softirq time in ksoftirqd context is not accounted in ns granularity
per cpu softirq stats, as we want that to be a part of ksoftirqd
exec_runtime.
Accounting them as softirq on /proc/stat separately.
Tested-by: Shaun Ruffell <sruffell@digium.com>
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1292980144-28796-6-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
CONFIG_IRQ_TIME_ACCOUNTING adds ns granularity irq time on each CPU.
This info is already used in scheduler to do proper task chargeback
(earlier patches). This patch retro-fits this ns granularity
hardirq and softirq information to /proc/stat irq and softirq fields.
The update is still done on timer tick, where we look at accumulated
ns hardirq/softirq time and account the tick to user/system/irq/hardirq/guest
accordingly.
No new interface added.
Earlier versions looked at adding this as new fields in some /proc
files. This one seems to be the best in terms of impact to existing
apps, even though it has somewhat more kernel code than earlier versions.
Tested-by: Shaun Ruffell <sruffell@digium.com>
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1292980144-28796-5-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Refactor account_system_time, to separate out the logic of
identifying the update needed and code that does actual update.
This is used by following patch for IRQ_TIME_ACCOUNTING,
which has different identification logic and same update logic.
Tested-by: Shaun Ruffell <sruffell@digium.com>
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1292980144-28796-4-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cleanup patch, freeing up PF_KSOFTIRQD and use per_cpu ksoftirqd pointer
instead, as suggested by Eric Dumazet.
Tested-by: Shaun Ruffell <sruffell@digium.com>
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1292980144-28796-2-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Since cfs->{load_stamp,load_last} are zero-initalized the initial load update
will consider the delta to be 'since the beginning of time'.
This results in a lot of pointless divisions to bring this large period to be
within the sysctl_sched_shares_window.
Fix this by initializing load_stamp to be 1 at cfs_rq initialization, this
allows for an initial load_stamp > load_last which then lets standard idle
truncation proceed.
We avoid spinning (and slightly improve consistency) by fixing delta to be
[period - 1] in this path resulting in a slightly more predictable shares ramp.
(Previously the amount of idle time preserved by the overflow would range between
[period/2,period-1].)
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20110122044852.102126037@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Re-visiting this: Since update_cfs_shares will now only ever re-weight an
entity that is a relative parent of the current entity in enqueue_entity; we
can safely issue the account_entity_enqueue relative to that cfs_rq and avoid
the requirement for special handling of the enqueue case in update_cfs_shares.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20110122044851.915214637@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'sched-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
sched, cgroup: Use exit hook to avoid use-after-free crash
sched: Fix signed unsigned comparison in check_preempt_tick()
sched: Replace rq->bkl_count with rq->rq_sched_info.bkl_count
sched, autogroup: Fix CONFIG_RT_GROUP_SCHED sched_setscheduler() failure
sched: Display autogroup names in /proc/sched_debug
sched: Reinstate group names in /proc/sched_debug
sched: Update effective_load() to use global share weights
By not notifying the controller of the on-exit move back to
init_css_set, we fail to move the task out of the previous
cgroup's cfs_rq. This leads to an opportunity for a
cgroup-destroy to come in and free the cgroup (there are no
active tasks left in it after all) to which the not-quite dead
task is still enqueued.
Reported-by: Miklos Vajna <vmiklos@frugalware.org>
Fixed-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: <stable@kernel.org>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
LKML-Reference: <1293206353.29444.205.camel@laptop>
Now rq->rq_sched_info.bkl_count is not used for rq, scroll
rq->bkl_count into it. Thus we can save some space for rq.
Signed-off-by: Yong Zhang <yong.zhang0@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294991859-13246-1-git-send-email-yong.zhang0@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
If CONFIG_RT_GROUP_SCHED is set, __sched_setscheduler() fails due to autogroup
not allocating rt_runtime. Free unused/unusable rt_se and rt_rq, redirect RT
tasks to the root task group, and tell __sched_setscheduler() that it's ok.
Reported-and-tested-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294890890.8089.39.camel@marge.simson.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
One of the operands, buf, is incorrect, since it is stripped and the
correct address for subsequent string comparing could change if
leading white spaces, if any, are removed from buf.
It is fixed by replacing buf with cmp.
Signed-off-by: Hillf Danton <dhillf@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <AANLkTinOPuYsVovrZpbuCCmG5deEyc8WgA_A1RJx_YK7@mail.gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Seems I lost a change somewhere, leaking memory.
sched: fix struct autogroup memory leak
Add missing change to actually use autogroup_free().
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294222285.8369.2.camel@marge.simson.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
root_task_group is the leftover of USER_SCHED, now it's always
same to init_task_group.
But as Mike suggested, root_task_group is maybe the suitable name
to keep for a tree.
So in this patch:
init_task_group --> root_task_group
init_task_group_load --> root_task_group_load
INIT_TASK_GROUP_LOAD --> ROOT_TASK_GROUP_LOAD
Suggested-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Yong Zhang <yong.zhang0@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20110107071736.GA32635@windriver.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (30 commits)
sched: Change wait_for_completion_*_timeout() to return a signed long
sched, autogroup: Fix reference leak
sched, autogroup: Fix potential access to freed memory
sched: Remove redundant CONFIG_CGROUP_SCHED ifdef
sched: Fix interactivity bug by charging unaccounted run-time on entity re-weight
sched: Move periodic share updates to entity_tick()
printk: Use this_cpu_{read|write} api on printk_pending
sched: Make pushable_tasks CONFIG_SMP dependant
sched: Add 'autogroup' scheduling feature: automated per session task groups
sched: Fix unregister_fair_sched_group()
sched: Remove unused argument dest_cpu to migrate_task()
mutexes, sched: Introduce arch_mutex_cpu_relax()
sched: Add some clock info to sched_debug
cpu: Remove incorrect BUG_ON
cpu: Remove unused variable
sched: Fix UP build breakage
sched: Make task dump print all 15 chars of proc comm
sched: Update tg->shares after cpu.shares write
sched: Allow update_cfs_load() to update global load
sched: Implement demand based update_cfs_load()
...
