Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755994Ab3FEOGu (ORCPT ); Wed, 5 Jun 2013 10:06:50 -0400 Received: from a9-62.smtp-out.amazonses.com ([54.240.9.62]:58680 "EHLO a9-62.smtp-out.amazonses.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755902Ab3FEOGt (ORCPT ); Wed, 5 Jun 2013 10:06:49 -0400 Date: Wed, 5 Jun 2013 14:06:47 +0000 From: Christoph Lameter X-X-Sender: cl@gentwo.org To: JoonSoo Kim cc: Steven Rostedt , Joonsoo Kim , LKML , RT , Thomas Gleixner , Clark Williams , Pekka Enberg Subject: Re: [RT LATENCY] 249 microsecond latency caused by slub's unfreeze_partials() code. In-Reply-To: Message-ID: <0000013f14a94e72-7933a6c6-c0db-40d1-b7ec-dba19e7393b5-000000@email.amazonses.com> References: <1364010673.6345.156.camel@gandalf.local.home> <0000013da2ace21a-9e87fe8a-75c2-4b7c-b5e1-37ad196ce012-000000@email.amazonses.com> <1364234613.6345.184.camel@gandalf.local.home> <0000013da2ce20f8-0e3a64ef-67ed-4ab4-9f20-b77980c876c3-000000@email.amazonses.com> <1364236355.6345.185.camel@gandalf.local.home> <20130327025957.GA17125@lge.com> <1364355032.6345.200.camel@gandalf.local.home> <20130327061351.GB17125@lge.com> <0000013db20ca149-0064fbb8-2f81-4323-9095-a38f6abb79c5-000000@email.amazonses.com> <1369751967.15552.12.camel@gandalf.local.home> <0000013eec62261a-77052f34-f1cb-41b2-ae90-7575a2d0472c-000000@email.amazonses.com> <0000013f0b7b993c-35844ed2-81be-4146-ab3c-6c803006576b-000000@email.amazonses.com> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SES-Outgoing: 2013.06.05-54.240.9.62 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 722 Lines: 17 On Wed, 5 Jun 2013, JoonSoo Kim wrote: > I knew that we need the other nodes pages if we consistently allocate > from there. > In allocation path, it is reasonable. > But, in free path, it is not reasonable that we put other nodes' page > to cpu partial list. It is reasonable. An "other nodes page" on the cpu partial list will decrease the synchronization necessary to free other objects to that page. And multiple objects from a remote node are typically freed in on go. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/