb_info *sbi; - - /* - * We need to make sure all of the "written" metapages - * actually make it to disk - */ - list_for_each_entry(sbi, &log->sb_list, log_list) { - if (sbi->flag & JFS_NOINTEGRITY) - continue; - filemap_fdatawrite(sbi->ipbmap->i_mapping); - filemap_fdatawrite(sbi->ipimap->i_mapping); - filemap_fdatawrite(sbi->sb->s_bdev->bd_inode->i_mapping); - } - list_for_each_entry(sbi, &log->sb_list, log_list) { - if (sbi->flag & JFS_NOINTEGRITY) - continue; - filemap_fdatawait(sbi->ipbmap->i_mapping); - filemap_fdatawait(sbi->ipimap->i_mapping); - filemap_fdatawait(sbi->sb->s_bdev->bd_inode->i_mapping); - } - lrd.logtid = 0; lrd.backchain = 0; lrd.type = cpu_to_le16(LOG_SYNCPT); @@ -1066,6 +1050,18 @@ static int lmLogSync(struct jfs_log * log, int nosyncwait) return lsn; } +/* + * NAME: jfs_syncpt + * + * FUNCTION: write log SYNCPT record for specified log + * + * PARAMETERS: log - log structure + */ +void jfs_syncpt(struct jfs_log *log) +{ LOG_LOCK(log); + lmLogSync(log, 1); + LOG_UNLOCK(log); +} /* * NAME: lmLogOpen() @@ -1547,6 +1543,7 @@ void jfs_flush_journal(struct jfs_log *log, int wait) { int i; struct tblock *target = NULL; + struct jfs_sb_info *sbi; /* jfs_write_inode may call us during read-only mount */ if (!log) @@ -1608,12 +1605,18 @@ void jfs_flush_journal(struct jfs_log *log, int wait) if (wait < 2) return; + list_for_each_entry(sbi, &log->sb_list, log_list) { + filemap_fdatawrite(sbi->ipbmap->i_mapping); + filemap_fdatawrite(sbi->ipimap->i_mapping); + filemap_fdatawrite(sbi->direct_inode->i_mapping); + } + /* * If there was recent activity, we may need to wait * for the lazycommit thread to catch up */ if ((!list_empty(&log->cqueue)) || !list_empty(&log->synclist)) { - for (i = 0; i < 800; i++) { /* Too much? */ + for (i = 0; i < 200; i++) { /* Too much? */ msleep(250); if (list_empty(&log->cqueue) && list_empty(&log->synclist)) @@ -1621,7 +1624,24 @@ void jfs_flush_journal(struct jfs_log *log, int wait) } } assert(list_empty(&log->cqueue)); - assert(list_empty(&log->synclist)); + if (!list_empty(&log->synclist)) { + struct logsyncblk *lp; + + list_for_each_entry(lp, &log->synclist, synclist) { + if (lp->xflag & COMMIT_PAGE) { + struct metapage *mp = (struct metapage *)lp; + dump_mem("orphan metapage", lp, + sizeof(struct metapage)); + dump_mem("page", mp->page, sizeof(struct page)); + } + else + dump_mem("orphan tblock", lp, + sizeof(struct tblock)); + } +// current->state = TASK_INTERRUPTIBLE; +// schedule(); + } + //assert(list_empty(&log->synclist)); clear_bit(log_FLUSH, &log->flag); } @@ -1669,6 +1689,7 @@ int lmLogShutdown(struct jfs_log * log) lp->h.eor = lp->t.eor = cpu_to_le16(bp->l_eor); lbmWrite(log, log->bp, lbmWRITE | lbmRELEASE | lbmSYNC, 0); lbmIOWait(log->bp, lbmFREE); + log->bp = NULL; /* * synchronous update log superblock @@ -1819,20 +1840,34 @@ static int lbmLogInit(struct jfs_log * log) log->lbuf_free = NULL; - for (i = 0; i < LOGPAGES; i++) { - lbuf = kmalloc(sizeof(struct lbuf), GFP_KERNEL); - if (lbuf == 0) - goto error; - lbuf->l_ldata = (char *) get_zeroed_page(GFP_KERNEL); - if (lbuf->l_ldata == 0) { - kfree(lbuf); + for (i = 0; i < LOGPAGES;) { + char *buffer; + uint offset; + struct page *page; + + buffer = (char *) get_zeroed_page(GFP_KERNEL); + if (buffer == NULL) goto error; + page = virt_to_page(buffer); + for (offset = 0; offset < PAGE_SIZE; offset += LOGPSIZE) { + lbuf = kmalloc(sizeof(struct lbuf), GFP_KERNEL); + if (lbuf == NULL) { + if (offset == 0) + free_page((unsigned long) buffer); + goto error; + } + if (offset) /* we already have one reference */ + get_page(page); + lbuf->l_offset = offset; + lbuf->l_ldata = buffer + offset; + lbuf->l_page = page; + lbuf->l_log = log; + init_waitqueue_head(&lbuf->l_ioevent); + + lbuf->l_freelist = log->lbuf_free; + log->lbuf_free = lbuf; + i++; } - lbuf->l_log = log; - init_waitqueue_head(&lbuf->l_ioevent); - - lbuf->l_freelist = log->lbuf_free; - log->lbuf_free = lbuf; } return (0); @@ -1857,12 +1892,10 @@ static void lbmLogShutdown(struct jfs_log * log) lbuf = log->lbuf_free; while (lbuf) { struct lbuf *next = lbuf->l_freelist; - free_page((unsigned long) lbuf->l_ldata); + __free_page(lbuf->l_page); kfree(lbuf); lbuf = next; } - - log->bp = NULL; } @@ -1974,9 +2007,9 @@ static int lbmRead(struct jfs_log * log, int pn, struct lbuf ** bpp) bio->bi_sector = bp->l_blkno << (log->l2bsize - 9); bio->bi_bdev = log->bdev; - bio->bi_io_vec[0].bv_page = virt_to_page(bp->l_ldata); + bio->bi_io_vec[0].bv_page = bp->l_page; bio->bi_io_vec[0].bv_len = LOGPSIZE; - bio->bi_io_vec[0].bv_offset = 0; + bio->bi_io_vec[0].bv_offset = bp->l_offset; bio->bi_vcnt = 1; bio->bi_idx = 0; @@ -2115,9 +2148,9 @@ static void lbmStartIO(struct lbuf * bp) bio = bio_alloc(GFP_NOFS, 1); bio->bi_sector = bp->l_blkno << (log->l2bsize - 9); bio->bi_bdev = log->bdev; - bio->bi_io_vec[0].bv_page = virt_to_page(bp->l_ldata); + bio->bi_io_vec[0].bv_page = bp->l_page; bio->bi_io_vec[0].bv_len = LOGPSIZE; - bio->bi_io_vec[0].bv_offset = 0; + bio->bi_io_vec[0].bv_offset = bp->l_offset; bio->bi_vcnt = 1; bio->bi_idx = 0; @@ -2127,16 +2160,13 @@ static void lbmStartIO(struct lbuf * bp) bio->bi_private = bp; /* check if journaling to disk has been disabled */ - if (!log->no_integrity) { + if (log->no_integrity) { + bio->bi_size = 0; + lbmIODone(bio, 0, 0); + } else { submit_bio(WRITE_SYNC, bio); INCREMENT(lmStat.submitted); } - else { - bio->bi_size = 0; - lbmIODone(bio, 0, 0); /* 2nd argument appears to not be used => 0 - * 3rd argument appears to not be used => 0 - */ - } } diff --git a/fs/jfs/jfs_logmgr.h b/fs/jfs/jfs_logmgr.h index 141ad74010c..51291fbc420 100644 --- a/fs/jfs/jfs_logmgr.h +++ b/fs/jfs/jfs_logmgr.h @@ -463,9 +463,10 @@ struct lbuf { s64 l_blkno; /* 8: log page block number */ caddr_t l_ldata; /* 4: data page */ + struct page *l_page; /* The page itself */ + uint l_offset; /* Offset of l_ldata within the page */ wait_queue_head_t l_ioevent; /* 4: i/o done event */ - struct page *l_page; /* The page itself */ }; /* Reuse l_freelist for redrive list */ @@ -489,8 +490,9 @@ struct logsyncblk { */ #define LOGSYNC_LOCK_INIT(log) spin_lock_init(&(log)->synclock) -#define LOGSYNC_LOCK(log) spin_lock(&(log)->synclock) -#define LOGSYNC_UNLOCK(log) spin_unlock(&(log)->synclock) +#define LOGSYNC_LOCK(log, flags) spin_lock_irqsave(&(log)->synclock, flags) +#define LOGSYNC_UNLOCK(log, flags) \ + spin_unlock_irqrestore(&(log)->synclock, flags) /* compute the difference in bytes of lsn from sync point */ #define logdiff(diff, lsn, log)\ @@ -506,5 +508,6 @@ extern int lmLogShutdown(struct jfs_log * log); extern int lmLogInit(struct jfs_log * log); extern int lmLogFormat(struct jfs_log *log, s64 logAddress, int logSize); extern void jfs_flush_journal(struct jfs_log * log, int wait); +extern void jfs_syncpt(struct jfs_log *log); #endif /* _H_JFS_LOGMGR */ diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 4c0a3ac75c0..41bf078dce0 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -1,5 +1,5 @@ /* - * Copyright (C) International Business Machines Corp., 2000-2003 + * Copyright (C) International Business Machines Corp., 2000-2005 * Portions Copyright (C) Christoph Hellwig, 2001-2002 * * This program is free software; you can redistribute it and/or modify @@ -18,10 +18,11 @@ */ #include <linux/fs.h> +#include <linux/mm.h> +#include <linux/bio.h> #include <linux/init.h> #include <linux/buffer_head.h> #include <linux/mempool.h> -#include <linux/delay.h> #include "jfs_incore.h" #include "jfs_superblock.h" #include "jfs_filsys.h" @@ -29,8 +30,6 @@ #include "jfs_txnmgr.h" #include "jfs_debug.h" -static DEFINE_SPINLOCK(meta_lock); - #ifdef CONFIG_JFS_STATISTICS static struct { uint pagealloc; /* # of page allocations */ @@ -39,22 +38,8 @@ static struct { } mpStat; #endif - -#define HASH_BITS 10 /* This makes hash_table 1 4K page */ -#define HASH_SIZE (1 << HASH_BITS) -static struct metapage **hash_table = NULL; -static unsigned long hash_order; - - -static inline int metapage_locked(struct metapage *mp) -{ - return test_bit(META_locked, &mp->flag); -} - -static inline int trylock_metapage(struct metapage *mp) -{ - return test_and_set_bit(META_locked, &mp->flag); -} +#define metapage_locked(mp) test_bit(META_locked, &(mp)->flag) +#define trylock_metapage(mp) test_and_set_bit(META_locked, &(mp)->flag) static inline void unlock_metapage(struct metapage *mp) { @@ -62,26 +47,26 @@ static inline void unlock_metapage(struct metapage *mp) wake_up(&mp->wait); } -static void __lock_metapage(struct metapage *mp) +static inline void __lock_metapage(struct metapage *mp) { DECLARE_WAITQUEUE(wait, current); - INCREMENT(mpStat.lockwait); - add_wait_queue_exclusive(&mp->wait, &wait); do { set_current_state(TASK_UNINTERRUPTIBLE); if (metapage_locked(mp)) { - spin_unlock(&meta_lock); + unlock_page(mp->page); schedule(); - spin_lock(&meta_lock); + lock_page(mp->page); } } while (trylock_metapage(mp)); __set_current_state(TASK_RUNNING); remove_wait_queue(&mp->wait, &wait); } -/* needs meta_lock */ +/* + * Must have mp->page locked + */ static inline void lock_metapage(struct metapage *mp) { if (trylock_metapage(mp)) @@ -92,6 +77,110 @@ static inline void lock_metapage(struct metapage *mp) static kmem_cache_t *metapage_cache; static mempool_t *metapage_mempool; +#define MPS_PER_PAGE (PAGE_CACHE_SIZE >> L2PSIZE) + +#if MPS_PER_PAGE > 1 + +struct meta_anchor { + int mp_count; + atomic_t io_count; + struct metapage *mp[MPS_PER_PAGE]; +}; +#define mp_anchor(page) ((struct meta_anchor *)page->private) + +static inline struct metapage *page_to_mp(struct page *page, uint offset) +{ + if (!PagePrivate(page)) + return NULL; + return mp_anchor(page)->mp[offset >> L2PSIZE]; +} + +static inline int insert_metapage(struct page *page, struct metapage *mp) +{ + struct meta_anchor *a; + int index; + int l2mp_blocks; /* log2 blocks per metapage */ + + if (PagePrivate(page)) + a = mp_anchor(page); + else { + a = kmalloc(sizeof(struct meta_anchor), GFP_NOFS); + if (!a) + return -ENOMEM; + memset(a, 0, sizeof(struct meta_anchor)); + page->private = (unsigned long)a; + SetPagePrivate(page); + kmap(page); + } + + if (mp) { + l2mp_blocks = L2PSIZE - page->mapping->host->i_blkbits; + index = (mp->index >> l2mp_blocks) & (MPS_PER_PAGE - 1); + a->mp_count++; + a->mp[index] = mp; + } + + return 0; +} + +static inline void remove_metapage(struct page *page, struct metapage *mp) +{ + struct meta_anchor *a = mp_anchor(page); + int l2mp_blocks = L2PSIZE - page->mapping->host->i_blkbits; + int index; + + index = (mp->index >> l2mp_blocks) & (MPS_PER_PAGE - 1); + + BUG_ON(a->mp[index] != mp); + + a->mp[index] = NULL; + if (--a->mp_count == 0) { + kfree(a); + page->private = 0; + ClearPagePrivate(page); + kunmap(page); + } +} + +static inline void inc_io(struct page *page) +{ + atomic_inc(&mp_anchor(page)->io_count); +} + +static inline void dec_io(struct page *page, void (*handler) (struct page *)) +{ + if (atomic_dec_and_test(&mp_anchor(page)->io_count)) + handler(page); +} + +#else +static inline struct metapage *page_to_mp(struct page *page, uint offset) +{ + return PagePrivate(page) ? (struct metapage *)page->private : NULL; +} + +static inline int insert_metapage(struct page *page, struct metapage *mp) +{ + if (mp) { + page->private = (unsigned long)mp; + SetPagePrivate(page); + kmap(page); + } + return 0; +} + +static inline void remove_metapage(struct page *page, struct metapage *mp) +{ + page->private = 0; + ClearPagePrivate(page); + kunmap(page); +} + +#define inc_io(page) do {} while(0) +#define dec_io(page, handler) handler(page) + +#endif + static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) { struct metapage *mp = (struct metapage *)foo; @@ -139,16 +228,6 @@ int __init metapage_init(void) kmem_cache_destroy(metapage_cache); return -ENOMEM; } - /* - * Now the hash list - */ - for (hash_order = 0; - ((PAGE_SIZE << hash_order) / sizeof(void *)) < HASH_SIZE; - hash_order++); - hash_table = - (struct metapage **) __get_free_pages(GFP_KERNEL, hash_order); - assert(hash_table); - memset(hash_table, 0, PAGE_SIZE << hash_order); return 0; } @@ -159,73 +238,388 @@ void metapage_exit(void) kmem_cache_destroy(metapage_cache); } +static inline void drop_metapage(struct page *page, struct metapage *mp) +{ + if (mp->count || mp->nohomeok || test_bit(META_dirty, &mp->flag) || + test_bit(META_io, &mp->flag)) + return; + remove_metapage(page, mp); + INCREMENT(mpStat.pagefree); + free_metapage(mp); +} + /* - * Basically same hash as in pagemap.h, but using our hash table + * Metapage address space operations */ -static struct metapage **meta_hash(struct address_space *mapping, - unsigned long index) + +static sector_t metapage_get_blocks(struct inode *inode, sector_t lblock, + unsigned int *len) { -#define i (((unsigned long)mapping)/ \ - (sizeof(struct inode) & ~(sizeof(struct inode) -1 ))) -#define s(x) ((x) + ((x) >> HASH_BITS)) - return hash_table + (s(i + index) & (HASH_SIZE - 1)); -#undef i -#undef s + int rc = 0; + int xflag; + s64 xaddr; + sector_t file_blocks = (inode->i_size + inode->i_blksize - 1) >> + inode->i_blkbits; + + if (lblock >= file_blocks) + return 0; + if (lblock + *len > file_blocks) + *len = file_blocks - lblock; + + if (inode->i_ino) { + rc = xtLookup(inode, (s64)lblock, *len, &xflag, &xaddr, len, 0); + if ((rc == 0) && *len) + lblock = (sector_t)xaddr; + else + lblock = 0; + } /* else no mapping */ + + return lblock; } -static struct metapage *search_hash(struct metapage ** hash_ptr, - struct address_space *mapping, - unsigned long index) +static void last_read_complete(struct page *page) { - struct metapage *ptr; + if (!PageError(page)) + SetPageUptodate(page); + unlock_page(page); +} + +static int metapage_read_end_io(struct bio *bio, unsigned int bytes_done, + int err) +{ + struct page *page = bio->bi_private; + + if (bio->bi_size) + return 1; - for (ptr = *hash_ptr; ptr; ptr = ptr->hash_next) { - if ((ptr->mapping == mapping) && (ptr->index == index)) - return ptr; + if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) { + printk(KERN_ERR "metapage_read_end_io: I/O error\n"); + SetPageError(page); } - return NULL; + dec_io(page, last_read_complete); + bio_put(bio); + + return 0; } -static void add_to_hash(struct metapage * mp, struct metapage ** hash_ptr) +static void remove_from_logsync(struct metapage *mp) { - if (*hash_ptr) - (*hash_ptr)->hash_prev = mp; + struct jfs_log *log = mp->log; + unsigned long flags; +/* + * This can race. Recheck that log hasn't been set to null, and after + * acquiring logsync lock, recheck lsn + */ + if (!log) + return; + + LOGSYNC_LOCK(log, flags); + if (mp->lsn) { + mp->log = NULL; + mp->lsn = 0; + mp->clsn = 0; + log->count--; + list_del(&mp->synclist); + } + LOGSYNC_UNLOCK(log, flags); +} - mp->hash_prev = NULL; - mp->hash_next = *hash_ptr; - *hash_ptr = mp; +static void last_write_complete(struct page *page) +{ + struct metapage *mp; + unsigned int offset; + + for (offset = 0; offset < PAGE_CACHE_SIZE; offset += PSIZE) { + mp = page_to_mp(page, offset); + if (mp && test_bit(META_io, &mp->flag)) { + if (mp->lsn) + remove_from_logsync(mp); + clear_bit(META_io, &mp->flag); + } + /* + * I'd like to call drop_metapage here, but I don't think it's + * safe unless I have the page locked + */ + } + end_page_writeback(page); } -static void remove_from_hash(struct metapage * mp, struct metapage ** hash_ptr) +static int metapage_write_end_io(struct bio *bio, unsigned int bytes_done, + int err) { - if (mp->hash_prev) - mp->hash_prev->hash_next = mp->hash_next; - else { - assert(*hash_ptr == mp); - *hash_ptr = mp->hash_next; + struct page *page = bio->bi_private; + + BUG_ON(!PagePrivate(page)); + + if (bio->bi_size) + return 1; + + if (! test_bit(BIO_UPTODATE, &bio->bi_flags)) { + printk(KERN_ERR "metapage_write_end_io: I/O error\n"); + SetPageError(page); + } + dec_io(page, last_write_complete); + bio_put(bio); + return 0; +} + +static int metapage_writepage(struct page *page, struct writeback_control *wbc) +{ + struct bio *bio = NULL; + unsigned int block_offset; /* block offset of mp within page */ + struct inode *inode = page->mapping->host; + unsigned int blocks_per_mp = JFS_SBI(inode->i_sb)->nbperpage; + unsigned int len; + unsigned int xlen; + struct metapage *mp; + int redirty = 0; + sector_t lblock; + sector_t pblock; + sector_t next_block = 0; + sector_t page_start; + unsigned long bio_bytes = 0; + unsigned long bio_offset = 0; + unsigned int offset; + + page_start = (sector_t)page->index << + (PAGE_CACHE_SHIFT - inode->i_blkbits); + BUG_ON(!PageLocked(page)); + BUG_ON(PageWriteback(page)); + + for (offset = 0; offset < PAGE_CACHE_SIZE; offset += PSIZE) { + mp = page_to_mp(page, offset); + + if (!mp || !test_bit(META_dirty, &mp->flag)) + continue; + + if (mp->nohomeok && !test_bit(META_forcewrite, &mp->flag)) { + redirty = 1; + continue; + } + + clear_bit(META_dirty, &mp->flag); + block_offset = offset >> inode->i_blkbits; + lblock = page_start + block_offset; + if (bio) { + if (xlen && lblock == next_block) { + /* Contiguous, in memory & on disk */ + len = min(xlen, blocks_per_mp); + xlen -= len; + bio_bytes += len << inode->i_blkbits; + set_bit(META_io, &mp->flag); + continue; + } + /* Not contiguous */ + if (bio_add_page(bio, page, bio_bytes, bio_offset) < + bio_bytes) + goto add_failed; + /* + * Increment counter before submitting i/o to keep + * count from hitting zero before we're through + */ + inc_io(page); + if (!bio->bi_size) + goto dump_bio; + submit_bio(WRITE, bio); + bio = NULL; + } else { + set_page_writeback(page); + inc_io(page); + } + xlen = (PAGE_CACHE_SIZE - offset) >> inode->i_blkbits; + pblock = metapage_get_blocks(inode, lblock, &xlen); + if (!pblock) { + /* Need better error handling */ + printk(KERN_ERR "JFS: metapage_get_blocks failed\n"); + dec_io(page, last_write_complete); + continue; + } + set_bit(META_io, &mp->flag); + len = min(xlen, (uint) JFS_SBI(inode->i_sb)->nbperpage); + + bio = bio_alloc(GFP_NOFS, 1); + bio->bi_bdev = inode->i_sb->s_bdev; + bio->bi_sector = pblock << (inode->i_blkbits - 9); + bio->bi_end_io = metapage_write_end_io; + bio->bi_private = page; + + /* Don't call bio_add_page yet, we may add to this vec */ + bio_offset = offset; + bio_bytes = len << inode->i_blkbits; + + xlen -= len; + next_block = lblock + len; + } + if (bio) { + if (bio_add_page(bio, page, bio_bytes, bio_offset) < bio_bytes) + goto add_failed; + if (!bio->bi_size) + goto dump_bio; + + submit_bio(WRITE, bio); + } + if (redirty) + redirty_page_for_writepage(wbc, page); + + unlock_page(page); + + return 0; +add_failed: + /* We should never reach here, since we're only adding one vec */ + printk(KERN_ERR "JFS: bio_add_page failed unexpectedly\n"); + goto skip; +dump_bio: + dump_mem("bio", bio, sizeof(*bio)); +skip: + bio_put(bio); + unlock_page(page); + dec_io(page, last_write_complete); + + return -EIO; +} + +static int metapage_readpage(struct file *fp, struct page *page) +{ + struct inode *inode = page->mapping->host; + struct bio *bio = NULL; + unsigned int block_offset; + unsigned int blocks_per_page = PAGE_CACHE_SIZE >> inode->i_blkbits; + sector_t page_start; /* address of page in fs blocks */ + sector_t pblock; + unsigned int xlen; + unsigned int len; + unsigned int offset; + + BUG_ON(!PageLocked(page)); + page_start = (sector_t)page->index << + (PAGE_CACHE_SHIFT - inode->i_blkbits); + + block_offset = 0; + while (block_offset < blocks_per_page) { + xlen = blocks_per_page - block_offset; + pblock = metapage_get_blocks(inode, page_start + block_offset, + &xlen); + if (pblock) { + if (!PagePrivate(page)) + insert_metapage(page, NULL); + inc_io(page); + if (bio) + submit_bio(READ, bio); + + bio = bio_alloc(GFP_NOFS, 1); + bio->bi_bdev = inode->i_sb->s_bdev; + bio->bi_sector = pblock << (inode->i_blkbits - 9); + bio->bi_end_io = metapage_read_end_io; + bio->bi_private = page; + len = xlen << inode->i_blkbits; + offset = block_offset << inode->i_blkbits; + if (bio_add_page(bio, page, len, offset) < len) + goto add_failed; + block_offset += xlen; + } else + block_offset++; } + if (bio) + submit_bio(READ, bio); + else + unlock_page(page); + + return 0; - if (mp->hash_next) - mp->hash_next->hash_prev = mp->hash_prev; +add_failed: + printk(KERN_ERR "JFS: bio_add_page failed unexpectedly\n"); + bio_put(bio); + dec_io(page, last_read_complete); + return -EIO; } +static int metapage_releasepage(struct page *page, int gfp_mask) +{ + struct metapage *mp; + int busy = 0; + unsigned int offset; + + for (offset = 0; offset < PAGE_CACHE_SIZE; offset += PSIZE) { + mp = page_to_mp(page, offset); + + if (!mp) + continue; + + jfs_info("metapage_releasepage: mp = 0x%p", mp); + if (mp->count || mp->nohomeok) { + jfs_info("count = %ld, nohomeok = %d", mp->count, + mp->nohomeok); + busy = 1; + continue; + } + wait_on_page_writeback(page); + //WARN_ON(test_bit(META_dirty, &mp->flag)); + if (test_bit(META_dirty, &mp->flag)) { + dump_mem("dirty mp in metapage_releasepage", mp, + sizeof(struct metapage)); + dump_mem("page", page, sizeof(struct page)); + dump_stack(); + } + WARN_ON(mp->lsn); + if (mp->lsn) + remove_from_logsync(mp); + remove_metapage(page, mp); + INCREMENT(mpStat.pagefree); + free_metapage(mp); + } + if (busy) + return -1; + + return 0; +} + +static int metapage_invalidatepage(struct page *page, unsigned long offset) +{ + BUG_ON(offset); + + if (PageWriteback(page)) + return 0; + + return metapage_releasepage(page, 0); +} + +struct address_space_operations jfs_metapage_aops = { + .readpage = metapage_readpage, + .writepage = metapage_writepage, + .sync_page = block_sync_page, + .releasepage = metapage_releasepage, + .invalidatepage = metapage_invalidatepage, + .set_page_dirty = __set_page_dirty_nobuffers, +}; + struct metapage *__get_metapage(struct inode *inode, unsigned long lblock, unsigned int size, int absolute, unsigned long new) { - struct metapage **hash_ptr; int l2BlocksPerPage; int l2bsize; struct address_space *mapping; - struct metapage *mp; + struct metapage *mp = NULL; + struct page *page; unsigned long page_index; unsigned long page_offset; - jfs_info("__get_metapage: inode = 0x%p, lblock = 0x%lx", inode, lblock); - + jfs_info("__get_metapage: ino = %ld, lblock = 0x%lx, abs=%d", + inode->i_ino, lblock, absolute); + + l2bsize = inode->i_blkbits; + l2BlocksPerPage = PAGE_CACHE_SHIFT - l2bsize; + page_index = lblock >> l2BlocksPerPage; + page_offset = (lblock - (page_index << l2BlocksPerPage)) << l2bsize; + if ((page_offset + size) > PAGE_CACHE_SIZE) { + jfs_err("MetaData crosses page boundary!!"); + jfs_err("lblock = %lx, size = %d", lblock, size); + dump_stack(); + return NULL; + } if (absolute) - mapping = inode->i_sb->s_bdev->bd_inode->i_mapping; + mapping = JFS_SBI(inode->i_sb)->direct_inode->i_mapping; else { /* * If an nfs client tries to read an inode that is larger @@ -237,312 +631,212 @@ struct metapage *__get_metapage(struct inode *inode, unsigned long lblock, mapping = inode->i_mapping; } - hash_ptr = meta_hash(mapping, lblock); -again: - spin_lock(&meta_lock); - mp = search_hash(hash_ptr, mapping, lblock); + if (new && (PSIZE == PAGE_CACHE_SIZE)) { + page = grab_cache_page(mapping, page_index); + if (!page) { + jfs_err("grab_cache_page failed!"); + return NULL; + } + SetPageUptodate(page); + } else { + page = read_cache_page(mapping, page_index, + (filler_t *)mapping->a_ops->readpage, NULL); + if (IS_ERR(page)) { + jfs_err("read_cache_page failed!"); + return NULL; + } + lock_page(page); + } + + mp = page_to_mp(page, page_offset); if (mp) { - page_found: - if (test_bit(META_stale, &mp->flag)) { - spin_unlock(&meta_lock); - msleep(1); - goto again; + if (mp->logical_size != size) { + jfs_error(inode->i_sb, + "__get_metapage: mp->logical_size != size"); + jfs_err("logical_size = %d, size = %d", + mp->logical_size, size); + dump_stack(); + goto unlock; } mp->count++; lock_metapage(mp); - spin_unlock(&meta_lock); if (test_bit(META_discard, &mp->flag)) { if (!new) { jfs_error(inode->i_sb, "__get_metapage: using a " "discarded metapage"); - release_metapage(mp); - return NULL; + discard_metapage(mp); + goto unlock; } clear_bit(META_discard, &mp->flag); } - jfs_info("__get_metapage: found 0x%p, in hash", mp); - if (mp->logical_size != size) { - jfs_error(inode->i_sb, - "__get_metapage: mp->logical_size != size"); - release_metapage(mp); - return NULL; - } } else { - l2bsize = inode->i_blkbits; - l2BlocksPerPage = PAGE_CACHE_SHIFT - l2bsize; - page_index = lblock >> l2BlocksPerPage; - page_offset = (lblock - (page_index << l2BlocksPerPage)) << - l2bsize; - if ((page_offset + size) > PAGE_CACHE_SIZE) { - spin_unlock(&meta_lock); - jfs_err("MetaData crosses page boundary!!"); - return NULL; - } - - /* - * Locks held on aggregate inode pages are usually - * not held long, and they are taken in critical code - * paths (committing dirty inodes, txCommit thread) - * - * Attempt to get metapage without blocking, tapping into - * reserves if necessary. - */ - mp = NULL; - if (JFS_IP(inode)->fileset == AGGREGATE_I) { - mp = alloc_metapage(GFP_ATOMIC); - if (!mp) { - /* - * mempool is supposed to protect us from - * failing here. We will try a blocking - * call, but a deadlock is possible here - */ - printk(KERN_WARNING - "__get_metapage: atomic call to mempool_alloc failed.\n"); - printk(KERN_WARNING - "Will attempt blocking call\n"); - } - } - if (!mp) { - struct metapage *mp2; - - spin_unlock(&meta_lock); - mp = alloc_metapage(GFP_NOFS); - spin_lock(&meta_lock); - - /* we dropped the meta_lock, we need to search the - * hash again. - */ - mp2 = search_hash(hash_ptr, mapping, lblock); - if (mp2) { - free_metapage(mp); - mp = mp2; - goto page_found; - } - } + INCREMENT(mpStat.pagealloc); + mp = alloc_metapage(GFP_NOFS); + mp->page = page; mp->flag = 0; - lock_metapage(mp); - if (absolute) - set_bit(META_absolute, &mp->flag); mp->xflag = COMMIT_PAGE; mp->count = 1; - atomic_set(&mp->nohomeok,0); - mp->mapping = mapping; - mp->index = lblock; - mp->page = NULL; + mp->nohomeok = 0; mp->logical_size = size; - add_to_hash(mp, hash_ptr); - spin_unlock(&meta_lock); - - if (new) { - jfs_info("__get_metapage: Calling grab_cache_page"); - mp->page = grab_cache_page(mapping, page_index); - if (!mp->page) { - jfs_err("grab_cache_page failed!"); - goto freeit; - } else { - INCREMENT(mpStat.pagealloc); - unlock_page(mp->page); - } - } else { - jfs_info("__get_metapage: Calling read_cache_page"); - mp->page = read_cache_page(mapping, lblock, - (filler_t *)mapping->a_ops->readpage, NULL); - if (IS_ERR(mp->page)) { - jfs_err("read_cache_page failed!"); - goto freeit; - } else - INCREMENT(mpStat.pagealloc); + mp->data = page_address(page) + page_offset; + mp->index = lblock; + if (unlikely(insert_metapage(page, mp))) { + free_metapage(mp); + goto unlock; } - mp->data = kmap(mp->page) + page_offset; + lock_metapage(mp); } - if (new) + if (new) { + jfs_info("zeroing mp = 0x%p", mp); memset(mp->data, 0, PSIZE); + } - jfs_info("__get_metapage: returning = 0x%p", mp); + unlock_page(page); + jfs_info("__get_metapage: returning = 0x%p data = 0x%p", mp, mp->data); return mp; -freeit: - spin_lock(&meta_lock); - remove_from_hash(mp, hash_ptr); - free_metapage(mp); - spin_unlock(&meta_lock); +unlock: + unlock_page(page); return NULL; } -void hold_metapage(struct metapage * mp, int force) +void grab_metapage(struct metapage * mp) { - spin_lock(&meta_lock); - + jfs_info("grab_metapage: mp = 0x%p", mp); + page_cache_get(mp->page); + lock_page(mp->page); mp->count++; - - if (force) { - ASSERT (!(test_bit(META_forced, &mp->flag))); - if (trylock_metapage(mp)) - set_bit(META_forced, &mp->flag); - } else - lock_metapage(mp); - - spin_unlock(&meta_lock); + lock_metapage(mp); + unlock_page(mp->page); } -static void __write_metapage(struct metapage * mp) +void force_metapage(struct metapage *mp) { - int l2bsize = mp->mapping->host->i_blkbits; - int l2BlocksPerPage = PAGE_CACHE_SHIFT - l2bsize; - unsigned long page_index; - unsigned long page_offset; - int rc; - - jfs_info("__write_metapage: mp = 0x%p", mp); - - page_index = mp->page->index; - page_offset = - (mp->index - (page_index << l2BlocksPerPage)) << l2bsize; + struct page *page = mp->page; + jfs_info("force_metapage: mp = 0x%p", mp); + set_bit(META_forcewrite, &mp->flag); + clear_bit(META_sync, &mp->flag); + page_cache_get(page); + lock_page(page); + set_page_dirty(page); + write_one_page(page, 1); + clear_bit(META_forcewrite, &mp->flag); + page_cache_release(page); +} +extern void hold_metapage(struct metapage *mp) +{ lock_page(mp->page); - rc = mp->mapping->a_ops->prepare_write(NULL, mp->page, page_offset, - page_offset + - mp->logical_size); - if (rc) { - jfs_err("prepare_write return %d!", rc); - ClearPageUptodate(mp->page); +} + +extern void put_metapage(struct metapage *mp) +{ + if (mp->count || mp->nohomeok) { + /* Someone else will release this */ unlock_page(mp->page); - clear_bit(META_dirty, &mp->flag); return; } - rc = mp->mapping->a_ops->commit_write(NULL, mp->page, page_offset, - page_offset + - mp->logical_size); - if (rc) { - jfs_err("commit_write returned %d", rc); - } - + page_cache_get(mp->page); + mp->count++; + lock_metapage(mp); unlock_page(mp->page); - clear_bit(META_dirty, &mp->flag); - - jfs_info("__write_metapage done"); -} - -static inline void sync_metapage(struct metapage *mp) -{ - struct page *page = mp->page; - - page_cache_get(page); - lock_page(page); - - /* we're done with this page - no need to check for errors */ - if (page_has_buffers(page)) - write_one_page(page, 1); - else - unlock_page(page); - page_cache_release(page); + release_metapage(mp); } void release_metapage(struct metapage * mp) { - struct jfs_log *log; - + struct page *page = mp->page; jfs_info("release_metapage: mp = 0x%p, flag = 0x%lx", mp, mp->flag); - spin_lock(&meta_lock); - if (test_bit(META_forced, &mp->flag)) { - clear_bit(META_forced, &mp->flag); - mp->count--; - spin_unlock(&meta_lock); - return; - } + BUG_ON(!page); + + lock_page(page); + unlock_metapage(mp); assert(mp->count); - if (--mp->count || atomic_read(&mp->nohomeok)) { - unlock_metapage(mp); - spin_unlock(&meta_lock); + if (--mp->count || mp->nohomeok) { + unlock_page(page); + page_cache_release(page); return; } - if (mp->page) { - set_bit(META_stale, &mp->flag); - spin_unlock(&meta_lock); - kunmap(mp->page); - mp->data = NULL; - if (test_bit(META_dirty, &mp->flag)) - __write_metapage(mp); + if (test_bit(META_dirty, &mp->flag)) { + set_page_dirty(page); if (test_bit(META_sync, &mp->flag)) { - sync_metapage(mp); clear_bit(META_sync, &mp->flag); + write_one_page(page, 1); + lock_page(page); /* write_one_page unlocks the page */ } + } else if (mp->lsn) /* discard_metapage doesn't remove it */ + remove_from_logsync(mp); - if (test_bit(META_discard, &mp->flag)) { - lock_page(mp->page); - block_invalidatepage(mp->page, 0); - unlock_page(mp->page); - } - - page_cache_release(mp->page); - mp->page = NULL; - INCREMENT(mpStat.pagefree); - spin_lock(&meta_lock); - } +#if MPS_PER_PAGE == 1 + /* + * If we know this is the only thing in the page, we can throw + * the page out of the page cache. If pages are larger, we + * don't want to do this. + */ - if (mp->lsn) { - /* - * Remove metapage from logsynclist. - */ - log = mp->log; - LOGSYNC_LOCK(log); - mp->log = NULL; - mp->lsn = 0; - mp->clsn = 0; - log->count--; - list_del(&mp->synclist); - LOGSYNC_UNLOCK(log); + /* Retest mp->count since we may have released page lock */ + if (test_bit(META_discard, &mp->flag) && !mp->count) { + clear_page_dirty(page); + ClearPageUptodate(page); +#ifdef _NOT_YET + if (page->mapping) { + /* Remove from page cache and page cache reference */ + remove_from_page_cache(page); + page_cache_release(page); + metapage_releasepage(page, 0); + } +#endif } - remove_from_hash(mp, meta_hash(mp->mapping, mp->index)); - spin_unlock(&meta_lock); - - free_metapage(mp); +#else + /* Try to keep metapages from using up too much memory */ + drop_metapage(page, mp); +#endif + unlock_page(page); + page_cache_release(page); } void __invalidate_metapages(struct inode *ip, s64 addr, int len) { - struct metapage **hash_ptr; - unsigned long lblock; + sector_t lblock; int l2BlocksPerPage = PAGE_CACHE_SHIFT - ip->i_blkbits; + int BlocksPerPage = 1 << l2BlocksPerPage; /* All callers are interested in block device's mapping */ - struct address_space *mapping = ip->i_sb->s_bdev->bd_inode->i_mapping; + struct address_space *mapping = + JFS_SBI(ip->i_sb)->direct_inode->i_mapping; struct metapage *mp; struct page *page; + unsigned int offset; /* - * First, mark metapages to discard. They will eventually be + * Mark metapages to discard. They will eventually be * released, but should not be written. */ - for (lblock = addr; lblock < addr + len; - lblock += 1 << l2BlocksPerPage) { - hash_ptr = meta_hash(mapping, lblock); -again: - spin_lock(&meta_lock); - mp = search_hash(hash_ptr, mapping, lblock); - if (mp) { - if (test_bit(META_stale, &mp->flag)) { - spin_unlock(&meta_lock); - msleep(1); - goto again; - } + for (lblock = addr & ~(BlocksPerPage - 1); lblock < addr + len; + lblock += BlocksPerPage) { + page = find_lock_page(mapping, lblock >> l2BlocksPerPage); + if (!page) + continue; + for (offset = 0; offset < PAGE_CACHE_SIZE; offset += PSIZE) { + mp = page_to_mp(page, offset); + if (!mp) + continue; + if (mp->index < addr) + continue; + if (mp->index >= addr + len) + break; clear_bit(META_dirty, &mp->flag); set_bit(META_discard, &mp->flag); - spin_unlock(&meta_lock); - } else { - spin_unlock(&meta_lock); - page = find_lock_page(mapping, lblock>>l2BlocksPerPage); - if (page) { - block_invalidatepage(page, 0); - unlock_page(page); - page_cache_release(page); - } + if (mp->lsn) + remove_from_logsync(mp); } + unlock_page(page); + page_cache_release(page); } } diff --git a/fs/jfs/jfs_metapage.h b/fs/jfs/jfs_metapage.h index 0e58aba58c3..991e9fb84c7 100644 --- a/fs/jfs/jfs_metapage.h +++ b/fs/jfs/jfs_metapage.h @@ -33,38 +33,27 @@ struct metapage { unsigned long flag; /* See Below */ unsigned long count; /* Reference count */ void *data; /* Data pointer */ - - /* list management stuff */ - struct metapage *hash_prev; - struct metapage *hash_next; /* Also used for free list */ - - /* - * mapping & index become redundant, but we need these here to - * add the metapage to the hash before we have the real page - */ - struct address_space *mapping; - unsigned long index; + sector_t index; /* block address of page */ wait_queue_head_t wait; /* implementation */ struct page *page; - unsigned long logical_size; + unsigned int logical_size; /* Journal management */ int clsn; - atomic_t nohomeok; + int nohomeok; struct jfs_log *log; }; /* metapage flag */ #define META_locked 0 -#define META_absolute 1 -#define META_free 2 -#define META_dirty 3 -#define META_sync 4 -#define META_discard 5 -#define META_forced 6 -#define META_stale 7 +#define META_free 1 +#define META_dirty 2 +#define META_sync 3 +#define META_discard 4 +#define META_forcewrite 5 +#define META_io 6 #define mark_metapage_dirty(mp) set_bit(META_dirty, &(mp)->flag) @@ -80,7 +69,16 @@ extern struct metapage *__get_metapage(struct inode *inode, __get_metapage(inode, lblock, size, absolute, TRUE) extern void release_metapage(struct metapage *); -extern void hold_metapage(struct metapage *, int); +extern void grab_metapage(struct metapage *); +extern void force_metapage(struct metapage *); + +/* + * hold_metapage and put_metapage are used in conjuction. The page lock + * is not dropped between the two, so no other threads can get or release + * the metapage + */ +extern void hold_metapage(struct metapage *); +extern void put_metapage(struct metapage *); static inline void write_metapage(struct metapage *mp) { @@ -101,6 +99,46 @@ static inline void discard_metapage(struct metapage *mp) release_metapage(mp); } +static inline void metapage_nohomeok(struct metapage *mp) +{ + struct page *page = mp->page; + lock_page(page); + if (!mp->nohomeok++) { + mark_metapage_dirty(mp); + page_cache_get(page); + wait_on_page_writeback(page); + } + unlock_page(page); +} + +/* + * This serializes access to mp->lsn when metapages are added to logsynclist + * without setting nohomeok. i.e. updating imap & dmap + */ +static inline void metapage_wait_for_io(struct metapage *mp) +{ + if (test_bit(META_io, &mp->flag)) + wait_on_page_writeback(mp->page); +} + +/* + * This is called when already holding the metapage + */ +static inline void _metapage_homeok(struct metapage *mp) +{ + if (!--mp->nohomeok) + page_cache_release(mp->page); +} + +static inline void metapage_homeok(struct metapage *mp) +{ + hold_metapage(mp); + _metapage_homeok(mp); + put_metapage(mp); +} + +extern struct address_space_operations jfs_metapage_aops; + /* * This routines invalidate all pages for an extent. */ diff --git a/fs/jfs/jfs_mount.c b/fs/jfs/jfs_mount.c index c535ffd638e..032d111bc33 100644 --- a/fs/jfs/jfs_mount.c +++ b/fs/jfs/jfs_mount.c @@ -285,11 +285,6 @@ int jfs_mount_rw(struct super_block *sb, int remount) */ logMOUNT(sb); - /* - * Set page cache allocation policy - */ - mapping_set_gfp_mask(sb->s_bdev->bd_inode->i_mapping, GFP_NOFS); - return rc; } diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index f40301d93f7..e93d01aa12c 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -227,6 +227,7 @@ static lid_t txLockAlloc(void) static void txLockFree(lid_t lid) { + TxLock[lid].tid = 0; TxLock[lid].next = TxAnchor.freelock; TxAnchor.freelock = lid; TxAnchor.tlocksInUse--; @@ -566,9 +567,6 @@ void txEnd(tid_t tid) * synchronize with logsync barrier */ if (test_bit(log_SYNCBARRIER, &log->flag)) { - /* forward log syncpt */ - /* lmSync(log); */ - jfs_info("log barrier off: 0x%x", log->lsn); /* enable new transactions start */ @@ -576,15 +574,22 @@ void txEnd(tid_t tid) /* wakeup all waitors for logsync barrier */ TXN_WAKEUP(&log->syncwait); + + TXN_UNLOCK(); + + /* forward log syncpt */ + jfs_syncpt(log); + + goto wakeup; } } + TXN_UNLOCK(); +wakeup: /* * wakeup all waitors for a free tblock */ TXN_WAKEUP(&TxAnchor.freewait); - - TXN_UNLOCK(); } @@ -633,8 +638,10 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, /* is page locked by the requester transaction ? */ tlck = lid_to_tlock(lid); - if ((xtid = tlck->tid) == tid) + if ((xtid = tlck->tid) == tid) { + TXN_UNLOCK(); goto grantLock; + } /* * is page locked by anonymous transaction/lock ? @@ -649,6 +656,7 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, */ if (xtid == 0) { tlck->tid = tid; + TXN_UNLOCK(); tblk = tid_to_tblock(tid); /* * The order of the tlocks in the transaction is important @@ -706,17 +714,18 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, */ tlck->tid = tid; + TXN_UNLOCK(); + /* mark tlock for meta-data page */ if (mp->xflag & COMMIT_PAGE) { tlck->flag = tlckPAGELOCK; /* mark the page dirty and nohomeok */ - mark_metapage_dirty(mp); - atomic_inc(&mp->nohomeok); + metapage_nohomeok(mp); jfs_info("locking mp = 0x%p, nohomeok = %d tid = %d tlck = 0x%p", - mp, atomic_read(&mp->nohomeok), tid, tlck); + mp, mp->nohomeok, tid, tlck); /* if anonymous transaction, and buffer is on the group * commit synclist, mark inode to show this. This will @@ -762,8 +771,10 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, if (tlck->next == 0) { /* This inode's first anonymous transaction */ jfs_ip->atltail = lid; + TXN_LOCK(); list_add_tail(&jfs_ip->anon_inode_list, &TxAnchor.anon_list); + TXN_UNLOCK(); } } @@ -821,8 +832,6 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, grantLock: tlck->type |= type; - TXN_UNLOCK(); - return tlck; /* @@ -841,11 +850,19 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, BUG(); } INCREMENT(stattx.waitlock); /* statistics */ + TXN_UNLOCK(); release_metapage(mp); + TXN_LOCK(); + xtid = tlck->tid; /* reaquire after dropping TXN_LOCK */ jfs_info("txLock: in waitLock, tid = %d, xtid = %d, lid = %d", tid, xtid, lid); - TXN_SLEEP_DROP_LOCK(&tid_to_tblock(xtid)->waitor); + + /* Recheck everything since dropping TXN_LOCK */ + if (xtid && (tlck->mp == mp) && (mp->lid == lid)) + TXN_SLEEP_DROP_LOCK(&tid_to_tblock(xtid)->waitor); + else + TXN_UNLOCK(); jfs_info("txLock: awakened tid = %d, lid = %d", tid, lid); return NULL; @@ -906,6 +923,7 @@ static void txUnlock(struct tblock * tblk) struct metapage *mp; struct jfs_log *log; int difft, diffp; + unsigned long flags; jfs_info("txUnlock: tblk = 0x%p", tblk); log = JFS_SBI(tblk->sb)->log; @@ -925,19 +943,14 @@ static void txUnlock(struct tblock * tblk) assert(mp->xflag & COMMIT_PAGE); /* hold buffer - * - * It's possible that someone else has the metapage. - * The only things were changing are nohomeok, which - * is handled atomically, and clsn which is protected - * by the LOGSYNC_LOCK. */ - hold_metapage(mp, 1); + hold_metapage(mp); - assert(atomic_read(&mp->nohomeok) > 0); - atomic_dec(&mp->nohomeok); + assert(mp->nohomeok > 0); + _metapage_homeok(mp); /* inherit younger/larger clsn */ - LOGSYNC_LOCK(log); + LOGSYNC_LOCK(log, flags); if (mp->clsn) { logdiff(difft, tblk->clsn, log); logdiff(diffp, mp->clsn, log); @@ -945,16 +958,11 @@ static void txUnlock(struct tblock * tblk) mp->clsn = tblk->clsn; } else mp->clsn = tblk->clsn; - LOGSYNC_UNLOCK(log); + LOGSYNC_UNLOCK(log, flags); assert(!(tlck->flag & tlckFREEPAGE)); - if (tlck->flag & tlckWRITEPAGE) { - write_metapage(mp); - } else { - /* release page which has been forced */ - release_metapage(mp); - } + put_metapage(mp); } /* insert tlock, and linelock(s) of the tlock if any, @@ -981,10 +989,10 @@ static void txUnlock(struct tblock * tblk) * has been inserted in logsync list at txUpdateMap()) */ if (tblk->lsn) { - LOGSYNC_LOCK(log); + LOGSYNC_LOCK(log, flags); log->count--; list_del(&tblk->synclist); - LOGSYNC_UNLOCK(log); + LOGSYNC_UNLOCK(log, flags); } } @@ -1573,8 +1581,8 @@ static int dataLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, * the last entry, so don't bother logging this */ mp->lid = 0; - hold_metapage(mp, 0); - atomic_dec(&mp->nohomeok); + grab_metapage(mp); + metapage_homeok(mp); discard_metapage(mp); tlck->mp = NULL; return 0; @@ -1712,7 +1720,7 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, struct maplock *maplock; struct xdlistlock *xadlock; struct pxd_lock *pxdlock; - pxd_t *pxd; + pxd_t *page_pxd; int next, lwm, hwm; ip = tlck->ip; @@ -1722,7 +1730,7 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, lrd->log.redopage.type = cpu_to_le16(LOG_XTREE); lrd->log.redopage.l2linesize = cpu_to_le16(L2XTSLOTSIZE); - pxd = &lrd->log.redopage.pxd; + page_pxd = &lrd->log.redopage.pxd; if (tlck->type & tlckBTROOT) { lrd->log.redopage.type |= cpu_to_le16(LOG_BTROOT); @@ -1752,9 +1760,9 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, * applying the after-image to the meta-data page. */ lrd->type = cpu_to_le16(LOG_REDOPAGE); -// *pxd = mp->cm_pxd; - PXDaddress(pxd, mp->index); - PXDlength(pxd, +// *page_pxd = mp->cm_pxd; + PXDaddress(page_pxd, mp->index); + PXDlength(page_pxd, mp->logical_size >> tblk->sb->s_blocksize_bits); lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); @@ -1776,25 +1784,31 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, tlck->flag |= tlckUPDATEMAP; xadlock->flag = mlckALLOCXADLIST; xadlock->count = next - lwm; - if ((xadlock->count <= 2) && (tblk->xflag & COMMIT_LAZY)) { + if ((xadlock->count <= 4) && (tblk->xflag & COMMIT_LAZY)) { int i; + pxd_t *pxd; /* * Lazy commit may allow xtree to be modified before * txUpdateMap runs. Copy xad into linelock to * preserve correct data. + * + * We can fit twice as may pxd's as xads in the lock */ - xadlock->xdlist = &xtlck->pxdlock; - memcpy(xadlock->xdlist, &p->xad[lwm], - sizeof(xad_t) * xadlock->count); - - for (i = 0; i < xadlock->count; i++) + xadlock->flag = mlckALLOCPXDLIST; + pxd = xadlock->xdlist = &xtlck->pxdlock; + for (i = 0; i < xadlock->count; i++) { + PXDaddress(pxd, addressXAD(&p->xad[lwm + i])); + PXDlength(pxd, lengthXAD(&p->xad[lwm + i])); p->xad[lwm + i].flag &= ~(XAD_NEW | XAD_EXTENDED); + pxd++; + } } else { /* * xdlist will point to into inode's xtree, ensure * that transaction is not committed lazily. */ + xadlock->flag = mlckALLOCXADLIST; xadlock->xdlist = &p->xad[lwm]; tblk->xflag &= ~COMMIT_LAZY; } @@ -1836,8 +1850,8 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, if (tblk->xflag & COMMIT_TRUNCATE) { /* write NOREDOPAGE for the page */ lrd->type = cpu_to_le16(LOG_NOREDOPAGE); - PXDaddress(pxd, mp->index); - PXDlength(pxd, + PXDaddress(page_pxd, mp->index); + PXDlength(page_pxd, mp->logical_size >> tblk->sb-> s_blocksize_bits); lrd->backchain = @@ -1872,22 +1886,32 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, * deleted page itself; */ tlck->flag |= tlckUPDATEMAP; - xadlock->flag = mlckFREEXADLIST; xadlock->count = hwm - XTENTRYSTART + 1; - if ((xadlock->count <= 2) && (tblk->xflag & COMMIT_LAZY)) { + if ((xadlock->count <= 4) && (tblk->xflag & COMMIT_LAZY)) { + int i; + pxd_t *pxd; /* * Lazy commit may allow xtree to be modified before * txUpdateMap runs. Copy xad into linelock to * preserve correct data. + * + * We can fit twice as may pxd's as xads in the lock */ - xadlock->xdlist = &xtlck->pxdlock; - memcpy(xadlock->xdlist, &p->xad[XTENTRYSTART], - sizeof(xad_t) * xadlock->count); + xadlock->flag = mlckFREEPXDLIST; + pxd = xadlock->xdlist = &xtlck->pxdlock; + for (i = 0; i < xadlock->count; i++) { + PXDaddress(pxd, + addressXAD(&p->xad[XTENTRYSTART + i])); + PXDlength(pxd, + lengthXAD(&p->xad[XTENTRYSTART + i])); + pxd++; + } } else { /* * xdlist will point to into inode's xtree, ensure * that transaction is not committed lazily. */ + xadlock->flag = mlckFREEXADLIST; xadlock->xdlist = &p->xad[XTENTRYSTART]; tblk->xflag &= ~COMMIT_LAZY; } @@ -1918,7 +1942,7 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, * header ? */ if (tlck->type & tlckTRUNCATE) { - pxd_t tpxd; /* truncated extent of xad */ + pxd_t pxd; /* truncated extent of xad */ int twm; /* @@ -1947,8 +1971,9 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, * applying the after-image to the meta-data page. */ lrd->type = cpu_to_le16(LOG_REDOPAGE); - PXDaddress(pxd, mp->index); - PXDlength(pxd, mp->logical_size >> tblk->sb->s_blocksize_bits); + PXDaddress(page_pxd, mp->index); + PXDlength(page_pxd, + mp->logical_size >> tblk->sb->s_blocksize_bits); lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, tlck)); /* @@ -1966,7 +1991,7 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, lrd->log.updatemap.type = cpu_to_le16(LOG_FREEPXD); lrd->log.updatemap.nxd = cpu_to_le16(1); lrd->log.updatemap.pxd = pxdlock->pxd; - tpxd = pxdlock->pxd; /* save to format maplock */ + pxd = pxdlock->pxd; /* save to format maplock */ lrd->backchain = cpu_to_le32(lmLog(log, tblk, lrd, NULL)); } @@ -2035,7 +2060,7 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, pxdlock = (struct pxd_lock *) xadlock; pxdlock->flag = mlckFREEPXD; pxdlock->count = 1; - pxdlock->pxd = tpxd; + pxdlock->pxd = pxd; jfs_info("xtLog: truncate ip:0x%p mp:0x%p count:%d " "hwm:%d", ip, mp, pxdlock->count, hwm); @@ -2253,7 +2278,8 @@ void txForce(struct tblock * tblk) tlck->flag &= ~tlckWRITEPAGE; /* do not release page to freelist */ - + force_metapage(mp); +#if 0 /* * The "right" thing to do here is to * synchronously write the metadata. @@ -2265,9 +2291,10 @@ void txForce(struct tblock * tblk) * we can get by with synchronously writing * the pages when they are released. */ - assert(atomic_read(&mp->nohomeok)); + assert(mp->nohomeok); set_bit(META_dirty, &mp->flag); set_bit(META_sync, &mp->flag); +#endif } } } @@ -2327,7 +2354,7 @@ static void txUpdateMap(struct tblock * tblk) */ mp = tlck->mp; ASSERT(mp->xflag & COMMIT_PAGE); - hold_metapage(mp, 0); + grab_metapage(mp); } /* @@ -2377,8 +2404,8 @@ static void txUpdateMap(struct tblock * tblk) ASSERT(mp->lid == lid); tlck->mp->lid = 0; } - assert(atomic_read(&mp->nohomeok) == 1); - atomic_dec(&mp->nohomeok); + assert(mp->nohomeok == 1); + metapage_homeok(mp); discard_metapage(mp); tlck->mp = NULL; } @@ -2844,24 +2871,9 @@ static void LogSyncRelease(struct metapage * mp) { struct jfs_log *log = mp->log; - assert(atomic_read(&mp->nohomeok)); + assert(mp->nohomeok); assert(log); - atomic_dec(&mp->nohomeok); - - if (atomic_read(&mp->nohomeok)) - return; - - hold_metapage(mp, 0); - - LOGSYNC_LOCK(log); - mp->log = NULL; - mp->lsn = 0; - mp->clsn = 0; - log->count--; - list_del_init(&mp->synclist); - LOGSYNC_UNLOCK(log); - - release_metapage(mp); + metapage_homeok(mp); } /* diff --git a/fs/jfs/jfs_umount.c b/fs/jfs/jfs_umount.c index f31a9e3f3fe..5cf91785b54 100644 --- a/fs/jfs/jfs_umount.c +++ b/fs/jfs/jfs_umount.c @@ -49,7 +49,6 @@ */ int jfs_umount(struct super_block *sb) { - struct address_space *bdev_mapping = sb->s_bdev->bd_inode->i_mapping; struct jfs_sb_info *sbi = JFS_SBI(sb); struct inode *ipbmap = sbi->ipbmap; struct inode *ipimap = sbi->ipimap; @@ -109,8 +108,8 @@ int jfs_umount(struct super_block *sb) * Make sure all metadata makes it to disk before we mark * the superblock as clean */ - filemap_fdatawrite(bdev_mapping); - filemap_fdatawait(bdev_mapping); + filemap_fdatawrite(sbi->direct_inode->i_mapping); + filemap_fdatawait(sbi->direct_inode->i_mapping); /* * ensure all file system file pages are propagated to their @@ -123,9 +122,6 @@ int jfs_umount(struct super_block *sb) if (log) { /* log = NULL if read-only mount */ updateSuper(sb, FM_CLEAN); - /* Restore default gfp_mask for bdev */ - mapping_set_gfp_mask(bdev_mapping, GFP_USER); - /* * close log: * @@ -140,7 +136,6 @@ int jfs_umount(struct super_block *sb) int jfs_umount_rw(struct super_block *sb) { - struct address_space *bdev_mapping = sb->s_bdev->bd_inode->i_mapping; struct jfs_sb_info *sbi = JFS_SBI(sb); struct jfs_log *log = sbi->log; @@ -166,13 +161,10 @@ int jfs_umount_rw(struct super_block *sb) * mark the superblock clean before everything is flushed to * disk. */ - filemap_fdatawrite(bdev_mapping); - filemap_fdatawait(bdev_mapping); + filemap_fdatawrite(sbi->direct_inode->i_mapping); + filemap_fdatawait(sbi->direct_inode->i_mapping); updateSuper(sb, FM_CLEAN); - /* Restore default gfp_mask for bdev */ - mapping_set_gfp_mask(bdev_mapping, GFP_USER); - return lmLogClose(sb); } diff --git a/fs/jfs/jfs_xtree.c b/fs/jfs/jfs_xtree.c index 11c58c54b81..31b34db4519 100644 --- a/fs/jfs/jfs_xtree.c +++ b/fs/jfs/jfs_xtree.c @@ -1,5 +1,5 @@ /* - * Copyright (C) International Business Machines Corp., 2000-2004 + * Copyright (C) International Business Machines Corp., 2000-2005 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -111,8 +111,8 @@ static struct { /* * forward references */ -static int xtSearch(struct inode *ip, - s64 xoff, int *cmpp, struct btstack * btstack, int flag); +static int xtSearch(struct inode *ip, s64 xoff, s64 *next, int *cmpp, + struct btstack * btstack, int flag); static int xtSplitUp(tid_t tid, struct inode *ip, @@ -159,11 +159,12 @@ int xtLookup(struct inode *ip, s64 lstart, xtpage_t *p; int index; xad_t *xad; - s64 size, xoff, xend; + s64 next, size, xoff, xend; int xlen; s64 xaddr; - *plen = 0; + *paddr = 0; + *plen = llen; if (!no_check) { /* is lookup offset beyond eof ? */ @@ -180,7 +181,7 @@ int xtLookup(struct inode *ip, s64 lstart, * search for the xad entry covering the logical extent */ //search: - if ((rc = xtSearch(ip, lstart, &cmp, &btstack, 0))) { + if ((rc = xtSearch(ip, lstart, &next, &cmp, &btstack, 0))) { jfs_err("xtLookup: xtSearch returned %d", rc); return rc; } @@ -198,8 +199,11 @@ int xtLookup(struct inode *ip, s64 lstart, * lstart is a page start address, * i.e., lstart cannot start in a hole; */ - if (cmp) + if (cmp) { + if (next) + *plen = min(next - lstart, llen); goto out; + } /* * lxd covered by xad @@ -284,7 +288,7 @@ int xtLookupList(struct inode *ip, struct lxdlist * lxdlist, if (lstart >= size) return 0; - if ((rc = xtSearch(ip, lstart, &cmp, &btstack, 0))) + if ((rc = xtSearch(ip, lstart, NULL, &cmp, &btstack, 0))) return rc; /* @@ -488,6 +492,7 @@ int xtLookupList(struct inode *ip, struct lxdlist * lxdlist, * parameters: * ip - file object; * xoff - extent offset; + * nextp - address of next extent (if any) for search miss * cmpp - comparison result: * btstack - traverse stack; * flag - search process flag (XT_INSERT); @@ -497,7 +502,7 @@ int xtLookupList(struct inode *ip, struct lxdlist * lxdlist, * *cmpp is set to result of comparison with the entry returned. * the page containing the entry is pinned at exit. */ -static int xtSearch(struct inode *ip, s64 xoff, /* offset of extent */ +static int xtSearch(struct inode *ip, s64 xoff, s64 *nextp, int *cmpp, struct btstack * btstack, int flag) { struct jfs_inode_info *jfs_ip = JFS_IP(ip); @@ -511,6 +516,7 @@ static int xtSearch(struct inode *ip, s64 xoff, /* offset of extent */ struct btframe *btsp; int nsplit = 0; /* number of pages to split */ s64 t64; + s64 next = 0; INCREMENT(xtStat.search); @@ -579,6 +585,7 @@ static int xtSearch(struct inode *ip, s64 xoff, /* offset of extent */ * previous and this entry */ *cmpp = 1; + next = t64; goto out; } @@ -623,6 +630,9 @@ static int xtSearch(struct inode *ip, s64 xoff, /* offset of extent */ /* update sequential access heuristics */ jfs_ip->btindex = index; + if (nextp) + *nextp = next; + INCREMENT(xtStat.fastSearch); return 0; } @@ -675,10 +685,11 @@ static int xtSearch(struct inode *ip, s64 xoff, /* offset of extent */ return 0; } - /* search hit - internal page: * descend/search its child page */ + if (index < le16_to_cpu(p->header.nextindex)-1) + next = offsetXAD(&p->xad[index + 1]); goto next; } @@ -694,6 +705,8 @@ static int xtSearch(struct inode *ip, s64 xoff, /* offset of extent */ * base is the smallest index with key (Kj) greater than * search key (K) and may be zero or maxentry index. */ + if (base < le16_to_cpu(p->header.nextindex)) + next = offsetXAD(&p->xad[base]); /* * search miss - leaf page: * @@ -727,6 +740,9 @@ static int xtSearch(struct inode *ip, s64 xoff, /* offset of extent */ jfs_ip->btorder = BT_RANDOM; jfs_ip->btindex = base; + if (nextp) + *nextp = next; + return 0; } @@ -793,6 +809,7 @@ int xtInsert(tid_t tid, /* transaction id */ struct xtsplit split; /* split information */ xad_t *xad; int cmp; + s64 next; struct tlock *tlck; struct xtlock *xtlck; @@ -806,7 +823,7 @@ int xtInsert(tid_t tid, /* transaction id */ * n.b. xtSearch() may return index of maxentry of * the full page. */ - if ((rc = xtSearch(ip, xoff, &cmp, &btstack, XT_INSERT))) + if ((rc = xtSearch(ip, xoff, &next, &cmp, &btstack, XT_INSERT))) return rc; /* retrieve search result */ @@ -814,7 +831,7 @@ int xtInsert(tid_t tid, /* transaction id */ /* This test must follow XT_GETSEARCH since mp must be valid if * we branch to out: */ - if (cmp == 0) { + if ((cmp == 0) || (next && (xlen > next - xoff))) { rc = -EEXIST; goto out; } @@ -1626,7 +1643,7 @@ int xtExtend(tid_t tid, /* transaction id */ jfs_info("xtExtend: nxoff:0x%lx nxlen:0x%x", (ulong) xoff, xlen); /* there must exist extent to be extended */ - if ((rc = xtSearch(ip, xoff - 1, &cmp, &btstack, XT_INSERT))) + if ((rc = xtSearch(ip, xoff - 1, NULL, &cmp, &btstack, XT_INSERT))) return rc; /* retrieve search result */ @@ -1794,7 +1811,7 @@ printf("xtTailgate: nxoff:0x%lx nxlen:0x%x nxaddr:0x%lx\n", */ /* there must exist extent to be tailgated */ - if ((rc = xtSearch(ip, xoff, &cmp, &btstack, XT_INSERT))) + if ((rc = xtSearch(ip, xoff, NULL, &cmp, &btstack, XT_INSERT))) return rc; /* retrieve search result */ @@ -1977,7 +1994,7 @@ int xtUpdate(tid_t tid, struct inode *ip, xad_t * nxad) nxlen = lengthXAD(nxad); nxaddr = addressXAD(nxad); - if ((rc = xtSearch(ip, nxoff, &cmp, &btstack, XT_INSERT))) + if ((rc = xtSearch(ip, nxoff, NULL, &cmp, &btstack, XT_INSERT))) return rc; /* retrieve search result */ @@ -2291,7 +2308,7 @@ int xtUpdate(tid_t tid, struct inode *ip, xad_t * nxad) if (nextindex == le16_to_cpu(p->header.maxentry)) { XT_PUTPAGE(mp); - if ((rc = xtSearch(ip, nxoff, &cmp, &btstack, XT_INSERT))) + if ((rc = xtSearch(ip, nxoff, NULL, &cmp, &btstack, XT_INSERT))) return rc; /* retrieve search result */ @@ -2438,6 +2455,7 @@ int xtAppend(tid_t tid, /* transaction id */ int nsplit, nblocks, xlen; struct pxdlist pxdlist; pxd_t *pxd; + s64 next; xaddr = *xaddrp; xlen = *xlenp; @@ -2452,7 +2470,7 @@ int xtAppend(tid_t tid, /* transaction id */ * n.b. xtSearch() may return index of maxentry of * the full page. */ - if ((rc = xtSearch(ip, xoff, &cmp, &btstack, XT_INSERT))) + if ((rc = xtSearch(ip, xoff, &next, &cmp, &btstack, XT_INSERT))) return rc; /* retrieve search result */ @@ -2462,6 +2480,9 @@ int xtAppend(tid_t tid, /* transaction id */ rc = -EEXIST; goto out; } + + if (next) + xlen = min(xlen, (int)(next - xoff)); //insert: /* * insert entry for new extent @@ -2600,7 +2621,7 @@ int xtDelete(tid_t tid, struct inode *ip, s64 xoff, s32 xlen, int flag) /* * find the matching entry; xtSearch() pins the page */ - if ((rc = xtSearch(ip, xoff, &cmp, &btstack, 0))) + if ((rc = xtSearch(ip, xoff, NULL, &cmp, &btstack, 0))) return rc; XT_GETSEARCH(ip, btstack.top, bn, mp, p, index); @@ -2852,7 +2873,7 @@ xtRelocate(tid_t tid, struct inode * ip, xad_t * oxad, /* old XAD */ */ if (xtype == DATAEXT) { /* search in leaf entry */ - rc = xtSearch(ip, xoff, &cmp, &btstack, 0); + rc = xtSearch(ip, xoff, NULL, &cmp, &btstack, 0); if (rc) return rc; @@ -2958,7 +2979,7 @@ xtRelocate(tid_t tid, struct inode * ip, xad_t * oxad, /* old XAD */ } /* get back parent page */ - if ((rc = xtSearch(ip, xoff, &cmp, &btstack, 0))) + if ((rc = xtSearch(ip, xoff, NULL, &cmp, &btstack, 0))) return rc; XT_GETSEARCH(ip, btstack.top, bn, pmp, pp, index); @@ -3991,7 +4012,7 @@ s64 xtTruncate_pmap(tid_t tid, struct inode *ip, s64 committed_size) if (committed_size) { xoff = (committed_size >> JFS_SBI(ip->i_sb)->l2bsize) - 1; - rc = xtSearch(ip, xoff, &cmp, &btstack, 0); + rc = xtSearch(ip, xoff, NULL, &cmp, &btstack, 0); if (rc) return rc; diff --git a/fs/jfs/resize.c b/fs/jfs/resize.c index 2eb6869b6e7..c6dc254d325 100644 --- a/fs/jfs/resize.c +++ b/fs/jfs/resize.c @@ -209,6 +209,9 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize) */ txQuiesce(sb); + /* Reset size of direct inode */ + sbi->direct_inode->i_size = sb->s_bdev->bd_inode->i_size; + if (sbi->mntflag & JFS_INLINELOG) { /* * deactivate old inline log diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 5856866e24f..5e774ed7fb6 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -210,6 +210,10 @@ static void jfs_put_super(struct super_block *sb) unload_nls(sbi->nls_tab); sbi->nls_tab = NULL; + truncate_inode_pages(sbi->direct_inode->i_mapping, 0); + iput(sbi->direct_inode); + sbi->direct_inode = NULL; + kfree(sbi); } @@ -358,6 +362,12 @@ static int jfs_remount(struct super_block *sb, int *flags, char *data) } if ((sb->s_flags & MS_RDONLY) && !(*flags & MS_RDONLY)) { + /* + * Invalidate any previously read metadata. fsck may have + * changed the on-disk data since we mounted r/o + */ + truncate_inode_pages(JFS_SBI(sb)->direct_inode->i_mapping, 0); + JFS_SBI(sb)->flag = flag; return jfs_mount_rw(sb, 1); } @@ -428,12 +438,26 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) sb->s_op = &jfs_super_operations; sb->s_export_op = &jfs_export_operations; + /* + * Initialize direct-mapping inode/address-space + */ + inode = new_inode(sb); + if (inode == NULL) + goto out_kfree; + inode->i_ino = 0; + inode->i_nlink = 1; + inode->i_size = sb->s_bdev->bd_inode->i_size; + inode->i_mapping->a_ops = &jfs_metapage_aops; + mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS); + + sbi->direct_inode = inode; + rc = jfs_mount(sb); if (rc) { if (!silent) { jfs_err("jfs_mount failed w/return code = %d", rc); } - goto out_kfree; + goto out_mount_failed; } if (sb->s_flags & MS_RDONLY) sbi->log = NULL; @@ -482,6 +506,13 @@ out_no_rw: if (rc) { jfs_err("jfs_umount failed with return code %d", rc); } +out_mount_failed: + filemap_fdatawrite(sbi->direct_inode->i_mapping); + filemap_fdatawait(sbi->direct_inode->i_mapping); + truncate_inode_pages(sbi->direct_inode->i_mapping, 0); + make_bad_inode(sbi->direct_inode); + iput(sbi->direct_inode); + sbi->direct_inode = NULL; out_kfree: if (sbi->nls_tab) unload_nls(sbi->nls_tab); @@ -527,8 +558,10 @@ static int jfs_sync_fs(struct super_block *sb, int wait) struct jfs_log *log = JFS_SBI(sb)->log; /* log == NULL indicates read-only mount */ - if (log) + if (log) { jfs_flush_journal(log, wait); + jfs_syncpt(log); + } return 0; } diff --git a/fs/locks.c b/fs/locks.c index 1792ce547af..3fa6a7ce57a 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -406,12 +406,12 @@ static void lease_release_private_callback(struct file_lock *fl) fl->fl_file->f_owner.signum = 0; } -int lease_mylease_callback(struct file_lock *fl, struct file_lock *try) +static int lease_mylease_callback(struct file_lock *fl, struct file_lock *try) { return fl->fl_file == try->fl_file; } -struct lock_manager_operations lease_manager_ops = { +static struct lock_manager_operations lease_manager_ops = { .fl_break = lease_break_callback, .fl_release_private = lease_release_private_callback, .fl_mylease = lease_mylease_callback, @@ -1274,7 +1274,7 @@ int fcntl_getlease(struct file *filp) * * Called with kernel lock held. */ -int __setlease(struct file *filp, long arg, struct file_lock **flp) +static int __setlease(struct file *filp, long arg, struct file_lock **flp) { struct file_lock *fl, **before, **my_before = NULL, *lease = *flp; struct dentry *dentry = filp->f_dentry; diff --git a/fs/mbcache.c b/fs/mbcache.c index f9e4d2700cd..c7170b9221a 100644 --- a/fs/mbcache.c +++ b/fs/mbcache.c @@ -57,7 +57,7 @@ #define MB_CACHE_WRITER ((unsigned short)~0U >> 1) -DECLARE_WAIT_QUEUE_HEAD(mb_cache_queue); +static DECLARE_WAIT_QUEUE_HEAD(mb_cache_queue); MODULE_AUTHOR("Andreas Gruenbacher <a.gruenbacher@computer.org>"); MODULE_DESCRIPTION("Meta block cache (for extended attributes)"); diff --git a/fs/mpage.c b/fs/mpage.c index e7d8d1a7760..b92c0e64aef 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -87,7 +87,7 @@ static int mpage_end_io_write(struct bio *bio, unsigned int bytes_done, int err) return 0; } -struct bio *mpage_bio_submit(int rw, struct bio *bio) +static struct bio *mpage_bio_submit(int rw, struct bio *bio) { bio->bi_end_io = mpage_end_io_read; if (rw == WRITE) @@ -160,52 +160,6 @@ map_buffer_to_page(struct page *page, struct buffer_head *bh, int page_block) } while (page_bh != head); } -/** - * mpage_readpages - populate an address space with some pages, and - * start reads against them. - * - * @mapping: the address_space - * @pages: The address of a list_head which contains the target pages. These - * pages have their ->index populated and are otherwise uninitialised. - * - * The page at @pages->prev has the lowest file offset, and reads should be - * issued in @pages->prev to @pages->next order. - * - * @nr_pages: The number of pages at *@pages - * @get_block: The filesystem's block mapper function. - * - * This function walks the pages and the blocks within each page, building and - * emitting large BIOs. - * - * If anything unusual happens, such as: - * - * - encountering a page which has buffers - * - encountering a page which has a non-hole after a hole - * - encountering a page with non-contiguous blocks - * - * then this code just gives up and calls the buffer_head-based read function. - * It does handle a page which has holes at the end - that is a common case: - * the end-of-file on blocksize < PAGE_CACHE_SIZE setups. - * - * BH_Boundary explanation: - * - * There is a problem. The mpage read code assembles several pages, gets all - * their disk mappings, and then submits them all. That's fine, but obtaining - * the disk mappings may require I/O. Reads of indirect blocks, for example. - * - * So an mpage read of the first 16 blocks of an ext2 file will cause I/O to be - * submitted in the following order: - * 12 0 1 2 3 4 5 6 7 8 9 10 11 13 14 15 16 - * because the indirect block has to be read to get the mappings of blocks - * 13,14,15,16. Obviously, this impacts performance. - * - * So what we do it to allow the filesystem's get_block() function to set - * BH_Boundary when it maps block 11. BH_Boundary says: mapping of the block - * after this one will require I/O against a block which is probably close to - * this one. So you should push what I/O you have currently accumulated. - * - * This all causes the disk requests to be issued in the correct order. - */ static struct bio * do_mpage_readpage(struct bio *bio, struct page *page, unsigned nr_pages, sector_t *last_block_in_bio, get_block_t get_block) @@ -320,6 +274,52 @@ confused: goto out; } +/** + * mpage_readpages - populate an address space with some pages, and + * start reads against them. + * + * @mapping: the address_space + * @pages: The address of a list_head which contains the target pages. These + * pages have their ->index populated and are otherwise uninitialised. + * + * The page at @pages->prev has the lowest file offset, and reads should be + * issued in @pages->prev to @pages->next order. + * + * @nr_pages: The number of pages at *@pages + * @get_block: The filesystem's block mapper function. + * + * This function walks the pages and the blocks within each page, building and + * emitting large BIOs. + * + * If anything unusual happens, such as: + * + * - encountering a page which has buffers + * - encountering a page which has a non-hole after a hole + * - encountering a page with non-contiguous blocks + * + * then this code just gives up and calls the buffer_head-based read function. + * It does handle a page which has holes at the end - that is a common case: + * the end-of-file on blocksize < PAGE_CACHE_SIZE setups. + * + * BH_Boundary explanation: + * + * There is a problem. The mpage read code assembles several pages, gets all + * their disk mappings, and then submits them all. That's fine, but obtaining + * the disk mappings may require I/O. Reads of indirect blocks, for example. + * + * So an mpage read of the first 16 blocks of an ext2 file will cause I/O to be + * submitted in the following order: + * 12 0 1 2 3 4 5 6 7 8 9 10 11 13 14 15 16 + * because the indirect block has to be read to get the mappings of blocks + * 13,14,15,16. Obviously, this impacts performance. + * + * So what we do it to allow the filesystem's get_block() function to set + * BH_Boundary when it maps block 11. BH_Boundary says: mapping of the block + * after this one will require I/O against a block which is probably close to + * this one. So you should push what I/O you have currently accumulated. + * + * This all causes the disk requests to be issued in the correct order. + */ int mpage_readpages(struct address_space *mapping, struct list_head *pages, unsigned nr_pages, get_block_t get_block) @@ -627,15 +627,6 @@ int mpage_writepages(struct address_space *mapping, struct writeback_control *wbc, get_block_t get_block) { - return __mpage_writepages(mapping, wbc, get_block, - mapping->a_ops->writepage); -} - -int -__mpage_writepages(struct address_space *mapping, - struct writeback_control *wbc, get_block_t get_block, - writepage_t writepage_fn) -{ struct backing_dev_info *bdi = mapping->backing_dev_info; struct bio *bio = NULL; sector_t last_block_in_bio = 0; @@ -725,8 +716,10 @@ retry: } else { bio = __mpage_writepage(bio, page, get_block, &last_block_in_bio, &ret, wbc, - writepage_fn); + page->mapping->a_ops->writepage); } + if (unlikely(ret == WRITEPAGE_ACTIVATE)) + unlock_page(page); if (ret || (--(wbc->nr_to_write) <= 0)) done = 1; if (wbc->nonblocking && bdi_write_congested(bdi)) { @@ -753,7 +746,6 @@ retry: return ret; } EXPORT_SYMBOL(mpage_writepages); -EXPORT_SYMBOL(__mpage_writepages); int mpage_writepage(struct page *page, get_block_t get_block, struct writeback_control *wbc) diff --git a/fs/namei.c b/fs/namei.c index 9e4aef2a1a2..defe6781e00 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -686,11 +686,11 @@ fail: /* * Name resolution. + * This is the basic name resolution function, turning a pathname into + * the final dentry. We expect 'base' to be positive and a directory. * - * This is the basic name resolution function, turning a pathname - * into the final dentry. - * - * We expect 'base' to be positive and a directory. + * Returns 0 and nd will have valid dentry and mnt on success. + * Returns error and drops reference to input namei data on failure. */ static fastcall int __link_path_walk(const char * name, struct nameidata *nd) { @@ -929,8 +929,10 @@ int fastcall path_walk(const char * name, struct nameidata *nd) return link_path_walk(name, nd); } -/* SMP-safe */ -/* returns 1 if everything is done */ +/* + * SMP-safe: Returns 1 and nd will have valid dentry and mnt, if + * everything is done. Returns 0 and drops input nd, if lookup failed; + */ static int __emul_lookup_dentry(const char *name, struct nameidata *nd) { if (path_walk(name, nd)) @@ -994,9 +996,10 @@ set_it: } } +/* Returns 0 and nd will be valid on success; Retuns error, otherwise. */ int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata *nd) { - int retval; + int retval = 0; nd->last_type = LAST_ROOT; /* if there are only slashes... */ nd->flags = flags; @@ -1009,7 +1012,7 @@ int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata nd->dentry = dget(current->fs->altroot); read_unlock(¤t->fs->lock); if (__emul_lookup_dentry(name,nd)) - return 0; + goto out; /* found in altroot */ read_lock(¤t->fs->lock); } nd->mnt = mntget(current->fs->rootmnt); @@ -1021,6 +1024,7 @@ int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata read_unlock(¤t->fs->lock); current->total_link_count = 0; retval = link_path_walk(name, nd); +out: if (unlikely(current->audit_context && nd && nd->dentry && nd->dentry->d_inode)) audit_inode(name, nd->dentry->d_inode); @@ -2067,8 +2071,8 @@ exit: * ->i_sem on parents, which works but leads to some truely excessive * locking]. */ -int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry) +static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) { int error = 0; struct inode *target; @@ -2112,8 +2116,8 @@ int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, return error; } -int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry) +static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) { struct inode *target; int error; diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index b74c4e3a64e..87f4f9aeac8 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -79,7 +79,7 @@ static ssize_t idmap_pipe_upcall(struct file *, struct rpc_pipe_msg *, char __user *, size_t); static ssize_t idmap_pipe_downcall(struct file *, const char __user *, size_t); -void idmap_pipe_destroy_msg(struct rpc_pipe_msg *); +static void idmap_pipe_destroy_msg(struct rpc_pipe_msg *); static unsigned int fnvhash32(const void *, size_t); @@ -434,7 +434,7 @@ out: return ret; } -void +static void idmap_pipe_destroy_msg(struct rpc_pipe_msg *msg) { struct idmap_msg *im = msg->data; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 6345f26e87e..f2317f3e29f 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1904,7 +1904,7 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) } } -int nfs_init_inodecache(void) +static int nfs_init_inodecache(void) { nfs_inode_cachep = kmem_cache_create("nfs_inode_cache", sizeof(struct nfs_inode), @@ -1916,7 +1916,7 @@ int nfs_init_inodecache(void) return 0; } -void nfs_destroy_inodecache(void) +static void nfs_destroy_inodecache(void) { if (kmem_cache_destroy(nfs_inode_cachep)) printk(KERN_INFO "nfs_inode_cache: not all structures were freed\n"); diff --git a/fs/nls/nls_base.c b/fs/nls/nls_base.c index 897512796ed..a912debcd20 100644 --- a/fs/nls/nls_base.c +++ b/fs/nls/nls_base.c @@ -243,7 +243,7 @@ void unload_nls(struct nls_table *nls) module_put(nls->owner); } -wchar_t charset2uni[256] = { +static wchar_t charset2uni[256] = { /* 0x00*/ 0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007, diff --git a/fs/proc/base.c b/fs/proc/base.c index 39fd336cfdb..e31903aadd9 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -820,7 +820,7 @@ static ssize_t proc_loginuid_write(struct file * file, const char __user * buf, goto out_free_page; } - length = audit_set_loginuid(task->audit_context, loginuid); + length = audit_set_loginuid(task, loginuid); if (likely(length == 0)) length = count; @@ -1419,6 +1419,8 @@ static struct file_operations proc_tgid_attr_operations; static struct inode_operations proc_tgid_attr_inode_operations; #endif +static int get_tid_list(int index, unsigned int *tids, struct inode *dir); + /* SMP-safe */ static struct dentry *proc_pident_lookup(struct inode *dir, struct dentry *dentry, @@ -1458,7 +1460,7 @@ static struct dentry *proc_pident_lookup(struct inode *dir, */ switch(p->type) { case PROC_TGID_TASK: - inode->i_nlink = 3; + inode->i_nlink = 2 + get_tid_list(2, NULL, dir); inode->i_op = &proc_task_inode_operations; inode->i_fop = &proc_task_operations; break; @@ -1701,13 +1703,13 @@ static struct inode_operations proc_self_inode_operations = { }; /** - * proc_pid_unhash - Unhash /proc/<pid> entry from the dcache. + * proc_pid_unhash - Unhash /proc/@pid entry from the dcache. * @p: task that should be flushed. * - * Drops the /proc/<pid> dcache entry from the hash chains. + * Drops the /proc/@pid dcache entry from the hash chains. * - * Dropping /proc/<pid> entries and detach_pid must be synchroneous, - * otherwise e.g. /proc/<pid>/exe might point to the wrong executable, + * Dropping /proc/@pid entries and detach_pid must be synchroneous, + * otherwise e.g. /proc/@pid/exe might point to the wrong executable, * if the pid value is immediately reused. This is enforced by * - caller must acquire spin_lock(p->proc_lock) * - must be called before detach_pid() @@ -1739,8 +1741,8 @@ struct dentry *proc_pid_unhash(struct task_struct *p) } /** - * proc_pid_flush - recover memory used by stale /proc/<pid>/x entries - * @proc_entry: directoy to prune. + * proc_pid_flush - recover memory used by stale /proc/@pid/x entries + * @proc_dentry: directoy to prune. * * Shrink the /proc directory that was used by the just killed thread. */ @@ -1800,8 +1802,12 @@ struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, struct inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO; inode->i_op = &proc_tgid_base_inode_operations; inode->i_fop = &proc_tgid_base_operations; - inode->i_nlink = 3; inode->i_flags|=S_IMMUTABLE; +#ifdef CONFIG_SECURITY + inode->i_nlink = 5; +#else + inode->i_nlink = 4; +#endif dentry->d_op = &pid_base_dentry_operations; @@ -1855,8 +1861,12 @@ static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO; inode->i_op = &proc_tid_base_inode_operations; inode->i_fop = &proc_tid_base_operations; - inode->i_nlink = 3; inode->i_flags|=S_IMMUTABLE; +#ifdef CONFIG_SECURITY + inode->i_nlink = 4; +#else + inode->i_nlink = 3; +#endif dentry->d_op = &pid_base_dentry_operations; @@ -1935,7 +1945,8 @@ static int get_tid_list(int index, unsigned int *tids, struct inode *dir) if (--index >= 0) continue; - tids[nr_tids] = tid; + if (tids != NULL) + tids[nr_tids] = tid; nr_tids++; if (nr_tids >= PROC_MAXPIDS) break; @@ -2035,6 +2046,7 @@ static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldi } nr_tids = get_tid_list(pos, tid_array, inode); + inode->i_nlink = pos + nr_tids; for (i = 0; i < nr_tids; i++) { unsigned long j = PROC_NUMBUF; diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c index a4e2ed544bb..49c479c9454 100644 --- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -260,8 +260,9 @@ static inline int block_group_used(struct super_block *s, u32 id) { /* * the packing is returned in disk byte order */ -u32 reiserfs_choose_packing(struct inode *dir) { - u32 packing; +__le32 reiserfs_choose_packing(struct inode *dir) +{ + __le32 packing; if (TEST_OPTION(packing_groups, dir->i_sb)) { u32 parent_dir = le32_to_cpu(INODE_PKEY(dir)->k_dir_id); /* @@ -655,7 +656,7 @@ static int get_left_neighbor(reiserfs_blocknr_hint_t *hint) struct buffer_head * bh; struct item_head * ih; int pos_in_item; - __u32 * item; + __le32 * item; int ret = 0; if (!hint->path) /* reiserfs code can call this function w/o pointer to path @@ -736,7 +737,7 @@ static inline int this_blocknr_allocation_would_make_it_a_large_file(reiserfs_bl #ifdef DISPLACE_NEW_PACKING_LOCALITIES static inline void displace_new_packing_locality (reiserfs_blocknr_hint_t *hint) { - struct reiserfs_key * key = &hint->key; + struct in_core_key * key = &hint->key; hint->th->displace_new_blocks = 0; hint->search_start = hint->beg + keyed_hash((char*)(&key->k_objectid),4) % (hint->end - hint->beg); @@ -777,7 +778,7 @@ static inline int old_way (reiserfs_blocknr_hint_t * hint) static inline void hundredth_slices (reiserfs_blocknr_hint_t * hint) { - struct reiserfs_key * key = &hint->key; + struct in_core_key * key = &hint->key; b_blocknr_t slice_start; slice_start = (keyed_hash((char*)(&key->k_dir_id),4) % 100) * (hint->end / 100); diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c index d1514a9b051..fbde4b01a32 100644 --- a/fs/reiserfs/dir.c +++ b/fs/reiserfs/dir.c @@ -209,8 +209,8 @@ static int reiserfs_readdir (struct file * filp, void * dirent, filldir_t filldi /* compose directory item containing "." and ".." entries (entries are not aligned to 4 byte boundary) */ /* the last four params are LE */ -void make_empty_dir_item_v1 (char * body, __u32 dirid, __u32 objid, - __u32 par_dirid, __u32 par_objid) +void make_empty_dir_item_v1 (char * body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid) { struct reiserfs_de_head * deh; @@ -242,8 +242,8 @@ void make_empty_dir_item_v1 (char * body, __u32 dirid, __u32 objid, } /* compose directory item containing "." and ".." entries */ -void make_empty_dir_item (char * body, __u32 dirid, __u32 objid, - __u32 par_dirid, __u32 par_objid) +void make_empty_dir_item (char * body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid) { struct reiserfs_de_head * deh; diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 26950113af8..2230afff187 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -166,7 +166,7 @@ static int reiserfs_allocate_blocks_for_region( struct cpu_key key; // cpu key of item that we are going to deal with struct item_head *ih; // pointer to item head that we are going to deal with struct buffer_head *bh; // Buffer head that contains items that we are going to deal with - __u32 * item; // pointer to item we are going to deal with + __le32 * item; // pointer to item we are going to deal with INITIALIZE_PATH(path); // path to item, that we are going to deal with. b_blocknr_t *allocated_blocks; // Pointer to a place where allocated blocknumbers would be stored. reiserfs_blocknr_hint_t hint; // hint structure for block allocator. @@ -891,7 +891,7 @@ static int reiserfs_prepare_file_region_for_write( struct item_head *ih = NULL; // pointer to item head that we are going to deal with struct buffer_head *itembuf=NULL; // Buffer head that contains items that we are going to deal with INITIALIZE_PATH(path); // path to item, that we are going to deal with. - __u32 * item=NULL; // pointer to item we are going to deal with + __le32 * item=NULL; // pointer to item we are going to deal with int item_pos=-1; /* Position in indirect item */ @@ -1284,10 +1284,11 @@ static ssize_t reiserfs_file_write( struct file *file, /* the file we are going reiserfs_claim_blocks_to_be_allocated(inode->i_sb, num_pages << (PAGE_CACHE_SHIFT - inode->i_blkbits)); reiserfs_write_unlock(inode->i_sb); - if ( !num_pages ) { /* If we do not have enough space even for */ - res = -ENOSPC; /* single page, return -ENOSPC */ - if ( pos > (inode->i_size & (inode->i_sb->s_blocksize-1))) - break; // In case we are writing past the file end, break. + if ( !num_pages ) { /* If we do not have enough space even for a single page... */ + if ( pos > inode->i_size+inode->i_sb->s_blocksize-(pos & (inode->i_sb->s_blocksize-1))) { + res = -ENOSPC; + break; // In case we are writing past the end of the last file block, break. + } // Otherwise we are possibly overwriting the file, so // let's set write size to be equal or less than blocksize. // This way we get it correctly for file holes. diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 7543031396f..2711dff1b7b 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -173,7 +173,7 @@ static inline void fix_tail_page_for_writing(struct page *page) { done already or non-hole position has been found in the indirect item */ static inline int allocation_needed (int retval, b_blocknr_t allocated, struct item_head * ih, - __u32 * item, int pos_in_item) + __le32 * item, int pos_in_item) { if (allocated) return 0; @@ -278,7 +278,7 @@ research: bh = get_last_bh (&path); ih = get_ih (&path); if (is_indirect_le_ih (ih)) { - __u32 * ind_item = (__u32 *)B_I_PITEM (bh, ih); + __le32 * ind_item = (__le32 *)B_I_PITEM (bh, ih); /* FIXME: here we could cache indirect item or part of it in the inode to avoid search_by_key in case of subsequent @@ -581,7 +581,7 @@ int reiserfs_get_block (struct inode * inode, sector_t block, struct cpu_key key; struct buffer_head * bh, * unbh = NULL; struct item_head * ih, tmp_ih; - __u32 * item; + __le32 * item; int done; int fs_gen; struct reiserfs_transaction_handle *th = NULL; @@ -746,7 +746,7 @@ start_trans: done = 0; do { if (is_statdata_le_ih (ih)) { - __u32 unp = 0; + __le32 unp = 0; struct cpu_key tmp_key; /* indirect item has to be inserted */ @@ -1341,8 +1341,8 @@ void reiserfs_read_locked_inode (struct inode * inode, struct reiserfs_iget_args key.version = KEY_FORMAT_3_5; key.on_disk_key.k_dir_id = dirino; key.on_disk_key.k_objectid = inode->i_ino; - key.on_disk_key.u.k_offset_v1.k_offset = SD_OFFSET; - key.on_disk_key.u.k_offset_v1.k_uniqueness = SD_UNIQUENESS; + key.on_disk_key.k_offset = 0; + key.on_disk_key.k_type = 0; /* look for the object's stat data */ retval = search_item (inode->i_sb, &key, &path_to_sd); @@ -2067,7 +2067,7 @@ static int map_block_for_writepage(struct inode *inode, struct item_head tmp_ih ; struct item_head *ih ; struct buffer_head *bh ; - __u32 *item ; + __le32 *item ; struct cpu_key key ; INITIALIZE_PATH(path) ; int pos_in_item ; diff --git a/fs/reiserfs/item_ops.c b/fs/reiserfs/item_ops.c index 9cf7c13b120..0ce33db1acd 100644 --- a/fs/reiserfs/item_ops.c +++ b/fs/reiserfs/item_ops.c @@ -296,10 +296,11 @@ static void print_sequence (__u32 start, int len) static void indirect_print_item (struct item_head * ih, char * item) { int j; - __u32 * unp, prev = INT_MAX; + __le32 * unp; + __u32 prev = INT_MAX; int num; - unp = (__u32 *)item; + unp = (__le32 *)item; if (ih_item_len(ih) % UNFM_P_SIZE) reiserfs_warning (NULL, "indirect_print_item: invalid item len"); diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index c9ad3a7849f..3072cfdee95 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2306,13 +2306,16 @@ static int journal_init_dev( struct super_block *super, if( !IS_ERR( journal -> j_dev_file ) ) { struct inode *jdev_inode = journal->j_dev_file->f_mapping->host; if( !S_ISBLK( jdev_inode -> i_mode ) ) { - reiserfs_warning (super, "journal_init_dev: '%s' is " - "not a block device", jdev_name ); + reiserfs_warning(super, "journal_init_dev: '%s' is " + "not a block device", jdev_name ); result = -ENOTBLK; + release_journal_dev( super, journal ); } else { /* ok */ journal->j_dev_bd = I_BDEV(jdev_inode); set_blocksize(journal->j_dev_bd, super->s_blocksize); + reiserfs_info(super, "journal_init_dev: journal device: %s\n", + bdevname(journal->j_dev_bd, b)); } } else { result = PTR_ERR( journal -> j_dev_file ); @@ -2321,11 +2324,6 @@ static int journal_init_dev( struct super_block *super, "journal_init_dev: Cannot open '%s': %i", jdev_name, result ); } - if( result != 0 ) { - release_journal_dev( super, journal ); - } - reiserfs_info(super, "journal_init_dev: journal device: %s\n", - bdevname(journal->j_dev_bd, b)); return result; } @@ -2393,7 +2391,7 @@ int journal_init(struct super_block *p_s_sb, const char * j_dev_name, int old_fo jh = (struct reiserfs_journal_header *)(bhjh->b_data); /* make sure that journal matches to the super block */ - if (is_reiserfs_jr(rs) && (jh->jh_journal.jp_journal_magic != sb_jp_journal_magic(rs))) { + if (is_reiserfs_jr(rs) && (le32_to_cpu(jh->jh_journal.jp_journal_magic) != sb_jp_journal_magic(rs))) { reiserfs_warning (p_s_sb, "sh-460: journal header magic %x " "(device %s) does not match to magic found in super " "block %x", diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index 80e92d9b81c..7d4dc5f5aa8 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -608,7 +608,7 @@ static int reiserfs_create (struct inode * dir, struct dentry *dentry, int mode, goto out_failed; } - retval = reiserfs_new_inode (&th, dir, mode, 0, 0/*i_size*/, dentry, inode); + retval = reiserfs_new_inode (&th, dir, mode, NULL, 0/*i_size*/, dentry, inode); if (retval) goto out_failed; diff --git a/fs/reiserfs/objectid.c b/fs/reiserfs/objectid.c index 0785c43a748..bfe8e25ef29 100644 --- a/fs/reiserfs/objectid.c +++ b/fs/reiserfs/objectid.c @@ -11,13 +11,13 @@ // find where objectid map starts #define objectid_map(s,rs) (old_format_only (s) ? \ - (__u32 *)((struct reiserfs_super_block_v1 *)(rs) + 1) :\ - (__u32 *)((rs) + 1)) + (__le32 *)((struct reiserfs_super_block_v1 *)(rs) + 1) :\ + (__le32 *)((rs) + 1)) #ifdef CONFIG_REISERFS_CHECK -static void check_objectid_map (struct super_block * s, __u32 * map) +static void check_objectid_map (struct super_block * s, __le32 * map) { if (le32_to_cpu (map[0]) != 1) reiserfs_panic (s, "vs-15010: check_objectid_map: map corrupted: %lx", @@ -27,7 +27,7 @@ static void check_objectid_map (struct super_block * s, __u32 * map) } #else -static void check_objectid_map (struct super_block * s, __u32 * map) +static void check_objectid_map (struct super_block * s, __le32 * map) {;} #endif @@ -52,7 +52,7 @@ __u32 reiserfs_get_unused_objectid (struct reiserfs_transaction_handle *th) { struct super_block * s = th->t_super; struct reiserfs_super_block * rs = SB_DISK_SUPER_BLOCK (s); - __u32 * map = objectid_map (s, rs); + __le32 * map = objectid_map (s, rs); __u32 unused_objectid; BUG_ON (!th->t_trans_id); @@ -97,7 +97,7 @@ void reiserfs_release_objectid (struct reiserfs_transaction_handle *th, { struct super_block * s = th->t_super; struct reiserfs_super_block * rs = SB_DISK_SUPER_BLOCK (s); - __u32 * map = objectid_map (s, rs); + __le32 * map = objectid_map (s, rs); int i = 0; BUG_ON (!th->t_trans_id); @@ -172,12 +172,12 @@ int reiserfs_convert_objectid_map_v1(struct super_block *s) { int new_size = (s->s_blocksize - SB_SIZE) / sizeof(__u32) / 2 * 2 ; int old_max = sb_oid_maxsize(disk_sb); struct reiserfs_super_block_v1 *disk_sb_v1 ; - __u32 *objectid_map, *new_objectid_map ; + __le32 *objectid_map, *new_objectid_map ; int i ; disk_sb_v1=(struct reiserfs_super_block_v1 *)(SB_BUFFER_WITH_SB(s)->b_data); - objectid_map = (__u32 *)(disk_sb_v1 + 1) ; - new_objectid_map = (__u32 *)(disk_sb + 1) ; + objectid_map = (__le32 *)(disk_sb_v1 + 1) ; + new_objectid_map = (__le32 *)(disk_sb + 1) ; if (cur_size > new_size) { /* mark everyone used that was listed as free at the end of the objectid diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index f4ea81ae0e0..e242ebc7f6f 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -73,8 +73,8 @@ int reiserfs_global_version_in_proc( char *buffer, char **start, off_t offset, #define DFL( x ) D4C( rs -> s_v1.x ) #define objectid_map( s, rs ) (old_format_only (s) ? \ - (__u32 *)((struct reiserfs_super_block_v1 *)rs + 1) : \ - (__u32 *)(rs + 1)) + (__le32 *)((struct reiserfs_super_block_v1 *)rs + 1) : \ + (__le32 *)(rs + 1)) #define MAP( i ) D4C( objectid_map( sb, rs )[ i ] ) #define DJF( x ) le32_to_cpu( rs -> x ) diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c index 73ec5212178..da23ba75f3d 100644 --- a/fs/reiserfs/stree.c +++ b/fs/reiserfs/stree.c @@ -87,22 +87,20 @@ inline void copy_item_head(struct item_head * p_v_to, inline int comp_short_keys (const struct reiserfs_key * le_key, const struct cpu_key * cpu_key) { - __u32 * p_s_le_u32, * p_s_cpu_u32; - int n_key_length = REISERFS_SHORT_KEY_LEN; - - p_s_le_u32 = (__u32 *)le_key; - p_s_cpu_u32 = (__u32 *)&cpu_key->on_disk_key; - for( ; n_key_length--; ++p_s_le_u32, ++p_s_cpu_u32 ) { - if ( le32_to_cpu (*p_s_le_u32) < *p_s_cpu_u32 ) + __u32 n; + n = le32_to_cpu(le_key->k_dir_id); + if (n < cpu_key->on_disk_key.k_dir_id) return -1; - if ( le32_to_cpu (*p_s_le_u32) > *p_s_cpu_u32 ) + if (n > cpu_key->on_disk_key.k_dir_id) + return 1; + n = le32_to_cpu(le_key->k_objectid); + if (n < cpu_key->on_disk_key.k_objectid) + return -1; + if (n > cpu_key->on_disk_key.k_objectid) return 1; - } - return 0; } - /* k1 is pointer to on-disk structure which is stored in little-endian form. k2 is pointer to cpu variable. Compare keys using all 4 key fields. @@ -152,18 +150,15 @@ inline int comp_short_le_keys (const struct reiserfs_key * key1, const struct re inline void le_key2cpu_key (struct cpu_key * to, const struct reiserfs_key * from) { + int version; to->on_disk_key.k_dir_id = le32_to_cpu (from->k_dir_id); to->on_disk_key.k_objectid = le32_to_cpu (from->k_objectid); // find out version of the key - to->version = le_key_version (from); - if (to->version == KEY_FORMAT_3_5) { - to->on_disk_key.u.k_offset_v1.k_offset = le32_to_cpu (from->u.k_offset_v1.k_offset); - to->on_disk_key.u.k_offset_v1.k_uniqueness = le32_to_cpu (from->u.k_offset_v1.k_uniqueness); - } else { - to->on_disk_key.u.k_offset_v2.k_offset = offset_v2_k_offset(&from->u.k_offset_v2); - to->on_disk_key.u.k_offset_v2.k_type = offset_v2_k_type(&from->u.k_offset_v2); - } + version = le_key_version (from); + to->version = version; + to->on_disk_key.k_offset = le_key_k_offset(version, from); + to->on_disk_key.k_type = le_key_k_type(version, from); } @@ -228,8 +223,14 @@ extern struct tree_balance * cur_tb; const struct reiserfs_key MIN_KEY = {0, 0, {{0, 0},}}; /* Maximal possible key. It is never in the tree. */ -const struct reiserfs_key MAX_KEY = {0xffffffff, 0xffffffff, {{0xffffffff, 0xffffffff},}}; +const struct reiserfs_key MAX_KEY = { + __constant_cpu_to_le32(0xffffffff), + __constant_cpu_to_le32(0xffffffff), + {{__constant_cpu_to_le32(0xffffffff), + __constant_cpu_to_le32(0xffffffff)},} +}; +const struct in_core_key MAX_IN_CORE_KEY = {~0U, ~0U, ~0ULL>>4, 15}; /* Get delimiting key of the buffer by looking for it in the buffers in the path, starting from the bottom of the path, and going upwards. We must check the path's validity at each step. If the key is not in @@ -997,7 +998,7 @@ static char prepare_for_delete_or_cut( int n_unfm_number, /* Number of the item unformatted nodes. */ n_counter, n_blk_size; - __u32 * p_n_unfm_pointer; /* Pointer to the unformatted node number. */ + __le32 * p_n_unfm_pointer; /* Pointer to the unformatted node number. */ __u32 tmp; struct item_head s_ih; /* Item header. */ char c_mode; /* Returned mode of the balance. */ @@ -1059,7 +1060,7 @@ static char prepare_for_delete_or_cut( /* pointers to be cut */ n_unfm_number -= pos_in_item (p_s_path); /* Set pointer to the last unformatted node pointer that is to be cut. */ - p_n_unfm_pointer = (__u32 *)B_I_PITEM(p_s_bh, &s_ih) + I_UNFM_NUM(&s_ih) - 1 - *p_n_removed; + p_n_unfm_pointer = (__le32 *)B_I_PITEM(p_s_bh, &s_ih) + I_UNFM_NUM(&s_ih) - 1 - *p_n_removed; /* We go through the unformatted nodes pointers of the indirect @@ -1081,8 +1082,8 @@ static char prepare_for_delete_or_cut( need_research = 1 ; break; } - RFALSE( p_n_unfm_pointer < (__u32 *)B_I_PITEM(p_s_bh, &s_ih) || - p_n_unfm_pointer > (__u32 *)B_I_PITEM(p_s_bh, &s_ih) + I_UNFM_NUM(&s_ih) - 1, + RFALSE( p_n_unfm_pointer < (__le32 *)B_I_PITEM(p_s_bh, &s_ih) || + p_n_unfm_pointer > (__le32 *)B_I_PITEM(p_s_bh, &s_ih) + I_UNFM_NUM(&s_ih) - 1, "vs-5265: pointer out of range"); /* Hole, nothing to remove. */ @@ -1431,7 +1432,7 @@ int reiserfs_delete_object (struct reiserfs_transaction_handle *th, struct inode #if defined( USE_INODE_GENERATION_COUNTER ) if( !old_format_only ( th -> t_super ) ) { - __u32 *inode_generation; + __le32 *inode_generation; inode_generation = &REISERFS_SB(th -> t_super) -> s_rs -> s_inode_generation; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index bcdf2438d15..31e75125f48 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -110,7 +110,7 @@ static void reiserfs_unlockfs(struct super_block *s) { reiserfs_allow_writes(s) ; } -extern const struct reiserfs_key MAX_KEY; +extern const struct in_core_key MAX_IN_CORE_KEY; /* this is used to delete "save link" when there are no items of a @@ -164,7 +164,7 @@ static int finish_unfinished (struct super_block * s) /* compose key to look for "save" links */ max_cpu_key.version = KEY_FORMAT_3_5; - max_cpu_key.on_disk_key = MAX_KEY; + max_cpu_key.on_disk_key = MAX_IN_CORE_KEY; max_cpu_key.key_length = 3; #ifdef CONFIG_QUOTA @@ -216,10 +216,10 @@ static int finish_unfinished (struct super_block * s) /* reiserfs_iget needs k_dirid and k_objectid only */ item = B_I_PITEM (bh, ih); - obj_key.on_disk_key.k_dir_id = le32_to_cpu (*(__u32 *)item); + obj_key.on_disk_key.k_dir_id = le32_to_cpu (*(__le32 *)item); obj_key.on_disk_key.k_objectid = le32_to_cpu (ih->ih_key.k_objectid); - obj_key.on_disk_key.u.k_offset_v1.k_offset = 0; - obj_key.on_disk_key.u.k_offset_v1.k_uniqueness = 0; + obj_key.on_disk_key.k_offset = 0; + obj_key.on_disk_key.k_type = 0; pathrelse (&path); @@ -304,7 +304,7 @@ void add_save_link (struct reiserfs_transaction_handle * th, int retval; struct cpu_key key; struct item_head ih; - __u32 link; + __le32 link; BUG_ON (!th->t_trans_id); @@ -889,12 +889,18 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st char * p; p = NULL; - /* "resize=NNN" */ - *blocks = simple_strtoul (arg, &p, 0); - if (*p != '\0') { - /* NNN does not look like a number */ - reiserfs_warning (s, "reiserfs_parse_options: bad value %s", arg); - return 0; + /* "resize=NNN" or "resize=auto" */ + + if (!strcmp(arg, "auto")) { + /* From JFS code, to auto-get the size.*/ + *blocks = s->s_bdev->bd_inode->i_size >> s->s_blocksize_bits; + } else { + *blocks = simple_strtoul (arg, &p, 0); + if (*p != '\0') { + /* NNN does not look like a number */ + reiserfs_warning (s, "reiserfs_parse_options: bad value %s", arg); + return 0; + } } } @@ -903,7 +909,8 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st unsigned long val = simple_strtoul (arg, &p, 0); /* commit=NNN (time in seconds) */ if ( *p != '\0' || val >= (unsigned int)-1) { - reiserfs_warning (s, "reiserfs_parse_options: bad value %s", arg); return 0; + reiserfs_warning (s, "reiserfs_parse_options: bad value %s", arg); + return 0; } *commit_max_age = (unsigned int)val; } @@ -1329,7 +1336,7 @@ static int read_super_block (struct super_block * s, int offset) return 1; } - if ( rs->s_v1.s_root_block == -1 ) { + if ( rs->s_v1.s_root_block == cpu_to_le32(-1) ) { brelse(bh) ; reiserfs_warning (s, "Unfinished reiserfsck --rebuild-tree run detected. Please run\n" "reiserfsck --rebuild-tree and wait for a completion. If that fails\n" diff --git a/fs/select.c b/fs/select.c index 25b1ccac2f2..b80e7eb0ac0 100644 --- a/fs/select.c +++ b/fs/select.c @@ -55,7 +55,8 @@ struct poll_table_page { * as all select/poll functions have to call it to add an entry to the * poll table. */ -void __pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p); +static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, + poll_table *p); void poll_initwait(struct poll_wqueues *pwq) { @@ -87,7 +88,8 @@ void poll_freewait(struct poll_wqueues *pwq) EXPORT_SYMBOL(poll_freewait); -void __pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *_p) +static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, + poll_table *_p) { struct poll_wqueues *p = container_of(_p, struct poll_wqueues, pt); struct poll_table_page *table = p->table; diff --git a/fs/seq_file.c b/fs/seq_file.c index 650c43ba86c..38ef913767f 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -51,7 +51,10 @@ EXPORT_SYMBOL(seq_open); /** * seq_read - ->read() method for sequential files. - * @file, @buf, @size, @ppos: see file_operations method + * @file: the file to read from + * @buf: the buffer to read to + * @size: the maximum number of bytes to read + * @ppos: the current position in the file * * Ready-made ->f_op->read() */ @@ -219,7 +222,9 @@ Eoverflow: /** * seq_lseek - ->llseek() method for sequential files. - * @file, @offset, @origin: see file_operations method + * @file: the file in question + * @offset: new position + * @origin: 0 for absolute, 1 for relative position * * Ready-made ->f_op->llseek() */ diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index da25aeb0e06..364208071e1 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -96,7 +96,7 @@ static int fill_read_buffer(struct dentry * dentry, struct sysfs_buffer * buffer /** * flush_read_buffer - push buffer to userspace. * @buffer: data buffer for file. - * @userbuf: user-passed buffer. + * @buf: user-passed buffer. * @count: number of bytes requested. * @ppos: file position. * @@ -164,7 +164,7 @@ out: /** * fill_write_buffer - copy buffer from userspace. * @buffer: data buffer for file. - * @userbuf: data from user. + * @buf: data from user. * @count: number of bytes in @userbuf. * * Allocate @buffer->page if it hasn't been already, then diff --git a/fs/udf/file.c b/fs/udf/file.c index 2faa4172b9f..bb40d63f328 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -49,8 +49,7 @@ static int udf_adinicb_readpage(struct file *file, struct page * page) struct inode *inode = page->mapping->host; char *kaddr; - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); kaddr = kmap(page); memset(kaddr, 0, PAGE_CACHE_SIZE); @@ -67,8 +66,7 @@ static int udf_adinicb_writepage(struct page *page, struct writeback_control *wb struct inode *inode = page->mapping->host; char *kaddr; - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); kaddr = kmap(page); memcpy(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), kaddr, inode->i_size); diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 0506e117378..3d68de39fad 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -167,8 +167,8 @@ void udf_expand_file_adinicb(struct inode * inode, int newsize, int * err) } page = grab_cache_page(inode->i_mapping, 0); - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); + if (!PageUptodate(page)) { kaddr = kmap(page); diff --git a/fs/udf/udftime.c b/fs/udf/udftime.c index c2634bda6b5..457a8fe2857 100644 --- a/fs/udf/udftime.c +++ b/fs/udf/udftime.c @@ -103,7 +103,7 @@ udf_stamp_to_time(time_t *dest, long *dest_usec, kernel_timestamp src) offset = 0; if ((src.year < EPOCH_YEAR) || - (src.year > EPOCH_YEAR+MAX_YEAR_SECONDS)) + (src.year >= EPOCH_YEAR+MAX_YEAR_SECONDS)) { *dest = -1; *dest_usec = -1; diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 554e4a18c15..d3ff7835463 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -49,7 +49,7 @@ ifeq ($(CONFIG_XFS_TRACE),y) EXTRA_CFLAGS += -DXFS_LOG_TRACE EXTRA_CFLAGS += -DXFS_RW_TRACE EXTRA_CFLAGS += -DPAGEBUF_TRACE - # EXTRA_CFLAGS += -DXFS_VNODE_TRACE + EXTRA_CFLAGS += -DXFS_VNODE_TRACE endif obj-$(CONFIG_XFS_FS) += xfs.o diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 76a84758073..9278e9aba9b 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -558,7 +558,8 @@ xfs_submit_page( int i; BUG_ON(PageWriteback(page)); - set_page_writeback(page); + if (bh_count) + set_page_writeback(page); if (clear_dirty) clear_page_dirty(page); unlock_page(page); @@ -578,9 +579,6 @@ xfs_submit_page( if (probed_page && clear_dirty) wbc->nr_to_write--; /* Wrote an "extra" page */ - } else { - end_page_writeback(page); - wbc->pages_skipped++; /* We didn't write this page */ } } @@ -602,21 +600,26 @@ xfs_convert_page( { struct buffer_head *bh_arr[MAX_BUF_PER_PAGE], *bh, *head; xfs_iomap_t *mp = iomapp, *tmp; - unsigned long end, offset; - pgoff_t end_index; - int i = 0, index = 0; + unsigned long offset, end_offset; + int index = 0; int bbits = inode->i_blkbits; + int len, page_dirty; - end_index = i_size_read(inode) >> PAGE_CACHE_SHIFT; - if (page->index < end_index) { - end = PAGE_CACHE_SIZE; - } else { - end = i_size_read(inode) & (PAGE_CACHE_SIZE-1); - } + end_offset = (i_size_read(inode) & (PAGE_CACHE_SIZE - 1)); + + /* + * page_dirty is initially a count of buffers on the page before + * EOF and is decrememted as we move each into a cleanable state. + */ + len = 1 << inode->i_blkbits; + end_offset = max(end_offset, PAGE_CACHE_SIZE); + end_offset = roundup(end_offset, len); + page_dirty = end_offset / len; + + offset = 0; bh = head = page_buffers(page); do { - offset = i << bbits; - if (offset >= end) + if (offset >= end_offset) break; if (!(PageUptodate(page) || buffer_uptodate(bh))) continue; @@ -625,6 +628,7 @@ xfs_convert_page( if (startio) { lock_buffer(bh); bh_arr[index++] = bh; + page_dirty--; } continue; } @@ -657,10 +661,11 @@ xfs_convert_page( unlock_buffer(bh); mark_buffer_dirty(bh); } - } while (i++, (bh = bh->b_this_page) != head); + page_dirty--; + } while (offset += len, (bh = bh->b_this_page) != head); - if (startio) { - xfs_submit_page(page, wbc, bh_arr, index, 1, index == i); + if (startio && index) { + xfs_submit_page(page, wbc, bh_arr, index, 1, !page_dirty); } else { unlock_page(page); } @@ -725,8 +730,11 @@ xfs_page_state_convert( __uint64_t end_offset; pgoff_t end_index, last_index, tlast; int len, err, i, cnt = 0, uptodate = 1; - int flags = startio ? 0 : BMAPI_TRYLOCK; - int page_dirty, delalloc = 0; + int flags; + int page_dirty; + + /* wait for other IO threads? */ + flags = (startio && wbc->sync_mode != WB_SYNC_NONE) ? 