Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753067AbZIYBeF (ORCPT ); Thu, 24 Sep 2009 21:34:05 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751613AbZIYBeD (ORCPT ); Thu, 24 Sep 2009 21:34:03 -0400 Received: from mga03.intel.com ([143.182.124.21]:44436 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751698AbZIYBeC (ORCPT ); Thu, 24 Sep 2009 21:34:02 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.44,449,1249282800"; d="scan'208";a="191464619" Date: Fri, 25 Sep 2009 09:33:50 +0800 From: Wu Fengguang To: Peter Zijlstra Cc: Jan Kara , Chris Mason , Artem Bityutskiy , Jens Axboe , "linux-kernel@vger.kernel.org" , "linux-fsdevel@vger.kernel.org" , "david@fromorbit.com" , "hch@infradead.org" , "akpm@linux-foundation.org" , "Theodore Ts'o" Subject: Re: [PATCH 8/8] vm: Add an tuning knob for vm.max_writeback_mb Message-ID: <20090925013350.GD6190@localhost> References: <1252428983.7746.140.camel@twins> <20090908172842.GC2975@think> <1252431974.7746.151.camel@twins> <1252432501.7746.156.camel@twins> <1252434746.7035.7.camel@laptop> <20090909142315.GA7949@duck.suse.cz> <1252597750.7205.82.camel@laptop> <20090914111721.GA24075@duck.suse.cz> <20090924083342.GA15918@localhost> <1253806696.18939.40.camel@laptop> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1253806696.18939.40.camel@laptop> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2093 Lines: 54 On Thu, Sep 24, 2009 at 11:38:16PM +0800, Peter Zijlstra wrote: > On Thu, 2009-09-24 at 16:33 +0800, Wu Fengguang wrote: > > > Yeah, FIFO queuing should be good enough. > > > > I'd like to propose one more data structure for evaluation :) > > > > - bdi->throttle_lock > > - bdi->throttle_list pages to sync for each waiting task, taken from sync_writeback_pages() > > - bdi->throttle_pages (counted down) pages to sync for the head task, shall be atomic_t > > > > In balance_dirty_pages(), it would do > > > > nr_to_sync = sync_writeback_pages() > > if (list_empty(bdi->throttle_list)) # I'm the only task > > bdi->throttle_pages = nr_to_sync > > append nr_to_sync to bdi->throttle_list > > kick off background writeback > > wait > > remove itself from bdi->throttle_list and wait list > > set bdi->throttle_pages for new head task (or LONG_MAX) > > > > In __bdi_writeout_inc(), it would do > > > > if (--bdi->throttle_pages <= 0) > > check and wake up head task > > > > In wb_writeback(), it would do > > > > if (args->for_background && exiting) > > wake up all throttled tasks > > To prevent wake up too many tasks at the same time, it can relax the > > background threshold a bit, so that __bdi_writeout_inc() become the > > only wake up point in normal cases. > > > > if (args->for_background && !list_empty(bdi->throttle_list) && > > over background_thresh - background_thresh / 32) > > keep write pages; I realized this last change is not necessary, because we already have a big enough buffer area: (dirty_thresh + background_thresh)/2 ==> background_thresh > Right, something like that ought to work well, or at least sounds like > worth a try ;-) Thanks :) -- 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/