From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wu Fengguang Subject: Re: [RFC][PATCH 2/7] writeback: fix queue_io() ordering Date: Thu, 10 Sep 2009 09:26:24 +0800 Message-ID: <20090910012624.GA10957@localhost> References: <20090909145141.293229693@intel.com> <20090909150600.451373732@intel.com> <20090909155330.GE7949@duck.suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Andrew Morton , Jens Axboe , Dave Chinner , Chris Mason , Peter Zijlstra , Christoph Hellwig , "jack@suse.cz" , Artem Bityutskiy , LKML , "linux-fsdevel@vger.kernel.org" To: Jan Kara Return-path: Received: from mga14.intel.com ([143.182.124.37]:29069 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753883AbZIJB0c (ORCPT ); Wed, 9 Sep 2009 21:26:32 -0400 Content-Disposition: inline In-Reply-To: <20090909155330.GE7949@duck.suse.cz> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Wed, Sep 09, 2009 at 11:53:30PM +0800, Jan Kara wrote: > On Wed 09-09-09 22:51:43, Wu Fengguang wrote: > > This was not a bug, since b_io is empty for kupdate writeback. > > The next patch will do requeue_io() for non-kupdate writeback, > > so let's fix it. > But doesn't this patch also need your "anti-starvation" patch? Honza, can you show me that patch? > Looking into the code, we put inode to b_more_io when nr_to_write > drops to zero and this way we'd just start writing it again > in the next round... I'm confused. It's OK to start it in next round. Starvation can occur if we start it immediately in the next writeback_inodes() invocation. How can that happen with this patch? Thanks, Fengguang > Honza > > > > CC: Dave Chinner > > Cc: Martin Bligh > > Cc: Michael Rubin > > Cc: Peter Zijlstra > > Signed-off-by: Fengguang Wu > > --- > > fs/fs-writeback.c | 7 +++++-- > > 1 file changed, 5 insertions(+), 2 deletions(-) > > > > --- linux.orig/fs/fs-writeback.c 2009-09-09 21:41:14.000000000 +0800 > > +++ linux/fs/fs-writeback.c 2009-09-09 21:45:15.000000000 +0800 > > @@ -313,11 +313,14 @@ static void move_expired_inodes(struct l > > } > > > > /* > > - * Queue all expired dirty inodes for io, eldest first. > > + * Queue all expired dirty inodes for io, eldest first: > > + * (newly dirtied) => b_dirty inodes > > + * => b_more_io inodes > > + * => remaining inodes in b_io => (dequeue for sync) > > */ > > static void queue_io(struct bdi_writeback *wb, unsigned long *older_than_this) > > { > > - list_splice_init(&wb->b_more_io, wb->b_io.prev); > > + list_splice_init(&wb->b_more_io, &wb->b_io); > > move_expired_inodes(&wb->b_dirty, &wb->b_io, older_than_this); > > } > > > > > > -- > > > -- > Jan Kara > SUSE Labs, CR