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (146 commits)
tools, perf: Documentation for the power events API
perf: Add calls to suspend trace point
perf script: Make some lists static
perf script: Use the default lost event handler
perf session: Warn about errors when processing pipe events too
perf tools: Fix perf_event.h header usage
perf test: Clarify some error reports in the open syscall test
x86, NMI: Add touch_nmi_watchdog to io_check_error delay
x86: Avoid calling arch_trigger_all_cpu_backtrace() at the same time
x86: Only call smp_processor_id in non-preempt cases
perf timechart: Adjust perf timechart to the new power events
perf: Clean up power events by introducing new, more generic ones
perf: Do not export power_frequency, but power_start event
perf test: Add test for counting open syscalls
perf evsel: Auto allocate resources needed for some methods
perf evsel: Use {cpu,thread}_map to shorten list of parameters
perf tools: Refactor all_tids to hold nr and the map
perf tools: Refactor cpumap to hold nr and the map
perf evsel: Introduce per cpu and per thread open helpers
perf evsel: Steal the counter reading routines from stat
...
wait_for_completion_*_timeout() can return:
0: if the wait timed out
-ve: if the wait was interrupted
+ve: if the completion was completed.
As they currently return an 'unsigned long', the last two cases
are not easily distinguished which can easily result in buggy
code, as is the case for the recently added
wait_for_completion_interruptible_timeout() call in
net/sunrpc/cache.c
So change them both to return 'long'. As MAX_SCHEDULE_TIMEOUT
is LONG_MAX, a large +ve return value should never overflow.
Signed-off-by: NeilBrown <neilb@suse.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: J. Bruce Fields <bfields@fieldses.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <20110105125016.64ccab0e@notabene.brown>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Conflicts:
MAINTAINERS
arch/arm/mach-omap2/pm24xx.c
drivers/scsi/bfa/bfa_fcpim.c
Needed to update to apply fixes for which the old branch was too
outdated.
Linus reported that the new warning introduced by commit f26f9aff6a
"Sched: fix skip_clock_update optimization" triggers. The need_resched
flag can be set by other CPUs asynchronously so this debug check is
bogus - remove it.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <AANLkTinJ8hAG1TpyC+CSYPR47p48+1=E7fiC45hMXT_1@mail.gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently we call perf_event_init() from sched_init(). In order to
make it more obvious move it to the cannnonical location.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101117222056.093629821@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Since the irqtime accounting is using non-atomic u64 and can be read
from remote cpus (writes are strictly cpu local, reads are not) we
have to deal with observing partial updates.
When we do observe partial updates the clock movement (in particular,
->clock_task movement) will go funny (in either direction), a
subsequent clock update (observing the full update) will make it go
funny in the oposite direction.
Since we rely on these clocks to be strictly monotonic we cannot
suffer backwards motion. One possible solution would be to simply
ignore all backwards deltas, but that will lead to accounting
artefacts, most notable: clock_task + irq_time != clock, this
inaccuracy would end up in user visible stats.
Therefore serialize the reads using a seqcount.
Reviewed-by: Venkatesh Pallipadi <venki@google.com>
Reported-by: Mikael Pettersson <mikpe@it.uu.se>
Tested-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1292242434.6803.200.camel@twins>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Some ARM systems have a short sched_clock() [ which needs to be fixed
too ], but this exposed a bug in the irq_time code as well, it doesn't
deal with wraps at all.
Fix the irq_time code to deal with u64 wraps by re-writing the code to
only use delta increments, which avoids the whole issue.
Reviewed-by: Venkatesh Pallipadi <venki@google.com>
Reported-by: Mikael Pettersson <mikpe@it.uu.se>
Tested-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1292242433.6803.199.camel@twins>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
As noted by Peter Zijlstra at https://lkml.org/lkml/2010/11/10/391
(while reviewing other stuff, though), tracking pushable tasks
only makes sense on SMP systems.
Signed-off-by: Dario Faggioli <raistlin@linux.it>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Gregory Haskins <ghaskins@novell.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1291143093.2697.298.camel@Palantir>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
idle_balance() drops/retakes rq->lock, leaving the previous task
vulnerable to set_tsk_need_resched(). Clear it after we return
from balancing instead, and in setup_thread_stack() as well, so
no successfully descheduled or never scheduled task has it set.
Need resched confused the skip_clock_update logic, which assumes
that the next call to update_rq_clock() will come nearly immediately
after being set. Make the optimization robust against the waking
a sleeper before it sucessfully deschedules case by checking that
the current task has not been dequeued before setting the flag,
since it is that useless clock update we're trying to save, and
clear unconditionally in schedule() proper instead of conditionally
in put_prev_task().
Signed-off-by: Mike Galbraith <efault@gmx.de>
Reported-by: Bjoern B. Brandenburg <bbb.lst@gmail.com>
Tested-by: Yong Zhang <yong.zhang0@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: stable@kernel.org
LKML-Reference: <1291802742.1417.9.camel@marge.simson.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There's a long-running regression that proved difficult to fix and
which is hitting certain people and is rather annoying in its effects.
Damien reported that after 74f5187ac8 (sched: Cure load average vs
NO_HZ woes) his load average is unnaturally high, he also noted that
even with that patch reverted the load avgerage numbers are not
correct.
The problem is that the previous patch only solved half the NO_HZ
problem, it addressed the part of going into NO_HZ mode, not of
comming out of NO_HZ mode. This patch implements that missing half.
When comming out of NO_HZ mode there are two important things to take
care of:
- Folding the pending idle delta into the global active count.
- Correctly aging the averages for the idle-duration.
So with this patch the NO_HZ interaction should be complete and
behaviour between CONFIG_NO_HZ=[yn] should be equivalent.
