Chengming Zhou 2fcd7bbae9 sched/psi: Fix avgs_work re-arm in psi_avgs_work() ... Pavan reported a problem that PSI avgs_work idle shutoff is not working at all. Because PSI_NONIDLE condition would be observed in psi_avgs_work()->collect_percpu_times()->get_recent_times() even if only the kworker running avgs_work on the CPU. Although commit 1b69ac6b40 ("psi: fix aggregation idle shut-off") avoided the ping-pong wake problem when the worker sleep, psi_avgs_work() still will always re-arm the avgs_work, so shutoff is not working. This patch changes to use PSI_STATE_RESCHEDULE to flag whether to re-arm avgs_work in get_recent_times(). For the current CPU, we re-arm avgs_work only when (NR_RUNNING > 1 || NR_IOWAIT > 0 || NR_MEMSTALL > 0), for other CPUs we can just check PSI_NONIDLE delta. The new flag is only used in psi_avgs_work(), so we check in get_recent_times() that current_work() is avgs_work. One potential problem is that the brief period of non-idle time incurred between the aggregation run and the kworker's dequeue will be stranded in the per-cpu buckets until avgs_work run next time. The buckets can hold 4s worth of time, and future activity will wake the avgs_work with a 2s delay, giving us 2s worth of data we can leave behind when shut off the avgs_work. If the kworker run other works after avgs_work shut off and doesn't have any scheduler activities for 2s, this maybe a problem. Reported-by: Pavan Kondeti <quic_pkondeti@quicinc.com> Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Suren Baghdasaryan <surenb@google.com> Tested-by: Chengming Zhou <zhouchengming@bytedance.com> Link: https://lore.kernel.org/r/20221014110551.22695-1-zhouchengming@bytedance.com		2022-10-30 10:12:14 +01:00
..
acpi	ACPI: APEI: Fix integer overflow in ghes_estatus_pool_init()	2022-10-13 20:40:09 +02:00
asm-generic	ftrace,kcfi: Separate ftrace_stub() and ftrace_stub_graph()	2022-10-20 17:10:27 +02:00
clocksource	clocksource/drivers/timer-ti-dm: Move struct omap_dm_timer fields to driver	2022-09-20 10:49:46 +02:00
crypto	crypto: scatterwalk - Remove unused inline function scatterwalk_aligned()	2022-09-30 13:59:13 +08:00
drm	Merge drm/drm-fixes into drm-misc-fixes	2022-10-20 09:09:00 +02:00
dt-bindings	These are the pin control changes for the v6.1 kernel cycle:	2022-10-11 10:59:59 -07:00
keys
kunit	kunit: declare kunit_assert structs as const	2022-10-07 10:19:18 -06:00
kvm
linux	sched/psi: Fix avgs_work re-arm in psi_avgs_work()	2022-10-30 10:12:14 +01:00
math-emu
media	media fixes for v6.1-rc2	2022-10-22 15:30:15 -07:00
memory
misc
net	Networking fixes for 6.1-rc2, including fixes from netfilter	2022-10-20 17:24:59 -07:00
pcmcia
ras
rdma	RDMA/core: Add UVERBS_ATTR_RAW_FD	2022-09-27 10:15:24 -03:00
rv
scsi	SCSI misc on 20221007	2022-10-07 12:33:18 -07:00
soc	RISC-V Patches for the 6.1 Merge Window, Part 2	2022-10-14 11:21:11 -07:00
sound	ALSA: hda: Update register polling macros	2022-10-09 12:34:32 +02:00
target
trace	linux-watchdog 6.1-rc2 tag	2022-10-21 12:25:39 -07:00
uapi	media fixes for v6.1-rc2	2022-10-22 15:30:15 -07:00
ufs	SCSI misc on 20221007	2022-10-07 12:33:18 -07:00
vdso
video
xen	xen/virtio: enable grant based virtio on x86	2022-10-10 14:31:26 +02:00