Commit graph

676 commits

Author SHA1 Message Date
Dave Jones
5c0d5d262a [PATCH] missing license tag in intermodule
It may suck something awful, but it shouldn't taint the kernel.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-05 11:06:52 -08:00
Chuck Ebbert
bd576c9523 [PATCH] sched: only print migration_cost once per boot
migration_cost prints after every CPU hotplug event.  Make it print only
once at boot.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-05 11:06:51 -08:00
Eric Dumazet
88a2a4ac6b [PATCH] percpu data: only iterate over possible CPUs
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>
2006-02-05 11:06:51 -08:00
Andrew Morton
514a01b880 [PATCH] uninline __sigqueue_free()
Five callsites.  I dunno how all this crap got back in there :(

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 08:32:10 -08:00
Randy Dunlap
fe85a998ca [PATCH] cpuset: fix sparse warning
kernel/cpuset.c:644:38: warning: non-ANSI function declaration of function 'cpuset_update_task_memory_state'

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Acked-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 08:32:06 -08:00
George Anzinger
88fc3897e3 [PATCH] Normalize timespec for negative values in ns_to_timespec
- In case of a negative nsec value the result of the division must be
  normalized.

- Remove inline from an exported function.

Signed-off-by: George Anzinger <george@wildturkeyranch.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 08:32:06 -08:00
Keith Owens
54e8ce463a [PATCH] Tell kallsyms_lookup_name() to ignore type U entries
When one module exports a function symbol and another module uses that
symbol then kallsyms shows the symbol twice.  Once from the consumer with a
type of 'U' and once from the provider with a type of 't' or 'T'.  On most
architectures, both entries have the same address so it does not matter
which one is returned by kallsyms_lookup_name().  But on architectures with
function descriptors, the 'U' entry points to the descriptor, not to the
code body, which is not what we want.

IA64 # grep -w qla2x00_remove_one /proc/kallsyms
a000000208c25ef8 U qla2x00_remove_one   [qla2300]   <= descriptor
a000000208bf44c0 t qla2x00_remove_one   [qla2xxx]   <= function body

Tell kallsyms_lookup_name() to ignore type U entries in modules.

Signed-off-by: Keith Owens <kaos@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 08:32:02 -08:00
Ananth N Mavinakayanahalli
278ff95370 [PATCH] Kprobes: Fix deadlock in function-return probes
When two function-return probes are inserted on kfree()[1] and the second
on say, sys_link()[2], and later [2] is unregistered, we have a deadlock as
kfree is called with the kretprobe_lock held and the function-return probe
on kfree will also try to grab the same lock.

However, we can move the kfree() during unregistration to outside the
spinlock as we are sure that no instances from the free list will be used
after synchronized_sched() returns during the unregistration process.
Thanks to Masami Hiramatsu for spotting this.

Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 08:32:00 -08:00
Adrian Bunk
e65cefe87b [PATCH] kernel/kprobes.c: fix a warning #ifndef ARCH_SUPPORTS_KRETPROBES
kernel/kprobes.c:353: warning: 'pre_handler_kretprobe' defined but not used

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Acked-by: "Keshavamurthy, Anil S" <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 08:32:00 -08:00
Linus Torvalds
59ed2f59e4 Merge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6 2006-02-01 22:06:15 -08:00
Christoph Lameter
2a11ff06d7 [PATCH] zone_reclaim: configurable off node allocation period.
Currently the zone_reclaim code has a fixed window of 30 seconds of off node
allocations should a local zone have no unused pagecache pages left.  Reclaim
will be attempted again after this timeout period to avoid repeated useless
scans for memory.  This is also useful to established sufficiently large off
node allocation chunks to relieve the local node.

It may be beneficial to adjust that time period for some special situations.
For example if memory use was exceeding node capacity one may want to give up
for longer periods of time.  If memory spikes intermittendly then one may want
to shorten the time period to reduce the number of off node allocations.

This patch allows just that....

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:16 -08:00
Christoph Lameter
c84db23c6e [PATCH] zone_reclaim: minor fixes
- If we only reclaim nr_pages then its okay to stay on node.
  Switch from > to >= for the comparison.

- vm_table[] entry for zone_reclaim_mode is a bit screwed up.

