Re: [PATCH] signals: work around random wakeups in sigsuspend()

[Peter Zijlstra]

On Thu, Feb 25, 2016 at 03:18:52AM +0000, Al Viro wrote:
> On Mon, Jan 25, 2016 at 08:09:15PM +0100, Oleg Nesterov wrote:
> > On 01/25, Sasha Levin wrote:
> > >
> > > A random wakeup can get us out of sigsuspend() without TIF_SIGPENDING
> > > being set.
> > 
> > and TIF_RESTORE_SIGMASK is just wrong in this case. I'd say this is the
> > bugfix, not work-around ;)
> > 
> > > Avoid that by making sure we were signaled, like sys_pause() does.
> > >
> > > Signed-off-by: Sasha Levin <sasha.levin@xxxxxxxxxx>
> > 
> > Acked-by: Oleg Nesterov <oleg@xxxxxxxxxx>
> > 
> > Thanks Sasha.
> 
> Out of curiousity - where did that stray wakeup come from?  PTRACE_KILL
> used to trigger those, but that got fixed.  How does one trigger that
> kind of bugs on the current kernels?

Its a regular TASK_INTERRUPTIBLE sleep, for those spurious wakeups are
not a bug, they're pretty fundamentally allowed.

See: lkml.kernel.org/r/CA+55aFwHkOo+YGWKYROmce1-H_uG3KfEUmCkJUerTj=ojY2H6Q@xxxxxxxxxxxxxx

But they became a lot more common because we explicitly used the delayed
wakeup pattern described there with wake_q. See commits:

  7675104990ed ("sched: Implement lockless wake-queues")
  1d0dcb3ad9d3 ("futex: Implement lockless wakeups")