Two new statistics are introduced to show the internal of burst feature
and explain why burst helps or not.
nr_bursts: number of periods bandwidth burst occurs
burst_time: cumulative wall-time (in nanoseconds) that any cpus has
used above quota in respective periods
Co-developed-by: Shanpei Chen <shanpeic@linux.alibaba.com>
Signed-off-by: Shanpei Chen <shanpeic@linux.alibaba.com>
Co-developed-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Signed-off-by: Huaixin Chang <changhuaixin@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20210830032215.16302-2-changhuaixin@linux.alibaba.com
Give reduced sleeper credit to SCHED_IDLE entities. As a result, woken
SCHED_IDLE entities will take longer to preempt normal entities.
The benefit of this change is to make it less likely that a newly woken
SCHED_IDLE entity will preempt a short-running normal entity before it
blocks.
We still give a small sleeper credit to SCHED_IDLE entities, so that
idle<->idle competition retains some fairness.
Example: With HZ=1000, spawned four threads affined to one cpu, one of
which was set to SCHED_IDLE. Without this patch, wakeup latency for the
SCHED_IDLE thread was ~1-2ms, with the patch the wakeup latency was
~5ms.
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Jiang Biao <benbjiang@tencent.com>
Link: https://lore.kernel.org/r/20210820010403.946838-5-joshdon@google.com
Use a small, non-scaled min granularity for SCHED_IDLE entities, when
competing with normal entities. This reduces the latency of getting
a normal entity back on cpu, at the expense of increased context
switch frequency of SCHED_IDLE entities.
The benefit of this change is to reduce the round-robin latency for
normal entities when competing with a SCHED_IDLE entity.
Example: on a machine with HZ=1000, spawned two threads, one of which is
SCHED_IDLE, and affined to one cpu. Without this patch, the SCHED_IDLE
thread runs for 4ms then waits for 1.4s. With this patch, it runs for
1ms and waits 340ms (as it round-robins with the other thread).
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20210820010403.946838-4-joshdon@google.com
Adds cfs_rq->idle_nr_running, which accounts the number of idle entities
directly enqueued on the cfs_rq.
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20210820010403.946838-3-joshdon@google.com
Consider a system with some NOHZ-idle CPUs, such that
nohz.idle_cpus_mask = S
nohz.next_balance = T
When a new CPU k goes NOHZ idle (nohz_balance_enter_idle()), we end up
with:
nohz.idle_cpus_mask = S \U {k}
nohz.next_balance = T
Note that the nohz.next_balance hasn't changed - it won't be updated until
a NOHZ balance is triggered. This is problematic if the newly NOHZ idle CPU
has an earlier rq.next_balance than the other NOHZ idle CPUs, IOW if:
cpu_rq(k).next_balance < nohz.next_balance
In such scenarios, the existing nohz.next_balance will prevent any NOHZ
balance from happening, which itself will prevent nohz.next_balance from
being updated to this new cpu_rq(k).next_balance. Unnecessary load balance
delays of over 12ms caused by this were observed on an arm64 RB5 board.
Use the new nohz.needs_update flag to mark the presence of newly-idle CPUs
that need their rq->next_balance to be collated into
nohz.next_balance. Trigger a NOHZ_NEXT_KICK when the flag is set.
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210823111700.2842997-3-valentin.schneider@arm.com
A following patch will trigger NOHZ idle balances as a means to update
nohz.next_balance. Vincent noted that blocked load updates can have
non-negligible overhead, which should be avoided if the intent is to only
update nohz.next_balance.
Add a new NOHZ balance kick flag, NOHZ_NEXT_KICK. Gate NOHZ blocked load
update by the presence of NOHZ_STATS_KICK - currently all NOHZ balance
kicks will have the NOHZ_STATS_KICK flag set, so no change in behaviour is
expected.
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210823111700.2842997-2-valentin.schneider@arm.com
Since commit a7b359fc6a37 ("sched/fair: Correctly insert cfs_rq's to
list on unthrottle") we add cfs_rqs with no runnable tasks but not fully
decayed into the load (leaf) list. We may ignore adding some ancestors
and therefore breaking tmp_alone_branch invariant. This broke LTP test
cfs_bandwidth01 and it was partially fixed in commit fdaba61ef8a2
("sched/fair: Ensure that the CFS parent is added after unthrottling").
I noticed the named test still fails even with the fix (but with low
probability, 1 in ~1000 executions of the test). The reason is when
bailing out of unthrottle_cfs_rq early, we may miss adding ancestors of
the unthrottled cfs_rq, thus, not joining tmp_alone_branch properly.
Fix this by adding ancestors if we notice the unthrottled cfs_rq was
added to the load list.
Fixes: a7b359fc6a37 ("sched/fair: Correctly insert cfs_rq's to list on unthrottle")
Signed-off-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Odin Ugedal <odin@uged.al>
Link: https://lore.kernel.org/r/20210917153037.11176-1-mkoutny@suse.com
- The biggest change in this cycle is scheduler support for asymmetric
scheduling affinity, to support the execution of legacy 32-bit tasks on
AArch32 systems that also have 64-bit-only CPUs.
Architectures can fill in this functionality by defining their
own task_cpu_possible_mask(p). When this is done, the scheduler will
make sure the task will only be scheduled on CPUs that support it.
(The actual arm64 specific changes are not part of this tree.)
For other architectures there will be no change in functionality.
- Add cgroup SCHED_IDLE support
- Increase node-distance flexibility & delay determining it until a CPU
is brought online. (This enables platforms where node distance isn't
final until the CPU is only.)
- Deadline scheduler enhancements & fixes
- Misc fixes & cleanups.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=VdGE
-----END PGP SIGNATURE-----
Merge tag 'sched-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler updates from Ingo Molnar:
- The biggest change in this cycle is scheduler support for asymmetric
scheduling affinity, to support the execution of legacy 32-bit tasks
on AArch32 systems that also have 64-bit-only CPUs.
Architectures can fill in this functionality by defining their own
task_cpu_possible_mask(p). When this is done, the scheduler will make
sure the task will only be scheduled on CPUs that support it.
(The actual arm64 specific changes are not part of this tree.)
For other architectures there will be no change in functionality.
- Add cgroup SCHED_IDLE support
- Increase node-distance flexibility & delay determining it until a CPU
is brought online. (This enables platforms where node distance isn't
final until the CPU is only.)
- Deadline scheduler enhancements & fixes
- Misc fixes & cleanups.
