Re: [PATCH v7 0/7] Linux RISC-V IOMMU Support

[xiebo bo]

Hi all,
I hope this email finds you well.

I have a question regarding the process of updating page table entries (PTE) in the system. Specifically, if there are ongoing accesses to a PTE that is not yet present in the IOTLB, and the PTE is updated while these accesses are still in progress, there could be a scenario where the system expects to retrieve the old PTE value but instead retrieves the new one, potentially leading to translation errors.

However, I have not seen any memory barriers or lock operations in the code to ensure proper synchronization during PTE updates. Could this cause any issues with the translation process?

Thank you for your time, and I look forward to your insights on this matter.

Best regards