0 : BMAPI_TRYLOCK; /* Is this page beyond the end of the file? */ offset = i_size_read(inode); @@ -740,19 +748,22 @@ xfs_page_state_convert( } } - offset = (loff_t)page->index << PAGE_CACHE_SHIFT; end_offset = min_t(unsigned long long, - offset + PAGE_CACHE_SIZE, i_size_read(inode)); - - bh = head = page_buffers(page); - iomp = NULL; + (loff_t)(page->index + 1) << PAGE_CACHE_SHIFT, offset); + offset = (loff_t)page->index << PAGE_CACHE_SHIFT; /* - * page_dirty is initially a count of buffers on the page and - * is decrememted as we move each into a cleanable state. + * page_dirty is initially a count of buffers on the page before + * EOF and is decrememted as we move each into a cleanable state. */ - len = bh->b_size; - page_dirty = PAGE_CACHE_SIZE / len; + len = 1 << inode->i_blkbits; + p_offset = max(p_offset, PAGE_CACHE_SIZE); + p_offset = roundup(p_offset, len); + page_dirty = p_offset / len; + + iomp = NULL; + p_offset = 0; + bh = head = page_buffers(page); do { if (offset >= end_offset) @@ -804,7 +815,6 @@ xfs_page_state_convert( */ } else if (buffer_delay(bh)) { if (!iomp) { - delalloc = 1; err = xfs_map_blocks(inode, offset, len, &iomap, BMAPI_ALLOCATE | flags); if (err) { @@ -875,14 +885,15 @@ xfs_page_state_convert( if (uptodate && bh == head) SetPageUptodate(page); - if (startio) - xfs_submit_page(page, wbc, bh_arr, cnt, 0, 1); + if (startio) { + WARN_ON(page_dirty); + xfs_submit_page(page, wbc, bh_arr, cnt, 0, !page_dirty); + } if (iomp) { - tlast = (iomp->iomap_offset + iomp->iomap_bsize - 1) >> + offset = (iomp->iomap_offset + iomp->iomap_bsize - 1) >> PAGE_CACHE_SHIFT; - if (delalloc && (tlast > last_index)) - tlast = last_index; + tlast = min_t(pgoff_t, offset, last_index); xfs_cluster_write(inode, page->index + 1, iomp, wbc, startio, unmapped, tlast); } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 23e0eb67fc2..997963e5362 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -1746,13 +1746,15 @@ STATIC DECLARE_COMPLETION(pagebuf_daemon_done); STATIC struct task_struct *pagebuf_daemon_task; STATIC int pagebuf_daemon_active; STATIC int force_flush; - +STATIC int force_sleep; STATIC int pagebuf_daemon_wakeup( int priority, unsigned int mask) { + if (force_sleep) + return 0; force_flush = 1; barrier(); wake_up_process(pagebuf_daemon_task); @@ -1778,7 +1780,12 @@ pagebuf_daemon( INIT_LIST_HEAD(&tmp); do { - try_to_freeze(PF_FREEZE); + if (unlikely(current->flags & PF_FREEZE)) { + force_sleep = 1; + refrigerator(PF_FREEZE); + } else { + force_sleep = 0; + } set_current_state(TASK_INTERRUPTIBLE); schedule_timeout((xfs_buf_timer_centisecs * HZ) / 100); diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index 9f057a4a5b0..d0d412afd26 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -515,10 +515,49 @@ open_exec_out: } #endif /* HAVE_FOP_OPEN_EXEC */ +/* + * Temporary workaround to the AIO direct IO write problem. + * This code can go and we can revert to do_sync_write once + * the writepage(s) rework is merged. + */ +STATIC ssize_t +linvfs_write( + struct file *filp, + const char __user *buf, + size_t len, + loff_t *ppos) +{ + struct kiocb kiocb; + ssize_t ret; + + init_sync_kiocb(&kiocb, filp); + kiocb.ki_pos = *ppos; + ret = __linvfs_write(&kiocb, buf, 0, len, kiocb.ki_pos); + *ppos = kiocb.ki_pos; + return ret; +} +STATIC ssize_t +linvfs_write_invis( + struct file *filp, + const char __user *buf, + size_t len, + loff_t *ppos) +{ + struct kiocb kiocb; + ssize_t ret; + + init_sync_kiocb(&kiocb, filp); + kiocb.ki_pos = *ppos; + ret = __linvfs_write(&kiocb, buf, IO_INVIS, len, kiocb.ki_pos); + *ppos = kiocb.ki_pos; + return ret; +} + + struct file_operations linvfs_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, - .write = do_sync_write, + .write = linvfs_write, .readv = linvfs_readv, .writev = linvfs_writev, .aio_read = linvfs_aio_read, @@ -540,7 +579,7 @@ struct file_operations linvfs_file_operations = { struct file_operations linvfs_invis_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, - .write = do_sync_write, + .write = linvfs_write_invis, .readv = linvfs_readv_invis, .writev = linvfs_writev_invis, .aio_read = linvfs_aio_read_invis, diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index ff145fd0d1a..aa9daaea6c3 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -683,6 +683,9 @@ xfs_write( (xip->i_d.di_flags & XFS_DIFLAG_REALTIME) ? mp->m_rtdev_targp : mp->m_ddev_targp; + if (ioflags & IO_ISAIO) + return XFS_ERROR(-ENOSYS); + if ((pos & target->pbr_smask) || (count & target->pbr_smask)) return XFS_ERROR(-EINVAL); diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c index 849c61c74f3..a832d165f24 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.c +++ b/fs/xfs/linux-2.6/xfs_vnode.c @@ -156,7 +156,6 @@ vn_initialize( #ifdef XFS_VNODE_TRACE vp->v_trace = ktrace_alloc(VNODE_TRACE_SIZE, KM_SLEEP); - printk("Allocated VNODE_TRACE at 0x%p\n", vp->v_trace); #endif /* XFS_VNODE_TRACE */ vn_trace_exit(vp, "vn_initialize", (inst_t *)__return_address); @@ -424,13 +423,13 @@ vn_remove( * Vnode tracing code. */ void -vn_trace_entry(vnode_t *vp, char *func, inst_t *ra) +vn_trace_entry(vnode_t *vp, const char *func, inst_t *ra) { KTRACE_ENTER(vp, VNODE_KTRACE_ENTRY, func, 0, ra); } void -vn_trace_exit(vnode_t *vp, char *func, inst_t *ra) +vn_trace_exit(vnode_t *vp, const char *func, inst_t *ra) { KTRACE_ENTER(vp, VNODE_KTRACE_EXIT, func, 0, ra); } diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h index da76c1f1e11..00466c3194a 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.h +++ b/fs/xfs/linux-2.6/xfs_vnode.h @@ -86,10 +86,11 @@ typedef struct vnode { vnumber_t v_number; /* in-core vnode number */ vn_bhv_head_t v_bh; /* behavior head */ spinlock_t v_lock; /* VN_LOCK/VN_UNLOCK */ - struct inode v_inode; /* Linux inode */ #ifdef XFS_VNODE_TRACE struct ktrace *v_trace; /* trace header structure */ #endif + struct inode v_inode; /* Linux inode */ + /* inode MUST be last */ } vnode_t; #define v_fbhv v_bh.bh_first /* first behavior */ @@ -409,7 +410,7 @@ typedef struct vattr { int va_mask; /* bit-mask of attributes present */ enum vtype va_type; /* vnode type (for create) */ mode_t va_mode; /* file access mode and type */ - nlink_t va_nlink; /* number of references to file */ + xfs_nlink_t va_nlink; /* number of references to file */ uid_t va_uid; /* owner user id */ gid_t va_gid; /* owner group id */ xfs_ino_t va_nodeid; /* file id */ @@ -625,6 +626,7 @@ static inline int VN_BAD(struct vnode *vp) #define ATTR_DMI 0x08 /* invocation from a DMI function */ #define ATTR_LAZY 0x80 /* set/get attributes lazily */ #define ATTR_NONBLOCK 0x100 /* return EAGAIN if operation would block */ +#define ATTR_NOLOCK 0x200 /* Don't grab any conflicting locks */ /* * Flags to VOP_FSYNC and VOP_RECLAIM. @@ -646,8 +648,8 @@ static inline int VN_BAD(struct vnode *vp) #define VNODE_KTRACE_REF 4 #define VNODE_KTRACE_RELE 5 -extern void vn_trace_entry(struct vnode *, char *, inst_t *); -extern void vn_trace_exit(struct vnode *, char *, inst_t *); +extern void vn_trace_entry(struct vnode *, const char *, inst_t *); +extern void vn_trace_exit(struct vnode *, const char *, inst_t *); extern void vn_trace_hold(struct vnode *, char *, int, inst_t *); extern void vn_trace_ref(struct vnode *, char *, int, inst_t *); extern void vn_trace_rele(struct vnode *, char *, int, inst_t *); diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index 08d551a1734..63abdc2ac7f 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -182,7 +182,7 @@ xfs_swapext( if (VN_CACHED(tvp) != 0) xfs_inval_cached_pages(XFS_ITOV(tip), &(tip->i_iocore), - (loff_t)0, 0, 0); + (xfs_off_t)0, 0, 0); /* Verify O_DIRECT for ftmp */ if (VN_CACHED(tvp) != 0) { diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 3a0ba1dfd0e..d3da00045f2 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -136,6 +136,40 @@ xfs_chash_free(xfs_mount_t *mp) } /* + * Try to move an inode to the front of its hash list if possible + * (and if its not there already). Called right after obtaining + * the list version number and then dropping the read_lock on the + * hash list in question (which is done right after looking up the + * inode in question...). + */ +STATIC void +xfs_ihash_promote( + xfs_ihash_t *ih, + xfs_inode_t *ip, + ulong version) +{ + xfs_inode_t *iq; + + if ((ip->i_prevp != &ih->ih_next) && write_trylock(&ih->ih_lock)) { + if (likely(version == ih->ih_version)) { + /* remove from list */ + if ((iq = ip->i_next)) { + iq->i_prevp = ip->i_prevp; + } + *ip->i_prevp = iq; + + /* insert at list head */ + iq = ih->ih_next; + iq->i_prevp = &ip->i_next; + ip->i_next = iq; + ip->i_prevp = &ih->ih_next; + ih->ih_next = ip; + } + write_unlock(&ih->ih_lock); + } +} + +/* * Look up an inode by number in the given file system. * The inode is looked up in the hash table for the file system * represented by the mount point parameter mp. Each bucket of @@ -229,7 +263,9 @@ again: XFS_STATS_INC(xs_ig_found); ip->i_flags &= ~XFS_IRECLAIMABLE; + version = ih->ih_version; read_unlock(&ih->ih_lock); + xfs_ihash_promote(ih, ip, version); XFS_MOUNT_ILOCK(mp); list_del_init(&ip->i_reclaim); @@ -259,8 +295,15 @@ again: inode_vp, vp); } + /* + * Inode cache hit: if ip is not at the front of + * its hash chain, move it there now. + * Do this with the lock held for update, but + * do statistics after releasing the lock. + */ + version = ih->ih_version; read_unlock(&ih->ih_lock); - + xfs_ihash_promote(ih, ip, version); XFS_STATS_INC(xs_ig_found); finish_inode: @@ -547,6 +590,7 @@ xfs_inode_incore(xfs_mount_t *mp, { xfs_ihash_t *ih; xfs_inode_t *ip; + ulong version; ih = XFS_IHASH(mp, ino); read_lock(&ih->ih_lock); @@ -554,11 +598,15 @@ xfs_inode_incore(xfs_mount_t *mp, if (ip->i_ino == ino) { /* * If we find it and tp matches, return it. + * Also move it to the front of the hash list + * if we find it and it is not already there. * Otherwise break from the loop and return * NULL. */ if (ip->i_transp == tp) { + version = ih->ih_version; read_unlock(&ih->ih_lock); + xfs_ihash_promote(ih, ip, version); return (ip); } break; @@ -685,6 +733,7 @@ xfs_iextract( iq->i_prevp = ip->i_prevp; } *ip->i_prevp = iq; + ih->ih_version++; write_unlock(&ih->ih_lock); /* diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 43c632ab86a..bc8c8c7f903 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1130,7 +1130,7 @@ xfs_ialloc( xfs_trans_t *tp, xfs_inode_t *pip, mode_t mode, - nlink_t nlink, + xfs_nlink_t nlink, xfs_dev_t rdev, cred_t *cr, xfs_prid_t prid, diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index a53b1ccf607..37e1c316f3b 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -495,9 +495,9 @@ int xfs_itobp(struct xfs_mount *, struct xfs_trans *, int xfs_iread(struct xfs_mount *, struct xfs_trans *, xfs_ino_t, xfs_inode_t **, xfs_daddr_t); int xfs_iread_extents(struct xfs_trans *, xfs_inode_t *, int); -int xfs_ialloc(struct xfs_trans *, xfs_inode_t *, mode_t, nlink_t, - xfs_dev_t, struct cred *, xfs_prid_t, int, - struct xfs_buf **, boolean_t *, xfs_inode_t **); +int xfs_ialloc(struct xfs_trans *, xfs_inode_t *, mode_t, + xfs_nlink_t, xfs_dev_t, struct cred *, xfs_prid_t, + int, struct xfs_buf **, boolean_t *, xfs_inode_t **); void xfs_xlate_dinode_core(xfs_caddr_t, struct xfs_dinode_core *, int); uint xfs_ip2xflags(struct xfs_inode *); diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 3826e8f0e28..991f8a61f7c 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -308,7 +308,8 @@ phase2: break; } - error = XFS_IOMAP_WRITE_ALLOCATE(mp, io, &imap, &nimaps); + error = XFS_IOMAP_WRITE_ALLOCATE(mp, io, offset, count, + &imap, &nimaps); break; case BMAPI_UNWRITTEN: lockmode = 0; @@ -365,7 +366,7 @@ xfs_flush_space( int xfs_iomap_write_direct( xfs_inode_t *ip, - loff_t offset, + xfs_off_t offset, size_t count, int flags, xfs_bmbt_irec_t *ret_imap, @@ -541,7 +542,7 @@ error_out: int xfs_iomap_write_delay( xfs_inode_t *ip, - loff_t offset, + xfs_off_t offset, size_t count, int ioflag, xfs_bmbt_irec_t *ret_imap, @@ -746,6 +747,8 @@ write_map: int xfs_iomap_write_allocate( xfs_inode_t *ip, + xfs_off_t offset, + size_t count, xfs_bmbt_irec_t *map, int *retmap) { @@ -770,9 +773,9 @@ xfs_iomap_write_allocate( if ((error = XFS_QM_DQATTACH(mp, ip, 0))) return XFS_ERROR(error); - offset_fsb = map->br_startoff; + offset_fsb = XFS_B_TO_FSBT(mp, offset); count_fsb = map->br_blockcount; - map_start_fsb = offset_fsb; + map_start_fsb = map->br_startoff; XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb)); @@ -868,9 +871,9 @@ xfs_iomap_write_allocate( imap[i].br_startoff, imap[i].br_blockcount,imap[i].br_state); } - if ((map->br_startoff >= imap[i].br_startoff) && - (map->br_startoff < (imap[i].br_startoff + - imap[i].br_blockcount))) { + if ((offset_fsb >= imap[i].br_startoff) && + (offset_fsb < (imap[i].br_startoff + + imap[i].br_blockcount))) { *map = imap[i]; *retmap = 1; XFS_STATS_INC(xs_xstrat_quick); @@ -883,9 +886,8 @@ xfs_iomap_write_allocate( * file, just surrounding data, try again. */ nimaps--; - offset_fsb = imap[nimaps].br_startoff + - imap[nimaps].br_blockcount; - map_start_fsb = offset_fsb; + map_start_fsb = imap[nimaps].br_startoff + + imap[nimaps].br_blockcount; } trans_cancel: @@ -899,7 +901,7 @@ error0: int xfs_iomap_write_unwritten( xfs_inode_t *ip, - loff_t offset, + xfs_off_t offset, size_t count) { xfs_mount_t *mp = ip->i_mount; diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h index 31c91087cb3..4daaa521210 100644 --- a/fs/xfs/xfs_iomap.h +++ b/fs/xfs/xfs_iomap.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2003,2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2003-2005 Silicon Graphics, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License as @@ -29,9 +29,6 @@ * * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ */ - - - #ifndef __XFS_IOMAP_H__ #define __XFS_IOMAP_H__ @@ -56,7 +53,7 @@ typedef enum { BMAPI_UNWRITTEN = (1 << 3), /* unwritten extents to real extents */ /* modifiers */ BMAPI_IGNSTATE = (1 << 4), /* ignore unwritten state on read */ - BMAPI_DIRECT = (1 << 5), /* direct instead of buffered write */ + BMAPI_DIRECT = (1 << 5), /* direct instead of buffered write */ BMAPI_MMAP = (1 << 6), /* allocate for mmap write */ BMAPI_SYNC = (1 << 7), /* sync write to flush delalloc space */ BMAPI_TRYLOCK = (1 << 8), /* non-blocking request */ @@ -67,13 +64,13 @@ typedef enum { /* * xfs_iomap_t: File system I/O map * - * The iomap_bn field is expressed in 512-byte blocks, and is where the + * The iomap_bn field is expressed in 512-byte blocks, and is where the * mapping starts on disk. * * The iomap_offset, iomap_bsize and iomap_delta fields are in bytes. * iomap_offset is the offset of the mapping in the file itself. - * iomap_bsize is the size of the mapping, iomap_delta is the - * desired data's offset into the mapping, given the offset supplied + * iomap_bsize is the size of the mapping, iomap_delta is the + * desired data's offset into the mapping, given the offset supplied * to the file I/O map routine. * * When a request is made to read beyond the logical end of the object, @@ -84,8 +81,8 @@ typedef enum { typedef struct xfs_iomap { xfs_daddr_t iomap_bn; /* first 512b blk of mapping */ xfs_buftarg_t *iomap_target; - loff_t iomap_offset; /* offset of mapping, bytes */ - loff_t iomap_bsize; /* size of mapping, bytes */ + xfs_off_t iomap_offset; /* offset of mapping, bytes */ + xfs_off_t iomap_bsize; /* size of mapping, bytes */ size_t iomap_delta; /* offset into mapping, bytes */ iomap_flags_t iomap_flags; } xfs_iomap_t; @@ -96,12 +93,12 @@ struct xfs_bmbt_irec; extern int xfs_iomap(struct xfs_iocore *, xfs_off_t, ssize_t, int, struct xfs_iomap *, int *); -extern int xfs_iomap_write_direct(struct xfs_inode *, loff_t, size_t, +extern int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t, int, struct xfs_bmbt_irec *, int *, int); -extern int xfs_iomap_write_delay(struct xfs_inode *, loff_t, size_t, int, +extern int xfs_iomap_write_delay(struct xfs_inode *, xfs_off_t, size_t, int, struct xfs_bmbt_irec *, int *); -extern int xfs_iomap_write_allocate(struct xfs_inode *, +extern int xfs_iomap_write_allocate(struct xfs_inode *, xfs_off_t, size_t, struct xfs_bmbt_irec *, int *); -extern int xfs_iomap_write_unwritten(struct xfs_inode *, loff_t, size_t); +extern int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, size_t); #endif /* __XFS_IOMAP_H__*/ diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index b57423caef9..2ec967d93e5 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -301,6 +301,15 @@ xfs_mount_validate_sb( } /* + * Version 1 directory format has never worked on Linux. + */ + if (unlikely(!XFS_SB_VERSION_HASDIRV2(sbp))) { + cmn_err(CE_WARN, + "XFS: Attempted to mount file system using version 1 directory format"); + return XFS_ERROR(ENOSYS); + } + + /* * Until this is fixed only page-sized or smaller data blocks work. */ if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) { diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 5fc6201dd8e..30dd08fb9f5 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -210,15 +210,16 @@ typedef int (*xfs_bmapi_t)(struct xfs_trans *, void *, struct xfs_bmap_free *); typedef int (*xfs_bmap_eof_t)(void *, xfs_fileoff_t, int, int *); typedef int (*xfs_iomap_write_direct_t)( - void *, loff_t, size_t, int, + void *, xfs_off_t, size_t, int, struct xfs_bmbt_irec *, int *, int); typedef int (*xfs_iomap_write_delay_t)( - void *, loff_t, size_t, int, + void *, xfs_off_t, size_t, int, struct xfs_bmbt_irec *, int *); typedef int (*xfs_iomap_write_allocate_t)( - void *, struct xfs_bmbt_irec *, int *); + void *, xfs_off_t, size_t, + struct xfs_bmbt_irec *, int *); typedef int (*xfs_iomap_write_unwritten_t)( - void *, loff_t, size_t); + void *, xfs_off_t, size_t); typedef uint (*xfs_lck_map_shared_t)(void *); typedef void (*xfs_lock_t)(void *, uint); typedef void (*xfs_lock_demote_t)(void *, uint); @@ -258,9 +259,9 @@ typedef struct xfs_ioops { #define XFS_IOMAP_WRITE_DELAY(mp, io, offset, count, flags, mval, nmap) \ (*(mp)->m_io_ops.xfs_iomap_write_delay) \ ((io)->io_obj, offset, count, flags, mval, nmap) -#define XFS_IOMAP_WRITE_ALLOCATE(mp, io, mval, nmap) \ +#define XFS_IOMAP_WRITE_ALLOCATE(mp, io, offset, count, mval, nmap) \ (*(mp)->m_io_ops.xfs_iomap_write_allocate) \ - ((io)->io_obj, mval, nmap) + ((io)->io_obj, offset, count, mval, nmap) #define XFS_IOMAP_WRITE_UNWRITTEN(mp, io, offset, count) \ (*(mp)->m_io_ops.xfs_iomap_write_unwritten) \ ((io)->io_obj, offset, count) @@ -428,10 +429,10 @@ typedef struct xfs_mount { #define XFS_WRITEIO_LOG_LARGE 16 /* - * Max and min values for UIO and mount-option defined I/O sizes; - * min value can't be less than a page. Currently unused. + * Max and min values for mount-option defined I/O + * preallocation sizes. */ -#define XFS_MAX_IO_LOG 16 /* 64K */ +#define XFS_MAX_IO_LOG 30 /* 1G */ #define XFS_MIN_IO_LOG PAGE_SHIFT /* diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h index 04609d27ea5..e4bf711e48f 100644 --- a/fs/xfs/xfs_types.h +++ b/fs/xfs/xfs_types.h @@ -63,6 +63,7 @@ typedef __u64 xfs_ino_t; /* <inode> type */ typedef __s64 xfs_daddr_t; /* <disk address> type */ typedef char * xfs_caddr_t; /* <core address> type */ typedef __u32 xfs_dev_t; +typedef __u32 xfs_nlink_t; /* __psint_t is the same size as a pointer */ #if (BITS_PER_LONG == 32) diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 816b945fa0e..d1f8146a06e 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -147,7 +147,7 @@ xfs_dir_ialloc( xfs_inode_t *dp, /* directory within whose allocate the inode. */ mode_t mode, - nlink_t nlink, + xfs_nlink_t nlink, xfs_dev_t rdev, cred_t *credp, prid_t prid, /* project id */ diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h index e1ed6a58800..01d98b4b7af 100644 --- a/fs/xfs/xfs_utils.h +++ b/fs/xfs/xfs_utils.h @@ -42,7 +42,7 @@ extern int xfs_get_dir_entry (vname_t *, xfs_inode_t **); extern int xfs_dir_lookup_int (bhv_desc_t *, uint, vname_t *, xfs_ino_t *, xfs_inode_t **); extern int xfs_truncate_file (xfs_mount_t *, xfs_inode_t *); -extern int xfs_dir_ialloc (xfs_trans_t **, xfs_inode_t *, mode_t, nlink_t, +extern int xfs_dir_ialloc (xfs_trans_t **, xfs_inode_t *, mode_t, xfs_nlink_t, xfs_dev_t, cred_t *, prid_t, int, xfs_inode_t **, int *); extern int xfs_droplink (xfs_trans_t *, xfs_inode_t *); diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index 00aae9c6a90..b5373665010 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -1649,6 +1649,7 @@ xfs_vget( #define MNTOPT_SWIDTH "swidth" /* data volume stripe width */ #define MNTOPT_NOUUID "nouuid" /* ignore filesystem UUID */ #define MNTOPT_MTPT "mtpt" /* filesystem mount point */ +#define MNTOPT_ALLOCSIZE "allocsize" /* preferred allocation size */ #define MNTOPT_IHASHSIZE "ihashsize" /* size of inode hash table */ #define MNTOPT_NORECOVERY "norecovery" /* don't run XFS recovery */ #define MNTOPT_NOLOGFLUSH "nologflush" /* don't hard flush on log writes */ @@ -1657,6 +1658,28 @@ xfs_vget( #define MNTOPT_IKEEP "ikeep" /* do not free empty inode clusters */ #define MNTOPT_NOIKEEP "noikeep" /* free empty inode clusters */ +STATIC unsigned long +suffix_strtoul(const char *cp, char **endp, unsigned int base) +{ + int last, shift_left_factor = 0; + char *value = (char *)cp; + + last = strlen(value) - 1; + if (value[last] == 'K' || value[last] == 'k') { + shift_left_factor = 10; + value[last] = '\0'; + } + if (value[last] == 'M' || value[last] == 'm') { + shift_left_factor = 20; + value[last] = '\0'; + } + if (value[last] == 'G' || value[last] == 'g') { + shift_left_factor = 30; + value[last] = '\0'; + } + + return simple_strtoul(cp, endp, base) << shift_left_factor; +} int xfs_parseargs( @@ -1688,60 +1711,60 @@ xfs_parseargs( if (!strcmp(this_char, MNTOPT_LOGBUFS)) { if (!value || !*value) { printk("XFS: %s option requires an argument\n", - MNTOPT_LOGBUFS); + this_char); return EINVAL; } args->logbufs = simple_strtoul(value, &eov, 10); } else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) { - int last, in_kilobytes = 0; - if (!value || !*value) { printk("XFS: %s option requires an argument\n", - MNTOPT_LOGBSIZE); + this_char); return EINVAL; } - last = strlen(value) - 1; - if (value[last] == 'K' || value[last] == 'k') { - in_kilobytes = 1; - value[last] = '\0'; - } - args->logbufsize = simple_strtoul(value, &eov, 10); - if (in_kilobytes) - args->logbufsize <<= 10; + args->logbufsize = suffix_strtoul(value, &eov, 10); } else if (!strcmp(this_char, MNTOPT_LOGDEV)) { if (!value || !*value) { printk("XFS: %s option requires an argument\n", - MNTOPT_LOGDEV); + this_char); return EINVAL; } strncpy(args->logname, value, MAXNAMELEN); } else if (!strcmp(this_char, MNTOPT_MTPT)) { if (!value || !*value) { printk("XFS: %s option requires an argument\n", - MNTOPT_MTPT); + this_char); return EINVAL; } strncpy(args->mtpt, value, MAXNAMELEN); } else if (!strcmp(this_char, MNTOPT_RTDEV)) { if (!value || !*value) { printk("XFS: %s option requires an argument\n", - MNTOPT_RTDEV); + this_char); return EINVAL; } strncpy(args->rtname, value, MAXNAMELEN); } else if (!strcmp(this_char, MNTOPT_BIOSIZE)) { if (!value || !*value) { printk("XFS: %s option requires an argument\n", - MNTOPT_BIOSIZE); + this_char); return EINVAL; } iosize = simple_strtoul(value, &eov, 10); args->flags |= XFSMNT_IOSIZE; args->iosizelog = (uint8_t) iosize; + } else if (!strcmp(this_char, MNTOPT_ALLOCSIZE)) { + if (!value || !*value) { + printk("XFS: %s option requires an argument\n", + this_char); + return EINVAL; + } + iosize = suffix_strtoul(value, &eov, 10); + args->flags |= XFSMNT_IOSIZE; + args->iosizelog = ffs(iosize) - 1; } else if (!strcmp(this_char, MNTOPT_IHASHSIZE)) { if (!value || !*value) { printk("XFS: %s option requires an argument\n", - this_char); + this_char); return EINVAL; } args->flags |= XFSMNT_IHASHSIZE; @@ -1756,7 +1779,7 @@ xfs_parseargs( args->flags |= XFSMNT_INO64; #if !XFS_BIG_INUMS printk("XFS: %s option not allowed on this system\n", - MNTOPT_INO64); + this_char); return EINVAL; #endif } else if (!strcmp(this_char, MNTOPT_NOALIGN)) { @@ -1766,14 +1789,14 @@ xfs_parseargs( } else if (!strcmp(this_char, MNTOPT_SUNIT)) { if (!value || !*value) { printk("XFS: %s option requires an argument\n", - MNTOPT_SUNIT); + this_char); return EINVAL; } dsunit = simple_strtoul(value, &eov, 10); } else if (!strcmp(this_char, MNTOPT_SWIDTH)) { if (!value || !*value) { printk("XFS: %s option requires an argument\n", - MNTOPT_SWIDTH); + this_char); return EINVAL; } dswidth = simple_strtoul(value, &eov, 10); @@ -1781,7 +1804,7 @@ xfs_parseargs( args->flags &= ~XFSMNT_32BITINODES; #if !XFS_BIG_INUMS printk("XFS: %s option not allowed on this system\n", - MNTOPT_64BITINODE); + this_char); return EINVAL; #endif } else if (!strcmp(this_char, MNTOPT_NOUUID)) { @@ -1877,7 +1900,7 @@ xfs_showargs( seq_printf(m, "," MNTOPT_IHASHSIZE "=%d", mp->m_ihsize); if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) - seq_printf(m, "," MNTOPT_BIOSIZE "=%d", mp->m_writeio_log); + seq_printf(m, "," MNTOPT_ALLOCSIZE "=%d", 1<<mp->m_writeio_log); if (mp->m_logbufs > 0) seq_printf(m, "," MNTOPT_LOGBUFS "=%d", mp->m_logbufs); diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 70092963ca9..25a526629b1 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -305,7 +305,7 @@ xfs_setattr( int mandlock_before, mandlock_after; struct xfs_dquot *udqp, *gdqp, *olddquot1, *olddquot2; int file_owner; - int need_iolock = (flags & ATTR_DMI) == 0; + int need_iolock = 1; vp = BHV_TO_VNODE(bdp); vn_trace_entry(vp, __FUNCTION__, (inst_t *)__return_address); @@ -384,6 +384,9 @@ xfs_setattr( */ tp = NULL; lock_flags = XFS_ILOCK_EXCL; + ASSERT(flags & ATTR_NOLOCK ? flags & ATTR_DMI : 1); + if (flags & ATTR_NOLOCK) + need_iolock = 0; if (!(mask & XFS_AT_SIZE)) { if ((mask != (XFS_AT_CTIME|XFS_AT_ATIME|XFS_AT_MTIME)) || (mp->m_flags & XFS_MOUNT_WSYNC)) { @@ -4320,7 +4323,7 @@ xfs_free_file_space( int rt; xfs_fileoff_t startoffset_fsb; xfs_trans_t *tp; - int need_iolock = (attr_flags & ATTR_DMI) == 0; + int need_iolock = 1; vn_trace_entry(XFS_ITOV(ip), __FUNCTION__, (inst_t *)__return_address); mp = ip->i_mount; @@ -4348,8 +4351,12 @@ xfs_free_file_space( return(error); } + ASSERT(attr_flags & ATTR_NOLOCK ? attr_flags & ATTR_DMI : 1); + if (attr_flags & ATTR_NOLOCK) + need_iolock = 0; if (need_iolock) xfs_ilock(ip, XFS_IOLOCK_EXCL); + rounding = MAX((__uint8_t)(1 << mp->m_sb.sb_blocklog), (__uint8_t)NBPP); ilen = len + (offset & (rounding - 1)); diff --git a/include/asm-alpha/bug.h b/include/asm-alpha/bug.h index ae1e0a5fa49..39a3e2a5017 100644 --- a/include/asm-alpha/bug.h +++ b/include/asm-alpha/bug.h @@ -1,6 +1,7 @@ #ifndef _ALPHA_BUG_H #define _ALPHA_BUG_H +#ifdef CONFIG_BUG #include <asm/pal.h> /* ??? Would be nice to use .gprel32 here, but we can't be sure that the @@ -10,6 +11,8 @@ : : "i" (PAL_bugchk), "i"(__LINE__), "i"(__FILE__)) #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-alpha/errno.h b/include/asm-alpha/errno.h index c85ab6b9d6c..69e2655249d 100644 --- a/include/asm-alpha/errno.h +++ b/include/asm-alpha/errno.h @@ -116,4 +116,8 @@ #define EKEYREVOKED 134 /* Key has been revoked */ #define EKEYREJECTED 135 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 136 /* Owner died */ +#define ENOTRECOVERABLE 137 /* State not recoverable */ + #endif diff --git a/include/asm-alpha/siginfo.h b/include/asm-alpha/siginfo.h index 86bcab59c52..9822362a842 100644 --- a/include/asm-alpha/siginfo.h +++ b/include/asm-alpha/siginfo.h @@ -4,8 +4,6 @@ #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) #define __ARCH_SI_TRAPNO -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) - #include <asm-generic/siginfo.h> #endif diff --git a/include/asm-alpha/signal.h b/include/asm-alpha/signal.h index 25f98bc5576..1a2c52a056f 100644 --- a/include/asm-alpha/signal.h +++ b/include/asm-alpha/signal.h @@ -109,34 +109,11 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 4096 #define SIGSTKSZ 16384 - -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x40000000 -#endif - #define SIG_BLOCK 1 /* for blocking signals */ #define SIG_UNBLOCK 2 /* for unblocking signals */ #define SIG_SETMASK 3 /* for setting the signal mask */ -/* Type of a signal handler. */ -typedef void __signalfn_t(int); -typedef __signalfn_t __user *__sighandler_t; - -typedef void __restorefn_t(void); -typedef __restorefn_t __user *__sigrestore_t; - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct osf_sigaction { diff --git a/include/asm-arm/arch-cl7500/vmalloc.h b/include/asm-arm/arch-cl7500/vmalloc.h index 91883def488..ba8d7a84456 100644 --- a/include/asm-arm/arch-cl7500/vmalloc.h +++ b/include/asm-arm/arch-cl7500/vmalloc.h @@ -1,15 +1,4 @@ /* * linux/include/asm-arm/arch-cl7500/vmalloc.h */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (PAGE_OFFSET + 0x1c000000) diff --git a/include/asm-arm/arch-clps711x/vmalloc.h b/include/asm-arm/arch-clps711x/vmalloc.h index 42571ed5e49..a5dfe96abc9 100644 --- a/include/asm-arm/arch-clps711x/vmalloc.h +++ b/include/asm-arm/arch-clps711x/vmalloc.h @@ -17,15 +17,4 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (PAGE_OFFSET + 0x10000000) diff --git a/include/asm-arm/arch-ebsa110/vmalloc.h b/include/asm-arm/arch-ebsa110/vmalloc.h index 759659be109..26674ba4683 100644 --- a/include/asm-arm/arch-ebsa110/vmalloc.h +++ b/include/asm-arm/arch-ebsa110/vmalloc.h @@ -7,15 +7,4 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (PAGE_OFFSET + 0x1f000000) diff --git a/include/asm-arm/arch-ebsa285/vmalloc.h b/include/asm-arm/arch-ebsa285/vmalloc.h index def705a3c20..d1ca955ce43 100644 --- a/include/asm-arm/arch-ebsa285/vmalloc.h +++ b/include/asm-arm/arch-ebsa285/vmalloc.h @@ -8,17 +8,6 @@ #include <linux/config.h> -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) - #ifdef CONFIG_ARCH_FOOTBRIDGE #define VMALLOC_END (PAGE_OFFSET + 0x30000000) #else diff --git a/include/asm-arm/arch-epxa10db/vmalloc.h b/include/asm-arm/arch-epxa10db/vmalloc.h index d31ef858476..546fb7d2b6a 100644 --- a/include/asm-arm/arch-epxa10db/vmalloc.h +++ b/include/asm-arm/arch-epxa10db/vmalloc.h @@ -17,15 +17,4 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (PAGE_OFFSET + 0x10000000) diff --git a/include/asm-arm/arch-h720x/vmalloc.h b/include/asm-arm/arch-h720x/vmalloc.h index 4af523a5e18..b4693cb821e 100644 --- a/include/asm-arm/arch-h720x/vmalloc.h +++ b/include/asm-arm/arch-h720x/vmalloc.h @@ -5,17 +5,6 @@ #ifndef __ARCH_ARM_VMALLOC_H #define __ARCH_ARM_VMALLOC_H -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) -#define VMALLOC_VMADDR(x) ((unsigned long)(x)) #define VMALLOC_END (PAGE_OFFSET + 0x10000000) #endif diff --git a/include/asm-arm/arch-imx/imxfb.h b/include/asm-arm/arch-imx/imxfb.h new file mode 100644 index 00000000000..2346d454ab9 --- /dev/null +++ b/include/asm-arm/arch-imx/imxfb.h @@ -0,0 +1,35 @@ +/* + * This structure describes the machine which we are running on. + */ +struct imxfb_mach_info { + u_long pixclock; + + u_short xres; + u_short yres; + + u_char bpp; + u_char hsync_len; + u_char left_margin; + u_char right_margin; + + u_char vsync_len; + u_char upper_margin; + u_char lower_margin; + u_char sync; + + u_int cmap_greyscale:1, + cmap_inverse:1, + cmap_static:1, + unused:29; + + u_int pcr; + u_int pwmr; + u_int lscr1; + + u_char * fixed_screen_cpu; + dma_addr_t fixed_screen_dma; + + void (*lcd_power)(int); + void (*backlight_power)(int); +}; +void set_imx_fb_info(struct imxfb_mach_info *hard_imx_fb_info); diff --git a/include/asm-arm/arch-imx/vmalloc.h b/include/asm-arm/arch-imx/vmalloc.h index 252038f4816..cb616912706 100644 --- a/include/asm-arm/arch-imx/vmalloc.h +++ b/include/asm-arm/arch-imx/vmalloc.h @@ -17,16 +17,4 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) -#define VMALLOC_VMADDR(x) ((unsigned long)(x)) #define VMALLOC_END (PAGE_OFFSET + 0x10000000) diff --git a/include/asm-arm/arch-integrator/cm.h b/include/asm-arm/arch-integrator/cm.h index d31c1a71f78..1ab353e2359 100644 --- a/include/asm-arm/arch-integrator/cm.h +++ b/include/asm-arm/arch-integrator/cm.h @@ -24,9 +24,9 @@ void cm_control(u32, u32); #define CM_CTRL_LCDBIASDN (1 << 10) #define CM_CTRL_LCDMUXSEL_MASK (7 << 11) #define CM_CTRL_LCDMUXSEL_GENLCD (1 << 11) -#define CM_CTRL_LCDMUXSEL_SHARPLCD1 (3 << 11) -#define CM_CTRL_LCDMUXSEL_SHARPLCD2 (4 << 11) -#define CM_CTRL_LCDMUXSEL_VGA (7 << 11) +#define CM_CTRL_LCDMUXSEL_VGA_16BPP (2 << 11) +#define CM_CTRL_LCDMUXSEL_SHARPLCD (3 << 11) +#define CM_CTRL_LCDMUXSEL_VGA_8421BPP (4 << 11) #define CM_CTRL_LCDEN0 (1 << 14) #define CM_CTRL_LCDEN1 (1 << 15) #define CM_CTRL_STATIC1 (1 << 16) diff --git a/include/asm-arm/arch-integrator/platform.h b/include/asm-arm/arch-integrator/platform.h index 6b67e41669f..bd364f5a99b 100644 --- a/include/asm-arm/arch-integrator/platform.h +++ b/include/asm-arm/arch-integrator/platform.h @@ -20,14 +20,14 @@ * * Copyright © ARM Limited 1998. All rights reserved. * ***********************************************************************/ /* ************************************************************************ - * + * * Integrator address map - * + * * NOTE: This is a multi-hosted header file for use with uHAL and * supported debuggers. - * + * * $Id: platform.s,v 1.32 2000/02/18 10:51:39 asims Exp $ - * + * * ***********************************************************************/ #ifndef __address_h @@ -40,22 +40,22 @@ * Memory definitions * ------------------------------------------------------------------------ * Integrator memory map - * + * */ #define INTEGRATOR_BOOT_ROM_LO 0x00000000 #define INTEGRATOR_BOOT_ROM_HI 0x20000000 #define INTEGRATOR_BOOT_ROM_BASE INTEGRATOR_BOOT_ROM_HI /* Normal position */ #define INTEGRATOR_BOOT_ROM_SIZE SZ_512K -/* +/* * New Core Modules have different amounts of SSRAM, the amount of SSRAM * fitted can be found in HDR_STAT. - * + * * The symbol INTEGRATOR_SSRAM_SIZE is kept, however this now refers to * the minimum amount of SSRAM fitted on any core module. - * + * * New Core Modules also alias the SSRAM. - * + * */ #define INTEGRATOR_SSRAM_BASE 0x00000000 #define INTEGRATOR_SSRAM_ALIAS_BASE 0x10800000 @@ -67,9 +67,9 @@ #define INTEGRATOR_MBRD_SSRAM_BASE 0x28000000 #define INTEGRATOR_MBRD_SSRAM_SIZE SZ_512K -/* +/* * SDRAM is a SIMM therefore the size is not known. - * + * */ #define INTEGRATOR_SDRAM_BASE 0x00040000 @@ -79,9 +79,9 @@ #define INTEGRATOR_HDR2_SDRAM_BASE 0xA0000000 #define INTEGRATOR_HDR3_SDRAM_BASE 0xB0000000 -/* +/* * Logic expansion modules - * + * */ #define INTEGRATOR_LOGIC_MODULES_BASE 0xC0000000 #define INTEGRATOR_LOGIC_MODULE0_BASE 0xC0000000 @@ -92,7 +92,7 @@ /* ------------------------------------------------------------------------ * Integrator header card registers * ------------------------------------------------------------------------ - * + * */ #define INTEGRATOR_HDR_ID_OFFSET 0x00 #define INTEGRATOR_HDR_PROC_OFFSET 0x04 @@ -185,12 +185,12 @@ /* ------------------------------------------------------------------------ * Integrator system registers * ------------------------------------------------------------------------ - * + * */ -/* +/* * System Controller - * + * */ #define INTEGRATOR_SC_ID_OFFSET 0x00 #define INTEGRATOR_SC_OSC_OFFSET 0x04 @@ -230,11 +230,11 @@ #define INTEGRATOR_SC_CTRL_URTS1 (1 << 6) #define INTEGRATOR_SC_CTRL_UDTR1 (1 << 7) -/* +/* * External Bus Interface - * + * */ -#define INTEGRATOR_EBI_BASE 0x12000000 +#define INTEGRATOR_EBI_BASE 0x12000000 #define INTEGRATOR_EBI_CSR0_OFFSET 0x00 #define INTEGRATOR_EBI_CSR1_OFFSET 0x04 @@ -279,9 +279,9 @@ #define INTEGRATOR_KBD_BASE 0x18000000 /* Keyboard */ #define INTEGRATOR_MOUSE_BASE 0x19000000 /* Mouse */ -/* +/* * LED's & Switches - * + * */ #define INTEGRATOR_DBG_ALPHA_OFFSET 0x00 #define INTEGRATOR_DBG_LEDS_OFFSET 0x04 @@ -300,7 +300,7 @@ * ------------------------------------------------------------------------ */ /* PS2 Keyboard interface */ -#define KMI0_BASE INTEGRATOR_KBD_BASE +#define KMI0_BASE INTEGRATOR_KBD_BASE /* PS2 Mouse interface */ #define KMI1_BASE INTEGRATOR_MOUSE_BASE @@ -313,7 +313,7 @@ * This represents a fairly liberal usage of address space. Even though * the V3 only has two windows (therefore we need to map stuff on the fly), * we maintain the same addresses, even if they're not mapped. - * + * */ #define PHYS_PCI_MEM_BASE 0x40000000 /* 512M to xxx */ /* unused 256M from A0000000-AFFFFFFF might be used for I2O ??? @@ -326,7 +326,7 @@ */ #define PHYS_PCI_V3_BASE 0x62000000 -#define PCI_DRAMSIZE INTEGRATOR_SSRAM_SIZE +#define PCI_DRAMSIZE INTEGRATOR_SSRAM_SIZE /* 'export' these to UHAL */ #define UHAL_PCI_IO PCI_IO_BASE @@ -334,7 +334,7 @@ #define UHAL_PCI_ALLOC_IO_BASE 0x00004000 #define UHAL_PCI_ALLOC_MEM_BASE PCI_MEM_BASE #define UHAL_PCI_MAX_SLOT 20 - + /* ======================================================================== * Start of uHAL definitions * ======================================================================== @@ -343,17 +343,17 @@ /* ------------------------------------------------------------------------ * Integrator Interrupt Controllers * ------------------------------------------------------------------------ - * - * Offsets from interrupt controller base - * + * + * Offsets from interrupt controller base + * * System Controller interrupt controller base is - * + * * INTEGRATOR_IC_BASE + (header_number << 6) - * + * * Core Module interrupt controller base is - * - * INTEGRATOR_HDR_IC - * + * + * INTEGRATOR_HDR_IC + * */ #define IRQ_STATUS 0 #define IRQ_RAW_STATUS 0x04 @@ -374,22 +374,22 @@ /* ------------------------------------------------------------------------ * Interrupts * ------------------------------------------------------------------------ - * - * + * + * * Each Core Module has two interrupts controllers, one on the core module * itself and one in the system controller on the motherboard. The * READ_INT macro in target.s reads both interrupt controllers and returns * a 32 bit bitmask, bits 0 to 23 are interrupts from the system controller * and bits 24 to 31 are from the core module. - * + * * The following definitions relate to the bitmask returned by READ_INT. - * + * */ /* ------------------------------------------------------------------------ * LED's - The header LED is not accessible via the uHAL API * ------------------------------------------------------------------------ - * + * */ #define GREEN_LED 0x01 #define YELLOW_LED 0x02 @@ -399,44 +399,44 @@ #define LED_BANK INTEGRATOR_DBG_LEDS -/* +/* * Memory definitions - run uHAL out of SSRAM. - * + * */ #define uHAL_MEMORY_SIZE INTEGRATOR_SSRAM_SIZE -/* +/* * Application Flash - * + * */ #define FLASH_BASE INTEGRATOR_FLASH_BASE #define FLASH_SIZE INTEGRATOR_FLASH_SIZE #define FLASH_END (FLASH_BASE + FLASH_SIZE - 1) #define FLASH_BLOCK_SIZE SZ_128K -/* +/* * Boot Flash - * + * */ #define EPROM_BASE INTEGRATOR_BOOT_ROM_HI #define EPROM_SIZE INTEGRATOR_BOOT_ROM_SIZE #define EPROM_END (EPROM_BASE + EPROM_SIZE - 1) -/* +/* * Clean base - dummy - * + * */ #define CLEAN_BASE EPROM_BASE -/* +/* * Timer definitions - * + * * Only use timer 1 & 2 * (both run at 24MHz and will need the clock divider set to 16). - * + * * Timer 0 runs at bus frequency and therefore could vary and currently * uHAL can't handle that. - * + * */ #define INTEGRATOR_TIMER0_BASE INTEGRATOR_CT_BASE @@ -447,9 +447,9 @@ #define MAX_PERIOD 699050 #define TICKS_PER_uSEC 24 -/* - * These are useconds NOT ticks. - * +/* + * These are useconds NOT ticks. + * */ #define mSEC_1 1000 #define mSEC_5 (mSEC_1 * 5) diff --git a/include/asm-arm/arch-integrator/vmalloc.h b/include/asm-arm/arch-integrator/vmalloc.h index 50e9aee7948..170cccece52 100644 --- a/include/asm-arm/arch-integrator/vmalloc.h +++ b/include/asm-arm/arch-integrator/vmalloc.h @@ -17,15 +17,4 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (PAGE_OFFSET + 0x10000000) diff --git a/include/asm-arm/arch-iop3xx/vmalloc.h b/include/asm-arm/arch-iop3xx/vmalloc.h index dc1d2a95716..0f2f6847f93 100644 --- a/include/asm-arm/arch-iop3xx/vmalloc.h +++ b/include/asm-arm/arch-iop3xx/vmalloc.h @@ -10,9 +10,6 @@ * The vmalloc() routines leaves a hole of 4kB between each vmalloced * area for the same reason. ;) */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) -#define VMALLOC_VMADDR(x) ((unsigned long)(x)) //#define VMALLOC_END (0xe8000000) /* increase usable physical RAM to ~992M per RMK */ #define VMALLOC_END (0xfe000000) diff --git a/include/asm-arm/arch-ixp2000/platform.h b/include/asm-arm/arch-ixp2000/platform.h index 509e44d528d..901bba6d02b 100644 --- a/include/asm-arm/arch-ixp2000/platform.h +++ b/include/asm-arm/arch-ixp2000/platform.h @@ -121,6 +121,7 @@ unsigned long ixp2000_gettimeoffset(void); struct pci_sys_data; +u32 *ixp2000_pci_config_addr(unsigned int bus, unsigned int devfn, int where); void ixp2000_pci_preinit(void); int ixp2000_pci_setup(int, struct pci_sys_data*); struct pci_bus* ixp2000_pci_scan_bus(int, struct pci_sys_data*); diff --git a/include/asm-arm/arch-ixp2000/vmalloc.h b/include/asm-arm/arch-ixp2000/vmalloc.h index 2e4bcbcf31f..473dff4ec56 100644 --- a/include/asm-arm/arch-ixp2000/vmalloc.h +++ b/include/asm-arm/arch-ixp2000/vmalloc.h @@ -17,7 +17,4 @@ * The vmalloc() routines leaves a hole of 4kB between each vmalloced * area for the same reason. ;) */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) -#define VMALLOC_VMADDR(x) ((unsigned long)(x)) #define VMALLOC_END 0xfaffefff diff --git a/include/asm-arm/arch-ixp4xx/vmalloc.h b/include/asm-arm/arch-ixp4xx/vmalloc.h index da46e560ad6..050d46e6b12 100644 --- a/include/asm-arm/arch-ixp4xx/vmalloc.h +++ b/include/asm-arm/arch-ixp4xx/vmalloc.h @@ -1,17 +1,5 @@ /* * linux/include/asm-arm/arch-ixp4xx/vmalloc.h */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) -#define VMALLOC_VMADDR(x) ((unsigned long)(x)) #define VMALLOC_END (0xFF000000) diff --git a/include/asm-arm/arch-l7200/vmalloc.h b/include/asm-arm/arch-l7200/vmalloc.h index edeaebe1f14..816231eedaa 100644 --- a/include/asm-arm/arch-l7200/vmalloc.h +++ b/include/asm-arm/arch-l7200/vmalloc.h @@ -1,15 +1,4 @@ /* * linux/include/asm-arm/arch-l7200/vmalloc.h */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (PAGE_OFFSET + 0x10000000) diff --git a/include/asm-arm/arch-lh7a40x/vmalloc.h b/include/asm-arm/arch-lh7a40x/vmalloc.h index 5ac607925be..8163e45109b 100644 --- a/include/asm-arm/arch-lh7a40x/vmalloc.h +++ b/include/asm-arm/arch-lh7a40x/vmalloc.h @@ -7,15 +7,4 @@ * version 2 as published by the Free Software Foundation. * */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after - * the physical memory until the kernel virtual memory starts. That - * means that any out-of-bounds memory accesses will hopefully be - * caught. The vmalloc() routines leaves a hole of 4kB (one page) - * between each vmalloced area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (0xe8000000) diff --git a/include/asm-arm/arch-omap/vmalloc.h b/include/asm-arm/arch-omap/vmalloc.h index c6a83581a2f..5b8bd8dae8b 100644 --- a/include/asm-arm/arch-omap/vmalloc.h +++ b/include/asm-arm/arch-omap/vmalloc.h @@ -17,17 +17,5 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) -#define VMALLOC_VMADDR(x) ((unsigned long)(x)) #define VMALLOC_END (PAGE_OFFSET + 0x10000000) diff --git a/include/asm-arm/arch-pxa/vmalloc.h b/include/asm-arm/arch-pxa/vmalloc.h index 3381af6ddb0..5bb450c7aa2 100644 --- a/include/asm-arm/arch-pxa/vmalloc.h +++ b/include/asm-arm/arch-pxa/vmalloc.h @@ -8,15 +8,4 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (0xe8000000) diff --git a/include/asm-arm/arch-rpc/vmalloc.h b/include/asm-arm/arch-rpc/vmalloc.h index a13c27f37d7..077046bb2f3 100644 --- a/include/asm-arm/arch-rpc/vmalloc.h +++ b/include/asm-arm/arch-rpc/vmalloc.h @@ -7,15 +7,4 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (PAGE_OFFSET + 0x1c000000) diff --git a/include/asm-arm/arch-s3c2410/regs-iis.h b/include/asm-arm/arch-s3c2410/regs-iis.h index 7ae8e1f45bc..385b07d510d 100644 --- a/include/asm-arm/arch-s3c2410/regs-iis.h +++ b/include/asm-arm/arch-s3c2410/regs-iis.h @@ -14,6 +14,7 @@ * 26-06-2003 BJD Finished off definitions for register addresses * 12-03-2004 BJD Updated include protection * 07-03-2005 BJD Added FIFO size flags and S3C2440 MPLL + * 05-04-2005 LCVR Added IISFCON definitions for the S3C2400 */ #ifndef __ASM_ARCH_REGS_IIS_H @@ -68,5 +69,14 @@ #define S3C2410_IISFCON_RXMASK (0x3f) #define S3C2410_IISFCON_RXSHIFT (0) +#define S3C2400_IISFCON_TXDMA (1<<11) +#define S3C2400_IISFCON_RXDMA (1<<10) +#define S3C2400_IISFCON_TXENABLE (1<<9) +#define S3C2400_IISFCON_RXENABLE (1<<8) +#define S3C2400_IISFCON_TXMASK (0x07 << 4) +#define S3C2400_IISFCON_TXSHIFT (4) +#define S3C2400_IISFCON_RXMASK (0x07) +#define S3C2400_IISFCON_RXSHIFT (0) + #define S3C2410_IISFIFO (0x10) #endif /* __ASM_ARCH_REGS_IIS_H */ diff --git a/include/asm-arm/arch-s3c2410/regs-mem.h b/include/asm-arm/arch-s3c2410/regs-mem.h index 1a1328ac0d7..a2d7d0cec04 100644 --- a/include/asm-arm/arch-s3c2410/regs-mem.h +++ b/include/asm-arm/arch-s3c2410/regs-mem.h @@ -12,6 +12,7 @@ * Changelog: * 29-Sep-2004 BJD Initial include for Linux * 10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA + * 04-Apr-2005 LCVR Added S3C2400 DRAM/BANKSIZE_MASK definitions * */ @@ -183,6 +184,12 @@ #define S3C2410_REFRESH_TRP_3clk (1<<20) #define S3C2410_REFRESH_TRP_4clk (2<<20) +#define S3C2400_REFRESH_DRAM_TRP_MASK (3<<20) +#define S3C2400_REFRESH_DRAM_TRP_1_5clk (0<<20) +#define S3C2400_REFRESH_DRAM_TRP_2_5clk (1<<20) +#define S3C2400_REFRESH_DRAM_TRP_3_5clk (2<<20) +#define S3C2400_REFRESH_DRAM_TRP_4_5clk (3<<20) + #define S3C2410_REFRESH_TSRC_MASK (3<<18) #define S3C2410_REFRESH_TSRC_4clk (0<<18) #define S3C2410_REFRESH_TSRC_5clk (1<<18) @@ -205,6 +212,7 @@ #define S3C2410_BANKSIZE_4M (0x5 << 0) #define S3C2410_BANKSIZE_2M (0x4 << 0) #define S3C2410_BANKSIZE_MASK (0x7 << 0) +#define S3C2400_BANKSIZE_MASK (0x4 << 0) #define S3C2410_BANKSIZE_SCLK_EN (1<<4) #define S3C2410_BANKSIZE_SCKE_EN (1<<5) #define S3C2410_BANKSIZE_BURST (1<<7) diff --git a/include/asm-arm/arch-s3c2410/regs-spi.h b/include/asm-arm/arch-s3c2410/regs-spi.h index cb502a88158..338217858c7 100644 --- a/include/asm-arm/arch-s3c2410/regs-spi.h +++ b/include/asm-arm/arch-s3c2410/regs-spi.h @@ -12,6 +12,7 @@ * 20-04-2004 KF Created file * 04-10-2004 BJD Removed VA address (no longer mapped) * tidied file for submission + * 03-04-2005 LCVR Added S3C2400_SPPIN_nCS definition */ #ifndef __ASM_ARCH_REGS_SPI_H @@ -46,6 +47,7 @@ #define S3C2410_SPPIN_ENMUL (1<<2) /* Multi Master Error detect */ #define S3C2410_SPPIN_RESERVED (1<<1) +#define S3C2400_SPPIN_nCS (1<<1) /* SPI Card Select */ #define S3C2410_SPPIN_KEEP (1<<0) /* Master Out keep */ diff --git a/include/asm-arm/arch-s3c2410/uncompress.h b/include/asm-arm/arch-s3c2410/uncompress.h index ad4252e2779..d7a4a8354fa 100644 --- a/include/asm-arm/arch-s3c2410/uncompress.h +++ b/include/asm-arm/arch-s3c2410/uncompress.h @@ -16,6 +16,7 @@ * 12-Oct-2004 BJD Take account of debug uart configuration * 15-Nov-2004 BJD Fixed uart configuration * 22-Feb-2005 BJD Added watchdog to uncompress + * 04-Apr-2005 LCVR Added support to S3C2400 (no cpuid at GSTATUS1) */ #ifndef __ASM_ARCH_UNCOMPRESS_H @@ -69,9 +70,12 @@ uart_rd(unsigned int reg) static void putc(char ch) { - int cpuid = *((volatile unsigned int *)S3C2410_GSTATUS1); + int cpuid = S3C2410_GSTATUS1_2410; +#ifndef CONFIG_CPU_S3C2400 + cpuid = *((volatile unsigned int *)S3C2410_GSTATUS1); cpuid &= S3C2410_GSTATUS1_IDMASK; +#endif if (ch == '\n') putc('\r'); /* expand newline to \r\n */ diff --git a/include/asm-arm/arch-s3c2410/vmalloc.h b/include/asm-arm/arch-s3c2410/vmalloc.h index 5fe72ad7090..33963cd5461 100644 --- a/include/asm-arm/arch-s3c2410/vmalloc.h +++ b/include/asm-arm/arch-s3c2410/vmalloc.h @@ -19,18 +19,6 @@ #ifndef __ASM_ARCH_VMALLOC_H #define __ASM_ARCH_VMALLOC_H -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ - -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) -#define VMALLOC_VMADDR(x) ((unsigned long)(x)) #define VMALLOC_END (0xE0000000) #endif /* __ASM_ARCH_VMALLOC_H */ diff --git a/include/asm-arm/arch-sa1100/vmalloc.h b/include/asm-arm/arch-sa1100/vmalloc.h index 135bc9493c0..2fb1c6f3aa1 100644 --- a/include/asm-arm/arch-sa1100/vmalloc.h +++ b/include/asm-arm/arch-sa1100/vmalloc.h @@ -1,15 +1,4 @@ /* * linux/include/asm-arm/arch-sa1100/vmalloc.h */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (0xe8000000) diff --git a/include/asm-arm/arch-shark/vmalloc.h b/include/asm-arm/arch-shark/vmalloc.h index 1cc20098f69..10db5d18823 100644 --- a/include/asm-arm/arch-shark/vmalloc.h +++ b/include/asm-arm/arch-shark/vmalloc.h @@ -1,15 +1,4 @@ /* * linux/include/asm-arm/arch-rpc/vmalloc.h */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) #define VMALLOC_END (PAGE_OFFSET + 0x10000000) diff --git a/include/asm-arm/arch-versatile/vmalloc.h b/include/asm-arm/arch-versatile/vmalloc.h index adfb34829bf..ac780df6215 100644 --- a/include/asm-arm/arch-versatile/vmalloc.h +++ b/include/asm-arm/arch-versatile/vmalloc.h @@ -18,16 +18,4 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ - -/* - * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 8MB value just means that there will be a 8MB "hole" after the - * physical memory until the kernel virtual memory starts. That means that - * any out-of-bounds memory accesses will hopefully be caught. - * The vmalloc() routines leaves a hole of 4kB between each vmalloced - * area for the same reason. ;) - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) -#define VMALLOC_VMADDR(x) ((unsigned long)(x)) #define VMALLOC_END (PAGE_OFFSET + 0x18000000) diff --git a/include/asm-arm/bug.h b/include/asm-arm/bug.h index 5e91b90a818..24d11672eb6 100644 --- a/include/asm-arm/bug.h +++ b/include/asm-arm/bug.h @@ -3,6 +3,7 @@ #include <linux/config.h> +#ifdef CONFIG_BUG #ifdef CONFIG_DEBUG_BUGVERBOSE extern volatile void __bug(const char *file, int line, void *data); @@ -17,6 +18,8 @@ extern volatile void __bug(const char *file, int line, void *data); #endif #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-arm/hardware/amba_clcd.h b/include/asm-arm/hardware/amba_clcd.h index 2149be7c702..ce4cf5c1c05 100644 --- a/include/asm-arm/hardware/amba_clcd.h +++ b/include/asm-arm/hardware/amba_clcd.h @@ -153,7 +153,7 @@ struct clcd_fb { static inline void clcdfb_decode(struct clcd_fb *fb, struct clcd_regs *regs) { - u32 val; + u32 val, cpl; /* * Program the CLCD controller registers and start the CLCD @@ -164,7 +164,10 @@ static inline void clcdfb_decode(struct clcd_fb *fb, struct clcd_regs *regs) val |= (fb->fb.var.left_margin - 1) << 24; regs->tim0 = val; - val = fb->fb.var.yres - 1; + val = fb->fb.var.yres; + if (fb->panel->cntl & CNTL_LCDDUAL) + val /= 2; + val -= 1; val |= (fb->fb.var.vsync_len - 1) << 10; val |= fb->fb.var.lower_margin << 16; val |= fb->fb.var.upper_margin << 24; @@ -174,13 +177,17 @@ static inline void clcdfb_decode(struct clcd_fb *fb, struct clcd_regs *regs) val |= fb->fb.var.sync & FB_SYNC_HOR_HIGH_ACT ? 0 : TIM2_IHS; val |= fb->fb.var.sync & FB_SYNC_VERT_HIGH_ACT ? 0 : TIM2_IVS; - if (fb->panel->cntl & CNTL_LCDTFT) - val |= (fb->fb.var.xres_virtual - 1) << 16; - else if (fb->panel->cntl & CNTL_LCDBW) - printk("what value for CPL for stnmono panels?"); - else - val |= ((fb->fb.var.xres_virtual * 8 / 3) - 1) << 16; - regs->tim2 = val; + cpl = fb->fb.var.xres_virtual; + if (fb->panel->cntl & CNTL_LCDTFT) /* TFT */ + /* / 1 */; + else if (!fb->fb.var.grayscale) /* STN color */ + cpl = cpl * 8 / 3; + else if (fb->panel->cntl & CNTL_LCDMONO8) /* STN monochrome, 8bit */ + cpl /= 8; + else /* STN monochrome, 4bit */ + cpl /= 4; + + regs->tim2 = val | ((cpl - 1) << 16); regs->tim3 = fb->panel->tim3; @@ -204,7 +211,7 @@ static inline void clcdfb_decode(struct clcd_fb *fb, struct clcd_regs *regs) case 16: val |= CNTL_LCDBPP16; break; - case 24: + case 32: val |= CNTL_LCDBPP24; break; } @@ -215,8 +222,8 @@ static inline void clcdfb_decode(struct clcd_fb *fb, struct clcd_regs *regs) static inline int clcdfb_check(struct clcd_fb *fb, struct fb_var_screeninfo *var) { - var->xres_virtual = var->xres = (var->xres + 7) & ~7; - var->yres_virtual = var->yres; + var->xres_virtual = var->xres = (var->xres + 15) & ~15; + var->yres_virtual = var->yres = (var->yres + 1) & ~1; #define CHECK(e,l,h) (var->e < l || var->e > h) if (CHECK(right_margin, (5+1), 256) || /* back porch */ diff --git a/include/asm-arm/hardware/clock.h b/include/asm-arm/hardware/clock.h index 4983449ff2c..19da861e523 100644 --- a/include/asm-arm/hardware/clock.h +++ b/include/asm-arm/hardware/clock.h @@ -26,10 +26,13 @@ struct clk; /** * clk_get - lookup and obtain a reference to a clock producer. * @dev: device for clock "consumer" - * @id: device ID + * @id: clock comsumer ID * * Returns a struct clk corresponding to the clock producer, or - * valid IS_ERR() condition containing errno. + * valid IS_ERR() condition containing errno. The implementation + * uses @dev and @id to determine the clock consumer, and thereby + * the clock producer. (IOW, @id may be identical strings, but + * clk_get may return different clock producers depending on @dev.) */ struct clk *clk_get(struct device *dev, const char *id); diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h index 69bc7a3e816..658ffa384fd 100644 --- a/include/asm-arm/io.h +++ b/include/asm-arm/io.h @@ -99,12 +99,16 @@ extern void __readwrite_bug(const char *fn); */ #ifdef __io #define outb(v,p) __raw_writeb(v,__io(p)) -#define outw(v,p) __raw_writew(cpu_to_le16(v),__io(p)) -#define outl(v,p) __raw_writel(cpu_to_le32(v),__io(p)) +#define outw(v,p) __raw_writew((__force __u16) \ + cpu_to_le16(v),__io(p)) +#define outl(v,p) __raw_writel((__force __u32) \ + cpu_to_le32(v),__io(p)) -#define inb(p) ({ unsigned int __v = __raw_readb(__io(p)); __v; }) -#define inw(p) ({ unsigned int __v = le16_to_cpu(__raw_readw(__io(p))); __v; }) -#define inl(p) ({ unsigned int __v = le32_to_cpu(__raw_readl(__io(p))); __v; }) +#define inb(p) ({ __u8 __v = __raw_readb(__io(p)); __v; }) +#define inw(p) ({ __u16 __v = le16_to_cpu((__force __le16) \ + __raw_readw(__io(p))); __v; }) +#define inl(p) ({ __u32 __v = le32_to_cpu((__force __le32) \ + __raw_readl(__io(p))); __v; }) #define outsb(p,d,l) __raw_writesb(__io(p),d,l) #define outsw(p,d,l) __raw_writesw(__io(p),d,l) @@ -149,9 +153,11 @@ extern void _memset_io(void __iomem *, int, size_t); * IO port primitives for more information. */ #ifdef __mem_pci -#define readb(c) ({ unsigned int __v = __raw_readb(__mem_pci(c)); __v; }) -#define readw(c) ({ unsigned int __v = le16_to_cpu(__raw_readw(__mem_pci(c))); __v; }) -#define readl(c) ({ unsigned int __v = le32_to_cpu(__raw_readl(__mem_pci(c))); __v; }) +#define readb(c) ({ __u8 __v = __raw_readb(__mem_pci(c)); __v; }) +#define readw(c) ({ __u16 __v = le16_to_cpu((__force __le16) \ + __raw_readw(__mem_pci(c))); __v; }) +#define readl(c) ({ __u32 __v = le32_to_cpu((__force __le32) \ + __raw_readl(__mem_pci(c))); __v; }) #define readb_relaxed(addr) readb(addr) #define readw_relaxed(addr) readw(addr) #define readl_relaxed(addr) readl(addr) @@ -161,8 +167,10 @@ extern void _memset_io(void __iomem *, int, size_t); #define readsl(p,d,l) __raw_readsl(__mem_pci(p),d,l) #define writeb(v,c) __raw_writeb(v,__mem_pci(c)) -#define writew(v,c) __raw_writew(cpu_to_le16(v),__mem_pci(c)) -#define writel(v,c) __raw_writel(cpu_to_le32(v),__mem_pci(c)) +#define writew(v,c) __raw_writew((__force __u16) \ + cpu_to_le16(v),__mem_pci(c)) +#define writel(v,c) __raw_writel((__force __u32) \ + cpu_to_le32(v),__mem_pci(c)) #define writesb(p,d,l) __raw_writesb(__mem_pci(p),d,l) #define writesw(p,d,l) __raw_writesw(__mem_pci(p),d,l) diff --git a/include/asm-arm/pgtable.h b/include/asm-arm/pgtable.h index 2df4eacf4fa..a9892eb42a2 100644 --- a/include/asm-arm/pgtable.h +++ b/include/asm-arm/pgtable.h @@ -17,6 +17,23 @@ #include <asm/arch/vmalloc.h> /* + * Just any arbitrary offset to the start of the vmalloc VM area: the + * current 8MB value just means that there will be a 8MB "hole" after the + * physical memory until the kernel virtual memory starts. That means that + * any out-of-bounds memory accesses will hopefully be caught. + * The vmalloc() routines leaves a hole of 4kB between each vmalloced + * area for the same reason. ;) + * + * Note that platforms may override VMALLOC_START, but they must provide + * VMALLOC_END. VMALLOC_END defines the (exclusive) limit of this space, + * which may not overlap IO space. + */ +#ifndef VMALLOC_START +#define VMALLOC_OFFSET (8*1024*1024) +#define VMALLOC_START (((unsigned long)high_memory + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1)) +#endif + +/* * Hardware-wise, we have a two level page table structure, where the first * level has 4096 entries, and the second level has 256 entries. Each entry * is one 32-bit word. Most of the bits in the second level entry are used diff --git a/include/asm-arm/processor.h b/include/asm-arm/processor.h index 4a9845997a7..7d4118e0905 100644 --- a/include/asm-arm/processor.h +++ b/include/asm-arm/processor.h @@ -23,8 +23,6 @@ #include <asm/procinfo.h> #include <asm/types.h> -#define KERNEL_STACK_SIZE PAGE_SIZE - union debug_insn { u32 arm; u16 thumb; @@ -87,8 +85,9 @@ unsigned long get_wchan(struct task_struct *p); */ extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags); -#define KSTK_EIP(tsk) (((unsigned long *)(4096+(unsigned long)(tsk)->thread_info))[1019]) -#define KSTK_ESP(tsk) (((unsigned long *)(4096+(unsigned long)(tsk)->thread_info))[1017]) +#define KSTK_REGS(tsk) (((struct pt_regs *)(THREAD_START_SP + (unsigned long)(tsk)->thread_info)) - 1) +#define KSTK_EIP(tsk) KSTK_REGS(tsk)->ARM_pc +#define KSTK_ESP(tsk) KSTK_REGS(tsk)->ARM_sp /* * Prefetching support - only ARMv5. diff --git a/include/asm-arm/rtc.h b/include/asm-arm/rtc.h index aa7e16b2e22..370dfe77589 100644 --- a/include/asm-arm/rtc.h +++ b/include/asm-arm/rtc.h @@ -18,9 +18,9 @@ struct rtc_ops { void (*release)(void); int (*ioctl)(unsigned int, unsigned long); - void (*read_time)(struct rtc_time *); + int (*read_time)(struct rtc_time *); int (*set_time)(struct rtc_time *); - void (*read_alarm)(struct rtc_wkalrm *); + int (*read_alarm)(struct rtc_wkalrm *); int (*set_alarm)(struct rtc_wkalrm *); int (*proc)(char *buf); }; diff --git a/include/asm-arm/signal.h b/include/asm-arm/signal.h index b033e5fd60f..46e69ae395a 100644 --- a/include/asm-arm/signal.h +++ b/include/asm-arm/signal.h @@ -114,34 +114,10 @@ typedef unsigned long sigset_t; #define SIGSTKSZ 8192 #ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE 0x80000000 -#define SA_SAMPLE_RANDOM 0x10000000 #define SA_IRQNOMASK 0x08000000 -#define SA_SHIRQ 0x04000000 #endif -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void __signalfn_t(int); -typedef __signalfn_t __user *__sighandler_t; - -typedef void __restorefn_t(void); -typedef __restorefn_t __user *__sigrestore_t; - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-arm/string.h b/include/asm-arm/string.h index 2a8ab162412..e50c4a39b69 100644 --- a/include/asm-arm/string.h +++ b/include/asm-arm/string.h @@ -29,15 +29,22 @@ extern void __memzero(void *ptr, __kernel_size_t n); #define memset(p,v,n) \ ({ \ - if ((n) != 0) { \ + void *__p = (p); size_t __n = n; \ + if ((__n) != 0) { \ if (__builtin_constant_p((v)) && (v) == 0) \ - __memzero((p),(n)); \ + __memzero((__p),(__n)); \ else \ - memset((p),(v),(n)); \ + memset((__p),(v),(__n)); \ } \ - (p); \ + (__p); \ }) -#define memzero(p,n) ({ if ((n) != 0) __memzero((p),(n)); (p); }) +#define memzero(p,n) \ + ({ \ + void *__p = (p); size_t __n = n; \ + if ((__n) != 0) \ + __memzero((__p),(__n)); \ + (__p); \ + }) #endif diff --git a/include/asm-arm/thread_info.h b/include/asm-arm/thread_info.h index a61618fb433..66c585c50cf 100644 --- a/include/asm-arm/thread_info.h +++ b/include/asm-arm/thread_info.h @@ -14,6 +14,10 @@ #include <asm/fpstate.h> +#define THREAD_SIZE_ORDER 1 +#define THREAD_SIZE 8192 +#define THREAD_START_SP (THREAD_SIZE - 8) + #ifndef __ASSEMBLY__ struct task_struct; @@ -77,8 +81,6 @@ struct thread_info { #define init_thread_info (init_thread_union.thread_info) #define init_stack (init_thread_union.stack) -#define THREAD_SIZE 8192 - /* * how to get the thread information struct from C */ diff --git a/include/asm-arm/unistd.h b/include/asm-arm/unistd.h index a19ec09eaa0..ace27480886 100644 --- a/include/asm-arm/unistd.h +++ b/include/asm-arm/unistd.h @@ -359,8 +359,7 @@ #define __ARM_NR_cacheflush (__ARM_NR_BASE+2) #define __ARM_NR_usr26 (__ARM_NR_BASE+3) #define __ARM_NR_usr32 (__ARM_NR_BASE+4) - -#define __ARM_NR_set_tls (__ARM_NR_BASE+0x800) +#define __ARM_NR_set_tls (__ARM_NR_BASE+5) #define __sys2(x) #x #define __sys1(x) __sys2(x) diff --git a/include/asm-arm26/bug.h b/include/asm-arm26/bug.h index 920b7053336..7177c739996 100644 --- a/include/asm-arm26/bug.h +++ b/include/asm-arm26/bug.h @@ -3,6 +3,7 @@ #include <linux/config.h> +#ifdef CONFIG_BUG #ifdef CONFIG_DEBUG_BUGVERBOSE extern volatile void __bug(const char *file, int line, void *data); /* give file/line information */ @@ -12,6 +13,8 @@ extern volatile void __bug(const char *file, int line, void *data); #endif #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-arm26/signal.h b/include/asm-arm26/signal.h index 6f62e51a2e5..dedb2928030 100644 --- a/include/asm-arm26/signal.h +++ b/include/asm-arm26/signal.h @@ -114,30 +114,10 @@ typedef unsigned long sigset_t; #define SIGSTKSZ 8192 #ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE 0x80000000 -#define SA_SAMPLE_RANDOM 0x10000000 #define SA_IRQNOMASK 0x08000000 -#define SA_SHIRQ 0x04000000 #endif -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-cris/page.h b/include/asm-cris/page.h index ddd8915e41e..c767da1ef8f 100644 --- a/include/asm-cris/page.h +++ b/include/asm-cris/page.h @@ -77,10 +77,6 @@ typedef struct { unsigned long pgprot; } pgprot_t; printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \ } while (0) -#define PAGE_BUG(page) do { \ - BUG(); \ -} while (0) - /* Pure 2^n version of get_order */ static inline int get_order(unsigned long size) { diff --git a/include/asm-cris/signal.h b/include/asm-cris/signal.h index 3f187ec4800..dfe039593a7 100644 --- a/include/asm-cris/signal.h +++ b/include/asm-cris/signal.h @@ -108,30 +108,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-frv/bug.h b/include/asm-frv/bug.h index 011860b2881..074c0d5770e 100644 --- a/include/asm-frv/bug.h +++ b/include/asm-frv/bug.h @@ -13,6 +13,7 @@ #include <linux/config.h> +#ifdef CONFIG_BUG /* * Tell the user there is some problem. */ @@ -45,6 +46,7 @@ do { \ #define HAVE_ARCH_KGDB_BAD_PAGE #define kgdb_bad_page(page) do { kgdb_raise(SIGABRT); } while(0) #endif +#endif #include <asm-generic/bug.h> diff --git a/include/asm-frv/pgtable.h b/include/asm-frv/pgtable.h index 3c6d42a22df..d0a9c2f9c13 100644 --- a/include/asm-frv/pgtable.h +++ b/include/asm-frv/pgtable.h @@ -349,9 +349,9 @@ static inline pmd_t *pmd_offset(pud_t *dir, unsigned long address) /* * Define this to warn about kernel memory accesses that are - * done without a 'verify_area(VERIFY_WRITE,..)' + * done without a 'access_ok(VERIFY_WRITE,..)' */ -#undef TEST_VERIFY_AREA +#undef TEST_ACCESS_OK #define pte_present(x) (pte_val(x) & _PAGE_PRESENT) #define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0) diff --git a/include/asm-frv/signal.h b/include/asm-frv/signal.h index f18952f86a8..d407bde57ec 100644 --- a/include/asm-frv/signal.h +++ b/include/asm-frv/signal.h @@ -107,30 +107,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index e5913c3b715..400c2b41896 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -4,6 +4,7 @@ #include <linux/compiler.h> #include <linux/config.h> +#ifdef CONFIG_BUG #ifndef HAVE_ARCH_BUG #define BUG() do { \ printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \ @@ -11,13 +12,6 @@ } while (0) #endif -#ifndef HAVE_ARCH_PAGE_BUG -#define PAGE_BUG(page) do { \ - printk("page BUG for page at %p\n", page); \ - BUG(); \ -} while (0) -#endif - #ifndef HAVE_ARCH_BUG_ON #define BUG_ON(condition) do { if (unlikely((condition)!=0)) BUG(); } while(0) #endif @@ -31,4 +25,18 @@ } while (0) #endif +#else /* !CONFIG_BUG */ +#ifndef HAVE_ARCH_BUG +#define BUG() +#endif + +#ifndef HAVE_ARCH_BUG_ON +#define BUG_ON(condition) do { if (condition) ; } while(0) +#endif + +#ifndef HAVE_ARCH_WARN_ON +#define WARN_ON(condition) do { if (condition) ; } while(0) +#endif +#endif + #endif diff --git a/include/asm-generic/errno.h b/include/asm-generic/errno.h index 4dd2384bc38..e8852c092fe 100644 --- a/include/asm-generic/errno.h +++ b/include/asm-generic/errno.h @@ -102,4 +102,8 @@ #define EKEYREVOKED 128 /* Key has been revoked */ #define EKEYREJECTED 129 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 130 /* Owner died */ +#define ENOTRECOVERABLE 131 /* State not recoverable */ + #endif diff --git a/include/asm-generic/resource.h b/include/asm-generic/resource.h index b1fcda9eac2..cfe3692b23e 100644 --- a/include/asm-generic/resource.h +++ b/include/asm-generic/resource.h @@ -41,8 +41,11 @@ #define RLIMIT_LOCKS 10 /* maximum file locks held */ #define RLIMIT_SIGPENDING 11 /* max number of pending signals */ #define RLIMIT_MSGQUEUE 12 /* maximum bytes in POSIX mqueues */ +#define RLIMIT_NICE 13 /* max nice prio allowed to raise to + 0-39 for nice level 19 .. -20 */ +#define RLIMIT_RTPRIO 14 /* maximum realtime priority */ -#define RLIM_NLIMITS 13 +#define RLIM_NLIMITS 15 /* * SuS says limits have to be unsigned. @@ -81,6 +84,8 @@ [RLIMIT_LOCKS] = { RLIM_INFINITY, RLIM_INFINITY }, \ [RLIMIT_SIGPENDING] = { 0, 0 }, \ [RLIMIT_MSGQUEUE] = { MQ_BYTES_MAX, MQ_BYTES_MAX }, \ + [RLIMIT_NICE] = { 0, 0 }, \ + [RLIMIT_RTPRIO] = { 0, 0 }, \ } #endif /* __KERNEL__ */ diff --git a/include/asm-generic/sections.h b/include/asm-generic/sections.h index 976ac29598b..195ccdc069e 100644 --- a/include/asm-generic/sections.h +++ b/include/asm-generic/sections.h @@ -8,6 +8,8 @@ extern char _data[], _sdata[], _edata[]; extern char __bss_start[], __bss_stop[]; extern char __init_begin[], __init_end[]; extern char _sinittext[], _einittext[]; +extern char _sextratext[] __attribute__((weak)); +extern char _eextratext[] __attribute__((weak)); extern char _end[]; #endif /* _ASM_GENERIC_SECTIONS_H_ */ diff --git a/include/asm-generic/siginfo.h b/include/asm-generic/siginfo.h index 9cac8e8dde5..8786e01e0db 100644 --- a/include/asm-generic/siginfo.h +++ b/include/asm-generic/siginfo.h @@ -236,11 +236,18 @@ typedef struct siginfo { #define SIGEV_THREAD 2 /* deliver via thread creation */ #define SIGEV_THREAD_ID 4 /* deliver to thread */ -#define SIGEV_MAX_SIZE 64 -#ifndef SIGEV_PAD_SIZE -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 3) +/* + * This works because the alignment is ok on all current architectures + * but we leave open this being overridden in the future + */ +#ifndef __ARCH_SIGEV_PREAMBLE_SIZE +#define __ARCH_SIGEV_PREAMBLE_SIZE (sizeof(int) * 2 + sizeof(sigval_t)) #endif +#define SIGEV_MAX_SIZE 64 +#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE - __ARCH_SIGEV_PREAMBLE_SIZE) \ + / sizeof(int)) + typedef struct sigevent { sigval_t sigev_value; int sigev_signo; diff --git a/include/asm-generic/signal.h b/include/asm-generic/signal.h new file mode 100644 index 00000000000..9418d6e9b8c --- /dev/null +++ b/include/asm-generic/signal.h @@ -0,0 +1,21 @@ +#ifndef SIG_BLOCK +#define SIG_BLOCK 0 /* for blocking signals */ +#endif +#ifndef SIG_UNBLOCK +#define SIG_UNBLOCK 1 /* for unblocking signals */ +#endif +#ifndef SIG_SETMASK +#define SIG_SETMASK 2 /* for setting the signal mask */ +#endif + +#ifndef __ASSEMBLY__ +typedef void __signalfn_t(int); +typedef __signalfn_t __user *__sighandler_t; + +typedef void __restorefn_t(void); +typedef __restorefn_t __user *__sigrestore_t; + +#define SIG_DFL ((__force __sighandler_t)0) /* default signal handling */ +#define SIG_IGN ((__force __sighandler_t)1) /* ignore signal */ +#define SIG_ERR ((__force __sighandler_t)-1) /* error return from signal */ +#endif diff --git a/include/asm-generic/unaligned.h b/include/asm-generic/unaligned.h index c856a43e3b4..6c90f0f36ee 100644 --- a/include/asm-generic/unaligned.h +++ b/include/asm-generic/unaligned.h @@ -76,46 +76,47 @@ static inline void __ustw(__u16 val, __u16 *addr) ptr->x = val; } -static inline unsigned long __get_unaligned(const void *ptr, size_t size) -{ - unsigned long val; - switch (size) { - case 1: - val = *(const __u8 *)ptr; - break; - case 2: - val = __uldw((const __u16 *)ptr); - break; - case 4: - val = __uldl((const __u32 *)ptr); - break; - case 8: - val = __uldq((const __u64 *)ptr); - break; - default: - bad_unaligned_access_length(); - }; - return val; -} - -static inline void __put_unaligned(unsigned long val, void *ptr, size_t size) -{ - switch (size) { - case 1: - *(__u8 *)ptr = val; - break; - case 2: - __ustw(val, (__u16 *)ptr); - break; - case 4: - __ustl(val, (__u32 *)ptr); - break; - case 8: - __ustq(val, (__u64 *)ptr); - break; - default: - bad_unaligned_access_length(); - }; -} +#define __get_unaligned(ptr, size) ({ \ + const void *__gu_p = ptr; \ + unsigned long val; \ + switch (size) { \ + case 1: \ + val = *(const __u8 *)__gu_p; \ + break; \ + case 2: \ + val = __uldw(__gu_p); \ + break; \ + case 4: \ + val = __uldl(__gu_p); \ + break; \ + case 8: \ + val = __uldq(__gu_p); \ + break; \ + default: \ + bad_unaligned_access_length(); \ + }; \ + val; \ +}) + +#define __put_unaligned(val, ptr, size) \ +do { \ + void *__gu_p = ptr; \ + switch (size) { \ + case 1: \ + *(__u8 *)__gu_p = val; \ + break; \ + case 2: \ + __ustw(val, __gu_p); \ + break; \ + case 4: \ + __ustl(val, __gu_p); \ + break; \ + case 8: \ + __ustq(val, __gu_p); \ + break; \ + default: \ + bad_unaligned_access_length(); \ + }; \ +} while(0) #endif /* _ASM_GENERIC_UNALIGNED_H */ diff --git a/include/asm-h8300/signal.h b/include/asm-h8300/signal.h index 3a08544a473..8eccdc17616 100644 --- a/include/asm-h8300/signal.h +++ b/include/asm-h8300/signal.h @@ -107,29 +107,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h index e1de67483f3..a5810cf7b57 100644 --- a/include/asm-i386/apic.h +++ b/include/asm-i386/apic.h @@ -109,7 +109,6 @@ extern int APIC_init_uniprocessor (void); extern void disable_APIC_timer(void); extern void enable_APIC_timer(void); -extern int check_nmi_watchdog (void); extern void enable_NMI_through_LVT0 (void * dummy); extern unsigned int nmi_watchdog; diff --git a/include/asm-i386/bug.h b/include/asm-i386/bug.h index 706eb511c33..8f79de19eb9 100644 --- a/include/asm-i386/bug.h +++ b/include/asm-i386/bug.h @@ -9,6 +9,8 @@ * undefined" opcode for parsing in the trap handler. */ +#ifdef CONFIG_BUG +#define HAVE_ARCH_BUG #ifdef CONFIG_DEBUG_BUGVERBOSE #define BUG() \ __asm__ __volatile__( "ud2\n" \ @@ -18,8 +20,7 @@ #else #define BUG() __asm__ __volatile__("ud2\n") #endif +#endif -#define HAVE_ARCH_BUG #include <asm-generic/bug.h> - #endif diff --git a/include/asm-i386/checksum.h b/include/asm-i386/checksum.h index d76a5f081c9..641342002bc 100644 --- a/include/asm-i386/checksum.h +++ b/include/asm-i386/checksum.h @@ -33,7 +33,7 @@ asmlinkage unsigned int csum_partial_copy_generic(const unsigned char *src, unsi * passed in an incorrect kernel address to one of these functions. * * If you use these functions directly please don't forget the - * verify_area(). + * access_ok(). */ static __inline__ unsigned int csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, diff --git a/include/asm-i386/cpufeature.h b/include/asm-i386/cpufeature.h index e147cabd3bf..ff1187e80c3 100644 --- a/include/asm-i386/cpufeature.h +++ b/include/asm-i386/cpufeature.h @@ -87,8 +87,8 @@ #define X86_FEATURE_XCRYPT_EN (5*32+ 7) /* on-CPU crypto enabled */ /* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */ -#define X86_FEATURE_LAHF_LM (5*32+ 0) /* LAHF/SAHF in long mode */ -#define X86_FEATURE_CMP_LEGACY (5*32+ 1) /* If yes HyperThreading not valid */ +#define X86_FEATURE_LAHF_LM (6*32+ 0) /* LAHF/SAHF in long mode */ +#define X86_FEATURE_CMP_LEGACY (6*32+ 1) /* If yes HyperThreading not valid */ #define cpu_has(c, bit) test_bit(bit, (c)->x86_capability) #define boot_cpu_has(bit) test_bit(bit, boot_cpu_data.x86_capability) diff --git a/include/asm-i386/e820.h b/include/asm-i386/e820.h index 5c285aee729..edf65be21a9 100644 --- a/include/asm-i386/e820.h +++ b/include/asm-i386/e820.h @@ -13,7 +13,7 @@ #define __E820_HEADER #define E820MAP 0x2d0 /* our map */ -#define E820MAX 32 /* number of entries in E820MAP */ +#define E820MAX 128 /* number of entries in E820MAP */ #define E820NR 0x1e8 /* # entries in E820MAP */ #define E820_RAM 1 diff --git a/include/asm-i386/floppy.h b/include/asm-i386/floppy.h index f4782284807..79727afb94c 100644 --- a/include/asm-i386/floppy.h +++ b/include/asm-i386/floppy.h @@ -257,7 +257,7 @@ static int hard_dma_setup(char *addr, unsigned long size, int mode, int io) return 0; } -struct fd_routine_l { +static struct fd_routine_l { int (*_request_dma)(unsigned int dmanr, const char * device_id); void (*_free_dma)(unsigned int dmanr); int (*_get_dma_residue)(unsigned int dummy); diff --git a/include/asm-i386/hpet.h b/include/asm-i386/hpet.h index 6e20b079f1d..16ef9f996e3 100644 --- a/include/asm-i386/hpet.h +++ b/include/asm-i386/hpet.h @@ -92,6 +92,7 @@ extern unsigned long hpet_tick; /* hpet clks count per tick */ extern unsigned long hpet_address; /* hpet memory map physical address */ +extern int hpet_use_timer; extern int hpet_rtc_timer_init(void); extern int hpet_enable(void); diff --git a/include/asm-i386/module.h b/include/asm-i386/module.h index 508865e2630..eb7f2b4234a 100644 --- a/include/asm-i386/module.h +++ b/include/asm-i386/module.h @@ -52,8 +52,8 @@ struct mod_arch_specific #define MODULE_PROC_FAMILY "CYRIXIII " #elif defined CONFIG_MVIAC3_2 #define MODULE_PROC_FAMILY "VIAC3-2 " -#elif CONFIG_MGEODE -#define MODULE_PROC_FAMILY "GEODE " +#elif CONFIG_MGEODEGX1 +#define MODULE_PROC_FAMILY "GEODEGX1 " #else #error unknown processor family #endif diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h index 5c725425d86..8d60c2b4b00 100644 --- a/include/asm-i386/pgtable.h +++ b/include/asm-i386/pgtable.h @@ -193,9 +193,9 @@ extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC; /* * Define this if things work differently on an i386 and an i486: * it will (on an i486) warn about kernel memory accesses that are - * done without a 'verify_area(VERIFY_WRITE,..)' + * done without a 'access_ok(VERIFY_WRITE,..)' */ -#undef TEST_VERIFY_AREA +#undef TEST_ACCESS_OK /* The boot page tables (all created as a single array) */ extern unsigned long pg0[]; diff --git a/include/asm-i386/setup.h b/include/asm-i386/setup.h index 8814b54c75d..7a32184d54b 100644 --- a/include/asm-i386/setup.h +++ b/include/asm-i386/setup.h @@ -16,7 +16,7 @@ #define MAXMEM_PFN PFN_DOWN(MAXMEM) #define MAX_NONPAE_PFN (1 << 20) -#define PARAM_SIZE 2048 +#define PARAM_SIZE 4096 #define COMMAND_LINE_SIZE 256 #define OLD_CL_MAGIC_ADDR 0x90020 diff --git a/include/asm-i386/signal.h b/include/asm-i386/signal.h index 7ef343b6812..cbb47d34aa3 100644 --- a/include/asm-i386/signal.h +++ b/include/asm-i386/signal.h @@ -110,34 +110,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void __signalfn_t(int); -typedef __signalfn_t __user *__sighandler_t; - -typedef void __restorefn_t(void); -typedef __restorefn_t __user *__sigrestore_t; - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-i386/string.h b/include/asm-i386/string.h index 1679983d053..6a78ac58c19 100644 --- a/include/asm-i386/string.h +++ b/include/asm-i386/string.h @@ -198,47 +198,80 @@ static inline void * __memcpy(void * to, const void * from, size_t n) int d0, d1, d2; __asm__ __volatile__( "rep ; movsl\n\t" - "testb $2,%b4\n\t" - "je 1f\n\t" - "movsw\n" - "1:\ttestb $1,%b4\n\t" - "je 2f\n\t" - "movsb\n" - "2:" + "movl %4,%%ecx\n\t" + "andl $3,%%ecx\n\t" +#if 1 /* want to pay 2 byte penalty for a chance to skip microcoded rep? */ + "jz 1f\n\t" +#endif + "rep ; movsb\n\t" + "1:" : "=&c" (d0), "=&D" (d1), "=&S" (d2) - :"0" (n/4), "q" (n),"1" ((long) to),"2" ((long) from) + : "0" (n/4), "g" (n), "1" ((long) to), "2" ((long) from) : "memory"); return (to); } /* - * This looks horribly ugly, but the compiler can optimize it totally, + * This looks ugly, but the compiler can optimize it totally, * as the count is constant. */ static inline void * __constant_memcpy(void * to, const void * from, size_t n) { - if (n <= 128) - return __builtin_memcpy(to, from, n); - -#define COMMON(x) \ -__asm__ __volatile__( \ - "rep ; movsl" \ - x \ - : "=&c" (d0), "=&D" (d1), "=&S" (d2) \ - : "0" (n/4),"1" ((long) to),"2" ((long) from) \ - : "memory"); -{ - int d0, d1, d2; + long esi, edi; + if (!n) return to; +#if 1 /* want to do small copies with non-string ops? */ + switch (n) { + case 1: *(char*)to = *(char*)from; return to; + case 2: *(short*)to = *(short*)from; return to; + case 4: *(int*)to = *(int*)from; return to; +#if 1 /* including those doable with two moves? */ + case 3: *(short*)to = *(short*)from; + *((char*)to+2) = *((char*)from+2); return to; + case 5: *(int*)to = *(int*)from; + *((char*)to+4) = *((char*)from+4); return to; + case 6: *(int*)to = *(int*)from; + *((short*)to+2) = *((short*)from+2); return to; + case 8: *(int*)to = *(int*)from; + *((int*)to+1) = *((int*)from+1); return to; +#endif + } +#endif + esi = (long) from; + edi = (long) to; + if (n >= 5*4) { + /* large block: use rep prefix */ + int ecx; + __asm__ __volatile__( + "rep ; movsl" + : "=&c" (ecx), "=&D" (edi), "=&S" (esi) + : "0" (n/4), "1" (edi),"2" (esi) + : "memory" + ); + } else { + /* small block: don't clobber ecx + smaller code */ + if (n >= 4*4) __asm__ __volatile__("movsl" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + if (n >= 3*4) __asm__ __volatile__("movsl" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + if (n >= 2*4) __asm__ __volatile__("movsl" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + if (n >= 1*4) __asm__ __volatile__("movsl" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + } switch (n % 4) { - case 0: COMMON(""); return to; - case 1: COMMON("\n\tmovsb"); return to; - case 2: COMMON("\n\tmovsw"); return to; - default: COMMON("\n\tmovsw\n\tmovsb"); return to; + /* tail */ + case 0: return to; + case 1: __asm__ __volatile__("movsb" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + return to; + case 2: __asm__ __volatile__("movsw" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + return to; + default: __asm__ __volatile__("movsw\n\tmovsb" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + return to; } } - -#undef COMMON -} #define __HAVE_ARCH_MEMCPY diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h index 6f74d4c44a0..3db717a244f 100644 --- a/include/asm-i386/system.h +++ b/include/asm-i386/system.h @@ -81,7 +81,7 @@ static inline unsigned long _get_base(char * addr) #define loadsegment(seg,value) \ asm volatile("\n" \ "1:\t" \ - "movl %0,%%" #seg "\n" \ + "mov %0,%%" #seg "\n" \ "2:\n" \ ".section .fixup,\"ax\"\n" \ "3:\t" \ @@ -93,13 +93,13 @@ static inline unsigned long _get_base(char * addr) ".align 4\n\t" \ ".long 1b,3b\n" \ ".previous" \ - : :"m" (*(unsigned int *)&(value))) + : :"m" (value)) /* * Save a segment register away */ #define savesegment(seg, value) \ - asm volatile("movl %%" #seg ",%0":"=m" (*(int *)&(value))) + asm volatile("mov %%" #seg ",%0":"=m" (value)) /* * Clear and set 'TS' bit respectively diff --git a/include/asm-ia64/bitops.h b/include/asm-ia64/bitops.h index 925d54cee47..7232528e2d0 100644 --- a/include/asm-ia64/bitops.h +++ b/include/asm-ia64/bitops.h @@ -314,8 +314,8 @@ __ffs (unsigned long x) #ifdef __KERNEL__ /* - * find_last_zero_bit - find the last zero bit in a 64 bit quantity - * @x: The value to search + * Return bit number of last (most-significant) bit set. Undefined + * for x==0. Bits are numbered from 0..63 (e.g., ia64_fls(9) == 3). */ static inline unsigned long ia64_fls (unsigned long x) @@ -327,10 +327,23 @@ ia64_fls (unsigned long x) return exp - 0xffff; } +/* + * Find the last (most significant) bit set. Returns 0 for x==0 and + * bits are numbered from 1..32 (e.g., fls(9) == 4). + */ static inline int -fls (int x) +fls (int t) { - return ia64_fls((unsigned int) x); + unsigned long x = t & 0xffffffffu; + + if (!x) + return 0; + x |= x >> 1; + x |= x >> 2; + x |= x >> 4; + x |= x >> 8; + x |= x >> 16; + return ia64_popcnt(x); } /* diff --git a/include/asm-ia64/bug.h b/include/asm-ia64/bug.h index 2c0cd51e885..3aa0a0a5474 100644 --- a/include/asm-ia64/bug.h +++ b/include/asm-ia64/bug.h @@ -1,6 +1,7 @@ #ifndef _ASM_IA64_BUG_H #define _ASM_IA64_BUG_H +#ifdef CONFIG_BUG #if (__GNUC__ > 3) || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1) # define ia64_abort() __builtin_trap() #else @@ -8,8 +9,10 @@ #endif #define BUG() do { printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); ia64_abort(); } while (0) -/* should this BUG should be made generic? */ +/* should this BUG be made generic? */ #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-ia64/gcc_intrin.h b/include/asm-ia64/gcc_intrin.h index 7c357dfbae5..4fb4e439b05 100644 --- a/include/asm-ia64/gcc_intrin.h +++ b/include/asm-ia64/gcc_intrin.h @@ -133,13 +133,17 @@ register unsigned long ia64_r13 asm ("r13") __attribute_used__; ia64_intri_res; \ }) -#define ia64_popcnt(x) \ -({ \ +#if __GNUC__ >= 4 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 4) +# define ia64_popcnt(x) __builtin_popcountl(x) +#else +# define ia64_popcnt(x) \ + ({ \ __u64 ia64_intri_res; \ asm ("popcnt %0=%1" : "=r" (ia64_intri_res) : "r" (x)); \ \ ia64_intri_res; \ -}) + }) +#endif #define ia64_getf_exp(x) \ ({ \ diff --git a/include/asm-ia64/hw_irq.h b/include/asm-ia64/hw_irq.h index 041ab8c51a6..cd4e06b74ab 100644 --- a/include/asm-ia64/hw_irq.h +++ b/include/asm-ia64/hw_irq.h @@ -81,6 +81,7 @@ extern __u8 isa_irq_to_vector_map[16]; extern struct hw_interrupt_type irq_type_ia64_lsapic; /* CPU-internal interrupt controller */ +extern int assign_irq_vector_nopanic (int irq); /* allocate a free vector without panic */ extern int assign_irq_vector (int irq); /* allocate a free vector */ extern void free_irq_vector (int vector); extern void ia64_send_ipi (int cpu, int vector, int delivery_mode, int redirect); diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h index 5dd477ffb88..2303a10ee59 100644 --- a/include/asm-ia64/pal.h +++ b/include/asm-ia64/pal.h @@ -67,6 +67,7 @@ #define PAL_REGISTER_INFO 39 /* return AR and CR register information*/ #define PAL_SHUTDOWN 40 /* enter processor shutdown state */ #define PAL_PREFETCH_VISIBILITY 41 /* Make Processor Prefetches Visible */ +#define PAL_LOGICAL_TO_PHYSICAL 42 /* returns information on logical to physical processor mapping */ #define PAL_COPY_PAL 256 /* relocate PAL procedures and PAL PMI */ #define PAL_HALT_INFO 257 /* return the low power capabilities of processor */ @@ -1559,6 +1560,73 @@ ia64_pal_prefetch_visibility (s64 trans_type) return iprv.status; } +/* data structure for getting information on logical to physical mappings */ +typedef union pal_log_overview_u { + struct { + u64 num_log :16, /* Total number of logical + * processors on this die + */ + tpc :8, /* Threads per core */ + reserved3 :8, /* Reserved */ + cpp :8, /* Cores per processor */ + reserved2 :8, /* Reserved */ + ppid :8, /* Physical processor ID */ + reserved1 :8; /* Reserved */ + } overview_bits; + u64 overview_data; +} pal_log_overview_t; + +typedef union pal_proc_n_log_info1_u{ + struct { + u64 tid :16, /* Thread id */ + reserved2 :16, /* Reserved */ + cid :16, /* Core id */ + reserved1 :16; /* Reserved */ + } ppli1_bits; + u64 ppli1_data; +} pal_proc_n_log_info1_t; + +typedef union pal_proc_n_log_info2_u { + struct { + u64 la :16, /* Logical address */ + reserved :48; /* Reserved */ + } ppli2_bits; + u64 ppli2_data; +} pal_proc_n_log_info2_t; + +typedef struct pal_logical_to_physical_s +{ + pal_log_overview_t overview; + pal_proc_n_log_info1_t ppli1; + pal_proc_n_log_info2_t ppli2; +} pal_logical_to_physical_t; + +#define overview_num_log overview.overview_bits.num_log +#define overview_tpc overview.overview_bits.tpc +#define overview_cpp overview.overview_bits.cpp +#define overview_ppid overview.overview_bits.ppid +#define log1_tid ppli1.ppli1_bits.tid +#define log1_cid ppli1.ppli1_bits.cid +#define log2_la ppli2.ppli2_bits.la + +/* Get information on logical to physical processor mappings. */ +static inline s64 +ia64_pal_logical_to_phys(u64 proc_number, pal_logical_to_physical_t *mapping) +{ + struct ia64_pal_retval iprv; + + PAL_CALL(iprv, PAL_LOGICAL_TO_PHYSICAL, proc_number, 0, 0); + + if (iprv.status == PAL_STATUS_SUCCESS) + { + if (proc_number == 0) + mapping->overview.overview_data = iprv.v0; + mapping->ppli1.ppli1_data = iprv.v1; + mapping->ppli2.ppli2_data = iprv.v2; + } + + return iprv.status; +} #endif /* __ASSEMBLY__ */ #endif /* _ASM_IA64_PAL_H */ diff --git a/include/asm-ia64/perfmon.h b/include/asm-ia64/perfmon.h index 136c60e6bfc..ed5416c5b1a 100644 --- a/include/asm-ia64/perfmon.h +++ b/include/asm-ia64/perfmon.h @@ -254,6 +254,18 @@ extern int pfm_mod_write_dbrs(struct task_struct *task, void *req, unsigned int #define PFM_CPUINFO_DCR_PP 0x2 /* if set the system wide session has started */ #define PFM_CPUINFO_EXCL_IDLE 0x4 /* the system wide session excludes the idle task */ +/* + * sysctl control structure. visible to sampling formats + */ +typedef struct { + int debug; /* turn on/off debugging via syslog */ + int debug_ovfl; /* turn on/off debug printk in overflow handler */ + int fastctxsw; /* turn on/off fast (unsecure) ctxsw */ + int expert_mode; /* turn on/off value checking */ +} pfm_sysctl_t; +extern pfm_sysctl_t pfm_sysctl; + + #endif /* __KERNEL__ */ #endif /* _ASM_IA64_PERFMON_H */ diff --git a/include/asm-ia64/pgalloc.h b/include/asm-ia64/pgalloc.h index 0f05dc8bd46..a5f214554af 100644 --- a/include/asm-ia64/pgalloc.h +++ b/include/asm-ia64/pgalloc.h @@ -22,146 +22,124 @@ #include <asm/mmu_context.h> -/* - * Very stupidly, we used to get new pgd's and pmd's, init their contents - * to point to the NULL versions of the next level page table, later on - * completely re-init them the same way, then free them up. This wasted - * a lot of work and caused unnecessary memory traffic. How broken... - * We fix this by caching them. - */ -#define pgd_quicklist (local_cpu_data->pgd_quick) -#define pmd_quicklist (local_cpu_data->pmd_quick) -#define pgtable_cache_size (local_cpu_data->pgtable_cache_sz) +DECLARE_PER_CPU(unsigned long *, __pgtable_quicklist); +#define pgtable_quicklist __ia64_per_cpu_var(__pgtable_quicklist) +DECLARE_PER_CPU(long, __pgtable_quicklist_size); +#define pgtable_quicklist_size __ia64_per_cpu_var(__pgtable_quicklist_size) -static inline pgd_t* -pgd_alloc_one_fast (struct mm_struct *mm) +static inline long pgtable_quicklist_total_size(void) +{ + long ql_size = 0; + int cpuid; + + for_each_online_cpu(cpuid) { + ql_size += per_cpu(__pgtable_quicklist_size, cpuid); + } + return ql_size; +} + +static inline void *pgtable_quicklist_alloc(void) { unsigned long *ret = NULL; preempt_disable(); - ret = pgd_quicklist; + ret = pgtable_quicklist; if (likely(ret != NULL)) { - pgd_quicklist = (unsigned long *)(*ret); + pgtable_quicklist = (unsigned long *)(*ret); ret[0] = 0; - --pgtable_cache_size; - } else - ret = NULL; - - preempt_enable(); + --pgtable_quicklist_size; + preempt_enable(); + } else { + preempt_enable(); + ret = (unsigned long *)__get_free_page(GFP_KERNEL | __GFP_ZERO); + } - return (pgd_t *) ret; + return ret; } -static inline pgd_t* -pgd_alloc (struct mm_struct *mm) +static inline void pgtable_quicklist_free(void *pgtable_entry) { - /* the VM system never calls pgd_alloc_one_fast(), so we do it here. */ - pgd_t *pgd = pgd_alloc_one_fast(mm); +#ifdef CONFIG_NUMA + unsigned long nid = page_to_nid(virt_to_page(pgtable_entry)); - if (unlikely(pgd == NULL)) { - pgd = (pgd_t *)__get_free_page(GFP_KERNEL|__GFP_ZERO); + if (unlikely(nid != numa_node_id())) { + free_page((unsigned long)pgtable_entry); + return; } - return pgd; -} +#endif -static inline void -pgd_free (pgd_t *pgd) -{ preempt_disable(); - *(unsigned long *)pgd = (unsigned long) pgd_quicklist; - pgd_quicklist = (unsigned long *) pgd; - ++pgtable_cache_size; + *(unsigned long *)pgtable_entry = (unsigned long)pgtable_quicklist; + pgtable_quicklist = (unsigned long *)pgtable_entry; + ++pgtable_quicklist_size; preempt_enable(); } -static inline void -pud_populate (struct mm_struct *mm, pud_t *pud_entry, pmd_t *pmd) +static inline pgd_t *pgd_alloc(struct mm_struct *mm) { - pud_val(*pud_entry) = __pa(pmd); + return pgtable_quicklist_alloc(); } -static inline pmd_t* -pmd_alloc_one_fast (struct mm_struct *mm, unsigned long addr) +static inline void pgd_free(pgd_t * pgd) { - unsigned long *ret = NULL; - - preempt_disable(); - - ret = (unsigned long *)pmd_quicklist; - if (likely(ret != NULL)) { - pmd_quicklist = (unsigned long *)(*ret); - ret[0] = 0; - --pgtable_cache_size; - } - - preempt_enable(); - - return (pmd_t *)ret; + pgtable_quicklist_free(pgd); } -static inline pmd_t* -pmd_alloc_one (struct mm_struct *mm, unsigned long addr) +static inline void +pud_populate(struct mm_struct *mm, pud_t * pud_entry, pmd_t * pmd) { - pmd_t *pmd = (pmd_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO); + pud_val(*pud_entry) = __pa(pmd); +} - return pmd; +static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr) +{ + return pgtable_quicklist_alloc(); } -static inline void -pmd_free (pmd_t *pmd) +static inline void pmd_free(pmd_t * pmd) { - preempt_disable(); - *(unsigned long *)pmd = (unsigned long) pmd_quicklist; - pmd_quicklist = (unsigned long *) pmd; - ++pgtable_cache_size; - preempt_enable(); + pgtable_quicklist_free(pmd); } #define __pmd_free_tlb(tlb, pmd) pmd_free(pmd) static inline void -pmd_populate (struct mm_struct *mm, pmd_t *pmd_entry, struct page *pte) +pmd_populate(struct mm_struct *mm, pmd_t * pmd_entry, struct page *pte) { pmd_val(*pmd_entry) = page_to_phys(pte); } static inline void -pmd_populate_kernel (struct mm_struct *mm, pmd_t *pmd_entry, pte_t *pte) +pmd_populate_kernel(struct mm_struct *mm, pmd_t * pmd_entry, pte_t * pte) { pmd_val(*pmd_entry) = __pa(pte); } -static inline struct page * -pte_alloc_one (struct mm_struct *mm, unsigned long addr) +static inline struct page *pte_alloc_one(struct mm_struct *mm, + unsigned long addr) { - struct page *pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0); - - return pte; + return virt_to_page(pgtable_quicklist_alloc()); } -static inline pte_t * -pte_alloc_one_kernel (struct mm_struct *mm, unsigned long addr) +static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, + unsigned long addr) { - pte_t *pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO); - - return pte; + return pgtable_quicklist_alloc(); } -static inline void -pte_free (struct page *pte) +static inline void pte_free(struct page *pte) { - __free_page(pte); + pgtable_quicklist_free(page_address(pte)); } -static inline void -pte_free_kernel (pte_t *pte) +static inline void pte_free_kernel(pte_t * pte) { - free_page((unsigned long) pte); + pgtable_quicklist_free(pte); } -#define __pte_free_tlb(tlb, pte) tlb_remove_page((tlb), (pte)) +#define __pte_free_tlb(tlb, pte) pte_free(pte) -extern void check_pgt_cache (void); +extern void check_pgt_cache(void); -#endif /* _ASM_IA64_PGALLOC_H */ +#endif /* _ASM_IA64_PGALLOC_H */ diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h index 2807f8d766d..9e1ba8b7fb6 100644 --- a/include/asm-ia64/processor.h +++ b/include/asm-ia64/processor.h @@ -137,9 +137,6 @@ struct cpuinfo_ia64 { __u64 nsec_per_cyc; /* (1000000000<<IA64_NSEC_PER_CYC_SHIFT)/itc_freq */ __u64 unimpl_va_mask; /* mask of unimplemented virtual address bits (from PAL) */ __u64 unimpl_pa_mask; /* mask of unimplemented physical address bits (from PAL) */ - __u64 *pgd_quick; - __u64 *pmd_quick; - __u64 pgtable_cache_sz; __u64 itc_freq; /* frequency of ITC counter */ __u64 proc_freq; /* frequency of processor */ __u64 cyc_per_usec; /* itc_freq/1000000 */ @@ -151,6 +148,13 @@ struct cpuinfo_ia64 { #ifdef CONFIG_SMP __u64 loops_per_jiffy; int cpu; + __u32 socket_id; /* physical processor socket id */ + __u16 core_id; /* core id */ + __u16 thread_id; /* thread id */ + __u16 num_log; /* Total number of logical processors on + * this socket that were successfully booted */ + __u8 cores_per_socket; /* Cores per processor socket */ + __u8 threads_per_core; /* Threads per core */ #endif /* CPUID-derived information: */ diff --git a/include/asm-ia64/sal.h b/include/asm-ia64/sal.h index ea1ed377de4..29df88bdd2b 100644 --- a/include/asm-ia64/sal.h +++ b/include/asm-ia64/sal.h @@ -91,6 +91,7 @@ extern spinlock_t sal_lock; #define SAL_PCI_CONFIG_READ 0x01000010 #define SAL_PCI_CONFIG_WRITE 0x01000011 #define SAL_FREQ_BASE 0x01000012 +#define SAL_PHYSICAL_ID_INFO 0x01000013 #define SAL_UPDATE_PAL 0x01000020 @@ -815,6 +816,17 @@ ia64_sal_update_pal (u64 param_buf, u64 scratch_buf, u64 scratch_buf_size, return isrv.status; } +/* Get physical processor die mapping in the platform. */ +static inline s64 +ia64_sal_physical_id_info(u16 *splid) +{ + struct ia64_sal_retval isrv; + SAL_CALL(isrv, SAL_PHYSICAL_ID_INFO, 0, 0, 0, 0, 0, 0, 0); + if (splid) + *splid = isrv.v0; + return isrv.status; +} + extern unsigned long sal_platform_features; extern int (*salinfo_platform_oemdata)(const u8 *, u8 **, u64 *); @@ -832,6 +844,44 @@ extern int ia64_sal_oemcall_nolock(struct ia64_sal_retval *, u64, u64, u64, u64, u64, u64, u64, u64); extern int ia64_sal_oemcall_reentrant(struct ia64_sal_retval *, u64, u64, u64, u64, u64, u64, u64, u64); +#ifdef CONFIG_HOTPLUG_CPU +/* + * System Abstraction Layer Specification + * Section 3.2.5.1: OS_BOOT_RENDEZ to SAL return State. + * Note: region regs are stored first in head.S _start. Hence they must + * stay up front. + */ +struct sal_to_os_boot { + u64 rr[8]; /* Region Registers */ + u64 br[6]; /* br0: return addr into SAL boot rendez routine */ + u64 gr1; /* SAL:GP */ + u64 gr12; /* SAL:SP */ + u64 gr13; /* SAL: Task Pointer */ + u64 fpsr; + u64 pfs; + u64 rnat; + u64 unat; + u64 bspstore; + u64 dcr; /* Default Control Register */ + u64 iva; + u64 pta; + u64 itv; + u64 pmv; + u64 cmcv; + u64 lrr[2]; + u64 gr[4]; + u64 pr; /* Predicate registers */ + u64 lc; /* Loop Count */ + struct ia64_fpreg fp[20]; +}; + +/* + * Global array allocated for NR_CPUS at boot time + */ +extern struct sal_to_os_boot sal_boot_rendez_state[NR_CPUS]; + +extern void ia64_jump_to_sal(struct sal_to_os_boot *); +#endif extern void ia64_sal_handler_init(void *entry_point, void *gpval); diff --git a/include/asm-ia64/siginfo.h b/include/asm-ia64/siginfo.h index d55f139cbcd..9294e4b0c8b 100644 --- a/include/asm-ia64/siginfo.h +++ b/include/asm-ia64/siginfo.h @@ -8,9 +8,7 @@ * David Mosberger-Tang <davidm@hpl.hp.com>, Hewlett-Packard Co */ -#define SI_PAD_SIZE ((SI_MAX_SIZE/sizeof(int)) - 4) - -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) +#define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) #define HAVE_ARCH_SIGINFO_T #define HAVE_ARCH_COPY_SIGINFO diff --git a/include/asm-ia64/signal.h b/include/asm-ia64/signal.h index 660a759744d..608168d713d 100644 --- a/include/asm-ia64/signal.h +++ b/include/asm-ia64/signal.h @@ -114,27 +114,11 @@ #define _NSIG_BPW 64 #define _NSIG_WORDS (_NSIG / _NSIG_BPW) -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 #define SA_PERCPU_IRQ 0x02000000 #endif /* __KERNEL__ */ -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> # ifndef __ASSEMBLY__ @@ -143,9 +127,6 @@ /* Avoid too many header ordering problems. */ struct siginfo; -/* Type of a signal handler. */ -typedef void __user (*__sighandler_t)(int); - typedef struct sigaltstack { void __user *ss_sp; int ss_flags; diff --git a/include/asm-ia64/smp.h b/include/asm-ia64/smp.h index c4a227acfeb..3ba1a061e4a 100644 --- a/include/asm-ia64/smp.h +++ b/include/asm-ia64/smp.h @@ -56,6 +56,10 @@ extern struct smp_boot_data { extern char no_int_routing __devinitdata; extern cpumask_t cpu_online_map; +extern cpumask_t cpu_core_map[NR_CPUS]; +extern cpumask_t cpu_sibling_map[NR_CPUS]; +extern int smp_num_siblings; +extern int smp_num_cpucores; extern void __iomem *ipi_base_addr; extern unsigned char smp_int_redirect; @@ -124,6 +128,7 @@ extern int smp_call_function_single (int cpuid, void (*func) (void *info), void extern void smp_send_reschedule (int cpu); extern void lock_ipi_calllock(void); extern void unlock_ipi_calllock(void); +extern void identify_siblings (struct cpuinfo_ia64 *); #else diff --git a/include/asm-ia64/sn/addrs.h b/include/asm-ia64/sn/addrs.h index c916bd22767..1bfdfb4d7b0 100644 --- a/include/asm-ia64/sn/addrs.h +++ b/include/asm-ia64/sn/addrs.h @@ -136,6 +136,7 @@ */ #define CAC_BASE (CACHED | AS_CAC_SPACE) #define AMO_BASE (UNCACHED | AS_AMO_SPACE) +#define AMO_PHYS_BASE (UNCACHED_PHYS | AS_AMO_SPACE) #define GET_BASE (CACHED | AS_GET_SPACE) /* @@ -154,12 +155,20 @@ * the chiplet id is zero. If we implement TIO-TIO dma, we might need * to insert a chiplet id into this macro. However, it is our belief * right now that this chiplet id will be ICE, which is also zero. + * Nasid starts on bit 40. */ -#define PHYS_TO_TIODMA(x) ( (((u64)(x) & NASID_MASK) << 2) | NODE_OFFSET(x)) +#define PHYS_TO_TIODMA(x) ( (((u64)(NASID_GET(x))) << 40) | NODE_OFFSET(x)) #define PHYS_TO_DMA(x) ( (((u64)(x) & NASID_MASK) >> 2) | NODE_OFFSET(x)) /* + * Macros to test for address type. + */ +#define IS_AMO_ADDRESS(x) (((u64)(x) & (REGION_BITS | AS_MASK)) == AMO_BASE) +#define IS_AMO_PHYS_ADDRESS(x) (((u64)(x) & (REGION_BITS | AS_MASK)) == AMO_PHYS_BASE) + + +/* * The following definitions pertain to the IO special address * space. They define the location of the big and little windows * of any given node. @@ -168,7 +177,10 @@ #define TIO_BWIN_SIZE_BITS 30 /* big window size: 1G */ #define NODE_SWIN_BASE(n, w) ((w == 0) ? NODE_BWIN_BASE((n), SWIN0_BIGWIN) \ : RAW_NODE_SWIN_BASE(n, w)) +#define TIO_SWIN_BASE(n, w) (TIO_IO_BASE(n) + \ + ((u64) (w) << TIO_SWIN_SIZE_BITS)) #define NODE_IO_BASE(n) (GLOBAL_MMR_SPACE | NASID_SPACE(n)) +#define TIO_IO_BASE(n) (UNCACHED | NASID_SPACE(n)) #define BWIN_SIZE (1UL << BWIN_SIZE_BITS) #define NODE_BWIN_BASE0(n) (NODE_IO_BASE(n) + BWIN_SIZE) #define NODE_BWIN_BASE(n, w) (NODE_BWIN_BASE0(n) + ((u64) (w) << BWIN_SIZE_BITS)) diff --git a/include/asm-ia64/sn/arch.h b/include/asm-ia64/sn/arch.h index 7c349f07916..635fdce854a 100644 --- a/include/asm-ia64/sn/arch.h +++ b/include/asm-ia64/sn/arch.h @@ -5,7 +5,7 @@ * * SGI specific setup. * - * Copyright (C) 1995-1997,1999,2001-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1995-1997,1999,2001-2005 Silicon Graphics, Inc. All rights reserved. * Copyright (C) 1999 Ralf Baechle (ralf@gnu.org) */ #ifndef _ASM_IA64_SN_ARCH_H @@ -47,6 +47,21 @@ DECLARE_PER_CPU(struct sn_hub_info_s, __sn_hub_info); #define MAX_COMPACT_NODES 2048 #define CPUS_PER_NODE 4 + +/* + * Compact node ID to nasid mappings kept in the per-cpu data areas of each + * cpu. + */ +DECLARE_PER_CPU(short, __sn_cnodeid_to_nasid[MAX_NUMNODES]); +#define sn_cnodeid_to_nasid (&__get_cpu_var(__sn_cnodeid_to_nasid[0])) + + + +extern u8 sn_partition_id; +extern u8 sn_system_size; +extern u8 sn_sharing_domain_size; +extern u8 sn_region_size; + extern void sn_flush_all_caches(long addr, long bytes); #endif /* _ASM_IA64_SN_ARCH_H */ diff --git a/include/asm-ia64/sn/bte.h b/include/asm-ia64/sn/bte.h index 0ec27f99c18..f50da3d91d0 100644 --- a/include/asm-ia64/sn/bte.h +++ b/include/asm-ia64/sn/bte.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. */ @@ -13,8 +13,12 @@ #include <linux/timer.h> #include <linux/spinlock.h> #include <linux/cache.h> +#include <asm/sn/pda.h> #include <asm/sn/types.h> +#include <asm/sn/shub_mmr.h> +#define IBCT_NOTIFY (0x1UL << 4) +#define IBCT_ZFIL_MODE (0x1UL << 0) /* #define BTE_DEBUG */ /* #define BTE_DEBUG_VERBOSE */ @@ -39,8 +43,36 @@ /* Define hardware */ -#define BTES_PER_NODE 2 +#define BTES_PER_NODE (is_shub2() ? 4 : 2) +#define MAX_BTES_PER_NODE 4 +#define BTE2OFF_CTRL (0) +#define BTE2OFF_SRC (SH2_BT_ENG_SRC_ADDR_0 - SH2_BT_ENG_CSR_0) +#define BTE2OFF_DEST (SH2_BT_ENG_DEST_ADDR_0 - SH2_BT_ENG_CSR_0) +#define BTE2OFF_NOTIFY (SH2_BT_ENG_NOTIF_ADDR_0 - SH2_BT_ENG_CSR_0) + +#define BTE_BASE_ADDR(interface) \ + (is_shub2() ? (interface == 0) ? SH2_BT_ENG_CSR_0 : \ + (interface == 1) ? SH2_BT_ENG_CSR_1 : \ + (interface == 2) ? SH2_BT_ENG_CSR_2 : \ + SH2_BT_ENG_CSR_3 \ + : (interface == 0) ? IIO_IBLS0 : IIO_IBLS1) + +#define BTE_SOURCE_ADDR(base) \ + (is_shub2() ? base + (BTE2OFF_SRC/8) \ + : base + (BTEOFF_SRC/8)) + +#define BTE_DEST_ADDR(base) \ + (is_shub2() ? base + (BTE2OFF_DEST/8) \ + : base + (BTEOFF_DEST/8)) + +#define BTE_CTRL_ADDR(base) \ + (is_shub2() ? base + (BTE2OFF_CTRL/8) \ + : base + (BTEOFF_CTRL/8)) + +#define BTE_NOTIF_ADDR(base) \ + (is_shub2() ? base + (BTE2OFF_NOTIFY/8) \ + : base + (BTEOFF_NOTIFY/8)) /* Define hardware modes */ #define BTE_NOTIFY (IBCT_NOTIFY) @@ -68,14 +100,18 @@ #define BTE_LNSTAT_STORE(_bte, _x) \ HUB_S(_bte->bte_base_addr, (_x)) #define BTE_SRC_STORE(_bte, _x) \ - HUB_S(_bte->bte_base_addr + (BTEOFF_SRC/8), (_x)) + HUB_S(_bte->bte_source_addr, (_x)) #define BTE_DEST_STORE(_bte, _x) \ - HUB_S(_bte->bte_base_addr + (BTEOFF_DEST/8), (_x)) + HUB_S(_bte->bte_destination_addr, (_x)) #define BTE_CTRL_STORE(_bte, _x) \ - HUB_S(_bte->bte_base_addr + (BTEOFF_CTRL/8), (_x)) + HUB_S(_bte->bte_control_addr, (_x)) #define BTE_NOTIF_STORE(_bte, _x) \ - HUB_S(_bte->bte_base_addr + (BTEOFF_NOTIFY/8), (_x)) + HUB_S(_bte->bte_notify_addr, (_x)) +#define BTE_START_TRANSFER(_bte, _len, _mode) \ + is_shub2() ? BTE_CTRL_STORE(_bte, IBLS_BUSY | (_mode << 24) | _len) \ + : BTE_LNSTAT_STORE(_bte, _len); \ + BTE_CTRL_STORE(_bte, _mode) /* Possible results from bte_copy and bte_unaligned_copy */ /* The following error codes map into the BTE hardware codes @@ -110,6 +146,10 @@ typedef enum { struct bteinfo_s { volatile u64 notify ____cacheline_aligned; u64 *bte_base_addr ____cacheline_aligned; + u64 *bte_source_addr; + u64 *bte_destination_addr; + u64 *bte_control_addr; + u64 *bte_notify_addr; spinlock_t spinlock; cnodeid_t bte_cnode; /* cnode */ int bte_error_count; /* Number of errors encountered */ @@ -117,6 +157,7 @@ struct bteinfo_s { int cleanup_active; /* Interface is locked for cleanup */ volatile bte_result_t bh_error; /* error while processing */ volatile u64 *most_rcnt_na; + struct bteinfo_s *btes_to_try[MAX_BTES_PER_NODE]; }; diff --git a/include/asm-ia64/sn/fetchop.h b/include/asm-ia64/sn/fetchop.h deleted file mode 100644 index 5f4ad8f4b5d..00000000000 --- a/include/asm-ia64/sn/fetchop.h +++ /dev/null @@ -1,85 +0,0 @@ -/* - * - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (c) 2001-2004 Silicon Graphics, Inc. All rights reserved. - */ - -#ifndef _ASM_IA64_SN_FETCHOP_H -#define _ASM_IA64_SN_FETCHOP_H - -#include <linux/config.h> - -#define FETCHOP_BASENAME "sgi_fetchop" -#define FETCHOP_FULLNAME "/dev/sgi_fetchop" - - - -#define FETCHOP_VAR_SIZE 64 /* 64 byte per fetchop variable */ - -#define FETCHOP_LOAD 0 -#define FETCHOP_INCREMENT 8 -#define FETCHOP_DECREMENT 16 -#define FETCHOP_CLEAR 24 - -#define FETCHOP_STORE 0 -#define FETCHOP_AND 24 -#define FETCHOP_OR 32 - -#define FETCHOP_CLEAR_CACHE 56 - -#define FETCHOP_LOAD_OP(addr, op) ( \ - *(volatile long *)((char*) (addr) + (op))) - -#define FETCHOP_STORE_OP(addr, op, x) ( \ - *(volatile long *)((char*) (addr) + (op)) = (long) (x)) - -#ifdef __KERNEL__ - -/* - * Convert a region 6 (kaddr) address to the address of the fetchop variable - */ -#define FETCHOP_KADDR_TO_MSPEC_ADDR(kaddr) TO_MSPEC(kaddr) - - -/* - * Each Atomic Memory Operation (AMO formerly known as fetchop) - * variable is 64 bytes long. The first 8 bytes are used. The - * remaining 56 bytes are unaddressable due to the operation taking - * that portion of the address. - * - * NOTE: The AMO_t _MUST_ be placed in either the first or second half - * of the cache line. The cache line _MUST NOT_ be used for anything - * other than additional AMO_t entries. This is because there are two - * addresses which reference the same physical cache line. One will - * be a cached entry with the memory type bits all set. This address - * may be loaded into processor cache. The AMO_t will be referenced - * uncached via the memory special memory type. If any portion of the - * cached cache-line is modified, when that line is flushed, it will - * overwrite the uncached value in physical memory and lead to - * inconsistency. - */ -typedef struct { - u64 variable; - u64 unused[7]; -} AMO_t; - - -/* - * The following APIs are externalized to the kernel to allocate/free pages of - * fetchop variables. - * fetchop_kalloc_page - Allocate/initialize 1 fetchop page on the - * specified cnode. - * fetchop_kfree_page - Free a previously allocated fetchop page - */ - -unsigned long fetchop_kalloc_page(int nid); -void fetchop_kfree_page(unsigned long maddr); - - -#endif /* __KERNEL__ */ - -#endif /* _ASM_IA64_SN_FETCHOP_H */ - diff --git a/include/asm-ia64/sn/geo.h b/include/asm-ia64/sn/geo.h index f566343d25f..84b254603b8 100644 --- a/include/asm-ia64/sn/geo.h +++ b/include/asm-ia64/sn/geo.h @@ -18,32 +18,34 @@ #define GEOID_SIZE 8 /* Would 16 be better? The size can be different on different platforms. */ -#define MAX_SLABS 0xe /* slabs per module */ +#define MAX_SLOTS 0xf /* slots per module */ +#define MAX_SLABS 0xf /* slabs per slot */ typedef unsigned char geo_type_t; /* Fields common to all substructures */ -typedef struct geo_any_s { +typedef struct geo_common_s { moduleid_t module; /* The module (box) this h/w lives in */ geo_type_t type; /* What type of h/w is named by this geoid_t */ - slabid_t slab; /* The logical assembly within the module */ -} geo_any_t; + slabid_t slab:4; /* slab (ASIC), 0 .. 15 within slot */ + slotid_t slot:4; /* slot (Blade), 0 .. 15 within module */ +} geo_common_t; /* Additional fields for particular types of hardware */ typedef struct geo_node_s { - geo_any_t any; /* No additional fields needed */ + geo_common_t common; /* No additional fields needed */ } geo_node_t; typedef struct geo_rtr_s { - geo_any_t any; /* No additional fields needed */ + geo_common_t common; /* No additional fields needed */ } geo_rtr_t; typedef struct geo_iocntl_s { - geo_any_t any; /* No additional fields needed */ + geo_common_t common; /* No additional fields needed */ } geo_iocntl_t; typedef struct geo_pcicard_s { - geo_iocntl_t any; + geo_iocntl_t common; char bus; /* Bus/widget number */ char slot; /* PCI slot number */ } geo_pcicard_t; @@ -62,14 +64,14 @@ typedef struct geo_mem_s { typedef union geoid_u { - geo_any_t any; - geo_node_t node; + geo_common_t common; + geo_node_t node; geo_iocntl_t iocntl; geo_pcicard_t pcicard; - geo_rtr_t rtr; - geo_cpu_t cpu; - geo_mem_t mem; - char padsize[GEOID_SIZE]; + geo_rtr_t rtr; + geo_cpu_t cpu; + geo_mem_t mem; + char padsize[GEOID_SIZE]; } geoid_t; @@ -104,19 +106,26 @@ typedef union geoid_u { #define INVALID_CNODEID ((cnodeid_t)-1) #define INVALID_PNODEID ((pnodeid_t)-1) #define INVALID_SLAB (slabid_t)-1 +#define INVALID_SLOT (slotid_t)-1 #define INVALID_MODULE ((moduleid_t)-1) #define INVALID_PARTID ((partid_t)-1) static inline slabid_t geo_slab(geoid_t g) { - return (g.any.type == GEO_TYPE_INVALID) ? - INVALID_SLAB : g.any.slab; + return (g.common.type == GEO_TYPE_INVALID) ? + INVALID_SLAB : g.common.slab; +} + +static inline slotid_t geo_slot(geoid_t g) +{ + return (g.common.type == GEO_TYPE_INVALID) ? + INVALID_SLOT : g.common.slot; } static inline moduleid_t geo_module(geoid_t g) { - return (g.any.type == GEO_TYPE_INVALID) ? - INVALID_MODULE : g.any.module; + return (g.common.type == GEO_TYPE_INVALID) ? + INVALID_MODULE : g.common.module; } extern geoid_t cnodeid_get_geoid(cnodeid_t cnode); diff --git a/include/asm-ia64/sn/l1.h b/include/asm-ia64/sn/l1.h index d5dbd55e44b..08050d37b66 100644 --- a/include/asm-ia64/sn/l1.h +++ b/include/asm-ia64/sn/l1.h @@ -29,8 +29,9 @@ #define L1_BRICKTYPE_CHI_CG 0x76 /* v */ #define L1_BRICKTYPE_X 0x78 /* x */ #define L1_BRICKTYPE_X2 0x79 /* y */ -#define L1_BRICKTYPE_SA 0x5e /* ^ */ /* TIO bringup brick */ +#define L1_BRICKTYPE_SA 0x5e /* ^ */ #define L1_BRICKTYPE_PA 0x6a /* j */ #define L1_BRICKTYPE_IA 0x6b /* k */ +#define L1_BRICKTYPE_ATHENA 0x2b /* + */ #endif /* _ASM_IA64_SN_L1_H */ diff --git a/include/asm-ia64/sn/nodepda.h b/include/asm-ia64/sn/nodepda.h index 2fbde33656e..7138b1eafd6 100644 --- a/include/asm-ia64/sn/nodepda.h +++ b/include/asm-ia64/sn/nodepda.