From mboxrd@z Thu Jan 1 00:00:00 1970 From: Malcolm Crossley Subject: Re: [PATCH] VTD/Intremap: Disable Intremap on Chipset 5500/5520/X58 due to errata Date: Wed, 16 Jan 2013 14:33:39 +0000 Message-ID: <50F6BA43.6000100@citrix.com> References: <50F6C3E002000078000B649F@nat28.tlf.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <50F6C3E002000078000B649F@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: Andrew Cooper , "xiantao.zhang@intel.com" , xen-devel List-Id: xen-devel@lists.xenproject.org On 16/01/13 14:14, Jan Beulich wrote: >>>> On 16.01.13 at 00:27, Malcolm Crossley wrote: >> Unfortunately this chipset is very common and many BIOSes are not >> disabling remapping. We can detect this in Xen and prevent turning on >> remapping in the first place. However, this will turn VT-d off on many >> systems by default. > Why would the IOMMU be turned off in that case as a whole? > Interrupt remapping is not a prerequisite iirc. You are correct. I read the description of XSA-3 but didn't the check the current xen-unstable code correctly. >> Users who still wish to use VT-d can use iommu=force if they are happy >> exposing the associated security risk. > "iommu=force" doesn't enable interrupt remapping if it was > forcibly turned off. This was based upon the faulty code reading above and therefore the comment is wrong . I will change the patch description. > Jan >