Re: [PATCH 1/1] signal: Adjust error codes according to restore_user_sigmask()
From: Deepa Dinamani
Date: Tue May 21 2019 - 20:47:14 EST
On Tue, May 21, 2019 at 5:35 PM Deepa Dinamani <deepa.kernel@xxxxxxxxx> wrote:
>
> > > > It's been 2 weeks and this fix hasn't appeared in mmots / mmotm.
> > > > I also noticed it's missing Cc: for stable@ (below)
> > >
> > > Why is a -stable backport needed? I see some talk above about lost
> > > signals but it is unclear whether these are being observed after fixing
> > > the regression caused by 854a6ed56839a.
> >
> > I guess Deepa's commit messages wasn't clear...
> > I suggest prepending this as the first paragraph to Deepa's
> > original message:
> >
> > This fixes a bug introduced with 854a6ed56839a which caused
> > EINTR to not be reported to userspace on epoll_pwait. Failure
> > to report EINTR to userspace caused problems with user code
> > which relies on EINTR to run signal handlers.
>
> This is not what the patch fixed.
>
> The notable change is userspace is that now whenever a signal is
> delivered, the return value is adjusted to reflect the signal
> delivery.
> Prior to this patch, there was a window, however small it might have
> been, when the signal was delivered but the errono was not adjusted
> appropriately.
> This is because of the regression caused by 854a6ed56839a, which
> extended the window of delivery of signals that was delivered to
> userspace.
> The patch also fixes more than sys_epoll_pwait().
>
> I will post a follow up patch.
>
> >
> > > IOW, can we please have a changelog which has a clear and complete
> > > description of the user-visible effects of the change.
> > >
> > > And please Cc Oleg.
>
> I will cc Oleg.
Also the commit message was brief because the issue was explained in
the link that was quoted in the commit message.
Detailed issue discussion permalink:
https://lore.kernel.org/linux-fsdevel/20190427093319.sgicqik2oqkez3wk@dcvr/
-Deepa