From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753062Ab3AXHrb (ORCPT ); Thu, 24 Jan 2013 02:47:31 -0500 Received: from moutng.kundenserver.de ([212.227.126.187]:55384 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752933Ab3AXHrV (ORCPT ); Thu, 24 Jan 2013 02:47:21 -0500 Message-ID: <1359013623.5833.52.camel@marge.simpson.net> Subject: Re: [RFC PATCH 0/2] sched: simplify the select_task_rq_fair() From: Mike Galbraith To: Michael Wang Cc: linux-kernel@vger.kernel.org, mingo@redhat.com, peterz@infradead.org, mingo@kernel.org, a.p.zijlstra@chello.nl Date: Thu, 24 Jan 2013 08:47:03 +0100 In-Reply-To: <5100DF95.3020909@linux.vnet.ibm.com> References: <1356588535-23251-1-git-send-email-wangyun@linux.vnet.ibm.com> <50FCF212.3010504@linux.vnet.ibm.com> <1358759355.4994.108.camel@marge.simpson.net> <50FD08E1.8000302@linux.vnet.ibm.com> <1358761496.4994.118.camel@marge.simpson.net> <50FE0ADC.6060701@linux.vnet.ibm.com> <1358841795.5782.255.camel@marge.simpson.net> <50FE5433.1070801@linux.vnet.ibm.com> <1358865692.5782.420.camel@marge.simpson.net> <50FF4EA0.1070000@linux.vnet.ibm.com> <1358915494.5752.46.camel@marge.simpson.net> <50FF7086.4020509@linux.vnet.ibm.com> <1358922520.5752.91.camel@marge.simpson.net> <50FF8CD8.4060105@linux.vnet.ibm.com> <1358929257.5752.109.camel@marge.simp! son.net> <50FF9F92.60202@linux.vnet.ibm.com> <1358930968.5752.123.camel@marge.simpson.net> <50FFA695.6010407@linux.vnet.ibm.com> <1358932694.5752.126.camel@marge.simpson.net> <1358933555.5752.132.camel@marge.simpson.net> <5100CE1F.7080704@linux.vnet.ibm.com> <1359010272.5833.39.camel@marge.simpson.net> <5100DF95.3020909@linux.vnet.ibm.com> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.2.3 Content-Transfer-Encoding: 7bit Mime-Version: 1.0 X-Provags-ID: V02:K0:2OhnS3MKkVLhfSnK2rb5GDASMkadrBJnrKzKGCJWjC/ uyOPs816b432vFaSHJ4wM2yUSOXWop4HJPwuqboXPzNK3BAgUR N3vZOeni7dL15DRDnEFNVW44SZD4w4XoKoJxDuyUX2EeCpQVbo CS//ahZ181zQ9q4R5HZbCLe/cXvsse8uSUromdXQIrgKa0rNux +1wandEwE8opJTVuvx7HlQjLZ2o0Ew2RduluPvZbSUtY7w27em Yu6GFI65krN88O6Pi0tt8MzJYQCUfXW6WXmHYKb3vgTwabtNWL Yofggw9bFZy+mDRRXDVqAh0Acg00lQr4hoT/eyVzUzxGewz825 0UvGxr/bDQZfCBvheux4781TfRcaqcF8oulCbp37hPB/dlu/aS xfoOvbBR2P5fA== Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 2013-01-24 at 15:15 +0800, Michael Wang wrote: > On 01/24/2013 02:51 PM, Mike Galbraith wrote: > > On Thu, 2013-01-24 at 14:01 +0800, Michael Wang wrote: > > > >> I've enabled WAKE flag on my box like you did, but still can't see > >> regression, and I've just tested on a power server with 64 cpu, also > >> failed to reproduce the issue (not compared with virgin yet, but can't > >> see collapse). > > > > I'm not surprised. I'm seeing enough inconsistent crap to come to the > > conclusion that stock scheduler knobs flat can't be used on a largish > > box, they're just too preempt-happy, leading to weird crap. > > > > My 2 missing nodes came back, and the very same kernel that highly > > repeatably collapsed with 2 nodes does not with 4 nodes, and 2 nodes > > does not collapse with only preemption knob tweaking, and that's > > bullshit. Virgin shows instability in the mid-range, make a tiny tweak > > that should have little if any effect there, and that instability > > vanishes entirely. Test runs are not consistent enough boot to boot etc > > etc. Either stock knobs suck on NUMA boxen, or this box is possessed. > > Mike, I wonder the reason why change back to the old way make collapse > away may not because there are logical error in new balance path, it's > just changed the cost of select_task_rq(), whatever it's more or less, > it's accidentally achieve the same effect as you tweak the knob, so > that's the reason why it looks like old is better than new. That's what I'm saying, it's a useless crap side-effect of a preempt happy kernel. Results with these knobs are just not stable. Results go wildly unstable with 2 nodes vs 4 in this box, but can be stabilized in all with preemption knob adjustment.. or phase of moon might make them appear stable.. or not. -Mike