Re: [Qemu-devel] [PATCH v7 0/5] IOMMU: intel_iommu support map and unmap notifications

[Lan Tianyu]

On 2016年12月06日 15:22, Peter Xu wrote:
> On Tue, Dec 06, 2016 at 03:06:20PM +0800, Lan Tianyu wrote:
>> On 2016年12月06日 14:51, Peter Xu wrote:
>>> On Tue, Dec 06, 2016 at 02:30:24PM +0800, Lan Tianyu wrote:
>>>
>>> [...]
>>>
>>>>>> 2. How to restore GPA->HPA mapping when IOVA is disabled by guest.
>>>>>> When guest enables IOVA for device, vIOMMU will invalidate all previous
>>>>>> GPA->HPA mapping and update IOVA->HPA mapping to pIOMMU via iommu
>>>>>> notifier. But if IOVA is disabled, I think we should restore GPA->HPA
>>>>>> mapping for the device otherwise the device won't work again in the VM.
>>>>>
>>>>> If we can have a workable replay mechanism, this problem will be
>>>>> solved IMHO.
>>>>
>>>> Basic idea is to replay related memory regions to restore GPA->HPA
>>>> mapping when guest disables IOVA.
>>>
>>> Btw, could you help elaborate in what case we will trigger such a
>>> condition? Or, can we dynamically enable/disable an IOMMU?
>>
>> First case I think of is nest virtualization and we assign physical
>> device to l2 guest.
> 
> If we assign physical device to l2 guest, then it will belongs to an
> IOMMU domain in either l1 guest and host, right? (assuming we are
> using vfio-pci to do device assignment)

Yes.

> 
>> User space driver(E.G DPDK) also can enable/disable
>> IOVA for device dynamically.
> 
> Could you provide more detailed (or any pointer) on how to do that? I
> did try to find it myself, I see an VFIO_IOMMU_ENABLE ioctl, but looks
> like it is for ppc only.

No, I just give an example that user space may do that but no more
research. But since Qemu already can enable device's IOVA, other user
application also should can do that with the same VFIO interface, right?

-- 
Best regards
Tianyu Lan