-----邮件原件----- 发件人: Peter Zijlstra <[email protected]> 发送时间: 2020年12月15日 16:33 收件人: [email protected] 抄送: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; heddchen(陈贺) <[email protected]>; xiaoggchen(陈小光) <[email protected]> 主题: Re: [PATCH] sched: don't check rq after newidle_balance return positive(Internet mail)
On Tue, Dec 15, 2020 at 02:48:50PM +0800, [email protected] wrote: >> From: Chen Xiaoguang <[email protected]> >> >> In pick_next_task_fair, if CPU is going to idle newidle_balance is >> called first trying to pull some tasks. >> When newidle_balance returns positive which means it does pulls tasks >> or some tasks enqueued then there is no need to check >> sched_fair_runnable again. > No, I think it actually does need to, because while it counts the number of > tasks it pulled, it doesn't verify it still has them after it re-acquires > rq->lock. That is, someone could've come along and stolen them > right from under our noses. Ah, yes, our change only make sense when pulling nothing in load_balance but some tasks enqueued this rq during the lock of this rq is released. Thanks. >> >> Signed-off-by: He Chen <[email protected]> >> Signed-off-by: Xiaoguang Chen <[email protected]> > This SoB chain is broken. The first SoB should be the author, but From does > not match. We will fix this next time.
