[{"id":1778957,"web_url":"http://patchwork.ozlabs.org/comment/1778957/","msgid":"<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>","list_archive_url":null,"date":"2017-10-03T13:18:17","subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","submitter":{"id":66979,"url":"http://patchwork.ozlabs.org/api/people/66979/","name":"Michal Hocko","email":"mhocko@kernel.org"},"content":"On Wed 20-09-17 16:17:10, Pavel Tatashin wrote:\n> Some memory is reserved but unavailable: not present in memblock.memory\n> (because not backed by physical pages), but present in memblock.reserved.\n> Such memory has backing struct pages, but they are not initialized by going\n> through __init_single_page().\n\nCould you be more specific where is such a memory reserved?","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 3y603H74XWz9sPr\n\tfor ;\n\tWed, 4 Oct 2017 00:18:35 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752213AbdJCNSW (ORCPT );\n\tTue, 3 Oct 2017 09:18:22 -0400","from mx2.suse.de ([195.135.220.15]:47751 \"EHLO mx1.suse.de\"\n\trhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP\n\tid S1751345AbdJCNSU (ORCPT );\n\tTue, 3 Oct 2017 09:18:20 -0400","from relay1.suse.de (charybdis-ext.suse.de [195.135.220.254])\n\tby mx1.suse.de (Postfix) with ESMTP id A3AF1AD6D;\n\tTue, 3 Oct 2017 13:18:18 +0000 (UTC)"],"X-Virus-Scanned":"by amavisd-new at test-mx.suse.de","Date":"Tue, 3 Oct 2017 15:18:17 +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 08/12] mm: zero reserved and unavailable struct pages","Message-ID":"<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>","References":"<20170920201714.19817-1-pasha.tatashin@oracle.com>\n\t<20170920201714.19817-9-pasha.tatashin@oracle.com>","MIME-Version":"1.0","Content-Type":"text/plain; charset=us-ascii","Content-Disposition":"inline","In-Reply-To":"<20170920201714.19817-9-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":1779081,"web_url":"http://patchwork.ozlabs.org/comment/1779081/","msgid":"<691dba28-718c-e9a9-d006-88505eb5cd7e@oracle.com>","list_archive_url":null,"date":"2017-10-03T15:29:16","subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","submitter":{"id":71010,"url":"http://patchwork.ozlabs.org/api/people/71010/","name":"Pavel Tatashin","email":"pasha.tatashin@oracle.com"},"content":"On 10/03/2017 09:18 AM, Michal Hocko wrote:\n> On Wed 20-09-17 16:17:10, Pavel Tatashin wrote:\n>> Some memory is reserved but unavailable: not present in memblock.memory\n>> (because not backed by physical pages), but present in memblock.reserved.\n>> Such memory has backing struct pages, but they are not initialized by going\n>> through __init_single_page().\n> \n> Could you be more specific where is such a memory reserved?\n> \n\nI know of one example: trim_low_memory_range() unconditionally reserves \nfrom pfn 0, but e820__memblock_setup() might provide the exiting memory \nfrom pfn 1 (i.e. KVM).\n\nBut, there could be more based on this comment from linux/page-flags.h:\n\n 19 * PG_reserved is set for special pages, which can never be swapped \nout. Some\n 20 * of them might not even exist (eg empty_bad_page)...\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 3y630D12Xcz9s9Y\n\tfor ;\n\tWed, 4 Oct 2017 02:31:08 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752342AbdJCPay (ORCPT );\n\tTue, 3 Oct 2017 11:30:54 -0400","from userp1040.oracle.com ([156.151.31.81]:48327 \"EHLO\n\tuserp1040.oracle.com\" rhost-flags-OK-OK-OK-OK) by vger.kernel.org\n\twith ESMTP id S1751953AbdJCPaw (ORCPT\n\t); Tue, 3 Oct 2017 11:30:52 -0400","from aserv0022.oracle.com (aserv0022.oracle.com [141.146.126.234])\n\tby userp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2)\n\twith ESMTP id v93FTR9t006397\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Tue, 3 Oct 2017 15:29:28 GMT","from userv0122.oracle.com (userv0122.oracle.com [156.151.31.75])\n\tby aserv0022.oracle.com (8.14.4/8.14.4) with ESMTP id v93FTQ1d020907\n\t(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Tue, 