Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751414AbeAEJUV (ORCPT + 1 other); Fri, 5 Jan 2018 04:20:21 -0500 Received: from mx0a-001b2d01.pphosted.com ([148.163.156.1]:59144 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751317AbeAEJUT (ORCPT ); Fri, 5 Jan 2018 04:20:19 -0500 Subject: Re: [PATCH 1/3] mm, numa: rework do_pages_move To: Michal Hocko , Anshuman Khandual References: <20180103082555.14592-1-mhocko@kernel.org> <20180103082555.14592-2-mhocko@kernel.org> <20180105091443.GJ2801@dhcp22.suse.cz> Cc: Andrew Morton , Zi Yan , Naoya Horiguchi , "Kirill A. Shutemov" , Vlastimil Babka , Andrea Reale , linux-mm@kvack.org, LKML From: Anshuman Khandual Date: Fri, 5 Jan 2018 14:50:04 +0530 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.5.1 MIME-Version: 1.0 In-Reply-To: <20180105091443.GJ2801@dhcp22.suse.cz> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-TM-AS-GCONF: 00 x-cbid: 18010509-0040-0000-0000-00000420E740 X-IBM-AV-DETECTION: SAVI=unused REMOTE=unused XFE=unused x-cbparentid: 18010509-0041-0000-0000-000020C4322C Message-Id: X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:,, definitions=2018-01-05_04:,, signatures=0 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 priorityscore=1501 malwarescore=0 suspectscore=0 phishscore=0 bulkscore=0 spamscore=0 clxscore=1015 lowpriorityscore=0 impostorscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1709140000 definitions=main-1801050131 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Return-Path: On 01/05/2018 02:44 PM, Michal Hocko wrote: > On Fri 05-01-18 09:22:22, Anshuman Khandual wrote: > [...] >> Hi Michal, >> >> After slightly modifying your test case (like fixing the page size for >> powerpc and just doing simple migration from node 0 to 8 instead of the >> interleaving), I tried to measure the migration speed with and without >> the patches on mainline. Its interesting.... >> >> 10000 pages | 100000 pages >> -------------------------- >> Mainline 165 ms 1674 ms >> Mainline + first patch (move_pages) 191 ms 1952 ms >> Mainline + all three patches 146 ms 1469 ms >> >> Though overall it gives performance improvement, some how it slows >> down migration after the first patch. Will look into this further. > > What are you measuring actually? All pages migrated to the same node? The mount of time move_pages() system call took to move these many pages from node 0 to node 8. Yeah they migrated to the same node. > Do you have any profiles? How stable are the results? No, are you referring to perf record kind profile ? Results were repeating.