Re: [RFC PATCH 0/6] Auxiliary IOMMU domains and Arm SMMUv3

[Lu Baolu <baolu.lu-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>]

Hi,

On 10/23/18 12:48 AM, Jordan Crouse wrote:
> On Fri, Oct 19, 2018 at 07:11:52PM +0100, Jean-Philippe Brucker wrote:
> 
>> (2) Allocate a domain and attach it to the device.
>>
>>        dom = iommu_domain_alloc()
>>        iommu_attach_device(dom, dev)
>>
>>      I still have concerns about this part, which are highlighted by the
>>      messy changes of patch 1. I think it would make more sense to
>>      introduce new attach/detach_dev_aux() functions instead of reusing
>>      attach/detach_dev()
>>
>>      Can we reconsider this and avoid unnecessary complications in IOMMU
>>      core and drivers? Does the VFIO code greatly benefit from using the
>>      same attach() function? It could as well use a different one for
>>      devices in AUXD mode, which the mediating driver could tell by
>>      adding a flag in mdev_set_iommu_device(), for example.
>>
>>      And I don't think other users of AUXD would benefit from using the
>>      same attach() function, since they will know whether they want to be
>>      using main or auxiliary domain when doing attach().
> 
> I second this. For the arm-smmu-v2 multiple-pagetable model we would either need
> new API functions or do something along the lines of
> 
> dom = iommu_domain_alloc()
> iommu_set_attr(dom, DOMAIN_ATTR_I_WANT_TO_BE_AUX)
> iommu_attach_device(dom,dev)
> 
> Either that or do some some dummy device magic that I haven't started to work
> out in my head. Having aux specific functions just for this step would make the
> arm-smmu-v2 version work out much cleaner.

Okay! If there are no more comments, I will prepare a new version with
aux specific functions.

Best regards,
Lu Baolu