Re: [LKP] [mm] c8c06efa8b5: -7.6% unixbench.score

[Peter Zijlstra <peterz@infradead.org>]

On Thu, Jan 08, 2015 at 12:59:59AM -0800, Davidlohr Bueso wrote:
> > > >     721721 ±  1%    +303.6%    2913110 ±  3%  unixbench.time.voluntary_context_switches
> > > >      11767 ±  0%      -7.6%      10867 ±  1%  unixbench.score

> heh I was actually looking at the reader code. We really do:
> 
> 	/* wait until we successfully acquire the lock */
> 	set_current_state(TASK_UNINTERRUPTIBLE);
> 	while (true) {
> 		if (rwsem_try_write_lock(count, sem))
> 			break;
> 		raw_spin_unlock_irq(&sem->wait_lock);
> 
> 		/* Block until there are no active lockers. */
> 		do {
> 			schedule();
> 			set_current_state(TASK_UNINTERRUPTIBLE);
> 		} while ((count = sem->count) & RWSEM_ACTIVE_MASK);
> 
> 		raw_spin_lock_irq(&sem->wait_lock);
> 	}
> 
> 
> Which still has similar issues with even two barriers, I guess for both
> the rwsem_try_write_lock call (less severe) and count checks. Anyway...

So its actually scheduling a lot more, this could also mean the opt
spinning thing isn't working as well (I've no real idea what the
workload is).

One thing I noticed is that we set sem->owner very late in comparison
with the mutex code, this could cause us to break out of the spin loop
prematurely.