Re: [PATCH v3] psi: get poll_work to run when calling poll syscall next time

[Suren Baghdasaryan]

On Thu, Aug 22, 2019 at 4:17 PM Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> On Thu, 22 Aug 2019 16:11:15 -0700 Suren Baghdasaryan <surenb@xxxxxxxxxx> wrote:
>
> > On Thu, Aug 22, 2019 at 3:21 PM Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
> > >
> > > On Wed, 21 Aug 2019 11:26:25 +0800 Joseph Qi <joseph.qi@xxxxxxxxxxxxxxxxx> wrote:
> > >
> > > > Only when calling the poll syscall the first time can user
> > > > receive POLLPRI correctly. After that, user always fails to
> > > > acquire the event signal.
> > > >
> > > > Reproduce case:
> > > > 1. Get the monitor code in Documentation/accounting/psi.txt
> > > > 2. Run it, and wait for the event triggered.
> > > > 3. Kill and restart the process.
> > > >
> > > > The question is why we can end up with poll_scheduled = 1 but the work
> > > > not running (which would reset it to 0). And the answer is because the
> > > > scheduling side sees group->poll_kworker under RCU protection and then
> > > > schedules it, but here we cancel the work and destroy the worker. The
> > > > cancel needs to pair with resetting the poll_scheduled flag.
> > >
> > > Should this be backported into -stable kernels?
> >
> > Adding GregKH and stable@xxxxxxxxxxxxxxx
> >
> > I was able to cleanly apply this patch to stable master and
> > linux-5.2.y branches (these are the only branches that have psi
> > triggers).
> > Greg, Andrew got this patch into -mm tree. Please advise on how we
> > should proceed to land it in stable 5.2.y and master.
>
> That isn't the point - we know how to merge patches ;)
>
> What I'm asking is whether it is desirable that -stable kernels have
> this patch.  It certainly sounds like it from the changelog, so I'm
> wondering if the omission of cc:stable was intentional?

Sorry for my misunderstanding. I believe cc:stable omission was
unintentional. It's a fix for a bug which exists in stable branches I
mentioned above.
Thanks!