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1992 - 1997, 2000-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1992 - 1997, 2000-2005 Silicon Graphics, Inc. All rights reserved. */ #ifndef _ASM_IA64_SN_NODEPDA_H #define _ASM_IA64_SN_NODEPDA_H @@ -13,7 +13,6 @@ #include <asm/irq.h> #include <asm/sn/arch.h> #include <asm/sn/intr.h> -#include <asm/sn/pda.h> #include <asm/sn/bte.h> /* @@ -43,7 +42,7 @@ struct nodepda_s { /* * The BTEs on this node are shared by the local cpus */ - struct bteinfo_s bte_if[BTES_PER_NODE]; /* Virtual Interface */ + struct bteinfo_s bte_if[MAX_BTES_PER_NODE]; /* Virtual Interface */ struct timer_list bte_recovery_timer; spinlock_t bte_recovery_lock; @@ -67,20 +66,18 @@ typedef struct nodepda_s nodepda_t; * The next set of definitions provides this. * Routines are expected to use * - * nodepda -> to access node PDA for the node on which code is running - * subnodepda -> to access subnode PDA for the subnode on which code is running - * - * NODEPDA(cnode) -> to access node PDA for cnodeid - * SUBNODEPDA(cnode,sn) -> to access subnode PDA for cnodeid/subnode + * sn_nodepda - to access node PDA for the node on which code is running + * NODEPDA(cnodeid) - to access node PDA for cnodeid */ -#define nodepda pda->p_nodepda /* Ptr to this node's PDA */ -#define NODEPDA(cnode) (nodepda->pernode_pdaindr[cnode]) +DECLARE_PER_CPU(struct nodepda_s *, __sn_nodepda); +#define sn_nodepda (__get_cpu_var(__sn_nodepda)) +#define NODEPDA(cnodeid) (sn_nodepda->pernode_pdaindr[cnodeid]) /* * Check if given a compact node id the corresponding node has all the * cpus disabled. */ -#define is_headless_node(cnode) (nr_cpus_node(cnode) == 0) +#define is_headless_node(cnodeid) (nr_cpus_node(cnodeid) == 0) #endif /* _ASM_IA64_SN_NODEPDA_H */ diff --git a/arch/ia64/sn/include/pci/pcibus_provider_defs.h b/include/asm-ia64/sn/pcibus_provider_defs.h index 07065615bbe..04e27d5b382 100644 --- a/arch/ia64/sn/include/pci/pcibus_provider_defs.h +++ b/include/asm-ia64/sn/pcibus_provider_defs.h @@ -17,6 +17,9 @@ #define PCIIO_ASIC_TYPE_PPB 1 #define PCIIO_ASIC_TYPE_PIC 2 #define PCIIO_ASIC_TYPE_TIOCP 3 +#define PCIIO_ASIC_TYPE_TIOCA 4 + +#define PCIIO_ASIC_MAX_TYPES 5 /* * Common pciio bus provider data. There should be one of these as the @@ -35,9 +38,15 @@ struct pcibus_bussoft { }; /* - * DMA mapping flags + * SN pci bus indirection */ -#define SN_PCIDMA_CONSISTENT 0x0001 +struct sn_pcibus_provider { + dma_addr_t (*dma_map)(struct pci_dev *, unsigned long, size_t); + dma_addr_t (*dma_map_consistent)(struct pci_dev *, unsigned long, size_t); + void (*dma_unmap)(struct pci_dev *, dma_addr_t, int); + void * (*bus_fixup)(struct pcibus_bussoft *); +}; +extern struct sn_pcibus_provider *sn_pci_provider[]; #endif /* _ASM_IA64_SN_PCI_PCIBUS_PROVIDER_H */ diff --git a/arch/ia64/sn/include/pci/pcidev.h b/include/asm-ia64/sn/pcidev.h index 81eb95d3bf4..ed4031d8081 100644 --- a/arch/ia64/sn/include/pci/pcidev.h +++ b/include/asm-ia64/sn/pcidev.h @@ -32,6 +32,9 @@ extern struct sn_irq_info **sn_irq; #define SN_PCIDEV_BUSSOFT(pci_dev) \ (SN_PCIDEV_INFO(pci_dev)->pdi_host_pcidev_info->pdi_pcibus_info) +#define SN_PCIDEV_BUSPROVIDER(pci_dev) \ + (SN_PCIDEV_INFO(pci_dev)->pdi_provider) + #define PCIIO_BUS_NONE 255 /* bus 255 reserved */ #define PCIIO_SLOT_NONE 255 #define PCIIO_FUNC_NONE 255 @@ -46,6 +49,7 @@ struct pcidev_info { struct pci_dev *pdi_linux_pcidev; /* Kernel pci_dev */ struct sn_irq_info *pdi_sn_irq_info; + struct sn_pcibus_provider *pdi_provider; /* sn pci ops */ }; extern void sn_irq_fixup(struct pci_dev *pci_dev, diff --git a/include/asm-ia64/sn/pda.h b/include/asm-ia64/sn/pda.h index e940d3647c8..ea5590c76ca 100644 --- a/include/asm-ia64/sn/pda.h +++ b/include/asm-ia64/sn/pda.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1992 - 1997, 2000-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1992 - 1997, 2000-2005 Silicon Graphics, Inc. All rights reserved. */ #ifndef _ASM_IA64_SN_PDA_H #define _ASM_IA64_SN_PDA_H @@ -11,7 +11,6 @@ #include <linux/cache.h> #include <asm/percpu.h> #include <asm/system.h> -#include <asm/sn/bte.h> /* @@ -25,14 +24,6 @@ typedef struct pda_s { - /* Having a pointer in the begining of PDA tends to increase - * the chance of having this pointer in cache. (Yes something - * else gets pushed out). Doing this reduces the number of memory - * access to all nodepda variables to be one - */ - struct nodepda_s *p_nodepda; /* Pointer to Per node PDA */ - struct subnodepda_s *p_subnodepda; /* Pointer to CPU subnode PDA */ - /* * Support for SN LEDs */ @@ -50,7 +41,6 @@ typedef struct pda_s { unsigned long sn_soft_irr[4]; unsigned long sn_in_service_ivecs[4]; - short cnodeid_to_nasid_table[MAX_NUMNODES]; int sn_lb_int_war_ticks; int sn_last_irq; int sn_first_irq; diff --git a/include/asm-ia64/sn/shub_mmr.h b/include/asm-ia64/sn/shub_mmr.h index 5c2fcf13d5c..323fa0cd8d8 100644 --- a/include/asm-ia64/sn/shub_mmr.h +++ b/include/asm-ia64/sn/shub_mmr.h @@ -4,7 +4,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (c) 2001-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (c) 2001-2005 Silicon Graphics, Inc. All rights reserved. */ #ifndef _ASM_IA64_SN_SHUB_MMR_H @@ -129,6 +129,23 @@ #define SH_EVENT_OCCURRED_II_INT1_SHFT 30 #define SH_EVENT_OCCURRED_II_INT1_MASK 0x0000000040000000 +/* SH2_EVENT_OCCURRED_EXTIO_INT2 */ +/* Description: Pending SHUB 2 EXT IO INT2 */ +#define SH2_EVENT_OCCURRED_EXTIO_INT2_SHFT 33 +#define SH2_EVENT_OCCURRED_EXTIO_INT2_MASK 0x0000000200000000 + +/* SH2_EVENT_OCCURRED_EXTIO_INT3 */ +/* Description: Pending SHUB 2 EXT IO INT3 */ +#define SH2_EVENT_OCCURRED_EXTIO_INT3_SHFT 34 +#define SH2_EVENT_OCCURRED_EXTIO_INT3_MASK 0x0000000400000000 + +#define SH_ALL_INT_MASK \ + (SH_EVENT_OCCURRED_UART_INT_MASK | SH_EVENT_OCCURRED_IPI_INT_MASK | \ + SH_EVENT_OCCURRED_II_INT0_MASK | SH_EVENT_OCCURRED_II_INT1_MASK | \ + SH_EVENT_OCCURRED_II_INT1_MASK | SH2_EVENT_OCCURRED_EXTIO_INT2_MASK | \ + SH2_EVENT_OCCURRED_EXTIO_INT3_MASK) + + /* ==================================================================== */ /* LEDS */ /* ==================================================================== */ @@ -368,6 +385,17 @@ #define SH_EVENT_OCCURRED_RTC3_INT_MASK 0x0000000004000000 /* ==================================================================== */ +/* Register "SH_IPI_ACCESS" */ +/* CPU interrupt Access Permission Bits */ +/* ==================================================================== */ + +#define SH1_IPI_ACCESS 0x0000000110060480 +#define SH2_IPI_ACCESS0 0x0000000010060c00 +#define SH2_IPI_ACCESS1 0x0000000010060c80 +#define SH2_IPI_ACCESS2 0x0000000010060d00 +#define SH2_IPI_ACCESS3 0x0000000010060d80 + +/* ==================================================================== */ /* Register "SH_INT_CMPB" */ /* RTC Compare Value for Processor B */ /* ==================================================================== */ @@ -412,6 +440,19 @@ #define SH_INT_CMPD_REAL_TIME_CMPD_SHFT 0 #define SH_INT_CMPD_REAL_TIME_CMPD_MASK 0x007fffffffffffff +/* ==================================================================== */ +/* Register "SH_MD_DQLP_MMR_DIR_PRIVEC0" */ +/* privilege vector for acc=0 */ +/* ==================================================================== */ + +#define SH1_MD_DQLP_MMR_DIR_PRIVEC0 0x0000000100030300 + +/* ==================================================================== */ +/* Register "SH_MD_DQRP_MMR_DIR_PRIVEC0" */ +/* privilege vector for acc=0 */ +/* ==================================================================== */ + +#define SH1_MD_DQRP_MMR_DIR_PRIVEC0 0x0000000100050300 /* ==================================================================== */ /* Some MMRs are functionally identical (or close enough) on both SHUB1 */ @@ -438,4 +479,22 @@ #define SH_INT_CMPC shubmmr(SH, INT_CMPC) #define SH_INT_CMPD shubmmr(SH, INT_CMPD) +/* ========================================================================== */ +/* Register "SH2_BT_ENG_CSR_0" */ +/* Engine 0 Control and Status Register */ +/* ========================================================================== */ + +#define SH2_BT_ENG_CSR_0 0x0000000030040000 +#define SH2_BT_ENG_SRC_ADDR_0 0x0000000030040080 +#define SH2_BT_ENG_DEST_ADDR_0 0x0000000030040100 +#define SH2_BT_ENG_NOTIF_ADDR_0 0x0000000030040180 + +/* ========================================================================== */ +/* BTE interfaces 1-3 */ +/* ========================================================================== */ + +#define SH2_BT_ENG_CSR_1 0x0000000030050000 +#define SH2_BT_ENG_CSR_2 0x0000000030060000 +#define SH2_BT_ENG_CSR_3 0x0000000030070000 + #endif /* _ASM_IA64_SN_SHUB_MMR_H */ diff --git a/include/asm-ia64/sn/shubio.h b/include/asm-ia64/sn/shubio.h index fbd880e6bb9..831b72111fd 100644 --- a/include/asm-ia64/sn/shubio.h +++ b/include/asm-ia64/sn/shubio.h @@ -3,292 +3,287 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1992 - 1997, 2000-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1992 - 1997, 2000-2005 Silicon Graphics, Inc. All rights reserved. */ #ifndef _ASM_IA64_SN_SHUBIO_H #define _ASM_IA64_SN_SHUBIO_H -#define HUB_WIDGET_ID_MAX 0xf -#define IIO_NUM_ITTES 7 -#define HUB_NUM_BIG_WINDOW (IIO_NUM_ITTES - 1) - -#define IIO_WID 0x00400000 /* Crosstalk Widget Identification */ - /* This register is also accessible from - * Crosstalk at address 0x0. */ -#define IIO_WSTAT 0x00400008 /* Crosstalk Widget Status */ -#define IIO_WCR 0x00400020 /* Crosstalk Widget Control Register */ -#define IIO_ILAPR 0x00400100 /* IO Local Access Protection Register */ -#define IIO_ILAPO 0x00400108 /* IO Local Access Protection Override */ -#define IIO_IOWA 0x00400110 /* IO Outbound Widget Access */ -#define IIO_IIWA 0x00400118 /* IO Inbound Widget Access */ -#define IIO_IIDEM 0x00400120 /* IO Inbound Device Error Mask */ -#define IIO_ILCSR 0x00400128 /* IO LLP Control and Status Register */ -#define IIO_ILLR 0x00400130 /* IO LLP Log Register */ -#define IIO_IIDSR 0x00400138 /* IO Interrupt Destination */ - -#define IIO_IGFX0 0x00400140 /* IO Graphics Node-Widget Map 0 */ -#define IIO_IGFX1 0x00400148 /* IO Graphics Node-Widget Map 1 */ - -#define IIO_ISCR0 0x00400150 /* IO Scratch Register 0 */ -#define IIO_ISCR1 0x00400158 /* IO Scratch Register 1 */ - -#define IIO_ITTE1 0x00400160 /* IO Translation Table Entry 1 */ -#define IIO_ITTE2 0x00400168 /* IO Translation Table Entry 2 */ -#define IIO_ITTE3 0x00400170 /* IO Translation Table Entry 3 */ -#define IIO_ITTE4 0x00400178 /* IO Translation Table Entry 4 */ -#define IIO_ITTE5 0x00400180 /* IO Translation Table Entry 5 */ -#define IIO_ITTE6 0x00400188 /* IO Translation Table Entry 6 */ -#define IIO_ITTE7 0x00400190 /* IO Translation Table Entry 7 */ - -#define IIO_IPRB0 0x00400198 /* IO PRB Entry 0 */ -#define IIO_IPRB8 0x004001A0 /* IO PRB Entry 8 */ -#define IIO_IPRB9 0x004001A8 /* IO PRB Entry 9 */ -#define IIO_IPRBA 0x004001B0 /* IO PRB Entry A */ -#define IIO_IPRBB 0x004001B8 /* IO PRB Entry B */ -#define IIO_IPRBC 0x004001C0 /* IO PRB Entry C */ -#define IIO_IPRBD 0x004001C8 /* IO PRB Entry D */ -#define IIO_IPRBE 0x004001D0 /* IO PRB Entry E */ -#define IIO_IPRBF 0x004001D8 /* IO PRB Entry F */ - -#define IIO_IXCC 0x004001E0 /* IO Crosstalk Credit Count Timeout */ -#define IIO_IMEM 0x004001E8 /* IO Miscellaneous Error Mask */ -#define IIO_IXTT 0x004001F0 /* IO Crosstalk Timeout Threshold */ -#define IIO_IECLR 0x004001F8 /* IO Error Clear Register */ -#define IIO_IBCR 0x00400200 /* IO BTE Control Register */ - -#define IIO_IXSM 0x00400208 /* IO Crosstalk Spurious Message */ -#define IIO_IXSS 0x00400210 /* IO Crosstalk Spurious Sideband */ - -#define IIO_ILCT 0x00400218 /* IO LLP Channel Test */ - -#define IIO_IIEPH1 0x00400220 /* IO Incoming Error Packet Header, Part 1 */ -#define IIO_IIEPH2 0x00400228 /* IO Incoming Error Packet Header, Part 2 */ - - -#define IIO_ISLAPR 0x00400230 /* IO SXB Local Access Protection Regster */ -#define IIO_ISLAPO 0x00400238 /* IO SXB Local Access Protection Override */ - -#define IIO_IWI 0x00400240 /* IO Wrapper Interrupt Register */ -#define IIO_IWEL 0x00400248 /* IO Wrapper Error Log Register */ -#define IIO_IWC 0x00400250 /* IO Wrapper Control Register */ -#define IIO_IWS 0x00400258 /* IO Wrapper Status Register */ -#define IIO_IWEIM 0x00400260 /* IO Wrapper Error Interrupt Masking Register */ - -#define IIO_IPCA 0x00400300 /* IO PRB Counter Adjust */ - -#define IIO_IPRTE0_A 0x00400308 /* IO PIO Read Address Table Entry 0, Part A */ -#define IIO_IPRTE1_A 0x00400310 /* IO PIO Read Address Table Entry 1, Part A */ -#define IIO_IPRTE2_A 0x00400318 /* IO PIO Read Address Table Entry 2, Part A */ -#define IIO_IPRTE3_A 0x00400320 /* IO PIO Read Address Table Entry 3, Part A */ -#define IIO_IPRTE4_A 0x00400328 /* IO PIO Read Address Table Entry 4, Part A */ -#define IIO_IPRTE5_A 0x00400330 /* IO PIO Read Address Table Entry 5, Part A */ -#define IIO_IPRTE6_A 0x00400338 /* IO PIO Read Address Table Entry 6, Part A */ -#define IIO_IPRTE7_A 0x00400340 /* IO PIO Read Address Table Entry 7, Part A */ - -#define IIO_IPRTE0_B 0x00400348 /* IO PIO Read Address Table Entry 0, Part B */ -#define IIO_IPRTE1_B 0x00400350 /* IO PIO Read Address Table Entry 1, Part B */ -#define IIO_IPRTE2_B 0x00400358 /* IO PIO Read Address Table Entry 2, Part B */ -#define IIO_IPRTE3_B 0x00400360 /* IO PIO Read Address Table Entry 3, Part B */ -#define IIO_IPRTE4_B 0x00400368 /* IO PIO Read Address Table Entry 4, Part B */ -#define IIO_IPRTE5_B 0x00400370 /* IO PIO Read Address Table Entry 5, Part B */ -#define IIO_IPRTE6_B 0x00400378 /* IO PIO Read Address Table Entry 6, Part B */ -#define IIO_IPRTE7_B 0x00400380 /* IO PIO Read Address Table Entry 7, Part B */ - -#define IIO_IPDR 0x00400388 /* IO PIO Deallocation Register */ -#define IIO_ICDR 0x00400390 /* IO CRB Entry Deallocation Register */ -#define IIO_IFDR 0x00400398 /* IO IOQ FIFO Depth Register */ -#define IIO_IIAP 0x004003A0 /* IO IIQ Arbitration Parameters */ -#define IIO_ICMR 0x004003A8 /* IO CRB Management Register */ -#define IIO_ICCR 0x004003B0 /* IO CRB Control Register */ -#define IIO_ICTO 0x004003B8 /* IO CRB Timeout */ -#define IIO_ICTP 0x004003C0 /* IO CRB Timeout Prescalar */ - -#define IIO_ICRB0_A 0x00400400 /* IO CRB Entry 0_A */ -#define IIO_ICRB0_B 0x00400408 /* IO CRB Entry 0_B */ -#define IIO_ICRB0_C 0x00400410 /* IO CRB Entry 0_C */ -#define IIO_ICRB0_D 0x00400418 /* IO CRB Entry 0_D */ -#define IIO_ICRB0_E 0x00400420 /* IO CRB Entry 0_E */ - -#define IIO_ICRB1_A 0x00400430 /* IO CRB Entry 1_A */ -#define IIO_ICRB1_B 0x00400438 /* IO CRB Entry 1_B */ -#define IIO_ICRB1_C 0x00400440 /* IO CRB Entry 1_C */ -#define IIO_ICRB1_D 0x00400448 /* IO CRB Entry 1_D */ -#define IIO_ICRB1_E 0x00400450 /* IO CRB Entry 1_E */ - -#define IIO_ICRB2_A 0x00400460 /* IO CRB Entry 2_A */ -#define IIO_ICRB2_B 0x00400468 /* IO CRB Entry 2_B */ -#define IIO_ICRB2_C 0x00400470 /* IO CRB Entry 2_C */ -#define IIO_ICRB2_D 0x00400478 /* IO CRB Entry 2_D */ -#define IIO_ICRB2_E 0x00400480 /* IO CRB Entry 2_E */ - -#define IIO_ICRB3_A 0x00400490 /* IO CRB Entry 3_A */ -#define IIO_ICRB3_B 0x00400498 /* IO CRB Entry 3_B */ -#define IIO_ICRB3_C 0x004004a0 /* IO CRB Entry 3_C */ -#define IIO_ICRB3_D 0x004004a8 /* IO CRB Entry 3_D */ -#define IIO_ICRB3_E 0x004004b0 /* IO CRB Entry 3_E */ - -#define IIO_ICRB4_A 0x004004c0 /* IO CRB Entry 4_A */ -#define IIO_ICRB4_B 0x004004c8 /* IO CRB Entry 4_B */ -#define IIO_ICRB4_C 0x004004d0 /* IO CRB Entry 4_C */ -#define IIO_ICRB4_D 0x004004d8 /* IO CRB Entry 4_D */ -#define IIO_ICRB4_E 0x004004e0 /* IO CRB Entry 4_E */ - -#define IIO_ICRB5_A 0x004004f0 /* IO CRB Entry 5_A */ -#define IIO_ICRB5_B 0x004004f8 /* IO CRB Entry 5_B */ -#define IIO_ICRB5_C 0x00400500 /* IO CRB Entry 5_C */ -#define IIO_ICRB5_D 0x00400508 /* IO CRB Entry 5_D */ -#define IIO_ICRB5_E 0x00400510 /* IO CRB Entry 5_E */ - -#define IIO_ICRB6_A 0x00400520 /* IO CRB Entry 6_A */ -#define IIO_ICRB6_B 0x00400528 /* IO CRB Entry 6_B */ -#define IIO_ICRB6_C 0x00400530 /* IO CRB Entry 6_C */ -#define IIO_ICRB6_D 0x00400538 /* IO CRB Entry 6_D */ -#define IIO_ICRB6_E 0x00400540 /* IO CRB Entry 6_E */ - -#define IIO_ICRB7_A 0x00400550 /* IO CRB Entry 7_A */ -#define IIO_ICRB7_B 0x00400558 /* IO CRB Entry 7_B */ -#define IIO_ICRB7_C 0x00400560 /* IO CRB Entry 7_C */ -#define IIO_ICRB7_D 0x00400568 /* IO CRB Entry 7_D */ -#define IIO_ICRB7_E 0x00400570 /* IO CRB Entry 7_E */ - -#define IIO_ICRB8_A 0x00400580 /* IO CRB Entry 8_A */ -#define IIO_ICRB8_B 0x00400588 /* IO CRB Entry 8_B */ -#define IIO_ICRB8_C 0x00400590 /* IO CRB Entry 8_C */ -#define IIO_ICRB8_D 0x00400598 /* IO CRB Entry 8_D */ -#define IIO_ICRB8_E 0x004005a0 /* IO CRB Entry 8_E */ - -#define IIO_ICRB9_A 0x004005b0 /* IO CRB Entry 9_A */ -#define IIO_ICRB9_B 0x004005b8 /* IO CRB Entry 9_B */ -#define IIO_ICRB9_C 0x004005c0 /* IO CRB Entry 9_C */ -#define IIO_ICRB9_D 0x004005c8 /* IO CRB Entry 9_D */ -#define IIO_ICRB9_E 0x004005d0 /* IO CRB Entry 9_E */ - -#define IIO_ICRBA_A 0x004005e0 /* IO CRB Entry A_A */ -#define IIO_ICRBA_B 0x004005e8 /* IO CRB Entry A_B */ -#define IIO_ICRBA_C 0x004005f0 /* IO CRB Entry A_C */ -#define IIO_ICRBA_D 0x004005f8 /* IO CRB Entry A_D */ -#define IIO_ICRBA_E 0x00400600 /* IO CRB Entry A_E */ - -#define IIO_ICRBB_A 0x00400610 /* IO CRB Entry B_A */ -#define IIO_ICRBB_B 0x00400618 /* IO CRB Entry B_B */ -#define IIO_ICRBB_C 0x00400620 /* IO CRB Entry B_C */ -#define IIO_ICRBB_D 0x00400628 /* IO CRB Entry B_D */ -#define IIO_ICRBB_E 0x00400630 /* IO CRB Entry B_E */ - -#define IIO_ICRBC_A 0x00400640 /* IO CRB Entry C_A */ -#define IIO_ICRBC_B 0x00400648 /* IO CRB Entry C_B */ -#define IIO_ICRBC_C 0x00400650 /* IO CRB Entry C_C */ -#define IIO_ICRBC_D 0x00400658 /* IO CRB Entry C_D */ -#define IIO_ICRBC_E 0x00400660 /* IO CRB Entry C_E */ - -#define IIO_ICRBD_A 0x00400670 /* IO CRB Entry D_A */ -#define IIO_ICRBD_B 0x00400678 /* IO CRB Entry D_B */ -#define IIO_ICRBD_C 0x00400680 /* IO CRB Entry D_C */ -#define IIO_ICRBD_D 0x00400688 /* IO CRB Entry D_D */ -#define IIO_ICRBD_E 0x00400690 /* IO CRB Entry D_E */ - -#define IIO_ICRBE_A 0x004006a0 /* IO CRB Entry E_A */ -#define IIO_ICRBE_B 0x004006a8 /* IO CRB Entry E_B */ -#define IIO_ICRBE_C 0x004006b0 /* IO CRB Entry E_C */ -#define IIO_ICRBE_D 0x004006b8 /* IO CRB Entry E_D */ -#define IIO_ICRBE_E 0x004006c0 /* IO CRB Entry E_E */ - -#define IIO_ICSML 0x00400700 /* IO CRB Spurious Message Low */ -#define IIO_ICSMM 0x00400708 /* IO CRB Spurious Message Middle */ -#define IIO_ICSMH 0x00400710 /* IO CRB Spurious Message High */ - -#define IIO_IDBSS 0x00400718 /* IO Debug Submenu Select */ - -#define IIO_IBLS0 0x00410000 /* IO BTE Length Status 0 */ -#define IIO_IBSA0 0x00410008 /* IO BTE Source Address 0 */ -#define IIO_IBDA0 0x00410010 /* IO BTE Destination Address 0 */ -#define IIO_IBCT0 0x00410018 /* IO BTE Control Terminate 0 */ -#define IIO_IBNA0 0x00410020 /* IO BTE Notification Address 0 */ -#define IIO_IBIA0 0x00410028 /* IO BTE Interrupt Address 0 */ -#define IIO_IBLS1 0x00420000 /* IO BTE Length Status 1 */ -#define IIO_IBSA1 0x00420008 /* IO BTE Source Address 1 */ -#define IIO_IBDA1 0x00420010 /* IO BTE Destination Address 1 */ -#define IIO_IBCT1 0x00420018 /* IO BTE Control Terminate 1 */ -#define IIO_IBNA1 0x00420020 /* IO BTE Notification Address 1 */ -#define IIO_IBIA1 0x00420028 /* IO BTE Interrupt Address 1 */ - -#define IIO_IPCR 0x00430000 /* IO Performance Control */ -#define IIO_IPPR 0x00430008 /* IO Performance Profiling */ - - -/************************************************************************ - * * +#define HUB_WIDGET_ID_MAX 0xf +#define IIO_NUM_ITTES 7 +#define HUB_NUM_BIG_WINDOW (IIO_NUM_ITTES - 1) + +#define IIO_WID 0x00400000 /* Crosstalk Widget Identification */ + /* This register is also accessible from + * Crosstalk at address 0x0. */ +#define IIO_WSTAT 0x00400008 /* Crosstalk Widget Status */ +#define IIO_WCR 0x00400020 /* Crosstalk Widget Control Register */ +#define IIO_ILAPR 0x00400100 /* IO Local Access Protection Register */ +#define IIO_ILAPO 0x00400108 /* IO Local Access Protection Override */ +#define IIO_IOWA 0x00400110 /* IO Outbound Widget Access */ +#define IIO_IIWA 0x00400118 /* IO Inbound Widget Access */ +#define IIO_IIDEM 0x00400120 /* IO Inbound Device Error Mask */ +#define IIO_ILCSR 0x00400128 /* IO LLP Control and Status Register */ +#define IIO_ILLR 0x00400130 /* IO LLP Log Register */ +#define IIO_IIDSR 0x00400138 /* IO Interrupt Destination */ + +#define IIO_IGFX0 0x00400140 /* IO Graphics Node-Widget Map 0 */ +#define IIO_IGFX1 0x00400148 /* IO Graphics Node-Widget Map 1 */ + +#define IIO_ISCR0 0x00400150 /* IO Scratch Register 0 */ +#define IIO_ISCR1 0x00400158 /* IO Scratch Register 1 */ + +#define IIO_ITTE1 0x00400160 /* IO Translation Table Entry 1 */ +#define IIO_ITTE2 0x00400168 /* IO Translation Table Entry 2 */ +#define IIO_ITTE3 0x00400170 /* IO Translation Table Entry 3 */ +#define IIO_ITTE4 0x00400178 /* IO Translation Table Entry 4 */ +#define IIO_ITTE5 0x00400180 /* IO Translation Table Entry 5 */ +#define IIO_ITTE6 0x00400188 /* IO Translation Table Entry 6 */ +#define IIO_ITTE7 0x00400190 /* IO Translation Table Entry 7 */ + +#define IIO_IPRB0 0x00400198 /* IO PRB Entry 0 */ +#define IIO_IPRB8 0x004001A0 /* IO PRB Entry 8 */ +#define IIO_IPRB9 0x004001A8 /* IO PRB Entry 9 */ +#define IIO_IPRBA 0x004001B0 /* IO PRB Entry A */ +#define IIO_IPRBB 0x004001B8 /* IO PRB Entry B */ +#define IIO_IPRBC 0x004001C0 /* IO PRB Entry C */ +#define IIO_IPRBD 0x004001C8 /* IO PRB Entry D */ +#define IIO_IPRBE 0x004001D0 /* IO PRB Entry E */ +#define IIO_IPRBF 0x004001D8 /* IO PRB Entry F */ + +#define IIO_IXCC 0x004001E0 /* IO Crosstalk Credit Count Timeout */ +#define IIO_IMEM 0x004001E8 /* IO Miscellaneous Error Mask */ +#define IIO_IXTT 0x004001F0 /* IO Crosstalk Timeout Threshold */ +#define IIO_IECLR 0x004001F8 /* IO Error Clear Register */ +#define IIO_IBCR 0x00400200 /* IO BTE Control Register */ + +#define IIO_IXSM 0x00400208 /* IO Crosstalk Spurious Message */ +#define IIO_IXSS 0x00400210 /* IO Crosstalk Spurious Sideband */ + +#define IIO_ILCT 0x00400218 /* IO LLP Channel Test */ + +#define IIO_IIEPH1 0x00400220 /* IO Incoming Error Packet Header, Part 1 */ +#define IIO_IIEPH2 0x00400228 /* IO Incoming Error Packet Header, Part 2 */ + +#define IIO_ISLAPR 0x00400230 /* IO SXB Local Access Protection Regster */ +#define IIO_ISLAPO 0x00400238 /* IO SXB Local Access Protection Override */ + +#define IIO_IWI 0x00400240 /* IO Wrapper Interrupt Register */ +#define IIO_IWEL 0x00400248 /* IO Wrapper Error Log Register */ +#define IIO_IWC 0x00400250 /* IO Wrapper Control Register */ +#define IIO_IWS 0x00400258 /* IO Wrapper Status Register */ +#define IIO_IWEIM 0x00400260 /* IO Wrapper Error Interrupt Masking Register */ + +#define IIO_IPCA 0x00400300 /* IO PRB Counter Adjust */ + +#define IIO_IPRTE0_A 0x00400308 /* IO PIO Read Address Table Entry 0, Part A */ +#define IIO_IPRTE1_A 0x00400310 /* IO PIO Read Address Table Entry 1, Part A */ +#define IIO_IPRTE2_A 0x00400318 /* IO PIO Read Address Table Entry 2, Part A */ +#define IIO_IPRTE3_A 0x00400320 /* IO PIO Read Address Table Entry 3, Part A */ +#define IIO_IPRTE4_A 0x00400328 /* IO PIO Read Address Table Entry 4, Part A */ +#define IIO_IPRTE5_A 0x00400330 /* IO PIO Read Address Table Entry 5, Part A */ +#define IIO_IPRTE6_A 0x00400338 /* IO PIO Read Address Table Entry 6, Part A */ +#define IIO_IPRTE7_A 0x00400340 /* IO PIO Read Address Table Entry 7, Part A */ + +#define IIO_IPRTE0_B 0x00400348 /* IO PIO Read Address Table Entry 0, Part B */ +#define IIO_IPRTE1_B 0x00400350 /* IO PIO Read Address Table Entry 1, Part B */ +#define IIO_IPRTE2_B 0x00400358 /* IO PIO Read Address Table Entry 2, Part B */ +#define IIO_IPRTE3_B 0x00400360 /* IO PIO Read Address Table Entry 3, Part B */ +#define IIO_IPRTE4_B 0x00400368 /* IO PIO Read Address Table Entry 4, Part B */ +#define IIO_IPRTE5_B 0x00400370 /* IO PIO Read Address Table Entry 5, Part B */ +#define IIO_IPRTE6_B 0x00400378 /* IO PIO Read Address Table Entry 6, Part B */ +#define IIO_IPRTE7_B 0x00400380 /* IO PIO Read Address Table Entry 7, Part B */ + +#define IIO_IPDR 0x00400388 /* IO PIO Deallocation Register */ +#define IIO_ICDR 0x00400390 /* IO CRB Entry Deallocation Register */ +#define IIO_IFDR 0x00400398 /* IO IOQ FIFO Depth Register */ +#define IIO_IIAP 0x004003A0 /* IO IIQ Arbitration Parameters */ +#define IIO_ICMR 0x004003A8 /* IO CRB Management Register */ +#define IIO_ICCR 0x004003B0 /* IO CRB Control Register */ +#define IIO_ICTO 0x004003B8 /* IO CRB Timeout */ +#define IIO_ICTP 0x004003C0 /* IO CRB Timeout Prescalar */ + +#define IIO_ICRB0_A 0x00400400 /* IO CRB Entry 0_A */ +#define IIO_ICRB0_B 0x00400408 /* IO CRB Entry 0_B */ +#define IIO_ICRB0_C 0x00400410 /* IO CRB Entry 0_C */ +#define IIO_ICRB0_D 0x00400418 /* IO CRB Entry 0_D */ +#define IIO_ICRB0_E 0x00400420 /* IO CRB Entry 0_E */ + +#define IIO_ICRB1_A 0x00400430 /* IO CRB Entry 1_A */ +#define IIO_ICRB1_B 0x00400438 /* IO CRB Entry 1_B */ +#define IIO_ICRB1_C 0x00400440 /* IO CRB Entry 1_C */ +#define IIO_ICRB1_D 0x00400448 /* IO CRB Entry 1_D */ +#define IIO_ICRB1_E 0x00400450 /* IO CRB Entry 1_E */ + +#define IIO_ICRB2_A 0x00400460 /* IO CRB Entry 2_A */ +#define IIO_ICRB2_B 0x00400468 /* IO CRB Entry 2_B */ +#define IIO_ICRB2_C 0x00400470 /* IO CRB Entry 2_C */ +#define IIO_ICRB2_D 0x00400478 /* IO CRB Entry 2_D */ +#define IIO_ICRB2_E 0x00400480 /* IO CRB Entry 2_E */ + +#define IIO_ICRB3_A 0x00400490 /* IO CRB Entry 3_A */ +#define IIO_ICRB3_B 0x00400498 /* IO CRB Entry 3_B */ +#define IIO_ICRB3_C 0x004004a0 /* IO CRB Entry 3_C */ +#define IIO_ICRB3_D 0x004004a8 /* IO CRB Entry 3_D */ +#define IIO_ICRB3_E 0x004004b0 /* IO CRB Entry 3_E */ + +#define IIO_ICRB4_A 0x004004c0 /* IO CRB Entry 4_A */ +#define IIO_ICRB4_B 0x004004c8 /* IO CRB Entry 4_B */ +#define IIO_ICRB4_C 0x004004d0 /* IO CRB Entry 4_C */ +#define IIO_ICRB4_D 0x004004d8 /* IO CRB Entry 4_D */ +#define IIO_ICRB4_E 0x004004e0 /* IO CRB Entry 4_E */ + +#define IIO_ICRB5_A 0x004004f0 /* IO CRB Entry 5_A */ +#define IIO_ICRB5_B 0x004004f8 /* IO CRB Entry 5_B */ +#define IIO_ICRB5_C 0x00400500 /* IO CRB Entry 5_C */ +#define IIO_ICRB5_D 0x00400508 /* IO CRB Entry 5_D */ +#define IIO_ICRB5_E 0x00400510 /* IO CRB Entry 5_E */ + +#define IIO_ICRB6_A 0x00400520 /* IO CRB Entry 6_A */ +#define IIO_ICRB6_B 0x00400528 /* IO CRB Entry 6_B */ +#define IIO_ICRB6_C 0x00400530 /* IO CRB Entry 6_C */ +#define IIO_ICRB6_D 0x00400538 /* IO CRB Entry 6_D */ +#define IIO_ICRB6_E 0x00400540 /* IO CRB Entry 6_E */ + +#define IIO_ICRB7_A 0x00400550 /* IO CRB Entry 7_A */ +#define IIO_ICRB7_B 0x00400558 /* IO CRB Entry 7_B */ +#define IIO_ICRB7_C 0x00400560 /* IO CRB Entry 7_C */ +#define IIO_ICRB7_D 0x00400568 /* IO CRB Entry 7_D */ +#define IIO_ICRB7_E 0x00400570 /* IO CRB Entry 7_E */ + +#define IIO_ICRB8_A 0x00400580 /* IO CRB Entry 8_A */ +#define IIO_ICRB8_B 0x00400588 /* IO CRB Entry 8_B */ +#define IIO_ICRB8_C 0x00400590 /* IO CRB Entry 8_C */ +#define IIO_ICRB8_D 0x00400598 /* IO CRB Entry 8_D */ +#define IIO_ICRB8_E 0x004005a0 /* IO CRB Entry 8_E */ + +#define IIO_ICRB9_A 0x004005b0 /* IO CRB Entry 9_A */ +#define IIO_ICRB9_B 0x004005b8 /* IO CRB Entry 9_B */ +#define IIO_ICRB9_C 0x004005c0 /* IO CRB Entry 9_C */ +#define IIO_ICRB9_D 0x004005c8 /* IO CRB Entry 9_D */ +#define IIO_ICRB9_E 0x004005d0 /* IO CRB Entry 9_E */ + +#define IIO_ICRBA_A 0x004005e0 /* IO CRB Entry A_A */ +#define IIO_ICRBA_B 0x004005e8 /* IO CRB Entry A_B */ +#define IIO_ICRBA_C 0x004005f0 /* IO CRB Entry A_C */ +#define IIO_ICRBA_D 0x004005f8 /* IO CRB Entry A_D */ +#define IIO_ICRBA_E 0x00400600 /* IO CRB Entry A_E */ + +#define IIO_ICRBB_A 0x00400610 /* IO CRB Entry B_A */ +#define IIO_ICRBB_B 0x00400618 /* IO CRB Entry B_B */ +#define IIO_ICRBB_C 0x00400620 /* IO CRB Entry B_C */ +#define IIO_ICRBB_D 0x00400628 /* IO CRB Entry B_D */ +#define IIO_ICRBB_E 0x00400630 /* IO CRB Entry B_E */ + +#define IIO_ICRBC_A 0x00400640 /* IO CRB Entry C_A */ +#define IIO_ICRBC_B 0x00400648 /* IO CRB Entry C_B */ +#define IIO_ICRBC_C 0x00400650 /* IO CRB Entry C_C */ +#define IIO_ICRBC_D 0x00400658 /* IO CRB Entry C_D */ +#define IIO_ICRBC_E 0x00400660 /* IO CRB Entry C_E */ + +#define IIO_ICRBD_A 0x00400670 /* IO CRB Entry D_A */ +#define IIO_ICRBD_B 0x00400678 /* IO CRB Entry D_B */ +#define IIO_ICRBD_C 0x00400680 /* IO CRB Entry D_C */ +#define IIO_ICRBD_D 0x00400688 /* IO CRB Entry D_D */ +#define IIO_ICRBD_E 0x00400690 /* IO CRB Entry D_E */ + +#define IIO_ICRBE_A 0x004006a0 /* IO CRB Entry E_A */ +#define IIO_ICRBE_B 0x004006a8 /* IO CRB Entry E_B */ +#define IIO_ICRBE_C 0x004006b0 /* IO CRB Entry E_C */ +#define IIO_ICRBE_D 0x004006b8 /* IO CRB Entry E_D */ +#define IIO_ICRBE_E 0x004006c0 /* IO CRB Entry E_E */ + +#define IIO_ICSML 0x00400700 /* IO CRB Spurious Message Low */ +#define IIO_ICSMM 0x00400708 /* IO CRB Spurious Message Middle */ +#define IIO_ICSMH 0x00400710 /* IO CRB Spurious Message High */ + +#define IIO_IDBSS 0x00400718 /* IO Debug Submenu Select */ + +#define IIO_IBLS0 0x00410000 /* IO BTE Length Status 0 */ +#define IIO_IBSA0 0x00410008 /* IO BTE Source Address 0 */ +#define IIO_IBDA0 0x00410010 /* IO BTE Destination Address 0 */ +#define IIO_IBCT0 0x00410018 /* IO BTE Control Terminate 0 */ +#define IIO_IBNA0 0x00410020 /* IO BTE Notification Address 0 */ +#define IIO_IBIA0 0x00410028 /* IO BTE Interrupt Address 0 */ +#define IIO_IBLS1 0x00420000 /* IO BTE Length Status 1 */ +#define IIO_IBSA1 0x00420008 /* IO BTE Source Address 1 */ +#define IIO_IBDA1 0x00420010 /* IO BTE Destination Address 1 */ +#define IIO_IBCT1 0x00420018 /* IO BTE Control Terminate 1 */ +#define IIO_IBNA1 0x00420020 /* IO BTE Notification Address 1 */ +#define IIO_IBIA1 0x00420028 /* IO BTE Interrupt Address 1 */ + +#define IIO_IPCR 0x00430000 /* IO Performance Control */ +#define IIO_IPPR 0x00430008 /* IO Performance Profiling */ + +/************************************************************************ + * * * Description: This register echoes some information from the * * LB_REV_ID register. It is available through Crosstalk as described * * above. The REV_NUM and MFG_NUM fields receive their values from * * the REVISION and MANUFACTURER fields in the LB_REV_ID register. * * The PART_NUM field's value is the Crosstalk device ID number that * * Steve Miller assigned to the SHub chip. * - * * + * * ************************************************************************/ typedef union ii_wid_u { - uint64_t ii_wid_regval; - struct { - uint64_t w_rsvd_1 : 1; - uint64_t w_mfg_num : 11; - uint64_t w_part_num : 16; - uint64_t w_rev_num : 4; - uint64_t w_rsvd : 32; + uint64_t ii_wid_regval; + struct { + uint64_t w_rsvd_1:1; + uint64_t w_mfg_num:11; + uint64_t w_part_num:16; + uint64_t w_rev_num:4; + uint64_t w_rsvd:32; } ii_wid_fld_s; } ii_wid_u_t; - /************************************************************************ - * * + * * * The fields in this register are set upon detection of an error * * and cleared by various mechanisms, as explained in the * * description. * - * * + * * ************************************************************************/ typedef union ii_wstat_u { - uint64_t ii_wstat_regval; - struct { - uint64_t w_pending : 4; - uint64_t w_xt_crd_to : 1; - uint64_t w_xt_tail_to : 1; - uint64_t w_rsvd_3 : 3; - uint64_t w_tx_mx_rty : 1; - uint64_t w_rsvd_2 : 6; - uint64_t w_llp_tx_cnt : 8; - uint64_t w_rsvd_1 : 8; - uint64_t w_crazy : 1; - uint64_t w_rsvd : 31; + uint64_t ii_wstat_regval; + struct { + uint64_t w_pending:4; + uint64_t w_xt_crd_to:1; + uint64_t w_xt_tail_to:1; + uint64_t w_rsvd_3:3; + uint64_t w_tx_mx_rty:1; + uint64_t w_rsvd_2:6; + uint64_t w_llp_tx_cnt:8; + uint64_t w_rsvd_1:8; + uint64_t w_crazy:1; + uint64_t w_rsvd:31; } ii_wstat_fld_s; } ii_wstat_u_t; - /************************************************************************ - * * + * * * Description: This is a read-write enabled register. It controls * * various aspects of the Crosstalk flow control. * - * * + * * ************************************************************************/ typedef union ii_wcr_u { - uint64_t ii_wcr_regval; - struct { - uint64_t w_wid : 4; - uint64_t w_tag : 1; - uint64_t w_rsvd_1 : 8; - uint64_t w_dst_crd : 3; - uint64_t w_f_bad_pkt : 1; - uint64_t w_dir_con : 1; - uint64_t w_e_thresh : 5; - uint64_t w_rsvd : 41; + uint64_t ii_wcr_regval; + struct { + uint64_t w_wid:4; + uint64_t w_tag:1; + uint64_t w_rsvd_1:8; + uint64_t w_dst_crd:3; + uint64_t w_f_bad_pkt:1; + uint64_t w_dir_con:1; + uint64_t w_e_thresh:5; + uint64_t w_rsvd:41; } ii_wcr_fld_s; } ii_wcr_u_t; - /************************************************************************ - * * + * * * Description: This register's value is a bit vector that guards * * access to local registers within the II as well as to external * * Crosstalk widgets. Each bit in the register corresponds to a * @@ -311,21 +306,18 @@ typedef union ii_wcr_u { * region ID bits are enabled in this same register. It can also be * * accessed through the IAlias space by the local processors. * * The reset value of this register allows access by all nodes. * - * * + * * ************************************************************************/ typedef union ii_ilapr_u { - uint64_t ii_ilapr_regval; - struct { - uint64_t i_region : 64; + uint64_t ii_ilapr_regval; + struct { + uint64_t i_region:64; } ii_ilapr_fld_s; } ii_ilapr_u_t; - - - /************************************************************************ - * * + * * * Description: A write to this register of the 64-bit value * * "SGIrules" in ASCII, will cause the bit in the ILAPR register * * corresponding to the region of the requestor to be set (allow * @@ -334,59 +326,54 @@ typedef union ii_ilapr_u { * This register can also be accessed through the IAlias space. * * However, this access will not change the access permissions in the * * ILAPR. * - * * + * * ************************************************************************/ typedef union ii_ilapo_u { - uint64_t ii_ilapo_regval; - struct { - uint64_t i_io_ovrride : 64; + uint64_t ii_ilapo_regval; + struct { + uint64_t i_io_ovrride:64; } ii_ilapo_fld_s; } ii_ilapo_u_t; - - /************************************************************************ - * * + * * * This register qualifies all the PIO and Graphics writes launched * * from the SHUB towards a widget. * - * * + * * ************************************************************************/ typedef union ii_iowa_u { - uint64_t ii_iowa_regval; - struct { - uint64_t i_w0_oac : 1; - uint64_t i_rsvd_1 : 7; - uint64_t i_wx_oac : 8; - uint64_t i_rsvd : 48; + uint64_t ii_iowa_regval; + struct { + uint64_t i_w0_oac:1; + uint64_t i_rsvd_1:7; + uint64_t i_wx_oac:8; + uint64_t i_rsvd:48; } ii_iowa_fld_s; } ii_iowa_u_t; - /************************************************************************ - * * + * * * Description: This register qualifies all the requests launched * * from a widget towards the Shub. This register is intended to be * * used by software in case of misbehaving widgets. * - * * - * * + * * + * * ************************************************************************/ typedef union ii_iiwa_u { - uint64_t ii_iiwa_regval; - struct { - uint64_t i_w0_iac : 1; - uint64_t i_rsvd_1 : 7; - uint64_t i_wx_iac : 8; - uint64_t i_rsvd : 48; + uint64_t ii_iiwa_regval; + struct { + uint64_t i_w0_iac:1; + uint64_t i_rsvd_1:7; + uint64_t i_wx_iac:8; + uint64_t i_rsvd:48; } ii_iiwa_fld_s; } ii_iiwa_u_t; - - /************************************************************************ - * * + * * * Description: This register qualifies all the operations launched * * from a widget towards the SHub. It allows individual access * * control for up to 8 devices per widget. A device refers to * @@ -401,72 +388,69 @@ typedef union ii_iiwa_u { * The bits in this field are set by writing a 1 to them. Incoming * * replies from Crosstalk are not subject to this access control * * mechanism. * - * * + * * ************************************************************************/ typedef union ii_iidem_u { - uint64_t ii_iidem_regval; - struct { - uint64_t i_w8_dxs : 8; - uint64_t i_w9_dxs : 8; - uint64_t i_wa_dxs : 8; - uint64_t i_wb_dxs : 8; - uint64_t i_wc_dxs : 8; - uint64_t i_wd_dxs : 8; - uint64_t i_we_dxs : 8; - uint64_t i_wf_dxs : 8; + uint64_t ii_iidem_regval; + struct { + uint64_t i_w8_dxs:8; + uint64_t i_w9_dxs:8; + uint64_t i_wa_dxs:8; + uint64_t i_wb_dxs:8; + uint64_t i_wc_dxs:8; + uint64_t i_wd_dxs:8; + uint64_t i_we_dxs:8; + uint64_t i_wf_dxs:8; } ii_iidem_fld_s; } ii_iidem_u_t; - /************************************************************************ - * * + * * * This register contains the various programmable fields necessary * * for controlling and observing the LLP signals. * - * * + * * ************************************************************************/ typedef union ii_ilcsr_u { - uint64_t ii_ilcsr_regval; - struct { - uint64_t i_nullto : 6; - uint64_t i_rsvd_4 : 2; - uint64_t i_wrmrst : 1; - uint64_t i_rsvd_3 : 1; - uint64_t i_llp_en : 1; - uint64_t i_bm8 : 1; - uint64_t i_llp_stat : 2; - uint64_t i_remote_power : 1; - uint64_t i_rsvd_2 : 1; - uint64_t i_maxrtry : 10; - uint64_t i_d_avail_sel : 2; - uint64_t i_rsvd_1 : 4; - uint64_t i_maxbrst : 10; - uint64_t i_rsvd : 22; + uint64_t ii_ilcsr_regval; + struct { + uint64_t i_nullto:6; + uint64_t i_rsvd_4:2; + uint64_t i_wrmrst:1; + uint64_t i_rsvd_3:1; + uint64_t i_llp_en:1; + uint64_t i_bm8:1; + uint64_t i_llp_stat:2; + uint64_t i_remote_power:1; + uint64_t i_rsvd_2:1; + uint64_t i_maxrtry:10; + uint64_t i_d_avail_sel:2; + uint64_t i_rsvd_1:4; + uint64_t i_maxbrst:10; + uint64_t i_rsvd:22; } ii_ilcsr_fld_s; } ii_ilcsr_u_t; - /************************************************************************ - * * + * * * This is simply a status registers that monitors the LLP error * - * rate. * - * * + * rate. * + * * ************************************************************************/ typedef union ii_illr_u { - uint64_t ii_illr_regval; - struct { - uint64_t i_sn_cnt : 16; - uint64_t i_cb_cnt : 16; - uint64_t i_rsvd : 32; + uint64_t ii_illr_regval; + struct { + uint64_t i_sn_cnt:16; + uint64_t i_cb_cnt:16; + uint64_t i_rsvd:32; } ii_illr_fld_s; } ii_illr_u_t; - /************************************************************************ - * * + * * * Description: All II-detected non-BTE error interrupts are * * specified via this register. * * NOTE: The PI interrupt register address is hardcoded in the II. If * @@ -476,107 +460,100 @@ typedef union ii_illr_u { * PI_ID==1, then the II sends the interrupt request to address * * offset 0x01A0_0090 within the local register address space of PI1 * * on the node specified by the NODE field. * - * * + * * ************************************************************************/ typedef union ii_iidsr_u { - uint64_t ii_iidsr_regval; - struct { - uint64_t i_level : 8; - uint64_t i_pi_id : 1; - uint64_t i_node : 11; - uint64_t i_rsvd_3 : 4; - uint64_t i_enable : 1; - uint64_t i_rsvd_2 : 3; - uint64_t i_int_sent : 2; - uint64_t i_rsvd_1 : 2; - uint64_t i_pi0_forward_int : 1; - uint64_t i_pi1_forward_int : 1; - uint64_t i_rsvd : 30; + uint64_t ii_iidsr_regval; + struct { + uint64_t i_level:8; + uint64_t i_pi_id:1; + uint64_t i_node:11; + uint64_t i_rsvd_3:4; + uint64_t i_enable:1; + uint64_t i_rsvd_2:3; + uint64_t i_int_sent:2; + uint64_t i_rsvd_1:2; + uint64_t i_pi0_forward_int:1; + uint64_t i_pi1_forward_int:1; + uint64_t i_rsvd:30; } ii_iidsr_fld_s; } ii_iidsr_u_t; - - /************************************************************************ - * * + * * * There are two instances of this register. This register is used * * for matching up the incoming responses from the graphics widget to * * the processor that initiated the graphics operation. The * * write-responses are converted to graphics credits and returned to * * the processor so that the processor interface can manage the flow * * control. * - * * + * * ************************************************************************/ typedef union ii_igfx0_u { - uint64_t ii_igfx0_regval; - struct { - uint64_t i_w_num : 4; - uint64_t i_pi_id : 1; - uint64_t i_n_num : 12; - uint64_t i_p_num : 1; - uint64_t i_rsvd : 46; + uint64_t ii_igfx0_regval; + struct { + uint64_t i_w_num:4; + uint64_t i_pi_id:1; + uint64_t i_n_num:12; + uint64_t i_p_num:1; + uint64_t i_rsvd:46; } ii_igfx0_fld_s; } ii_igfx0_u_t; - /************************************************************************ - * * + * * * There are two instances of this register. This register is used * * for matching up the incoming responses from the graphics widget to * * the processor that initiated the graphics operation. The * * write-responses are converted to graphics credits and returned to * * the processor so that the processor interface can manage the flow * * control. * - * * + * * ************************************************************************/ typedef union ii_igfx1_u { - uint64_t ii_igfx1_regval; - struct { - uint64_t i_w_num : 4; - uint64_t i_pi_id : 1; - uint64_t i_n_num : 12; - uint64_t i_p_num : 1; - uint64_t i_rsvd : 46; + uint64_t ii_igfx1_regval; + struct { + uint64_t i_w_num:4; + uint64_t i_pi_id:1; + uint64_t i_n_num:12; + uint64_t i_p_num:1; + uint64_t i_rsvd:46; } ii_igfx1_fld_s; } ii_igfx1_u_t; - /************************************************************************ - * * + * * * There are two instances of this registers. These registers are * * used as scratch registers for software use. * - * * + * * ************************************************************************/ typedef union ii_iscr0_u { - uint64_t ii_iscr0_regval; - struct { - uint64_t i_scratch : 64; + uint64_t ii_iscr0_regval; + struct { + uint64_t i_scratch:64; } ii_iscr0_fld_s; } ii_iscr0_u_t; - - /************************************************************************ - * * + * * * There are two instances of this registers. These registers are * * used as scratch registers for software use. * - * * + * * ************************************************************************/ typedef union ii_iscr1_u { - uint64_t ii_iscr1_regval; - struct { - uint64_t i_scratch : 64; + uint64_t ii_iscr1_regval; + struct { + uint64_t i_scratch:64; } ii_iscr1_fld_s; } ii_iscr1_u_t; - /************************************************************************ - * * + * * * Description: There are seven instances of translation table entry * * registers. Each register maps a Shub Big Window to a 48-bit * * address on Crosstalk. * @@ -599,23 +576,22 @@ typedef union ii_iscr1_u { * Crosstalk space addressable by the Shub is thus the lower * * 8-GBytes per widget (N-mode), only <SUP >7</SUP>/<SUB >32nds</SUB> * * of this space can be accessed. * - * * + * * ************************************************************************/ typedef union ii_itte1_u { - uint64_t ii_itte1_regval; - struct { - uint64_t i_offset : 5; - uint64_t i_rsvd_1 : 3; - uint64_t i_w_num : 4; - uint64_t i_iosp : 1; - uint64_t i_rsvd : 51; + uint64_t ii_itte1_regval; + struct { + uint64_t i_offset:5; + uint64_t i_rsvd_1:3; + uint64_t i_w_num:4; + uint64_t i_iosp:1; + uint64_t i_rsvd:51; } ii_itte1_fld_s; } ii_itte1_u_t; - /************************************************************************ - * * + * * * Description: There are seven instances of translation table entry * * registers. Each register maps a Shub Big Window to a 48-bit * * address on Crosstalk. * @@ -638,23 +614,22 @@ typedef union ii_itte1_u { * Crosstalk space addressable by the Shub is thus the lower * * 8-GBytes per widget (N-mode), only <SUP >7</SUP>/<SUB >32nds</SUB> * * of this space can be accessed. * - * * + * * ************************************************************************/ typedef union ii_itte2_u { - uint64_t ii_itte2_regval; - struct { - uint64_t i_offset : 5; - uint64_t i_rsvd_1 : 3; - uint64_t i_w_num : 4; - uint64_t i_iosp : 1; - uint64_t i_rsvd : 51; + uint64_t ii_itte2_regval; + struct { + uint64_t i_offset:5; + uint64_t i_rsvd_1:3; + uint64_t i_w_num:4; + uint64_t i_iosp:1; + uint64_t i_rsvd:51; } ii_itte2_fld_s; } ii_itte2_u_t; - /************************************************************************ - * * + * * * Description: There are seven instances of translation table entry * * registers. Each register maps a Shub Big Window to a 48-bit * * address on Crosstalk. * @@ -677,23 +652,22 @@ typedef union ii_itte2_u { * Crosstalk space addressable by the SHub is thus the lower * * 8-GBytes per widget (N-mode), only <SUP >7</SUP>/<SUB >32nds</SUB> * * of this space can be accessed. * - * * + * * ************************************************************************/ typedef union ii_itte3_u { - uint64_t ii_itte3_regval; - struct { - uint64_t i_offset : 5; - uint64_t i_rsvd_1 : 3; - uint64_t i_w_num : 4; - uint64_t i_iosp : 1; - uint64_t i_rsvd : 51; + uint64_t ii_itte3_regval; + struct { + uint64_t i_offset:5; + uint64_t i_rsvd_1:3; + uint64_t i_w_num:4; + uint64_t i_iosp:1; + uint64_t i_rsvd:51; } ii_itte3_fld_s; } ii_itte3_u_t; - /************************************************************************ - * * + * * * Description: There are seven instances of translation table entry * * registers. Each register maps a SHub Big Window to a 48-bit * * address on Crosstalk. * @@ -716,23 +690,22 @@ typedef union ii_itte3_u { * Crosstalk space addressable by the SHub is thus the lower * * 8-GBytes per widget (N-mode), only <SUP >7</SUP>/<SUB >32nds</SUB> * * of this space can be accessed. * - * * + * * ************************************************************************/ typedef union ii_itte4_u { - uint64_t ii_itte4_regval; - struct { - uint64_t i_offset : 5; - uint64_t i_rsvd_1 : 3; - uint64_t i_w_num : 4; - uint64_t i_iosp : 1; - uint64_t i_rsvd : 51; + uint64_t ii_itte4_regval; + struct { + uint64_t i_offset:5; + uint64_t i_rsvd_1:3; + uint64_t i_w_num:4; + uint64_t i_iosp:1; + uint64_t i_rsvd:51; } ii_itte4_fld_s; } ii_itte4_u_t; - /************************************************************************ - * * + * * * Description: There are seven instances of translation table entry * * registers. Each register maps a SHub Big Window to a 48-bit * * address on Crosstalk. * @@ -755,23 +728,22 @@ typedef union ii_itte4_u { * Crosstalk space addressable by the Shub is thus the lower * * 8-GBytes per widget (N-mode), only <SUP >7</SUP>/<SUB >32nds</SUB> * * of this space can be accessed. * - * * + * * ************************************************************************/ typedef union ii_itte5_u { - uint64_t ii_itte5_regval; - struct { - uint64_t i_offset : 5; - uint64_t i_rsvd_1 : 3; - uint64_t i_w_num : 4; - uint64_t i_iosp : 1; - uint64_t i_rsvd : 51; + uint64_t ii_itte5_regval; + struct { + uint64_t i_offset:5; + uint64_t i_rsvd_1:3; + uint64_t i_w_num:4; + uint64_t i_iosp:1; + uint64_t i_rsvd:51; } ii_itte5_fld_s; } ii_itte5_u_t; - /************************************************************************ - * * + * * * Description: There are seven instances of translation table entry * * registers. Each register maps a Shub Big Window to a 48-bit * * address on Crosstalk. * @@ -794,23 +766,22 @@ typedef union ii_itte5_u { * Crosstalk space addressable by the Shub is thus the lower * * 8-GBytes per widget (N-mode), only <SUP >7</SUP>/<SUB >32nds</SUB> * * of this space can be accessed. * - * * + * * ************************************************************************/ typedef union ii_itte6_u { - uint64_t ii_itte6_regval; - struct { - uint64_t i_offset : 5; - uint64_t i_rsvd_1 : 3; - uint64_t i_w_num : 4; - uint64_t i_iosp : 1; - uint64_t i_rsvd : 51; + uint64_t ii_itte6_regval; + struct { + uint64_t i_offset:5; + uint64_t i_rsvd_1:3; + uint64_t i_w_num:4; + uint64_t i_iosp:1; + uint64_t i_rsvd:51; } ii_itte6_fld_s; } ii_itte6_u_t; - /************************************************************************ - * * + * * * Description: There are seven instances of translation table entry * * registers. Each register maps a Shub Big Window to a 48-bit * * address on Crosstalk. * @@ -833,23 +804,22 @@ typedef union ii_itte6_u { * Crosstalk space addressable by the SHub is thus the lower * * 8-GBytes per widget (N-mode), only <SUP >7</SUP>/<SUB >32nds</SUB> * * of this space can be accessed. * - * * + * * ************************************************************************/ typedef union ii_itte7_u { - uint64_t ii_itte7_regval; - struct { - uint64_t i_offset : 5; - uint64_t i_rsvd_1 : 3; - uint64_t i_w_num : 4; - uint64_t i_iosp : 1; - uint64_t i_rsvd : 51; + uint64_t ii_itte7_regval; + struct { + uint64_t i_offset:5; + uint64_t i_rsvd_1:3; + uint64_t i_w_num:4; + uint64_t i_iosp:1; + uint64_t i_rsvd:51; } ii_itte7_fld_s; } ii_itte7_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of SHub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -868,33 +838,32 @@ typedef union ii_itte7_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * . * - * * + * . * + * * ************************************************************************/ typedef union ii_iprb0_u { - uint64_t ii_iprb0_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; + uint64_t ii_iprb0_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; } ii_iprb0_fld_s; } ii_iprb0_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of SHub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -913,33 +882,32 @@ typedef union ii_iprb0_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * . * - * * + * . * + * * ************************************************************************/ typedef union ii_iprb8_u { - uint64_t ii_iprb8_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; + uint64_t ii_iprb8_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; } ii_iprb8_fld_s; } ii_iprb8_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of SHub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -958,33 +926,32 @@ typedef union ii_iprb8_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * . * - * * + * . * + * * ************************************************************************/ typedef union ii_iprb9_u { - uint64_t ii_iprb9_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; + uint64_t ii_iprb9_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; } ii_iprb9_fld_s; } ii_iprb9_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of SHub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -1003,33 +970,32 @@ typedef union ii_iprb9_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * * - * * + * * + * * ************************************************************************/ typedef union ii_iprba_u { - uint64_t ii_iprba_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; + uint64_t ii_iprba_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; } ii_iprba_fld_s; } ii_iprba_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of SHub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -1048,33 +1014,32 @@ typedef union ii_iprba_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * . * - * * + * . * + * * ************************************************************************/ typedef union ii_iprbb_u { - uint64_t ii_iprbb_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; + uint64_t ii_iprbb_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; } ii_iprbb_fld_s; } ii_iprbb_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of SHub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -1093,33 +1058,32 @@ typedef union ii_iprbb_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * . * - * * + * . * + * * ************************************************************************/ typedef union ii_iprbc_u { - uint64_t ii_iprbc_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; + uint64_t ii_iprbc_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; } ii_iprbc_fld_s; } ii_iprbc_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of SHub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -1138,33 +1102,32 @@ typedef union ii_iprbc_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * . * - * * + * . * + * * ************************************************************************/ typedef union ii_iprbd_u { - uint64_t ii_iprbd_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; + uint64_t ii_iprbd_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; } ii_iprbd_fld_s; } ii_iprbd_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of SHub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -1183,33 +1146,32 @@ typedef union ii_iprbd_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * . * - * * + * . * + * * ************************************************************************/ typedef union ii_iprbe_u { - uint64_t ii_iprbe_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; + uint64_t ii_iprbe_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; } ii_iprbe_fld_s; } ii_iprbe_u_t; - /************************************************************************ - * * + * * * Description: There are 9 instances of this register, one per * * actual widget in this implementation of Shub and Crossbow. * * Note: Crossbow only has ports for Widgets 8 through F, widget 0 * @@ -1228,33 +1190,32 @@ typedef union ii_iprbe_u { * register; the write will correct the C field and capture its new * * value in the internal register. Even if IECLR[E_PRB_x] is set, the * * SPUR_WR bit will persist if IPRBx hasn't yet been written. * - * . * - * * + * . * + * * ************************************************************************/ typedef union ii_iprbf_u { - uint64_t ii_iprbf_regval; - struct { - uint64_t i_c : 8; - uint64_t i_na : 14; - uint64_t i_rsvd_2 : 2; - uint64_t i_nb : 14; - uint64_t i_rsvd_1 : 2; - uint64_t i_m : 2; - uint64_t i_f : 1; - uint64_t i_of_cnt : 5; - uint64_t i_error : 1; - uint64_t i_rd_to : 1; - uint64_t i_spur_wr : 1; - uint64_t i_spur_rd : 1; - uint64_t i_rsvd : 11; - uint64_t i_mult_err : 1; - } ii_iprbe_fld_s; + uint64_t ii_iprbf_regval; + struct { + uint64_t i_c:8; + uint64_t i_na:14; + uint64_t i_rsvd_2:2; + uint64_t i_nb:14; + uint64_t i_rsvd_1:2; + uint64_t i_m:2; + uint64_t i_f:1; + uint64_t i_of_cnt:5; + uint64_t i_error:1; + uint64_t i_rd_to:1; + uint64_t i_spur_wr:1; + uint64_t i_spur_rd:1; + uint64_t i_rsvd:11; + uint64_t i_mult_err:1; + } ii_iprbe_fld_s; } ii_iprbf_u_t; - /************************************************************************ - * * + * * * This register specifies the timeout value to use for monitoring * * Crosstalk credits which are used outbound to Crosstalk. An * * internal counter called the Crosstalk Credit Timeout Counter * @@ -1267,20 +1228,19 @@ typedef union ii_iprbf_u { * Crosstalk Credit Timeout has occurred. The internal counter is not * * readable from software, and stops counting at its maximum value, * * so it cannot cause more than one interrupt. * - * * + * * ************************************************************************/ typedef union ii_ixcc_u { - uint64_t ii_ixcc_regval; - struct { - uint64_t i_time_out : 26; - uint64_t i_rsvd : 38; + uint64_t ii_ixcc_regval; + struct { + uint64_t i_time_out:26; + uint64_t i_rsvd:38; } ii_ixcc_fld_s; } ii_ixcc_u_t; - /************************************************************************ - * * + * * * Description: This register qualifies all the PIO and DMA * * operations launched from widget 0 towards the SHub. In * * addition, it also qualifies accesses by the BTE streams. * @@ -1292,27 +1252,25 @@ typedef union ii_ixcc_u { * the Wx_IAC field. The bits in this field are set by writing a 1 to * * them. Incoming replies from Crosstalk are not subject to this * * access control mechanism. * - * * + * * ************************************************************************/ typedef union ii_imem_u { - uint64_t ii_imem_regval; - struct { - uint64_t i_w0_esd : 1; - uint64_t i_rsvd_3 : 3; - uint64_t i_b0_esd : 1; - uint64_t i_rsvd_2 : 3; - uint64_t i_b1_esd : 1; - uint64_t i_rsvd_1 : 3; - uint64_t i_clr_precise : 1; - uint64_t i_rsvd : 51; + uint64_t ii_imem_regval; + struct { + uint64_t i_w0_esd:1; + uint64_t i_rsvd_3:3; + uint64_t i_b0_esd:1; + uint64_t i_rsvd_2:3; + uint64_t i_b1_esd:1; + uint64_t i_rsvd_1:3; + uint64_t i_clr_precise:1; + uint64_t i_rsvd:51; } ii_imem_fld_s; } ii_imem_u_t; - - /************************************************************************ - * * + * * * Description: This register specifies the timeout value to use for * * monitoring Crosstalk tail flits coming into the Shub in the * * TAIL_TO field. An internal counter associated with this register * @@ -1332,90 +1290,87 @@ typedef union ii_imem_u { * the value in the RRSP_TO field, a Read Response Timeout has * * occurred, and error handling occurs as described in the Error * * Handling section of this document. * - * * + * * ************************************************************************/ typedef union ii_ixtt_u { - uint64_t ii_ixtt_regval; - struct { - uint64_t i_tail_to : 26; - uint64_t i_rsvd_1 : 6; - uint64_t i_rrsp_ps : 23; - uint64_t i_rrsp_to : 5; - uint64_t i_rsvd : 4; + uint64_t ii_ixtt_regval; + struct { + uint64_t i_tail_to:26; + uint64_t i_rsvd_1:6; + uint64_t i_rrsp_ps:23; + uint64_t i_rrsp_to:5; + uint64_t i_rsvd:4; } ii_ixtt_fld_s; } ii_ixtt_u_t; - /************************************************************************ - * * + * * * Writing a 1 to the fields of this register clears the appropriate * * error bits in other areas of SHub. Note that when the * * E_PRB_x bits are used to clear error bits in PRB registers, * * SPUR_RD and SPUR_WR may persist, because they require additional * * action to clear them. See the IPRBx and IXSS Register * * specifications. * - * * + * * ************************************************************************/ typedef union ii_ieclr_u { - uint64_t ii_ieclr_regval; - struct { - uint64_t i_e_prb_0 : 1; - uint64_t i_rsvd : 7; - uint64_t i_e_prb_8 : 1; - uint64_t i_e_prb_9 : 1; - uint64_t i_e_prb_a : 1; - uint64_t i_e_prb_b : 1; - uint64_t i_e_prb_c : 1; - uint64_t i_e_prb_d : 1; - uint64_t i_e_prb_e : 1; - uint64_t i_e_prb_f : 1; - uint64_t i_e_crazy : 1; - uint64_t i_e_bte_0 : 1; - uint64_t i_e_bte_1 : 1; - uint64_t i_reserved_1 : 10; - uint64_t i_spur_rd_hdr : 1; - uint64_t i_cam_intr_to : 1; - uint64_t i_cam_overflow : 1; - uint64_t i_cam_read_miss : 1; - uint64_t i_ioq_rep_underflow : 1; - uint64_t i_ioq_req_underflow : 1; - uint64_t i_ioq_rep_overflow : 1; - uint64_t i_ioq_req_overflow : 1; - uint64_t i_iiq_rep_overflow : 1; - uint64_t i_iiq_req_overflow : 1; - uint64_t i_ii_xn_rep_cred_overflow : 1; - uint64_t i_ii_xn_req_cred_overflow : 1; - uint64_t i_ii_xn_invalid_cmd : 1; - uint64_t i_xn_ii_invalid_cmd : 1; - uint64_t i_reserved_2 : 21; + uint64_t ii_ieclr_regval; + struct { + uint64_t i_e_prb_0:1; + uint64_t i_rsvd:7; + uint64_t i_e_prb_8:1; + uint64_t i_e_prb_9:1; + uint64_t i_e_prb_a:1; + uint64_t i_e_prb_b:1; + uint64_t i_e_prb_c:1; + uint64_t i_e_prb_d:1; + uint64_t i_e_prb_e:1; + uint64_t i_e_prb_f:1; + uint64_t i_e_crazy:1; + uint64_t i_e_bte_0:1; + uint64_t i_e_bte_1:1; + uint64_t i_reserved_1:10; + uint64_t i_spur_rd_hdr:1; + uint64_t i_cam_intr_to:1; + uint64_t i_cam_overflow:1; + uint64_t i_cam_read_miss:1; + uint64_t i_ioq_rep_underflow:1; + uint64_t i_ioq_req_underflow:1; + uint64_t i_ioq_rep_overflow:1; + uint64_t i_ioq_req_overflow:1; + uint64_t i_iiq_rep_overflow:1; + uint64_t i_iiq_req_overflow:1; + uint64_t i_ii_xn_rep_cred_overflow:1; + uint64_t i_ii_xn_req_cred_overflow:1; + uint64_t i_ii_xn_invalid_cmd:1; + uint64_t i_xn_ii_invalid_cmd:1; + uint64_t i_reserved_2:21; } ii_ieclr_fld_s; } ii_ieclr_u_t; - /************************************************************************ - * * + * * * This register controls both BTEs. SOFT_RESET is intended for * * recovery after an error. COUNT controls the total number of CRBs * * that both BTEs (combined) can use, which affects total BTE * * bandwidth. * - * * + * * ************************************************************************/ typedef union ii_ibcr_u { - uint64_t ii_ibcr_regval; - struct { - uint64_t i_count : 4; - uint64_t i_rsvd_1 : 4; - uint64_t i_soft_reset : 1; - uint64_t i_rsvd : 55; + uint64_t ii_ibcr_regval; + struct { + uint64_t i_count:4; + uint64_t i_rsvd_1:4; + uint64_t i_soft_reset:1; + uint64_t i_rsvd:55; } ii_ibcr_fld_s; } ii_ibcr_u_t; - /************************************************************************ - * * + * * * This register contains the header of a spurious read response * * received from Crosstalk. A spurious read response is defined as a * * read response received by II from a widget for which (1) the SIDN * @@ -1440,49 +1395,47 @@ typedef union ii_ibcr_u { * will be set. Any SPUR_RD bits in any other PRB registers indicate * * spurious messages from other widets which were detected after the * * header was captured.. * - * * + * * ************************************************************************/ typedef union ii_ixsm_u { - uint64_t ii_ixsm_regval; - struct { - uint64_t i_byte_en : 32; - uint64_t i_reserved : 1; - uint64_t i_tag : 3; - uint64_t i_alt_pactyp : 4; - uint64_t i_bo : 1; - uint64_t i_error : 1; - uint64_t i_vbpm : 1; - uint64_t i_gbr : 1; - uint64_t i_ds : 2; - uint64_t i_ct : 1; - uint64_t i_tnum : 5; - uint64_t i_pactyp : 4; - uint64_t i_sidn : 4; - uint64_t i_didn : 4; + uint64_t ii_ixsm_regval; + struct { + uint64_t i_byte_en:32; + uint64_t i_reserved:1; + uint64_t i_tag:3; + uint64_t i_alt_pactyp:4; + uint64_t i_bo:1; + uint64_t i_error:1; + uint64_t i_vbpm:1; + uint64_t i_gbr:1; + uint64_t i_ds:2; + uint64_t i_ct:1; + uint64_t i_tnum:5; + uint64_t i_pactyp:4; + uint64_t i_sidn:4; + uint64_t i_didn:4; } ii_ixsm_fld_s; } ii_ixsm_u_t; - /************************************************************************ - * * + * * * This register contains the sideband bits of a spurious read * * response received from Crosstalk. * - * * + * * ************************************************************************/ typedef union ii_ixss_u { - uint64_t ii_ixss_regval; - struct { - uint64_t i_sideband : 8; - uint64_t i_rsvd : 55; - uint64_t i_valid : 1; + uint64_t ii_ixss_regval; + struct { + uint64_t i_sideband:8; + uint64_t i_rsvd:55; + uint64_t i_valid:1; } ii_ixss_fld_s; } ii_ixss_u_t; - /************************************************************************ - * * + * * * This register enables software to access the II LLP's test port. * * Refer to the LLP 2.5 documentation for an explanation of the test * * port. Software can write to this register to program the values * @@ -1490,27 +1443,26 @@ typedef union ii_ixss_u { * TestMask and TestSeed). Similarly, software can read from this * * register to obtain the values of the test port's status outputs * * (TestCBerr, TestValid and TestData). * - * * + * * ************************************************************************/ typedef union ii_ilct_u { - uint64_t ii_ilct_regval; - struct { - uint64_t i_test_seed : 20; - uint64_t i_test_mask : 8; - uint64_t i_test_data : 20; - uint64_t i_test_valid : 1; - uint64_t i_test_cberr : 1; - uint64_t i_test_flit : 3; - uint64_t i_test_clear : 1; - uint64_t i_test_err_capture : 1; - uint64_t i_rsvd : 9; + uint64_t ii_ilct_regval; + struct { + uint64_t i_test_seed:20; + uint64_t i_test_mask:8; + uint64_t i_test_data:20; + uint64_t i_test_valid:1; + uint64_t i_test_cberr:1; + uint64_t i_test_flit:3; + uint64_t i_test_clear:1; + uint64_t i_test_err_capture:1; + uint64_t i_rsvd:9; } ii_ilct_fld_s; } ii_ilct_u_t; - /************************************************************************ - * * + * * * If the II detects an illegal incoming Duplonet packet (request or * * reply) when VALID==0 in the IIEPH1 register, then it saves the * * contents of the packet's header flit in the IIEPH1 and IIEPH2 * @@ -1526,575 +1478,549 @@ typedef union ii_ilct_u { * packet when VALID==1 in the IIEPH1 register, then it merely sets * * the OVERRUN bit to indicate that a subsequent error has happened, * * and does nothing further. * - * * + * * ************************************************************************/ typedef union ii_iieph1_u { - uint64_t ii_iieph1_regval; - struct { - uint64_t i_command : 7; - uint64_t i_rsvd_5 : 1; - uint64_t i_suppl : 14; - uint64_t i_rsvd_4 : 1; - uint64_t i_source : 14; - uint64_t i_rsvd_3 : 1; - uint64_t i_err_type : 4; - uint64_t i_rsvd_2 : 4; - uint64_t i_overrun : 1; - uint64_t i_rsvd_1 : 3; - uint64_t i_valid : 1; - uint64_t i_rsvd : 13; + uint64_t ii_iieph1_regval; + struct { + uint64_t i_command:7; + uint64_t i_rsvd_5:1; + uint64_t i_suppl:14; + uint64_t i_rsvd_4:1; + uint64_t i_source:14; + uint64_t i_rsvd_3:1; + uint64_t i_err_type:4; + uint64_t i_rsvd_2:4; + uint64_t i_overrun:1; + uint64_t i_rsvd_1:3; + uint64_t i_valid:1; + uint64_t i_rsvd:13; } ii_iieph1_fld_s; } ii_iieph1_u_t; - /************************************************************************ - * * + * * * This register holds the Address field from the header flit of an * * incoming erroneous Duplonet packet, along with the tail bit which * * accompanied this header flit. This register is essentially an * * extension of IIEPH1. Two registers were necessary because the 64 * * bits available in only a single register were insufficient to * * capture the entire header flit of an erroneous packet. * - * * + * * ************************************************************************/ typedef union ii_iieph2_u { - uint64_t ii_iieph2_regval; - struct { - uint64_t i_rsvd_0 : 3; - uint64_t i_address : 47; - uint64_t i_rsvd_1 : 10; - uint64_t i_tail : 1; - uint64_t i_rsvd : 3; + uint64_t ii_iieph2_regval; + struct { + uint64_t i_rsvd_0:3; + uint64_t i_address:47; + uint64_t i_rsvd_1:10; + uint64_t i_tail:1; + uint64_t i_rsvd:3; } ii_iieph2_fld_s; } ii_iieph2_u_t; - /******************************/ - - /************************************************************************ - * * + * * * This register's value is a bit vector that guards access from SXBs * * to local registers within the II as well as to external Crosstalk * * widgets * - * * + * * ************************************************************************/ typedef union ii_islapr_u { - uint64_t ii_islapr_regval; - struct { - uint64_t i_region : 64; + uint64_t ii_islapr_regval; + struct { + uint64_t i_region:64; } ii_islapr_fld_s; } ii_islapr_u_t; - /************************************************************************ - * * + * * * A write to this register of the 56-bit value "Pup+Bun" will cause * * the bit in the ISLAPR register corresponding to the region of the * * requestor to be set (access allowed). ( - * * + * * ************************************************************************/ typedef union ii_islapo_u { - uint64_t ii_islapo_regval; - struct { - uint64_t i_io_sbx_ovrride : 56; - uint64_t i_rsvd : 8; + uint64_t ii_islapo_regval; + struct { + uint64_t i_io_sbx_ovrride:56; + uint64_t i_rsvd:8; } ii_islapo_fld_s; } ii_islapo_u_t; /************************************************************************ - * * + * * * Determines how long the wrapper will wait aftr an interrupt is * * initially issued from the II before it times out the outstanding * * interrupt and drops it from the interrupt queue. * - * * + * * ************************************************************************/ typedef union ii_iwi_u { - uint64_t ii_iwi_regval; - struct { - uint64_t i_prescale : 24; - uint64_t i_rsvd : 8; - uint64_t i_timeout : 8; - uint64_t i_rsvd1 : 8; - uint64_t i_intrpt_retry_period : 8; - uint64_t i_rsvd2 : 8; + uint64_t ii_iwi_regval; + struct { + uint64_t i_prescale:24; + uint64_t i_rsvd:8; + uint64_t i_timeout:8; + uint64_t i_rsvd1:8; + uint64_t i_intrpt_retry_period:8; + uint64_t i_rsvd2:8; } ii_iwi_fld_s; } ii_iwi_u_t; /************************************************************************ - * * + * * * Log errors which have occurred in the II wrapper. The errors are * * cleared by writing to the IECLR register. * - * * + * * ************************************************************************/ typedef union ii_iwel_u { - uint64_t ii_iwel_regval; - struct { - uint64_t i_intr_timed_out : 1; - uint64_t i_rsvd : 7; - uint64_t i_cam_overflow : 1; - uint64_t i_cam_read_miss : 1; - uint64_t i_rsvd1 : 2; - uint64_t i_ioq_rep_underflow : 1; - uint64_t i_ioq_req_underflow : 1; - uint64_t i_ioq_rep_overflow : 1; - uint64_t i_ioq_req_overflow : 1; - uint64_t i_iiq_rep_overflow : 1; - uint64_t i_iiq_req_overflow : 1; - uint64_t i_rsvd2 : 6; - uint64_t i_ii_xn_rep_cred_over_under: 1; - uint64_t i_ii_xn_req_cred_over_under: 1; - uint64_t i_rsvd3 : 6; - uint64_t i_ii_xn_invalid_cmd : 1; - uint64_t i_xn_ii_invalid_cmd : 1; - uint64_t i_rsvd4 : 30; + uint64_t ii_iwel_regval; + struct { + uint64_t i_intr_timed_out:1; + uint64_t i_rsvd:7; + uint64_t i_cam_overflow:1; + uint64_t i_cam_read_miss:1; + uint64_t i_rsvd1:2; + uint64_t i_ioq_rep_underflow:1; + uint64_t i_ioq_req_underflow:1; + uint64_t i_ioq_rep_overflow:1; + uint64_t i_ioq_req_overflow:1; + uint64_t i_iiq_rep_overflow:1; + uint64_t i_iiq_req_overflow:1; + uint64_t i_rsvd2:6; + uint64_t i_ii_xn_rep_cred_over_under:1; + uint64_t i_ii_xn_req_cred_over_under:1; + uint64_t i_rsvd3:6; + uint64_t i_ii_xn_invalid_cmd:1; + uint64_t i_xn_ii_invalid_cmd:1; + uint64_t i_rsvd4:30; } ii_iwel_fld_s; } ii_iwel_u_t; /************************************************************************ - * * + * * * Controls the II wrapper. * - * * + * * ************************************************************************/ typedef union ii_iwc_u { - uint64_t ii_iwc_regval; - struct { - uint64_t i_dma_byte_swap : 1; - uint64_t i_rsvd : 3; - uint64_t i_cam_read_lines_reset : 1; - uint64_t i_rsvd1 : 3; - uint64_t i_ii_xn_cred_over_under_log: 1; - uint64_t i_rsvd2 : 19; - uint64_t i_xn_rep_iq_depth : 5; - uint64_t i_rsvd3 : 3; - uint64_t i_xn_req_iq_depth : 5; - uint64_t i_rsvd4 : 3; - uint64_t i_iiq_depth : 6; - uint64_t i_rsvd5 : 12; - uint64_t i_force_rep_cred : 1; - uint64_t i_force_req_cred : 1; + uint64_t ii_iwc_regval; + struct { + uint64_t i_dma_byte_swap:1; + uint64_t i_rsvd:3; + uint64_t i_cam_read_lines_reset:1; + uint64_t i_rsvd1:3; + uint64_t i_ii_xn_cred_over_under_log:1; + uint64_t i_rsvd2:19; + uint64_t i_xn_rep_iq_depth:5; + uint64_t i_rsvd3:3; + uint64_t i_xn_req_iq_depth:5; + uint64_t i_rsvd4:3; + uint64_t i_iiq_depth:6; + uint64_t i_rsvd5:12; + uint64_t i_force_rep_cred:1; + uint64_t i_force_req_cred:1; } ii_iwc_fld_s; } ii_iwc_u_t; /************************************************************************ - * * + * * * Status in the II wrapper. * - * * + * * ************************************************************************/ typedef union ii_iws_u { - uint64_t ii_iws_regval; - struct { - uint64_t i_xn_rep_iq_credits : 5; - uint64_t i_rsvd : 3; - uint64_t i_xn_req_iq_credits : 5; - uint64_t i_rsvd1 : 51; + uint64_t ii_iws_regval; + struct { + uint64_t i_xn_rep_iq_credits:5; + uint64_t i_rsvd:3; + uint64_t i_xn_req_iq_credits:5; + uint64_t i_rsvd1:51; } ii_iws_fld_s; } ii_iws_u_t; /************************************************************************ - * * + * * * Masks errors in the IWEL register. * - * * + * * ************************************************************************/ typedef union ii_iweim_u { - uint64_t ii_iweim_regval; - struct { - uint64_t i_intr_timed_out : 1; - uint64_t i_rsvd : 7; - uint64_t i_cam_overflow : 1; - uint64_t i_cam_read_miss : 1; - uint64_t i_rsvd1 : 2; - uint64_t i_ioq_rep_underflow : 1; - uint64_t i_ioq_req_underflow : 1; - uint64_t i_ioq_rep_overflow : 1; - uint64_t i_ioq_req_overflow : 1; - uint64_t i_iiq_rep_overflow : 1; - uint64_t i_iiq_req_overflow : 1; - uint64_t i_rsvd2 : 6; - uint64_t i_ii_xn_rep_cred_overflow : 1; - uint64_t i_ii_xn_req_cred_overflow : 1; - uint64_t i_rsvd3 : 6; - uint64_t i_ii_xn_invalid_cmd : 1; - uint64_t i_xn_ii_invalid_cmd : 1; - uint64_t i_rsvd4 : 30; + uint64_t ii_iweim_regval; + struct { + uint64_t i_intr_timed_out:1; + uint64_t i_rsvd:7; + uint64_t i_cam_overflow:1; + uint64_t i_cam_read_miss:1; + uint64_t i_rsvd1:2; + uint64_t i_ioq_rep_underflow:1; + uint64_t i_ioq_req_underflow:1; + uint64_t i_ioq_rep_overflow:1; + uint64_t i_ioq_req_overflow:1; + uint64_t i_iiq_rep_overflow:1; + uint64_t i_iiq_req_overflow:1; + uint64_t i_rsvd2:6; + uint64_t i_ii_xn_rep_cred_overflow:1; + uint64_t i_ii_xn_req_cred_overflow:1; + uint64_t i_rsvd3:6; + uint64_t i_ii_xn_invalid_cmd:1; + uint64_t i_xn_ii_invalid_cmd:1; + uint64_t i_rsvd4:30; } ii_iweim_fld_s; } ii_iweim_u_t; - /************************************************************************ - * * + * * * A write to this register causes a particular field in the * * corresponding widget's PRB entry to be adjusted up or down by 1. * * This counter should be used when recovering from error and reset * * conditions. Note that software would be capable of causing * * inadvertent overflow or underflow of these counters. * - * * + * * ************************************************************************/ typedef union ii_ipca_u { - uint64_t ii_ipca_regval; - struct { - uint64_t i_wid : 4; - uint64_t i_adjust : 1; - uint64_t i_rsvd_1 : 3; - uint64_t i_field : 2; - uint64_t i_rsvd : 54; + uint64_t ii_ipca_regval; + struct { + uint64_t i_wid:4; + uint64_t i_adjust:1; + uint64_t i_rsvd_1:3; + uint64_t i_field:2; + uint64_t i_rsvd:54; } ii_ipca_fld_s; } ii_ipca_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ - typedef union ii_iprte0a_u { - uint64_t ii_iprte0a_regval; - struct { - uint64_t i_rsvd_1 : 54; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; + uint64_t ii_iprte0a_regval; + struct { + uint64_t i_rsvd_1:54; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; } ii_iprte0a_fld_s; } ii_iprte0a_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte1a_u { - uint64_t ii_iprte1a_regval; - struct { - uint64_t i_rsvd_1 : 54; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; + uint64_t ii_iprte1a_regval; + struct { + uint64_t i_rsvd_1:54; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; } ii_iprte1a_fld_s; } ii_iprte1a_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte2a_u { - uint64_t ii_iprte2a_regval; - struct { - uint64_t i_rsvd_1 : 54; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; + uint64_t ii_iprte2a_regval; + struct { + uint64_t i_rsvd_1:54; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; } ii_iprte2a_fld_s; } ii_iprte2a_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte3a_u { - uint64_t ii_iprte3a_regval; - struct { - uint64_t i_rsvd_1 : 54; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; + uint64_t ii_iprte3a_regval; + struct { + uint64_t i_rsvd_1:54; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; } ii_iprte3a_fld_s; } ii_iprte3a_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte4a_u { - uint64_t ii_iprte4a_regval; - struct { - uint64_t i_rsvd_1 : 54; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; + uint64_t ii_iprte4a_regval; + struct { + uint64_t i_rsvd_1:54; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; } ii_iprte4a_fld_s; } ii_iprte4a_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte5a_u { - uint64_t ii_iprte5a_regval; - struct { - uint64_t i_rsvd_1 : 54; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; + uint64_t ii_iprte5a_regval; + struct { + uint64_t i_rsvd_1:54; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; } ii_iprte5a_fld_s; } ii_iprte5a_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte6a_u { - uint64_t ii_iprte6a_regval; - struct { - uint64_t i_rsvd_1 : 54; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; + uint64_t ii_iprte6a_regval; + struct { + uint64_t i_rsvd_1:54; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; } ii_iprte6a_fld_s; } ii_iprte6a_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte7a_u { - uint64_t ii_iprte7a_regval; - struct { - uint64_t i_rsvd_1 : 54; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; - } ii_iprtea7_fld_s; + uint64_t ii_iprte7a_regval; + struct { + uint64_t i_rsvd_1:54; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; + } ii_iprtea7_fld_s; } ii_iprte7a_u_t; - - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ - typedef union ii_iprte0b_u { - uint64_t ii_iprte0b_regval; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_address : 47; - uint64_t i_init : 3; - uint64_t i_source : 11; + uint64_t ii_iprte0b_regval; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_address:47; + uint64_t i_init:3; + uint64_t i_source:11; } ii_iprte0b_fld_s; } ii_iprte0b_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte1b_u { - uint64_t ii_iprte1b_regval; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_address : 47; - uint64_t i_init : 3; - uint64_t i_source : 11; + uint64_t ii_iprte1b_regval; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_address:47; + uint64_t i_init:3; + uint64_t i_source:11; } ii_iprte1b_fld_s; } ii_iprte1b_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte2b_u { - uint64_t ii_iprte2b_regval; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_address : 47; - uint64_t i_init : 3; - uint64_t i_source : 11; + uint64_t ii_iprte2b_regval; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_address:47; + uint64_t i_init:3; + uint64_t i_source:11; } ii_iprte2b_fld_s; } ii_iprte2b_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte3b_u { - uint64_t ii_iprte3b_regval; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_address : 47; - uint64_t i_init : 3; - uint64_t i_source : 11; + uint64_t ii_iprte3b_regval; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_address:47; + uint64_t i_init:3; + uint64_t i_source:11; } ii_iprte3b_fld_s; } ii_iprte3b_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte4b_u { - uint64_t ii_iprte4b_regval; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_address : 47; - uint64_t i_init : 3; - uint64_t i_source : 11; + uint64_t ii_iprte4b_regval; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_address:47; + uint64_t i_init:3; + uint64_t i_source:11; } ii_iprte4b_fld_s; } ii_iprte4b_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte5b_u { - uint64_t ii_iprte5b_regval; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_address : 47; - uint64_t i_init : 3; - uint64_t i_source : 11; + uint64_t ii_iprte5b_regval; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_address:47; + uint64_t i_init:3; + uint64_t i_source:11; } ii_iprte5b_fld_s; } ii_iprte5b_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte6b_u { - uint64_t ii_iprte6b_regval; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_address : 47; - uint64_t i_init : 3; - uint64_t i_source : 11; + uint64_t ii_iprte6b_regval; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_address:47; + uint64_t i_init:3; + uint64_t i_source:11; } ii_iprte6b_fld_s; } ii_iprte6b_u_t; - /************************************************************************ - * * + * * * There are 8 instances of this register. This register contains * * the information that the II has to remember once it has launched a * * PIO Read operation. The contents are used to form the correct * * Router Network packet and direct the Crosstalk reply to the * * appropriate processor. * - * * + * * ************************************************************************/ typedef union ii_iprte7b_u { - uint64_t ii_iprte7b_regval; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_address : 47; - uint64_t i_init : 3; - uint64_t i_source : 11; - } ii_iprte7b_fld_s; + uint64_t ii_iprte7b_regval; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_address:47; + uint64_t i_init:3; + uint64_t i_source:11; + } ii_iprte7b_fld_s; } ii_iprte7b_u_t; - /************************************************************************ - * * + * * * Description: SHub II contains a feature which did not exist in * * the Hub which automatically cleans up after a Read Response * * timeout, including deallocation of the IPRTE and recovery of IBuf * @@ -2108,23 +2034,22 @@ typedef union ii_iprte7b_u { * Note that this register does not affect the contents of the IPRTE * * registers. The Valid bits in those registers have to be * * specifically turned off by software. * - * * + * * ************************************************************************/ typedef union ii_ipdr_u { - uint64_t ii_ipdr_regval; - struct { - uint64_t i_te : 3; - uint64_t i_rsvd_1 : 1; - uint64_t i_pnd : 1; - uint64_t i_init_rpcnt : 1; - uint64_t i_rsvd : 58; + uint64_t ii_ipdr_regval; + struct { + uint64_t i_te:3; + uint64_t i_rsvd_1:1; + uint64_t i_pnd:1; + uint64_t i_init_rpcnt:1; + uint64_t i_rsvd:58; } ii_ipdr_fld_s; } ii_ipdr_u_t; - /************************************************************************ - * * + * * * A write to this register causes a CRB entry to be returned to the * * queue of free CRBs. The entry should have previously been cleared * * (mark bit) via backdoor access to the pertinent CRB entry. This * @@ -2137,21 +2062,20 @@ typedef union ii_ipdr_u { * software clears the mark bit, and finally 4) software writes to * * the ICDR register to return the CRB entry to the list of free CRB * * entries. * - * * + * * ************************************************************************/ typedef union ii_icdr_u { - uint64_t ii_icdr_regval; - struct { - uint64_t i_crb_num : 4; - uint64_t i_pnd : 1; - uint64_t i_rsvd : 59; + uint64_t ii_icdr_regval; + struct { + uint64_t i_crb_num:4; + uint64_t i_pnd:1; + uint64_t i_rsvd:59; } ii_icdr_fld_s; } ii_icdr_u_t; - /************************************************************************ - * * + * * * This register provides debug access to two FIFOs inside of II. * * Both IOQ_MAX* fields of this register contain the instantaneous * * depth (in units of the number of available entries) of the * @@ -2164,130 +2088,124 @@ typedef union ii_icdr_u { * this register is written. If there are any active entries in any * * of these FIFOs when this register is written, the results are * * undefined. * - * * + * * ************************************************************************/ typedef union ii_ifdr_u { - uint64_t ii_ifdr_regval; - struct { - uint64_t i_ioq_max_rq : 7; - uint64_t i_set_ioq_rq : 1; - uint64_t i_ioq_max_rp : 7; - uint64_t i_set_ioq_rp : 1; - uint64_t i_rsvd : 48; + uint64_t ii_ifdr_regval; + struct { + uint64_t i_ioq_max_rq:7; + uint64_t i_set_ioq_rq:1; + uint64_t i_ioq_max_rp:7; + uint64_t i_set_ioq_rp:1; + uint64_t i_rsvd:48; } ii_ifdr_fld_s; } ii_ifdr_u_t; - /************************************************************************ - * * + * * * This register allows the II to become sluggish in removing * * messages from its inbound queue (IIQ). This will cause messages to * * back up in either virtual channel. Disabling the "molasses" mode * * subsequently allows the II to be tested under stress. In the * * sluggish ("Molasses") mode, the localized effects of congestion * * can be observed. * - * * + * * ************************************************************************/ typedef union ii_iiap_u { - uint64_t ii_iiap_regval; - struct { - uint64_t i_rq_mls : 6; - uint64_t i_rsvd_1 : 2; - uint64_t i_rp_mls : 6; - uint64_t i_rsvd : 50; - } ii_iiap_fld_s; + uint64_t ii_iiap_regval; + struct { + uint64_t i_rq_mls:6; + uint64_t i_rsvd_1:2; + uint64_t i_rp_mls:6; + uint64_t i_rsvd:50; + } ii_iiap_fld_s; } ii_iiap_u_t; - /************************************************************************ - * * + * * * This register allows several parameters of CRB operation to be * * set. Note that writing to this register can have catastrophic side * * effects, if the CRB is not quiescent, i.e. if the CRB is * * processing protocol messages when the write occurs. * - * * + * * ************************************************************************/ typedef union ii_icmr_u { - uint64_t ii_icmr_regval; - struct { - uint64_t i_sp_msg : 1; - uint64_t i_rd_hdr : 1; - uint64_t i_rsvd_4 : 2; - uint64_t i_c_cnt : 4; - uint64_t i_rsvd_3 : 4; - uint64_t i_clr_rqpd : 1; - uint64_t i_clr_rppd : 1; - uint64_t i_rsvd_2 : 2; - uint64_t i_fc_cnt : 4; - uint64_t i_crb_vld : 15; - uint64_t i_crb_mark : 15; - uint64_t i_rsvd_1 : 2; - uint64_t i_precise : 1; - uint64_t i_rsvd : 11; + uint64_t ii_icmr_regval; + struct { + uint64_t i_sp_msg:1; + uint64_t i_rd_hdr:1; + uint64_t i_rsvd_4:2; + uint64_t i_c_cnt:4; + uint64_t i_rsvd_3:4; + uint64_t i_clr_rqpd:1; + uint64_t i_clr_rppd:1; + uint64_t i_rsvd_2:2; + uint64_t i_fc_cnt:4; + uint64_t i_crb_vld:15; + uint64_t i_crb_mark:15; + uint64_t i_rsvd_1:2; + uint64_t i_precise:1; + uint64_t i_rsvd:11; } ii_icmr_fld_s; } ii_icmr_u_t; - /************************************************************************ - * * + * * * This register allows control of the table portion of the CRB * * logic via software. Control operations from this register have * * priority over all incoming Crosstalk or BTE requests. * - * * + * * ************************************************************************/ typedef union ii_iccr_u { - uint64_t ii_iccr_regval; - struct { - uint64_t i_crb_num : 4; - uint64_t i_rsvd_1 : 4; - uint64_t i_cmd : 8; - uint64_t i_pending : 1; - uint64_t i_rsvd : 47; + uint64_t ii_iccr_regval; + struct { + uint64_t i_crb_num:4; + uint64_t i_rsvd_1:4; + uint64_t i_cmd:8; + uint64_t i_pending:1; + uint64_t i_rsvd:47; } ii_iccr_fld_s; } ii_iccr_u_t; - /************************************************************************ - * * + * * * This register allows the maximum timeout value to be programmed. * - * * + * * ************************************************************************/ typedef union ii_icto_u { - uint64_t ii_icto_regval; - struct { - uint64_t i_timeout : 8; - uint64_t i_rsvd : 56; + uint64_t ii_icto_regval; + struct { + uint64_t i_timeout:8; + uint64_t i_rsvd:56; } ii_icto_fld_s; } ii_icto_u_t; - /************************************************************************ - * * + * * * This register allows the timeout prescalar to be programmed. An * * internal counter is associated with this register. When the * * internal counter reaches the value of the PRESCALE field, the * * timer registers in all valid CRBs are incremented (CRBx_D[TIMEOUT] * * field). The internal counter resets to zero, and then continues * * counting. * - * * + * * ************************************************************************/ typedef union ii_ictp_u { - uint64_t ii_ictp_regval; - struct { - uint64_t i_prescale : 24; - uint64_t i_rsvd : 40; + uint64_t ii_ictp_regval; + struct { + uint64_t i_prescale:24; + uint64_t i_rsvd:40; } ii_ictp_fld_s; } ii_ictp_u_t; - /************************************************************************ - * * + * * * Description: There are 15 CRB Entries (ICRB0 to ICRBE) that are * * used for Crosstalk operations (both cacheline and partial * * operations) or BTE/IO. Because the CRB entries are very wide, five * @@ -2306,243 +2224,234 @@ typedef union ii_ictp_u { * recovering any potential error state from before the reset). * * The following four tables summarize the format for the four * * registers that are used for each ICRB# Entry. * - * * + * * ************************************************************************/ typedef union ii_icrb0_a_u { - uint64_t ii_icrb0_a_regval; - struct { - uint64_t ia_iow : 1; - uint64_t ia_vld : 1; - uint64_t ia_addr : 47; - uint64_t ia_tnum : 5; - uint64_t ia_sidn : 4; - uint64_t ia_rsvd : 6; + uint64_t ii_icrb0_a_regval; + struct { + uint64_t ia_iow:1; + uint64_t ia_vld:1; + uint64_t ia_addr:47; + uint64_t ia_tnum:5; + uint64_t ia_sidn:4; + uint64_t ia_rsvd:6; } ii_icrb0_a_fld_s; } ii_icrb0_a_u_t; - /************************************************************************ - * * + * * * Description: There are 15 CRB Entries (ICRB0 to ICRBE) that are * * used for Crosstalk operations (both cacheline and partial * * operations) or BTE/IO. Because the CRB entries are very wide, five * * registers (_A to _E) are required to read and write each entry. * - * * + * * ************************************************************************/ typedef union ii_icrb0_b_u { - uint64_t ii_icrb0_b_regval; - struct { - uint64_t ib_xt_err : 1; - uint64_t ib_mark : 1; - uint64_t ib_ln_uce : 1; - uint64_t ib_errcode : 3; - uint64_t ib_error : 1; - uint64_t ib_stall__bte_1 : 1; - uint64_t ib_stall__bte_0 : 1; - uint64_t ib_stall__intr : 1; - uint64_t ib_stall_ib : 1; - uint64_t ib_intvn : 1; - uint64_t ib_wb : 1; - uint64_t ib_hold : 1; - uint64_t ib_ack : 1; - uint64_t ib_resp : 1; - uint64_t ib_ack_cnt : 11; - uint64_t ib_rsvd : 7; - uint64_t ib_exc : 5; - uint64_t ib_init : 3; - uint64_t ib_imsg : 8; - uint64_t ib_imsgtype : 2; - uint64_t ib_use_old : 1; - uint64_t ib_rsvd_1 : 11; + uint64_t ii_icrb0_b_regval; + struct { + uint64_t ib_xt_err:1; + uint64_t ib_mark:1; + uint64_t ib_ln_uce:1; + uint64_t ib_errcode:3; + uint64_t ib_error:1; + uint64_t ib_stall__bte_1:1; + uint64_t ib_stall__bte_0:1; + uint64_t ib_stall__intr:1; + uint64_t ib_stall_ib:1; + uint64_t ib_intvn:1; + uint64_t ib_wb:1; + uint64_t ib_hold:1; + uint64_t ib_ack:1; + uint64_t ib_resp:1; + uint64_t ib_ack_cnt:11; + uint64_t ib_rsvd:7; + uint64_t ib_exc:5; + uint64_t ib_init:3; + uint64_t ib_imsg:8; + uint64_t ib_imsgtype:2; + uint64_t ib_use_old:1; + uint64_t ib_rsvd_1:11; } ii_icrb0_b_fld_s; } ii_icrb0_b_u_t; - /************************************************************************ - * * + * * * Description: There are 15 CRB Entries (ICRB0 to ICRBE) that are * * used for Crosstalk operations (both cacheline and partial * * operations) or BTE/IO. Because the CRB entries are very wide, five * * registers (_A to _E) are required to read and write each entry. * - * * + * * ************************************************************************/ typedef union ii_icrb0_c_u { - uint64_t ii_icrb0_c_regval; - struct { - uint64_t ic_source : 15; - uint64_t ic_size : 2; - uint64_t ic_ct : 1; - uint64_t ic_bte_num : 1; - uint64_t ic_gbr : 1; - uint64_t ic_resprqd : 1; - uint64_t ic_bo : 1; - uint64_t ic_suppl : 15; - uint64_t ic_rsvd : 27; + uint64_t ii_icrb0_c_regval; + struct { + uint64_t ic_source:15; + uint64_t ic_size:2; + uint64_t ic_ct:1; + uint64_t ic_bte_num:1; + uint64_t ic_gbr:1; + uint64_t ic_resprqd:1; + uint64_t ic_bo:1; + uint64_t ic_suppl:15; + uint64_t ic_rsvd:27; } ii_icrb0_c_fld_s; } ii_icrb0_c_u_t; - /************************************************************************ - * * + * * * Description: There are 15 CRB Entries (ICRB0 to ICRBE) that are * * used for Crosstalk operations (both cacheline and partial * * operations) or BTE/IO. Because the CRB entries are very wide, five * * registers (_A to _E) are required to read and write each entry. * - * * + * * ************************************************************************/ typedef union ii_icrb0_d_u { - uint64_t ii_icrb0_d_regval; - struct { - uint64_t id_pa_be : 43; - uint64_t id_bte_op : 1; - uint64_t id_pr_psc : 4; - uint64_t id_pr_cnt : 4; - uint64_t id_sleep : 1; - uint64_t id_rsvd : 11; + uint64_t ii_icrb0_d_regval; + struct { + uint64_t id_pa_be:43; + uint64_t id_bte_op:1; + uint64_t id_pr_psc:4; + uint64_t id_pr_cnt:4; + uint64_t id_sleep:1; + uint64_t id_rsvd:11; } ii_icrb0_d_fld_s; } ii_icrb0_d_u_t; - /************************************************************************ - * * + * * * Description: There are 15 CRB Entries (ICRB0 to ICRBE) that are * * used for Crosstalk operations (both cacheline and partial * * operations) or BTE/IO. Because the CRB entries are very wide, five * * registers (_A to _E) are required to read and write each entry. * - * * + * * ************************************************************************/ typedef union ii_icrb0_e_u { - uint64_t ii_icrb0_e_regval; - struct { - uint64_t ie_timeout : 8; - uint64_t ie_context : 15; - uint64_t ie_rsvd : 1; - uint64_t ie_tvld : 1; - uint64_t ie_cvld : 1; - uint64_t ie_rsvd_0 : 38; + uint64_t ii_icrb0_e_regval; + struct { + uint64_t ie_timeout:8; + uint64_t ie_context:15; + uint64_t ie_rsvd:1; + uint64_t ie_tvld:1; + uint64_t ie_cvld:1; + uint64_t ie_rsvd_0:38; } ii_icrb0_e_fld_s; } ii_icrb0_e_u_t; - /************************************************************************ - * * + * * * This register contains the lower 64 bits of the header of the * * spurious message captured by II. Valid when the SP_MSG bit in ICMR * * register is set. * - * * + * * ************************************************************************/ typedef union ii_icsml_u { - uint64_t ii_icsml_regval; - struct { - uint64_t i_tt_addr : 47; - uint64_t i_newsuppl_ex : 14; - uint64_t i_reserved : 2; - uint64_t i_overflow : 1; + uint64_t ii_icsml_regval; + struct { + uint64_t i_tt_addr:47; + uint64_t i_newsuppl_ex:14; + uint64_t i_reserved:2; + uint64_t i_overflow:1; } ii_icsml_fld_s; } ii_icsml_u_t; - /************************************************************************ - * * + * * * This register contains the middle 64 bits of the header of the * * spurious message captured by II. Valid when the SP_MSG bit in ICMR * * register is set. * - * * + * * ************************************************************************/ typedef union ii_icsmm_u { - uint64_t ii_icsmm_regval; - struct { - uint64_t i_tt_ack_cnt : 11; - uint64_t i_reserved : 53; + uint64_t ii_icsmm_regval; + struct { + uint64_t i_tt_ack_cnt:11; + uint64_t i_reserved:53; } ii_icsmm_fld_s; } ii_icsmm_u_t; - /************************************************************************ - * * + * * * This register contains the microscopic state, all the inputs to * * the protocol table, captured with the spurious message. Valid when * * the SP_MSG bit in the ICMR register is set. * - * * + * * ************************************************************************/ typedef union ii_icsmh_u { - uint64_t ii_icsmh_regval; - struct { - uint64_t i_tt_vld : 1; - uint64_t i_xerr : 1; - uint64_t i_ft_cwact_o : 1; - uint64_t i_ft_wact_o : 1; - uint64_t i_ft_active_o : 1; - uint64_t i_sync : 1; - uint64_t i_mnusg : 1; - uint64_t i_mnusz : 1; - uint64_t i_plusz : 1; - uint64_t i_plusg : 1; - uint64_t i_tt_exc : 5; - uint64_t i_tt_wb : 1; - uint64_t i_tt_hold : 1; - uint64_t i_tt_ack : 1; - uint64_t i_tt_resp : 1; - uint64_t i_tt_intvn : 1; - uint64_t i_g_stall_bte1 : 1; - uint64_t i_g_stall_bte0 : 1; - uint64_t i_g_stall_il : 1; - uint64_t i_g_stall_ib : 1; - uint64_t i_tt_imsg : 8; - uint64_t i_tt_imsgtype : 2; - uint64_t i_tt_use_old : 1; - uint64_t i_tt_respreqd : 1; - uint64_t i_tt_bte_num : 1; - uint64_t i_cbn : 1; - uint64_t i_match : 1; - uint64_t i_rpcnt_lt_34 : 1; - uint64_t i_rpcnt_ge_34 : 1; - uint64_t i_rpcnt_lt_18 : 1; - uint64_t i_rpcnt_ge_18 : 1; - uint64_t i_rpcnt_lt_2 : 1; - uint64_t i_rpcnt_ge_2 : 1; - uint64_t i_rqcnt_lt_18 : 1; - uint64_t i_rqcnt_ge_18 : 1; - uint64_t i_rqcnt_lt_2 : 1; - uint64_t i_rqcnt_ge_2 : 1; - uint64_t i_tt_device : 7; - uint64_t i_tt_init : 3; - uint64_t i_reserved : 5; + uint64_t ii_icsmh_regval; + struct { + uint64_t i_tt_vld:1; + uint64_t i_xerr:1; + uint64_t i_ft_cwact_o:1; + uint64_t i_ft_wact_o:1; + uint64_t i_ft_active_o:1; + uint64_t i_sync:1; + uint64_t i_mnusg:1; + uint64_t i_mnusz:1; + uint64_t i_plusz:1; + uint64_t i_plusg:1; + uint64_t i_tt_exc:5; + uint64_t i_tt_wb:1; + uint64_t i_tt_hold:1; + uint64_t i_tt_ack:1; + uint64_t i_tt_resp:1; + uint64_t i_tt_intvn:1; + uint64_t i_g_stall_bte1:1; + uint64_t i_g_stall_bte0:1; + uint64_t i_g_stall_il:1; + uint64_t i_g_stall_ib:1; + uint64_t i_tt_imsg:8; + uint64_t i_tt_imsgtype:2; + uint64_t i_tt_use_old:1; + uint64_t i_tt_respreqd:1; + uint64_t i_tt_bte_num:1; + uint64_t i_cbn:1; + uint64_t i_match:1; + uint64_t i_rpcnt_lt_34:1; + uint64_t i_rpcnt_ge_34:1; + uint64_t i_rpcnt_lt_18:1; + uint64_t i_rpcnt_ge_18:1; + uint64_t i_rpcnt_lt_2:1; + uint64_t i_rpcnt_ge_2:1; + uint64_t i_rqcnt_lt_18:1; + uint64_t i_rqcnt_ge_18:1; + uint64_t i_rqcnt_lt_2:1; + uint64_t i_rqcnt_ge_2:1; + uint64_t i_tt_device:7; + uint64_t i_tt_init:3; + uint64_t i_reserved:5; } ii_icsmh_fld_s; } ii_icsmh_u_t; - /************************************************************************ - * * + * * * The Shub DEBUG unit provides a 3-bit selection signal to the * * II core and a 3-bit selection signal to the fsbclk domain in the II * * wrapper. * - * * + * * ************************************************************************/ typedef union ii_idbss_u { - uint64_t ii_idbss_regval; - struct { - uint64_t i_iioclk_core_submenu : 3; - uint64_t i_rsvd : 5; - uint64_t i_fsbclk_wrapper_submenu : 3; - uint64_t i_rsvd_1 : 5; - uint64_t i_iioclk_menu : 5; - uint64_t i_rsvd_2 : 43; + uint64_t ii_idbss_regval; + struct { + uint64_t i_iioclk_core_submenu:3; + uint64_t i_rsvd:5; + uint64_t i_fsbclk_wrapper_submenu:3; + uint64_t i_rsvd_1:5; + uint64_t i_iioclk_menu:5; + uint64_t i_rsvd_2:43; } ii_idbss_fld_s; } ii_idbss_u_t; - /************************************************************************ - * * + * * * Description: This register is used to set up the length for a * * transfer and then to monitor the progress of that transfer. This * * register needs to be initialized before a transfer is started. A * @@ -2553,63 +2462,60 @@ typedef union ii_idbss_u { * transfer completes, hardware will clear the Busy bit. The length * * field will also contain the number of cache lines left to be * * transferred. * - * * + * * ************************************************************************/ typedef union ii_ibls0_u { - uint64_t ii_ibls0_regval; - struct { - uint64_t i_length : 16; - uint64_t i_error : 1; - uint64_t i_rsvd_1 : 3; - uint64_t i_busy : 1; - uint64_t i_rsvd : 43; + uint64_t ii_ibls0_regval; + struct { + uint64_t i_length:16; + uint64_t i_error:1; + uint64_t i_rsvd_1:3; + uint64_t i_busy:1; + uint64_t i_rsvd:43; } ii_ibls0_fld_s; } ii_ibls0_u_t; - /************************************************************************ - * * + * * * This register should be loaded before a transfer is started. The * * address to be loaded in bits 39:0 is the 40-bit TRex+ physical * * address as described in Section 1.3, Figure2 and Figure3. Since * * the bottom 7 bits of the address are always taken to be zero, BTE * * transfers are always cacheline-aligned. * - * * + * * ************************************************************************/ typedef union ii_ibsa0_u { - uint64_t ii_ibsa0_regval; - struct { - uint64_t i_rsvd_1 : 7; - uint64_t i_addr : 42; - uint64_t i_rsvd : 15; + uint64_t ii_ibsa0_regval; + struct { + uint64_t i_rsvd_1:7; + uint64_t i_addr:42; + uint64_t i_rsvd:15; } ii_ibsa0_fld_s; } ii_ibsa0_u_t; - /************************************************************************ - * * + * * * This register should be loaded before a transfer is started. The * * address to be loaded in bits 39:0 is the 40-bit TRex+ physical * * address as described in Section 1.3, Figure2 and Figure3. Since * * the bottom 7 bits of the address are always taken to be zero, BTE * * transfers are always cacheline-aligned. * - * * + * * ************************************************************************/ typedef union ii_ibda0_u { - uint64_t ii_ibda0_regval; - struct { - uint64_t i_rsvd_1 : 7; - uint64_t i_addr : 42; - uint64_t i_rsvd : 15; + uint64_t ii_ibda0_regval; + struct { + uint64_t i_rsvd_1:7; + uint64_t i_addr:42; + uint64_t i_rsvd:15; } ii_ibda0_fld_s; } ii_ibda0_u_t; - /************************************************************************ - * * + * * * Writing to this register sets up the attributes of the transfer * * and initiates the transfer operation. Reading this register has * * the side effect of terminating any transfer in progress. Note: * @@ -2617,61 +2523,58 @@ typedef union ii_ibda0_u { * other BTE. If a BTE stream has to be stopped (due to error * * handling for example), both BTE streams should be stopped and * * their transfers discarded. * - * * + * * ************************************************************************/ typedef union ii_ibct0_u { - uint64_t ii_ibct0_regval; - struct { - uint64_t i_zerofill : 1; - uint64_t i_rsvd_2 : 3; - uint64_t i_notify : 1; - uint64_t i_rsvd_1 : 3; - uint64_t i_poison : 1; - uint64_t i_rsvd : 55; + uint64_t ii_ibct0_regval; + struct { + uint64_t i_zerofill:1; + uint64_t i_rsvd_2:3; + uint64_t i_notify:1; + uint64_t i_rsvd_1:3; + uint64_t i_poison:1; + uint64_t i_rsvd:55; } ii_ibct0_fld_s; } ii_ibct0_u_t; - /************************************************************************ - * * + * * * This register contains the address to which the WINV is sent. * * This address has to be cache line aligned. * - * * + * * ************************************************************************/ typedef union ii_ibna0_u { - uint64_t ii_ibna0_regval; - struct { - uint64_t i_rsvd_1 : 7; - uint64_t i_addr : 42; - uint64_t i_rsvd : 15; + uint64_t ii_ibna0_regval; + struct { + uint64_t i_rsvd_1:7; + uint64_t i_addr:42; + uint64_t i_rsvd:15; } ii_ibna0_fld_s; } ii_ibna0_u_t; - /************************************************************************ - * * + * * * This register contains the programmable level as well as the node * * ID and PI unit of the processor to which the interrupt will be * - * sent. * - * * + * sent. * + * * ************************************************************************/ typedef union ii_ibia0_u { - uint64_t ii_ibia0_regval; - struct { - uint64_t i_rsvd_2 : 1; - uint64_t i_node_id : 11; - uint64_t i_rsvd_1 : 4; - uint64_t i_level : 7; - uint64_t i_rsvd : 41; + uint64_t ii_ibia0_regval; + struct { + uint64_t i_rsvd_2:1; + uint64_t i_node_id:11; + uint64_t i_rsvd_1:4; + uint64_t i_level:7; + uint64_t i_rsvd:41; } ii_ibia0_fld_s; } ii_ibia0_u_t; - /************************************************************************ - * * + * * * Description: This register is used to set up the length for a * * transfer and then to monitor the progress of that transfer. This * * register needs to be initialized before a transfer is started. A * @@ -2682,63 +2585,60 @@ typedef union ii_ibia0_u { * transfer completes, hardware will clear the Busy bit. The length * * field will also contain the number of cache lines left to be * * transferred. * - * * + * * ************************************************************************/ typedef union ii_ibls1_u { - uint64_t ii_ibls1_regval; - struct { - uint64_t i_length : 16; - uint64_t i_error : 1; - uint64_t i_rsvd_1 : 3; - uint64_t i_busy : 1; - uint64_t i_rsvd : 43; + uint64_t ii_ibls1_regval; + struct { + uint64_t i_length:16; + uint64_t i_error:1; + uint64_t i_rsvd_1:3; + uint64_t i_busy:1; + uint64_t i_rsvd:43; } ii_ibls1_fld_s; } ii_ibls1_u_t; - /************************************************************************ - * * + * * * This register should be loaded before a transfer is started. The * * address to be loaded in bits 39:0 is the 40-bit TRex+ physical * * address as described in Section 1.3, Figure2 and Figure3. Since * * the bottom 7 bits of the address are always taken to be zero, BTE * * transfers are always cacheline-aligned. * - * * + * * ************************************************************************/ typedef union ii_ibsa1_u { - uint64_t ii_ibsa1_regval; - struct { - uint64_t i_rsvd_1 : 7; - uint64_t i_addr : 33; - uint64_t i_rsvd : 24; + uint64_t ii_ibsa1_regval; + struct { + uint64_t i_rsvd_1:7; + uint64_t i_addr:33; + uint64_t i_rsvd:24; } ii_ibsa1_fld_s; } ii_ibsa1_u_t; - /************************************************************************ - * * + * * * This register should be loaded before a transfer is started. The * * address to be loaded in bits 39:0 is the 40-bit TRex+ physical * * address as described in Section 1.3, Figure2 and Figure3. Since * * the bottom 7 bits of the address are always taken to be zero, BTE * * transfers are always cacheline-aligned. * - * * + * * ************************************************************************/ typedef union ii_ibda1_u { - uint64_t ii_ibda1_regval; - struct { - uint64_t i_rsvd_1 : 7; - uint64_t i_addr : 33; - uint64_t i_rsvd : 24; + uint64_t ii_ibda1_regval; + struct { + uint64_t i_rsvd_1:7; + uint64_t i_addr:33; + uint64_t i_rsvd:24; } ii_ibda1_fld_s; } ii_ibda1_u_t; - /************************************************************************ - * * + * * * Writing to this register sets up the attributes of the transfer * * and initiates the transfer operation. Reading this register has * * the side effect of terminating any transfer in progress. Note: * @@ -2746,61 +2646,58 @@ typedef union ii_ibda1_u { * other BTE. If a BTE stream has to be stopped (due to error * * handling for example), both BTE streams should be stopped and * * their transfers discarded. * - * * + * * ************************************************************************/ typedef union ii_ibct1_u { - uint64_t ii_ibct1_regval; - struct { - uint64_t i_zerofill : 1; - uint64_t i_rsvd_2 : 3; - uint64_t i_notify : 