Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755550Ab0DOR1M (ORCPT ); Thu, 15 Apr 2010 13:27:12 -0400 Received: from elvis.mu.org ([192.203.228.196]:58288 "EHLO elvis.mu.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755336Ab0DOR1K (ORCPT ); Thu, 15 Apr 2010 13:27:10 -0400 Cc: KOSAKI Motohiro , Mel Gorman , Chris Mason , linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, suleiman@google.com Message-Id: <85DB7083-8E78-4884-9E76-5BD803C530EF@freebsd.org> From: Suleiman Souhlal To: Dave Chinner In-Reply-To: <20100415093214.GV2493@dastard> Content-Type: text/plain; charset=US-ASCII; format=flowed Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v936) Subject: Re: [PATCH 1/4] vmscan: delegate pageout io to flusher thread if current is kswapd Date: Thu, 15 Apr 2010 10:27:09 -0700 References: <20100415013436.GO2493@dastard> <20100415130212.D16E.A69D9226@jp.fujitsu.com> <20100415131106.D174.A69D9226@jp.fujitsu.com> <64BE60A8-EEF9-4AC6-AF0A-0ED3CB544726@freebsd.org> <20100415093214.GV2493@dastard> X-Mailer: Apple Mail (2.936) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1521 Lines: 39 On Apr 15, 2010, at 2:32 AM, Dave Chinner wrote: > On Thu, Apr 15, 2010 at 01:05:57AM -0700, Suleiman Souhlal wrote: >> >> On Apr 14, 2010, at 9:11 PM, KOSAKI Motohiro wrote: >> >>> Now, vmscan pageout() is one of IO throuput degression source. >>> Some IO workload makes very much order-0 allocation and reclaim >>> and pageout's 4K IOs are making annoying lots seeks. >>> >>> At least, kswapd can avoid such pageout() because kswapd don't >>> need to consider OOM-Killer situation. that's no risk. >>> >>> Signed-off-by: KOSAKI Motohiro >> >> What's your opinion on trying to cluster the writes done by pageout, >> instead of not doing any paging out in kswapd? > > XFS already does this in ->writepage to try to minimise the impact > of the way pageout issues IO. It helps, but it is still not as good > as having all the writeback come from the flusher threads because > it's still pretty much random IO. Doesn't the randomness become irrelevant if you can cluster enough pages? > And, FWIW, it doesn't solve the stack usage problems, either. In > fact, it will make them worse as write_one_page() puts another > struct writeback_control on the stack... Sorry, this patch was not meant to solve the stack usage problems. -- Suleiman -- 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/