Furthermore, this patch slightly changes the load average computation
by adding a rounding term to the fixed point multiplication.
Reported-by: Damien Wyart <damien.wyart@free.fr>
Reported-by: Tim McGrath <tmhikaru@gmail.com>
Tested-by: Damien Wyart <damien.wyart@free.fr>
Tested-by: Orion Poplawski <orion@cora.nwra.com>
Tested-by: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: stable@kernel.org
Cc: Chase Douglas <chase.douglas@canonical.com>
LKML-Reference: <1291129145.32004.874.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
A recurring complaint from CFS users is that parallel kbuild has
a negative impact on desktop interactivity. This patch
implements an idea from Linus, to automatically create task
groups. Currently, only per session autogroups are implemented,
but the patch leaves the way open for enhancement.
Implementation: each task's signal struct contains an inherited
pointer to a refcounted autogroup struct containing a task group
pointer, the default for all tasks pointing to the
init_task_group. When a task calls setsid(), a new task group
is created, the process is moved into the new task group, and a
reference to the preveious task group is dropped. Child
processes inherit this task group thereafter, and increase it's
refcount. When the last thread of a process exits, the
process's reference is dropped, such that when the last process
referencing an autogroup exits, the autogroup is destroyed.
At runqueue selection time, IFF a task has no cgroup assignment,
its current autogroup is used.
Autogroup bandwidth is controllable via setting it's nice level
through the proc filesystem:
cat /proc/<pid>/autogroup
Displays the task's group and the group's nice level.
echo <nice level> > /proc/<pid>/autogroup
Sets the task group's shares to the weight of nice <level> task.
Setting nice level is rate limited for !admin users due to the
abuse risk of task group locking.
The feature is enabled from boot by default if
CONFIG_SCHED_AUTOGROUP=y is selected, but can be disabled via
the boot option noautogroup, and can also be turned on/off on
the fly via:
echo [01] > /proc/sys/kernel/sched_autogroup_enabled
... which will automatically move tasks to/from the root task group.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Markus Trippelsdorf <markus@trippelsdorf.de>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Paul Turner <pjt@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
[ Removed the task_group_path() debug code, and fixed !EVENTFD build failure. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
LKML-Reference: <1290281700.28711.9.camel@maggy.simson.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In the flipping and flopping between calling
unregister_fair_sched_group() on a per-cpu versus per-group basis
we ended up in a bad state.
Remove from the list for the passed cpu as opposed to some
arbitrary index.
( This fixes explosions w/ autogroup as well as a group
creation/destruction stress test. )
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <20101130005740.080828123@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The first version of synchronize_sched_expedited() used the migration
code in the scheduler, and was therefore implemented in kernel/sched.c.
However, the more recent version of this code no longer uses the
migration code, so this commit moves it to the main RCU source files.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Remove unused argument, 'dest_cpu' of migrate_task(), and pass runqueue,
as it is always known at the call site.
Signed-off-by: Nikanth Karthikesan <knikanth@suse.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <201011261237.09187.knikanth@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The spinning mutex implementation uses cpu_relax() in busy loops as a
compiler barrier. Depending on the architecture, cpu_relax() may do more
than needed in this specific mutex spin loops. On System z we also give
up the time slice of the virtual cpu in cpu_relax(), which prevents
effective spinning on the mutex.
This patch replaces cpu_relax() in the spinning mutex code with
arch_mutex_cpu_relax(), which can be defined by each architecture that
selects HAVE_ARCH_MUTEX_CPU_RELAX. The default is still cpu_relax(), so
this patch should not affect other architectures than System z for now.
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1290437256.7455.4.camel@thinkpad>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Formerly sched_group_set_shares would force a rebalance by overflowing domain
share sums. Now that per-cpu averages are maintained we can set the true value
by issuing an update_cfs_shares() following a tg->shares update.
Also initialize tg se->load to 0 for consistency since we'll now set correct
weights on enqueue.
Signed-off-by: Paul Turner <pjt@google.com?>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101115234938.465521344@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When the system is busy, dilation of rq->next_balance makes lb->update_shares()
insufficiently frequent for threads which don't sleep (no dequeue/enqueue
updates). Adjust for this by making demand based updates based on the
accumulation of execution time sufficient to wrap our averaging window.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101115234938.291159744@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Using cfs_rq->nr_running is not sufficient to synchronize update_cfs_load with
the put path since nr_running accounting occurs at deactivation.
It's also not safe to make the removal decision based on load_avg as this fails
with both high periods and low shares. Resolve this by clipping history after
4 periods without activity.
Note: the above will always occur from update_shares() since in the
last-task-sleep-case that task will still be cfs_rq->curr when update_cfs_load
is called.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101115234937.933428187@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Make tg_shares_up() use the active cgroup list, this means we cannot
do a strict bottom-up walk of the hierarchy, but assuming its a very
wide tree with a small number of active groups it should be a win.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101115234937.754159484@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Make certain load-balance actions scale per number of active cgroups
instead of the number of existing cgroups.
This makes wakeup/sleep paths more expensive, but is a win for systems
where the vast majority of existing cgroups are idle.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101115234937.666535048@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
By tracking a per-cpu load-avg for each cfs_rq and folding it into a
global task_group load on each tick we can rework tg_shares_up to be
strictly per-cpu.
This should improve cpu-cgroup performance for smp systems
significantly.
[ Paul: changed to use queueing cfs_rq + bug fixes ]
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101115234937.580480400@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
While discussing the need for sched_idle_next(), Oleg remarked that
since try_to_wake_up() ensures sleeping tasks will end up running on a
sane cpu, we can do away with migrate_live_tasks().
If we then extend the existing hack of migrating current from
CPU_DYING to migrating the full rq worth of tasks from CPU_DYING, the
need for the sched_idle_next() abomination disappears as well, since
idle will be the only possible thread left after the migration thread
stops.
