This patch touches the RT group scheduling case.
Functions inc_rt_prio_smp() and dec_rt_prio_smp() change (global) rq's
priority, while rt_rq passed to them may be not the top-level rt_rq.
This is wrong, because changing of priority on a child level does not
guarantee that the priority is the highest all over the rq. So, this
leak makes RT balancing unusable.
The short example: the task having the highest priority among all rq's
RT tasks (no one other task has the same priority) are waking on a
throttle rt_rq. The rq's cpupri is set to the task's priority
equivalent, but real rq->rt.highest_prio.curr is less.
The patch below fixes the problem.
Signed-off-by: Kirill Tkhai <tkhai@yandex.ru>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
CC: Steven Rostedt <rostedt@goodmis.org>
CC: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/49231385567953@web4m.yandex.ru
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit 42eb088e (sched: Avoid NULL dereference on sd_busy) corrected a NULL
dereference on sd_busy but the fix also altered what scheduling domain it
used for the 'sd_llc' percpu variable.
One impact of this is that a task selecting a runqueue may consider
idle CPUs that are not cache siblings as candidates for running.
Tasks are then running on CPUs that are not cache hot.
This was found through bisection where ebizzy threads were not seeing equal
performance and it looked like a scheduling fairness issue. This patch
mitigates but does not completely fix the problem on all machines tested
implying there may be an additional bug or a common root cause. Here are
the average range of performance seen by individual ebizzy threads. It
was tested on top of candidate patches related to x86 TLB range flushing.
4-core machine
3.13.0-rc3 3.13.0-rc3
vanilla fixsd-v3r3
Mean 1 0.00 ( 0.00%) 0.00 ( 0.00%)
Mean 2 0.34 ( 0.00%) 0.10 ( 70.59%)
Mean 3 1.29 ( 0.00%) 0.93 ( 27.91%)
Mean 4 7.08 ( 0.00%) 0.77 ( 89.12%)
Mean 5 193.54 ( 0.00%) 2.14 ( 98.89%)
Mean 6 151.12 ( 0.00%) 2.06 ( 98.64%)
Mean 7 115.38 ( 0.00%) 2.04 ( 98.23%)
Mean 8 108.65 ( 0.00%) 1.92 ( 98.23%)
8-core machine
Mean 1 0.00 ( 0.00%) 0.00 ( 0.00%)
Mean 2 0.40 ( 0.00%) 0.21 ( 47.50%)
Mean 3 23.73 ( 0.00%) 0.89 ( 96.25%)
Mean 4 12.79 ( 0.00%) 1.04 ( 91.87%)
Mean 5 13.08 ( 0.00%) 2.42 ( 81.50%)
Mean 6 23.21 ( 0.00%) 69.46 (-199.27%)
Mean 7 15.85 ( 0.00%) 101.72 (-541.77%)
Mean 8 109.37 ( 0.00%) 19.13 ( 82.51%)
Mean 12 124.84 ( 0.00%) 28.62 ( 77.07%)
Mean 16 113.50 ( 0.00%) 24.16 ( 78.71%)
It's eliminated for one machine and reduced for another.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Alex Shi <alex.shi@linaro.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: H Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/20131217092124.GV11295@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Christian suffers from a bad BIOS that wrecks his i5's TSC sync. This
results in him occasionally seeing time going backwards - which
crashes the scheduler ...
Most of our time accounting can actually handle that except the most
common one; the tick time update of sched_fair.
There is a further problem with that code; previously we assumed that
because we get a tick every TICK_NSEC our time delta could never
exceed 32bits and math was simpler.
However, ever since Frederic managed to get NO_HZ_FULL merged; this is
no longer the case since now a task can run for a long time indeed
without getting a tick. It only takes about ~4.2 seconds to overflow
our u32 in nanoseconds.
This means we not only need to better deal with time going backwards;
but also means we need to be able to deal with large deltas.
This patch reworks the entire code and uses mul_u64_u32_shr() as
proposed by Andy a long while ago.
We express our virtual time scale factor in a u32 multiplier and shift
right and the 32bit mul_u64_u32_shr() implementation reduces to a
single 32x32->64 multiply if the time delta is still short (common
case).
For 64bit a 64x64->128 multiply can be used if ARCH_SUPPORTS_INT128.
