From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx161.postini.com [74.125.245.161]) by kanga.kvack.org (Postfix) with SMTP id 0BBF56B002C for ; Thu, 1 Mar 2012 03:06:51 -0500 (EST) Received: by bkwq16 with SMTP id q16so286490bkw.14 for ; Thu, 01 Mar 2012 00:06:49 -0800 (PST) Message-ID: <4F4F2E16.5080703@openvz.org> Date: Thu, 01 Mar 2012 12:06:46 +0400 From: Konstantin Khlebnikov MIME-Version: 1.0 Subject: Re: [PATCH resend] mm: drain percpu lru add/rotate page-vectors on cpu hot-unplug References: <20120228193620.32063.83425.stgit@zurg> <20120229123818.61a61e9d.akpm@linux-foundation.org> In-Reply-To: <20120229123818.61a61e9d.akpm@linux-foundation.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Andrew Morton Cc: "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , KAMEZAWA Hiroyuki , Johannes Weiner , KOSAKI Motohiro Andrew Morton wrote: > On Tue, 28 Feb 2012 23:40:45 +0400 > Konstantin Khlebnikov wrote: > >> This cpu hotplug hook was accidentally removed in commit v2.6.30-rc4-18-g00a62ce >> ("mm: fix Committed_AS underflow on large NR_CPUS environment") > > That was a long time ago - maybe we should leave it removed ;) I mean, > did this bug(?) have any visible effect? If so, what was that effect? It's because cpu hotplug/unplug isn't widely used feature. Visible effect -- some pages are borrowed in per-cpu page-vectors. Truncate can deal with it, but these pages cannot be reused while this cpu is offline. So this is like temporary memory leak. > > IOW, the changelog didn't give anyone any reason to apply the patch to > anything! Sorry, I'm just stuck in pile of patches. It seems I should stop and send them one by one. This one isn't critical, so there no reasons for pushing it into stable branches. -- 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/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ 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 S1755043Ab2CAIGw (ORCPT ); Thu, 1 Mar 2012 03:06:52 -0500 Received: from mail-bk0-f46.google.com ([209.85.214.46]:40176 "EHLO mail-bk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751028Ab2CAIGv (ORCPT ); Thu, 1 Mar 2012 03:06:51 -0500 Authentication-Results: mr.google.com; spf=pass (google.com: domain of koct9i@gmail.com designates 10.204.145.70 as permitted sender) smtp.mail=koct9i@gmail.com; dkim=pass header.i=koct9i@gmail.com Message-ID: <4F4F2E16.5080703@openvz.org> Date: Thu, 01 Mar 2012 12:06:46 +0400 From: Konstantin Khlebnikov User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.19) Gecko/20120201 Iceape/2.0.14 MIME-Version: 1.0 To: Andrew Morton CC: "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , KAMEZAWA Hiroyuki , Johannes Weiner , KOSAKI Motohiro Subject: Re: [PATCH resend] mm: drain percpu lru add/rotate page-vectors on cpu hot-unplug References: <20120228193620.32063.83425.stgit@zurg> <20120229123818.61a61e9d.akpm@linux-foundation.org> In-Reply-To: <20120229123818.61a61e9d.akpm@linux-foundation.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Andrew Morton wrote: > On Tue, 28 Feb 2012 23:40:45 +0400 > Konstantin Khlebnikov wrote: > >> This cpu hotplug hook was accidentally removed in commit v2.6.30-rc4-18-g00a62ce >> ("mm: fix Committed_AS underflow on large NR_CPUS environment") > > That was a long time ago - maybe we should leave it removed ;) I mean, > did this bug(?) have any visible effect? If so, what was that effect? It's because cpu hotplug/unplug isn't widely used feature. Visible effect -- some pages are borrowed in per-cpu page-vectors. Truncate can deal with it, but these pages cannot be reused while this cpu is offline. So this is like temporary memory leak. > > IOW, the changelog didn't give anyone any reason to apply the patch to > anything! Sorry, I'm just stuck in pile of patches. It seems I should stop and send them one by one. This one isn't critical, so there no reasons for pushing it into stable branches.