sbitmap: replace CAS with atomic and

sbitmap_deferred_clear() does CAS loop to propagate cleared bits,
replace it with equivalent atomic bitwise and. That's slightly faster
and makes wait-free instead of lock-free as before.

The atomic can be relaxed (i.e. barrier-less) because following
sbitmap_get*() deal with synchronisation, see comments in
sbitmap_queue_clear().

It's ok to cast to atomic_long_t, that's what bitops/lock.h does.

Signed-off-by: Pavel Begunkov <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>

diff --git a/lib/sbitmap.c b/lib/sbitmap.c
index 4fd877048ba82dde1fab62a73c97c1013f63029f..c18b518a16ba1beaf58cc2b5ac51aac01d654ddc 100644 (file)
--- a/lib/sbitmap.c
+++ b/lib/sbitmap.c
@@ -14,7 +14,7 @@
  */
 static inline bool sbitmap_deferred_clear(struct sbitmap_word *map)
 {
-       unsigned long mask, val;
+       unsigned long mask;
 
        if (!READ_ONCE(map->cleared))
                return false;
@@ -27,10 +27,8 @@ static inline bool sbitmap_deferred_clear(struct sbitmap_word *map)
        /*
         * Now clear the masked bits in our free word
         */
-       do {
-               val = map->word;
-       } while (cmpxchg(&map->word, val, val & ~mask) != val);
-
+       atomic_long_andnot(mask, (atomic_long_t *)&map->word);
+       BUILD_BUG_ON(sizeof(atomic_long_t) != sizeof(map->word));
        return true;
 }