From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932265AbWHaAeB (ORCPT ); Wed, 30 Aug 2006 20:34:01 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751429AbWHaAeA (ORCPT ); Wed, 30 Aug 2006 20:34:00 -0400 Received: from ozlabs.tip.net.au ([203.10.76.45]:4275 "EHLO ozlabs.org") by vger.kernel.org with ESMTP id S1750774AbWHaAd7 (ORCPT ); Wed, 30 Aug 2006 20:33:59 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <17654.11889.933070.539839@cargo.ozlabs.ibm.com> Date: Thu, 31 Aug 2006 10:33:53 +1000 From: Paul Mackerras To: Dave Hansen Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org Subject: Re: [RFC][PATCH 0/9] generic PAGE_SIZE infrastructure (v4) In-Reply-To: <20060830221604.E7320C0F@localhost.localdomain> References: <20060830221604.E7320C0F@localhost.localdomain> X-Mailer: VM 7.19 under Emacs 21.4.1 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Dave Hansen writes: > Why am I doing this? The OpenVZ beancounter patch hooks into the > alloc_thread_info() path, but only in two architectures. It is silly > to patch each and every architecture when they all just do the same > thing. This is the first step to have a single place in which to > do alloc_thread_info(). Oh, and this series removes about 300 lines > of code. ... at the price of making the Kconfig help text more generic and therefore possibly confusing on some platforms. I really don't see much value in doing all this. Paul.