Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751501AbdG0IMm (ORCPT ); Thu, 27 Jul 2017 04:12:42 -0400 Received: from mx2.suse.de ([195.135.220.15]:49492 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750836AbdG0IMk (ORCPT ); Thu, 27 Jul 2017 04:12:40 -0400 Date: Thu, 27 Jul 2017 10:12:36 +0200 From: Michal Hocko To: "Aneesh Kumar K.V" Cc: Luiz Capitulino , Mike Kravetz , linux-mm@kvack.org, LKML Subject: Re: gigantic hugepages vs. movable zones Message-ID: <20170727081236.GK20970@dhcp22.suse.cz> References: <20170726105004.GI2981@dhcp22.suse.cz> <87inie1uwf.fsf@linux.vnet.ibm.com> <20170727072857.GI20970@dhcp22.suse.cz> <1529e986-5f28-35dd-c82e-a4b5801b4afd@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1529e986-5f28-35dd-c82e-a4b5801b4afd@linux.vnet.ibm.com> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1472 Lines: 37 On Thu 27-07-17 13:30:31, Aneesh Kumar K.V wrote: > > > On 07/27/2017 12:58 PM, Michal Hocko wrote: > >On Thu 27-07-17 07:52:08, Aneesh Kumar K.V wrote: > >>Michal Hocko writes: > >> > >>>Hi, > >>>I've just noticed that alloc_gigantic_page ignores movability of the > >>>gigantic page and it uses any existing zone. Considering that > >>>hugepage_migration_supported only supports 2MB and pgd level hugepages > >>>then 1GB pages are not migratable and as such allocating them from a > >>>movable zone will break the basic expectation of this zone. Standard > >>>hugetlb allocations try to avoid that by using htlb_alloc_mask and I > >>>believe we should do the same for gigantic pages as well. > >>> > >>>I suspect this behavior is not intentional. What do you think about the > >>>following untested patch? > >> > >> > >>I also noticed an unrelated issue with the usage of > >>start_isolate_page_range. On error we set the migrate type to > >>MIGRATE_MOVABLE. > > > >Why that should be a problem? I think it is perfectly OK to have > >MIGRATE_MOVABLE pageblocks inside kernel zones. > > > > we can pick pages with migrate type movable and if we fail to isolate won't > we set the migrate type of that pages to MOVABLE ? I do not see an immediate problem. GFP_KERNEL allocations can fallback to movable migrate pageblocks AFAIR. But I am not very much familiar with migratetypes. Vlastimil, could you have a look please? -- Michal Hocko SUSE Labs