From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail137.messagelabs.com (mail137.messagelabs.com [216.82.249.19]) by kanga.kvack.org (Postfix) with SMTP id 094046B01EE for ; Tue, 27 Apr 2010 20:09:05 -0400 (EDT) Date: Wed, 28 Apr 2010 02:08:33 +0200 From: Andrea Arcangeli Subject: Re: [PATCH 3/3] mm,migration: Remove straggling migration PTEs when page tables are being moved after the VMA has already moved Message-ID: <20100428000833.GE510@random.random> References: <1272403852-10479-1-git-send-email-mel@csn.ul.ie> <1272403852-10479-4-git-send-email-mel@csn.ul.ie> <20100428090302.5e69721f.kamezawa.hiroyu@jp.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100428090302.5e69721f.kamezawa.hiroyu@jp.fujitsu.com> Sender: owner-linux-mm@kvack.org To: KAMEZAWA Hiroyuki Cc: Mel Gorman , Linux-MM , LKML , Minchan Kim , Christoph Lameter , Rik van Riel , Andrew Morton List-ID: On Wed, Apr 28, 2010 at 09:03:02AM +0900, KAMEZAWA Hiroyuki wrote: > I bet calling __get_user_pages_fast() before vma_adjust() is the way to go. > When page_count(page) != page_mapcount(page) +1, migration skip it. My proposed fix avoids to walk the pagetables once more time and to mangle over the page counts. Can you check it? It works but it needs more review. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org