* tag 'sched-core-2021-08-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (27 commits)
eventfd: Make signal recursion protection a task bit
sched/fair: Mark tg_is_idle() an inline in the !CONFIG_FAIR_GROUP_SCHED case
sched: Introduce dl_task_check_affinity() to check proposed affinity
sched: Allow task CPU affinity to be restricted on asymmetric systems
sched: Split the guts of sched_setaffinity() into a helper function
sched: Introduce task_struct::user_cpus_ptr to track requested affinity
sched: Reject CPU affinity changes based on task_cpu_possible_mask()
cpuset: Cleanup cpuset_cpus_allowed_fallback() use in select_fallback_rq()
cpuset: Honour task_cpu_possible_mask() in guarantee_online_cpus()
cpuset: Don't use the cpu_possible_mask as a last resort for cgroup v1
sched: Introduce task_cpu_possible_mask() to limit fallback rq selection
sched: Cgroup SCHED_IDLE support
sched/topology: Skip updating masks for non-online nodes
sched: Replace deprecated CPU-hotplug functions.
sched: Skip priority checks with SCHED_FLAG_KEEP_PARAMS
sched: Fix UCLAMP_FLAG_IDLE setting
sched/deadline: Fix missing clock update in migrate_task_rq_dl()
sched/fair: Avoid a second scan of target in select_idle_cpu
sched/fair: Use prev instead of new target as recent_used_cpu
sched: Don't report SCHED_FLAG_SUGOV in sched_getattr()
...
It's not actually used in the !CONFIG_FAIR_GROUP_SCHED case:
kernel/sched/fair.c:488:12: warning: ‘tg_is_idle’ defined but not used [-Wunused-function]
Keep around a placeholder nevertheless, for API completeness. Mark it inline,
so the compiler doesn't think it must be used.
Fixes: 304000390f88: ("sched: Cgroup SCHED_IDLE support")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Josh Don <joshdon@google.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
This extends SCHED_IDLE to cgroups.
Interface: cgroup/cpu.idle.
0: default behavior
1: SCHED_IDLE
Extending SCHED_IDLE to cgroups means that we incorporate the existing
aspects of SCHED_IDLE; a SCHED_IDLE cgroup will count all of its
descendant threads towards the idle_h_nr_running count of all of its
ancestor cgroups. Thus, sched_idle_rq() will work properly.
Additionally, SCHED_IDLE cgroups are configured with minimum weight.
There are two key differences between the per-task and per-cgroup
SCHED_IDLE interface:
- The cgroup interface allows tasks within a SCHED_IDLE hierarchy to
maintain their relative weights. The entity that is "idle" is the
cgroup, not the tasks themselves.
- Since the idle entity is the cgroup, our SCHED_IDLE wakeup preemption
decision is not made by comparing the current task with the woken
task, but rather by comparing their matching sched_entity.
A typical use-case for this is a user that creates an idle and a
non-idle subtree. The non-idle subtree will dominate competition vs
the idle subtree, but the idle subtree will still be high priority vs
other users on the system. The latter is accomplished via comparing
matching sched_entity in the waken preemption path (this could also be
improved by making the sched_idle_rq() decision dependent on the
perspective of a specific task).
For now, we maintain the existing SCHED_IDLE semantics. Future patches
may make improvements that extend how we treat SCHED_IDLE entities.
The per-task_group idle field is an integer that currently only holds
either a 0 or a 1. This is explicitly typed as an integer to allow for
further extensions to this API. For example, a negative value may
indicate a highly latency-sensitive cgroup that should be preferred
for preemption/placement/etc.
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20210730020019.1487127-2-joshdon@google.com
When select_idle_cpu starts scanning for an idle CPU, it starts with
a target CPU that has already been checked by select_idle_sibling.
This patch starts with the next CPU instead.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210804115857.6253-3-mgorman@techsingularity.net
After select_idle_sibling, p->recent_used_cpu is set to the
new target. However on the next wakeup, prev will be the same as
recent_used_cpu unless the load balancer has moved the task since the
last wakeup. It still works, but is less efficient than it could be.
This patch preserves recent_used_cpu for longer.
The impact on SIS efficiency is tiny so the SIS statistic patches were
used to track the hit rate for using recent_used_cpu. With perf bench
pipe on a 2-socket Cascadelake machine, the hit rate went from 57.14%
to 85.32%. For more intensive wakeup loads like hackbench, the hit rate
is almost negligible but rose from 0.21% to 6.64%. For scaling loads
like tbench, the hit rate goes from almost 0% to 25.42% overall. Broadly
speaking, on tbench, the success rate is much higher for lower thread
counts and drops to almost 0 as the workload scales to towards saturation.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210804115857.6253-2-mgorman@techsingularity.net
Use the loop variable instead of the function argument to test the
other SMT siblings for idle.
Fixes: ff7db0bf24db ("sched/numa: Prefer using an idle CPU as a migration target instead of comparing tasks")
Signed-off-by: Mika Penttilä <mika.penttila@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Pankaj Gupta <pankaj.gupta@ionos.com>
Link: https://lkml.kernel.org/r/20210722063946.28951-1-mika.penttila@gmail.com
The time remaining until expiry of the refresh_timer can be negative.
Casting the type to an unsigned 64-bit value will cause integer
underflow, making the runtime_refresh_within return false instead of
true. These situations are rare, but they do happen.
This does not cause user-facing issues or errors; other than
possibly unthrottling cfs_rq's using runtime from the previous period(s),
making the CFS bandwidth enforcement less strict in those (special)
situations.
Signed-off-by: Odin Ugedal <odin@uged.al>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Link: https://lore.kernel.org/r/20210629121452.18429-1-odin@uged.al
commit 9e077b52d86a ("sched/pelt: Check that *_avg are null when *_sum are")
reported some inconsitencies between *_avg and *_sum.
commit 1c35b07e6d39 ("sched/fair: Ensure _sum and _avg values stay consistent")
fixed some but one remains when dequeuing load.
sync the cfs's load_sum with its load_avg after dequeuing the load of a
sched_entity.
Fixes: 9e077b52d86a ("sched/pelt: Check that *_avg are null when *_sum are")
Reported-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Odin Ugedal <odin@uged.al>
Tested-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Link: https://lore.kernel.org/r/20210701171837.32156-1-vincent.guittot@linaro.org
new warning that several people reported.
- Flip CONFIG_SCHED_CORE to default-disabled, and update the
Kconfig help text.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=Jgwc
-----END PGP SIGNATURE-----
Merge tag 'sched-urgent-2021-06-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
- Fix a small inconsistency (bug) in load tracking, caught by a new
warning that several people reported.
