When re-computing the shares for each task group's cpu
representation we need the ratio of weight on each cpu vs the
total weight of the sched domain.
Since load-balancing is loosely (read not) synchronized, the
weight of individual cpus can change between doing the sum and
calculating the ratio.
The previous patch dealt with only one of the race scenarios,
this patch side steps them all by saving a snapshot of all the
individual cpu weights, thereby always working on a consistent
set.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: torvalds@linux-foundation.org
Cc: jes@sgi.com
Cc: jens.axboe@oracle.com
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <1251371336.18584.77.camel@twins>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Patch a5004278f0 (sched: Fix
cgroup smp fairness) introduced the possibility of a
divide-by-zero because load-balancing is not synchronized
between sched_domains.
This can cause the state of cpus to change between the first
and second loop over the sched domain in tg_shares_up().
Reported-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Jes Sorensen <jes@sgi.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <1250855934.7538.30.camel@twins>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The field stack_canary is only used with CC_STACKPROTECTOR.
This patch reduces task_struct size without CC_STACKPROTECTOR.
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
LKML-Reference: <4A8A44CA.2020701@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The PREEMPT_ACTIVE setting doesn't actually need to be
arch-specific, so set up a sane default for all arches to
(hopefully) migrate to.
> if we look at linux/hardirq.h, it makes this claim:
> * - bit 28 is the PREEMPT_ACTIVE flag
> if that's true, then why are we letting any arch set this define ? a
> quick survey shows that half the arches (11) are using 0x10000000 (bit
> 28) while the other half (10) are using 0x4000000 (bit 26). and then
> there is the ia64 oddity which uses bit 30. the exact value here
> shouldnt really matter across arches though should it ?
actually alpha, arm and avr32 also use bit 30 (0x40000000),
there are only five (or eight, depending on how you count)
architectures (blackfin, h8300, m68k, s390 and sparc) using bit
26.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This build bug:
In file included from kernel/sched.c:1765:
kernel/sched_rt.c: In function ‘has_pushable_tasks’:
kernel/sched_rt.c:1069: error: ‘struct rt_rq’ has no member named ‘pushable_tasks’
kernel/sched_rt.c: In function ‘pick_next_task_rt’:
kernel/sched_rt.c:1084: error: ‘struct rq’ has no member named ‘post_schedule’
Triggers because both pushable_tasks and post_schedule are
SMP-only fields.
Move pushable_tasks() to the SMP section and #ifdef the post_schedule use.
Cc: Gregory Haskins <ghaskins@novell.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090729150422.17691.55590.stgit@dev.haskins.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Like sched_migrate_task(), set_cpus_allowed_ptr() should hold
onto the migration thread too.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
A frequent mistake appears to be to call task_of() on a
scheduler entity that is not actually a task, which can result
in a wild pointer.
Add a check to catch these mistakes.
Suggested-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Reflect "active" cpus in the rq->rd->online field, instead of
the online_map.
The motivation is that things that use the root-domain code
(such as cpupri) only care about cpus classified as "active"
anyway. By synchronizing the root-domain state with the active
map, we allow several optimizations.
For instance, we can remove an extra cpumask_and from the
scheduler hotpath by utilizing rq->rd->online (since it is now
a cached version of cpu_active_map & rq->rd->span).
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090730145723.25226.24493.stgit@dev.haskins.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We currently have an explicit "needs_post" vtable method which
returns a stack variable for whether we should later run
post-schedule. This leads to an awkward exchange of the
variable as it bubbles back up out of the context switch. Peter
Zijlstra observed that this information could be stored in the
run-queue itself instead of handled on the stack.
Therefore, we revert to the method of having context_switch
return void, and update an internal rq->post_schedule variable
when we require further processing.
In addition, we fix a race condition where we try to access
current->sched_class without holding the rq->lock. This is
technically racy, as the sched-class could change out from
under us. Instead, we reference the per-rq post_schedule
variable with the runqueue unlocked, but with preemption
disabled to see if we need to reacquire the rq->lock.
Finally, we clean the code up slightly by removing the #ifdef
CONFIG_SMP conditionals from the schedule() call, and implement
some inline helper functions instead.
This patch passes checkpatch, and rt-migrate.
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090729150422.17691.55590.stgit@dev.haskins.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We need to add the new prio to the cpupri accounting before
removing the old prio. This is because removing the old prio
first will open a race window where the cpu will be removed
from pri_active. In this case the cpu will not be visible for
RT push and pulls. This could cause a RT task to not migrate
appropriately, and create a very large latency.
