[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20200627073159.2447325-9-hch@lst.de>
Date: Sat, 27 Jun 2020 09:31:53 +0200
From: Christoph Hellwig <hch@....de>
To: Jens Axboe <axboe@...nel.dk>, Tejun Heo <tj@...nel.org>
Cc: Dennis Zhou <dennis@...nel.org>, Li Zefan <lizefan@...wei.com>,
Johannes Weiner <hannes@...xchg.org>, dm-devel@...hat.com,
linux-block@...r.kernel.org, linux-kernel@...r.kernel.org,
cgroups@...r.kernel.org, linux-mm@...ck.org
Subject: [PATCH 08/14] block: bypass blkg_tryget_closest for the root_blkg
The root_blkg is only torn down at the very end of removing a queue.
So in the I/O submission path is always has a life reference and we
can just grab another one using blkg_get instead of doing a tryget
and parent walk that won't lead anywhere.
Signed-off-by: Christoph Hellwig <hch@....de>
---
block/blk-cgroup.c | 15 ++++++++++-----
1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 6aedb73ffbf473..0912820d4f8194 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -1758,16 +1758,21 @@ void bio_associate_blkg_from_css(struct bio *bio,
struct cgroup_subsys_state *css)
{
struct request_queue *q = bio->bi_disk->queue;
- struct blkcg_gq *blkg = q->root_blkg;
if (bio->bi_blkg)
blkg_put(bio->bi_blkg);
- rcu_read_lock();
- if (css && css->parent)
+ if (css && css->parent) {
+ struct blkcg_gq *blkg;
+
+ rcu_read_lock();
blkg = blkg_lookup_create(css_to_blkcg(css), q);
- bio->bi_blkg = blkg_tryget_closest(blkg);
- rcu_read_unlock();
+ bio->bi_blkg = blkg_tryget_closest(blkg);
+ rcu_read_unlock();
+ } else {
+ blkg_get(q->root_blkg);
+ bio->bi_blkg = q->root_blkg;
+ }
}
EXPORT_SYMBOL_GPL(bio_associate_blkg_from_css);
--
2.26.2
Powered by blists - more mailing lists