From: Trond Myklebust <trond.myklebust@primarydata.com>
To: Bruce Fields <bfields@fieldses.org>
Cc: linux-nfs@vger.kernel.org
Subject: [PATCH 1/3] SUNRPC: Reduce contention in svc_xprt_enqueue()
Date: Thu, 24 Jul 2014 23:59:31 -0400 [thread overview]
Message-ID: <1406260773-16231-1-git-send-email-trond.myklebust@primarydata.com> (raw)
Ensure that all calls to svc_xprt_enqueue() except svc_xprt_received()
check the value of XPT_BUSY, before attempting to grab spinlocks etc.
This is to avoid situations such as the following "perf" trace,
which shows heavy contention on the pool spinlock:
54.15% nfsd [kernel.kallsyms] [k] _raw_spin_lock_bh
|
--- _raw_spin_lock_bh
|
|--71.43%-- svc_xprt_enqueue
| |
| |--50.31%-- svc_reserve
| |
| |--31.35%-- svc_xprt_received
| |
| |--18.34%-- svc_tcp_data_ready
...
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
net/sunrpc/svc_xprt.c | 25 +++++++++++++++++--------
1 file changed, 17 insertions(+), 8 deletions(-)
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index b4737fbdec13..54a761fa6351 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -23,6 +23,7 @@ static int svc_deferred_recv(struct svc_rqst *rqstp);
static struct cache_deferred_req *svc_defer(struct cache_req *req);
static void svc_age_temp_xprts(unsigned long closure);
static void svc_delete_xprt(struct svc_xprt *xprt);
+static void svc_xprt_do_enqueue(struct svc_xprt *xprt);
/* apparently the "standard" is that clients close
* idle connections after 5 minutes, servers after
@@ -222,11 +223,12 @@ static void svc_xprt_received(struct svc_xprt *xprt)
if (!test_bit(XPT_BUSY, &xprt->xpt_flags))
return;
/* As soon as we clear busy, the xprt could be closed and
- * 'put', so we need a reference to call svc_xprt_enqueue with:
+ * 'put', so we need a reference to call svc_xprt_do_enqueue with:
*/
svc_xprt_get(xprt);
+ smp_mb__before_clear_bit();
clear_bit(XPT_BUSY, &xprt->xpt_flags);
- svc_xprt_enqueue(xprt);
+ svc_xprt_do_enqueue(xprt);
svc_xprt_put(xprt);
}
@@ -335,12 +337,7 @@ static bool svc_xprt_has_something_to_do(struct svc_xprt *xprt)
return false;
}
-/*
- * Queue up a transport with data pending. If there are idle nfsd
- * processes, wake 'em up.
- *
- */
-void svc_xprt_enqueue(struct svc_xprt *xprt)
+static void svc_xprt_do_enqueue(struct svc_xprt *xprt)
{
struct svc_pool *pool;
struct svc_rqst *rqstp;
@@ -398,6 +395,18 @@ void svc_xprt_enqueue(struct svc_xprt *xprt)
out_unlock:
spin_unlock_bh(&pool->sp_lock);
}
+
+/*
+ * Queue up a transport with data pending. If there are idle nfsd
+ * processes, wake 'em up.
+ *
+ */
+void svc_xprt_enqueue(struct svc_xprt *xprt)
+{
+ if (test_bit(XPT_BUSY, &xprt->xpt_flags))
+ return;
+ svc_xprt_do_enqueue(xprt);
+}
EXPORT_SYMBOL_GPL(svc_xprt_enqueue);
/*
--
1.9.3
next reply other threads:[~2014-07-25 3:59 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-25 3:59 Trond Myklebust [this message]
2014-07-25 3:59 ` [PATCH 2/3] SUNRPC: svc_tcp_write_space: don't clear SOCK_NOSPACE prematurely Trond Myklebust
2014-07-25 3:59 ` [PATCH 3/3] SUNRPC: Allow svc_reserve() to notify TCP socket that space has been freed Trond Myklebust
2014-07-29 19:31 ` [PATCH 1/3] SUNRPC: Reduce contention in svc_xprt_enqueue() Bruce Fields
2014-07-29 19:57 ` Peter Zijlstra
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=1406260773-16231-1-git-send-email-trond.myklebust@primarydata.com \
--to=trond.myklebust@primarydata.com \
--cc=bfields@fieldses.org \
--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).