aboutsummaryrefslogtreecommitdiff
path: root/drivers
AgeCommit message (Collapse)Author
2006-06-01Merge branch 'merge'Paul Mackerras
2006-05-31[PATCH] md: Fix badness in sysfs_notify caused by md_new_eventNeilBrown
From: NeilBrown <neilb@suse.de> If an error is reported by a drive in a RAID array (which is done via bi_end_io - in interrupt context), we call md_error and md_new_event which calls sysfs_notify. However sysfs_notify grabs a mutex and so cannot be called in interrupt context. This patch just creates a variant of md_new_event which avoids the sysfs call, and uses that. A better fix for later is to arrange for the event to be called from user-context. Note: avoiding the sysfs call isn't a problem as an error will not, by itself, modify the sync_action attribute. (We do still need to wake_up(&md_event_waiters) as an error by itself will modify /proc/mdstat). Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-31[PATCH] sgiioc4: use mmio ops instead of port ioJeremy Higdon
From: Jeremy Higdon <jeremy@sgi.com> This patch fixes a bug in sgiioc4 where it was using the default IDE port I/O operations instead of MMIO. The IDE part of the IOC4 chip uses MMIO to map the chip registers. Unfortunately, the sgiioc4 driver uses the default port IO operations, which happens to have worked for the past few years. That's about to change, however, thus this change from inX/outX to readX/writeX. Signed-off-by: Jeremy Higdon <jeremy@sgi.com> Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-31[PATCH] maxinefb: Fix compilation errorMartin Michlmayr
From: Martin Michlmayr <tbm@cyrius.com> Fix the following compilation error: CC drivers/video/maxinefb.o drivers/video/maxinefb.c:58: warning: initializer-string for array of chars is too long drivers/video/maxinefb.c:58: warning: (near initialization for \u2018maxinefb_fix.id\u2019) drivers/video/maxinefb.c:110: error: unknown field \u2018fb_get_fix\u2019 specified in initializer drivers/video/maxinefb.c:110: error: \u2018gen_get_fix\u2019 undeclared here (not in a function) drivers/video/maxinefb.c:111: error: unknown field \u2018fb_get_var\u2019 specified in initializer drivers/video/maxinefb.c:111: error: \u2018gen_get_var\u2019 undeclared here (not in a function) make[2]: *** [drivers/video/maxinefb.o] Error 1 Signed-off-by: Martin Michlmayr <tbm@cyrius.com> Signed-off-by: Antonino Daplas <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-31[PATCH] au1100fb: Fix compilationRodolfo Giometti
From: Rodolfo Giometti <giometti@linux.it> Fix the following warning on compilation: drivers/video/au1100fb.c: In function `au1100fb_fb_setcolreg': drivers/video/au1100fb.c:219: warning: ISO C90 forbids mixed declarations and code drivers/video/au1100fb.c: In function `au1100fb_fb_pan_display': drivers/video/au1100fb.c:321: warning: ISO C90 forbids mixed declarations and code drivers/video/au1100fb.c: In function `au1100fb_fb_mmap': drivers/video/au1100fb.c:387: warning: ISO C90 forbids mixed declarations and code drivers/video/au1100fb.c: In function `au1100fb_drv_probe': drivers/video/au1100fb.c:471: warning: unsigned int format, long unsigned int arg (arg 2) drivers/video/au1100fb.c: At top level: drivers/video/au1100fb.c:617: warning: initialization from incompatible pointer type drivers/video/au1100fb.c:618: warning: initialization from incompatible pointer type Signed-off-by: Rodolfo Giometti <giometti@linux.it> Signed-off-by: Antonino Daplas <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-31[PATCH] revert "swsusp add check for suspension of X controlled devices"Andrew Morton
From: Andrew Morton <akpm@osdl.org> Revert commit ff4da2e262d2509fe1bacff70dd00934be569c66. It broke APM suspend, probably because APM doesn't switch back to a VT when suspending. Tracked down by Matt Mackall <mpm@selenic.com> Rafael sayeth: "It only fixed the theoretical issue that a quick-handed user could switch to X after processes have been frozen and before the devices are suspended. With the current userland suspend tools it shouldn't be necessary." Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-31[PATCH] IPMI: reserve I/O ports separatelyCorey Minyard
