From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id n6OC4qM5126717 for ; Fri, 24 Jul 2009 07:04:53 -0500 Received: from acsinet11.oracle.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 261041B59CD7 for ; Fri, 24 Jul 2009 05:05:36 -0700 (PDT) Received: from acsinet11.oracle.com (acsinet11.oracle.com [141.146.126.233]) by cuda.sgi.com with ESMTP id rfhSgim7X7rUi3ZI for ; Fri, 24 Jul 2009 05:05:36 -0700 (PDT) Date: Fri, 24 Jul 2009 08:05:19 -0400 From: Chris Mason Subject: Re: [PATCH] bump up nr_to_write in xfs_vm_writepage Message-ID: <20090724120519.GB16192@think> References: <20090709110342.2386.A69D9226@jp.fujitsu.com> <20090709130134.GH18008@think> <20090710153349.17EC.A69D9226@jp.fujitsu.com> <7149D747-2769-4559-BAF6-AAD2B6C6C941@sgi.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <7149D747-2769-4559-BAF6-AAD2B6C6C941@sgi.com> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: Felix Blyakher Cc: Eric Sandeen , KOSAKI Motohiro , xfs mailing list , Christoph Hellwig , linux-mm@kvack.org, Olaf Weber On Fri, Jul 24, 2009 at 12:20:32AM -0500, Felix Blyakher wrote: > > On Jul 10, 2009, at 2:12 AM, KOSAKI Motohiro wrote: >> 3. Current wbc->nr_to_write value is not proper? >> >> Current writeback_set_ratelimit() doesn't permit that ratelimit_pages >> exceed >> 4M byte. but it is too low restriction for nowadays. >> (that's my understand. right?) >> >> ======================================================= >> void writeback_set_ratelimit(void) >> { >> ratelimit_pages = vm_total_pages / (num_online_cpus() * 32); >> if (ratelimit_pages < 16) >> ratelimit_pages = 16; >> if (ratelimit_pages * PAGE_CACHE_SIZE > 4096 * 1024) >> ratelimit_pages = (4096 * 1024) / PAGE_CACHE_SIZE; >> } >> ======================================================= >> >> Yes, 4M bytes are pretty magical constant. We have three choice >> A. Remove magical 4M constant simple (a bit danger) > > That's will be outside the xfs, and seems like there is no much interest > from mm people. > >> B. Decide high border from IO capability It is worth pointing out that Jens Axboe is planning on more feedback controlled knobs as part of pdflush rework. -chris _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs