The patch titled
Subject: ipc-semc-fix-complex_count-vs-simple-op-race-update
has been removed from the -mm tree. Its filename was
ipc-semc-fix-complex_count-vs-simple-op-race-update.patch
This patch was dropped because it was folded into ipc-semc-fix-complex_count-vs-simple-op-race.patch
------------------------------------------------------
From: Manfred Spraul <manfred@xxxxxxxxxxxxxxxx>
Subject: ipc-semc-fix-complex_count-vs-simple-op-race-update
Link: http://lkml.kernel.org/r/1468386412-3608-2-git-send-email-manfred@xxxxxxxxxxxxxxxx
Signed-off-by: Manfred Spraul <manfred@xxxxxxxxxxxxxxxx>
Cc: Davidlohr Bueso <dave@xxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
ipc/sem.c | 12 ++++++++++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff -puN ipc/sem.c~ipc-semc-fix-complex_count-vs-simple-op-race-update ipc/sem.c
--- a/ipc/sem.c~ipc-semc-fix-complex_count-vs-simple-op-race-update
+++ a/ipc/sem.c
@@ -291,6 +291,14 @@ static void complexmode_enter(struct sem
sem = sma->sem_base + i;
spin_unlock_wait(&sem->lock);
}
+ /*
+ * spin_unlock_wait() is not a memory barriers, it is only a
+ * control barrier. The code must pair with spin_unlock(&sem->lock),
+ * thus just the control barrier is insufficient.
+ *
+ * smp_rmb() is sufficient, as writes cannot pass the control barrier.
+ */
+ smp_rmb();
}
/*
@@ -346,7 +354,7 @@ static inline int sem_lock(struct sem_ar
/*
* Initial check for complex_mode. Just an optimization,
- * no locking.
+ * no locking, no memory barrier.
*/
if (!READ_ONCE(sma->complex_mode)) {
/*
@@ -361,7 +369,7 @@ static inline int sem_lock(struct sem_ar
*/
smp_mb();
- if (!READ_ONCE(sma->complex_mode)) {
+ if (!smp_load_acquire(&sma->complex_mode)) {
/* fast path successful! */
return sops->sem_num;
}
_
Patches currently in -mm which might be from manfred@xxxxxxxxxxxxxxxx are
ipc-semc-fix-complex_count-vs-simple-op-race.patch
ipc-semc-remove-duplicated-memory-barriers.patch
ipc-sem-sem_lock-with-hysteresis.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