This greatly simplifies the hot-unplug task migration path, as can be
seen from the resulting code reduction (and about half the new lines
are comments).
Suggested-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1289851597.2109.547.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Instead of dealing with sched classes inside each check_preempt_curr()
implementation, pull out this logic into the generic wakeup preemption
path.
This fixes a hang in KVM (and others) where we are waiting for the
stop machine thread to run ...
Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Tested-by: Marcelo Tosatti <mtosatti@redhat.com>
Tested-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1288891946.2039.31.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently we consider a sched domain to be well balanced when the imbalance
is less than the domain's imablance_pct. As the number of cores and threads
are increasing, current values of imbalance_pct (for example 25% for a
NUMA domain) are not enough to detect imbalances like:
a) On a WSM-EP system (two sockets, each having 6 cores and 12 logical threads),
24 cpu-hogging tasks get scheduled as 13 on one socket and 11 on another
socket. Leading to an idle HT cpu.
b) On a hypothetial 2 socket NHM-EX system (each socket having 8 cores and
16 logical threads), 16 cpu-hogging tasks can get scheduled as 9 on one
socket and 7 on another socket. Leaving one core in a socket idle
whereas in another socket we have a core having both its HT siblings busy.
While this issue can be fixed by decreasing the domain's imbalance_pct
(by making it a function of number of logical cpus in the domain), it
can potentially cause more task migrations across sched groups in an
overloaded case.
Fix this by using imbalance_pct only during newly_idle and busy
load balancing. And during idle load balancing, check if there
is an imbalance in number of idle cpu's across the busiest and this
sched_group or if the busiest group has more tasks than its weight that
the idle cpu in this_group can pull.
Reported-by: Nikhil Rao <ncrao@google.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1284760952.2676.11.camel@sbsiddha-MOBL3.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Andrew Morton pointed out almost all sched_setscheduler() callers are
using fixed parameters and can be converted to static. It reduces runtime
memory use a little.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: James Morris <jmorris@namei.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Dima noticed that we fail to correct the ->vruntime of sleeping tasks
when we move them between cgroups.
Reported-by: Dima Zavin <dima@android.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Mike Galbraith <efault@gmx.de>
LKML-Reference: <1287150604.29097.1513.camel@twins>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (29 commits)
sched: Export account_system_vtime()
sched: Call tick_check_idle before __irq_enter
sched: Remove irq time from available CPU power
sched: Do not account irq time to current task
x86: Add IRQ_TIME_ACCOUNTING
sched: Add IRQ_TIME_ACCOUNTING, finer accounting of irq time
sched: Add a PF flag for ksoftirqd identification
sched: Consolidate account_system_vtime extern declaration
sched: Fix softirq time accounting
sched: Drop group_capacity to 1 only if local group has extra capacity
sched: Force balancing on newidle balance if local group has capacity
sched: Set group_imb only a task can be pulled from the busiest cpu
sched: Do not consider SCHED_IDLE tasks to be cache hot
sched: Drop all load weight manipulation for RT tasks
sched: Create special class for stop/migrate work
sched: Unindent labels
sched: Comment updates: fix default latency and granularity numbers
tracing/sched: Add sched_pi_setprio tracepoint
sched: Give CPU bound RT tasks preference
sched: Try not to migrate higher priority RT tasks
...
All security modules shouldn't change sched_param parameter of
security_task_setscheduler(). This is not only meaningless, but also
make a harmful result if caller pass a static variable.
This patch remove policy and sched_param parameter from
security_task_setscheduler() becuase none of security module is
using it.
Cc: James Morris <jmorris@namei.org>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: James Morris <jmorris@namei.org>
When CPU is idle and on first interrupt, irq_enter calls tick_check_idle()
to notify interruption from idle. But, there is a problem if this call
is done after __irq_enter, as all routines in __irq_enter may find
stale time due to yet to be done tick_check_idle.
Specifically, trace calls in __irq_enter when they use global clock and also
account_system_vtime change in this patch as it wants to use sched_clock_cpu()
to do proper irq timing.
But, tick_check_idle was moved after __irq_enter intentionally to
prevent problem of unneeded ksoftirqd wakeups by the commit ee5f80a:
irq: call __irq_enter() before calling the tick_idle_check
Impact: avoid spurious ksoftirqd wakeups
Moving tick_check_idle() before __irq_enter and wrapping it with
local_bh_enable/disable would solve both the problems.
Fixed-by: Yong Zhang <yong.zhang0@gmail.com>
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-9-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The idea was suggested by Peter Zijlstra here:
http://marc.info/?l=linux-kernel&m=127476934517534&w=2
irq time is technically not available to the tasks running on the CPU.
This patch removes irq time from CPU power piggybacking on
sched_rt_avg_update().
Tested this by keeping CPU X busy with a network intensive task having 75%
oa a single CPU irq processing (hard+soft) on a 4-way system. And start seven
cycle soakers on the system. Without this change, there will be two tasks on
each CPU. With this change, there is a single task on irq busy CPU X and
remaining 7 tasks are spread around among other 3 CPUs.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-8-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Scheduler accounts both softirq and interrupt processing times to the
currently running task. This means, if the interrupt processing was
for some other task in the system, then the current task ends up being
penalized as it gets shorter runtime than otherwise.
Change sched task accounting to acoount only actual task time from
currently running task. Now update_curr(), modifies the delta_exec to
depend on rq->clock_task.
Note that this change only handles CONFIG_IRQ_TIME_ACCOUNTING case. We can
extend this to CONFIG_VIRT_CPU_ACCOUNTING with minimal effort. But, thats
for later.
This change will impact scheduling behavior in interrupt heavy conditions.
Tested on a 4-way system with eth0 handled by CPU 2 and a network heavy
task (nc) running on CPU 3 (and no RSS/RFS). With that I have CPU 2
spending 75%+ of its time in irq processing. CPU 3 spending around 35%
time running nc task.
