Return-Path: <linux-kernel-owner+willy=40w.ods.org@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id <S313070AbSGKVMv>; Thu, 11 Jul 2002 17:12:51 -0400
Received: (majordomo@vger.kernel.org) by vger.kernel.org
	id <S317909AbSGKVMu>; Thu, 11 Jul 2002 17:12:50 -0400
Received: from perninha.conectiva.com.br ([200.250.58.156]:9491 "HELO
	perninha.conectiva.com.br") by vger.kernel.org with SMTP
	id <S313070AbSGKVMu>; Thu, 11 Jul 2002 17:12:50 -0400
Date: Thu, 11 Jul 2002 17:21:24 -0300 (BRT)
From: Marcelo Tosatti <marcelo@conectiva.com.br>
X-X-Sender: marcelo@freak.distro.conectiva
To: Andrea Arcangeli <andrea@suse.de>
Cc: Andrew Morton <akpm@zip.com.au>, <linux-kernel@vger.kernel.org>,
       "Carter K. George" <carter@polyserve.com>,
       Don Norton <djn@polyserve.com>, "James S. Tybur" <jtybur@polyserve.com>
Subject: Re: fsync fixes for 2.4
In-Reply-To: <20020710202036.GA1342@dualathlon.random>
Message-ID: <Pine.LNX.4.44.0207111710470.21365-100000@freak.distro.conectiva>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 1190
Lines: 40



On Wed, 10 Jul 2002, Andrea Arcangeli wrote:

> At polyserve they found a severe problem with fsync in 2.4.
>
> In short the write_buffer (ll_rw_block of mainline) is a noop if old I/O
> is in flight. You know the buffer can be made dirty while I/O is in
> flight, and in such case fsync would return without flushing the dirty
> buffers at all. Their proposed fix is strightforward, just a
> wait_on_buffer before the ll_rw_block will guarantee somebody marked the
> bh locked _after_ we wrote to it.

>From what I can see the problem goes like:


thread1				thread2
				writepage(page) (marks the buffers clean, page is
				locked for IO)

mark_buffer_dirty()

fsync()

fsync_buffers_list() finds
the dirtied buffer, but since
its locked ll_rw_block() returns
without queueing the data.

fsync_buffers_list() waits on the writepage()'s
write to return but not on latest data write.


Is that what you mean or I'm misunderstanding something?

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/