lkml.org 
[lkml]   [2021]   [Jan]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH -tip V3 0/8] workqueue: break affinity initiatively
On Mon, Jan 11, 2021 at 11:07:34AM +0100, Thomas Gleixner wrote:
> On Fri, Jan 08 2021 at 12:46, Peter Zijlstra wrote:
> > On Sat, Dec 26, 2020 at 10:51:08AM +0800, Lai Jiangshan wrote:
> >> From: Lai Jiangshan <laijs@linux.alibaba.com>
> >>
> >> 06249738a41a ("workqueue: Manually break affinity on hotplug")
> >> said that scheduler will not force break affinity for us.
> >
> > So I've been looking at this the past day or so, and the more I look,
> > the more I think commit:
> >
> > 1cf12e08bc4d ("sched/hotplug: Consolidate task migration on CPU unplug")
> >
> > is a real problem and we need to revert it (at least for now).
> >
> > Let me attempt a brain dump:
> >
> > - the assumption that per-cpu kernel threads are 'well behaved' on
> > hot-plug has, I think, been proven incorrect, it's far worse than
> > just bounded workqueue. Therefore, it makes sense to provide the old
> > semantics.
>
> I disagree. Per-cpu kernel threads which are magically stopped during
> hotplug and then migrated to a random other CPU are just wrong.
>
> We really need to fix that and not proliferate the sloppy and ill
> defined behaviour.

Well yes, but afaict the workqueue stuff hasn't been settled yet, and
the rcutorture patch Paul did was just plain racy and who knows what
other daft kthread users are out there. That and we're at -rc3.

So I'm really tempted to revert for now and try again later.

\
 
 \ /
  Last update: 2021-01-11 12:04    [W:0.130 / U:0.136 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site