3 Oct 2017 15:29:27 GMT","from abhmp0002.oracle.com (abhmp0002.oracle.com [141.146.116.8])\n\tby userv0122.oracle.com (8.14.4/8.14.4) with ESMTP id v93FTPvc004296; \n\tTue, 3 Oct 2017 15:29:25 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 08:29:25 -0700"],"Subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","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<20170920201714.19817-9-pasha.tatashin@oracle.com>\n\t<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>","From":"Pasha Tatashin ","Message-ID":"<691dba28-718c-e9a9-d006-88505eb5cd7e@oracle.com>","Date":"Tue, 3 Oct 2017 11:29:16 -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":"<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Language":"en-US","Content-Transfer-Encoding":"7bit","X-Source-IP":"aserv0022.oracle.com [141.146.126.234]","Sender":"sparclinux-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"sparclinux@vger.kernel.org"}},{"id":1779537,"web_url":"http://patchwork.ozlabs.org/comment/1779537/","msgid":"<20171004085636.w2rnwf5xxhahzuy7@dhcp22.suse.cz>","list_archive_url":null,"date":"2017-10-04T08:56:36","subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","submitter":{"id":66979,"url":"http://patchwork.ozlabs.org/api/people/66979/","name":"Michal Hocko","email":"mhocko@kernel.org"},"content":"On Tue 03-10-17 11:29:16, Pasha Tatashin wrote:\n> On 10/03/2017 09:18 AM, Michal Hocko wrote:\n> > On Wed 20-09-17 16:17:10, Pavel Tatashin wrote:\n> > > Some memory is reserved but unavailable: not present in memblock.memory\n> > > (because not backed by physical pages), but present in memblock.reserved.\n> > > Such memory has backing struct pages, but they are not initialized by going\n> > > through __init_single_page().\n> > \n> > Could you be more specific where is such a memory reserved?\n> > \n> \n> I know of one example: trim_low_memory_range() unconditionally reserves from\n> pfn 0, but e820__memblock_setup() might provide the exiting memory from pfn\n> 1 (i.e. KVM).\n\nThen just initialize struct pages for that mapping rigth there where a\nspecial API is used.\n\n> But, there could be more based on this comment from linux/page-flags.h:\n> \n> 19 * PG_reserved is set for special pages, which can never be swapped out.\n> Some\n> 20 * of them might not even exist (eg empty_bad_page)...\n\nI have no idea wht empty_bad_page is but a quick grep shows that this is\nnever used. I might be wrong here but if somebody is reserving a memory\nin a special way then we should handle the initialization right there.\nE.g. create an API for special memblock reservations.","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 3y6VBd1cyQz9sRV\n\tfor ;\n\tWed, 4 Oct 2017 19:56:41 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752006AbdJDI4k (ORCPT );\n\tWed, 4 Oct 2017 04:56:40 -0400","from mx2.suse.de ([195.135.220.15]:35076 \"EHLO mx1.suse.de\"\n\trhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP\n\tid S1751736AbdJDI4i (ORCPT );\n\tWed, 4 Oct 2017 04:56:38 -0400","from relay1.suse.de (charybdis-ext.suse.de [195.135.220.254])\n\tby mx1.suse.de (Postfix) with ESMTP id 07AE7AAEF;\n\tWed, 4 Oct 2017 08:56:36 +0000 (UTC)"],"X-Virus-Scanned":"by amavisd-new at test-mx.suse.de","Date":"Wed, 4 Oct 2017 10:56:36 +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 08/12] mm: zero reserved and unavailable struct pages","Message-ID":"<20171004085636.w2rnwf5xxhahzuy7@dhcp22.suse.cz>","References":"<20170920201714.19817-1-pasha.tatashin@oracle.com>\n\t<20170920201714.19817-9-pasha.tatashin@oracle.com>\n\t<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>\n\t<691dba28-718c-e9a9-d006-88505eb5cd7e@oracle.com>","MIME-Version":"1.0","Content-Type":"text/plain; charset=us-ascii","Content-Disposition":"inline","In-Reply-To":"<691dba28-718c-e9a9-d006-88505eb5cd7e@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":1779772,"web_url":"http://patchwork.ozlabs.org/comment/1779772/","msgid":"<9198a33d-cd40-dd70-4823-7f70c57ef9a2@oracle.com>","list_archive_url":null,"date":"2017-10-04T12:40:11","subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","submitter":{"id":71010,"url":"http://patchwork.ozlabs.org/api/people/71010/","name":"Pavel Tatashin","email":"pasha.tatashin@oracle.com"},"content":">>> Could you be more specific where is such a memory reserved?