Re: [PATCH v2 2/5] mutex: Modify the way optimistic spinners arequeued

From: Paul E. McKenney
Date: Tue Jan 28 2014 - 15:23:49 EST


On Tue, Jan 28, 2014 at 11:13:13AM -0800, Jason Low wrote:
> The mutex->spin_mlock was introduced in order to ensure that only 1 thread
> spins for lock acquisition at a time to reduce cache line contention. When
> lock->owner is NULL and the lock->count is still not 1, the spinner(s) will
> continually release and obtain the lock->spin_mlock. This can generate
> quite a bit of overhead/contention, and also might just delay the spinner
> from getting the lock.
>
> This patch modifies the way optimistic spinners are queued by queuing before
> entering the optimistic spinning loop as oppose to acquiring before every
> call to mutex_spin_on_owner(). So in situations where the spinner requires
> a few extra spins before obtaining the lock, then there will only be 1 spinner
> trying to get the lock and it will avoid the overhead from unnecessarily
> unlocking and locking the spin_mlock.
>
> Signed-off-by: Jason Low <jason.low2@xxxxxx>

One question below. Also, this might well have a visible effect on
performance, so would be good to see the numbers.

Thanx, Paul

> ---
> kernel/locking/mutex.c | 16 +++++++---------
> 1 files changed, 7 insertions(+), 9 deletions(-)
>
> diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c
> index 85c6be1..7519d27 100644
> --- a/kernel/locking/mutex.c
> +++ b/kernel/locking/mutex.c
> @@ -419,6 +419,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
> struct mutex_waiter waiter;
> unsigned long flags;
> int ret;
> + struct mspin_node node;
>
> preempt_disable();
> mutex_acquire_nest(&lock->dep_map, subclass, 0, nest_lock, ip);
> @@ -449,9 +450,9 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
> if (!mutex_can_spin_on_owner(lock))
> goto slowpath;
>
> + mspin_lock(MLOCK(lock), &node);
> for (;;) {
> struct task_struct *owner;
> - struct mspin_node node;
>
> if (use_ww_ctx && ww_ctx->acquired > 0) {
> struct ww_mutex *ww;
> @@ -466,19 +467,16 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
> * performed the optimistic spinning cannot be done.
> */
> if (ACCESS_ONCE(ww->ctx))
> - goto slowpath;
> + break;
> }
>
> /*
> * If there's an owner, wait for it to either
> * release the lock or go to sleep.
> */
> - mspin_lock(MLOCK(lock), &node);
> owner = ACCESS_ONCE(lock->owner);
> - if (owner && !mutex_spin_on_owner(lock, owner)) {
> - mspin_unlock(MLOCK(lock), &node);
> - goto slowpath;
> - }
> + if (owner && !mutex_spin_on_owner(lock, owner))
> + break;
>
> if ((atomic_read(&lock->count) == 1) &&
> (atomic_cmpxchg(&lock->count, 1, 0) == 1)) {
> @@ -495,7 +493,6 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
> preempt_enable();
> return 0;
> }
> - mspin_unlock(MLOCK(lock), &node);
>
> /*
> * When there's no owner, we might have preempted between the
> @@ -504,7 +501,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
> * the owner complete.
> */
> if (!owner && (need_resched() || rt_task(task)))
> - goto slowpath;
> + break;
>
> /*
> * The cpu_relax() call is a compiler barrier which forces
> @@ -514,6 +511,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
> */
> arch_mutex_cpu_relax();
> }
> + mspin_unlock(MLOCK(lock), &node);
> slowpath:

Are there any remaining goto statements to slowpath? If so, they need
to release the lock. If not, this label should be removed.

> #endif
> spin_lock_mutex(&lock->wait_lock, flags);
> --
> 1.7.1
>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/