1; - uint64_t i_rsvd_1 : 3; - uint64_t i_poison : 1; - uint64_t i_rsvd : 55; + uint64_t ii_ibct1_regval; + struct { + uint64_t i_zerofill:1; + uint64_t i_rsvd_2:3; + uint64_t i_notify:1; + uint64_t i_rsvd_1:3; + uint64_t i_poison:1; + uint64_t i_rsvd:55; } ii_ibct1_fld_s; } ii_ibct1_u_t; - /************************************************************************ - * * + * * * This register contains the address to which the WINV is sent. * * This address has to be cache line aligned. * - * * + * * ************************************************************************/ typedef union ii_ibna1_u { - uint64_t ii_ibna1_regval; - struct { - uint64_t i_rsvd_1 : 7; - uint64_t i_addr : 33; - uint64_t i_rsvd : 24; + uint64_t ii_ibna1_regval; + struct { + uint64_t i_rsvd_1:7; + uint64_t i_addr:33; + uint64_t i_rsvd:24; } ii_ibna1_fld_s; } ii_ibna1_u_t; - /************************************************************************ - * * + * * * This register contains the programmable level as well as the node * * ID and PI unit of the processor to which the interrupt will be * - * sent. * - * * + * sent. * + * * ************************************************************************/ typedef union ii_ibia1_u { - uint64_t ii_ibia1_regval; - struct { - uint64_t i_pi_id : 1; - uint64_t i_node_id : 8; - uint64_t i_rsvd_1 : 7; - uint64_t i_level : 7; - uint64_t i_rsvd : 41; + uint64_t ii_ibia1_regval; + struct { + uint64_t i_pi_id:1; + uint64_t i_node_id:8; + uint64_t i_rsvd_1:7; + uint64_t i_level:7; + uint64_t i_rsvd:41; } ii_ibia1_fld_s; } ii_ibia1_u_t; - /************************************************************************ - * * + * * * This register defines the resources that feed information into * * the two performance counters located in the IO Performance * * Profiling Register. There are 17 different quantities that can be * @@ -2811,133 +2708,129 @@ typedef union ii_ibia1_u { * other is available from the other performance counter. Hence, the * * II supports all 17*16=272 possible combinations of quantities to * * measure. * - * * + * * ************************************************************************/ typedef union ii_ipcr_u { - uint64_t ii_ipcr_regval; - struct { - uint64_t i_ippr0_c : 4; - uint64_t i_ippr1_c : 4; - uint64_t i_icct : 8; - uint64_t i_rsvd : 48; + uint64_t ii_ipcr_regval; + struct { + uint64_t i_ippr0_c:4; + uint64_t i_ippr1_c:4; + uint64_t i_icct:8; + uint64_t i_rsvd:48; } ii_ipcr_fld_s; } ii_ipcr_u_t; - /************************************************************************ - * * - * * - * * + * * + * * + * * ************************************************************************/ typedef union ii_ippr_u { - uint64_t ii_ippr_regval; - struct { - uint64_t i_ippr0 : 32; - uint64_t i_ippr1 : 32; + uint64_t ii_ippr_regval; + struct { + uint64_t i_ippr0:32; + uint64_t i_ippr1:32; } ii_ippr_fld_s; } ii_ippr_u_t; - - -/************************************************************************** - * * - * The following defines which were not formed into structures are * - * probably indentical to another register, and the name of the * - * register is provided against each of these registers. This * - * information needs to be checked carefully * - * * - * IIO_ICRB1_A IIO_ICRB0_A * - * IIO_ICRB1_B IIO_ICRB0_B * - * IIO_ICRB1_C IIO_ICRB0_C * - * IIO_ICRB1_D IIO_ICRB0_D * - * IIO_ICRB1_E IIO_ICRB0_E * - * IIO_ICRB2_A IIO_ICRB0_A * - * IIO_ICRB2_B IIO_ICRB0_B * - * IIO_ICRB2_C IIO_ICRB0_C * - * IIO_ICRB2_D IIO_ICRB0_D * - * IIO_ICRB2_E IIO_ICRB0_E * - * IIO_ICRB3_A IIO_ICRB0_A * - * IIO_ICRB3_B IIO_ICRB0_B * - * IIO_ICRB3_C IIO_ICRB0_C * - * IIO_ICRB3_D IIO_ICRB0_D * - * IIO_ICRB3_E IIO_ICRB0_E * - * IIO_ICRB4_A IIO_ICRB0_A * - * IIO_ICRB4_B IIO_ICRB0_B * - * IIO_ICRB4_C IIO_ICRB0_C * - * IIO_ICRB4_D IIO_ICRB0_D * - * IIO_ICRB4_E IIO_ICRB0_E * - * IIO_ICRB5_A IIO_ICRB0_A * - * IIO_ICRB5_B IIO_ICRB0_B * - * IIO_ICRB5_C IIO_ICRB0_C * - * IIO_ICRB5_D IIO_ICRB0_D * - * IIO_ICRB5_E IIO_ICRB0_E * - * IIO_ICRB6_A IIO_ICRB0_A * - * IIO_ICRB6_B IIO_ICRB0_B * - * IIO_ICRB6_C IIO_ICRB0_C * - * IIO_ICRB6_D IIO_ICRB0_D * - * IIO_ICRB6_E IIO_ICRB0_E * - * IIO_ICRB7_A IIO_ICRB0_A * - * IIO_ICRB7_B IIO_ICRB0_B * - * IIO_ICRB7_C IIO_ICRB0_C * - * IIO_ICRB7_D IIO_ICRB0_D * - * IIO_ICRB7_E IIO_ICRB0_E * - * IIO_ICRB8_A IIO_ICRB0_A * - * IIO_ICRB8_B IIO_ICRB0_B * - * IIO_ICRB8_C IIO_ICRB0_C * - * IIO_ICRB8_D IIO_ICRB0_D * - * IIO_ICRB8_E IIO_ICRB0_E * - * IIO_ICRB9_A IIO_ICRB0_A * - * IIO_ICRB9_B IIO_ICRB0_B * - * IIO_ICRB9_C IIO_ICRB0_C * - * IIO_ICRB9_D IIO_ICRB0_D * - * IIO_ICRB9_E IIO_ICRB0_E * - * IIO_ICRBA_A IIO_ICRB0_A * - * IIO_ICRBA_B IIO_ICRB0_B * - * IIO_ICRBA_C IIO_ICRB0_C * - * IIO_ICRBA_D IIO_ICRB0_D * - * IIO_ICRBA_E IIO_ICRB0_E * - * IIO_ICRBB_A IIO_ICRB0_A * - * IIO_ICRBB_B IIO_ICRB0_B * - * IIO_ICRBB_C IIO_ICRB0_C * - * IIO_ICRBB_D IIO_ICRB0_D * - * IIO_ICRBB_E IIO_ICRB0_E * - * IIO_ICRBC_A IIO_ICRB0_A * - * IIO_ICRBC_B IIO_ICRB0_B * - * IIO_ICRBC_C IIO_ICRB0_C * - * IIO_ICRBC_D IIO_ICRB0_D * - * IIO_ICRBC_E IIO_ICRB0_E * - * IIO_ICRBD_A IIO_ICRB0_A * - * IIO_ICRBD_B IIO_ICRB0_B * - * IIO_ICRBD_C IIO_ICRB0_C * - * IIO_ICRBD_D IIO_ICRB0_D * - * IIO_ICRBD_E IIO_ICRB0_E * - * IIO_ICRBE_A IIO_ICRB0_A * - * IIO_ICRBE_B IIO_ICRB0_B * - * IIO_ICRBE_C IIO_ICRB0_C * - * IIO_ICRBE_D IIO_ICRB0_D * - * IIO_ICRBE_E IIO_ICRB0_E * - * * - **************************************************************************/ - +/************************************************************************ + * * + * The following defines which were not formed into structures are * + * probably indentical to another register, and the name of the * + * register is provided against each of these registers. This * + * information needs to be checked carefully * + * * + * IIO_ICRB1_A IIO_ICRB0_A * + * IIO_ICRB1_B IIO_ICRB0_B * + * IIO_ICRB1_C IIO_ICRB0_C * + * IIO_ICRB1_D IIO_ICRB0_D * + * IIO_ICRB1_E IIO_ICRB0_E * + * IIO_ICRB2_A IIO_ICRB0_A * + * IIO_ICRB2_B IIO_ICRB0_B * + * IIO_ICRB2_C IIO_ICRB0_C * + * IIO_ICRB2_D IIO_ICRB0_D * + * IIO_ICRB2_E IIO_ICRB0_E * + * IIO_ICRB3_A IIO_ICRB0_A * + * IIO_ICRB3_B IIO_ICRB0_B * + * IIO_ICRB3_C IIO_ICRB0_C * + * IIO_ICRB3_D IIO_ICRB0_D * + * IIO_ICRB3_E IIO_ICRB0_E * + * IIO_ICRB4_A IIO_ICRB0_A * + * IIO_ICRB4_B IIO_ICRB0_B * + * IIO_ICRB4_C IIO_ICRB0_C * + * IIO_ICRB4_D IIO_ICRB0_D * + * IIO_ICRB4_E IIO_ICRB0_E * + * IIO_ICRB5_A IIO_ICRB0_A * + * IIO_ICRB5_B IIO_ICRB0_B * + * IIO_ICRB5_C IIO_ICRB0_C * + * IIO_ICRB5_D IIO_ICRB0_D * + * IIO_ICRB5_E IIO_ICRB0_E * + * IIO_ICRB6_A IIO_ICRB0_A * + * IIO_ICRB6_B IIO_ICRB0_B * + * IIO_ICRB6_C IIO_ICRB0_C * + * IIO_ICRB6_D IIO_ICRB0_D * + * IIO_ICRB6_E IIO_ICRB0_E * + * IIO_ICRB7_A IIO_ICRB0_A * + * IIO_ICRB7_B IIO_ICRB0_B * + * IIO_ICRB7_C IIO_ICRB0_C * + * IIO_ICRB7_D IIO_ICRB0_D * + * IIO_ICRB7_E IIO_ICRB0_E * + * IIO_ICRB8_A IIO_ICRB0_A * + * IIO_ICRB8_B IIO_ICRB0_B * + * IIO_ICRB8_C IIO_ICRB0_C * + * IIO_ICRB8_D IIO_ICRB0_D * + * IIO_ICRB8_E IIO_ICRB0_E * + * IIO_ICRB9_A IIO_ICRB0_A * + * IIO_ICRB9_B IIO_ICRB0_B * + * IIO_ICRB9_C IIO_ICRB0_C * + * IIO_ICRB9_D IIO_ICRB0_D * + * IIO_ICRB9_E IIO_ICRB0_E * + * IIO_ICRBA_A IIO_ICRB0_A * + * IIO_ICRBA_B IIO_ICRB0_B * + * IIO_ICRBA_C IIO_ICRB0_C * + * IIO_ICRBA_D IIO_ICRB0_D * + * IIO_ICRBA_E IIO_ICRB0_E * + * IIO_ICRBB_A IIO_ICRB0_A * + * IIO_ICRBB_B IIO_ICRB0_B * + * IIO_ICRBB_C IIO_ICRB0_C * + * IIO_ICRBB_D IIO_ICRB0_D * + * IIO_ICRBB_E IIO_ICRB0_E * + * IIO_ICRBC_A IIO_ICRB0_A * + * IIO_ICRBC_B IIO_ICRB0_B * + * IIO_ICRBC_C IIO_ICRB0_C * + * IIO_ICRBC_D IIO_ICRB0_D * + * IIO_ICRBC_E IIO_ICRB0_E * + * IIO_ICRBD_A IIO_ICRB0_A * + * IIO_ICRBD_B IIO_ICRB0_B * + * IIO_ICRBD_C IIO_ICRB0_C * + * IIO_ICRBD_D IIO_ICRB0_D * + * IIO_ICRBD_E IIO_ICRB0_E * + * IIO_ICRBE_A IIO_ICRB0_A * + * IIO_ICRBE_B IIO_ICRB0_B * + * IIO_ICRBE_C IIO_ICRB0_C * + * IIO_ICRBE_D IIO_ICRB0_D * + * IIO_ICRBE_E IIO_ICRB0_E * + * * + ************************************************************************/ /* * Slightly friendlier names for some common registers. */ -#define IIO_WIDGET IIO_WID /* Widget identification */ -#define IIO_WIDGET_STAT IIO_WSTAT /* Widget status register */ -#define IIO_WIDGET_CTRL IIO_WCR /* Widget control register */ -#define IIO_PROTECT IIO_ILAPR /* IO interface protection */ -#define IIO_PROTECT_OVRRD IIO_ILAPO /* IO protect override */ -#define IIO_OUTWIDGET_ACCESS IIO_IOWA /* Outbound widget access */ -#define IIO_INWIDGET_ACCESS IIO_IIWA /* Inbound widget access */ -#define IIO_INDEV_ERR_MASK IIO_IIDEM /* Inbound device error mask */ -#define IIO_LLP_CSR IIO_ILCSR /* LLP control and status */ -#define IIO_LLP_LOG IIO_ILLR /* LLP log */ -#define IIO_XTALKCC_TOUT IIO_IXCC /* Xtalk credit count timeout*/ -#define IIO_XTALKTT_TOUT IIO_IXTT /* Xtalk tail timeout */ -#define IIO_IO_ERR_CLR IIO_IECLR /* IO error clear */ +#define IIO_WIDGET IIO_WID /* Widget identification */ +#define IIO_WIDGET_STAT IIO_WSTAT /* Widget status register */ +#define IIO_WIDGET_CTRL IIO_WCR /* Widget control register */ +#define IIO_PROTECT IIO_ILAPR /* IO interface protection */ +#define IIO_PROTECT_OVRRD IIO_ILAPO /* IO protect override */ +#define IIO_OUTWIDGET_ACCESS IIO_IOWA /* Outbound widget access */ +#define IIO_INWIDGET_ACCESS IIO_IIWA /* Inbound widget access */ +#define IIO_INDEV_ERR_MASK IIO_IIDEM /* Inbound device error mask */ +#define IIO_LLP_CSR IIO_ILCSR /* LLP control and status */ +#define IIO_LLP_LOG IIO_ILLR /* LLP log */ +#define IIO_XTALKCC_TOUT IIO_IXCC /* Xtalk credit count timeout */ +#define IIO_XTALKTT_TOUT IIO_IXTT /* Xtalk tail timeout */ +#define IIO_IO_ERR_CLR IIO_IECLR /* IO error clear */ #define IIO_IGFX_0 IIO_IGFX0 #define IIO_IGFX_1 IIO_IGFX1 #define IIO_IBCT_0 IIO_IBCT0 @@ -2957,12 +2850,12 @@ typedef union ii_ippr_u { #define IIO_PRTE_A(_x) (IIO_IPRTE0_A + (8 * (_x))) #define IIO_PRTE_B(_x) (IIO_IPRTE0_B + (8 * (_x))) #define IIO_NUM_PRTES 8 /* Total number of PRB table entries */ -#define IIO_WIDPRTE_A(x) IIO_PRTE_A(((x) - 8)) /* widget ID to its PRTE num */ -#define IIO_WIDPRTE_B(x) IIO_PRTE_B(((x) - 8)) /* widget ID to its PRTE num */ +#define IIO_WIDPRTE_A(x) IIO_PRTE_A(((x) - 8)) /* widget ID to its PRTE num */ +#define IIO_WIDPRTE_B(x) IIO_PRTE_B(((x) - 8)) /* widget ID to its PRTE num */ -#define IIO_NUM_IPRBS (9) +#define IIO_NUM_IPRBS 9 -#define IIO_LLP_CSR_IS_UP 0x00002000 +#define IIO_LLP_CSR_IS_UP 0x00002000 #define IIO_LLP_CSR_LLP_STAT_MASK 0x00003000 #define IIO_LLP_CSR_LLP_STAT_SHFT 12 @@ -2970,30 +2863,29 @@ typedef union ii_ippr_u { #define IIO_LLP_SN_MAX 0xffff /* in ILLR SN_CNT, Max Sequence Number errors */ /* key to IIO_PROTECT_OVRRD */ -#define IIO_PROTECT_OVRRD_KEY 0x53474972756c6573ull /* "SGIrules" */ +#define IIO_PROTECT_OVRRD_KEY 0x53474972756c6573ull /* "SGIrules" */ /* BTE register names */ -#define IIO_BTE_STAT_0 IIO_IBLS_0 /* Also BTE length/status 0 */ -#define IIO_BTE_SRC_0 IIO_IBSA_0 /* Also BTE source address 0 */ -#define IIO_BTE_DEST_0 IIO_IBDA_0 /* Also BTE dest. address 0 */ -#define IIO_BTE_CTRL_0 IIO_IBCT_0 /* Also BTE control/terminate 0 */ -#define IIO_BTE_NOTIFY_0 IIO_IBNA_0 /* Also BTE notification 0 */ -#define IIO_BTE_INT_0 IIO_IBIA_0 /* Also BTE interrupt 0 */ -#define IIO_BTE_OFF_0 0 /* Base offset from BTE 0 regs. */ -#define IIO_BTE_OFF_1 (IIO_IBLS_1 - IIO_IBLS_0) /* Offset from base to BTE 1 */ +#define IIO_BTE_STAT_0 IIO_IBLS_0 /* Also BTE length/status 0 */ +#define IIO_BTE_SRC_0 IIO_IBSA_0 /* Also BTE source address 0 */ +#define IIO_BTE_DEST_0 IIO_IBDA_0 /* Also BTE dest. address 0 */ +#define IIO_BTE_CTRL_0 IIO_IBCT_0 /* Also BTE control/terminate 0 */ +#define IIO_BTE_NOTIFY_0 IIO_IBNA_0 /* Also BTE notification 0 */ +#define IIO_BTE_INT_0 IIO_IBIA_0 /* Also BTE interrupt 0 */ +#define IIO_BTE_OFF_0 0 /* Base offset from BTE 0 regs. */ +#define IIO_BTE_OFF_1 (IIO_IBLS_1 - IIO_IBLS_0) /* Offset from base to BTE 1 */ /* BTE register offsets from base */ #define BTEOFF_STAT 0 -#define BTEOFF_SRC (IIO_BTE_SRC_0 - IIO_BTE_STAT_0) -#define BTEOFF_DEST (IIO_BTE_DEST_0 - IIO_BTE_STAT_0) -#define BTEOFF_CTRL (IIO_BTE_CTRL_0 - IIO_BTE_STAT_0) -#define BTEOFF_NOTIFY (IIO_BTE_NOTIFY_0 - IIO_BTE_STAT_0) -#define BTEOFF_INT (IIO_BTE_INT_0 - IIO_BTE_STAT_0) - +#define BTEOFF_SRC (IIO_BTE_SRC_0 - IIO_BTE_STAT_0) +#define BTEOFF_DEST (IIO_BTE_DEST_0 - IIO_BTE_STAT_0) +#define BTEOFF_CTRL (IIO_BTE_CTRL_0 - IIO_BTE_STAT_0) +#define BTEOFF_NOTIFY (IIO_BTE_NOTIFY_0 - IIO_BTE_STAT_0) +#define BTEOFF_INT (IIO_BTE_INT_0 - IIO_BTE_STAT_0) /* names used in shub diags */ -#define IIO_BASE_BTE0 IIO_IBLS_0 -#define IIO_BASE_BTE1 IIO_IBLS_1 +#define IIO_BASE_BTE0 IIO_IBLS_0 +#define IIO_BASE_BTE1 IIO_IBLS_1 /* * Macro which takes the widget number, and returns the @@ -3001,10 +2893,9 @@ typedef union ii_ippr_u { * value _x is expected to be a widget number in the range * 0, 8 - 0xF */ -#define IIO_IOPRB(_x) (IIO_IOPRB_0 + ( ( (_x) < HUB_WIDGET_ID_MIN ? \ - (_x) : \ - (_x) - (HUB_WIDGET_ID_MIN-1)) << 3) ) - +#define IIO_IOPRB(_x) (IIO_IOPRB_0 + ( ( (_x) < HUB_WIDGET_ID_MIN ? \ + (_x) : \ + (_x) - (HUB_WIDGET_ID_MIN-1)) << 3) ) /* GFX Flow Control Node/Widget Register */ #define IIO_IGFX_W_NUM_BITS 4 /* size of widget num field */ @@ -3025,7 +2916,6 @@ typedef union ii_ippr_u { (((node) & IIO_IGFX_N_NUM_MASK) << IIO_IGFX_N_NUM_SHIFT) | \ (((cpu) & IIO_IGFX_P_NUM_MASK) << IIO_IGFX_P_NUM_SHIFT)) - /* Scratch registers (all bits available) */ #define IIO_SCRATCH_REG0 IIO_ISCR0 #define IIO_SCRATCH_REG1 IIO_ISCR1 @@ -3046,21 +2936,21 @@ typedef union ii_ippr_u { #define IIO_SCRATCH_BIT1_0 0x0000000000000001UL #define IIO_SCRATCH_BIT1_1 0x0000000000000002UL /* IO Translation Table Entries */ -#define IIO_NUM_ITTES 7 /* ITTEs numbered 0..6 */ - /* Hw manuals number them 1..7! */ +#define IIO_NUM_ITTES 7 /* ITTEs numbered 0..6 */ + /* Hw manuals number them 1..7! */ /* * IIO_IMEM Register fields. */ -#define IIO_IMEM_W0ESD 0x1UL /* Widget 0 shut down due to error */ -#define IIO_IMEM_B0ESD (1UL << 4) /* BTE 0 shut down due to error */ -#define IIO_IMEM_B1ESD (1UL << 8) /* BTE 1 Shut down due to error */ +#define IIO_IMEM_W0ESD 0x1UL /* Widget 0 shut down due to error */ +#define IIO_IMEM_B0ESD (1UL << 4) /* BTE 0 shut down due to error */ +#define IIO_IMEM_B1ESD (1UL << 8) /* BTE 1 Shut down due to error */ /* * As a permanent workaround for a bug in the PI side of the shub, we've * redefined big window 7 as small window 0. XXX does this still apply for SN1?? */ -#define HUB_NUM_BIG_WINDOW (IIO_NUM_ITTES - 1) +#define HUB_NUM_BIG_WINDOW (IIO_NUM_ITTES - 1) /* * Use the top big window as a surrogate for the first small window @@ -3071,11 +2961,11 @@ typedef union ii_ippr_u { /* * CRB manipulation macros - * The CRB macros are slightly complicated, since there are up to - * four registers associated with each CRB entry. + * The CRB macros are slightly complicated, since there are up to + * four registers associated with each CRB entry. */ -#define IIO_NUM_CRBS 15 /* Number of CRBs */ -#define IIO_NUM_PC_CRBS 4 /* Number of partial cache CRBs */ +#define IIO_NUM_CRBS 15 /* Number of CRBs */ +#define IIO_NUM_PC_CRBS 4 /* Number of partial cache CRBs */ #define IIO_ICRB_OFFSET 8 #define IIO_ICRB_0 IIO_ICRB0_A #define IIO_ICRB_ADDR_SHFT 2 /* Shift to get proper address */ @@ -3083,43 +2973,43 @@ typedef union ii_ippr_u { #define IIO_FIRST_PC_ENTRY 12 */ -#define IIO_ICRB_A(_x) ((u64)(IIO_ICRB_0 + (6 * IIO_ICRB_OFFSET * (_x)))) -#define IIO_ICRB_B(_x) ((u64)((char *)IIO_ICRB_A(_x) + 1*IIO_ICRB_OFFSET)) -#define IIO_ICRB_C(_x) ((u64)((char *)IIO_ICRB_A(_x) + 2*IIO_ICRB_OFFSET)) -#define IIO_ICRB_D(_x) ((u64)((char *)IIO_ICRB_A(_x) + 3*IIO_ICRB_OFFSET)) -#define IIO_ICRB_E(_x) ((u64)((char *)IIO_ICRB_A(_x) + 4*IIO_ICRB_OFFSET)) +#define IIO_ICRB_A(_x) ((u64)(IIO_ICRB_0 + (6 * IIO_ICRB_OFFSET * (_x)))) +#define IIO_ICRB_B(_x) ((u64)((char *)IIO_ICRB_A(_x) + 1*IIO_ICRB_OFFSET)) +#define IIO_ICRB_C(_x) ((u64)((char *)IIO_ICRB_A(_x) + 2*IIO_ICRB_OFFSET)) +#define IIO_ICRB_D(_x) ((u64)((char *)IIO_ICRB_A(_x) + 3*IIO_ICRB_OFFSET)) +#define IIO_ICRB_E(_x) ((u64)((char *)IIO_ICRB_A(_x) + 4*IIO_ICRB_OFFSET)) #define TNUM_TO_WIDGET_DEV(_tnum) (_tnum & 0x7) /* * values for "ecode" field */ -#define IIO_ICRB_ECODE_DERR 0 /* Directory error due to IIO access */ -#define IIO_ICRB_ECODE_PERR 1 /* Poison error on IO access */ -#define IIO_ICRB_ECODE_WERR 2 /* Write error by IIO access - * e.g. WINV to a Read only line. */ -#define IIO_ICRB_ECODE_AERR 3 /* Access error caused by IIO access */ -#define IIO_ICRB_ECODE_PWERR 4 /* Error on partial write */ -#define IIO_ICRB_ECODE_PRERR 5 /* Error on partial read */ -#define IIO_ICRB_ECODE_TOUT 6 /* CRB timeout before deallocating */ -#define IIO_ICRB_ECODE_XTERR 7 /* Incoming xtalk pkt had error bit */ +#define IIO_ICRB_ECODE_DERR 0 /* Directory error due to IIO access */ +#define IIO_ICRB_ECODE_PERR 1 /* Poison error on IO access */ +#define IIO_ICRB_ECODE_WERR 2 /* Write error by IIO access + * e.g. WINV to a Read only line. */ +#define IIO_ICRB_ECODE_AERR 3 /* Access error caused by IIO access */ +#define IIO_ICRB_ECODE_PWERR 4 /* Error on partial write */ +#define IIO_ICRB_ECODE_PRERR 5 /* Error on partial read */ +#define IIO_ICRB_ECODE_TOUT 6 /* CRB timeout before deallocating */ +#define IIO_ICRB_ECODE_XTERR 7 /* Incoming xtalk pkt had error bit */ /* * Values for field imsgtype */ -#define IIO_ICRB_IMSGT_XTALK 0 /* Incoming Meessage from Xtalk */ -#define IIO_ICRB_IMSGT_BTE 1 /* Incoming message from BTE */ -#define IIO_ICRB_IMSGT_SN1NET 2 /* Incoming message from SN1 net */ -#define IIO_ICRB_IMSGT_CRB 3 /* Incoming message from CRB ??? */ +#define IIO_ICRB_IMSGT_XTALK 0 /* Incoming Meessage from Xtalk */ +#define IIO_ICRB_IMSGT_BTE 1 /* Incoming message from BTE */ +#define IIO_ICRB_IMSGT_SN1NET 2 /* Incoming message from SN1 net */ +#define IIO_ICRB_IMSGT_CRB 3 /* Incoming message from CRB ??? */ /* * values for field initiator. */ -#define IIO_ICRB_INIT_XTALK 0 /* Message originated in xtalk */ -#define IIO_ICRB_INIT_BTE0 0x1 /* Message originated in BTE 0 */ -#define IIO_ICRB_INIT_SN1NET 0x2 /* Message originated in SN1net */ -#define IIO_ICRB_INIT_CRB 0x3 /* Message originated in CRB ? */ -#define IIO_ICRB_INIT_BTE1 0x5 /* MEssage originated in BTE 1 */ +#define IIO_ICRB_INIT_XTALK 0 /* Message originated in xtalk */ +#define IIO_ICRB_INIT_BTE0 0x1 /* Message originated in BTE 0 */ +#define IIO_ICRB_INIT_SN1NET 0x2 /* Message originated in SN1net */ +#define IIO_ICRB_INIT_CRB 0x3 /* Message originated in CRB ? */ +#define IIO_ICRB_INIT_BTE1 0x5 /* MEssage originated in BTE 1 */ /* * Number of credits Hub widget has while sending req/response to @@ -3127,8 +3017,8 @@ typedef union ii_ippr_u { * Value of 3 is required by Xbow 1.1 * We may be able to increase this to 4 with Xbow 1.2. */ -#define HUBII_XBOW_CREDIT 3 -#define HUBII_XBOW_REV2_CREDIT 4 +#define HUBII_XBOW_CREDIT 3 +#define HUBII_XBOW_REV2_CREDIT 4 /* * Number of credits that xtalk devices should use when communicating @@ -3159,28 +3049,28 @@ typedef union ii_ippr_u { */ #define IIO_ICMR_CRB_VLD_SHFT 20 -#define IIO_ICMR_CRB_VLD_MASK (0x7fffUL << IIO_ICMR_CRB_VLD_SHFT) +#define IIO_ICMR_CRB_VLD_MASK (0x7fffUL << IIO_ICMR_CRB_VLD_SHFT) #define IIO_ICMR_FC_CNT_SHFT 16 -#define IIO_ICMR_FC_CNT_MASK (0xf << IIO_ICMR_FC_CNT_SHFT) +#define IIO_ICMR_FC_CNT_MASK (0xf << IIO_ICMR_FC_CNT_SHFT) #define IIO_ICMR_C_CNT_SHFT 4 -#define IIO_ICMR_C_CNT_MASK (0xf << IIO_ICMR_C_CNT_SHFT) +#define IIO_ICMR_C_CNT_MASK (0xf << IIO_ICMR_C_CNT_SHFT) -#define IIO_ICMR_PRECISE (1UL << 52) -#define IIO_ICMR_CLR_RPPD (1UL << 13) -#define IIO_ICMR_CLR_RQPD (1UL << 12) +#define IIO_ICMR_PRECISE (1UL << 52) +#define IIO_ICMR_CLR_RPPD (1UL << 13) +#define IIO_ICMR_CLR_RQPD (1UL << 12) /* * IIO PIO Deallocation register field masks : (IIO_IPDR) XXX present but not needed in bedrock? See the manual. */ -#define IIO_IPDR_PND (1 << 4) +#define IIO_IPDR_PND (1 << 4) /* * IIO CRB deallocation register field masks: (IIO_ICDR) */ -#define IIO_ICDR_PND (1 << 4) +#define IIO_ICDR_PND (1 << 4) /* * IO BTE Length/Status (IIO_IBLS) register bit field definitions @@ -3223,35 +3113,35 @@ typedef union ii_ippr_u { /* * IO Error Clear register bit field definitions */ -#define IECLR_PI1_FWD_INT (1UL << 31) /* clear PI1_FORWARD_INT in iidsr */ -#define IECLR_PI0_FWD_INT (1UL << 30) /* clear PI0_FORWARD_INT in iidsr */ -#define IECLR_SPUR_RD_HDR (1UL << 29) /* clear valid bit in ixss reg */ -#define IECLR_BTE1 (1UL << 18) /* clear bte error 1 */ -#define IECLR_BTE0 (1UL << 17) /* clear bte error 0 */ -#define IECLR_CRAZY (1UL << 16) /* clear crazy bit in wstat reg */ -#define IECLR_PRB_F (1UL << 15) /* clear err bit in PRB_F reg */ -#define IECLR_PRB_E (1UL << 14) /* clear err bit in PRB_E reg */ -#define IECLR_PRB_D (1UL << 13) /* clear err bit in PRB_D reg */ -#define IECLR_PRB_C (1UL << 12) /* clear err bit in PRB_C reg */ -#define IECLR_PRB_B (1UL << 11) /* clear err bit in PRB_B reg */ -#define IECLR_PRB_A (1UL << 10) /* clear err bit in PRB_A reg */ -#define IECLR_PRB_9 (1UL << 9) /* clear err bit in PRB_9 reg */ -#define IECLR_PRB_8 (1UL << 8) /* clear err bit in PRB_8 reg */ -#define IECLR_PRB_0 (1UL << 0) /* clear err bit in PRB_0 reg */ +#define IECLR_PI1_FWD_INT (1UL << 31) /* clear PI1_FORWARD_INT in iidsr */ +#define IECLR_PI0_FWD_INT (1UL << 30) /* clear PI0_FORWARD_INT in iidsr */ +#define IECLR_SPUR_RD_HDR (1UL << 29) /* clear valid bit in ixss reg */ +#define IECLR_BTE1 (1UL << 18) /* clear bte error 1 */ +#define IECLR_BTE0 (1UL << 17) /* clear bte error 0 */ +#define IECLR_CRAZY (1UL << 16) /* clear crazy bit in wstat reg */ +#define IECLR_PRB_F (1UL << 15) /* clear err bit in PRB_F reg */ +#define IECLR_PRB_E (1UL << 14) /* clear err bit in PRB_E reg */ +#define IECLR_PRB_D (1UL << 13) /* clear err bit in PRB_D reg */ +#define IECLR_PRB_C (1UL << 12) /* clear err bit in PRB_C reg */ +#define IECLR_PRB_B (1UL << 11) /* clear err bit in PRB_B reg */ +#define IECLR_PRB_A (1UL << 10) /* clear err bit in PRB_A reg */ +#define IECLR_PRB_9 (1UL << 9) /* clear err bit in PRB_9 reg */ +#define IECLR_PRB_8 (1UL << 8) /* clear err bit in PRB_8 reg */ +#define IECLR_PRB_0 (1UL << 0) /* clear err bit in PRB_0 reg */ /* * IIO CRB control register Fields: IIO_ICCR */ -#define IIO_ICCR_PENDING (0x10000) -#define IIO_ICCR_CMD_MASK (0xFF) -#define IIO_ICCR_CMD_SHFT (7) -#define IIO_ICCR_CMD_NOP (0x0) /* No Op */ -#define IIO_ICCR_CMD_WAKE (0x100) /* Reactivate CRB entry and process */ -#define IIO_ICCR_CMD_TIMEOUT (0x200) /* Make CRB timeout & mark invalid */ -#define IIO_ICCR_CMD_EJECT (0x400) /* Contents of entry written to memory +#define IIO_ICCR_PENDING 0x10000 +#define IIO_ICCR_CMD_MASK 0xFF +#define IIO_ICCR_CMD_SHFT 7 +#define IIO_ICCR_CMD_NOP 0x0 /* No Op */ +#define IIO_ICCR_CMD_WAKE 0x100 /* Reactivate CRB entry and process */ +#define IIO_ICCR_CMD_TIMEOUT 0x200 /* Make CRB timeout & mark invalid */ +#define IIO_ICCR_CMD_EJECT 0x400 /* Contents of entry written to memory * via a WB */ -#define IIO_ICCR_CMD_FLUSH (0x800) +#define IIO_ICCR_CMD_FLUSH 0x800 /* * @@ -3283,8 +3173,8 @@ typedef union ii_ippr_u { * Easy access macros for CRBs, all 5 registers (A-E) */ typedef ii_icrb0_a_u_t icrba_t; -#define a_sidn ii_icrb0_a_fld_s.ia_sidn -#define a_tnum ii_icrb0_a_fld_s.ia_tnum +#define a_sidn ii_icrb0_a_fld_s.ia_sidn +#define a_tnum ii_icrb0_a_fld_s.ia_tnum #define a_addr ii_icrb0_a_fld_s.ia_addr #define a_valid ii_icrb0_a_fld_s.ia_vld #define a_iow ii_icrb0_a_fld_s.ia_iow @@ -3324,14 +3214,13 @@ typedef ii_icrb0_c_u_t icrbc_t; #define c_source ii_icrb0_c_fld_s.ic_source #define c_regvalue ii_icrb0_c_regval - typedef ii_icrb0_d_u_t icrbd_t; #define d_sleep ii_icrb0_d_fld_s.id_sleep #define d_pricnt ii_icrb0_d_fld_s.id_pr_cnt #define d_pripsc ii_icrb0_d_fld_s.id_pr_psc #define d_bteop ii_icrb0_d_fld_s.id_bte_op -#define d_bteaddr ii_icrb0_d_fld_s.id_pa_be /* ic_pa_be fld has 2 names*/ -#define d_benable ii_icrb0_d_fld_s.id_pa_be /* ic_pa_be fld has 2 names*/ +#define d_bteaddr ii_icrb0_d_fld_s.id_pa_be /* ic_pa_be fld has 2 names */ +#define d_benable ii_icrb0_d_fld_s.id_pa_be /* ic_pa_be fld has 2 names */ #define d_regvalue ii_icrb0_d_regval typedef ii_icrb0_e_u_t icrbe_t; @@ -3341,7 +3230,6 @@ typedef ii_icrb0_e_u_t icrbe_t; #define icrbe_timeout ii_icrb0_e_fld_s.ie_timeout #define e_regvalue ii_icrb0_e_regval - /* Number of widgets supported by shub */ #define HUB_NUM_WIDGET 9 #define HUB_WIDGET_ID_MIN 0x8 @@ -3367,27 +3255,27 @@ typedef ii_icrb0_e_u_t icrbe_t; #define LNK_STAT_WORKING 0x2 /* LLP is working */ -#define IIO_WSTAT_ECRAZY (1ULL << 32) /* Hub gone crazy */ -#define IIO_WSTAT_TXRETRY (1ULL << 9) /* Hub Tx Retry timeout */ -#define IIO_WSTAT_TXRETRY_MASK (0x7F) /* should be 0xFF?? */ -#define IIO_WSTAT_TXRETRY_SHFT (16) -#define IIO_WSTAT_TXRETRY_CNT(w) (((w) >> IIO_WSTAT_TXRETRY_SHFT) & \ - IIO_WSTAT_TXRETRY_MASK) +#define IIO_WSTAT_ECRAZY (1ULL << 32) /* Hub gone crazy */ +#define IIO_WSTAT_TXRETRY (1ULL << 9) /* Hub Tx Retry timeout */ +#define IIO_WSTAT_TXRETRY_MASK 0x7F /* should be 0xFF?? */ +#define IIO_WSTAT_TXRETRY_SHFT 16 +#define IIO_WSTAT_TXRETRY_CNT(w) (((w) >> IIO_WSTAT_TXRETRY_SHFT) & \ + IIO_WSTAT_TXRETRY_MASK) /* Number of II perf. counters we can multiplex at once */ #define IO_PERF_SETS 32 /* Bit for the widget in inbound access register */ -#define IIO_IIWA_WIDGET(_w) ((uint64_t)(1ULL << _w)) +#define IIO_IIWA_WIDGET(_w) ((uint64_t)(1ULL << _w)) /* Bit for the widget in outbound access register */ -#define IIO_IOWA_WIDGET(_w) ((uint64_t)(1ULL << _w)) +#define IIO_IOWA_WIDGET(_w) ((uint64_t)(1ULL << _w)) /* NOTE: The following define assumes that we are going to get * widget numbers from 8 thru F and the device numbers within * widget from 0 thru 7. */ -#define IIO_IIDEM_WIDGETDEV_MASK(w, d) ((uint64_t)(1ULL << (8 * ((w) - 8) + (d)))) +#define IIO_IIDEM_WIDGETDEV_MASK(w, d) ((uint64_t)(1ULL << (8 * ((w) - 8) + (d)))) /* IO Interrupt Destination Register */ #define IIO_IIDSR_SENT_SHIFT 28 @@ -3402,11 +3290,11 @@ typedef ii_icrb0_e_u_t icrbe_t; #define IIO_IIDSR_LVL_MASK 0x000000ff /* Xtalk timeout threshhold register (IIO_IXTT) */ -#define IXTT_RRSP_TO_SHFT 55 /* read response timeout */ +#define IXTT_RRSP_TO_SHFT 55 /* read response timeout */ #define IXTT_RRSP_TO_MASK (0x1FULL << IXTT_RRSP_TO_SHFT) -#define IXTT_RRSP_PS_SHFT 32 /* read responsed TO prescalar */ +#define IXTT_RRSP_PS_SHFT 32 /* read responsed TO prescalar */ #define IXTT_RRSP_PS_MASK (0x7FFFFFULL << IXTT_RRSP_PS_SHFT) -#define IXTT_TAIL_TO_SHFT 0 /* tail timeout counter threshold */ +#define IXTT_TAIL_TO_SHFT 0 /* tail timeout counter threshold */ #define IXTT_TAIL_TO_MASK (0x3FFFFFFULL << IXTT_TAIL_TO_SHFT) /* @@ -3414,17 +3302,17 @@ typedef ii_icrb0_e_u_t icrbe_t; */ typedef union hubii_wcr_u { - uint64_t wcr_reg_value; - struct { - uint64_t wcr_widget_id: 4, /* LLP crossbar credit */ - wcr_tag_mode: 1, /* Tag mode */ - wcr_rsvd1: 8, /* Reserved */ - wcr_xbar_crd: 3, /* LLP crossbar credit */ - wcr_f_bad_pkt: 1, /* Force bad llp pkt enable */ - wcr_dir_con: 1, /* widget direct connect */ - wcr_e_thresh: 5, /* elasticity threshold */ - wcr_rsvd: 41; /* unused */ - } wcr_fields_s; + uint64_t wcr_reg_value; + struct { + uint64_t wcr_widget_id:4, /* LLP crossbar credit */ + wcr_tag_mode:1, /* Tag mode */ + wcr_rsvd1:8, /* Reserved */ + wcr_xbar_crd:3, /* LLP crossbar credit */ + wcr_f_bad_pkt:1, /* Force bad llp pkt enable */ + wcr_dir_con:1, /* widget direct connect */ + wcr_e_thresh:5, /* elasticity threshold */ + wcr_rsvd:41; /* unused */ + } wcr_fields_s; } hubii_wcr_t; #define iwcr_dir_con wcr_fields_s.wcr_dir_con @@ -3436,41 +3324,35 @@ performance registers */ performed */ typedef union io_perf_sel { - uint64_t perf_sel_reg; - struct { - uint64_t perf_ippr0 : 4, - perf_ippr1 : 4, - perf_icct : 8, - perf_rsvd : 48; - } perf_sel_bits; + uint64_t perf_sel_reg; + struct { + uint64_t perf_ippr0:4, perf_ippr1:4, perf_icct:8, perf_rsvd:48; + } perf_sel_bits; } io_perf_sel_t; /* io_perf_cnt is to extract the count from the shub registers. Due to hardware problems there is only one counter, not two. */ typedef union io_perf_cnt { - uint64_t perf_cnt; - struct { - uint64_t perf_cnt : 20, - perf_rsvd2 : 12, - perf_rsvd1 : 32; - } perf_cnt_bits; + uint64_t perf_cnt; + struct { + uint64_t perf_cnt:20, perf_rsvd2:12, perf_rsvd1:32; + } perf_cnt_bits; } io_perf_cnt_t; typedef union iprte_a { - uint64_t entry; - struct { - uint64_t i_rsvd_1 : 3; - uint64_t i_addr : 38; - uint64_t i_init : 3; - uint64_t i_source : 8; - uint64_t i_rsvd : 2; - uint64_t i_widget : 4; - uint64_t i_to_cnt : 5; - uint64_t i_vld : 1; + uint64_t entry; + struct { + uint64_t i_rsvd_1:3; + uint64_t i_addr:38; + uint64_t i_init:3; + uint64_t i_source:8; + uint64_t i_rsvd:2; + uint64_t i_widget:4; + uint64_t i_to_cnt:5; + uint64_t i_vld:1; } iprte_fields; } iprte_a_t; -#endif /* _ASM_IA64_SN_SHUBIO_H */ - +#endif /* _ASM_IA64_SN_SHUBIO_H */ diff --git a/include/asm-ia64/sn/sn_cpuid.h b/include/asm-ia64/sn/sn_cpuid.h index 685435af170..20b30018766 100644 --- a/include/asm-ia64/sn/sn_cpuid.h +++ b/include/asm-ia64/sn/sn_cpuid.h @@ -4,7 +4,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 2000-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 2000-2005 Silicon Graphics, Inc. All rights reserved. */ @@ -92,24 +92,24 @@ * NOTE: on non-MP systems, only cpuid 0 exists */ -extern short physical_node_map[]; /* indexed by nasid to get cnode */ +extern short physical_node_map[]; /* indexed by nasid to get cnode */ /* * Macros for retrieving info about current cpu */ -#define get_nasid() (nodepda->phys_cpuid[smp_processor_id()].nasid) -#define get_subnode() (nodepda->phys_cpuid[smp_processor_id()].subnode) -#define get_slice() (nodepda->phys_cpuid[smp_processor_id()].slice) -#define get_cnode() (nodepda->phys_cpuid[smp_processor_id()].cnode) -#define get_sapicid() ((ia64_getreg(_IA64_REG_CR_LID) >> 16) & 0xffff) +#define get_nasid() (sn_nodepda->phys_cpuid[smp_processor_id()].nasid) +#define get_subnode() (sn_nodepda->phys_cpuid[smp_processor_id()].subnode) +#define get_slice() (sn_nodepda->phys_cpuid[smp_processor_id()].slice) +#define get_cnode() (sn_nodepda->phys_cpuid[smp_processor_id()].cnode) +#define get_sapicid() ((ia64_getreg(_IA64_REG_CR_LID) >> 16) & 0xffff) /* * Macros for retrieving info about an arbitrary cpu * cpuid - logical cpu id */ -#define cpuid_to_nasid(cpuid) (nodepda->phys_cpuid[cpuid].nasid) -#define cpuid_to_subnode(cpuid) (nodepda->phys_cpuid[cpuid].subnode) -#define cpuid_to_slice(cpuid) (nodepda->phys_cpuid[cpuid].slice) +#define cpuid_to_nasid(cpuid) (sn_nodepda->phys_cpuid[cpuid].nasid) +#define cpuid_to_subnode(cpuid) (sn_nodepda->phys_cpuid[cpuid].subnode) +#define cpuid_to_slice(cpuid) (sn_nodepda->phys_cpuid[cpuid].slice) #define cpuid_to_cnodeid(cpuid) (physical_node_map[cpuid_to_nasid(cpuid)]) @@ -123,11 +123,8 @@ extern int nasid_slice_to_cpuid(int, int); /* * cnodeid_to_nasid - convert a cnodeid to a NASID - * Macro relies on pg_data for a node being on the node itself. - * Just extract the NASID from the pointer. - * */ -#define cnodeid_to_nasid(cnodeid) pda->cnodeid_to_nasid_table[cnodeid] +#define cnodeid_to_nasid(cnodeid) (sn_cnodeid_to_nasid[cnodeid]) /* * nasid_to_cnodeid - convert a NASID to a cnodeid diff --git a/include/asm-ia64/sn/sn_fru.h b/include/asm-ia64/sn/sn_fru.h deleted file mode 100644 index 8c21ac3f015..00000000000 --- a/include/asm-ia64/sn/sn_fru.h +++ /dev/null @@ -1,44 +0,0 @@ -/* - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (C) 1992-1997,1999-2004 Silicon Graphics, Inc. All rights reserved. - */ -#ifndef _ASM_IA64_SN_SN_FRU_H -#define _ASM_IA64_SN_SN_FRU_H - -#define MAX_DIMMS 8 /* max # of dimm banks */ -#define MAX_PCIDEV 8 /* max # of pci devices on a pci bus */ - -typedef unsigned char confidence_t; - -typedef struct kf_mem_s { - confidence_t km_confidence; /* confidence level that the memory is bad - * is this necessary ? - */ - confidence_t km_dimm[MAX_DIMMS]; - /* confidence level that dimm[i] is bad - *I think this is the right number - */ - -} kf_mem_t; - -typedef struct kf_cpu_s { - confidence_t kc_confidence; /* confidence level that cpu is bad */ - confidence_t kc_icache; /* confidence level that instr. cache is bad */ - confidence_t kc_dcache; /* confidence level that data cache is bad */ - confidence_t kc_scache; /* confidence level that sec. cache is bad */ - confidence_t kc_sysbus; /* confidence level that sysad/cmd/state bus is bad */ -} kf_cpu_t; - - -typedef struct kf_pci_bus_s { - confidence_t kpb_belief; /* confidence level that the pci bus is bad */ - confidence_t kpb_pcidev_belief[MAX_PCIDEV]; - /* confidence level that the pci dev is bad */ -} kf_pci_bus_t; - - -#endif /* _ASM_IA64_SN_SN_FRU_H */ - diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index 88c31b53dc0..56d74ca76b5 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -8,7 +8,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All rights reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. All rights reserved. */ @@ -35,8 +35,8 @@ #define SN_SAL_PRINT_ERROR 0x02000012 #define SN_SAL_SET_ERROR_HANDLING_FEATURES 0x0200001a // reentrant #define SN_SAL_GET_FIT_COMPT 0x0200001b // reentrant -#define SN_SAL_GET_SN_INFO 0x0200001c #define SN_SAL_GET_SAPIC_INFO 0x0200001d +#define SN_SAL_GET_SN_INFO 0x0200001e #define SN_SAL_CONSOLE_PUTC 0x02000021 #define SN_SAL_CONSOLE_GETC 0x02000022 #define SN_SAL_CONSOLE_PUTS 0x02000023 @@ -64,6 +64,7 @@ #define SN_SAL_SYSCTL_IOBRICK_PCI_OP 0x02000042 // reentrant #define SN_SAL_IROUTER_OP 0x02000043 +#define SN_SAL_SYSCTL_EVENT 0x02000044 #define SN_SAL_IOIF_INTERRUPT 0x0200004a #define SN_SAL_HWPERF_OP 0x02000050 // lock #define SN_SAL_IOIF_ERROR_INTERRUPT 0x02000051 @@ -76,7 +77,8 @@ #define SN_SAL_IOIF_GET_WIDGET_DMAFLUSH_LIST 0x02000058 #define SN_SAL_HUB_ERROR_INTERRUPT 0x02000060 - +#define SN_SAL_BTE_RECOVER 0x02000061 +#define SN_SAL_IOIF_GET_PCI_TOPOLOGY 0x02000062 /* * Service-specific constants @@ -555,7 +557,8 @@ static inline u64 ia64_sn_partition_serial_get(void) { struct ia64_sal_retval ret_stuff; - SAL_CALL(ret_stuff, SN_SAL_PARTITION_SERIAL_GET, 0, 0, 0, 0, 0, 0, 0); + ia64_sal_oemcall_reentrant(&ret_stuff, SN_SAL_PARTITION_SERIAL_GET, 0, + 0, 0, 0, 0, 0, 0); if (ret_stuff.status != 0) return 0; return ret_stuff.v0; @@ -563,11 +566,10 @@ ia64_sn_partition_serial_get(void) static inline u64 sn_partition_serial_number_val(void) { - if (sn_partition_serial_number) { - return(sn_partition_serial_number); - } else { - return(sn_partition_serial_number = ia64_sn_partition_serial_get()); + if (unlikely(sn_partition_serial_number == 0)) { + sn_partition_serial_number = ia64_sn_partition_serial_get(); } + return sn_partition_serial_number; } /* @@ -578,8 +580,8 @@ static inline partid_t ia64_sn_sysctl_partition_get(nasid_t nasid) { struct ia64_sal_retval ret_stuff; - SAL_CALL(ret_stuff, SN_SAL_SYSCTL_PARTITION_GET, nasid, - 0, 0, 0, 0, 0, 0); + ia64_sal_oemcall_nolock(&ret_stuff, SN_SAL_SYSCTL_PARTITION_GET, nasid, + 0, 0, 0, 0, 0, 0); if (ret_stuff.status != 0) return INVALID_PARTID; return ((partid_t)ret_stuff.v0); @@ -593,11 +595,38 @@ extern partid_t sn_partid; static inline partid_t sn_local_partid(void) { - if (sn_partid < 0) { - return (sn_partid = ia64_sn_sysctl_partition_get(cpuid_to_nasid(smp_processor_id()))); - } else { - return sn_partid; + if (unlikely(sn_partid < 0)) { + sn_partid = ia64_sn_sysctl_partition_get(cpuid_to_nasid(smp_processor_id())); } + return sn_partid; +} + +/* + * Returns the physical address of the partition's reserved page through + * an iterative number of calls. + * + * On first call, 'cookie' and 'len' should be set to 0, and 'addr' + * set to the nasid of the partition whose reserved page's address is + * being sought. + * On subsequent calls, pass the values, that were passed back on the + * previous call. + * + * While the return status equals SALRET_MORE_PASSES, keep calling + * this function after first copying 'len' bytes starting at 'addr' + * into 'buf'. Once the return status equals SALRET_OK, 'addr' will + * be the physical address of the partition's reserved page. If the + * return status equals neither of these, an error as occurred. + */ +static inline s64 +sn_partition_reserved_page_pa(u64 buf, u64 *cookie, u64 *addr, u64 *len) +{ + struct ia64_sal_retval rv; + ia64_sal_oemcall_reentrant(&rv, SN_SAL_GET_PARTITION_ADDR, *cookie, + *addr, buf, *len, 0, 0, 0); + *cookie = rv.v0; + *addr = rv.v1; + *len = rv.v2; + return rv.status; } /* @@ -619,8 +648,8 @@ static inline int sn_register_xp_addr_region(u64 paddr, u64 len, int operation) { struct ia64_sal_retval ret_stuff; - SAL_CALL(ret_stuff, SN_SAL_XP_ADDR_REGION, paddr, len, (u64)operation, - 0, 0, 0, 0); + ia64_sal_oemcall(&ret_stuff, SN_SAL_XP_ADDR_REGION, paddr, len, + (u64)operation, 0, 0, 0, 0); return ret_stuff.status; } @@ -644,8 +673,8 @@ sn_register_nofault_code(u64 start_addr, u64 end_addr, u64 return_addr, } else { call = SN_SAL_NO_FAULT_ZONE_PHYSICAL; } - SAL_CALL(ret_stuff, call, start_addr, end_addr, return_addr, (u64)1, - 0, 0, 0); + ia64_sal_oemcall(&ret_stuff, call, start_addr, end_addr, return_addr, + (u64)1, 0, 0, 0); return ret_stuff.status; } @@ -666,8 +695,8 @@ static inline int sn_change_coherence(u64 *new_domain, u64 *old_domain) { struct ia64_sal_retval ret_stuff; - SAL_CALL(ret_stuff, SN_SAL_COHERENCE, new_domain, old_domain, 0, 0, - 0, 0, 0); + ia64_sal_oemcall(&ret_stuff, SN_SAL_COHERENCE, (u64)new_domain, + (u64)old_domain, 0, 0, 0, 0, 0); return ret_stuff.status; } @@ -686,8 +715,8 @@ sn_change_memprotect(u64 paddr, u64 len, u64 perms, u64 *nasid_array) cnodeid = nasid_to_cnodeid(get_node_number(paddr)); // spin_lock(&NODEPDA(cnodeid)->bist_lock); local_irq_save(irq_flags); - SAL_CALL_NOLOCK(ret_stuff, SN_SAL_MEMPROTECT, paddr, len, nasid_array, - perms, 0, 0, 0); + ia64_sal_oemcall_nolock(&ret_stuff, SN_SAL_MEMPROTECT, paddr, len, + (u64)nasid_array, perms, 0, 0, 0); local_irq_restore(irq_flags); // spin_unlock(&NODEPDA(cnodeid)->bist_lock); return ret_stuff.status; @@ -849,6 +878,19 @@ ia64_sn_irtr_intr_disable(nasid_t nasid, int subch, u64 intr) return (int) rv.v0; } +/* + * Set up a node as the point of contact for system controller + * environmental event delivery. + */ +static inline int +ia64_sn_sysctl_event_init(nasid_t nasid) +{ + struct ia64_sal_retval rv; + SAL_CALL_REENTRANT(rv, SN_SAL_SYSCTL_EVENT, (u64) nasid, + 0, 0, 0, 0, 0, 0); + return (int) rv.v0; +} + /** * ia64_sn_get_fit_compt - read a FIT entry from the PROM header * @nasid: NASID of node to read @@ -1012,4 +1054,29 @@ ia64_sn_hwperf_op(nasid_t nasid, u64 opcode, u64 a0, u64 a1, u64 a2, return (int) rv.status; } +static inline int +ia64_sn_ioif_get_pci_topology(u64 rack, u64 bay, u64 slot, u64 slab, + u64 buf, u64 len) +{ + struct ia64_sal_retval rv; + SAL_CALL_NOLOCK(rv, SN_SAL_IOIF_GET_PCI_TOPOLOGY, + rack, bay, slot, slab, buf, len, 0); + return (int) rv.status; +} + +/* + * BTE error recovery is implemented in SAL + */ +static inline int +ia64_sn_bte_recovery(nasid_t nasid) +{ + struct ia64_sal_retval rv; + + rv.status = 0; + SAL_CALL_NOLOCK(rv, SN_SAL_BTE_RECOVER, 0, 0, 0, 0, 0, 0, 0); + if (rv.status == SALRET_NOT_IMPLEMENTED) + return 0; + return (int) rv.status; +} + #endif /* _ASM_IA64_SN_SN_SAL_H */ diff --git a/include/asm-ia64/sn/sndrv.h b/include/asm-ia64/sn/sndrv.h deleted file mode 100644 index aa00d42cde3..00000000000 --- a/include/asm-ia64/sn/sndrv.h +++ /dev/null @@ -1,47 +0,0 @@ -/* - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (c) 2002-2004 Silicon Graphics, Inc. All Rights Reserved. - */ - -#ifndef _ASM_IA64_SN_SNDRV_H -#define _ASM_IA64_SN_SNDRV_H - -/* ioctl commands */ -#define SNDRV_GET_ROUTERINFO 1 -#define SNDRV_GET_INFOSIZE 2 -#define SNDRV_GET_HUBINFO 3 -#define SNDRV_GET_FLASHLOGSIZE 4 -#define SNDRV_SET_FLASHSYNC 5 -#define SNDRV_GET_FLASHLOGDATA 6 -#define SNDRV_GET_FLASHLOGALL 7 - -#define SNDRV_SET_HISTOGRAM_TYPE 14 - -#define SNDRV_ELSC_COMMAND 19 -#define SNDRV_CLEAR_LOG 20 -#define SNDRV_INIT_LOG 21 -#define SNDRV_GET_PIMM_PSC 22 -#define SNDRV_SET_PARTITION 23 -#define SNDRV_GET_PARTITION 24 - -/* see synergy_perf_ioctl() */ -#define SNDRV_GET_SYNERGY_VERSION 30 -#define SNDRV_GET_SYNERGY_STATUS 31 -#define SNDRV_GET_SYNERGYINFO 32 -#define SNDRV_SYNERGY_APPEND 33 -#define SNDRV_SYNERGY_ENABLE 34 -#define SNDRV_SYNERGY_FREQ 35 - -/* Devices */ -#define SNDRV_UKNOWN_DEVICE -1 -#define SNDRV_ROUTER_DEVICE 1 -#define SNDRV_HUB_DEVICE 2 -#define SNDRV_ELSC_NVRAM_DEVICE 3 -#define SNDRV_ELSC_CONTROLLER_DEVICE 4 -#define SNDRV_SYSCTL_SUBCH 5 -#define SNDRV_SYNERGY_DEVICE 6 - -#endif /* _ASM_IA64_SN_SNDRV_H */ diff --git a/include/asm-ia64/sn/tioca.h b/include/asm-ia64/sn/tioca.h new file mode 100644 index 00000000000..bc1aacfb948 --- /dev/null +++ b/include/asm-ia64/sn/tioca.h @@ -0,0 +1,596 @@ +#ifndef _ASM_IA64_SN_TIO_TIOCA_H +#define _ASM_IA64_SN_TIO_TIOCA_H + +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2003-2005 Silicon Graphics, Inc. All rights reserved. + */ + + +#define TIOCA_PART_NUM 0xE020 +#define TIOCA_MFGR_NUM 0x24 +#define TIOCA_REV_A 0x1 + +/* + * Register layout for TIO:CA. See below for bitmasks for each register. + */ + +struct tioca { + uint64_t ca_id; /* 0x000000 */ + uint64_t ca_control1; /* 0x000008 */ + uint64_t ca_control2; /* 0x000010 */ + uint64_t ca_status1; /* 0x000018 */ + uint64_t ca_status2; /* 0x000020 */ + uint64_t ca_gart_aperature; /* 0x000028 */ + uint64_t ca_gfx_detach; /* 0x000030 */ + uint64_t ca_inta_dest_addr; /* 0x000038 */ + uint64_t ca_intb_dest_addr; /* 0x000040 */ + uint64_t ca_err_int_dest_addr; /* 0x000048 */ + uint64_t ca_int_status; /* 0x000050 */ + uint64_t ca_int_status_alias; /* 0x000058 */ + uint64_t ca_mult_error; /* 0x000060 */ + uint64_t ca_mult_error_alias; /* 0x000068 */ + uint64_t ca_first_error; /* 0x000070 */ + uint64_t ca_int_mask; /* 0x000078 */ + uint64_t ca_crm_pkterr_type; /* 0x000080 */ + uint64_t ca_crm_pkterr_type_alias; /* 0x000088 */ + uint64_t ca_crm_ct_error_detail_1; /* 0x000090 */ + uint64_t ca_crm_ct_error_detail_2; /* 0x000098 */ + uint64_t ca_crm_tnumto; /* 0x0000A0 */ + uint64_t ca_gart_err; /* 0x0000A8 */ + uint64_t ca_pcierr_type; /* 0x0000B0 */ + uint64_t ca_pcierr_addr; /* 0x0000B8 */ + + uint64_t ca_pad_0000C0[3]; /* 0x0000{C0..D0} */ + + uint64_t ca_pci_rd_buf_flush; /* 0x0000D8 */ + uint64_t ca_pci_dma_addr_extn; /* 0x0000E0 */ + uint64_t ca_agp_dma_addr_extn; /* 0x0000E8 */ + uint64_t ca_force_inta; /* 0x0000F0 */ + uint64_t ca_force_intb; /* 0x0000F8 */ + uint64_t ca_debug_vector_sel; /* 0x000100 */ + uint64_t ca_debug_mux_core_sel; /* 0x000108 */ + uint64_t ca_debug_mux_pci_sel; /* 0x000110 */ + uint64_t ca_debug_domain_sel; /* 0x000118 */ + + uint64_t ca_pad_000120[28]; /* 0x0001{20..F8} */ + + uint64_t ca_gart_ptr_table; /* 0x200 */ + uint64_t ca_gart_tlb_addr[8]; /* 0x2{08..40} */ +}; + +/* + * Mask/shift definitions for TIO:CA registers. The convention here is + * to mainly use the names as they appear in the "TIO AEGIS Programmers' + * Reference" with a CA_ prefix added. Some exceptions were made to fix + * duplicate field names or to generalize fields that are common to + * different registers (ca_debug_mux_core_sel and ca_debug_mux_pci_sel for + * example). + * + * Fields consisting of a single bit have a single #define have a single + * macro declaration to mask the bit. Fields consisting of multiple bits + * have two declarations: one to mask the proper bits in a register, and + * a second with the suffix "_SHFT" to identify how far the mask needs to + * be shifted right to get its base value. + */ + +/* ==== ca_control1 */ +#define CA_SYS_BIG_END (1ull << 0) +#define CA_DMA_AGP_SWAP (1ull << 1) +#define CA_DMA_PCI_SWAP (1ull << 2) +#define CA_PIO_IO_SWAP (1ull << 3) +#define CA_PIO_MEM_SWAP (1ull << 4) +#define CA_GFX_WR_SWAP (1ull << 5) +#define CA_AGP_FW_ENABLE (1ull << 6) +#define CA_AGP_CAL_CYCLE (0x7ull << 7) +#define CA_AGP_CAL_CYCLE_SHFT 7 +#define CA_AGP_CAL_PRSCL_BYP (1ull << 10) +#define CA_AGP_INIT_CAL_ENB (1ull << 11) +#define CA_INJ_ADDR_PERR (1ull << 12) +#define CA_INJ_DATA_PERR (1ull << 13) + /* bits 15:14 unused */ +#define CA_PCIM_IO_NBE_AD (0x7ull << 16) +#define CA_PCIM_IO_NBE_AD_SHFT 16 +#define CA_PCIM_FAST_BTB_ENB (1ull << 19) + /* bits 23:20 unused */ +#define CA_PIO_ADDR_OFFSET (0xffull << 24) +#define CA_PIO_ADDR_OFFSET_SHFT 24 + /* bits 35:32 unused */ +#define CA_AGPDMA_OP_COMBDELAY (0x1full << 36) +#define CA_AGPDMA_OP_COMBDELAY_SHFT 36 + /* bit 41 unused */ +#define CA_AGPDMA_OP_ENB_COMBDELAY (1ull << 42) +#define CA_PCI_INT_LPCNT (0xffull << 44) +#define CA_PCI_INT_LPCNT_SHFT 44 + /* bits 63:52 unused */ + +/* ==== ca_control2 */ +#define CA_AGP_LATENCY_TO (0xffull << 0) +#define CA_AGP_LATENCY_TO_SHFT 0 +#define CA_PCI_LATENCY_TO (0xffull << 8) +#define CA_PCI_LATENCY_TO_SHFT 8 +#define CA_PCI_MAX_RETRY (0x3ffull << 16) +#define CA_PCI_MAX_RETRY_SHFT 16 + /* bits 27:26 unused */ +#define CA_RT_INT_EN (0x3ull << 28) +#define CA_RT_INT_EN_SHFT 28 +#define CA_MSI_INT_ENB (1ull << 30) +#define CA_PCI_ARB_ERR_ENB (1ull << 31) +#define CA_GART_MEM_PARAM (0x3ull << 32) +#define CA_GART_MEM_PARAM_SHFT 32 +#define CA_GART_RD_PREFETCH_ENB (1ull << 34) +#define CA_GART_WR_PREFETCH_ENB (1ull << 35) +#define CA_GART_FLUSH_TLB (1ull << 36) + /* bits 39:37 unused */ +#define CA_CRM_TNUMTO_PERIOD (0x1fffull << 40) +#define CA_CRM_TNUMTO_PERIOD_SHFT 40 + /* bits 55:53 unused */ +#define CA_CRM_TNUMTO_ENB (1ull << 56) +#define CA_CRM_PRESCALER_BYP (1ull << 57) + /* bits 59:58 unused */ +#define CA_CRM_MAX_CREDIT (0x7ull << 60) +#define CA_CRM_MAX_CREDIT_SHFT 60 + /* bit 63 unused */ + +/* ==== ca_status1 */ +#define CA_CORELET_ID (0x3ull << 0) +#define CA_CORELET_ID_SHFT 0 +#define CA_INTA_N (1ull << 2) +#define CA_INTB_N (1ull << 3) +#define CA_CRM_CREDIT_AVAIL (0x7ull << 4) +#define CA_CRM_CREDIT_AVAIL_SHFT 4 + /* bit 7 unused */ +#define CA_CRM_SPACE_AVAIL (0x7full << 8) +#define CA_CRM_SPACE_AVAIL_SHFT 8 + /* bit 15 unused */ +#define CA_GART_TLB_VAL (0xffull << 16) +#define CA_GART_TLB_VAL_SHFT 16 + /* bits 63:24 unused */ + +/* ==== ca_status2 */ +#define CA_GFX_CREDIT_AVAIL (0xffull << 0) +#define CA_GFX_CREDIT_AVAIL_SHFT 0 +#define CA_GFX_OPQ_AVAIL (0xffull << 8) +#define CA_GFX_OPQ_AVAIL_SHFT 8 +#define CA_GFX_WRBUFF_AVAIL (0xffull << 16) +#define CA_GFX_WRBUFF_AVAIL_SHFT 16 +#define CA_ADMA_OPQ_AVAIL (0xffull << 24) +#define CA_ADMA_OPQ_AVAIL_SHFT 24 +#define CA_ADMA_WRBUFF_AVAIL (0xffull << 32) +#define CA_ADMA_WRBUFF_AVAIL_SHFT 32 +#define CA_ADMA_RDBUFF_AVAIL (0x7full << 40) +#define CA_ADMA_RDBUFF_AVAIL_SHFT 40 +#define CA_PCI_PIO_OP_STAT (1ull << 47) +#define CA_PDMA_OPQ_AVAIL (0xfull << 48) +#define CA_PDMA_OPQ_AVAIL_SHFT 48 +#define CA_PDMA_WRBUFF_AVAIL (0xfull << 52) +#define CA_PDMA_WRBUFF_AVAIL_SHFT 52 +#define CA_PDMA_RDBUFF_AVAIL (0x3ull << 56) +#define CA_PDMA_RDBUFF_AVAIL_SHFT 56 + /* bits 63:58 unused */ + +/* ==== ca_gart_aperature */ +#define CA_GART_AP_ENB_AGP (1ull << 0) +#define CA_GART_PAGE_SIZE (1ull << 1) +#define CA_GART_AP_ENB_PCI (1ull << 2) + /* bits 11:3 unused */ +#define CA_GART_AP_SIZE (0x3ffull << 12) +#define CA_GART_AP_SIZE_SHFT 12 +#define CA_GART_AP_BASE (0x3ffffffffffull << 22) +#define CA_GART_AP_BASE_SHFT 22 + +/* ==== ca_inta_dest_addr + ==== ca_intb_dest_addr + ==== ca_err_int_dest_addr */ + /* bits 2:0 unused */ +#define CA_INT_DEST_ADDR (0x7ffffffffffffull << 3) +#define CA_INT_DEST_ADDR_SHFT 3 + /* bits 55:54 unused */ +#define CA_INT_DEST_VECT (0xffull << 56) +#define CA_INT_DEST_VECT_SHFT 56 + +/* ==== ca_int_status */ +/* ==== ca_int_status_alias */ +/* ==== ca_mult_error */ +/* ==== ca_mult_error_alias */ +/* ==== ca_first_error */ +/* ==== ca_int_mask */ +#define CA_PCI_ERR (1ull << 0) + /* bits 3:1 unused */ +#define CA_GART_FETCH_ERR (1ull << 4) +#define CA_GFX_WR_OVFLW (1ull << 5) +#define CA_PIO_REQ_OVFLW (1ull << 6) +#define CA_CRM_PKTERR (1ull << 7) +#define CA_CRM_DVERR (1ull << 8) +#define CA_TNUMTO (1ull << 9) +#define CA_CXM_RSP_CRED_OVFLW (1ull << 10) +#define CA_CXM_REQ_CRED_OVFLW (1ull << 11) +#define CA_PIO_INVALID_ADDR (1ull << 12) +#define CA_PCI_ARB_TO (1ull << 13) +#define CA_AGP_REQ_OFLOW (1ull << 14) +#define CA_SBA_TYPE1_ERR (1ull << 15) + /* bit 16 unused */ +#define CA_INTA (1ull << 17) +#define CA_INTB (1ull << 18) +#define CA_MULT_INTA (1ull << 19) +#define CA_MULT_INTB (1ull << 20) +#define CA_GFX_CREDIT_OVFLW (1ull << 21) + /* bits 63:22 unused */ + +/* ==== ca_crm_pkterr_type */ +/* ==== ca_crm_pkterr_type_alias */ +#define CA_CRM_PKTERR_SBERR_HDR (1ull << 0) +#define CA_CRM_PKTERR_DIDN (1ull << 1) +#define CA_CRM_PKTERR_PACTYPE (1ull << 2) +#define CA_CRM_PKTERR_INV_TNUM (1ull << 3) +#define CA_CRM_PKTERR_ADDR_RNG (1ull << 4) +#define CA_CRM_PKTERR_ADDR_ALGN (1ull << 5) +#define CA_CRM_PKTERR_HDR_PARAM (1ull << 6) +#define CA_CRM_PKTERR_CW_ERR (1ull << 7) +#define CA_CRM_PKTERR_SBERR_NH (1ull << 8) +#define CA_CRM_PKTERR_EARLY_TERM (1ull << 9) +#define CA_CRM_PKTERR_EARLY_TAIL (1ull << 10) +#define CA_CRM_PKTERR_MSSNG_TAIL (1ull << 11) +#define CA_CRM_PKTERR_MSSNG_HDR (1ull << 12) + /* bits 15:13 unused */ +#define CA_FIRST_CRM_PKTERR_SBERR_HDR (1ull << 16) +#define CA_FIRST_CRM_PKTERR_DIDN (1ull << 17) +#define CA_FIRST_CRM_PKTERR_PACTYPE (1ull << 18) +#define CA_FIRST_CRM_PKTERR_INV_TNUM (1ull << 19) +#define CA_FIRST_CRM_PKTERR_ADDR_RNG (1ull << 20) +#define CA_FIRST_CRM_PKTERR_ADDR_ALGN (1ull << 21) +#define CA_FIRST_CRM_PKTERR_HDR_PARAM (1ull << 22) +#define CA_FIRST_CRM_PKTERR_CW_ERR (1ull << 23) +#define CA_FIRST_CRM_PKTERR_SBERR_NH (1ull << 24) +#define CA_FIRST_CRM_PKTERR_EARLY_TERM (1ull << 25) +#define CA_FIRST_CRM_PKTERR_EARLY_TAIL (1ull << 26) +#define CA_FIRST_CRM_PKTERR_MSSNG_TAIL (1ull << 27) +#define CA_FIRST_CRM_PKTERR_MSSNG_HDR (1ull << 28) + /* bits 63:29 unused */ + +/* ==== ca_crm_ct_error_detail_1 */ +#define CA_PKT_TYPE (0xfull << 0) +#define CA_PKT_TYPE_SHFT 0 +#define CA_SRC_ID (0x3ull << 4) +#define CA_SRC_ID_SHFT 4 +#define CA_DATA_SZ (0x3ull << 6) +#define CA_DATA_SZ_SHFT 6 +#define CA_TNUM (0xffull << 8) +#define CA_TNUM_SHFT 8 +#define CA_DW_DATA_EN (0xffull << 16) +#define CA_DW_DATA_EN_SHFT 16 +#define CA_GFX_CRED (0xffull << 24) +#define CA_GFX_CRED_SHFT 24 +#define CA_MEM_RD_PARAM (0x3ull << 32) +#define CA_MEM_RD_PARAM_SHFT 32 +#define CA_PIO_OP (1ull << 34) +#define CA_CW_ERR (1ull << 35) + /* bits 62:36 unused */ +#define CA_VALID (1ull << 63) + +/* ==== ca_crm_ct_error_detail_2 */ + /* bits 2:0 unused */ +#define CA_PKT_ADDR (0x1fffffffffffffull << 3) +#define CA_PKT_ADDR_SHFT 3 + /* bits 63:56 unused */ + +/* ==== ca_crm_tnumto */ +#define CA_CRM_TNUMTO_VAL (0xffull << 0) +#define CA_CRM_TNUMTO_VAL_SHFT 0 +#define CA_CRM_TNUMTO_WR (1ull << 8) + /* bits 63:9 unused */ + +/* ==== ca_gart_err */ +#define CA_GART_ERR_SOURCE (0x3ull << 0) +#define CA_GART_ERR_SOURCE_SHFT 0 + /* bits 3:2 unused */ +#define CA_GART_ERR_ADDR (0xfffffffffull << 4) +#define CA_GART_ERR_ADDR_SHFT 4 + /* bits 63:40 unused */ + +/* ==== ca_pcierr_type */ +#define CA_PCIERR_DATA (0xffffffffull << 0) +#define CA_PCIERR_DATA_SHFT 0 +#define CA_PCIERR_ENB (0xfull << 32) +#define CA_PCIERR_ENB_SHFT 32 +#define CA_PCIERR_CMD (0xfull << 36) +#define CA_PCIERR_CMD_SHFT 36 +#define CA_PCIERR_A64 (1ull << 40) +#define CA_PCIERR_SLV_SERR (1ull << 41) +#define CA_PCIERR_SLV_WR_PERR (1ull << 42) +#define CA_PCIERR_SLV_RD_PERR (1ull << 43) +#define CA_PCIERR_MST_SERR (1ull << 44) +#define CA_PCIERR_MST_WR_PERR (1ull << 45) +#define CA_PCIERR_MST_RD_PERR (1ull << 46) +#define CA_PCIERR_MST_MABT (1ull << 47) +#define CA_PCIERR_MST_TABT (1ull << 48) +#define CA_PCIERR_MST_RETRY_TOUT (1ull << 49) + +#define CA_PCIERR_TYPES \ + (CA_PCIERR_A64|CA_PCIERR_SLV_SERR| \ + CA_PCIERR_SLV_WR_PERR|CA_PCIERR_SLV_RD_PERR| \ + CA_PCIERR_MST_SERR|CA_PCIERR_MST_WR_PERR|CA_PCIERR_MST_RD_PERR| \ + CA_PCIERR_MST_MABT|CA_PCIERR_MST_TABT|CA_PCIERR_MST_RETRY_TOUT) + + /* bits 63:50 unused */ + +/* ==== ca_pci_dma_addr_extn */ +#define CA_UPPER_NODE_OFFSET (0x3full << 0) +#define CA_UPPER_NODE_OFFSET_SHFT 0 + /* bits 7:6 unused */ +#define CA_CHIPLET_ID (0x3ull << 8) +#define CA_CHIPLET_ID_SHFT 8 + /* bits 11:10 unused */ +#define CA_PCI_DMA_NODE_ID (0xffffull << 12) +#define CA_PCI_DMA_NODE_ID_SHFT 12 + /* bits 27:26 unused */ +#define CA_PCI_DMA_PIO_MEM_TYPE (1ull << 28) + /* bits 63:29 unused */ + + +/* ==== ca_agp_dma_addr_extn */ + /* bits 19:0 unused */ +#define CA_AGP_DMA_NODE_ID (0xffffull << 20) +#define CA_AGP_DMA_NODE_ID_SHFT 20 + /* bits 27:26 unused */ +#define CA_AGP_DMA_PIO_MEM_TYPE (1ull << 28) + /* bits 63:29 unused */ + +/* ==== ca_debug_vector_sel */ +#define CA_DEBUG_MN_VSEL (0xfull << 0) +#define CA_DEBUG_MN_VSEL_SHFT 0 +#define CA_DEBUG_PP_VSEL (0xfull << 4) +#define CA_DEBUG_PP_VSEL_SHFT 4 +#define CA_DEBUG_GW_VSEL (0xfull << 8) +#define CA_DEBUG_GW_VSEL_SHFT 8 +#define CA_DEBUG_GT_VSEL (0xfull << 12) +#define CA_DEBUG_GT_VSEL_SHFT 12 +#define CA_DEBUG_PD_VSEL (0xfull << 16) +#define CA_DEBUG_PD_VSEL_SHFT 16 +#define CA_DEBUG_AD_VSEL (0xfull << 20) +#define CA_DEBUG_AD_VSEL_SHFT 20 +#define CA_DEBUG_CX_VSEL (0xfull << 24) +#define CA_DEBUG_CX_VSEL_SHFT 24 +#define CA_DEBUG_CR_VSEL (0xfull << 28) +#define CA_DEBUG_CR_VSEL_SHFT 28 +#define CA_DEBUG_BA_VSEL (0xfull << 32) +#define CA_DEBUG_BA_VSEL_SHFT 32 +#define CA_DEBUG_PE_VSEL (0xfull << 36) +#define CA_DEBUG_PE_VSEL_SHFT 36 +#define CA_DEBUG_BO_VSEL (0xfull << 40) +#define CA_DEBUG_BO_VSEL_SHFT 40 +#define CA_DEBUG_BI_VSEL (0xfull << 44) +#define CA_DEBUG_BI_VSEL_SHFT 44 +#define CA_DEBUG_AS_VSEL (0xfull << 48) +#define CA_DEBUG_AS_VSEL_SHFT 48 +#define CA_DEBUG_PS_VSEL (0xfull << 52) +#define CA_DEBUG_PS_VSEL_SHFT 52 +#define CA_DEBUG_PM_VSEL (0xfull << 56) +#define CA_DEBUG_PM_VSEL_SHFT 56 + /* bits 63:60 unused */ + +/* ==== ca_debug_mux_core_sel */ +/* ==== ca_debug_mux_pci_sel */ +#define CA_DEBUG_MSEL0 (0x7ull << 0) +#define CA_DEBUG_MSEL0_SHFT 0 + /* bit 3 unused */ +#define CA_DEBUG_NSEL0 (0x7ull << 4) +#define CA_DEBUG_NSEL0_SHFT 4 + /* bit 7 unused */ +#define CA_DEBUG_MSEL1 (0x7ull << 8) +#define CA_DEBUG_MSEL1_SHFT 8 + /* bit 11 unused */ +#define CA_DEBUG_NSEL1 (0x7ull << 12) +#define CA_DEBUG_NSEL1_SHFT 12 + /* bit 15 unused */ +#define CA_DEBUG_MSEL2 (0x7ull << 16) +#define CA_DEBUG_MSEL2_SHFT 16 + /* bit 19 unused */ +#define CA_DEBUG_NSEL2 (0x7ull << 20) +#define CA_DEBUG_NSEL2_SHFT 20 + /* bit 23 unused */ +#define CA_DEBUG_MSEL3 (0x7ull << 24) +#define CA_DEBUG_MSEL3_SHFT 24 + /* bit 27 unused */ +#define CA_DEBUG_NSEL3 (0x7ull << 28) +#define CA_DEBUG_NSEL3_SHFT 28 + /* bit 31 unused */ +#define CA_DEBUG_MSEL4 (0x7ull << 32) +#define CA_DEBUG_MSEL4_SHFT 32 + /* bit 35 unused */ +#define CA_DEBUG_NSEL4 (0x7ull << 36) +#define CA_DEBUG_NSEL4_SHFT 36 + /* bit 39 unused */ +#define CA_DEBUG_MSEL5 (0x7ull << 40) +#define CA_DEBUG_MSEL5_SHFT 40 + /* bit 43 unused */ +#define CA_DEBUG_NSEL5 (0x7ull << 44) +#define CA_DEBUG_NSEL5_SHFT 44 + /* bit 47 unused */ +#define CA_DEBUG_MSEL6 (0x7ull << 48) +#define CA_DEBUG_MSEL6_SHFT 48 + /* bit 51 unused */ +#define CA_DEBUG_NSEL6 (0x7ull << 52) +#define CA_DEBUG_NSEL6_SHFT 52 + /* bit 55 unused */ +#define CA_DEBUG_MSEL7 (0x7ull << 56) +#define CA_DEBUG_MSEL7_SHFT 56 + /* bit 59 unused */ +#define CA_DEBUG_NSEL7 (0x7ull << 60) +#define CA_DEBUG_NSEL7_SHFT 60 + /* bit 63 unused */ + + +/* ==== ca_debug_domain_sel */ +#define CA_DEBUG_DOMAIN_L (1ull << 0) +#define CA_DEBUG_DOMAIN_H (1ull << 1) + /* bits 63:2 unused */ + +/* ==== ca_gart_ptr_table */ +#define CA_GART_PTR_VAL (1ull << 0) + /* bits 11:1 unused */ +#define CA_GART_PTR_ADDR (0xfffffffffffull << 12) +#define CA_GART_PTR_ADDR_SHFT 12 + /* bits 63:56 unused */ + +/* ==== ca_gart_tlb_addr[0-7] */ +#define CA_GART_TLB_ADDR (0xffffffffffffffull << 0) +#define CA_GART_TLB_ADDR_SHFT 0 + /* bits 62:56 unused */ +#define CA_GART_TLB_ENTRY_VAL (1ull << 63) + +/* + * PIO address space ranges for TIO:CA + */ + +/* CA internal registers */ +#define CA_PIO_ADMIN 0x00000000 +#define CA_PIO_ADMIN_LEN 0x00010000 + +/* GFX Write Buffer - Diagnostics */ +#define CA_PIO_GFX 0x00010000 +#define CA_PIO_GFX_LEN 0x00010000 + +/* AGP DMA Write Buffer - Diagnostics */ +#define CA_PIO_AGP_DMAWRITE 0x00020000 +#define CA_PIO_AGP_DMAWRITE_LEN 0x00010000 + +/* AGP DMA READ Buffer - Diagnostics */ +#define CA_PIO_AGP_DMAREAD 0x00030000 +#define CA_PIO_AGP_DMAREAD_LEN 0x00010000 + +/* PCI Config Type 0 */ +#define CA_PIO_PCI_TYPE0_CONFIG 0x01000000 +#define CA_PIO_PCI_TYPE0_CONFIG_LEN 0x01000000 + +/* PCI Config Type 1 */ +#define CA_PIO_PCI_TYPE1_CONFIG 0x02000000 +#define CA_PIO_PCI_TYPE1_CONFIG_LEN 0x01000000 + +/* PCI I/O Cycles - mapped to PCI Address 0x00000000-0x04ffffff */ +#define CA_PIO_PCI_IO 0x03000000 +#define CA_PIO_PCI_IO_LEN 0x05000000 + +/* PCI MEM Cycles - mapped to PCI with CA_PIO_ADDR_OFFSET of ca_control1 */ +/* use Fast Write if enabled and coretalk packet type is a GFX request */ +#define CA_PIO_PCI_MEM_OFFSET 0x08000000 +#define CA_PIO_PCI_MEM_OFFSET_LEN 0x08000000 + +/* PCI MEM Cycles - mapped to PCI Address 0x00000000-0xbfffffff */ +/* use Fast Write if enabled and coretalk packet type is a GFX request */ +#define CA_PIO_PCI_MEM 0x40000000 +#define CA_PIO_PCI_MEM_LEN 0xc0000000 + +/* + * DMA space + * + * The CA aperature (ie. bus address range) mapped by the GART is segmented into + * two parts. The lower portion of the aperature is used for mapping 32 bit + * PCI addresses which are managed by the dma interfaces in this file. The + * upper poprtion of the aperature is used for mapping 48 bit AGP addresses. + * The AGP portion of the aperature is managed by the agpgart_be.c driver + * in drivers/linux/agp. There are ca-specific hooks in that driver to + * manipulate the gart, but management of the AGP portion of the aperature + * is the responsibility of that driver. + * + * CA allows three main types of DMA mapping: + * + * PCI 64-bit Managed by this driver + * PCI 32-bit Managed by this driver + * AGP 48-bit Managed by hooks in the /dev/agpgart driver + * + * All of the above can optionally be remapped through the GART. The following + * table lists the combinations of addressing types and GART remapping that + * is currently supported by the driver (h/w supports all, s/w limits this): + * + * PCI64 PCI32 AGP48 + * GART no yes yes + * Direct yes yes no + * + * GART remapping of PCI64 is not done because there is no need to. The + * 64 bit PCI address holds all of the information necessary to target any + * memory in the system. + * + * AGP48 is always mapped through the GART. Management of the AGP48 portion + * of the aperature is the responsibility of code in the agpgart_be driver. + * + * The non-64 bit bus address space will currently be partitioned like this: + * + * 0xffff_ffff_ffff +-------- + * | AGP48 direct + * | Space managed by this driver + * CA_AGP_DIRECT_BASE +-------- + * | AGP GART mapped (gfx aperature) + * | Space managed by /dev/agpgart driver + * | This range is exposed to the agpgart + * | driver as the "graphics aperature" + * CA_AGP_MAPPED_BASE +----- + * | PCI GART mapped + * | Space managed by this driver + * CA_PCI32_MAPPED_BASE +---- + * | PCI32 direct + * | Space managed by this driver + * 0xC000_0000 +-------- + * (CA_PCI32_DIRECT_BASE) + * + * The bus address range CA_PCI32_MAPPED_BASE through CA_AGP_DIRECT_BASE + * is what we call the CA aperature. Addresses falling in this range will + * be remapped using the GART. + * + * The bus address range CA_AGP_MAPPED_BASE through CA_AGP_DIRECT_BASE + * is what we call the graphics aperature. This is a subset of the CA + * aperature and is under the control of the agpgart_be driver. + * + * CA_PCI32_MAPPED_BASE, CA_AGP_MAPPED_BASE, and CA_AGP_DIRECT_BASE are + * somewhat arbitrary values. The known constraints on choosing these is: + * + * 1) CA_AGP_DIRECT_BASE-CA_PCI32_MAPPED_BASE+1 (the CA aperature size) + * must be one of the values supported by the ca_gart_aperature register. + * Currently valid values are: 4MB through 4096MB in powers of 2 increments + * + * 2) CA_AGP_DIRECT_BASE-CA_AGP_MAPPED_BASE+1 (the gfx aperature size) + * must be in MB units since that's what the agpgart driver assumes. + */ + +/* + * Define Bus DMA ranges. These are configurable (see constraints above) + * and will probably need tuning based on experience. + */ + + +/* + * 11/24/03 + * CA has an addressing glitch w.r.t. PCI direct 32 bit DMA that makes it + * generally unusable. The problem is that for PCI direct 32 + * DMA's, all 32 bits of the bus address are used to form the lower 32 bits + * of the coretalk address, and coretalk bits 38:32 come from a register. + * Since only PCI bus addresses 0xC0000000-0xFFFFFFFF (1GB) are available + * for DMA (the rest is allocated to PIO), host node addresses need to be + * such that their lower 32 bits fall in the 0xC0000000-0xffffffff range + * as well. So there can be no PCI32 direct DMA below 3GB!! For this + * reason we set the CA_PCI32_DIRECT_SIZE to 0 which essentially makes + * tioca_dma_direct32() a noop but preserves the code flow should this issue + * be fixed in a respin. + * + * For now, all PCI32 DMA's must be mapped through the GART. + */ + +#define CA_PCI32_DIRECT_BASE 0xC0000000UL /* BASE not configurable */ +#define CA_PCI32_DIRECT_SIZE 0x00000000UL /* 0 MB */ + +#define CA_PCI32_MAPPED_BASE 0xC0000000UL +#define CA_PCI32_MAPPED_SIZE 0x40000000UL /* 2GB */ + +#define CA_AGP_MAPPED_BASE 0x80000000UL +#define CA_AGP_MAPPED_SIZE 0x40000000UL /* 2GB */ + +#define CA_AGP_DIRECT_BASE 0x40000000UL /* 2GB */ +#define CA_AGP_DIRECT_SIZE 0x40000000UL + +#define CA_APERATURE_BASE (CA_AGP_MAPPED_BASE) +#define CA_APERATURE_SIZE (CA_AGP_MAPPED_SIZE+CA_PCI32_MAPPED_SIZE) + +#endif /* _ASM_IA64_SN_TIO_TIOCA_H */ diff --git a/include/asm-ia64/sn/tioca_provider.h b/include/asm-ia64/sn/tioca_provider.h new file mode 100644 index 00000000000..b6acc22ab23 --- /dev/null +++ b/include/asm-ia64/sn/tioca_provider.h @@ -0,0 +1,206 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2003-2005 Silicon Graphics, Inc. All rights reserved. + */ + +#ifndef _ASM_IA64_SN_TIO_CA_AGP_PROVIDER_H +#define _ASM_IA64_SN_TIO_CA_AGP_PROVIDER_H + +#include <asm/sn/tioca.h> + +/* + * WAR enables + * Defines for individual WARs. Each is a bitmask of applicable + * part revision numbers. (1 << 1) == rev A, (1 << 2) == rev B, + * (3 << 1) == (rev A or rev B), etc + */ + +#define TIOCA_WAR_ENABLED(pv, tioca_common) \ + ((1 << tioca_common->ca_rev) & pv) + + /* TIO:ICE:FRZ:Freezer loses a PIO data ucred on PIO RD RSP with CW error */ +#define PV907908 (1 << 1) + /* ATI config space problems after BIOS execution starts */ +#define PV908234 (1 << 1) + /* CA:AGPDMA write request data mismatch with ABC1CL merge */ +#define PV895469 (1 << 1) + /* TIO:CA TLB invalidate of written GART entries possibly not occuring in CA*/ +#define PV910244 (1 << 1) + +struct tioca_dmamap{ + struct list_head cad_list; /* headed by ca_list */ + + dma_addr_t cad_dma_addr; /* Linux dma handle */ + uint cad_gart_entry; /* start entry in ca_gart_pagemap */ + uint cad_gart_size; /* #entries for this map */ +}; + +/* + * Kernel only fields. Prom may look at this stuff for debugging only. + * Access this structure through the ca_kernel_private ptr. + */ + +struct tioca_common ; + +struct tioca_kernel { + struct tioca_common *ca_common; /* tioca this belongs to */ + struct list_head ca_list; /* list of all ca's */ + struct list_head ca_dmamaps; + spinlock_t ca_lock; /* Kernel lock */ + cnodeid_t ca_closest_node; + struct list_head *ca_devices; /* bus->devices */ + + /* + * General GART stuff + */ + uint64_t ca_ap_size; /* size of aperature in bytes */ + uint32_t ca_gart_entries; /* # uint64_t entries in gart */ + uint32_t ca_ap_pagesize; /* aperature page size in bytes */ + uint64_t ca_ap_bus_base; /* bus address of CA aperature */ + uint64_t ca_gart_size; /* gart size in bytes */ + uint64_t *ca_gart; /* gart table vaddr */ + uint64_t ca_gart_coretalk_addr; /* gart coretalk addr */ + uint8_t ca_gart_iscoherent; /* used in tioca_tlbflush */ + + /* PCI GART convenience values */ + uint64_t ca_pciap_base; /* pci aperature bus base address */ + uint64_t ca_pciap_size; /* pci aperature size (bytes) */ + uint64_t ca_pcigart_base; /* gfx GART bus base address */ + uint64_t *ca_pcigart; /* gfx GART vm address */ + uint32_t ca_pcigart_entries; + uint32_t ca_pcigart_start; /* PCI start index in ca_gart */ + void *ca_pcigart_pagemap; + + /* AGP GART convenience values */ + uint64_t ca_gfxap_base; /* gfx aperature bus base address */ + uint64_t ca_gfxap_size; /* gfx aperature size (bytes) */ + uint64_t ca_gfxgart_base; /* gfx GART bus base address */ + uint64_t *ca_gfxgart; /* gfx GART vm address */ + uint32_t ca_gfxgart_entries; + uint32_t ca_gfxgart_start; /* agpgart start index in ca_gart */ +}; + +/* + * Common tioca info shared between kernel and prom + * + * DO NOT CHANGE THIS STRUCT WITHOUT MAKING CORRESPONDING CHANGES + * TO THE PROM VERSION. + */ + +struct tioca_common { + struct pcibus_bussoft ca_common; /* common pciio header */ + + uint32_t ca_rev; + uint32_t ca_closest_nasid; + + uint64_t ca_prom_private; + uint64_t ca_kernel_private; +}; + +/** + * tioca_paddr_to_gart - Convert an SGI coretalk address to a CA GART entry + * @paddr: page address to convert + * + * Convert a system [coretalk] address to a GART entry. GART entries are + * formed using the following: + * + * data = ( (1<<63) | ( (REMAP_NODE_ID << 40) | (MD_CHIPLET_ID << 38) | + * (REMAP_SYS_ADDR) ) >> 12 ) + * + * DATA written to 1 GART TABLE Entry in system memory is remapped system + * addr for 1 page + * + * The data is for coretalk address format right shifted 12 bits with a + * valid bit. + * + * GART_TABLE_ENTRY [ 25:0 ] -- REMAP_SYS_ADDRESS[37:12]. + * GART_TABLE_ENTRY [ 27:26 ] -- SHUB MD chiplet id. + * GART_TABLE_ENTRY [ 41:28 ] -- REMAP_NODE_ID. + * GART_TABLE_ENTRY [ 63 ] -- Valid Bit + */ +static inline u64 +tioca_paddr_to_gart(unsigned long paddr) +{ + /* + * We are assuming right now that paddr already has the correct + * format since the address from xtalk_dmaXXX should already have + * NODE_ID, CHIPLET_ID, and SYS_ADDR in the correct locations. + */ + + return ((paddr) >> 12) | (1UL << 63); +} + +/** + * tioca_physpage_to_gart - Map a host physical page for SGI CA based DMA + * @page_addr: system page address to map + */ + +static inline unsigned long +tioca_physpage_to_gart(uint64_t page_addr) +{ + uint64_t coretalk_addr; + + coretalk_addr = PHYS_TO_TIODMA(page_addr); + if (!coretalk_addr) { + return 0; + } + + return tioca_paddr_to_gart(coretalk_addr); +} + +/** + * tioca_tlbflush - invalidate cached SGI CA GART TLB entries + * @tioca_kernel: CA context + * + * Invalidate tlb entries for a given CA GART. Main complexity is to account + * for revA bug. + */ +static inline void +tioca_tlbflush(struct tioca_kernel *tioca_kernel) +{ + volatile uint64_t tmp; + volatile struct tioca *ca_base; + struct tioca_common *tioca_common; + + tioca_common = tioca_kernel->ca_common; + ca_base = (struct tioca *)tioca_common->ca_common.bs_base; + + /* + * Explicit flushes not needed if GART is in cached mode + */ + if (tioca_kernel->ca_gart_iscoherent) { + if (TIOCA_WAR_ENABLED(PV910244, tioca_common)) { + /* + * PV910244: RevA CA needs explicit flushes. + * Need to put GART into uncached mode before + * flushing otherwise the explicit flush is ignored. + * + * Alternate WAR would be to leave GART cached and + * touch every CL aligned GART entry. + */ + + ca_base->ca_control2 &= ~(CA_GART_MEM_PARAM); + ca_base->ca_control2 |= CA_GART_FLUSH_TLB; + ca_base->ca_control2 |= + (0x2ull << CA_GART_MEM_PARAM_SHFT); + tmp = ca_base->ca_control2; + } + + return; + } + + /* + * Gart in uncached mode ... need an explicit flush. + */ + + ca_base->ca_control2 |= CA_GART_FLUSH_TLB; + tmp = ca_base->ca_control2; +} + +extern uint32_t tioca_gart_found; +extern int tioca_init_provider(void); +extern void tioca_fastwrite_enable(struct tioca_kernel *tioca_kern); +#endif /* _ASM_IA64_SN_TIO_CA_AGP_PROVIDER_H */ diff --git a/include/asm-ia64/sn/tiocx.h b/include/asm-ia64/sn/tiocx.h new file mode 100644 index 00000000000..c5447a50450 --- /dev/null +++ b/include/asm-ia64/sn/tiocx.h @@ -0,0 +1,71 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (c) 2005 Silicon Graphics, Inc. All rights reserved. + */ + +#ifndef _ASM_IA64_SN_TIO_TIOCX_H +#define _ASM_IA64_SN_TIO_TIOCX_H + +#ifdef __KERNEL__ + +struct cx_id_s { + unsigned int part_num; + unsigned int mfg_num; + int nasid; +}; + +struct cx_dev { + struct cx_id_s cx_id; + void *soft; /* driver specific */ + struct hubdev_info *hubdev; + struct device dev; + struct cx_drv *driver; +}; + +struct cx_device_id { + unsigned int part_num; + unsigned int mfg_num; +}; + +struct cx_drv { + char *name; + const struct cx_device_id *id_table; + struct device_driver driver; + int (*probe) (struct cx_dev * dev, const struct cx_device_id * id); + int (*remove) (struct cx_dev * dev); +}; + +/* create DMA address by stripping AS bits */ +#define TIOCX_DMA_ADDR(a) (uint64_t)((uint64_t)(a) & 0xffffcfffffffffUL) + +#define TIOCX_TO_TIOCX_DMA_ADDR(a) (uint64_t)(((uint64_t)(a) & 0xfffffffff) | \ + ((((uint64_t)(a)) & 0xffffc000000000UL) <<2)) + +#define TIO_CE_ASIC_PARTNUM 0xce00 +#define TIOCX_CORELET 3 + +/* These are taken from tio_mmr_as.h */ +#define TIO_ICE_FRZ_CFG TIO_MMR_ADDR_MOD(0x00000000b0008100UL) +#define TIO_ICE_PMI_TX_CFG TIO_MMR_ADDR_MOD(0x00000000b000b100UL) +#define TIO_ICE_PMI_TX_DYN_CREDIT_STAT_CB3 TIO_MMR_ADDR_MOD(0x00000000b000be18UL) +#define TIO_ICE_PMI_TX_DYN_CREDIT_STAT_CB3_CREDIT_CNT_MASK 0x000000000000000fUL + +#define to_cx_dev(n) container_of(n, struct cx_dev, dev) +#define to_cx_driver(drv) container_of(drv, struct cx_drv, driver) + +extern struct sn_irq_info *tiocx_irq_alloc(nasid_t, int, int, nasid_t, int); +extern void tiocx_irq_free(struct sn_irq_info *); +extern int cx_device_unregister(struct cx_dev *); +extern int cx_device_register(nasid_t, int, int, struct hubdev_info *); +extern int cx_driver_unregister(struct cx_drv *); +extern int cx_driver_register(struct cx_drv *); +extern uint64_t tiocx_dma_addr(uint64_t addr); +extern uint64_t tiocx_swin_base(int nasid); +extern void tiocx_mmr_store(int nasid, uint64_t offset, uint64_t value); +extern uint64_t tiocx_mmr_load(int nasid, uint64_t offset); + +#endif // __KERNEL__ +#endif // _ASM_IA64_SN_TIO_TIOCX__ diff --git a/include/asm-ia64/sn/types.h b/include/asm-ia64/sn/types.h index 586ed47cae9..8e04ee211e5 100644 --- a/include/asm-ia64/sn/types.h +++ b/include/asm-ia64/sn/types.h @@ -16,7 +16,8 @@ typedef signed short nasid_t; /* node id in numa-as-id space */ typedef signed char partid_t; /* partition ID type */ typedef unsigned int moduleid_t; /* user-visible module number type */ typedef unsigned int cmoduleid_t; /* kernel compact module id type */ -typedef signed char slabid_t; +typedef unsigned char slotid_t; /* slot (blade) within module */ +typedef unsigned char slabid_t; /* slab (asic) within slot */ typedef u64 nic_t; typedef unsigned long iopaddr_t; typedef unsigned long paddr_t; diff --git a/include/asm-ia64/sn/xp.h b/include/asm-ia64/sn/xp.h new file mode 100644 index 00000000000..9902185c028 --- /dev/null +++ b/include/asm-ia64/sn/xp.h @@ -0,0 +1,436 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2004-2005 Silicon Graphics, Inc. All rights reserved. + */ + + +/* + * External Cross Partition (XP) structures and defines. + */ + + +#ifndef _ASM_IA64_SN_XP_H +#define _ASM_IA64_SN_XP_H + + +#include <linux/version.h> +#include <linux/cache.h> +#include <linux/hardirq.h> +#include <asm/sn/types.h> +#include <asm/sn/bte.h> + + +#ifdef USE_DBUG_ON +#define DBUG_ON(condition) BUG_ON(condition) +#else +#define DBUG_ON(condition) +#endif + + +/* + * Define the maximum number of logically defined partitions the system + * can support. It is constrained by the maximum number of hardware + * partitionable regions. The term 'region' in this context refers to the + * minimum number of nodes that can comprise an access protection grouping. + * The access protection is in regards to memory, IPI and IOI. + * + * The maximum number of hardware partitionable regions is equal to the + * maximum number of nodes in the entire system divided by the minimum number + * of nodes that comprise an access protection grouping. + */ +#define XP_MAX_PARTITIONS 64 + + +/* + * Define the number of u64s required to represent all the C-brick nasids + * as a bitmap. The cross-partition kernel modules deal only with + * C-brick nasids, thus the need for bitmaps which don't account for + * odd-numbered (non C-brick) nasids. + */ +#define XP_MAX_PHYSNODE_ID (MAX_PHYSNODE_ID / 2) +#define XP_NASID_MASK_BYTES ((XP_MAX_PHYSNODE_ID + 7) / 8) +#define XP_NASID_MASK_WORDS ((XP_MAX_PHYSNODE_ID + 63) / 64) + + +/* + * Wrapper for bte_copy() that should it return a failure status will retry + * the bte_copy() once in the hope that the failure was due to a temporary + * aberration (i.e., the link going down temporarily). + * + * See bte_copy for definition of the input parameters. + * + * Note: xp_bte_copy() should never be called while holding a spinlock. + */ +static inline bte_result_t +xp_bte_copy(u64 src, u64 dest, u64 len, u64 mode, void *notification) +{ + bte_result_t ret; + + + ret = bte_copy(src, dest, len, mode, notification); + + if (ret != BTE_SUCCESS) { + if (!in_interrupt()) { + cond_resched(); + } + ret = bte_copy(src, dest, len, mode, notification); + } + + return ret; +} + + +/* + * XPC establishes channel connections between the local partition and any + * other partition that is currently up. Over these channels, kernel-level + * `users' can communicate with their counterparts on the other partitions. + * + * The maxinum number of channels is limited to eight. For performance reasons, + * the internal cross partition structures require sixteen bytes per channel, + * and eight allows all of this interface-shared info to fit in one cache line. + * + * XPC_NCHANNELS reflects the total number of channels currently defined. + * If the need for additional channels arises, one can simply increase + * XPC_NCHANNELS accordingly. If the day should come where that number + * exceeds the MAXIMUM number of channels allowed (eight), then one will need + * to make changes to the XPC code to allow for this. + */ +#define XPC_MEM_CHANNEL 0 /* memory channel number */ +#define XPC_NET_CHANNEL 1 /* network channel number */ + +#define XPC_NCHANNELS 2 /* #of defined channels */ +#define XPC_MAX_NCHANNELS 8 /* max #of channels allowed */ + +#if XPC_NCHANNELS > XPC_MAX_NCHANNELS +#error XPC_NCHANNELS exceeds MAXIMUM allowed. +#endif + + +/* + * The format of an XPC message is as follows: + * + * +-------+--------------------------------+ + * | flags |////////////////////////////////| + * +-------+--------------------------------+ + * | message # | + * +----------------------------------------+ + * | payload (user-defined message) | + * | | + * : + * | | + * +----------------------------------------+ + * + * The size of the payload is defined by the user via xpc_connect(). A user- + * defined message resides in the payload area. + * + * The user should have no dealings with the message header, but only the + * message's payload. When a message entry is allocated (via xpc_allocate()) + * a pointer to the payload area is returned and not the actual beginning of + * the XPC message. The user then constructs a message in the payload area + * and passes that pointer as an argument on xpc_send() or xpc_send_notify(). + * + * The size of a message entry (within a message queue) must be a cacheline + * sized multiple in order to facilitate the BTE transfer of messages from one + * message queue to another. A macro, XPC_MSG_SIZE(), is provided for the user + * that wants to fit as many msg entries as possible in a given memory size + * (e.g. a memory page). + */ +struct xpc_msg { + u8 flags; /* FOR XPC INTERNAL USE ONLY */ + u8 reserved[7]; /* FOR XPC INTERNAL USE ONLY */ + s64 number; /* FOR XPC INTERNAL USE ONLY */ + + u64 payload; /* user defined portion of message */ +}; + + +#define XPC_MSG_PAYLOAD_OFFSET (u64) (&((struct xpc_msg *)0)->payload) +#define XPC_MSG_SIZE(_payload_size) \ + L1_CACHE_ALIGN(XPC_MSG_PAYLOAD_OFFSET + (_payload_size)) + + +/* + * Define the return values and values passed to user's callout functions. + * (It is important to add new value codes at the end just preceding + * xpcUnknownReason, which must have the highest numerical value.) + */ +enum xpc_retval { + xpcSuccess = 0, + + xpcNotConnected, /* 1: channel is not connected */ + xpcConnected, /* 2: channel connected (opened) */ + xpcRETIRED1, /* 3: (formerly xpcDisconnected) */ + + xpcMsgReceived, /* 4: message received */ + xpcMsgDelivered, /* 5: message delivered and acknowledged */ + + xpcRETIRED2, /* 6: (formerly xpcTransferFailed) */ + + xpcNoWait, /* 7: operation would require wait */ + xpcRetry, /* 8: retry operation */ + xpcTimeout, /* 9: timeout in xpc_allocate_msg_wait() */ + xpcInterrupted, /* 10: interrupted wait */ + + xpcUnequalMsgSizes, /* 11: message size disparity between sides */ + xpcInvalidAddress, /* 12: invalid address */ + + xpcNoMemory, /* 13: no memory available for XPC structures */ + xpcLackOfResources, /* 14: insufficient resources for operation */ + xpcUnregistered, /* 15: channel is not registered */ + xpcAlreadyRegistered, /* 16: channel is already registered */ + + xpcPartitionDown, /* 17: remote partition is down */ + xpcNotLoaded, /* 18: XPC module is not loaded */ + xpcUnloading, /* 19: this side is unloading XPC module */ + + xpcBadMagic, /* 20: XPC MAGIC string not found */ + + xpcReactivating, /* 21: remote partition was reactivated */ + + xpcUnregistering, /* 22: this side is unregistering channel */ + xpcOtherUnregistering, /* 23: other side is unregistering channel */ + + xpcCloneKThread, /* 24: cloning kernel thread */ + xpcCloneKThreadFailed, /* 25: cloning kernel thread failed */ + + xpcNoHeartbeat, /* 26: remote partition has no heartbeat */ + + xpcPioReadError, /* 27: PIO read error */ + xpcPhysAddrRegFailed, /* 28: registration of phys addr range failed */ + + xpcBteDirectoryError, /* 29: maps to BTEFAIL_DIR */ + xpcBtePoisonError, /* 30: maps to BTEFAIL_POISON */ + xpcBteWriteError, /* 31: maps to BTEFAIL_WERR */ + xpcBteAccessError, /* 32: maps to BTEFAIL_ACCESS */ + xpcBtePWriteError, /* 33: maps to BTEFAIL_PWERR */ + xpcBtePReadError, /* 34: maps to BTEFAIL_PRERR */ + xpcBteTimeOutError, /* 35: maps to BTEFAIL_TOUT */ + xpcBteXtalkError, /* 36: maps to BTEFAIL_XTERR */ + xpcBteNotAvailable, /* 37: maps to BTEFAIL_NOTAVAIL */ + xpcBteUnmappedError, /* 38: unmapped BTEFAIL_ error */ + + xpcBadVersion, /* 39: bad version number */ + xpcVarsNotSet, /* 40: the XPC variables are not set up */ + xpcNoRsvdPageAddr, /* 41: unable to get rsvd page's phys addr */ + xpcInvalidPartid, /* 42: invalid partition ID */ + xpcLocalPartid, /* 43: local partition ID */ + + xpcUnknownReason /* 44: unknown reason -- must be last in list */ +}; + + +/* + * Define the callout function types used by XPC to update the user on + * connection activity and state changes (via the user function registered by + * xpc_connect()) and to notify them of messages received and delivered (via + * the user function registered by xpc_send_notify()). + * + * The two function types are xpc_channel_func and xpc_notify_func and + * both share the following arguments, with the exception of "data", which + * only xpc_channel_func has. + * + * Arguments: + * + * reason - reason code. (See following table.) + * partid - partition ID associated with condition. + * ch_number - channel # associated with condition. + * data - pointer to optional data. (See following table.) + * key - pointer to optional user-defined value provided as the "key" + * argument to xpc_connect() or xpc_send_notify(). + * + * In the following table the "Optional Data" column applies to callouts made + * to functions registered by xpc_connect(). A "NA" in that column indicates + * that this reason code can be passed to functions registered by + * xpc_send_notify() (i.e. they don't have data arguments). + * + * Also, the first three reason codes in the following table indicate + * success, whereas the others indicate failure. When a failure reason code + * is received, one can assume that the channel is not connected. + * + * + * Reason Code | Cause | Optional Data + * =====================+================================+===================== + * xpcConnected | connection has been established| max #of entries + * | to the specified partition on | allowed in message + * | the specified channel | queue + * ---------------------+--------------------------------+--------------------- + * xpcMsgReceived | an XPC message arrived from | address of payload + * | the specified partition on the | + * | specified channel | [the user must call + * | | xpc_received() when + * | | finished with the + * | | payload] + * ---------------------+--------------------------------+--------------------- + * xpcMsgDelivered | notification that the message | NA + * | was delivered to the intended | + * | recipient and that they have | + * | acknowledged its receipt by | + * | calling xpc_received() | + * =====================+================================+===================== + * xpcUnequalMsgSizes | can't connect to the specified | NULL + * | partition on the specified | + * | channel because of mismatched | + * | message sizes | + * ---------------------+--------------------------------+--------------------- + * xpcNoMemory | insufficient memory avaiable | NULL + * | to allocate message queue | + * ---------------------+--------------------------------+--------------------- + * xpcLackOfResources | lack of resources to create | NULL + * | the necessary kthreads to | + * | support the channel | + * ---------------------+--------------------------------+--------------------- + * xpcUnregistering | this side's user has | NULL or NA + * | unregistered by calling | + * | xpc_disconnect() | + * ---------------------+--------------------------------+--------------------- + * xpcOtherUnregistering| the other side's user has | NULL or NA + * | unregistered by calling | + * | xpc_disconnect() | + * ---------------------+--------------------------------+--------------------- + * xpcNoHeartbeat | the other side's XPC is no | NULL or NA + * | longer heartbeating | + * | | + * ---------------------+--------------------------------+--------------------- + * xpcUnloading | this side's XPC module is | NULL or NA + * | being unloaded | + * | | + * ---------------------+--------------------------------+--------------------- + * xpcOtherUnloading | the other side's XPC module is | NULL or NA + * | is being unloaded | + * | | + * ---------------------+--------------------------------+--------------------- + * xpcPioReadError | xp_nofault_PIOR() returned an | NULL or NA + * | error while sending an IPI | + * | | + * ---------------------+--------------------------------+--------------------- + * xpcInvalidAddress | the address either received or | NULL or NA + * | sent by the specified partition| + * | is invalid | + * ---------------------+--------------------------------+--------------------- + * xpcBteNotAvailable | attempt to pull data from the | NULL or NA + * xpcBtePoisonError | specified partition over the | + * xpcBteWriteError | specified channel via a | + * xpcBteAccessError | bte_copy() failed | + * xpcBteTimeOutError | | + * xpcBteXtalkError | | + * xpcBteDirectoryError | | + * xpcBteGenericError | | + * xpcBteUnmappedError | | + * ---------------------+--------------------------------+--------------------- + * xpcUnknownReason | the specified channel to the | NULL or NA + * | specified partition was | + * | unavailable for unknown reasons| + * =====================+================================+===================== + */ + +typedef void (*xpc_channel_func)(enum xpc_retval reason, partid_t partid, + int ch_number, void *data, void *key); + +typedef void (*xpc_notify_func)(enum xpc_retval reason, partid_t partid, + int ch_number, void *key); + + +/* + * The following is a registration entry. There is a global array of these, + * one per channel. It is used to record the connection registration made + * by the users of XPC. As long as a registration entry exists, for any + * partition that comes up, XPC will attempt to establish a connection on + * that channel. Notification that a connection has been made will occur via + * the xpc_channel_func function. + * + * The 'func' field points to the function to call when aynchronous + * notification is required for such events as: a connection established/lost, + * or an incomming message received, or an error condition encountered. A + * non-NULL 'func' field indicates that there is an active registration for + * the channel. + */ +struct xpc_registration { + struct semaphore sema; + xpc_channel_func func; /* function to call */ + void *key; /* pointer to user's key */ + u16 nentries; /* #of msg entries in local msg queue */ + u16 msg_size; /* message queue's message size */ + u32 assigned_limit; /* limit on #of assigned kthreads */ + u32 idle_limit; /* limit on #of idle kthreads */ +} ____cacheline_aligned; + + +#define XPC_CHANNEL_REGISTERED(_c) (xpc_registrations[_c].func != NULL) + + +/* the following are valid xpc_allocate() flags */ +#define XPC_WAIT 0 /* wait flag */ +#define XPC_NOWAIT 1 /* no wait flag */ + + +struct xpc_interface { + void (*connect)(int); + void (*disconnect)(int); + enum xpc_retval (*allocate)(partid_t, int, u32, void **); + enum xpc_retval (*send)(partid_t, int, void *); + enum xpc_retval (*send_notify)(partid_t, int, void *, + xpc_notify_func, void *); + void (*received)(partid_t, int, void *); + enum xpc_retval (*partid_to_nasids)(partid_t, void *); +}; + + +extern struct xpc_interface xpc_interface; + +extern void xpc_set_interface(void (*)(int), + void (*)(int), + enum xpc_retval (*)(partid_t, int, u32, void **), + enum xpc_retval (*)(partid_t, int, void *), + enum xpc_retval (*)(partid_t, int, void *, xpc_notify_func, + void *), + void (*)(partid_t, int, void *), + enum xpc_retval (*)(partid_t, void *)); +extern void xpc_clear_interface(void); + + +extern enum xpc_retval xpc_connect(int, xpc_channel_func, void *, u16, + u16, u32, u32); +extern void xpc_disconnect(int); + +static inline enum xpc_retval +xpc_allocate(partid_t partid, int ch_number, u32 flags, void **payload) +{ + return xpc_interface.allocate(partid, ch_number, flags, payload); +} + +static inline enum xpc_retval +xpc_send(partid_t partid, int ch_number, void *payload) +{ + return xpc_interface.send(partid, ch_number, payload); +} + +static inline enum xpc_retval +xpc_send_notify(partid_t partid, int ch_number, void *payload, + xpc_notify_func func, void *key) +{ + return xpc_interface.send_notify(partid, ch_number, payload, func, key); +} + +static inline void +xpc_received(partid_t partid, int ch_number, void *payload) +{ + return xpc_interface.received(partid, ch_number, payload); +} + +static inline enum xpc_retval +xpc_partid_to_nasids(partid_t partid, void *nasids) +{ + return xpc_interface.partid_to_nasids(partid, nasids); +} + + +extern u64 xp_nofault_PIOR_target; +extern int xp_nofault_PIOR(void *); +extern int xp_error_PIOR(void); + + +#endif /* _ASM_IA64_SN_XP_H */ + diff --git a/include/asm-m32r/signal.h b/include/asm-m32r/signal.h index ce46eaea449..95f69b19195 100644 --- a/include/asm-m32r/signal.h +++ b/include/asm-m32r/signal.h @@ -114,34 +114,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void __signalfn_t(int); -typedef __signalfn_t __user *__sighandler_t; - -typedef void __restorefn_t(void); -typedef __restorefn_t __user *__sigrestore_t; - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-m68k/bug.h b/include/asm-m68k/bug.h index 3e1d2266fa6..072ce274d53 100644 --- a/include/asm-m68k/bug.h +++ b/include/asm-m68k/bug.h @@ -3,6 +3,7 @@ #include <linux/config.h> +#ifdef CONFIG_BUG #ifdef CONFIG_DEBUG_BUGVERBOSE #ifndef CONFIG_SUN3 #define BUG() do { \ @@ -22,6 +23,8 @@ #endif #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-m68k/signal.h b/include/asm-m68k/signal.h index 6681bb6a552..a0cdf908237 100644 --- a/include/asm-m68k/signal.h +++ b/include/asm-m68k/signal.h @@ -105,42 +105,20 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { __sighandler_t sa_handler; old_sigset_t sa_mask; unsigned long sa_flags; - void (*sa_restorer)(void); + __sigrestore_t sa_restorer; }; struct sigaction { __sighandler_t sa_handler; unsigned long sa_flags; - void (*sa_restorer)(void); + __sigrestore_t sa_restorer; sigset_t sa_mask; /* mask last for extensibility */ }; diff --git a/include/asm-m68knommu/MC68328.h b/include/asm-m68knommu/MC68328.h index 4f5a9845f5b..a337e56d09b 100644 --- a/include/asm-m68knommu/MC68328.h +++ b/include/asm-m68knommu/MC68328.h @@ -993,7 +993,7 @@ typedef volatile struct { volatile unsigned short int pad1; volatile unsigned short int pad2; volatile unsigned short int pad3; -} m68328_uart __attribute__((packed)); +} __attribute__((packed)) m68328_uart; /********** diff --git a/include/asm-m68knommu/MC68EZ328.h b/include/asm-m68knommu/MC68EZ328.h index 801933da4c7..69b7f9139e5 100644 --- a/include/asm-m68knommu/MC68EZ328.h +++ b/include/asm-m68knommu/MC68EZ328.h @@ -815,7 +815,7 @@ typedef volatile struct { volatile unsigned short int nipr; volatile unsigned short int pad1; volatile unsigned short int pad2; -} m68328_uart __attribute__((packed)); +} __attribute__((packed)) m68328_uart; /********** diff --git a/include/asm-m68knommu/MC68VZ328.h b/include/asm-m68knommu/MC68VZ328.h index df74322f37e..2b9bf626a0a 100644 --- a/include/asm-m68knommu/MC68VZ328.h +++ b/include/asm-m68knommu/MC68VZ328.h @@ -909,7 +909,7 @@ typedef struct { volatile unsigned short int nipr; volatile unsigned short int hmark; volatile unsigned short int unused; -} m68328_uart __attribute__((packed)); +} __attribute__((packed)) m68328_uart; diff --git a/include/asm-m68knommu/signal.h b/include/asm-m68knommu/signal.h index 486cbb0dc08..1d13187f606 100644 --- a/include/asm-m68knommu/signal.h +++ b/include/asm-m68knommu/signal.h @@ -105,29 +105,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-mips/bug.h b/include/asm-mips/bug.h index eb94bb96cfb..3f594b440ab 100644 --- a/include/asm-mips/bug.h +++ b/include/asm-mips/bug.h @@ -3,12 +3,14 @@ #include <asm/break.h> +#ifdef CONFIG_BUG +#define HAVE_ARCH_BUG #define BUG() \ do { \ __asm__ __volatile__("break %0" : : "i" (BRK_BUG)); \ } while (0) +#endif -#define HAVE_ARCH_BUG #include <asm-generic/bug.h> #endif diff --git a/include/asm-mips/errno.h b/include/asm-mips/errno.h index 2b458f9538c..3c0d840e457 100644 --- a/include/asm-mips/errno.h +++ b/include/asm-mips/errno.h @@ -115,6 +115,10 @@ #define EKEYREVOKED 163 /* Key has been revoked */ #define EKEYREJECTED 164 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 165 /* Owner died */ +#define ENOTRECOVERABLE 166 /* State not recoverable */ + #define EDQUOT 1133 /* Quota exceeded */ #ifdef __KERNEL__ diff --git a/include/asm-mips/siginfo.h b/include/asm-mips/siginfo.h index 8ddd3c99bcf..a0e26e6c994 100644 --- a/include/asm-mips/siginfo.h +++ b/include/asm-mips/siginfo.h @@ -11,8 +11,6 @@ #include <linux/config.h> -#define SIGEV_HEAD_SIZE (sizeof(long) + 2*sizeof(int)) -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE-SIGEV_HEAD_SIZE) / sizeof(int)) #undef __ARCH_SI_TRAPNO /* exception code needs to fill this ... */ #define HAVE_ARCH_SIGINFO_T diff --git a/include/asm-mips/signal.h b/include/asm-mips/signal.h index 994987db61b..f2c470f1d36 100644 --- a/include/asm-mips/signal.h +++ b/include/asm-mips/signal.h @@ -98,34 +98,12 @@ typedef unsigned long old_sigset_t; /* at least 32 bits */ #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ flag is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x02000000 - -#endif /* __KERNEL__ */ - #define SIG_BLOCK 1 /* for blocking signals */ #define SIG_UNBLOCK 2 /* for unblocking signals */ #define SIG_SETMASK 3 /* for setting the signal mask */ #define SIG_SETMASK32 256 /* Goodie from SGI for BSD compatibility: set only the low 32 bit of the sigset. */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -/* Fake signal functions */ -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> struct sigaction { unsigned int sa_flags; diff --git a/include/asm-parisc/bug.h b/include/asm-parisc/bug.h index e72f6e2b4b9..695588da41f 100644 --- a/include/asm-parisc/bug.h +++ b/include/asm-parisc/bug.h @@ -1,12 +1,14 @@ #ifndef _PARISC_BUG_H #define _PARISC_BUG_H +#ifdef CONFIG_BUG #define HAVE_ARCH_BUG #define BUG() do { \ printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \ dump_stack(); \ panic("BUG!"); \ } while (0) +#endif #include <asm-generic/bug.h> #endif diff --git a/include/asm-parisc/errno.h b/include/asm-parisc/errno.h index a10f109770f..08464c40547 100644 --- a/include/asm-parisc/errno.h +++ b/include/asm-parisc/errno.h @@ -115,5 +115,9 @@ #define ENOTSUP 252 /* Function not implemented (POSIX.4 / HPUX) */ #define ECANCELLED 253 /* aio request was canceled before complete (POSIX.4 / HPUX) */ +/* for robust mutexes */ +#define EOWNERDEAD 254 /* Owner died */ +#define ENOTRECOVERABLE 255 /* State not recoverable */ + #endif diff --git a/include/asm-parisc/floppy.h b/include/asm-parisc/floppy.h index 47f53df2cef..ca3aed768cd 100644 --- a/include/asm-parisc/floppy.h +++ b/include/asm-parisc/floppy.h @@ -235,7 +235,7 @@ static int hard_dma_setup(char *addr, unsigned long size, int mode, int io) return 0; } -struct fd_routine_l { +static struct fd_routine_l { int (*_request_dma)(unsigned int dmanr, const char * device_id); void (*_free_dma)(unsigned int dmanr); int (*_get_dma_residue)(unsigned int dummy); diff --git a/include/asm-parisc/signal.h b/include/asm-parisc/signal.h index 358f577c8eb..25cb23ef7dd 100644 --- a/include/asm-parisc/signal.h +++ b/include/asm-parisc/signal.h @@ -89,17 +89,6 @@ #define _NSIG_BPW BITS_PER_LONG #define _NSIG_WORDS (_NSIG / _NSIG_BPW) -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 - #endif /* __KERNEL__ */ #define SIG_BLOCK 0 /* for blocking signals */ diff --git a/include/asm-parisc/uaccess.h b/include/asm-parisc/uaccess.h index 8a08423b757..c1b5bdea53e 100644 --- a/include/asm-parisc/uaccess.h +++ b/include/asm-parisc/uaccess.h @@ -24,7 +24,7 @@ /* * Note that since kernel addresses are in a separate address space on - * parisc, we don't need to do anything for access_ok() or verify_area(). + * parisc, we don't need to do anything for access_ok(). * We just let the page fault handler do the right thing. This also means * that put_user is the same as __put_user, etc. */ diff --git a/include/asm-ppc/bug.h b/include/asm-ppc/bug.h index e99c6cb9d61..8b34fd682b0 100644 --- a/include/asm-ppc/bug.h +++ b/include/asm-ppc/bug.h @@ -14,6 +14,7 @@ struct bug_entry { */ #define BUG_WARNING_TRAP 0x1000000 +#ifdef CONFIG_BUG #define BUG() do { \ __asm__ __volatile__( \ "1: twi 31,0,0\n" \ @@ -50,6 +51,8 @@ struct bug_entry { #define HAVE_ARCH_BUG #define HAVE_ARCH_BUG_ON #define HAVE_ARCH_WARN_ON +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-ppc/hydra.h b/include/asm-ppc/hydra.h index 1134431431d..833a8aff2a8 100644 --- a/include/asm-ppc/hydra.h +++ b/include/asm-ppc/hydra.h @@ -51,7 +51,7 @@ struct Hydra { char OpenPIC[0x40000]; }; -extern volatile struct Hydra *Hydra; +extern volatile struct Hydra __iomem *Hydra; /* diff --git a/include/asm-ppc/pci-bridge.h b/include/asm-ppc/pci-bridge.h index 78e9be619f1..ffa423456c2 100644 --- a/include/asm-ppc/pci-bridge.h +++ b/include/asm-ppc/pci-bridge.h @@ -12,7 +12,7 @@ struct pci_controller; * pci_io_base returns the memory address at which you can access * the I/O space for PCI bus number `bus' (or NULL on error). */ -extern void *pci_bus_io_base(unsigned int bus); +extern void __iomem *pci_bus_io_base(unsigned int bus); extern unsigned long pci_bus_io_base_phys(unsigned int bus); extern unsigned long pci_bus_mem_base_phys(unsigned int bus); @@ -48,7 +48,7 @@ struct pci_controller { int last_busno; int bus_offset; - void *io_base_virt; + void __iomem *io_base_virt; unsigned long io_base_phys; /* Some machines (PReP) have a non 1:1 mapping of diff --git a/include/asm-ppc/pmac_feature.h b/include/asm-ppc/pmac_feature.h index 639b690ce6f..8beb162873f 100644 --- a/include/asm-ppc/pmac_feature.h +++ b/include/asm-ppc/pmac_feature.h @@ -316,6 +316,9 @@ extern void pmac_register_agp_pm(struct pci_dev *bridge, extern void pmac_suspend_agp_for_card(struct pci_dev *dev); extern void pmac_resume_agp_for_card(struct pci_dev *dev); +/* Used by the via-pmu driver for suspend/resume + */ +extern void pmac_tweak_clock_spreading(int enable); /* * The part below is for use by macio_asic.c only, do not rely diff --git a/include/asm-ppc/reg_booke.h b/include/asm-ppc/reg_booke.h index e70c25f3c33..45c5e6f2b7a 100644 --- a/include/asm-ppc/reg_booke.h +++ b/include/asm-ppc/reg_booke.h @@ -305,6 +305,7 @@ do { \ #define ESR_PIL 0x08000000 /* Program Exception - Illegal */ #define ESR_PPR 0x04000000 /* Program Exception - Priveleged */ #define ESR_PTR 0x02000000 /* Program Exception - Trap */ +#define ESR_FP 0x01000000 /* Floating Point Operation */ #define ESR_DST 0x00800000 /* Storage Exception - Data miss */ #define ESR_DIZ 0x00400000 /* Storage Exception - Zone fault */ #define ESR_ST 0x00800000 /* Store Operation */ diff --git a/include/asm-ppc/sigcontext.h b/include/asm-ppc/sigcontext.h index fc5e358c65f..f82dcccdee1 100644 --- a/include/asm-ppc/sigcontext.h +++ b/include/asm-ppc/sigcontext.h @@ -9,7 +9,7 @@ struct sigcontext { int signal; unsigned long handler; unsigned long oldmask; - struct pt_regs *regs; + struct pt_regs __user *regs; }; #endif diff --git a/include/asm-ppc/signal.h b/include/asm-ppc/signal.h index 8cc8b88d4ed..caf6ede3710 100644 --- a/include/asm-ppc/signal.h +++ b/include/asm-ppc/signal.h @@ -99,34 +99,8 @@ typedef struct { #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif /* __KERNEL__ */ - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void __signalfn_t(int); -typedef __signalfn_t __user *__sighandler_t; - -typedef void __restorefn_t(void); -typedef __restorefn_t __user *__sigrestore_t; -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> struct old_sigaction { __sighandler_t sa_handler; diff --git a/include/asm-ppc64/a.out.h b/include/asm-ppc64/a.out.h index 802338efcb1..3871e252a6f 100644 --- a/include/asm-ppc64/a.out.h +++ b/include/asm-ppc64/a.out.h @@ -1,8 +1,6 @@ #ifndef __PPC64_A_OUT_H__ #define __PPC64_A_OUT_H__ -#include <asm/ppcdebug.h> - /* * c 2001 PPC 64 Team, IBM Corp * diff --git a/include/asm-ppc64/bug.h b/include/asm-ppc64/bug.h index db31dd22233..169868fa307 100644 --- a/include/asm-ppc64/bug.h +++ b/include/asm-ppc64/bug.h @@ -26,6 +26,8 @@ struct bug_entry *find_bug(unsigned long bugaddr); */ #define BUG_WARNING_TRAP 0x1000000 +#ifdef CONFIG_BUG + #define BUG() do { \ __asm__ __volatile__( \ "1: twi 31,0,0\n" \ @@ -55,11 +57,12 @@ struct bug_entry *find_bug(unsigned long bugaddr); "i" (__FILE__), "i" (__FUNCTION__)); \ } while (0) -#endif - #define HAVE_ARCH_BUG #define HAVE_ARCH_BUG_ON #define HAVE_ARCH_WARN_ON +#endif +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-ppc64/elf.h b/include/asm-ppc64/elf.h index 8457d90244f..6c42d61bedd 100644 --- a/include/asm-ppc64/elf.h +++ b/include/asm-ppc64/elf.h @@ -229,9 +229,13 @@ do { \ /* * An executable for which elf_read_implies_exec() returns TRUE will - * have the READ_IMPLIES_EXEC personality flag set automatically. + * have the READ_IMPLIES_EXEC personality flag set automatically. This + * is only required to work around bugs in old 32bit toolchains. Since + * the 64bit ABI has never had these issues dont enable the workaround + * even if we have an executable stack. */ -#define elf_read_implies_exec(ex, exec_stk) (exec_stk != EXSTACK_DISABLE_X) +#define elf_read_implies_exec(ex, exec_stk) (test_thread_flag(TIF_32BIT) ? \ + (exec_stk != EXSTACK_DISABLE_X) : 0) #endif diff --git a/include/asm-ppc64/imalloc.h b/include/asm-ppc64/imalloc.h new file mode 100644 index 00000000000..3a45e918bf1 --- /dev/null +++ b/include/asm-ppc64/imalloc.h @@ -0,0 +1,24 @@ +#ifndef _PPC64_IMALLOC_H +#define _PPC64_IMALLOC_H + +/* + * Define the address range of the imalloc VM area. + */ +#define PHBS_IO_BASE IOREGIONBASE +#define IMALLOC_BASE (IOREGIONBASE + 0x80000000ul) /* Reserve 2 gigs for PHBs */ +#define IMALLOC_END (IOREGIONBASE + EADDR_MASK) + + +/* imalloc region types */ +#define IM_REGION_UNUSED 0x1 +#define IM_REGION_SUBSET 0x2 +#define IM_REGION_EXISTS 0x4 +#define IM_REGION_OVERLAP 0x8 +#define IM_REGION_SUPERSET 0x10 + +extern struct vm_struct * im_get_free_area(unsigned long size); +extern struct vm_struct * im_get_area(unsigned long v_addr, unsigned long size, + int region_type); +unsigned long im_free(void *addr); + +#endif /* _PPC64_IMALLOC_H */ diff --git a/include/asm-ppc64/mmu.h b/include/asm-ppc64/mmu.h index 188987e9d9d..c78282a67d8 100644 --- a/include/asm-ppc64/mmu.h +++ b/include/asm-ppc64/mmu.h @@ -15,19 +15,10 @@ #include <linux/config.h> #include <asm/page.h> -#include <linux/stringify.h> -#ifndef __ASSEMBLY__ - -/* Time to allow for more things here */ -typedef unsigned long mm_context_id_t; -typedef struct { - mm_context_id_t id; -#ifdef CONFIG_HUGETLB_PAGE - pgd_t *huge_pgdir; - u16 htlb_segs; /* bitmask */ -#endif -} mm_context_t; +/* + * Segment table + */ #define STE_ESID_V 0x80 #define STE_ESID_KS 0x20 @@ -36,15 +27,48 @@ typedef struct { #define STE_VSID_SHIFT 12 -struct stab_entry { - unsigned long esid_data; - unsigned long vsid_data; -}; +/* Location of cpu0's segment table */ +#define STAB0_PAGE 0x9 +#define STAB0_PHYS_ADDR (STAB0_PAGE<<PAGE_SHIFT) +#define STAB0_VIRT_ADDR (KERNELBASE+STAB0_PHYS_ADDR) + +/* + * SLB + */ -/* Hardware Page Table Entry */ +#define SLB_NUM_BOLTED 3 +#define SLB_CACHE_ENTRIES 8 + +/* Bits in the SLB ESID word */ +#define SLB_ESID_V ASM_CONST(0x0000000008000000) /* valid */ + +/* Bits in the SLB VSID word */ +#define SLB_VSID_SHIFT 12 +#define SLB_VSID_KS ASM_CONST(0x0000000000000800) +#define SLB_VSID_KP ASM_CONST(0x0000000000000400) +#define SLB_VSID_N ASM_CONST(0x0000000000000200) /* no-execute */ +#define SLB_VSID_L ASM_CONST(0x0000000000000100) /* largepage 16M */ +#define SLB_VSID_C ASM_CONST(0x0000000000000080) /* class */ + +#define SLB_VSID_KERNEL (SLB_VSID_KP|SLB_VSID_C) +#define SLB_VSID_USER (SLB_VSID_KP|SLB_VSID_KS) + +/* + * Hash table + */ #define HPTES_PER_GROUP 8 +/* Values for PP (assumes Ks=0, Kp=1) */ +/* pp0 will always be 0 for linux */ +#define PP_RWXX 0 /* Supervisor read/write, User none */ +#define PP_RWRX 1 /* Supervisor read/write, User read */ +#define PP_RWRW 2 /* Supervisor read/write, User read/write */ +#define PP_RXRX 3 /* Supervisor read, User read */ + +#ifndef __ASSEMBLY__ + +/* Hardware Page Table Entry */ typedef struct { unsigned long avpn:57; /* vsid | api == avpn */ unsigned long : 2; /* Software use */ @@ -90,14 +114,6 @@ typedef struct { } dw1; } HPTE; -/* Values for PP (assumes Ks=0, Kp=1) */ -/* pp0 will always be 0 for linux */ -#define PP_RWXX 0 /* Supervisor read/write, User none */ -#define PP_RWRX 1 /* Supervisor read/write, User read */ -#define PP_RWRW 2 /* Supervisor read/write, User read/write */ -#define PP_RXRX 3 /* Supervisor read, User read */ - - extern HPTE * htab_address; extern unsigned long htab_hash_mask; @@ -174,31 +190,70 @@ extern int __hash_page(unsigned long ea, unsigned long access, extern void htab_finish_init(void); +extern void hpte_init_native(void); +extern void hpte_init_lpar(void); +extern void hpte_init_iSeries(void); + +extern long pSeries_lpar_hpte_insert(unsigned long hpte_group, + unsigned long va, unsigned long prpn, + int secondary, unsigned long hpteflags, + int bolted, int large); +extern long native_hpte_insert(unsigned long hpte_group, unsigned long va, + unsigned long prpn, int secondary, + unsigned long hpteflags, int bolted, int large); + #endif /* __ASSEMBLY__ */ /* - * Location of cpu0's segment table + * VSID allocation + * + * We first generate a 36-bit "proto-VSID". For kernel addresses this + * is equal to the ESID, for user addresses it is: + * (context << 15) | (esid & 0x7fff) + * + * The two forms are distinguishable because the top bit is 0 for user + * addresses, whereas the top two bits are 1 for kernel addresses. + * Proto-VSIDs with the top two bits equal to 0b10 are reserved for + * now. + * + * The proto-VSIDs are then scrambled into real VSIDs with the + * multiplicative hash: + * + * VSID = (proto-VSID * VSID_MULTIPLIER) % VSID_MODULUS + * where VSID_MULTIPLIER = 268435399 = 0xFFFFFC7 + * VSID_MODULUS = 2^36-1 = 0xFFFFFFFFF + * + * This scramble is only well defined for proto-VSIDs below + * 0xFFFFFFFFF, so both proto-VSID and actual VSID 