[{"id":1778950,"web_url":"http://patchwork.ozlabs.org/comment/1778950/","msgid":"<20171003130857.vohli6lnqj4tdmhl@dhcp22.suse.cz>","list_archive_url":null,"date":"2017-10-03T13:08:57","subject":"Re: [PATCH v9 06/12] mm: zero struct pages during initialization","submitter":{"id":66979,"url":"http://patchwork.ozlabs.org/api/people/66979/","name":"Michal Hocko","email":"mhocko@kernel.org"},"content":"On Wed 20-09-17 16:17:08, Pavel Tatashin wrote:\n> Add struct page zeroing as a part of initialization of other fields in\n> __init_single_page().\n> \n> This single thread performance collected on: Intel(R) Xeon(R) CPU E7-8895\n> v3 @ 2.60GHz with 1T of memory (268400646 pages in 8 nodes):\n> \n> BASE FIX\n> sparse_init 11.244671836s 0.007199623s\n> zone_sizes_init 4.879775891s 8.355182299s\n> --------------------------\n> Total 16.124447727s 8.362381922s\n\nHmm, this is confusing. This assumes that sparse_init doesn't zero pages\nanymore, right? So these number depend on the last patch in the series?\n\n> sparse_init is where memory for struct pages is zeroed, and the zeroing\n> part is moved later in this patch into __init_single_page(), which is\n> called from zone_sizes_init().\n> \n> Signed-off-by: Pavel Tatashin \n> Reviewed-by: Steven Sistare \n> Reviewed-by: Daniel Jordan \n> Reviewed-by: Bob Picco \n> Acked-by: Michal Hocko \n> ---\n> include/linux/mm.h | 9 +++++++++\n> mm/page_alloc.c | 1 +\n> 2 files changed, 10 insertions(+)\n> \n> diff --git a/include/linux/mm.h b/include/linux/mm.h\n> index f8c10d336e42..50b74d628243 100644\n> --- a/include/linux/mm.h\n> +++ b/include/linux/mm.h\n> @@ -94,6 +94,15 @@ extern int mmap_rnd_compat_bits __read_mostly;\n> #define mm_forbids_zeropage(X)\t(0)\n> #endif\n> \n> +/*\n> + * On some architectures it is expensive to call memset() for small sizes.\n> + * Those architectures should provide their own implementation of \"struct page\"\n> + * zeroing by defining this macro in .\n> + */\n> +#ifndef mm_zero_struct_page\n> +#define mm_zero_struct_page(pp) ((void)memset((pp), 0, sizeof(struct page)))\n> +#endif\n> +\n> /*\n> * Default maximum number of active map areas, this limits the number of vmas\n> * per mm struct. Users can overwrite this number by sysctl but there is a\n> diff --git a/mm/page_alloc.c b/mm/page_alloc.c\n> index a8dbd405ed94..4b630ee91430 100644\n> --- a/mm/page_alloc.c\n> +++ b/mm/page_alloc.c\n> @@ -1170,6 +1170,7 @@ static void free_one_page(struct zone *zone,\n> static void __meminit __init_single_page(struct page *page, unsigned long pfn,\n> \t\t\t\tunsigned long zone, int nid)\n> {\n> +\tmm_zero_struct_page(page);\n> \tset_page_links(page, zone, nid, pfn);\n> \tinit_page_count(page);\n> \tpage_mapcount_reset(page);\n> -- \n> 2.14.1","headers":{"Return-Path":"","X-Original-To":"patchwork-incoming@ozlabs.org","Delivered-To":"patchwork-incoming@ozlabs.org","Authentication-Results":"ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=vger.kernel.org\n\t(client-ip=209.132.180.67; helo=vger.kernel.org;\n\tenvelope-from=sparclinux-owner@vger.kernel.org;\n\treceiver=)","Received":["from vger.kernel.org (vger.kernel.org [209.132.180.67])\n\tby ozlabs.org (Postfix) with ESMTP id 3y5zrb5dY2z9s06\n\tfor ;\n\tWed, 4 Oct 2017 00:09:19 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752911AbdJCNJB (ORCPT );\n\tTue, 3 Oct 2017 09:09:01 -0400","from mx2.suse.de ([195.135.220.15]:46826 \"EHLO mx1.suse.de\"\n\trhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP\n\tid S1753173AbdJCNJA (ORCPT );\n\tTue, 3 