Re: [PATCH -v4][RFC]: mutex: implement adaptive spinning

[Peter Zijlstra <peterz@infradead.org>]

On Wed, 2009-01-07 at 15:50 +0100, Fr=C3=A9d=C3=A9ric Weisbecker wrote:
> 2009/1/7 Peter Zijlstra <peterz@infradead.org>:
> > Change mutex contention behaviour such that it will sometimes busy =
wait on
> > acquisition - moving its behaviour closer to that of spinlocks.
> >
> > This concept got ported to mainline from the -rt tree, where it was=
 originally
> > implemented for rtmutexes by Steven Rostedt, based on work by Grego=
ry Haskins.
> >
> > Testing with Ingo's test-mutex application (http://lkml.org/lkml/20=
06/1/8/50)
> > gave a 8% boost for VFS scalability on my testbox:
> >
> >  # echo MUTEX_SPIN > /debug/sched_features
> >  # ./test-mutex V 16 10
> >  2 CPUs, running 16 parallel test-tasks.
> >  checking VFS performance.
> >
> >  avg ops/sec:                74910
> >
> >  # echo NO_MUTEX_SPIN > /debug/sched_features
> >  # ./test-mutex V 16 10
> >  2 CPUs, running 16 parallel test-tasks.
> >  checking VFS performance.
> >
> >  avg ops/sec:                68804
> >
> > The key criteria for the busy wait is that the lock owner has to be=
 running on
> > a (different) cpu. The idea is that as long as the owner is running=
, there is a
> > fair chance it'll release the lock soon, and thus we'll be better o=
ff spinning
> > instead of blocking/scheduling.
> >
> > Since regular mutexes (as opposed to rtmutexes) do not atomically t=
rack the
> > owner, we add the owner in a non-atomic fashion and deal with the r=
aces in
> > the slowpath.
> >
> > Furthermore, to ease the testing of the performance impact of this =
new code,
> > there is means to disable this behaviour runtime (without having to=
 reboot
> > the system), when scheduler debugging is enabled (CONFIG_SCHED_DEBU=
G=3Dy),
> > by issuing the following command:
> >
> >  # echo NO_MUTEX_SPIN > /debug/sched_features
> >
> > This command re-enables spinning again (this is also the default):
> >
> >  # echo MUTEX_SPIN > /debug/sched_features
> >
> > There's also a few new statistic fields in /proc/sched_debug
> > (available if CONFIG_SCHED_DEBUG=3Dy and CONFIG_SCHEDSTATS=3Dy):
> >
> >  # grep mtx /proc/sched_debug
> >  .mtx_spin                      : 2387
> >  .mtx_sched                     : 2283
> >  .mtx_spin                      : 1277
> >  .mtx_sched                     : 1700
> >
> > Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> > Reviewed-and-signed-off-by: Ingo Molnar <mingo@elte.hu>
> > ---

> Sorry I haven't read all the previous talk about the older version.
> But it is possible that, in hopefully rare cases, you enter
> mutex_spin_or_schedule
> multiple times, and try to spin for the same lock each of these times=
=2E
>=20
> For each of the above break,
>=20
> _if you exit the spin because the mutex is unlocked, and someone else
> grab it before you
> _ or simply the owner changed...
>=20
> then you will enter again in mutex_spin_or_schedule, you have some ch=
ances that
> rq->curr =3D=3D the new owner, and then you will spin again.
> And this situation can almost really make you behave like a spinlock.=
=2E.

You understand correctly, that is indeed possible.

> Shouldn't it actually try only one time to spin, and if it calls agai=
n
> mutex_spin_or_schedule()
> then it would be better to schedule()  ?

I don't know, maybe code it up and find a benchmark where it makes a
difference. :-)
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel=
" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html