Reported-and-Tested-by: Christian Engelmayer <cengelma@gmx.at>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: fweisbec@gmail.com
Cc: Paul Turner <pjt@google.com>
Cc: Stanislaw Gruszka <sgruszka@redhat.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20131118172706.GI3866@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Yinghai reported that he saw a /0 in sg_capacity on his EX parts.
Make sure to always initialize power_orig now that we actually use it.
Ideally build_sched_domains() -> init_sched_groups_power() would also
initialize this; but for some yet unexplained reason some setups seem
to miss updates there.
Reported-by: Yinghai Lu <yinghai@kernel.org>
Tested-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-l8ng2m9uml6fhibln8wqpom7@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In preparation of conversion to kernfs, cgroup file handling is
updated so that it can be easily mapped to kernfs. This patch
replaces cftype->read_seq_string() with cftype->seq_show() which is
not limited to single_open() operation and will map directcly to
kernfs seq_file interface.
The conversions are mechanical. As ->seq_show() doesn't have @css and
@cft, the functions which make use of them are converted to use
seq_css() and seq_cft() respectively. In several occassions, e.f. if
it has seq_string in its name, the function name is updated to fit the
new method better.
This patch does not introduce any behavior changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Aristeu Rozanski <arozansk@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Acked-by: Daniel Wagner <daniel.wagner@bmw-carit.de>
Acked-by: Li Zefan <lizefan@huawei.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Neil Horman <nhorman@tuxdriver.com>
In preparation of conversion to kernfs, cgroup file handling is being
consolidated so that it can be easily mapped to the seq_file based
interface of kernfs.
cftype->read_map() doesn't add any value and being replaced with
->read_seq_string(). Update cpu_stats_show() and cpuacct_stats_show()
accordingly.
This patch doesn't make any visible behavior changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Li Zefan <lizefan@huawei.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Tony reported that aa0d532605 ("ia64: Use preempt_schedule_irq")
broke PREEMPT=n builds on ia64.
Ok, wrapped my brain around it. I tripped over the magic asm foo which
has a single need_resched check and schedule point for both sys call
return and interrupt return.
So you need the schedule_preempt_irq() for kernel preemption from
interrupt return while on a normal syscall preemption a schedule would
be sufficient. But using schedule_preempt_irq() is not harmful here in
any way. It just sets the preempt_active bit also in cases where it
would not be required.
Even on preempt=n kernels adding the preempt_active bit is completely
harmless. So instead of having an extra function, moving the existing
one out of the ifdef PREEMPT looks like the sanest thing to do.
It would also allow getting rid of various other sti/schedule/cli asm
magic in other archs.
Reported-and-Tested-by: Tony Luck <tony.luck@gmail.com>
Fixes: aa0d532605 ("ia64: Use preempt_schedule_irq")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
[slightly edited Changelog]
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/alpine.DEB.2.02.1311211230030.30673@ionos.tec.linutronix.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Because we're completely unserialized against hotplug its well
possible to try and generate numa stats for an offlined node.
Bail out early (and avoid a /0) in this case. The resulting stats are
all 0 which should result in an undesirable balance target -- not to
mention that actually trying to migrate to an offline CPU will fail.
Reported-by: Prarit Bhargava <prarit@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Link: http://lkml.kernel.org/n/tip-orja0qylcvyhxfsuebcyL5sI@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The cpusets code can split up the scheduler's domain tree into
smaller domains. Some of those smaller domains may not cross
NUMA nodes at all, leading to a NULL pointer dereference on the
per-cpu sd_numa pointer.
Tasks cannot be migrated out of their domain, so the patch
also sets p->numa_preferred_nid to whereever they are, to
prevent the migration from being retried over and over again.
Reported-by: Prarit Bhargava <prarit@redhat.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Link: http://lkml.kernel.org/n/tip-oosqomw0Jput0Jkvoowhrqtu@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
nr_busy_cpus parameter is used by nohz_kick_needed() to find out the
number of busy cpus in a sched domain which has SD_SHARE_PKG_RESOURCES
flag set. Therefore instead of updating nr_busy_cpus at every level
of sched domain, since it is irrelevant, we can update this parameter
only at the parent domain of the sd which has this flag set. Introduce
a per-cpu parameter sd_busy which represents this parent domain.
In nohz_kick_needed() we directly query the nr_busy_cpus parameter
associated with the groups of sd_busy.