0xFFFFFFFFF are + * reserved. VSID_MULTIPLIER is prime, so in particular it is + * co-prime to VSID_MODULUS, making this a 1:1 scrambling function. + * Because the modulus is 2^n-1 we can compute it efficiently without + * a divide or extra multiply (see below). + * + * This scheme has several advantages over older methods: + * + * - We have VSIDs allocated for every kernel address + * (i.e. everything above 0xC000000000000000), except the very top + * segment, which simplifies several things. + * + * - We allow for 15 significant bits of ESID and 20 bits of + * context for user addresses. i.e. 8T (43 bits) of address space for + * up to 1M contexts (although the page table structure and context + * allocation will need changes to take advantage of this). + * + * - The scramble function gives robust scattering in the hash + * table (at least based on some initial results). The previous + * method was more susceptible to pathological cases giving excessive + * hash collisions. + */ +/* + * WARNING - If you change these you must make sure the asm + * implementations in slb_allocate (slb_low.S), do_stab_bolted + * (head.S) and ASM_VSID_SCRAMBLE (below) are changed accordingly. + * + * You'll also need to change the precomputed VSID values in head.S + * which are used by the iSeries firmware. */ -#define STAB0_PAGE 0x9 -#define STAB0_PHYS_ADDR (STAB0_PAGE<<PAGE_SHIFT) -#define STAB0_VIRT_ADDR (KERNELBASE+STAB0_PHYS_ADDR) - -#define SLB_NUM_BOLTED 3 -#define SLB_CACHE_ENTRIES 8 - -/* Bits in the SLB ESID word */ -#define SLB_ESID_V 0x0000000008000000 /* entry is valid */ - -/* Bits in the SLB VSID word */ -#define SLB_VSID_SHIFT 12 -#define SLB_VSID_KS 0x0000000000000800 -#define SLB_VSID_KP 0x0000000000000400 -#define SLB_VSID_N 0x0000000000000200 /* no-execute */ -#define SLB_VSID_L 0x0000000000000100 /* largepage (4M) */ -#define SLB_VSID_C 0x0000000000000080 /* class */ - -#define SLB_VSID_KERNEL (SLB_VSID_KP|SLB_VSID_C) -#define SLB_VSID_USER (SLB_VSID_KP|SLB_VSID_KS) #define VSID_MULTIPLIER ASM_CONST(200730139) /* 28-bit prime */ #define VSID_BITS 36 @@ -239,4 +294,50 @@ extern void htab_finish_init(void); srdi rx,rx,VSID_BITS; /* extract 2^36 bit */ \ add rt,rt,rx + +#ifndef __ASSEMBLY__ + +typedef unsigned long mm_context_id_t; + +typedef struct { + mm_context_id_t id; +#ifdef CONFIG_HUGETLB_PAGE + pgd_t *huge_pgdir; + u16 htlb_segs; /* bitmask */ +#endif +} mm_context_t; + + +static inline unsigned long vsid_scramble(unsigned long protovsid) +{ +#if 0 + /* The code below is equivalent to this function for arguments + * < 2^VSID_BITS, which is all this should ever be called + * with. However gcc is not clever enough to compute the + * modulus (2^n-1) without a second multiply. */ + return ((protovsid * VSID_MULTIPLIER) % VSID_MODULUS); +#else /* 1 */ + unsigned long x; + + x = protovsid * VSID_MULTIPLIER; + x = (x >> VSID_BITS) + (x & VSID_MODULUS); + return (x + ((x+1) >> VSID_BITS)) & VSID_MODULUS; +#endif /* 1 */ +} + +/* This is only valid for addresses >= KERNELBASE */ +static inline unsigned long get_kernel_vsid(unsigned long ea) +{ + return vsid_scramble(ea >> SID_SHIFT); +} + +/* This is only valid for user addresses (which are below 2^41) */ +static inline unsigned long get_vsid(unsigned long context, unsigned long ea) +{ + return vsid_scramble((context << USER_ESID_BITS) + | (ea >> SID_SHIFT)); +} + +#endif /* __ASSEMBLY */ + #endif /* _PPC64_MMU_H_ */ diff --git a/include/asm-ppc64/mmu_context.h b/include/asm-ppc64/mmu_context.h index c2e8e046638..77a743402db 100644 --- a/include/asm-ppc64/mmu_context.h +++ b/include/asm-ppc64/mmu_context.h @@ -84,86 +84,4 @@ static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next) local_irq_restore(flags); } -/* VSID allocation - * =============== - * - * We first generate a 36-bit "proto-VSID". For kernel addresses this - * is equal to the ESID, for user addresses it is: - * (context << 15) | (esid & 0x7fff) - * - * The two forms are distinguishable because the top bit is 0 for user - * addresses, whereas the top two bits are 1 for kernel addresses. - * Proto-VSIDs with the top two bits equal to 0b10 are reserved for - * now. - * - * The proto-VSIDs are then scrambled into real VSIDs with the - * multiplicative hash: - * - * VSID = (proto-VSID * VSID_MULTIPLIER) % VSID_MODULUS - * where VSID_MULTIPLIER = 268435399 = 0xFFFFFC7 - * VSID_MODULUS = 2^36-1 = 0xFFFFFFFFF - * - * This scramble is only well defined for proto-VSIDs below - * 0xFFFFFFFFF, so both proto-VSID and actual VSID 0xFFFFFFFFF are - * reserved. VSID_MULTIPLIER is prime, so in particular it is - * co-prime to VSID_MODULUS, making this a 1:1 scrambling function. - * Because the modulus is 2^n-1 we can compute it efficiently without - * a divide or extra multiply (see below). - * - * This scheme has several advantages over older methods: - * - * - We have VSIDs allocated for every kernel address - * (i.e. everything above 0xC000000000000000), except the very top - * segment, which simplifies several things. - * - * - We allow for 15 significant bits of ESID and 20 bits of - * context for user addresses. i.e. 8T (43 bits) of address space for - * up to 1M contexts (although the page table structure and context - * allocation will need changes to take advantage of this). - * - * - The scramble function gives robust scattering in the hash - * table (at least based on some initial results). The previous - * method was more susceptible to pathological cases giving excessive - * hash collisions. - */ - -/* - * WARNING - If you change these you must make sure the asm - * implementations in slb_allocate(), do_stab_bolted and mmu.h - * (ASM_VSID_SCRAMBLE macro) are changed accordingly. - * - * You'll also need to change the precomputed VSID values in head.S - * which are used by the iSeries firmware. - */ - -static inline unsigned long vsid_scramble(unsigned long protovsid) -{ -#if 0 - /* The code below is equivalent to this function for arguments - * < 2^VSID_BITS, which is all this should ever be called - * with. However gcc is not clever enough to compute the - * modulus (2^n-1) without a second multiply. */ - return ((protovsid * VSID_MULTIPLIER) % VSID_MODULUS); -#else /* 1 */ - unsigned long x; - - x = protovsid * VSID_MULTIPLIER; - x = (x >> VSID_BITS) + (x & VSID_MODULUS); - return (x + ((x+1) >> VSID_BITS)) & VSID_MODULUS; -#endif /* 1 */ -} - -/* This is only valid for addresses >= KERNELBASE */ -static inline unsigned long get_kernel_vsid(unsigned long ea) -{ - return vsid_scramble(ea >> SID_SHIFT); -} - -/* This is only valid for user addresses (which are below 2^41) */ -static inline unsigned long get_vsid(unsigned long context, unsigned long ea) -{ - return vsid_scramble((context << USER_ESID_BITS) - | (ea >> SID_SHIFT)); -} - #endif /* __PPC64_MMU_CONTEXT_H */ diff --git a/include/asm-ppc64/page.h b/include/asm-ppc64/page.h index 20e0f19324e..bcd21789d3b 100644 --- a/include/asm-ppc64/page.h +++ b/include/asm-ppc64/page.h @@ -23,7 +23,6 @@ #define PAGE_SHIFT 12 #define PAGE_SIZE (ASM_CONST(1) << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) -#define PAGE_OFFSET_MASK (PAGE_SIZE-1) #define SID_SHIFT 28 #define SID_MASK 0xfffffffffUL @@ -85,9 +84,6 @@ /* align addr on a size boundary - adjust address up if needed */ #define _ALIGN(addr,size) _ALIGN_UP(addr,size) -/* to align the pointer to the (next) double word boundary */ -#define DOUBLEWORD_ALIGN(addr) _ALIGN(addr,sizeof(unsigned long)) - /* to align the pointer to the (next) page boundary */ #define PAGE_ALIGN(addr) _ALIGN(addr, PAGE_SIZE) @@ -100,7 +96,6 @@ #define REGION_SIZE 4UL #define REGION_SHIFT 60UL #define REGION_MASK (((1UL<<REGION_SIZE)-1UL)<<REGION_SHIFT) -#define REGION_STRIDE (1UL << REGION_SHIFT) static __inline__ void clear_page(void *addr) { @@ -209,13 +204,13 @@ extern u64 ppc64_pft_size; /* Log 2 of page table size */ #define VMALLOCBASE ASM_CONST(0xD000000000000000) #define IOREGIONBASE ASM_CONST(0xE000000000000000) -#define IO_REGION_ID (IOREGIONBASE>>REGION_SHIFT) -#define VMALLOC_REGION_ID (VMALLOCBASE>>REGION_SHIFT) -#define KERNEL_REGION_ID (KERNELBASE>>REGION_SHIFT) +#define IO_REGION_ID (IOREGIONBASE >> REGION_SHIFT) +#define VMALLOC_REGION_ID (VMALLOCBASE >> REGION_SHIFT) +#define KERNEL_REGION_ID (KERNELBASE >> REGION_SHIFT) #define USER_REGION_ID (0UL) -#define REGION_ID(X) (((unsigned long)(X))>>REGION_SHIFT) +#define REGION_ID(ea) (((unsigned long)(ea)) >> REGION_SHIFT) -#define __bpn_to_ba(x) ((((unsigned long)(x))<<PAGE_SHIFT) + KERNELBASE) +#define __bpn_to_ba(x) ((((unsigned long)(x)) << PAGE_SHIFT) + KERNELBASE) #define __ba_to_bpn(x) ((((unsigned long)(x)) & ~REGION_MASK) >> PAGE_SHIFT) #define __va(x) ((void *)((unsigned long)(x) + KERNELBASE)) @@ -252,10 +247,19 @@ extern u64 ppc64_pft_size; /* Log 2 of page table size */ /* * This is the default if a program doesn't have a PT_GNU_STACK - * program header entry. + * program header entry. The PPC64 ELF ABI has a non executable stack + * stack by default, so in the absense of a PT_GNU_STACK program header + * we turn execute permission off. */ -#define VM_STACK_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ - VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) +#define VM_STACK_DEFAULT_FLAGS32 (VM_READ | VM_WRITE | VM_EXEC | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + +#define VM_STACK_DEFAULT_FLAGS64 (VM_READ | VM_WRITE | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + +#define VM_STACK_DEFAULT_FLAGS \ + (test_thread_flag(TIF_32BIT) ? \ + VM_STACK_DEFAULT_FLAGS32 : VM_STACK_DEFAULT_FLAGS64) #endif /* __KERNEL__ */ #endif /* _PPC64_PAGE_H */ diff --git a/include/asm-ppc64/pgalloc.h b/include/asm-ppc64/pgalloc.h index 16232d74017..4fc4b739b38 100644 --- a/include/asm-ppc64/pgalloc.h +++ b/include/asm-ppc64/pgalloc.h @@ -27,7 +27,7 @@ pgd_free(pgd_t *pgd) kmem_cache_free(zero_cache, pgd); } -#define pgd_populate(MM, PGD, PMD) pgd_set(PGD, PMD) +#define pud_populate(MM, PUD, PMD) pud_set(PUD, PMD) static inline pmd_t * pmd_alloc_one(struct mm_struct *mm, unsigned long addr) diff --git a/include/asm-ppc64/pgtable.h b/include/asm-ppc64/pgtable.h index a26120517c5..264c4f7993b 100644 --- a/include/asm-ppc64/pgtable.h +++ b/include/asm-ppc64/pgtable.h @@ -1,8 +1,6 @@ #ifndef _PPC64_PGTABLE_H #define _PPC64_PGTABLE_H -#include <asm-generic/4level-fixup.h> - /* * This file contains the functions and defines necessary to modify and use * the ppc64 hashed page table. @@ -17,15 +15,7 @@ #include <asm/tlbflush.h> #endif /* __ASSEMBLY__ */ -/* PMD_SHIFT determines what a second-level page table entry can map */ -#define PMD_SHIFT (PAGE_SHIFT + PAGE_SHIFT - 3) -#define PMD_SIZE (1UL << PMD_SHIFT) -#define PMD_MASK (~(PMD_SIZE-1)) - -/* PGDIR_SHIFT determines what a third-level page table entry can map */ -#define PGDIR_SHIFT (PAGE_SHIFT + (PAGE_SHIFT - 3) + (PAGE_SHIFT - 2)) -#define PGDIR_SIZE (1UL << PGDIR_SHIFT) -#define PGDIR_MASK (~(PGDIR_SIZE-1)) +#include <asm-generic/pgtable-nopud.h> /* * Entries per page directory level. The PTE level must use a 64b record @@ -40,40 +30,30 @@ #define PTRS_PER_PMD (1 << PMD_INDEX_SIZE) #define PTRS_PER_PGD (1 << PGD_INDEX_SIZE) -#define USER_PTRS_PER_PGD (1024) -#define FIRST_USER_ADDRESS 0 +/* PMD_SHIFT determines what a second-level page table entry can map */ +#define PMD_SHIFT (PAGE_SHIFT + PTE_INDEX_SIZE) +#define PMD_SIZE (1UL << PMD_SHIFT) +#define PMD_MASK (~(PMD_SIZE-1)) -#define EADDR_SIZE (PTE_INDEX_SIZE + PMD_INDEX_SIZE + \ - PGD_INDEX_SIZE + PAGE_SHIFT) +/* PGDIR_SHIFT determines what a third-level page table entry can map */ +#define PGDIR_SHIFT (PMD_SHIFT + PMD_INDEX_SIZE) +#define PGDIR_SIZE (1UL << PGDIR_SHIFT) +#define PGDIR_MASK (~(PGDIR_SIZE-1)) + +#define FIRST_USER_ADDRESS 0 /* * Size of EA range mapped by our pagetables. */ -#define PGTABLE_EA_BITS 41 -#define PGTABLE_EA_MASK ((1UL<<PGTABLE_EA_BITS)-1) +#define EADDR_SIZE (PTE_INDEX_SIZE + PMD_INDEX_SIZE + \ + PGD_INDEX_SIZE + PAGE_SHIFT) +#define EADDR_MASK ((1UL << EADDR_SIZE) - 1) /* * Define the address range of the vmalloc VM area. */ #define VMALLOC_START (0xD000000000000000ul) -#define VMALLOC_END (VMALLOC_START + PGTABLE_EA_MASK) - -/* - * Define the address range of the imalloc VM area. - * (used for ioremap) - */ -#define IMALLOC_START (ioremap_bot) -#define IMALLOC_VMADDR(x) ((unsigned long)(x)) -#define PHBS_IO_BASE (0xE000000000000000ul) /* Reserve 2 gigs for PHBs */ -#define IMALLOC_BASE (0xE000000080000000ul) -#define IMALLOC_END (IMALLOC_BASE + PGTABLE_EA_MASK) - -/* - * Define the user address range - */ -#define USER_START (0UL) -#define USER_END (USER_START + PGTABLE_EA_MASK) - +#define VMALLOC_END (VMALLOC_START + EADDR_MASK) /* * Bits in a linux-style PTE. These match the bits in the @@ -168,10 +148,6 @@ extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)]; /* shift to put page number into pte */ #define PTE_SHIFT (17) -/* We allow 2^41 bytes of real memory, so we need 29 bits in the PMD - * to give the PTE page number. The bottom two bits are for flags. */ -#define PMD_TO_PTEPAGE_SHIFT (2) - #ifdef CONFIG_HUGETLB_PAGE #ifndef __ASSEMBLY__ @@ -200,13 +176,14 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm); */ #define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot)) -#define pfn_pte(pfn,pgprot) \ -({ \ - pte_t pte; \ - pte_val(pte) = ((unsigned long)(pfn) << PTE_SHIFT) | \ - pgprot_val(pgprot); \ - pte; \ -}) +static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot) +{ + pte_t pte; + + + pte_val(pte) = (pfn << PTE_SHIFT) | pgprot_val(pgprot); + return pte; +} #define pte_modify(_pte, newprot) \ (__pte((pte_val(_pte) & _PAGE_CHG_MASK) | pgprot_val(newprot))) @@ -220,20 +197,20 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm); #define pte_page(x) pfn_to_page(pte_pfn(x)) #define pmd_set(pmdp, ptep) \ - (pmd_val(*(pmdp)) = (__ba_to_bpn(ptep) << PMD_TO_PTEPAGE_SHIFT)) + (pmd_val(*(pmdp)) = __ba_to_bpn(ptep)) #define pmd_none(pmd) (!pmd_val(pmd)) #define pmd_bad(pmd) (pmd_val(pmd) == 0) #define pmd_present(pmd) (pmd_val(pmd) != 0) #define pmd_clear(pmdp) (pmd_val(*(pmdp)) = 0) -#define pmd_page_kernel(pmd) \ - (__bpn_to_ba(pmd_val(pmd) >> PMD_TO_PTEPAGE_SHIFT)) +#define pmd_page_kernel(pmd) (__bpn_to_ba(pmd_val(pmd))) #define pmd_page(pmd) virt_to_page(pmd_page_kernel(pmd)) -#define pgd_set(pgdp, pmdp) (pgd_val(*(pgdp)) = (__ba_to_bpn(pmdp))) -#define pgd_none(pgd) (!pgd_val(pgd)) -#define pgd_bad(pgd) ((pgd_val(pgd)) == 0) -#define pgd_present(pgd) (pgd_val(pgd) != 0UL) -#define pgd_clear(pgdp) (pgd_val(*(pgdp)) = 0UL) -#define pgd_page(pgd) (__bpn_to_ba(pgd_val(pgd))) + +#define pud_set(pudp, pmdp) (pud_val(*(pudp)) = (__ba_to_bpn(pmdp))) +#define pud_none(pud) (!pud_val(pud)) +#define pud_bad(pud) ((pud_val(pud)) == 0UL) +#define pud_present(pud) (pud_val(pud) != 0UL) +#define pud_clear(pudp) (pud_val(*(pudp)) = 0UL) +#define pud_page(pud) (__bpn_to_ba(pud_val(pud))) /* * Find an entry in a page-table-directory. We combine the address region @@ -245,12 +222,13 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm); #define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address)) /* Find an entry in the second-level page table.. */ -#define pmd_offset(dir,addr) \ - ((pmd_t *) pgd_page(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1))) +#define pmd_offset(pudp,addr) \ + ((pmd_t *) pud_page(*(pudp)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1))) /* Find an entry in the third-level page table.. */ #define pte_offset_kernel(dir,addr) \ - ((pte_t *) pmd_page_kernel(*(dir)) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))) + ((pte_t *) pmd_page_kernel(*(dir)) \ + + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))) #define pte_offset_map(dir,addr) pte_offset_kernel((dir), (addr)) #define pte_offset_map_nested(dir,addr) pte_offset_kernel((dir), (addr)) @@ -264,8 +242,6 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm); /* to find an entry in the ioremap page-table-directory */ #define pgd_offset_i(address) (ioremap_pgd + pgd_index(address)) -#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT)) - /* * The following only work if pte_present() is true. * Undefined behaviour if not.. @@ -440,7 +416,7 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr, pte_clear(mm, addr, ptep); flush_tlb_pending(); } - *ptep = __pte(pte_val(pte)) & ~_PAGE_HPTEFLAGS; + *ptep = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS); } /* Set the dirty and/or accessed bits atomically in a linux PTE, this @@ -485,18 +461,13 @@ extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long addr, extern unsigned long ioremap_bot, ioremap_base; -#define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT) -#define KERNEL_PGD_PTRS (PTRS_PER_PGD-USER_PGD_PTRS) - -#define pte_ERROR(e) \ - printk("%s:%d: bad pte %016lx.\n", __FILE__, __LINE__, pte_val(e)) #define pmd_ERROR(e) \ printk("%s:%d: bad pmd %08x.\n", __FILE__, __LINE__, pmd_val(e)) #define pgd_ERROR(e) \ printk("%s:%d: bad pgd %08x.\n", __FILE__, __LINE__, pgd_val(e)) -extern pgd_t swapper_pg_dir[1024]; -extern pgd_t ioremap_dir[1024]; +extern pgd_t swapper_pg_dir[]; +extern pgd_t ioremap_dir[]; extern void paging_init(void); @@ -538,43 +509,11 @@ extern void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t); */ #define kern_addr_valid(addr) (1) -#define io_remap_page_range(vma, vaddr, paddr, size, prot) \ - remap_pfn_range(vma, vaddr, (paddr) >> PAGE_SHIFT, size, prot) - #define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \ remap_pfn_range(vma, vaddr, pfn, size, prot) -#define MK_IOSPACE_PFN(space, pfn) (pfn) -#define GET_IOSPACE(pfn) 0 -#define GET_PFN(pfn) (pfn) - void pgtable_cache_init(void); -extern void hpte_init_native(void); -extern void hpte_init_lpar(void); -extern void hpte_init_iSeries(void); - -/* imalloc region types */ -#define IM_REGION_UNUSED 0x1 -#define IM_REGION_SUBSET 0x2 -#define IM_REGION_EXISTS 0x4 -#define IM_REGION_OVERLAP 0x8 -#define IM_REGION_SUPERSET 0x10 - -extern struct vm_struct * im_get_free_area(unsigned long size); -extern struct vm_struct * im_get_area(unsigned long v_addr, unsigned long size, - int region_type); -unsigned long im_free(void *addr); - -extern long pSeries_lpar_hpte_insert(unsigned long hpte_group, - unsigned long va, unsigned long prpn, - int secondary, unsigned long hpteflags, - int bolted, int large); - -extern long native_hpte_insert(unsigned long hpte_group, unsigned long va, - unsigned long prpn, int secondary, - unsigned long hpteflags, int bolted, int large); - /* * find_linux_pte returns the address of a linux pte for a given * effective address and directory. If not found, it returns zero. @@ -582,19 +521,22 @@ extern long native_hpte_insert(unsigned long hpte_group, unsigned long va, static inline pte_t *find_linux_pte(pgd_t *pgdir, unsigned long ea) { pgd_t *pg; + pud_t *pu; pmd_t *pm; pte_t *pt = NULL; pte_t pte; pg = pgdir + pgd_index(ea); if (!pgd_none(*pg)) { - - pm = pmd_offset(pg, ea); - if (pmd_present(*pm)) { - pt = pte_offset_kernel(pm, ea); - pte = *pt; - if (!pte_present(pte)) - pt = NULL; + pu = pud_offset(pg, ea); + if (!pud_none(*pu)) { + pm = pmd_offset(pu, ea); + if (pmd_present(*pm)) { + pt = pte_offset_kernel(pm, ea); + pte = *pt; + if (!pte_present(pte)) + pt = NULL; + } } } diff --git a/include/asm-ppc64/signal.h b/include/asm-ppc64/signal.h index fe5401adb41..432df7dd355 100644 --- a/include/asm-ppc64/signal.h +++ b/include/asm-ppc64/signal.h @@ -96,47 +96,20 @@ typedef struct { #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void __sigfunction(int); -typedef __sigfunction __user * __sighandler_t; - -/* Type of the restorer function */ -typedef void __sigrestorer(void); -typedef __sigrestorer __user * __sigrestorer_t; - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> struct old_sigaction { __sighandler_t sa_handler; old_sigset_t sa_mask; unsigned long sa_flags; - __sigrestorer_t sa_restorer; + __sigrestore_t sa_restorer; }; struct sigaction { __sighandler_t sa_handler; unsigned long sa_flags; - __sigrestorer_t sa_restorer; + __sigrestore_t sa_restorer; sigset_t sa_mask; /* mask last for extensibility */ }; diff --git a/include/asm-ppc64/spinlock.h b/include/asm-ppc64/spinlock.h index a9b2a1162cf..acd11564dd7 100644 --- a/include/asm-ppc64/spinlock.h +++ b/include/asm-ppc64/spinlock.h @@ -110,7 +110,7 @@ static void __inline__ _raw_spin_lock(spinlock_t *lock) HMT_low(); if (SHARED_PROCESSOR) __spin_yield(lock); - } while (likely(lock->lock != 0)); + } while (unlikely(lock->lock != 0)); HMT_medium(); } } @@ -128,7 +128,7 @@ static void __inline__ _raw_spin_lock_flags(spinlock_t *lock, unsigned long flag HMT_low(); if (SHARED_PROCESSOR) __spin_yield(lock); - } while (likely(lock->lock != 0)); + } while (unlikely(lock->lock != 0)); HMT_medium(); local_irq_restore(flags_dis); } @@ -194,7 +194,7 @@ static void __inline__ _raw_read_lock(rwlock_t *rw) HMT_low(); if (SHARED_PROCESSOR) __rw_yield(rw); - } while (likely(rw->lock < 0)); + } while (unlikely(rw->lock < 0)); HMT_medium(); } } @@ -251,7 +251,7 @@ static void __inline__ _raw_write_lock(rwlock_t *rw) HMT_low(); if (SHARED_PROCESSOR) __rw_yield(rw); - } while (likely(rw->lock != 0)); + } while (unlikely(rw->lock != 0)); HMT_medium(); } } diff --git a/include/asm-ppc64/xics.h b/include/asm-ppc64/xics.h index 0027da4364a..fdec5e7a7af 100644 --- a/include/asm-ppc64/xics.h +++ b/include/asm-ppc64/xics.h @@ -30,7 +30,4 @@ struct xics_ipi_struct { extern struct xics_ipi_struct xics_ipi_message[NR_CPUS] __cacheline_aligned; -extern unsigned int default_distrib_server; -extern unsigned int interrupt_server_size; - #endif /* _PPC64_KERNEL_XICS_H */ diff --git a/include/asm-s390/bug.h b/include/asm-s390/bug.h index 2b8d6d4dffc..a2e7430aafa 100644 --- a/include/asm-s390/bug.h +++ b/include/asm-s390/bug.h @@ -3,12 +3,15 @@ #include <linux/kernel.h> +#ifdef CONFIG_BUG #define BUG() do { \ printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \ __asm__ __volatile__(".long 0"); \ } while (0) #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-s390/cmb.h b/include/asm-s390/cmb.h index 1bfe2bd630b..dae1dd4fb93 100644 --- a/include/asm-s390/cmb.h +++ b/include/asm-s390/cmb.h @@ -52,7 +52,7 @@ struct cmbdata { #define BIODASDREADALLCMB _IOWR(DASD_IOCTL_LETTER,33,struct cmbdata) #ifdef __KERNEL__ - +struct ccw_device; /** * enable_cmf() - switch on the channel measurement for a specific device * @cdev: The ccw device to be enabled diff --git a/include/asm-s390/debug.h b/include/asm-s390/debug.h index 28ef2354b1b..6bbcdea42a8 100644 --- a/include/asm-s390/debug.h +++ b/include/asm-s390/debug.h @@ -43,7 +43,7 @@ struct __debug_entry{ #define DEBUG_OFF_LEVEL -1 /* level where debug is switched off */ #define DEBUG_FLUSH_ALL -1 /* parameter to flush all areas */ #define DEBUG_MAX_VIEWS 10 /* max number of views in proc fs */ -#define DEBUG_MAX_PROCF_LEN 16 /* max length for a proc file name */ +#define DEBUG_MAX_PROCF_LEN 64 /* max length for a proc file name */ #define DEBUG_DEFAULT_LEVEL 3 /* initial debug level */ #define DEBUG_DIR_ROOT "s390dbf" /* name of debug root directory in proc fs */ diff --git a/include/asm-s390/page.h b/include/asm-s390/page.h index 614e2a93c70..2be287b9df8 100644 --- a/include/asm-s390/page.h +++ b/include/asm-s390/page.h @@ -16,6 +16,8 @@ #define PAGE_SHIFT 12 #define PAGE_SIZE (1UL << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) +#define PAGE_DEFAULT_ACC 0 +#define PAGE_DEFAULT_KEY (PAGE_DEFAULT_ACC << 4) #ifdef __KERNEL__ #ifndef __ASSEMBLY__ diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h index 88c272ca48b..fb46e9090b5 100644 --- a/include/asm-s390/processor.h +++ b/include/asm-s390/processor.h @@ -245,7 +245,7 @@ static inline void enabled_wait(void) psw_t wait_psw; wait_psw.mask = PSW_BASE_BITS | PSW_MASK_IO | PSW_MASK_EXT | - PSW_MASK_MCHECK | PSW_MASK_WAIT; + PSW_MASK_MCHECK | PSW_MASK_WAIT | PSW_DEFAULT_KEY; #ifndef __s390x__ asm volatile ( " basr %0,0\n" diff --git a/include/asm-s390/ptrace.h b/include/asm-s390/ptrace.h index 1dc80666e97..4eff8f2e3bf 100644 --- a/include/asm-s390/ptrace.h +++ b/include/asm-s390/ptrace.h @@ -185,6 +185,7 @@ #include <linux/stddef.h> #include <linux/types.h> #include <asm/setup.h> +#include <asm/page.h> typedef union { @@ -235,6 +236,7 @@ typedef struct #define PSW_ADDR_INSN 0x7FFFFFFFUL #define PSW_BASE_BITS 0x00080000UL +#define PSW_DEFAULT_KEY (((unsigned long) PAGE_DEFAULT_ACC) << 20) #define PSW_ASC_PRIMARY 0x00000000UL #define PSW_ASC_ACCREG 0x00004000UL @@ -260,6 +262,7 @@ typedef struct #define PSW_BASE_BITS 0x0000000180000000UL #define PSW_BASE32_BITS 0x0000000080000000UL +#define PSW_DEFAULT_KEY (((unsigned long) PAGE_DEFAULT_ACC) << 52) #define PSW_ASC_PRIMARY 0x0000000000000000UL #define PSW_ASC_ACCREG 0x0000400000000000UL @@ -268,14 +271,15 @@ typedef struct #define PSW_USER32_BITS (PSW_BASE32_BITS | PSW_MASK_DAT | PSW_ASC_HOME | \ PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK | \ - PSW_MASK_PSTATE) + PSW_MASK_PSTATE | PSW_DEFAULT_KEY) #endif /* __s390x__ */ -#define PSW_KERNEL_BITS (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_PRIMARY) +#define PSW_KERNEL_BITS (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_PRIMARY | \ + PSW_DEFAULT_KEY) #define PSW_USER_BITS (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_HOME | \ PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK | \ - PSW_MASK_PSTATE) + PSW_MASK_PSTATE | PSW_DEFAULT_KEY) /* This macro merges a NEW PSW mask specified by the user into the currently active PSW mask CURRENT, modifying only those @@ -470,6 +474,12 @@ struct user_regs_struct extern void show_regs(struct pt_regs * regs); #endif +static inline void +psw_set_key(unsigned int key) +{ + asm volatile ( "spka 0(%0)" : : "d" (key) ); +} + #endif /* __ASSEMBLY__ */ #endif /* _S390_PTRACE_H */ diff --git a/include/asm-s390/siginfo.h b/include/asm-s390/siginfo.h index 72303537b73..e0ff1ab054b 100644 --- a/include/asm-s390/siginfo.h +++ b/include/asm-s390/siginfo.h @@ -13,12 +13,6 @@ #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) #endif -#ifdef CONFIG_ARCH_S390X -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) -#else -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 3) -#endif - #include <asm-generic/siginfo.h> #endif diff --git a/include/asm-s390/signal.h b/include/asm-s390/signal.h index f273cdcd1cf..3d6e11c6c1f 100644 --- a/include/asm-s390/signal.h +++ b/include/asm-s390/signal.h @@ -117,30 +117,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-sh/bug.h b/include/asm-sh/bug.h index 70172217140..70508a360cd 100644 --- a/include/asm-sh/bug.h +++ b/include/asm-sh/bug.h @@ -3,6 +3,7 @@ #include <linux/config.h> +#ifdef CONFIG_BUG /* * Tell the user there is some problem. */ @@ -12,6 +13,8 @@ } while (0) #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-sh/checksum.h b/include/asm-sh/checksum.h index 5113c7f8a73..5ebd0f24299 100644 --- a/include/asm-sh/checksum.h +++ b/include/asm-sh/checksum.h @@ -42,7 +42,7 @@ asmlinkage unsigned int csum_partial_copy_generic(const unsigned char *src, unsi * passed in an incorrect kernel address to one of these functions. * * If you use these functions directly please don't forget the - * verify_area(). + * access_ok(). */ static __inline__ unsigned int csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, diff --git a/include/asm-sh/floppy.h b/include/asm-sh/floppy.h index f030ca08052..38d7a294247 100644 --- a/include/asm-sh/floppy.h +++ b/include/asm-sh/floppy.h @@ -227,7 +227,7 @@ static int hard_dma_setup(char *addr, unsigned long size, int mode, int io) return 0; } -struct fd_routine_l { +static struct fd_routine_l { int (*_request_dma)(unsigned int dmanr, const char * device_id); void (*_free_dma)(unsigned int dmanr); int (*_get_dma_residue)(unsigned int dummy); diff --git a/include/asm-sh/signal.h b/include/asm-sh/signal.h index 0a7ff717c24..d6e8eb0e65c 100644 --- a/include/asm-sh/signal.h +++ b/include/asm-sh/signal.h @@ -108,30 +108,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-sh64/bug.h b/include/asm-sh64/bug.h index 3acd54d5956..5d659ec28e1 100644 --- a/include/asm-sh64/bug.h +++ b/include/asm-sh64/bug.h @@ -17,10 +17,6 @@ BUG(); \ } while(0) -#define PAGE_BUG(page) do { \ - BUG(); \ -} while (0) - #define WARN_ON(condition) do { \ if (unlikely((condition)!=0)) { \ printk("Badness in %s at %s:%d\n", __FUNCTION__, __FILE__, __LINE__); \ diff --git a/include/asm-sh64/checksum.h b/include/asm-sh64/checksum.h index aa3911a9949..fd034e9ae6e 100644 --- a/include/asm-sh64/checksum.h +++ b/include/asm-sh64/checksum.h @@ -34,7 +34,7 @@ asmlinkage unsigned int csum_partial(const unsigned char *buff, int len, * passed in an incorrect kernel address to one of these functions. * * If you use these functions directly please don't forget the - * verify_area(). + * access_ok(). */ diff --git a/include/asm-sh64/signal.h b/include/asm-sh64/signal.h index 77957e9b92d..2400dc688a6 100644 --- a/include/asm-sh64/signal.h +++ b/include/asm-sh64/signal.h @@ -107,30 +107,7 @@ typedef struct { #define MINSIGSTKSZ 2048 #define SIGSTKSZ THREAD_SIZE -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct old_sigaction { diff --git a/include/asm-sparc/bug.h b/include/asm-sparc/bug.h index 0d30a67d87a..04151208189 100644 --- a/include/asm-sparc/bug.h +++ b/include/asm-sparc/bug.h @@ -1,6 +1,7 @@ #ifndef _SPARC_BUG_H #define _SPARC_BUG_H +#ifdef CONFIG_BUG /* Only use the inline asm until a gcc release that can handle __builtin_trap * -rob 2003-06-25 * @@ -26,6 +27,8 @@ extern void do_BUG(const char *file, int line); #endif #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-sparc/errno.h b/include/asm-sparc/errno.h index 8c01c5f3b06..ed41c8bac1f 100644 --- a/include/asm-sparc/errno.h +++ b/include/asm-sparc/errno.h @@ -107,4 +107,8 @@ #define EKEYREVOKED 130 /* Key has been revoked */ #define EKEYREJECTED 131 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 132 /* Owner died */ +#define ENOTRECOVERABLE 133 /* State not recoverable */ + #endif diff --git a/include/asm-sparc/floppy.h b/include/asm-sparc/floppy.h index 780ee7ff9dc..caf92611650 100644 --- a/include/asm-sparc/floppy.h +++ b/include/asm-sparc/floppy.h @@ -227,7 +227,7 @@ static __inline__ void sun_fd_disable_dma(void) doing_pdma = 0; if (pdma_base) { mmu_unlockarea(pdma_base, pdma_areasize); - pdma_base = 0; + pdma_base = NULL; } } diff --git a/include/asm-sparc/mxcc.h b/include/asm-sparc/mxcc.h index efe4e843122..60ef9d6fe7b 100644 --- a/include/asm-sparc/mxcc.h +++ b/include/asm-sparc/mxcc.h @@ -115,8 +115,8 @@ extern __inline__ unsigned long mxcc_get_creg(void) { unsigned long mxcc_control; - __asm__ __volatile__("set -1, %%g2\n\t" - "set -1, %%g3\n\t" + __asm__ __volatile__("set 0xffffffff, %%g2\n\t" + "set 0xffffffff, %%g3\n\t" "stda %%g2, [%1] %2\n\t" "lda [%3] %2, %0\n\t" : "=r" (mxcc_control) : diff --git a/include/asm-sparc/signal.h b/include/asm-sparc/signal.h index d8211cb6e6b..aa9960ad0ca 100644 --- a/include/asm-sparc/signal.h +++ b/include/asm-sparc/signal.h @@ -143,7 +143,6 @@ struct sigstack { #define SA_ONESHOT _SV_RESET #define SA_INTERRUPT 0x10u #define SA_NOMASK 0x20u -#define SA_SHIRQ 0x40u #define SA_NOCLDWAIT 0x100u #define SA_SIGINFO 0x200u @@ -162,11 +161,6 @@ struct sigstack { #ifdef __KERNEL__ /* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * * DJHR * SA_STATIC_ALLOC is used for the SPARC system to indicate that this * interrupt handler's irq structure should be statically allocated @@ -177,21 +171,10 @@ struct sigstack { * statically allocated data.. which is NOT GOOD. * */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART #define SA_STATIC_ALLOC 0x80 #endif -/* Type of a signal handler. */ -#ifdef __KERNEL__ -typedef void (*__sighandler_t)(int, int, struct sigcontext *, char *); -#else -typedef void (*__sighandler_t)(int); -#endif - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> #ifdef __KERNEL__ struct __new_sigaction { diff --git a/include/asm-sparc/uaccess.h b/include/asm-sparc/uaccess.h index 3f47889883b..f461144067e 100644 --- a/include/asm-sparc/uaccess.h +++ b/include/asm-sparc/uaccess.h @@ -18,7 +18,7 @@ #ifndef __ASSEMBLY__ -/* Sparc is not segmented, however we need to be able to fool verify_area() +/* Sparc is not segmented, however we need to be able to fool access_ok() * when doing system calls from kernel mode legitimately. * * "For historical reasons, these macros are grossly misnamed." -Linus diff --git a/include/asm-sparc64/bug.h b/include/asm-sparc64/bug.h index 25c5b1dfe37..516bb27f3fc 100644 --- a/include/asm-sparc64/bug.h +++ b/include/asm-sparc64/bug.h @@ -1,6 +1,7 @@ #ifndef _SPARC64_BUG_H #define _SPARC64_BUG_H +#ifdef CONFIG_BUG #include <linux/compiler.h> #ifdef CONFIG_DEBUG_BUGVERBOSE @@ -14,6 +15,8 @@ extern void do_BUG(const char *file, int line); #endif #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif diff --git a/include/asm-sparc64/errno.h b/include/asm-sparc64/errno.h index cc98a73b55a..ea3509ee1b0 100644 --- a/include/asm-sparc64/errno.h +++ b/include/asm-sparc64/errno.h @@ -107,4 +107,8 @@ #define EKEYREVOKED 130 /* Key has been revoked */ #define EKEYREJECTED 131 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 132 /* Owner died */ +#define ENOTRECOVERABLE 133 /* State not recoverable */ + #endif /* !(_SPARC64_ERRNO_H) */ diff --git a/include/asm-sparc64/mostek.h b/include/asm-sparc64/mostek.h index ccf2f5f82d7..09b5aba6678 100644 --- a/include/asm-sparc64/mostek.h +++ b/include/asm-sparc64/mostek.h @@ -38,7 +38,7 @@ * * We now deal with physical addresses for I/O to the chip. -DaveM */ -static __inline__ u8 mostek_read(unsigned long addr) +static __inline__ u8 mostek_read(void __iomem *addr) { u8 ret; @@ -48,7 +48,7 @@ static __inline__ u8 mostek_read(unsigned long addr) return ret; } -static __inline__ void mostek_write(unsigned long addr, u8 val) +static __inline__ void mostek_write(void __iomem *addr, u8 val) { __asm__ __volatile__("stba %0, [%1] %2" : /* no outputs */ @@ -67,7 +67,7 @@ static __inline__ void mostek_write(unsigned long addr, u8 val) #define MOSTEK_YEAR 0x07ffUL extern spinlock_t mostek_lock; -extern unsigned long mstk48t02_regs; +extern void __iomem *mstk48t02_regs; /* Control register values. */ #define MSTK_CREG_WRITE 0x80 /* Must set this before placing values. */ diff --git a/include/asm-sparc64/parport.h b/include/asm-sparc64/parport.h index ab88349ddad..b7e635544ce 100644 --- a/include/asm-sparc64/parport.h +++ b/include/asm-sparc64/parport.h @@ -13,6 +13,12 @@ #define PARPORT_PC_MAX_PORTS PARPORT_MAX +/* + * While sparc64 doesn't have an ISA DMA API, we provide something that looks + * close enough to make parport_pc happy + */ +#define HAS_DMA + static struct sparc_ebus_info { struct ebus_dma_info info; unsigned int addr; diff --git a/include/asm-sparc64/pgalloc.h b/include/asm-sparc64/pgalloc.h index 2c28e1f605b..b9b1914aae6 100644 --- a/include/asm-sparc64/pgalloc.h +++ b/include/asm-sparc64/pgalloc.h @@ -122,17 +122,12 @@ static __inline__ void free_pmd_slow(pmd_t *pmd) #define pmd_populate(MM,PMD,PTE_PAGE) \ pmd_populate_kernel(MM,PMD,page_address(PTE_PAGE)) -extern pte_t *__pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address); - -static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) -{ - return __pte_alloc_one_kernel(mm, address); -} +extern pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address); static inline struct page * pte_alloc_one(struct mm_struct *mm, unsigned long addr) { - pte_t *pte = __pte_alloc_one_kernel(mm, addr); + pte_t *pte = pte_alloc_one_kernel(mm, addr); if (pte) return virt_to_page(pte); diff --git a/include/asm-sparc64/pgtable.h b/include/asm-sparc64/pgtable.h index af9bf175a22..ae2cd5b09a7 100644 --- a/include/asm-sparc64/pgtable.h +++ b/include/asm-sparc64/pgtable.h @@ -416,6 +416,11 @@ extern int io_remap_pfn_range(struct vm_area_struct *vma, unsigned long from, unsigned long pfn, unsigned long size, pgprot_t prot); +/* Clear virtual and physical cachability, set side-effect bit. */ +#define pgprot_noncached(prot) \ + (__pgprot((pgprot_val(prot) & ~(_PAGE_CP | _PAGE_CV)) | \ + _PAGE_E)) + /* * For sparc32&64, the pfn in io_remap_pfn_range() carries <iospace> in * its high 4 bits. These macros/functions put it there or get it from there. diff --git a/include/asm-sparc64/siginfo.h b/include/asm-sparc64/siginfo.h index 7160449e7ca..df17e47abc1 100644 --- a/include/asm-sparc64/siginfo.h +++ b/include/asm-sparc64/siginfo.h @@ -3,8 +3,6 @@ #define SI_PAD_SIZE32 ((SI_MAX_SIZE/sizeof(int)) - 3) -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) - #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) #define __ARCH_SI_TRAPNO #define __ARCH_SI_BAND_T int diff --git a/include/asm-sparc64/signal.h b/include/asm-sparc64/signal.h index 6428e366c38..becdf1bc592 100644 --- a/include/asm-sparc64/signal.h +++ b/include/asm-sparc64/signal.h @@ -145,7 +145,6 @@ struct sigstack { #define SA_ONESHOT _SV_RESET #define SA_INTERRUPT 0x10u #define SA_NOMASK 0x20u -#define SA_SHIRQ 0x40u #define SA_NOCLDWAIT 0x100u #define SA_SIGINFO 0x200u @@ -165,11 +164,6 @@ struct sigstack { #ifdef __KERNEL__ /* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * * DJHR * SA_STATIC_ALLOC is used for the SPARC system to indicate that this * interrupt handler's irq structure should be statically allocated @@ -180,26 +174,10 @@ struct sigstack { * statically allocated data.. which is NOT GOOD. * */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART #define SA_STATIC_ALLOC 0x80 #endif -/* Type of a signal handler. */ -#ifdef __KERNEL__ -typedef void __signalfn_t(int); -typedef __signalfn_t __user *__sighandler_t; - -typedef void __restorefn_t(void); -typedef __restorefn_t __user *__sigrestore_t; -#else -typedef void (*__sighandler_t)(int); -typedef void (*__sigrestore_t)(void); -#endif - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ +#include <asm-generic/signal.h> struct __new_sigaction { __sighandler_t sa_handler; diff --git a/include/asm-sparc64/spinlock.h b/include/asm-sparc64/spinlock.h index d1f91a4f24a..db7581bdb53 100644 --- a/include/asm-sparc64/spinlock.h +++ b/include/asm-sparc64/spinlock.h @@ -44,7 +44,7 @@ typedef struct { #define spin_unlock_wait(lp) \ do { membar("#LoadLoad"); \ -} while(lp->lock) +} while((lp)->lock) static inline void _raw_spin_lock(spinlock_t *lock) { @@ -149,7 +149,7 @@ typedef struct { unsigned int break_lock; #endif } rwlock_t; -#define RW_LOCK_UNLOCKED {0,} +#define RW_LOCK_UNLOCKED (rwlock_t) {0,} #define rwlock_init(lp) do { *(lp) = RW_LOCK_UNLOCKED; } while(0) static void inline __read_lock(rwlock_t *lock) diff --git a/include/asm-um/arch-signal-i386.h b/include/asm-um/arch-signal-i386.h index 99a9de4728d..e69de29bb2d 100644 --- a/include/asm-um/arch-signal-i386.h +++ b/include/asm-um/arch-signal-i386.h @@ -1,24 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#ifndef __UM_ARCH_SIGNAL_I386_H -#define __UM_ARCH_SIGNAL_I386_H - -struct arch_signal_context { - unsigned long extrasigs[_NSIG_WORDS]; -}; - -#endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/include/asm-um/archparam-i386.h b/include/asm-um/archparam-i386.h index 6f78de5b621..49e89b8d7e5 100644 --- a/include/asm-um/archparam-i386.h +++ b/include/asm-um/archparam-i386.h @@ -6,143 +6,6 @@ #ifndef __UM_ARCHPARAM_I386_H #define __UM_ARCHPARAM_I386_H -/********* Bits for asm-um/elf.h ************/ - -#include <asm/user.h> - -extern char * elf_aux_platform; -#define ELF_PLATFORM (elf_aux_platform) - -#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) - -typedef struct user_i387_struct elf_fpregset_t; -typedef unsigned long elf_greg_t; - -#define ELF_NGREG (sizeof (struct user_regs_struct) / sizeof(elf_greg_t)) -typedef elf_greg_t elf_gregset_t[ELF_NGREG]; - -#define ELF_DATA ELFDATA2LSB -#define ELF_ARCH EM_386 - -#define ELF_PLAT_INIT(regs, load_addr) do { \ - PT_REGS_EBX(regs) = 0; \ - PT_REGS_ECX(regs) = 0; \ - PT_REGS_EDX(regs) = 0; \ - PT_REGS_ESI(regs) = 0; \ - PT_REGS_EDI(regs) = 0; \ - PT_REGS_EBP(regs) = 0; \ - PT_REGS_EAX(regs) = 0; \ -} while(0) - -/* Shamelessly stolen from include/asm-i386/elf.h */ - -#define ELF_CORE_COPY_REGS(pr_reg, regs) do { \ - pr_reg[0] = PT_REGS_EBX(regs); \ - pr_reg[1] = PT_REGS_ECX(regs); \ - pr_reg[2] = PT_REGS_EDX(regs); \ - pr_reg[3] = PT_REGS_ESI(regs); \ - pr_reg[4] = PT_REGS_EDI(regs); \ - pr_reg[5] = PT_REGS_EBP(regs); \ - pr_reg[6] = PT_REGS_EAX(regs); \ - pr_reg[7] = PT_REGS_DS(regs); \ - pr_reg[8] = PT_REGS_ES(regs); \ - /* fake once used fs and gs selectors? */ \ - pr_reg[9] = PT_REGS_DS(regs); \ - pr_reg[10] = PT_REGS_DS(regs); \ - pr_reg[11] = PT_REGS_SYSCALL_NR(regs); \ - pr_reg[12] = PT_REGS_IP(regs); \ - pr_reg[13] = PT_REGS_CS(regs); \ - pr_reg[14] = PT_REGS_EFLAGS(regs); \ - pr_reg[15] = PT_REGS_SP(regs); \ - pr_reg[16] = PT_REGS_SS(regs); \ -} while(0); - - -extern unsigned long vsyscall_ehdr; -extern unsigned long vsyscall_end; -extern unsigned long __kernel_vsyscall; - -#define VSYSCALL_BASE vsyscall_ehdr -#define VSYSCALL_END vsyscall_end - -/* - * This is the range that is readable by user mode, and things - * acting like user mode such as get_user_pages. - */ -#define FIXADDR_USER_START VSYSCALL_BASE -#define FIXADDR_USER_END VSYSCALL_END - -/* - * Architecture-neutral AT_ values in 0-17, leave some room - * for more of them, start the x86-specific ones at 32. - */ -#define AT_SYSINFO 32 -#define AT_SYSINFO_EHDR 33 - -#define ARCH_DLINFO \ -do { \ - if ( vsyscall_ehdr ) { \ - NEW_AUX_ENT(AT_SYSINFO, __kernel_vsyscall); \ - NEW_AUX_ENT(AT_SYSINFO_EHDR, vsyscall_ehdr); \ - } \ -} while (0) - -/* - * These macros parameterize elf_core_dump in fs/binfmt_elf.c to write out - * extra segments containing the vsyscall DSO contents. Dumping its - * contents makes post-mortem fully interpretable later without matching up - * the same kernel and hardware config to see what PC values meant. - * Dumping its extra ELF program headers includes all the other information - * a debugger needs to easily find how the vsyscall DSO was being used. - */ -#define ELF_CORE_EXTRA_PHDRS \ - (vsyscall_ehdr ? (((struct elfhdr *)vsyscall_ehdr)->e_phnum) : 0 ) - -#define ELF_CORE_WRITE_EXTRA_PHDRS \ -if ( vsyscall_ehdr ) { \ - const struct elfhdr *const ehdrp = (struct elfhdr *)vsyscall_ehdr; \ - const struct elf_phdr *const phdrp = \ - (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); \ - int i; \ - Elf32_Off ofs = 0; \ - for (i = 0; i < ehdrp->e_phnum; ++i) { \ - struct elf_phdr phdr = phdrp[i]; \ - if (phdr.p_type == PT_LOAD) { \ - ofs = phdr.p_offset = offset; \ - offset += phdr.p_filesz; \ - } \ - else \ - phdr.p_offset += ofs; \ - phdr.p_paddr = 0; /* match other core phdrs */ \ - DUMP_WRITE(&phdr, sizeof(phdr)); \ - } \ -} -#define ELF_CORE_WRITE_EXTRA_DATA \ -if ( vsyscall_ehdr ) { \ - const struct elfhdr *const ehdrp = (struct elfhdr *)vsyscall_ehdr; \ - const struct elf_phdr *const phdrp = \ - (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); \ - int i; \ - for (i = 0; i < ehdrp->e_phnum; ++i) { \ - if (phdrp[i].p_type == PT_LOAD) \ - DUMP_WRITE((void *) phdrp[i].p_vaddr, \ - phdrp[i].p_filesz); \ - } \ -} - -#define R_386_NONE 0 -#define R_386_32 1 -#define R_386_PC32 2 -#define R_386_GOT32 3 -#define R_386_PLT32 4 -#define R_386_COPY 5 -#define R_386_GLOB_DAT 6 -#define R_386_JMP_SLOT 7 -#define R_386_RELATIVE 8 -#define R_386_GOTOFF 9 -#define R_386_GOTPC 10 -#define R_386_NUM 11 - /********* Nothing for asm-um/hardirq.h **********/ /********* Nothing for asm-um/hw_irq.h **********/ diff --git a/include/asm-um/archparam-ppc.h b/include/asm-um/archparam-ppc.h index 0ebced92a76..172cd6ffacc 100644 --- a/include/asm-um/archparam-ppc.h +++ b/include/asm-um/archparam-ppc.h @@ -1,26 +1,6 @@ #ifndef __UM_ARCHPARAM_PPC_H #define __UM_ARCHPARAM_PPC_H -/********* Bits for asm-um/elf.h ************/ - -#define ELF_PLATFORM (0) - -#define ELF_ET_DYN_BASE (0x08000000) - -/* the following stolen from asm-ppc/elf.h */ -#define ELF_NGREG 48 /* includes nip, msr, lr, etc. */ -#define ELF_NFPREG 33 /* includes fpscr */ -/* General registers */ -typedef unsigned long elf_greg_t; -typedef elf_greg_t elf_gregset_t[ELF_NGREG]; - -/* Floating point registers */ -typedef double elf_fpreg_t; -typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG]; - -#define ELF_DATA ELFDATA2MSB -#define ELF_ARCH EM_PPC - /********* Bits for asm-um/hw_irq.h **********/ struct hw_interrupt_type; diff --git a/include/asm-um/archparam-x86_64.h b/include/asm-um/archparam-x86_64.h index 96321c4892f..270ed9586b6 100644 --- a/include/asm-um/archparam-x86_64.h +++ b/include/asm-um/archparam-x86_64.h @@ -7,42 +7,6 @@ #ifndef __UM_ARCHPARAM_X86_64_H #define __UM_ARCHPARAM_X86_64_H -#include <asm/user.h> - -#define ELF_PLATFORM "x86_64" - -#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) - -typedef unsigned long elf_greg_t; -typedef struct { } elf_fpregset_t; - -#define ELF_NGREG (sizeof (struct user_regs_struct) / sizeof(elf_greg_t)) -typedef elf_greg_t elf_gregset_t[ELF_NGREG]; - -#define ELF_DATA ELFDATA2LSB -#define ELF_ARCH EM_X86_64 - -#define ELF_PLAT_INIT(regs, load_addr) do { \ - PT_REGS_RBX(regs) = 0; \ - PT_REGS_RCX(regs) = 0; \ - PT_REGS_RDX(regs) = 0; \ - PT_REGS_RSI(regs) = 0; \ - PT_REGS_RDI(regs) = 0; \ - PT_REGS_RBP(regs) = 0; \ - PT_REGS_RAX(regs) = 0; \ - PT_REGS_R8(regs) = 0; \ - PT_REGS_R9(regs) = 0; \ - PT_REGS_R10(regs) = 0; \ - PT_REGS_R11(regs) = 0; \ - PT_REGS_R12(regs) = 0; \ - PT_REGS_R13(regs) = 0; \ - PT_REGS_R14(regs) = 0; \ - PT_REGS_R15(regs) = 0; \ -} while (0) - -#ifdef TIF_IA32 /* XXX */ - clear_thread_flag(TIF_IA32); -#endif /* No user-accessible fixmap addresses, i.e. vsyscall */ #define FIXADDR_USER_START 0 diff --git a/include/asm-um/common.lds.S b/include/asm-um/common.lds.S index a3d6aab0e74..1010153faaf 100644 --- a/include/asm-um/common.lds.S +++ b/include/asm-um/common.lds.S @@ -8,11 +8,6 @@ _sdata = .; PROVIDE (sdata = .); - . = ALIGN(16); /* Exception table */ - __start___ex_table = .; - __ex_table : { *(__ex_table) } - __stop___ex_table = .; - RODATA .unprotected : { *(.unprotected) } @@ -20,6 +15,10 @@ PROVIDE (_unprotected_end = .); . = ALIGN(4096); + __start___ex_table = .; + __ex_table : { *(__ex_table) } + __stop___ex_table = .; + __uml_setup_start = .; .uml.setup.init : { *(.uml.setup.init) } __uml_setup_end = .; diff --git a/include/asm-um/delay.h b/include/asm-um/delay.h index 40695576ca6..0985bda6675 100644 --- a/include/asm-um/delay.h +++ b/include/asm-um/delay.h @@ -4,4 +4,6 @@ #include "asm/arch/delay.h" #include "asm/archparam.h" +#define MILLION 1000000 + #endif diff --git a/include/asm-um/elf-i386.h b/include/asm-um/elf-i386.h new file mode 100644 index 00000000000..b72e23519e0 --- /dev/null +++ b/include/asm-um/elf-i386.h @@ -0,0 +1,169 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ +#ifndef __UM_ELF_I386_H +#define __UM_ELF_I386_H + +#include "user.h" + +#define R_386_NONE 0 +#define R_386_32 1 +#define R_386_PC32 2 +#define R_386_GOT32 3 +#define R_386_PLT32 4 +#define R_386_COPY 5 +#define R_386_GLOB_DAT 6 +#define R_386_JMP_SLOT 7 +#define R_386_RELATIVE 8 +#define R_386_GOTOFF 9 +#define R_386_GOTPC 10 +#define R_386_NUM 11 + +typedef unsigned long elf_greg_t; + +#define ELF_NGREG (sizeof (struct user_regs_struct) / sizeof(elf_greg_t)) +typedef elf_greg_t elf_gregset_t[ELF_NGREG]; + +typedef struct user_i387_struct elf_fpregset_t; + +/* + * This is used to ensure we don't load something for the wrong architecture. + */ +#define elf_check_arch(x) \ + (((x)->e_machine == EM_386) || ((x)->e_machine == EM_486)) + +#define ELF_CLASS ELFCLASS32 +#define ELF_DATA ELFDATA2LSB +#define ELF_ARCH EM_386 + +#define ELF_PLAT_INIT(regs, load_addr) do { \ + PT_REGS_EBX(regs) = 0; \ + PT_REGS_ECX(regs) = 0; \ + PT_REGS_EDX(regs) = 0; \ + PT_REGS_ESI(regs) = 0; \ + PT_REGS_EDI(regs) = 0; \ + PT_REGS_EBP(regs) = 0; \ + PT_REGS_EAX(regs) = 0; \ +} while(0) + +#define USE_ELF_CORE_DUMP +#define ELF_EXEC_PAGESIZE 4096 + +#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) + +/* Shamelessly stolen from include/asm-i386/elf.h */ + +#define ELF_CORE_COPY_REGS(pr_reg, regs) do { \ + pr_reg[0] = PT_REGS_EBX(regs); \ + pr_reg[1] = PT_REGS_ECX(regs); \ + pr_reg[2] = PT_REGS_EDX(regs); \ + pr_reg[3] = PT_REGS_ESI(regs); \ + pr_reg[4] = PT_REGS_EDI(regs); \ + pr_reg[5] = PT_REGS_EBP(regs); \ + pr_reg[6] = PT_REGS_EAX(regs); \ + pr_reg[7] = PT_REGS_DS(regs); \ + pr_reg[8] = PT_REGS_ES(regs); \ + /* fake once used fs and gs selectors? */ \ + pr_reg[9] = PT_REGS_DS(regs); \ + pr_reg[10] = PT_REGS_DS(regs); \ + pr_reg[11] = PT_REGS_SYSCALL_NR(regs); \ + pr_reg[12] = PT_REGS_IP(regs); \ + pr_reg[13] = PT_REGS_CS(regs); \ + pr_reg[14] = PT_REGS_EFLAGS(regs); \ + pr_reg[15] = PT_REGS_SP(regs); \ + pr_reg[16] = PT_REGS_SS(regs); \ +} while(0); + +extern long elf_aux_hwcap; +#define ELF_HWCAP (elf_aux_hwcap) + +extern char * elf_aux_platform; +#define ELF_PLATFORM (elf_aux_platform) + +#define SET_PERSONALITY(ex, ibcs2) do ; while(0) + +extern unsigned long vsyscall_ehdr; +extern unsigned long vsyscall_end; +extern unsigned long __kernel_vsyscall; + +#define VSYSCALL_BASE vsyscall_ehdr +#define VSYSCALL_END vsyscall_end + +/* + * This is the range that is readable by user mode, and things + * acting like user mode such as get_user_pages. + */ +#define FIXADDR_USER_START VSYSCALL_BASE +#define FIXADDR_USER_END VSYSCALL_END + +/* + * Architecture-neutral AT_ values in 0-17, leave some room + * for more of them, start the x86-specific ones at 32. + */ +#define AT_SYSINFO 32 +#define AT_SYSINFO_EHDR 33 + +#define ARCH_DLINFO \ +do { \ + if ( vsyscall_ehdr ) { \ + NEW_AUX_ENT(AT_SYSINFO, __kernel_vsyscall); \ + NEW_AUX_ENT(AT_SYSINFO_EHDR, vsyscall_ehdr); \ + } \ +} while (0) + +/* + * These macros parameterize elf_core_dump in fs/binfmt_elf.c to write out + * extra segments containing the vsyscall DSO contents. Dumping its + * contents makes post-mortem fully interpretable later without matching up + * the same kernel and hardware config to see what PC values meant. + * Dumping its extra ELF program headers includes all the other information + * a debugger needs to easily find how the vsyscall DSO was being used. + */ +#define ELF_CORE_EXTRA_PHDRS \ + (vsyscall_ehdr ? (((struct elfhdr *)vsyscall_ehdr)->e_phnum) : 0 ) + +#define ELF_CORE_WRITE_EXTRA_PHDRS \ +if ( vsyscall_ehdr ) { \ + const struct elfhdr *const ehdrp = (struct elfhdr *)vsyscall_ehdr; \ + const struct elf_phdr *const phdrp = \ + (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); \ + int i; \ + Elf32_Off ofs = 0; \ + for (i = 0; i < ehdrp->e_phnum; ++i) { \ + struct elf_phdr phdr = phdrp[i]; \ + if (phdr.p_type == PT_LOAD) { \ + ofs = phdr.p_offset = offset; \ + offset += phdr.p_filesz; \ + } \ + else \ + phdr.p_offset += ofs; \ + phdr.p_paddr = 0; /* match other core phdrs */ \ + DUMP_WRITE(&phdr, sizeof(phdr)); \ + } \ +} +#define ELF_CORE_WRITE_EXTRA_DATA \ +if ( vsyscall_ehdr ) { \ + const struct elfhdr *const ehdrp = (struct elfhdr *)vsyscall_ehdr; \ + const struct elf_phdr *const phdrp = \ + (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); \ + int i; \ + for (i = 0; i < ehdrp->e_phnum; ++i) { \ + if (phdrp[i].p_type == PT_LOAD) \ + DUMP_WRITE((void *) phdrp[i].p_vaddr, \ + phdrp[i].p_filesz); \ + } \ +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/include/asm-um/elf-ppc.h b/include/asm-um/elf-ppc.h new file mode 100644 index 00000000000..2998cf92504 --- /dev/null +++ b/include/asm-um/elf-ppc.h @@ -0,0 +1,54 @@ +#ifndef __UM_ELF_PPC_H +#define __UM_ELF_PPC_H + +#include "linux/config.h" + +extern long elf_aux_hwcap; +#define ELF_HWCAP (elf_aux_hwcap) + +#define SET_PERSONALITY(ex, ibcs2) do ; while(0) + +#define ELF_EXEC_PAGESIZE 4096 + +#define elf_check_arch(x) (1) + +#ifdef CONFIG_64_BIT +#define ELF_CLASS ELFCLASS64 +#else +#define ELF_CLASS ELFCLASS32 +#endif + +#define USE_ELF_CORE_DUMP + +#define R_386_NONE 0 +#define R_386_32 1 +#define R_386_PC32 2 +#define R_386_GOT32 3 +#define R_386_PLT32 4 +#define R_386_COPY 5 +#define R_386_GLOB_DAT 6 +#define R_386_JMP_SLOT 7 +#define R_386_RELATIVE 8 +#define R_386_GOTOFF 9 +#define R_386_GOTPC 10 +#define R_386_NUM 11 + +#define ELF_PLATFORM (0) + +#define ELF_ET_DYN_BASE (0x08000000) + +/* the following stolen from asm-ppc/elf.h */ +#define ELF_NGREG 48 /* includes nip, msr, lr, etc. */ +#define ELF_NFPREG 33 /* includes fpscr */ +/* General registers */ +typedef unsigned long elf_greg_t; +typedef elf_greg_t elf_gregset_t[ELF_NGREG]; + +/* Floating point registers */ +typedef double elf_fpreg_t; +typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG]; + +#define ELF_DATA ELFDATA2MSB +#define ELF_ARCH EM_PPC + +#endif diff --git a/include/asm-um/elf-x86_64.h b/include/asm-um/elf-x86_64.h new file mode 100644 index 00000000000..19309d001aa --- /dev/null +++ b/include/asm-um/elf-x86_64.h @@ -0,0 +1,73 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ +#ifndef __UM_ELF_X86_64_H +#define __UM_ELF_X86_64_H + +#include <asm/user.h> + +typedef unsigned long elf_greg_t; + +#define ELF_NGREG (sizeof (struct user_regs_struct) / sizeof(elf_greg_t)) +typedef elf_greg_t elf_gregset_t[ELF_NGREG]; + +typedef struct { } elf_fpregset_t; + +/* + * This is used to ensure we don't load something for the wrong architecture. + */ +#define elf_check_arch(x) \ + ((x)->e_machine == EM_X86_64) + +#define ELF_CLASS ELFCLASS64 +#define ELF_DATA ELFDATA2LSB +#define ELF_ARCH EM_X86_64 + +#define ELF_PLAT_INIT(regs, load_addr) do { \ + PT_REGS_RBX(regs) = 0; \ + PT_REGS_RCX(regs) = 0; \ + PT_REGS_RDX(regs) = 0; \ + PT_REGS_RSI(regs) = 0; \ + PT_REGS_RDI(regs) = 0; \ + PT_REGS_RBP(regs) = 0; \ + PT_REGS_RAX(regs) = 0; \ + PT_REGS_R8(regs) = 0; \ + PT_REGS_R9(regs) = 0; \ + PT_REGS_R10(regs) = 0; \ + PT_REGS_R11(regs) = 0; \ + PT_REGS_R12(regs) = 0; \ + PT_REGS_R13(regs) = 0; \ + PT_REGS_R14(regs) = 0; \ + PT_REGS_R15(regs) = 0; \ +} while (0) + +#ifdef TIF_IA32 /* XXX */ + clear_thread_flag(TIF_IA32); \ +#endif + +#define USE_ELF_CORE_DUMP +#define ELF_EXEC_PAGESIZE 4096 + +#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) + +extern long elf_aux_hwcap; +#define ELF_HWCAP (elf_aux_hwcap) + +#define ELF_PLATFORM "x86_64" + +#define SET_PERSONALITY(ex, ibcs2) do ; while(0) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/include/asm-um/elf.h b/include/asm-um/elf.h index b3a7258f997..ebf7c63886e 100644 --- a/include/asm-um/elf.h +++ b/include/asm-um/elf.h @@ -1,25 +1,13 @@ -#ifndef __UM_ELF_H -#define __UM_ELF_H +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ +#ifndef __UM_ELF_I386_H +#define __UM_ELF_I386_H -#include "linux/config.h" -#include "asm/archparam.h" +#include "user.h" -extern long elf_aux_hwcap; -#define ELF_HWCAP (elf_aux_hwcap) - -#define SET_PERSONALITY(ex, ibcs2) do ; while(0) - -#define ELF_EXEC_PAGESIZE 4096 - -#define elf_check_arch(x) (1) - -#ifdef CONFIG_64_BIT -#define ELF_CLASS ELFCLASS64 -#else -#define ELF_CLASS ELFCLASS32 -#endif - -#define USE_ELF_CORE_DUMP +#if defined(CONFIG_UML_X86) && !defined(CONFIG_64BIT) #define R_386_NONE 0 #define R_386_32 1 @@ -34,4 +22,175 @@ extern long elf_aux_hwcap; #define R_386_GOTPC 10 #define R_386_NUM 11 +#elif defined(CONFIG_UML_X86) && defined(CONFIG_64BIT) + +/* x86-64 relocation types */ +#define R_X86_64_NONE 0 /* No reloc */ +#define R_X86_64_64 1 /* Direct 64 bit */ +#define R_X86_64_PC32 2 /* PC relative 32 bit signed */ +#define R_X86_64_GOT32 3 /* 32 bit GOT entry */ +#define R_X86_64_PLT32 4 /* 32 bit PLT address */ +#define R_X86_64_COPY 5 /* Copy symbol at runtime */ +#define R_X86_64_GLOB_DAT 6 /* Create GOT entry */ +#define R_X86_64_JUMP_SLOT 7 /* Create PLT entry */ +#define R_X86_64_RELATIVE 8 /* Adjust by program base */ +#define R_X86_64_GOTPCREL 9 /* 32 bit signed pc relative + offset to GOT */ +#define R_X86_64_32 10 /* Direct 32 bit zero extended */ +#define R_X86_64_32S 11 /* Direct 32 bit sign extended */ +#define R_X86_64_16 12 /* Direct 16 bit zero extended */ +#define R_X86_64_PC16 13 /* 16 bit sign extended pc relative */ +#define R_X86_64_8 14 /* Direct 8 bit sign extended */ +#define R_X86_64_PC8 15 /* 8 bit sign extended pc relative */ + +#define R_X86_64_NUM 16 + +#endif + +typedef unsigned long elf_greg_t; + +#define ELF_NGREG (sizeof (struct user_regs_struct) / sizeof(elf_greg_t)) +typedef elf_greg_t elf_gregset_t[ELF_NGREG]; + +typedef struct user_i387_struct elf_fpregset_t; + +/* + * This is used to ensure we don't load something for the wrong architecture. + */ +#define elf_check_arch(x) \ + (((x)->e_machine == EM_386) || ((x)->e_machine == EM_486)) + +#define ELF_CLASS ELFCLASS32 +#define ELF_DATA ELFDATA2LSB +#define ELF_ARCH EM_386 + +#define ELF_PLAT_INIT(regs, load_addr) do { \ + PT_REGS_EBX(regs) = 0; \ + PT_REGS_ECX(regs) = 0; \ + PT_REGS_EDX(regs) = 0; \ + PT_REGS_ESI(regs) = 0; \ + PT_REGS_EDI(regs) = 0; \ + PT_REGS_EBP(regs) = 0; \ + PT_REGS_EAX(regs) = 0; \ +} while(0) + +#define USE_ELF_CORE_DUMP +#define ELF_EXEC_PAGESIZE 4096 + +#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) + +/* Shamelessly stolen from include/asm-i386/elf.h */ + +#define ELF_CORE_COPY_REGS(pr_reg, regs) do { \ + pr_reg[0] = PT_REGS_EBX(regs); \ + pr_reg[1] = PT_REGS_ECX(regs); \ + pr_reg[2] = PT_REGS_EDX(regs); \ + pr_reg[3] = PT_REGS_ESI(regs); \ + pr_reg[4] = PT_REGS_EDI(regs); \ + pr_reg[5] = PT_REGS_EBP(regs); \ + pr_reg[6] = PT_REGS_EAX(regs); \ + pr_reg[7] = PT_REGS_DS(regs); \ + pr_reg[8] = PT_REGS_ES(regs); \ + /* fake once used fs and gs selectors? */ \ + pr_reg[9] = PT_REGS_DS(regs); \ + pr_reg[10] = PT_REGS_DS(regs); \ + pr_reg[11] = PT_REGS_SYSCALL_NR(regs); \ + pr_reg[12] = PT_REGS_IP(regs); \ + pr_reg[13] = PT_REGS_CS(regs); \ + pr_reg[14] = PT_REGS_EFLAGS(regs); \ + pr_reg[15] = PT_REGS_SP(regs); \ + pr_reg[16] = PT_REGS_SS(regs); \ +} while(0); + +extern long elf_aux_hwcap; +#define ELF_HWCAP (elf_aux_hwcap) + +extern char * elf_aux_platform; +#define ELF_PLATFORM (elf_aux_platform) + +#define SET_PERSONALITY(ex, ibcs2) do ; while(0) + +extern unsigned long vsyscall_ehdr; +extern unsigned long vsyscall_end; +extern unsigned long __kernel_vsyscall; + +#define VSYSCALL_BASE vsyscall_ehdr +#define VSYSCALL_END vsyscall_end + +/* + * This is the range that is readable by user mode, and things + * acting like user mode such as get_user_pages. + */ +#define FIXADDR_USER_START VSYSCALL_BASE +#define FIXADDR_USER_END VSYSCALL_END + +/* + * Architecture-neutral AT_ values in 0-17, leave some room + * for more of them, start the x86-specific ones at 32. + */ +#define AT_SYSINFO 32 +#define AT_SYSINFO_EHDR 33 + +#define ARCH_DLINFO \ +do { \ + if ( vsyscall_ehdr ) { \ + NEW_AUX_ENT(AT_SYSINFO, __kernel_vsyscall); \ + NEW_AUX_ENT(AT_SYSINFO_EHDR, vsyscall_ehdr); \ + } \ +} while (0) + +/* + * These macros parameterize elf_core_dump in fs/binfmt_elf.c to write out + * extra segments containing the vsyscall DSO contents. Dumping its + * contents makes post-mortem fully interpretable later without matching up + * the same kernel and hardware config to see what PC values meant. + * Dumping its extra ELF program headers includes all the other information + * a debugger needs to easily find how the vsyscall DSO was being used. + */ +#define ELF_CORE_EXTRA_PHDRS \ + (vsyscall_ehdr ? (((struct elfhdr *)vsyscall_ehdr)->e_phnum) : 0 ) + +#define ELF_CORE_WRITE_EXTRA_PHDRS \ +if ( vsyscall_ehdr ) { \ + const struct elfhdr *const ehdrp = (struct elfhdr *)vsyscall_ehdr; \ + const struct elf_phdr *const phdrp = \ + (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); \ + int i; \ + Elf32_Off ofs = 0; \ + for (i = 0; i < ehdrp->e_phnum; ++i) { \ + struct elf_phdr phdr = phdrp[i]; \ + if (phdr.p_type == PT_LOAD) { \ + ofs = phdr.p_offset = offset; \ + offset += phdr.p_filesz; \ + } \ + else \ + phdr.p_offset += ofs; \ + phdr.p_paddr = 0; /* match other core phdrs */ \ + DUMP_WRITE(&phdr, sizeof(phdr)); \ + } \ +} +#define ELF_CORE_WRITE_EXTRA_DATA \ +if ( vsyscall_ehdr ) { \ + const struct elfhdr *const ehdrp = (struct elfhdr *)vsyscall_ehdr; \ + const struct elf_phdr *const phdrp = \ + (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); \ + int i; \ + for (i = 0; i < ehdrp->e_phnum; ++i) { \ + if (phdrp[i].p_type == PT_LOAD) \ + DUMP_WRITE((void *) phdrp[i].p_vaddr, \ + phdrp[i].p_filesz); \ + } \ +} + #endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/include/asm-um/fixmap.h b/include/asm-um/fixmap.h index 900f3fbb9fa..ae0ca3932d5 100644 --- a/include/asm-um/fixmap.h +++ b/include/asm-um/fixmap.h @@ -4,6 +4,7 @@ #include <linux/config.h> #include <asm/kmap_types.h> #include <asm/archparam.h> +#include <asm/elf.h> /* * Here we define all the compile-time 'special' virtual diff --git a/include/asm-um/ipc.h b/include/asm-um/ipc.h index e2ddc47f3e5..a46e3d9c2a3 100644 --- a/include/asm-um/ipc.h +++ b/include/asm-um/ipc.h @@ -1,6 +1 @@ -#ifndef __UM_IPC_H -#define __UM_IPC_H - -#include "asm/arch/ipc.h" - -#endif +#include <asm-generic/ipc.h> diff --git a/include/asm-um/linkage.h b/include/asm-um/linkage.h index 27011652b01..7dfce37adc8 100644 --- a/include/asm-um/linkage.h +++ b/include/asm-um/linkage.h @@ -1,7 +1,6 @@ -#ifndef __ASM_LINKAGE_H -#define __ASM_LINKAGE_H +#ifndef __ASM_UM_LINKAGE_H +#define __ASM_UM_LINKAGE_H -#define FASTCALL(x) x __attribute__((regparm(3))) -#define fastcall __attribute__((regparm(3))) +#include "asm/arch/linkage.h" #endif diff --git a/include/asm-um/page.h b/include/asm-um/page.h index 3620a08dc9f..504ea8e486b 100644 --- a/include/asm-um/page.h +++ b/include/asm-um/page.h @@ -27,7 +27,7 @@ struct page; #define clear_user_page(page, vaddr, pg) clear_page(page) #define copy_user_page(to, from, vaddr, pg) copy_page(to, from) -#if defined(CONFIG_3_LEVEL_PGTABLES) && !defined(CONFIG_64_BIT) +#if defined(CONFIG_3_LEVEL_PGTABLES) && !defined(CONFIG_64BIT) typedef struct { unsigned long pte_low, pte_high; } pte_t; typedef struct { unsigned long long pmd; } pmd_t; @@ -45,6 +45,9 @@ typedef struct { unsigned long pgd; } pgd_t; ({ (pte).pte_high = (phys) >> 32; \ (pte).pte_low = (phys) | pgprot_val(prot); }) +#define pmd_val(x) ((x).pmd) +#define __pmd(x) ((pmd_t) { (x) } ) + typedef unsigned long long pfn_t; typedef unsigned long long phys_t; diff --git a/include/asm-um/pgtable-3level.h b/include/asm-um/pgtable-3level.h index bdbc3f97e20..65e8bfc55fc 100644 --- a/include/asm-um/pgtable-3level.h +++ b/include/asm-um/pgtable-3level.h @@ -145,11 +145,11 @@ static inline pmd_t pfn_pmd(pfn_t page_nr, pgprot_t pgprot) */ #define PTE_FILE_MAX_BITS 32 -#ifdef CONFIG_64_BIT +#ifdef CONFIG_64BIT #define pte_to_pgoff(p) ((p).pte >> 32) -#define pgoff_to_pte(off) ((pte_t) { ((off) < 32) | _PAGE_FILE }) +#define pgoff_to_pte(off) ((pte_t) { ((off) << 32) | _PAGE_FILE }) #else diff --git a/include/asm-um/pgtable.h b/include/asm-um/pgtable.h index 71f9c0c78c0..510e513c7f8 100644 --- a/include/asm-um/pgtable.h +++ b/include/asm-um/pgtable.h @@ -106,7 +106,7 @@ extern unsigned long end_iomem; /* * Define this if things work differently on an i386 and an i486: * it will (on an i486) warn about kernel memory accesses that are - * done without a 'verify_area(VERIFY_WRITE,..)' + * done without a 'access_ok(VERIFY_WRITE,..)' */ #undef TEST_VERIFY_AREA diff --git a/include/asm-um/processor-generic.h b/include/asm-um/processor-generic.h index 038ba6fc88b..b2fc94fbc2d 100644 --- a/include/asm-um/processor-generic.h +++ b/include/asm-um/processor-generic.h @@ -17,12 +17,13 @@ struct task_struct; struct mm_struct; struct thread_struct { + /* This flag is set to 1 before calling do_fork (and analyzed in + * copy_thread) to mark that we are begin called from userspace (fork / + * vfork / clone), and reset to 0 after. It is left to 0 when called + * from kernelspace (i.e. kernel_thread() or fork_idle(), as of 2.6.11). */ int forking; int nsyscalls; struct pt_regs regs; - unsigned long cr2; - int err; - unsigned long trap_no; int singlestep_syscall; void *fault_addr; void *fault_catcher; @@ -70,8 +71,6 @@ struct thread_struct { .forking = 0, \ .nsyscalls = 0, \ .regs = EMPTY_REGS, \ - .cr2 = 0, \ - .err = 0, \ .fault_addr = NULL, \ .prev_sched = NULL, \ .temp_stack = 0, \ @@ -89,7 +88,11 @@ extern struct task_struct *alloc_task_struct(void); extern void release_thread(struct task_struct *); extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags); extern void dump_thread(struct pt_regs *regs, struct user *u); -extern void prepare_to_copy(struct task_struct *tsk); + +static inline void prepare_to_copy(struct task_struct *tsk) +{ +} + extern unsigned long thread_saved_pc(struct task_struct *t); diff --git a/include/asm-um/processor-i386.h b/include/asm-um/processor-i386.h index 2deb8f1adbf..431bad3ae9d 100644 --- a/include/asm-um/processor-i386.h +++ b/include/asm-um/processor-i386.h @@ -9,13 +9,18 @@ extern int host_has_xmm; extern int host_has_cmov; +/* include faultinfo structure */ +#include "sysdep/faultinfo.h" + struct arch_thread { unsigned long debugregs[8]; int debugregs_seq; + struct faultinfo faultinfo; }; #define INIT_ARCH_THREAD { .debugregs = { [ 0 ... 7 ] = 0 }, \ - .debugregs_seq = 0 } + .debugregs_seq = 0, \ + .faultinfo = { 0, 0, 0 } } #include "asm/arch/user.h" diff --git a/include/asm-um/processor-x86_64.h b/include/asm-um/processor-x86_64.h index a1ae3a4cd93..0beb9a42ae0 100644 --- a/include/asm-um/processor-x86_64.h +++ b/include/asm-um/processor-x86_64.h @@ -7,9 +7,13 @@ #ifndef __UM_PROCESSOR_X86_64_H #define __UM_PROCESSOR_X86_64_H -#include "asm/arch/user.h" +/* include faultinfo structure */ +#include "sysdep/faultinfo.h" struct arch_thread { + unsigned long debugregs[8]; + int debugregs_seq; + struct faultinfo faultinfo; }; /* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */ @@ -20,7 +24,11 @@ extern inline void rep_nop(void) #define cpu_relax() rep_nop() -#define INIT_ARCH_THREAD { } +#define INIT_ARCH_THREAD { .debugregs = { [ 0 ... 7 ] = 0 }, \ + .debugregs_seq = 0, \ + .faultinfo = { 0, 0, 0 } } + +#include "asm/arch/user.h" #define current_text_addr() \ ({ void *pc; __asm__("movq $1f,%0\n1:":"=g" (pc)); pc; }) diff --git a/include/asm-um/ptrace-i386.h b/include/asm-um/ptrace-i386.h index 9e47590ec29..04222f35c43 100644 --- a/include/asm-um/ptrace-i386.h +++ b/include/asm-um/ptrace-i386.h @@ -6,6 +6,8 @@ #ifndef __UM_PTRACE_I386_H #define __UM_PTRACE_I386_H +#define HOST_AUDIT_ARCH AUDIT_ARCH_I386 + #include "sysdep/ptrace.h" #include "asm/ptrace-generic.h" diff --git a/include/asm-um/ptrace-x86_64.h b/include/asm-um/ptrace-x86_64.h index c34be39b78b..be51219a8ff 100644 --- a/include/asm-um/ptrace-x86_64.h +++ b/include/asm-um/ptrace-x86_64.h @@ -14,6 +14,8 @@ #include "asm/ptrace-generic.h" #undef signal_fault +#define HOST_AUDIT_ARCH AUDIT_ARCH_X86_64 + void signal_fault(struct pt_regs_subarch *regs, void *frame, char *where); #define FS_BASE (21 * sizeof(unsigned long)) diff --git a/include/asm-um/thread_info.h b/include/asm-um/thread_info.h index bffb577bc54..a10ea155907 100644 --- a/include/asm-um/thread_info.h +++ b/include/asm-um/thread_info.h @@ -72,12 +72,14 @@ static inline struct thread_info *current_thread_info(void) */ #define TIF_RESTART_BLOCK 4 #define TIF_MEMDIE 5 +#define TIF_SYSCALL_AUDIT 6 #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE) #define _TIF_SIGPENDING (1 << TIF_SIGPENDING) #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED) #define _TIF_POLLING_NRFLAG (1 << TIF_POLLING_NRFLAG) -#define _TIF_RESTART_BLOCK (1 << TIF_RESTART_BLOCK) +#define _TIF_MEMDIE (1 << TIF_MEMDIE) +#define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) #endif diff --git a/include/asm-v850/bug.h b/include/asm-v850/bug.h index c778916bf7f..b0ed2d35f3e 100644 --- a/include/asm-v850/bug.h +++ b/include/asm-v850/bug.h @@ -14,9 +14,12 @@ #ifndef __V850_BUG_H__ #define __V850_BUG_H__ +#ifdef CONFIG_BUG extern void __bug (void) __attribute__ ((noreturn)); #define BUG() __bug() #define HAVE_ARCH_BUG +#endif + #include <asm-generic/bug.h> #endif /* __V850_BUG_H__ */ diff --git a/include/asm-v850/signal.h b/include/asm-v850/signal.h index 407db875899..cb52caa6992 100644 --- a/include/asm-v850/signal.h +++ b/include/asm-v850/signal.h @@ -110,32 +110,7 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 - -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif /* __KERNEL__ */ - - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ - -/* Type of a signal handler. */ -typedef void (*__sighandler_t)(int); - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ - +#include <asm-generic/signal.h> #ifdef __KERNEL__ diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h index c025cc3ef78..e4b1017b8b2 100644 --- a/include/asm-x86_64/apic.h +++ b/include/asm-x86_64/apic.h @@ -99,7 +99,6 @@ extern void disable_APIC_timer(void); extern void enable_APIC_timer(void); extern void clustered_apic_check(void); -extern int check_nmi_watchdog(void); extern void nmi_watchdog_default(void); extern int setup_nmi_watchdog(char *); diff --git a/include/asm-x86_64/bootsetup.h b/include/asm-x86_64/bootsetup.h index b570a484dc5..b829f7b534b 100644 --- a/include/asm-x86_64/bootsetup.h +++ b/include/asm-x86_64/bootsetup.h @@ -2,7 +2,8 @@ #ifndef _X86_64_BOOTSETUP_H #define _X86_64_BOOTSETUP_H 1 -extern char x86_boot_params[2048]; +#define BOOT_PARAM_SIZE 4096 +extern char x86_boot_params[BOOT_PARAM_SIZE]; /* * This is set up by the setup-routine at boot-time diff --git a/include/asm-x86_64/bug.h b/include/asm-x86_64/bug.h index 19aed6e78fe..bdbf66eab6e 100644 --- a/include/asm-x86_64/bug.h +++ b/include/asm-x86_64/bug.h @@ -15,11 +15,13 @@ struct bug_frame { unsigned short line; } __attribute__((packed)); +#ifdef CONFIG_BUG #define HAVE_ARCH_BUG #define BUG() \ asm volatile("ud2 ; .quad %c1 ; .short %c0" :: \ "i"(__LINE__), "i" (__stringify(__FILE__))) void out_of_line_bug(void); -#include <asm-generic/bug.h> +#endif +#include <asm-generic/bug.h> #endif diff --git a/include/asm-x86_64/cpufeature.h b/include/asm-x86_64/cpufeature.h index e68ad97a631..aea308c6570 100644 --- a/include/asm-x86_64/cpufeature.h +++ b/include/asm-x86_64/cpufeature.h @@ -7,7 +7,7 @@ #ifndef __ASM_X8664_CPUFEATURE_H #define __ASM_X8664_CPUFEATURE_H -#define NCAPINTS 6 +#define NCAPINTS 7 /* N 32-bit words worth of info */ /* Intel-defined CPU features, CPUID level 0x00000001, word 0 */ #define X86_FEATURE_FPU (0*32+ 0) /* Onboard FPU */ @@ -74,9 +74,15 @@ #define X86_FEATURE_CX16 (4*32+13) /* CMPXCHG16B */ #define X86_FEATURE_XTPR (4*32+14) /* Send Task Priority Messages */ -/* More extended AMD flags: CPUID level 0x80000001, ecx, word 5 */ -#define X86_FEATURE_LAHF_LM (5*32+ 0) /* LAHF/SAHF in long mode */ -#define X86_FEATURE_CMP_LEGACY (5*32+ 1) /* If yes HyperThreading not valid */ +/* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */ +#define X86_FEATURE_XSTORE (5*32+ 2) /* on-CPU RNG present (xstore insn) */ +#define X86_FEATURE_XSTORE_EN (5*32+ 3) /* on-CPU RNG enabled */ +#define X86_FEATURE_XCRYPT (5*32+ 6) /* on-CPU crypto (xcrypt insn) */ +#define X86_FEATURE_XCRYPT_EN (5*32+ 7) /* on-CPU crypto enabled */ + +/* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */ +#define X86_FEATURE_LAHF_LM (6*32+ 0) /* LAHF/SAHF in long mode */ +#define X86_FEATURE_CMP_LEGACY (6*32+ 1) /* If yes HyperThreading not valid */ #define cpu_has(c, bit) test_bit(bit, (c)->x86_capability) #define boot_cpu_has(bit) test_bit(bit, boot_cpu_data.x86_capability) diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h index 08f83a4b4f4..8e94edf0b98 100644 --- a/include/asm-x86_64/e820.h +++ b/include/asm-x86_64/e820.h @@ -14,7 +14,7 @@ #include <linux/mmzone.h> #define E820MAP 0x2d0 /* our map */ -#define E820MAX 32 /* number of entries in E820MAP */ +#define E820MAX 128 /* number of entries in E820MAP */ #define E820NR 0x1e8 /* # entries in E820MAP */ #define E820_RAM 1 diff --git a/include/asm-x86_64/floppy.h b/include/asm-x86_64/floppy.h index bca9b28a1a0..af7ded63b51 100644 --- a/include/asm-x86_64/floppy.h +++ b/include/asm-x86_64/floppy.h @@ -223,7 +223,7 @@ static int hard_dma_setup(char *addr, unsigned long size, int mode, int io) return 0; } -struct fd_routine_l { +static struct fd_routine_l { int (*_request_dma)(unsigned int dmanr, const char * device_id); void (*_free_dma)(unsigned int dmanr); int (*_get_dma_residue)(unsigned int dummy); diff --git a/include/asm-x86_64/siginfo.h b/include/asm-x86_64/siginfo.h index 7bc15985f12..d09a1e6e724 100644 --- a/include/asm-x86_64/siginfo.h +++ b/include/asm-x86_64/siginfo.h @@ -3,8 +3,6 @@ #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) - #include <asm-generic/siginfo.h> #endif diff --git a/include/asm-x86_64/signal.h b/include/asm-x86_64/signal.h index 643a20d7376..fe9b96d9481 100644 --- a/include/asm-x86_64/signal.h +++ b/include/asm-x86_64/signal.h @@ -116,35 +116,9 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - -#define SIG_BLOCK 0 /* for blocking signals */ -#define SIG_UNBLOCK 1 /* for unblocking signals */ -#define SIG_SETMASK 2 /* for setting the signal mask */ +#include <asm-generic/signal.h> #ifndef __ASSEMBLY__ -/* Type of a signal handler. */ -typedef void __signalfn_t(int); -typedef __signalfn_t __user *__sighandler_t; - -typedef void __restorefn_t(void); -typedef __restorefn_t __user *__sigrestore_t; - -#define SIG_DFL ((__sighandler_t)0) /* default signal handling */ -#define SIG_IGN ((__sighandler_t)1) /* ignore signal */ -#define SIG_ERR ((__sighandler_t)-1) /* error return from signal */ struct sigaction { __sighandler_t sa_handler; diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h index 3d65d240dc9..3c9af6fd433 100644 --- a/include/asm-x86_64/unistd.h +++ b/include/asm-x86_64/unistd.h @@ -76,7 +76,7 @@ __SYSCALL(__NR_madvise, sys_madvise) #define __NR_shmget 29 __SYSCALL(__NR_shmget, sys_shmget) #define __NR_shmat 30 -__SYSCALL(__NR_shmat, wrap_sys_shmat) +__SYSCALL(__NR_shmat, sys_shmat) #define __NR_shmctl 31 __SYSCALL(__NR_shmctl, sys_shmctl) diff --git a/include/linux/audit.h b/include/linux/audit.h index 3628f7cfb17..19f04b04979 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -1,4 +1,4 @@ -/* audit.h -- Auditing support -*- linux-c -*- +/* audit.h -- Auditing support * * Copyright 2003-2004 Red Hat Inc., Durham, North Carolina. * All Rights Reserved. @@ -24,6 +24,9 @@ #ifndef _LINUX_AUDIT_H_ #define _LINUX_AUDIT_H_ +#include <linux/sched.h> +#include <linux/elf.h> + /* Request and reply types */ #define AUDIT_GET 1000 /* Get status */ #define AUDIT_SET 1001 /* Set status (enable/disable/auditd) */ @@ -67,6 +70,7 @@ #define AUDIT_FSGID 8 #define AUDIT_LOGINUID 9 #define AUDIT_PERS 10 +#define AUDIT_ARCH 11 /* These are ONLY useful when checking * at syscall exit time (AUDIT_AT_EXIT). */ @@ -96,6 +100,38 @@ #define AUDIT_FAIL_PRINTK 1 #define AUDIT_FAIL_PANIC 2 +/* distinguish syscall tables */ +#define __AUDIT_ARCH_64BIT 0x80000000 +#define __AUDIT_ARCH_LE 0x40000000 +#define AUDIT_ARCH_ALPHA (EM_ALPHA|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_ARM (EM_ARM|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_ARMEB (EM_ARM) +#define AUDIT_ARCH_CRIS (EM_CRIS|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_FRV (EM_FRV) +#define AUDIT_ARCH_H8300 (EM_H8_300) +#define AUDIT_ARCH_I386 (EM_386|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_IA64 (EM_IA_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_M32R (EM_M32R) +#define AUDIT_ARCH_M68K (EM_68K) +#define AUDIT_ARCH_MIPS (EM_MIPS) +#define AUDIT_ARCH_MIPSEL (EM_MIPS|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_MIPS64 (EM_MIPS|__AUDIT_ARCH_64BIT) +#define AUDIT_ARCH_MIPSEL64 (EM_MIPS|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_PARISC (EM_PARISC) +#define AUDIT_ARCH_PARISC64 (EM_PARISC|__AUDIT_ARCH_64BIT) +#define AUDIT_ARCH_PPC (EM_PPC) +#define AUDIT_ARCH_PPC64 (EM_PPC64|__AUDIT_ARCH_64BIT) +#define AUDIT_ARCH_S390 (EM_S390) +#define AUDIT_ARCH_S390X (EM_S390|__AUDIT_ARCH_64BIT) +#define AUDIT_ARCH_SH (EM_SH) +#define AUDIT_ARCH_SHEL (EM_SH|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_SH64 (EM_SH|__AUDIT_ARCH_64BIT) +#define AUDIT_ARCH_SHEL64 (EM_SH|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_SPARC (EM_SPARC) +#define AUDIT_ARCH_SPARC64 (EM_SPARC64|__AUDIT_ARCH_64BIT) +#define AUDIT_ARCH_V850 (EM_V850|__AUDIT_ARCH_LE) +#define AUDIT_ARCH_X86_64 (EM_X86_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) + #ifndef __KERNEL__ struct audit_message { struct nlmsghdr nlh; @@ -129,32 +165,36 @@ struct audit_buffer; struct audit_context; struct inode; +#define AUDITSC_INVALID 0 +#define AUDITSC_SUCCESS 1 +#define AUDITSC_FAILURE 2 +#define AUDITSC_RESULT(x) ( ((long)(x))<0?AUDITSC_FAILURE:AUDITSC_SUCCESS ) #ifdef CONFIG_AUDITSYSCALL /* These are defined in auditsc.c */ /* Public API */ extern int audit_alloc(struct task_struct *task); extern void audit_free(struct task_struct *task); -extern void audit_syscall_entry(struct task_struct *task, +extern void audit_syscall_entry(struct task_struct *task, int arch, int major, unsigned long a0, unsigned long a1, unsigned long a2, unsigned long a3); -extern void audit_syscall_exit(struct task_struct *task, int return_code); +extern void audit_syscall_exit(struct task_struct *task, int failed, long return_code); extern void audit_getname(const char *name); extern void audit_putname(const char *name); extern void audit_inode(const char *name, const struct inode *inode); /* Private API (for audit.c only) */ extern int audit_receive_filter(int type, int pid, int uid, int seq, - void *data); + void *data, uid_t loginuid); extern void audit_get_stamp(struct audit_context *ctx, - struct timespec *t, int *serial); -extern int audit_set_loginuid(struct audit_context *ctx, uid_t loginuid); + struct timespec *t, unsigned int *serial); +extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); #else #define audit_alloc(t) ({ 0; }) #define audit_free(t) do { ; } while (0) -#define audit_syscall_entry(t,a,b,c,d,e) do { ; } while (0) -#define audit_syscall_exit(t,r) do { ; } while (0) +#define audit_syscall_entry(t,ta,a,b,c,d,e) do { ; } while (0) +#define audit_syscall_exit(t,f,r) do { ; } while (0) #define audit_getname(n) do { ; } while (0) #define audit_putname(n) do { ; } while (0) #define audit_inode(n,i) do { ; } while (0) @@ -174,11 +214,15 @@ extern void audit_log_format(struct audit_buffer *ab, const char *fmt, ...) __attribute__((format(printf,2,3))); extern void audit_log_end(struct audit_buffer *ab); +extern void audit_log_hex(struct audit_buffer *ab, + const unsigned char *buf, + size_t len); +extern void audit_log_untrustedstring(struct audit_buffer *ab, + const char *string); extern void audit_log_d_path(struct audit_buffer *ab, const char *prefix, struct dentry *dentry, struct vfsmount *vfsmnt); - /* Private API (for auditsc.c only) */ extern void audit_send_reply(int pid, int seq, int type, int done, int multi, @@ -190,6 +234,8 @@ extern void audit_log_lost(const char *message); #define audit_log_vformat(b,f,a) do { ; } while (0) #define audit_log_format(b,f,...) do { ; } while (0) #define audit_log_end(b) do { ; } while (0) +#define audit_log_hex(a,b,l) do { ; } while (0) +#define audit_log_untrustedstring(a,s) do { ; } while (0) #define audit_log_d_path(b,p,d,v) do { ; } while (0) #endif #endif diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h index d1c7b0ec7c2..a1657fb9951 100644 --- a/include/linux/auto_fs4.h +++ b/include/linux/auto_fs4.h @@ -23,7 +23,7 @@ #define AUTOFS_MIN_PROTO_VERSION 3 #define AUTOFS_MAX_PROTO_VERSION 4 -#define AUTOFS_PROTO_SUBVERSION 5 +#define AUTOFS_PROTO_SUBVERSION 6 /* Mask for expire behaviour */ #define AUTOFS_EXP_IMMEDIATE 1 diff --git a/include/linux/awe_voice.h b/include/linux/awe_voice.h index da0e27de752..4bf9f33048e 100644 --- a/include/linux/awe_voice.h +++ b/include/linux/awe_voice.h @@ -29,9 +29,9 @@ #define SAMPLE_TYPE_AWE32 0x20 #endif -#ifndef _PATCHKEY -#define _PATCHKEY(id) ((id<<8)|0xfd) -#endif +#define _LINUX_PATCHKEY_H_INDIRECT +#include <linux/patchkey.h> +#undef _LINUX_PATCHKEY_H_INDIRECT /*---------------------------------------------------------------- * patch information record diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index 54f820832c7..7e736e201c4 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -77,7 +77,6 @@ extern int flush_old_exec(struct linux_binprm * bprm); extern int setup_arg_pages(struct linux_binprm * bprm, unsigned long stack_top, int executable_stack); -extern int copy_strings(int argc,char __user * __user * argv,struct linux_binprm *bprm); extern int copy_strings_kernel(int argc,char ** argv,struct linux_binprm *bprm); extern void compute_creds(struct linux_binprm *binprm); extern int do_coredump(long signr, int exit_code, struct pt_regs * regs); diff --git a/include/linux/compiler-gcc2.h b/include/linux/compiler-gcc2.h index 5a359153ffd..ebed17660c5 100644 --- a/include/linux/compiler-gcc2.h +++ b/include/linux/compiler-gcc2.h @@ -22,3 +22,8 @@ # define __attribute_pure__ __attribute__((pure)) # define __attribute_const__ __attribute__((__const__)) #endif + +/* GCC 2.95.x/2.96 recognize __va_copy, but not va_copy. Actually later GCC's + * define both va_copy and __va_copy, but the latter may go away, so limit this + * to this header */ +#define va_copy __va_copy diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 487725cf0d0..d7378215b85 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -90,6 +90,12 @@ extern void __chk_io_ptr(void __iomem *); # define __deprecated /* unimplemented */ #endif +#ifdef MODULE +#define __deprecated_for_modules __deprecated +#else +#define __deprecated_for_modules +#endif + #ifndef __must_check #define __must_check #endif diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 910eca35583..f21af067d01 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -103,6 +103,7 @@ struct cpufreq_policy { #define CPUFREQ_PRECHANGE (0) #define CPUFREQ_POSTCHANGE (1) #define CPUFREQ_RESUMECHANGE (8) +#define CPUFREQ_SUSPENDCHANGE (9) struct cpufreq_freqs { unsigned int cpu; /* cpu nr */ @@ -200,6 +201,7 @@ struct cpufreq_driver { /* optional */ int (*exit) (struct cpufreq_policy *policy); + int (*suspend) (struct cpufreq_policy *policy, u32 state); int (*resume) (struct cpufreq_policy *policy); struct freq_attr **attr; }; @@ -211,7 +213,8 @@ struct cpufreq_driver { #define CPUFREQ_CONST_LOOPS 0x02 /* loops_per_jiffy or other kernel * "constants" aren't affected by * frequency transitions */ - +#define CPUFREQ_PM_NO_WARN 0x04 /* don't warn on suspend/resume speed + * mismatches */ int cpufreq_register_driver(struct cpufreq_driver *driver_data); int cpufreq_unregister_driver(struct cpufreq_driver *driver_data); diff --git a/include/linux/edd.h b/include/linux/edd.h index c6e6747a401..162512b886f 100644 --- a/include/linux/edd.h +++ b/include/linux/edd.h @@ -32,7 +32,7 @@ #define EDDNR 0x1e9 /* addr of number of edd_info structs at EDDBUF in boot_params - treat this as 1 byte */ -#define EDDBUF 0x600 /* addr of edd_info structs in boot_params */ +#define EDDBUF 0xd00 /* addr of edd_info structs in boot_params */ #define EDDMAXNR 6 /* number of edd_info structs starting at EDDBUF */ #define EDDEXTSIZE 8 /* change these if you muck with the structures */ #define EDDPARMSIZE 74 diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index 396c48cbaeb..220748b7abe 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -7,7 +7,7 @@ * * Version: @(#)eth.h 1.0.4 05/13/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * Relocated to include/linux where it belongs by Alan Cox diff --git a/include/linux/fb.h b/include/linux/fb.h index b45d3e2d711..b468bf49654 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -563,6 +563,9 @@ struct fb_ops { int (*fb_setcolreg)(unsigned regno, unsigned red, unsigned green, unsigned blue, unsigned transp, struct fb_info *info); + /* set color registers in batch */ + int (*fb_setcmap)(struct fb_cmap *cmap, struct fb_info *info); + /* blank display */ int (*fb_blank)(int blank, struct fb_info *info); diff --git a/include/linux/fddidevice.h b/include/linux/fddidevice.h index 2e5ee47f3e1..002f6367697 100644 --- a/include/linux/fddidevice.h +++ b/include/linux/fddidevice.h @@ -10,7 +10,7 @@ * Author: Lawrence V. Stefani, <stefani@lkg.dec.com> * * fddidevice.h is based on previous trdevice.h work by - * Ross Biro, <bir7@leland.Stanford.Edu> + * Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Alan Cox, <gw4pts@gw4pts.ampr.org> * diff --git a/include/linux/fs.h b/include/linux/fs.h index 5df687d940f..0180102dace 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1053,83 +1053,87 @@ static inline void file_accessed(struct file *file) int sync_inode(struct inode *inode, struct writeback_control *wbc); /** - * &export_operations - for nfsd to communicate with file systems - * decode_fh: decode a file handle fragment and return a &struct dentry - * encode_fh: encode a file handle fragment from a dentry - * get_name: find the name for a given inode in a given directory - * get_parent: find the parent of a given directory - * get_dentry: find a dentry for the inode given a file handle sub-fragment + * struct export_operations - for nfsd to communicate with file systems + * @decode_fh: decode a file handle fragment and return a &struct dentry + * @encode_fh: encode a file handle fragment from a dentry + * @get_name: find the name for a given inode in a given directory + * @get_parent: find the parent of a given directory + * @get_dentry: find a dentry for the inode given a file handle sub-fragment * * Description: * The export_operations structure provides a means for nfsd to communicate * with a particular exported file system - particularly enabling nfsd and * the filesystem to co-operate when dealing with file handles. * - * export_operations contains two basic operation for dealing with file handles, - * decode_fh() and encode_fh(), and allows for some other operations to be defined - * which standard helper routines use to get specific information from the - * filesystem. + * export_operations contains two basic operation for dealing with file + * handles, decode_fh() and encode_fh(), and allows for some other + * operations to be defined which standard helper routines use to get + * specific information from the filesystem. * * nfsd encodes information use to determine which filesystem a filehandle - * applies to in the initial part of the file handle. The remainder, termed a - * file handle fragment, is controlled completely by the filesystem. - * The standard helper routines assume that this fragment will contain one or two - * sub-fragments, one which identifies the file, and one which may be used to - * identify the (a) directory containing the file. + * applies to in the initial part of the file handle. The remainder, termed + * a file handle fragment, is controlled completely by the filesystem. The + * standard helper routines assume that this fragment will contain one or + * two sub-fragments, one which identifies the file, and one which may be + * used to identify the (a) directory containing the file. * * In some situations, nfsd needs to get a dentry which is connected into a - * specific part of the file tree. To allow for this, it passes the function - * acceptable() together with a @context which can be used to see if the dentry - * is acceptable. As there can be multiple dentrys for a given file, the filesystem - * should check each one for acceptability before looking for the next. As soon - * as an acceptable one is found, it should be returned. + * specific part of the file tree. To allow for this, it passes the + * function acceptable() together with a @context which can be used to see + * if the dentry is acceptable. As there can be multiple dentrys for a + * given file, the filesystem should check each one for acceptability before + * looking for the next. As soon as an acceptable one is found, it should + * be returned. * * decode_fh: - * @decode_fh is given a &struct super_block (@sb), a file handle fragment (@fh, @fh_len) - * and an acceptability testing function (@acceptable, @context). It should return - * a &struct dentry which refers to the same file that the file handle fragment refers - * to, and which passes the acceptability test. If it cannot, it should return - * a %NULL pointer if the file was found but no acceptable &dentries were available, or - * a %ERR_PTR error code indicating why it couldn't be found (e.g. %ENOENT or %ENOMEM). + * @decode_fh is given a &struct super_block (@sb), a file handle fragment + * (@fh, @fh_len) and an acceptability testing function (@acceptable, + * @context). It should return a &struct dentry which refers to the same + * file that the file handle fragment refers to, and which passes the + * acceptability test. If it cannot, it should return a %NULL pointer if + * the file was found but no acceptable &dentries were available, or a + * %ERR_PTR error code indicating why it couldn't be found (e.g. %ENOENT or + * %ENOMEM). * * encode_fh: - * @encode_fh should store in the file handle fragment @fh (using at most @max_len bytes) - * information that can be used by @decode_fh to recover the file refered to by the - * &struct dentry @de. If the @connectable flag is set, the encode_fh() should store - * sufficient information so that a good attempt can be made to find not only - * the file but also it's place in the filesystem. This typically means storing - * a reference to de->d_parent in the filehandle fragment. - * encode_fh() should return the number of bytes stored or a negative error code - * such as %-ENOSPC + * @encode_fh should store in the file handle fragment @fh (using at most + * @max_len bytes) information that can be used by @decode_fh to recover the + * file refered to by the &struct dentry @de. If the @connectable flag is + * set, the encode_fh() should store sufficient information so that a good + * attempt can be made to find not only the file but also it's place in the + * filesystem. This typically means storing a reference to de->d_parent in + * the filehandle fragment. encode_fh() should return the number of bytes + * stored or a negative error code such as %-ENOSPC * * get_name: - * @get_name should find a name for the given @child in the given @parent directory. - * The name should be stored in the @name (with the understanding that it is already - * pointing to a a %NAME_MAX+1 sized buffer. get_name() should return %0 on success, - * a negative error code or error. - * @get_name will be called without @parent->i_sem held. + * @get_name should find a name for the given @child in the given @parent + * directory. The name should be stored in the @name (with the + * understanding that it is already pointing to a a %NAME_MAX+1 sized + * buffer. get_name() should return %0 on success, a negative error code + * or error. @get_name will be called without @parent->i_sem held. * * get_parent: - * @get_parent should find the parent directory for the given @child which is also - * a directory. In the event that it cannot be found, or storage space cannot be - * allocated, a %ERR_PTR should be returned. + * @get_parent should find the parent directory for the given @child which + * is also a directory. In the event that it cannot be found, or storage + * space cannot be allocated, a %ERR_PTR should be returned. * * get_dentry: - * Given a &super_block (@sb) and a pointer to a file-system specific inode identifier, - * possibly an inode number, (@inump) get_dentry() should find the identified inode and - * return a dentry for that inode. - * Any suitable dentry can be returned including, if necessary, a new dentry created - * with d_alloc_root. The caller can then find any other extant dentrys by following the - * d_alias links. If a new dentry was created using d_alloc_root, DCACHE_NFSD_DISCONNECTED - * should be set, and the dentry should be d_rehash()ed. + * Given a &super_block (@sb) and a pointer to a file-system specific inode + * identifier, possibly an inode number, (@inump) get_dentry() should find + * the identified inode and return a dentry for that inode. Any suitable + * dentry can be returned including, if necessary, a new dentry created with + * d_alloc_root. The caller can then find any other extant dentrys by + * following the d_alias links. If a new dentry was created using + * d_alloc_root, DCACHE_NFSD_DISCONNECTED should be set, and the dentry + * should be d_rehash()ed. * - * If the inode cannot be found, either a %NULL pointer or an %ERR_PTR code can be returned. - * The @inump will be whatever was passed to nfsd_find_fh_dentry() in either the - * @obj or @parent parameters. + * If the inode cannot be found, either a %NULL pointer or an %ERR_PTR code + * can be returned. The @inump will be whatever was passed to + * nfsd_find_fh_dentry() in either the @obj or @parent parameters. * * Locking rules: - * get_parent is called with child->d_inode->i_sem down - * get_name is not (which is possibly inconsistent) + * get_parent is called with child->d_inode->i_sem down + * get_name is not (which is possibly inconsistent) */ struct export_operations { @@ -1337,7 +1341,7 @@ extern int fs_may_remount_ro(struct super_block *); extern int check_disk_change(struct block_device *); extern int invalidate_inodes(struct super_block *); -extern int __invalidate_device(struct block_device *, int); +extern int __invalidate_device(struct block_device *); extern int invalidate_partition(struct gendisk *, int); unsigned long invalidate_mapping_pages(struct address_space *mapping, pgoff_t start, pgoff_t end); diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 848a1baac07..af7407e8cfc 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -38,14 +38,16 @@ struct vm_area_struct; #define __GFP_NO_GROW 0x2000u /* Slab internal usage */ #define __GFP_COMP 0x4000u /* Add compound page metadata */ #define __GFP_ZERO 0x8000u /* Return zeroed page on success */ +#define __GFP_NOMEMALLOC 0x10000u /* Don't use emergency reserves */ -#define __GFP_BITS_SHIFT 16 /* Room for 16 __GFP_FOO bits */ +#define __GFP_BITS_SHIFT 20 /* Room for 20 __GFP_FOO bits */ #define __GFP_BITS_MASK ((1 << __GFP_BITS_SHIFT) - 1) /* if you forget to add the bitmask here kernel will crash, period */ #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \ __GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \ - __GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP) + __GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP| \ + __GFP_NOMEMALLOC) #define GFP_ATOMIC (__GFP_HIGH) #define GFP_NOIO (__GFP_WAIT) diff --git a/include/linux/hippidevice.h b/include/linux/hippidevice.h index 89b3a4a5b76..9debe6bbe5f 100644 --- a/include/linux/hippidevice.h +++ b/include/linux/hippidevice.h @@ -10,7 +10,7 @@ * Author: Jes Sorensen, <Jes.Sorensen@cern.ch> * * hippidevice.h is based on previous fddidevice.h work by - * Ross Biro, <bir7@leland.Stanford.Edu> + * Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Alan Cox, <gw4pts@gw4pts.ampr.org> * Lawrence V. Stefani, <stefani@lkg.dec.com> diff --git a/include/linux/if.h b/include/linux/if.h index 110282dbd3e..d73a9d62f20 100644 --- a/include/linux/if.h +++ b/include/linux/if.h @@ -8,7 +8,7 @@ * Version: @(#)if.h 1.0.2 04/18/93 * * Authors: Original taken from Berkeley UNIX 4.3, (c) UCB 1982-1988 - * Ross Biro, <bir7@leland.Stanford.Edu> + * Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * This program is free software; you can redistribute it and/or diff --git a/include/linux/if_arp.h b/include/linux/if_arp.h index bbf49bcd770..0856548a2a0 100644 --- a/include/linux/if_arp.h +++ b/include/linux/if_arp.h @@ -9,7 +9,7 @@ * * Authors: Original taken from Berkeley UNIX 4.3, (c) UCB 1986-1988 * Portions taken from the KA9Q/NOS (v2.00m PA0GRI) source. - * Ross Biro, <bir7@leland.Stanford.Edu> + * Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Florian La Roche, * Jonathan Layes <layes@loran.com> diff --git a/include/linux/if_ltalk.h b/include/linux/if_ltalk.h index e75e832b7ff..76525760ba4 100644 --- a/include/linux/if_ltalk.h +++ b/include/linux/if_ltalk.h @@ -6,7 +6,7 @@ #define LTALK_ALEN 1 #ifdef __KERNEL__ -extern void ltalk_setup(struct net_device *); +extern struct net_device *alloc_ltalkdev(int sizeof_priv); #endif #endif diff --git a/include/linux/input.h b/include/linux/input.h index b70df8fe60e..72731d7d189 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -328,6 +328,11 @@ struct input_absinfo { #define KEY_BRIGHTNESSUP 225 #define KEY_MEDIA 226 +#define KEY_SWITCHVIDEOMODE 227 +#define KEY_KBDILLUMTOGGLE 228 +#define KEY_KBDILLUMDOWN 229 +#define KEY_KBDILLUMUP 230 + #define KEY_UNKNOWN 240 #define BTN_MISC 0x100 diff --git a/include/linux/iso_fs.h b/include/linux/iso_fs.h index 099039d4b10..47967878bfe 100644 --- a/include/linux/iso_fs.h +++ b/include/linux/iso_fs.h @@ -1,4 +1,3 @@ - #ifndef _ISOFS_FS_H #define _ISOFS_FS_H @@ -163,150 +162,4 @@ struct iso_directory_record { #define ISOFS_SUPER_MAGIC 0x9660 -#ifdef __KERNEL__ -/* Number conversion inlines, named after the section in ISO 9660 - they correspond to. */ - -#include <asm/byteorder.h> -#include <asm/unaligned.h> -#include <linux/iso_fs_i.h> -#include <linux/iso_fs_sb.h> - -static inline struct isofs_sb_info *ISOFS_SB(struct super_block *sb) -{ - return sb->s_fs_info; -} - -static inline struct iso_inode_info *ISOFS_I(struct inode *inode) -{ - return container_of(inode, struct iso_inode_info, vfs_inode); -} - -static inline int isonum_711(char *p) -{ - return *(u8 *)p; -} -static inline int isonum_712(char *p) -{ - return *(s8 *)p; -} -static inline unsigned int isonum_721(char *p) -{ - return le16_to_cpu(get_unaligned((__le16 *)p)); -} -static inline unsigned int isonum_722(char *p) -{ - return be16_to_cpu(get_unaligned((__le16 *)p)); -} -static inline unsigned int isonum_723(char *p) -{ - /* Ignore bigendian datum due to broken mastering programs */ - return le16_to_cpu(get_unaligned((__le16 *)p)); -} -static inline unsigned int isonum_731(char *p) -{ - return le32_to_cpu(get_unaligned((__le32 *)p)); -} -static inline unsigned int isonum_732(char *p) -{ - return be32_to_cpu(get_unaligned((__le32 *)p)); -} -static inline unsigned int isonum_733(char *p) -{ - /* Ignore bigendian datum due to broken mastering programs */ - return le32_to_cpu(get_unaligned((__le32 *)p)); -} -extern int iso_date(char *, int); - -struct inode; /* To make gcc happy */ - -extern int parse_rock_ridge_inode(struct iso_directory_record *, struct inode *); -extern int get_rock_ridge_filename(struct iso_directory_record *, char *, struct inode *); -extern int isofs_name_translate(struct iso_directory_record *, char *, struct inode *); - -int get_joliet_filename(struct iso_directory_record *, unsigned char *, struct inode *); -int get_acorn_filename(struct iso_directory_record *, char *, struct inode *); - -extern struct dentry *isofs_lookup(struct inode *, struct dentry *, struct nameidata *); -extern struct buffer_head *isofs_bread(struct inode *, sector_t); -extern int isofs_get_blocks(struct inode *, sector_t, struct buffer_head **, unsigned long); - -extern struct inode *isofs_iget(struct super_block *sb, - unsigned long block, - unsigned long offset); - -/* Because the inode number is no longer relevant to finding the - * underlying meta-data for an inode, we are free to choose a more - * convenient 32-bit number as the inode number. The inode numbering - * scheme was recommended by Sergey Vlasov and Eric Lammerts. */ -static inline unsigned long isofs_get_ino(unsigned long block, - unsigned long offset, - unsigned long bufbits) -{ - return (block << (bufbits - 5)) | (offset >> 5); -} - -/* Every directory can have many redundant directory entries scattered - * throughout the directory tree. First there is the directory entry - * with the name of the directory stored in the parent directory. - * Then, there is the "." directory entry stored in the directory - * itself. Finally, there are possibly many ".." directory entries - * stored in all the subdirectories. - * - * In order for the NFS get_parent() method to work and for the - * general consistency of the dcache, we need to make sure the - * "i_iget5_block" and "i_iget5_offset" all point to exactly one of - * the many redundant entries for each directory. We normalize the - * block and offset by always making them point to the "." directory. - * - * Notice that we do not use the entry for the directory with the name - * that is located in the parent directory. Even though choosing this - * first directory is more natural, it is much easier to find the "." - * entry in the NFS get_parent() method because it is implicitly - * encoded in the "extent + ext_attr_length" fields of _all_ the - * redundant entries for the directory. Thus, it can always be - * reached regardless of which directory entry you have in hand. - * - * This works because the "." entry is simply the first directory - * record when you start reading the file that holds all the directory - * records, and this file starts at "extent + ext_attr_length" blocks. - * Because the "." entry is always the first entry listed in the - * directories file, the normalized "offset" value is always 0. - * - * You should pass the directory entry in "de". On return, "block" - * and "offset" will hold normalized values. Only directories are - * affected making it safe to call even for non-directory file - * types. */ -static inline void -isofs_normalize_block_and_offset(struct iso_directory_record* de, - unsigned long *block, - unsigned long *offset) -{ - /* Only directories are normalized. */ - if (de->flags[0] & 2) { - *offset = 0; - *block = (unsigned long)isonum_733(de->extent) - + (unsigned long)isonum_711(de->ext_attr_length); - } -} - -extern struct inode_operations isofs_dir_inode_operations; -extern struct file_operations isofs_dir_operations; -extern struct address_space_operations isofs_symlink_aops; -extern struct export_operations isofs_export_ops; - -/* The following macros are used to check for memory leaks. */ -#ifdef LEAK_CHECK -#define free_s leak_check_free_s -#define malloc leak_check_malloc -#define sb_bread leak_check_bread -#define brelse leak_check_brelse -extern void * leak_check_malloc(unsigned int size); -extern void leak_check_free_s(void * obj, int size); -extern struct buffer_head * leak_check_bread(struct super_block *sb, int block); -extern void leak_check_brelse(struct buffer_head * bh); -#endif /* LEAK_CHECK */ - -#endif /* __KERNEL__ */ - #endif diff --git a/include/linux/iso_fs_i.h b/include/linux/iso_fs_i.h deleted file mode 100644 index 59065e939ea..00000000000 --- a/include/linux/iso_fs_i.h +++ /dev/null @@ -1,27 +0,0 @@ -#ifndef _ISO_FS_I -#define _ISO_FS_I - -#include <linux/fs.h> - -enum isofs_file_format { - isofs_file_normal = 0, - isofs_file_sparse = 1, - isofs_file_compressed = 2, -}; - -/* - * iso fs inode data in memory - */ -struct iso_inode_info { - unsigned long i_iget5_block; - unsigned long i_iget5_offset; - unsigned int i_first_extent; - unsigned char i_file_format; - unsigned char i_format_parm[3]; - unsigned long i_next_section_block; - unsigned long i_next_section_offset; - off_t i_section_size; - struct inode vfs_inode; -}; - -#endif diff --git a/include/linux/iso_fs_sb.h b/include/linux/iso_fs_sb.h deleted file mode 100644 index 043b97b55b8..00000000000 --- a/include/linux/iso_fs_sb.h +++ /dev/null @@ -1,34 +0,0 @@ -#ifndef _ISOFS_FS_SB -#define _ISOFS_FS_SB - -/* - * iso9660 super-block data in memory - */ -struct isofs_sb_info { - unsigned long s_ninodes; - unsigned long s_nzones; - unsigned long s_firstdatazone; - unsigned long s_log_zone_size; - unsigned long s_max_size; - - unsigned char s_high_sierra; /* A simple flag */ - unsigned char s_mapping; - int s_rock_offset; /* offset of SUSP fields within SU area */ - unsigned char s_rock; - unsigned char s_joliet_level; - unsigned char s_utf8; - unsigned char s_cruft; /* Broken disks with high - byte of length containing - junk */ - unsigned char s_unhide; - unsigned char s_nosuid; - unsigned char s_nodev; - unsigned char s_nocompress; - - mode_t s_mode; - gid_t s_gid; - uid_t s_uid; - struct nls_table *s_nls_iocharset; /* Native language support table */ -}; - -#endif diff --git a/include/linux/ixjuser.h b/include/linux/ixjuser.h index 88121166d71..fd1756d3a47 100644 --- a/include/linux/ixjuser.h +++ b/include/linux/ixjuser.h @@ -42,8 +42,6 @@ * *****************************************************************************/ -static char ixjuser_h_rcsid[] = "$Id: ixjuser.h,v 4.1 2001/08/05 00:17:37 craigs Exp $"; - #include <linux/telephony.h> diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 7c1cba4a527..e25b97062ce 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -115,10 +115,19 @@ extern int __kernel_text_address(unsigned long addr); extern int kernel_text_address(unsigned long addr); extern int session_of_pgrp(int pgrp); +#ifdef CONFIG_PRINTK asmlinkage int vprintk(const char *fmt, va_list args) __attribute__ ((format (printf, 1, 0))); asmlinkage int printk(const char * fmt, ...) __attribute__ ((format (printf, 1, 2))); +#else +static inline int vprintk(const char *s, va_list args) + __attribute__ ((format (printf, 1, 0))); +static inline int vprintk(const char *s, va_list args) { return 0; } +static inline int printk(const char *s, ...) + __attribute__ ((format (printf, 1, 2))); +static inline int printk(const char *s, ...) { return 0; } +#endif unsigned long int_sqrt(unsigned long); diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index f20c163de4f..99ddba5a4e0 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -43,6 +43,9 @@ typedef int (*kprobe_fault_handler_t) (struct kprobe *, struct pt_regs *, struct kprobe { struct hlist_node hlist; + /* list of kprobes for multi-handler support */ + struct list_head list; + /* location of the probe point */ kprobe_opcode_t *addr; diff --git a/include/linux/mm.h b/include/linux/mm.h index c74a74ca401..17518fe0b31 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -637,9 +637,9 @@ extern unsigned long do_mremap(unsigned long addr, * These functions are passed a count `nr_to_scan' and a gfpmask. They should * scan `nr_to_scan' objects, attempting to free them. * - * The callback must the number of objects which remain in the cache. + * The callback must return the number of objects which remain in the cache. * - * The callback will be passes nr_to_scan == 0 when the VM is querying the + * The callback will be passed nr_to_scan == 0 when the VM is querying the * cache size, so a fastpath for that case is appropriate. */ typedef int (*shrinker_t)(int nr_to_scan, unsigned int gfp_mask); @@ -726,6 +726,7 @@ extern void __vma_link_rb(struct mm_struct *, struct vm_area_struct *, extern struct vm_area_struct *copy_vma(struct vm_area_struct **, unsigned long addr, unsigned long len, pgoff_t pgoff); extern void exit_mmap(struct mm_struct *); +extern int may_expand_vm(struct mm_struct *mm, unsigned long npages); extern unsigned long get_unmapped_area(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); diff --git a/include/linux/mpage.h b/include/linux/mpage.h index dea1b008366..3ca880463c4 100644 --- a/include/linux/mpage.h +++ b/include/linux/mpage.h @@ -20,9 +20,6 @@ int mpage_writepages(struct address_space *mapping, struct writeback_control *wbc, get_block_t get_block); int mpage_writepage(struct page *page, get_block_t *get_block, struct writeback_control *wbc); -int __mpage_writepages(struct address_space *mapping, - struct writeback_control *wbc, get_block_t get_block, - writepage_t writepage); static inline int generic_writepages(struct address_space *mapping, struct writeback_control *wbc) diff --git a/include/linux/net.h b/include/linux/net.h index 7823b348250..6d997ff3f10 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -7,7 +7,7 @@ * Version: @(#)net.h 1.0.3 05/25/93 * * Authors: Orest Zborowski, <obz@Kodak.COM> - * Ross Biro, <bir7@leland.Stanford.Edu> + * Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * This program is free software; you can redistribute it and/or @@ -64,19 +64,19 @@ typedef enum { #define SOCK_PASSCRED 3 #ifndef ARCH_HAS_SOCKET_TYPES -/** sock_type - Socket types - * +/** + * enum sock_type - Socket types + * @SOCK_STREAM: stream (connection) socket + * @SOCK_DGRAM: datagram (conn.less) socket + * @SOCK_RAW: raw socket + * @SOCK_RDM: reliably-delivered message + * @SOCK_SEQPACKET: sequential packet socket + * @SOCK_PACKET: linux specific way of getting packets at the dev level. + * For writing rarp and other similar things on the user level. + * * When adding some new socket type please * grep ARCH_HAS_SOCKET_TYPE include/asm-* /socket.h, at least MIPS * overrides this enum for binary compat reasons. - * - * @SOCK_STREAM - stream (connection) socket - * @SOCK_DGRAM - datagram (conn.less) socket - * @SOCK_RAW - raw socket - * @SOCK_RDM - reliably-delivered message - * @SOCK_SEQPACKET - sequential packet socket - * @SOCK_PACKET - linux specific way of getting packets at the dev level. - * For writing rarp and other similar things on the user level. */ enum sock_type { SOCK_STREAM = 1, @@ -93,15 +93,15 @@ enum sock_type { /** * struct socket - general BSD socket - * @state - socket state (%SS_CONNECTED, etc) - * @flags - socket flags (%SOCK_ASYNC_NOSPACE, etc) - * @ops - protocol specific socket operations - * @fasync_list - Asynchronous wake up list - * @file - File back pointer for gc - * @sk - internal networking protocol agnostic socket representation - * @wait - wait queue for several uses - * @type - socket type (%SOCK_STREAM, etc) - * @passcred - credentials (used only in Unix Sockets (aka PF_LOCAL)) + * @state: socket state (%SS_CONNECTED, etc) + * @flags: socket flags (%SOCK_ASYNC_NOSPACE, etc) + * @ops: protocol specific socket operations + * @fasync_list: Asynchronous wake up list + * @file: File back pointer for gc + * @sk: internal networking protocol agnostic socket representation + * @wait: wait queue for several uses + * @type: socket type (%SOCK_STREAM, etc) + * @passcred: credentials (used only in Unix Sockets (aka PF_LOCAL)) */ struct socket { socket_state state; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 8d775be6783..ac11d73be4c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -7,7 +7,7 @@ * * Version: @(#)dev.h 1.0.10 08/12/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Corey Minyard <wf-rch!minyard@relay.EU.net> * Donald J. Becker, <becker@cesdis.gsfc.nasa.gov> diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h index c9bacf9b243..9e5750079e0 100644 --- a/include/linux/netfilter_ipv4.h +++ b/include/linux/netfilter_ipv4.h @@ -62,6 +62,9 @@ enum nf_ip_hook_priorities { NF_IP_PRI_FILTER = 0, NF_IP_PRI_NAT_SRC = 100, NF_IP_PRI_SELINUX_LAST = 225, + NF_IP_PRI_CONNTRACK_HELPER = INT_MAX - 2, + NF_IP_PRI_NAT_SEQ_ADJUST = INT_MAX - 1, + NF_IP_PRI_CONNTRACK_CONFIRM = INT_MAX, NF_IP_PRI_LAST = INT_MAX, }; diff --git a/include/linux/netlink.h b/include/linux/netlink.h index f731abdc1a2..b2738ac8bc9 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -110,6 +110,7 @@ struct netlink_skb_parms __u32 dst_pid; __u32 dst_groups; kernel_cap_t eff_cap; + __u32 loginuid; /* Login (audit) uid */ }; #define NETLINK_CB(skb) (*(struct netlink_skb_parms*)&((skb)->cb)) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 6b74fcf5bb6..39ab8c6b565 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -131,6 +131,7 @@ struct page_state { unsigned long allocstall; /* direct reclaim calls */ unsigned long pgrotated; /* pages rotated to tail of the LRU */ + unsigned long nr_bounce; /* pages for bounce buffers */ }; extern void get_page_state(struct page_state *ret); diff --git a/include/linux/patchkey.h b/include/linux/patchkey.h new file mode 100644 index 00000000000..d974a6e9237 --- /dev/null +++ b/include/linux/patchkey.h @@ -0,0 +1,45 @@ +/* + * <linux/patchkey.h> -- definition of _PATCHKEY macro + * + * Copyright (C) 2005 Stuart Brady + * + * This exists because awe_voice.h defined its own _PATCHKEY and it wasn't + * clear whether removing this would break anything in userspace. + * + * Do not include this file directly. Please use <sys/soundcard.h> instead. + * For kernel code, use <linux/soundcard.h> + */ + +#ifndef _LINUX_PATCHKEY_H_INDIRECT +#error "patchkey.h included directly" +#endif + +#ifndef _LINUX_PATCHKEY_H +#define _LINUX_PATCHKEY_H + +/* Endian macros. */ +#ifdef __KERNEL__ +# include <asm/byteorder.h> +#else +# include <endian.h> +#endif + +#if defined(__KERNEL__) +# if defined(__BIG_ENDIAN) +# define _PATCHKEY(id) (0xfd00|id) +# elif defined(__LITTLE_ENDIAN) +# define _PATCHKEY(id) ((id<<8)|0x00fd) +# else +# error "could not determine byte order" +# endif +#elif defined(__BYTE_ORDER) +# if __BYTE_ORDER == __BIG_ENDIAN +# define _PATCHKEY(id) (0xfd00|id) +# elif __BYTE_ORDER == __LITTLE_ENDIAN +# define _PATCHKEY(id) ((id<<8)|0x00fd) +# else +# error "could not determine byte order" +# endif +#endif + +#endif /* _LINUX_PATCHKEY_H */ diff --git a/include/linux/pci.h b/include/linux/pci.h index 3c89148ae28..b5238bd1883 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -671,6 +671,7 @@ struct pci_driver { int (*suspend) (struct pci_dev *dev, pm_message_t state); /* Device suspended */ int (*resume) (struct pci_dev *dev); /* Device woken up */ int (*enable_wake) (struct pci_dev *dev, pci_power_t state, int enable); /* Enable wake event */ + void (*shutdown) (struct pci_dev *dev); struct device_driver driver; struct pci_dynids dynids; @@ -810,7 +811,6 @@ void pci_set_master(struct pci_dev *dev); int pci_set_mwi(struct pci_dev *dev); void pci_clear_mwi(struct pci_dev *dev); int pci_set_dma_mask(struct pci_dev *dev, u64 mask); -int pci_dac_set_dma_mask(struct pci_dev *dev, u64 mask); int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask); int pci_assign_resource(struct pci_dev *dev, int i); @@ -941,7 +941,6 @@ static inline void pci_set_master(struct pci_dev *dev) { } static inline int pci_enable_device(struct pci_dev *dev) { return -EIO; } static inline void pci_disable_device(struct pci_dev *dev) { } static inline int pci_set_dma_mask(struct pci_dev *dev, u64 mask) { return -EIO; } -static inline int pci_dac_set_dma_mask(struct pci_dev *dev, u64 mask) { return -EIO; } static inline int pci_assign_resource(struct pci_dev *dev, int i) { return -EBUSY;} static inline int pci_register_driver(struct pci_driver *drv) { return 0;} static inline void pci_unregister_driver(struct pci_driver *drv) { } diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f1f75fde8cd..ae27792b5aa 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -854,6 +854,7 @@ #define PCI_DEVICE_ID_MYLEX_DAC960_LA 0x0020 #define PCI_DEVICE_ID_MYLEX_DAC960_LP 0x0050 #define PCI_DEVICE_ID_MYLEX_DAC960_BA 0xBA56 +#define PCI_DEVICE_ID_MYLEX_DAC960_GEM 0xB166 #define PCI_VENDOR_ID_PICOP 0x1066 #define PCI_DEVICE_ID_PICOP_PT86C52X 0x0001 @@ -2062,6 +2063,7 @@ #define PCI_DEVICE_ID_AFAVLAB_P030 0x2182 #define PCI_VENDOR_ID_BROADCOM 0x14e4 +#define PCI_DEVICE_ID_TIGON3_5752 0x1600 #define PCI_DEVICE_ID_TIGON3_5700 0x1644 #define PCI_DEVICE_ID_TIGON3_5701 0x1645 #define PCI_DEVICE_ID_TIGON3_5702 0x1646 @@ -2414,6 +2416,8 @@ #define PCI_DEVICE_ID_INTEL_ICH7_1 0x27b9 #define PCI_DEVICE_ID_INTEL_ICH7_2 0x27c0 #define PCI_DEVICE_ID_INTEL_ICH7_3 0x27c1 +#define PCI_DEVICE_ID_INTEL_ICH7_30 0x27b0 +#define PCI_DEVICE_ID_INTEL_ICH7_31 0x27bd #define PCI_DEVICE_ID_INTEL_ICH7_5 0x27c4 #define PCI_DEVICE_ID_INTEL_ICH7_6 0x27c5 #define PCI_DEVICE_ID_INTEL_ICH7_7 0x27c8 diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 4d747433916..fd276adf0fd 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -157,9 +157,9 @@ static inline int rcu_pending(int cpu) /** * rcu_read_lock - mark the beginning of an RCU read-side critical section. * - * When synchronize_kernel() is invoked on one CPU while other CPUs + * When synchronize_rcu() is invoked on one CPU while other CPUs * are within RCU read-side critical sections, then the - * synchronize_kernel() is guaranteed to block until after all the other + * synchronize_rcu() is guaranteed to block until after all the other * CPUs exit their critical sections. Similarly, if call_rcu() is invoked * on one CPU while other CPUs are within RCU read-side critical * sections, invocation of the corresponding RCU callback is deferred @@ -256,6 +256,21 @@ static inline int rcu_pending(int cpu) (p) = (v); \ }) +/** + * synchronize_sched - block until all CPUs have exited any non-preemptive + * kernel code sequences. + * + * This means that all preempt_disable code sequences, including NMI and + * hardware-interrupt handlers, in progress on entry will have completed + * before this primitive returns. However, this does not guarantee that + * softirq handlers will have completed, since in some kernels + * + * This primitive provides the guarantees made by the (deprecated) + * synchronize_kernel() API. In contrast, synchronize_rcu() only + * guarantees that rcu_read_lock() sections will have completed. + */ +#define synchronize_sched() synchronize_rcu() + extern void rcu_init(void); extern void rcu_check_callbacks(int cpu, int user); extern void rcu_restart_cpu(int cpu); @@ -265,7 +280,9 @@ extern void FASTCALL(call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *head))); extern void FASTCALL(call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *head))); -extern void synchronize_kernel(void); +extern __deprecated_for_modules void synchronize_kernel(void); +extern void synchronize_rcu(void); +void synchronize_idle(void); #endif /* __KERNEL__ */ #endif /* __LINUX_RCUPDATE_H */ diff --git a/include/linux/reboot_fixups.h b/include/linux/reboot_fixups.h new file mode 100644 index 00000000000..480ea2d489d --- /dev/null +++ b/include/linux/reboot_fixups.h @@ -0,0 +1,10 @@ +#ifndef _LINUX_REBOOT_FIXUPS_H +#define _LINUX_REBOOT_FIXUPS_H + +#ifdef CONFIG_X86_REBOOTFIXUPS +extern void mach_reboot_fixups(void); +#else +#define mach_reboot_fixups() ((void)(0)) +#endif + +#endif /* _LINUX_REBOOT_FIXUPS_H */ diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h index a57e973af0b..2aef9c3f5ce 100644 --- a/include/linux/reiserfs_acl.h +++ b/include/linux/reiserfs_acl.h @@ -5,18 +5,18 @@ #define REISERFS_ACL_VERSION 0x0001 typedef struct { - __u16 e_tag; - __u16 e_perm; - __u32 e_id; + __le16 e_tag; + __le16 e_perm; + __le32 e_id; } reiserfs_acl_entry; typedef struct { - __u16 e_tag; - __u16 e_perm; + __le16 e_tag; + __le16 e_perm; } reiserfs_acl_entry_short; typedef struct { - __u32 a_version; + __le32 a_version; } reiserfs_acl_header; static inline size_t reiserfs_acl_size(int count) diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index bccff8b17dc..32148625fc2 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -114,47 +114,47 @@ if( !