Commit-ID: 5dae63c442131f1b0a66abd43fdc861031f13ca6 Gitweb: http://git.kernel.org/tip/5dae63c442131f1b0a66abd43fdc861031f13ca6 Author: Yuanhan Liu <yuanhan....@linux.intel.com> AuthorDate: Fri, 1 Feb 2013 18:59:16 +0800 Committer: Ingo Molnar <mi...@kernel.org> CommitDate: Mon, 18 Feb 2013 10:10:21 +0100
rwsem-spinlock: Implement writer lock-stealing for better scalability We (Linux Kernel Performance project) found a regression introduced by commit: 5a505085f043 mm/rmap: Convert the struct anon_vma::mutex to an rwsem which converted all anon_vma::mutex locks rwsem write locks. The semantics are the same, but the behavioral difference is quite huge in some cases. After investigating it we found the root cause: mutexes support lock stealing while rwsems don't. Here is the link for the detailed regression report: https://lkml.org/lkml/2013/1/29/84 Ingo suggested adding write lock stealing to rwsems: "I think we should allow lock-steal between rwsem writers - that will not hurt fairness as most rwsem fairness concerns relate to reader vs. writer fairness" And here is the rwsem-spinlock version. With this patch, we got a double performance increase in one test box with following aim7 workfile: FILESIZE: 1M POOLSIZE: 10M 10 fork_test /usr/bin/time output w/o patch /usr/bin/time_output with patch -- Percent of CPU this job got: 369% Percent of CPU this job got: 537% Voluntary context switches: 640595016 Voluntary context switches: 157915561 We got a 45% increase in CPU usage and saved about 3/4 voluntary context switches. Reported-by: LKP project <l...@linux.intel.com> Suggested-by: Ingo Molnar <mi...@kernel.org> Signed-off-by: Yuanhan Liu <yuanhan....@linux.intel.com> Cc: Alex Shi <alex....@intel.com> Cc: David Howells <dhowe...@redhat.com> Cc: Michel Lespinasse <wal...@google.com> Cc: Linus Torvalds <torva...@linux-foundation.org> Cc: Andrew Morton <a...@linux-foundation.org> Cc: Peter Zijlstra <a.p.zijls...@chello.nl> Cc: Anton Blanchard <an...@samba.org> Cc: Arjan van de Ven <ar...@linux.intel.com> Cc: paul.gortma...@windriver.com Link: http://lkml.kernel.org/r/1359716356-23865-1-git-send-email-yuanhan....@linux.intel.com Signed-off-by: Ingo Molnar <mi...@kernel.org> --- lib/rwsem-spinlock.c | 69 ++++++++++++++++++---------------------------------- 1 file changed, 24 insertions(+), 45 deletions(-) diff --git a/lib/rwsem-spinlock.c b/lib/rwsem-spinlock.c index 7e0d6a5..7542afb 100644 --- a/lib/rwsem-spinlock.c +++ b/lib/rwsem-spinlock.c @@ -73,20 +73,13 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wakewrite) goto dont_wake_writers; } - /* if we are allowed to wake writers try to grant a single write lock - * if there's a writer at the front of the queue - * - we leave the 'waiting count' incremented to signify potential - * contention + /* + * as we support write lock stealing, we can't set sem->activity + * to -1 here to indicate we get the lock. Instead, we wake it up + * to let it go get it again. */ if (waiter->flags & RWSEM_WAITING_FOR_WRITE) { - sem->activity = -1; - list_del(&waiter->list); - tsk = waiter->task; - /* Don't touch waiter after ->task has been NULLed */ - smp_mb(); - waiter->task = NULL; - wake_up_process(tsk); - put_task_struct(tsk); + wake_up_process(waiter->task); goto out; } @@ -121,18 +114,10 @@ static inline struct rw_semaphore * __rwsem_wake_one_writer(struct rw_semaphore *sem) { struct rwsem_waiter *waiter; - struct task_struct *tsk; - - sem->activity = -1; waiter = list_entry(sem->wait_list.next, struct rwsem_waiter, list); - list_del(&waiter->list); + wake_up_process(waiter->task); - tsk = waiter->task; - smp_mb(); - waiter->task = NULL; - wake_up_process(tsk); - put_task_struct(tsk); return sem; } @@ -204,7 +189,6 @@ int __down_read_trylock(struct rw_semaphore *sem) /* * get a write lock on the semaphore - * - we increment the waiting count anyway to indicate an exclusive lock */ void __sched __down_write_nested(struct rw_semaphore *sem, int subclass) { @@ -214,37 +198,32 @@ void __sched __down_write_nested(struct rw_semaphore *sem, int subclass) raw_spin_lock_irqsave(&sem->wait_lock, flags); - if (sem->activity == 0 && list_empty(&sem->wait_list)) { - /* granted */ - sem->activity = -1; - raw_spin_unlock_irqrestore(&sem->wait_lock, flags); - goto out; - } - - tsk = current; - set_task_state(tsk, TASK_UNINTERRUPTIBLE); - /* set up my own style of waitqueue */ + tsk = current; waiter.task = tsk; waiter.flags = RWSEM_WAITING_FOR_WRITE; - get_task_struct(tsk); - list_add_tail(&waiter.list, &sem->wait_list); - /* we don't need to touch the semaphore struct anymore */ - raw_spin_unlock_irqrestore(&sem->wait_lock, flags); - - /* wait to be given the lock */ + /* wait for someone to release the lock */ for (;;) { - if (!waiter.task) + /* + * That is the key to support write lock stealing: allows the + * task already on CPU to get the lock soon rather than put + * itself into sleep and waiting for system woke it or someone + * else in the head of the wait list up. + */ + if (sem->activity == 0) break; - schedule(); set_task_state(tsk, TASK_UNINTERRUPTIBLE); + raw_spin_unlock_irqrestore(&sem->wait_lock, flags); + schedule(); + raw_spin_lock_irqsave(&sem->wait_lock, flags); } + /* got the lock */ + sem->activity = -1; + list_del(&waiter.list); - tsk->state = TASK_RUNNING; - out: - ; + raw_spin_unlock_irqrestore(&sem->wait_lock, flags); } void __sched __down_write(struct rw_semaphore *sem) @@ -262,8 +241,8 @@ int __down_write_trylock(struct rw_semaphore *sem) raw_spin_lock_irqsave(&sem->wait_lock, flags); - if (sem->activity == 0 && list_empty(&sem->wait_list)) { - /* granted */ + if (sem->activity == 0) { + /* got the lock */ sem->activity = -1; ret = 1; } -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
