On 02/12/2016 05:02 PM, Davidlohr Bueso wrote:
On Fri, 12 Feb 2016, Waiman Long wrote:
This patch adds a new waiter parameter to the mutex_optimistic_spin()
function to prepare it to be used by a waiter-spinner that doesn't
need to go into the OSQ as there can only be one waiter-spinner which
is the head of the waiting queue.
Signed-off-by: Waiman Long <waiman.l...@hp.com>
---
kernel/locking/mutex.c | 55
++++++++++++++++++++++++++++++++---------------
1 files changed, 37 insertions(+), 18 deletions(-)
diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c
index 0551c21..3c41448 100644
--- a/kernel/locking/mutex.c
+++ b/kernel/locking/mutex.c
@@ -273,11 +273,15 @@ static inline int
mutex_can_spin_on_owner(struct mutex *lock)
/*
* Atomically try to take the lock when it is available
+ *
+ * For waiter-spinner, the count needs to be set to -1 first which
will be
+ * cleared to 0 later on if the list becomes empty. For regular
spinner,
+ * the count will be set to 0.
*/
-static inline bool mutex_try_to_acquire(struct mutex *lock)
+static inline bool mutex_try_to_acquire(struct mutex *lock, int waiter)
{
return !mutex_is_locked(lock) &&
- (atomic_cmpxchg_acquire(&lock->count, 1, 0) == 1);
+ (atomic_cmpxchg_acquire(&lock->count, 1, waiter ? -1 : 0) ==
1);
}
This can be a really hot path, could we get rid of the waiter check
and just
introduce mutex_tro_to_acquire_waiter() or such and set the counter to
-1 there?
Thanks,
Davidlohr
It is hot in the sense that the lock cacheline is highly contested. On
x86, the ?: statement will most likely be translated to a cmov
instruction before doing the cmpxchg. The cmov instruction won't affect
the amount of cacheline contention on that lock cacheline. So I don't
see there is any problem here.
Cheers,
Longman
