Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934676AbdHYQoT (ORCPT ); Fri, 25 Aug 2017 12:44:19 -0400 Received: from mga06.intel.com ([134.134.136.31]:18364 "EHLO mga06.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934456AbdHYQoQ (ORCPT ); Fri, 25 Aug 2017 12:44:16 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.41,426,1498546800"; d="scan'208";a="1007655356" Subject: Re: [PATCH 1/2] sched/wait: Break up long wake list walk To: Mel Gorman , Linus Torvalds Cc: "Liang, Kan" , Mel Gorman , "Kirill A. Shutemov" , Peter Zijlstra , Ingo Molnar , Andi Kleen , Andrew Morton , Johannes Weiner , Jan Kara , linux-mm , Linux Kernel Mailing List References: <37D7C6CF3E00A74B8858931C1DB2F0775378A24A@SHSMSX103.ccr.corp.intel.com> <37D7C6CF3E00A74B8858931C1DB2F0775378A377@SHSMSX103.ccr.corp.intel.com> <37D7C6CF3E00A74B8858931C1DB2F0775378A8AB@SHSMSX103.ccr.corp.intel.com> <6e8b81de-e985-9222-29c5-594c6849c351@linux.intel.com> <85fb2a78-cbb7-dceb-12e8-7d18519c30a0@linux.intel.com> <20170824204448.if2mve3iy5k425di@techsingularity.net> From: Tim Chen Message-ID: <63454831-3259-c758-d164-3b2ff2a04b7e@linux.intel.com> Date: Fri, 25 Aug 2017 09:44:06 -0700 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.1.0 MIME-Version: 1.0 In-Reply-To: <20170824204448.if2mve3iy5k425di@techsingularity.net> Content-Type: text/plain; charset=iso-8859-15 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1061 Lines: 26 On 08/24/2017 01:44 PM, Mel Gorman wrote: > On Thu, Aug 24, 2017 at 11:16:15AM -0700, Linus Torvalds wrote: >> On Thu, Aug 24, 2017 at 10:49 AM, Tim Chen wrote: >>> >>> These changes look fine. We are testing them now. >>> Does the second patch in the series look okay to you? >> >> I didn't really have any reaction to that one, as long as Mel&co are >> ok with it, I'm fine with it. >> > > I've no strong objections or concerns. I'm disappointed that the > original root cause for this could not be found but hope that eventually a > reproducible test case will eventually be available. Despite having access > to a 4-socket box, I was still unable to create a workload that caused > large delays on wakeup. I'm going to have to stop as I don't think it's > possible to create on that particular machine for whatever reason. > Kan helped to test the updated patch 1 from Linus. It worked fine. I've refreshed the patch set that includes all the changes and send a version 2 refresh of the patch set separately. Thanks. Tim