Re: [PATCH] iommu/arm-smmu-v3: Simplify stage selection logic

[Robin Murphy <robin.murphy@arm.com>]

On 2023-08-17 17:03, Michael Shavit wrote:
> It is invalid for an arm-smmu-v3 to have neither FEAT_TRANS_S1 nor
> FEAT_TRANS_S2 bits set, and this is even checked in the probe.
> 
> Only set ARM_SMMU_DOMAIN_S2 if FEAT_TRANS_S1 isn't supported, otherwise set
> ARM_SMMU_DOMAIN_S1. This is clearer as the existing code implies that
> something more sophisticated is going on with the stage selection logic.

The reason it's like this is because of arm_smmu_enable_nesting(), which 
*is* the additional thing that's going on with the stage selection logic.

Thanks,
Robin.

> Signed-off-by: Michael Shavit <mshavit@google.com>
> ---
> 
>   drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
> index 9b0dc35056019..e74c8c630adfa 100644
> --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
> +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
> @@ -2206,7 +2206,7 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>   	/* Restrict the stage to what we can actually support */
>   	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1))
>   		smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
> -	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S2))
> +	else
>   		smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
>   
>   	switch (smmu_domain->stage) {
> 
> base-commit: 6eaae198076080886b9e7d57f4ae06fa782f90ef