Re: [RFCv2 PATCH 5/7] iommufd: Introduce data struct for AMD nested domain allocation

[Jason Gunthorpe]

On Thu, Jan 11, 2024 at 06:06:44PM -0600, Suravee Suthikulpanit wrote:
> Introduce IOMMU_HWPT_DATA_AMD_V2 data type for AMD IOMMU v2 page table,
> which is used for stage-1 in nested translation. The data structure
> contains information necessary for setting up the AMD HW-vIOMMU support.
> 
> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@xxxxxxx>
> ---
>  include/uapi/linux/iommufd.h | 25 +++++++++++++++++++++++++
>  1 file changed, 25 insertions(+)
> 
> diff --git a/include/uapi/linux/iommufd.h b/include/uapi/linux/iommufd.h
> index 9901b9f4abe2..b28ec5947571 100644
> --- a/include/uapi/linux/iommufd.h
> +++ b/include/uapi/linux/iommufd.h
> @@ -389,14 +389,39 @@ struct iommu_hwpt_vtd_s1 {
>  	__u32 __reserved;
>  };
>  
> +/**
> + * struct iommu_hwpt_amd_v2 - AMD IOMMU specific user-managed
> + *                            v2 I/O page table data
> + * @gcr3: GCR3 guest physical ddress
> + * @flags.glx: GCR3 table levels
> + * @flags.giov: GIOV mode
> + * @flags.guest_paging_mode: Guest v2 page table paging mode
> + * @flags.reserved : Must be 0
> + * @gdom_id: Guest domain ID
> + * @__reserved: Must be 0
> + */
> +struct iommu_hwpt_amd_v2 {
> +	__aligned_u64 gcr3;
> +	struct {
> +		__aligned_u64 glx  : 1,
> +			      giov : 1,
> +			      guest_paging_mode : 2,
> +			      reserved : 60;
> +	} flags;

IIRC you should not put bitfileds in UAPI headers..

I suggest this should just be something like:

__aligned_u64 dte[3];

And you actually just copy bits directly from the vDTE to the pDTE.

Jason