|author||Will Deacon <firstname.lastname@example.org>||2012-08-13 17:38:48 +0000|
|committer||Varun Colbert <email@example.com>||2012-09-10 14:29:19 -0700|
ARM: mutex: use generic atomic_dec-based implementation for ARMv6+
Commit a76d7bd96d65 ("ARM: 7467/1: mutex: use generic xchg-based implementation for ARMv6+") removed the barrier-less, ARM-specific mutex implementation in favour of the generic xchg-based code. Since then, a bug was uncovered in the xchg code when running on SMP platforms, due to interactions between the locking paths and the MUTEX_SPIN_ON_OWNER code. This was fixed in 0bce9c46bf3b ("mutex: place lock in contended state after fastpath_lock failure"), however, the atomic_dec-based mutex algorithm is now marginally more efficient for ARM (~0.5% improvement in hackbench scores on dual A15). This patch moves ARMv6+ platforms to the atomic_dec-based mutex code. Change-Id: I8f64e98ccb61cc1cb9cb68ee15e55d8a792792f5 Cc: Nicolas Pitre <firstname.lastname@example.org> Signed-off-by: Will Deacon <email@example.com> Acked-by: Nicolas Pitre <firstname.lastname@example.org> Reviewed-on: http://git-master/r/130941 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Bharat Nihalani <email@example.com> Tested-by: Bharat Nihalani <firstname.lastname@example.org>
1 files changed, 7 insertions, 2 deletions
diff --git a/arch/arm/include/asm/mutex.h b/arch/arm/include/asm/mutex.h
index b1479fd04a95..6e813d93a736 100644
@@ -9,8 +9,13 @@
* On pre-ARMv6 hardware this results in a swp-based implementation,
- * which is the most efficient. For ARMv6+, we emit a pair of exclusive
- * accesses instead.
+ * which is the most efficient. For ARMv6+, we have exclusive memory
+ * accessors and use atomic_dec to avoid the extra xchg operations
+ * on the locking slowpaths.
+#if __LINUX_ARM_ARCH__ < 6