This bug was found with the use of ftrace sched events and
trace_printk.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090729042526.438281019@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The current method for pushing RT tasks after scheduling only
happens after a context switch. But we found cases where a task
is set up on a run queue to be pushed but the push never
happens because the schedule chooses the same task.
This bug was found with the help of Gregory Haskins and the use
of ftrace (trace_printk). It tooks several days for both of us
analyzing the code and the trace output to find this.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090729042526.205923666@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When cgroup group scheduling is built in, skip some code paths
if we don't have any (but the root) cgroups configured.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit ec4e0e2fe0 ("fix
inconsistency when redistribute per-cpu tg->cfs_rq shares")
broke cgroup smp fairness.
In order to avoid starvation of newly placed tasks, we never
quite set the share of an empty cpu group-task to 0, but
instead we set it as if there's a single NICE-0 task present.
If however we actually set this in cfs_rq[cpu]->shares, that
means the total shares for that group will be slightly inflated
every time we balance, causing the observed unfairness.
Fix this by setting cfs_rq[cpu]->shares to 0 but actually
setting the effective weight of the related se to the inflated
number.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1248696557.6987.1615.camel@twins>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Background:
Several race conditions in the scheduler have cropped up
recently, which Steven and I have tracked down using ftrace.
The most recent one turns out to be a race in how the scheduler
determines a suitable migration target for RT tasks, introduced
recently with commit:
commit 68e74568fb
Date: Tue Nov 25 02:35:13 2008 +1030
sched: convert struct cpupri_vec cpumask_var_t.
The original design of cpupri allowed lockless readers to
quickly determine a best-estimate target. Races between the
pri_active bitmap and the vec->mask were handled in the
original code because we would detect and return "0" when this
occured. The design was predicated on the *effective*
atomicity (*) of caching the result of cpus_and() between the
cpus_allowed and the vec->mask.
Commit 68e74568 changed the behavior such that vec->mask is
accessed multiple times. This introduces a subtle race, the
result of which means we can have a result that returns "1",
but with an empty bitmap.
*) yes, we know cpus_and() is not a locked operator across the
entire composite array, but it is implicitly atomic on a
per-word basis which is all the design required to work.
Implementation:
Rather than forgoing the lockless design, or reverting to a
stack-based cpumask_t, we simply check for when the race has
been encountered and continue processing in the event that the
race is hit. This renders the removal race as if the priority
bit had been atomically cleared as well, and allows the
algorithm to execute correctly.
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
CC: Rusty Russell <rusty@rustcorp.com.au>
CC: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090730145728.25226.92769.stgit@dev.haskins.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The latencytop and sleep accounting code assumes that any
scheduler entity represents a task, this is not so.
Cc: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The might_sleep() test inside cond_resched_lock() assumes the
spinlock is held and then preemption is disabled. This is true
with CONFIG_PREEMPT but the preempt_count() doesn't change
otherwise.
Check by starting from the appropriate preempt offset depending
on the config.
Reported-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1248458723-12146-1-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
io context: fix ref counting
block: make the end_io functions be non-GPL exports
block: fix improper kobject release in blk_integrity_unregister
block: always assign default lock to queues
mg_disk: Add missing ready status check on mg_write()
mg_disk: fix issue with data integrity on error in mg_write()
mg_disk: fix reading invalid status when use polling driver
mg_disk: remove prohibited sleep operation
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc:
mmc: orphan subsystem
imxmmc: Remove unnecessary semicolons
cb710: use SG_MITER_TO_SG/SG_MITER_FROM_SG
sdhci: use SG_MITER_TO_SG/SG_MITER_FROM_SG
lib/scatterlist: add a flags to signalize mapping direction
the code allready uses flush_kernel_dcache_page(). This patch updates the
driver to the recent sg API changes which require that either SG_MITER_TO_SG
or SG_MITER_FROM_SG is set. SG_MITER_TO_SG calls flush_kernel_dcache_page()
in sg_mitter_stop()
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Acked-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: Pierre Ossman <pierre@ossman.eu>
so the page will be flushed on unmap on ARCH which need it.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Pierre Ossman <pierre@ossman.eu>
sg_miter_start() is currently unaware of the direction of the copy
process (to or from the scatter list). It is important to know the
direction because the page has to be flushed in case the data written
is seen on a different mapping in user land on cache incoherent
architectures.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Acked-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Pierre Ossman <pierre@ossman.eu>
Check that the result of kzalloc is not NULL before a dereference.
The semantic match that finds this problem is as follows:
(http://www.emn.fr/x-info/coccinelle/)
// <smpl>
@@
expression *x;
identifier f;
constant char *C;
@@
x = \(kmalloc\|kcalloc\|kzalloc\)(...);
... when != x == NULL
when != x != NULL
when != (x || ...)
