From: Tim Chen <tim.c.chen@linux.intel.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: "Yan, Zheng" <zheng.z.yan@intel.com>,
"Yan, Zheng" <yanzheng@21cn.com>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"davem@davemloft.net" <davem@davemloft.net>,
"sfr@canb.auug.org.au" <sfr@canb.auug.org.au>,
"jirislaby@gmail.com" <jirislaby@gmail.com>,
"sedat.dilek@gmail.com" <sedat.dilek@gmail.com>,
"Shi, Alex" <alex.shi@intel.com>,
Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Subject: Re: [PATCH net-next] af_unix: dont send SCM_CREDENTIALS by default
Date: Mon, 12 Sep 2011 12:15:46 -0700 [thread overview]
Message-ID: <1315854946.2576.3122.camel@schen9-DESK> (raw)
In-Reply-To: <1315544777.5410.19.camel@edumazet-laptop>
On Fri, 2011-09-09 at 07:06 +0200, Eric Dumazet wrote:
> Le jeudi 08 septembre 2011 à 02:24 -0700, Tim Chen a écrit :
>
> > Looking forward to the patch. This should improve the scalability of
> > af_unix.
>
> Here it is, based on top on previous one [af_unix: Fix use-after-free
> crashes]
>
> Thanks
>
> [PATCH net-next] af_unix: dont send SCM_CREDENTIALS by default
>
> Since commit 7361c36c5224 (af_unix: Allow credentials to work across
> user and pid namespaces) af_unix performance dropped a lot.
>
> This is because we now take a reference on pid and cred in each write(),
> and release them in read(), usually done from another process,
> eventually from another cpu. This triggers false sharing.
>
> # Events: 154K cycles
> #
> # Overhead Command Shared Object Symbol
> # ........ ....... .................. ...................................
> #
> 10.40% hackbench [kernel.kallsyms] [k] put_pid
> 8.60% hackbench [kernel.kallsyms] [k] unix_stream_recvmsg
> 7.87% hackbench [kernel.kallsyms] [k] unix_stream_sendmsg
> 6.11% hackbench [kernel.kallsyms] [k] do_raw_spin_lock
> 4.95% hackbench [kernel.kallsyms] [k] unix_scm_to_skb
> 4.87% hackbench [kernel.kallsyms] [k] pid_nr_ns
> 4.34% hackbench [kernel.kallsyms] [k] cred_to_ucred
> 2.39% hackbench [kernel.kallsyms] [k] unix_destruct_scm
> 2.24% hackbench [kernel.kallsyms] [k] sub_preempt_count
> 1.75% hackbench [kernel.kallsyms] [k] fget_light
> 1.51% hackbench [kernel.kallsyms] [k] __mutex_lock_interruptible_slowpath
> 1.42% hackbench [kernel.kallsyms] [k] sock_alloc_send_pskb
>
>
> This patch includes SCM_CREDENTIALS information in a af_unix message/skb
> only if requested by the sender, [man 7 unix for details how to include
> ancillary data using sendmsg() system call]
>
> Note: This might break buggy applications that expected SCM_CREDENTIAL
> from an unaware write() system call.
>
> Performance boost in hackbench : more than 50% gain on a 16 thread
> machine (2 quad-core cpus, 2 threads per core)
>
> hackbench 20 thread 2000
>
> 4.224 sec instead of 9.102 sec
>
>
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
This speeds up the case where we're not passing SCM_CREDENTIALS nicely.
