From: "NeilBrown" <neilb@suse.de>
To: Chuck Lever <chuck.lever@oracle.com>,
"J. Bruce Fields" <bfields@fieldses.org>
Cc: linux-nfs@vger.kernel.org
Subject: [PATCH] SUNRPC: use congestion_wait() in svc_alloc_args()
Date: Mon, 06 Sep 2021 14:44:08 +1000 [thread overview]
Message-ID: <163090344807.19339.10071205771966144716@noble.neil.brown.name> (raw)
Many places that need to wait before retrying a memory allocation use
congestion_wait(). xfs_buf_alloc_pages() is a good example which
follows a similar pattern to that in svc_alloc_args().
It make sense to do the same thing in svc_alloc_args(); This will allow
the allocation to be retried sooner if some backing device becomes
non-congested before the timeout.
Every call to congestion_wait() in the entire kernel passes BLK_RW_ASYNC
as the first argument, so we should so.
The second argument - an upper limit for waiting - seem fairly
arbitrary. Many places use "HZ/50" or "HZ/10". As there is no obvious
choice, it seems reasonable to leave the maximum time unchanged.
If a service using svc_alloc_args() is terminated, it may now have to
wait up to the full 500ms before termination completes as
congestion_wait() cannot be interrupted. I don't believe this will be a
problem in practice, though it might be justification for using a
smaller timeout.
Signed-off-by: NeilBrown <neilb@suse.de>
---
I happened to notice this inconsistency between svc_alloc_args() and
xfs_buf_alloc_pages() despite them doing very similar things, so thought
I'd send a patch.
NeilBrown
net/sunrpc/svc_xprt.c | 9 ++++-----
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index 796eebf1787d..161433ae0fab 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -10,6 +10,7 @@
#include <linux/freezer.h>
#include <linux/kthread.h>
#include <linux/slab.h>
+#include <linux/backing-dev.h>
#include <net/sock.h>
#include <linux/sunrpc/addr.h>
#include <linux/sunrpc/stats.h>
@@ -682,12 +683,10 @@ static int svc_alloc_arg(struct svc_rqst *rqstp)
/* Made progress, don't sleep yet */
continue;
- set_current_state(TASK_INTERRUPTIBLE);
- if (signalled() || kthread_should_stop()) {
- set_current_state(TASK_RUNNING);
+ if (signalled() || kthread_should_stop())
return -EINTR;
- }
- schedule_timeout(msecs_to_jiffies(500));
+
+ congestion_wait(BLK_RW_ASYNC, msecs_to_jiffies(500));
}
rqstp->rq_page_end = &rqstp->rq_pages[pages];
rqstp->rq_pages[pages] = NULL; /* this might be seen in nfsd_splice_actor() */
--
2.33.0
next reply other threads:[~2021-09-06 4:44 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-06 4:44 NeilBrown [this message]
2021-09-06 15:46 ` [PATCH] SUNRPC: use congestion_wait() in svc_alloc_args() Chuck Lever III
2021-09-06 20:20 ` Matthew Wilcox
2021-09-06 22:13 ` Bruce Fields
2021-09-06 22:22 ` NeilBrown
2021-09-07 0:41 ` NeilBrown
2021-09-07 14:53 ` Chuck Lever III
2021-09-07 15:39 ` Bruce Fields
2021-09-07 15:41 ` Mel Gorman
2021-09-07 16:21 ` Chuck Lever III
2021-09-07 21:47 ` NeilBrown
2021-09-07 8:17 ` Mel Gorman
2021-09-06 21:52 ` NeilBrown
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=163090344807.19339.10071205771966144716@noble.neil.brown.name \
--to=neilb@suse.de \
--cc=bfields@fieldses.org \
--cc=chuck.lever@oracle.com \
--cc=linux-nfs@vger.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).