- Add empty lines around shrink_zone to show that this is the
  central function to be called.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:15 -08:00
Rafael J. Wysocki
f7b8988ff5 [PATCH] swsusp: do not change log level during suspend/resume
Prevent the kernel from setting the log level to 10 unconditionally during
suspend/resume which was needed in the past for debugging, but generally is
undesirable.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:14 -08:00
Jack Steiner
2f7016d917 [PATCH] sys_sched_getaffinity() & hotplug
Change sched_getaffinity() so that it returns a bitmap that indicates the
legally schedulable cpus that a task is allowed to run on.

Without this patch, if CONFIG_HOTPLUG_CPU is enabled, sched_getaffinity()
unconditionally returns (at least on IA64) a mask with NR_CPUS bits set.
This conveys no useful infornmation except for a kernel compile option.

This fixes a breakage we obseved running recent kernels. We have MPI jobs
that use sched_getaffinity() to determine where to place their threads.
Placing them on non-existant cpus is problematic :-)

Signed-off-by: Jack Steiner <steiner@sgi.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Nathan Lynch <ntl@pobox.com>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:13 -08:00
Adrian Bunk
493f01d1d0 [PATCH] kernel/posix-timers.c: remove do_posix_clock_notimer_create()
This function is neither used nor has any real contents.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:13 -08:00
Thomas Gleixner
952bbc87f0 [PATCH] hrtimers: set correct initial expiry time for relative SIGEV_NONE timers
The expiry time for relative timers with SIGEV_NONE set was never
updated to the correct value.

Pointed out by George Anzinger.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:13 -08:00
Thomas Gleixner
66188fae3b [PATCH] hrtimers: add back lost credit lines
At some point we added credits to people who actively helped to bring
k/hr-timers along.  This was lost in the big code revamp.  Add it back.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:13 -08:00
George Anzinger
7978672c4d [PATCH] hrtimers: cleanups and simplifications
Clean up the interface to hrtimers by changing the init code to pass the mode
as well as the clock.  This allow the init code to select the correct base and
eliminates extra timer re-init code in posix-timers.  We also simplify the
restart interface nanosleep use.

Signed-off-by: George Anzinger <george@mvista.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:13 -08:00
akpm@osdl.org
ff60a5dc4f [PATCH] hrtimers: fix posix-timer requeue race
From: Steven Rostedtrostedt@goodmis.org <rostedt@goodmis.org>

CPU0 expires a posix-timer and runs the callback function.  The signal is
queued.

After releasing the posix-timer lock and before returning to hrtimer_run_queue
CPU0 gets interrupted.  CPU1 delivers the queued signal and rearms the timer.
CPU0 comes back to hrtimer_run_queue and sets the timer state to expired.

The next modification of the timer can result in an oops, because the state
information is wrong.

Keep track of state = RUNNING and check if the state has been in the return
path of hrtimer_run_queue.  In case the state has been changed, ignore a
restart request and do not touch the state variable.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:13 -08:00
Thomas Gleixner
a16a1c095a [PATCH] hrtimers: fix oldvalue return in setitimer
This resolves bugzilla bug#5617.  The oldvalue of the timer was read after the
timer was cancelled, so the remaining time was always zero.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:12 -08:00
Thomas Gleixner
b6557fbca8 [PATCH] hrtimers: fix possible use of NULL pointer in posix-timers
Fixup the conversion of posix-timers to hrtimers.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:12 -08:00
Thomas Gleixner
bc1978d404 [PATCH] hrtimers: fixup itimer conversion
The itimer conversion removed the locking which protects the timer and
variables in the shared signal structure.  Steven Rostedt found the problem in
the latest -rt patches.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:12 -08:00
Rafael J. Wysocki
853609b61e [PATCH] swsusp: use bytes as image size units
Make swsusp use bytes as the image size units, which is needed for future
compatibility.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01 08:53:12 -08:00
Andrew Morton
3fa97c9db4 [PATCH] "Fix uidhash_lock <-> RXU deadlock" fix
I get storms of warnings from local_bh_enable().  Better-tested patches,
please.

Cc: Ingo Molnar <mingo@elte.hu>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-31 16:49:43 -08:00
Ingo Molnar
adac166523 [PATCH] rcu_torture_lock deadlock fix
rcu_torture_lock is used in a softirq-unsafe manner, but it is also
taken by rcu_torture_cb(), which may execute in softirq-context,
resulting in potential deadlocks.

The fix is to acquire rcu_torture_lock in a softirq-safe manner.  With
this fix applied, the rcu-torture code passes validation.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Paul E. McKenney <paulmck@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-31 11:30:18 -08:00
Ingo Molnar
4021cb279a [PATCH] fix uidhash_lock <-> RCU deadlock
RCU task-struct freeing can call free_uid(), which is taking
uidhash_lock - while other users of uidhash_lock are softirq-unsafe.

