Messages in this thread | | | From | "byungchul.park" <> | Subject | RE: [PATCH] sched: fix lose fair sleeper bonus in switch_to_fair() | Date | Tue, 8 Sep 2015 17:55:02 +0900 |
| |
> -----Original Message----- > From: Wanpeng Li [mailto:wanpeng.li@hotmail.com] > Sent: Tuesday, September 08, 2015 5:46 PM > To: Byungchul Park > Cc: Peter Zijlstra; Ingo Molnar; linux-kernel@vger.kernel.org; > yuyang.du@intel.com > Subject: Re: [PATCH] sched: fix lose fair sleeper bonus in switch_to_fair() > > On 9/8/15 4:38 PM, Wanpeng Li wrote: > > On 9/8/15 4:22 PM, Byungchul Park wrote: > >> On Tue, Sep 08, 2015 at 04:04:49PM +0800, Wanpeng Li wrote: > >>>>> However, if se->vruntime -= cfs_rq->min_vruntime is positive, the > >>>>> behavior is different after your patch. e.g. se->vruntime(the > >>>>> relative vruntime in switched_to_fair()) < min_vruntime - > >>>>> sysctl_sched_latency/2 > >>>>> > >>>>> before your patch: > >>>>> > >>>>> se->vruntime = min_vruntime - sysctl_sched_latency/2 (place_entity()) > >>>> my patch is based on ff277d4 commit at tip/sched/core. > >>>> > >>>> there's no change between before and after. > >>>> > >>>> check it please. > >>>> > >>>> and this logic seems to be no problem to me. :( > >>> Your logic will lose fair sleeper bonus in the scenario which I > >>> pointed out. > >> i mean in ff277d4 commit: > > > > Please include the commit subject when you point out a commit, do you > > mean this one? > > > > commit ff277d4250fe715b6666219b1a3423b863418794 > > Author: Andrea Parri <parri.andrea@gmail.com> > > Date: Wed Aug 5 15:56:19 2015 +0200 > > > > sched/deadline: Fix comment in enqueue_task_dl() > > > > The "dl_boosted" flag is set by comparing *absolute* deadlines > > (c.f., rt_mutex_setprio()). > > > > > > What's the relationship w/ this patch? > > I think you mean your commit: > > commit 7855a35ac07a350e2cd26f09568a6d8e372be358 > Author: Byungchul Park <byungchul.park@lge.com> > Date: Mon Aug 10 18:02:55 2015 +0900 > > sched: Ensure a task has a non-normalized vruntime when returning > back to CFS > > > However, that is wrong in the scenario which I mentioned.
Ah.. Ok. What you are pointing out against, is this commit. I will focus on verifying if this commit is wrong or not from now. Let me think about this commit more.
Thanks, byungchul
> > Regards, > Wanpeng Li
| |