| Age | Commit message (Collapse) | Author |
|---|
|
I'm seeing a kernel panic on an ES7000-600 when booting in virtual wire
mode. The panic happens because smp_read_mpc() is passed a physical
address, and it should be virtual. I tested the attached patch on the
ES7000-600 and on a 2 cpu Dell box, and saw no problems on either.
Signed-off-by: Dan Yeisley <dan.yeisley@unisys.com>
Acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
AMD SimNow!'s JIT doesn't like them at all in the guest. For distribution
installation it's easiest if it's a boot time option.
Also I moved the variable to a more appropiate place and make
it independent from sysctl
And marked __read_mostly which it is.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Add some more gitignore files for i386 architecture. This files are
created during the build process of a i386 kernel.
Signed-off-by: Thomas Meyer <thomas@m3y3r.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
This path isn't obvious. It looks as if the kernel will be taking three
args from the user stack, but it only takes one from there.
Signed-off-by: Albert Cahalan <acahalan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Fix x86 oprofile regression introduced by:
commit c34d1b4d165c67b966bca4aba026443d7ff161eb
[PATCH] mm: kill check_user_page_readable
That commit reorganized tests for the userspace stack walking moving all
those tests into dump_backtrace(), however, dump_backtrace() was used for
both userspace and kernel stalk walking. The result is typically no
recorded callgraph information for kernel samples.
Revive the original function as dump_kernel_backtrace() and rename the
other to dump_user_backtrace() to avoid future confusion.
Signed-off-by: Gerald Britton <gbritton@alum.mit.edu>
Apology-from: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
|
|
Lost a few hours debugging an early-bootup fault within printk itself,
which manifested itself as a hard to debug early hang.
This patch makes it much easier by printing out early faults via
early_printk(), which function is a lot simpler than a full printk, and
hence more likely to succeed in emergencies. (We do not recover from early
faults anyway, so there's no loss from not having these messages in the
normal printk buffer.)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
The *at patches introduced fstatat and, due to inusfficient research, I
used the newfstat functions generally as the guideline. The result is that
on 32-bit platforms we don't have all the information needed to implement
fstatat64.
This patch modifies the code to pass up 64-bit information if
__ARCH_WANT_STAT64 is defined. I renamed the syscall entry point to make
this clear. Other archs will continue to use the existing code. On x86-64
the compat code is implemented using a new sys32_ function. this is what
is done for the other stat syscalls as well.
This patch might break some other archs (those which define
__ARCH_WANT_STAT64 and which already wired up the syscall). Yet others
might need changes to accomodate the compatibility mode. I really don't
want to do that work because all this stat handling is a mess (more so in
glibc, but the kernel is also affected). It should be done by the arch
maintainers. I'll provide some stand-alone test shortly. Those who are
eager could compile glibc and run 'make check' (no installation needed).
The patch below has been tested on x86 and x86-64.
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Initialising cpu_possible_map to all-ones with CONFIG_HOTPLUG_CPU means that
a) All for_each_cpu() loops will iterate across all NR_CPUS CPUs, rather
than over possible ones. That can be quite expensive.
b) Soon we'll be allocating per-cpu areas only for possible CPUs. So with
CPU_MASK_ALL, we'll be wasting memory.
I also switched voyager over to not use CPU_MASK_ALL in the non-CPU-hotplug
case. Should be OK..
I note that parisc is also using CPU_MASK_ALL. Suggest that it stop doing
that.
Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Paul Jackson <pj@sgi.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Cc: Zwane Mwaikambo <zwane@linuxpower.ca>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Registers system call for the i386 architecture.
Signed-off-by: Janak Desai <janak@us.ibm.com>
Cc: Al Viro <viro@ftp.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Cc: Andi Kleen <ak@muc.de>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Fix wrong '!' in bad apic fix
I forgot to remove the ! when moving the code from x86-64 to i386 x86-64
tested !disable_apic, but of course for cpu_has_apic it shouldn't be
negated.
