From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx111.postini.com [74.125.245.111]) by kanga.kvack.org (Postfix) with SMTP id 054B28D0001 for ; Fri, 11 May 2012 17:11:43 -0400 (EDT) From: Satoru Moriya Date: Fri, 11 May 2012 17:11:35 -0400 Subject: RE: [RFC][PATCH] avoid swapping out with swappiness==0 Message-ID: <65795E11DBF1E645A09CEC7EAEE94B9C01583B4D7C@USINDEVS02.corp.hds.com> References: <65795E11DBF1E645A09CEC7EAEE94B9CB9455FE2@USINDEVS02.corp.hds.com> <20120305215602.GA1693@redhat.com> <4F5798B1.5070005@jp.fujitsu.com> <65795E11DBF1E645A09CEC7EAEE94B9CB951A45F@USINDEVS02.corp.hds.com> <65795E11DBF1E645A09CEC7EAEE94B9C01454D13A6@USINDEVS02.corp.hds.com> <4F7ADE1A.2050004@redhat.com> <4F7C870B.6020807@gmail.com> <65795E11DBF1E645A09CEC7EAEE94B9C014575D8CF@USINDEVS02.corp.hds.com> In-Reply-To: <65795E11DBF1E645A09CEC7EAEE94B9C014575D8CF@USINDEVS02.corp.hds.com> Content-Language: en-US Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Sender: owner-linux-mm@kvack.org List-ID: To: Satoru Moriya , KOSAKI Motohiro , Jerome Marchand Cc: KOSAKI Motohiro , "jweiner@redhat.com" , "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , "riel@redhat.com" , "lwoodman@redhat.com" , "dle-develop@lists.sourceforge.net" , Seiji Aguchi On 04/20/2012 08:21 PM, Satoru Moriya wrote: > On 04/04/2012 01:38 PM, KOSAKI Motohiro wrote: >> (4/3/12 4:25 AM), Jerome Marchand wrote: >>> On 04/02/2012 07:10 PM, KOSAKI Motohiro wrote: >>>> 2012/3/30 Satoru Moriya: >>>>> So the kernel reclaims pages like following. >>>>> >>>>> nr_free + nr_filebacked>=3D watermark_high: reclaim only filebacked p= ages >>>>> nr_free + nr_filebacked< watermark_high: reclaim only anonymous pag= es >>>> >>>> How? >>> >>> get_scan_count() checks that case explicitly: >>> >>> if (global_reclaim(sc)) { >>> free =3D zone_page_state(mz->zone, NR_FREE_PAGES); >>> /* If we have very few page cache pages, >>> force-scan anon pages. */ >>> if (unlikely(file + free<=3D high_wmark_pages(mz->zone))) { >>> fraction[0] =3D 1; >>> fraction[1] =3D 0; >>> denominator =3D 1; >>> goto out; >>> } >>> } >> >> Eek. This is silly. Nowaday many people enabled THP and it increase zone= watermark. >> so, high watermask is not good threshold anymore. >=20 > Ah yes, it is not so small now. > On 4GB server, without THP min_free_kbytes is 8113 but with THP it is=20 > 67584. >=20 > How about using low watermark or min watermark? > Are they still big? >=20 > ...or should we use other value?=20 What do you think of the idea above? By the way, I'd like to discuss this topic in other thread because discussion about optimal threshold where the kernel changes its reclaim policy does not affect only swappiness=3D=3D0 case but also all other settings.=20 So, I propose that we start with applying this patch first and then discuss/improve the threshold. The patch may not be perfect but, at least, we can improve the kernel behavior in the enough filebacked memory case with this patch. I believe it's better than nothing. Regards, Satoru -- 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