From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756089Ab0J0Pka (ORCPT ); Wed, 27 Oct 2010 11:40:30 -0400 Received: from hera.kernel.org ([140.211.167.34]:58611 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752503Ab0J0Pk2 (ORCPT ); Wed, 27 Oct 2010 11:40:28 -0400 Message-ID: <4CC847D9.3080805@kernel.org> Date: Wed, 27 Oct 2010 17:40:09 +0200 From: Tejun Heo User-Agent: Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.9.2.11) Gecko/20101013 Lightning/1.0b2 Thunderbird/3.1.5 MIME-Version: 1.0 To: Ingo Molnar CC: Linus Torvalds , Brian Gerst , x86@kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] x86-32: Allocate irq stacks seperate from percpu area References: <1288158182-1753-1-git-send-email-brgerst@gmail.com> <20101027153053.GA20243@elte.hu> <20101027153331.GA21789@elte.hu> In-Reply-To: <20101027153331.GA21789@elte.hu> X-Enigmail-Version: 1.1.1 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.3 (hera.kernel.org [127.0.0.1]); Wed, 27 Oct 2010 15:40:10 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello, Ingo, Linus. On 10/27/2010 05:33 PM, Ingo Molnar wrote: > Btw., the NUMA stuff never really worked percpu alloc, as per Eric's observation: Oh, it works in general. In this case, it's probably because the NUMA configuration is rather weird in 32bit (but then again NUMA on 32bit is supposed to be so). I think it's just 32bit init code setting up early_cpu_to_node() differently. > | Anyway, I just discovered per_cpu data on my machine (NUMA capable) all sit on a > | single node, if 32bit kernel used. > > ... and in practice it's not really relevant on 32-bit anyway. So we can decouple > the two issues just fine. Yeah, the two issues are separate and as I wrote in the other message the end result is correct (as much as it could be). I'll try to dig down what exactly is going on but I don't think there's anything to be alarmed about. Thanks. -- tejun