- Flip CONFIG_SCHED_CORE to default-disabled, and update the Kconfig
help text.
* tag 'sched-urgent-2021-06-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/core: Disable CONFIG_SCHED_CORE by default
sched/fair: Ensure _sum and _avg values stay consistent
- Changes to core scheduling facilities:
- Add "Core Scheduling" via CONFIG_SCHED_CORE=y, which enables
coordinated scheduling across SMT siblings. This is a much
requested feature for cloud computing platforms, to allow
the flexible utilization of SMT siblings, without exposing
untrusted domains to information leaks & side channels, plus
to ensure more deterministic computing performance on SMT
systems used by heterogenous workloads.
There's new prctls to set core scheduling groups, which
allows more flexible management of workloads that can share
siblings.
- Fix task->state access anti-patterns that may result in missed
wakeups and rename it to ->__state in the process to catch new
abuses.
- Load-balancing changes:
- Tweak newidle_balance for fair-sched, to improve
'memcache'-like workloads.
- "Age" (decay) average idle time, to better track & improve workloads
such as 'tbench'.
- Fix & improve energy-aware (EAS) balancing logic & metrics.
- Fix & improve the uclamp metrics.
- Fix task migration (taskset) corner case on !CONFIG_CPUSET.
- Fix RT and deadline utilization tracking across policy changes
- Introduce a "burstable" CFS controller via cgroups, which allows
bursty CPU-bound workloads to borrow a bit against their future
quota to improve overall latencies & batching. Can be tweaked
via /sys/fs/cgroup/cpu/<X>/cpu.cfs_burst_us.
- Rework assymetric topology/capacity detection & handling.
- Scheduler statistics & tooling:
- Disable delayacct by default, but add a sysctl to enable
it at runtime if tooling needs it. Use static keys and
other optimizations to make it more palatable.
- Use sched_clock() in delayacct, instead of ktime_get_ns().
- Misc cleanups and fixes.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=3VDr
-----END PGP SIGNATURE-----
Merge tag 'sched-core-2021-06-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler udpates from Ingo Molnar:
- Changes to core scheduling facilities:
- Add "Core Scheduling" via CONFIG_SCHED_CORE=y, which enables
coordinated scheduling across SMT siblings. This is a much
requested feature for cloud computing platforms, to allow the
flexible utilization of SMT siblings, without exposing untrusted
domains to information leaks & side channels, plus to ensure more
deterministic computing performance on SMT systems used by
heterogenous workloads.
There are new prctls to set core scheduling groups, which allows
more flexible management of workloads that can share siblings.
- Fix task->state access anti-patterns that may result in missed
wakeups and rename it to ->__state in the process to catch new
abuses.
- Load-balancing changes:
- Tweak newidle_balance for fair-sched, to improve 'memcache'-like
workloads.
- "Age" (decay) average idle time, to better track & improve
workloads such as 'tbench'.
- Fix & improve energy-aware (EAS) balancing logic & metrics.
- Fix & improve the uclamp metrics.
- Fix task migration (taskset) corner case on !CONFIG_CPUSET.
- Fix RT and deadline utilization tracking across policy changes
- Introduce a "burstable" CFS controller via cgroups, which allows
bursty CPU-bound workloads to borrow a bit against their future
quota to improve overall latencies & batching. Can be tweaked via
/sys/fs/cgroup/cpu/<X>/cpu.cfs_burst_us.
- Rework assymetric topology/capacity detection & handling.
- Scheduler statistics & tooling:
- Disable delayacct by default, but add a sysctl to enable it at
runtime if tooling needs it. Use static keys and other
optimizations to make it more palatable.
- Use sched_clock() in delayacct, instead of ktime_get_ns().
- Misc cleanups and fixes.
* tag 'sched-core-2021-06-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (72 commits)
sched/doc: Update the CPU capacity asymmetry bits
sched/topology: Rework CPU capacity asymmetry detection
sched/core: Introduce SD_ASYM_CPUCAPACITY_FULL sched_domain flag
psi: Fix race between psi_trigger_create/destroy
sched/fair: Introduce the burstable CFS controller
sched/uclamp: Fix uclamp_tg_restrict()
sched/rt: Fix Deadline utilization tracking during policy change
sched/rt: Fix RT utilization tracking during policy change
sched: Change task_struct::state
sched,arch: Remove unused TASK_STATE offsets
sched,timer: Use __set_current_state()
sched: Add get_current_state()
sched,perf,kvm: Fix preemption condition
sched: Introduce task_is_running()
sched: Unbreak wakeups
sched/fair: Age the average idle time
sched/cpufreq: Consider reduced CPU capacity in energy calculation
sched/fair: Take thermal pressure into account while estimating energy
thermal/cpufreq_cooling: Update offline CPUs per-cpu thermal_pressure
sched/fair: Return early from update_tg_cfs_load() if delta == 0
...
On a 128 cores AMD machine, there are 8 cores in nohz_full mode, and
the others are used for housekeeping. When many housekeeping cpus are
in idle state, we can observe huge time burn in the loop for searching
nearest busy housekeeper cpu by ftrace.
9) | get_nohz_timer_target() {
9) | housekeeping_test_cpu() {
9) 0.390 us | housekeeping_get_mask.part.1();
9) 0.561 us | }
9) 0.090 us | __rcu_read_lock();
9) 0.090 us | housekeeping_cpumask();
9) 0.521 us | housekeeping_cpumask();
9) 0.140 us | housekeeping_cpumask();
...
9) 0.500 us | housekeeping_cpumask();
9) | housekeeping_any_cpu() {
9) 0.090 us | housekeeping_get_mask.part.1();
9) 0.100 us | sched_numa_find_closest();
9) 0.491 us | }
9) 0.100 us | __rcu_read_unlock();
9) + 76.163 us | }
for_each_cpu_and() is a micro function, so in get_nohz_timer_target()
function the
for_each_cpu_and(i, sched_domain_span(sd),
housekeeping_cpumask(HK_FLAG_TIMER))
equals to below:
for (i = -1; i = cpumask_next_and(i, sched_domain_span(sd),
housekeeping_cpumask(HK_FLAG_TIMER)), i < nr_cpu_ids;)
That will cause that housekeeping_cpumask() will be invoked many times.
The housekeeping_cpumask() function returns a const value, so it is
unnecessary to invoke it every time. This patch can minimize the worst
searching time from ~76us to ~16us in my testing.
