Re: [PATCH v3 1/5] iommu/amd: Remove iommu_v2 module

[Baolu Lu <baolu.lu@linux.intel.com>]

On 2023/9/22 20:18, Jason Gunthorpe wrote:
> On Fri, Sep 22, 2023 at 08:13:02PM +0800, Baolu Lu wrote:
> 
>>> You might make this argument if ACS is also non-isolating though..
>>>
>>> Regardless I think we need to get into a position where the iommu core
>>> is deciding if PRI or ATS is enabled for a device, not the iommu
>>> driver.
>>
>> Agreed. I ever had a series to achieve this and it may be time to
>> revisit them.
>>
>> One additional concern is how the core knows whether ATS should be
>> enabled. In my previous design, the IOMMU core turns on ATS by default
>> if the device is capable of it, but the driver could enable/disable it
>> from its driver probe() callback.
> 
> I think that is a good place to start, maybe we have some command line
> and/or sysfs like we have for the default domain policy.

Yes. It sounds reasonable.

> 
> IMHO iommu drivers should not override this.
> 
> I think it was a mistake that iommu drivers could override to identity
> domains. If we need quirks then the core code should have the quirk
> list, similar to how PCI works.

If I remember it correctly, you have posted a series to address this. Or
I might misunderstood it.

Best regards,
baolu