[PATCH] block: adjust the judgment order in submit_bio

[PATCH] block: adjust the judgment order in submit_bio

[Liu Song]

From: Liu Song <liusong@linux.alibaba.com>

BIO_WORKINGSET is rarer than read, so adjust to the first one to judge.

Signed-off-by: Liu Song <liusong@linux.alibaba.com>
---
 block/blk-core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index 06ff5bb..7b6809b 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -901,8 +901,8 @@ void submit_bio(struct bio *bio)
 	 * the submitting cgroup IO-throttled, submission can be a significant
 	 * part of overall IO time.
 	 */
-	if (unlikely(bio_op(bio) == REQ_OP_READ &&
-	    bio_flagged(bio, BIO_WORKINGSET))) {
+	if (unlikely(bio_flagged(bio, BIO_WORKINGSET) &&
+		bio_op(bio) == REQ_OP_READ)) {
 		unsigned long pflags;
 
 		psi_memstall_enter(&pflags);
-- 
1.8.3.1

Re: [PATCH] block: adjust the judgment order in submit_bio

[Christoph Hellwig]

On Wed, Jun 15, 2022 at 05:47:22PM +0800, Liu Song wrote:
> From: Liu Song <liusong@linux.alibaba.com>
> 
> BIO_WORKINGSET is rarer than read, so adjust to the first one to judge.

Can you explain how this matters at all?  The right thing is of course
to get rid of this check and I need to finish the series to do that.

Re: [PATCH] block: adjust the judgment order in submit_bio

[liusong]

>> From: Liu Song <liusong@linux.alibaba.com>
>> 
>> BIO_WORKINGSET is rarer than read, so adjust to the first one to judge.
>
> Can you explain how this matters at all?  The right thing is of course
> to get rid of this check and I need to finish the series to do that.

Hi
    This change do not involve functional changes, I just feel that it needs to
be emphasized that for example, the io triggered by refault is much less than
the read.

Thanks