Re: [PATCH -v4 6/8] locking/mutex: Restructure wait loop

[Peter Zijlstra]

On Mon, Oct 17, 2016 at 07:16:50PM -0400, Waiman Long wrote:
> >+++ b/kernel/locking/mutex.c
> >@@ -631,13 +631,21 @@ __mutex_lock_common(struct mutex *lock,
> >
> >  	lock_contended(&lock->dep_map, ip);
> >
> >+	set_task_state(task, state);
> 
> Do we want to set the state here? I am not sure if it is OK to set the task
> state without ever calling schedule().

That's entirely fine, note how we'll set it back to RUNNING at the end.

> >  	for (;;) {
> >+		/*
> >+		 * Once we hold wait_lock, we're serialized against
> >+		 * mutex_unlock() handing the lock off to us, do a trylock
> >+		 * before testing the error conditions to make sure we pick up
> >+		 * the handoff.
> >+		 */
> >  		if (__mutex_trylock(lock, first))
> >-			break;
> >+			goto acquired;
> >
> >  		/*
> >-		 * got a signal? (This code gets eliminated in the
> >-		 * TASK_UNINTERRUPTIBLE case.)
> >+		 * Check for signals and wound conditions while holding
> >+		 * wait_lock. This ensures the lock cancellation is ordered
> >+		 * against mutex_unlock() and wake-ups do not go missing.
> >  		 */
> >  		if (unlikely(signal_pending_state(state, task))) {
> >  			ret = -EINTR;
> >@@ -650,16 +658,27 @@ __mutex_lock_common(struct mutex *lock,
> >  				goto err;
> >  		}
> >
> >-		__set_task_state(task, state);
> >  		spin_unlock_mutex(&lock->wait_lock, flags);
> >  		schedule_preempt_disabled();
> >-		spin_lock_mutex(&lock->wait_lock, flags);
> >
> >  		if (!first&&  __mutex_waiter_is_first(lock,&waiter)) {
> >  			first = true;
> >  			__mutex_set_flag(lock, MUTEX_FLAG_HANDOFF);
> >  		}
> >+
> >+		set_task_state(task, state);
> 
> I would suggest keep the __set_task_state() above and change
> set_task_state(task, state) to set_task_state(task, TASK_RUNNING) to provide
> the memory barrier. Then we don't need adding __set_task_state() calls
> below.

set_task_state(RUNNING) doesn't make sense, ever.

See the comment near set_task_state() for the reason it has a barrier.

We need it here because when we do that trylock (or optimistic spin) we
need to have set the state and done a barrier, otherwise we can miss a
wakeup and get stuck.

> >+		/*
> >+		 * Here we order against unlock; we must either see it change
> >+		 * state back to RUNNING and fall through the next schedule(),
> >+		 * or we must see its unlock and acquire.
> >+		 */
> >+		if (__mutex_trylock(lock, first))
> >+			break;
> >+
> 
> I don't think we need a trylock here since we are going to do it at the top
> of the loop within wait_lock anyway.

The idea was to avoid the wait-time of that lock acquire, also, this is
a place-holder for the optimistic spin site for the next patch.