Re: [PATCH v2 4/5] iommu: Support mm PASID 1:n with sva domains

[Vasant Hegde]

On 8/31/2023 2:06 AM, Jason Gunthorpe wrote:
> On Mon, Aug 28, 2023 at 02:02:52PM +0530, Vasant Hegde wrote:
> 
>> I am not sure whether I understood the flow completely. Just wondering why you
>> are not freeing pasid in iommu_sva_unbind_device().
>> I mean once iommu_mm->sva_domains becomes free shouldn't we free the
>> iommu_mm->pasid?
> 
> No, for Intel use cases that PASID permanently becomes part of the
> ENQCMD MSR and cannot be revoked once it has been set
>  

Fair enough.

Patch description did explain it to some extent. ("The PASID is released in
__mmdrop()"). May be this needs to be expanded to cover why pasid is not
released in iommu_sva_unbind_device().


>> Also in this function (mm_pasid_drop()), should we check/free sva domains?
> 
> No, the driver must support a SVA domain with an empty mm_struct, eg
> by hooking release.

Sorry. Looks like confused you. Looking into code I got this.

My question was: when PASID is enabled, is there any chance of mm_pasid_drop()
getting called before freeing all SVA domains?

-Vasant