Martin Maas

Thanks for bringing this up, and also for talking through it in yesterday's J Extension meeting. The resolution was that these are indeed the two main options. The rationale why...

Closing the loop on the issue: I will leave the issue open for now in case there is more discussion around these topics, but currently, there are no plans to...

You are correct – pointer masking does indeed not apply to DMA controllers and other devices. It is therefore the responsibility of the kernel to manually untag these addresses. The...

I think there is no downside to adding it as a clarification (I don't believe there are any worries about the word limit). I will leave this issue open as...

The spec has been updated to clarify that pointer masking does not apply to DMA (654a5c4).

+1 to Samuel's description. > What does this really mean? Does it mean these tags are names to be used in profiles to indicate that the higher privilege level as...

The spec has been updated to provide an additional explanation (654a5c4).

For pointer masking, there are implementations for Spike and QEMU. However, these implementations are based on a very old version of the standard and will need to be updated once...

The decision to split the decision was made by the overarching RISC-V architecture review committee. The reason for these different extensions is to bring them into a format that they...

This issue has been addressed in 654a5c4. We added language to the spec to explain the rationale behind these extensions, and why they are necessary.