[PATCH RFC] blk-mq: no need to modify bt->wake_index if someone has updated it

[Wenbo Wang]

If someone else has found active waitqueue and updated
bt->wake_index, do not modify it again. This saves an
atomic read.

Signed-off-by: Wenbo Wang 
CC: linux-bl...@vger.kernel.org
CC: linux-kernel@vger.kernel.org
---
 block/blk-mq-tag.c | 9 -
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
index abdbb47..5ed9111 100644
--- a/block/blk-mq-tag.c
+++ b/block/blk-mq-tag.c
@@ -356,16 +356,15 @@ unsigned int blk_mq_get_tag(struct blk_mq_alloc_data 
*data)
 
 static struct bt_wait_state *bt_wake_ptr(struct blk_mq_bitmap_tags *bt)
 {
-   int i, wake_index;
+   int i, wake_index, orig_wake_index;
 
-   wake_index = atomic_read(>wake_index);
+   orig_wake_index = wake_index = atomic_read(>wake_index);
for (i = 0; i < BT_WAIT_QUEUES; i++) {
struct bt_wait_state *bs = >bs[wake_index];
 
if (waitqueue_active(>wait)) {
-   int o = atomic_read(>wake_index);
-   if (wake_index != o)
-   atomic_cmpxchg(>wake_index, o, wake_index);
+   if (wake_index != orig_wake_index)
+   atomic_cmpxchg(>wake_index, 
orig_wake_index, wake_index);
 
return bs;
}
-- 
1.8.3.1
---
Not seeing too much benifit from this patch, but it makes the logic here more 
clear

[PATCH RFC] blk-mq: no need to modify bt->wake_index if someone has updated it

[Wenbo Wang]

If someone else has found active waitqueue and updated
bt->wake_index, do not modify it again. This saves an
atomic read.

Signed-off-by: Wenbo Wang 
CC: linux-bl...@vger.kernel.org
CC: linux-kernel@vger.kernel.org
---
 block/blk-mq-tag.c | 9 -
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
index abdbb47..5ed9111 100644
--- a/block/blk-mq-tag.c
+++ b/block/blk-mq-tag.c
@@ -356,16 +356,15 @@ unsigned int blk_mq_get_tag(struct blk_mq_alloc_data 
*data)
 
 static struct bt_wait_state *bt_wake_ptr(struct blk_mq_bitmap_tags *bt)
 {
-   int i, wake_index;
+   int i, wake_index, orig_wake_index;
 
-   wake_index = atomic_read(>wake_index);
+   orig_wake_index = wake_index = atomic_read(>wake_index);
for (i = 0; i < BT_WAIT_QUEUES; i++) {
struct bt_wait_state *bs = >bs[wake_index];
 
if (waitqueue_active(>wait)) {
-   int o = atomic_read(>wake_index);
-   if (wake_index != o)
-   atomic_cmpxchg(>wake_index, o, wake_index);
+   if (wake_index != orig_wake_index)
+   atomic_cmpxchg(>wake_index, 
orig_wake_index, wake_index);
 
return bs;
}
-- 
1.8.3.1
---
Not seeing too much benifit from this patch, but it makes the logic here more 
clear