linux_dsm_epyc7002/kernel/sched
Xuewen Yan 143a6b8ec5 sched/uclamp: Ignore max aggregation if rq is idle
[ Upstream commit 3e1493f46390618ea78607cb30c58fc19e2a5035 ]

When a task wakes up on an idle rq, uclamp_rq_util_with() would max
aggregate with rq value. But since there is no task enqueued yet, the
values are stale based on the last task that was running. When the new
task actually wakes up and enqueued, then the rq uclamp values should
reflect that of the newly woken up task effective uclamp values.

This is a problem particularly for uclamp_max because it default to
1024. If a task p with uclamp_max = 512 wakes up, then max aggregation
would ignore the capping that should apply when this task is enqueued,
which is wrong.

Fix that by ignoring max aggregation if the rq is idle since in that
case the effective uclamp value of the rq will be the ones of the task
that will wake up.

Fixes: 9d20ad7dfc ("sched/uclamp: Add uclamp_util_with()")
Signed-off-by: Xuewen Yan <xuewen.yan@unisoc.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
[qias: Changelog]
Reviewed-by: Qais Yousef <qais.yousef@arm.com>
Link: https://lore.kernel.org/r/20210630141204.8197-1-xuewen.yan94@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-07-20 16:05:58 +02:00
..
autogroup.c
autogroup.h
clock.c
completion.c
core.c sched/uclamp: Fix uclamp_tg_restrict() 2021-07-14 16:56:10 +02:00
cpuacct.c
cpudeadline.c sched/deadline: Implement fallback mechanism for !fit case 2020-06-15 14:10:05 +02:00
cpudeadline.h
cpufreq_schedutil.c cpufreq: Introduce governor flags 2020-11-10 18:31:17 +01:00
cpufreq.c
cpupri.c
cpupri.h
cputime.c
deadline.c sched/rt: Fix Deadline utilization tracking during policy change 2021-07-14 16:56:09 +02:00
debug.c sched/fair: Fix util_est UTIL_AVG_UNCHANGED handling 2021-06-16 12:01:46 +02:00
fair.c sched/fair: Ensure _sum and _avg values stay consistent 2021-07-19 09:44:54 +02:00
features.h sched,fair: Alternative sched_slice() 2021-05-11 14:47:31 +02:00
idle.c rcu/nocb: Perform deferred wake up before last idle's need_resched() check 2021-03-04 11:38:35 +01:00
isolation.c isolcpus: Affine unbound kernel threads to housekeeping cpus 2020-06-15 14:10:03 +02:00
loadavg.c sched: nohz: stop passing around unused "ticks" parameter. 2020-07-22 10:22:04 +02:00
Makefile
membarrier.c sched/membarrier: fix missing local execution of ipi_sync_rq_state() 2021-03-17 17:06:35 +01:00
pelt.c sched: Add a tracepoint to track rq->nr_running 2020-07-08 11:39:02 +02:00
pelt.h sched/fair: Fix util_est UTIL_AVG_UNCHANGED handling 2021-06-16 12:01:46 +02:00
psi.c psi: Fix race between psi_trigger_create/destroy 2021-07-14 16:56:10 +02:00
rt.c sched/rt: Fix RT utilization tracking during policy change 2021-07-14 16:56:09 +02:00
sched-pelt.h
sched.h sched/uclamp: Ignore max aggregation if rq is idle 2021-07-20 16:05:58 +02:00
smp.h
stats.c
stats.h
stop_task.c treewide: Convert macro and uses of __section(foo) to __section("foo") 2020-10-25 14:51:49 -07:00
swait.c
topology.c Scheduler changes for v5.10: 2020-10-12 12:56:01 -07:00
wait_bit.c
wait.c rq-qos: fix missed wake-ups in rq_qos_throttle try two 2021-07-19 09:45:00 +02:00