[PATCH] locks should not be needed for reading completion state

[PATCH] locks should not be needed for reading completion state

[Nicholas Mc Guire]

Signed-off-by: Nicholas Mc Guire <der.herr@hofr.at>
---

Should that not be enough to wrap it into ACCESS_ONCE() here ?

Not a wild performance benefit - but might be worth it anyway

 kernel/sched/completion.c |    9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c
index 607f852..7c5cd70 100644
--- a/kernel/sched/completion.c
+++ b/kernel/sched/completion.c
@@ -288,13 +288,6 @@ EXPORT_SYMBOL(try_wait_for_completion);
  */
 bool completion_done(struct completion *x)
 {
-	unsigned long flags;
-	int ret = 1;
-
-	spin_lock_irqsave(&x->wait.lock, flags);
-	if (!x->done)
-		ret = 0;
-	spin_unlock_irqrestore(&x->wait.lock, flags);
-	return ret;
+	return !ACCESS_ONCE(x->done);
 }
 EXPORT_SYMBOL(completion_done);
-- 
1.7.10.4

[tip:locking/core] sched/completion: Remove unnecessary -> wait.lock serialization when reading completion state

[tip-bot for Nicholas Mc Guire]

Commit-ID:  de30ec47302c101c7badc8fe687641fd75e596e7
Gitweb:     http://git.kernel.org/tip/de30ec47302c101c7badc8fe687641fd75e596e7
Author:     Nicholas Mc Guire <der.herr@hofr.at>
AuthorDate: Sat, 17 Jan 2015 05:05:34 +0100
Committer:  Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 4 Feb 2015 07:57:36 +0100

sched/completion: Remove unnecessary ->wait.lock serialization when reading completion state

Signed-off-by: Nicholas Mc Guire <der.herr@hofr.at>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1421467534-22834-1-git-send-email-der.herr@hofr.at
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
 kernel/sched/completion.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c
index 607f852..9d1fe32d 100644
--- a/kernel/sched/completion.c
+++ b/kernel/sched/completion.c
@@ -288,13 +288,6 @@ EXPORT_SYMBOL(try_wait_for_completion);
  */
 bool completion_done(struct completion *x)
 {
-	unsigned long flags;
-	int ret = 1;
-
-	spin_lock_irqsave(&x->wait.lock, flags);
-	if (!x->done)
-		ret = 0;
-	spin_unlock_irqrestore(&x->wait.lock, flags);
-	return ret;
+	return !!ACCESS_ONCE(x->done);
 }
 EXPORT_SYMBOL(completion_done);