Similarly, the find_new_ilb() function has the same problem.
Co-developed-by: Li RongQing <lirongqing@baidu.com>
Signed-off-by: Li RongQing <lirongqing@baidu.com>
Signed-off-by: Yuan ZhaoXiong <yuanzhaoxiong@baidu.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1622985115-51007-1-git-send-email-yuanzhaoxiong@baidu.com
The _sum and _avg values are in general sync together with the PELT
divider. They are however not always completely in perfect sync,
resulting in situations where _sum gets to zero while _avg stays
positive. Such situations are undesirable.
This comes from the fact that PELT will increase period_contrib, also
increasing the PELT divider, without updating _sum and _avg values to
stay in perfect sync where (_sum == _avg * divider). However, such PELT
change will never lower _sum, making it impossible to end up in a
situation where _sum is zero and _avg is not.
Therefore, we need to ensure that when subtracting load outside PELT,
that when _sum is zero, _avg is also set to zero. This occurs when
(_sum < _avg * divider), and the subtracted (_avg * divider) is bigger
or equal to the current _sum, while the subtracted _avg is smaller than
the current _avg.
Reported-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Signed-off-by: Odin Ugedal <odin@uged.al>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Link: https://lore.kernel.org/r/20210624111815.57937-1-odin@uged.al
The CFS bandwidth controller limits CPU requests of a task group to
quota during each period. However, parallel workloads might be bursty
so that they get throttled even when their average utilization is under
quota. And they are latency sensitive at the same time so that
throttling them is undesired.
We borrow time now against our future underrun, at the cost of increased
interference against the other system users. All nicely bounded.
Traditional (UP-EDF) bandwidth control is something like:
(U = \Sum u_i) <= 1
This guaranteeds both that every deadline is met and that the system is
stable. After all, if U were > 1, then for every second of walltime,
we'd have to run more than a second of program time, and obviously miss
our deadline, but the next deadline will be further out still, there is
never time to catch up, unbounded fail.
This work observes that a workload doesn't always executes the full
quota; this enables one to describe u_i as a statistical distribution.
For example, have u_i = {x,e}_i, where x is the p(95) and x+e p(100)
(the traditional WCET). This effectively allows u to be smaller,
increasing the efficiency (we can pack more tasks in the system), but at
the cost of missing deadlines when all the odds line up. However, it
does maintain stability, since every overrun must be paired with an
underrun as long as our x is above the average.
That is, suppose we have 2 tasks, both specify a p(95) value, then we
have a p(95)*p(95) = 90.25% chance both tasks are within their quota and
everything is good. At the same time we have a p(5)p(5) = 0.25% chance
both tasks will exceed their quota at the same time (guaranteed deadline
fail). Somewhere in between there's a threshold where one exceeds and
the other doesn't underrun enough to compensate; this depends on the
specific CDFs.
At the same time, we can say that the worst case deadline miss, will be
\Sum e_i; that is, there is a bounded tardiness (under the assumption
that x+e is indeed WCET).
The benefit of burst is seen when testing with schbench. Default value of
kernel.sched_cfs_bandwidth_slice_us(5ms) and CONFIG_HZ(1000) is used.
mkdir /sys/fs/cgroup/cpu/test
echo $$ > /sys/fs/cgroup/cpu/test/cgroup.procs
echo 100000 > /sys/fs/cgroup/cpu/test/cpu.cfs_quota_us
echo 100000 > /sys/fs/cgroup/cpu/test/cpu.cfs_burst_us
./schbench -m 1 -t 3 -r 20 -c 80000 -R 10
The average CPU usage is at 80%. I run this for 10 times, and got long tail
latency for 6 times and got throttled for 8 times.
Tail latencies are shown below, and it wasn't the worst case.
Latency percentiles (usec)
50.0000th: 19872
75.0000th: 21344
90.0000th: 22176
95.0000th: 22496
*99.0000th: 22752
99.5000th: 22752
99.9000th: 22752
min=0, max=22727
rps: 9.90 p95 (usec) 22496 p99 (usec) 22752 p95/cputime 28.12% p99/cputime 28.44%
The interferenece when using burst is valued by the possibilities for
missing the deadline and the average WCET. Test results showed that when
there many cgroups or CPU is under utilized, the interference is
limited. More details are shown in:
https://lore.kernel.org/lkml/5371BD36-55AE-4F71-B9D7-B86DC32E3D2B@linux.alibaba.com/
Co-developed-by: Shanpei Chen <shanpeic@linux.alibaba.com>
Signed-off-by: Shanpei Chen <shanpeic@linux.alibaba.com>
Co-developed-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Signed-off-by: Huaixin Chang <changhuaixin@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20210621092800.23714-2-changhuaixin@linux.alibaba.com
Ensure that a CFS parent will be in the list whenever one of its children is also
in the list.
A warning on rq->tmp_alone_branch != &rq->leaf_cfs_rq_list has been
reported while running LTP test cfs_bandwidth01.
Odin Ugedal found the root cause:
$ tree /sys/fs/cgroup/ltp/ -d --charset=ascii
/sys/fs/cgroup/ltp/
|-- drain
`-- test-6851
`-- level2
|-- level3a
| |-- worker1
| `-- worker2
`-- level3b
`-- worker3
Timeline (ish):
- worker3 gets throttled
- level3b is decayed, since it has no more load
- level2 get throttled
- worker3 get unthrottled
- level2 get unthrottled
- worker3 is added to list
- level3b is not added to list, since nr_running==0 and is decayed
[ Vincent Guittot: Rebased and updated to fix for the reported warning. ]
Fixes: a7b359fc6a37 ("sched/fair: Correctly insert cfs_rq's to list on unthrottle")
Reported-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Rik van Riel <riel@surriel.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Acked-by: Odin Ugedal <odin@uged.al>
Link: https://lore.kernel.org/r/20210621174330.11258-1-vincent.guittot@linaro.org
Change the type and name of task_struct::state. Drop the volatile and
shrink it to an 'unsigned int'. Rename it in order to find all uses
such that we can use READ_ONCE/WRITE_ONCE as appropriate.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Bristot de Oliveira <bristot@redhat.com>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Daniel Thompson <daniel.thompson@linaro.org>
Link: https://lore.kernel.org/r/20210611082838.550736351@infradead.org
This commit in sched/urgent moved the cfs_rq_is_decayed() function:
a7b359fc6a37: ("sched/fair: Correctly insert cfs_rq's to list on unthrottle")
and this fresh commit in sched/core modified it in the old location:
9e077b52d86a: ("sched/pelt: Check that *_avg are null when *_sum are")
Merge the two variants.
