From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 25 Oct 2002 08:33:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 25 Oct 2002 08:33:54 -0400 Received: from zero.aec.at ([193.170.194.10]:12810 "EHLO zero.aec.at") by vger.kernel.org with ESMTP id ; Fri, 25 Oct 2002 08:33:53 -0400 Date: Fri, 25 Oct 2002 14:40:01 +0200 From: Andi Kleen To: Mikael Pettersson Cc: Andi Kleen , linux-kernel@vger.kernel.org Subject: Re: [PATCH] x86 performance counters driver 3.0-pre2 for 2.5.44: [2/4] x86 support Message-ID: <20021025124001.GA29937@averell> References: <200210241500.RAA03585@kim.it.uu.se> <15801.14413.909403.323948@kim.it.uu.se> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <15801.14413.909403.323948@kim.it.uu.se> User-Agent: Mutt/1.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > For what values of cpu is per_cpu(var,cpu) valid? For those where > cpu_online(cpu) is true, or those where cpu_possible(cpu) is true? > (I need to convert a memset() on the per_cpu_cache[] array to the > per_cpu(,) framework.) Currently for cpu_possible(), but there is a patchkit around that makes it only true for cpu_online() so better assume that. Of course that would require a hotplug CPU handler, but I would just ignore that for now. -Andi