( cond ) ) \ struct journal_params { - __u32 jp_journal_1st_block; /* where does journal start from on its + __le32 jp_journal_1st_block; /* where does journal start from on its * device */ - __u32 jp_journal_dev; /* journal device st_rdev */ - __u32 jp_journal_size; /* size of the journal */ - __u32 jp_journal_trans_max; /* max number of blocks in a transaction. */ - __u32 jp_journal_magic; /* random value made on fs creation (this + __le32 jp_journal_dev; /* journal device st_rdev */ + __le32 jp_journal_size; /* size of the journal */ + __le32 jp_journal_trans_max; /* max number of blocks in a transaction. */ + __le32 jp_journal_magic; /* random value made on fs creation (this * was sb_journal_block_count) */ - __u32 jp_journal_max_batch; /* max number of blocks to batch into a + __le32 jp_journal_max_batch; /* max number of blocks to batch into a * trans */ - __u32 jp_journal_max_commit_age; /* in seconds, how old can an async + __le32 jp_journal_max_commit_age; /* in seconds, how old can an async * commit be */ - __u32 jp_journal_max_trans_age; /* in seconds, how old can a transaction + __le32 jp_journal_max_trans_age; /* in seconds, how old can a transaction * be */ }; /* this is the super from 3.5.X, where X >= 10 */ struct reiserfs_super_block_v1 { - __u32 s_block_count; /* blocks count */ - __u32 s_free_blocks; /* free blocks count */ - __u32 s_root_block; /* root block number */ + __le32 s_block_count; /* blocks count */ + __le32 s_free_blocks; /* free blocks count */ + __le32 s_root_block; /* root block number */ struct journal_params s_journal; - __u16 s_blocksize; /* block size */ - __u16 s_oid_maxsize; /* max size of object id array, see + __le16 s_blocksize; /* block size */ + __le16 s_oid_maxsize; /* max size of object id array, see * get_objectid() commentary */ - __u16 s_oid_cursize; /* current size of object id array */ - __u16 s_umount_state; /* this is set to 1 when filesystem was + __le16 s_oid_cursize; /* current size of object id array */ + __le16 s_umount_state; /* this is set to 1 when filesystem was * umounted, to 2 - when not */ char s_magic[10]; /* reiserfs magic string indicates that * file system is reiserfs: * "ReIsErFs" or "ReIsEr2Fs" or "ReIsEr3Fs" */ - __u16 s_fs_state; /* it is set to used by fsck to mark which + __le16 s_fs_state; /* it is set to used by fsck to mark which * phase of rebuilding is done */ - __u32 s_hash_function_code; /* indicate, what hash function is being use + __le32 s_hash_function_code; /* indicate, what hash function is being use * to sort names in a directory*/ - __u16 s_tree_height; /* height of disk tree */ - __u16 s_bmap_nr; /* amount of bitmap blocks needed to address + __le16 s_tree_height; /* height of disk tree */ + __le16 s_bmap_nr; /* amount of bitmap blocks needed to address * each block of file system */ - __u16 s_version; /* this field is only reliable on filesystem + __le16 s_version; /* this field is only reliable on filesystem * with non-standard journal */ - __u16 s_reserved_for_journal; /* size in blocks of journal area on main + __le16 s_reserved_for_journal; /* size in blocks of journal area on main * device, we need to keep after * making fs with non-standard journal */ } __attribute__ ((__packed__)); @@ -165,8 +165,8 @@ struct reiserfs_super_block_v1 struct reiserfs_super_block { struct reiserfs_super_block_v1 s_v1; - __u32 s_inode_generation; - __u32 s_flags; /* Right now used only by inode-attributes, if enabled */ + __le32 s_inode_generation; + __le32 s_flags; /* Right now used only by inode-attributes, if enabled */ unsigned char s_uuid[16]; /* filesystem unique identifier */ unsigned char s_label[16]; /* filesystem volume label */ char s_unused[88] ; /* zero filled by mkreiserfs and @@ -225,7 +225,7 @@ struct reiserfs_super_block #define SB_ONDISK_JOURNAL_DEVICE(s) \ le32_to_cpu ((SB_ONDISK_JP(s)->jp_journal_dev)) #define SB_ONDISK_RESERVED_FOR_JOURNAL(s) \ - le32_to_cpu ((SB_V1_DISK_SUPER_BLOCK(s)->s_reserved_for_journal)) + le16_to_cpu ((SB_V1_DISK_SUPER_BLOCK(s)->s_reserved_for_journal)) #define is_block_in_log_or_reserved_area(s, block) \ block >= SB_JOURNAL_1st_RESERVED_BLOCK(s) \ @@ -269,7 +269,7 @@ int is_reiserfs_jr (struct reiserfs_super_block * rs); #define QUOTA_EXCEEDED -6 typedef __u32 b_blocknr_t; -typedef __u32 unp_t; +typedef __le32 unp_t; struct unfm_nodeinfo { unp_t unfm_nodenum; @@ -376,78 +376,57 @@ static inline struct reiserfs_sb_info *REISERFS_SB(const struct super_block *sb) // directories use this key as well as old files // struct offset_v1 { - __u32 k_offset; - __u32 k_uniqueness; + __le32 k_offset; + __le32 k_uniqueness; } __attribute__ ((__packed__)); struct offset_v2 { -#ifdef __LITTLE_ENDIAN - /* little endian version */ - __u64 k_offset:60; - __u64 k_type: 4; -#else - /* big endian version */ - __u64 k_type: 4; - __u64 k_offset:60; -#endif + __le64 v; } __attribute__ ((__packed__)); -#ifndef __LITTLE_ENDIAN -typedef union { - struct offset_v2 offset_v2; - __u64 linear; -} __attribute__ ((__packed__)) offset_v2_esafe_overlay; - static inline __u16 offset_v2_k_type( const struct offset_v2 *v2 ) { - offset_v2_esafe_overlay tmp = *(const offset_v2_esafe_overlay *)v2; - tmp.linear = le64_to_cpu( tmp.linear ); - return (tmp.offset_v2.k_type <= TYPE_MAXTYPE)?tmp.offset_v2.k_type:TYPE_ANY; + __u8 type = le64_to_cpu(v2->v) >> 60; + return (type <= TYPE_MAXTYPE)?type:TYPE_ANY; } static inline void set_offset_v2_k_type( struct offset_v2 *v2, int type ) { - offset_v2_esafe_overlay *tmp = (offset_v2_esafe_overlay *)v2; - tmp->linear = le64_to_cpu(tmp->linear); - tmp->offset_v2.k_type = type; - tmp->linear = cpu_to_le64(tmp->linear); + v2->v = (v2->v & cpu_to_le64(~0ULL>>4)) | cpu_to_le64((__u64)type<<60); } static inline loff_t offset_v2_k_offset( const struct offset_v2 *v2 ) { - offset_v2_esafe_overlay tmp = *(const offset_v2_esafe_overlay *)v2; - tmp.linear = le64_to_cpu( tmp.linear ); - return tmp.offset_v2.k_offset; + return le64_to_cpu(v2->v) & (~0ULL>>4); } static inline void set_offset_v2_k_offset( struct offset_v2 *v2, loff_t offset ){ - offset_v2_esafe_overlay *tmp = (offset_v2_esafe_overlay *)v2; - tmp->linear = le64_to_cpu(tmp->linear); - tmp->offset_v2.k_offset = offset; - tmp->linear = cpu_to_le64(tmp->linear); + offset &= (~0ULL>>4); + v2->v = (v2->v & cpu_to_le64(15ULL<<60)) | cpu_to_le64(offset); } -#else -# define offset_v2_k_type(v2) ((v2)->k_type) -# define set_offset_v2_k_type(v2,val) (offset_v2_k_type(v2) = (val)) -# define offset_v2_k_offset(v2) ((v2)->k_offset) -# define set_offset_v2_k_offset(v2,val) (offset_v2_k_offset(v2) = (val)) -#endif /* Key of an item determines its location in the S+tree, and is composed of 4 components */ struct reiserfs_key { - __u32 k_dir_id; /* packing locality: by default parent + __le32 k_dir_id; /* packing locality: by default parent directory object id */ - __u32 k_objectid; /* object identifier */ + __le32 k_objectid; /* object identifier */ union { struct offset_v1 k_offset_v1; struct offset_v2 k_offset_v2; } __attribute__ ((__packed__)) u; } __attribute__ ((__packed__)); +struct in_core_key { + __u32 k_dir_id; /* packing locality: by default parent + directory object id */ + __u32 k_objectid; /* object identifier */ + __u64 k_offset; + __u8 k_type; +}; struct cpu_key { - struct reiserfs_key on_disk_key; + struct in_core_key on_disk_key; int version; int key_length; /* 3 in all cases but direct2indirect and indirect2direct conversion */ @@ -508,15 +487,15 @@ struct item_head item. Note that the key, not this field, is used to determine the item type, and thus which field this union contains. */ - __u16 ih_free_space_reserved; + __le16 ih_free_space_reserved; /* Iff this is a directory item, this field equals the number of directory entries in the directory item. */ - __u16 ih_entry_count; + __le16 ih_entry_count; } __attribute__ ((__packed__)) u; - __u16 ih_item_len; /* total size of the item body */ - __u16 ih_item_location; /* an offset to the item body + __le16 ih_item_len; /* total size of the item body */ + __le16 ih_item_location; /* an offset to the item body * within the block */ - __u16 ih_version; /* 0 for all old items, 2 for new + __le16 ih_version; /* 0 for all old items, 2 for new ones. Highest bit is set by fsck temporary, cleaned after all done */ @@ -670,43 +649,29 @@ static inline void set_le_ih_k_type (struct item_head * ih, int type) // static inline loff_t cpu_key_k_offset (const struct cpu_key * key) { - return (key->version == KEY_FORMAT_3_5) ? - key->on_disk_key.u.k_offset_v1.k_offset : - key->on_disk_key.u.k_offset_v2.k_offset; + return key->on_disk_key.k_offset; } static inline loff_t cpu_key_k_type (const struct cpu_key * key) { - return (key->version == KEY_FORMAT_3_5) ? - uniqueness2type (key->on_disk_key.u.k_offset_v1.k_uniqueness) : - key->on_disk_key.u.k_offset_v2.k_type; + return key->on_disk_key.k_type; } static inline void set_cpu_key_k_offset (struct cpu_key * key, loff_t offset) { - (key->version == KEY_FORMAT_3_5) ? - (key->on_disk_key.u.k_offset_v1.k_offset = offset) : - (key->on_disk_key.u.k_offset_v2.k_offset = offset); + key->on_disk_key.k_offset = offset; } - static inline void set_cpu_key_k_type (struct cpu_key * key, int type) { - (key->version == KEY_FORMAT_3_5) ? - (key->on_disk_key.u.k_offset_v1.k_uniqueness = type2uniqueness (type)): - (key->on_disk_key.u.k_offset_v2.k_type = type); + key->on_disk_key.k_type = type; } - static inline void cpu_key_k_offset_dec (struct cpu_key * key) { - if (key->version == KEY_FORMAT_3_5) - key->on_disk_key.u.k_offset_v1.k_offset --; - else - key->on_disk_key.u.k_offset_v2.k_offset --; + key->on_disk_key.k_offset --; } - #define is_direntry_cpu_key(key) (cpu_key_k_type (key) == TYPE_DIRENTRY) #define is_direct_cpu_key(key) (cpu_key_k_type (key) == TYPE_DIRECT) #define is_indirect_cpu_key(key) (cpu_key_k_type (key) == TYPE_INDIRECT) @@ -752,10 +717,10 @@ extern struct reiserfs_key root_key; /* Header of a disk block. More precisely, header of a formatted leaf or internal node, and not the header of an unformatted node. */ struct block_head { - __u16 blk_level; /* Level of a block in the tree. */ - __u16 blk_nr_item; /* Number of keys/items in a block. */ - __u16 blk_free_space; /* Block free space in bytes. */ - __u16 blk_reserved; + __le16 blk_level; /* Level of a block in the tree. */ + __le16 blk_nr_item; /* Number of keys/items in a block. */ + __le16 blk_free_space; /* Block free space in bytes. */ + __le16 blk_reserved; /* dump this in v4/planA */ struct reiserfs_key blk_right_delim_key; /* kept only for compatibility */ }; @@ -819,19 +784,19 @@ struct block_head { // struct stat_data_v1 { - __u16 sd_mode; /* file type, permissions */ - __u16 sd_nlink; /* number of hard links */ - __u16 sd_uid; /* owner */ - __u16 sd_gid; /* group */ - __u32 sd_size; /* file size */ - __u32 sd_atime; /* time of last access */ - __u32 sd_mtime; /* time file was last modified */ - __u32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ + __le16 sd_mode; /* file type, permissions */ + __le16 sd_nlink; /* number of hard links */ + __le16 sd_uid; /* owner */ + __le16 sd_gid; /* group */ + __le32 sd_size; /* file size */ + __le32 sd_atime; /* time of last access */ + __le32 sd_mtime; /* time file was last modified */ + __le32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ union { - __u32 sd_rdev; - __u32 sd_blocks; /* number of blocks file uses */ + __le32 sd_rdev; + __le32 sd_blocks; /* number of blocks file uses */ } __attribute__ ((__packed__)) u; - __u32 sd_first_direct_byte; /* first byte of file which is stored + __le32 sd_first_direct_byte; /* first byte of file which is stored in a direct item: except that if it equals 1 it is a symlink and if it equals ~(__u32)0 there is no @@ -897,20 +862,20 @@ struct stat_data_v1 /* Stat Data on disk (reiserfs version of UFS disk inode minus the address blocks) */ struct stat_data { - __u16 sd_mode; /* file type, permissions */ - __u16 sd_attrs; /* persistent inode flags */ - __u32 sd_nlink; /* number of hard links */ - __u64 sd_size; /* file size */ - __u32 sd_uid; /* owner */ - __u32 sd_gid; /* group */ - __u32 sd_atime; /* time of last access */ - __u32 sd_mtime; /* time file was last modified */ - __u32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ - __u32 sd_blocks; + __le16 sd_mode; /* file type, permissions */ + __le16 sd_attrs; /* persistent inode flags */ + __le32 sd_nlink; /* number of hard links */ + __le64 sd_size; /* file size */ + __le32 sd_uid; /* owner */ + __le32 sd_gid; /* group */ + __le32 sd_atime; /* time of last access */ + __le32 sd_mtime; /* time file was last modified */ + __le32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ + __le32 sd_blocks; union { - __u32 sd_rdev; - __u32 sd_generation; - //__u32 sd_first_direct_byte; + __le32 sd_rdev; + __le32 sd_generation; + //__le32 sd_first_direct_byte; /* first byte of file which is stored in a direct item: except that if it equals 1 it is a symlink and if it equals @@ -993,12 +958,12 @@ struct stat_data { struct reiserfs_de_head { - __u32 deh_offset; /* third component of the directory entry key */ - __u32 deh_dir_id; /* objectid of the parent directory of the object, that is referenced + __le32 deh_offset; /* third component of the directory entry key */ + __le32 deh_dir_id; /* objectid of the parent directory of the object, that is referenced by directory entry */ - __u32 deh_objectid; /* objectid of the object, that is referenced by directory entry */ - __u16 deh_location; /* offset of name in the whole item */ - __u16 deh_state; /* whether 1) entry contains stat data (for future), and 2) whether + __le32 deh_objectid; /* objectid of the object, that is referenced by directory entry */ + __le16 deh_location; /* offset of name in the whole item */ + __le16 deh_state; /* whether 1) entry contains stat data (for future), and 2) whether entry is hidden (unlinked) */ } __attribute__ ((__packed__)); #define DEH_SIZE sizeof(struct reiserfs_de_head) @@ -1058,10 +1023,10 @@ struct reiserfs_de_head #define de_visible(deh) test_bit_unaligned (DEH_Visible, &((deh)->deh_state)) #define de_hidden(deh) !test_bit_unaligned (DEH_Visible, &((deh)->deh_state)) -extern void make_empty_dir_item_v1 (char * body, __u32 dirid, __u32 objid, - __u32 par_dirid, __u32 par_objid); -extern void make_empty_dir_item (char * body, __u32 dirid, __u32 objid, - __u32 par_dirid, __u32 par_objid); +extern void make_empty_dir_item_v1 (char * body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid); +extern void make_empty_dir_item (char * body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid); /* array of the entry headers */ /* get item body */ @@ -1160,9 +1125,9 @@ struct reiserfs_dir_entry /* Disk child pointer: The pointer from an internal node of the tree to a node that is on disk. */ struct disk_child { - __u32 dc_block_number; /* Disk child's block number. */ - __u16 dc_size; /* Disk child's used space. */ - __u16 dc_reserved; + __le32 dc_block_number; /* Disk child's block number. */ + __le16 dc_size; /* Disk child's used space. */ + __le16 dc_reserved; }; #define DC_SIZE (sizeof(struct disk_child)) @@ -1476,7 +1441,7 @@ struct tree_balance int fs_gen; /* saved value of `reiserfs_generation' counter see FILESYSTEM_CHANGED() macro in reiserfs_fs.h */ #ifdef DISPLACE_NEW_PACKING_LOCALITIES - struct reiserfs_key key; /* key pointer, to pass to block allocator or + struct in_core_key key; /* key pointer, to pass to block allocator or another low-level subsystem */ #endif } ; @@ -1630,10 +1595,10 @@ struct reiserfs_iget_args { /* first block written in a commit. */ struct reiserfs_journal_desc { - __u32 j_trans_id ; /* id of commit */ - __u32 j_len ; /* length of commit. len +1 is the commit block */ - __u32 j_mount_id ; /* mount id of this trans*/ - __u32 j_realblock[1] ; /* real locations for each block */ + __le32 j_trans_id ; /* id of commit */ + __le32 j_len ; /* length of commit. len +1 is the commit block */ + __le32 j_mount_id ; /* mount id of this trans*/ + __le32 j_realblock[1] ; /* real locations for each block */ } ; #define get_desc_trans_id(d) le32_to_cpu((d)->j_trans_id) @@ -1646,9 +1611,9 @@ struct reiserfs_journal_desc { /* last block written in a commit */ struct reiserfs_journal_commit { - __u32 j_trans_id ; /* must match j_trans_id from the desc block */ - __u32 j_len ; /* ditto */ - __u32 j_realblock[1] ; /* real locations for each block */ + __le32 j_trans_id ; /* must match j_trans_id from the desc block */ + __le32 j_len ; /* ditto */ + __le32 j_realblock[1] ; /* real locations for each block */ } ; #define get_commit_trans_id(c) le32_to_cpu((c)->j_trans_id) @@ -1663,9 +1628,9 @@ struct reiserfs_journal_commit { ** and this transaction does not need to be replayed. */ struct reiserfs_journal_header { - __u32 j_last_flush_trans_id ; /* id of last fully flushed transaction */ - __u32 j_first_unflushed_offset ; /* offset in the log of where to start replay after a crash */ - __u32 j_mount_id ; + __le32 j_last_flush_trans_id ; /* id of last fully flushed transaction */ + __le32 j_first_unflushed_offset ; /* offset in the log of where to start replay after a crash */ + __le32 j_mount_id ; /* 12 */ struct journal_params jh_journal; } ; @@ -2117,7 +2082,7 @@ struct buffer_head * get_FEB (struct tree_balance *); struct __reiserfs_blocknr_hint { struct inode * inode; /* inode passed to allocator, if we allocate unf. nodes */ long block; /* file offset, in blocks */ - struct reiserfs_key key; + struct in_core_key key; struct path * path; /* search path, used by allocator to deternine search_start by * various ways */ struct reiserfs_transaction_handle * th; /* transaction handle is needed to log super blocks and @@ -2144,7 +2109,7 @@ void reiserfs_init_alloc_options (struct super_block *s); * to use for a new object underneat it. The locality is returned * in disk byte order (le). */ -u32 reiserfs_choose_packing(struct inode *dir); +__le32 reiserfs_choose_packing(struct inode *dir); int is_reusable (struct super_block * s, b_blocknr_t block, int bit_value); void reiserfs_free_block (struct reiserfs_transaction_handle *th, struct inode *, b_blocknr_t, int for_unformatted); diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h index 1eaa48eca81..9244c574882 100644 --- a/include/linux/reiserfs_xattr.h +++ b/include/linux/reiserfs_xattr.h @@ -10,8 +10,8 @@ #define REISERFS_XATTR_MAGIC 0x52465841 /* "RFXA" */ struct reiserfs_xattr_header { - __u32 h_magic; /* magic number for identification */ - __u32 h_hash; /* hash of the value */ + __le32 h_magic; /* magic number for identification */ + __le32 h_hash; /* hash of the value */ }; #ifdef __KERNEL__ diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 32e52769a00..91ac97c2077 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -89,10 +89,14 @@ enum { RTM_GETANYCAST = 62, #define RTM_GETANYCAST RTM_GETANYCAST - RTM_MAX, -#define RTM_MAX RTM_MAX + __RTM_MAX, +#define RTM_MAX (((__RTM_MAX + 3) & ~3) - 1) }; +#define RTM_NR_MSGTYPES (RTM_MAX + 1 - RTM_BASE) +#define RTM_NR_FAMILIES (RTM_NR_MSGTYPES >> 2) +#define RTM_FAM(cmd) (((cmd) - RTM_BASE) >> 2) + /* Generic structure for encapsulation of optional route information. It is reminiscent of sockaddr, but with sa_family replaced diff --git a/include/linux/sched.h b/include/linux/sched.h index 1cced971232..4dbb109022f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -578,7 +578,7 @@ struct task_struct { unsigned long flags; /* per process flags, defined below */ unsigned long ptrace; - int lock_depth; /* Lock depth */ + int lock_depth; /* BKL lock depth */ int prio, static_prio; struct list_head run_list; @@ -661,7 +661,10 @@ struct task_struct { struct key *thread_keyring; /* keyring private to this thread */ #endif int oomkilladj; /* OOM kill score adjustment (bit shift). */ - char comm[TASK_COMM_LEN]; + char comm[TASK_COMM_LEN]; /* executable name excluding path + - access with [gs]et_task_comm (which lock + it with task_lock()) + - initialized normally by flush_old_exec */ /* file system info */ int link_count, total_link_count; /* ipc stuff */ @@ -845,6 +848,7 @@ extern void sched_idle_next(void); extern void set_user_nice(task_t *p, long nice); extern int task_prio(const task_t *p); extern int task_nice(const task_t *p); +extern int can_nice(const task_t *p, const int nice); extern int task_curr(const task_t *p); extern int idle_cpu(int cpu); extern int sched_setscheduler(struct task_struct *, int, struct sched_param *); @@ -1011,7 +1015,6 @@ extern int copy_thread(int, unsigned long, unsigned long, unsigned long, struct extern void flush_thread(void); extern void exit_thread(void); -extern void exit_mm(struct task_struct *); extern void exit_files(struct task_struct *); extern void exit_signal(struct task_struct *); extern void __exit_signal(struct task_struct *); diff --git a/include/linux/signal.h b/include/linux/signal.h index 99c97ad026c..0a98f5ec5ca 100644 --- a/include/linux/signal.h +++ b/include/linux/signal.h @@ -9,6 +9,17 @@ #ifdef __KERNEL__ /* + * These values of sa_flags are used only by the kernel as part of the + * irq handling routines. + * + * SA_INTERRUPT is also used by the irq handling routines. + * SA_SHIRQ is for shared interrupt support on PCI and EISA. + */ +#define SA_PROBE SA_ONESHOT +#define SA_SAMPLE_RANDOM SA_RESTART +#define SA_SHIRQ 0x04000000 + +/* * Real Time signals may be queued. */ @@ -209,6 +220,12 @@ static inline void init_sigpending(struct sigpending *sig) INIT_LIST_HEAD(&sig->list); } +/* Test if 'sig' is valid signal. Use this instead of testing _NSIG directly */ +static inline int valid_signal(unsigned long sig) +{ + return sig <= _NSIG ? 1 : 0; +} + extern int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p); extern int __group_send_sig_info(int, struct siginfo *, struct task_struct *); extern long do_sigpending(void __user *, unsigned long); diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 22b70181961..cc04f5cd228 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -167,13 +167,14 @@ struct skb_shared_info { * @h: Transport layer header * @nh: Network layer header * @mac: Link layer header - * @dst: FIXME: Describe this field + * @dst: destination entry + * @sp: the security path, used for xfrm * @cb: Control buffer. Free for use by every layer. Put private vars here * @len: Length of actual data * @data_len: Data length * @mac_len: Length of link layer header * @csum: Checksum - * @__unused: Dead field, may be reused + * @local_df: allow local fragmentation * @cloned: Head may be cloned (check refcnt to be sure) * @nohdr: Payload reference only, must not modify header * @pkt_type: Packet class @@ -968,6 +969,7 @@ static inline void __skb_queue_purge(struct sk_buff_head *list) kfree_skb(skb); } +#ifndef CONFIG_HAVE_ARCH_DEV_ALLOC_SKB /** * __dev_alloc_skb - allocate an skbuff for sending * @length: length to allocate @@ -980,7 +982,6 @@ static inline void __skb_queue_purge(struct sk_buff_head *list) * * %NULL is returned in there is no free memory. */ -#ifndef CONFIG_HAVE_ARCH_DEV_ALLOC_SKB static inline struct sk_buff *__dev_alloc_skb(unsigned int length, int gfp_mask) { diff --git a/include/linux/slab.h b/include/linux/slab.h index 3e3c3ab8ff9..7d66385ae75 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -62,16 +62,9 @@ extern kmem_cache_t *kmem_cache_create(const char *, size_t, size_t, unsigned lo extern int kmem_cache_destroy(kmem_cache_t *); extern int kmem_cache_shrink(kmem_cache_t *); extern void *kmem_cache_alloc(kmem_cache_t *, unsigned int __nocast); -#ifdef CONFIG_NUMA -extern void *kmem_cache_alloc_node(kmem_cache_t *, int); -#else -static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, int node) -{ - return kmem_cache_alloc(cachep, GFP_KERNEL); -} -#endif extern void kmem_cache_free(kmem_cache_t *, void *); extern unsigned int kmem_cache_size(kmem_cache_t *); +extern kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags); /* Size description struct for general caches. */ struct cache_sizes { @@ -109,6 +102,20 @@ extern void *kcalloc(size_t, size_t, unsigned int __nocast); extern void kfree(const void *); extern unsigned int ksize(const void *); +#ifdef CONFIG_NUMA +extern void *kmem_cache_alloc_node(kmem_cache_t *, int flags, int node); +extern void *kmalloc_node(size_t size, int flags, int node); +#else +static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int node) +{ + return kmem_cache_alloc(cachep, flags); +} +static inline void *kmalloc_node(size_t size, int flags, int node) +{ + return kmalloc(size, flags); +} +#endif + extern int FASTCALL(kmem_cache_reap(int)); extern int FASTCALL(kmem_ptr_validate(kmem_cache_t *cachep, void *ptr)); diff --git a/include/linux/sockios.h b/include/linux/sockios.h index 5eb33205cc0..e6b9d1d36ea 100644 --- a/include/linux/sockios.h +++ b/include/linux/sockios.h @@ -7,7 +7,7 @@ * * Version: @(#)sockios.h 1.0.2 03/09/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * This program is free software; you can redistribute it and/or diff --git a/include/linux/soundcard.h b/include/linux/soundcard.h index 28d2d188197..523d069c862 100644 --- a/include/linux/soundcard.h +++ b/include/linux/soundcard.h @@ -39,6 +39,13 @@ /* In Linux we need to be prepared for cross compiling */ #include <linux/ioctl.h> +/* Endian macros. */ +#ifdef __KERNEL__ +# include <asm/byteorder.h> +#else +# include <endian.h> +#endif + /* * Supported card ID numbers (Should be somewhere else?) */ @@ -179,13 +186,26 @@ typedef struct seq_event_rec { * Some big endian/little endian handling macros */ -#if defined(_AIX) || defined(AIX) || defined(sparc) || defined(__sparc__) || defined(HPPA) || defined(PPC) || defined(__mc68000__) -/* Big endian machines */ -# define _PATCHKEY(id) (0xfd00|id) -# define AFMT_S16_NE AFMT_S16_BE -#else -# define _PATCHKEY(id) ((id<<8)|0xfd) -# define AFMT_S16_NE AFMT_S16_LE +#define _LINUX_PATCHKEY_H_INDIRECT +#include <linux/patchkey.h> +#undef _LINUX_PATCHKEY_H_INDIRECT + +#if defined(__KERNEL__) +# if defined(__BIG_ENDIAN) +# define AFMT_S16_NE AFMT_S16_BE +# elif defined(__LITTLE_ENDIAN) +# define AFMT_S16_NE AFMT_S16_LE +# else +# error "could not determine byte order" +# endif +#elif defined(__BYTE_ORDER) +# if __BYTE_ORDER == __BIG_ENDIAN +# define AFMT_S16_NE AFMT_S16_BE +# elif __BYTE_ORDER == __LITTLE_ENDIAN +# define AFMT_S16_NE AFMT_S16_LE +# else +# error "could not determine byte order" +# endif #endif /* diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 757cd9be774..c39f6f72cbb 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -456,8 +456,7 @@ asmlinkage long sys_semctl(int semid, int semnum, int cmd, union semun arg); asmlinkage long sys_semtimedop(int semid, struct sembuf __user *sops, unsigned nsops, const struct timespec __user *timeout); -asmlinkage long sys_shmat(int shmid, char __user *shmaddr, - int shmflg, unsigned long __user *addr); +asmlinkage long sys_shmat(int shmid, char __user *shmaddr, int shmflg); asmlinkage long sys_shmget(key_t key, size_t size, int flag); asmlinkage long sys_shmdt(char __user *shmaddr); asmlinkage long sys_shmctl(int shmid, int cmd, struct shmid_ds __user *buf); diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 358d52b0c44..772998147e3 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -643,6 +643,7 @@ enum { NET_SCTP_MAX_BURST = 12, NET_SCTP_ADDIP_ENABLE = 13, NET_SCTP_PRSCTP_ENABLE = 14, + NET_SCTP_SNDBUF_POLICY = 15, }; /* /proc/sys/net/bridge */ diff --git a/include/linux/tc_act/tc_defact.h b/include/linux/tc_act/tc_defact.h new file mode 100644 index 00000000000..964f473af0f --- /dev/null +++ b/include/linux/tc_act/tc_defact.h @@ -0,0 +1,21 @@ +#ifndef __LINUX_TC_DEF_H +#define __LINUX_TC_DEF_H + +#include <linux/pkt_cls.h> + +struct tc_defact +{ + tc_gen; +}; + +enum +{ + TCA_DEF_UNSPEC, + TCA_DEF_TM, + TCA_DEF_PARMS, + TCA_DEF_DATA, + __TCA_DEF_MAX +}; +#define TCA_DEF_MAX (__TCA_DEF_MAX - 1) + +#endif diff --git a/include/linux/trdevice.h b/include/linux/trdevice.h index aaa1f337edc..99e02ef54c4 100644 --- a/include/linux/trdevice.h +++ b/include/linux/trdevice.h @@ -7,7 +7,7 @@ * * Version: @(#)eth.h 1.0.4 05/13/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * Relocated to include/linux where it belongs by Alan Cox diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index f0df02ae68a..fd2ef742a9f 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -140,8 +140,11 @@ enum { XFRM_MSG_FLUSHPOLICY, #define XFRM_MSG_FLUSHPOLICY XFRM_MSG_FLUSHPOLICY - XFRM_MSG_MAX + __XFRM_MSG_MAX }; +#define XFRM_MSG_MAX (__XFRM_MSG_MAX - 1) + +#define XFRM_NR_MSGTYPES (XFRM_MSG_MAX + 1 - XFRM_MSG_BASE) struct xfrm_user_tmpl { struct xfrm_id id; diff --git a/include/net/act_generic.h b/include/net/act_generic.h new file mode 100644 index 00000000000..95b120781c1 --- /dev/null +++ b/include/net/act_generic.h @@ -0,0 +1,142 @@ +/* + * include/net/act_generic.h + * +*/ +#ifndef ACT_GENERIC_H +#define ACT_GENERIC_H +static inline int tcf_defact_release(struct tcf_defact *p, int bind) +{ + int ret = 0; + if (p) { + if (bind) { + p->bindcnt--; + } + p->refcnt--; + if (p->bindcnt <= 0 && p->refcnt <= 0) { + kfree(p->defdata); + tcf_hash_destroy(p); + ret = 1; + } + } + return ret; +} + +static inline int +alloc_defdata(struct tcf_defact *p, u32 datalen, void *defdata) +{ + p->defdata = kmalloc(datalen, GFP_KERNEL); + if (p->defdata == NULL) + return -ENOMEM; + p->datalen = datalen; + memcpy(p->defdata, defdata, datalen); + return 0; +} + +static inline int +realloc_defdata(struct tcf_defact *p, u32 datalen, void *defdata) +{ + /* safer to be just brute force for now */ + kfree(p->defdata); + return alloc_defdata(p, datalen, defdata); +} + +static inline int +tcf_defact_init(struct rtattr *rta, struct rtattr *est, + struct tc_action *a, int ovr, int bind) +{ + struct rtattr *tb[TCA_DEF_MAX]; + struct tc_defact *parm; + struct tcf_defact *p; + void *defdata; + u32 datalen = 0; + int ret = 0; + + if (rta == NULL || rtattr_parse_nested(tb, TCA_DEF_MAX, rta) < 0) + return -EINVAL; + + if (tb[TCA_DEF_PARMS - 1] == NULL || + RTA_PAYLOAD(tb[TCA_DEF_PARMS - 1]) < sizeof(*parm)) + return -EINVAL; + + parm = RTA_DATA(tb[TCA_DEF_PARMS - 1]); + defdata = RTA_DATA(tb[TCA_DEF_DATA - 1]); + if (defdata == NULL) + return -EINVAL; + + datalen = RTA_PAYLOAD(tb[TCA_DEF_DATA - 1]); + if (datalen <= 0) + return -EINVAL; + + p = tcf_hash_check(parm->index, a, ovr, bind); + if (p == NULL) { + p = tcf_hash_create(parm->index, est, a, sizeof(*p), ovr, bind); + if (p == NULL) + return -ENOMEM; + + ret = alloc_defdata(p, datalen, defdata); + if (ret < 0) { + kfree(p); + return ret; + } + ret = ACT_P_CREATED; + } else { + if (!ovr) { + tcf_defact_release(p, bind); + return -EEXIST; + } + realloc_defdata(p, datalen, defdata); + } + + spin_lock_bh(&p->lock); + p->action = parm->action; + spin_unlock_bh(&p->lock); + if (ret == ACT_P_CREATED) + tcf_hash_insert(p); + return ret; +} + +static inline int tcf_defact_cleanup(struct tc_action *a, int bind) +{ + struct tcf_defact *p = PRIV(a, defact); + + if (p != NULL) + return tcf_defact_release(p, bind); + return 0; +} + +static inline int +tcf_defact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) +{ + unsigned char *b = skb->tail; + struct tc_defact opt; + struct tcf_defact *p = PRIV(a, defact); + struct tcf_t t; + + opt.index = p->index; + opt.refcnt = p->refcnt - ref; + opt.bindcnt = p->bindcnt - bind; + opt.action = p->action; + RTA_PUT(skb, TCA_DEF_PARMS, sizeof(opt), &opt); + RTA_PUT(skb, TCA_DEF_DATA, p->datalen, p->defdata); + t.install = jiffies_to_clock_t(jiffies - p->tm.install); + t.lastuse = jiffies_to_clock_t(jiffies - p->tm.lastuse); + t.expires = jiffies_to_clock_t(p->tm.expires); + RTA_PUT(skb, TCA_DEF_TM, sizeof(t), &t); + return skb->len; + +rtattr_failure: + skb_trim(skb, b - skb->data); + return -1; +} + +#define tca_use_default_ops \ + .dump = tcf_defact_dump, \ + .cleanup = tcf_defact_cleanup, \ + .init = tcf_defact_init, \ + .walk = tcf_generic_walker, \ + +#define tca_use_default_defines(name) \ + static u32 idx_gen; \ + static struct tcf_defact *tcf_##name_ht[MY_TAB_SIZE]; \ + static DEFINE_RWLOCK(##name_lock); +#endif /* _NET_ACT_GENERIC_H */ diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 7af9a13cb9b..a0ed9367217 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -17,6 +17,8 @@ #define IPV6_MAX_ADDRESSES 16 +#include <linux/in6.h> + struct prefix_info { __u8 type; __u8 length; @@ -43,9 +45,9 @@ struct prefix_info { #ifdef __KERNEL__ -#include <linux/in6.h> #include <linux/netdevice.h> #include <net/if_inet6.h> +#include <net/ipv6.h> #define IN6_ADDR_HSIZE 16 diff --git a/include/net/ax25.h b/include/net/ax25.h index fb95ecb6fe0..9e6368a5454 100644 --- a/include/net/ax25.h +++ b/include/net/ax25.h @@ -220,6 +220,14 @@ static __inline__ void ax25_cb_put(ax25_cb *ax25) } } +static inline unsigned short ax25_type_trans(struct sk_buff *skb, struct net_device *dev) +{ + skb->dev = dev; + skb->pkt_type = PACKET_HOST; + skb->mac.raw = skb->data; + return htons(ETH_P_AX25); +} + /* af_ax25.c */ extern struct hlist_head ax25_list; extern spinlock_t ax25_list_lock; @@ -305,7 +313,7 @@ extern ax25_cb *ax25_send_frame(struct sk_buff *, int, ax25_address *, ax25_addr extern void ax25_output(ax25_cb *, int, struct sk_buff *); extern void ax25_kick(ax25_cb *); extern void ax25_transmit_buffer(ax25_cb *, struct sk_buff *, int); -extern void ax25_queue_xmit(struct sk_buff *); +extern void ax25_queue_xmit(struct sk_buff *skb, struct net_device *dev); extern int ax25_check_iframes_acked(ax25_cb *, unsigned short); /* ax25_route.c */ diff --git a/include/net/icmp.h b/include/net/icmp.h index 3fc192478aa..e5ef0d15fb4 100644 --- a/include/net/icmp.h +++ b/include/net/icmp.h @@ -7,7 +7,7 @@ * * Version: @(#)icmp.h 1.0.4 05/13/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * This program is free software; you can redistribute it and/or diff --git a/include/net/ip.h b/include/net/ip.h index b4db1375da2..3f63992eb71 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -7,7 +7,7 @@ * * Version: @(#)ip.h 1.0.2 05/07/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Alan Cox, <gw4pts@gw4pts.ampr.org> * diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 87c45cbfbaf..771b47e30f8 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -416,7 +416,7 @@ extern void ipv6_push_frag_opts(struct sk_buff *skb, u8 *proto); extern int ipv6_skip_exthdr(const struct sk_buff *, int start, - u8 *nexthdrp, int len); + u8 *nexthdrp); extern int ipv6_ext_hdr(u8 nexthdr); diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 87496e3aa33..fcb05a387db 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -140,7 +140,7 @@ psched_tod_diff(int delta_sec, int bound) if (bound <= 1000000 || delta_sec > (0x7FFFFFFF/1000000)-1) return bound; delta = delta_sec * 1000000; - if (delta > bound) + if (delta > bound || delta < 0) delta = bound; return delta; } @@ -156,7 +156,9 @@ psched_tod_diff(int delta_sec, int bound) __delta += 1000000; \ case 1: \ __delta += 1000000; \ - case 0: ; \ + case 0: \ + if (__delta > bound || __delta < 0) \ + __delta = bound; \ } \ __delta; \ }) diff --git a/include/net/route.h b/include/net/route.h index 22da7579d5d..efe92b239ef 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -7,7 +7,7 @@ * * Version: @(#)route.h 1.0.4 05/27/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Fixes: * Alan Cox : Reformatted. Added ip_rt_local() diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index 5576db56324..f4fcee10470 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -407,32 +407,38 @@ sctp_vtag_verify(const struct sctp_chunk *chunk, return 0; } -/* Check VTAG of the packet matches the sender's own tag OR its peer's - * tag and the T bit is set in the Chunk Flags. +/* Check VTAG of the packet matches the sender's own tag and the T bit is + * not set, OR its peer's tag and the T bit is set in the Chunk Flags. */ static inline int sctp_vtag_verify_either(const struct sctp_chunk *chunk, const struct sctp_association *asoc) { - /* RFC 2960 Section 8.5.1, sctpimpguide-06 Section 2.13.2 + /* RFC 2960 Section 8.5.1, sctpimpguide Section 2.41 * - * B) The receiver of a ABORT shall accept the packet if the - * Verification Tag field of the packet matches its own tag OR it - * is set to its peer's tag and the T bit is set in the Chunk - * Flags. Otherwise, the receiver MUST silently discard the packet - * and take no further action. - * - * (C) The receiver of a SHUTDOWN COMPLETE shall accept the - * packet if the Verification Tag field of the packet - * matches its own tag OR it is set to its peer's tag and - * the T bit is set in the Chunk Flags. Otherwise, the - * receiver MUST silently discard the packet and take no - * further action.... + * B) The receiver of a ABORT MUST accept the packet + * if the Verification Tag field of the packet matches its own tag + * and the T bit is not set + * OR + * it is set to its peer's tag and the T bit is set in the Chunk + * Flags. + * Otherwise, the receiver MUST silently discard the packet + * and take no further action. * + * C) The receiver of a SHUTDOWN COMPLETE shall accept the packet + * if the Verification Tag field of the packet matches its own tag + * and the T bit is not set + * OR + * it is set to its peer's tag and the T bit is set in the Chunk + * Flags. + * Otherwise, the receiver MUST silently discard the packet + * and take no further action. An endpoint MUST ignore the + * SHUTDOWN COMPLETE if it is not in the SHUTDOWN-ACK-SENT state. */ - if ((ntohl(chunk->sctp_hdr->vtag) == asoc->c.my_vtag) || - (sctp_test_T_bit(chunk) && (ntohl(chunk->sctp_hdr->vtag) - == asoc->c.peer_vtag))) { + if ((!sctp_test_T_bit(chunk) && + (ntohl(chunk->sctp_hdr->vtag) == asoc->c.my_vtag)) || + (sctp_test_T_bit(chunk) && + (ntohl(chunk->sctp_hdr->vtag) == asoc->c.peer_vtag))) { return 1; } diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 7e64cf6bda1..6c24d9cd3d6 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -154,6 +154,13 @@ extern struct sctp_globals { int max_retrans_path; int max_retrans_init; + /* + * Policy for preforming sctp/socket accounting + * 0 - do socket level accounting, all assocs share sk_sndbuf + * 1 - do sctp accounting, each asoc may use sk_sndbuf bytes + */ + int sndbuf_policy; + /* HB.interval - 30 seconds */ int hb_interval; @@ -207,6 +214,7 @@ extern struct sctp_globals { #define sctp_valid_cookie_life (sctp_globals.valid_cookie_life) #define sctp_cookie_preserve_enable (sctp_globals.cookie_preserve_enable) #define sctp_max_retrans_association (sctp_globals.max_retrans_association) +#define sctp_sndbuf_policy (sctp_globals.sndbuf_policy) #define sctp_max_retrans_path (sctp_globals.max_retrans_path) #define sctp_max_retrans_init (sctp_globals.max_retrans_init) #define sctp_hb_interval (sctp_globals.hb_interval) @@ -1212,7 +1220,8 @@ struct sctp_endpoint { /* Default timeouts. */ int timeouts[SCTP_NUM_TIMEOUT_TYPES]; - /* Various thresholds. */ + /* sendbuf acct. policy. */ + __u32 sndbuf_policy; /* Name for debugging output... */ char *debug_name; diff --git a/include/net/sock.h b/include/net/sock.h index be81cabd0da..a9ef3a6a13f 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -7,7 +7,7 @@ * * Version: @(#)sock.h 1.0.4 05/13/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Corey Minyard <wf-rch!minyard@relay.EU.net> * Florian La Roche <flla@stud.uni-sb.de> @@ -90,17 +90,17 @@ do { spin_lock_init(&((__sk)->sk_lock.slock)); \ struct sock; /** - * struct sock_common - minimal network layer representation of sockets - * @skc_family - network address family - * @skc_state - Connection state - * @skc_reuse - %SO_REUSEADDR setting - * @skc_bound_dev_if - bound device index if != 0 - * @skc_node - main hash linkage for various protocol lookup tables - * @skc_bind_node - bind hash linkage for various protocol lookup tables - * @skc_refcnt - reference count - * - * This is the minimal network layer representation of sockets, the header - * for struct sock and struct tcp_tw_bucket. + * struct sock_common - minimal network layer representation of sockets + * @skc_family: network address family + * @skc_state: Connection state + * @skc_reuse: %SO_REUSEADDR setting + * @skc_bound_dev_if: bound device index if != 0 + * @skc_node: main hash linkage for various protocol lookup tables + * @skc_bind_node: bind hash linkage for various protocol lookup tables + * @skc_refcnt: reference count + * + * This is the minimal network layer representation of sockets, the header + * for struct sock and struct tcp_tw_bucket. */ struct sock_common { unsigned short skc_family; @@ -114,60 +114,62 @@ struct sock_common { /** * struct sock - network layer representation of sockets - * @__sk_common - shared layout with tcp_tw_bucket - * @sk_shutdown - mask of %SEND_SHUTDOWN and/or %RCV_SHUTDOWN - * @sk_userlocks - %SO_SNDBUF and %SO_RCVBUF settings - * @sk_lock - synchronizer - * @sk_rcvbuf - size of receive buffer in bytes - * @sk_sleep - sock wait queue - * @sk_dst_cache - destination cache - * @sk_dst_lock - destination cache lock - * @sk_policy - flow policy - * @sk_rmem_alloc - receive queue bytes committed - * @sk_receive_queue - incoming packets - * @sk_wmem_alloc - transmit queue bytes committed - * @sk_write_queue - Packet sending queue - * @sk_omem_alloc - "o" is "option" or "other" - * @sk_wmem_queued - persistent queue size - * @sk_forward_alloc - space allocated forward - * @sk_allocation - allocation mode - * @sk_sndbuf - size of send buffer in bytes - * @sk_flags - %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE, %SO_OOBINLINE settings - * @sk_no_check - %SO_NO_CHECK setting, wether or not checkup packets - * @sk_route_caps - route capabilities (e.g. %NETIF_F_TSO) - * @sk_lingertime - %SO_LINGER l_linger setting - * @sk_hashent - hash entry in several tables (e.g. tcp_ehash) - * @sk_backlog - always used with the per-socket spinlock held - * @sk_callback_lock - used with the callbacks in the end of this struct - * @sk_error_queue - rarely used - * @sk_prot - protocol handlers inside a network family - * @sk_err - last error - * @sk_err_soft - errors that don't cause failure but are the cause of a persistent failure not just 'timed out' - * @sk_ack_backlog - current listen backlog - * @sk_max_ack_backlog - listen backlog set in listen() - * @sk_priority - %SO_PRIORITY setting - * @sk_type - socket type (%SOCK_STREAM, etc) - * @sk_protocol - which protocol this socket belongs in this network family - * @sk_peercred - %SO_PEERCRED setting - * @sk_rcvlowat - %SO_RCVLOWAT setting - * @sk_rcvtimeo - %SO_RCVTIMEO setting - * @sk_sndtimeo - %SO_SNDTIMEO setting - * @sk_filter - socket filtering instructions - * @sk_protinfo - private area, net family specific, when not using slab - * @sk_timer - sock cleanup timer - * @sk_stamp - time stamp of last packet received - * @sk_socket - Identd and reporting IO signals - * @sk_user_data - RPC layer private data - * @sk_sndmsg_page - cached page for sendmsg - * @sk_sndmsg_off - cached offset for sendmsg - * @sk_send_head - front of stuff to transmit - * @sk_write_pending - a write to stream socket waits to start - * @sk_state_change - callback to indicate change in the state of the sock - * @sk_data_ready - callback to indicate there is data to be processed - * @sk_write_space - callback to indicate there is bf sending space available - * @sk_error_report - callback to indicate errors (e.g. %MSG_ERRQUEUE) - * @sk_backlog_rcv - callback to process the backlog - * @sk_destruct - called at sock freeing time, i.e. when all refcnt == 0 + * @__sk_common: shared layout with tcp_tw_bucket + * @sk_shutdown: mask of %SEND_SHUTDOWN and/or %RCV_SHUTDOWN + * @sk_userlocks: %SO_SNDBUF and %SO_RCVBUF settings + * @sk_lock: synchronizer + * @sk_rcvbuf: size of receive buffer in bytes + * @sk_sleep: sock wait queue + * @sk_dst_cache: destination cache + * @sk_dst_lock: destination cache lock + * @sk_policy: flow policy + * @sk_rmem_alloc: receive queue bytes committed + * @sk_receive_queue: incoming packets + * @sk_wmem_alloc: transmit queue bytes committed + * @sk_write_queue: Packet sending queue + * @sk_omem_alloc: "o" is "option" or "other" + * @sk_wmem_queued: persistent queue size + * @sk_forward_alloc: space allocated forward + * @sk_allocation: allocation mode + * @sk_sndbuf: size of send buffer in bytes + * @sk_flags: %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE, %SO_OOBINLINE settings + * @sk_no_check: %SO_NO_CHECK setting, wether or not checkup packets + * @sk_route_caps: route capabilities (e.g. %NETIF_F_TSO) + * @sk_lingertime: %SO_LINGER l_linger setting + * @sk_hashent: hash entry in several tables (e.g. tcp_ehash) + * @sk_backlog: always used with the per-socket spinlock held + * @sk_callback_lock: used with the callbacks in the end of this struct + * @sk_error_queue: rarely used + * @sk_prot: protocol handlers inside a network family + * @sk_prot_creator: sk_prot of original sock creator (see ipv6_setsockopt, IPV6_ADDRFORM for instance) + * @sk_err: last error + * @sk_err_soft: errors that don't cause failure but are the cause of a persistent failure not just 'timed out' + * @sk_ack_backlog: current listen backlog + * @sk_max_ack_backlog: listen backlog set in listen() + * @sk_priority: %SO_PRIORITY setting + * @sk_type: socket type (%SOCK_STREAM, etc) + * @sk_protocol: which protocol this socket belongs in this network family + * @sk_peercred: %SO_PEERCRED setting + * @sk_rcvlowat: %SO_RCVLOWAT setting + * @sk_rcvtimeo: %SO_RCVTIMEO setting + * @sk_sndtimeo: %SO_SNDTIMEO setting + * @sk_filter: socket filtering instructions + * @sk_protinfo: private area, net family specific, when not using slab + * @sk_timer: sock cleanup timer + * @sk_stamp: time stamp of last packet received + * @sk_socket: Identd and reporting IO signals + * @sk_user_data: RPC layer private data + * @sk_sndmsg_page: cached page for sendmsg + * @sk_sndmsg_off: cached offset for sendmsg + * @sk_send_head: front of stuff to transmit + * @sk_security: used by security modules + * @sk_write_pending: a write to stream socket waits to start + * @sk_state_change: callback to indicate change in the state of the sock + * @sk_data_ready: callback to indicate there is data to be processed + * @sk_write_space: callback to indicate there is bf sending space available + * @sk_error_report: callback to indicate errors (e.g. %MSG_ERRQUEUE) + * @sk_backlog_rcv: callback to process the backlog + * @sk_destruct: called at sock freeing time, i.e. when all refcnt == 0 */ struct sock { /* @@ -217,6 +219,7 @@ struct sock { } sk_backlog; struct sk_buff_head sk_error_queue; struct proto *sk_prot; + struct proto *sk_prot_creator; rwlock_t sk_callback_lock; int sk_err, sk_err_soft; @@ -1223,8 +1226,8 @@ sock_recv_timestamp(struct msghdr *msg, struct sock *sk, struct sk_buff *skb) /** * sk_eat_skb - Release a skb if it is no longer needed - * @sk - socket to eat this skb from - * @skb - socket buffer to eat + * @sk: socket to eat this skb from + * @skb: socket buffer to eat * * This routine must be called with interrupts disabled or with the socket * locked so that the sk_buff queue operation is ok. diff --git a/include/net/tc_act/tc_defact.h b/include/net/tc_act/tc_defact.h new file mode 100644 index 00000000000..463aa671f95 --- /dev/null +++ b/include/net/tc_act/tc_defact.h @@ -0,0 +1,13 @@ +#ifndef __NET_TC_DEF_H +#define __NET_TC_DEF_H + +#include <net/act_api.h> + +struct tcf_defact +{ + tca_gen(defact); + u32 datalen; + void *defdata; +}; + +#endif diff --git a/include/net/tcp.h b/include/net/tcp.h index 503810a70e2..e71f8ba3e10 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -7,7 +7,7 @@ * * Version: @(#)tcp.h 1.0.5 05/23/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * This program is free software; you can redistribute it and/or @@ -1417,19 +1417,20 @@ tcp_nagle_check(const struct tcp_sock *tp, const struct sk_buff *skb, tcp_minshall_check(tp)))); } -extern void tcp_set_skb_tso_segs(struct sk_buff *, unsigned int); +extern void tcp_set_skb_tso_segs(struct sock *, struct sk_buff *); /* This checks if the data bearing packet SKB (usually sk->sk_send_head) * should be put on the wire right now. */ -static __inline__ int tcp_snd_test(const struct tcp_sock *tp, +static __inline__ int tcp_snd_test(struct sock *sk, struct sk_buff *skb, unsigned cur_mss, int nonagle) { + struct tcp_sock *tp = tcp_sk(sk); int pkts = tcp_skb_pcount(skb); if (!pkts) { - tcp_set_skb_tso_segs(skb, tp->mss_cache_std); + tcp_set_skb_tso_segs(sk, skb); pkts = tcp_skb_pcount(skb); } @@ -1490,7 +1491,7 @@ static __inline__ void __tcp_push_pending_frames(struct sock *sk, if (skb) { if (!tcp_skb_is_last(sk, skb)) nonagle = TCP_NAGLE_PUSH; - if (!tcp_snd_test(tp, skb, cur_mss, nonagle) || + if (!tcp_snd_test(sk, skb, cur_mss, nonagle) || tcp_write_xmit(sk, nonagle)) tcp_check_probe_timer(sk, tp); } @@ -1508,7 +1509,7 @@ static __inline__ int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp) struct sk_buff *skb = sk->sk_send_head; return (skb && - tcp_snd_test(tp, skb, tcp_current_mss(sk, 1), + tcp_snd_test(sk, skb, tcp_current_mss(sk, 1), tcp_skb_is_last(sk, skb) ? TCP_NAGLE_PUSH : tp->nonagle)); } diff --git a/include/net/udp.h b/include/net/udp.h index c496d10101d..ac229b761db 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -7,7 +7,7 @@ * * Version: @(#)udp.h 1.0.2 05/07/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * Fixes: diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 73e9a8ca3d3..e142a256d5d 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1,6 +1,7 @@ #ifndef _NET_XFRM_H #define _NET_XFRM_H +#include <linux/compiler.h> #include <linux/xfrm.h> #include <linux/spinlock.h> #include <linux/list.h> @@ -516,6 +517,15 @@ struct xfrm_dst u32 child_mtu_cached; }; +static inline void xfrm_dst_destroy(struct xfrm_dst *xdst) +{ + dst_release(xdst->route); + if (likely(xdst->u.dst.xfrm)) + xfrm_state_put(xdst->u.dst.xfrm); +} + +extern void xfrm_dst_ifdown(struct dst_entry *dst, struct net_device *dev); + /* Decapsulation state, used by the input to store data during * decapsulation procedure, to be used later (during the policy * check diff --git a/include/video/edid.h b/include/video/edid.h index abc1b489c0d..b913f196131 100644 --- a/include/video/edid.h +++ b/include/video/edid.h @@ -4,9 +4,6 @@ #ifdef __KERNEL__ #include <linux/config.h> -#ifdef CONFIG_PPC_OF -#include <linux/pci.h> -#endif #ifdef CONFIG_X86 struct edid_info { @@ -14,14 +11,8 @@ struct edid_info { }; extern struct edid_info edid_info; -extern char *get_EDID_from_BIOS(void *); - #endif /* CONFIG_X86 */ -#ifdef CONFIG_PPC_OF -extern char *get_EDID_from_OF(struct pci_dev *pdev); -#endif - #endif /* __KERNEL__ */ #endif /* __linux_video_edid_h__ */ diff --git a/include/video/tdfx.h b/include/video/tdfx.h index a896e444206..04237676b17 100644 --- a/include/video/tdfx.h +++ b/include/video/tdfx.h @@ -99,6 +99,8 @@ #define MISCINIT1_2DBLOCK_DIS BIT(15) #define DRAMINIT0_SGRAM_NUM BIT(26) #define DRAMINIT0_SGRAM_TYPE BIT(27) +#define DRAMINIT0_SGRAM_TYPE_MASK (BIT(27)|BIT(28)|BIT(29)) +#define DRAMINIT0_SGRAM_TYPE_SHIFT 27 #define DRAMINIT1_MEM_SDRAM BIT(30) #define VGAINIT0_VGA_DISABLE BIT(0) #define VGAINIT0_EXT_TIMING BIT(1) diff --git a/init/Kconfig b/init/Kconfig index abe2682a6ca..d920baed109 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -173,7 +173,7 @@ config AUDIT config AUDITSYSCALL bool "Enable system-call auditing support" - depends on AUDIT && (X86 || PPC64 || ARCH_S390 || IA64) + depends on AUDIT && (X86 || PPC64 || ARCH_S390 || IA64 || UML) default y if SECURITY_SELINUX help Enable low-overhead system-call auditing infrastructure that @@ -275,6 +275,27 @@ config KALLSYMS_EXTRA_PASS reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while you wait for kallsyms to be fixed. + +config PRINTK + default y + bool "Enable support for printk" if EMBEDDED + help + This option enables normal printk support. Removing it + eliminates most of the message strings from the kernel image + and makes the kernel more or less silent. As this makes it + very difficult to diagnose system problems, saying N here is + strongly discouraged. + +config BUG + bool "BUG() support" if EMBEDDED + default y + help + Disabling this option eliminates support for BUG and WARN, reducing + the size of your kernel image and potentially quietly ignoring + numerous fatal conditions. You should only consider disabling this + option for embedded systems with no facilities for reporting errors. + Just say Y. + config BASE_FULL default y bool "Enable full-sized data structures for core" if EMBEDDED diff --git a/ipc/mqueue.c b/ipc/mqueue.c index cb0cd3cf3b5..0acf245f441 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -23,6 +23,7 @@ #include <linux/skbuff.h> #include <linux/netlink.h> #include <linux/syscalls.h> +#include <linux/signal.h> #include <net/sock.h> #include "util.h" @@ -767,7 +768,7 @@ static inline void pipelined_send(struct mqueue_inode_info *info, list_del(&receiver->list); receiver->state = STATE_PENDING; wake_up_process(receiver->task); - wmb(); + smp_wmb(); receiver->state = STATE_READY; } @@ -786,7 +787,7 @@ static inline void pipelined_receive(struct mqueue_inode_info *info) list_del(&sender->list); sender->state = STATE_PENDING; wake_up_process(sender->task); - wmb(); + smp_wmb(); sender->state = STATE_READY; } @@ -976,8 +977,7 @@ asmlinkage long sys_mq_notify(mqd_t mqdes, notification.sigev_notify != SIGEV_THREAD)) return -EINVAL; if (notification.sigev_notify == SIGEV_SIGNAL && - (notification.sigev_signo < 0 || - notification.sigev_signo > _NSIG)) { + !valid_signal(notification.sigev_signo)) { return -EINVAL; } if (notification.sigev_notify == SIGEV_THREAD) { diff --git a/ipc/shm.c b/ipc/shm.c index 06cd5c91056..cce022435db 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -28,6 +28,8 @@ #include <linux/security.h> #include <linux/syscalls.h> #include <linux/audit.h> +#include <linux/ptrace.h> + #include <asm/uaccess.h> #include "util.h" @@ -771,6 +773,18 @@ out: return err; } +asmlinkage long sys_shmat(int shmid, char __user *shmaddr, int shmflg) +{ + unsigned long ret; + long err; + + err = do_shmat(shmid, shmaddr, shmflg, &ret); + if (err) + return err; + force_successful_syscall_return(); + return (long)ret; +} + /* * detach and kill segment if marked destroyed. * The work is done in shm_close. diff --git a/kernel/Makefile b/kernel/Makefile index eb88b446c2c..b01d26fe8db 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -29,7 +29,7 @@ obj-$(CONFIG_SYSFS) += ksysfs.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o -ifneq ($(CONFIG_IA64),y) +ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y) # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is # needed for x86 only. Why this used to be enabled for all architectures is beyond # me. I suspect most platforms don't need this, but until we know that for sure diff --git a/kernel/audit.c b/kernel/audit.c index 0f84dd7af2c..9c4f1af0c79 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -1,4 +1,4 @@ -/* audit.c -- Auditing support -*- linux-c -*- +/* audit.c -- Auditing support * Gateway between the kernel (e.g., selinux) and the user-space audit daemon. * System-call specific features have moved to auditsc.c * @@ -38,7 +38,7 @@ * 6) Support low-overhead kernel-based filtering to minimize the * information that must be passed to user-space. * - * Example user-space utilities: http://people.redhat.com/faith/audit/ + * Example user-space utilities: http://people.redhat.com/sgrubb/audit/ */ #include <linux/init.h> @@ -142,7 +142,6 @@ struct audit_buffer { int total; int type; int pid; - int count; /* Times requeued */ }; void audit_set_type(struct audit_buffer *ab, int type) @@ -239,36 +238,36 @@ void audit_log_lost(const char *message) } -static int audit_set_rate_limit(int limit) +static int audit_set_rate_limit(int limit, uid_t loginuid) { int old = audit_rate_limit; audit_rate_limit = limit; - audit_log(current->audit_context, "audit_rate_limit=%d old=%d", - audit_rate_limit, old); + audit_log(NULL, "audit_rate_limit=%d old=%d by auid %u", + audit_rate_limit, old, loginuid); return old; } -static int audit_set_backlog_limit(int limit) +static int audit_set_backlog_limit(int limit, uid_t loginuid) { int old = audit_backlog_limit; audit_backlog_limit = limit; - audit_log(current->audit_context, "audit_backlog_limit=%d old=%d", - audit_backlog_limit, old); + audit_log(NULL, "audit_backlog_limit=%d old=%d by auid %u", + audit_backlog_limit, old, loginuid); return old; } -static int audit_set_enabled(int state) +static int audit_set_enabled(int state, uid_t loginuid) { int old = audit_enabled; if (state != 0 && state != 1) return -EINVAL; audit_enabled = state; - audit_log(current->audit_context, "audit_enabled=%d old=%d", - audit_enabled, old); + audit_log(NULL, "audit_enabled=%d old=%d by auid %u", + audit_enabled, old, loginuid); return old; } -static int audit_set_failure(int state) +static int audit_set_failure(int state, uid_t loginuid) { int old = audit_failure; if (state != AUDIT_FAIL_SILENT @@ -276,8 +275,8 @@ static int audit_set_failure(int state) && state != AUDIT_FAIL_PANIC) return -EINVAL; audit_failure = state; - audit_log(current->audit_context, "audit_failure=%d old=%d", - audit_failure, old); + audit_log(NULL, "audit_failure=%d old=%d by auid %u", + audit_failure, old, loginuid); return old; } @@ -344,6 +343,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) int err; struct audit_buffer *ab; u16 msg_type = nlh->nlmsg_type; + uid_t loginuid; /* loginuid of sender */ err = audit_netlink_ok(NETLINK_CB(skb).eff_cap, msg_type); if (err) @@ -351,6 +351,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) pid = NETLINK_CREDS(skb)->pid; uid = NETLINK_CREDS(skb)->uid; + loginuid = NETLINK_CB(skb).loginuid; seq = nlh->nlmsg_seq; data = NLMSG_DATA(nlh); @@ -371,34 +372,36 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) return -EINVAL; status_get = (struct audit_status *)data; if (status_get->mask & AUDIT_STATUS_ENABLED) { - err = audit_set_enabled(status_get->enabled); + err = audit_set_enabled(status_get->enabled, loginuid); if (err < 0) return err; } if (status_get->mask & AUDIT_STATUS_FAILURE) { - err = audit_set_failure(status_get->failure); + err = audit_set_failure(status_get->failure, loginuid); if (err < 0) return err; } if (status_get->mask & AUDIT_STATUS_PID) { int old = audit_pid; audit_pid = status_get->pid; - audit_log(current->audit_context, - "audit_pid=%d old=%d", audit_pid, old); + audit_log(NULL, "audit_pid=%d old=%d by auid %u", + audit_pid, old, loginuid); } if (status_get->mask & AUDIT_STATUS_RATE_LIMIT) - audit_set_rate_limit(status_get->rate_limit); + audit_set_rate_limit(status_get->rate_limit, loginuid); if (status_get->mask & AUDIT_STATUS_BACKLOG_LIMIT) - audit_set_backlog_limit(status_get->backlog_limit); + audit_set_backlog_limit(status_get->backlog_limit, + loginuid); break; case AUDIT_USER: ab = audit_log_start(NULL); if (!ab) break; /* audit_panic has been called */ audit_log_format(ab, - "user pid=%d uid=%d length=%d msg='%.1024s'", + "user pid=%d uid=%d length=%d loginuid=%u" + " msg='%.1024s'", pid, uid, (int)(nlh->nlmsg_len - ((char *)data - (char *)nlh)), - (char *)data); + loginuid, (char *)data); ab->type = AUDIT_USER; ab->pid = pid; audit_log_end(ab); @@ -411,7 +414,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) case AUDIT_LIST: #ifdef CONFIG_AUDITSYSCALL err = audit_receive_filter(nlh->nlmsg_type, NETLINK_CB(skb).pid, - uid, seq, data); + uid, seq, data, loginuid); #else err = -EOPNOTSUPP; #endif @@ -427,7 +430,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) /* Get message from skb (based on rtnetlink_rcv_skb). Each message is * processed by audit_receive_msg. Malformed skbs with wrong length are * discarded silently. */ -static int audit_receive_skb(struct sk_buff *skb) +static void audit_receive_skb(struct sk_buff *skb) { int err; struct nlmsghdr *nlh; @@ -436,7 +439,7 @@ static int audit_receive_skb(struct sk_buff *skb) while (skb->len >= NLMSG_SPACE(0)) { nlh = (struct nlmsghdr *)skb->data; if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len) - return 0; + return; rlen = NLMSG_ALIGN(nlh->nlmsg_len); if (rlen > skb->len) rlen = skb->len; @@ -446,23 +449,20 @@ static int audit_receive_skb(struct sk_buff *skb) netlink_ack(skb, nlh, 0); skb_pull(skb, rlen); } - return 0; } /* Receive messages from netlink socket. */ static void audit_receive(struct sock *sk, int length) { struct sk_buff *skb; + unsigned int qlen; - if (down_trylock(&audit_netlink_sem)) - return; + down(&audit_netlink_sem); - /* FIXME: this must not cause starvation */ - while ((skb = skb_dequeue(&sk->sk_receive_queue))) { - if (audit_receive_skb(skb) && skb->len) - skb_queue_head(&sk->sk_receive_queue, skb); - else - kfree_skb(skb); + for (qlen = skb_queue_len(&sk->sk_receive_queue); qlen; qlen--) { + skb = skb_dequeue(&sk->sk_receive_queue); + audit_receive_skb(skb); + kfree_skb(skb); } up(&audit_netlink_sem); } @@ -483,7 +483,7 @@ static void audit_log_move(struct audit_buffer *ab) if (ab->len == 0) return; - skb = skb_peek(&ab->sklist); + skb = skb_peek_tail(&ab->sklist); if (!skb || skb_tailroom(skb) <= ab->len + extra) { skb = alloc_skb(2 * ab->len + extra, GFP_ATOMIC); if (!skb) { @@ -522,9 +522,9 @@ static inline int audit_log_drain(struct audit_buffer *ab) retval = netlink_unicast(audit_sock, skb, audit_pid, MSG_DONTWAIT); } - if (retval == -EAGAIN && ab->count < 5) { - ++ab->count; - skb_queue_tail(&ab->sklist, skb); + if (retval == -EAGAIN && + (atomic_read(&audit_backlog)) < audit_backlog_limit) { + skb_queue_head(&ab->sklist, skb); audit_log_end_irq(ab); return 1; } @@ -540,8 +540,8 @@ static inline int audit_log_drain(struct audit_buffer *ab) if (!audit_pid) { /* No daemon */ int offset = ab->nlh ? NLMSG_SPACE(0) : 0; int len = skb->len - offset; - printk(KERN_ERR "%*.*s\n", - len, len, skb->data + offset); + skb->data[offset + len] = '\0'; + printk(KERN_ERR "%s\n", skb->data + offset); } kfree_skb(skb); ab->nlh = NULL; @@ -620,7 +620,7 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx) struct audit_buffer *ab = NULL; unsigned long flags; struct timespec t; - int serial = 0; + unsigned int serial; if (!audit_initialized) return NULL; @@ -662,15 +662,16 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx) ab->total = 0; ab->type = AUDIT_KERNEL; ab->pid = 0; - ab->count = 0; #ifdef CONFIG_AUDITSYSCALL if (ab->ctx) audit_get_stamp(ab->ctx, &t, &serial); else #endif + { t = CURRENT_TIME; - + serial = 0; + } audit_log_format(ab, "audit(%lu.%03lu:%u): ", t.tv_sec, t.tv_nsec/1000000, serial); return ab; @@ -720,6 +721,29 @@ void audit_log_format(struct audit_buffer *ab, const char *fmt, ...) va_end(args); } +void audit_log_hex(struct audit_buffer *ab, const unsigned char *buf, size_t len) +{ + int i; + + for (i=0; i<len; i++) + audit_log_format(ab, "%02x", buf[i]); +} + +void audit_log_untrustedstring(struct audit_buffer *ab, const char *string) +{ + const unsigned char *p = string; + + while (*p) { + if (*p == '"' || *p == ' ' || *p < 0x20 || *p > 0x7f) { + audit_log_hex(ab, string, strlen(string)); + return; + } + p++; + } + audit_log_format(ab, "\"%s\"", string); +} + + /* This is a helper-function to print the d_path without using a static * buffer or allocating another buffer in addition to the one in * audit_buffer. */ diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 6f1931381bc..37b3ac94bc4 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -1,4 +1,4 @@ -/* auditsc.c -- System-call auditing support -*- linux-c -*- +/* auditsc.c -- System-call auditing support * Handles all system-call specific auditing features. * * Copyright 2003-2004 Red Hat Inc., Durham, North Carolina. @@ -123,7 +123,7 @@ struct audit_context { int major; /* syscall number */ unsigned long argv[4]; /* syscall arguments */ int return_valid; /* return code is valid */ - int return_code;/* syscall return code */ + long return_code;/* syscall return code */ int auditable; /* 1 if record should be written */ int name_count; struct audit_names names[AUDIT_NAMES]; @@ -135,6 +135,7 @@ struct audit_context { uid_t uid, euid, suid, fsuid; gid_t gid, egid, sgid, fsgid; unsigned long personality; + int arch; #if AUDIT_DEBUG int put_count; @@ -250,7 +251,8 @@ static int audit_copy_rule(struct audit_rule *d, struct audit_rule *s) return 0; } -int audit_receive_filter(int type, int pid, int uid, int seq, void *data) +int audit_receive_filter(int type, int pid, int uid, int seq, void *data, + uid_t loginuid) { u32 flags; struct audit_entry *entry; @@ -285,6 +287,7 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data) err = audit_add_rule(entry, &audit_entlist); if (!err && (flags & AUDIT_AT_EXIT)) err = audit_add_rule(entry, &audit_extlist); + audit_log(NULL, "auid %u added an audit rule\n", loginuid); break; case AUDIT_DEL: flags =((struct audit_rule *)data)->flags; @@ -294,6 +297,7 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data) err = audit_del_rule(data, &audit_entlist); if (!err && (flags & AUDIT_AT_EXIT)) err = audit_del_rule(data, &audit_extlist); + audit_log(NULL, "auid %u removed an audit rule\n", loginuid); break; default: return -EINVAL; @@ -348,6 +352,10 @@ static int audit_filter_rules(struct task_struct *tsk, case AUDIT_PERS: result = (tsk->personality == value); break; + case AUDIT_ARCH: + if (ctx) + result = (ctx->arch == value); + break; case AUDIT_EXIT: if (ctx && ctx->return_valid) @@ -355,7 +363,7 @@ static int audit_filter_rules(struct task_struct *tsk, break; case AUDIT_SUCCESS: if (ctx && ctx->return_valid) - result = (ctx->return_code >= 0); + result = (ctx->return_valid == AUDITSC_SUCCESS); break; case AUDIT_DEVMAJOR: if (ctx) { @@ -648,8 +656,11 @@ static void audit_log_exit(struct audit_context *context) audit_log_format(ab, "syscall=%d", context->major); if (context->personality != PER_LINUX) audit_log_format(ab, " per=%lx", context->personality); + audit_log_format(ab, " arch=%x", context->arch); if (context->return_valid) - audit_log_format(ab, " exit=%d", context->return_code); + audit_log_format(ab, " success=%s exit=%ld", + (context->return_valid==AUDITSC_SUCCESS)?"yes":"no", + context->return_code); audit_log_format(ab, " a0=%lx a1=%lx a2=%lx a3=%lx items=%d" " pid=%d loginuid=%d uid=%d gid=%d" @@ -696,9 +707,10 @@ static void audit_log_exit(struct audit_context *context) if (!ab) continue; /* audit_panic has been called */ audit_log_format(ab, "item=%d", i); - if (context->names[i].name) - audit_log_format(ab, " name=%s", - context->names[i].name); + if (context->names[i].name) { + audit_log_format(ab, " name="); + audit_log_untrustedstring(ab, context->names[i].name); + } if (context->names[i].ino != (unsigned long)-1) audit_log_format(ab, " inode=%lu dev=%02x:%02x mode=%#o" " uid=%d gid=%d rdev=%02x:%02x", @@ -772,7 +784,7 @@ static inline unsigned int audit_serial(void) * then the record will be written at syscall exit time (otherwise, it * will only be written if another part of the kernel requests that it * be written). */ -void audit_syscall_entry(struct task_struct *tsk, int major, +void audit_syscall_entry(struct task_struct *tsk, int arch, int major, unsigned long a1, unsigned long a2, unsigned long a3, unsigned long a4) { @@ -826,6 +838,7 @@ void audit_syscall_entry(struct task_struct *tsk, int major, if (!audit_enabled) return; + context->arch = arch; context->major = major; context->argv[0] = a1; context->argv[1] = a2; @@ -849,13 +862,13 @@ void audit_syscall_entry(struct task_struct *tsk, int major, * filtering, or because some other part of the kernel write an audit * message), then write out the syscall information. In call cases, * free the names stored from getname(). */ -void audit_syscall_exit(struct task_struct *tsk, int return_code) +void audit_syscall_exit(struct task_struct *tsk, int valid, long return_code) { struct audit_context *context; get_task_struct(tsk); task_lock(tsk); - context = audit_get_context(tsk, 1, return_code); + context = audit_get_context(tsk, valid, return_code); task_unlock(tsk); /* Not having a context here is ok, since the parent may have @@ -868,6 +881,7 @@ void audit_syscall_exit(struct task_struct *tsk, int return_code) context->in_syscall = 0; context->auditable = 0; + if (context->previous) { struct audit_context *new_context = context->previous; context->previous = NULL; @@ -981,7 +995,7 @@ void audit_inode(const char *name, const struct inode *inode) } void audit_get_stamp(struct audit_context *ctx, - struct timespec *t, int *serial) + struct timespec *t, unsigned int *serial) { if (ctx) { t->tv_sec = ctx->ctime.tv_sec; @@ -996,20 +1010,21 @@ void audit_get_stamp(struct audit_context *ctx, extern int audit_set_type(struct audit_buffer *ab, int type); -int audit_set_loginuid(struct audit_context *ctx, uid_t loginuid) +int audit_set_loginuid(struct task_struct *task, uid_t loginuid) { - if (ctx) { + if (task->audit_context) { struct audit_buffer *ab; ab = audit_log_start(NULL); if (ab) { audit_log_format(ab, "login pid=%d uid=%u " "old loginuid=%u new loginuid=%u", - ctx->pid, ctx->uid, ctx->loginuid, loginuid); + task->pid, task->uid, + task->audit_context->loginuid, loginuid); audit_set_type(ab, AUDIT_LOGIN); audit_log_end(ab); } - ctx->loginuid = loginuid; + task->audit_context->loginuid = loginuid; } return 0; } diff --git a/kernel/exit.c b/kernel/exit.c index 39d35935b37..edaa50b5bbf 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -27,6 +27,7 @@ #include <linux/mempolicy.h> #include <linux/cpuset.h> #include <linux/syscalls.h> +#include <linux/signal.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -38,6 +39,8 @@ extern struct task_struct *child_reaper; int getrusage(struct task_struct *, int, struct rusage __user *); +static void exit_mm(struct task_struct * tsk); + static void __unhash_process(struct task_struct *p) { nr_threads--; @@ -209,7 +212,7 @@ static inline int has_stopped_jobs(int pgrp) } /** - * reparent_to_init() - Reparent the calling kernel thread to the init task. + * reparent_to_init - Reparent the calling kernel thread to the init task. * * If a kernel thread is launched as a result of a system call, or if * it ever exits, it should generally reparent itself to init so that @@ -277,7 +280,7 @@ void set_special_pids(pid_t session, pid_t pgrp) */ int allow_signal(int sig) { - if (sig < 1 || sig > _NSIG) + if (!valid_signal(sig) || sig < 1) return -EINVAL; spin_lock_irq(¤t->sighand->siglock); @@ -298,7 +301,7 @@ EXPORT_SYMBOL(allow_signal); int disallow_signal(int sig) { - if (sig < 1 || sig > _NSIG) + if (!valid_signal(sig) || sig < 1) return -EINVAL; spin_lock_irq(¤t->sighand->siglock); @@ -473,7 +476,7 @@ EXPORT_SYMBOL_GPL(exit_fs); * Turn us into a lazy TLB process if we * aren't already.. */ -void exit_mm(struct task_struct * tsk) +static void exit_mm(struct task_struct * tsk) { struct mm_struct *mm = tsk->mm; @@ -517,8 +520,6 @@ static inline void choose_new_parent(task_t *p, task_t *reaper, task_t *child_re */ BUG_ON(p == reaper || reaper->exit_state >= EXIT_ZOMBIE); p->real_parent = reaper; - if (p->parent == p->real_parent) - BUG(); } static inline void reparent_thread(task_t *p, task_t *father, int traced) @@ -845,6 +846,8 @@ fastcall NORET_TYPE void do_exit(long code) for (;;) ; } +EXPORT_SYMBOL_GPL(do_exit); + NORET_TYPE void complete_and_exit(struct completion *comp, long code) { if (comp) diff --git a/kernel/futex.c b/kernel/futex.c index 7b54a672d0a..c7130f86106 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -39,6 +39,7 @@ #include <linux/mount.h> #include <linux/pagemap.h> #include <linux/syscalls.h> +#include <linux/signal.h> #define FUTEX_HASHBITS (CONFIG_BASE_SMALL ? 4 : 8) @@ -654,7 +655,7 @@ static int futex_fd(unsigned long uaddr, int signal) int ret, err; ret = -EINVAL; - if (signal < 0 || signal > _NSIG) + if (!valid_signal(signal)) goto out; ret = get_unused_fd(); diff --git a/kernel/itimer.c b/kernel/itimer.c index e9a40e947e0..1dc988e0d2c 100644 --- a/kernel/itimer.c +++ b/kernel/itimer.c @@ -123,7 +123,11 @@ static inline void it_real_arm(struct task_struct *p, unsigned long interval) return; if (interval > (unsigned long) LONG_MAX) interval = LONG_MAX; - p->signal->real_timer.expires = jiffies + interval; + /* the "+ 1" below makes sure that the timer doesn't go off before + * the interval requested. This could happen if + * time requested % (usecs per jiffy) is more than the usecs left + * in the current jiffy */ + p->signal->real_timer.expires = jiffies + interval + 1; add_timer(&p->signal->real_timer); } diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index 1627f8d6e0c..13bcec151b5 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -46,6 +46,14 @@ static inline int is_kernel_inittext(unsigned long addr) return 0; } +static inline int is_kernel_extratext(unsigned long addr) +{ + if (addr >= (unsigned long)_sextratext + && addr <= (unsigned long)_eextratext) + return 1; + return 0; +} + static inline int is_kernel_text(unsigned long addr) { if (addr >= (unsigned long)_stext && addr <= (unsigned long)_etext) @@ -169,8 +177,9 @@ const char *kallsyms_lookup(unsigned long addr, namebuf[0] = 0; if ((all_var && is_kernel(addr)) || - (!all_var && (is_kernel_text(addr) || is_kernel_inittext(addr)))) { - unsigned long symbol_end=0; + (!all_var && (is_kernel_text(addr) || is_kernel_inittext(addr) || + is_kernel_extratext(addr)))) { + unsigned long symbol_end = 0; /* do a binary search on the sorted kallsyms_addresses array */ low = 0; diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 1d5dd1337bd..037142b72a4 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -44,6 +44,7 @@ static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE]; unsigned int kprobe_cpu = NR_CPUS; static DEFINE_SPINLOCK(kprobe_lock); +static struct kprobe *curr_kprobe; /* Locks kprobe: irqs must be disabled */ void lock_kprobes(void) @@ -73,22 +74,139 @@ struct kprobe *get_kprobe(void *addr) return NULL; } +/* + * Aggregate handlers for multiple kprobes support - these handlers + * take care of invoking the individual kprobe handlers on p->list + */ +int aggr_pre_handler(struct kprobe *p, struct pt_regs *regs) +{ + struct kprobe *kp; + + list_for_each_entry(kp, &p->list, list) { + if (kp->pre_handler) { + curr_kprobe = kp; + kp->pre_handler(kp, regs); + curr_kprobe = NULL; + } + } + return 0; +} + +void aggr_post_handler(struct kprobe *p, struct pt_regs *regs, + unsigned long flags) +{ + struct kprobe *kp; + + list_for_each_entry(kp, &p->list, list) { + if (kp->post_handler) { + curr_kprobe = kp; + kp->post_handler(kp, regs, flags); + curr_kprobe = NULL; + } + } + return; +} + +int aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr) +{ + /* + * if we faulted "during" the execution of a user specified + * probe handler, invoke just that probe's fault handler + */ + if (curr_kprobe && curr_kprobe->fault_handler) { + if (curr_kprobe->fault_handler(curr_kprobe, regs, trapnr)) + return 1; + } + return 0; +} + +/* + * Fill in the required fields of the "manager kprobe". Replace the + * earlier kprobe in the hlist with the manager kprobe + */ +static inline void add_aggr_kprobe(struct kprobe *ap, struct kprobe *p) +{ + ap->addr = p->addr; + ap->opcode = p->opcode; + memcpy(&ap->ainsn, &p->ainsn, sizeof(struct arch_specific_insn)); + + ap->pre_handler = aggr_pre_handler; + ap->post_handler = aggr_post_handler; + ap->fault_handler = aggr_fault_handler; + + INIT_LIST_HEAD(&ap->list); + list_add(&p->list, &ap->list); + + INIT_HLIST_NODE(&ap->hlist); + hlist_del(&p->hlist); + hlist_add_head(&ap->hlist, + &kprobe_table[hash_ptr(ap->addr, KPROBE_HASH_BITS)]); +} + +/* + * This is the second or subsequent kprobe at the address - handle + * the intricacies + * TODO: Move kcalloc outside the spinlock + */ +static int register_aggr_kprobe(struct kprobe *old_p, struct kprobe *p) +{ + int ret = 0; + struct kprobe *ap; + + if (old_p->break_handler || p->break_handler) { + ret = -EEXIST; /* kprobe and jprobe can't (yet) coexist */ + } else if (old_p->pre_handler == aggr_pre_handler) { + list_add(&p->list, &old_p->list); + } else { + ap = kcalloc(1, sizeof(struct kprobe), GFP_ATOMIC); + if (!ap) + return -ENOMEM; + add_aggr_kprobe(ap, old_p); + list_add(&p->list, &ap->list); + } + return ret; +} + +/* kprobe removal house-keeping routines */ +static inline void cleanup_kprobe(struct kprobe *p, unsigned long flags) +{ + *p->addr = p->opcode; + hlist_del(&p->hlist); + flush_icache_range((unsigned long) p->addr, + (unsigned long) p->addr + sizeof(kprobe_opcode_t)); + spin_unlock_irqrestore(&kprobe_lock, flags); + arch_remove_kprobe(p); +} + +static inline void cleanup_aggr_kprobe(struct kprobe *old_p, + struct kprobe *p, unsigned long flags) +{ + list_del(&p->list); + if (list_empty(&old_p->list)) { + cleanup_kprobe(old_p, flags); + kfree(old_p); + } else + spin_unlock_irqrestore(&kprobe_lock, flags); +} + int register_kprobe(struct kprobe *p) { int ret = 0; unsigned long flags = 0; + struct kprobe *old_p; if ((ret = arch_prepare_kprobe(p)) != 0) { goto rm_kprobe; } spin_lock_irqsave(&kprobe_lock, flags); - INIT_HLIST_NODE(&p->hlist); - if (get_kprobe(p->addr)) { - ret = -EEXIST; + old_p = get_kprobe(p->addr); + if (old_p) { + ret = register_aggr_kprobe(old_p, p); goto out; } - arch_copy_kprobe(p); + arch_copy_kprobe(p); + INIT_HLIST_NODE(&p->hlist); hlist_add_head(&p->hlist, &kprobe_table[hash_ptr(p->addr, KPROBE_HASH_BITS)]); @@ -107,13 +225,17 @@ rm_kprobe: void unregister_kprobe(struct kprobe *p) { unsigned long flags; - arch_remove_kprobe(p); + struct kprobe *old_p; + spin_lock_irqsave(&kprobe_lock, flags); - *p->addr = p->opcode; - hlist_del(&p->hlist); - flush_icache_range((unsigned long) p->addr, - (unsigned long) p->addr + sizeof(kprobe_opcode_t)); - spin_unlock_irqrestore(&kprobe_lock, flags); + old_p = get_kprobe(p->addr); + if (old_p) { + if (old_p->pre_handler == aggr_pre_handler) + cleanup_aggr_kprobe(old_p, p, flags); + else + cleanup_kprobe(p, flags); + } else + spin_unlock_irqrestore(&kprobe_lock, flags); } static struct notifier_block kprobe_exceptions_nb = { diff --git a/kernel/kthread.c b/kernel/kthread.c index e377e224410..f50f174e92d 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -174,7 +174,7 @@ int kthread_stop(struct task_struct *k) /* Must init completion *before* thread sees kthread_stop_info.k */ init_completion(&kthread_stop_info.done); - wmb(); + smp_wmb(); /* Now set kthread_should_stop() to true, and wake it up. */ kthread_stop_info.k = k; diff --git a/kernel/module.c b/kernel/module.c index 2dbfa0773fa..5734ab09d3f 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1801,7 +1801,7 @@ sys_init_module(void __user *umod, /* Init routine failed: abort. Try to protect us from buggy refcounters. */ mod->state = MODULE_STATE_GOING; - synchronize_kernel(); + synchronize_sched(); if (mod->unsafe) printk(KERN_ERR "%s: module is now stuck!\n", mod->name); diff --git a/kernel/panic.c b/kernel/panic.c index 0fa3f3a66fb..081f7465fc8 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -102,9 +102,9 @@ NORET_TYPE void panic(const char * fmt, ...) #ifdef __sparc__ { extern int stop_a_enabled; - /* Make sure the user can actually press L1-A */ + /* Make sure the user can actually press Stop-A (L1-A) */ stop_a_enabled = 1; - printk(KERN_EMERG "Press L1-A to return to the boot prom\n"); + printk(KERN_EMERG "Press Stop-A (L1-A) to return to the boot prom\n"); } #endif #if defined(CONFIG_ARCH_S390) diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index ae5bebc3b18..90b3b68dee3 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -1099,7 +1099,7 @@ static struct pbe * swsusp_pagedir_relocate(struct pbe *pblist) return pblist; } -/** +/* * Using bio to read from swap. * This code requires a bit more work than just using buffer heads * but, it is the recommended way for 2.5/2.6. diff --git a/kernel/printk.c b/kernel/printk.c index 1498689548d..290a07ce2c8 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -85,10 +85,6 @@ static int console_locked; */ static DEFINE_SPINLOCK(logbuf_lock); -static char __log_buf[__LOG_BUF_LEN]; -static char *log_buf = __log_buf; -static int log_buf_len = __LOG_BUF_LEN; - #define LOG_BUF_MASK (log_buf_len-1) #define LOG_BUF(idx) (log_buf[(idx) & LOG_BUF_MASK]) @@ -99,7 +95,6 @@ static int log_buf_len = __LOG_BUF_LEN; static unsigned long log_start; /* Index into log_buf: next char to be read by syslog() */ static unsigned long con_start; /* Index into log_buf: next char to be sent to consoles */ static unsigned long log_end; /* Index into log_buf: most-recently-written-char + 1 */ -static unsigned long logged_chars; /* Number of chars produced since last read+clear operation */ /* * Array of consoles built from command line options (console=) @@ -120,6 +115,13 @@ static int preferred_console = -1; /* Flag: console code may call schedule() */ static int console_may_schedule; +#ifdef CONFIG_PRINTK + +static char __log_buf[__LOG_BUF_LEN]; +static char *log_buf = __log_buf; +static int log_buf_len = __LOG_BUF_LEN; +static unsigned long logged_chars; /* Number of chars produced since last read+clear operation */ + /* * Setup a list of consoles. Called from init/main.c */ @@ -535,6 +537,7 @@ __setup("time", printk_time_setup); * then changes console_loglevel may break. This is because console_loglevel * is inspected when the actual printing occurs. */ + asmlinkage int printk(const char *fmt, ...) { va_list args; @@ -655,6 +658,18 @@ out: EXPORT_SYMBOL(printk); EXPORT_SYMBOL(vprintk); +#else + +asmlinkage long sys_syslog(int type, char __user * buf, int len) +{ + return 0; +} + +int do_syslog(int type, char __user * buf, int len) { return 0; } +static void call_console_drivers(unsigned long start, unsigned long end) {} + +#endif + /** * acquire_console_sem - lock the console system for exclusive use. * @@ -931,7 +946,7 @@ int unregister_console(struct console * console) return res; } EXPORT_SYMBOL(unregister_console); - + /** * tty_write_message - write a message to a certain tty, not just the console. * diff --git a/kernel/profile.c b/kernel/profile.c index a38fa70075f..0221a50ca86 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -184,7 +184,7 @@ void unregister_timer_hook(int (*hook)(struct pt_regs *)) WARN_ON(hook != timer_hook); timer_hook = NULL; /* make sure all CPUs see the NULL hook */ - synchronize_kernel(); + synchronize_sched(); /* Allow ongoing interrupts to complete. */ } EXPORT_SYMBOL_GPL(register_timer_hook); @@ -522,7 +522,7 @@ static int __init create_hash_tables(void) return 0; out_cleanup: prof_on = 0; - mb(); + smp_mb(); on_each_cpu(profile_nop, NULL, 0, 1); for_each_online_cpu(cpu) { struct page *page; diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 88b306c4e84..8dcb8f6288b 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -16,6 +16,7 @@ #include <linux/smp_lock.h> #include <linux/ptrace.h> #include <linux/security.h> +#include <linux/signal.h> #include <asm/pgtable.h> #include <asm/uaccess.h> @@ -135,7 +136,7 @@ int ptrace_attach(struct task_struct *task) (current->gid != task->sgid) || (current->gid != task->gid)) && !capable(CAP_SYS_PTRACE)) goto bad; - rmb(); + smp_rmb(); if (!task->mm->dumpable && !capable(CAP_SYS_PTRACE)) goto bad; /* the same process cannot be attached many times */ @@ -166,7 +167,7 @@ bad: int ptrace_detach(struct task_struct *child, unsigned int data) { - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) return -EIO; /* Architecture-specific hardware disable .. */ diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index d00eded75d7..f436993bd59 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -444,15 +444,18 @@ static void wakeme_after_rcu(struct rcu_head *head) } /** - * synchronize_kernel - wait until a grace period has elapsed. + * synchronize_rcu - wait until a grace period has elapsed. * * Control will return to the caller some time after a full grace * period has elapsed, in other words after all currently executing RCU * read-side critical sections have completed. RCU read-side critical * sections are delimited by rcu_read_lock() and rcu_read_unlock(), * and may be nested. + * + * If your read-side code is not protected by rcu_read_lock(), do -not- + * use synchronize_rcu(). */ -void synchronize_kernel(void) +void synchronize_rcu(void) { struct rcu_synchronize rcu; @@ -464,7 +467,16 @@ void synchronize_kernel(void) wait_for_completion(&rcu.completion); } +/* + * Deprecated, use synchronize_rcu() or synchronize_sched() instead. + */ +void synchronize_kernel(void) +{ + synchronize_rcu(); +} + module_param(maxbatch, int, 0); -EXPORT_SYMBOL_GPL(call_rcu); -EXPORT_SYMBOL_GPL(call_rcu_bh); -EXPORT_SYMBOL_GPL(synchronize_kernel); +EXPORT_SYMBOL(call_rcu); /* WARNING: GPL-only in April 2006. */ +EXPORT_SYMBOL(call_rcu_bh); /* WARNING: GPL-only in April 2006. */ +EXPORT_SYMBOL_GPL(synchronize_rcu); +EXPORT_SYMBOL(synchronize_kernel); /* WARNING: GPL-only in April 2006. */ diff --git a/kernel/sched.c b/kernel/sched.c index 9bb7489ee64..0dc3158667a 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2906,6 +2906,7 @@ static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, * @q: the waitqueue * @mode: which threads * @nr_exclusive: how many wake-one or wake-many threads to wake up + * @key: is directly passed to the wakeup function */ void fastcall __wake_up(wait_queue_head_t *q, unsigned int mode, int nr_exclusive, void *key) @@ -2928,7 +2929,7 @@ void fastcall __wake_up_locked(wait_queue_head_t *q, unsigned int mode) } /** - * __wake_up - sync- wake up threads blocked on a waitqueue. + * __wake_up_sync - wake up threads blocked on a waitqueue. * @q: the waitqueue * @mode: which threads * @nr_exclusive: how many wake-one or wake-many threads to wake up @@ -3223,6 +3224,19 @@ out_unlock: EXPORT_SYMBOL(set_user_nice); +/* + * can_nice - check if a task can reduce its nice value + * @p: task + * @nice: nice value + */ +int can_nice(const task_t *p, const int nice) +{ + /* convert nice value [19,-20] to rlimit style value [0,39] */ + int nice_rlim = 19 - nice; + return (nice_rlim <= p->signal->rlim[RLIMIT_NICE].rlim_cur || + capable(CAP_SYS_NICE)); +} + #ifdef __ARCH_WANT_SYS_NICE /* @@ -3242,12 +3256,8 @@ asmlinkage long sys_nice(int increment) * We don't have to worry. Conceptually one call occurs first * and we have a single winner. */ - if (increment < 0) { - if (!capable(CAP_SYS_NICE)) - return -EPERM; - if (increment < -40) - increment = -40; - } + if (increment < -40) + increment = -40; if (increment > 40) increment = 40; @@ -3257,6 +3267,9 @@ asmlinkage long sys_nice(int increment) if (nice > 19) nice = 19; + if (increment < 0 && !can_nice(current, nice)) + return -EPERM; + retval = security_task_setnice(current, nice); if (retval) return retval; @@ -3372,6 +3385,7 @@ recheck: return -EINVAL; if ((policy == SCHED_FIFO || policy == SCHED_RR) && + param->sched_priority > p->signal->rlim[RLIMIT_RTPRIO].rlim_cur && !capable(CAP_SYS_NICE)) return -EPERM; if ((current->euid != p->euid) && (current->euid != p->uid) && diff --git a/kernel/signal.c b/kernel/signal.c index e6567d7f2b6..8f3debc77c5 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -23,6 +23,7 @@ #include <linux/syscalls.h> #include <linux/ptrace.h> #include <linux/posix-timers.h> +#include <linux/signal.h> #include <asm/param.