From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753105Ab0LGOVs (ORCPT ); Tue, 7 Dec 2010 09:21:48 -0500 Received: from hera.kernel.org ([140.211.167.34]:54369 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751853Ab0LGOVr (ORCPT ); Tue, 7 Dec 2010 09:21:47 -0500 Message-ID: <4CFE4294.30001@kernel.org> Date: Tue, 07 Dec 2010 15:20:04 +0100 From: Tejun Heo User-Agent: Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.9.2.12) Gecko/20101027 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: Christoph Lameter CC: akpm@linux-foundation.org, Yinghai Lu , Ingo Molnar , Pekka Enberg , linux-kernel@vger.kernel.org, Eric Dumazet , Mathieu Desnoyers Subject: Re: [Use cpuops V1 04/11] x86: Use this_cpu_ops for current_cpu_data accesses References: <20101206171618.302060721@linux.com> <20101206171638.595205962@linux.com> <4CFE41F8.7040400@kernel.org> In-Reply-To: <4CFE41F8.7040400@kernel.org> 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]); Tue, 07 Dec 2010 14:20:06 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 12/07/2010 03:17 PM, Tejun Heo wrote: > On 12/06/2010 06:16 PM, Christoph Lameter wrote: >> Current_cpu_data accesses are per cpu accesses. We can also use >> this_cpu_ops if a scalar is retrieved. >> >> Cc: Yinghai Lu >> Cc: Ingo Molnar >> Signed-off-by: Christoph Lameter >> >> --- >> arch/x86/kernel/cpu/cpufreq/powernow-k8.c | 2 +- >> arch/x86/kernel/cpu/intel_cacheinfo.c | 4 ++-- >> arch/x86/kernel/smpboot.c | 10 +++++----- >> 3 files changed, 8 insertions(+), 8 deletions(-) >> >> Index: linux-2.6/arch/x86/kernel/smpboot.c >> =================================================================== >> --- linux-2.6.orig/arch/x86/kernel/smpboot.c 2010-11-30 11:53:03.000000000 -0600 >> +++ linux-2.6/arch/x86/kernel/smpboot.c 2010-11-30 11:57:02.000000000 -0600 >> @@ -430,7 +430,7 @@ void __cpuinit set_cpu_sibling_map(int c >> >> cpumask_set_cpu(cpu, c->llc_shared_map); >> >> - if (current_cpu_data.x86_max_cores == 1) { >> + if (__this_cpu_read(cpu_info.x86_max_cores) == 1) { >> cpumask_copy(cpu_core_mask(cpu), cpu_sibling_mask(cpu)); >> c->booted_cores = 1; >> return; >> @@ -1377,7 +1377,7 @@ void play_dead_common(void) >> >> mb(); >> /* Ack it */ >> - __get_cpu_var(cpu_state) = CPU_DEAD; >> + __this_cpu_write(cpu_state, CPU_DEAD); > > This belongs to the previous patch, right? I'll move it over and > apply 03 and 04. I think routing these through percpu is okay but if > anyone wants these to go through x86, scream. Ooh, was too fast. I think mixing the use of percpu accesses to cpu_info and the wrapper macro current_cpu_data is quite confusing. There aren't too many current_cpu_data users in x86 anyway. Can you please make the conversion complete? I'm moving the above misplaced chunk into 03 and not applying 04 for now. Thank you. -- tejun