\n>>>\n>>\n>> I know of one example: trim_low_memory_range() unconditionally reserves from\n>> pfn 0, but e820__memblock_setup() might provide the exiting memory from pfn\n>> 1 (i.e. KVM).\n> \n> Then just initialize struct pages for that mapping rigth there where a\n> special API is used.\n> \n>> But, there could be more based on this comment from linux/page-flags.h:\n>>\n>> 19 * PG_reserved is set for special pages, which can never be swapped out.\n>> Some\n>> 20 * of them might not even exist (eg empty_bad_page)...\n> \n> I have no idea wht empty_bad_page is but a quick grep shows that this is\n> never used. I might be wrong here but if somebody is reserving a memory\n> in a special way then we should handle the initialization right there.\n> E.g. create an API for special memblock reservations.\n> \n\nHi Michal,\n\nThe reservations happen before struct pages are allocated and mapped. \nSo, it is not always possible to do it at call sites.\n\nPreviously, I have solved this problem like this:\n\nhttps://patchwork.kernel.org/patch/9886163\n\nBut, I was not too happy with that approach, so I replaced it with the \ncurrent approach as it is more generic, and solves similar issues if \nthey happen in other places. Also, the comment in page-flags got me \nscared that there are probably other places perhaps on other \narchitectures that can have the similar issue.\n\nIn addition, I did not like my solution, I was simply shrinking the low \nreservation from:\n[0 - reserve_low) to [min_pfn - reserve_low), but if min_pfn > \nreserve_low can we skip low reservation entirely? I was not sure.\n\nThe current approach notifies us if there are such pages, and we can \nfix/remove them in the future without crashing kernel in the meantime.\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 3y6bB63hZZz9t16\n\tfor ;\n\tWed, 4 Oct 2017 23:41:34 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1751996AbdJDMld (ORCPT );\n\tWed, 4 Oct 2017 08:41:33 -0400","from aserp1040.oracle.com ([141.146.126.69]:30455 \"EHLO\n\taserp1040.oracle.com\" rhost-flags-OK-OK-OK-OK) by vger.kernel.org\n\twith ESMTP id S1751699AbdJDMlc (ORCPT\n\t); Wed, 4 Oct 2017 08:41:32 -0400","from aserv0021.oracle.com (aserv0021.oracle.com [141.146.126.233])\n\tby aserp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2)\n\twith ESMTP id v94CeGIH001666\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Wed, 4 Oct 2017 12:40:16 GMT","from userv0121.oracle.com (userv0121.oracle.com [156.151.31.72])\n\tby aserv0021.oracle.com (8.14.4/8.14.4) with ESMTP id v94CeF12017474\n\t(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Wed, 4 Oct 2017 12:40:15 GMT","from abhmp0012.oracle.com (abhmp0012.oracle.com [141.146.116.18])\n\tby userv0121.oracle.com (8.14.4/8.13.8) with ESMTP id\n\tv94CeDVh010487; Wed, 4 Oct 2017 12:40:14 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 05:40:13 -0700"],"Subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","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<20170920201714.19817-9-pasha.tatashin@oracle.com>\n\t<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>\n\t<691dba28-718c-e9a9-d006-88505eb5cd7e@oracle.com>\n\t<20171004085636.w2rnwf5xxhahzuy7@dhcp22.suse.cz>","From":"Pasha Tatashin ","Message-ID":"<9198a33d-cd40-dd70-4823-7f70c57ef9a2@oracle.com>","Date":"Wed, 4 Oct 2017 08:40:11 -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":"<20171004085636.w2rnwf5xxhahzuy7@dhcp22.suse.cz>","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Language":"en-US","Content-Transfer-Encoding":"7bit","X-Source-IP":"aserv0021.oracle.com [141.146.126.233]","Sender":"sparclinux-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"sparclinux@vger.kernel.org"}},{"id":1779786,"web_url":"http://patchwork.ozlabs.org/comment/1779786/","msgid":"<20171004125743.fm6mf2artbga76et@dhcp22.suse.cz>","list_archive_url":null,"date":"2017-10-04T12:57:43","subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","submitter":{"id":66979,"url":"http://patchwork.ozlabs.org/api/people/66979/","name":"Michal Hocko","email":"mhocko@kernel.org"},"content":"On Wed 04-10-17 08:40:11, Pasha Tatashin wrote:\n> > > > Could you be more specific where is such a memory reserved?