From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934538AbbCPPjv (ORCPT ); Mon, 16 Mar 2015 11:39:51 -0400 Received: from resqmta-ch2-12v.sys.comcast.net ([69.252.207.44]:33754 "EHLO resqmta-ch2-12v.sys.comcast.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932942AbbCPPjp (ORCPT ); Mon, 16 Mar 2015 11:39:45 -0400 Date: Mon, 16 Mar 2015 10:39:42 -0500 (CDT) From: Christoph Lameter X-X-Sender: cl@gentwo.org To: David Rientjes cc: Eric B Munson , Rik van Riel , Andrew Morton , Vlastimil Babka , Thomas Gleixner , Peter Zijlstra , Mel Gorman , Michal Hocko , linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH V5] Allow compaction of unevictable pages In-Reply-To: Message-ID: References: <1426267597-25811-1-git-send-email-emunson@akamai.com> <550332CE.7040404@redhat.com> <20150313190915.GA12589@akamai.com> Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 13 Mar 2015, David Rientjes wrote: > It would be really disappointing to not enable this by default for !rt > kernels. We haven't migrated mlocked pages in the past by way of memory > compaction because it can theoretically result in consistent minor page > faults, but I haven't yet heard a !rt objection to enabling this. > > If the rt patchset is going to carry a patch to disable this, then the > question arises: why not just carry an ISOLATE_UNEVICTABLE patch instead? > I think you've done the due diligence required to allow this to be > disabled at any time in a very easy way from userspace by the new tunable. > I think it should be enabled and I'd be very surprised to hear any other > objection about it other than it's different from the status quo. Compaction can alrady be disabled and thus you can also disable migration of mlocked pages. In general low latency requires that no expensive kernel processing is being done. Thus the rest of compaction processing also needs to be disabled. That means that allowing compaction handling mlocked pages would be ok. RT loads and low latency configurations (like my environment) will selective disable compaction to avoid creating additional latencies. This could be done only for specific nodes and processors if necessary.