(
kfree(x)
|
f(...,C,...,x,...)
|
*f(...,x,...)
|
*x->f
)
// </smpl>
Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The name buf with size 16 is too short for some codec names, e.g.
truncated like "ALC861-VD Analo". Now the size is doubled.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
VM calculation for nr_to_write seems off. Bump it way
up, this gets simple streaming writes zippy again.
To be reviewed again after Jens' writeback changes.
Signed-off-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Eric Sandeen <sandeen@sandeen.net>
Cc: Chris Mason <chris.mason@oracle.com>
Reviewed-by: Felix Blyakher <felixb@sgi.com>
Signed-off-by: Felix Blyakher <felixb@sgi.com>
commit 6321e3ed2a caused
the full bmv_count's worth of getbmapx structures to get
allocated; telling it to do MAXEXTNUM was a bit insane,
resulting in ENOMEM every time.
Chop it down to something reasonable, the number of slots
in the caller's input buffer. If this is too large the
caller may get ENOMEM but the reason should not be a
mystery, and they can try again with something smaller.
We add 1 to the value because in the normal getbmap
world, bmv_count includes the header and xfs_getbmap does:
nex = bmv->bmv_count - 1;
if (nex <= 0)
return XFS_ERROR(EINVAL);
Signed-off-by: Eric Sandeen <sandeen@sandeen.net>
Reviewed-by: Olaf Weber <olaf@sgi.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Felix Blyakher <felixb@sgi.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: be more polite in the async caching threads
Btrfs: preserve commit_root for async caching
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx:
dmaengine: at_hdmac: add DMA slave transfers
dmaengine: at_hdmac: new driver for the Atmel AHB DMA Controller
dmaengine: dmatest: correct thread_count while using multiple thread per channel
dmaengine: dmatest: add a maximum number of test iterations
drivers/dma: Remove unnecessary semicolons
drivers/dma/fsldma.c: Remove unnecessary semicolons
dmaengine: move HIGHMEM64G restriction to ASYNC_TX_DMA
fsldma: do not clear bandwidth control bits on the 83xx controller
fsldma: enable external start for the 83xx controller
fsldma: use PCI Read Multiple command
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-udf-2.6:
udf: Fix loading of VAT inode when drive wrongly reports number of recorded blocks
* git://git.infradead.org/users/cbou/battery-2.6.31:
Add ds2782 battery gas gauge driver
olpc_battery: Ensure that the TRICKLE bit is checked
olpc_battery: Fix up eeprom read function
* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-fixes:
GFS2: remove dcache entries for remote deleted inodes
GFS2: Fix incorrent statfs consistency check
GFS2: Don't put unlikely reclaim candidates on the reclaim list.
GFS2: Don't try and dealloc own inode
GFS2: Fix panic in glock memory shrinker
GFS2: keep statfs info in sync on grows
GFS2: Shrink the shrinker
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
powerpc: Update defconfigs for embedded 6xx/7xxx, 8xx, 8{3,5,6}xxx
powerpc/86xx: Update GE Fanuc sbc310 default configuration
powerpc/86xx: Update defconfig for GE Fanuc's PPC9A
cpm_uart: Don't use alloc_bootmem in cpm_uart_cpm2.c
powerpc/83xx: Fix PCI IO base address on MPC837xE-RDB boards
powerpc/85xx: Don't scan for TBI PHY addresses on MPC8569E-MDS boards
powerpc/85xx: Fix ethernet link detection on MPC8569E-MDS boards
powerpc/mm: Fix SMP issue with MMU context handling code
* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
lguest and virtio: cleanup struct definitions to Linux style.
lguest: update commentry
lguest: fix comment style
virtio: refactor find_vqs
virtio: delete vq from list
virtio: fix memory leak on device removal
lguest: fix descriptor corruption in example launcher
lguest: dereferencing freed mem in add_eventfd()
Use kernel_text_address() for checking probe address instead of
__kernel_text_address(), because __kernel_text_address() returns true
for init functions even after relaseing those functions.
That will hit a BUG() in text_poke().
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Not that anybody can blame him. It's a morass. But hey, it's way
better than it _used_ to be, though, so thanks for all the fish.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit d01730d74d didn't completely fix
the problem since we still take dqio_mutex and i_mutex in the wrong
order. Move taking of i_mutex further down (luckily it's needed only
for updating inode flags) below where dqio_mutex is taken.
Tested-by: Valdis Kletnieks <valdis.kletnieks@vt.edu>
Signed-off-by: Jan Kara <jack@suse.cz>
