Hi Peter, On 2019/4/25 16:00, Peter Zijlstra wrote: > On Thu, Apr 25, 2019 at 11:52:28AM +0800, Xie XiuQi wrote: >> On 2019/4/24 2:44, Peter Zijlstra wrote: > >>> I'll try and come up with a better Changelog tomorrow. > > I actually did, but forgot to send out. I have the below. > Does that work for you?
It works for me, thank you very much. > > --- > Subject: sched/numa: Fix a possible divide-by-zero > From: Xie XiuQi <xiexi...@huawei.com> > Date: Sat, 20 Apr 2019 16:34:16 +0800 > > sched_clock_cpu() may not be consistent between CPUs. If a task > migrates to another CPU, then se.exec_start is set to that CPU's > rq_clock_task() by update_stats_curr_start(). Specifically, the new > value might be before the old value due to clock skew. > > So then if in numa_get_avg_runtime() the expression: > > 'now - p->last_task_numa_placement' > > ends up as -1, then the divider '*period + 1' in task_numa_placement() > is 0 and things go bang. Similar to update_curr(), check if time goes > backwards to avoid this. > > Cc: mi...@redhat.com > Cc: cj.chengj...@huawei.com > Signed-off-by: Xie XiuQi <xiexi...@huawei.com> > [peterz: simplified changelog] > Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org> > Link: https://lkml.kernel.org/r/20190420083416.170446-1-xiexi...@huawei.com > --- > kernel/sched/fair.c | 4 ++++ > 1 file changed, 4 insertions(+) > > --- a/kernel/sched/fair.c > +++ b/kernel/sched/fair.c > @@ -2007,6 +2007,10 @@ static u64 numa_get_avg_runtime(struct t > if (p->last_task_numa_placement) { > delta = runtime - p->last_sum_exec_runtime; > *period = now - p->last_task_numa_placement; > + > + /* Avoid backward, and prevent potential divide error */ > + if (unlikely((s64)*period < 0)) > + *period = 0; > } else { > delta = p->se.avg.load_sum; > *period = LOAD_AVG_MAX; > > . > -- Thanks, Xie XiuQi