Oct 2017 09:09:00 -0400","from relay1.suse.de (charybdis-ext.suse.de [195.135.220.254])\n\tby mx1.suse.de (Postfix) with ESMTP id 4E0E3AD6D;\n\tTue, 3 Oct 2017 13:08:58 +0000 (UTC)"],"X-Virus-Scanned":"by amavisd-new at test-mx.suse.de","Date":"Tue, 3 Oct 2017 15:08:57 +0200","From":"Michal Hocko ","To":"Pavel Tatashin ","Cc":"linux-kernel@vger.kernel.org, sparclinux@vger.kernel.org,\n\tlinux-mm@kvack.org, linuxppc-dev@lists.ozlabs.org,\n\tlinux-s390@vger.kernel.org, linux-arm-kernel@lists.infradead.org,\n\tx86@kernel.org, kasan-dev@googlegroups.com, borntraeger@de.ibm.com,\n\theiko.carstens@de.ibm.com, davem@davemloft.net,\n\twilly@infradead.org, ard.biesheuvel@linaro.org,\n\tmark.rutland@arm.com, will.deacon@arm.com, catalin.marinas@arm.com,\n\tsam@ravnborg.org, mgorman@techsingularity.net,\n\tsteven.sistare@oracle.com, daniel.m.jordan@oracle.com,\n\tbob.picco@oracle.com","Subject":"Re: [PATCH v9 06/12] mm: zero struct pages during initialization","Message-ID":"<20171003130857.vohli6lnqj4tdmhl@dhcp22.suse.cz>","References":"<20170920201714.19817-1-pasha.tatashin@oracle.com>\n\t<20170920201714.19817-7-pasha.tatashin@oracle.com>","MIME-Version":"1.0","Content-Type":"text/plain; charset=us-ascii","Content-Disposition":"inline","In-Reply-To":"<20170920201714.19817-7-pasha.tatashin@oracle.com>","User-Agent":"NeoMutt/20170609 (1.8.3)","Sender":"sparclinux-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"sparclinux@vger.kernel.org"}},{"id":1779075,"web_url":"http://patchwork.ozlabs.org/comment/1779075/","msgid":"<73ea1215-7aa2-39e1-b820-30f58119183e@oracle.com>","list_archive_url":null,"date":"2017-10-03T15:22:35","subject":"Re: [PATCH v9 06/12] mm: zero struct pages during initialization","submitter":{"id":71010,"url":"http://patchwork.ozlabs.org/api/people/71010/","name":"Pavel Tatashin","email":"pasha.tatashin@oracle.com"},"content":"On 10/03/2017 09:08 AM, Michal Hocko wrote:\n> On Wed 20-09-17 16:17:08, Pavel Tatashin wrote:\n>> Add struct page zeroing as a part of initialization of other fields in\n>> __init_single_page().\n>>\n>> This single thread performance collected on: Intel(R) Xeon(R) CPU E7-8895\n>> v3 @ 2.60GHz with 1T of memory (268400646 pages in 8 nodes):\n>>\n>> BASE FIX\n>> sparse_init 11.244671836s 0.007199623s\n>> zone_sizes_init 4.879775891s 8.355182299s\n>> --------------------------\n>> Total 16.124447727s 8.362381922s\n> \n> Hmm, this is confusing. This assumes that sparse_init doesn't zero pages\n> anymore, right? So these number depend on the last patch in the series?\n\nCorrect, without the last patch sparse_init time won't change.\n\nPasha\n--\nTo unsubscribe from this list: send the line \"unsubscribe sparclinux\" in\nthe body of a message to majordomo@vger.kernel.org\nMore majordomo info at http://vger.kernel.org/majordomo-info.html","headers":{"Return-Path":"","X-Original-To":"patchwork-incoming@ozlabs.org","Delivered-To":"patchwork-incoming@ozlabs.org","Authentication-Results":"ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=vger.kernel.org\n\t(client-ip=209.132.180.67; helo=vger.kernel.org;\n\tenvelope-from=sparclinux-owner@vger.kernel.org;\n\treceiver=)","Received":["from vger.kernel.org (vger.kernel.org [209.132.180.67])\n\tby ozlabs.org (Postfix) with ESMTP id 3y62qs2mx2z9sPk\n\tfor ;\n\tWed, 4 Oct 2017 02:23:53 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752355AbdJCPXw (ORCPT );\n\tTue, 3 Oct 2017 11:23:52 -0400","from aserp1040.oracle.com ([141.146.126.69]:38905 \"EHLO\n\taserp1040.oracle.com\" rhost-flags-OK-OK-OK-OK) by