From: Corey Minyard <minyard@acm.org> This patch is pretty important to get in for IPMI, new systems have been changing the way ACPI and IPMI interact, and this works around the problems for now. This is a temporary fix until we get proper ACPI handling in IPMI. Fixed releasing already-allocated regions when a later request fails, and forward-ported it to HEAD. Some BIOSes reserve disjoint I/O regions in their ACPI tables for the IPMI controller. This causes problems when trying to register the entire I/O region. Therefore we must register each I/O port separately. Signed-off-by: Jordan Hargrave <Jordan_Hargrave@dell.com> Signed-off-by: Matt Domsch <Matt_Domsch@dell.com> Signed-off-by: Corey Minyard <minyard@acm.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-31[PATCH] tpm: more bios log parsing fixesSeiji Munetoh
From: Seiji Munetoh <seiji.munetoh@gmail.com> Change the binary output format to actual ACPI TCPA log structure since the current format does not contain all event-data information that need to verify the PCRs in TPM. tpm_binary_bios_measurements_show() uses get_event_name() to convert the binary event-data to ascii format, and puts them as binary. However, to verify the PCRs, the event-data must be a actual binary event-data used by SHA1 calc. in BIOS. So, I think actual ACPI TCPA log is good for this binary output format. That way, any userland tools easily parse this data with reference to TCG PC specification. Signed-off-by: Seiji Munetoh <seiji.munetoh@gmail.com> Acked-by: Kylene Hall <kjhall@us.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-31[PATCH] tpm: bios log parsing fixesSeiji Munetoh
From: Seiji Munetoh <seiji.munetoh@gmail.com> Fix "tcpa_pc_event" misalignment between enum, strings and TCG PC spec and output of the event which contains a hash data. Signed-off-by: Seiji Munetoh <seiji.munetoh@gmail.com> Acked-by: Kylene Hall <kjhall@us.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-31[PATCH] fbcon: fix scrollback with logo issue immediately after bootDavid Hollister
From: David Hollister <david.hollister@amd.com> After the system boots with the logo, if the first action is a scrollback, the screen may become garbled. This patch ensures that the softback_curr value is updated along with softback_in following the scrollback. Signed-off-by: David Hollister <david.hollister@amd.com> Signed-off-by: Jordan Crouse <jordan.crouse@amd.com> Cc: "Antonino A. Daplas" <adaplas@gmail.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-30Merge master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgartLinus Torvalds
* master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgart: [AGPGART] VIA PT880 Ultra support. [AGPGART] Fix Nforce3 suspend on amd64. [AGPGART] Enable SIS AGP driver on x86-64 for EM64T systems
2006-05-29Input: psmouse - DMI updates for lifebook protocolKenan Esau
Added different lifebook-versions and the CF-18 to the corresponding dmi-table. Signed-off-by: Kenan Esau <kenan.esau@conan.de> Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
2006-05-29Input: change from numbered to named switchesRichard Purdie
Remove the numbered SW_* entries from the input system and assign names to the existing users. Signed-off-by: Richard Purdie <rpurdie@rpsys.net> Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
2006-05-29Input: alps - fix old protocol decodingYotam Medini
Correct touchpad left & right keys assignments for ALPS_OLDPROTO that were swapped. Old protocol is used on UMAX ActionBook-530T notebook. Signed-off-by: Yotam Medini <yotam.medini@gmail.com> Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
2006-05-29Input: wistron - add support for AOpen Barebook 1559asmasc@theaterzentrum.at
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
2006-05-29Input: psmouse - add detection of Logitech TrackMan Wheel trackballZbigniew Luszpinski
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
2006-05-29Input: sidewinder - fix memory leakJesper Juhl
In sw_connect we leak 'buf' and 'idbuf' when we do not leave via one of the fail* labels. This was spotted by the coverity checker. Patch is compile tested only due to lack of hardware. Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com> Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
