-----Original Message----- From: Mike Galbraith [mailto:efault@gmx.de] Sent: Tuesday, April 27, 2021 5:55 PM To: Song Bao Hua (Barry Song) song.bao.hua@hisilicon.com; vincent.guittot@linaro.org; mingo@redhat.com; peterz@infradead.org; dietmar.eggemann@arm.com; rostedt@goodmis.org; bsegall@google.com; mgorman@suse.de Cc: valentin.schneider@arm.com; juri.lelli@redhat.com; bristot@redhat.com; linux-arm-kernel@lists.infradead.org; linux-kernel@vger.kernel.org; xuwei (O) xuwei5@huawei.com; Zengtao (B) prime.zeng@hisilicon.com; guodong.xu@linaro.org; yangyicong yangyicong@huawei.com; Liguozhu (Kenneth) liguozhu@hisilicon.com; linuxarm@openeuler.org; wanghuiqiang wanghuiqiang@huawei.com; xieyongjia (A) xieyongjia1@huawei.com Subject: Re: [PATCH] sched/fair: don't use waker's cpu if the waker of sync wake-up is interrupt
On Tue, 2021-04-27 at 04:44 +0000, Song Bao Hua (Barry Song) wrote:
I agree sync hint might have been overused by other kernel subsystem. But this patch will at least fix a case: sync waker is interrupt, in this case, the existing task has nothing to do with waker and wakee, so this case should be excluded from wake_affine_idle().
I long ago tried filtering interrupt wakeups, and met some surprises. Wakeup twiddling always managing to end up being a rob Peter to pay Paul operation despite our best efforts, here's hoping that your pile of stolen cycles is small enough to escape performance bot notice :)
Would you like to share the link you did before to filter interrupt wakeups?
The wake up path has hundreds of lines of code, so I don't expect that reading preempt_count will cause visible performance losses to bot. But who knows :-)
-Mike
Thanks Barry