Credit goes to Jan Beulich for spotting it with eagle eyes.
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Due to the usage of set_64bit in include/asm-i386/pgtable-3level.h,
HIGHMEM64G must depend on X86_CMPXCHG64.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Show first field of kernel version in register dumps like x86_64 does.
Changes output from e.g.:
(2.6.16-rc1)
to:
(2.6.16-rc1 #12)
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
i386 CPU init code accesses freed init memory when booting a newly-started
processor after CPU hotplug. The cpu_devs array is searched to find the
vendor and it contains pointers to freed data.
Fix that by:
1. Zeroing entries for freed vendor data after bootup.
2. Changing Transmeta, NSC and UMC to all __init[data].
3. Printing a warning (once only) and setting this_cpu
to a safe default when the vendor is not found.
This does not change behavior for AMD systems. They were broken already
but no error was reported.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
dump_stack() on page allocation failure presently has an irritating habit
of shouting just "====" at everyone: please stop it.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
percpu_data blindly allocates bootmem memory to store NR_CPUS instances of
cpudata, instead of allocating memory only for possible cpus.
As a preparation for changing that, we need to convert various 0 -> NR_CPUS
loops to use for_each_cpu().
(The above only applies to users of asm-generic/percpu.h. powerpc has gone it
alone and is presently only allocating memory for present CPUs, so it's
currently corrupting memory).
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Jens Axboe <axboe@suse.de>
Cc: Anton Blanchard <anton@samba.org>
Acked-by: William Irwin <wli@holomorphy.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
not enabled
It's bad juju to touch the APIC when it hasn't been enabled.
I also moved ack_bad_irq for x86-64 out of line following i386.
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Some broken BIOS's had processors disabled, but
same apic id as a valid processor. This causes
acpi_processor_start() to think this disabled
cpu is ok, and croak. So we dont record bad
apicid's anymore.
http://bugzilla.kernel.org/show_bug.cgi?id=5930
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Fix the problem in kernel 2.6.15.1 (and early versions) that OProfile on
x86_64 does not correctly collect the stack traces for kernel functions.
The original code in valid_kernel_stack() in arch/i386/oprofile/backtrace.c
assumes that the frame pointer (headaddr) should be greater than stack
(i.e., regs).
This assumption is wrong for x86_64 because NMIs in x86_64 use a seperate
stack different from the kernel stack. Therefore, the variable stack now
points to some location on the NMI stack, which turns out to be at a higher
address than the frame pointer (headaddr) on the kernel stack. The correct
comparison here should be between headaddr and regs->rsp for x86_64.
Signed-off-by: Tong Li <tong.n.li@intel.com>
Cc: John Levon <levon@movementarian.org>
Cc: Philippe Elie <phil.el@wanadoo.fr>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Test for old_freq equals 0 to insure not to divide by 0:
______________________________________________
Check for not initialized freq on cpufreq changes
Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Dave Jones <davej@redhat.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6
|
|
Enable selection of different user/kernel VM splits for i386, including an
optimized mode for 1GB physical RAM, which gives the kernel a direct (non
HIGHMEM) mapping to the entire 1GB rather than just the first 896MB.
There is a similarly a similarly optimized mode for machines with exactly 2GB
of physical RAM.
This can speed up the kernel by avoiding having to create/destroy temporary
HIGHMEM mappings, and by not having to include HIGHMEM support at all on such
machines. The flip side is that there's less virtual addressing left for
userspace in these alternatives, and some binary-only kernel modules may
misbehave unless rebuilt with the same VMSPLIT option as the main kernel
image.
Original idea/patch from Jens Axboe, modified based on suggestions from Linus
et al.
Signed-off-by: Mark Lord <mlord@pobox.com>
Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Avoid lost tick compensation early in boot before the TSCs are
synchronized. Currently timekeeping is enabled before the TSCs are
synchronized, thus when the TSCs are synched (reset to zero), it appears
that a number of lost ticks have occurred. This can cause premature expiry
of timers and in extreme cases can cause the soft lockup detection to fire.
