From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Cooper Subject: Re: question on share ept page table with vtd Date: Sat, 1 Feb 2014 00:56:48 +0000 Message-ID: <52EC4650.7020302@citrix.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from mail6.bemta14.messagelabs.com ([193.109.254.103]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1W9Ot9-0004J8-IH for xen-devel@lists.xenproject.org; Sat, 01 Feb 2014 00:56:51 +0000 In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: "Zhang, Yang Z" , "xen-devel@lists.xenproject.org" Cc: "Tim Deegan (tim@xen.org)" , "'Jan Beulich (JBeulich@suse.com)'" , "Zhang, Xiantao" List-Id: xen-devel@lists.xenproject.org On 01/02/2014 00:29, Zhang, Yang Z wrote: > Hi all > > I have a question with the page table share between EPT and VTD which is enabled by default. Currently, we will enable log_dirty mode to track vram. And when first enable the log dirty mode, it will set all guest ram to readonly in EPT entry. The question is that if a page is current used as DMA write buffer, isn't it a problem? > > best regards > yang Looking at the code, it would indeed appear to be a problem. Is has presumably gone unnoticed until now as all writes into the vram would have been from software, and hit the log_dirty logic, rather than DMA writes hitting the IOMMU. ~Andrew