virtio — archive

[Cornelia Huck]

All those clauses actually apply whenever the feature is negotiated, not merely offered. Rename to clarify things. Signed-off-by: Cornelia Huck <cohuck@redhat.com> --- virtio-iommu.tex 10 +++++ ----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/virtio-iommu.tex b/virtio-iommu.tex index f8cbe8895f0d..efe1000ac2d2 100644 --- a/virtio-iommu.tex +++ b/virtio-iommu.tex @@ -232,22 +232,22 @@ subsection{Device operations}label{sec:Device Types / IOMMU Device / Device op Creating mappings aligned on large page sizes can improve performance since they require fewer page table and TLB entries. -item If the VIRTIO_IOMMU_F_DOMAIN_RANGE feature is offered, +item If the VIRTIO_IOMMU_F_DOMAIN_RANGE feature is negotiated, field{domain_range} describes the values supported in a field{domain} - field. If the feature is not offered, any field{domain} value is valid. + field. If the feature is not negotiated, any field{domain} value is valid. -item If the VIRTIO_IOMMU_F_INPUT_RANGE feature is offered, +item If the VIRTIO_IOMMU_F_INPUT_RANGE feature is negotiated, field{input_range} contains the virtual address range that the IOMMU is able to translate. Any mapping request to virtual addresses outside of this range fails. - If the feature is not offered, virtual mappings span over the whole + If the feature is not negotiated, virtual mappings span over the whole 64-bit address space ( exttt{start = 0, end = 0xffffffff ffffffff}) end{itemize} An endpoint is in bypass mode if: egin{itemize} - item the VIRTIO_IOMMU_F_BYPASS_CONFIG feature is offered and: + item the VIRTIO_IOMMU_F_BYPASS_CONFIG feature is negotiated and: egin{itemize} item config field field{bypass} is 1 and the endpoint is not attached to a domain. This applies even if the driver -- 2.34.1