vger.kernel.org\n\twith ESMTP id S1752351AbdJCPXv (ORCPT\n\t); Tue, 3 Oct 2017 11:23:51 -0400","from userv0022.oracle.com (userv0022.oracle.com [156.151.31.74])\n\tby aserp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2) with\n\tESMTP id v93FMjsX006181\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Tue, 3 Oct 2017 15:22:46 GMT","from aserv0121.oracle.com (aserv0121.oracle.com [141.146.126.235])\n\tby userv0022.oracle.com (8.14.4/8.14.4) with ESMTP id\n\tv93FMi3q021707\n\t(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Tue, 3 Oct 2017 15:22:45 GMT","from abhmp0018.oracle.com (abhmp0018.oracle.com [141.146.116.24])\n\tby aserv0121.oracle.com (8.14.4/8.13.8) with ESMTP id\n\tv93FMiZr004139; Tue, 3 Oct 2017 15:22:44 GMT","from [192.168.1.10] (/98.216.35.41)\n\tby default (Oracle Beehive Gateway v4.0)\n\twith ESMTP ; Tue, 03 Oct 2017 08:22:44 -0700"],"Subject":"Re: [PATCH v9 06/12] mm: zero struct pages during initialization","To":"Michal Hocko ","Cc":"linux-kernel@vger.kernel.org, sparclinux@vger.kernel.org,\n\tlinux-mm@kvack.org, linuxppc-dev@lists.ozlabs.org,\n\tlinux-s390@vger.kernel.org, linux-arm-kernel@lists.infradead.org,\n\tx86@kernel.org, kasan-dev@googlegroups.com, borntraeger@de.ibm.com,\n\theiko.carstens@de.ibm.com, davem@davemloft.net,\n\twilly@infradead.org, ard.biesheuvel@linaro.org,\n\tmark.rutland@arm.com, will.deacon@arm.com, catalin.marinas@arm.com,\n\tsam@ravnborg.org, mgorman@techsingularity.net,\n\tsteven.sistare@oracle.com, daniel.m.jordan@oracle.com,\n\tbob.picco@oracle.com","References":"<20170920201714.19817-1-pasha.tatashin@oracle.com>\n\t<20170920201714.19817-7-pasha.tatashin@oracle.com>\n\t<20171003130857.vohli6lnqj4tdmhl@dhcp22.suse.cz>","From":"Pasha Tatashin ","Message-ID":"<73ea1215-7aa2-39e1-b820-30f58119183e@oracle.com>","Date":"Tue, 3 Oct 2017 11:22:35 -0400","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.3.0","MIME-Version":"1.0","In-Reply-To":"<20171003130857.vohli6lnqj4tdmhl@dhcp22.suse.cz>","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Language":"en-US","Content-Transfer-Encoding":"7bit","X-Source-IP":"userv0022.oracle.com [156.151.31.74]","Sender":"sparclinux-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"sparclinux@vger.kernel.org"}},{"id":1779530,"web_url":"http://patchwork.ozlabs.org/comment/1779530/","msgid":"<20171004084554.fhxpmywtovs5umnm@dhcp22.suse.cz>","list_archive_url":null,"date":"2017-10-04T08:45:54","subject":"Re: [PATCH v9 06/12] mm: zero struct pages during initialization","submitter":{"id":66979,"url":"http://patchwork.ozlabs.org/api/people/66979/","name":"Michal Hocko","email":"mhocko@kernel.org"},"content":"On Tue 03-10-17 11:22:35, Pasha Tatashin wrote:\n> \n> \n> On 10/03/2017 09:08 AM, Michal Hocko wrote:\n> > On Wed 20-09-17 16:17:08, Pavel Tatashin wrote:\n> > > Add struct page zeroing as a part of initialization of other fields in\n> > > __init_single_page().\n> > > \n> > > This single thread performance collected on: Intel(R) Xeon(R) CPU E7-8895\n> > > v3 @ 2.60GHz with 1T of memory (268400646 pages in 8 nodes):\n> > > \n> > > BASE FIX\n> > > sparse_init 11.244671836s 0.007199623s\n> > > zone_sizes_init 4.879775891s 8.355182299s\n> > > --------------------------\n> > > Total 16.124447727s 8.362381922s\n> > \n> > Hmm, this is confusing. This assumes that sparse_init doesn't zero pages\n> > anymore, right? So these number depend on the last patch in the series?\n> \n> Correct, without the last patch sparse_init time won't change.