From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-yw0-f200.google.com (mail-yw0-f200.google.com [209.85.161.200]) by kanga.kvack.org (Postfix) with ESMTP id 1FD346B0260 for ; Tue, 19 Apr 2016 12:15:06 -0400 (EDT) Received: by mail-yw0-f200.google.com with SMTP id o131so44815502ywc.2 for ; Tue, 19 Apr 2016 09:15:06 -0700 (PDT) Received: from mx1.redhat.com (mx1.redhat.com. [209.132.183.28]) by mx.google.com with ESMTPS id u11si5797736qhu.103.2016.04.19.09.15.05 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 19 Apr 2016 09:15:05 -0700 (PDT) Date: Tue, 19 Apr 2016 19:15:00 +0300 From: "Michael S. Tsirkin" Subject: Re: [PATCH kernel 1/2] mm: add the related functions to build the free page bitmap Message-ID: <20160419191111-mutt-send-email-mst@redhat.com> References: <1461076474-3864-1-git-send-email-liang.z.li@intel.com> <1461076474-3864-2-git-send-email-liang.z.li@intel.com> <1461077659.3200.8.camel@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org List-ID: To: "Li, Liang Z" Cc: Rik van Riel , "viro@zeniv.linux.org.uk" , "linux-kernel@vger.kernel.org" , "quintela@redhat.com" , "amit.shah@redhat.com" , "pbonzini@redhat.com" , "dgilbert@redhat.com" , "linux-mm@kvack.org" , "kvm@vger.kernel.org" , "qemu-devel@nongnu.org" , "agraf@suse.de" , "borntraeger@de.ibm.com" On Tue, Apr 19, 2016 at 03:02:09PM +0000, Li, Liang Z wrote: > > On Tue, 2016-04-19 at 22:34 +0800, Liang Li wrote: > > > The free page bitmap will be sent to QEMU through virtio interface and > > > used for live migration optimization. > > > Drop the cache before building the free page bitmap can get more free > > > pages. Whether dropping the cache is decided by user. > > > > > > > How do you prevent the guest from using those recently-freed pages for > > something else, between when you build the bitmap and the live migration > > completes? > > Because the dirty page logging is enabled before building the bitmap, there is no need > to prevent the guest from using the recently-freed pages ... > > Liang Well one point of telling host that page is free is so that it can mark it clean even if it was dirty previously. So I think you must pass the pages to guest under the lock. This will allow host optimizations such as marking these pages MADV_DONTNEED or MADV_FREE. Otherwise it's all too tied up to a specific usecase - you aren't telling host that a page is free, you are telling it that a page was free in the past. -- MST -- 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