Conflicts:
kernel/sched/fair.c
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This is a partial forward-port of Peter Ziljstra's work first posted
at:
https://lore.kernel.org/lkml/20180530142236.667774973@infradead.org/
Currently select_idle_cpu()'s proportional scheme uses the average idle
time *for when we are idle*, that is temporally challenged. When a CPU
is not at all idle, we'll happily continue using whatever value we did
see when the CPU goes idle. To fix this, introduce a separate average
idle and age it (the existing value still makes sense for things like
new-idle balancing, which happens when we do go idle).
The overall goal is to not spend more time scanning for idle CPUs than
we're idle for. Otherwise we're inhibiting work. This means that we need to
consider the cost over all the wake-ups between consecutive idle periods.
To track this, the scan cost is subtracted from the estimated average
idle time.
The impact of this patch is related to workloads that have domains that
are fully busy or overloaded. Without the patch, the scan depth may be
too high because a CPU is not reaching idle.
Due to the nature of the patch, this is a regression magnet. It
potentially wins when domains are almost fully busy or overloaded --
at that point searches are likely to fail but idle is not being aged
as CPUs are active so search depth is too large and useless. It will
potentially show regressions when there are idle CPUs and a deep search is
beneficial. This tbench result on a 2-socket broadwell machine partially
illustates the problem
5.13.0-rc2 5.13.0-rc2
vanilla sched-avgidle-v1r5
Hmean 1 445.02 ( 0.00%) 451.36 * 1.42%*
Hmean 2 830.69 ( 0.00%) 846.03 * 1.85%*
Hmean 4 1350.80 ( 0.00%) 1505.56 * 11.46%*
Hmean 8 2888.88 ( 0.00%) 2586.40 * -10.47%*
Hmean 16 5248.18 ( 0.00%) 5305.26 * 1.09%*
Hmean 32 8914.03 ( 0.00%) 9191.35 * 3.11%*
Hmean 64 10663.10 ( 0.00%) 10192.65 * -4.41%*
Hmean 128 18043.89 ( 0.00%) 18478.92 * 2.41%*
Hmean 256 16530.89 ( 0.00%) 17637.16 * 6.69%*
Hmean 320 16451.13 ( 0.00%) 17270.97 * 4.98%*
Note that 8 was a regression point where a deeper search would have helped
but it gains for high thread counts when searches are useless. Hackbench
is a more extreme example although not perfect as the tasks idle rapidly
hackbench-process-pipes
5.13.0-rc2 5.13.0-rc2
vanilla sched-avgidle-v1r5
Amean 1 0.3950 ( 0.00%) 0.3887 ( 1.60%)
Amean 4 0.9450 ( 0.00%) 0.9677 ( -2.40%)
Amean 7 1.4737 ( 0.00%) 1.4890 ( -1.04%)
Amean 12 2.3507 ( 0.00%) 2.3360 * 0.62%*
Amean 21 4.0807 ( 0.00%) 4.0993 * -0.46%*
Amean 30 5.6820 ( 0.00%) 5.7510 * -1.21%*
Amean 48 8.7913 ( 0.00%) 8.7383 ( 0.60%)
Amean 79 14.3880 ( 0.00%) 13.9343 * 3.15%*
Amean 110 21.2233 ( 0.00%) 19.4263 * 8.47%*
Amean 141 28.2930 ( 0.00%) 25.1003 * 11.28%*
Amean 172 34.7570 ( 0.00%) 30.7527 * 11.52%*
Amean 203 41.0083 ( 0.00%) 36.4267 * 11.17%*
Amean 234 47.7133 ( 0.00%) 42.0623 * 11.84%*
Amean 265 53.0353 ( 0.00%) 47.7720 * 9.92%*
Amean 296 60.0170 ( 0.00%) 53.4273 * 10.98%*
Stddev 1 0.0052 ( 0.00%) 0.0025 ( 51.57%)
Stddev 4 0.0357 ( 0.00%) 0.0370 ( -3.75%)
Stddev 7 0.0190 ( 0.00%) 0.0298 ( -56.64%)
Stddev 12 0.0064 ( 0.00%) 0.0095 ( -48.38%)
Stddev 21 0.0065 ( 0.00%) 0.0097 ( -49.28%)
Stddev 30 0.0185 ( 0.00%) 0.0295 ( -59.54%)
Stddev 48 0.0559 ( 0.00%) 0.0168 ( 69.92%)
Stddev 79 0.1559 ( 0.00%) 0.0278 ( 82.17%)
Stddev 110 1.1728 ( 0.00%) 0.0532 ( 95.47%)
Stddev 141 0.7867 ( 0.00%) 0.0968 ( 87.69%)
Stddev 172 1.0255 ( 0.00%) 0.0420 ( 95.91%)
Stddev 203 0.8106 ( 0.00%) 0.1384 ( 82.92%)
Stddev 234 1.1949 ( 0.00%) 0.1328 ( 88.89%)
Stddev 265 0.9231 ( 0.00%) 0.0820 ( 91.11%)
Stddev 296 1.0456 ( 0.00%) 0.1327 ( 87.31%)
Again, higher thread counts benefit and the standard deviation
shows that results are also a lot more stable when the idle
time is aged.
The patch potentially matters when a socket was multiple LLCs as the
maximum search depth is lower. However, some of the test results were
suspiciously good (e.g. specjbb2005 gaining 50% on a Zen1 machine) and
other results were not dramatically different to other mcahines.
Given the nature of the patch, Peter's full series is not being forward
ported as each part should stand on its own. Preferably they would be
merged at different times to reduce the risk of false bisections.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210615111611.GH30378@techsingularity.net
Energy Aware Scheduling (EAS) needs to predict the decisions made by
SchedUtil. The map_util_freq() exists to do that.
There are corner cases where the max allowed frequency might be reduced
(due to thermal). SchedUtil as a CPUFreq governor, is aware of that
but EAS is not. This patch aims to address it.
SchedUtil stores the maximum allowed frequency in
'sugov_policy::next_freq' field. EAS has to predict that value, which is
the real used frequency. That value is made after a call to
cpufreq_driver_resolve_freq() which clamps to the CPUFreq policy limits.
In the existing code EAS is not able to predict that real frequency.
This leads to energy estimation errors.
