From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wei Wang Subject: Re: [PATCH 00 of 14 V3] amd iommu: support ATS device passthru on IOMMUv2 systems Date: Wed, 11 Jan 2012 18:36:41 +0100 Message-ID: <4F0DC8A9.8060106@amd.com> References: <4F0DB31D020000780006BDB2@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: <4F0DB31D020000780006BDB2@nat28.tlf.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Jan Beulich Cc: keir@xen.org, xen-devel@lists.xensource.com, Ian.Jackson@eu.citrix.com, Ian.Campbell@citrix.com List-Id: xen-devel@lists.xenproject.org Am 11.01.2012 16:04, schrieb Jan Beulich: >>>> On 10.01.12 at 18:07, Wei Wang wrote: >> Hi all, this is patch v3. >> ATS devices with PRI and PASID capabilities can communicate with iommuv2 to >> perform two level (nested) address translation and demand paging for DMA. >> To passthru such devices, iommu driver has to been enabled in guest OS. >> This patch set adds initial iommu emulation for hvm guests to support ATS >> device passthru. > I would look into committing 1-6 and 10 (if that one is independent of > 7-9), if you can confirm that those on their own provide meaningful > benefit (enabling the ppr log probably is what I'm after, but I'd still > like your confirmation - patch 3 in particular doesn't look very useful > without the later ones). So ideally the ones leading up to the ppr log > enabling would all be first (or even a separate series), and the guest > iommu ones would follow (as those make only sense when the tools > maintainers are okay with the changes too) Hi Jan, Thanks for doing this. It sounds great! Even without guest iommu being enabled, patch 1-6 and 10 would still be useful for turn on ppr and GT features. Patch 6 will call some functions from patch 3. If you want to leave patch 3 behind, I could send a new version to move patch 6 ahead. Thanks, Wei > Jan > >> changes in v3: >> * Use xenstore to receive guest iommu configuration instead of adding in a >> new field in hvm_info_table. >> * Support pci segment in vbdf to mbdf bind. >> * Make hypercalls visible for non-x86 platforms. >> * A few code cleanups according to comments from Jan and Ian. >> >> Changes in v2: >> * Do not use linked list to access guest iommu tables. >> * Do not parse iommu parameter in libxl_device_model_info again. >> * Fix incorrect logical calculation in patch 11. >> * Fix hypercall definition for non-x86 systems. >> >> Thanks, >> Wei > > >