From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932308Ab2IQPyM (ORCPT ); Mon, 17 Sep 2012 11:54:12 -0400 Received: from cantor2.suse.de ([195.135.220.15]:39377 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932093Ab2IQPyK (ORCPT ); Mon, 17 Sep 2012 11:54:10 -0400 Date: Mon, 17 Sep 2012 17:54:06 +0200 From: Jan Kara To: OGAWA Hirofumi Cc: Jan Kara , Fengguang Wu , viro@zeniv.linux.org.uk, hch@lst.de, linux-kernel@vger.kernel.org Subject: Re: [PATCH] Fix queueing work if !bdi_cap_writeback_dirty() Message-ID: <20120917155406.GC9150@quack.suse.cz> References: <20120914131952.GA4952@quack.suse.cz> <87ipbgn2gz.fsf@devron.myhome.or.jp> <20120914144543.GB4952@quack.suse.cz> <878vccmygy.fsf@devron.myhome.or.jp> <20120916214912.GA7503@quack.suse.cz> <87wqzt7drb.fsf@devron.myhome.or.jp> <20120917084853.GA9150@quack.suse.cz> <87627d6lae.fsf@devron.myhome.or.jp> <20120917095623.GB9150@quack.suse.cz> <87obl55405.fsf@devron.myhome.or.jp> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <87obl55405.fsf@devron.myhome.or.jp> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon 17-09-12 19:37:46, OGAWA Hirofumi wrote: > Jan Kara writes: > > > On Mon 17-09-12 18:39:05, OGAWA Hirofumi wrote: > >> Jan Kara writes: > >> > >> >> I think you know how to solve it though. You can add the periodic flush > >> >> in own task. And you can check bdi->dirty_exceeded in any handlers. > >> > Sure, you can have your private thread. That is possible but you will > >> > have to duplicate flusher logic and you will still get odd behavior e.g. > >> > when your filesystem is on one partition and another filesystem is on a > >> > different partition of the same disk. > >> > >> Right. But it is what current FSes are doing more or less. > > It's not. Page writeback is respected by all filesystems in most cases > > AFAIK. Inode writeback is a different issue but that's not so interesting > > from mm point of view... > > Duplicate flusher - many FSes has own task to flush. Odd behavior in > the case of partition - agree, but I'm not sure why metadata is ok, and > it is not odd behavior. Well, because there is much more of data pages then there is metadata. So when you do strange things (like refuse to write / reclaim) with metadata, it usually ends up in the noise. But when you start doing similar things with data pages, people will notice. > Sorry, I'm not sure your point in latest comment. You are just saying FS > must flush pages on writepages()? Yes. > And if alternative plan is acceptable, maybe I will not have interest to > this anymore. Yes, the alternative plan looks better to me. But all in all I don't want to stop you from your experiments :) I mostly just wanted to point out that disabling flusher thread for a filesystem has a complex consequences which IMHO bring more bad than good. Honza -- Jan Kara SUSE Labs, CR