By associating sd_busy with the highest domain which has
SD_SHARE_PKG_RESOURCES flag set, we cover all lower level domains
which could have this flag set and trigger nohz_idle_balancing if any
of the levels have more than one busy cpu.
sd_busy is irrelevant for asymmetric load balancing. However sd_asym
has been introduced to represent the highest sched domain which has
SD_ASYM_PACKING flag set so that it can be queried directly when
required.
While we are at it, we might as well change the nohz_idle parameter to
be updated at the sd_busy domain level alone and not the base domain
level of a CPU. This will unify the concept of busy cpus at just one
level of sched domain where it is currently used.
Signed-off-by: Preeti U Murthy<preeti@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: svaidy@linux.vnet.ibm.com
Cc: vincent.guittot@linaro.org
Cc: bitbucket@online.de
Cc: benh@kernel.crashing.org
Cc: anton@samba.org
Cc: Morten.Rasmussen@arm.com
Cc: pjt@google.com
Cc: peterz@infradead.org
Cc: mikey@neuling.org
Link: http://lkml.kernel.org/r/20131030031252.23426.4417.stgit@preeti.in.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
throttle_cfs_rq() doesn't check to make sure that period_timer is running,
and while update_curr/assign_cfs_runtime does, a concurrently running
period_timer on another cpu could cancel itself between this cpu's
update_curr and throttle_cfs_rq(). If there are no other cfs_rqs running
in the tg to restart the timer, this causes the cfs_rq to be stranded
forever.
Fix this by calling __start_cfs_bandwidth() in throttle if the timer is
inactive.
(Also add some sched_debug lines for cfs_bandwidth.)
Tested: make a run/sleep task in a cgroup, loop switching the cgroup
between 1ms/100ms quota and unlimited, checking for timer_active=0 and
throttled=1 as a failure. With the throttle_cfs_rq() change commented out
this fails, with the full patch it passes.
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181632.22647.84174.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently, group entity load-weights are initialized to zero. This
admits some races with respect to the first time they are re-weighted in
earlty use. ( Let g[x] denote the se for "g" on cpu "x". )
Suppose that we have root->a and that a enters a throttled state,
immediately followed by a[0]->t1 (the only task running on cpu[0])
blocking:
put_prev_task(group_cfs_rq(a[0]), t1)
put_prev_entity(..., t1)
check_cfs_rq_runtime(group_cfs_rq(a[0]))
throttle_cfs_rq(group_cfs_rq(a[0]))
Then, before unthrottling occurs, let a[0]->b[0]->t2 wake for the first
time:
enqueue_task_fair(rq[0], t2)
enqueue_entity(group_cfs_rq(b[0]), t2)
enqueue_entity_load_avg(group_cfs_rq(b[0]), t2)
account_entity_enqueue(group_cfs_ra(b[0]), t2)
update_cfs_shares(group_cfs_rq(b[0]))
< skipped because b is part of a throttled hierarchy >
enqueue_entity(group_cfs_rq(a[0]), b[0])
...
We now have b[0] enqueued, yet group_cfs_rq(a[0])->load.weight == 0
which violates invariants in several code-paths. Eliminate the
possibility of this by initializing group entity weight.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20131016181627.22647.47543.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
__start_cfs_bandwidth calls hrtimer_cancel while holding rq->lock,
waiting for the hrtimer to finish. However, if sched_cfs_period_timer
runs for another loop iteration, the hrtimer can attempt to take
rq->lock, resulting in deadlock.
Fix this by ensuring that cfs_b->timer_active is cleared only if the
_latest_ call to do_sched_cfs_period_timer is returning as idle. Then
__start_cfs_bandwidth can just call hrtimer_try_to_cancel and wait for
that to succeed or timer_active == 1.
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181622.22647.16643.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When we transition cfs_bandwidth_used to false, any currently
throttled groups will incorrectly return false from cfs_rq_throttled.
While tg_set_cfs_bandwidth will unthrottle them eventually, currently
running code (including at least dequeue_task_fair and
distribute_cfs_runtime) will cause errors.
Fix this by turning off cfs_bandwidth_used only after unthrottling all
cfs_rqs.
Tested: toggle bandwidth back and forth on a loaded cgroup. Caused
crashes in minutes without the patch, hasn't crashed with it.
Signed-off-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181611.22647.80365.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Remove get_online_cpus() usage from the scheduler; there's 4 sites that
use it:
- sched_init_smp(); where its completely superfluous since we're in
'early' boot and there simply cannot be any hotplugging.
