Re: [RFC v2 12/20] dma-iommu: Implement NESTED_MSI cookie

2018-10-27 Thread Auger Eric
Hi Robin, On 10/25/18 12:05 AM, Robin Murphy wrote: > On 2018-10-24 7:44 pm, Auger Eric wrote: >> Hi Robin, >> >> On 10/24/18 8:02 PM, Robin Murphy wrote: >>> Hi Eric, >>> >>> On 2018-09-18 3:24 pm, Eric Auger wrote: Up to now, when the type was UNMANAGED, we used to allocate IOVA pages

Re: [RFC v2 12/20] dma-iommu: Implement NESTED_MSI cookie

2018-10-24 Thread Robin Murphy
On 2018-10-24 7:44 pm, Auger Eric wrote: Hi Robin, On 10/24/18 8:02 PM, Robin Murphy wrote: Hi Eric, On 2018-09-18 3:24 pm, Eric Auger wrote: Up to now, when the type was UNMANAGED, we used to allocate IOVA pages within a range provided by the user. This does not work in nested mode. If both

Re: [RFC v2 12/20] dma-iommu: Implement NESTED_MSI cookie

2018-10-24 Thread Auger Eric
Hi Robin, On 10/24/18 8:02 PM, Robin Murphy wrote: > Hi Eric, > > On 2018-09-18 3:24 pm, Eric Auger wrote: >> Up to now, when the type was UNMANAGED, we used to >> allocate IOVA pages within a range provided by the user. >> This does not work in nested mode. >> >> If both the host and the guest a

Re: [RFC v2 12/20] dma-iommu: Implement NESTED_MSI cookie

2018-10-24 Thread Robin Murphy
Hi Eric, On 2018-09-18 3:24 pm, Eric Auger wrote: Up to now, when the type was UNMANAGED, we used to allocate IOVA pages within a range provided by the user. This does not work in nested mode. If both the host and the guest are exposed with SMMUs, each would allocate an IOVA. The guest allocate

[RFC v2 12/20] dma-iommu: Implement NESTED_MSI cookie

2018-09-18 Thread Eric Auger
Up to now, when the type was UNMANAGED, we used to allocate IOVA pages within a range provided by the user. This does not work in nested mode. If both the host and the guest are exposed with SMMUs, each would allocate an IOVA. The guest allocates an IOVA (gIOVA) to map onto the guest MSI doorbell