[PATCH] powerpc/qspinlock: Fix deadlock in MCS queue

[Nysal Jan K.A.]

On Wed, Aug 28, 2024 at 01:52:33PM GMT, Michael Ellerman wrote:
> "Nysal Jan K.A." <nysal at linux.ibm.com> writes:
> > If an interrupt occurs in queued_spin_lock_slowpath() after we increment
> > qnodesp->count and before node->lock is initialized, another CPU might
> > see stale lock values in get_tail_qnode(). If the stale lock value happens
> > to match the lock on that CPU, then we write to the "next" pointer of
> > the wrong qnode. This causes a deadlock as the former CPU, once it becomes
> > the head of the MCS queue, will spin indefinitely until it's "next" pointer
> > is set by its successor in the queue. This results in lockups similar to
> > the following.
> ...
> >
> > Thanks to Saket Kumar Bhaskar for help with recreating the issue
> >
> > Fixes: 84990b169557 ("powerpc/qspinlock: add mcs queueing for contended waiters")
> > Cc: stable at vger.kernel.org # v6.2+
> > Reported-by: Geetika Moolchandani <geetika at linux.ibm.com>
> > Reported-by: Vaishnavi Bhat <vaish123 at in.ibm.com>
> > Reported-by: Jijo Varghese <vargjijo at in.ibm.com>
>  
> Do we have links for any of these reports?

They are all internal reports on LTC bugzilla. I don't see one that is public.

> 
> cheers

Regards
--Nysal