From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keir Fraser Subject: Re: [PATCH V2] amd iommu: re-enable iommu msi if dom0 disabled it Date: Thu, 21 Jun 2012 17:31:20 +0100 Message-ID: References: <4FE35EC7020000780008B2AD@nat28.tlf.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4FE35EC7020000780008B2AD@nat28.tlf.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Jan Beulich Cc: Wei Wang , Andrew Cooper , Jeremy Fitzhardinge , "xen-devel@lists.xensource.com" , Konrad Rzeszutek Wilk List-Id: xen-devel@lists.xenproject.org On 21/06/2012 16:49, "Jan Beulich" wrote: > the question now arises whether we really want this, and > particularly this late before 4.2. The Linux folks don't seem to > be willing to take the strait forward workaround for the > problem introduced at their end, so we will likely need > something (the more that the questionable fix already made > it into various stable trees) before 4.2 goes out (and even > the older trees would be affected, just that putting a change > like this there is even more questionable). I'd be okay seeing the proposed patch go in ahead of 4.2.0-rc1. -- Keir > There are obviously more potential problems in this area: If > any of the MMIO addresses used by AMD's IOMMU is > configurable through one of the BARs, and if Dom0 decides to > re-assign MMIO space, neither allowing the writes nor simply > dropping them as done here will work. Whether that's a real > problem I don't know - Wei? And there might be other issues > arising from dropping all writes - we might just currently not be > aware of them.