This resolves issues reported by Andy Whitcroft as well as bug #5366
reported by Tim Mann.
Signed-off-by: John Stultz <johnstul@us.ibm.com>
Acked-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Move CONFIG_DOUBLEFAULT from the main Kconfig menu (!) into its proper
place: the "Processor Type and features" submenu.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Handle more bogus MCFG entries
Some Asus P4 boards seem to have broken MCFG tables with
only a single entry for busses 0-0. Special case these
and assume they mean all busses can be accessed.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
|
|
This patch adds the Intel ICH8 DID's to the irq.c and pci_ids.h files.
Signed-off-by: Jason Gaston <Jason.d.gaston@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
|
|
|
|
|
|
From http://www.intel.com/design/xeon/applnots/24161830.pdf
16MB of 16-way assoc 64 byte per cacheline L3 cache anyone? Yum.
Signed-off-by: Dave Jones <davej@redhat.com>
|
|
Ignore clock frequencies below 2Ghz for CPU's detected with N60 errata bug.
Signed-off-by: Ben Collins <bcollins@ubuntu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dave Jones <davej@redhat.com>
|
|
branches into release
Signed-off-by: Len Brown <len.brown@intel.com>
|
|
This patch fixes the following compile error:
...
CC arch/i386/kernel/cpu/cpufreq/gx-suspmod.o
arch/i386/kernel/cpu/cpufreq/gx-suspmod.c: In function 'gx_detect_chipset':
arch/i386/kernel/cpu/cpufreq/gx-suspmod.c:193: error: implicit declaration of function 'pci_match_id'
arch/i386/kernel/cpu/cpufreq/gx-suspmod.c:193: warning: comparison between pointer and integer
make[3]: *** [arch/i386/kernel/cpu/cpufreq/gx-suspmod.o] Error 1
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Dave Jones <davej@redhat.com>
|
|
This is a subset of the bluesmoke project core code, stripped of the NMI work
which isn't ready to merge and some of the "interesting" proc functionality
that needs reworking or just has no place in kernel. It requires no core
kernel changes except the added scrub functions already posted.
The goal is to merge further functionality only after the core code is
accepted and proven in the base kernel, and only at the point the upstream
extras are really ready to merge.
From: doug thompson <norsk5@xmission.com>
This converts EDAC to sysfs and is the final chunk neccessary before EDAC
has a stable user space API and can be considered for submission into the
base kernel.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: doug thompson <norsk5@xmission.com>
Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Add the sys_pselect6() and sys_poll() calls to the i386 syscall table.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Handle TIF_RESTORE_SIGMASK as added by David Woodhouse's patch entitled:
[PATCH] 2/3 Add TIF_RESTORE_SIGMASK support for arch/powerpc
[PATCH] 3/3 Generic sys_rt_sigsuspend
It does the following:
(1) Declares TIF_RESTORE_SIGMASK for i386.
(2) Invokes it over to do_signal() when TIF_RESTORE_SIGMASK is set.
(3) Makes do_signal() support TIF_RESTORE_SIGMASK, using the signal mask saved
in current->saved_sigmask.
(4) Discards sys_rt_sigsuspend() from the arch, using the generic one instead.
(5) Makes sys_sigsuspend() save the signal mask and set TIF_RESTORE_SIGMASK
rather than attempting to fudge the return registers.
(6) Makes sys_sigsuspend() return -ERESTARTNOHAND rather than looping
intrinsically.
(7) Makes setup_frame(), setup_rt_frame() and handle_signal() return 0 or
-EFAULT rather than true/false to be consistent with the rest of the
kernel.
Due to the fact do_signal() is then only called from one place:
(8) Makes do_signal() no longer have a return value is it was just being
ignored; force_sig() takes care of this.
(9) Discards the old sigmask argument to do_signal() as it's no longer
necessary.
(10) Makes do_signal() static.