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -646,7 +647,7 @@ static int check_kill_permission(int sig, struct siginfo *info, struct task_struct *t) { int error = -EINVAL; - if (sig < 0 || sig > _NSIG) + if (!valid_signal(sig)) return error; error = -EPERM; if ((!info || ((unsigned long)info != 1 && @@ -1245,7 +1246,7 @@ send_sig_info(int sig, struct siginfo *info, struct task_struct *p) * Make sure legacy kernel users don't send in bad values * (normal paths check this in check_kill_permission). */ - if (sig < 0 || sig > _NSIG) + if (!valid_signal(sig)) return -EINVAL; /* @@ -1520,7 +1521,7 @@ void do_notify_parent(struct task_struct *tsk, int sig) if (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN) sig = 0; } - if (sig > 0 && sig <= _NSIG) + if (valid_signal(sig) && sig > 0) __group_send_sig_info(sig, &info, tsk->parent); __wake_up_parent(tsk, tsk->parent); spin_unlock_irqrestore(&psig->siglock, flags); @@ -2364,7 +2365,7 @@ do_sigaction(int sig, const struct k_sigaction *act, struct k_sigaction *oact) { struct k_sigaction *k; - if (sig < 1 || sig > _NSIG || (act && sig_kernel_only(sig))) + if (!valid_signal(sig) || sig < 1 || (act && sig_kernel_only(sig))) return -EINVAL; k = ¤t->sighand->action[sig-1]; diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index c39ed70af17..6116b25aa7c 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -33,7 +33,7 @@ static int stopmachine(void *cpu) set_cpus_allowed(current, cpumask_of_cpu((int)(long)cpu)); /* Ack: we are alive */ - mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */ + smp_mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */ atomic_inc(&stopmachine_thread_ack); /* Simple state machine */ @@ -43,14 +43,14 @@ static int stopmachine(void *cpu) local_irq_disable(); irqs_disabled = 1; /* Ack: irqs disabled. */ - mb(); /* Must read state first. */ + smp_mb(); /* Must read state first. */ atomic_inc(&stopmachine_thread_ack); } else if (stopmachine_state == STOPMACHINE_PREPARE && !prepared) { /* Everyone is in place, hold CPU. */ preempt_disable(); prepared = 1; - mb(); /* Must read state first. */ + smp_mb(); /* Must read state first. */ atomic_inc(&stopmachine_thread_ack); } /* Yield in first stage: migration threads need to @@ -62,7 +62,7 @@ static int stopmachine(void *cpu) } /* Ack: we are exiting. */ - mb(); /* Must read state first. */ + smp_mb(); /* Must read state first. */ atomic_inc(&stopmachine_thread_ack); if (irqs_disabled) @@ -77,7 +77,7 @@ static int stopmachine(void *cpu) static void stopmachine_set_state(enum stopmachine_state state) { atomic_set(&stopmachine_thread_ack, 0); - wmb(); + smp_wmb(); stopmachine_state = state; while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) cpu_relax(); diff --git a/kernel/sys.c b/kernel/sys.c index 462d78d5589..f006632c2ba 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -25,6 +25,7 @@ #include <linux/dcookies.h> #include <linux/suspend.h> #include <linux/tty.h> +#include <linux/signal.h> #include <linux/compat.h> #include <linux/syscalls.h> @@ -227,7 +228,7 @@ static int set_one_prio(struct task_struct *p, int niceval, int error) error = -EPERM; goto out; } - if (niceval < task_nice(p) && !capable(CAP_SYS_NICE)) { + if (niceval < task_nice(p) && !can_nice(p, niceval)) { error = -EACCES; goto out; } @@ -525,7 +526,7 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) if (new_egid != old_egid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } if (rgid != (gid_t) -1 || (egid != (gid_t) -1 && egid != old_rgid)) @@ -556,7 +557,7 @@ asmlinkage long sys_setgid(gid_t gid) if(old_egid != gid) { current->mm->dumpable=0; - wmb(); + smp_wmb(); } current->gid = current->egid = current->sgid = current->fsgid = gid; } @@ -565,7 +566,7 @@ asmlinkage long sys_setgid(gid_t gid) if(old_egid != gid) { current->mm->dumpable=0; - wmb(); + smp_wmb(); } current->egid = current->fsgid = gid; } @@ -596,7 +597,7 @@ static int set_user(uid_t new_ruid, int dumpclear) if(dumpclear) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->uid = new_ruid; return 0; @@ -653,7 +654,7 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) if (new_euid != old_euid) { current->mm->dumpable=0; - wmb(); + smp_wmb(); } current->fsuid = current->euid = new_euid; if (ruid != (uid_t) -1 || @@ -703,7 +704,7 @@ asmlinkage long sys_setuid(uid_t uid) if (old_euid != uid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->fsuid = current->euid = uid; current->suid = new_suid; @@ -748,7 +749,7 @@ asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid) if (euid != current->euid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->euid = euid; } @@ -798,7 +799,7 @@ asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid) if (egid != current->egid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->egid = egid; } @@ -845,7 +846,7 @@ asmlinkage long sys_setfsuid(uid_t uid) if (uid != old_fsuid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->fsuid = uid; } @@ -875,7 +876,7 @@ asmlinkage long sys_setfsgid(gid_t gid) if (gid != old_fsgid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->fsgid = gid; key_fsgid_changed(current); @@ -1194,7 +1195,7 @@ static int groups_from_user(struct group_info *group_info, return 0; } -/* a simple shell-metzner sort */ +/* a simple Shell sort */ static void groups_sort(struct group_info *group_info) { int base, max, stride; @@ -1637,7 +1638,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, switch (option) { case PR_SET_PDEATHSIG: sig = arg2; - if (sig < 0 || sig > _NSIG) { + if (!valid_signal(sig)) { error = -EINVAL; break; } diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 1802a311dd3..0dda70ed1f9 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -52,6 +52,7 @@ cond_syscall(sys_msgsnd); cond_syscall(sys_msgrcv); cond_syscall(sys_msgctl); cond_syscall(sys_shmget); +cond_syscall(sys_shmat); cond_syscall(sys_shmdt); cond_syscall(sys_shmctl); cond_syscall(sys_mq_open); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 79dbd93bd69..701d12c6306 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1991,6 +1991,8 @@ int proc_dointvec_userhz_jiffies(ctl_table *table, int write, struct file *filp, * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer + * @ppos: file position + * @ppos: the current position in the file * * Reads/writes up to table->maxlen/sizeof(unsigned int) integer * values from/to the user buffer, treated as an ASCII string. diff --git a/kernel/time.c b/kernel/time.c index 96fd0f49963..d4335c1c884 100644 --- a/kernel/time.c +++ b/kernel/time.c @@ -516,14 +516,6 @@ int do_settimeofday (struct timespec *tv) write_seqlock_irq(&xtime_lock); { - /* - * This is revolting. We need to set "xtime" correctly. However, the value - * in this location is the value at the most recent update of wall time. - * Discover what correction gettimeofday would have done, and then undo - * it! - */ - nsec -= time_interpolator_get_offset(); - wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec); wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec); diff --git a/kernel/timer.c b/kernel/timer.c index ecb3d67c0e1..207aa4f0aa1 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -1007,7 +1007,7 @@ asmlinkage long sys_getppid(void) * Make sure we read the pid before re-reading the * parent pointer: */ - rmb(); + smp_rmb(); parent = me->group_leader->real_parent; if (old != parent) continue; diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 426a0cf7b11..ac23847ce0e 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -108,6 +108,7 @@ config DEBUG_HIGHMEM config DEBUG_BUGVERBOSE bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EMBEDDED + depends on BUG depends on ARM || ARM26 || M32R || M68K || SPARC32 || SPARC64 || (X86 && !X86_64) || FRV default !EMBEDDED help diff --git a/lib/kobject.c b/lib/kobject.c index 5df8441c44e..94048826624 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -216,13 +216,12 @@ int kobject_register(struct kobject * kobj) /** * kobject_set_name - Set the name of an object * @kobj: object. - * @name: name. + * @fmt: format string used to build the name * * If strlen(name) >= KOBJ_NAME_LEN, then use a dynamically allocated * string that @kobj->k_name points to. Otherwise, use the static * @kobj->name array. */ - int kobject_set_name(struct kobject * kobj, const char * fmt, ...) { int error = 0; diff --git a/lib/rwsem-spinlock.c b/lib/rwsem-spinlock.c index 21f0db2c971..40ffde940a8 100644 --- a/lib/rwsem-spinlock.c +++ b/lib/rwsem-spinlock.c @@ -76,7 +76,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wakewrite) list_del(&waiter->list); tsk = waiter->task; /* Don't touch waiter after ->task has been NULLed */ - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); @@ -91,7 +91,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wakewrite) list_del(&waiter->list); tsk = waiter->task; - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); @@ -123,7 +123,7 @@ __rwsem_wake_one_writer(struct rw_semaphore *sem) list_del(&waiter->list); tsk = waiter->task; - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); diff --git a/lib/rwsem.c b/lib/rwsem.c index 7644089ec8f..62fa4eba9ff 100644 --- a/lib/rwsem.c +++ b/lib/rwsem.c @@ -74,7 +74,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int downgrading) */ list_del(&waiter->list); tsk = waiter->task; - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); @@ -117,7 +117,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int downgrading) waiter = list_entry(next, struct rwsem_waiter, list); next = waiter->list.next; tsk = waiter->task; - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); diff --git a/lib/sort.c b/lib/sort.c index ea3caedeabd..b73dbb0e7c8 100644 --- a/lib/sort.c +++ b/lib/sort.c @@ -90,7 +90,7 @@ int cmpint(const void *a, const void *b) static int sort_test(void) { - int *a, i, r = 0; + int *a, i, r = 1; a = kmalloc(1000 * sizeof(int), GFP_KERNEL); BUG_ON(!a); diff --git a/lib/string.c b/lib/string.c index 4bb93ad23c6..d886ef157c1 100644 --- a/lib/string.c +++ b/lib/string.c @@ -65,6 +65,7 @@ EXPORT_SYMBOL(strnicmp); * @dest: Where to copy the string to * @src: Where to copy the string from */ +#undef strcpy char * strcpy(char * dest,const char *src) { char *tmp = dest; @@ -85,6 +86,10 @@ EXPORT_SYMBOL(strcpy); * * The result is not %NUL-terminated if the source exceeds * @count bytes. + * + * In the case where the length of @src is less than that of + * count, the remainder of @dest will be padded with %NUL. + * */ char * strncpy(char * dest,const char *src,size_t count) { @@ -132,6 +137,7 @@ EXPORT_SYMBOL(strlcpy); * @dest: The string to be appended to * @src: The string to append to it */ +#undef strcat char * strcat(char * dest, const char * src) { char *tmp = dest; @@ -209,6 +215,7 @@ EXPORT_SYMBOL(strlcat); * @cs: One string * @ct: Another string */ +#undef strcmp int strcmp(const char * cs,const char * ct) { register signed char __res; @@ -514,6 +521,7 @@ EXPORT_SYMBOL(memmove); * @ct: Another area of memory * @count: The size of the area. */ +#undef memcmp int memcmp(const void * cs,const void * ct,size_t count) { const unsigned char *su1, *su2; diff --git a/mm/filemap.c b/mm/filemap.c index 93595c327bb..47263ac3e4e 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -29,11 +29,6 @@ #include <linux/security.h> #include <linux/syscalls.h> /* - * This is needed for the following functions: - * - try_to_release_page - * - block_invalidatepage - * - generic_osync_inode - * * FIXME: remove all knowledge of the buffer layer from the core VM */ #include <linux/buffer_head.h> /* for generic_osync_inode */ @@ -123,8 +118,7 @@ void remove_from_page_cache(struct page *page) { struct address_space *mapping = page->mapping; - if (unlikely(!PageLocked(page))) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); write_lock_irq(&mapping->tree_lock); __remove_from_page_cache(page); @@ -139,7 +133,25 @@ static int sync_page(void *word) page = container_of((page_flags_t *)word, struct page, flags); /* - * FIXME, fercrissake. What is this barrier here for? + * page_mapping() is being called without PG_locked held. + * Some knowledge of the state and use of the page is used to + * reduce the requirements down to a memory barrier. + * The danger here is of a stale page_mapping() return value + * indicating a struct address_space different from the one it's + * associated with when it is associated with one. + * After smp_mb(), it's either the correct page_mapping() for + * the page, or an old page_mapping() and the page's own + * page_mapping() has gone NULL. + * The ->sync_page() address_space operation must tolerate + * page_mapping() going NULL. By an amazing coincidence, + * this comes about because none of the users of the page + * in the ->sync_page() methods make essential use of the + * page_mapping(), merely passing the page down to the backing + * device's unplug functions when it's non-NULL, which in turn + * ignore it for all cases but swap, where only page->private is + * of interest. When page_mapping() does go NULL, the entire + * call stack gracefully ignores the page and returns. + * -- wli */ smp_mb(); mapping = page_mapping(page); @@ -152,9 +164,10 @@ static int sync_page(void *word) /** * filemap_fdatawrite_range - start writeback against all of a mapping's * dirty pages that lie within the byte offsets <start, end> - * @mapping: address space structure to write - * @start: offset in bytes where the range starts - * @end : offset in bytes where the range ends + * @mapping: address space structure to write + * @start: offset in bytes where the range starts + * @end: offset in bytes where the range ends + * @sync_mode: enable synchronous operation * * If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as * opposed to a regular memory * cleansing writeback. The difference between @@ -518,8 +531,8 @@ EXPORT_SYMBOL(find_trylock_page); /** * find_lock_page - locate, pin and lock a pagecache page * - * @mapping - the address_space to search - * @offset - the page index + * @mapping: the address_space to search + * @offset: the page index * * Locates the desired pagecache page, locks it, increments its reference * count and returns its address. @@ -558,9 +571,9 @@ EXPORT_SYMBOL(find_lock_page); /** * find_or_create_page - locate or add a pagecache page * - * @mapping - the page's address_space - * @index - the page's index into the mapping - * @gfp_mask - page allocation mode + * @mapping: the page's address_space + * @index: the page's index into the mapping + * @gfp_mask: page allocation mode * * Locates a page in the pagecache. If the page is not present, a new page * is allocated using @gfp_mask and is added to the pagecache and to the VM's @@ -1949,7 +1962,7 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, buf = iov->iov_base + written; else { filemap_set_next_iovec(&cur_iov, &iov_base, written); - buf = iov->iov_base + iov_base; + buf = cur_iov->iov_base + iov_base; } do { @@ -2007,9 +2020,11 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, count -= status; pos += status; buf += status; - if (unlikely(nr_segs > 1)) + if (unlikely(nr_segs > 1)) { filemap_set_next_iovec(&cur_iov, &iov_base, status); + buf = cur_iov->iov_base + iov_base; + } } } if (unlikely(copied != bytes)) diff --git a/mm/highmem.c b/mm/highmem.c index d01276506b0..40091159946 100644 --- a/mm/highmem.c +++ b/mm/highmem.c @@ -325,6 +325,7 @@ static void bounce_end_io(struct bio *bio, mempool_t *pool, int err) continue; mempool_free(bvec->bv_page, pool); + dec_page_state(nr_bounce); } bio_endio(bio_orig, bio_orig->bi_size, err); @@ -405,6 +406,7 @@ static void __blk_queue_bounce(request_queue_t *q, struct bio **bio_orig, to->bv_page = mempool_alloc(pool, q->bounce_gfp); to->bv_len = from->bv_len; to->bv_offset = from->bv_offset; + inc_page_state(nr_bounce); if (rw == WRITE) { char *vto, *vfrom; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index a3b44a671ce..08c41da429c 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -661,7 +661,7 @@ static struct zonelist *zonelist_policy(unsigned int __nocast gfp, struct mempol case MPOL_BIND: /* Lower zones don't get a policy applied */ /* Careful: current->mems_allowed might have moved */ - if (gfp >= policy_zone) + if ((gfp & GFP_ZONEMASK) >= policy_zone) if (cpuset_zonelist_valid_mems_allowed(policy->v.zonelist)) return policy->v.zonelist; /*FALL THROUGH*/ diff --git a/mm/mempool.c b/mm/mempool.c index b014ffeaa41..c9f3d462042 100644 --- a/mm/mempool.c +++ b/mm/mempool.c @@ -198,31 +198,22 @@ void * mempool_alloc(mempool_t *pool, unsigned int __nocast gfp_mask) void *element; unsigned long flags; DEFINE_WAIT(wait); - int gfp_nowait = gfp_mask & ~(__GFP_WAIT | __GFP_IO); + int gfp_temp; might_sleep_if(gfp_mask & __GFP_WAIT); + + gfp_mask |= __GFP_NOMEMALLOC; /* don't allocate emergency reserves */ + gfp_mask |= __GFP_NORETRY; /* don't loop in __alloc_pages */ + gfp_mask |= __GFP_NOWARN; /* failures are OK */ + + gfp_temp = gfp_mask & ~(__GFP_WAIT|__GFP_IO); + repeat_alloc: - element = pool->alloc(gfp_nowait|__GFP_NOWARN, pool->pool_data); + + element = pool->alloc(gfp_temp, pool->pool_data); if (likely(element != NULL)) return element; - /* - * If the pool is less than 50% full and we can perform effective - * page reclaim then try harder to allocate an element. - */ - mb(); - if ((gfp_mask & __GFP_FS) && (gfp_mask != gfp_nowait) && - (pool->curr_nr <= pool->min_nr/2)) { - element = pool->alloc(gfp_mask, pool->pool_data); - if (likely(element != NULL)) - return element; - } - - /* - * Kick the VM at this point. - */ - wakeup_bdflush(0); - spin_lock_irqsave(&pool->lock, flags); if (likely(pool->curr_nr)) { element = remove_element(pool); @@ -235,8 +226,10 @@ repeat_alloc: if (!(gfp_mask & __GFP_WAIT)) return NULL; + /* Now start performing page reclaim */ + gfp_temp = gfp_mask; prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE); - mb(); + smp_mb(); if (!pool->curr_nr) io_schedule(); finish_wait(&pool->wait, &wait); @@ -257,7 +250,7 @@ void mempool_free(void *element, mempool_t *pool) { unsigned long flags; - mb(); + smp_mb(); if (pool->curr_nr < pool->min_nr) { spin_lock_irqsave(&pool->lock, flags); if (pool->curr_nr < pool->min_nr) { diff --git a/mm/mmap.c b/mm/mmap.c index 6ea204cc751..01f9793591f 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -937,9 +937,10 @@ unsigned long do_mmap_pgoff(struct file * file, unsigned long addr, /* mlock MCL_FUTURE? */ if (vm_flags & VM_LOCKED) { unsigned long locked, lock_limit; - locked = mm->locked_vm << PAGE_SHIFT; + locked = len >> PAGE_SHIFT; + locked += mm->locked_vm; lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; - locked += len; + lock_limit >>= PAGE_SHIFT; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; } @@ -1009,8 +1010,7 @@ munmap_back: } /* Check against address space limit. */ - if ((mm->total_vm << PAGE_SHIFT) + len - > current->signal->rlim[RLIMIT_AS].rlim_cur) + if (!may_expand_vm(mm, len >> PAGE_SHIFT)) return -ENOMEM; if (accountable && (!(flags & MAP_NORESERVE) || @@ -1421,7 +1421,7 @@ static int acct_stack_growth(struct vm_area_struct * vma, unsigned long size, un struct rlimit *rlim = current->signal->rlim; /* address space limit tests */ - if (mm->total_vm + grow > rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT) + if (!may_expand_vm(mm, grow)) return -ENOMEM; /* Stack limit test */ @@ -1823,9 +1823,10 @@ unsigned long do_brk(unsigned long addr, unsigned long len) */ if (mm->def_flags & VM_LOCKED) { unsigned long locked, lock_limit; - locked = mm->locked_vm << PAGE_SHIFT; + locked = len >> PAGE_SHIFT; + locked += mm->locked_vm; lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; - locked += len; + lock_limit >>= PAGE_SHIFT; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; } @@ -1848,8 +1849,7 @@ unsigned long do_brk(unsigned long addr, unsigned long len) } /* Check against address space limits *after* clearing old maps... */ - if ((mm->total_vm << PAGE_SHIFT) + len - > current->signal->rlim[RLIMIT_AS].rlim_cur) + if (!may_expand_vm(mm, len >> PAGE_SHIFT)) return -ENOMEM; if (mm->map_count > sysctl_max_map_count) @@ -2019,3 +2019,19 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, } return new_vma; } + +/* + * Return true if the calling process may expand its vm space by the passed + * number of pages + */ +int may_expand_vm(struct mm_struct *mm, unsigned long npages) +{ + unsigned long cur = mm->total_vm; /* pages */ + unsigned long lim; + + lim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT; + + if (cur + npages > lim) + return 0; + return 1; +} diff --git a/mm/mremap.c b/mm/mremap.c index 0d1c1b9c7a0..0dd7ace94e5 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -347,10 +347,10 @@ unsigned long do_mremap(unsigned long addr, if (locked > lock_limit && !capable(CAP_IPC_LOCK)) goto out; } - ret = -ENOMEM; - if ((current->mm->total_vm << PAGE_SHIFT) + (new_len - old_len) - > current->signal->rlim[RLIMIT_AS].rlim_cur) + if (!may_expand_vm(current->mm, (new_len - old_len) >> PAGE_SHIFT)) { + ret = -ENOMEM; goto out; + } if (vma->vm_flags & VM_ACCOUNT) { charged = (new_len - old_len) >> PAGE_SHIFT; diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 6ddd6a29c73..613b99a5591 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -255,7 +255,7 @@ static void balance_dirty_pages(struct address_space *mapping) /** * balance_dirty_pages_ratelimited - balance dirty memory state - * @mapping - address_space which was dirtied + * @mapping: address_space which was dirtied * * Processes which are dirtying memory should call in here once for each page * which was newly dirtied. The function will periodically check the system's @@ -562,8 +562,8 @@ int do_writepages(struct address_space *mapping, struct writeback_control *wbc) /** * write_one_page - write out a single page and optionally wait on I/O * - * @page - the page to write - * @wait - if true, wait on writeout + * @page: the page to write + * @wait: if true, wait on writeout * * The page must be locked by the caller and will be unlocked upon return. * diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c73dbbc1cd8..b1061b1962f 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -43,7 +43,9 @@ * initializer cleaner */ nodemask_t node_online_map = { { [0] = 1UL } }; +EXPORT_SYMBOL(node_online_map); nodemask_t node_possible_map = NODE_MASK_ALL; +EXPORT_SYMBOL(node_possible_map); struct pglist_data *pgdat_list; unsigned long totalram_pages; unsigned long totalhigh_pages; @@ -799,14 +801,18 @@ __alloc_pages(unsigned int __nocast gfp_mask, unsigned int order, } /* This allocation should allow future memory freeing. */ - if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE))) && !in_interrupt()) { - /* go through the zonelist yet again, ignoring mins */ - for (i = 0; (z = zones[i]) != NULL; i++) { - if (!cpuset_zone_allowed(z)) - continue; - page = buffered_rmqueue(z, order, gfp_mask); - if (page) - goto got_pg; + + if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE))) + && !in_interrupt()) { + if (!(gfp_mask & __GFP_NOMEMALLOC)) { + /* go through the zonelist yet again, ignoring mins */ + for (i = 0; (z = zones[i]) != NULL; i++) { + if (!cpuset_zone_allowed(z)) + continue; + page = buffered_rmqueue(z, order, gfp_mask); + if (page) + goto got_pg; + } } goto nopage; } @@ -1351,8 +1357,7 @@ static int __init build_zonelists_node(pg_data_t *pgdat, struct zonelist *zoneli #define MAX_NODE_LOAD (num_online_nodes()) static int __initdata node_load[MAX_NUMNODES]; /** - * find_next_best_node - find the next node that should appear in a given - * node's fallback list + * find_next_best_node - find the next node that should appear in a given node's fallback list * @node: node whose fallback list we're appending * @used_node_mask: nodemask_t of already used nodes * @@ -1671,6 +1676,18 @@ static void __init free_area_init_core(struct pglist_data *pgdat, if (batch < 1) batch = 1; + /* + * Clamp the batch to a 2^n - 1 value. Having a power + * of 2 value was found to be more likely to have + * suboptimal cache aliasing properties in some cases. + * + * For example if 2 tasks are alternately allocating + * batches of pages, one task can end up with a lot + * of pages of one half of the possible page colors + * and the other with pages of the other colors. + */ + batch = (1 << fls(batch + batch/2)) - 1; + for (cpu = 0; cpu < NR_CPUS; cpu++) { struct per_cpu_pages *pcp; @@ -1881,6 +1898,7 @@ static char *vmstat_text[] = { "allocstall", "pgrotated", + "nr_bounce", }; static void *vmstat_start(struct seq_file *m, loff_t *pos) diff --git a/mm/rmap.c b/mm/rmap.c index 884d6d1928b..378de234c12 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -243,6 +243,42 @@ unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma) } /* + * Check that @page is mapped at @address into @mm. + * + * On success returns with mapped pte and locked mm->page_table_lock. + */ +static pte_t *page_check_address(struct page *page, struct mm_struct *mm, + unsigned long address) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + /* + * We need the page_table_lock to protect us from page faults, + * munmap, fork, etc... + */ + spin_lock(&mm->page_table_lock); + pgd = pgd_offset(mm, address); + if (likely(pgd_present(*pgd))) { + pud = pud_offset(pgd, address); + if (likely(pud_present(*pud))) { + pmd = pmd_offset(pud, address); + if (likely(pmd_present(*pmd))) { + pte = pte_offset_map(pmd, address); + if (likely(pte_present(*pte) && + page_to_pfn(page) == pte_pfn(*pte))) + return pte; + pte_unmap(pte); + } + } + } + spin_unlock(&mm->page_table_lock); + return ERR_PTR(-ENOENT); +} + +/* * Subfunctions of page_referenced: page_referenced_one called * repeatedly from either page_referenced_anon or page_referenced_file. */ @@ -251,9 +287,6 @@ static int page_referenced_one(struct page *page, { struct mm_struct *mm = vma->vm_mm; unsigned long address; - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; pte_t *pte; int referenced = 0; @@ -263,39 +296,18 @@ static int page_referenced_one(struct page *page, if (address == -EFAULT) goto out; - spin_lock(&mm->page_table_lock); - - pgd = pgd_offset(mm, address); - if (!pgd_present(*pgd)) - goto out_unlock; - - pud = pud_offset(pgd, address); - if (!pud_present(*pud)) - goto out_unlock; - - pmd = pmd_offset(pud, address); - if (!pmd_present(*pmd)) - goto out_unlock; - - pte = pte_offset_map(pmd, address); - if (!pte_present(*pte)) - goto out_unmap; - - if (page_to_pfn(page) != pte_pfn(*pte)) - goto out_unmap; - - if (ptep_clear_flush_young(vma, address, pte)) - referenced++; - - if (mm != current->mm && !ignore_token && has_swap_token(mm)) - referenced++; + pte = page_check_address(page, mm, address); + if (!IS_ERR(pte)) { + if (ptep_clear_flush_young(vma, address, pte)) + referenced++; - (*mapcount)--; + if (mm != current->mm && !ignore_token && has_swap_token(mm)) + referenced++; -out_unmap: - pte_unmap(pte); -out_unlock: - spin_unlock(&mm->page_table_lock); + (*mapcount)--; + pte_unmap(pte); + spin_unlock(&mm->page_table_lock); + } out: return referenced; } @@ -502,9 +514,6 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma) { struct mm_struct *mm = vma->vm_mm; unsigned long address; - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; pte_t *pte; pte_t pteval; int ret = SWAP_AGAIN; @@ -515,30 +524,9 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma) if (address == -EFAULT) goto out; - /* - * We need the page_table_lock to protect us from page faults, - * munmap, fork, etc... - */ - spin_lock(&mm->page_table_lock); - - pgd = pgd_offset(mm, address); - if (!pgd_present(*pgd)) - goto out_unlock; - - pud = pud_offset(pgd, address); - if (!pud_present(*pud)) - goto out_unlock; - - pmd = pmd_offset(pud, address); - if (!pmd_present(*pmd)) - goto out_unlock; - - pte = pte_offset_map(pmd, address); - if (!pte_present(*pte)) - goto out_unmap; - - if (page_to_pfn(page) != pte_pfn(*pte)) - goto out_unmap; + pte = page_check_address(page, mm, address); + if (IS_ERR(pte)) + goto out; /* * If the page is mlock()d, we cannot swap it out. @@ -604,7 +592,6 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma) out_unmap: pte_unmap(pte); -out_unlock: spin_unlock(&mm->page_table_lock); out: return ret; @@ -708,7 +695,6 @@ static void try_to_unmap_cluster(unsigned long cursor, } pte_unmap(pte); - out_unlock: spin_unlock(&mm->page_table_lock); } @@ -860,3 +846,4 @@ int try_to_unmap(struct page *page) ret = SWAP_SUCCESS; return ret; } + diff --git a/mm/slab.c b/mm/slab.c index ec660d85ddd..84074264115 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -583,7 +583,7 @@ static inline struct array_cache *ac_data(kmem_cache_t *cachep) return cachep->array[smp_processor_id()]; } -static inline kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags) +static inline kmem_cache_t *__find_general_cachep(size_t size, int gfpflags) { struct cache_sizes *csizep = malloc_sizes; @@ -607,6 +607,12 @@ static inline kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags) return csizep->cs_cachep; } +kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags) +{ + return __find_general_cachep(size, gfpflags); +} +EXPORT_SYMBOL(kmem_find_general_cachep); + /* Cal the num objs, wastage, and bytes left over for a given slab size. */ static void cache_estimate(unsigned long gfporder, size_t size, size_t align, int flags, size_t *left_over, unsigned int *num) @@ -672,14 +678,11 @@ static struct array_cache *alloc_arraycache(int cpu, int entries, int memsize = sizeof(void*)*entries+sizeof(struct array_cache); struct array_cache *nc = NULL; - if (cpu != -1) { - kmem_cache_t *cachep; - cachep = kmem_find_general_cachep(memsize, GFP_KERNEL); - if (cachep) - nc = kmem_cache_alloc_node(cachep, cpu_to_node(cpu)); - } - if (!nc) + if (cpu == -1) nc = kmalloc(memsize, GFP_KERNEL); + else + nc = kmalloc_node(memsize, GFP_KERNEL, cpu_to_node(cpu)); + if (nc) { nc->avail = 0; nc->limit = entries; @@ -1663,7 +1666,7 @@ int kmem_cache_destroy(kmem_cache_t * cachep) } if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU)) - synchronize_kernel(); + synchronize_rcu(); /* no cpu_online check required here since we clear the percpu * array on cpu offline and set this to NULL. @@ -2361,7 +2364,7 @@ out: * and can sleep. And it will allocate memory on the given node, which * can improve the performance for cpu bound structures. */ -void *kmem_cache_alloc_node(kmem_cache_t *cachep, int nodeid) +void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int nodeid) { int loop; void *objp; @@ -2393,7 +2396,7 @@ void *kmem_cache_alloc_node(kmem_cache_t *cachep, int nodeid) spin_unlock_irq(&cachep->spinlock); local_irq_disable(); - if (!cache_grow(cachep, GFP_KERNEL, nodeid)) { + if (!cache_grow(cachep, flags, nodeid)) { local_irq_enable(); return NULL; } @@ -2435,6 +2438,16 @@ got_slabp: } EXPORT_SYMBOL(kmem_cache_alloc_node); +void *kmalloc_node(size_t size, int flags, int node) +{ + kmem_cache_t *cachep; + + cachep = kmem_find_general_cachep(size, flags); + if (unlikely(cachep == NULL)) + return NULL; + return kmem_cache_alloc_node(cachep, flags, node); +} +EXPORT_SYMBOL(kmalloc_node); #endif /** @@ -2462,7 +2475,12 @@ void *__kmalloc(size_t size, unsigned int __nocast flags) { kmem_cache_t *cachep; - cachep = kmem_find_general_cachep(size, flags); + /* If you want to save a few bytes .text space: replace + * __ with kmem_. + * Then kmalloc uses the uninlined functions instead of the inline + * functions. + */ + cachep = __find_general_cachep(size, flags); if (unlikely(cachep == NULL)) return NULL; return __cache_alloc(cachep, flags); @@ -2489,9 +2507,8 @@ void *__alloc_percpu(size_t size, size_t align) for (i = 0; i < NR_CPUS; i++) { if (!cpu_possible(i)) continue; - pdata->ptrs[i] = kmem_cache_alloc_node( - kmem_find_general_cachep(size, GFP_KERNEL), - cpu_to_node(i)); + pdata->ptrs[i] = kmalloc_node(size, GFP_KERNEL, + cpu_to_node(i)); if (!pdata->ptrs[i]) goto unwind_oom; diff --git a/mm/swap_state.c b/mm/swap_state.c index a063a902ed0..4f251775ef9 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -143,7 +143,6 @@ void __delete_from_swap_cache(struct page *page) int add_to_swap(struct page * page) { swp_entry_t entry; - int pf_flags; int err; if (!PageLocked(page)) @@ -154,29 +153,19 @@ int add_to_swap(struct page * page) if (!entry.val) return 0; - /* Radix-tree node allocations are performing - * GFP_ATOMIC allocations under PF_MEMALLOC. - * They can completely exhaust the page allocator. - * - * So PF_MEMALLOC is dropped here. This causes the slab - * allocations to fail earlier, so radix-tree nodes will - * then be allocated from the mempool reserves. + /* + * Radix-tree node allocations from PF_MEMALLOC contexts could + * completely exhaust the page allocator. __GFP_NOMEMALLOC + * stops emergency reserves from being allocated. * - * We're still using __GFP_HIGH for radix-tree node - * allocations, so some of the emergency pools are available, - * just not all of them. + * TODO: this could cause a theoretical memory reclaim + * deadlock in the swap out path. */ - - pf_flags = current->flags; - current->flags &= ~PF_MEMALLOC; - /* * Add it to the swap cache and mark it dirty */ - err = __add_to_swap_cache(page, entry, GFP_ATOMIC|__GFP_NOWARN); - - if (pf_flags & PF_MEMALLOC) - current->flags |= PF_MEMALLOC; + err = __add_to_swap_cache(page, entry, + GFP_ATOMIC|__GFP_NOMEMALLOC|__GFP_NOWARN); switch (err) { case 0: /* Success */ diff --git a/mm/truncate.c b/mm/truncate.c index c9a63f0b69a..60c8764bfac 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -242,7 +242,7 @@ EXPORT_SYMBOL(invalidate_inode_pages); /** * invalidate_inode_pages2_range - remove range of pages from an address_space - * @mapping - the address_space + * @mapping: the address_space * @start: the page offset 'from' which to invalidate * @end: the page offset 'to' which to invalidate (inclusive) * @@ -322,7 +322,7 @@ EXPORT_SYMBOL_GPL(invalidate_inode_pages2_range); /** * invalidate_inode_pages2 - remove all pages from an address_space - * @mapping - the address_space + * @mapping: the address_space * * Any pages which are found to be mapped into pagetables are unmapped prior to * invalidation. diff --git a/mm/vmalloc.c b/mm/vmalloc.c index c6182f6f130..2bd83e5c2bb 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -475,6 +475,10 @@ void *vmalloc(unsigned long size) EXPORT_SYMBOL(vmalloc); +#ifndef PAGE_KERNEL_EXEC +# define PAGE_KERNEL_EXEC PAGE_KERNEL +#endif + /** * vmalloc_exec - allocate virtually contiguous, executable memory * @@ -488,10 +492,6 @@ EXPORT_SYMBOL(vmalloc); * use __vmalloc() instead. */ -#ifndef PAGE_KERNEL_EXEC -# define PAGE_KERNEL_EXEC PAGE_KERNEL -#endif - void *vmalloc_exec(unsigned long size) { return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC); diff --git a/net/802/fddi.c b/net/802/fddi.c index f9a31a9f70f..ebcf4830d6f 100644 --- a/net/802/fddi.c +++ b/net/802/fddi.c @@ -10,7 +10,7 @@ * Authors: Lawrence V. Stefani, <stefani@lkg.dec.com> * * fddi.c is based on previous eth.c and tr.c work by - * Ross Biro, <bir7@leland.Stanford.Edu> + * Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * Florian La Roche, <rzsfl@rz.uni-sb.de> diff --git a/net/802/hippi.c b/net/802/hippi.c index 4eb135c0afb..051e8af56a7 100644 --- a/net/802/hippi.c +++ b/net/802/hippi.c @@ -7,7 +7,7 @@ * * Version: @(#)hippi.c 1.0.0 05/29/97 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * Florian La Roche, <rzsfl@rz.uni-sb.de> diff --git a/net/8021q/vlanproc.c b/net/8021q/vlanproc.c index c32d27af0a3..7b214cffc95 100644 --- a/net/8021q/vlanproc.c +++ b/net/8021q/vlanproc.c @@ -23,7 +23,7 @@ #include <linux/errno.h> /* return codes */ #include <linux/kernel.h> #include <linux/slab.h> /* kmalloc(), kfree() */ -#include <linux/mm.h> /* verify_area(), etc. */ +#include <linux/mm.h> #include <linux/string.h> /* inline mem*, str* functions */ #include <linux/init.h> /* __initfunc et al. */ #include <asm/byteorder.h> /* htons(), etc. */ diff --git a/net/appletalk/dev.c b/net/appletalk/dev.c index 76598445d84..1237e208e24 100644 --- a/net/appletalk/dev.c +++ b/net/appletalk/dev.c @@ -19,7 +19,7 @@ static int ltalk_mac_addr(struct net_device *dev, void *addr) return -EINVAL; } -void ltalk_setup(struct net_device *dev) +static void ltalk_setup(struct net_device *dev) { /* Fill in the fields of the device structure with localtalk-generic values. */ @@ -40,4 +40,22 @@ void ltalk_setup(struct net_device *dev) dev->flags = IFF_BROADCAST|IFF_MULTICAST|IFF_NOARP; } -EXPORT_SYMBOL(ltalk_setup); + +/** + * alloc_ltalkdev - Allocates and sets up an localtalk device + * @sizeof_priv: Size of additional driver-private structure to be allocated + * for this localtalk device + * + * Fill in the fields of the device structure with localtalk-generic + * values. Basically does everything except registering the device. + * + * Constructs a new net device, complete with a private data area of + * size @sizeof_priv. A 32-byte (not bit) alignment is enforced for + * this private data area. + */ + +struct net_device *alloc_ltalkdev(int sizeof_priv) +{ + return alloc_netdev(sizeof_priv, "lt%d", ltalk_setup); +} +EXPORT_SYMBOL(alloc_ltalkdev); diff --git a/net/atm/common.c b/net/atm/common.c index 6d16be334ea..e93e838069e 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -12,7 +12,7 @@ #include <linux/socket.h> /* SOL_SOCKET */ #include <linux/errno.h> /* error codes */ #include <linux/capability.h> -#include <linux/mm.h> /* verify_area */ +#include <linux/mm.h> #include <linux/sched.h> #include <linux/time.h> /* struct timeval */ #include <linux/skbuff.h> @@ -540,7 +540,7 @@ int vcc_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m, error = -EMSGSIZE; goto out; } - /* verify_area is done by net/socket.c */ + eff = (size+3) & ~3; /* align to word boundary */ prepare_to_wait(sk->sk_sleep, &wait, TASK_INTERRUPTIBLE); error = 0; diff --git a/net/atm/resources.c b/net/atm/resources.c index 33f1685dbb7..a57a9268bd2 100644 --- a/net/atm/resources.c +++ b/net/atm/resources.c @@ -44,11 +44,6 @@ static struct atm_dev *__alloc_atm_dev(const char *type) return dev; } -static void __free_atm_dev(struct atm_dev *dev) -{ - kfree(dev); -} - static struct atm_dev *__atm_dev_lookup(int number) { struct atm_dev *dev; @@ -90,7 +85,7 @@ struct atm_dev *atm_dev_register(const char *type, const struct atmdev_ops *ops, if ((inuse = __atm_dev_lookup(number))) { atm_dev_put(inuse); spin_unlock(&atm_dev_lock); - __free_atm_dev(dev); + kfree(dev); return NULL; } dev->number = number; @@ -119,7 +114,7 @@ struct atm_dev *atm_dev_register(const char *type, const struct atmdev_ops *ops, spin_lock(&atm_dev_lock); list_del(&dev->dev_list); spin_unlock(&atm_dev_lock); - __free_atm_dev(dev); + kfree(dev); return NULL; } @@ -148,7 +143,7 @@ void atm_dev_deregister(struct atm_dev *dev) } } - __free_atm_dev(dev); + kfree(dev); } void shutdown_atm_dev(struct atm_dev *dev) diff --git a/net/atm/signaling.c b/net/atm/signaling.c index 6ff803154c0..f7c449ac180 100644 --- a/net/atm/signaling.c +++ b/net/atm/signaling.c @@ -134,6 +134,7 @@ static int sigd_send(struct atm_vcc *vcc,struct sk_buff *skb) break; case as_indicate: vcc = *(struct atm_vcc **) &msg->listen_vcc; + sk = sk_atm(vcc); DPRINTK("as_indicate!!!\n"); lock_sock(sk); if (sk_acceptq_is_full(sk)) { diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index 33b1a376302..707097deac3 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -1587,9 +1587,7 @@ static int ax25_sendmsg(struct kiocb *iocb, struct socket *sock, *asmptr = AX25_UI; /* Datagram frames go straight out of the door as UI */ - skb->dev = ax25->ax25_dev->dev; - - ax25_queue_xmit(skb); + ax25_queue_xmit(skb, ax25->ax25_dev->dev); err = len; diff --git a/net/ax25/ax25_ds_subr.c b/net/ax25/ax25_ds_subr.c index 10ffd2beba3..1d4ab641f82 100644 --- a/net/ax25/ax25_ds_subr.c +++ b/net/ax25/ax25_ds_subr.c @@ -143,8 +143,7 @@ static void ax25_kiss_cmd(ax25_dev *ax25_dev, unsigned char cmd, unsigned char p *p++ = cmd; *p++ = param; - skb->dev = ax25_dev->dev; - skb->protocol = htons(ETH_P_AX25); + skb->protocol = ax25_type_trans(skb, ax25_dev->dev); dev_queue_xmit(skb); } diff --git a/net/ax25/ax25_ip.c b/net/ax25/ax25_ip.c index 04d711344d5..bba0173e2d6 100644 --- a/net/ax25/ax25_ip.c +++ b/net/ax25/ax25_ip.c @@ -199,9 +199,7 @@ int ax25_rebuild_header(struct sk_buff *skb) skb = ourskb; } - skb->dev = dev; - - ax25_queue_xmit(skb); + ax25_queue_xmit(skb, dev); put: ax25_put_route(route); diff --git a/net/ax25/ax25_out.c b/net/ax25/ax25_out.c index 3475a3ac934..5fc048dcd39 100644 --- a/net/ax25/ax25_out.c +++ b/net/ax25/ax25_out.c @@ -340,21 +340,18 @@ void ax25_transmit_buffer(ax25_cb *ax25, struct sk_buff *skb, int type) ax25_addr_build(ptr, &ax25->source_addr, &ax25->dest_addr, ax25->digipeat, type, ax25->modulus); - skb->dev = ax25->ax25_dev->dev; - - ax25_queue_xmit(skb); + ax25_queue_xmit(skb, ax25->ax25_dev->dev); } /* * A small shim to dev_queue_xmit to add the KISS control byte, and do * any packet forwarding in operation. */ -void ax25_queue_xmit(struct sk_buff *skb) +void ax25_queue_xmit(struct sk_buff *skb, struct net_device *dev) { unsigned char *ptr; - skb->protocol = htons(ETH_P_AX25); - skb->dev = ax25_fwd_dev(skb->dev); + skb->protocol = ax25_type_trans(skb, ax25_fwd_dev(dev)); ptr = skb_push(skb, 1); *ptr = 0x00; /* KISS */ diff --git a/net/ax25/ax25_subr.c b/net/ax25/ax25_subr.c index 8cf72707af8..99694b57f6f 100644 --- a/net/ax25/ax25_subr.c +++ b/net/ax25/ax25_subr.c @@ -220,9 +220,7 @@ void ax25_return_dm(struct net_device *dev, ax25_address *src, ax25_address *des dptr = skb_push(skb, ax25_addr_size(digi)); dptr += ax25_addr_build(dptr, dest, src, &retdigi, AX25_RESPONSE, AX25_MODULUS); - skb->dev = dev; - - ax25_queue_xmit(skb); + ax25_queue_xmit(skb, dev); } /* diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 1650c6bf699..12b43345b54 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -31,7 +31,6 @@ #include <linux/list.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/skbuff.h> diff --git a/net/bluetooth/bnep/sock.c b/net/bluetooth/bnep/sock.c index 9a8d99a39b6..9778c6acd53 100644 --- a/net/bluetooth/bnep/sock.c +++ b/net/bluetooth/bnep/sock.c @@ -34,7 +34,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/cmtp/capi.c b/net/bluetooth/cmtp/capi.c index 1e5c030b72a..b2e7e38531c 100644 --- a/net/bluetooth/cmtp/capi.c +++ b/net/bluetooth/cmtp/capi.c @@ -26,7 +26,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/cmtp/core.c b/net/bluetooth/cmtp/core.c index 20ce04f2be8..2e341de3e76 100644 --- a/net/bluetooth/cmtp/core.c +++ b/net/bluetooth/cmtp/core.c @@ -26,7 +26,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/cmtp/sock.c b/net/bluetooth/cmtp/sock.c index 4c7f9e20dad..beb045bf571 100644 --- a/net/bluetooth/cmtp/sock.c +++ b/net/bluetooth/cmtp/sock.c @@ -26,7 +26,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 71762d7e997..a31244e5888 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -30,7 +30,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 860dba7bdd8..fb5524365bc 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -31,7 +31,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 8ccba8ee997..c4b592b4ef1 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -30,7 +30,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index c9792ba7512..ebdcce5e7ca 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -30,7 +30,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 2cf98ceabcc..affbc55462e 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -26,7 +26,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/hidp/sock.c b/net/bluetooth/hidp/sock.c index fabb36d4666..f8986f88143 100644 --- a/net/bluetooth/hidp/sock.c +++ b/net/bluetooth/hidp/sock.c @@ -26,7 +26,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index c12babcf0b3..32fccfb5bfa 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -30,7 +30,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index 640028a2183..f3f6355a278 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -33,7 +33,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 3e750ef09e6..746c11fc017 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -30,7 +30,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> diff --git a/net/core/datagram.c b/net/core/datagram.c index d1bfd279cc1..fcee054b6f7 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -115,10 +115,10 @@ out_noerr: /** * skb_recv_datagram - Receive a datagram skbuff - * @sk - socket - * @flags - MSG_ flags - * @noblock - blocking operation? - * @err - error code returned + * @sk: socket + * @flags: MSG_ flags + * @noblock: blocking operation? + * @err: error code returned * * Get a datagram skbuff, understands the peeking, nonblocking wakeups * and possible races. This replaces identical code in packet, raw and @@ -201,10 +201,10 @@ void skb_free_datagram(struct sock *sk, struct sk_buff *skb) /** * skb_copy_datagram_iovec - Copy a datagram to an iovec. - * @skb - buffer to copy - * @offset - offset in the buffer to start copying from - * @iovec - io vector to copy to - * @len - amount of data to copy from buffer to iovec + * @skb: buffer to copy + * @offset: offset in the buffer to start copying from + * @to: io vector to copy to + * @len: amount of data to copy from buffer to iovec * * Note: the iovec is modified during the copy. */ @@ -377,9 +377,9 @@ fault: /** * skb_copy_and_csum_datagram_iovec - Copy and checkum skb to user iovec. - * @skb - skbuff - * @hlen - hardware length - * @iovec - io vector + * @skb: skbuff + * @hlen: hardware length + * @iov: io vector * * Caller _must_ check that skb will fit to this iovec. * @@ -425,9 +425,9 @@ fault: /** * datagram_poll - generic datagram poll - * @file - file struct - * @sock - socket - * @wait - poll table + * @file: file struct + * @sock: socket + * @wait: poll table * * Datagram poll: Again totally generic. This also handles * sequenced packet sockets providing the socket receive queue diff --git a/net/core/dev.c b/net/core/dev.c index 42344d90369..d4d9e2680ad 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -7,7 +7,7 @@ * 2 of the License, or (at your option) any later version. * * Derived from the non IP parts of dev.c 1.0.19 - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * @@ -1214,6 +1214,19 @@ int __skb_linearize(struct sk_buff *skb, int gfp_mask) * A negative errno code is returned on a failure. A success does not * guarantee the frame will be transmitted as it may be dropped due * to congestion or traffic shaping. + * + * ----------------------------------------------------------------------------------- + * I notice this method can also return errors from the queue disciplines, + * including NET_XMIT_DROP, which is a positive value. So, errors can also + * be positive. + * + * Regardless of the return value, the skb is consumed, so it is currently + * difficult to retry a send to this method. (You can bump the ref count + * before sending to hold a reference for retry if you are careful.) + * + * When calling this method, interrupts MUST be enabled. This is because + * the BH enable code must have IRQs enabled so that it will not deadlock. + * --BLG */ int dev_queue_xmit(struct sk_buff *skb) @@ -3078,7 +3091,7 @@ void free_netdev(struct net_device *dev) void synchronize_net(void) { might_sleep(); - synchronize_kernel(); + synchronize_rcu(); } /** diff --git a/net/core/iovec.c b/net/core/iovec.c index d57ace949ab..65e4b56fbc7 100644 --- a/net/core/iovec.c +++ b/net/core/iovec.c @@ -33,7 +33,7 @@ * Verify iovec. The caller must ensure that the iovec is big enough * to hold the message iovec. * - * Save time not doing verify_area. copy_*_user will make this work + * Save time not doing access_ok. copy_*_user will make this work * in any case. */ diff --git a/net/core/link_watch.c b/net/core/link_watch.c index 4859b7446c6..d43d1201275 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -16,6 +16,7 @@ #include <linux/netdevice.h> #include <linux/if.h> #include <net/sock.h> +#include <net/pkt_sched.h> #include <linux/rtnetlink.h> #include <linux/jiffies.h> #include <linux/spinlock.h> @@ -74,6 +75,12 @@ void linkwatch_run_queue(void) clear_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state); if (dev->flags & IFF_UP) { + if (netif_carrier_ok(dev)) { + WARN_ON(dev->qdisc_sleeping == &noop_qdisc); + dev_activate(dev); + } else + dev_deactivate(dev); + netdev_state_change(dev); } diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 0a2f67bbef2..43bdc521e20 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1953,7 +1953,7 @@ static int neigh_stat_seq_show(struct seq_file *seq, void *v) struct neigh_statistics *st = v; if (v == SEQ_START_TOKEN) { - seq_printf(seq, "entries allocs destroys hash_grows lookups hits res_failed rcv_probes_mcast rcv_probes_ucast periodic_gc_runs forced_gc_runs forced_gc_goal_miss\n"); + seq_printf(seq, "entries allocs destroys hash_grows lookups hits res_failed rcv_probes_mcast rcv_probes_ucast periodic_gc_runs forced_gc_runs\n"); return 0; } diff --git a/net/core/netfilter.c b/net/core/netfilter.c index e51cfa46950..22a8f127c4a 100644 --- a/net/core/netfilter.c +++ b/net/core/netfilter.c @@ -217,21 +217,10 @@ void nf_debug_ip_local_deliver(struct sk_buff *skb) * NF_IP_RAW_INPUT and NF_IP_PRE_ROUTING. */ if (!skb->dev) { printk("ip_local_deliver: skb->dev is NULL.\n"); - } - else if (strcmp(skb->dev->name, "lo") == 0) { - if (skb->nf_debug != ((1 << NF_IP_LOCAL_OUT) - | (1 << NF_IP_POST_ROUTING) - | (1 << NF_IP_PRE_ROUTING) - | (1 << NF_IP_LOCAL_IN))) { - printk("ip_local_deliver: bad loopback skb: "); - debug_print_hooks_ip(skb->nf_debug); - nf_dump_skb(PF_INET, skb); - } - } - else { + } else { if (skb->nf_debug != ((1<<NF_IP_PRE_ROUTING) | (1<<NF_IP_LOCAL_IN))) { - printk("ip_local_deliver: bad non-lo skb: "); + printk("ip_local_deliver: bad skb: "); debug_print_hooks_ip(skb->nf_debug); nf_dump_skb(PF_INET, skb); } @@ -247,8 +236,6 @@ void nf_debug_ip_loopback_xmit(struct sk_buff *newskb) debug_print_hooks_ip(newskb->nf_debug); nf_dump_skb(PF_INET, newskb); } - /* Clear to avoid confusing input check */ - newskb->nf_debug = 0; } void nf_debug_ip_finish_output2(struct sk_buff *skb) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 44dfaf8f04a..00caf4b318b 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -22,7 +22,6 @@ #include <linux/types.h> #include <linux/socket.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/timer.h> #include <linux/string.h> @@ -87,30 +86,33 @@ struct sock *rtnl; struct rtnetlink_link * rtnetlink_links[NPROTO]; -static const int rtm_min[(RTM_MAX+1-RTM_BASE)/4] = +static const int rtm_min[RTM_NR_FAMILIES] = { - NLMSG_LENGTH(sizeof(struct ifinfomsg)), - NLMSG_LENGTH(sizeof(struct ifaddrmsg)), - NLMSG_LENGTH(sizeof(struct rtmsg)), - NLMSG_LENGTH(sizeof(struct ndmsg)), - NLMSG_LENGTH(sizeof(struct rtmsg)), - NLMSG_LENGTH(sizeof(struct tcmsg)), - NLMSG_LENGTH(sizeof(struct tcmsg)), - NLMSG_LENGTH(sizeof(struct tcmsg)), - NLMSG_LENGTH(sizeof(struct tcamsg)) + [RTM_FAM(RTM_NEWLINK)] = NLMSG_LENGTH(sizeof(struct ifinfomsg)), + [RTM_FAM(RTM_NEWADDR)] = NLMSG_LENGTH(sizeof(struct ifaddrmsg)), + [RTM_FAM(RTM_NEWROUTE)] = NLMSG_LENGTH(sizeof(struct rtmsg)), + [RTM_FAM(RTM_NEWNEIGH)] = NLMSG_LENGTH(sizeof(struct ndmsg)), + [RTM_FAM(RTM_NEWRULE)] = NLMSG_LENGTH(sizeof(struct rtmsg)), + [RTM_FAM(RTM_NEWQDISC)] = NLMSG_LENGTH(sizeof(struct tcmsg)), + [RTM_FAM(RTM_NEWTCLASS)] = NLMSG_LENGTH(sizeof(struct tcmsg)), + [RTM_FAM(RTM_NEWTFILTER)] = NLMSG_LENGTH(sizeof(struct tcmsg)), + [RTM_FAM(RTM_NEWACTION)] = NLMSG_LENGTH(sizeof(struct tcamsg)), + [RTM_FAM(RTM_NEWPREFIX)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), + [RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), + [RTM_FAM(RTM_GETANYCAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), }; -static const int rta_max[(RTM_MAX+1-RTM_BASE)/4] = +static const int rta_max[RTM_NR_FAMILIES] = { - IFLA_MAX, - IFA_MAX, - RTA_MAX, - NDA_MAX, - RTA_MAX, - TCA_MAX, - TCA_MAX, - TCA_MAX, - TCAA_MAX + [RTM_FAM(RTM_NEWLINK)] = IFLA_MAX, + [RTM_FAM(RTM_NEWADDR)] = IFA_MAX, + [RTM_FAM(RTM_NEWROUTE)] = RTA_MAX, + [RTM_FAM(RTM_NEWNEIGH)] = NDA_MAX, + [RTM_FAM(RTM_NEWRULE)] = RTA_MAX, + [RTM_FAM(RTM_NEWQDISC)] = TCA_MAX, + [RTM_FAM(RTM_NEWTCLASS)] = TCA_MAX, + [RTM_FAM(RTM_NEWTFILTER)] = TCA_MAX, + [RTM_FAM(RTM_NEWACTION)] = TCAA_MAX, }; void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data) @@ -607,27 +609,33 @@ static inline int rtnetlink_rcv_skb(struct sk_buff *skb) /* * rtnetlink input queue processing routine: - * - try to acquire shared lock. If it is failed, defer processing. + * - process as much as there was in the queue upon entry. * - feed skbs to rtnetlink_rcv_skb, until it refuse a message, - * that will occur, when a dump started and/or acquisition of - * exclusive lock failed. + * that will occur, when a dump started. */ static void rtnetlink_rcv(struct sock *sk, int len) { + unsigned int qlen = skb_queue_len(&sk->sk_receive_queue); + do { struct sk_buff *skb; - if (rtnl_shlock_nowait()) - return; + rtnl_lock(); + + if (qlen > skb_queue_len(&sk->sk_receive_queue)) + qlen = skb_queue_len(&sk->sk_receive_queue); - while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) { + for (; qlen; qlen--) { + skb = skb_dequeue(&sk->sk_receive_queue); if (rtnetlink_rcv_skb(skb)) { if (skb->len) skb_queue_head(&sk->sk_receive_queue, skb); - else + else { kfree_skb(skb); + qlen--; + } break; } kfree_skb(skb); @@ -636,10 +644,10 @@ static void rtnetlink_rcv(struct sock *sk, int len) up(&rtnl_sem); netdev_run_todo(); - } while (rtnl && rtnl->sk_receive_queue.qlen); + } while (qlen); } -static struct rtnetlink_link link_rtnetlink_table[RTM_MAX-RTM_BASE+1] = +static struct rtnetlink_link link_rtnetlink_table[RTM_NR_MSGTYPES] = { [RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo }, [RTM_SETLINK - RTM_BASE] = { .doit = do_setlink }, diff --git a/net/core/scm.c b/net/core/scm.c index a2ebf30f6aa..e887d19be50 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -15,7 +15,6 @@ #include <linux/sched.h> #include <linux/mm.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/stat.h> #include <linux/socket.h> #include <linux/file.h> diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 1b64817d7de..f65b3de590a 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -86,8 +86,10 @@ static kmem_cache_t *skbuff_head_cache; */ void skb_over_panic(struct sk_buff *skb, int sz, void *here) { - printk(KERN_INFO "skput:over: %p:%d put:%d dev:%s", - here, skb->len, sz, skb->dev ? skb->dev->name : "<NULL>"); + printk(KERN_EMERG "skb_over_panic: text:%p len:%d put:%d head:%p " + "data:%p tail:%p end:%p dev:%s\n", + here, skb->len, sz, skb->head, skb->data, skb->tail, skb->end, + skb->dev ? skb->dev->name : "<NULL>"); BUG(); } @@ -102,8 +104,10 @@ void skb_over_panic(struct sk_buff *skb, int sz, void *here) void skb_under_panic(struct sk_buff *skb, int sz, void *here) { - printk(KERN_INFO "skput:under: %p:%d put:%d dev:%s", - here, skb->len, sz, skb->dev ? skb->dev->name : "<NULL>"); + printk(KERN_EMERG "skb_under_panic: text:%p len:%d put:%d head:%p " + "data:%p tail:%p end:%p dev:%s\n", + here, skb->len, sz, skb->head, skb->data, skb->tail, skb->end, + skb->dev ? skb->dev->name : "<NULL>"); BUG(); } diff --git a/net/core/sock.c b/net/core/sock.c index 4df4fa3c5de..96e00b08698 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -9,7 +9,7 @@ * * Version: $Id: sock.c,v 1.117 2002/02/01 22:01:03 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Florian La Roche, <flla@stud.uni-sb.de> * Alan Cox, <A.Cox@swansea.ac.uk> @@ -97,7 +97,6 @@ #include <linux/socket.h> #include <linux/in.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/module.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> @@ -617,10 +616,10 @@ lenout: /** * sk_alloc - All socket objects are allocated here - * @family - protocol family - * @priority - for allocation (%GFP_KERNEL, %GFP_ATOMIC, etc) - * @prot - struct proto associated with this new sock instance - * @zero_it - if we should zero the newly allocated sock + * @family: protocol family + * @priority: for allocation (%GFP_KERNEL, %GFP_ATOMIC, etc) + * @prot: struct proto associated with this new sock instance + * @zero_it: if we should zero the newly allocated sock */ struct sock *sk_alloc(int family, int priority, struct proto *prot, int zero_it) { @@ -636,7 +635,11 @@ struct sock *sk_alloc(int family, int priority, struct proto *prot, int zero_it) if (zero_it) { memset(sk, 0, prot->obj_size); sk->sk_family = family; - sk->sk_prot = prot; + /* + * See comment in struct sock definition to understand + * why we need sk_prot_creator -acme + */ + sk->sk_prot = sk->sk_prot_creator = prot; sock_lock_init(sk); } @@ -655,7 +658,7 @@ struct sock *sk_alloc(int family, int priority, struct proto *prot, int zero_it) void sk_free(struct sock *sk) { struct sk_filter *filter; - struct module *owner = sk->sk_prot->owner; + struct module *owner = sk->sk_prot_creator->owner; if (sk->sk_destruct) sk->sk_destruct(sk); @@ -673,8 +676,8 @@ void sk_free(struct sock *sk) __FUNCTION__, atomic_read(&sk->sk_omem_alloc)); security_sk_free(sk); - if (sk->sk_prot->slab != NULL) - kmem_cache_free(sk->sk_prot->slab, sk); + if (sk->sk_prot_creator->slab != NULL) + kmem_cache_free(sk->sk_prot_creator->slab, sk); else kfree(sk); module_put(owner); @@ -971,8 +974,8 @@ static void __release_sock(struct sock *sk) /** * sk_wait_data - wait for data to arrive at sk_receive_queue - * sk - sock to wait on - * timeo - for how long + * @sk: sock to wait on + * @timeo: for how long * * Now socket state including sk->sk_err is changed only under lock, * hence we may omit checks after joining wait queue. diff --git a/net/core/stream.c b/net/core/stream.c index 1e27a57b5a9..ac9edfdf874 100644 --- a/net/core/stream.c +++ b/net/core/stream.c @@ -21,7 +21,7 @@ /** * sk_stream_write_space - stream socket write_space callback. - * sk - socket + * @sk: socket * * FIXME: write proper description */ @@ -43,8 +43,8 @@ EXPORT_SYMBOL(sk_stream_write_space); /** * sk_stream_wait_connect - Wait for a socket to get into the connected state - * @sk - sock to wait on - * @timeo_p - for how long to wait + * @sk: sock to wait on + * @timeo_p: for how long to wait * * Must be called with the socket locked. */ @@ -79,7 +79,7 @@ EXPORT_SYMBOL(sk_stream_wait_connect); /** * sk_stream_closing - Return 1 if we still have things to send in our buffers. - * @sk - socket to verify + * @sk: socket to verify */ static inline int sk_stream_closing(struct sock *sk) { @@ -107,8 +107,8 @@ EXPORT_SYMBOL(sk_stream_wait_close); /** * sk_stream_wait_memory - Wait for more memory for a socket - * @sk - socket to wait for memory - * @timeo_p - for how long + * @sk: socket to wait for memory + * @timeo_p: for how long */ int sk_stream_wait_memory(struct sock *sk, long *timeo_p) { diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index c2a0346f423..ee7bf46eb78 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -1411,21 +1411,22 @@ static struct file_operations dn_dev_seq_fops = { #endif /* CONFIG_PROC_FS */ -static struct rtnetlink_link dnet_rtnetlink_table[RTM_MAX-RTM_BASE+1] = +static struct rtnetlink_link dnet_rtnetlink_table[RTM_NR_MSGTYPES] = { - [4] = { .doit = dn_dev_rtm_newaddr, }, - [5] = { .doit = dn_dev_rtm_deladdr, }, - [6] = { .dumpit = dn_dev_dump_ifaddr, }, - + [RTM_NEWADDR - RTM_BASE] = { .doit = dn_dev_rtm_newaddr, }, + [RTM_DELADDR - RTM_BASE] = { .doit = dn_dev_rtm_deladdr, }, + [RTM_GETADDR - RTM_BASE] = { .dumpit = dn_dev_dump_ifaddr, }, #ifdef CONFIG_DECNET_ROUTER - [8] = { .doit = dn_fib_rtm_newroute, }, - [9] = { .doit = dn_fib_rtm_delroute, }, - [10] = { .doit = dn_cache_getroute, .dumpit = dn_fib_dump, }, - [16] = { .doit = dn_fib_rtm_newrule, }, - [17] = { .doit = dn_fib_rtm_delrule, }, - [18] = { .dumpit = dn_fib_dump_rules, }, + [RTM_NEWROUTE - RTM_BASE] = { .doit = dn_fib_rtm_newroute, }, + [RTM_DELROUTE - RTM_BASE] = { .doit = dn_fib_rtm_delroute, }, + [RTM_GETROUTE - RTM_BASE] = { .doit = dn_cache_getroute, + .dumpit = dn_fib_dump, }, + [RTM_NEWRULE - RTM_BASE] = { .doit = dn_fib_rtm_newrule, }, + [RTM_DELRULE - RTM_BASE] = { .doit = dn_fib_rtm_delrule, }, + [RTM_GETRULE - RTM_BASE] = { .dumpit = dn_fib_dump_rules, }, #else - [10] = { .doit = dn_cache_getroute, .dumpit = dn_cache_dump, }, + [RTM_GETROUTE - RTM_BASE] = { .doit = dn_cache_getroute, + .dumpit = dn_cache_dump, }, #endif }; diff --git a/net/decnet/netfilter/dn_rtmsg.c b/net/decnet/netfilter/dn_rtmsg.c index f86a6259fd1..284a9998e53 100644 --- a/net/decnet/netfilter/dn_rtmsg.c +++ b/net/decnet/netfilter/dn_rtmsg.c @@ -119,8 +119,9 @@ static inline void dnrmg_receive_user_skb(struct sk_buff *skb) static void dnrmg_receive_user_sk(struct sock *sk, int len) { struct sk_buff *skb; + unsigned int qlen = skb_queue_len(&sk->sk_receive_queue); - while((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) { + for (; qlen && (skb = skb_dequeue(&sk->sk_receive_queue)); qlen--) { dnrmg_receive_user_skb(skb); kfree_skb(skb); } diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c index 16c4234cbe1..6617ea47d36 100644 --- a/net/ethernet/eth.c +++ b/net/ethernet/eth.c @@ -7,7 +7,7 @@ * * Version: @(#)eth.c 1.0.7 05/25/93 * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * Florian La Roche, <rzsfl@rz.uni-sb.de> diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index c34dab67e46..b3cb49ce5fa 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -7,7 +7,7 @@ * * Version: $Id: af_inet.c,v 1.137 2002/02/01 22:01:03 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Florian La Roche, <flla@stud.uni-sb.de> * Alan Cox, <A.Cox@swansea.ac.uk> @@ -73,7 +73,6 @@ #include <linux/socket.h> #include <linux/in.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/module.h> #include <linux/sched.h> #include <linux/timer.h> diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index eea7ef01077..3cc96730c4e 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -9,7 +9,7 @@ * 2 of the License, or (at your option) any later version. * * Derived from the IP parts of dev.c 1.0.19 - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * @@ -1107,17 +1107,18 @@ static void rtmsg_ifa(int event, struct in_ifaddr* ifa) } } -static struct rtnetlink_link inet_rtnetlink_table[RTM_MAX - RTM_BASE + 1] = { - [4] = { .doit = inet_rtm_newaddr, }, - [5] = { .doit = inet_rtm_deladdr, }, - [6] = { .dumpit = inet_dump_ifaddr, }, - [8] = { .doit = inet_rtm_newroute, }, - [9] = { .doit = inet_rtm_delroute, }, - [10] = { .doit = inet_rtm_getroute, .dumpit = inet_dump_fib, }, +static struct rtnetlink_link inet_rtnetlink_table[RTM_NR_MSGTYPES] = { + [RTM_NEWADDR - RTM_BASE] = { .doit = inet_rtm_newaddr, }, + [RTM_DELADDR - RTM_BASE] = { .doit = inet_rtm_deladdr, }, + [RTM_GETADDR - RTM_BASE] = { .dumpit = inet_dump_ifaddr, }, + [RTM_NEWROUTE - RTM_BASE] = { .doit = inet_rtm_newroute, }, + [RTM_DELROUTE - RTM_BASE] = { .doit = inet_rtm_delroute, }, + [RTM_GETROUTE - RTM_BASE] = { .doit = inet_rtm_getroute, + .dumpit = inet_dump_fib, }, #ifdef CONFIG_IP_MULTIPLE_TABLES - [16] = { .doit = inet_rtm_newrule, }, - [17] = { .doit = inet_rtm_delrule, }, - [18] = { .dumpit = inet_dump_rules, }, + [RTM_NEWRULE - RTM_BASE] = { .doit = inet_rtm_newrule, }, + [RTM_DELRULE - RTM_BASE] = { .doit = inet_rtm_delrule, }, + [RTM_GETRULE - RTM_BASE] = { .dumpit = inet_dump_rules, }, #endif }; diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index a0d0833034b..4e47a2658c7 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -7,7 +7,7 @@ * * Version: $Id: ip_input.c,v 1.55 2002/01/12 07:39:45 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Donald Becker, <becker@super.org> * Alan Cox, <Alan.Cox@linux.org> diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 30ab7b6ab76..daebd93fd8a 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -7,7 +7,7 @@ * * Version: $Id: ip_output.c,v 1.100 2002/02/01 22:01:03 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Donald Becker, <becker@super.org> * Alan Cox, <Alan.Cox@linux.org> @@ -111,6 +111,7 @@ static int ip_dev_loopback_xmit(struct sk_buff *newskb) #ifdef CONFIG_NETFILTER_DEBUG nf_debug_ip_loopback_xmit(newskb); #endif + nf_reset(newskb); netif_rx(newskb); return 0; } @@ -195,6 +196,8 @@ static inline int ip_finish_output2(struct sk_buff *skb) nf_debug_ip_finish_output2(skb); #endif /*CONFIG_NETFILTER_DEBUG*/ + nf_reset(skb); + if (hh) { int hh_alen; diff --git a/net/ipv4/multipath_wrandom.c b/net/ipv4/multipath_wrandom.c index 10b23e1bece..c3d2ca1a678 100644 --- a/net/ipv4/multipath_wrandom.c +++ b/net/ipv4/multipath_wrandom.c @@ -172,7 +172,7 @@ static void wrandom_select_route(const struct flowi *flp, multipath_comparekeys(&rt->fl, flp)) { struct multipath_candidate* mpc = (struct multipath_candidate*) - kmalloc(size_mpc, GFP_KERNEL); + kmalloc(size_mpc, GFP_ATOMIC); if (!mpc) return; @@ -244,7 +244,7 @@ static void wrandom_set_nhinfo(__u32 network, if (!target_route) { const size_t size_rt = sizeof(struct multipath_route); target_route = (struct multipath_route *) - kmalloc(size_rt, GFP_KERNEL); + kmalloc(size_rt, GFP_ATOMIC); target_route->gw = nh->nh_gw; target_route->oif = nh->nh_oif; @@ -265,7 +265,7 @@ static void wrandom_set_nhinfo(__u32 network, if (!target_dest) { const size_t size_dst = sizeof(struct multipath_dest); target_dest = (struct multipath_dest*) - kmalloc(size_dst, GFP_KERNEL); + kmalloc(size_dst, GFP_ATOMIC); target_dest->nh_info = nh; target_dest->network = network; diff --git a/net/ipv4/netfilter/ip_conntrack_ftp.c b/net/ipv4/netfilter/ip_conntrack_ftp.c index 12b88cbb11d..dd86503aa78 100644 --- a/net/ipv4/netfilter/ip_conntrack_ftp.c +++ b/net/ipv4/netfilter/ip_conntrack_ftp.c @@ -252,7 +252,7 @@ static int find_pattern(const char *data, size_t dlen, } /* Look up to see if we're just after a \n. */ -static int find_nl_seq(u16 seq, const struct ip_ct_ftp_master *info, int dir) +static int find_nl_seq(u32 seq, const struct ip_ct_ftp_master *info, int dir) { unsigned int i; @@ -263,7 +263,7 @@ static int find_nl_seq(u16 seq, const struct ip_ct_ftp_master *info, int dir) } /* We don't update if it's older than what we have. */ -static void update_nl_seq(u16 nl_seq, struct ip_ct_ftp_master *info, int dir) +static void update_nl_seq(u32 nl_seq, struct ip_ct_ftp_master *info, int dir) { unsigned int i, oldest = NUM_SEQ_TO_REMEMBER; diff --git a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c index e800b16fc92..721ddbf522b 100644 --- a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c +++ b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c @@ -770,6 +770,7 @@ static u8 tcp_valid_flags[(TH_FIN|TH_SYN|TH_RST|TH_PUSH|TH_ACK|TH_URG) + 1] = { [TH_SYN] = 1, [TH_SYN|TH_ACK] = 1, + [TH_SYN|TH_ACK|TH_PUSH] = 1, [TH_RST] = 1, [TH_RST|TH_ACK] = 1, [TH_RST|TH_ACK|TH_PUSH] = 1, @@ -818,6 +819,7 @@ static int tcp_error(struct sk_buff *skb, */ /* FIXME: Source route IP option packets --RR */ if (hooknum == NF_IP_PRE_ROUTING + && skb->ip_summed != CHECKSUM_UNNECESSARY && csum_tcpudp_magic(iph->saddr, iph->daddr, tcplen, IPPROTO_TCP, skb->ip_summed == CHECKSUM_HW ? skb->csum : skb_checksum(skb, iph->ihl*4, tcplen, 0))) { diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c index 80a7bde2a57..46ca45f74d8 100644 --- a/net/ipv4/netfilter/ip_conntrack_standalone.c +++ b/net/ipv4/netfilter/ip_conntrack_standalone.c @@ -401,6 +401,16 @@ static unsigned int ip_confirm(unsigned int hooknum, const struct net_device *out, int (*okfn)(struct sk_buff *)) { + /* We've seen it coming out the other side: confirm it */ + return ip_conntrack_confirm(pskb); +} + +static unsigned int ip_conntrack_help(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ struct ip_conntrack *ct; enum ip_conntrack_info ctinfo; @@ -412,9 +422,7 @@ static unsigned int ip_confirm(unsigned int hooknum, if (ret != NF_ACCEPT) return ret; } - - /* We've seen it coming out the other side: confirm it */ - return ip_conntrack_confirm(pskb); + return NF_ACCEPT; } static unsigned int ip_conntrack_defrag(unsigned int hooknum, @@ -423,13 +431,6 @@ static unsigned int ip_conntrack_defrag(unsigned int hooknum, const struct net_device *out, int (*okfn)(struct sk_buff *)) { -#if !defined(CONFIG_IP_NF_NAT) && !defined(CONFIG_IP_NF_NAT_MODULE) - /* Previously seen (loopback)? Ignore. Do this before - fragment check. */ - if ((*pskb)->nfct) - return NF_ACCEPT; -#endif - /* Gather fragments. */ if ((*pskb)->nh.iph->frag_off & htons(IP_MF|IP_OFFSET)) { *pskb = ip_ct_gather_frags(*pskb, @@ -516,13 +517,30 @@ static struct nf_hook_ops ip_conntrack_local_out_ops = { .priority = NF_IP_PRI_CONNTRACK, }; +/* helpers */ +static struct nf_hook_ops ip_conntrack_helper_out_ops = { + .hook = ip_conntrack_help, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_CONNTRACK_HELPER, +}; + +static struct nf_hook_ops ip_conntrack_helper_in_ops = { + .hook = ip_conntrack_help, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_CONNTRACK_HELPER, +}; + /* Refragmenter; last chance. */ static struct nf_hook_ops ip_conntrack_out_ops = { .hook = ip_refrag, .owner = THIS_MODULE, .pf = PF_INET, .hooknum = NF_IP_POST_ROUTING, - .priority = NF_IP_PRI_LAST, + .priority = NF_IP_PRI_CONNTRACK_CONFIRM, }; static struct nf_hook_ops ip_conntrack_local_in_ops = { @@ -530,7 +548,7 @@ static struct nf_hook_ops ip_conntrack_local_in_ops = { .owner = THIS_MODULE, .pf = PF_INET, .hooknum = NF_IP_LOCAL_IN, - .priority = NF_IP_PRI_LAST-1, + .priority = NF_IP_PRI_CONNTRACK_CONFIRM, }; /* Sysctl support */ @@ -831,10 +849,20 @@ static int init_or_cleanup(int init) printk("ip_conntrack: can't register local out hook.\n"); goto cleanup_inops; } + ret = nf_register_hook(&ip_conntrack_helper_in_ops); + if (ret < 0) { + printk("ip_conntrack: can't register local in helper hook.\n"); + goto cleanup_inandlocalops; + } + ret = nf_register_hook(&ip_conntrack_helper_out_ops); + if (ret < 0) { + printk("ip_conntrack: can't register postrouting helper hook.\n"); + goto cleanup_helperinops; + } ret = nf_register_hook(&ip_conntrack_out_ops); if (ret < 0) { printk("ip_conntrack: can't register post-routing hook.\n"); - goto cleanup_inandlocalops; + goto cleanup_helperoutops; } ret = nf_register_hook(&ip_conntrack_local_in_ops); if (ret < 0) { @@ -860,6 +888,10 @@ static int init_or_cleanup(int init) nf_unregister_hook(&ip_conntrack_local_in_ops); cleanup_inoutandlocalops: nf_unregister_hook(&ip_conntrack_out_ops); + cleanup_helperoutops: + nf_unregister_hook(&ip_conntrack_helper_out_ops); + cleanup_helperinops: + nf_unregister_hook(&ip_conntrack_helper_in_ops); cleanup_inandlocalops: nf_unregister_hook(&ip_conntrack_local_out_ops); cleanup_inops: diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c index 162ceacfc29..9fc6f93af0d 100644 --- a/net/ipv4/netfilter/ip_nat_core.c +++ b/net/ipv4/netfilter/ip_nat_core.c @@ -356,15 +356,6 @@ unsigned int nat_packet(struct ip_conntrack *ct, unsigned long statusbit; enum ip_nat_manip_type mtype = HOOK2MANIP(hooknum); - if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) - && (hooknum == NF_IP_POST_ROUTING || hooknum == NF_IP_LOCAL_IN)) { - DEBUGP("ip_nat_core: adjusting sequence number\n"); - /* future: put this in a l4-proto specific function, - * and call this function here. */ - if (!ip_nat_seq_adjust(pskb, ct, ctinfo)) - return NF_DROP; - } - if (mtype == IP_NAT_MANIP_SRC) statusbit = IPS_SRC_NAT; else diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c index dec4a74212c..79f56f662b3 100644 --- a/net/ipv4/netfilter/ip_nat_standalone.c +++ b/net/ipv4/netfilter/ip_nat_standalone.c @@ -230,6 +230,25 @@ ip_nat_local_fn(unsigned int hooknum, return ret; } +static unsigned int +ip_nat_adjust(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct ip_conntrack *ct; + enum ip_conntrack_info ctinfo; + + ct = ip_conntrack_get(*pskb, &ctinfo); + if (ct && test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) { + DEBUGP("ip_nat_standalone: adjusting sequence number\n"); + if (!ip_nat_seq_adjust(pskb, ct, ctinfo)) + return NF_DROP; + } + return NF_ACCEPT; +} + /* We must be after connection tracking and before packet filtering. */ /* Before packet filtering, change destination */ @@ -250,6 +269,15 @@ static struct nf_hook_ops ip_nat_out_ops = { .priority = NF_IP_PRI_NAT_SRC, }; +/* After conntrack, adjust sequence number */ +static struct nf_hook_ops ip_nat_adjust_out_ops = { + .hook = ip_nat_adjust, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_NAT_SEQ_ADJUST, +}; + /* Before packet filtering, change destination */ static struct nf_hook_ops ip_nat_local_out_ops = { .hook = ip_nat_local_fn, @@ -268,6 +296,16 @@ static struct nf_hook_ops ip_nat_local_in_ops = { .priority = NF_IP_PRI_NAT_SRC, }; +/* After conntrack, adjust sequence number */ +static struct nf_hook_ops ip_nat_adjust_in_ops = { + .hook = ip_nat_adjust, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_NAT_SEQ_ADJUST, +}; + + static int init_or_cleanup(int init) { int ret = 0; @@ -296,10 +334,20 @@ static int init_or_cleanup(int init) printk("ip_nat_init: can't register out hook.\n"); goto cleanup_inops; } + ret = nf_register_hook(&ip_nat_adjust_in_ops); + if (ret < 0) { + printk("ip_nat_init: can't register adjust in hook.\n"); + goto cleanup_outops; + } + ret = nf_register_hook(&ip_nat_adjust_out_ops); + if (ret < 0) { + printk("ip_nat_init: can't register adjust out hook.\n"); + goto cleanup_adjustin_ops; + } ret = nf_register_hook(&ip_nat_local_out_ops); if (ret < 0) { printk("ip_nat_init: can't register local out hook.\n"); - goto cleanup_outops; + goto cleanup_adjustout_ops;; } ret = nf_register_hook(&ip_nat_local_in_ops); if (ret < 0) { @@ -312,6 +360,10 @@ static int init_or_cleanup(int init) nf_unregister_hook(&ip_nat_local_in_ops); cleanup_localoutops: nf_unregister_hook(&ip_nat_local_out_ops); + cleanup_adjustout_ops: + nf_unregister_hook(&ip_nat_adjust_out_ops); + cleanup_adjustin_ops: + nf_unregister_hook(&ip_nat_adjust_in_ops); cleanup_outops: nf_unregister_hook(&ip_nat_out_ops); cleanup_inops: diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c index 9e40dffc204..e5746b67441 100644 --- a/net/ipv4/netfilter/ip_queue.c +++ b/net/ipv4/netfilter/ip_queue.c @@ -546,20 +546,18 @@ ipq_rcv_skb(struct sk_buff *skb) static void ipq_rcv_sk(struct sock *sk, int len) { - do { - struct sk_buff *skb; + struct sk_buff *skb; + unsigned int qlen; - if (down_trylock(&ipqnl_sem)) - return; + down(&ipqnl_sem); - while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) { - ipq_rcv_skb(skb); - kfree_skb(skb); - } + for (qlen = skb_queue_len(&sk->sk_receive_queue); qlen; qlen--) { + skb = skb_dequeue(&sk->sk_receive_queue); + ipq_rcv_skb(skb); + kfree_skb(skb); + } - up(&ipqnl_sem); - - } while (ipqnl && ipqnl->sk_receive_queue.qlen); + up(&ipqnl_sem); } static int diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index 01b4a3c814d..47449ba83eb 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -103,13 +103,15 @@ static struct nf_hook_ops ipt_ops[] = { .hook = ipt_hook, .pf = PF_INET, .hooknum = NF_IP_PRE_ROUTING, - .priority = NF_IP_PRI_RAW + .priority = NF_IP_PRI_RAW, + .owner = THIS_MODULE, }, { .hook = ipt_hook, .pf = PF_INET, .hooknum = NF_IP_LOCAL_OUT, - .priority = NF_IP_PRI_RAW + .priority = NF_IP_PRI_RAW, + .owner = THIS_MODULE, }, }; diff --git a/net/ipv4/protocol.c b/net/ipv4/protocol.c index 90a587cacaa..0db405a869f 100644 --- a/net/ipv4/protocol.c +++ b/net/ipv4/protocol.c @@ -7,7 +7,7 @@ * * Version: $Id: protocol.c,v 1.14 2001/05/18 02:25:49 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * Fixes: diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 93624a32eb9..5b1ec586bae 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -7,7 +7,7 @@ * * Version: $Id: raw.c,v 1.64 2002/02/01 22:01:04 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * Fixes: diff --git a/net/ipv4/route.c b/net/ipv4/route.c index bb90a0c3a91..a682d28e247 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -7,7 +7,7 @@ * * Version: $Id: route.c,v 1.103 2002/01/12 07:44:09 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Alan Cox, <gw4pts@gw4pts.ampr.org> * Linus Torvalds, <Linus.Torvalds@helsinki.fi> @@ -397,7 +397,7 @@ static int rt_cpu_seq_show(struct seq_file *seq, void *v) struct rt_cache_stat *st = v; if (v == SEQ_START_TOKEN) { - seq_printf(seq, "entries in_hit in_slow_tot in_no_route in_brd in_martian_dst in_martian_src out_hit out_slow_tot out_slow_mc gc_total gc_ignored gc_goal_miss gc_dst_overflow in_hlist_search out_hlist_search\n"); + seq_printf(seq, "entries in_hit in_slow_tot in_slow_mc in_no_route in_brd in_martian_dst in_martian_src out_hit out_slow_tot out_slow_mc gc_total gc_ignored gc_goal_miss gc_dst_overflow in_hlist_search out_hlist_search\n"); return 0; } @@ -2843,7 +2843,7 @@ ctl_table ipv4_route_table[] = { .procname = "flush", .data = &flush_delay, .maxlen = sizeof(int), - .mode = 0644, + .mode = 0200, .proc_handler = &ipv4_sysctl_rtcache_flush, .strategy = &ipv4_sysctl_rtcache_flush_strategy, }, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 5cff56af785..a037bafcba3 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -7,7 +7,7 @@ * * Version: $Id: tcp.c,v 1.216 2002/02/01 22:01:04 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * Corey Minyard <wf-rch!minyard@relay.EU.net> diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 313c1408da3..8faa8948f75 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -777,8 +777,9 @@ static inline void tcpdiag_rcv_skb(struct sk_buff *skb) static void tcpdiag_rcv(struct sock *sk, int len) { struct sk_buff *skb; + unsigned int qlen = skb_queue_len(&sk->sk_receive_queue); - while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) { + while (qlen-- && (skb = skb_dequeue(&sk->sk_receive_queue))) { tcpdiag_rcv_skb(skb); kfree_skb(skb); } diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 25049273590..79835a67a27 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -7,7 +7,7 @@ * * Version: $Id: tcp_input.c,v 1.243 2002/02/01 22:01:04 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * Corey Minyard <wf-rch!minyard@relay.EU.net> @@ -3517,7 +3517,6 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) goto drop; - th = skb->h.th; __skb_pull(skb, th->doff*4); TCP_ECN_accept_cwr(tp, skb); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 3ac6659869c..dad98e4a504 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -222,10 +222,13 @@ static int tcp_v4_get_port(struct sock *sk, unsigned short snum) int rover; spin_lock(&tcp_portalloc_lock); - rover = tcp_port_rover; + if (tcp_port_rover < low) + rover = low; + else + rover = tcp_port_rover; do { rover++; - if (rover < low || rover > high) + if (rover > high) rover = low; head = &tcp_bhash[tcp_bhashfn(rover)]; spin_lock(&head->lock); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index fd70509f0d5..eea1a17a9ac 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -7,7 +7,7 @@ * * Version: $Id: tcp_minisocks.c,v 1.15 2002/02/01 22:01:04 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * Corey Minyard <wf-rch!minyard@relay.EU.net> diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 13c14cb6dee..fa24e7ae1f4 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -7,7 +7,7 @@ * * Version: $Id: tcp_output.c,v 1.146 2002/02/01 22:01:04 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * Corey Minyard <wf-rch!minyard@relay.EU.net> @@ -427,7 +427,7 @@ void tcp_push_one(struct sock *sk, unsigned cur_mss) struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb = sk->sk_send_head; - if (tcp_snd_test(tp, skb, cur_mss, TCP_NAGLE_PUSH)) { + if (tcp_snd_test(sk, skb, cur_mss, TCP_NAGLE_PUSH)) { /* Send it out now. */ TCP_SKB_CB(skb)->when = tcp_time_stamp; tcp_tso_set_push(skb); @@ -440,9 +440,12 @@ void tcp_push_one(struct sock *sk, unsigned cur_mss) } } -void tcp_set_skb_tso_segs(struct sk_buff *skb, unsigned int mss_std) +void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb) { - if (skb->len <= mss_std) { + struct tcp_sock *tp = tcp_sk(sk); + + if (skb->len <= tp->mss_cache_std || + !(sk->sk_route_caps & NETIF_F_TSO)) { /* Avoid the costly divide in the normal * non-TSO case. */ @@ -451,10 +454,10 @@ void tcp_set_skb_tso_segs(struct sk_buff *skb, unsigned int mss_std) } else { unsigned int factor; - factor = skb->len + (mss_std - 1); - factor /= mss_std; + factor = skb->len + (tp->mss_cache_std - 1); + factor /= tp->mss_cache_std; skb_shinfo(skb)->tso_segs = factor; - skb_shinfo(skb)->tso_size = mss_std; + skb_shinfo(skb)->tso_size = tp->mss_cache_std; } } @@ -525,8 +528,8 @@ static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len) } /* Fix up tso_factor for both original and new SKB. */ - tcp_set_skb_tso_segs(skb, tp->mss_cache_std); - tcp_set_skb_tso_segs(buff, tp->mss_cache_std); + tcp_set_skb_tso_segs(sk, skb); + tcp_set_skb_tso_segs(sk, buff); if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) { tp->lost_out += tcp_skb_pcount(skb); @@ -601,7 +604,7 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len) * factor and mss. */ if (tcp_skb_pcount(skb) > 1) - tcp_set_skb_tso_segs(skb, tcp_skb_mss(skb)); + tcp_set_skb_tso_segs(sk, skb); return 0; } @@ -752,7 +755,7 @@ int tcp_write_xmit(struct sock *sk, int nonagle) mss_now = tcp_current_mss(sk, 1); while ((skb = sk->sk_send_head) && - tcp_snd_test(tp, skb, mss_now, + tcp_snd_test(sk, skb, mss_now, tcp_skb_is_last(sk, skb) ? nonagle : TCP_NAGLE_PUSH)) { if (skb->len > mss_now) { @@ -1676,7 +1679,7 @@ int tcp_write_wakeup(struct sock *sk) tp->mss_cache = tp->mss_cache_std; } } else if (!tcp_skb_pcount(skb)) - tcp_set_skb_tso_segs(skb, tp->mss_cache_std); + tcp_set_skb_tso_segs(sk, skb); TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH; TCP_SKB_CB(skb)->when = tcp_time_stamp; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 85b279f1e93..799ebe061e2 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -7,7 +7,7 @@ * * Version: $Id: tcp_timer.c,v 1.88 2002/02/01 22:01:04 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Mark Evans, <evansmp@uhura.aston.ac.uk> * Corey Minyard <wf-rch!minyard@relay.EU.net> diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 8a213238f28..4a6952e3fee 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -7,7 +7,7 @@ * * Version: $Id: udp.c,v 1.102 2002/02/01 22:01:04 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Arnt Gulbrandsen, <agulbra@nvg.unit.no> * Alan Cox, <Alan.Cox@linux.org> diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 7fe2afd2e66..b2b60f3e9cd 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -8,7 +8,10 @@ * */ +#include <asm/bug.h> +#include <linux/compiler.h> #include <linux/config.h> +#include <linux/inetdevice.h> #include <net/xfrm.h> #include <net/ip.h> @@ -152,6 +155,8 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int x->u.rt.rt_dst = rt0->rt_dst; x->u.rt.rt_gateway = rt->rt_gateway; x->u.rt.rt_spec_dst = rt0->rt_spec_dst; + x->u.rt.idev = rt0->idev; + in_dev_hold(rt0->idev); header_len -= x->u.dst.xfrm->props.header_len; trailer_len -= x->u.dst.xfrm->props.trailer_len; } @@ -243,11 +248,48 @@ static void xfrm4_update_pmtu(struct dst_entry *dst, u32 mtu) path->ops->update_pmtu(path, mtu); } +static void xfrm4_dst_destroy(struct dst_entry *dst) +{ + struct xfrm_dst *xdst = (struct xfrm_dst *)dst; + + if (likely(xdst->u.rt.idev)) + in_dev_put(xdst->u.rt.idev); + xfrm_dst_destroy(xdst); +} + +static void xfrm4_dst_ifdown(struct dst_entry *dst, struct net_device *dev, + int unregister) +{ + struct xfrm_dst *xdst; + + if (!unregister) + return; + + xdst = (struct xfrm_dst *)dst; + if (xdst->u.rt.idev->dev == dev) { + struct in_device *loopback_idev = in_dev_get(&loopback_dev); + BUG_ON(!loopback_idev); + + do { + in_dev_put(xdst->u.rt.idev); + xdst->u.rt.idev = loopback_idev; + in_dev_hold(loopback_idev); + xdst = (struct xfrm_dst *)xdst->u.dst.child; + } while (xdst->u.dst.xfrm); + + __in_dev_put(loopback_idev); + } + + xfrm_dst_ifdown(dst, dev); +} + static struct dst_ops xfrm4_dst_ops = { .family = AF_INET, .protocol = __constant_htons(ETH_P_IP), .gc = xfrm4_garbage_collect, .update_pmtu = xfrm4_update_pmtu, + .destroy = xfrm4_dst_destroy, + .ifdown = xfrm4_dst_ifdown, .gc_thresh = 1024, .entry_size = sizeof(struct xfrm_dst), }; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 7196ac2f2d1..7744a259269 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3076,7 +3076,7 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev, netlink_broadcast(rtnl, skb, 0, RTMGRP_IPV6_PREFIX, GFP_ATOMIC); } -static struct rtnetlink_link inet6_rtnetlink_table[RTM_MAX - RTM_BASE + 1] = { +static struct rtnetlink_link inet6_rtnetlink_table[RTM_NR_MSGTYPES] = { [RTM_GETLINK - RTM_BASE] = { .dumpit = inet6_dump_ifinfo, }, [RTM_NEWADDR - RTM_BASE] = { .doit = inet6_rtm_newaddr, }, [RTM_DELADDR - RTM_BASE] = { .doit = inet6_rtm_deladdr, }, diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 768b11703da..2b193e3df49 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -28,7 +28,6 @@ #include <linux/socket.h> #include <linux/in.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/sched.h> #include <linux/timer.h> #include <linux/string.h> @@ -88,6 +87,7 @@ int sysctl_ipv6_bindv6only; #ifdef INET_REFCNT_DEBUG atomic_t inet6_sock_nr; +EXPORT_SYMBOL(inet6_sock_nr); #endif /* The inetsw table contains everything that inet_create needs to diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c index 6dda815c013..315bc1fbec3 100644 --- a/net/ipv6/exthdrs_core.c +++ b/net/ipv6/exthdrs_core.c @@ -41,8 +41,8 @@ int ipv6_ext_hdr(u8 nexthdr) * when Linux implements ESP (and maybe AUTH) headers. * --AK * - * This function parses (probably truncated) exthdr set "hdr" - * of length "len". "nexthdrp" initially points to some place, + * This function parses (probably truncated) exthdr set "hdr". + * "nexthdrp" initially points to some place, * where type of the first header can be found. * * It skips all well-known exthdrs, and returns pointer to the start @@ -63,7 +63,7 @@ int ipv6_ext_hdr(u8 nexthdr) * --ANK (980726) */ -int ipv6_skip_exthdr(const struct sk_buff *skb, int start, u8 *nexthdrp, int len) +int ipv6_skip_exthdr(const struct sk_buff *skb, int start, u8 *nexthdrp) { u8 nexthdr = *nexthdrp; @@ -71,13 +71,11 @@ int ipv6_skip_exthdr(const struct sk_buff *skb, int start, u8 *nexthdrp, int len struct ipv6_opt_hdr _hdr, *hp; int hdrlen; - if (len < (int)sizeof(struct ipv6_opt_hdr)) - return -1; if (nexthdr == NEXTHDR_NONE) return -1; hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr); if (hp == NULL) - BUG(); + return -1; if (nexthdr == NEXTHDR_FRAGMENT) { unsigned short _frag_off, *fp; fp = skb_header_pointer(skb, @@ -97,7 +95,6 @@ int ipv6_skip_exthdr(const struct sk_buff *skb, int start, u8 *nexthdrp, int len hdrlen = ipv6_optlen(hp); nexthdr = hp->nexthdr; - len -= hdrlen; start += hdrlen; } diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 87b9082ceab..8e0f569b883 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -135,7 +135,7 @@ static int is_ineligible(struct sk_buff *skb) if (len < 0) return 1; - ptr = ipv6_skip_exthdr(skb, ptr, &nexthdr, len); + ptr = ipv6_skip_exthdr(skb, ptr, &nexthdr); if (ptr < 0) return 0; if (nexthdr == IPPROTO_ICMPV6) { @@ -514,7 +514,7 @@ static void icmpv6_notify(struct sk_buff *skb, int type, int code, u32 info) nexthdr = ((struct ipv6hdr *)skb->data)->nexthdr; if (ipv6_ext_hdr(nexthdr)) { /* now skip over extension headers */ - inner_offset = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr, skb->len - sizeof(struct ipv6hdr)); + inner_offset = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr); if (inner_offset<0) return; } else { diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c index c54830b8959..750943e2d34 100644 --- a/net/ipv6/netfilter/ip6_queue.c +++ b/net/ipv6/netfilter/ip6_queue.c @@ -549,20 +549,18 @@ ipq_rcv_skb(struct sk_buff *skb) static void ipq_rcv_sk(struct sock *sk, int len) { - do { - struct sk_buff *skb; + struct sk_buff *skb; + unsigned int qlen; - if (down_trylock(&ipqnl_sem)) - return; + down(&ipqnl_sem); - while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) { - ipq_rcv_skb(skb); - kfree_skb(skb); - } + for (qlen = skb_queue_len(&sk->sk_receive_queue); qlen; qlen--) { + skb = skb_dequeue(&sk->sk_receive_queue); + ipq_rcv_skb(skb); + kfree_skb(skb); + } - up(&ipqnl_sem); - - } while (ipqnl && ipqnl->sk_receive_queue.qlen); + up(&ipqnl_sem); } static int diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 1352c1d9bf4..617645bc5ed 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -455,11 +455,11 @@ csum_copy_err: static int rawv6_push_pending_frames(struct sock *sk, struct flowi *fl, struct raw6_sock *rp) { - struct inet_sock *inet = inet_sk(sk); struct sk_buff *skb; int err = 0; int offset; int len; + int total_len; u32 tmp_csum; u16 csum; @@ -470,7 +470,8 @@ static int rawv6_push_pending_frames(struct sock *sk, struct flowi *fl, goto out; offset = rp->offset; - if (offset >= inet->cork.length - 1) { + total_len = inet_sk(sk)->cork.length - (skb->nh.raw - skb->data); + if (offset >= total_len - 1) { err = -EINVAL; ip6_flush_pending_frames(sk); goto out; @@ -514,7 +515,7 @@ static int rawv6_push_pending_frames(struct sock *sk, struct flowi *fl, tmp_csum = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst, - inet->cork.length, fl->proto, tmp_csum); + total_len, fl->proto, tmp_csum); if (tmp_csum == 0) tmp_csum = -1; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 183802902c0..3bf8a0254f8 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2005,7 +2005,7 @@ ctl_table ipv6_route_table[] = { .procname = "flush", .data = &flush_delay, .maxlen = sizeof(int), - .mode = 0644, + .mode = 0200, .proc_handler = &ipv6_sysctl_rtcache_flush }, { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 4760c85e19d..0f69e800a0a 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -139,9 +139,12 @@ static int tcp_v6_get_port(struct sock *sk, unsigned short snum) int rover; spin_lock(&tcp_portalloc_lock); - rover = tcp_port_rover; + if (tcp_port_rover < low) + rover = low; + else + rover = tcp_port_rover; do { rover++; - if ((rover < low) || (rover > high)) + if (rover > high) rover = low; head = &tcp_bhash[tcp_bhashfn(rover)]; spin_lock(&head->lock); diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 8a4f37de4d2..4429b1a1fe5 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -11,7 +11,11 @@ * */ +#include <asm/bug.h> +#include <linux/compiler.h> #include <linux/config.h> +#include <linux/netdevice.h> +#include <net/addrconf.h> #include <net/xfrm.h> #include <net/ip.h> #include <net/ipv6.h> @@ -166,6 +170,8 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int memcpy(&x->u.rt6.rt6i_gateway, &rt0->rt6i_gateway, sizeof(x->u.rt6.rt6i_gateway)); x->u.rt6.rt6i_dst = rt0->rt6i_dst; x->u.rt6.rt6i_src = rt0->rt6i_src; + x->u.rt6.rt6i_idev = rt0->rt6i_idev; + in6_dev_hold(rt0->rt6i_idev); header_len -= x->u.dst.xfrm->props.header_len; trailer_len -= x->u.dst.xfrm->props.trailer_len; } @@ -251,11 +257,48 @@ static void xfrm6_update_pmtu(struct dst_entry *dst, u32 mtu) path->ops->update_pmtu(path, mtu); } +static void xfrm6_dst_destroy(struct dst_entry *dst) +{ + struct xfrm_dst *xdst = (struct xfrm_dst *)dst; + + if (likely(xdst->u.rt6.rt6i_idev)) + in6_dev_put(xdst->u.rt6.rt6i_idev); + xfrm_dst_destroy(xdst); +} + +static void xfrm6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, + int unregister) +{ + struct xfrm_dst *xdst; + + if (!unregister) + return; + + xdst = (struct xfrm_dst *)dst; + if (xdst->u.rt6.rt6i_idev->dev == dev) { + struct inet6_dev *loopback_idev = in6_dev_get(&loopback_dev); + BUG_ON(!loopback_idev); + + do { + in6_dev_put(xdst->u.rt6.rt6i_idev); + xdst->u.rt6.rt6i_idev = loopback_idev; + in6_dev_hold(loopback_idev); + xdst = (struct xfrm_dst *)xdst->u.dst.child; + } while (xdst->u.dst.xfrm); + + __in6_dev_put(loopback_idev); + } + + xfrm_dst_ifdown(dst, dev); +} + static struct dst_ops xfrm6_dst_ops = { .family = AF_INET6, .protocol = __constant_htons(ETH_P_IPV6), .gc = xfrm6_garbage_collect, .update_pmtu = xfrm6_update_pmtu, + .destroy = xfrm6_dst_destroy, + .ifdown = xfrm6_dst_ifdown, .gc_thresh = 1024, .entry_size = sizeof(struct xfrm_dst), }; diff --git a/net/irda/irda_device.c b/net/irda/irda_device.c index d6ccd3239dc..70543d89438 100644 --- a/net/irda/irda_device.c +++ b/net/irda/irda_device.c @@ -470,6 +470,7 @@ void irda_device_unregister_dongle(struct dongle_reg *dongle) } EXPORT_SYMBOL(irda_device_unregister_dongle); +#ifdef CONFIG_ISA_DMA_API /* * Function setup_dma (idev, buffer, count, mode) * @@ -492,3 +493,4 @@ void irda_setup_dma(int channel, dma_addr_t buffer, int count, int mode) release_dma_lock(flags); } EXPORT_SYMBOL(irda_setup_dma); +#endif diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 1d5905c90cd..733bf52cef3 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -21,7 +21,6 @@ #include <linux/kernel.h> #include <linux/init.h> -#include <linux/major.h> #include <linux/signal.h> #include <linux/sched.h> #include <linux/errno.h> @@ -50,6 +49,8 @@ #include <linux/bitops.h> #include <linux/mm.h> #include <linux/types.h> +#include <linux/audit.h> + #include <net/sock.h> #include <net/scm.h> @@ -374,7 +375,6 @@ static int netlink_release(struct socket *sock) nlk->cb->done(nlk->cb); netlink_destroy_callback(nlk->cb); nlk->cb = NULL; - __sock_put(sk); } spin_unlock(&nlk->cb_lock); @@ -906,6 +906,7 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, NETLINK_CB(skb).groups = nlk->groups; NETLINK_CB(skb).dst_pid = dst_pid; NETLINK_CB(skb).dst_groups = dst_groups; + NETLINK_CB(skb).loginuid = audit_get_loginuid(current->audit_context); memcpy(NETLINK_CREDS(skb), &siocb->scm->creds, sizeof(struct ucred)); /* What can I do? Netlink is asynchronous, so that @@ -1100,7 +1101,6 @@ static int netlink_dump(struct sock *sk) spin_unlock(&nlk->cb_lock); netlink_destroy_callback(cb); - __sock_put(sk); return 0; } @@ -1139,7 +1139,6 @@ int netlink_dump_start(struct sock *ssk, struct sk_buff *skb, return -EBUSY; } nlk->cb = cb; - sock_hold(sk); spin_unlock(&nlk->cb_lock); netlink_dump(sk); diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 64acea0adaa..0269616e75a 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -7,7 +7,7 @@ * * Version: $Id: af_packet.c,v 1.61 2002/02/08 03:57:19 davem Exp $ * - * Authors: Ross Biro, <bir7@leland.Stanford.Edu> + * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Alan Cox, <gw4pts@gw4pts.ampr.org> * diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 3d1d902dd1a..b0941186f86 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -185,7 +185,7 @@ config NET_SCH_GRED depends on NET_SCHED help Say Y here if you want to use the Generic Random Early Detection - (RED) packet scheduling algorithm for some of your network devices + (GRED) packet scheduling algorithm for some of your network devices (see the top of <file:net/sched/sch_red.c> for details and references about the algorithm). @@ -506,3 +506,13 @@ config NET_CLS_POLICE Say Y to support traffic policing (bandwidth limits). Needed for ingress and egress rate limiting. +config NET_ACT_SIMP + tristate "Simple action" + depends on NET_CLS_ACT + ---help--- + You must have new iproute2 to use this feature. + This adds a very simple action for demonstration purposes + The idea is to give action authors a basic example to look at. + All this action will do is print on the console the configured + policy string followed by _ then packet count. + diff --git a/net/sched/Makefile b/net/sched/Makefile index 431e55786ef..eb3fe583eba 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -6,13 +6,14 @@ obj-y := sch_generic.o obj-$(CONFIG_NET_SCHED) += sch_api.o sch_fifo.o obj-$(CONFIG_NET_CLS) += cls_api.o -obj-$(CONFIG_NET_CLS_ACT) += act_api.o +obj-$(CONFIG_NET_CLS_ACT) += act_api.o obj-$(CONFIG_NET_ACT_POLICE) += police.o obj-$(CONFIG_NET_CLS_POLICE) += police.o -obj-$(CONFIG_NET_ACT_GACT) += gact.o -obj-$(CONFIG_NET_ACT_MIRRED) += mirred.o -obj-$(CONFIG_NET_ACT_IPT) += ipt.o -obj-$(CONFIG_NET_ACT_PEDIT) += pedit.o +obj-$(CONFIG_NET_ACT_GACT) += gact.o +obj-$(CONFIG_NET_ACT_MIRRED) += mirred.o +obj-$(CONFIG_NET_ACT_IPT) += ipt.o +obj-$(CONFIG_NET_ACT_PEDIT) += pedit.o +obj-$(CONFIG_NET_ACT_SIMP) += simple.o obj-$(CONFIG_NET_SCH_CBQ) += sch_cbq.o obj-$(CONFIG_NET_SCH_HTB) += sch_htb.o obj-$(CONFIG_NET_SCH_HPFQ) += sch_hpfq.o diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 5e6cc371b39..cafcb084098 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -171,10 +171,10 @@ repeat: skb->tc_verd = SET_TC_OK2MUNGE(skb->tc_verd); skb->tc_verd = CLR_TC_MUNGED(skb->tc_verd); } - if (ret != TC_ACT_PIPE) - goto exec_done; if (ret == TC_ACT_REPEAT) goto repeat; /* we need a ttl - JHS */ + if (ret != TC_ACT_PIPE) + goto exec_done; } act = a->next; } diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c index fdfc83af3d1..29d8b9a4d16 100644 --- a/net/sched/cls_fw.c +++ b/net/sched/cls_fw.c @@ -46,9 +46,11 @@ #include <net/act_api.h> #include <net/pkt_cls.h> +#define HTSIZE (PAGE_SIZE/sizeof(struct fw_filter *)) + struct fw_head { - struct fw_filter *ht[256]; + struct fw_filter *ht[HTSIZE]; }; struct fw_filter @@ -69,7 +71,28 @@ static struct tcf_ext_map fw_ext_map = { static __inline__ int fw_hash(u32 handle) { - return handle&0xFF; + if (HTSIZE == 4096) + return ((handle >> 24) & 0xFFF) ^ + ((handle >> 12) & 0xFFF) ^ + (handle & 0xFFF); + else if (HTSIZE == 2048) + return ((handle >> 22) & 0x7FF) ^ + ((handle >> 11) & 0x7FF) ^ + (handle & 0x7FF); + else if (HTSIZE == 1024) + return ((handle >> 20) & 0x3FF) ^ + ((handle >> 10) & 0x3FF) ^ + (handle & 0x3FF); + else if (HTSIZE == 512) + return (handle >> 27) ^ + ((handle >> 18) & 0x1FF) ^ + ((handle >> 9) & 0x1FF) ^ + (handle & 0x1FF); + else if (HTSIZE == 256) { + u8 *t = (u8 *) &handle; + return t[0] ^ t[1] ^ t[2] ^ t[3]; + } else + return handle & (HTSIZE - 1); } static int fw_classify(struct sk_buff *skb, struct tcf_proto *tp, @@ -152,7 +175,7 @@ static void fw_destroy(struct tcf_proto *tp) if (head == NULL) return; - for (h=0; h<256; h++) { + for (h=0; h<HTSIZE; h++) { while ((f=head->ht[h]) != NULL) { head->ht[h] = f->next; fw_delete_filter(tp, f); @@ -291,7 +314,7 @@ static void fw_walk(struct tcf_proto *tp, struct tcf_walker *arg) if (arg->stop) return; - for (h = 0; h < 256; h++) { + for (h = 0; h < HTSIZE; h++) { struct fw_filter *f; for (f = head->ht[h]; f; f = f->next) { diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 4323a74eea3..07977f8f267 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1289,6 +1289,7 @@ static int __init pktsched_init(void) subsys_initcall(pktsched_init); +EXPORT_SYMBOL(qdisc_lookup); EXPORT_SYMBOL(qdisc_get_rtab); EXPORT_SYMBOL(qdisc_put_rtab); EXPORT_SYMBOL(register_qdisc); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 8c01e023f02..87e48a4e105 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -179,6 +179,7 @@ requeue: netif_schedule(dev); return 1; } + BUG_ON((int) q->q.qlen < 0); return q->q.qlen; } @@ -539,6 +540,10 @@ void dev_activate(struct net_device *dev) write_unlock_bh(&qdisc_tree_lock); } + if (!netif_carrier_ok(dev)) + /* Delay activation until next carrier-on event */ + return; + spin_lock_bh(&dev->queue_lock); rcu_assign_pointer(dev->qdisc, dev->qdisc_sleeping); if (dev->qdisc != &noqueue_qdisc) { diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index a85935e7d53..558cc087e60 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -717,6 +717,10 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (q->direct_queue.qlen < q->direct_qlen) { __skb_queue_tail(&q->direct_queue, skb); q->direct_pkts++; + } else { + kfree_skb(skb); + sch->qstats.drops++; + return NET_XMIT_DROP; } #ifdef CONFIG_NET_CLS_ACT } else if (!cl) { diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 31c29deb139..e0c9fbe73b1 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -138,38 +138,77 @@ static long tabledist(unsigned long mu, long sigma, } /* Put skb in the private delayed queue. */ -static int delay_skb(struct Qdisc *sch, struct sk_buff *skb) +static int netem_delay(struct Qdisc *sch, struct sk_buff *skb) { struct netem_sched_data *q = qdisc_priv(sch); - struct netem_skb_cb *cb = (struct netem_skb_cb *)skb->cb; psched_tdiff_t td; psched_time_t now; PSCHED_GET_TIME(now); td = tabledist(q->latency, q->jitter, &q->delay_cor, q->delay_dist); - PSCHED_TADD2(now, td, cb->time_to_send); /* Always queue at tail to keep packets in order */ if (likely(q->delayed.qlen < q->limit)) { + struct netem_skb_cb *cb = (struct netem_skb_cb *)skb->cb; + + PSCHED_TADD2(now, td, cb->time_to_send); + + pr_debug("netem_delay: skb=%p now=%llu tosend=%llu\n", skb, + now, cb->time_to_send); + __skb_queue_tail(&q->delayed, skb); - if (!timer_pending(&q->timer)) { - q->timer.expires = jiffies + PSCHED_US2JIFFIE(td); - add_timer(&q->timer); - } return NET_XMIT_SUCCESS; } + pr_debug("netem_delay: queue over limit %d\n", q->limit); + sch->qstats.overlimits++; kfree_skb(skb); return NET_XMIT_DROP; } +/* + * Move a packet that is ready to send from the delay holding + * list to the underlying qdisc. + */ +static int netem_run(struct Qdisc *sch) +{ + struct netem_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + psched_time_t now; + + PSCHED_GET_TIME(now); + + skb = skb_peek(&q->delayed); + if (skb) { + const struct netem_skb_cb *cb + = (const struct netem_skb_cb *)skb->cb; + long delay + = PSCHED_US2JIFFIE(PSCHED_TDIFF(cb->time_to_send, now)); + pr_debug("netem_run: skb=%p delay=%ld\n", skb, delay); + + /* if more time remaining? */ + if (delay > 0) { + mod_timer(&q->timer, jiffies + delay); + return 1; + } + + __skb_unlink(skb, &q->delayed); + + if (q->qdisc->enqueue(skb, q->qdisc)) { + sch->q.qlen--; + sch->qstats.drops++; + } + } + + return 0; +} + static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); - struct sk_buff *skb2; int ret; - pr_debug("netem_enqueue skb=%p @%lu\n", skb, jiffies); + pr_debug("netem_enqueue skb=%p\n", skb); /* Random packet drop 0 => none, ~0 => all */ if (q->loss && q->loss >= get_crandom(&q->loss_cor)) { @@ -180,11 +219,21 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) } /* Random duplication */ - if (q->duplicate && q->duplicate >= get_crandom(&q->dup_cor) - && (skb2 = skb_clone(skb, GFP_ATOMIC)) != NULL) { - pr_debug("netem_enqueue: dup %p\n", skb2); + if (q->duplicate && q->duplicate >= get_crandom(&q->dup_cor)) { + struct sk_buff *skb2; + + skb2 = skb_clone(skb, GFP_ATOMIC); + if (skb2 && netem_delay(sch, skb2) == NET_XMIT_SUCCESS) { + struct Qdisc *qp; + + /* Since one packet can generate two packets in the + * queue, the parent's qlen accounting gets confused, + * so fix it. + */ + qp = qdisc_lookup(sch->dev, TC_H_MAJ(sch->parent)); + if (qp) + qp->q.qlen++; - if (delay_skb(sch, skb2)) { sch->q.qlen++; sch->bstats.bytes += skb2->len; sch->bstats.packets++; @@ -202,7 +251,8 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) ret = q->qdisc->enqueue(skb, q->qdisc); } else { q->counter = 0; - ret = delay_skb(sch, skb); + ret = netem_delay(sch, skb); + netem_run(sch); } if (likely(ret == NET_XMIT_SUCCESS)) { @@ -212,6 +262,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) } else sch->qstats.drops++; + pr_debug("netem: enqueue ret %d\n", ret); return ret; } @@ -241,56 +292,35 @@ static unsigned int netem_drop(struct Qdisc* sch) return len; } -/* Dequeue packet. - * Move all packets that are ready to send from the delay holding - * list to the underlying qdisc, then just call dequeue - */ static struct sk_buff *netem_dequeue(struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); struct sk_buff *skb; + int pending; + + pending = netem_run(sch); skb = q->qdisc->dequeue(q->qdisc); - if (skb) + if (skb) { + pr_debug("netem_dequeue: return skb=%p\n", skb); sch->q.qlen--; + sch->flags &= ~TCQ_F_THROTTLED; + } + else if (pending) { + pr_debug("netem_dequeue: throttling\n"); + sch->flags |= TCQ_F_THROTTLED; + } + return skb; } static void netem_watchdog(unsigned long arg) { struct Qdisc *sch = (struct Qdisc *)arg; - struct netem_sched_data *q = qdisc_priv(sch); - struct net_device *dev = sch->dev; - struct sk_buff *skb; - psched_time_t now; - - pr_debug("netem_watchdog: fired @%lu\n", jiffies); - - spin_lock_bh(&dev->queue_lock); - PSCHED_GET_TIME(now); - - while ((skb = skb_peek(&q->delayed)) != NULL) { - const struct netem_skb_cb *cb - = (const struct netem_skb_cb *)skb->cb; - long delay - = PSCHED_US2JIFFIE(PSCHED_TDIFF(cb->time_to_send, now)); - pr_debug("netem_watchdog: skb %p@%lu %ld\n", - skb, jiffies, delay); - /* if more time remaining? */ - if (delay > 0) { - mod_timer(&q->timer, jiffies + delay); - break; - } - __skb_unlink(skb, &q->delayed); - - if (q->qdisc->enqueue(skb, q->qdisc)) { - sch->q.qlen--; - sch->qstats.drops++; - } - } - qdisc_run(dev); - spin_unlock_bh(&dev->queue_lock); + pr_debug("netem_watchdog qlen=%d\n", sch->q.qlen); + sch->flags &= ~TCQ_F_THROTTLED; + netif_schedule(sch->dev); } static void netem_reset(struct Qdisc *sch) @@ -301,6 +331,7 @@ static void netem_reset(struct Qdisc *sch) skb_queue_purge(&q->delayed); sch->q.qlen = 0; + sch->flags &= ~TCQ_F_THROTTLED; del_timer_sync(&q->timer); } diff --git a/net/sched/simple.c b/net/sched/simple.c new file mode 100644 index 00000000000..3ab4c675ab5 --- /dev/null +++ b/net/sched/simple.c @@ -0,0 +1,93 @@ +/* + * net/sched/simp.c Simple example of an action + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Authors: Jamal Hadi Salim (2005) + * + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/netdevice.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <net/pkt_sched.h> + +#define TCA_ACT_SIMP 22 + +/* XXX: Hide all these common elements under some macro + * probably +*/ +#include <linux/tc_act/tc_defact.h> +#include <net/tc_act/tc_defact.h> + +/* use generic hash table with 8 buckets */ +#define MY_TAB_SIZE 8 +#define MY_TAB_MASK (MY_TAB_SIZE - 1) +static u32 idx_gen; +static struct tcf_defact *tcf_simp_ht[MY_TAB_SIZE]; +static DEFINE_RWLOCK(simp_lock); + +/* override the defaults */ +#define tcf_st tcf_defact +#define tc_st tc_defact +#define tcf_t_lock simp_lock +#define tcf_ht tcf_simp_ht + +#define CONFIG_NET_ACT_INIT 1 +#include <net/pkt_act.h> +#include <net/act_generic.h> + +static int tcf_simp(struct sk_buff **pskb, struct tc_action *a) +{ + struct sk_buff *skb = *pskb; + struct tcf_defact *p = PRIV(a, defact); + + spin_lock(&p->lock); + p->tm.lastuse = jiffies; + p->bstats.bytes += skb->len; + p->bstats.packets++; + + /* print policy string followed by _ then packet count + * Example if this was the 3rd packet and the string was "hello" + * then it would look like "hello_3" (without quotes) + **/ + printk("simple: %s_%d\n", (char *)p->defdata, p->bstats.packets); + spin_unlock(&p->lock); + return p->action; +} + +static struct tc_action_ops act_simp_ops = { + .kind = "simple", + .type = TCA_ACT_SIMP, + .capab = TCA_CAP_NONE, + .owner = THIS_MODULE, + .act = tcf_simp, + tca_use_default_ops +}; + +MODULE_AUTHOR("Jamal Hadi Salim(2005)"); +MODULE_DESCRIPTION("Simple example action"); +MODULE_LICENSE("GPL"); + +static int __init simp_init_module(void) +{ + int ret = tcf_register_action(&act_simp_ops); + if (!ret) + printk("Simple TC action Loaded\n"); + return ret; +} + +static void __exit simp_cleanup_module(void) +{ + tcf_unregister_action(&act_simp_ops); +} + +module_init(simp_init_module); +module_exit(simp_cleanup_module); diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 544b75077db..334f61773e6 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -125,6 +125,7 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, sp->autoclose * HZ; /* Use SCTP specific send buffer space queues. */ + ep->sndbuf_policy = sctp_sndbuf_policy; sk->sk_write_space = sctp_write_space; sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index e42c74e3ec1..c9d9ea06473 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -496,9 +496,7 @@ static void sctp_v6_inaddr_any(union sctp_addr *addr, unsigned short port) /* Is this a wildcard address? */ static int sctp_v6_is_any(const union sctp_addr *addr) { - int type; - type = ipv6_addr_type((struct in6_addr *)&addr->v6.sin6_addr); - return IPV6_ADDR_ANY == type; + return ipv6_addr_any(&addr->v6.sin6_addr); } /* Should this be available for binding? */ diff --git a/net/sctp/output.c b/net/sctp/output.c index 9013f64f521..84b5b370b09 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -313,12 +313,12 @@ int sctp_packet_transmit(struct sctp_packet *packet) sk = chunk->skb->sk; /* Allocate the new skb. */ - nskb = dev_alloc_skb(packet->size); + nskb = alloc_skb(packet->size + LL_MAX_HEADER, GFP_ATOMIC); if (!nskb) goto nomem; /* Make sure the outbound skb has enough header room reserved. */ - skb_reserve(nskb, packet->overhead); + skb_reserve(nskb, packet->overhead + LL_MAX_HEADER); /* Set the owning socket so that we know where to get the * destination IP address. diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index b9813cf3d91..2e1f9c3556f 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1043,6 +1043,9 @@ SCTP_STATIC __init int sctp_init(void) sctp_max_retrans_path = 5; sctp_max_retrans_init = 8; + /* Sendbuffer growth - do per-socket accounting */ + sctp_sndbuf_policy = 0; + /* HB.interval - 30 seconds */ sctp_hb_interval = 30 * HZ; @@ -1159,8 +1162,6 @@ SCTP_STATIC __init int sctp_init(void) status = 0; out: return status; -err_add_protocol: - proto_unregister(&sctp_prot); err_ctl_sock_init: sctp_v6_exit(); err_v6_init: @@ -1188,6 +1189,8 @@ err_bucket_cachep: inet_del_protocol(&sctp_protocol, IPPROTO_SCTP); inet_unregister_protosw(&sctp_seqpacket_protosw); inet_unregister_protosw(&sctp_stream_protosw); +err_add_protocol: + proto_unregister(&sctp_prot); goto out; } diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 1db12cc18cf..33ac8bf47b0 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -710,7 +710,9 @@ struct sctp_chunk *sctp_make_shutdown_complete( struct sctp_chunk *retval; __u8 flags = 0; - /* Maybe set the T-bit if we have no association. */ + /* Set the T-bit if we have no association (vtag will be + * reflected) + */ flags |= asoc ? 