\n> > > > \n> > > \n> > > I know of one example: trim_low_memory_range() unconditionally reserves from\n> > > pfn 0, but e820__memblock_setup() might provide the exiting memory from pfn\n> > > 1 (i.e. KVM).\n> > \n> > Then just initialize struct pages for that mapping rigth there where a\n> > special API is used.\n> > \n> > > But, there could be more based on this comment from linux/page-flags.h:\n> > > \n> > > 19 * PG_reserved is set for special pages, which can never be swapped out.\n> > > Some\n> > > 20 * of them might not even exist (eg empty_bad_page)...\n> > \n> > I have no idea wht empty_bad_page is but a quick grep shows that this is\n> > never used. I might be wrong here but if somebody is reserving a memory\n> > in a special way then we should handle the initialization right there.\n> > E.g. create an API for special memblock reservations.\n> > \n> \n> Hi Michal,\n> \n> The reservations happen before struct pages are allocated and mapped. So, it\n> is not always possible to do it at call sites.\n\nOK, I didn't realize that.\n \n> Previously, I have solved this problem like this:\n> \n> https://patchwork.kernel.org/patch/9886163\n> \n> But, I was not too happy with that approach, so I replaced it with the\n> current approach as it is more generic, and solves similar issues if they\n> happen in other places. Also, the comment in page-flags got me scared that\n> there are probably other places perhaps on other architectures that can have\n> the similar issue.\n\nI believe the comment is just stale. I have looked into empty_bad_page\nand it is just a relict. I plan to post a patch soon.\n\n> In addition, I did not like my solution, I was simply shrinking the low\n> reservation from:\n> [0 - reserve_low) to [min_pfn - reserve_low), but if min_pfn > reserve_low\n> can we skip low reservation entirely? I was not sure.\n> \n> The current approach notifies us if there are such pages, and we can\n> fix/remove them in the future without crashing kernel in the meantime.\n\nI am not really familiar with the trim_low_memory_range code path. I am\nnot even sure we have to care about it because nobody should be walking\npfns outside of any zone. I am worried that this patch adds a code which\nis not really used and it will just stay that way for ever because\nnobody will dare to change it as it is too obscure and not explained\nvery well. trim_low_memory_range is a good example of this. Why do we\neven reserve this range from the memory block allocator? The memory\nshouldn't be backed by any real memory and thus not in the allocator in\nthe first place, no?","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 3y6bYC5Z2sz9t2S\n\tfor ;\n\tWed, 4 Oct 2017 23:58:07 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1751969AbdJDM5y (ORCPT );\n\tWed, 4 Oct 2017 08:57:54 -0400","from mx2.suse.de ([195.135.220.15]:53221 \"EHLO mx1.suse.de\"\n\trhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP\n\tid S1751742AbdJDM5x (ORCPT );\n\tWed, 4 Oct 2017 08:57:53 -0400","from relay2.suse.de (charybdis-ext.suse.de [195.135.220.254])\n\tby mx1.suse.de (Postfix) with ESMTP id 33BFBACFA;\n\tWed, 4 Oct 2017 12:57:50 +0000 (UTC)"],"X-Virus-Scanned":"by amavisd-new at test-mx.suse.de","Date":"Wed, 4 Oct 2017 14:57:43 +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 08/12] mm: zero reserved and unavailable struct