Now, if I run another CPU intensive task on CPU 2, without this change
/proc/<pid>/schedstat shows 100% of time accounted to this task. With this
change, it rightly shows less than 25% accounted to this task as remaining
time is actually spent on irq processing.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-7-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
s390/powerpc/ia64 have support for CONFIG_VIRT_CPU_ACCOUNTING which does
the fine granularity accounting of user, system, hardirq, softirq times.
Adding that option on archs like x86 will be challenging however, given the
state of TSC reliability on various platforms and also the overhead it will
add in syscall entry exit.
Instead, add a lighter variant that only does finer accounting of
hardirq and softirq times, providing precise irq times (instead of timer tick
based samples). This accounting is added with a new config option
CONFIG_IRQ_TIME_ACCOUNTING so that there won't be any overhead for users not
interested in paying the perf penalty.
This accounting is based on sched_clock, with the code being generic.
So, other archs may find it useful as well.
This patch just adds the core logic and does not enable this logic yet.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-5-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Peter Zijlstra found a bug in the way softirq time is accounted in
VIRT_CPU_ACCOUNTING on this thread:
http://lkml.indiana.edu/hypermail//linux/kernel/1009.2/01366.html
The problem is, softirq processing uses local_bh_disable internally. There
is no way, later in the flow, to differentiate between whether softirq is
being processed or is it just that bh has been disabled. So, a hardirq when bh
is disabled results in time being wrongly accounted as softirq.
Looking at the code a bit more, the problem exists in !VIRT_CPU_ACCOUNTING
as well. As account_system_time() in normal tick based accouting also uses
softirq_count, which will be set even when not in softirq with bh disabled.
Peter also suggested solution of using 2*SOFTIRQ_OFFSET as irq count
for local_bh_{disable,enable} and using just SOFTIRQ_OFFSET while softirq
processing. The patch below does that and adds API in_serving_softirq() which
returns whether we are currently processing softirq or not.
Also changes one of the usages of softirq_count in net/sched/cls_cgroup.c
to in_serving_softirq.
Looks like many usages of in_softirq really want in_serving_softirq. Those
changes can be made individually on a case by case basis.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-2-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch adds a check in task_hot to return if the task has SCHED_IDLE
policy. SCHED_IDLE tasks have very low weight, and when run with regular
workloads, are typically scheduled many milliseconds apart. There is no
need to consider these tasks hot for load balancing.
Signed-off-by: Nikhil Rao <ncrao@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1287173550-30365-2-git-send-email-ncrao@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Load weights are for the CFS, they do not belong in the RT task. This makes all
RT scheduling classes leave the CFS weights alone.
This fixes a real bug as well: I noticed the following phonomena: a process
elevated to SCHED_RR forks with SCHED_RESET_ON_FORK set, and the child is
indeed SCHED_OTHER, and the niceval is indeed reset to 0. However the weight
inserted by set_load_weight() remains at 0, giving the task insignificat
priority.
With this fix, the weight is reset to what the task had before being elevated
to SCHED_RR/SCHED_FIFO.
Cc: Lennart Poettering <lennart@poettering.net>
Cc: stable@kernel.org
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286807811-10568-1-git-send-email-linus.walleij@stericsson.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In order to separate the stop/migrate work thread from the SCHED_FIFO
implementation, create a special class for it that is of higher priority than
SCHED_FIFO itself.
This currently solves a problem where cpu-hotplug consumes so much cpu-time
that the SCHED_FIFO class gets throttled, but has the bandwidth replenishment
timer pending on the now dead cpu.
It is also required for when we add the planned deadline scheduling class above
SCHED_FIFO, as the stop/migrate thread still needs to transcent those tasks.
Tested-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1285165776.2275.1022.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Labels should be on column 0.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This addresses the following RCU lockdep splat:
[0.051203] CPU0: AMD QEMU Virtual CPU version 0.12.4 stepping 03
[0.052999] lockdep: fixing up alternatives.
[0.054105]
[0.054106] ===================================================
[0.054999] [ INFO: suspicious rcu_dereference_check() usage. ]
[0.054999] ---------------------------------------------------
[0.054999] kernel/sched.c:616 invoked rcu_dereference_check() without protection!
[0.054999]
[0.054999] other info that might help us debug this:
[0.054999]
[0.054999]
[0.054999] rcu_scheduler_active = 1, debug_locks = 1
[0.054999] 3 locks held by swapper/1:
[0.054999] #0: (cpu_add_remove_lock){+.+.+.}, at: [<ffffffff814be933>] cpu_up+0x42/0x6a
[0.054999] #1: (cpu_hotplug.lock){+.+.+.}, at: [<ffffffff810400d8>] cpu_hotplug_begin+0x2a/0x51
[0.054999] #2: (&rq->lock){-.-...}, at: [<ffffffff814be2f7>] init_idle+0x2f/0x113
[0.054999]
[0.054999] stack backtrace:
[0.054999] Pid: 1, comm: swapper Not tainted 2.6.35 #1
[0.054999] Call Trace:
[0.054999] [<ffffffff81068054>] lockdep_rcu_dereference+0x9b/0xa3
[0.054999] [<ffffffff810325c3>] task_group+0x7b/0x8a
[0.054999] [<ffffffff810325e5>] set_task_rq+0x13/0x40
[0.054999] [<ffffffff814be39a>] init_idle+0xd2/0x113
[0.054999] [<ffffffff814be78a>] fork_idle+0xb8/0xc7
[0.054999] [<ffffffff81068717>] ? mark_held_locks+0x4d/0x6b
[0.054999] [<ffffffff814bcebd>] do_fork_idle+0x17/0x2b
[0.054999] [<ffffffff814bc89b>] native_cpu_up+0x1c1/0x724
[0.054999] [<ffffffff814bcea6>] ? do_fork_idle+0x0/0x2b
[0.054999] [<ffffffff814be876>] _cpu_up+0xac/0x127
[0.054999] [<ffffffff814be946>] cpu_up+0x55/0x6a
[0.054999] [<ffffffff81ab562a>] kernel_init+0xe1/0x1ff
[0.054999] [<ffffffff81003854>] kernel_thread_helper+0x4/0x10
[0.054999] [<ffffffff814c353c>] ? restore_args+0x0/0x30
[0.054999] [<ffffffff81ab5549>] ? kernel_init+0x0/0x1ff
[0.054999] [<ffffffff81003850>] ? kernel_thread_helper+0x0/0x10
[0.056074] Booting Node 0, Processors #1lockdep: fixing up alternatives.
