Re: [PATCH for 8.0 0/2] virtio-iommu: Fix Replay

2022-12-20 Thread Eric Auger
Hi Michael, Peter, On 12/20/22 15:59, Michael S. Tsirkin wrote: > On Thu, Dec 08, 2022 at 09:58:06AM -0500, Peter Xu wrote: >> On Thu, Dec 08, 2022 at 08:48:09AM +0100, Eric Auger wrote: >>> Hi Peter, >> Hi, Eric, >> >>> On 12/8/22 00:49, Peter Xu wrote: Hi, Eric, On Wed, Dec 07, 202

Re: [PATCH for 8.0 0/2] virtio-iommu: Fix Replay

2022-12-20 Thread Michael S. Tsirkin
On Thu, Dec 08, 2022 at 09:58:06AM -0500, Peter Xu wrote: > On Thu, Dec 08, 2022 at 08:48:09AM +0100, Eric Auger wrote: > > Hi Peter, > > Hi, Eric, > > > > > On 12/8/22 00:49, Peter Xu wrote: > > > Hi, Eric, > > > > > > On Wed, Dec 07, 2022 at 02:36:44PM +0100, Eric Auger wrote: > > >> When assi

Re: [PATCH for 8.0 0/2] virtio-iommu: Fix Replay

2022-12-08 Thread Peter Xu
On Thu, Dec 08, 2022 at 08:48:09AM +0100, Eric Auger wrote: > Hi Peter, Hi, Eric, > > On 12/8/22 00:49, Peter Xu wrote: > > Hi, Eric, > > > > On Wed, Dec 07, 2022 at 02:36:44PM +0100, Eric Auger wrote: > >> When assigning VFIO devices protected by a virtio-iommu we need to replay > >> the mappin

Re: [PATCH for 8.0 0/2] virtio-iommu: Fix Replay

2022-12-07 Thread Eric Auger
Hi Peter, On 12/8/22 00:49, Peter Xu wrote: > Hi, Eric, > > On Wed, Dec 07, 2022 at 02:36:44PM +0100, Eric Auger wrote: >> When assigning VFIO devices protected by a virtio-iommu we need to replay >> the mappings when adding a new IOMMU MR and when attaching a device to >> a domain. While we do a

Re: [PATCH for 8.0 0/2] virtio-iommu: Fix Replay

2022-12-07 Thread Peter Xu
Hi, Eric, On Wed, Dec 07, 2022 at 02:36:44PM +0100, Eric Auger wrote: > When assigning VFIO devices protected by a virtio-iommu we need to replay > the mappings when adding a new IOMMU MR and when attaching a device to > a domain. While we do a "remap" we currently fail to first unmap the > existi

[PATCH for 8.0 0/2] virtio-iommu: Fix Replay

2022-12-07 Thread Eric Auger
When assigning VFIO devices protected by a virtio-iommu we need to replay the mappings when adding a new IOMMU MR and when attaching a device to a domain. While we do a "remap" we currently fail to first unmap the existing IOVA mapping and just map the new one. With some device/group topology this