[08/11] block: always attach cgroup info into bio

Commit Message

Shaohua Li June 2, 2017, 9:54 p.m. UTC

From: Shaohua Li <[e-mail]>

blkcg_bio_issue_check() already gets blkcg for a BIO.
bio_associate_blkcg() uses a percpu refcounter, so it's a very cheap
operation. There is no point we don't attach the cgroup info into bio at
blkcg_bio_issue_check. This also makes blktrace outputs correct cgroup
info.

Signed-off-by: Shaohua Li <[e-mail]>
---
 block/blk-throttle.c       | 7 +------
 include/linux/blk-cgroup.h | 2 ++
 2 files changed, 3 insertions(+), 6 deletions(-)

Comments

Tejun Heo June 12, 2017, 6:49 p.m. UTC | #1

On Fri, Jun 02, 2017 at 02:54:01PM -0700, Shaohua Li wrote:
> @@ -691,6 +691,8 @@ static inline bool blkcg_bio_issue_check(struct request_queue *q,
>  	rcu_read_lock();
>  	blkcg = bio_blkcg(bio);
>  
> +	bio_associate_blkcg(bio, &blkcg->css);
> +

Let's please note that this only established the fallback mapping when
the bio hasn't been associated yet and doesn't override the existing
association.

Thanks.

Patch

diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index b78db2e..53d3e3d 100644
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -2033,14 +2033,9 @@  static inline void throtl_update_latency_buckets(struct throtl_data *td)
 static void blk_throtl_assoc_bio(struct throtl_grp *tg, struct bio *bio)
 {
 #ifdef CONFIG_BLK_DEV_THROTTLING_LOW
-	int ret;
-
-	ret = bio_associate_current(bio);
-	if (ret == 0 || ret == -EBUSY)
+	if (bio->bi_css)
 		bio->bi_cg_private = tg;
 	blk_stat_set_issue(&bio->bi_issue_stat, bio_sectors(bio));
-#else
-	bio_associate_current(bio);
 #endif
 }
 
diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h
index 01b62e7..d176247 100644
--- a/include/linux/blk-cgroup.h
+++ b/include/linux/blk-cgroup.h
@@ -691,6 +691,8 @@  static inline bool blkcg_bio_issue_check(struct request_queue *q,
 	rcu_read_lock();
 	blkcg = bio_blkcg(bio);
 
+	bio_associate_blkcg(bio, &blkcg->css);
+
 	blkg = blkg_lookup(blkcg, q);
 	if (unlikely(!blkg)) {
 		spin_lock_irq(q->queue_lock);