0 : SCTP_CHUNK_FLAG_T; retval = sctp_make_chunk(asoc, SCTP_CID_SHUTDOWN_COMPLETE, flags, 0); @@ -732,7 +734,7 @@ struct sctp_chunk *sctp_make_shutdown_complete( } /* Create an ABORT. Note that we set the T bit if we have no - * association. + * association, except when responding to an INIT (sctpimpguide 2.41). */ struct sctp_chunk *sctp_make_abort(const struct sctp_association *asoc, const struct sctp_chunk *chunk, @@ -741,8 +743,16 @@ struct sctp_chunk *sctp_make_abort(const struct sctp_association *asoc, struct sctp_chunk *retval; __u8 flags = 0; - /* Maybe set the T-bit if we have no association. */ - flags |= asoc ? 0 : SCTP_CHUNK_FLAG_T; + /* Set the T-bit if we have no association and 'chunk' is not + * an INIT (vtag will be reflected). + */ + if (!asoc) { + if (chunk && chunk->chunk_hdr && + chunk->chunk_hdr->type == SCTP_CID_INIT) + flags = 0; + else + flags = SCTP_CHUNK_FLAG_T; + } retval = sctp_make_chunk(asoc, SCTP_CID_ABORT, flags, hint); @@ -2744,7 +2754,6 @@ struct sctp_chunk *sctp_make_fwdtsn(const struct sctp_association *asoc, hint = (nstreams + 1) * sizeof(__u32); - /* Maybe set the T-bit if we have no association. */ retval = sctp_make_chunk(asoc, SCTP_CID_FWD_TSN, 0, hint); if (!retval) diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 278c56a2d07..8e01b8f09ac 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -126,15 +126,18 @@ sctp_chunk_length_valid(struct sctp_chunk *chunk, * should stop the T2-shutdown timer and remove all knowledge of the * association (and thus the association enters the CLOSED state). * - * Verification Tag: 8.5.1(C) + * Verification Tag: 8.5.1(C), sctpimpguide 2.41. * C) Rules for packet carrying SHUTDOWN COMPLETE: * ... - * - The receiver of a SHUTDOWN COMPLETE shall accept the packet if the - * Verification Tag field of the packet matches its own tag OR it is - * set to its peer's tag and the T bit is set in the Chunk Flags. - * Otherwise, the receiver MUST silently discard the packet and take - * no further action. An endpoint MUST ignore the SHUTDOWN COMPLETE if - * it is not in the SHUTDOWN-ACK-SENT state. + * - The receiver of a SHUTDOWN COMPLETE shall accept the packet + * if the Verification Tag field of the packet matches its own tag and + * the T bit is not set + * OR + * it is set to its peer's tag and the T bit is set in the Chunk + * Flags. + * Otherwise, the receiver MUST silently discard the packet + * and take no further action. An endpoint MUST ignore the + * SHUTDOWN COMPLETE if it is not in the SHUTDOWN-ACK-SENT state. * * Inputs * (endpoint, asoc, chunk) @@ -2858,16 +2861,16 @@ sctp_disposition_t sctp_sf_eat_sack_6_2(const struct sctp_endpoint *ep, /* * Generate an ABORT in response to a packet. * - * Section: 8.4 Handle "Out of the blue" Packets + * Section: 8.4 Handle "Out of the blue" Packets, sctpimpguide 2.41 * - * 8) The receiver should respond to the sender of the OOTB packet - * with an ABORT. When sending the ABORT, the receiver of the - * OOTB packet MUST fill in the Verification Tag field of the - * outbound packet with the value found in the Verification Tag - * field of the OOTB packet and set the T-bit in the Chunk Flags - * to indicate that no TCB was found. After sending this ABORT, - * the receiver of the OOTB packet shall discard the OOTB packet - * and take no further action. + * 8) The receiver should respond to the sender of the OOTB packet with + * an ABORT. When sending the ABORT, the receiver of the OOTB packet + * MUST fill in the Verification Tag field of the outbound packet + * with the value found in the Verification Tag field of the OOTB + * packet and set the T-bit in the Chunk Flags to indicate that the + * Verification Tag is reflected. After sending this ABORT, the + * receiver of the OOTB packet shall discard the OOTB packet and take + * no further action. * * Verification Tag: * @@ -2895,6 +2898,10 @@ sctp_disposition_t sctp_sf_tabort_8_4_8(const struct sctp_endpoint *ep, return SCTP_DISPOSITION_NOMEM; } + /* Reflect vtag if T-Bit is set */ + if (sctp_test_T_bit(abort)) + packet->vtag = ntohl(chunk->sctp_hdr->vtag); + /* Set the skb to the belonging sock for accounting. */ abort->skb->sk = ep->base.sk; @@ -3026,22 +3033,24 @@ nomem: } /* - * RFC 2960, 8.4 - Handle "Out of the blue" Packets + * RFC 2960, 8.4 - Handle "Out of the blue" Packets, sctpimpguide 2.41. + * * 5) If the packet contains a SHUTDOWN ACK chunk, the receiver should * respond to the sender of the OOTB packet with a SHUTDOWN COMPLETE. * When sending the SHUTDOWN COMPLETE, the receiver of the OOTB * packet must fill in the Verification Tag field of the outbound * packet with the Verification Tag received in the SHUTDOWN ACK and - * set the T-bit in the Chunk Flags to indicate that no TCB was - * found. Otherwise, + * set the T-bit in the Chunk Flags to indicate that the Verification + * Tag is reflected. * * 8) The receiver should respond to the sender of the OOTB packet with * an ABORT. When sending the ABORT, the receiver of the OOTB packet * MUST fill in the Verification Tag field of the outbound packet * with the value found in the Verification Tag field of the OOTB - * packet and set the T-bit in the Chunk Flags to indicate that no - * TCB was found. After sending this ABORT, the receiver of the OOTB - * packet shall discard the OOTB packet and take no further action. + * packet and set the T-bit in the Chunk Flags to indicate that the + * Verification Tag is reflected. After sending this ABORT, the + * receiver of the OOTB packet shall discard the OOTB packet and take + * no further action. */ sctp_disposition_t sctp_sf_ootb(const struct sctp_endpoint *ep, const struct sctp_association *asoc, @@ -3090,13 +3099,15 @@ sctp_disposition_t sctp_sf_ootb(const struct sctp_endpoint *ep, /* * Handle an "Out of the blue" SHUTDOWN ACK. * - * Section: 8.4 5) + * Section: 8.4 5, sctpimpguide 2.41. + * * 5) If the packet contains a SHUTDOWN ACK chunk, the receiver should - * respond to the sender of the OOTB packet with a SHUTDOWN COMPLETE. - * When sending the SHUTDOWN COMPLETE, the receiver of the OOTB packet - * must fill in the Verification Tag field of the outbound packet with - * the Verification Tag received in the SHUTDOWN ACK and set the - * T-bit in the Chunk Flags to indicate that no TCB was found. + * respond to the sender of the OOTB packet with a SHUTDOWN COMPLETE. + * When sending the SHUTDOWN COMPLETE, the receiver of the OOTB + * packet must fill in the Verification Tag field of the outbound + * packet with the Verification Tag received in the SHUTDOWN ACK and + * set the T-bit in the Chunk Flags to indicate that the Verification + * Tag is reflected. * * Inputs * (endpoint, asoc, type, arg, commands) @@ -3128,6 +3139,10 @@ static sctp_disposition_t sctp_sf_shut_8_4_5(const struct sctp_endpoint *ep, return SCTP_DISPOSITION_NOMEM; } + /* Reflect vtag if T-Bit is set */ + if (sctp_test_T_bit(shut)) + packet->vtag = ntohl(chunk->sctp_hdr->vtag); + /* Set the skb to the belonging sock for accounting. */ shut->skb->sk = ep->base.sk; @@ -3591,7 +3606,6 @@ sctp_disposition_t sctp_sf_discard_chunk(const struct sctp_endpoint *ep, * * 2) If the OOTB packet contains an ABORT chunk, the receiver MUST * silently discard the OOTB packet and take no further action. - * Otherwise, * * Verification Tag: No verification necessary * @@ -4961,6 +4975,11 @@ static struct sctp_packet *sctp_abort_pkt_new(const struct sctp_endpoint *ep, sctp_ootb_pkt_free(packet); return NULL; } + + /* Reflect vtag if T-Bit is set */ + if (sctp_test_T_bit(abort)) + packet->vtag = ntohl(chunk->sctp_hdr->vtag); + /* Add specified error causes, i.e., payload, to the * end of the chunk. */ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index e8c21018257..0b338eca6dc 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -115,9 +115,17 @@ static inline int sctp_wspace(struct sctp_association *asoc) struct sock *sk = asoc->base.sk; int amt = 0; - amt = sk->sk_sndbuf - asoc->sndbuf_used; + if (asoc->ep->sndbuf_policy) { + /* make sure that no association uses more than sk_sndbuf */ + amt = sk->sk_sndbuf - asoc->sndbuf_used; + } else { + /* do socket level accounting */ + amt = sk->sk_sndbuf - atomic_read(&sk->sk_wmem_alloc); + } + if (amt < 0) amt = 0; + return amt; } @@ -138,12 +146,21 @@ static inline void sctp_set_owner_w(struct sctp_chunk *chunk) /* The sndbuf space is tracked per association. */ sctp_association_hold(asoc); + skb_set_owner_w(chunk->skb, sk); + chunk->skb->destructor = sctp_wfree; /* Save the chunk pointer in skb for sctp_wfree to use later. */ *((struct sctp_chunk **)(chunk->skb->cb)) = chunk; - asoc->sndbuf_used += SCTP_DATA_SNDSIZE(chunk); - sk->sk_wmem_queued += SCTP_DATA_SNDSIZE(chunk); + asoc->sndbuf_used += SCTP_DATA_SNDSIZE(chunk) + + sizeof(struct sk_buff) + + sizeof(struct sctp_chunk); + + sk->sk_wmem_queued += SCTP_DATA_SNDSIZE(chunk) + + sizeof(struct sk_buff) + + sizeof(struct sctp_chunk); + + atomic_add(sizeof(struct sctp_chunk), &sk->sk_wmem_alloc); } /* Verify that this is a valid address. */ @@ -3473,7 +3490,7 @@ static int sctp_getsockopt_associnfo(struct sock *sk, int len, return -EINVAL; /* Values correspoinding to the specific association */ - if (assocparams.sasoc_assoc_id != 0) { + if (asoc) { assocparams.sasoc_asocmaxrxt = asoc->max_retrans; assocparams.sasoc_peer_rwnd = asoc->peer.rwnd; assocparams.sasoc_local_rwnd = asoc->a_rwnd; @@ -4422,8 +4439,17 @@ static void sctp_wfree(struct sk_buff *skb) chunk = *((struct sctp_chunk **)(skb->cb)); asoc = chunk->asoc; sk = asoc->base.sk; - asoc->sndbuf_used -= SCTP_DATA_SNDSIZE(chunk); - sk->sk_wmem_queued -= SCTP_DATA_SNDSIZE(chunk); + asoc->sndbuf_used -= SCTP_DATA_SNDSIZE(chunk) + + sizeof(struct sk_buff) + + sizeof(struct sctp_chunk); + + sk->sk_wmem_queued -= SCTP_DATA_SNDSIZE(chunk) + + sizeof(struct sk_buff) + + sizeof(struct sctp_chunk); + + atomic_sub(sizeof(struct sctp_chunk), &sk->sk_wmem_alloc); + + sock_wfree(skb); __sctp_write_space(asoc); sctp_association_put(asoc); diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index 89fa20c73a5..7fc31849312 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -110,6 +110,14 @@ static ctl_table sctp_table[] = { .proc_handler = &proc_dointvec }, { + .ctl_name = NET_SCTP_SNDBUF_POLICY, + .procname = "sndbuf_policy", + .data = &sctp_sndbuf_policy, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { .ctl_name = NET_SCTP_PATH_MAX_RETRANS, .procname = "path_max_retrans", .data = &sctp_max_retrans_path, diff --git a/net/socket.c b/net/socket.c index 2cd44990d8d..cec0cb38b9c 100644 --- a/net/socket.c +++ b/net/socket.c @@ -4,7 +4,7 @@ * Version: @(#)socket.c 1.1.93 18/02/95 * * Authors: Orest Zborowski, <obz@Kodak.COM> - * Ross Biro, <bir7@leland.Stanford.Edu> + * Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * * Fixes: diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index 4484931018e..67b9f035ba8 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -46,9 +46,9 @@ xdr_decode_netobj(u32 *p, struct xdr_netobj *obj) /** * xdr_encode_opaque_fixed - Encode fixed length opaque data - * @p - pointer to current position in XDR buffer. - * @ptr - pointer to data to encode (or NULL) - * @nbytes - size of data. + * @p: pointer to current position in XDR buffer. + * @ptr: pointer to data to encode (or NULL) + * @nbytes: size of data. * * Copy the array of data of length nbytes at ptr to the XDR buffer * at position p, then align to the next 32-bit boundary by padding @@ -76,9 +76,9 @@ EXPORT_SYMBOL(xdr_encode_opaque_fixed); /** * xdr_encode_opaque - Encode variable length opaque data - * @p - pointer to current position in XDR buffer. - * @ptr - pointer to data to encode (or NULL) - * @nbytes - size of data. + * @p: pointer to current position in XDR buffer. + * @ptr: pointer to data to encode (or NULL) + * @nbytes: size of data. * * Returns the updated current XDR buffer position */ diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index acc73fe6869..c478fc8db77 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -85,7 +85,6 @@ #include <linux/module.h> #include <linux/config.h> #include <linux/kernel.h> -#include <linux/major.h> #include <linux/signal.h> #include <linux/sched.h> #include <linux/errno.h> diff --git a/net/wanrouter/wanmain.c b/net/wanrouter/wanmain.c index 956c17f6c54..d6844ac226f 100644 --- a/net/wanrouter/wanmain.c +++ b/net/wanrouter/wanmain.c @@ -48,8 +48,8 @@ #include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> /* support for loadable modules */ -#include <linux/slab.h> /* kmalloc(), kfree() */ -#include <linux/mm.h> /* verify_area(), etc. */ +#include <linux/slab.h> /* kmalloc(), kfree() */ +#include <linux/mm.h> #include <linux/string.h> /* inline mem*, str* functions */ #include <asm/byteorder.h> /* htons(), etc. */ diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 80828078733..55ed979db14 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -1028,30 +1028,15 @@ static int stale_bundle(struct dst_entry *dst) return !xfrm_bundle_ok((struct xfrm_dst *)dst, NULL, AF_UNSPEC); } -static void xfrm_dst_destroy(struct dst_entry *dst) +void xfrm_dst_ifdown(struct dst_entry *dst, struct net_device *dev) { - struct xfrm_dst *xdst = (struct xfrm_dst *)dst; - - dst_release(xdst->route); - - if (!dst->xfrm) - return; - xfrm_state_put(dst->xfrm); - dst->xfrm = NULL; -} - -static void xfrm_dst_ifdown(struct dst_entry *dst, struct net_device *dev, - int unregister) -{ - if (!unregister) - return; - while ((dst = dst->child) && dst->xfrm && dst->dev == dev) { dst->dev = &loopback_dev; dev_hold(&loopback_dev); dev_put(dev); } } +EXPORT_SYMBOL(xfrm_dst_ifdown); static void xfrm_link_failure(struct sk_buff *skb) { @@ -1262,10 +1247,6 @@ int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo) dst_ops->kmem_cachep = xfrm_dst_cache; if (likely(dst_ops->check == NULL)) dst_ops->check = xfrm_dst_check; - if (likely(dst_ops->destroy == NULL)) - dst_ops->destroy = xfrm_dst_destroy; - if (likely(dst_ops->ifdown == NULL)) - dst_ops->ifdown = xfrm_dst_ifdown; if (likely(dst_ops->negative_advice == NULL)) dst_ops->negative_advice = xfrm_negative_advice; if (likely(dst_ops->link_failure == NULL)) @@ -1297,8 +1278,6 @@ int xfrm_policy_unregister_afinfo(struct xfrm_policy_afinfo *afinfo) xfrm_policy_afinfo[afinfo->family] = NULL; dst_ops->kmem_cachep = NULL; dst_ops->check = NULL; - dst_ops->destroy = NULL; - dst_ops->ifdown = NULL; dst_ops->negative_advice = NULL; dst_ops->link_failure = NULL; dst_ops->get_mss = NULL; diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 1db59f11f37..d11747c2a76 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -357,8 +357,9 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, x = best; if (!x && !error && !acquire_in_progress) { - x0 = afinfo->state_lookup(&tmpl->id.daddr, tmpl->id.spi, tmpl->id.proto); - if (x0 != NULL) { + if (tmpl->id.spi && + (x0 = afinfo->state_lookup(daddr, tmpl->id.spi, + tmpl->id.proto)) != NULL) { xfrm_state_put(x0); error = -EEXIST; goto out; diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 63661b0fd73..5ddda2c98af 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -855,47 +855,44 @@ static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **x return 0; } -static const int xfrm_msg_min[(XFRM_MSG_MAX + 1 - XFRM_MSG_BASE)] = { - NLMSG_LENGTH(sizeof(struct xfrm_usersa_info)), /* NEW SA */ - NLMSG_LENGTH(sizeof(struct xfrm_usersa_id)), /* DEL SA */ - NLMSG_LENGTH(sizeof(struct xfrm_usersa_id)), /* GET SA */ - NLMSG_LENGTH(sizeof(struct xfrm_userpolicy_info)),/* NEW POLICY */ - NLMSG_LENGTH(sizeof(struct xfrm_userpolicy_id)), /* DEL POLICY */ - NLMSG_LENGTH(sizeof(struct xfrm_userpolicy_id)), /* GET POLICY */ - NLMSG_LENGTH(sizeof(struct xfrm_userspi_info)), /* ALLOC SPI */ - NLMSG_LENGTH(sizeof(struct xfrm_user_acquire)), /* ACQUIRE */ - NLMSG_LENGTH(sizeof(struct xfrm_user_expire)), /* EXPIRE */ - NLMSG_LENGTH(sizeof(struct xfrm_userpolicy_info)),/* UPD POLICY */ - NLMSG_LENGTH(sizeof(struct xfrm_usersa_info)), /* UPD SA */ - NLMSG_LENGTH(sizeof(struct xfrm_user_polexpire)), /* POLEXPIRE */ - NLMSG_LENGTH(sizeof(struct xfrm_usersa_flush)), /* FLUSH SA */ - NLMSG_LENGTH(0), /* FLUSH POLICY */ +#define XMSGSIZE(type) NLMSG_LENGTH(sizeof(struct type)) + +static const int xfrm_msg_min[XFRM_NR_MSGTYPES] = { + [XFRM_MSG_NEWSA - XFRM_MSG_BASE] = XMSGSIZE(xfrm_usersa_info), + [XFRM_MSG_DELSA - XFRM_MSG_BASE] = XMSGSIZE(xfrm_usersa_id), + [XFRM_MSG_GETSA - XFRM_MSG_BASE] = XMSGSIZE(xfrm_usersa_id), + [XFRM_MSG_NEWPOLICY - XFRM_MSG_BASE] = XMSGSIZE(xfrm_userpolicy_info), + [XFRM_MSG_DELPOLICY - XFRM_MSG_BASE] = XMSGSIZE(xfrm_userpolicy_id), + [XFRM_MSG_GETPOLICY - XFRM_MSG_BASE] = XMSGSIZE(xfrm_userpolicy_id), + [XFRM_MSG_ALLOCSPI - XFRM_MSG_BASE] = XMSGSIZE(xfrm_userspi_info), + [XFRM_MSG_ACQUIRE - XFRM_MSG_BASE] = XMSGSIZE(xfrm_user_acquire), + [XFRM_MSG_EXPIRE - XFRM_MSG_BASE] = XMSGSIZE(xfrm_user_expire), + [XFRM_MSG_UPDPOLICY - XFRM_MSG_BASE] = XMSGSIZE(xfrm_userpolicy_info), + [XFRM_MSG_UPDSA - XFRM_MSG_BASE] = XMSGSIZE(xfrm_usersa_info), + [XFRM_MSG_POLEXPIRE - XFRM_MSG_BASE] = XMSGSIZE(xfrm_user_polexpire), + [XFRM_MSG_FLUSHSA - XFRM_MSG_BASE] = XMSGSIZE(xfrm_usersa_flush), + [XFRM_MSG_FLUSHPOLICY - XFRM_MSG_BASE] = NLMSG_LENGTH(0), }; +#undef XMSGSIZE + static struct xfrm_link { int (*doit)(struct sk_buff *, struct nlmsghdr *, void **); int (*dump)(struct sk_buff *, struct netlink_callback *); -} xfrm_dispatch[] = { - { .doit = xfrm_add_sa, }, - { .doit = xfrm_del_sa, }, - { - .doit = xfrm_get_sa, - .dump = xfrm_dump_sa, - }, - { .doit = xfrm_add_policy }, - { .doit = xfrm_get_policy }, - { - .doit = xfrm_get_policy, - .dump = xfrm_dump_policy, - }, - { .doit = xfrm_alloc_userspi }, - {}, - {}, - { .doit = xfrm_add_policy }, - { .doit = xfrm_add_sa, }, - {}, - { .doit = xfrm_flush_sa }, - { .doit = xfrm_flush_policy }, +} xfrm_dispatch[XFRM_NR_MSGTYPES] = { + [XFRM_MSG_NEWSA - XFRM_MSG_BASE] = { .doit = xfrm_add_sa }, + [XFRM_MSG_DELSA - XFRM_MSG_BASE] = { .doit = xfrm_del_sa }, + [XFRM_MSG_GETSA - XFRM_MSG_BASE] = { .doit = xfrm_get_sa, + .dump = xfrm_dump_sa }, + [XFRM_MSG_NEWPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_add_policy }, + [XFRM_MSG_DELPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_get_policy }, + [XFRM_MSG_GETPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_get_policy, + .dump = xfrm_dump_policy }, + [XFRM_MSG_ALLOCSPI - XFRM_MSG_BASE] = { .doit = xfrm_alloc_userspi }, + [XFRM_MSG_UPDPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_add_policy }, + [XFRM_MSG_UPDSA - XFRM_MSG_BASE] = { .doit = xfrm_add_sa }, + [XFRM_MSG_FLUSHSA - XFRM_MSG_BASE] = { .doit = xfrm_flush_sa }, + [XFRM_MSG_FLUSHPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_flush_policy }, }; static int xfrm_done(struct netlink_callback *cb) @@ -931,7 +928,9 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *err return -1; } - if ((type == 2 || type == 5) && (nlh->nlmsg_flags & NLM_F_DUMP)) { + if ((type == (XFRM_MSG_GETSA - XFRM_MSG_BASE) || + type == (XFRM_MSG_GETPOLICY - XFRM_MSG_BASE)) && + (nlh->nlmsg_flags & NLM_F_DUMP)) { u32 rlen; if (link->dump == NULL) @@ -1009,18 +1008,26 @@ static int xfrm_user_rcv_skb(struct sk_buff *skb) static void xfrm_netlink_rcv(struct sock *sk, int len) { + unsigned int qlen = skb_queue_len(&sk->sk_receive_queue); + do { struct sk_buff *skb; down(&xfrm_cfg_sem); - while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) { + if (qlen > skb_queue_len(&sk->sk_receive_queue)) + qlen = skb_queue_len(&sk->sk_receive_queue); + + for (; qlen; qlen--) { + skb = skb_dequeue(&sk->sk_receive_queue); if (xfrm_user_rcv_skb(skb)) { if (skb->len) skb_queue_head(&sk->sk_receive_queue, skb); - else + else { kfree_skb(skb); + qlen--; + } break; } kfree_skb(skb); @@ -1028,7 +1035,7 @@ static void xfrm_netlink_rcv(struct sock *sk, int len) up(&xfrm_cfg_sem); - } while (xfrm_nl && xfrm_nl->sk_receive_queue.qlen); + } while (qlen); } static int build_expire(struct sk_buff *skb, struct xfrm_state *x, int hard) diff --git a/scripts/kallsyms.c b/scripts/kallsyms.c index 090ffda4adb..d3d2e534105 100644 --- a/scripts/kallsyms.c +++ b/scripts/kallsyms.c @@ -67,8 +67,9 @@ struct sym_entry { static struct sym_entry *table; static int size, cnt; -static unsigned long long _stext, _etext, _sinittext, _einittext; +static unsigned long long _stext, _etext, _sinittext, _einittext, _sextratext, _eextratext; static int all_symbols = 0; +static char symbol_prefix_char = '\0'; struct token { unsigned char data[MAX_TOK_SIZE]; @@ -93,7 +94,7 @@ unsigned char best_table_len[256]; static void usage(void) { - fprintf(stderr, "Usage: kallsyms [--all-symbols] < in.map > out.S\n"); + fprintf(stderr, "Usage: kallsyms [--all-symbols] [--symbol-prefix=<prefix char>] < in.map > out.S\n"); exit(1); } @@ -112,6 +113,7 @@ static int read_symbol(FILE *in, struct sym_entry *s) { char str[500]; + char *sym; int rc; rc = fscanf(in, "%llx %c %499s\n", &s->addr, &s->type, str); @@ -123,27 +125,36 @@ read_symbol(FILE *in, struct sym_entry *s) return -1; } + sym = str; + /* skip prefix char */ + if (symbol_prefix_char && str[0] == symbol_prefix_char) + sym++; + /* Ignore most absolute/undefined (?) symbols. */ - if (strcmp(str, "_stext") == 0) + if (strcmp(sym, "_stext") == 0) _stext = s->addr; - else if (strcmp(str, "_etext") == 0) + else if (strcmp(sym, "_etext") == 0) _etext = s->addr; - else if (strcmp(str, "_sinittext") == 0) + else if (strcmp(sym, "_sinittext") == 0) _sinittext = s->addr; - else if (strcmp(str, "_einittext") == 0) + else if (strcmp(sym, "_einittext") == 0) _einittext = s->addr; + else if (strcmp(sym, "_sextratext") == 0) + _sextratext = s->addr; + else if (strcmp(sym, "_eextratext") == 0) + _eextratext = s->addr; else if (toupper(s->type) == 'A') { /* Keep these useful absolute symbols */ - if (strcmp(str, "__kernel_syscall_via_break") && - strcmp(str, "__kernel_syscall_via_epc") && - strcmp(str, "__kernel_sigtramp") && - strcmp(str, "__gp")) + if (strcmp(sym, "__kernel_syscall_via_break") && + strcmp(sym, "__kernel_syscall_via_epc") && + strcmp(sym, "__kernel_sigtramp") && + strcmp(sym, "__gp")) return -1; } else if (toupper(s->type) == 'U' || - is_arm_mapping_symbol(str)) + is_arm_mapping_symbol(sym)) return -1; /* include the type field in the symbol name, so that it gets @@ -177,30 +188,37 @@ symbol_valid(struct sym_entry *s) "_SDA2_BASE_", /* ppc */ NULL }; int i; + int offset = 1; + + /* skip prefix char */ + if (symbol_prefix_char && *(s->sym + 1) == symbol_prefix_char) + offset++; /* if --all-symbols is not specified, then symbols outside the text * and inittext sections are discarded */ if (!all_symbols) { if ((s->addr < _stext || s->addr > _etext) - && (s->addr < _sinittext || s->addr > _einittext)) + && (s->addr < _sinittext || s->addr > _einittext) + && (s->addr < _sextratext || s->addr > _eextratext)) return 0; /* Corner case. Discard any symbols with the same value as - * _etext or _einittext, they can move between pass 1 and 2 - * when the kallsyms data is added. If these symbols move then - * they may get dropped in pass 2, which breaks the kallsyms - * rules. + * _etext _einittext or _eextratext; they can move between pass + * 1 and 2 when the kallsyms data are added. If these symbols + * move then they may get dropped in pass 2, which breaks the + * kallsyms rules. */ - if ((s->addr == _etext && strcmp(s->sym + 1, "_etext")) || - (s->addr == _einittext && strcmp(s->sym + 1, "_einittext"))) + if ((s->addr == _etext && strcmp(s->sym + offset, "_etext")) || + (s->addr == _einittext && strcmp(s->sym + offset, "_einittext")) || + (s->addr == _eextratext && strcmp(s->sym + offset, "_eextratext"))) return 0; } /* Exclude symbols which vary between passes. */ - if (strstr(s->sym + 1, "_compiled.")) + if (strstr(s->sym + offset, "_compiled.")) return 0; for (i = 0; special_symbols[i]; i++) - if( strcmp(s->sym + 1, special_symbols[i]) == 0 ) + if( strcmp(s->sym + offset, special_symbols[i]) == 0 ) return 0; return 1; @@ -225,9 +243,15 @@ read_map(FILE *in) static void output_label(char *label) { - printf(".globl %s\n",label); + if (symbol_prefix_char) + printf(".globl %c%s\n", symbol_prefix_char, label); + else + printf(".globl %s\n", label); printf("\tALGN\n"); - printf("%s:\n",label); + if (symbol_prefix_char) + printf("%c%s:\n", symbol_prefix_char, label); + else + printf("%s:\n", label); } /* uncompress a compressed symbol. When this function is called, the best table @@ -665,6 +689,13 @@ static void optimize_token_table(void) insert_real_symbols_in_table(); + /* When valid symbol is not registered, exit to error */ + if (good_head.left == good_head.right && + bad_head.left == bad_head.right) { + fprintf(stderr, "No valid symbol.\n"); + exit(1); + } + optimize_result(); } @@ -672,9 +703,21 @@ static void optimize_token_table(void) int main(int argc, char **argv) { - if (argc == 2 && strcmp(argv[1], "--all-symbols") == 0) - all_symbols = 1; - else if (argc != 1) + if (argc >= 2) { + int i; + for (i = 1; i < argc; i++) { + if(strcmp(argv[i], "--all-symbols") == 0) + all_symbols = 1; + else if (strncmp(argv[i], "--symbol-prefix=", 16) == 0) { + char *p = &argv[i][16]; + /* skip quote */ + if ((*p == '"' && *(p+2) == '"') || (*p == '\'' && *(p+2) == '\'')) + p++; + symbol_prefix_char = *p; + } else + usage(); + } + } else if (argc != 1) usage(); read_map(stdin); @@ -683,4 +726,3 @@ main(int argc, char **argv) return 0; } - diff --git a/scripts/kconfig/Makefile b/scripts/kconfig/Makefile index 5a5ddc40f36..09abb891d11 100644 --- a/scripts/kconfig/Makefile +++ b/scripts/kconfig/Makefile @@ -2,7 +2,7 @@ # Kernel configuration targets # These targets are used from top-level makefile -.PHONY: oldconfig xconfig gconfig menuconfig config silentoldconfig +.PHONY: oldconfig xconfig gconfig menuconfig config silentoldconfig update-po-config xconfig: $(obj)/qconf $< arch/$(ARCH)/Kconfig @@ -23,6 +23,13 @@ oldconfig: $(obj)/conf silentoldconfig: $(obj)/conf $< -s arch/$(ARCH)/Kconfig +update-po-config: $(obj)/kxgettext + xgettext --default-domain=linux \ + --add-comments --keyword=_ --keyword=N_ \ + --files-from=scripts/kconfig/POTFILES.in \ + -o scripts/kconfig/linux.pot + scripts/kconfig/kxgettext arch/$(ARCH)/Kconfig >> scripts/kconfig/linux.pot + .PHONY: randconfig allyesconfig allnoconfig allmodconfig defconfig randconfig: $(obj)/conf @@ -72,9 +79,10 @@ help: # Based on GTK which needs to be installed to compile it # object files used by all kconfig flavours -hostprogs-y := conf mconf qconf gconf +hostprogs-y := conf mconf qconf gconf kxgettext conf-objs := conf.o zconf.tab.o mconf-objs := mconf.o zconf.tab.o +kxgettext-objs := kxgettext.o zconf.tab.o ifeq ($(MAKECMDGOALS),xconfig) qconf-target := 1 @@ -107,7 +115,7 @@ HOSTLOADLIBES_gconf = `pkg-config gtk+-2.0 gmodule-2.0 libglade-2.0 --libs` HOSTCFLAGS_gconf.o = `pkg-config gtk+-2.0 gmodule-2.0 libglade-2.0 --cflags` \ -D LKC_DIRECT_LINK -$(obj)/conf.o $(obj)/mconf.o $(obj)/qconf.o $(obj)/gconf.o: $(obj)/zconf.tab.h +$(obj)/conf.o $(obj)/mconf.o $(obj)/qconf.o $(obj)/gconf.o $(obj)/kxgettext: $(obj)/zconf.tab.h $(obj)/zconf.tab.h: $(src)/zconf.tab.h_shipped $(obj)/zconf.tab.c: $(src)/zconf.tab.c_shipped diff --git a/scripts/kconfig/POTFILES.in b/scripts/kconfig/POTFILES.in new file mode 100644 index 00000000000..cc94e46a79e --- /dev/null +++ b/scripts/kconfig/POTFILES.in @@ -0,0 +1,5 @@ +scripts/kconfig/mconf.c +scripts/kconfig/conf.c +scripts/kconfig/confdata.c +scripts/kconfig/gconf.c +scripts/kconfig/qconf.cc diff --git a/scripts/kconfig/conf.c b/scripts/kconfig/conf.c index a494d1aeb9f..70e7264c694 100644 --- a/scripts/kconfig/conf.c +++ b/scripts/kconfig/conf.c @@ -34,7 +34,7 @@ static int conf_cnt; static signed char line[128]; static struct menu *rootEntry; -static char nohelp_text[] = "Sorry, no help available for this option yet.\n"; +static char nohelp_text[] = N_("Sorry, no help available for this option yet.\n"); static void strip(signed char *str) { @@ -56,9 +56,9 @@ static void strip(signed char *str) static void check_stdin(void) { if (!valid_stdin && input_mode == ask_silent) { - printf("aborted!\n\n"); - printf("Console input/output is redirected. "); - printf("Run 'make oldconfig' to update configuration.\n\n"); + printf(_("aborted!\n\n")); + printf(_("Console input/output is redirected. ")); + printf(_("Run 'make oldconfig' to update configuration.\n\n")); exit(1); } } @@ -470,7 +470,7 @@ static void check_conf(struct menu *menu) if (sym) { if (sym_is_changable(sym) && !sym_has_value(sym)) { if (!conf_cnt++) - printf("*\n* Restart config...\n*\n"); + printf(_("*\n* Restart config...\n*\n")); rootEntry = menu_get_parent_menu(menu); conf(rootEntry); } @@ -504,7 +504,7 @@ int main(int ac, char **av) input_mode = set_default; defconfig_file = av[i++]; if (!defconfig_file) { - printf("%s: No default config file specified\n", + printf(_("%s: No default config file specified\n"), av[0]); exit(1); } @@ -530,7 +530,7 @@ int main(int ac, char **av) } name = av[i]; if (!name) { - printf("%s: Kconfig file missing\n", av[0]); + printf(_("%s: Kconfig file missing\n"), av[0]); } conf_parse(name); //zconfdump(stdout); @@ -547,12 +547,12 @@ int main(int ac, char **av) break; case ask_silent: if (stat(".config", &tmpstat)) { - printf("***\n" + printf(_("***\n" "*** You have not yet configured your kernel!\n" "***\n" "*** Please run some configurator (e.g. \"make oldconfig\" or\n" "*** \"make menuconfig\" or \"make xconfig\").\n" - "***\n"); + "***\n")); exit(1); } case ask_all: @@ -576,7 +576,7 @@ int main(int ac, char **av) check_conf(&rootmenu); } while (conf_cnt); if (conf_write(NULL)) { - fprintf(stderr, "\n*** Error during writing of the kernel configuration.\n\n"); + fprintf(stderr, _("\n*** Error during writing of the kernel configuration.\n\n")); return 1; } return 0; diff --git a/scripts/kconfig/confdata.c b/scripts/kconfig/confdata.c index 1e82ae390a6..2755c459d78 100644 --- a/scripts/kconfig/confdata.c +++ b/scripts/kconfig/confdata.c @@ -88,9 +88,9 @@ int conf_read(const char *name) name = conf_expand_value(name); in = zconf_fopen(name); if (in) { - printf("#\n" - "# using defaults found in %s\n" - "#\n", name); + printf(_("#\n" + "# using defaults found in %s\n" + "#\n"), name); break; } } @@ -312,11 +312,11 @@ int conf_write(const char *name) if (env && *env) use_timestamp = 0; - fprintf(out, "#\n" - "# Automatically generated make config: don't edit\n" - "# Linux kernel version: %s\n" - "%s%s" - "#\n", + fprintf(out, _("#\n" + "# Automatically generated make config: don't edit\n" + "# Linux kernel version: %s\n" + "%s%s" + "#\n"), sym_get_string_value(sym), use_timestamp ? "# " : "", use_timestamp ? ctime(&now) : ""); diff --git a/scripts/kconfig/gconf.c b/scripts/kconfig/gconf.c index 6fdbe6e3ce0..ad6b1204387 100644 --- a/scripts/kconfig/gconf.c +++ b/scripts/kconfig/gconf.c @@ -41,7 +41,7 @@ static gboolean resizeable = FALSE; static gboolean config_changed = FALSE; static char nohelp_text[] = - "Sorry, no help available for this option yet.\n"; + N_("Sorry, no help available for this option yet.\n"); GtkWidget *main_wnd = NULL; GtkWidget *tree1_w = NULL; // left frame @@ -193,7 +193,7 @@ void init_main_window(const gchar * glade_file) xml = glade_xml_new(glade_file, "window1", NULL); if (!xml) - g_error("GUI loading failed !\n"); + g_error(_("GUI loading failed !\n")); glade_xml_signal_autoconnect(xml); main_wnd = glade_xml_get_widget(xml, "window1"); @@ -275,7 +275,7 @@ void init_main_window(const gchar * glade_file) /*"style", PANGO_STYLE_OBLIQUE, */ NULL); - sprintf(title, "Linux Kernel v%s Configuration", + sprintf(title, _("Linux Kernel v%s Configuration"), getenv("KERNELRELEASE")); gtk_window_set_title(GTK_WINDOW(main_wnd), title); @@ -325,7 +325,7 @@ void init_left_tree(void) column = gtk_tree_view_column_new(); gtk_tree_view_append_column(view, column); - gtk_tree_view_column_set_title(column, "Options"); + gtk_tree_view_column_set_title(column, _("Options")); renderer = gtk_cell_renderer_toggle_new(); gtk_tree_view_column_pack_start(GTK_TREE_VIEW_COLUMN(column), @@ -370,7 +370,7 @@ void init_right_tree(void) column = gtk_tree_view_column_new(); gtk_tree_view_append_column(view, column); - gtk_tree_view_column_set_title(column, "Options"); + gtk_tree_view_column_set_title(column, _("Options")); renderer = gtk_cell_renderer_pixbuf_new(); gtk_tree_view_column_pack_start(GTK_TREE_VIEW_COLUMN(column), @@ -401,7 +401,7 @@ void init_right_tree(void) renderer = gtk_cell_renderer_text_new(); gtk_tree_view_insert_column_with_attributes(view, -1, - "Name", renderer, + _("Name"), renderer, "text", COL_NAME, "foreground-gdk", COL_COLOR, NULL); @@ -425,7 +425,7 @@ void init_right_tree(void) COL_COLOR, NULL); renderer = gtk_cell_renderer_text_new(); gtk_tree_view_insert_column_with_attributes(view, -1, - "Value", renderer, + _("Value"), renderer, "text", COL_VALUE, "editable", COL_EDIT, @@ -466,15 +466,15 @@ static void text_insert_help(struct menu *menu) GtkTextIter start, end; const char *prompt = menu_get_prompt(menu); gchar *name; - const char *help = nohelp_text; + const char *help = _(nohelp_text); if (!menu->sym) help = ""; else if (menu->sym->help) - help = menu->sym->help; + help = _(menu->sym->help); if (menu->sym && menu->sym->name) - name = g_strdup_printf(menu->sym->name); + name = g_strdup_printf(_(menu->sym->name)); else name = g_strdup(""); @@ -530,7 +530,7 @@ gboolean on_window1_delete_event(GtkWidget * widget, GdkEvent * event, if (config_changed == FALSE) return FALSE; - dialog = gtk_dialog_new_with_buttons("Warning !", + dialog = gtk_dialog_new_with_buttons(_("Warning !"), GTK_WINDOW(main_wnd), (GtkDialogFlags) (GTK_DIALOG_MODAL | @@ -544,7 +544,7 @@ gboolean on_window1_delete_event(GtkWidget * widget, GdkEvent * event, gtk_dialog_set_default_response(GTK_DIALOG(dialog), GTK_RESPONSE_CANCEL); - label = gtk_label_new("\nSave configuration ?\n"); + label = gtk_label_new(_("\nSave configuration ?\n")); gtk_container_add(GTK_CONTAINER(GTK_DIALOG(dialog)->vbox), label); gtk_widget_show(label); @@ -604,7 +604,7 @@ load_filename(GtkFileSelection * file_selector, gpointer user_data) (user_data)); if (conf_read(fn)) - text_insert_msg("Error", "Unable to load configuration !"); + text_insert_msg(_("Error"), _("Unable to load configuration !")); else display_tree(&rootmenu); } @@ -613,7 +613,7 @@ void on_load1_activate(GtkMenuItem * menuitem, gpointer user_data) { GtkWidget *fs; - fs = gtk_file_selection_new("Load file..."); + fs = gtk_file_selection_new(_("Load file...")); g_signal_connect(GTK_OBJECT(GTK_FILE_SELECTION(fs)->ok_button), "clicked", G_CALLBACK(load_filename), (gpointer) fs); @@ -632,7 +632,7 @@ void on_load1_activate(GtkMenuItem * menuitem, gpointer user_data) void on_save1_activate(GtkMenuItem * menuitem, gpointer user_data) { if (conf_write(NULL)) - text_insert_msg("Error", "Unable to save configuration !"); + text_insert_msg(_("Error"), _("Unable to save configuration !")); config_changed = FALSE; } @@ -647,7 +647,7 @@ store_filename(GtkFileSelection * file_selector, gpointer user_data) (user_data)); if (conf_write(fn)) - text_insert_msg("Error", "Unable to save configuration !"); + text_insert_msg(_("Error"), _("Unable to save configuration !")); gtk_widget_destroy(GTK_WIDGET(user_data)); } @@ -656,7 +656,7 @@ void on_save_as1_activate(GtkMenuItem * menuitem, gpointer user_data) { GtkWidget *fs; - fs = gtk_file_selection_new("Save file as..."); + fs = gtk_file_selection_new(_("Save file as...")); g_signal_connect(GTK_OBJECT(GTK_FILE_SELECTION(fs)->ok_button), "clicked", G_CALLBACK(store_filename), (gpointer) fs); @@ -740,7 +740,7 @@ on_show_debug_info1_activate(GtkMenuItem * menuitem, gpointer user_data) void on_introduction1_activate(GtkMenuItem * menuitem, gpointer user_data) { GtkWidget *dialog; - const gchar *intro_text = + const gchar *intro_text = _( "Welcome to gkc, the GTK+ graphical kernel configuration tool\n" "for Linux.\n" "For each option, a blank box indicates the feature is disabled, a\n" @@ -756,7 +756,7 @@ void on_introduction1_activate(GtkMenuItem * menuitem, gpointer user_data) "option.\n" "\n" "Toggling Show Debug Info under the Options menu will show \n" - "the dependencies, which you can then match by examining other options."; + "the dependencies, which you can then match by examining other options."); dialog = gtk_message_dialog_new(GTK_WINDOW(main_wnd), GTK_DIALOG_DESTROY_WITH_PARENT, @@ -773,8 +773,8 @@ void on_about1_activate(GtkMenuItem * menuitem, gpointer user_data) { GtkWidget *dialog; const gchar *about_text = - "gkc is copyright (c) 2002 Romain Lievin <roms@lpg.ticalc.org>.\n" - "Based on the source code from Roman Zippel.\n"; + _("gkc is copyright (c) 2002 Romain Lievin <roms@lpg.ticalc.org>.\n" + "Based on the source code from Roman Zippel.\n"); dialog = gtk_message_dialog_new(GTK_WINDOW(main_wnd), GTK_DIALOG_DESTROY_WITH_PARENT, @@ -791,9 +791,9 @@ void on_license1_activate(GtkMenuItem * menuitem, gpointer user_data) { GtkWidget *dialog; const gchar *license_text = - "gkc is released under the terms of the GNU GPL v2.\n" - "For more information, please see the source code or\n" - "visit http://www.fsf.org/licenses/licenses.html\n"; + _("gkc is released under the terms of the GNU GPL v2.\n" + "For more information, please see the source code or\n" + "visit http://www.fsf.org/licenses/licenses.html\n"); dialog = gtk_message_dialog_new(GTK_WINDOW(main_wnd), GTK_DIALOG_DESTROY_WITH_PARENT, @@ -1579,6 +1579,10 @@ int main(int ac, char *av[]) kconfig_load(); #endif + bindtextdomain(PACKAGE, LOCALEDIR); + bind_textdomain_codeset(PACKAGE, "UTF-8"); + textdomain(PACKAGE); + /* GTK stuffs */ gtk_set_locale(); gtk_init(&ac, &av); diff --git a/scripts/kconfig/kxgettext.c b/scripts/kconfig/kxgettext.c new file mode 100644 index 00000000000..1c88d7c6d5a --- /dev/null +++ b/scripts/kconfig/kxgettext.c @@ -0,0 +1,221 @@ +/* + * Arnaldo Carvalho de Melo <acme@conectiva.com.br>, 2005 + * + * Released under the terms of the GNU GPL v2.0 + */ + +#include <stdlib.h> +#include <string.h> + +#define LKC_DIRECT_LINK +#include "lkc.h" + +static char *escape(const char* text, char *bf, int len) +{ + char *bfp = bf; + int multiline = strchr(text, '\n') != NULL; + + *bfp++ = '"'; + --len; + + if (multiline) { + *bfp++ = '"'; + *bfp++ = '\n'; + *bfp++ = '"'; + len -= 3; + } + + while (*text != '\0' && len > 1) { + if (*text == '"') + *bfp++ = '\\'; + else if (*text == '\n') { + *bfp++ = '\\'; + *bfp++ = 'n'; + *bfp++ = '"'; + *bfp++ = '\n'; + *bfp++ = '"'; + len -= 5; + ++text; + goto next; + } + *bfp++ = *text++; +next: + --len; + } + + if (multiline) + bfp -= 3; + + *bfp++ = '"'; + *bfp = '\0'; + + return bf; +} + +struct file_line { + struct file_line *next; + char* file; + int lineno; +}; + +static struct file_line *file_line__new(char *file, int lineno) +{ + struct file_line *self = malloc(sizeof(*self)); + + if (self == NULL) + goto out; + + self->file = file; + self->lineno = lineno; + self->next = NULL; +out: + return self; +} + +struct message { + const char *msg; + const char *option; + struct message *next; + struct file_line *files; +}; + +static struct message *message__list; + +static struct message *message__new(const char *msg, char *option, char *file, int lineno) +{ + struct message *self = malloc(sizeof(*self)); + + if (self == NULL) + goto out; + + self->files = file_line__new(file, lineno); + if (self->files == NULL) + goto out_fail; + + self->msg = strdup(msg); + if (self->msg == NULL) + goto out_fail_msg; + + self->option = option; + self->next = NULL; +out: + return self; +out_fail_msg: + free(self->files); +out_fail: + free(self); + self = NULL; + goto out; +} + +static struct message *mesage__find(const char *msg) +{ + struct message *m = message__list; + + while (m != NULL) { + if (strcmp(m->msg, msg) == 0) + break; + m = m->next; + } + + return m; +} + +static int message__add_file_line(struct message *self, char *file, int lineno) +{ + int rc = -1; + struct file_line *fl = file_line__new(file, lineno); + + if (fl == NULL) + goto out; + + fl->next = self->files; + self->files = fl; + rc = 0; +out: + return rc; +} + +static int message__add(const char *msg, char *option, char *file, int lineno) +{ + int rc = 0; + char bf[16384]; + char *escaped = escape(msg, bf, sizeof(bf)); + struct message *m = mesage__find(escaped); + + if (m != NULL) + rc = message__add_file_line(m, file, lineno); + else { + m = message__new(escaped, option, file, lineno); + + if (m != NULL) { + m->next = message__list; + message__list = m; + } else + rc = -1; + } + return rc; +} + +void menu_build_message_list(struct menu *menu) +{ + struct menu *child; + + message__add(menu_get_prompt(menu), NULL, + menu->file == NULL ? "Root Menu" : menu->file->name, + menu->lineno); + + if (menu->sym != NULL && menu->sym->help != NULL) + message__add(menu->sym->help, menu->sym->name, + menu->file == NULL ? "Root Menu" : menu->file->name, + menu->lineno); + + for (child = menu->list; child != NULL; child = child->next) + if (child->prompt != NULL) + menu_build_message_list(child); +} + +static void message__print_file_lineno(struct message *self) +{ + struct file_line *fl = self->files; + + printf("\n#: %s:%d", fl->file, fl->lineno); + fl = fl->next; + + while (fl != NULL) { + printf(", %s:%d", fl->file, fl->lineno); + fl = fl->next; + } + + if (self->option != NULL) + printf(", %s:00000", self->option); + + putchar('\n'); +} + +static void message__print_gettext_msgid_msgstr(struct message *self) +{ + message__print_file_lineno(self); + + printf("msgid %s\n" + "msgstr \"\"\n", self->msg); +} + +void menu__xgettext(void) +{ + struct message *m = message__list; + + while (m != NULL) { + message__print_gettext_msgid_msgstr(m); + m = m->next; + } +} + +int main(int ac, char **av) +{ + conf_parse(av[1]); + + menu_build_message_list(menu_get_root_menu(NULL)); + menu__xgettext(); + return 0; +} diff --git a/scripts/kconfig/lkc.h b/scripts/kconfig/lkc.h index b8a67fc9d64..8b84c42b49b 100644 --- a/scripts/kconfig/lkc.h +++ b/scripts/kconfig/lkc.h @@ -8,6 +8,8 @@ #include "expr.h" +#include <libintl.h> + #ifdef __cplusplus extern "C" { #endif @@ -23,6 +25,12 @@ extern "C" { #define SRCTREE "srctree" +#define PACKAGE "linux" +#define LOCALEDIR "/usr/share/locale" + +#define _(text) gettext(text) +#define N_(text) (text) + int zconfparse(void); void zconfdump(FILE *out); diff --git a/scripts/kconfig/mconf.c b/scripts/kconfig/mconf.c index 730d316fe7f..e5db10ca956 100644 --- a/scripts/kconfig/mconf.c +++ b/scripts/kconfig/mconf.c @@ -4,6 +4,8 @@ * * Introduced single menu mode (show all sub-menus in one large tree). * 2002-11-06 Petr Baudis <pasky@ucw.cz> + * + * i18n, 2005, Arnaldo Carvalho de Melo <acme@conectiva.com.br> */ #include <sys/ioctl.h> @@ -23,7 +25,7 @@ #include "lkc.h" static char menu_backtitle[128]; -static const char mconf_readme[] = +static const char mconf_readme[] = N_( "Overview\n" "--------\n" "Some kernel features may be built directly into the kernel.\n" @@ -156,39 +158,39 @@ static const char mconf_readme[] = "\n" "Note that this mode can eventually be a little more CPU expensive\n" "(especially with a larger number of unrolled categories) than the\n" -"default mode.\n", -menu_instructions[] = +"default mode.\n"), +menu_instructions[] = N_( "Arrow keys navigate the menu. " "<Enter> selects submenus --->. " "Highlighted letters are hotkeys. " "Pressing <Y> includes, <N> excludes, <M> modularizes features. " "Press <Esc><Esc> to exit, <?> for Help, </> for Search. " - "Legend: [*] built-in [ ] excluded <M> module < > module capable", -radiolist_instructions[] = + "Legend: [*] built-in [ ] excluded <M> module < > module capable"), +radiolist_instructions[] = N_( "Use the arrow keys to navigate this window or " "press the hotkey of the item you wish to select " "followed by the <SPACE BAR>. " - "Press <?> for additional information about this option.", -inputbox_instructions_int[] = + "Press <?> for additional information about this option."), +inputbox_instructions_int[] = N_( "Please enter a decimal value. " "Fractions will not be accepted. " - "Use the <TAB> key to move from the input field to the buttons below it.", -inputbox_instructions_hex[] = + "Use the <TAB> key to move from the input field to the buttons below it."), +inputbox_instructions_hex[] = N_( "Please enter a hexadecimal value. " - "Use the <TAB> key to move from the input field to the buttons below it.", -inputbox_instructions_string[] = + "Use the <TAB> key to move from the input field to the buttons below it."), +inputbox_instructions_string[] = N_( "Please enter a string value. " - "Use the <TAB> key to move from the input field to the buttons below it.", -setmod_text[] = + "Use the <TAB> key to move from the input field to the buttons below it."), +setmod_text[] = N_( "This feature depends on another which has been configured as a module.\n" - "As a result, this feature will be built as a module.", -nohelp_text[] = - "There is no help available for this kernel option.\n", -load_config_text[] = + "As a result, this feature will be built as a module."), +nohelp_text[] = N_( + "There is no help available for this kernel option.\n"), +load_config_text[] = N_( "Enter the name of the configuration file you wish to load. " "Accept the name shown to restore the configuration you " - "last retrieved. Leave blank to abort.", -load_config_help[] = + "last retrieved. Leave blank to abort."), +load_config_help[] = N_( "\n" "For various reasons, one may wish to keep several different kernel\n" "configurations available on a single machine.\n" @@ -198,11 +200,11 @@ load_config_help[] = "to modify that configuration.\n" "\n" "If you are uncertain, then you have probably never used alternate\n" - "configuration files. You should therefor leave this blank to abort.\n", -save_config_text[] = + "configuration files. You should therefor leave this blank to abort.\n"), +save_config_text[] = N_( "Enter a filename to which this configuration should be saved " - "as an alternate. Leave blank to abort.", -save_config_help[] = + "as an alternate. Leave blank to abort."), +save_config_help[] = N_( "\n" "For various reasons, one may wish to keep different kernel\n" "configurations available on a single machine.\n" @@ -212,8 +214,8 @@ save_config_help[] = "configuration options you have selected at that time.\n" "\n" "If you are uncertain what all this means then you should probably\n" - "leave this blank.\n", -search_help[] = + "leave this blank.\n"), +search_help[] = N_( "\n" "Search for CONFIG_ symbols and display their relations.\n" "Example: search for \"^FOO\"\n" @@ -250,7 +252,7 @@ search_help[] = "Examples: USB => find all CONFIG_ symbols containing USB\n" " ^USB => find all CONFIG_ symbols starting with USB\n" " USB$ => find all CONFIG_ symbols ending with USB\n" - "\n"; + "\n"); static signed char buf[4096], *bufptr = buf; static signed char input_buf[4096]; @@ -305,8 +307,8 @@ static void init_wsize(void) } if (rows < 19 || cols < 80) { - fprintf(stderr, "Your display is too small to run Menuconfig!\n"); - fprintf(stderr, "It must be at least 19 lines by 80 columns.\n"); + fprintf(stderr, N_("Your display is too small to run Menuconfig!\n")); + fprintf(stderr, N_("It must be at least 19 lines by 80 columns.\n")); exit(1); } @@ -526,9 +528,9 @@ static void search_conf(void) again: cprint_init(); cprint("--title"); - cprint("Search Configuration Parameter"); + cprint(_("Search Configuration Parameter")); cprint("--inputbox"); - cprint("Enter Keyword"); + cprint(_("Enter Keyword")); cprint("10"); cprint("75"); cprint(""); @@ -539,7 +541,7 @@ again: case 0: break; case 1: - show_helptext("Search Configuration", search_help); + show_helptext(_("Search Configuration"), search_help); goto again; default: return; @@ -548,7 +550,7 @@ again: sym_arr = sym_re_search(input_buf); res = get_relations_str(sym_arr); free(sym_arr); - show_textbox("Search Results", str_get(&res), 0, 0); + show_textbox(_("Search Results"), str_get(&res), 0, 0); str_free(&res); } @@ -721,9 +723,9 @@ static void conf(struct menu *menu) while (1) { cprint_init(); cprint("--title"); - cprint("%s", prompt ? prompt : "Main Menu"); + cprint("%s", prompt ? prompt : _("Main Menu")); cprint("--menu"); - cprint(menu_instructions); + cprint(_(menu_instructions)); cprint("%d", rows); cprint("%d", cols); cprint("%d", rows - 10); @@ -736,9 +738,9 @@ static void conf(struct menu *menu) cprint(":"); cprint("--- "); cprint("L"); - cprint(" Load an Alternate Configuration File"); + cprint(_(" Load an Alternate Configuration File")); cprint("S"); - cprint(" Save Configuration to an Alternate File"); + cprint(_(" Save Configuration to an Alternate File")); } stat = exec_conf(); if (stat < 0) @@ -793,7 +795,7 @@ static void conf(struct menu *menu) if (sym) show_help(submenu); else - show_helptext("README", mconf_readme); + show_helptext("README", _(mconf_readme)); break; case 3: if (type == 't') { @@ -849,7 +851,7 @@ static void show_help(struct menu *menu) { if (sym->name) { str_printf(&help, "CONFIG_%s:\n\n", sym->name); - str_append(&help, sym->help); + str_append(&help, _(sym->help)); str_append(&help, "\n"); } } else { @@ -886,9 +888,9 @@ static void conf_choice(struct menu *menu) while (1) { cprint_init(); cprint("--title"); - cprint("%s", prompt ? prompt : "Main Menu"); + cprint("%s", prompt ? prompt : _("Main Menu")); cprint("--radiolist"); - cprint(radiolist_instructions); + cprint(_(radiolist_instructions)); cprint("15"); cprint("70"); cprint("6"); @@ -935,17 +937,17 @@ static void conf_string(struct menu *menu) while (1) { cprint_init(); cprint("--title"); - cprint("%s", prompt ? prompt : "Main Menu"); + cprint("%s", prompt ? prompt : _("Main Menu")); cprint("--inputbox"); switch (sym_get_type(menu->sym)) { case S_INT: - cprint(inputbox_instructions_int); + cprint(_(inputbox_instructions_int)); break; case S_HEX: - cprint(inputbox_instructions_hex); + cprint(_(inputbox_instructions_hex)); break; case S_STRING: - cprint(inputbox_instructions_string); + cprint(_(inputbox_instructions_string)); break; default: /* panic? */; @@ -958,7 +960,7 @@ static void conf_string(struct menu *menu) case 0: if (sym_set_string_value(menu->sym, input_buf)) return; - show_textbox(NULL, "You have made an invalid entry.", 5, 43); + show_textbox(NULL, _("You have made an invalid entry."), 5, 43); break; case 1: show_help(menu); @@ -987,10 +989,10 @@ static void conf_load(void) return; if (!conf_read(input_buf)) return; - show_textbox(NULL, "File does not exist!", 5, 38); + show_textbox(NULL, _("File does not exist!"), 5, 38); break; case 1: - show_helptext("Load Alternate Configuration", load_config_help); + show_helptext(_("Load Alternate Configuration"), load_config_help); break; case 255: return; @@ -1016,10 +1018,10 @@ static void conf_save(void) return; if (!conf_write(input_buf)) return; - show_textbox(NULL, "Can't create file! Probably a nonexistent directory.", 5, 60); + show_textbox(NULL, _("Can't create file! Probably a nonexistent directory."), 5, 60); break; case 1: - show_helptext("Save Alternate Configuration", save_config_help); + show_helptext(_("Save Alternate Configuration"), save_config_help); break; case 255: return; @@ -1040,12 +1042,16 @@ int main(int ac, char **av) char *mode; int stat; + setlocale(LC_ALL, ""); + bindtextdomain(PACKAGE, LOCALEDIR); + textdomain(PACKAGE); + conf_parse(av[1]); conf_read(NULL); sym = sym_lookup("KERNELRELEASE", 0); sym_calc_value(sym); - sprintf(menu_backtitle, "Linux Kernel v%s Configuration", + sprintf(menu_backtitle, _("Linux Kernel v%s Configuration"), sym_get_string_value(sym)); mode = getenv("MENUCONFIG_MODE"); @@ -1062,7 +1068,7 @@ int main(int ac, char **av) do { cprint_init(); cprint("--yesno"); - cprint("Do you wish to save your new kernel configuration?"); + cprint(_("Do you wish to save your new kernel configuration?")); cprint("5"); cprint("60"); stat = exec_conf(); @@ -1070,20 +1076,20 @@ int main(int ac, char **av) if (stat == 0) { if (conf_write(NULL)) { - fprintf(stderr, "\n\n" + fprintf(stderr, _("\n\n" "Error during writing of the kernel configuration.\n" "Your kernel configuration changes were NOT saved." - "\n\n"); + "\n\n")); return 1; } - printf("\n\n" + printf(_("\n\n" "*** End of Linux kernel configuration.\n" "*** Execute 'make' to build the kernel or try 'make help'." - "\n\n"); + "\n\n")); } else { - fprintf(stderr, "\n\n" + fprintf(stderr, _("\n\n" "Your kernel configuration changes were NOT saved." - "\n\n"); + "\n\n")); } return 0; diff --git a/scripts/kconfig/menu.c b/scripts/kconfig/menu.c index 0c13156f334..8c59b212722 100644 --- a/scripts/kconfig/menu.c +++ b/scripts/kconfig/menu.c @@ -365,9 +365,9 @@ bool menu_is_visible(struct menu *menu) const char *menu_get_prompt(struct menu *menu) { if (menu->prompt) - return menu->prompt->text; + return _(menu->prompt->text); else if (menu->sym) - return menu->sym->name; + return _(menu->sym->name); return NULL; } diff --git a/scripts/kconfig/qconf.cc b/scripts/kconfig/qconf.cc index 0cdbf9dbbd5..4590cd31623 100644 --- a/scripts/kconfig/qconf.cc +++ b/scripts/kconfig/qconf.cc @@ -26,8 +26,23 @@ #include "qconf.moc" #include "images.c" +#ifdef _ +# undef _ +# define _ qgettext +#endif + static QApplication *configApp; +static inline QString qgettext(const char* str) +{ + return QString::fromLocal8Bit(gettext(str)); +} + +static inline QString qgettext(const QString& str) +{ + return QString::fromLocal8Bit(gettext(str.latin1())); +} + ConfigSettings::ConfigSettings() : showAll(false), showName(false), showRange(false), showData(false) { @@ -177,7 +192,7 @@ void ConfigItem::updateMenu(void) sym = menu->sym; prop = menu->prompt; - prompt = menu_get_prompt(menu); + prompt = QString::fromLocal8Bit(menu_get_prompt(menu)); if (prop) switch (prop->type) { case P_MENU: @@ -203,7 +218,7 @@ void ConfigItem::updateMenu(void) if (!sym) goto set_prompt; - setText(nameColIdx, sym->name); + setText(nameColIdx, QString::fromLocal8Bit(sym->name)); type = sym_get_type(sym); switch (type) { @@ -213,9 +228,9 @@ void ConfigItem::updateMenu(void) if (!sym_is_changable(sym) && !list->showAll) { setPixmap(promptColIdx, 0); - setText(noColIdx, 0); - setText(modColIdx, 0); - setText(yesColIdx, 0); + setText(noColIdx, QString::null); + setText(modColIdx, QString::null); + setText(yesColIdx, QString::null); break; } expr = sym_get_tristate_value(sym); @@ -257,6 +272,7 @@ void ConfigItem::updateMenu(void) const char* data; data = sym_get_string_value(sym); + #if QT_VERSION >= 300 int i = list->mapIdx(dataColIdx); if (i >= 0) @@ -264,9 +280,9 @@ void ConfigItem::updateMenu(void) #endif setText(dataColIdx, data); if (type == S_STRING) - prompt.sprintf("%s: %s", prompt.latin1(), data); + prompt = QString("%1: %2").arg(prompt).arg(data); else - prompt.sprintf("(%s) %s", data, prompt.latin1()); + prompt = QString("(%2) %1").arg(prompt).arg(data); break; } if (!sym_has_value(sym) && visible) @@ -343,9 +359,9 @@ void ConfigLineEdit::show(ConfigItem* i) { item = i; if (sym_get_string_value(item->menu->sym)) - setText(sym_get_string_value(item->menu->sym)); + setText(QString::fromLocal8Bit(sym_get_string_value(item->menu->sym))); else - setText(0); + setText(QString::null); Parent::show(); setFocus(); } @@ -961,7 +977,7 @@ ConfigMainWindow::ConfigMainWindow(void) delete configSettings; } -static QString print_filter(const char *str) +static QString print_filter(const QString &str) { QRegExp re("[<>&\"\\n]"); QString res = str; @@ -994,7 +1010,7 @@ static QString print_filter(const char *str) static void expr_print_help(void *data, const char *str) { - ((QString*)data)->append(print_filter(str)); + reinterpret_cast<QString*>(data)->append(print_filter(str)); } /* @@ -1009,7 +1025,7 @@ void ConfigMainWindow::setHelp(QListViewItem* item) if (item) menu = ((ConfigItem*)item)->menu; if (!menu) { - helpText->setText(NULL); + helpText->setText(QString::null); return; } @@ -1019,16 +1035,16 @@ void ConfigMainWindow::setHelp(QListViewItem* item) if (sym) { if (menu->prompt) { head += "<big><b>"; - head += print_filter(menu->prompt->text); + head += print_filter(_(menu->prompt->text)); head += "</b></big>"; if (sym->name) { head += " ("; - head += print_filter(sym->name); + head += print_filter(_(sym->name)); head += ")"; } } else if (sym->name) { head += "<big><b>"; - head += print_filter(sym->name); + head += print_filter(_(sym->name)); head += "</b></big>"; } head += "<br><br>"; @@ -1049,7 +1065,7 @@ void ConfigMainWindow::setHelp(QListViewItem* item) case P_PROMPT: case P_MENU: debug += "prompt: "; - debug += print_filter(prop->text); + debug += print_filter(_(prop->text)); debug += "<br>"; break; case P_DEFAULT: @@ -1088,10 +1104,10 @@ void ConfigMainWindow::setHelp(QListViewItem* item) debug += "<br>"; } - help = print_filter(sym->help); + help = print_filter(_(sym->help)); } else if (menu->prompt) { head += "<big><b>"; - head += print_filter(menu->prompt->text); + head += print_filter(_(menu->prompt->text)); head += "</b></big><br><br>"; if (showDebug) { if (menu->prompt->visible.expr) { @@ -1111,7 +1127,7 @@ void ConfigMainWindow::loadConfig(void) QString s = QFileDialog::getOpenFileName(".config", NULL, this); if (s.isNull()) return; - if (conf_read(s.latin1())) + if (conf_read(QFile::encodeName(s))) QMessageBox::information(this, "qconf", "Unable to load configuration!"); ConfigView::updateListAll(); } @@ -1127,7 +1143,7 @@ void ConfigMainWindow::saveConfigAs(void) QString s = QFileDialog::getSaveFileName(".config", NULL, this); if (s.isNull()) return; - if (conf_write(s.latin1())) + if (conf_write(QFile::encodeName(s))) QMessageBox::information(this, "qconf", "Unable to save configuration!"); } @@ -1372,6 +1388,9 @@ int main(int ac, char** av) ConfigMainWindow* v; const char *name; + bindtextdomain(PACKAGE, LOCALEDIR); + textdomain(PACKAGE); + #ifndef LKC_DIRECT_LINK kconfig_load(); #endif diff --git a/scripts/kernel-doc b/scripts/kernel-doc index 8b1dab63f11..0835dc2a8aa 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -553,15 +553,20 @@ sub output_section_xml(%) { # print out each section $lineprefix=" "; foreach $section (@{$args{'sectionlist'}}) { - print "<refsect1>\n <title>$section</title>\n <para>\n"; + print "<refsect1>\n"; + print "<title>$section</title>\n"; if ($section =~ m/EXAMPLE/i) { - print "<example><para>\n"; + print "<informalexample><programlisting>\n"; + } else { + print "<para>\n"; } output_highlight($args{'sections'}{$section}); if ($section =~ m/EXAMPLE/i) { - print "</para></example>\n"; + print "</programlisting></informalexample>\n"; + } else { + print "</para>\n"; } - print " </para>\n</refsect1>\n"; + print "</refsect1>\n"; } } @@ -576,8 +581,14 @@ sub output_function_xml(%) { $id =~ s/[^A-Za-z0-9]/-/g; print "<refentry>\n"; + print "<refentryinfo>\n"; + print " <title>LINUX</title>\n"; + print " <productname>Kernel Hackers Manual</productname>\n"; + print " <date>$man_date</date>\n"; + print "</refentryinfo>\n"; print "<refmeta>\n"; - print "<refentrytitle><phrase id=\"$id\">".$args{'function'}."</phrase></refentrytitle>\n"; + print " <refentrytitle><phrase id=\"$id\">".$args{'function'}."</phrase></refentrytitle>\n"; + print " <manvolnum>9</manvolnum>\n"; print "</refmeta>\n"; print "<refnamediv>\n"; print " <refname>".$args{'function'}."</refname>\n"; @@ -607,7 +618,7 @@ sub output_function_xml(%) { } } } else { - print " <void>\n"; + print " <void/>\n"; } print " </funcprototype></funcsynopsis>\n"; print "</refsynopsisdiv>\n"; @@ -646,8 +657,14 @@ sub output_struct_xml(%) { $id =~ s/[^A-Za-z0-9]/-/g; print "<refentry>\n"; + print "<refentryinfo>\n"; + print " <title>LINUX</title>\n"; + print " <productname>Kernel Hackers Manual</productname>\n"; + print " <date>$man_date</date>\n"; + print "</refentryinfo>\n"; print "<refmeta>\n"; - print "<refentrytitle><phrase id=\"$id\">".$args{'type'}." ".$args{'struct'}."</phrase></refentrytitle>\n"; + print " <refentrytitle><phrase id=\"$id\">".$args{'type'}." ".$args{'struct'}."</phrase></refentrytitle>\n"; + print " <manvolnum>9</manvolnum>\n"; print "</refmeta>\n"; print "<refnamediv>\n"; print " <refname>".$args{'type'}." ".$args{'struct'}."</refname>\n"; @@ -724,8 +741,14 @@ sub output_enum_xml(%) { $id =~ s/[^A-Za-z0-9]/-/g; print "<refentry>\n"; + print "<refentryinfo>\n"; + print " <title>LINUX</title>\n"; + print " <productname>Kernel Hackers Manual</productname>\n"; + print " <date>$man_date</date>\n"; + print "</refentryinfo>\n"; print "<refmeta>\n"; - print "<refentrytitle><phrase id=\"$id\">enum ".$args{'enum'}."</phrase></refentrytitle>\n"; + print " <refentrytitle><phrase id=\"$id\">enum ".$args{'enum'}."</phrase></refentrytitle>\n"; + print " <manvolnum>9</manvolnum>\n"; print "</refmeta>\n"; print "<refnamediv>\n"; print " <refname>enum ".$args{'enum'}."</refname>\n"; @@ -784,8 +807,14 @@ sub output_typedef_xml(%) { $id =~ s/[^A-Za-z0-9]/-/g; print "<refentry>\n"; + print "<refentryinfo>\n"; + print " <title>LINUX</title>\n"; + print " <productname>Kernel Hackers Manual</productname>\n"; + print " <date>$man_date</date>\n"; + print "</refentryinfo>\n"; print "<refmeta>\n"; - print "<refentrytitle><phrase id=\"$id\">typedef ".$args{'typedef'}."</phrase></refentrytitle>\n"; + print " <refentrytitle><phrase id=\"$id\">typedef ".$args{'typedef'}."</phrase></refentrytitle>\n"; + print " <manvolnum>9</manvolnum>\n"; print "</refmeta>\n"; print "<refnamediv>\n"; print " <refname>typedef ".$args{'typedef'}."</refname>\n"; @@ -1465,6 +1494,8 @@ sub dump_function($$) { $prototype =~ s/^static +//; $prototype =~ s/^extern +//; + $prototype =~ s/^fastcall +//; + $prototype =~ s/^asmlinkage +//; $prototype =~ s/^inline +//; $prototype =~ s/^__inline__ +//; $prototype =~ s/^#define +//; #ak added diff --git a/scripts/makeman b/scripts/makeman deleted file mode 100755 index db3af647ee1..00000000000 --- a/scripts/makeman +++ /dev/null @@ -1,185 +0,0 @@ -#!/usr/bin/perl - -use strict; - -## Copyright (C) Michael Still (mikal@stillhq.com) -## Released under the terms of the GNU GPL -## -## A script to make or install the manpages extracted by split-man -## -## Arguements: $1 -- the word "convert" or "install" -## $2 -- the directory containing the SGML files for the manpages -## $3 -- the filename which contained the sgmldoc output -## (I need this so I know which manpages to convert) - -my($LISTING, $GENERATED, $INPUT, $OUTPUT, $front, $mode, $filename, $tmpdir); - -if($ARGV[0] eq ""){ - die "Usage: makeman [convert | install] <dir> <file>\n"; -} - -if( ! -d "$ARGV[1]" ){ - die "Output directory \"$ARGV[1]\" does not exist\n"; -} - -if($ENV{"TMPDIR"} ne ""){ - $tmpdir = $ENV{"TMPDIR"}; -} -else{ - $tmpdir = "/tmp"; -} - -if($ARGV[0] eq "convert"){ - open LISTING, "grep \"<refentrytitle>\" $ARGV[2] |"; - while(<LISTING>){ - s/<\/.*$//; - s/^.*>//; - s/\.sgml//; - s/struct //; - s/typedef //; - - chomp; - $filename = $_; - print "Processing $filename\n"; - - # Open the input file to extract the front matter, generate the man page, - # and open it, and the rearrange everything until it is happy - open INPUT, "< $ARGV[1]/$filename.sgml"; - $front = ""; - $mode = 0; - - # The modes used here are: - # mode = 0 - # <!-- BEGINFRONTTAG --> - # <!-- <bookinfo> mode = 1 - # <!-- <legalnotice> mode = 2 - # <!-- ...GPL or whatever... - # <!-- </legalnotice> mode = 4 - # <!-- </bookinfo> mode = 3 - # <!-- ENDFRONTTAG --> - # - # ...doco... - - # I know that some of the if statements in this while loop are in a funny - # order, but that is deliberate... - while(<INPUT>){ - if($mode > 0){ - s/<!-- //; - s/ -->//; - s/<docinfo>//i; - s<\/docinfo>//i; - s/^[ \t]*//i; - } - - if($mode == 2){ - if(/<para>/i){ - } - elsif(/<\/para>/i){ - $front = "$front.\\\" \n"; - } - elsif(/<\/legalnotice>/i){ - $mode = 4; - } - elsif(/^[ \t]*$/){ - } - else{ - $front = "$front.\\\" $_"; - } - } - - if($mode == 1){ - if(/<title>(.*)<\/title>/i){ - $front = "$front.\\\" This documentation was generated from the book titled \"$1\", which is part of the Linux kernel source.\n.\\\" \n"; - } - elsif(/<legalnotice>/i){ - $front = "$front.\\\" This documentation comes with the following legal notice:\n.\\\" \n"; - $mode = 2; - } - - elsif(/<author>/i){ - $front = "$front.\\\" Documentation by: "; - } - elsif(/<firstname>(.*)<\/firstname>/i){ - $front = "$front$1 "; - } - elsif(/<surname>(.*)<\/surname>/i){ - $front = "$front$1 "; - } - elsif(/<email>(.*)<\/email>/i){ - $front = "$front($1)"; - } - elsif(/\/author>/i){ - $front = "$front\n"; - } - - elsif(/<copyright>/i){ - $front = "$front.\\\" Documentation copyright: "; - } - elsif(/<holder>(.*)<\/holder>/i){ - $front = "$front$1 "; - } - elsif(/<year>(.*)<\/year>/i){ - $front = "$front$1 "; - } - elsif(/\/copyright>/i){ - $front = "$front\n"; - } - - elsif(/^[ \t]*$/ - || /<affiliation>/i - || /<\/affiliation>/i - || /<address>/i - || /<\/address>/i - || /<authorgroup>/i - || /<\/authorgroup>/i - || /<\/legalnotice>/i - || /<date>/i - || /<\/date>/i - || /<edition>/i - || /<\/edition>/i - || /<pubdate>/i - || /<\/pubdate>/i){ - } - else{ - print "Unknown tag in manpage conversion: $_"; - } - } - - if($mode == 0){ - if(/<bookinfo>/i){ - $mode = 1; - } - } - - if($mode == 4){ - if(/<\/bookinfo>/i){ - $mode = 3; - } - } - } - close INPUT; - - system("cd $ARGV[1]; docbook2man $filename.sgml; mv $filename.9 $tmpdir/$$.9\n"); - open GENERATED, "< $tmpdir/$$.9"; - open OUTPUT, "> $ARGV[1]/$filename.9"; - - print OUTPUT "$front"; - print OUTPUT ".\\\" For comments on the formatting of this manpage, please contact Michael Still <mikal\@stillhq.com>\n\n"; - while(<GENERATED>){ - print OUTPUT "$_"; - } - close OUTPUT; - close GENERATED; - - system("gzip -f $ARGV[1]/$filename.9\n"); - unlink("$tmpdir/$$.9"); - } -} -elsif($ARGV[0] eq "install"){ - system("mkdir -p /usr/local/man/man9/; install $ARGV[1]/*.9.gz /usr/local/man/man9/"); -} -else{ - die "Usage: makeman [convert | install] <dir> <file>\n"; -} - -print "Done\n"; diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c index d54b52d3bb6..32197efe67e 100644 --- a/scripts/mod/file2alias.c +++ b/scripts/mod/file2alias.c @@ -47,32 +47,31 @@ do { \ sprintf(str + strlen(str), "*"); \ } while(0) -/* Looks like "usb:vNpNdlNdhNdcNdscNdpNicNiscNipN" */ -static int do_usb_entry(const char *filename, - struct usb_device_id *id, char *alias) +/* USB is special because the bcdDevice can be matched against a numeric range */ +/* Looks like "usb:vNpNdNdcNdscNdpNicNiscNipN" */ +static void do_usb_entry(struct usb_device_id *id, + unsigned int bcdDevice_initial, int bcdDevice_initial_digits, + unsigned char range_lo, unsigned char range_hi, + struct module *mod) { - id->match_flags = TO_NATIVE(id->match_flags); - id->idVendor = TO_NATIVE(id->idVendor); - id->idProduct = TO_NATIVE(id->idProduct); - id->bcdDevice_lo = TO_NATIVE(id->bcdDevice_lo); - id->bcdDevice_hi = TO_NATIVE(id->bcdDevice_hi); - - /* - * Some modules (visor) have empty slots as placeholder for - * run-time specification that results in catch-all alias - */ - if (!(id->idVendor | id->bDeviceClass | id->bInterfaceClass)) - return 1; - + char alias[500]; strcpy(alias, "usb:"); ADD(alias, "v", id->match_flags&USB_DEVICE_ID_MATCH_VENDOR, id->idVendor); ADD(alias, "p", id->match_flags&USB_DEVICE_ID_MATCH_PRODUCT, id->idProduct); - ADD(alias, "dl", id->match_flags&USB_DEVICE_ID_MATCH_DEV_LO, - id->bcdDevice_lo); - ADD(alias, "dh", id->match_flags&USB_DEVICE_ID_MATCH_DEV_HI, - id->bcdDevice_hi); + + strcat(alias, "d"); + if (bcdDevice_initial_digits) + sprintf(alias + strlen(alias), "%0*X", + bcdDevice_initial_digits, bcdDevice_initial); + if (range_lo == range_hi) + sprintf(alias + strlen(alias), "%u", range_lo); + else if (range_lo > 0 || range_hi < 9) + sprintf(alias + strlen(alias), "[%u-%u]", range_lo, range_hi); + if (bcdDevice_initial_digits < (sizeof(id->bcdDevice_lo) * 2 - 1)) + strcat(alias, "*"); + ADD(alias, "dc", id->match_flags&USB_DEVICE_ID_MATCH_DEV_CLASS, id->bDeviceClass); ADD(alias, "dsc", @@ -90,7 +89,73 @@ static int do_usb_entry(const char *filename, ADD(alias, "ip", id->match_flags&USB_DEVICE_ID_MATCH_INT_PROTOCOL, id->bInterfaceProtocol); - return 1; + + /* Always end in a wildcard, for future extension */ + if (alias[strlen(alias)-1] != '*') + strcat(alias, "*"); + buf_printf(&mod->dev_table_buf, + "MODULE_ALIAS(\"%s\");\n", alias); +} + +static void do_usb_entry_multi(struct usb_device_id *id, struct module *mod) +{ + unsigned int devlo, devhi; + unsigned char chi, clo; + int ndigits; + + id->match_flags = TO_NATIVE(id->match_flags); + id->idVendor = TO_NATIVE(id->idVendor); + id->idProduct = TO_NATIVE(id->idProduct); + + devlo = id->match_flags & USB_DEVICE_ID_MATCH_DEV_LO ? + TO_NATIVE(id->bcdDevice_lo) : 0x0U; + devhi = id->match_flags & USB_DEVICE_ID_MATCH_DEV_HI ? + TO_NATIVE(id->bcdDevice_hi) : ~0x0U; + + /* + * Some modules (visor) have empty slots as placeholder for + * run-time specification that results in catch-all alias + */ + if (!(id->idVendor | id->bDeviceClass | id->bInterfaceClass)) + return; + + /* Convert numeric bcdDevice range into fnmatch-able pattern(s) */ + for (ndigits = sizeof(id->bcdDevice_lo) * 2 - 1; devlo <= devhi; ndigits--) { + clo = devlo & 0xf; + chi = devhi & 0xf; + if (chi > 9) /* it's bcd not hex */ + chi = 9; + devlo >>= 4; + devhi >>= 4; + + if (devlo == devhi || !ndigits) { + do_usb_entry(id, devlo, ndigits, clo, chi, mod); + break; + } + + if (clo > 0) + do_usb_entry(id, devlo++, ndigits, clo, 9, mod); + + if (chi < 9) + do_usb_entry(id, devhi--, ndigits, 0, chi, mod); + } +} + +static void do_usb_table(void *symval, unsigned long size, + struct module *mod) +{ + unsigned int i; + const unsigned long id_size = sizeof(struct usb_device_id); + + if (size % id_size || size < id_size) { + fprintf(stderr, "*** Warning: %s ids %lu bad size " + "(each on %lu)\n", mod->name, size, id_size); + } + /* Leave last one: it's the terminator. */ + size -= id_size; + + for (i = 0; i < size; i += id_size) + do_usb_entry_multi(symval + i, mod); } /* Looks like: ieee1394:venNmoNspNverN */ @@ -280,8 +345,8 @@ void handle_moddevtable(struct module *mod, struct elf_info *info, do_table(symval, sym->st_size, sizeof(struct pci_device_id), do_pci_entry, mod); else if (sym_is(symname, "__mod_usb_device_table")) - do_table(symval, sym->st_size, sizeof(struct usb_device_id), - do_usb_entry, mod); + /* special case to handle bcdDevice ranges */ + do_usb_table(symval, sym->st_size, mod); else if (sym_is(symname, "__mod_ieee1394_device_table")) do_table(symval, sym->st_size, sizeof(struct ieee1394_device_id), do_ieee1394_entry, mod); diff --git a/scripts/patch-kernel b/scripts/patch-kernel index 43af0107561..f2d47ca9c8f 100755 --- a/scripts/patch-kernel +++ b/scripts/patch-kernel @@ -46,6 +46,19 @@ # fix some whitespace damage; # be smarter about stopping when current version is larger than requested; # Randy Dunlap <rddunlap@osdl.org>, 2004-AUG-18. +# +# Add better support for (non-incremental) 2.6.x.y patches; +# If an ending version number if not specified, the script automatically +# increments the SUBLEVEL (x in 2.6.x.y) until no more patch files are found; +# however, EXTRAVERSION (y in 2.6.x.y) is never automatically incremented +# but must be specified fully. +# +# patch-kernel does not normally support reverse patching, but does so when +# applying EXTRAVERSION (x.y) patches, so that moving from 2.6.11.y to 2.6.11.z +# is easy and handled by the script (reverse 2.6.11.y and apply 2.6.11.z). +# Randy Dunlap <rddunlap@osdl.org>, 2005-APR-08. + +PNAME=patch-kernel # Set directories from arguments, or use defaults. sourcedir=${1-/usr/src/linux} @@ -54,7 +67,7 @@ stopvers=${3-default} if [ "$1" == -h -o "$1" == --help -o ! -r "$sourcedir/Makefile" ]; then cat << USAGE -usage: patch-kernel [-h] [ sourcedir [ patchdir [ stopversion ] [ -acxx ] ] ] +usage: $PNAME [-h] [ sourcedir [ patchdir [ stopversion ] [ -acxx ] ] ] source directory defaults to /usr/src/linux, patch directory defaults to the current directory, stopversion defaults to <all in patchdir>. @@ -73,6 +86,19 @@ do done # --------------------------------------------------------------------------- +# arg1 is filename +noFile () { + echo "cannot find patch file: ${patch}" + exit 1 +} + +# --------------------------------------------------------------------------- +backwards () { + echo "$PNAME does not support reverse patching" + exit 1 +} + +# --------------------------------------------------------------------------- # Find a file, first parameter is basename of file # it tries many compression mechanisms and sets variables to say how to get it findFile () { @@ -133,6 +159,28 @@ applyPatch () { return 0; } +# --------------------------------------------------------------------------- +# arg1 is patch filename +reversePatch () { + echo -n "Reversing $1 (${name}) ... " + if $uncomp ${patchdir}/"$1"${ext} | patch -p1 -Rs -N -E -d $sourcedir + then + echo "done." + else + echo "failed. Clean it up." + exit 1 + fi + if [ "`find $sourcedir/ '(' -name '*.rej' -o -name '.*.rej' ')' -print`" ] + then + echo "Aborting. Reject files found." + return 1 + fi + # Remove backup files + find $sourcedir/ '(' -name '*.orig' -o -name '.*.orig' ')' -exec rm -f {} \; + + return 0 +} + # set current VERSION, PATCHLEVEL, SUBLEVEL, EXTRAVERSION TMPFILE=`mktemp .tmpver.XXXXXX` || { echo "cannot make temp file" ; exit 1; } grep -E "^(VERSION|PATCHLEVEL|SUBLEVEL|EXTRAVERSION)" $sourcedir/Makefile > $TMPFILE @@ -160,53 +208,57 @@ then EXTRAVER=$EXTRAVERSION fi EXTRAVER=${EXTRAVER%%[[:punct:]]*} - #echo "patch-kernel: changing EXTRAVERSION from $EXTRAVERSION to $EXTRAVER" + #echo "$PNAME: changing EXTRAVERSION from $EXTRAVERSION to $EXTRAVER" fi #echo "stopvers=$stopvers" if [ $stopvers != "default" ]; then STOPSUBLEVEL=`echo $stopvers | cut -d. -f3` STOPEXTRA=`echo $stopvers | cut -d. -f4` - #echo "STOPSUBLEVEL=$STOPSUBLEVEL, STOPEXTRA=$STOPEXTRA" + #echo "#___STOPSUBLEVEL=/$STOPSUBLEVEL/, STOPEXTRA=/$STOPEXTRA/" else STOPSUBLEVEL=9999 STOPEXTRA=9999 fi -while : # incrementing SUBLEVEL (s in v.p.s) -do - if [ x$EXTRAVER != "x" ]; then +# This all assumes a 2.6.x[.y] kernel tree. +# Don't allow backwards/reverse patching. +if [ $STOPSUBLEVEL -lt $SUBLEVEL ]; then + backwards +fi + +if [ x$EXTRAVER != "x" ]; then CURRENTFULLVERSION="$VERSION.$PATCHLEVEL.$SUBLEVEL.$EXTRAVER" - else +else CURRENTFULLVERSION="$VERSION.$PATCHLEVEL.$SUBLEVEL" - fi +fi + +if [ x$EXTRAVER != "x" ]; then + echo "backing up to: $VERSION.$PATCHLEVEL.$SUBLEVEL" + patch="patch-${CURRENTFULLVERSION}" + findFile $patchdir/${patch} || noFile ${patch} + reversePatch ${patch} || exit 1 +fi +# now current is 2.6.x, with no EXTRA applied, +# so update to target SUBLEVEL (2.6.SUBLEVEL) +# and then to target EXTRAVER (2.6.SUB.EXTRAVER) if requested. +# If not ending sublevel is specified, it is incremented until +# no further sublevels are found. + +if [ $STOPSUBLEVEL -gt $SUBLEVEL ]; then +while : # incrementing SUBLEVEL (s in v.p.s) +do + CURRENTFULLVERSION="$VERSION.$PATCHLEVEL.$SUBLEVEL" + EXTRAVER= if [ $stopvers == $CURRENTFULLVERSION ]; then echo "Stopping at $CURRENTFULLVERSION base as requested." break fi - while : # incrementing EXTRAVER (x in v.p.s.x) - do - EXTRAVER=$((EXTRAVER + 1)) - FULLVERSION="$VERSION.$PATCHLEVEL.$SUBLEVEL.$EXTRAVER" - #echo "... trying $FULLVERSION ..." - - patch=patch-$FULLVERSION - - # See if the file exists and find extension - findFile $patchdir/${patch} || break - - # Apply the patch and check all is OK - applyPatch $patch || break - - continue 2 - done - - EXTRAVER= SUBLEVEL=$((SUBLEVEL + 1)) FULLVERSION="$VERSION.$PATCHLEVEL.$SUBLEVEL" - #echo "___ trying $FULLVERSION ___" + #echo "#___ trying $FULLVERSION ___" if [ $((SUBLEVEL)) -gt $((STOPSUBLEVEL)) ]; then echo "Stopping since sublevel ($SUBLEVEL) is beyond stop-sublevel ($STOPSUBLEVEL)" @@ -214,14 +266,33 @@ do fi patch=patch-$FULLVERSION - # See if the file exists and find extension - findFile $patchdir/${patch} || break + findFile $patchdir/${patch} || noFile ${patch} # Apply the patch and check all is OK applyPatch $patch || break done -#echo "base all done" +#echo "#___sublevel all done" +fi + +# There is no incremental searching for extraversion... +if [ "$STOPEXTRA" != "" ]; then +while : # just to allow break +do +# apply STOPEXTRA directly (not incrementally) (x in v.p.s.x) + FULLVERSION="$VERSION.$PATCHLEVEL.$SUBLEVEL.$STOPEXTRA" + #echo "#... trying $FULLVERSION ..." + patch=patch-$FULLVERSION + + # See if the file exists and find extension + findFile $patchdir/${patch} || noFile ${patch} + + # Apply the patch and check all is OK + applyPatch $patch || break + #echo "#___extraver all done" + break +done +fi if [ x$gotac != x ]; then # Out great user wants the -ac patches diff --git a/scripts/split-man b/scripts/split-man deleted file mode 100755 index 03897fe6a75..00000000000 --- a/scripts/split-man +++ /dev/null @@ -1,112 +0,0 @@ -#!/usr/bin/perl - -use strict; - -## Copyright (C) Michael Still (mikal@stillhq.com) -## Released under the terms of the GNU GPL -## -## Hoon through the specified DocBook SGML file, and split out the -## man pages. These can then be processed into groff format, and -## installed if desired... -## -## Arguements: $1 -- the name of the sgml file -## $2 -- the directory to put the generated SGML files in -## $3 -- kernel version - -my($SGML, $REF, $front, $refdata, $mode, $filename); - -if(($ARGV[0] eq "") || ($ARGV[1] eq "") || ($ARGV[2] eq "")){ - die "Usage: split-man <sgml file> <output dir> <kernel version>\n"; -} - -open SGML, "< $ARGV[0]" or die "Could not open input file \"$ARGV[0]\"\n"; -if( ! -d "$ARGV[1]" ){ - die "Output directory \"$ARGV[1]\" does not exist\n"; -} - -# Possible modes: -# 0: Looking for input I care about -# 1: Inside book front matter -# 2: Inside a refentry -# 3: Inside a refentry, and we know the filename - -$mode = 0; -$refdata = ""; -$front = ""; -while(<SGML>){ - # Starting modes - if(/<bookinfo>/ || /<docinfo>/){ - $mode = 1; - } - elsif(/<refentry>/){ - $mode = 2; - } - elsif(/<refentrytitle><phrase[^>]*>([^<]*)<.*$/){ - $mode = 3; - $filename = $1; - - $filename =~ s/struct //; - $filename =~ s/typedef //; - - print "Found manpage for $filename\n"; - open REF, "> $ARGV[1]/$filename.sgml" or - die "Couldn't open output file \"$ARGV[1]/$filename.sgml\": $!\n"; - print REF <<EOF; -<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook V4.1//EN"> - -<!-- BEGINFRONTTAG: The following is front matter for the parent book --> -$front -<!-- ENDFRONTTAG: End front matter --> - -$refdata -EOF - $refdata = ""; - } - - # Extraction - if($mode == 1){ - chomp $_; - $front = "$front<!-- $_ -->\n"; - } - elsif($mode == 2){ - $refdata = "$refdata$_"; - } - elsif($mode == 3){ - # There are some fixups which need to be applied - if(/<\/refmeta>/){ - print REF "<manvolnum>9</manvolnum>\n"; - } - if(/<\/refentry>/){ - print REF <<EOF; -<refsect1><title>About this document</title> -<para> -This documentation was generated with kernel version $ARGV[2]. -</para> -</refsect1> -EOF - } - - # For some reason, we title the synopsis twice in the main DocBook - if(! /<title>Synopsis<\/title>/){ - if(/<refentrytitle>/){ - s/struct //; - s/typedef //; - } - - print REF "$_"; - } - } - - # Ending modes - if(/<\/bookinfo>/ || /<\/docinfo>/){ - $mode = 0; - } - elsif(/<\/refentry>/){ - $mode = 0; - close REF; - } -} - -# And make sure we don't process this unnessesarily -$ARGV[0] =~ s/\.sgml/.9/; -`touch $ARGV[0]`; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 2ae7d3cb8df..aae1e794fe4 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -2855,8 +2855,7 @@ static int selinux_parse_skb_ipv6(struct sk_buff *skb, struct avc_audit_data *ad nexthdr = ip6->nexthdr; offset += sizeof(_ipv6h); - offset = ipv6_skip_exthdr(skb, offset, &nexthdr, - skb->tail - skb->head - offset); + offset = ipv6_skip_exthdr(skb, offset, &nexthdr); if (offset < 0) goto out; @@ -3668,7 +3667,7 @@ static void msg_msg_free_security(struct msg_msg *msg) } static int ipc_has_perm(struct kern_ipc_perm *ipc_perms, - u16 sclass, u32 perms) + u32 perms) { struct task_security_struct *tsec; struct ipc_security_struct *isec; @@ -3680,7 +3679,7 @@ static int ipc_has_perm(struct kern_ipc_perm *ipc_perms, AVC_AUDIT_DATA_INIT(&ad, IPC); ad.u.ipc_id = ipc_perms->key; - return avc_has_perm(tsec->sid, isec->sid, sclass, perms, &ad); + return avc_has_perm(tsec->sid, isec->sid, isec->sclass, perms, &ad); } static int selinux_msg_msg_alloc_security(struct msg_msg *msg) @@ -3765,7 +3764,7 @@ static int selinux_msg_queue_msgctl(struct msg_queue *msq, int cmd) return 0; } - err = ipc_has_perm(&msq->q_perm, SECCLASS_MSGQ, perms); + err = ipc_has_perm(&msq->q_perm, perms); return err; } @@ -3917,7 +3916,7 @@ static int selinux_shm_shmctl(struct shmid_kernel *shp, int cmd) return 0; } - err = ipc_has_perm(&shp->shm_perm, SECCLASS_SHM, perms); + err = ipc_has_perm(&shp->shm_perm, perms); return err; } @@ -3936,7 +3935,7 @@ static int selinux_shm_shmat(struct shmid_kernel *shp, else perms = SHM__READ | SHM__WRITE; - return ipc_has_perm(&shp->shm_perm, SECCLASS_SHM, perms); + return ipc_has_perm(&shp->shm_perm, perms); } /* Semaphore security operations */ @@ -4025,7 +4024,7 @@ static int selinux_sem_semctl(struct sem_array *sma, int cmd) return 0; } - err = ipc_has_perm(&sma->sem_perm, SECCLASS_SEM, perms); + err = ipc_has_perm(&sma->sem_perm, perms); return err; } @@ -4039,18 +4038,13 @@ static int selinux_sem_semop(struct sem_array *sma, else perms = SEM__READ; - return ipc_has_perm(&sma->sem_perm, SECCLASS_SEM, perms); + return ipc_has_perm(&sma->sem_perm, perms); } static int selinux_ipc_permission(struct kern_ipc_perm *ipcp, short flag) { - struct ipc_security_struct *isec = ipcp->security; - u16 sclass = SECCLASS_IPC; u32 av = 0; - if (isec && isec->magic == SELINUX_MAGIC) - sclass = isec->sclass; - av = 0; if (flag & S_IRUGO) av |= IPC__UNIX_READ; @@ -4060,7 +4054,7 @@ static int selinux_ipc_permission(struct kern_ipc_perm *ipcp, short flag) if (av == 0) return 0; - return ipc_has_perm(ipcp, sclass, av); + return ipc_has_perm(ipcp, av); } /* module stacking operations */ diff --git a/security/selinux/include/av_perm_to_string.h b/security/selinux/include/av_perm_to_string.h index eb340b45bc6..8928bb4d3c5 100644 --- a/security/selinux/include/av_perm_to_string.h +++ b/security/selinux/include/av_perm_to_string.h @@ -220,6 +220,8 @@ S_(SECCLASS_NETLINK_XFRM_SOCKET, NETLINK_XFRM_SOCKET__NLMSG_WRITE, "nlmsg_write") S_(SECCLASS_NETLINK_AUDIT_SOCKET, NETLINK_AUDIT_SOCKET__NLMSG_READ, "nlmsg_read") S_(SECCLASS_NETLINK_AUDIT_SOCKET, NETLINK_AUDIT_SOCKET__NLMSG_WRITE, "nlmsg_write") + S_(SECCLASS_NETLINK_AUDIT_SOCKET, NETLINK_AUDIT_SOCKET__NLMSG_RELAY, "nlmsg_relay") + S_(SECCLASS_NETLINK_AUDIT_SOCKET, NETLINK_AUDIT_SOCKET__NLMSG_READPRIV, "nlmsg_readpriv") S_(SECCLASS_NETLINK_IP6FW_SOCKET, NETLINK_IP6FW_SOCKET__NLMSG_READ, "nlmsg_read") S_(SECCLASS_NETLINK_IP6FW_SOCKET, NETLINK_IP6FW_SOCKET__NLMSG_WRITE, "nlmsg_write") S_(SECCLASS_DBUS, DBUS__ACQUIRE_SVC, "acquire_svc") diff --git a/security/selinux/include/av_permissions.h b/security/selinux/include/av_permissions.h index f9de0f96655..bdfce4ca8f8 100644 --- a/security/selinux/include/av_permissions.h +++ b/security/selinux/include/av_permissions.h @@ -840,6 +840,8 @@ #define NETLINK_AUDIT_SOCKET__NLMSG_READ 0x00400000UL #define NETLINK_AUDIT_SOCKET__NLMSG_WRITE 0x00800000UL +#define NETLINK_AUDIT_SOCKET__NLMSG_RELAY 0x01000000UL +#define NETLINK_AUDIT_SOCKET__NLMSG_READPRIV 0x02000000UL #define NETLINK_IP6FW_SOCKET__IOCTL 0x00000001UL #define NETLINK_IP6FW_SOCKET__READ 0x00000002UL diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index f7940825273..b3adb481bc2 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -91,13 +91,12 @@ static struct nlmsg_perm nlmsg_xfrm_perms[] = static struct nlmsg_perm nlmsg_audit_perms[] = { - { AUDIT_GET, NETLINK_AUDIT_SOCKET__NLMSG_READ }, - { AUDIT_SET, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, - { AUDIT_LIST, NETLINK_AUDIT_SOCKET__NLMSG_READ }, - { AUDIT_ADD, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, - { AUDIT_DEL, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, - { AUDIT_USER, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, - { AUDIT_LOGIN, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, + { AUDIT_GET, NETLINK_AUDIT_SOCKET__NLMSG_READ }, + { AUDIT_SET, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, + { AUDIT_LIST, NETLINK_AUDIT_SOCKET__NLMSG_READPRIV }, + { AUDIT_ADD, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, + { AUDIT_DEL, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, + { AUDIT_USER, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, }; diff --git a/sound/isa/Kconfig b/sound/isa/Kconfig index 4a6be966bd9..3a3228b1872 100644 --- a/sound/isa/Kconfig +++ b/sound/isa/Kconfig @@ -164,6 +164,7 @@ config SND_INTERWAVE select SND_RAWMIDI select SND_CS4231_LIB select SND_GUS_SYNTH + select ISAPNP help Say Y here to include support for AMD InterWave based soundcards (Gravis UltraSound Plug & Play, STB SoundRage32, diff --git a/sound/oss/ac97_codec.c b/sound/oss/ac97_codec.c index 124b1e10a13..3ecef4689f1 100644 --- a/sound/oss/ac97_codec.c +++ b/sound/oss/ac97_codec.c @@ -155,6 +155,7 @@ static const struct { {0x43525931, "Cirrus Logic CS4299 rev A", &crystal_digital_ops}, {0x43525933, "Cirrus Logic CS4299 rev C", &crystal_digital_ops}, {0x43525934, "Cirrus Logic CS4299 rev D", &crystal_digital_ops}, + {0x43585430, "CXT48", &default_ops, AC97_DELUDED_MODEM }, {0x43585442, "CXT66", &default_ops, AC97_DELUDED_MODEM }, {0x44543031, "Diamond Technology DT0893", &default_ops}, {0x45838308, "ESS Allegro ES1988", &null_ops}, diff --git a/sound/oss/msnd_pinnacle.c b/sound/oss/msnd_pinnacle.c index 6ba03f89fc4..0c2db657bad 100644 --- a/sound/oss/msnd_pinnacle.c +++ b/sound/oss/msnd_pinnacle.c @@ -892,7 +892,7 @@ static __inline__ int pack_DAPF_to_DAPQ(register int start) static int dsp_read(char __user *buf, size_t len) { int count = len; - char *page = (char *)__get_free_page(PAGE_SIZE); + char *page = (char *)__get_free_page(GFP_KERNEL); if (!page) return -ENOMEM; diff --git a/sound/oss/sonicvibes.c b/sound/oss/sonicvibes.c index e1d69611a25..06047e7979a 100644 --- a/sound/oss/sonicvibes.c +++ b/sound/oss/sonicvibes.c @@ -1149,7 +1149,7 @@ static int mixer_ioctl(struct sv_state *s, unsigned int cmd, unsigned long arg) if (mixtable[i].rec) break; } - if (!mixtable[i].rec) + if (i == SOUND_MIXER_NRDEVICES) return 0; spin_lock_irqsave(&s->lock, flags); frobindir(s, SV_CIMIX_ADCINL, 0x1f, mixtable[i].rec << 5); diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index d89647a3d44..959953ca320 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -64,7 +64,8 @@ MODULE_PARM_DESC(model, "Use the given board model."); MODULE_LICENSE("GPL"); MODULE_SUPPORTED_DEVICE("{{Intel, ICH6}," "{Intel, ICH6M}," - "{Intel, ICH7}}"); + "{Intel, ICH7}," + "{Intel, ESB2}}"); MODULE_DESCRIPTION("Intel HDA driver"); #define SFX "hda-intel: " @@ -1422,6 +1423,7 @@ static void __devexit azx_remove(struct pci_dev *pci) static struct pci_device_id azx_ids[] = { { 0x8086, 0x2668, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ICH6 */ { 0x8086, 0x27d8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ICH7 */ + { 0x8086, 0x269a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ESB2 */ { 0, } }; MODULE_DEVICE_TABLE(pci, azx_ids); diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c index d143d2c7898..8b33b12fa5d 100644 --- a/sound/pci/intel8x0.c +++ b/sound/pci/intel8x0.c @@ -125,8 +125,8 @@ MODULE_PARM_DESC(xbox, "Set to 1 for Xbox, if you have problems with the AC'97 c #ifndef PCI_DEVICE_ID_INTEL_ICH7_20 #define PCI_DEVICE_ID_INTEL_ICH7_20 0x27de #endif -#ifndef PCI_DEVICE_ID_INTEL_ESB2_13 -#define PCI_DEVICE_ID_INTEL_ESB2_13 0x2698 +#ifndef PCI_DEVICE_ID_INTEL_ESB2_14 +#define PCI_DEVICE_ID_INTEL_ESB2_14 0x2698 #endif #ifndef PCI_DEVICE_ID_SI_7012 #define PCI_DEVICE_ID_SI_7012 0x7012 @@ -2741,7 +2741,7 @@ static struct shortname_table { { PCI_DEVICE_ID_INTEL_ESB_5, "Intel 6300ESB" }, { PCI_DEVICE_ID_INTEL_ICH6_18, "Intel ICH6" }, { PCI_DEVICE_ID_INTEL_ICH7_20, "Intel ICH7" }, - { PCI_DEVICE_ID_INTEL_ESB2_13, "Intel ESB2" }, + { PCI_DEVICE_ID_INTEL_ESB2_14, "Intel ESB2" }, { PCI_DEVICE_ID_SI_7012, "SiS SI7012" }, { PCI_DEVICE_ID_NVIDIA_MCP_AUDIO, "NVidia nForce" }, { PCI_DEVICE_ID_NVIDIA_MCP2_AUDIO, "NVidia nForce2" }, diff --git a/sound/pci/via82xx.c b/sound/pci/via82xx.c index f1ce808501d..9b4d74d49f9 100644 --- a/sound/pci/via82xx.c +++ b/sound/pci/via82xx.c @@ -1836,7 +1836,7 @@ static void __devinit snd_via82xx_proc_init(via82xx_t *chip) * */ -static int __devinit snd_via82xx_chip_init(via82xx_t *chip) +static int snd_via82xx_chip_init(via82xx_t *chip) { unsigned int val; int max_count; diff --git a/sound/ppc/Kconfig b/sound/ppc/Kconfig index b0a9ebf8bf3..75213bf4d56 100644 --- a/sound/ppc/Kconfig +++ b/sound/ppc/Kconfig @@ -11,7 +11,7 @@ comment "ALSA PowerMac requires INPUT" config SND_POWERMAC tristate "PowerMac (AWACS, DACA, Burgundy, Tumbler, Keywest)" - depends on SND && I2C && INPUT + depends on SND && I2C && INPUT && PPC_PMAC select SND_PCM help Say Y here to include support for the integrated sound device. diff --git a/sound/ppc/Makefile b/sound/ppc/Makefile index 4d95c652c8c..d6ba9959097 100644 --- a/sound/ppc/Makefile +++ b/sound/ppc/Makefile @@ -3,7 +3,7 @@ # Copyright (c) 2001 by Jaroslav Kysela <perex@suse.cz> # -snd-powermac-objs := powermac.o pmac.o awacs.o burgundy.o daca.o tumbler.o keywest.o beep.o +snd-powermac-objs := powermac.o pmac.o awacs.o burgundy.o daca.o tumbler.o toonie.o keywest.o beep.o # Toplevel Module Dependency obj-$(CONFIG_SND_POWERMAC) += snd-powermac.o diff --git a/sound/ppc/pmac.c b/sound/ppc/pmac.c index 3bf5f069a03..32d94754acf 100644 --- a/sound/ppc/pmac.c +++ b/sound/ppc/pmac.c @@ -986,7 +986,13 @@ static int __init snd_pmac_detect(pmac_t *chip) chip->num_freqs = ARRAY_SIZE(tumbler_freqs); chip->model = PMAC_SNAPPER; chip->can_byte_swap = 0; /* FIXME: check this */ - chip->control_mask = MASK_IEPC | 0x11; /* disable IEE */ + chip->control_mask = MASK_IEPC | 0x11;/* disable IEE */ + break; + case 0x3a: + chip->num_freqs = ARRAY_SIZE(tumbler_freqs); + chip->model = PMAC_TOONIE; + chip->can_byte_swap = 0; /* FIXME: check this */ + chip->control_mask = MASK_IEPC | 0x11;/* disable IEE */ break; } } diff --git a/sound/ppc/pmac.h b/sound/ppc/pmac.h index dc6c99dd14e..0a84c05f714 100644 --- a/sound/ppc/pmac.h +++ b/sound/ppc/pmac.h @@ -94,7 +94,8 @@ struct snd_pmac_stream { */ enum snd_pmac_model { - PMAC_AWACS, PMAC_SCREAMER, PMAC_BURGUNDY, PMAC_DACA, PMAC_TUMBLER, PMAC_SNAPPER + PMAC_AWACS, PMAC_SCREAMER, PMAC_BURGUNDY, PMAC_DACA, PMAC_TUMBLER, + PMAC_SNAPPER, PMAC_TOONIE }; struct snd_pmac { @@ -191,6 +192,7 @@ int snd_pmac_burgundy_init(pmac_t *chip); int snd_pmac_daca_init(pmac_t *chip); int snd_pmac_tumbler_init(pmac_t *chip); int snd_pmac_tumbler_post_init(void); +int snd_pmac_toonie_init(pmac_t *chip); /* i2c functions */ typedef struct snd_pmac_keywest { diff --git a/sound/ppc/powermac.c b/sound/ppc/powermac.c index 8f1953a8e29..231f6432ea6 100644 --- a/sound/ppc/powermac.c +++ b/sound/ppc/powermac.c @@ -95,6 +95,13 @@ static int __init snd_pmac_probe(void) if ( snd_pmac_tumbler_init(chip) < 0 || snd_pmac_tumbler_post_init() < 0) goto __error; break; + case PMAC_TOONIE: + strcpy(card->driver, "PMac Toonie"); + strcpy(card->shortname, "PowerMac Toonie"); + strcpy(card->longname, card->shortname); + if ((err = snd_pmac_toonie_init(chip)) < 0) + goto __error; + break; case PMAC_AWACS: case PMAC_SCREAMER: name_ext = chip->model == PMAC_SCREAMER ? "Screamer" : "AWACS"; diff --git a/sound/ppc/toonie.c b/sound/ppc/toonie.c new file mode 100644 index 00000000000..082bc4babab --- /dev/null +++ b/sound/ppc/toonie.c @@ -0,0 +1,379 @@ +/* + * Mac Mini "toonie" mixer control + * + * Copyright (c) 2005 by Benjamin Herrenschmidt <benh@kernel.crashing.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include <sound/driver.h> +#include <linux/init.h> +#include <linux/delay.h> +#include <linux/i2c.h> +#include <linux/i2c-dev.h> +#include <linux/kmod.h> +#include <linux/slab.h> +#include <linux/interrupt.h> +#include <sound/core.h> +#include <asm/io.h> +#include <asm/irq.h> +#include <asm/machdep.h> +#include <asm/pmac_feature.h> +#include "pmac.h" + +#undef DEBUG + +#ifdef DEBUG +#define DBG(fmt...) printk(fmt) +#else +#define DBG(fmt...) +#endif + +struct pmac_gpio { + unsigned int addr; + u8 active_val; + u8 inactive_val; + u8 active_state; +}; + +struct pmac_toonie +{ + struct pmac_gpio hp_detect_gpio; + struct pmac_gpio hp_mute_gpio; + struct pmac_gpio amp_mute_gpio; + int hp_detect_irq; + int auto_mute_notify; + struct work_struct detect_work; +}; + + +/* + * gpio access + */ +#define do_gpio_write(gp, val) \ + pmac_call_feature(PMAC_FTR_WRITE_GPIO, NULL, (gp)->addr, val) +#define do_gpio_read(gp) \ + pmac_call_feature(PMAC_FTR_READ_GPIO, NULL, (gp)->addr, 0) +#define tumbler_gpio_free(gp) /* NOP */ + +static void write_audio_gpio(struct pmac_gpio *gp, int active) +{ + if (! gp->addr) + return; + active = active ? gp->active_val : gp->inactive_val; + do_gpio_write(gp, active); + DBG("(I) gpio %x write %d\n", gp->addr, active); +} + +static int check_audio_gpio(struct pmac_gpio *gp) +{ + int ret; + + if (! gp->addr) + return 0; + + ret = do_gpio_read(gp); + + return (ret & 0xd) == (gp->active_val & 0xd); +} + +static int read_audio_gpio(struct pmac_gpio *gp) +{ + int ret; + if (! gp->addr) + return 0; + ret = ((do_gpio_read(gp) & 0x02) !=0); + return ret == gp->active_state; +} + + +enum { TOONIE_MUTE_HP, TOONIE_MUTE_AMP }; + +static int toonie_get_mute_switch(snd_kcontrol_t *kcontrol, + snd_ctl_elem_value_t *ucontrol) +{ + pmac_t *chip = snd_kcontrol_chip(kcontrol); + struct pmac_toonie *mix = chip->mixer_data; + struct pmac_gpio *gp; + + if (mix == NULL) + return -ENODEV; + switch(kcontrol->private_value) { + case TOONIE_MUTE_HP: + gp = &mix->hp_mute_gpio; + break; + case TOONIE_MUTE_AMP: + gp = &mix->amp_mute_gpio; + break; + default: + return -EINVAL;; + } + ucontrol->value.integer.value[0] = !check_audio_gpio(gp); + return 0; +} + +static int toonie_put_mute_switch(snd_kcontrol_t *kcontrol, + snd_ctl_elem_value_t *ucontrol) +{ + pmac_t *chip = snd_kcontrol_chip(kcontrol); + struct pmac_toonie *mix = chip->mixer_data; + struct pmac_gpio *gp; + int val; + + if (chip->update_automute && chip->auto_mute) + return 0; /* don't touch in the auto-mute mode */ + + if (mix == NULL) + return -ENODEV; + + switch(kcontrol->private_value) { + case TOONIE_MUTE_HP: + gp = &mix->hp_mute_gpio; + break; + case TOONIE_MUTE_AMP: + gp = &mix->amp_mute_gpio; + break; + default: + return -EINVAL;; + } + val = ! check_audio_gpio(gp); + if (val != ucontrol->value.integer.value[0]) { + write_audio_gpio(gp, ! ucontrol->value.integer.value[0]); + return 1; + } + return 0; +} + +static snd_kcontrol_new_t toonie_hp_sw __initdata = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Headphone Playback Switch", + .info = snd_pmac_boolean_mono_info, + .get = toonie_get_mute_switch, + .put = toonie_put_mute_switch, + .private_value = TOONIE_MUTE_HP, +}; +static snd_kcontrol_new_t toonie_speaker_sw __initdata = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "PC Speaker Playback Switch", + .info = snd_pmac_boolean_mono_info, + .get = toonie_get_mute_switch, + .put = toonie_put_mute_switch, + .private_value = TOONIE_MUTE_AMP, +}; + +/* + * auto-mute stuffs + */ +static int toonie_detect_headphone(pmac_t *chip) +{ + struct pmac_toonie *mix = chip->mixer_data; + int detect = 0; + + if (mix->hp_detect_gpio.addr) + detect |= read_audio_gpio(&mix->hp_detect_gpio); + return detect; +} + +static void toonie_check_mute(pmac_t *chip, struct pmac_gpio *gp, int val, + int do_notify, snd_kcontrol_t *sw) +{ + if (check_audio_gpio(gp) != val) { + write_audio_gpio(gp, val); + if (do_notify) + snd_ctl_notify(chip->card, SNDRV_CTL_EVENT_MASK_VALUE, + &sw->id); + } +} + +static void toonie_detect_handler(void *self) +{ + pmac_t *chip = (pmac_t*) self; + struct pmac_toonie *mix; + int headphone; + + if (!chip) + return; + + mix = chip->mixer_data; + snd_assert(mix, return); + + headphone = toonie_detect_headphone(chip); + + DBG("headphone: %d, lineout: %d\n", headphone, lineout); + + if (headphone) { + /* unmute headphone/lineout & mute speaker */ + toonie_check_mute(chip, &mix->hp_mute_gpio, 0, + mix->auto_mute_notify, chip->master_sw_ctl); + toonie_check_mute(chip, &mix->amp_mute_gpio, 1, + mix->auto_mute_notify, chip->speaker_sw_ctl); + } else { + /* unmute speaker, mute others */ + toonie_check_mute(chip, &mix->amp_mute_gpio, 0, + mix->auto_mute_notify, chip->speaker_sw_ctl); + toonie_check_mute(chip, &mix->hp_mute_gpio, 1, + mix->auto_mute_notify, chip->master_sw_ctl); + } + if (mix->auto_mute_notify) { + snd_ctl_notify(chip->card, SNDRV_CTL_EVENT_MASK_VALUE, + &chip->hp_detect_ctl->id); + } +} + +static void toonie_update_automute(pmac_t *chip, int do_notify) +{ + if (chip->auto_mute) { + struct pmac_toonie *mix; + mix = chip->mixer_data; + snd_assert(mix, return); + mix->auto_mute_notify = do_notify; + schedule_work(&mix->detect_work); + } +} + +/* interrupt - headphone plug changed */ +static irqreturn_t toonie_hp_intr(int irq, void *devid, struct pt_regs *regs) +{ + pmac_t *chip = devid; + + if (chip->update_automute && chip->initialized) { + chip->update_automute(chip, 1); + return IRQ_HANDLED; + } + return IRQ_NONE; +} + +/* look for audio gpio device */ +static int find_audio_gpio(const char *name, const char *platform, + struct pmac_gpio *gp) +{ + struct device_node *np; + u32 *base, addr; + + if (! (np = find_devices("gpio"))) + return -ENODEV; + + for (np = np->child; np; np = np->sibling) { + char *property = get_property(np, "audio-gpio", NULL); + if (property && strcmp(property, name) == 0) + break; + if (device_is_compatible(np, name)) + break; + } + if (np == NULL) + return -ENODEV; + + base = (u32 *)get_property(np, "AAPL,address", NULL); + if (! base) { + base = (u32 *)get_property(np, "reg", NULL); + if (!base) { + DBG("(E) cannot find address for device %s !\n", name); + return -ENODEV; + } + addr = *base; + if (addr < 0x50) + addr += 0x50; + } else + addr = *base; + + gp->addr = addr & 0x0000ffff; + + /* Try to find the active state, default to 0 ! */ + base = (u32 *)get_property(np, "audio-gpio-active-state", NULL); + if (base) { + gp->active_state = *base; + gp->active_val = (*base) ? 0x5 : 0x4; + gp->inactive_val = (*base) ? 0x4 : 0x5; + } else { + u32 *prop = NULL; + gp->active_state = 0; + gp->active_val = 0x4; + gp->inactive_val = 0x5; + /* Here are some crude hacks to extract the GPIO polarity and + * open collector informations out of the do-platform script + * as we don't yet have an interpreter for these things + */ + if (platform) + prop = (u32 *)get_property(np, platform, NULL); + if (prop) { + if (prop[3] == 0x9 && prop[4] == 0x9) { + gp->active_val = 0xd; + gp->inactive_val = 0xc; + } + if (prop[3] == 0x1 && prop[4] == 0x1) { + gp->active_val = 0x5; + gp->inactive_val = 0x4; + } + } + } + + DBG("(I) GPIO device %s found, offset: %x, active state: %d !\n", + name, gp->addr, gp->active_state); + + return (np->n_intrs > 0) ? np->intrs[0].line : 0; +} + +static void toonie_cleanup(pmac_t *chip) +{ + struct pmac_toonie *mix = chip->mixer_data; + if (! mix) + return; + if (mix->hp_detect_irq >= 0) + free_irq(mix->hp_detect_irq, chip); + kfree(mix); + chip->mixer_data = NULL; +} + +int snd_pmac_toonie_init(pmac_t *chip) +{ + struct pmac_toonie *mix; + + mix = kmalloc(sizeof(*mix), GFP_KERNEL); + if (! mix) + return -ENOMEM; + + chip->mixer_data = mix; + chip->mixer_free = toonie_cleanup; + + find_audio_gpio("headphone-mute", NULL, &mix->hp_mute_gpio); + find_audio_gpio("amp-mute", NULL, &mix->amp_mute_gpio); + mix->hp_detect_irq = find_audio_gpio("headphone-detect", + NULL, &mix->hp_detect_gpio); + + strcpy(chip->card->mixername, "PowerMac Toonie"); + + chip->master_sw_ctl = snd_ctl_new1(&toonie_hp_sw, chip); + snd_ctl_add(chip->card, chip->master_sw_ctl); + + chip->speaker_sw_ctl = snd_ctl_new1(&toonie_speaker_sw, chip); + snd_ctl_add(chip->card, chip->speaker_sw_ctl); + + INIT_WORK(&mix->detect_work, toonie_detect_handler, (void *)chip); + + if (mix->hp_detect_irq >= 0) { + snd_pmac_add_automute(chip); + + chip->detect_headphone = toonie_detect_headphone; + chip->update_automute = toonie_update_automute; + toonie_update_automute(chip, 0); + + if (request_irq(mix->hp_detect_irq, toonie_hp_intr, 0, + "Sound Headphone Detection", chip) < 0) + mix->hp_detect_irq = -1; + } + + return 0; +} + diff --git a/sound/ppc/tumbler.c b/sound/ppc/tumbler.c index c71807e069e..9332237cb6a 100644 --- a/sound/ppc/tumbler.c +++ b/sound/ppc/tumbler.c @@ -99,6 +99,7 @@ typedef struct pmac_tumbler_t { pmac_gpio_t hp_detect; int headphone_irq; int lineout_irq; + unsigned int save_master_vol[2]; unsigned int master_vol[2]; unsigned int save_master_switch[2]; unsigned int master_switch[2]; @@ -177,11 +178,22 @@ static void write_audio_gpio(pmac_gpio_t *gp, int active) if (! gp->addr) return; active = active ? gp->active_val : gp->inactive_val; - do_gpio_write(gp, active); DBG("(I) gpio %x write %d\n", gp->addr, active); } +static int check_audio_gpio(pmac_gpio_t *gp) +{ + int ret; + + if (! gp->addr) + return 0; + + ret = do_gpio_read(gp); + + return (ret & 0xd) == (gp->active_val & 0xd); +} + static int read_audio_gpio(pmac_gpio_t *gp) { int ret; @@ -683,7 +695,7 @@ static int tumbler_get_mute_switch(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_ } if (gp == NULL) return -EINVAL; - ucontrol->value.integer.value[0] = ! read_audio_gpio(gp); + ucontrol->value.integer.value[0] = !check_audio_gpio(gp); return 0; } @@ -711,7 +723,7 @@ static int tumbler_put_mute_switch(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_ } if (gp == NULL) return -EINVAL; - val = ! read_audio_gpio(gp); + val = ! check_audio_gpio(gp); if (val != ucontrol->value.integer.value[0]) { write_audio_gpio(gp, ! ucontrol->value.integer.value[0]); return 1; @@ -897,11 +909,11 @@ static int tumbler_detect_lineout(pmac_t *chip) static void check_mute(pmac_t *chip, pmac_gpio_t *gp, int val, int do_notify, snd_kcontrol_t *sw) { - //pmac_tumbler_t *mix = chip->mixer_data; - if (val != read_audio_gpio(gp)) { + if (check_audio_gpio(gp) != val) { write_audio_gpio(gp, val); if (do_notify) - snd_ctl_notify(chip->card, SNDRV_CTL_EVENT_MASK_VALUE, &sw->id); + snd_ctl_notify(chip->card, SNDRV_CTL_EVENT_MASK_VALUE, + &sw->id); } } @@ -1128,6 +1140,8 @@ static void tumbler_suspend(pmac_t *chip) disable_irq(mix->lineout_irq); mix->save_master_switch[0] = mix->master_switch[0]; mix->save_master_switch[1] = mix->master_switch[1]; + mix->save_master_vol[0] = mix->master_vol[0]; + mix->save_master_vol[1] = mix->master_vol[1]; mix->master_switch[0] = mix->master_switch[1] = 0; tumbler_set_master_volume(mix); if (!mix->anded_reset) { @@ -1155,6 +1169,8 @@ static void tumbler_resume(pmac_t *chip) mix->acs &= ~1; mix->master_switch[0] = mix->save_master_switch[0]; mix->master_switch[1] = mix->save_master_switch[1]; + mix->master_vol[0] = mix->save_master_vol[0]; + mix->master_vol[1] = mix->save_master_vol[1]; tumbler_reset_audio(chip); if (mix->i2c.client && mix->i2c.init_client) { if (mix->i2c.init_client(&mix->i2c) < 0) |