Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754533AbaJHHKR (ORCPT ); Wed, 8 Oct 2014 03:10:17 -0400 Received: from resqmta-po-03v.sys.comcast.net ([96.114.154.162]:39432 "EHLO resqmta-po-03v.sys.comcast.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753868AbaJHHKP (ORCPT ); Wed, 8 Oct 2014 03:10:15 -0400 Date: Wed, 8 Oct 2014 02:10:13 -0500 (CDT) From: Christoph Lameter X-X-Sender: cl@gentwo.org To: Dave Hansen cc: Sasha Levin , akpm@linux-foundation.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, hughd@google.com, mgorman@suse.de Subject: Re: [PATCH 5/5] mm: poison page struct In-Reply-To: <5434630C.3070006@intel.com> Message-ID: References: <1412041639-23617-1-git-send-email-sasha.levin@oracle.com> <1412041639-23617-6-git-send-email-sasha.levin@oracle.com> <5434630C.3070006@intel.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 Tue, 7 Oct 2014, Dave Hansen wrote: > Does this break slub's __cmpxchg_double_slab trick? I thought it > required page->freelist and page->counters to be doubleword-aligned. Sure that would be required for it to work. > It's not like we really require this optimization when we're debugging, > but trying to use it will unnecessarily slow things down. Debugging by inserting more data into the page struct will already cause a significant slow down because the cache footprint of key functions will increase significantly. I would think that using the fallback functions is reasonable in this scenario, -- 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/