From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932214AbWIPGiZ (ORCPT ); Sat, 16 Sep 2006 02:38:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932208AbWIPGiY (ORCPT ); Sat, 16 Sep 2006 02:38:24 -0400 Received: from gatekeeper.ncic.ac.cn ([159.226.41.188]:7122 "HELO ncic.ac.cn") by vger.kernel.org with SMTP id S932214AbWIPGiY convert rfc822-to-8bit (ORCPT ); Sat, 16 Sep 2006 02:38:24 -0400 Date: Sat, 16 Sep 2006 14:38:19 +0800 From: "Zhou Yingchao" To: "Hugh Dickins" Cc: "linux-kernel" , "akpm" , "alan" , "zxc" Subject: Re: Re: Re: Re: [RFC] PAGE_RW Should be added to PAGE_COPY ? X-mailer: Foxmail 5.0 [cn] Mime-Version: 1.0 Content-Type: text/plain; charset="gb2312" X-Antivirus: avast! (VPS 0637-2, 2006-09-15), Outbound message X-Antivirus-Status: Clean Message-Id: <20060916063823.14432FB044@ncic.ac.cn> Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > >I replied too hastily. I've since looked back to your original mail Thank you for your patience. >(in which you do mention Infiniband as an example, which has now >helped to jog my brain), and realized that I was misinterpreting >what you meant by "user-level network driver". > >The problem, as I now see it, is precisely with do_wp_page()'s >TestSetPageLocked, as you first said. Yeah, I think you get it. ^_^ > >Yes, it would be good if we could do that check in some other, >reliable way. The problem is that can_share_swap_page has to >check page_mapcount (and PageSwapCache) and page_swapcount in >an atomic way: the page lock is what we have used to guard the >movement between mapcount and swapcount. > >I'll try to think whether we can do that better, >but not until next week. Ok, I expect for good news. > >Hugh Yingchao Zhou