From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756308AbZETMGT (ORCPT ); Wed, 20 May 2009 08:06:19 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754979AbZETMGJ (ORCPT ); Wed, 20 May 2009 08:06:09 -0400 Received: from cantor2.suse.de ([195.135.220.15]:58671 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753735AbZETMGI (ORCPT ); Wed, 20 May 2009 08:06:08 -0400 Date: Wed, 20 May 2009 14:06:08 +0200 From: Jan Kara To: Jens Axboe Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, chris.mason@oracle.com, david@fromorbit.com, hch@infradead.org, akpm@linux-foundation.org, yanmin_zhang@linux.intel.com Subject: Re: [PATCH 04/11] writeback: separate the flushing state/task from the bdi Message-ID: <20090520120608.GE3760@duck.suse.cz> References: <1242649192-16263-1-git-send-email-jens.axboe@oracle.com> <1242649192-16263-5-git-send-email-jens.axboe@oracle.com> <20090520113430.GD3760@duck.suse.cz> <20090520113918.GU11363@kernel.dk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090520113918.GU11363@kernel.dk> User-Agent: Mutt/1.5.17 (2007-11-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed 20-05-09 13:39:18, Jens Axboe wrote: > On Wed, May 20 2009, Jan Kara wrote: > > On Mon 18-05-09 14:19:45, Jens Axboe wrote: > > > Add a struct bdi_writeback for tracking and handling dirty IO. This > > > is in preparation for adding > 1 flusher task per bdi. > > Some changes (IMO the most complicated ones ;) in this patch set seem to > > be just reordering / cleanup of changes which happened in patch #2. Could > > you maybe move it there. Commented below... > > Some of it, most of it is due to switching from one fixed thread to the > potential of having lots more. The moving code around is mostly due to > other callers now having to use functions that were below them, and I'd > rather move them around instead of having prototypes at the top. I meant mainly the changes in forker thread and such. > It would be easy to unify the two patches, but I wanted to separate the > switch from pdflush to 1 bdi thread from the transition from 1 bdi > thread to several. Yes, this is probably desirable. Honza -- Jan Kara SUSE Labs, CR