To avoid wrong energy estimation in EAS (due to frequency miss prediction)
make sure that the step which calculates Performance Domain frequency,
is also aware of the allowed CPU capacity.
Furthermore, modify map_util_freq() to not extend the frequency value.
Instead, use map_util_perf() to extend the util value in both places:
SchedUtil and EAS, but for EAS clamp it to max allowed CPU capacity.
In the end, we achieve the same desirable behavior for both subsystems
and alignment in regards to the real CPU frequency.
Signed-off-by: Lukasz Luba <lukasz.luba@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> (For the schedutil part)
Link: https://lore.kernel.org/r/20210614191238.23224-1-lukasz.luba@arm.com
Energy Aware Scheduling (EAS) needs to be able to predict the frequency
requests made by the SchedUtil governor to properly estimate energy used
in the future. It has to take into account CPUs utilization and forecast
Performance Domain (PD) frequency. There is a corner case when the max
allowed frequency might be reduced due to thermal. SchedUtil is aware of
that reduced frequency, so it should be taken into account also in EAS
estimations.
SchedUtil, as a CPUFreq governor, knows the maximum allowed frequency of
a CPU, thanks to cpufreq_driver_resolve_freq() and internal clamping
to 'policy::max'. SchedUtil is responsible to respect that upper limit
while setting the frequency through CPUFreq drivers. This effective
frequency is stored internally in 'sugov_policy::next_freq' and EAS has
to predict that value.
In the existing code the raw value of arch_scale_cpu_capacity() is used
for clamping the returned CPU utilization from effective_cpu_util().
This patch fixes issue with too big single CPU utilization, by introducing
clamping to the allowed CPU capacity. The allowed CPU capacity is a CPU
capacity reduced by thermal pressure raw value.
Thanks to knowledge about allowed CPU capacity, we don't get too big value
for a single CPU utilization, which is then added to the util sum. The
util sum is used as a source of information for estimating whole PD energy.
To avoid wrong energy estimation in EAS (due to capped frequency), make
sure that the calculation of util sum is aware of allowed CPU capacity.
This thermal pressure might be visible in scenarios where the CPUs are not
heavily loaded, but some other component (like GPU) drastically reduced
available power budget and increased the SoC temperature. Thus, we still
use EAS for task placement and CPUs are not over-utilized.
Signed-off-by: Lukasz Luba <lukasz.luba@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20210614191128.22735-1-lukasz.luba@arm.com
In case the _avg delta is 0 there is no need to update se's _avg
(level n) nor cfs_rq's _avg (level n-1). These values stay the same.
Since cfs_rq's _avg isn't changed, i.e. no load is propagated down,
cfs_rq's _sum should stay the same as well.
So bail out after se's _sum has been updated.
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20210601083616.804229-1-dietmar.eggemann@arm.com
Check that we never break the rule that pelt's avg values are null if
pelt's sum are.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Odin Ugedal <odin@uged.al>
Link: https://lore.kernel.org/r/20210601155328.19487-1-vincent.guittot@linaro.org
Fix an issue where fairness is decreased since cfs_rq's can end up not
being decayed properly. For two sibling control groups with the same
priority, this can often lead to a load ratio of 99/1 (!!).
This happens because when a cfs_rq is throttled, all the descendant
cfs_rq's will be removed from the leaf list. When they initial cfs_rq
is unthrottled, it will currently only re add descendant cfs_rq's if
they have one or more entities enqueued. This is not a perfect
heuristic.
Instead, we insert all cfs_rq's that contain one or more enqueued
entities, or it its load is not completely decayed.
Can often lead to situations like this for equally weighted control
groups:
$ ps u -C stress
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 10009 88.8 0.0 3676 100 pts/1 R+ 11:04 0:13 stress --cpu 1
root 10023 3.0 0.0 3676 104 pts/1 R+ 11:04 0:00 stress --cpu 1
Fixes: 31bc6aeaab1d ("sched/fair: Optimize update_blocked_averages()")
[vingo: !SMP build fix]
Signed-off-by: Odin Ugedal <odin@uged.al>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20210612112815.61678-1-odin@uged.al
The util_est internal UTIL_AVG_UNCHANGED flag which is used to prevent
unnecessary util_est updates uses the LSB of util_est.enqueued. It is
exposed via _task_util_est() (and task_util_est()).
Commit 92a801e5d5b7 ("sched/fair: Mask UTIL_AVG_UNCHANGED usages")
mentions that the LSB is lost for util_est resolution but
find_energy_efficient_cpu() checks if task_util_est() returns 0 to
return prev_cpu early.
_task_util_est() returns the max value of util_est.ewma and
util_est.enqueued or'ed w/ UTIL_AVG_UNCHANGED.
So task_util_est() returning the max of task_util() and
_task_util_est() will never return 0 under the default
SCHED_FEAT(UTIL_EST, true).
To fix this use the MSB of util_est.enqueued instead and keep the flag
util_est internal, i.e. don't export it via _task_util_est().
The maximal possible util_avg value for a task is 1024 so the MSB of
'unsigned int util_est.enqueued' isn't used to store a util value.
As a caveat the code behind the util_est_se trace point has to filter
UTIL_AVG_UNCHANGED to see the real util_est.enqueued value which should
be easy to do.
This also fixes an issue report by Xuewen Yan that util_est_update()
only used UTIL_AVG_UNCHANGED for the subtrahend of the equation:
last_enqueued_diff = ue.enqueued - (task_util() | UTIL_AVG_UNCHANGED)
Fixes: b89997aa88f0b sched/pelt: Fix task util_est update filtering
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Xuewen Yan <xuewen.yan@unisoc.com>
Reviewed-by: Vincent Donnefort <vincent.donnefort@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20210602145808.1562603-1-dietmar.eggemann@arm.com
Rounding in PELT calculation happening when entities are attached/detached
of a cfs_rq can result into situations where util/runnable_avg is not null
but util/runnable_sum is. This is normally not possible so we need to
ensure that util/runnable_sum stays synced with util/runnable_avg.
detach_entity_load_avg() is the last place where we don't sync
util/runnable_sum with util/runnbale_avg when moving some sched_entities
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210601085832.12626-1-vincent.guittot@linaro.org
When using something other than 8 spaces per tab, this ascii art
makes not sense, and the reader might end up wondering what this
advanced equation "is".