pages","Message-ID":"<20171004125743.fm6mf2artbga76et@dhcp22.suse.cz>","References":"<20170920201714.19817-1-pasha.tatashin@oracle.com>\n\t<20170920201714.19817-9-pasha.tatashin@oracle.com>\n\t<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>\n\t<691dba28-718c-e9a9-d006-88505eb5cd7e@oracle.com>\n\t<20171004085636.w2rnwf5xxhahzuy7@dhcp22.suse.cz>\n\t<9198a33d-cd40-dd70-4823-7f70c57ef9a2@oracle.com>","MIME-Version":"1.0","Content-Type":"text/plain; charset=us-ascii","Content-Disposition":"inline","In-Reply-To":"<9198a33d-cd40-dd70-4823-7f70c57ef9a2@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":1779818,"web_url":"http://patchwork.ozlabs.org/comment/1779818/","msgid":"","list_archive_url":null,"date":"2017-10-04T13:28:55","subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","submitter":{"id":71010,"url":"http://patchwork.ozlabs.org/api/people/71010/","name":"Pavel Tatashin","email":"pasha.tatashin@oracle.com"},"content":"> I am not really familiar with the trim_low_memory_range code path. I am\n> not even sure we have to care about it because nobody should be walking\n> pfns outside of any zone.\n\nAccording to commit comments first 4K belongs to BIOS, so I think the \nmemory exists but BIOS may or may not report it to Linux. So, reserve it \nto make sure we never touch it.\n\n I am worried that this patch adds a code which\n> is not really used and it will just stay that way for ever because\n> nobody will dare to change it as it is too obscure and not explained\n> very well.\n\nI could explain mine code better. Perhaps add more comments, and explain \nwhen it can be removed?\n\n trim_low_memory_range is a good example of this. Why do we\n> even reserve this range from the memory block allocator? The memory\n> shouldn't be backed by any real memory and thus not in the allocator in\n> the first place, no?\n> \n\nSince it is not enforced in memblock that everything in reserved list \nmust be part of memory list, we can have it, and we need to make sure \nkernel does not panic. Otherwise, it is very hard to detect such bugs.\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 3y6cGn3SBfz9t2l\n\tfor ;\n\tThu, 5 Oct 2017 00:30:41 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752332AbdJDNa3 (ORCPT );\n\tWed, 4 Oct 2017 09:30:29 -0400","from userp1040.oracle.com ([156.151.31.81]:31943 \"EHLO\n\tuserp1040.oracle.com\" rhost-flags-OK-OK-OK-OK) by vger.kernel.org\n\twith ESMTP id S1752292AbdJDNa1 (ORCPT\n\t); Wed, 4 Oct 2017 09:30:27 -0400","from userv0022.oracle.com (userv0022.oracle.com [156.151.31.74])\n\tby userp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2) with\n\tESMTP id v94DT1AU020593\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Wed, 4 Oct 2017 13:29:01 GMT","from aserv0122.oracle.com (aserv0122.oracle.com [141.146.126.236])\n\tby userv0022.oracle.com (8.14.4/8.14.4) with ESMTP id\n\tv94DT0sH012100\n\t(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Wed, 4 Oct 2017 13:29:00 GMT","from abhmp0005.oracle.com (abhmp0005.oracle.com [141.146.116.11])\n\tby aserv0122.oracle.com (8.14.4/8.14.4) with ESMTP id\n\tv94DSwLu020790; Wed, 4 Oct 2017 13:28:58 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 06:28:58 -0700"],"Subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","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<20170920201714.19817-9-pasha.tatashin@oracle.com>\n\t<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>\n\t<691dba28-718c-e9a9-d006-88505eb5cd7e@oracle.com>\n\t<20171004085636.w2rnwf5xxhahzuy7@dhcp22.suse.cz>\n\t<9198a33d-cd40-dd70-4823-7f70c57ef9a2@oracle.com>\n\t<20171004125743.fm6mf2artbga76et@dhcp22.suse.cz>","From":"Pasha Tatashin ","Message-ID":"","Date":"Wed, 4 Oct 2017 09:28:55 -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":"<20171004125743.fm6mf2artbga76et@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":1779849,"web_url":"http://patchwork.ozlabs.org/comment/1779849/","msgid":"<20171004140410.2w2zf2gbutdxunir@dhcp22.suse.cz>","list_archive_url":null,"date":"2017-10-04T14:04:10","subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","submitter":{"id":66979,"url":"http://patchwork.ozlabs.org/api/people/66979/","name":"Michal Hocko","email":"mhocko@kernel.org"},"content":"On Wed 04-10-17 09:28:55, Pasha Tatashin wrote:\n> \n> > I am not really familiar with the trim_low_memory_range code path. I am\n> > not even sure we have to care about it because nobody should be walking\n> > pfns outside of any zone.