From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S268051AbUHSDlh (ORCPT ); Wed, 18 Aug 2004 23:41:37 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S268042AbUHSDlh (ORCPT ); Wed, 18 Aug 2004 23:41:37 -0400 Received: from holomorphy.com ([207.189.100.168]:52924 "EHLO holomorphy.com") by vger.kernel.org with ESMTP id S268025AbUHSDlf (ORCPT ); Wed, 18 Aug 2004 23:41:35 -0400 Date: Wed, 18 Aug 2004 20:41:09 -0700 From: William Lee Irwin III To: Rajesh Venkatasubramanian , Hugh Dickins , "David S. Miller" , raybry@sgi.com, ak@muc.de, benh@kernel.crashing.org, manfred@colorfullife.com, linux-ia64@vger.kernel.org, LKML Subject: Re: page fault fastpath patch v2: fix race conditions, stats for 8,32 and 512 cpu SMP Message-ID: <20040819034109.GC11200@holomorphy.com> Mail-Followup-To: William Lee Irwin III , Rajesh Venkatasubramanian , Hugh Dickins , "David S. Miller" , raybry@sgi.com, ak@muc.de, benh@kernel.crashing.org, manfred@colorfullife.com, linux-ia64@vger.kernel.org, LKML References: <2uexw-1Nn-1@gated-at.bofh.it> <2uCTq-2wa-55@gated-at.bofh.it> <20040819000151.GU11200@holomorphy.com> <20040819002038.GW11200@holomorphy.com> <20040819033127.GB11200@holomorphy.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040819033127.GB11200@holomorphy.com> User-Agent: Mutt/1.5.6+20040722i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 18 Aug 2004, William Lee Irwin III wrote: >>> Something like this? On Wed, Aug 18, 2004 at 11:19:25PM -0400, Rajesh Venkatasubramanian wrote: >> Yeah, something similar... A small nitpick: page_table_lock >> nests within i_mmap_lock and anon_lock. >> do_unmap() also frees page tables first and then removes vmas from >> i_mmap (and/or) anon_vma list. Is there a reason to preserve >> this ordering ? On Wed, Aug 18, 2004 at 08:31:27PM -0700, William Lee Irwin III wrote: > I don't see a reason to rearrange it. We can just as easily drop and > reacquire ->page_table_lock. On closer examination it's not that simple. -- wli