Signed-off-by: Odin Ugedal <odin@uged.al>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210518125202.78658-4-odin@uged.al
During the update of fair blocked load (__update_blocked_fair()), we
update the contribution of the cfs in tg->load_avg if cfs_rq's pelt
has decayed. Nevertheless, the pelt values of a cfs_rq could have
been recently updated while propagating the change of a child. In this
case, cfs_rq's pelt will not decayed because it has already been
updated and we don't update tg->load_avg.
__update_blocked_fair
...
for_each_leaf_cfs_rq_safe: child cfs_rq
update cfs_rq_load_avg() for child cfs_rq
...
update_load_avg(cfs_rq_of(se), se, 0)
...
update cfs_rq_load_avg() for parent cfs_rq
-propagation of child's load makes parent cfs_rq->load_sum
becoming null
-UPDATE_TG is not set so it doesn't update parent
cfs_rq->tg_load_avg_contrib
..
for_each_leaf_cfs_rq_safe: parent cfs_rq
update cfs_rq_load_avg() for parent cfs_rq
- nothing to do because parent cfs_rq has already been updated
recently so cfs_rq->tg_load_avg_contrib is not updated
...
parent cfs_rq is decayed
list_del_leaf_cfs_rq parent cfs_rq
- but it still contibutes to tg->load_avg
we must set UPDATE_TG flags when propagting pending load to the parent
Fixes: 039ae8bcf7a5 ("sched/fair: Fix O(nr_cgroups) in the load balancing path")
Reported-by: Odin Ugedal <odin@uged.al>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Odin Ugedal <odin@uged.al>
Link: https://lkml.kernel.org/r/20210527122916.27683-3-vincent.guittot@linaro.org
when removing a cfs_rq from the list we only check _sum value so we must
ensure that _avg and _sum stay synced so load_sum can't be null whereas
load_avg is not after propagating load in the cgroup hierarchy.
Use load_avg to compute load_sum similarly to what is done for util_sum
and runnable_sum.
Fixes: 0e2d2aaaae52 ("sched/fair: Rewrite PELT migration propagation")
Reported-by: Odin Ugedal <odin@uged.al>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Odin Ugedal <odin@uged.al>
Link: https://lkml.kernel.org/r/20210527122916.27683-2-vincent.guittot@linaro.org
- Don't migrate if there is a cookie mismatch
Load balance tries to move task from busiest CPU to the
destination CPU. When core scheduling is enabled, if the
task's cookie does not match with the destination CPU's
core cookie, this task may be skipped by this CPU. This
mitigates the forced idle time on the destination CPU.
- Select cookie matched idle CPU
In the fast path of task wakeup, select the first cookie matched
idle CPU instead of the first idle CPU.
- Find cookie matched idlest CPU
In the slow path of task wakeup, find the idlest CPU whose core
cookie matches with task's cookie
Signed-off-by: Aubrey Li <aubrey.li@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Don Hiatt <dhiatt@digitalocean.com>
Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.860083871@infradead.org
During force-idle, we end up doing cross-cpu comparison of vruntimes
during pick_next_task. If we simply compare (vruntime-min_vruntime)
across CPUs, and if the CPUs only have 1 task each, we will always
end up comparing 0 with 0 and pick just one of the tasks all the time.
This starves the task that was not picked. To fix this, take a snapshot
of the min_vruntime when entering force idle and use it for comparison.
This min_vruntime snapshot will only be used for cross-CPU vruntime
comparison, and nothing else.
A note about the min_vruntime snapshot and force idling:
During selection:
When we're not fi, we need to update snapshot.
when we're fi and we were not fi, we must update snapshot.
When we're fi and we were already fi, we must not update snapshot.
Which gives:
fib fi update
0 0 1
0 1 1
1 0 1
1 1 0
Where:
fi: force-idled now
fib: force-idled before
So the min_vruntime snapshot needs to be updated when: !(fib && fi).
Also, the cfs_prio_less() function needs to be aware of whether the
core is in force idle or not, since it will be use this information to
know whether to advance a cfs_rq's min_vruntime_fi in the hierarchy.
So pass this information along via pick_task() -> prio_less().
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Don Hiatt <dhiatt@digitalocean.com>
Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.738542617@infradead.org
If there is only one long running local task and the sibling is
forced idle, it might not get a chance to run until a schedule
event happens on any cpu in the core.
So we check for this condition during a tick to see if a sibling
is starved and then give it a chance to schedule.
Signed-off-by: Vineeth Pillai <viremana@linux.microsoft.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Don Hiatt <dhiatt@digitalocean.com>
Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.617407840@infradead.org
Introduce task_struct::core_cookie as an opaque identifier for core
scheduling. When enabled; core scheduling will only allow matching
task to be on the core; where idle matches everything.
When task_struct::core_cookie is set (and core scheduling is enabled)
these tasks are indexed in a second RB-tree, first on cookie value
then on scheduling function, such that matching task selection always
finds the most elegible match.
NOTE: *shudder* at the overhead...
NOTE: *sigh*, a 3rd copy of the scheduling function; the alternative
is per class tracking of cookies and that just duplicates a lot of
stuff for no raisin (the 2nd copy lives in the rt-mutex PI code).
[Joel: folded fixes]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Don Hiatt <dhiatt@digitalocean.com>
Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.496975854@infradead.org
Because sched_class::pick_next_task() also implies
sched_class::set_next_task() (and possibly put_prev_task() and
newidle_balance) it is not state invariant. This makes it unsuitable
for remote task selection.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
[Vineeth: folded fixes]
Signed-off-by: Vineeth Remanan Pillai <viremana@linux.microsoft.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Don Hiatt <dhiatt@digitalocean.com>
Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.437092775@infradead.org
rq_lockp() includes a static_branch(), which is asm-goto, which is
asm volatile which defeats regular CSE. This means that:
if (!static_branch(&foo))
return simple;
if (static_branch(&foo) && cond)
return complex;
Doesn't fold and we get horrible code. Introduce __rq_lockp() without
the static_branch() on.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Don Hiatt <dhiatt@digitalocean.com>
Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.316696988@infradead.org
In preparation of playing games with rq->lock, abstract the thing
using an accessor.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Don Hiatt <dhiatt@digitalocean.com>
Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.136465446@infradead.org
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Don Hiatt <dhiatt@digitalocean.com>
Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.015639083@infradead.org
find_energy_efficient_cpu() (feec()) searches the best energy CPU
to place a task on. To do so, compute_energy() estimates the energy
impact of placing the task on a CPU, based on CPU and task utilization
signals.
Utilization signals can be concurrently updated while evaluating a
performance domain (pd). In some cases, this leads to having a
'negative delta', i.e. placing the task in the pd is seen as an
energy gain. Thus, any further energy comparison is biased.
