From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752801Ab1GUIoL (ORCPT ); Thu, 21 Jul 2011 04:44:11 -0400 Received: from mail-ww0-f44.google.com ([74.125.82.44]:63705 "EHLO mail-ww0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752679Ab1GUIoI (ORCPT ); Thu, 21 Jul 2011 04:44:08 -0400 Subject: Re: [PATCH] mm-slab: allocate kmem_cache with __GFP_REPEAT From: Eric Dumazet To: Christoph Lameter Cc: Mel Gorman , Pekka Enberg , Konstantin Khlebnikov , Andrew Morton , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Matt Mackall In-Reply-To: References: <20110720121612.28888.38970.stgit@localhost6> <20110720134342.GK5349@suse.de> <1311170893.2338.29.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Content-Type: text/plain; charset="UTF-8" Date: Thu, 21 Jul 2011 10:43:59 +0200 Message-ID: <1311237839.2422.10.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Mime-Version: 1.0 X-Mailer: Evolution 2.32.2 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Le mercredi 20 juillet 2011 à 09:52 -0500, Christoph Lameter a écrit : > We should be making it a per cpu pointer like slub then. I looked at what > it would take to do so a couple of month ago but it was quite invasive. > I took a look at this too, but using percpu data would consume more memory, because percpu allocator allocates memory blobs for all possible cpus, while current code handles online/offline cpu nicely. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail172.messagelabs.com (mail172.messagelabs.com [216.82.254.3]) by kanga.kvack.org (Postfix) with ESMTP id EFB2C6B004A for ; Thu, 21 Jul 2011 04:44:09 -0400 (EDT) Received: by wyg36 with SMTP id 36so899776wyg.14 for ; Thu, 21 Jul 2011 01:44:07 -0700 (PDT) Subject: Re: [PATCH] mm-slab: allocate kmem_cache with __GFP_REPEAT From: Eric Dumazet In-Reply-To: References: <20110720121612.28888.38970.stgit@localhost6> <20110720134342.GK5349@suse.de> <1311170893.2338.29.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Content-Type: text/plain; charset="UTF-8" Date: Thu, 21 Jul 2011 10:43:59 +0200 Message-ID: <1311237839.2422.10.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Mime-Version: 1.0 Content-Transfer-Encoding: 8bit Sender: owner-linux-mm@kvack.org List-ID: To: Christoph Lameter Cc: Mel Gorman , Pekka Enberg , Konstantin Khlebnikov , Andrew Morton , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Matt Mackall Le mercredi 20 juillet 2011 A 09:52 -0500, Christoph Lameter a A(C)crit : > We should be making it a per cpu pointer like slub then. I looked at what > it would take to do so a couple of month ago but it was quite invasive. > I took a look at this too, but using percpu data would consume more memory, because percpu allocator allocates memory blobs for all possible cpus, while current code handles online/offline cpu nicely. -- 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