Re: [PATCH] mm: vmscan: memcontrol: remove mem_cgroup_select_victim_node()

[Shakeel Butt]

On Tue, Oct 29, 2019 at 5:55 PM Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> On Tue, 29 Oct 2019 16:47:53 -0700 Shakeel Butt <shakeelb@xxxxxxxxxx> wrote:
>
> > Since commit 1ba6fc9af35b ("mm: vmscan: do not share cgroup iteration
> > between reclaimers"), the memcg reclaim does not bail out earlier based
> > on sc->nr_reclaimed and will traverse all the nodes. All the reclaimable
> > pages of the memcg on all the nodes will be scanned relative to the
> > reclaim priority. So, there is no need to maintain state regarding which
> > node to start the memcg reclaim from. Also KCSAN complains data races in
> > the code maintaining the state.
> >
> > This patch effectively reverts the commit 889976dbcb12 ("memcg: reclaim
> > memory from nodes in round-robin order") and the commit 453a9bf347f1
> > ("memcg: fix numa scan information update to be triggered by memory
> > event").
> >
> > Signed-off-by: Shakeel Butt <shakeelb@xxxxxxxxxx>
> > Reported-by: <syzbot+13f93c99c06988391efe@xxxxxxxxxxxxxxxxxxxxxxxxx>
>
> I can't find the original sysbot email.  Help?
>
> iirc the incidentally-fixed issue is a rather theoretical data race and
> the patch isn't a high priority thing?
>

Let me check why the link is not working. However you are right that
the fix is for a theoretical data race and no need to be backported to
stable trees.

Shakeel