2006-05-28Merge branch 'upstream-fixes' of ↵Linus Torvalds
master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev * 'upstream-fixes' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev: [PATCH] the latest consensus libata resume fix
2006-05-28[PATCH] the latest consensus libata resume fixMark Lord
Okay, just to sum things up. This forces libata to wait for up to 2 seconds for BUSY|DRQ to clear on resume before continuing. [jgarzik adds...] During testing we never saw DRQ asserted, but nonetheless (a) this works and (b) testing for DRQ won't hurt. Signed-off-by: Mark Lord <liml@rtr.ca> Acked-by: Jens Axboe <axboe@suse.de> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-27Merge branch 'merge' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc * 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc: [PATCH] powerpc: fix RTC/NVRAM accesses on Maple [PATCH] ppc32 CPM_UART: various fixes for pq2 uart users [PATCH] powerpc: linuxppc64.org no more
2006-05-26[netdrvr s/390] trim trailing whitespaceJeff Garzik
Previous fix patches added a bunch of trailing whitespace, which git-applymbox complained loudly about.
2006-05-26[PATCH] s390: lcs driver bug fixes and improvements [2/2]Klaus Wacker
This is the second lcs driver patch containing the rest of lcs fixes. Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-26[PATCH] s390: lcs driver bug fixes and improvements [1/2]Klaus Wacker
Several problems occured with lcs device driver: - device not operational anymore after cable pull/plug-in. - unpredictable results occured, e.g. kernel panic using cards of type QD8F. - STOPLAN and delete multicast address command were not proper recognized by OSA card under heavy network workload. - channel/device error checks missing in interrupt handler. To fix all problems at once recovery of lcs devices has been improved. missing error checks in lcs interrupt handler has been added. Once a hardware problem occurs lcs will recover the device now properly. Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-26[PATCH] s390: qeth driver fixesUrsula Braun
From: Frank Blaschka <Frank.Blaschka@de.ibm.com> From: Frank Pavlic <fpavlic@de.ibm.com> - fix fake_ll during initial device bringup. fake_ll was not active after first start of the device. Problem only occured when qeth was built without IPV6 support. - avoid skb usage after invocation of qeth_flush_buffers, because skb might already be freed. - remove yet another useless netif_wake_queue in qeth_softsetup_ipv6 since this function is only called when device is going online. In this case card->state will never be in state UP. So let the net_device queue down . Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-26[PATCH] s390: qeth driver fixesUrsula Braun
From: Frank Pavlic <fpavlic@de.ibm.com> - correct checking of sscanf-%n value in qeth_string_to_ipaddr(). - don't use netif_stop_queue outside the hard_start_xmit routine. Rather use netif_tx_disable. - don't call qeth_netdev_init on a recovery. Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-26[PATCH] s390: minor fix in cu3088Cornelia Huck
In case of a parse error for the cu3088 group attribute, return -EINVAL instead of count. Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-26[PATCH] pcnet32: remove incorrect pcnet32_free_ringDon Fry
During a code scan for another change I discovered that this call to pcnet32_free_ring must be removed. If the open fails due to a lack of memory all the ring structures are removed via the call to free_ring and a subsequent call to open will dereference a null pointer in pcnet32_init_ring. Please apply to 2.6.17. Signed-off-by: Don Fry <brazilnut@us.ibm.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-26e1000: add shutdown handler back to fix WOLAuke Kok
Someone was waaay too aggressive and removed e1000's reboot notifier instead of porting it to the new way of the shutdown handler. This change broke wake on lan. Add the shutdown handler back in using the same method as e100 uses. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> (cherry picked from c653e6351e371b33b29871e5eedf610ffb3be037 commit)
