From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752325Ab0KEI2U (ORCPT ); Fri, 5 Nov 2010 04:28:20 -0400 Received: from hera.kernel.org ([140.211.167.34]:54979 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752022Ab0KEI2R (ORCPT ); Fri, 5 Nov 2010 04:28:17 -0400 Message-ID: <4CD3BFEC.4080405@kernel.org> Date: Fri, 05 Nov 2010 09:27:24 +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: Eric Dumazet CC: Andrew Morton , linux-kernel , Christoph Lameter , Ingo Molnar , Andi Kleen , Thomas Gleixner Subject: Re: [PATCH] irq: use per_cpu kstat_irqs References: <1288892908.2659.175.camel@edumazet-laptop> In-Reply-To: <1288892908.2659.175.camel@edumazet-laptop> X-Enigmail-Version: 1.1.1 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.3 (hera.kernel.org [127.0.0.1]); Fri, 05 Nov 2010 08:27:27 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 11/04/2010 06:48 PM, Eric Dumazet wrote: > Use modern per_cpu API to increment {soft|hard}irq counters, and > use per_cpu allocation for (struct irq_desc)->kstats_irq instead of an > array. > > This gives better SMP/NUMA locality and saves few instructions per irq. > > With small nr_cpuids values (8 for example), kstats_irq was a small > array (less than L1_CACHE_BYTES), potentially source of false sharing. > > Signed-off-by: Eric Dumazet > Cc: Christoph Lameter > Cc: Ingo Molnar > Cc: Andi Kleen > Cc: Tejun Heo > Cc: Thomas Gleixner Looks good to me although it would be nice to note the white space fixes in the comment. Reviewed-by: Tejun Heo Thanks. -- tejun