Hi Nicolin,
On Tue, Jul 15, 2025 at 11:42:24AM -0700, Nicolin Chen wrote:
Sorry for the late response.
No problem at all.
On Fri, Jul 11, 2025 at 05:14:27PM +0100, Will Deacon wrote:
/* MMIO registers */ #define ARM_SMMU_IDR0 0x0 @@ -720,6 +721,10 @@ struct arm_smmu_impl_ops { int (*init_structures)(struct arm_smmu_device *smmu); struct arm_smmu_cmdq *(*get_secondary_cmdq)( struct arm_smmu_device *smmu, struct arm_smmu_cmdq_ent *ent);
- const size_t vsmmu_size;
- const enum iommu_viommu_type vsmmu_type;
- int (*vsmmu_init)(struct arm_vsmmu *vsmmu,
const struct iommu_user_data *user_data);It would be nice to avoid adding data members to the ops structure, if
You mean the "vsmmu_size" and "vsmmu_type" right?
Yup.
So you want them to be removed, by having two impl_ops: size_t get_vsmmu_size(enum iommu_viommu_type vsmmu_type); int (*vsmmu_init)(struct arm_vsmmu *vsmmu, const struct iommu_user_data *user_data);
right?
Yes, please.
at all possible. The easiest thing would probably be to add a function for getting the vsmmu size and then pushing the two checks against 'vsmmu_type' down into the impl_ops callbacks so that:
- If the type is IOMMU_VIOMMU_TYPE_ARM_SMMUV3, we don't bother with the impl_ops at all in arm_vsmmu_init() and arm_smmu_get_viommu_size()
Hmm, I was hoping for an implementation could support the default IOMMU_VIOMMU_TYPE_ARM_SMMUV3 while having its own viommu_ops or so. But I think your suggestion is fine since there is no such a use case at this moment :)
- Otherwise, we pass the type into the impl_ops and they can check it
Of course, that can be a patch on top of the series as there's no point respinning the whole just for this.
Thanks for that! I can draft a patch to send later this week once the requirements are confirmed.
Thank you!
Will