From xen-devel-bounces@lists.xen.org Tue Feb 18 10:31:20 2014 Received: (at maildrop) by bugs.xenproject.org; 18 Feb 2014 10:31:20 +0000 Received: from lists.xen.org ([50.57.142.19]) by bugs.xenproject.org with esmtp (Exim 4.80) (envelope-from ) id 1WFhxQ-0005N6-Ft for xen-devel-maildrop-Eithu9ie@bugs.xenproject.org; Tue, 18 Feb 2014 10:31:20 +0000 Received: from localhost ([127.0.0.1] helo=lists.xen.org) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1WFhsa-0007lO-2g; Tue, 18 Feb 2014 10:26:20 +0000 Received: from mail6.bemta14.messagelabs.com ([193.109.254.103]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1WFhsY-0007l8-M0 for xen-devel@lists.xen.org; Tue, 18 Feb 2014 10:26:18 +0000 Received: from [193.109.254.147:63550] by server-11.bemta-14.messagelabs.com id 2E/A7-24604-A4533035; Tue, 18 Feb 2014 10:26:18 +0000 X-Env-Sender: George.Dunlap@citrix.com X-Msg-Ref: server-16.tower-27.messagelabs.com!1392719175!5049190!1 X-Originating-IP: [66.165.176.63] X-SpamReason: No, hits=0.0 required=7.0 tests=sa_preprocessor: VHJ1c3RlZCBJUDogNjYuMTY1LjE3Ni42MyA9PiAzMDYwNDg=\n X-StarScan-Received: X-StarScan-Version: 6.9.16; banners=-,-,- X-VirusChecked: Checked Received: (qmail 14388 invoked from network); 18 Feb 2014 10:26:17 -0000 Received: from smtp02.citrix.com (HELO SMTP02.CITRIX.COM) (66.165.176.63) by server-16.tower-27.messagelabs.com with RC4-SHA encrypted SMTP; 18 Feb 2014 10:26:17 -0000 X-IronPort-AV: E=Sophos;i="4.97,501,1389744000"; d="scan'208";a="101691543" Received: from accessns.citrite.net (HELO FTLPEX01CL02.citrite.net) ([10.9.154.239]) by FTLPIPO02.CITRIX.COM with ESMTP; 18 Feb 2014 10:26:15 +0000 Received: from ukmail1.uk.xensource.com (10.80.16.128) by smtprelay.citrix.com (10.13.107.79) with Microsoft SMTP Server id 14.2.342.4; Tue, 18 Feb 2014 05:26:15 -0500 Received: from elijah.uk.xensource.com ([10.80.2.24]) by ukmail1.uk.xensource.com with esmtp (Exim 4.69) (envelope-from ) id 1WFhsU-0000I9-9F; Tue, 18 Feb 2014 10:26:14 +0000 Message-ID: <53033544.2000409@eu.citrix.com> Date: Tue, 18 Feb 2014 10:26:12 +0000 From: George Dunlap User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0 MIME-Version: 1.0 To: "Zhang, Yang Z" , Jan Beulich , Tim Deegan References: <20140210080314.GA758@deinos.phlegethon.org> <20140211090202.GC92054@deinos.phlegethon.org> <20140211115553.GB97288@deinos.phlegethon.org> <52FA2C63020000780011B201@nat28.tlf.novell.com> <52FA480D.9040707@eu.citrix.com> <52FCE8BE.8050105@eu.citrix.com> <52FCF90F020000780011C29A@nat28.tlf.novell.com> <20140213162022.GE82703@deinos.phlegethon.org> <5301F000020000780011CCE0@nat28.tlf.novell.com> <53022209.1060005@eu.citrix.com> <5302332D020000780011CEF1@nat28.tlf.novell.com> In-Reply-To: X-DLP: MIA1 Cc: "andrew.cooper3@citrix.com" , "Dugger, Donald D" , "Zhang, Xiantao" , "xen-devel@lists.xen.org" Subject: Re: [Xen-devel] [PATCH] Don't track all memory when enabling log dirty to track vram X-BeenThere: xen-devel@lists.xen.org X-Mailman-Version: 2.1.13 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org On 02/18/2014 03:14 AM, Zhang, Yang Z wrote: > Jan Beulich wrote on 2014-02-17: >>>>> On 17.02.14 at 15:51, George Dunlap >> wrote: >>> On 02/17/2014 10:18 AM, Jan Beulich wrote: >>>> Actually I'm afraid there are two problems with this patch: >>>> >>>> For one, is enabling "global" log dirty mode still going to work >>>> after VRAM-only mode already got enabled? I ask because the >>>> paging_mode_log_dirty() check which paging_log_dirty_enable() does >>>> first thing suggests otherwise to me (i.e. the now conditional >>>> setting of all p2m entries to p2m_ram_logdirty would seem to never >>>> get executed). IOW I would think that we're now lacking a control >>>> operation allowing the transition from dirty VRAM tracking mode to >>>> full log dirty mode. >>> Hrm, will so far playing with this I've been unable to get a >>> localhost migrate to fail with the vncviewer attached. Which seems a bit strange... >> Not necessarily - it may depend on how the tools actually do this: >> They might temporarily disable log dirty mode altogether, just to >> re-enable full mode again right away. But this specific usage of the >> hypervisor interface wouldn't (to me) mean that other tool stacks >> might not be doing this differently. > You are right. Before migration, libxc will disable log dirty mode if it already enabled before and re-enable it. So when I am developing this patch, I think it is ok for migration. > > If there really have other tool stacks also will use this interface (Is it true?), perhaps my original patch is better which will check paging_mode_log_dirty(d) && log_global: It turns out that the reason I couldn't get a crash was because libxc was actually paying attention to the -EINVAL return value, and disabling and then re-enabling logdirty. That's what would happen before your dirty vram patch, and that's what happens after. And arguably, that's the correct behavior for any toolstack, given that the interface returns an error. This patch would actually change the interface; if we check this in, then if you enable logdirty when dirty vram tracking is enabled, you *won't* get an error, and thus *won't* disable and re-enable logdirty mode. So actually, this patch would be more disruptive. Thanks for the patch, though. :-) -George > > diff --git a/xen/arch/x86/mm/paging.c b/xen/arch/x86/mm/paging.c > index ab5eacb..368c975 100644 > --- a/xen/arch/x86/mm/paging.c > +++ b/xen/arch/x86/mm/paging.c > @@ -168,7 +168,7 @@ int paging_log_dirty_enable(struct domain *d, bool_t log_global) > { > int ret; > > - if ( paging_mode_log_dirty(d) ) > + if ( paging_mode_log_dirty(d) && !log_global ) > return -EINVAL; > > domain_pause(d); > > >> Jan > > Best regards, > Yang > > _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel