[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <e2645a659b6fc1c592e2f2dc45273c361a2d2f47.camel@gmx.de>
Date: Fri, 06 Oct 2023 16:39:09 +0200
From: Mike Galbraith <efault@....de>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Marek Szyprowski <m.szyprowski@...sung.com>,
Biju Das <biju.das.jz@...renesas.com>,
"bsegall@...gle.com" <bsegall@...gle.com>,
"bristot@...hat.com" <bristot@...hat.com>,
"chris.hyser@...cle.com" <chris.hyser@...cle.com>,
"corbet@....net" <corbet@....net>,
"dietmar.eggemann@....com" <dietmar.eggemann@....com>,
"gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
"joel@...lfernandes.org" <joel@...lfernandes.org>,
"joshdon@...gle.com" <joshdon@...gle.com>,
"juri.lelli@...hat.com" <juri.lelli@...hat.com>,
"kprateek.nayak@....com" <kprateek.nayak@....com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"mingo@...nel.org" <mingo@...nel.org>,
"patrick.bellasi@...bug.net" <patrick.bellasi@...bug.net>,
Pavel Machek <pavel@....cz>, "pjt@...gle.com" <pjt@...gle.com>,
"qperret@...gle.com" <qperret@...gle.com>,
"qyousef@...alina.io" <qyousef@...alina.io>,
"rostedt@...dmis.org" <rostedt@...dmis.org>,
"tglx@...utronix.de" <tglx@...utronix.de>,
"tim.c.chen@...ux.intel.com" <tim.c.chen@...ux.intel.com>,
"timj@....org" <timj@....org>,
"vincent.guittot@...aro.org" <vincent.guittot@...aro.org>,
"youssefesmat@...omium.org" <youssefesmat@...omium.org>,
"yu.c.chen@...el.com" <yu.c.chen@...el.com>,
"mgorman@...e.de" <mgorman@...e.de>,
"linux-renesas-soc@...r.kernel.org"
<linux-renesas-soc@...r.kernel.org>
Subject: Re: [PATCH] sched/fair: fix pick_eevdf to always find the correct se
On Fri, 2023-10-06 at 16:00 +0200, Peter Zijlstra wrote:
> On Fri, Oct 06, 2023 at 12:31:28PM +0200, Mike Galbraith wrote:
> > On Fri, 2023-10-06 at 10:35 +0200, Marek Szyprowski wrote:
> > >
> > >
> > > Just to note, I've run into this issue on the QEmu's 'arm64/virt'
> > > platform, not on the Samsung specific hardware.
> >
> > It doesn't appear to be arch specific, all I have to do is enable
> > autogroup, raspberry or x86_64 desktop box, the occasional failure
> > tripping over task groups, leaving both best and best_left with
> > identical but not what we're looking for ->min_deadline.
>
> OK, autogroups enabled and booted, /debug/sched/debug shows autogroups
> are indeed in existence.
>
> I've ran hackbench and a kernel build, but no whoopsie yet :-(
>
> I suppose I'll kick some benchmarks and go make a cup 'o tea or
> something.
Hm, just booting gets me a handful, and generic desktop activity
produces a fairly regular supply. This is virgin 6.6.0.ga9e6eb3-tip.
[ 340.473123] EEVDF scheduling fail, picking leftmost
[ 340.473132] EEVDF scheduling fail, picking leftmost
[ 343.656549] EEVDF scheduling fail, picking leftmost
[ 343.656557] EEVDF scheduling fail, picking leftmost
[ 343.690463] EEVDF scheduling fail, picking leftmost
[ 343.690472] EEVDF scheduling fail, picking leftmost
[ 426.224039] EEVDF scheduling fail, picking leftmost
[ 426.224080] EEVDF scheduling fail, picking leftmost
[ 426.224084] EEVDF scheduling fail, picking leftmost
[ 426.363323] EEVDF scheduling fail, picking leftmost
[ 426.759244] EEVDF scheduling fail, picking leftmost
[ 426.759256] EEVDF scheduling fail, picking leftmost
[ 441.872524] EEVDF scheduling fail, picking leftmost
[ 441.872556] EEVDF scheduling fail, picking leftmost
Powered by blists - more mailing lists