\n\nTHen this is just misleading.","headers":{"Return-Path":"","X-Original-To":"patchwork-incoming@ozlabs.org","Delivered-To":"patchwork-incoming@ozlabs.org","Authentication-Results":"ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=vger.kernel.org\n\t(client-ip=209.132.180.67; helo=vger.kernel.org;\n\tenvelope-from=sparclinux-owner@vger.kernel.org;\n\treceiver=)","Received":["from vger.kernel.org (vger.kernel.org [209.132.180.67])\n\tby ozlabs.org (Postfix) with ESMTP id 3y6Tyc17TYz9t1G\n\tfor ;\n\tWed, 4 Oct 2017 19:46:16 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1751988AbdJDIp6 (ORCPT );\n\tWed, 4 Oct 2017 04:45:58 -0400","from mx2.suse.de ([195.135.220.15]:60334 \"EHLO mx1.suse.de\"\n\trhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP\n\tid S1751736AbdJDIp5 (ORCPT );\n\tWed, 4 Oct 2017 04:45:57 -0400","from relay2.suse.de (charybdis-ext.suse.de [195.135.220.254])\n\tby mx1.suse.de (Postfix) with ESMTP id E2CE0AAEF;\n\tWed, 4 Oct 2017 08:45:55 +0000 (UTC)"],"X-Virus-Scanned":"by amavisd-new at test-mx.suse.de","Date":"Wed, 4 Oct 2017 10:45:54 +0200","From":"Michal Hocko ","To":"Pasha Tatashin ","Cc":"linux-kernel@vger.kernel.org, sparclinux@vger.kernel.org,\n\tlinux-mm@kvack.org, linuxppc-dev@lists.ozlabs.org,\n\tlinux-s390@vger.kernel.org, linux-arm-kernel@lists.infradead.org,\n\tx86@kernel.org, kasan-dev@googlegroups.com, borntraeger@de.ibm.com,\n\theiko.carstens@de.ibm.com, davem@davemloft.net,\n\twilly@infradead.org, ard.biesheuvel@linaro.org,\n\tmark.rutland@arm.com, will.deacon@arm.com, catalin.marinas@arm.com,\n\tsam@ravnborg.org, mgorman@techsingularity.net,\n\tsteven.sistare@oracle.com, daniel.m.jordan@oracle.com,\n\tbob.picco@oracle.com","Subject":"Re: [PATCH v9 06/12] mm: zero struct pages during initialization","Message-ID":"<20171004084554.fhxpmywtovs5umnm@dhcp22.suse.cz>","References":"<20170920201714.19817-1-pasha.tatashin@oracle.com>\n\t<20170920201714.19817-7-pasha.tatashin@oracle.com>\n\t<20171003130857.vohli6lnqj4tdmhl@dhcp22.suse.cz>\n\t<73ea1215-7aa2-39e1-b820-30f58119183e@oracle.com>","MIME-Version":"1.0","Content-Type":"text/plain; charset=us-ascii","Content-Disposition":"inline","In-Reply-To":"<73ea1215-7aa2-39e1-b820-30f58119183e@oracle.com>","User-Agent":"NeoMutt/20170609 (1.8.3)","Sender":"sparclinux-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"sparclinux@vger.kernel.org"}},{"id":1779754,"web_url":"http://patchwork.ozlabs.org/comment/1779754/","msgid":"<2fd7b718-c254-612b-e100-a457094e7c38@oracle.com>","list_archive_url":null,"date":"2017-10-04T12:26:57","subject":"Re: [PATCH v9 06/12] mm: zero struct pages during initialization","submitter":{"id":71010,"url":"http://patchwork.ozlabs.org/api/people/71010/","name":"Pavel Tatashin","email":"pasha.tatashin@oracle.com"},"content":"On 10/04/2017 04:45 AM, Michal Hocko wrote:\n> On Tue 03-10-17 11:22:35, Pasha Tatashin wrote:\n>>\n>>\n>> On 10/03/2017 09:08 AM, Michal Hocko wrote:\n>>> On Wed 20-09-17 16:17:08, Pavel Tatashin wrote:\n>>>> Add struct page zeroing as a part of initialization of other fields in\n>>>> __init_single_page().\n>>>>\n>>>> This single thread performance collected on: Intel(R) Xeon(R) CPU E7-8895\n>>>> v3 @ 2.60GHz with 1T of memory (268400646 pages in 8 nodes):\n>>>>\n>>>> BASE FIX\n>>>> sparse_init 11.244671836s 0.007199623s\n>>>> zone_sizes_init 4.879775891s 8.355182299s\n>>>> --------------------------\n>>>> Total 16.124447727s 8.362381922s\n>>>\n>>> Hmm, this is confusing. This assumes that sparse_init doesn't zero pages\n>>> anymore, right? So these number depend on the last patch in the series?