From mboxrd@z Thu Jan 1 00:00:00 1970 From: "KOSAKI Motohiro" Subject: Re: No, really, stop trying to delete slab until you've finished making slub perform as well Date: Tue, 5 Aug 2008 01:47:42 +0900 Message-ID: <2f11576a0808040947r69076eecv9ff92ecf583f7af2@mail.gmail.com> References: <20080801182324.572058187@lameter.com> <20080803015847.GD26461@parisc-linux.org> <48970779.80902@linux-foundation.org> 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, "Rik van Riel" , kosaki.motohiro@jp.fujitsu.com To: "Christoph Lameter" Return-path: Received: from py-out-1112.google.com ([64.233.166.178]:21756 "EHLO py-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753206AbYHDQro (ORCPT ); Mon, 4 Aug 2008 12:47:44 -0400 Received: by py-out-1112.google.com with SMTP id p76so1102638pyb.10 for ; Mon, 04 Aug 2008 09:47:43 -0700 (PDT) In-Reply-To: <48970779.80902@linux-foundation.org> Content-Disposition: inline Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Hi > Could you address the performance issues in different ways? F.e. try to free > when the object is hot or free from multiple processors? SLAB has to take the > list_lock rather frequently under high concurrent loads (depends on queue > size). That will not occur with SLUB. So you actually can free (and allocate) > concurrently with high performance. just information. (offtopic?) When hackbench running, SLUB consume memory very largely than SLAB. then, SLAB often outperform SLUB in memory stavation state. I don't know why memory comsumption different. Anyone know it?