\n> \n> According to commit comments first 4K belongs to BIOS, so I think the memory\n> exists but BIOS may or may not report it to Linux. So, reserve it to make\n> sure we never touch it.\n\nYes and that memory should be outside of any zones, no?\n\n> > I am worried that this patch adds a code which\n> > is not really used and it will just stay that way for ever because\n> > nobody will dare to change it as it is too obscure and not explained\n> > very well.\n> \n> I could explain mine code better. Perhaps add more comments, and explain\n> when it can be removed?\n\nMore explanation would be definitely helpful\n\n> > trim_low_memory_range is a good example of this. Why do we\n> > even reserve this range from the memory block allocator? The memory\n> > shouldn't be backed by any real memory and thus not in the allocator in\n> > the first place, no?\n> > \n> \n> Since it is not enforced in memblock that everything in reserved list must\n> be part of memory list, we can have it, and we need to make sure kernel does\n> not panic. Otherwise, it is very hard to detect such bugs.\n\nSo, should we report such a memblock reservation API (ab)use to the log?\nAre you actually sure that trim_low_memory_range is doing a sane and\nreally needed thing? In other words do we have a zone which contains\nthis no-memory backed pfns?","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 3y6d1X0gmgz9t16\n\tfor ;\n\tThu, 5 Oct 2017 01:04:16 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752085AbdJDOEP (ORCPT );\n\tWed, 4 Oct 2017 10:04:15 -0400","from mx2.suse.de ([195.135.220.15]:34887 \"EHLO mx1.suse.de\"\n\trhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP\n\tid S1751176AbdJDOEO (ORCPT );\n\tWed, 4 Oct 2017 10:04:14 -0400","from relay1.suse.de (charybdis-ext.suse.de [195.135.220.254])\n\tby mx1.suse.de (Postfix) with ESMTP id 2DAB1ADED;\n\tWed, 4 Oct 2017 14:04:12 +0000 (UTC)"],"X-Virus-Scanned":"by amavisd-new at test-mx.suse.de","Date":"Wed, 4 Oct 2017 16:04:10 +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 08/12] mm: zero reserved and unavailable struct pages","Message-ID":"<20171004140410.2w2zf2gbutdxunir@dhcp22.suse.cz>","References":"<20170920201714.19817-1-pasha.tatashin@oracle.com>\n\t<20170920201714.19817-9-pasha.tatashin@oracle.com>\n\t<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>\n\t<691dba28-718c-e9a9-d006-88505eb5cd7e@oracle.com>\n\t<20171004085636.w2rnwf5xxhahzuy7@dhcp22.suse.cz>\n\t<9198a33d-cd40-dd70-4823-7f70c57ef9a2@oracle.com>\n\t<20171004125743.fm6mf2artbga76et@dhcp22.suse.cz>\n\t","MIME-Version":"1.0","Content-Type":"text/plain; charset=us-ascii","Content-Disposition":"inline","In-Reply-To":"","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":1779923,"web_url":"http://patchwork.ozlabs.org/comment/1779923/","msgid":"","list_archive_url":null,"date":"2017-10-04T15:08:59","subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","submitter":{"id":71010,"url":"http://patchwork.ozlabs.org/api/people/71010/","name":"Pavel Tatashin","email":"pasha.tatashin@oracle.com"},"content":"On 10/04/2017 10:04 AM, Michal Hocko wrote:\n> On Wed 04-10-17 09:28:55, Pasha Tatashin wrote:\n>>\n>>> I am not really familiar with the trim_low_memory_range code path. I am\n>>> not even sure we have to care about it because nobody should be walking\n>>> pfns outside of any zone.\n>>\n>> According to commit comments first 4K belongs to BIOS, so I think the memory\n>> exists but BIOS may or may not report it to Linux. So, reserve it to make\n>> sure we never touch it.