(11) Marks the second argument to do_notify_resume() as unused. The unused
argument should remain in the middle as the arguments are passed in as
registers, and the ordering is specific in entry.S
Given the way do_signal() is now no longer called from sys_{,rt_}sigsuspend(),
they no longer need access to the exception frame, and so can just take
arguments normally.
This patch depends on sys_rt_sigsuspend patch.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Wire up the x86 syscalls
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Adds the ability to disability packet split at compile time and use the legacy receive path on PCI express hardware. Made this a CONFIG option and modified the Kconfig, to reflect the new option.
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
|
|
Add __meminit to the __init lineup to ensure functions default
to __init when memory hotplug is not enabled. Replace __devinit
with __meminit on functions that were changed when the memory
hotplug code was introduced.
Signed-off-by: Matt Tolentino <matthew.e.tolentino@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Since we do no longer support any gcc < 3.0, there's no need to check
for it..
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
cpufreq init can be called when a CPU is set online.
Need to make powernow-k8's initialisation functions __cpuinit to
prevents oopses when a CPU is off/onlined on a AMD system
Cc: trenn@suse.de
Cc: mark.langsdorf@amd.com
Cc: davej@redhat.com
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
|
|
Allow gcc4 compilers to optimize unit-at-a-time.
This flag enables gcc to "see" the entire C file before making optimisation
decisions such as inline, which results in gcc making better decisions. One
of the immediate effects of this is that static functions that are used only
once now get inlined.
gcc 3.4 has this flag as well, however gcc 3.x have a problem with inlining
and stacks and as a result, enabling this flag there would cause excessive and
unacceptable stack use. This problem is fixed in the gcc 4.x series. The
x86-64 architecture already enables this feature so it's well tested already.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Seems that the CS5530A chip used in Geode GX1 systems has some crazy feature
that causes SMI traps when accessing the PCI configuration space of the video
device. Various GX1 BIOSes seem to use this 'feature' to hide the real BARs
of the device. This patch disables these traps (in an early PCI fixup) so
that Linux sees the real, physical BARs and not the virtual ones provided by
the BIOS.
This should allow the GX1 framebuffer driver to work on more systems that have
different BIOSes as the driver no longer guesses at what the virtual BARs
mean.
I'm not entirely sure it the correct solution as I can neither test regular
VGA console nor the X's 'cyrix' video driver so there might be some breakage
there -- probably best to get some more testers before applying 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>
|
|
Recent changes caused part of stack traces from SysRq-T to print at
KERN_EMERG loglevel. Also, parts of stack dump during oops were failing to
print at that level when they should.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Move the HOTPLUG_CPU option under "Processor type" instead of under "Bus
options". This makes it the same for i386 as most other processor types
(arm, ia64, parisc, ppc, s390, & x86_64; but not for powerpc). Besides, it
takes me too long to find it under Bus options. I can't be the only person
who has trouble finding it.
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
I tried to send the forcedeth maintainer an email, but it came back with:
"The mail address manfreds@colorfullife.com is not read anymore.
Please resent your mail to manfred@ instead of manfreds@."
This patch fixes this.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
|
|
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
)
From: Al Viro <viro@ftp.linux.org.uk>
task_pt_regs() needs the same offset-by-8 to match copy_thread()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
|
)
From: Ingo Molnar <mingo@elte.hu>
This is the latest version of the scheduler cache-hot-auto-tune patch.
The first problem was that detection time scaled with O(N^2), which is
unacceptable on larger SMP and NUMA systems. To solve this:
- I've added a 'domain distance' function, which is used to cache
measurement results. Each distance is only measured once. This means
that e.g. on NUMA distances of 0, 1 and 2 might be measured, on HT
distances 0 and 1, and on SMP distance 0 is measured. The code walks
the domain tree to determine the distance, so it automatically follows
whatever hierarchy an architecture sets up. This cuts down on the boot
time significantly and removes the O(N^2) limit. The only assumption
is that migration costs can be expressed as a function of domain
distance - this covers the overwhelming majority of existing systems,
and is a good guess even for more assymetric systems.
