From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx187.postini.com [74.125.245.187]) by kanga.kvack.org (Postfix) with SMTP id 641B36B0031 for ; Fri, 13 Sep 2013 22:57:54 -0400 (EDT) Message-ID: <5233D09F.6040307@oracle.com> Date: Sat, 14 Sep 2013 10:57:35 +0800 From: Bob Liu MIME-Version: 1.0 Subject: Re: [PATCH 0/50] Basic scheduler support for automatic NUMA balancing V7 References: <1378805550-29949-1-git-send-email-mgorman@suse.de> In-Reply-To: <1378805550-29949-1-git-send-email-mgorman@suse.de> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Mel Gorman Cc: Peter Zijlstra , Rik van Riel , Srikar Dronamraju , Ingo Molnar , Andrea Arcangeli , Johannes Weiner , Linux-MM , LKML Hi Mel, On 09/10/2013 05:31 PM, Mel Gorman wrote: > It has been a long time since V6 of this series and time for an update. Much > of this is now stabilised with the most important addition being the inclusion > of Peter and Rik's work on grouping tasks that share pages together. > > This series has a number of goals. It reduces overhead of automatic balancing > through scan rate reduction and the avoidance of TLB flushes. It selects a > preferred node and moves tasks towards their memory as well as moving memory > toward their task. It handles shared pages and groups related tasks together. > I found sometimes numa balancing will be broken after khugepaged started, because khugepaged always allocate huge page from the node of the first scanned normal page during collapsing. A simple use case is when a user run his application interleaving all nodes using "numactl --interleave=all xxxx". But after khugepaged started most pages of his application will be located to only one specific node. I have a simple patch fix this issue in thread: [PATCH 2/2] mm: thp: khugepaged: add policy for finding target node I think this may related with this topic, I don't know whether this series can also fix the issue I mentioned. -- Regards, -Bob -- 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 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756403Ab3INC6D (ORCPT ); Fri, 13 Sep 2013 22:58:03 -0400 Received: from aserp1040.oracle.com ([141.146.126.69]:50812 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756179Ab3INC6B (ORCPT ); Fri, 13 Sep 2013 22:58:01 -0400 Message-ID: <5233D09F.6040307@oracle.com> Date: Sat, 14 Sep 2013 10:57:35 +0800 From: Bob Liu User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130308 Thunderbird/17.0.4 MIME-Version: 1.0 To: Mel Gorman CC: Peter Zijlstra , Rik van Riel , Srikar Dronamraju , Ingo Molnar , Andrea Arcangeli , Johannes Weiner , Linux-MM , LKML Subject: Re: [PATCH 0/50] Basic scheduler support for automatic NUMA balancing V7 References: <1378805550-29949-1-git-send-email-mgorman@suse.de> In-Reply-To: <1378805550-29949-1-git-send-email-mgorman@suse.de> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Source-IP: ucsinet22.oracle.com [156.151.31.94] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Mel, On 09/10/2013 05:31 PM, Mel Gorman wrote: > It has been a long time since V6 of this series and time for an update. Much > of this is now stabilised with the most important addition being the inclusion > of Peter and Rik's work on grouping tasks that share pages together. > > This series has a number of goals. It reduces overhead of automatic balancing > through scan rate reduction and the avoidance of TLB flushes. It selects a > preferred node and moves tasks towards their memory as well as moving memory > toward their task. It handles shared pages and groups related tasks together. > I found sometimes numa balancing will be broken after khugepaged started, because khugepaged always allocate huge page from the node of the first scanned normal page during collapsing. A simple use case is when a user run his application interleaving all nodes using "numactl --interleave=all xxxx". But after khugepaged started most pages of his application will be located to only one specific node. I have a simple patch fix this issue in thread: [PATCH 2/2] mm: thp: khugepaged: add policy for finding target node I think this may related with this topic, I don't know whether this series can also fix the issue I mentioned. -- Regards, -Bob