The fix is to always take the uidhash_spinlock in a softirq-safe manner.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Paul E. McKenney <paulmck@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-31 11:30:18 -08:00
Ingo Molnar
70b4d63e98 [PATCH] Fix boot-time slowdown for measure_migration_cost
This reduces the amount of time the migration cost calculations cost
during bootup. Based on numbers by Tony Luck <tony.luck@intel.com>.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2006-01-31 10:23:31 -08:00
Linus Torvalds
951069e311 Don't try to "validate" a non-existing timeval.
settime() with a NULL timeval is silly but legal.

Noticed by Dave Jones <davej@redhat.com>

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-31 10:16:55 -08:00
Len Brown
9fdb62af92 [ACPI] merge 3549 4320 4485 4588 4980 5483 5651 acpica asus fops pnpacpi branches into release
Signed-off-by: Len Brown <len.brown@intel.com>
2006-01-24 17:52:48 -05:00
Alan Cox
715b49ef2d [PATCH] EDAC: atomic scrub operations
EDAC requires a way to scrub memory if an ECC error is found and the chipset
does not do the work automatically.  That means rewriting memory locations
atomically with respect to all CPUs _and_ bus masters.  That means we can't
use atomic_add(foo, 0) as it gets optimised for non-SMP

This adds a function to include/asm-foo/atomic.h for the platforms currently
supported which implements a scrub of a mapped block.

It also adjusts a few other files include order where atomic.h is included
before types.h as this now causes an error as atomic_scrub uses u32.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-18 19:20:30 -08:00
David Woodhouse
150256d8aa [PATCH] Generic sys_rt_sigsuspend()
The TIF_RESTORE_SIGMASK flag allows us to have a generic implementation of
sys_rt_sigsuspend() instead of duplicating it for each architecture.  This
provides such an implementation and makes arch/powerpc use it.

It also tidies up the ppc32 sys_sigsuspend() to use TIF_RESTORE_SIGMASK.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-18 19:20:29 -08:00
Jason Baron
c21761f168 [PATCH] fix sched_setscheduler semantics
Currently, a negative policy argument passed into the
'sys_sched_setscheduler()' system call, will return with success.  However,
the manpage for 'sys_sched_setscheduler' says:

EINVAL The scheduling policy is not one of the recognized policies, or the
              parameter p does not make sense for the policy.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-18 19:20:22 -08:00
Christoph Lameter
1743660b91 [PATCH] Zone reclaim: proc override
proc support for zone reclaim

This patch creates a proc entry /proc/sys/vm/zone_reclaim_mode that may be
used to override the automatic determination of the zone reclaim made on
bootup.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-18 19:20:17 -08:00
Ingo Molnar
ea13dbc89c [PATCH] kernel/hrtimer.c sparse warning fix
fix the following sparse warning:

 kernel/hrtimer.c:665:34: warning: incorrect type in argument 2 (different address spaces)
 kernel/hrtimer.c:665:34:    expected void const *from
 kernel/hrtimer.c:665:34:    got struct timespec [noderef] *<noident><asn:1>
 kernel/hrtimer.c:664:2: warning: dereference of noderef expression

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-16 23:21:12 -08:00
Adrian Bunk
fd279197b1 [PATCH] build kernel/intermodule.c only when required
Build kernel/intermodule.c only when required.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-16 23:15:26 -08:00
Jonathan Corbet
8dca6f33f0 [PATCH] hrtimer comment tweak
Fix a comment which missed an update cycle somewhere.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-16 20:27:03 -08:00
Linus Torvalds
3f02d072d4 Merge git://git.kernel.org/pub/scm/linux/kernel/git/bunk/trivial 2006-01-15 16:43:29 -08:00
Paul Jackson
505970b96e [PATCH] cpuset oom lock fix
The problem, reported in:

  http://bugzilla.kernel.org/show_bug.cgi?id=5859

and by various other email messages and lkml posts is that the cpuset hook
in the oom (out of memory) code can try to take a cpuset semaphore while
holding the tasklist_lock (a spinlock).

One must not sleep while holding a spinlock.

The fix seems easy enough - move the cpuset semaphore region outside the
tasklist_lock region.