\n>>\n>> Correct, without the last patch sparse_init time won't change.\n> \n> THen this is just misleading.\n> \n\nOK, I will re-arrange patches the way you suggested earlier.\n\nPasha\n--\nTo unsubscribe from this list: send the line \"unsubscribe sparclinux\" in\nthe body of a message to majordomo@vger.kernel.org\nMore majordomo info at http://vger.kernel.org/majordomo-info.html","headers":{"Return-Path":"","X-Original-To":"patchwork-incoming@ozlabs.org","Delivered-To":"patchwork-incoming@ozlabs.org","Authentication-Results":"ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=vger.kernel.org\n\t(client-ip=209.132.180.67; helo=vger.kernel.org;\n\tenvelope-from=sparclinux-owner@vger.kernel.org;\n\treceiver=)","Received":["from vger.kernel.org (vger.kernel.org [209.132.180.67])\n\tby ozlabs.org (Postfix) with ESMTP id 3y6Zv03GPbz9t16\n\tfor ;\n\tWed, 4 Oct 2017 23:28:28 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752040AbdJDM21 (ORCPT );\n\tWed, 4 Oct 2017 08:28:27 -0400","from aserp1040.oracle.com ([141.146.126.69]:20221 \"EHLO\n\taserp1040.oracle.com\" rhost-flags-OK-OK-OK-OK) by vger.kernel.org\n\twith ESMTP id S1751947AbdJDM20 (ORCPT\n\t); Wed, 4 Oct 2017 08:28:26 -0400","from userv0021.oracle.com (userv0021.oracle.com [156.151.31.71])\n\tby aserp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2) with\n\tESMTP id v94CR8ao010922\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Wed, 4 Oct 2017 12:27:09 GMT","from aserv0121.oracle.com (aserv0121.oracle.com [141.146.126.235])\n\tby userv0021.oracle.com (8.14.4/8.14.4) with ESMTP id\n\tv94CR6O4023745\n\t(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Wed, 4 Oct 2017 12:27:06 GMT","from abhmp0017.oracle.com (abhmp0017.oracle.com [141.146.116.23])\n\tby aserv0121.oracle.com (8.14.4/8.13.8) with ESMTP id\n\tv94CR3g8026336; Wed, 4 Oct 2017 12:27:03 GMT","from [192.168.1.10] (/98.216.35.41)\n\tby default (Oracle Beehive Gateway v4.0)\n\twith ESMTP ; Wed, 04 Oct 2017 05:27:02 -0700"],"Subject":"Re: [PATCH v9 06/12] mm: zero struct pages during initialization","To":"Michal Hocko ","Cc":"linux-kernel@vger.kernel.org, sparclinux@vger.kernel.org,\n\tlinux-mm@kvack.org, linuxppc-dev@lists.ozlabs.org,\n\tlinux-s390@vger.kernel.org, linux-arm-kernel@lists.infradead.org,\n\tx86@kernel.org, kasan-dev@googlegroups.com, borntraeger@de.ibm.com,\n\theiko.carstens@de.ibm.com, davem@davemloft.net,\n\twilly@infradead.org, ard.biesheuvel@linaro.org,\n\tmark.rutland@arm.com, will.deacon@arm.com, catalin.marinas@arm.com,\n\tsam@ravnborg.org, mgorman@techsingularity.net,\n\tsteven.sistare@oracle.com, daniel.m.jordan@oracle.com,\n\tbob.picco@oracle.com","References":"<20170920201714.19817-1-pasha.tatashin@oracle.com>\n\t<20170920201714.19817-7-pasha.tatashin@oracle.com>\n\t<20171003130857.vohli6lnqj4tdmhl@dhcp22.suse.cz>\n\t<73ea1215-7aa2-39e1-b820-30f58119183e@oracle.com>\n\t<20171004084554.fhxpmywtovs5umnm@dhcp22.suse.cz>","From":"Pasha Tatashin ","Message-ID":"<2fd7b718-c254-612b-e100-a457094e7c38@oracle.com>","Date":"Wed, 4 Oct 2017 08:26:57 -0400","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.3.0","MIME-Version":"1.0","In-Reply-To":"<20171004084554.fhxpmywtovs5umnm@dhcp22.suse.cz>","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Language":"en-US","Content-Transfer-Encoding":"7bit","X-Source-IP":"userv0021.oracle.com [156.151.31.71]","Sender":"sparclinux-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"sparclinux@vger.kernel.org"}}]