- sched_getaffinity(); we already take a raw spinlock to protect the
task cpus_allowed mask, this disables preemption and therefore
also stabilizes cpu_online_mask as that's modified using
stop_machine. However switch to active mask for symmetry with
sched_setaffinity()/set_cpus_allowed_ptr(). We guarantee active
mask stability by inserting sync_rcu/sched() into _cpu_down.
- sched_setaffinity(); we don't appear to need get_online_cpus()
either, there's two sites where hotplug appears relevant:
* cpuset_cpus_allowed(); for the !cpuset case we use possible_mask,
for the cpuset case we hold task_lock, which is a spinlock and
thus for mainline disables preemption (might cause pain on RT).
* set_cpus_allowed_ptr(); Holds all scheduler locks and thus has
preemption properly disabled; also it already deals with hotplug
races explicitly where it releases them.
- migrate_swap(); we can make stop_two_cpus() do the heavy lifting for
us with a little trickery. By adding a sync_sched/rcu() after the
CPU_DOWN_PREPARE notifier we can provide preempt/rcu guarantees for
cpu_active_mask. Use these to validate that both our cpus are active
when queueing the stop work before we queue the stop_machine works
for take_cpu_down().
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com>
Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20131011123820.GV3081@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
There is a subtle race in migrate_swap, when task P, on CPU A, decides to swap
places with task T, on CPU B.
Task P:
- call migrate_swap
Task T:
- go to sleep, removing itself from the runqueue
Task P:
- double lock the runqueues on CPU A & B
Task T:
- get woken up, place itself on the runqueue of CPU C
Task P:
- see that task T is on a runqueue, and pretend to remove it
from the runqueue on CPU B
Now CPUs B & C both have corrupted scheduler data structures.
This patch fixes it, by holding the pi_lock for both of the tasks
involved in the migrate swap. This prevents task T from waking up,
and placing itself onto another runqueue, until after migrate_swap
has released all locks.
This means that, when migrate_swap checks, task T will be either
on the runqueue where it was originally seen, or not on any
runqueue at all. Migrate_swap deals correctly with of those cases.
Tested-by: Joe Mario <jmario@redhat.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: hannes@cmpxchg.org
Cc: aarcange@redhat.com
Cc: srikar@linux.vnet.ibm.com
Cc: tglx@linutronix.de
Cc: hpa@zytor.com
Link: http://lkml.kernel.org/r/20131010181722.GO13848@laptop.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reflow the function a bit because GCC gets confused:
kernel/sched/fair.c: In function ‘task_numa_fault’:
kernel/sched/fair.c:1448:3: warning: ‘my_grp’ may be used uninitialized in this function [-Wmaybe-uninitialized]
kernel/sched/fair.c:1463:27: note: ‘my_grp’ was declared here
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-6ebt6x7u64pbbonq1khqu2z9@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Short spikes of CPU load can lead to a task being migrated
away from its preferred node for temporary reasons.
It is important that the task is migrated back to where it
belongs, in order to avoid migrating too much memory to its
new location, and generally disturbing a task's NUMA location.
This patch fixes NUMA placement for 4 specjbb instances on
a 4 node system. Without this patch, things take longer to
converge, and processes are not always completely on their
own node.
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-64-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Shared faults can lead to lots of unnecessary page migrations,
slowing down the system, and causing private faults to hit the
per-pgdat migration ratelimit.
This patch adds sysctl numa_balancing_migrate_deferred, which specifies
how many shared page migrations to skip unconditionally, after each page
migration that is skipped because it is a shared fault.
This reduces the number of page migrations back and forth in
shared fault situations. It also gives a strong preference to
the tasks that are already running where most of the memory is,
and to moving the other tasks to near the memory.
Testing this with a much higher scan rate than the default
still seems to result in fewer page migrations than before.
Memory seems to be somewhat better consolidated than previously,
with multi-instance specjbb runs on a 4 node system.
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-62-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Adjust numa_scan_period in task_numa_placement, depending on how much
useful work the numa code can do. The more local faults there are in a
given scan window the longer the period (and hence the slower the scan rate)
during the next window. If there are excessive shared faults then the scan
period will decrease with the amount of scaling depending on whether the
ratio of shared/private faults. If the preferred node changes then the
scan rate is reset to recheck if the task is properly placed.
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-59-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
