The lockref cmpxchg loop is unbound as long as the spinlock is not
taken. Depending on the hardware implementation of compare-and-swap
a high number of loop retries might happen.
Add an upper bound to the loop to force the fallback to spinlocks
after some time. A retry value of 100 should not impact any hardware
that does not have this issue.
With the retry limit the performance of an open-close testcase
improved between 60-70% on ThunderX2.
Suggested-by: Linus Torvalds <torva...@linux-foundation.org>
Signed-off-by: Jan Glauber <jglau...@marvell.com>
---
lib/lockref.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/lib/lockref.c b/lib/lockref.c
index 3d468b53d4c9..5b34bbd3eba8 100644
--- a/lib/lockref.c
+++ b/lib/lockref.c
@@ -9,6 +9,7 @@
* failure case.
*/
#define CMPXCHG_LOOP(CODE, SUCCESS) do {
\
+ int retry = 100;
\
struct lockref old;
\
BUILD_BUG_ON(sizeof(old) != 8);
\
old.lock_count = READ_ONCE(lockref->lock_count);
\
@@ -21,6 +22,8 @@
if (likely(old.lock_count == prev.lock_count)) {
\
SUCCESS;
\
}
\
+ if (!--retry)
\
+ break;
\
cpu_relax();
\
}
\
} while (0)
--
2.21.0