[0.130045] #2lockdep: fixing up alternatives.
[0.203089] #3 Ok.
[0.275286] Brought up 4 CPUs
[0.276005] Total of 4 processors activated (16017.17 BogoMIPS).
The cgroup_subsys_state structures referenced by idle tasks are never
freed, because the idle tasks should be part of the root cgroup,
which is not removable.
The problem is that while we do in-fact hold rq->lock, the newly spawned
idle thread's cpu is not yet set to the correct cpu so the lockdep check
in task_group():
lockdep_is_held(&task_rq(p)->lock)
will fail.
But this is a chicken and egg problem. Setting the CPU's runqueue requires
that the CPU's runqueue already be set. ;-)
So insert an RCU read-side critical section to avoid the complaint.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Add a tracepoint that shows the priority of a task being boosted
via priority inheritance.
Cc: Gregory Haskins <ghaskins@novell.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Revert the timer per cpu-context timers because of unfortunate
nohz interaction. Fixing that would have been somewhat ugly, so
go back to driving things from the regular tick. Provide a
jiffies interval feature for people who want slower rotations.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20100917093009.519845633@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
With 710390d9 "sched: Optimize branch hint in context_switch()"
the branch hint logic within context_switch() got inversed.
In fact the hints "if (likely(!mm))" and "if (likely(!prev->mm))"
mean that it is likely that the previous and next task are kernel
threads.
That assumption is certainly counter intuitive, but Tim has shown
that at least with his workload this is true. Nevertheless the
truth is: it depends on the current workload. So just remove the
annotations which also improves readability.
Reported-by: Tim Blechmann <tim@klingt.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <20100916124225.GA2209@osiris.boeblingen.de.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We have 32-bit variable overflow possibility when multiply in
task_times() and thread_group_times() functions. When the
overflow happens then the scaled utime value becomes erroneously
small and the scaled stime becomes i erroneously big.
Reported here:
https://bugzilla.redhat.com/show_bug.cgi?id=633037https://bugzilla.kernel.org/show_bug.cgi?id=16559
Reported-by: Michael Chapman <redhat-bugzilla@very.puzzling.org>
Reported-by: Ciriaco Garcia de Celis <sysman@etherpilot.com>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: <stable@kernel.org> # 2.6.32.19+ (partially) and 2.6.33+
LKML-Reference: <20100914143513.GB8415@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix incorrect handling of the following case:
INTERACTIVE
INTERACTIVE_SOMETHING_ELSE
The comparison only checks up to each element's length.
Changelog since v1:
- Embellish using some Rostedtisms.
[ mingo: ^^ == smaller and cleaner ]
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Cc: <stable@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tony Lindgren <tony@atomide.com>
LKML-Reference: <20100913214700.GB16118@Krystal>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Give each cpu-context its own timer so that it is a self contained
entity, this eases the way for per-pmu-per-cpu contexts as well as
provides the basic infrastructure to allow different rotation
times per pmu.
Things to look at:
- folding the tick and these TICK_NSEC timers
- separate task context rotation
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
On top of the SMT and MC scheduling domains this adds the BOOK scheduling
domain. This is useful for NUMA like machines which do not have an interface
which tells which piece of memory is attached to which node or where the
hardware performs striping.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20100831082844.253053798@de.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Merge and simplify the two cpu_to_core_group variants so that the
resulting function follows the same pattern like cpu_to_phys_group.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20100831082843.953617555@de.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently sched_avg_update() (which updates rt_avg stats in the rq)
is getting called from scale_rt_power() (in the load balance context)
which doesn't take rq->lock.
Fix it by moving the sched_avg_update() to more appropriate
update_cpu_load() where the CFS load gets updated as well.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1282596171.2694.3.camel@sbsiddha-MOBL3>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The CONFIG_SMP ifdef isn't necessary at this point, because it
is checked in an outer ifdef level already and has no effect
here.
Cleanup only, no functional effect.
Signed-off-by: Christian Dietrich <qy03fugy@stud.informatik.uni-erlangen.de>
Cc: vamos-dev@i4.informatik.uni-erlangen.de
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Tejun Heo <tj@kernel.org>
LKML-Reference: <7a3a39ef3f765a4473cb026b1f204059568a7098.1283782701.git.qy03fugy@stud.informatik.uni-erlangen.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There is a scalability issue for current implementation of optimistic
mutex spin in the kernel. It is found on a 8 node 64 core Nehalem-EX
system (HT mode).
The intention of the optimistic mutex spin is to busy wait and spin on a
mutex if the owner of the mutex is running, in the hope that the mutex
will be released soon and be acquired, without the thread trying to
acquire mutex going to sleep. However, when we have a large number of
threads, contending for the mutex, we could have the mutex grabbed by
other thread, and then another ……, and we will keep spinning, wasting cpu
cycles and adding to the contention. One possible fix is to quit
spinning and put the current thread on wait-list if mutex lock switch to
a new owner while we spin, indicating heavy contention (see the patch
included).