In case of a 'negative delta', return prev_cpu since:
1. a 'negative delta' happens in less than 0.5% of feec() calls,
on a Juno with 6 CPUs (4 little, 2 big)
2. it is unlikely to have two consecutive 'negative delta' for
a task, so if the first call fails, feec() will correctly
place the task in the next feec() call
3. EAS current behavior tends to select prev_cpu if the task
doesn't raise the OPP of its current pd. prev_cpu is EAS's
generic decision
4. prev_cpu should be preferred to returning an error code.
In the latter case, select_idle_sibling() would do the placement,
selecting a big (and not energy efficient) CPU. As 3., the task
would potentially reside on the big CPU for a long time
Reported-by: Xuewen Yan <xuewen.yan@unisoc.com>
Suggested-by: Xuewen Yan <xuewen.yan@unisoc.com>
Signed-off-by: Pierre Gondois <Pierre.Gondois@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Vincent Donnefort <vincent.donnefort@arm.com>
Link: https://lkml.kernel.org/r/20210504090743.9688-3-Pierre.Gondois@arm.com
find_energy_efficient_cpu() searches the best energy CPU
to place a task on. To do so, the energy of each performance domain
(pd) is computed w/ and w/o the task placed on it.
The energy of a pd w/o the task (base_energy_pd) is computed prior
knowing whether a CPU is available in the pd.
Move the base_energy_pd computation after looping through the CPUs
of a pd and only compute it if at least one CPU is available.
Suggested-by: Xuewen Yan <xuewen.yan@unisoc.com>
Signed-off-by: Pierre Gondois <Pierre.Gondois@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Vincent Donnefort <vincent.donnefort@arm.com>
Link: https://lkml.kernel.org/r/20210504090743.9688-2-Pierre.Gondois@arm.com
The try_to_wake_up function has an optimization where it can queue
a task for wakeup on its previous CPU, if the task is still in the
middle of going to sleep inside schedule().
Once schedule() re-enables IRQs, the task will be woken up with an
IPI, and placed back on the runqueue.
If we have such a wakeup pending, there is no need to search other
CPUs for runnable tasks. Just skip (or bail out early from) newidle
balancing, and run the just woken up task.
For a memcache like workload test, this reduces total CPU use by
about 2%, proportionally split between user and system time,
and p99 and p95 application response time by 10% on average.
The schedstats run_delay number shows a similar improvement.
Signed-off-by: Rik van Riel <riel@surriel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Link: https://lkml.kernel.org/r/20210422130236.0bb353df@imladris.surriel.com
In commit:
9fe1f127b913 ("sched/fair: Merge select_idle_core/cpu()")
in select_idle_cpu(), we check if an idle core is present in the LLC
of the target CPU via the flag "has_idle_cores". We look for the idle
core in select_idle_cores(). If select_idle_cores() isn't able to find
an idle core/CPU, we need to unset the has_idle_cores flag in the LLC
of the target to prevent other CPUs from going down this route.
However, the current code is unsetting it in the LLC of the current
CPU instead of the target CPU. This patch fixes this issue.
Fixes: 9fe1f127b913 ("sched/fair: Merge select_idle_core/cpu()")
Signed-off-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Link: https://lore.kernel.org/r/1620746169-13996-1-git-send-email-ego@linux.vnet.ibm.com
This fixes an issue where old load on a cfs_rq is not properly decayed,
resulting in strange behavior where fairness can decrease drastically.
Real workloads with equally weighted control groups have ended up
getting a respective 99% and 1%(!!) of cpu time.
When an idle task is attached to a cfs_rq by attaching a pid to a cgroup,
the old load of the task is attached to the new cfs_rq and sched_entity by
attach_entity_cfs_rq. If the task is then moved to another cpu (and
therefore cfs_rq) before being enqueued/woken up, the load will be moved
to cfs_rq->removed from the sched_entity. Such a move will happen when
enforcing a cpuset on the task (eg. via a cgroup) that force it to move.
The load will however not be removed from the task_group itself, making
it look like there is a constant load on that cfs_rq. This causes the
vruntime of tasks on other sibling cfs_rq's to increase faster than they
are supposed to; causing severe fairness issues. If no other task is
started on the given cfs_rq, and due to the cpuset it would not happen,
this load would never be properly unloaded. With this patch the load
will be properly removed inside update_blocked_averages. This also
applies to tasks moved to the fair scheduling class and moved to another
cpu, and this path will also fix that. For fork, the entity is queued
right away, so this problem does not affect that.
This applies to cases where the new process is the first in the cfs_rq,
issue introduced 3d30544f0212 ("sched/fair: Apply more PELT fixes"), and
when there has previously been load on the cgroup but the cgroup was
removed from the leaflist due to having null PELT load, indroduced
in 039ae8bcf7a5 ("sched/fair: Fix O(nr_cgroups) in the load balancing
path").
For a simple cgroup hierarchy (as seen below) with two equally weighted
groups, that in theory should get 50/50 of cpu time each, it often leads
to a load of 60/40 or 70/30.
parent/
cg-1/
cpu.weight: 100
cpuset.cpus: 1
cg-2/
cpu.weight: 100
cpuset.cpus: 1
If the hierarchy is deeper (as seen below), while keeping cg-1 and cg-2
equally weighted, they should still get a 50/50 balance of cpu time.
This however sometimes results in a balance of 10/90 or 1/99(!!) between
the task groups.
$ ps u -C stress
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 18568 1.1 0.0 3684 100 pts/12 R+ 13:36 0:00 stress --cpu 1
root 18580 99.3 0.0 3684 100 pts/12 R+ 13:36 0:09 stress --cpu 1
parent/
cg-1/
cpu.weight: 100
sub-group/
cpu.weight: 1
cpuset.cpus: 1
cg-2/
cpu.weight: 100
sub-group/
cpu.weight: 10000
cpuset.cpus: 1
This can be reproduced by attaching an idle process to a cgroup and
moving it to a given cpuset before it wakes up. The issue is evident in
many (if not most) container runtimes, and has been reproduced
with both crun and runc (and therefore docker and all its "derivatives"),
and with both cgroup v1 and v2.
Fixes: 3d30544f0212 ("sched/fair: Apply more PELT fixes")
Fixes: 039ae8bcf7a5 ("sched/fair: Fix O(nr_cgroups) in the load balancing path")
Signed-off-by: Odin Ugedal <odin@uged.al>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210501141950.23622-2-odin@uged.al
