Re: [PATCH -v5 12/14] futex,rt_mutex: Restructure rt_mutex_finish_proxy_lock()

[Peter Zijlstra]

On Tue, Mar 07, 2017 at 03:18:46PM +0100, Thomas Gleixner wrote:
> On Sat, 4 Mar 2017, Peter Zijlstra wrote:
> > +/**
> > + * rt_mutex_cleanup_proxy_lock() - Cleanup failed lock acquisition
> > + * @lock:		the rt_mutex we were woken on
> > + * @waiter:		the pre-initialized rt_mutex_waiter
> > + *
> > + * Clean up the failed lock acquisition as per rt_mutex_wait_proxy_lock().
> > + *
> > + * Returns:
> > + *  true  - did the cleanup, we done.
> > + *  false - we acquired the lock after rt_mutex_wait_proxy_lock() returned,
> > + *          caller should disregards its return value.
> 
> Hmm. How would that happen? Magic owner assignement to a non waiter? The
> callsite only calls here in the failed case.

Ah, but until the remove_waiter() below, we _still_ are a waiter, and
thus can get assigned ownership.

> > + *
> > + * Special API call for PI-futex support
> > + */
> > +bool rt_mutex_cleanup_proxy_lock(struct rt_mutex *lock,
> > +				 struct rt_mutex_waiter *waiter)
> > +{
> > +	bool cleanup = false;
> > +
> > +	raw_spin_lock_irq(&lock->wait_lock);
> > +	/*
> > +	 * If we acquired the lock, no cleanup required.
> > +	 */
> > +	if (rt_mutex_owner(lock) != current) {
> > +		remove_waiter(lock, waiter);

See, up till this point, we still a waiter and any unlock can see us
being one.

> > +		fixup_rt_mutex_waiters(lock);
> > +		cleanup = true;
> > +	}
> > +	raw_spin_unlock_irq(&lock->wait_lock);
> > +
> > +	return cleanup;
> > +}