From: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
To: Jens Axboe <axboe@fb.com>,
linux-kernel@vger.kernel.org, linux-block@vger.kernel.org,
linux-mm@kvack.org, Tejun Heo <tj@kernel.org>,
cgroups@vger.kernel.org
Subject: [PATCH] cgroup writeback: use online cgroup when switching from dying bdi_writebacks
Date: Fri, 19 Jul 2019 20:46:35 +0300 [thread overview]
Message-ID: <156355839560.2063.5265687291430814589.stgit@buzz> (raw)
Offline memory cgroups forbids creation new bdi_writebacks.
Each try wastes cpu cycles and increases contention around cgwb_lock.
For example each O_DIRECT read calls filemap_write_and_wait_range()
if inode has cached pages which tries to switch from dying writeback.
This patch switches inode writeback to closest online parent cgroup.
Fixes: e8a7abf5a5bd ("writeback: disassociate inodes from dying bdi_writebacks")
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
---
fs/fs-writeback.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 542b02d170f8..3af44591a106 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -505,7 +505,7 @@ static void inode_switch_wbs(struct inode *inode, int new_wb_id)
/* find and pin the new wb */
rcu_read_lock();
memcg_css = css_from_id(new_wb_id, &memory_cgrp_subsys);
- if (memcg_css)
+ if (memcg_css && (memcg_css->flags & CSS_ONLINE))
isw->new_wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC);
rcu_read_unlock();
if (!isw->new_wb)
@@ -579,9 +579,16 @@ void wbc_attach_and_unlock_inode(struct writeback_control *wbc,
/*
* A dying wb indicates that the memcg-blkcg mapping has changed
* and a new wb is already serving the memcg. Switch immediately.
+ * If memory cgroup is offline switch to closest online parent.
*/
- if (unlikely(wb_dying(wbc->wb)))
- inode_switch_wbs(inode, wbc->wb_id);
+ if (unlikely(wb_dying(wbc->wb))) {
+ struct cgroup_subsys_state *memcg_css = wbc->wb->memcg_css;
+
+ while (!(memcg_css->flags & CSS_ONLINE))
+ memcg_css = memcg_css->parent;
+
+ inode_switch_wbs(inode, memcg_css->id);
+ }
}
EXPORT_SYMBOL_GPL(wbc_attach_and_unlock_inode);
next reply other threads:[~2019-07-19 17:46 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-19 17:46 Konstantin Khlebnikov [this message]
2019-07-19 21:13 ` [PATCH] cgroup writeback: use online cgroup when switching from dying bdi_writebacks Dennis Zhou
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=156355839560.2063.5265687291430814589.stgit@buzz \
--to=khlebnikov@yandex-team.ru \
--cc=axboe@fb.com \
--cc=cgroups@vger.kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=tj@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).