Acked-by: Tim Chen <tim.c.chen@linux.intel.com>
next prev parent reply other threads:[~2011-09-12 19:12 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-09-04 5:44 [PATCH -next v2] unix stream: Fix use-after-free crashes Yan, Zheng
2011-09-04 7:12 ` Sedat Dilek
2011-09-04 8:23 ` Yan, Zheng
2011-09-04 15:50 ` Joe Perches
2011-09-06 16:39 ` Tim Chen
2011-09-06 16:25 ` Tim Chen
2011-09-06 17:40 ` Eric Dumazet
2011-09-06 18:50 ` Tim Chen
2011-09-06 19:01 ` Eric Dumazet
2011-09-06 19:33 ` Tim Chen
2011-09-06 19:43 ` Eric Dumazet
2011-09-06 19:59 ` Tim Chen
2011-09-06 20:19 ` Eric Dumazet
2011-09-06 22:08 ` Tim Chen
2011-09-07 2:35 ` Eric Dumazet
2011-09-06 23:09 ` Yan, Zheng
2011-09-07 2:55 ` Eric Dumazet
2011-09-16 23:35 ` David Miller
2011-09-16 16:50 ` Tim Chen
2011-09-19 7:57 ` Eric Dumazet
2011-09-07 4:36 ` Yan, Zheng
2011-09-07 5:08 ` Eric Dumazet
2011-09-07 5:20 ` Yan, Zheng
[not found] ` <1315381503.3400.85.camel@edumazet-laptop>
2011-09-07 12:01 ` Tim Chen
2011-09-07 20:12 ` Sedat Dilek
2011-09-07 20:30 ` Sedat Dilek
2011-09-07 14:37 ` Tim Chen
2011-09-08 0:27 ` Yan, Zheng
2011-09-07 21:06 ` Tim Chen
2011-09-07 21:15 ` Tim Chen
2011-09-08 6:21 ` Eric Dumazet
2011-09-08 4:18 ` Yan, Zheng
2011-09-08 5:59 ` Eric Dumazet
2011-09-08 6:22 ` Yan, Zheng
2011-09-08 7:11 ` Eric Dumazet
2011-09-08 7:23 ` Yan, Zheng
2011-09-08 7:33 ` Eric Dumazet
2011-09-08 9:59 ` Sedat Dilek
2011-09-08 13:21 ` [PATCH net-next v3] af_unix: " Eric Dumazet
2011-09-08 8:37 ` Tim Chen
2011-09-09 6:51 ` Eric Dumazet
2011-09-09 7:58 ` [PATCH net-next] af_unix: fix use after free in unix_stream_recvmsg() Eric Dumazet
2011-09-09 10:39 ` Tim Chen
2011-09-09 10:41 ` [PATCH net-next v3] af_unix: Fix use-after-free crashes Tim Chen
2011-09-08 7:56 ` [PATCH -next v2] unix stream: " Jiri Slaby
2011-09-08 8:43 ` Sedat Dilek
2011-09-08 7:02 ` Sedat Dilek
2011-09-07 21:26 ` Eric Dumazet
2011-09-08 13:28 ` Eric Dumazet
2011-09-08 9:24 ` Tim Chen
2011-09-09 5:06 ` [PATCH net-next] af_unix: dont send SCM_CREDENTIALS by default Eric Dumazet
2011-09-12 19:15 ` Tim Chen [this message]
2011-09-19 1:07 ` David Miller
2011-09-19 4:28 ` Eric Dumazet
2011-09-19 15:02 ` Eric Dumazet
2011-09-19 15:52 ` [PATCH v2 " Eric Dumazet
2011-09-19 21:39 ` Tim Chen
2011-09-20 2:10 ` Valdis.Kletnieks
2011-09-20 4:16 ` Eric Dumazet
2011-09-22 16:15 ` tim
2011-11-28 13:23 ` Michal Schmidt
2011-11-28 13:38 ` Eric Dumazet
2011-09-28 17:30 ` David Miller
2011-09-08 10:05 ` [PATCH -next v2] unix stream: Fix use-after-free crashes Sedat Dilek
2011-09-08 8:50 ` Tim Chen
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=1315854946.2576.3122.camel@schen9-DESK \
--to=tim.c.chen@linux.intel.com \
--cc=Valdis.Kletnieks@vt.edu \
--cc=alex.shi@intel.com \
--cc=davem@davemloft.net \
--cc=eric.dumazet@gmail.com \
--cc=jirislaby@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=sedat.dilek@gmail.com \
--cc=sfr@canb.auug.org.au \
--cc=yanzheng@21cn.com \
--cc=zheng.z.yan@intel.com \
/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).