2006-05-26Merge branch 'upstream-fixes' of ↵Jeff Garzik
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6 into upstream-fixes
2006-05-26Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6Linus Torvalds
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: [NET]: dev.c comment fixes [IPV6] ROUTE: Don't try less preferred routes for on-link routes. [IRDA]: *_DONGLE should depend on IRTTY_SIR [MAINTAINERS]: Add entry for netem
2006-05-26[PATCH] arlan: fix section mismatch warningsRandy Dunlap
Fix section mismatch warnings: WARNING: drivers/net/wireless/arlan.o - Section mismatch: reference to .init.text:arlan_probe from .text between 'init_module' (at offset 0x3526) and 'cleanup_module' WARNING: drivers/net/wireless/arlan.o - Section mismatch: reference to .init.text:init_arlan_proc from .text between 'init_module' (at offset 0x3539) and 'cleanup_module' WARNING: drivers/net/wireless/arlan.o - Section mismatch: reference to .exit.text:cleanup_arlan_proc from .text between 'cleanup_module' (at offset 0x356c) and 'arlan_diagnostic_info_string' Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-05-26[PATCH] wavelan: fix section mismatchRandy Dunlap
Fix section mismatch warning: WARNING: drivers/net/wireless/wavelan.o - Section mismatch: reference to .init.text: from .text between 'init_module' (at offset 0x371e) and 'cleanup_module' Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2006-05-26[PATCH] tpm: fix bug for TPM on ThinkPad T60 and Z60Kylene Jo Hall
The TPM chip on the ThinkPad T60 and Z60 machines is returning 0xFFFF for the vendor ID which is a check the driver made to double check it was actually talking to the memory mapped space of a TPM. This patch removes the check since it isn't absolutely necessary and was causing device discovery to fail on these machines. Signed-off-by: Kylene Hall <kjhall@us.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-26[PATCH] scx200_acb: fix section mismatch warningRandy Dunlap
WARNING: drivers/i2c/busses/scx200_acb.o - Section mismatch: reference to .init.text: from .text after 'scx200_add_cs553x' (at offset 0x528) Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-26[PATCH] s3c24xx: fix spi driver with CONFIG_PMBen Dooks
Fix compile bug with the S3C24XX SPI driver when CONFIG_PM is set. Signed-off-by: Ben Dooks <ben-linux@fluff.org> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-26[PATCH] ads7846 conversion accuracyDavid Brownell
This improves accuracy of the touchscreen and hwmon sensor readings, addressing an issue noted by Imre Deak: there's an extra bit written before the sample (12 bits) gets written out. It also catches up to various comments, and makes the /proc/interrupts entry sensible again. Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Cc: Imre Deak <imre.deak@nokia.com> Cc: Jean Delvare <khali@linux-fr.org> Cc: Dmitry Torokhov <dtor_core@ameritech.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-26[PATCH] Unlock md devices when stopping them on reboot.Neil Brown
otherwise we get nasty messages about locks not being released. Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-26Merge branch 'for-linus' of ↵Linus Torvalds
master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband * 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband: IB/mthca: Fix posting lists of 256 receive requests to SRQ for Tavor
2006-05-26[PATCH] ppc32 CPM_UART: various fixes for pq2 uart usersVitaly Bordug
This fixes various odd things that missed update together with cpm_uart platform_device move. Unified resources names, restructurisation, etc. Also, addressed issue with recent phys/virt translation rework. Being cache-coherent, CPM2's do alloc_bootmem() for the console stuff, and it was used to treat console buffer descriptor mapping 1:1 (as in CPM1 case), which is definitely wrong. Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-05-25[IRDA]: *_DONGLE should depend on IRTTY_SIRSamuel Ortiz
