Re: [PATCH] mm, memory_hotplug: do not back off draining pcp free pages from kworker context

From	Tejun Heo <tj@kernel.org>
Newsgroups	linux.kernel
Subject	Re: [PATCH] mm, memory_hotplug: do not back off draining pcp free pages from kworker context
Date	2017-09-19 05:40 +0200
Message-ID	<urhyb-7dq-9@gated-at.bofh.it> (permalink)
References	<ujoGu-3Tj-31@gated-at.bofh.it> <ujARj-2QC-3@gated-at.bofh.it> <ujMSu-1Zf-21@gated-at.bofh.it> <ujMSu-1Zf-23@gated-at.bofh.it> <ukqmR-1y0-5@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

Hello, Sorry about the delay.

On Thu, Aug 31, 2017 at 07:33:42AM +0200, Michal Hocko wrote:
> > > Michal, are you sure that this patch does not cause deadlock?
> > > 
> > > As shown in "[PATCH] mm: Use WQ_HIGHPRI for mm_percpu_wq." thread, currently work
> > > items on mm_percpu_wq seem to be blocked by other work items not on mm_percpu_wq.

IIUC that wasn't a deadlock but more a legitimate starvation from too
many tasks trying to reclaim directly.

> > But we have a rescuer so we should make a forward progress eventually.
> > Or am I missing something. Tejun, could you have a look please?
> 
> ping... I would really appreaciate if you could double check my thinking
> Tejun. This is a tricky area and I would like to prevent further subtle
> issues here.

So, this shouldn't be an issue.  This may get affected by direct
reclaim frenzy but it's only a small piece of the whole symptom and we
gotta fix that at the source.

Thanks.

-- 
tejun

Back to linux.kernel | Previous | Next — Next in thread | Find similar | Unroll thread

Thread

Re: [PATCH] mm, memory_hotplug: do not back off draining pcp free  pages from kworker context Tejun Heo <tj@kernel.org> - 2017-09-19 05:40 +0200
  Re: [PATCH] mm, memory_hotplug: do not back off draining pcp free  pages from kworker context Michal Hocko <mhocko@kernel.org> - 2017-09-19 11:50 +0200
    Re: [PATCH] mm, memory_hotplug: do not back off draining pcp free  pages from kworker context Tejun Heo <tj@kernel.org> - 2017-09-19 21:00 +0200

csiph-web