This required a few lines of mechanism to implement.  The oom code where
the locking needs to be changed does not have access to the cpuset locks,
which are internal to kernel/cpuset.c only.  So I provided a couple more
cpuset interface routines, available to the rest of the kernel, which
simple take and drop the lock needed here (cpusets callback_sem).

Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-14 18:27:10 -08:00
Martin Schwidefsky
0152fb3760 [PATCH] s390: spinlock fixes
Remove useless spin_retry_counter and fix compilation for UP kernels.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-14 18:27:09 -08:00
Arjan van de Ven
858119e159 [PATCH] Unlinline a bunch of other functions
Remove the "inline" keyword from a bunch of big functions in the kernel with
the goal of shrinking it by 30kb to 40kb

Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-14 18:27:06 -08:00
Ingo Molnar
b0a9499c3d [PATCH] sched: add new SCHED_BATCH policy
Add a new SCHED_BATCH (3) scheduling policy: such tasks are presumed
CPU-intensive, and will acquire a constant +5 priority level penalty.  Such
policy is nice for workloads that are non-interactive, but which do not
want to give up their nice levels.  The policy is also useful for workloads
that want a deterministic scheduling policy without interactivity causing
extra preemptions (between that workload's tasks).

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-14 18:25:20 -08:00
Christian Kujau
624dffcbcf correct email address of Manfred Spraul
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>
2006-01-15 02:43:54 +01:00
Adrian Bunk
750c902ef4 SOFTWARE_SUSPEND: fix a typo in the dependencies
This patch fixes a typo in the dependencies of SOFTWARE_SUSPEND.

This patch is based on a report by
Jean-Luc Leger <reiga@dspnet.fr.eu.org>.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Pavel Machek <pavel@ucw.cz>
2006-01-15 02:01:39 +01:00
Linus Torvalds
661dd5c840 Merge master.kernel.org:/pub/scm/linux/kernel/git/tglx/hrtimer-2.6 2006-01-12 10:22:11 -08:00
akpm@osdl.org
d7102e95b7 [PATCH] sched: filter affine wakeups
)

From: Nick Piggin <nickpiggin@yahoo.com.au>

Track the last waker CPU, and only consider wakeup-balancing if there's a
match between current waker CPU and the previous waker CPU.  This ensures
that there is some correlation between two subsequent wakeup events before
we move the task.  Should help random-wakeup workloads on large SMP
systems, by reducing the migration attempts by a factor of nr_cpus.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-12 09:08:50 -08:00
akpm@osdl.org
198e2f1811 [PATCH] scheduler cache-hot-autodetect
)

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>
2006-01-12 09:08:50 -08:00
Thomas Gleixner
c9db4fa115 [hrtimer] Enforce resolution as lower limit of intervals
Roman Zippel pointed out that the missing lower limit of intervals
leads to an accounting error in the overrun count. Enforce the lower
limit of intervals to resolution in the timer forwarding code.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2006-01-12 11:47:34 +01:00
Thomas Gleixner
e2787630c1 [hrtimer] Change resolution storage to ktime_t format
Change the storage format of the per base resolution to ktime_t to
make it easier accessible in the hrtimers code.

Change the resolution from (NSEC_PER_SEC/HZ) to TICK_NSEC as Roman
pointed out. TICK_NSEC is closer to the real resolution.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2006-01-12 11:36:14 +01:00
Thomas Gleixner
288867ec5c [hrtimer] Remove listhead from hrtimer struct
The list_head in the hrtimer structure was introduced for easy access
to the first timer with the further extensions of real high resolution
timers in mind, but it turned out in the course of development that
it is not necessary for the standard use case. Remove the list head
and access the first expiry timer by a datafield in the timer base.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2006-01-12 11:25:54 +01:00
Ravikiran G Thirumalai
5fd63b3085 [PATCH] x86_64: Inclusion of ScaleMP vSMP architecture patches - vsmp_align
vSMP specific alignment patch to
1. Define INTERNODE_CACHE_SHIFT for vSMP
2. Use this for alignment of critical structures
3. Use INTERNODE_CACHE_SHIFT for ARCH_MIN_TASKALIGN,
   and let the slab align task_struct allocations to the internode cacheline size
4. Introduce and use ARCH_MIN_MMSTRUCT_ALIGN for mm_struct slab allocations.

Signed-off-by: Ravikiran Thirumalai <kiran@scalemp.com>
Signed-off-by: Shai Fultheim <shai@scalemp.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-11 19:05:01 -08:00