lkml.org 
[lkml]   [2013]   [Mar]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH 08/12] rwsem: avoid taking wait_lock in rwsem_down_write_failed
Date
In rwsem_down_write_failed(), if there are active locks after we wake up
(i.e. the lock got stolen from us), skip taking the wait_lock and go back
to sleep immediately.

Signed-off-by: Michel Lespinasse <walken@google.com>

---
lib/rwsem.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/lib/rwsem.c b/lib/rwsem.c
index 64c2dc007be2..edf3d9ca670e 100644
--- a/lib/rwsem.c
+++ b/lib/rwsem.c
@@ -214,8 +214,8 @@ struct rw_semaphore __sched *rwsem_down_write_failed(struct rw_semaphore *sem)
sem = __rwsem_do_wake(sem, RWSEM_WAKE_READ_OWNED);

/* wait until we successfully acquire the lock */
+ set_task_state(tsk, TASK_UNINTERRUPTIBLE);
while (true) {
- set_task_state(tsk, TASK_UNINTERRUPTIBLE);

/* Try acquiring the write lock. */
count = RWSEM_ACTIVE_WRITE_BIAS;
@@ -226,7 +226,13 @@ struct rw_semaphore __sched *rwsem_down_write_failed(struct rw_semaphore *sem)
break;

raw_spin_unlock_irq(&sem->wait_lock);
- schedule();
+
+ /* Block until there are no active lockers. */
+ do {
+ schedule();
+ set_task_state(tsk, TASK_UNINTERRUPTIBLE);
+ } while (sem->count & RWSEM_ACTIVE_MASK);
+
raw_spin_lock_irq(&sem->wait_lock);
}

--
1.8.1.3

\
 
 \ /
  Last update: 2013-03-07 01:01    [W:0.130 / U:0.240 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site