If a SIR dongle is built in the kernel while IRTTY_SIR is built as a module, kernel compilation will fail. Thus, the SIR dongle config should depend on the IRTTY_SIR. Closes kernel bug# 6512 (http://bugzilla.kernel.org/show_bug.cgi?id=6512) Signed-off-by: Samuel Ortiz <samuel@sortiz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-05-24[PATCH] V4L/DVB (4045): Fixes recursive dependency for I2CJean Delvare
Mixing "depends on I2C" and "select I2C" within the media subsystem leads to the following problem: Warning! Found recursive dependency: I2C DVB_BUDGET DVB_BUDGET_PATCH DVB_AV7110 VIDEO_SAA7146_VV VIDEO_SAA7146 I2C Signed-off-by: Jean Delvare <khali@linux-fr.org> Acked-by: Manu Abraham <manu@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-24IB/mthca: Fix posting lists of 256 receive requests to SRQ for TavorMichael S. Tsirkin
If we post a list of length exactly a multiple of 256, nreq in doorbell gets set to 256 which is wrong: it should be encoded by 0. This is because we only zero it out on the next WR, which may not be there. The solution is to ring the doorbell after posting a WQE, not before posting the next one. This is the same bug that we just fixed for QPs with non-shared RQ. Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2006-05-24Merge branch 'for-linus' of ↵Linus Torvalds
master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband * 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband: IB/ipath: deref correct pointer when using kernel SMA IB/ipath: fix null deref during rdma ops IB/ipath: register as IB device owner IB/ipath: enable PE800 receive interrupts on user ports IB/ipath: enable GPIO interrupt on HT-460 IB/ipath: fix NULL dereference during cleanup IB/ipath: replace uses of LIST_POISON IB/ipath: fix reporting of driver version to userspace IB/ipath: don't modify QP if changes fail IB/ipath: fix spinlock recursion bug
2006-05-24Merge branch 'upstream-linus' of ↵Linus Torvalds
master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev * 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev: [PATCH] libata: add pio flush for via atapi (was: Re: TR: ASUS A8V Deluxe, x86_64)
2006-05-24Merge branch 'upstream-linus' of ↵Linus Torvalds
master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6 * 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6: [PATCH] sky2: fix jumbo packet support
2006-05-24[MMC] Fix premature use of md->diskPierre Ossman
md->disk was being used in a debug message before it was allocated. Signed-off-by: Pierre Ossman <drzeus@drzeus.cx> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-24[PATCH] libata: add pio flush for via atapi (was: Re: TR: ASUS A8V Deluxe, ↵Albert Lee
x86_64) Backport the "pio flush" from the libata major update to 2.6.17 for via atapi. Signed-off-by: Albert Lee <albertcc@tw.ibm.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-24[PATCH] sky2: fix jumbo packet supportStephen Hemminger
The truncate threshold calculation to prevent receiver from getting stuck was incorrect, and it didn't take into account the upper limit on bits in the register so the jumbo packet support was broken. Signed-off-by: Stephen Hemminger <shemminger@osdl.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-05-23IB/ipath: deref correct pointer when using kernel SMABryan O'Sullivan
At this point, the core QP structure hasn't been initialized, so what's in there isn't valid. Get the same information elsewhere. Signed-off-by: Bryan O'Sullivan <bos@pathscale.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2006-05-23IB/ipath: fix null deref during rdma opsBryan O'Sullivan
The problem was that node A's sending thread, which handles sending RDMA read response data, would write the trigger word, the last packet would be sent, node B would send a new RDMA read request, node A's interrupt handler would initialize s_rdma_sge, then node A's sending thread would update s_rdma_sge. This didn't happen very often naturally but was more frequent with 1 byte RDMA reads. Rather than adding more locking or increasing the QP structure size and copying sge data, I modified the copy routine to update the pointers before writing the trigger word to avoid the update race. Signed-off-by: Ralph Campbell <ralphc@pathscale.com> Signed-off-by: Bryan O'Sullivan <bos@pathscale.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>