I did some testing on a 8 socket Nehalem-EX system with a total of 64
cores. Using Ingo's test-mutex program that creates/delete files with 256
threads (http://lkml.org/lkml/2006/1/8/50) , I see the following speed up
after putting in the mutex spin fix:
./mutex-test V 256 10
Ops/sec
2.6.34 62864
With fix 197200
Repeating the test with Aim7 fserver workload, again there is a speed up
with the fix:
Jobs/min
2.6.34 91657
With fix 149325
To look at the impact on the distribution of mutex acquisition time, I
collected the mutex acquisition time on Aim7 fserver workload with some
instrumentation. The average acquisition time is reduced by 48% and
number of contentions reduced by 32%.
#contentions Time to acquire mutex (cycles)
2.6.34 72973 44765791
With fix 49210 23067129
The histogram of mutex acquisition time is listed below. The acquisition
time is in 2^bin cycles. We see that without the fix, the acquisition
time is mostly around 2^26 cycles. With the fix, we the distribution get
spread out a lot more towards the lower cycles, starting from 2^13.
However, there is an increase of the tail distribution with the fix at
2^28 and 2^29 cycles. It seems a small price to pay for the reduced
average acquisition time and also getting the cpu to do useful work.
Mutex acquisition time distribution (acq time = 2^bin cycles):
2.6.34 With Fix
bin #occurrence % #occurrence %
11 2 0.00% 120 0.24%
12 10 0.01% 790 1.61%
13 14 0.02% 2058 4.18%
14 86 0.12% 3378 6.86%
15 393 0.54% 4831 9.82%
16 710 0.97% 4893 9.94%
17 815 1.12% 4667 9.48%
18 790 1.08% 5147 10.46%
19 580 0.80% 6250 12.70%
20 429 0.59% 6870 13.96%
21 311 0.43% 1809 3.68%
22 255 0.35% 2305 4.68%
23 317 0.44% 916 1.86%
24 610 0.84% 233 0.47%
25 3128 4.29% 95 0.19%
26 63902 87.69% 122 0.25%
27 619 0.85% 286 0.58%
28 0 0.00% 3536 7.19%
29 0 0.00% 903 1.83%
30 0 0.00% 0 0.00%
I've done similar experiments with 2.6.35 kernel on smaller boxes as
well. One is on a dual-socket Westmere box (12 cores total, with HT).
Another experiment is on an old dual-socket Core 2 box (4 cores total, no
HT)
On the 12-core Westmere box, I see a 250% increase for Ingo's mutex-test
program with my mutex patch but no significant difference in aim7's
fserver workload.
On the 4-core Core 2 box, I see the difference with the patch for both
mutex-test and aim7 fserver are negligible.
So far, it seems like the patch has not caused regression on smaller
systems.
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: <stable@kernel.org> # .35.x
LKML-Reference: <1282168827.9542.72.camel@schen9-DESK>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (27 commits)
sched: Use correct macro to display sched_child_runs_first in /proc/sched_debug
sched: No need for bootmem special cases
sched: Revert nohz_ratelimit() for now
sched: Reduce update_group_power() calls
sched: Update rq->clock for nohz balanced cpus
sched: Fix spelling of sibling
sched, cpuset: Drop __cpuexit from cpu hotplug callbacks
sched: Fix the racy usage of thread_group_cputimer() in fastpath_timer_check()
sched: run_posix_cpu_timers: Don't check ->exit_state, use lock_task_sighand()
sched: thread_group_cputime: Simplify, document the "alive" check
sched: Remove the obsolete exit_state/signal hacks
sched: task_tick_rt: Remove the obsolete ->signal != NULL check
sched: __sched_setscheduler: Read the RLIMIT_RTPRIO value lockless
sched: Fix comments to make them DocBook happy
sched: Fix fix_small_capacity
powerpc: Exclude arch_sd_sibiling_asym_packing() on UP
powerpc: Enable asymmetric SMT scheduling on POWER7
sched: Add asymmetric group packing option for sibling domain
sched: Fix capacity calculations for SMT4
sched: Change nohz idle load balancing logic to push model
...
As of commit dcce284 ("mm: Extend gfp masking to the page
allocator") and commit 7e85ee0 ("slab,slub: don't enable
interrupts during early boot"), the slab allocator makes
sure we don't attempt to sleep during boot.
Therefore, remove bootmem special cases from the scheduler
and use plain GFP_KERNEL instead.
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1279225102-2572-1-git-send-email-penberg@cs.helsinki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Norbert reported that nohz_ratelimit() causes his laptop to burn about
4W (40%) extra. For now back out the change and see if we can adjust
the power management code to make better decisions.
Reported-by: Norbert Preining <preining@logic.at>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Mike Galbraith <efault@gmx.de>
Cc: Arjan van de Ven <arjan@infradead.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit 0224cf4c5e (sched: Intoduce get_cpu_iowait_time_us())
broke things by not making sure preemption was indeed disabled
by the callers of nr_iowait_cpu() which took the iowait value of
the current cpu.
This resulted in a heap of preempt warnings. Cure this by making
nr_iowait_cpu() take a cpu number and fix up the callers to pass
in the right number.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Maxim Levitsky <maximlevitsky@gmail.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: linux-pm@lists.linux-foundation.org
LKML-Reference: <1277968037.1868.120.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
GCC 4.4.1 on ARM has been observed to replace the while loop in
sched_avg_update with a call to uldivmod, resulting in the
following build failure at link-time:
kernel/built-in.o: In function `sched_avg_update':
kernel/sched.c:1261: undefined reference to `__aeabi_uldivmod'
kernel/sched.c:1261: undefined reference to `__aeabi_uldivmod'
make: *** [.tmp_vmlinux1] Error 1
This patch introduces a fake data hazard to the loop body to
prevent the compiler optimising the loop away.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Because cgroup_fork() is ran before sched_fork() [ from copy_process() ]
and the child's pid is not yet visible the child is pinned to its
cgroup. Therefore we can silence this warning.
A nicer solution would be moving cgroup_fork() to right after
dup_task_struct() and exclude PF_STARTING from task_subsys_state().
