From: David Miller <davem@davemloft.net>
To: miklos@szeredi.hu
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] fix race in AF_UNIX
Date: Wed, 06 Jun 2007 01:12:58 -0700 (PDT) [thread overview]
Message-ID: <20070606.011258.71090697.davem@davemloft.net> (raw)
In-Reply-To: <E1HvqZB-0003Vc-00@dorka.pomaz.szeredi.hu>
From: Miklos Szeredi <miklos@szeredi.hu>
Date: Wed, 06 Jun 2007 10:08:29 +0200
> > > > Holding a global mutex over recvmsg() calls under AF_UNIX is pretty
> > > > much a non-starter, this will kill performance for multi-threaded
> > > > apps.
> > >
> > > That's an rwsem held for read. It's held for write in unix_gc() only
> > > for a short duration, and unix_gc() should only rarely be called. So
> > > I don't think there's any performance problem here.
> >
> > It pulls a non-local cacheline into the local thread, that's extremely
> > expensive on SMP.
>
> OK, here's an updated patch, that uses ->readlock, and passes my
> testing.
Thanks a lot, I'll review this as soon as possible unless
someone else beats me to it :)
next prev parent reply other threads:[~2007-06-06 8:12 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-06-02 21:50 [PATCH] fix race in AF_UNIX Miklos Szeredi
2007-06-02 22:11 ` Arnaldo Carvalho de Melo
2007-06-04 9:45 ` Miklos Szeredi
2007-06-05 7:02 ` David Miller
2007-06-05 7:42 ` Miklos Szeredi
2007-06-05 7:55 ` David Miller
2007-06-05 8:11 ` Miklos Szeredi
2007-06-05 8:19 ` David Miller
2007-06-05 20:11 ` David Miller
2007-06-06 0:31 ` David Miller
2007-06-06 5:26 ` Miklos Szeredi
2007-06-06 5:41 ` David Miller
2007-06-06 8:08 ` Miklos Szeredi
2007-06-06 8:12 ` David Miller [this message]
2007-06-08 1:47 ` David Miller
2007-06-11 9:57 ` Miklos Szeredi
2007-06-18 7:49 ` Miklos Szeredi
2007-06-18 7:57 ` David Miller
2007-06-18 8:20 ` Miklos Szeredi
2007-06-18 9:18 ` David Miller
2007-06-18 9:29 ` Miklos Szeredi
2007-06-18 9:35 ` David Miller
2007-06-18 9:44 ` Miklos Szeredi
2007-06-18 9:48 ` David Miller
2007-06-18 9:55 ` Miklos Szeredi
2007-06-18 9:59 ` David Miller
2007-06-18 10:32 ` Thomas Graf
2007-06-18 10:39 ` Miklos Szeredi
2007-06-18 10:43 ` Thomas Graf
2007-06-18 12:01 ` Alan Cox
2007-06-18 10:40 ` Thomas Graf
2007-06-18 10:47 ` Miklos Szeredi
2007-06-18 10:51 ` David Miller
2007-06-18 10:55 ` Miklos Szeredi
2007-06-18 11:02 ` David Miller
2007-06-18 11:06 ` Miklos Szeredi
2007-06-18 11:09 ` David Miller
2007-06-18 11:46 ` Miklos Szeredi
2007-06-18 11:47 ` Alan Cox
2007-06-18 11:45 ` Jan Engelhardt
2007-06-18 12:00 ` Miklos Szeredi
2007-06-21 15:18 ` Eric W. Biederman
2007-06-23 8:48 ` Miklos Szeredi
2007-06-23 16:42 ` Eric W. Biederman
2007-06-26 8:54 ` Miklos Szeredi
2007-06-26 15:24 ` Eric W. Biederman
2007-06-04 9:53 ` Miklos Szeredi
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=20070606.011258.71090697.davem@davemloft.net \
--to=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=miklos@szeredi.hu \
--cc=netdev@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).