\n> \n> Yes and that memory should be outside of any zones, no?\n\nI am not totally sure, I think some x86 expert could help us here. But, \nin either case this issue can be fixed separately from the rest of the \nseries.\n\n> \n>>> I am worried that this patch adds a code which\n>>> is not really used and it will just stay that way for ever because\n>>> nobody will dare to change it as it is too obscure and not explained\n>>> very well.\n>>\n>> I could explain mine code better. Perhaps add more comments, and explain\n>> when it can be removed?\n> \n> More explanation would be definitely helpful\n> \n>>> trim_low_memory_range is a good example of this. Why do we\n>>> even reserve this range from the memory block allocator? The memory\n>>> shouldn't be backed by any real memory and thus not in the allocator in\n>>> the first place, no?\n>>>\n>>\n>> Since it is not enforced in memblock that everything in reserved list must\n>> be part of memory list, we can have it, and we need to make sure kernel does\n>> not panic. Otherwise, it is very hard to detect such bugs.\n> \n> So, should we report such a memblock reservation API (ab)use to the log?\n> Are you actually sure that trim_low_memory_range is doing a sane and\n> really needed thing? In other words do we have a zone which contains\n> this no-memory backed pfns?\n> \n\nAnd, this patch reports it already:\n\n+\tpr_info(\"Reserved but unavailable: %lld pages\", pgcnt);\n\nI could add a comment above this print call, explain that such memory is \nprobably bogus and must be studied/fixed. Also, add that this code can \nbe removed once memblock is changed to allow reserve only memory that is \nbacked by physical memory i.e. in \"memory\" list.\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 3y6fV56z3Pz9t1G\n\tfor ;\n\tThu, 5 Oct 2017 02:10:37 +1100 (AEDT)","(majordomo@vger.kernel.org) by vger.kernel.org via listexpand\n\tid S1752353AbdJDPKh (ORCPT );\n\tWed, 4 Oct 2017 11:10:37 -0400","from aserp1040.oracle.com ([141.146.126.69]:37727 \"EHLO\n\taserp1040.oracle.com\" rhost-flags-OK-OK-OK-OK) by vger.kernel.org\n\twith ESMTP id S1752330AbdJDPKg (ORCPT\n\t); Wed, 4 Oct 2017 11:10:36 -0400","from aserv0022.oracle.com (aserv0022.oracle.com [141.146.126.234])\n\tby aserp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2)\n\twith ESMTP id v94F951W031914\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Wed, 4 Oct 2017 15:09:06 GMT","from aserv0122.oracle.com (aserv0122.oracle.com [141.146.126.236])\n\tby aserv0022.oracle.com (8.14.4/8.14.4) with ESMTP id\n\tv94F94Cv017140\n\t(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256\n\tverify=OK); Wed, 4 Oct 2017 15:09:04 GMT","from abhmp0009.oracle.com (abhmp0009.oracle.com [141.146.116.15])\n\tby aserv0122.oracle.com (8.14.4/8.14.4) with ESMTP id\n\tv94F91Xp031016; Wed, 4 Oct 2017 15:09:02 GMT","from [10.154.139.198] (/10.154.139.198)\n\tby default (Oracle Beehive Gateway v4.0)\n\twith ESMTP ; Wed, 04 Oct 2017 08:09:01 -0700"],"Subject":"Re: [PATCH v9 08/12] mm: zero reserved and unavailable struct pages","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<20170920201714.19817-9-pasha.tatashin@oracle.com>\n\t<20171003131817.omzbam3js67edp3s@dhcp22.suse.cz>\n\t<691dba28-718c-e9a9-d006-88505eb5cd7e@oracle.com>\n\t<20171004085636.w2rnwf5xxhahzuy7@dhcp22.suse.cz>\n\t<9198a33d-cd40-dd70-4823-7f70c57ef9a2@oracle.com>\n\t<20171004125743.fm6mf2artbga76et@dhcp22.suse.cz>\n\t\n\t<20171004140410.2w2zf2gbutdxunir@dhcp22.suse.cz>","From":"Pasha Tatashin ","Message-ID":"","Date":"Wed, 4 Oct 2017 11:08:59 -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":"<20171004140410.2w2zf2gbutdxunir@dhcp22.suse.cz>","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Language":"en-US","Content-Transfer-Encoding":"7bit","X-Source-IP":"aserv0022.oracle.com [141.146.126.234]","Sender":"sparclinux-owner@vger.kernel.org","Precedence":"bulk","List-ID":"","X-Mailing-List":"sparclinux@vger.kernel.org"}}]