From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755200Ab2CEV4R (ORCPT ); Mon, 5 Mar 2012 16:56:17 -0500 Received: from mx1.redhat.com ([209.132.183.28]:51219 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751846Ab2CEV4Q (ORCPT ); Mon, 5 Mar 2012 16:56:16 -0500 Date: Mon, 5 Mar 2012 22:56:02 +0100 From: Johannes Weiner To: Satoru Moriya Cc: "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , Rik van Riel , "lwoodman@redhat.com" , "shaohua.li@intel.com" , KOSAKI Motohiro , "dle-develop@lists.sourceforge.net" , Seiji Aguchi Subject: Re: [RFC][PATCH] avoid swapping out with swappiness==0 Message-ID: <20120305215602.GA1693@redhat.com> References: <65795E11DBF1E645A09CEC7EAEE94B9CB9455FE2@USINDEVS02.corp.hds.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <65795E11DBF1E645A09CEC7EAEE94B9CB9455FE2@USINDEVS02.corp.hds.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Mar 02, 2012 at 12:36:40PM -0500, Satoru Moriya wrote: > Sometimes we'd like to avoid swapping out anonymous memory > in particular, avoid swapping out pages of important process or > process groups while there is a reasonable amount of pagecache > on RAM so that we can satisfy our customers' requirements. > > OTOH, we can control how aggressive the kernel will swap memory pages > with /proc/sys/vm/swappiness for global and > /sys/fs/cgroup/memory/memory.swappiness for each memcg. > > But with current reclaim implementation, the kernel may swap out > even if we set swappiness==0 and there is pagecache on RAM. > > This patch changes the behavior with swappiness==0. If we set > swappiness==0, the kernel does not swap out completely > (for global reclaim until the amount of free pages and filebacked > pages in a zone has been reduced to something very very small > (nr_free + nr_filebacked < high watermark)). > > Any comments are welcome. Last time I tried that (getting rid of sc->may_swap, using !swappiness), it was rejected it as there were users who relied on swapping very slowly with this setting. KOSAKI-san, do I remember correctly? Do you still think it's an issue? Personally, I still think it's illogical that !swappiness allows swapping and would love to see this patch go in. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx143.postini.com [74.125.245.143]) by kanga.kvack.org (Postfix) with SMTP id EAAD36B002C for ; Mon, 5 Mar 2012 16:56:08 -0500 (EST) Date: Mon, 5 Mar 2012 22:56:02 +0100 From: Johannes Weiner Subject: Re: [RFC][PATCH] avoid swapping out with swappiness==0 Message-ID: <20120305215602.GA1693@redhat.com> References: <65795E11DBF1E645A09CEC7EAEE94B9CB9455FE2@USINDEVS02.corp.hds.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <65795E11DBF1E645A09CEC7EAEE94B9CB9455FE2@USINDEVS02.corp.hds.com> Sender: owner-linux-mm@kvack.org List-ID: To: Satoru Moriya Cc: "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , Rik van Riel , "lwoodman@redhat.com" , "shaohua.li@intel.com" , KOSAKI Motohiro , "dle-develop@lists.sourceforge.net" , Seiji Aguchi On Fri, Mar 02, 2012 at 12:36:40PM -0500, Satoru Moriya wrote: > Sometimes we'd like to avoid swapping out anonymous memory > in particular, avoid swapping out pages of important process or > process groups while there is a reasonable amount of pagecache > on RAM so that we can satisfy our customers' requirements. > > OTOH, we can control how aggressive the kernel will swap memory pages > with /proc/sys/vm/swappiness for global and > /sys/fs/cgroup/memory/memory.swappiness for each memcg. > > But with current reclaim implementation, the kernel may swap out > even if we set swappiness==0 and there is pagecache on RAM. > > This patch changes the behavior with swappiness==0. If we set > swappiness==0, the kernel does not swap out completely > (for global reclaim until the amount of free pages and filebacked > pages in a zone has been reduced to something very very small > (nr_free + nr_filebacked < high watermark)). > > Any comments are welcome. Last time I tried that (getting rid of sc->may_swap, using !swappiness), it was rejected it as there were users who relied on swapping very slowly with this setting. KOSAKI-san, do I remember correctly? Do you still think it's an issue? Personally, I still think it's illogical that !swappiness allows swapping and would love to see this patch go in. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: email@kvack.org