From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932130AbWDJU2E (ORCPT ); Mon, 10 Apr 2006 16:28:04 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932133AbWDJU2E (ORCPT ); Mon, 10 Apr 2006 16:28:04 -0400 Received: from ms-smtp-02.texas.rr.com ([24.93.47.41]:29896 "EHLO ms-smtp-02.texas.rr.com") by vger.kernel.org with ESMTP id S932130AbWDJU2B (ORCPT ); Mon, 10 Apr 2006 16:28:01 -0400 Date: Mon, 10 Apr 2006 15:27:48 -0500 From: Dave McCracken To: Christoph Lameter cc: Hugh Dickins , Linux Kernel Mailing List , Linux Memory Management , Adam Litke , wli@holomorphy.com Subject: Re: [RFC/PATCH] Shared Page Tables [0/2] Message-ID: <1083771BBA2E79327C70D039@[10.1.1.4]> In-Reply-To: References: <1144685588.570.35.camel@wildcat.int.mccr.org> <200ED4FEFEB8AA8427120DE7@[10.1.1.4]> X-Mailer: Mulberry/4.0.0b4 (Linux/x86) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --On Monday, April 10, 2006 13:20:59 -0700 Christoph Lameter wrote: >> The lock changes to hugetlb are only to support sharing of pmd pages when >> they contain hugetlb pages. They just substitute the struct page lock >> for the page_table_lock, and are only about 30 lines of code. Is this >> really worth separating out? > > Ia64 does not use pmd pages for huge pages. It relies instead on a > separate region. I wonder if this works on IA64. Sharing of hugetlb page tables is enabled on a per-architecture basis, so if ia64 doesn't use pmd pages we shouldn't try to enable it. If it's not enabled all the locking in hugetlb resolves to using page_table_lock, so the original semantics will be preserved. Dave McCracken