Re: [PATCH 1/1] iommu/vt-d: Fix RID2PASID setup failure

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

On 2022/6/22 10:56, Ethan Zhao wrote:
> 在 2022/6/20 16:17, Lu Baolu 写道:
>> The IOMMU driver shares the pasid table for PCI alias devices. When the
>> RID2PASID entry of the shared pasid table has been filled by the first
>> device, the subsequent devices will encounter the "DMAR: Setup RID2PASID
>> failed" failure as the pasid entry has already been marke as present. As
>> the result, the IOMMU probing process will be aborted.
>>
>> This fixes it by skipping RID2PASID setting if the pasid entry has been
>> populated. This works because the IOMMU core ensures that only the same
>> IOMMU domain can be attached to all PCI alias devices at the same time.
>> Therefore the subsequent devices just try to setup the RID2PASID entry
>> with the same domain, which is negligible.
>      We have two customers reported the issue "DMAR: Setup RID2PASID 
> failed",
> 
> Two ASPEED devices locate behind one PCIe-PCI bridge and iommu SM, PT 
> mode is enabled.  Most
> 
> Interesting thing is the second device is only used by BIOS, and BIOS 
> left it to OS without shutting down,
> 
> and it is useless for OS.

This sounds odd. Isn't this a bug?


> Is there practical case multi devices behind 
> PCIe-PCI bridge share the same
> 
> PASID entry without any security concern ? these two customer's case is 
> not.

The devices underneath the PCIe-PCI bridge are alias devices of the
bridge. PCI alias devices always sit in the same group (the minimal unit
that IOMMU guarantees isolation) and can only be attached with a same
domain (managed I/O address space). Hence, there's no security concern
if they further share the pasid table.

Best regards,
baolu