[ People hacking systems that have assymetries that break this
assumption (e.g. different CPU speeds) should experiment a bit with
the cpu_distance() function. Adding a ->migration_distance factor to
the domain structure would be one possible solution - but lets first
see the problem systems, if they exist at all. Lets not overdesign. ]
Another problem was that only a single cache-size was used for measuring
the cost of migration, and most architectures didnt set that variable
up. Furthermore, a single cache-size does not fit NUMA hierarchies with
L3 caches and does not fit HT setups, where different CPUs will often
have different 'effective cache sizes'. To solve this problem:
- Instead of relying on a single cache-size provided by the platform and
sticking to it, the code now auto-detects the 'effective migration
cost' between two measured CPUs, via iterating through a wide range of
cachesizes. The code searches for the maximum migration cost, which
occurs when the working set of the test-workload falls just below the
'effective cache size'. I.e. real-life optimized search is done for
the maximum migration cost, between two real CPUs.
This, amongst other things, has the positive effect hat if e.g. two
CPUs share a L2/L3 cache, a different (and accurate) migration cost
will be found than between two CPUs on the same system that dont share
any caches.
(The reliable measurement of migration costs is tricky - see the source
for details.)
Furthermore i've added various boot-time options to override/tune
migration behavior.
Firstly, there's a blanket override for autodetection:
migration_cost=1000,2000,3000
will override the depth 0/1/2 values with 1msec/2msec/3msec values.
Secondly, there's a global factor that can be used to increase (or
decrease) the autodetected values:
migration_factor=120
will increase the autodetected values by 20%. This option is useful to
tune things in a workload-dependent way - e.g. if a workload is
cache-insensitive then CPU utilization can be maximized by specifying
migration_factor=0.
I've tested the autodetection code quite extensively on x86, on 3
P3/Xeon/2MB, and the autodetected values look pretty good:
Dual Celeron (128K L2 cache):
---------------------
migration cost matrix (max_cache_size: 131072, cpu: 467 MHz):
---------------------
[00] [01]
[00]: - 1.7(1)
[01]: 1.7(1) -
---------------------
cacheflush times [2]: 0.0 (0) 1.7 (1784008)
---------------------
Here the slow memory subsystem dominates system performance, and even
though caches are small, the migration cost is 1.7 msecs.
Dual HT P4 (512K L2 cache):
---------------------
migration cost matrix (max_cache_size: 524288, cpu: 2379 MHz):
---------------------
[00] [01] [02] [03]
[00]: - 0.4(1) 0.0(0) 0.4(1)
[01]: 0.4(1) - 0.4(1) 0.0(0)
[02]: 0.0(0) 0.4(1) - 0.4(1)
[03]: 0.4(1) 0.0(0) 0.4(1) -
---------------------
cacheflush times [2]: 0.0 (33900) 0.4 (448514)
---------------------
Here it can be seen that there is no migration cost between two HT
siblings (CPU#0/2 and CPU#1/3 are separate physical CPUs). A fast memory
system makes inter-physical-CPU migration pretty cheap: 0.4 msecs.
8-way P3/Xeon [2MB L2 cache]:
---------------------
migration cost matrix (max_cache_size: 2097152, cpu: 700 MHz):
---------------------
[00] [01] [02] [03] [04] [05] [06] [07]
[00]: - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
[01]: 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
[02]: 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
[03]: 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1) 19.2(1)
[04]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1) 19.2(1)
[05]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1) 19.2(1)
[06]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) - 19.2(1)
[07]: 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) -
---------------------
cacheflush times [2]: 0.0 (0) 19.2 (19281756)
---------------------
This one has huge caches and a relatively slow memory subsystem - so the
migration cost is 19 msecs.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Cc: <wilder@us.ibm.com>
Signed-off-by: John Hawkes <hawkes@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
|
