Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S939240AbZDJSor (ORCPT ); Fri, 10 Apr 2009 14:44:47 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1762038AbZDJSoh (ORCPT ); Fri, 10 Apr 2009 14:44:37 -0400 Received: from hera.kernel.org ([140.211.167.34]:39270 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752434AbZDJSog (ORCPT ); Fri, 10 Apr 2009 14:44:36 -0400 Message-ID: <49DF935E.8030607@kernel.org> Date: Fri, 10 Apr 2009 11:43:42 -0700 From: Tejun Heo User-Agent: Thunderbird 2.0.0.19 (X11/20081227) MIME-Version: 1.0 To: Linus Torvalds CC: Ivan Kokshaysky , Al Viro , Ingo Molnar , Mike Travis , Christoph Lameter , Heiko Carstens , Martin Schwidefsky , Linux Kernel Mailing List Subject: Re: [PATCH] unbreak alpha percpu References: <20090410151224.GM26366@ZenIV.linux.org.uk> <20090410154704.GN26366@ZenIV.linux.org.uk> <20090410165030.GQ26366@ZenIV.linux.org.uk> <20090410170507.GR26366@ZenIV.linux.org.uk> <20090410171415.GA6033@ZenIV.linux.org.uk> <20090410180532.GA4419@jurassic.park.msu.ru> In-Reply-To: X-Enigmail-Version: 0.95.7 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.0 (hera.kernel.org [127.0.0.1]); Fri, 10 Apr 2009 18:43:02 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1224 Lines: 31 Linus Torvalds wrote: > > On Fri, 10 Apr 2009, Ivan Kokshaysky wrote: >> Actually, all of this has been discussed on lkml; here is the latest >> variant that makes everybody more or less happy, at least there are >> no objections from percpu folks and Martin (similar fix should work >> for s390 as well). > > Is there any reason why this version of DEFINE_PER_CPU_SECTION wouldn't > work on _any_ architecture? IOW, do we even need the #ifdef's and > per-arch #define? It should work for all archs but only alpha and s390 require __weak__ and other archs can use actual static or global definitions, but then again we'll need to add the __per_cpu_multiple_def_ thing to make sure no two static definitions clash anyway, so there isn't much point in keeping things separate. I was waiting for responses on the original thread. Is everyone okay with having the 'static per-cpu variables in different compile units can't have the same name' restriction? Thanks. -- tejun -- 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/