Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932913AbbLOAEQ (ORCPT ); Mon, 14 Dec 2015 19:04:16 -0500 Received: from mail-qk0-f181.google.com ([209.85.220.181]:35035 "EHLO mail-qk0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932197AbbLOAEP (ORCPT ); Mon, 14 Dec 2015 19:04:15 -0500 MIME-Version: 1.0 In-Reply-To: <1449931045.7411.8.camel@gmail.com> References: <1449903863.9638.8.camel@gmail.com> <1449931045.7411.8.camel@gmail.com> Date: Tue, 15 Dec 2015 01:04:14 +0100 Message-ID: Subject: Re: sched : performance regression 24% between 4.4rc4 and 4.3 kernel From: Jirka Hladky To: Mike Galbraith Cc: linux-kernel@vger.kernel.org, Peter Zijlstra , Ingo Molnar , Rik van Riel Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2749 Lines: 66 Hi all, I have the results of bisecting: first bad commit: [973759c80db96ed4b4c5cb85ac7d48107f801371] Merge tag 'v4.3-rc1' into sched/core, to refresh the branch Could you please have a look at this commit why it has caused the performance regression when running 4 stream benchmarks in parallel on 4 NUMA node server? Please let me know if you need additional data. git bisect log is bellow. Thanks a lot! Jirka $ git bisect log git bisect start '--' 'kernel/sched' # good: [6a13feb9c82803e2b815eca72fa7a9f5561d7861] Linux 4.3 git bisect good 6a13feb9c82803e2b815eca72fa7a9f5561d7861 # bad: [527e9316f8ec44bd53d90fb9f611fa7ffff52bb9] Linux 4.4-rc4 git bisect bad 527e9316f8ec44bd53d90fb9f611fa7ffff52bb9 # bad: [b99def8b961448f5b9a550dddeeb718e3975e7a6] sched/core: Rework TASK_DEAD preemption exception git bisect bad b99def8b961448f5b9a550dddeeb718e3975e7a6 # skip: [8cd5601c50603caa195ce86cc465cb04079ed488] sched/fair: Convert arch_scale_cpu_capacity() from weak function to #define git bisect skip 8cd5601c50603caa195ce86cc465cb04079ed488 # bad: [fe19159225d8516f3f57a5fe8f735c01684f0ddd] Merge branch 'sched/urgent' into sched/core, to pick up fixes before applying new changes git bisect bad fe19159225d8516f3f57a5fe8f735c01684f0ddd # bad: [78a9c54649ea220065aad9902460a1d137c7eafd] sched/numa: Rename numabalancing_enabled to sched_numa_balancing git bisect bad 78a9c54649ea220065aad9902460a1d137c7eafd # bad: [6efdb105d392da3ad5cb4ef951aed373cd049813] sched/fair: Fix switched_to_fair()'s per entity load tracking git bisect bad 6efdb105d392da3ad5cb4ef951aed373cd049813 # bad: [50a2a3b246149d041065a67ccb3e98145f780a2f] sched/fair: Have task_move_group_fair() unconditionally add the entity load to the runqueue git bisect bad 50a2a3b246149d041065a67ccb3e98145f780a2f # bad: [973759c80db96ed4b4c5cb85ac7d48107f801371] Merge tag 'v4.3-rc1' into sched/core, to refresh the branch git bisect bad 973759c80db96ed4b4c5cb85ac7d48107f801371 # first bad commit: [973759c80db96ed4b4c5cb85ac7d48107f801371] Merge tag 'v4.3-rc1' into sched/core, to refresh the branch On Sat, Dec 12, 2015 at 3:37 PM, Mike Galbraith wrote: > On Sat, 2015-12-12 at 15:16 +0100, Jirka Hladky wrote: >> > A bisection doesn't require any special skills, but may give busy >> > maintainers a single change to eyeball vs the entire lot. >> >> They have been couple of merges which makes git revert difficult... > > You could try https://git-scm.com/docs/git-bisect instead. > > -Mike -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/