From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-f71.google.com (mail-wm0-f71.google.com [74.125.82.71]) by kanga.kvack.org (Postfix) with ESMTP id 727216B0033 for ; Tue, 17 Jan 2017 03:13:12 -0500 (EST) Received: by mail-wm0-f71.google.com with SMTP id d140so31968182wmd.4 for ; Tue, 17 Jan 2017 00:13:12 -0800 (PST) Received: from mx2.suse.de (mx2.suse.de. [195.135.220.15]) by mx.google.com with ESMTPS id t7si24195304wrb.5.2017.01.17.00.13.10 for (version=TLS1 cipher=AES128-SHA bits=128/128); Tue, 17 Jan 2017 00:13:11 -0800 (PST) Subject: Re: getting oom/stalls for ltp test cpuset01 with latest/4.9 kernel References: <075075cc-3149-0df3-dd45-a81df1f1a506@suse.cz> <0ea1cfeb-7c4a-3a3e-9be9-967298ba303c@suse.cz> <20170111164616.GJ16365@dhcp22.suse.cz> <45ed555a-c6a3-fc8e-1e87-c347c8ed086b@suse.cz> <89fec1bd-52b7-7861-2e02-a719c5631610@suse.cz> From: Vlastimil Babka Message-ID: <7c8ed170-ced9-aa6a-25e0-47cdc6d66eb2@suse.cz> Date: Tue, 17 Jan 2017 09:13:09 +0100 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Ganapatrao Kulkarni Cc: Michal Hocko , linux-mm@kvack.org, "linux-kernel@vger.kernel.org" , Mel Gorman On 01/16/2017 02:22 PM, Vlastimil Babka wrote: >>> >>> no_zone: >>> -- >>> 2.11.0 >>> >> >> this patch did not fix the issue. >> issue still exists! > > Hmm, that's unfortunate. > >> i did bisect and this test passes in 4.4,4.5 and 4.6 >> test failing since 4.7-rc1 > > 4.7 would match the commit I was trying to fix. But I don't see other > problems now. Could you bisect to a single commit then, to be sure? Thanks. Ah, nevermind, I can reproduce the issue easily. After some more poking I now think the bisect would lead to the OOM rework, but it would be a red herring. Seems like this is an interaction between bind mempolicy and cpuset and I see several potential bugs in that area. Which also means there's a non-null nodemask and thus the code in the commit I originally suspected (replacing NULL nodemask with cpuset's mems_allowed) doesn't trigger at all here. Thanks, Vlastimil >> thanks >> Ganapat >>> >>> >>> > > -- > 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/ . > Don't email: email@kvack.org > -- 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/ . Don't email: email@kvack.org