From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Lameter Subject: Re: No, really, stop trying to delete slab until you've finished making slub perform as well Date: Mon, 04 Aug 2008 11:02:29 -0500 Message-ID: <48972815.6030108@linux-foundation.org> References: <20080801182324.572058187@lameter.com> <20080803015847.GD26461@parisc-linux.org> <48970779.80902@linux-foundation.org> <20080804111158.7f108047@cuia.bos.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Matthew Wilcox , Pekka Enberg , akpm@linux-foundation.org, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, Mel Gorman , andi@firstfloor.org To: Rik van Riel Return-path: Received: from smtp1.linux-foundation.org ([140.211.169.13]:47897 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753552AbYHDQDR (ORCPT ); Mon, 4 Aug 2008 12:03:17 -0400 In-Reply-To: <20080804111158.7f108047@cuia.bos.redhat.com> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Rik van Riel wrote: > I guess you could bypass the queueing on free for objects that > come from a "local" SLUB page, only queueing objects that go > onto remote pages. Tried that already. The logic to decide if an object is local is creating significant overhead. Plus you need queues for the remote nodes. Back to alien queues?