From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from zps18.corp.google.com (zps18.corp.google.com [172.25.146.18]) by smtp-out.google.com with ESMTP id l9QGpeO9010823 for ; Fri, 26 Oct 2007 17:51:40 +0100 Received: from nf-out-0910.google.com (nfhf5.prod.google.com [10.48.233.5]) by zps18.corp.google.com with ESMTP id l9QGocpF016580 for ; Fri, 26 Oct 2007 09:51:39 -0700 Received: by nf-out-0910.google.com with SMTP id f5so666214nfh for ; Fri, 26 Oct 2007 09:51:38 -0700 (PDT) Message-ID: Date: Fri, 26 Oct 2007 12:51:38 -0400 From: "Ross Biro" Subject: Re: RFC/POC Make Page Tables Relocatable In-Reply-To: <20071026161007.GA19443@skynet.ie> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <1193330774.4039.136.camel@localhost> <1193335725.24087.19.camel@localhost> <20071026161007.GA19443@skynet.ie> Sender: owner-linux-mm@kvack.org Return-Path: To: Mel Gorman Cc: Dave Hansen , linux-mm@kvack.org, Mel Gorman List-ID: On 10/26/07, Mel Gorman wrote: > I suspect this might be overkill from a memory fragmentation > perspective. When grouping pages by mobility, page table pages are > currently considered MIGRATE_UNMOVABLE. From what I have seen, they are I may be being dense, but the page migration code looks to me like it just moves pages in a process from one node to another node with no effort to touch the page tables. It would be easy to hook the code I wrote into the page migration code, what I don't understand is when the page tables should be migrated? Only when the whole process is being migrated? When all the pages pointed to a page table are being migrated? When any page pointed to by the page table is being migrated? Ross -- 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