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Commit 3a101d05 (sched: adjust when cpu_active and cpuset
configurations are updated during cpu on/offlining) added
hotplug notifiers marked with __cpuexit; however, ia64 drops
text in __cpuexit during link unlike x86.
This means that functions which are referenced during init but used
only for cpu hot unplugging afterwards shouldn't be marked with
__cpuexit. Drop __cpuexit from those functions.
Reported-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Tony Luck <tony.luck@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <4C1FDF5B.1040301@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
__sched_setscheduler() takes lock_task_sighand() to access task->signal.
This is not needed since ea6d290c, ->signal can't go away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20100610230944.GA25903@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit e70971591 ("sched: Optimize unused cgroup configuration") introduced
an imbalanced scheduling bug.
If we do not use CGROUP, function update_h_load won't update h_load. When the
system has a large number of tasks far more than logical CPU number, the
incorrect cfs_rq[cpu]->h_load value will cause load_balance() to pull too
many tasks to the local CPU from the busiest CPU. So the busiest CPU keeps
going in a round robin. That will hurt performance.
The issue was found originally by a scientific calculation workload that
developed by Yanmin. With that commit, the workload performance drops
about 40%.
CPU before after
00 : 2 : 7
01 : 1 : 7
02 : 11 : 6
03 : 12 : 7
04 : 6 : 6
05 : 11 : 7
06 : 10 : 6
07 : 12 : 7
08 : 11 : 6
09 : 12 : 6
10 : 1 : 6
11 : 1 : 6
12 : 6 : 6
13 : 2 : 6
14 : 2 : 6
15 : 1 : 6
Reviewed-by: Yanmin zhang <yanmin.zhang@intel.com>
Signed-off-by: Alex Shi <alex.shi@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1276754893.9452.5442.camel@debian>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In the new push model, all idle CPUs indeed go into nohz mode. There is
still the concept of idle load balancer (performing the load balancing
on behalf of all the idle cpu's in the system). Busy CPU kicks the nohz
balancer when any of the nohz CPUs need idle load balancing.
The kickee CPU does the idle load balancing on behalf of all idle CPUs
instead of the normal idle balance.
This addresses the below two problems with the current nohz ilb logic:
* the idle load balancer continued to have periodic ticks during idle and
wokeup frequently, even though it did not have any rebalancing to do on
behalf of any of the idle CPUs.
* On x86 and CPUs that have APIC timer stoppage on idle CPUs, this
periodic wakeup can result in a periodic additional interrupt on a CPU
doing the timer broadcast.
Also currently we are migrating the unpinned timers from an idle to the cpu
doing idle load balancing (when all the cpus in the system are idle,
there is no idle load balancing cpu and timers get added to the same idle cpu
where the request was made. So the existing optimization works only on semi idle
system).
And In semi idle system, we no longer have periodic ticks on the idle load
balancer CPU. Using that cpu will add more delays to the timers than intended
(as that cpu's timer base may not be uptodate wrt jiffies etc). This was
causing mysterious slowdowns during boot etc.
For now, in the semi idle case, use the nearest busy cpu for migrating timers
from an idle cpu. This is good for power-savings anyway.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <1274486981.2840.46.camel@sbs-t61.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
tickless idle has a negative side effect on update_cpu_load(), which
in turn can affect load balancing behavior.
update_cpu_load() is supposed to be called every tick, to keep track
of various load indicies. With tickless idle, there are no scheduler
ticks called on the idle CPUs. Idle CPUs may still do load balancing
(with idle_load_balance CPU) using the stale cpu_load. It will also
cause problems when all CPUs go idle for a while and become active
again. In this case loads would not degrade as expected.
This is how rq->nr_load_updates change looks like under different
conditions:
<cpu_num> <nr_load_updates change>
All CPUS idle for 10 seconds (HZ=1000)
0 1621
10 496
11 139
12 875
13 1672
14 12
15 21
1 1472
2 2426
3 1161
4 2108
5 1525
6 701
7 249
8 766
9 1967
One CPU busy rest idle for 10 seconds
0 10003
10 601
11 95
12 966
13 1597
14 114
15 98
1 3457
2 93
3 6679
4 1425
5 1479
6 595
7 193
8 633
9 1687
All CPUs busy for 10 seconds
0 10026
10 10026
11 10026
12 10026
13 10025
14 10025
15 10025
1 10026
2 10026
3 10026
4 10026
5 10026
6 10026
7 10026
8 10026
9 10026
That is update_cpu_load works properly only when all CPUs are busy.
If all are idle, all the CPUs get way lower updates. And when few
CPUs are busy and rest are idle, only busy and ilb CPU does proper
updates and rest of the idle CPUs will do lower updates.
The patch keeps track of when a last update was done and fixes up
the load avg based on current time.
On one of my test system SPECjbb with warehouse 1..numcpus, patch
improves throughput numbers by ~1% (average of 6 runs). On another
test system (with different domain hierarchy) there is no noticable
change in perf.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <AANLkTilLtDWQsAUrIxJ6s04WTgmw9GuOODc5AOrYsaR5@mail.gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
- Contrary to what 6d558c3a says, there is no need to reload
prev = rq->curr after the context switch. You always schedule
back to where you came from, prev must be equal to current
even if cpu/rq was changed.
- This also means reacquire_kernel_lock() can use prev instead
of current.
- No need to reassign switch_count if reacquire_kernel_lock()
reports need_resched(), we can just move the initial assignment
down, under the "need_resched_nonpreemptible:" label.
- Try to update the comment after context_switch().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20100519125711.GA30199@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
For people who otherwise get to write: cpu_clock(smp_processor_id()),
there is now: local_clock().
Also, as per suggestion from Andrew, provide some documentation on
the various clock interfaces, and minimize the unsigned long long vs
u64 mess.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Jens Axboe <jaxboe@fusionio.com>
LKML-Reference: <1275052414.1645.52.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>