From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753760AbYEQFeL (ORCPT ); Sat, 17 May 2008 01:34:11 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750937AbYEQFeA (ORCPT ); Sat, 17 May 2008 01:34:00 -0400 Received: from ozlabs.org ([203.10.76.45]:54737 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750769AbYEQFd7 convert rfc822-to-8bit (ORCPT ); Sat, 17 May 2008 01:33:59 -0400 From: Rusty Russell To: Mike Travis Subject: Re: [PATCH] modules: Use a better scheme for refcounting Date: Sat, 17 May 2008 15:33:46 +1000 User-Agent: KMail/1.9.9 Cc: Eric Dumazet , Andrew Morton , linux kernel , Christoph Lameter References: <482C9FC5.2070508@cosmosbay.com> <482D1BCE.3060501@cosmosbay.com> <482D8EFC.8040109@sgi.com> In-Reply-To: <482D8EFC.8040109@sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8BIT Content-Disposition: inline Message-Id: <200805171533.46295.rusty@rustcorp.com.au> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Friday 16 May 2008 23:41:16 Mike Travis wrote: > Eric Dumazet wrote: > > Rusty Russell a écrit : > >> Any chance I can ask you look at the issue of full dynamic per-cpu > >> allocation? > > > > You mean using alloc_percpu() ? Problem is that current implementation > > is expensive, I mean rewriting alloc_percpu :) > > We probably can change this to dynamic per-cpu as soon as Mike or > > Christopher finish their work on new dynamic per-cpu implementation ? > > Yes, the zero-based percpu variables followed by the cpu_alloc patch should > provide this and shrink the code quite well, including in some cases > removing locking requirements (because the resultant instructions will be > atomic.) Ah, I hadn't realized that Mike was already working on this. Mike, have you published patches already? Thanks, Rusty.