public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Nick Piggin <nickpiggin@yahoo.com.au>
To: Eric Dumazet <dada1@cosmosbay.com>
Cc: Andrew Morton <akpm@osdl.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Ingo Molnar <mingo@elte.hu>, Jens Axboe <jens.axboe@oracle.com>
Subject: Re: [PATCH] splice : two smp_mb() can be omitted
Date: Tue, 31 Oct 2006 20:40:21 +1100	[thread overview]
Message-ID: <45471A05.20205@yahoo.com.au> (raw)
In-Reply-To: <4546FA81.1020804@cosmosbay.com>

Eric Dumazet wrote:
> This patch deletes two calls to smp_mb() that were done after 
> mutex_unlock() that contains an implicit memory barrier.

Uh, there is nothing that says mutex_unlock or any unlock
functions contain an implicit smp_mb(). What is given is that the
lock and unlock obey aquire and release memory ordering,
respectively.

a = x;
xxx_unlock
b = y;

In this situation, the load of y can be executed before that of x.
And some architectures will even do so (i386 can, because the
unlock is an unprefixed store; ia64 can, because it uses a release
barrier in the unlock).

Whenever you rely on orderings of things *outside* locks (even
partially outside), you do need to be very careful about barriers
and can't rely on locks to do the right thing for you.

> 
> The first one in splice_to_pipe(), where 'do_wakeup' is set to true only 
> if pipe->inode is set (and in this case the
> if (pipe->inode)
>    mutex_unlock(&pipe->inode->i_mutex);
> is done too)
> 
> The second one in link_pipe(), following inode_double_unlock() that 
> contains calls to mutex_unlock() too.

It *may* be the case that these can be removed, but not by virtue
of the fact that the smp_mb is redundant.

> 
> Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
> 
> 
> ------------------------------------------------------------------------
> 
> --- linux/fs/splice.c	2006-10-31 07:49:52.000000000 +0100
> +++ linux-ed/fs/splice.c	2006-10-31 08:04:58.000000000 +0100
> @@ -248,7 +248,6 @@
>  		mutex_unlock(&pipe->inode->i_mutex);
>  
>  	if (do_wakeup) {
> -		smp_mb();
>  		if (waitqueue_active(&pipe->wait))
>  			wake_up_interruptible(&pipe->wait);
>  		kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
> @@ -1518,7 +1517,6 @@
>  	 * If we put data in the output pipe, wakeup any potential readers.
>  	 */
>  	if (ret > 0) {
> -		smp_mb();
>  		if (waitqueue_active(&opipe->wait))
>  			wake_up_interruptible(&opipe->wait);
>  		kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN);


-- 
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com 

  parent reply	other threads:[~2006-10-31  9:40 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-10-30  9:03 [PATCH 1/2] lockdep: spin_lock_irqsave_nested() Peter Zijlstra
2006-10-30  9:06 ` [PATCH 2/2] lockdep: annotate bcsp driver Peter Zijlstra
2006-10-30  9:06   ` Ingo Molnar
2006-10-30  9:30   ` Marcel Holtmann
2006-10-30  9:31   ` [PATCH 2/2] lockdep: annotate bcsp driver - v2 Peter Zijlstra
2006-10-30  9:07 ` [PATCH 1/2] lockdep: spin_lock_irqsave_nested() Ingo Molnar
2006-10-30 13:12 ` Jarek Poplawski
2006-10-30 13:27   ` Jarek Poplawski
2006-10-30 13:40   ` [PATCH 1/2] lockdep: spin_lock_irqsave_nested() -v2 Peter Zijlstra
2006-10-30 14:12     ` Jarek Poplawski
2006-10-31  6:48 ` [PATCH 1/2] lockdep: spin_lock_irqsave_nested() Andrew Morton
2006-10-31  7:25   ` [PATCH] splice : two smp_mb() can be omitted Eric Dumazet
2006-10-31  7:32     ` Jens Axboe
2006-10-31  7:41       ` Eric Dumazet
2006-10-31  7:46         ` Jens Axboe
2006-10-31  9:40     ` Nick Piggin [this message]
2006-10-31  9:49       ` Jens Axboe
2006-10-31 10:51       ` Eric Dumazet
2006-10-31 22:16         ` Nick Piggin
2006-10-31 23:08           ` Eric Dumazet
2006-10-31 23:45             ` Nick Piggin
2006-11-02 17:02         ` [PATCH] splice : Must fully check for fifos Eric Dumazet
2006-11-02 17:05           ` Eric Dumazet
2006-11-02 19:07             ` Jens Axboe
2006-11-03  8:50               ` Jens Axboe

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=45471A05.20205@yahoo.com.au \
    --to=nickpiggin@yahoo.com.au \
    --cc=akpm@osdl.org \
    --cc=dada1@cosmosbay.com \
    --cc=jens.axboe@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    /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