From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: [PATCH 06/45] writeback: use larger ratelimit when dirty_exceeded Date: Wed, 07 Oct 2009 10:53:19 +0200 Message-ID: <1254905599.26976.217.camel@twins> References: <20091007073818.318088777@intel.com> <20091007074901.884235522@intel.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Andrew Morton , Theodore Tso , Christoph Hellwig , Dave Chinner , Chris Mason , Li Shaohua , Myklebust Trond , "jens.axboe@oracle.com" , Jan Kara , Nick Piggin , linux-fsdevel@vger.kernel.org, Richard Kennedy To: Wu Fengguang Return-path: Received: from viefep20-int.chello.at ([62.179.121.40]:52570 "EHLO viefep20-int.chello.at" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934193AbZJGI5E (ORCPT ); Wed, 7 Oct 2009 04:57:04 -0400 In-Reply-To: <20091007074901.884235522@intel.com> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Wed, 2009-10-07 at 15:38 +0800, Wu Fengguang wrote: > plain text document attachment > (writeback-ratelimit-on-dirty-exceeded.patch) > When dirty_exceeded, use ratelimit = ratelimit_pages/8, allowing it to > scale up to 512KB for memory bounty systems. This is more efficient than > the original 8 pages, and won't risk exceeding the dirty limit too much. > > Given the larger ratelimit value, we can safely ignore the low bound > check in sync_writeback_pages. > > dirty_exceeded is more likely to be seen when there are multiple dirty > processes. In which case the lowered ratelimit will help reduce their > overall wait time (latency) in the throttled queue. Don't forget that ratelimit_pages is a per-cpu limit. So the total error on the dirty limit scales with the number of cpus. Other than that, I guess this patch needs numbers ;-)