On Tue, Feb 10, 2026 at 8:25 AM Richard Henderson
<[email protected]> wrote:
...
> Hmm.  This really overlaps the secure and space fields from arm, and possibly 
> some of the
> others as well (e.g. user, requester_id, pid).
>
> I don't really have a good suggestion for that right now, but it would be 
> nice to not keep
> expanding the count of these sorts of fields that somehow specify the 
> originator, but
> clearly cannot overlap.
>
> I'm reasonably sure we've had this discussion before, but nothing has come of 
> it.
>
> Time to paint the bikeshed again?
>

I can union the 'secure' and 'world_id' fields, as they are not used together.
However, I have no idea about the 'space' fields.

I have seen CPUTLBEntryFull has the extra union to support ARM-specific members.
Another idea is that also adding the extra union to MemTxAttrs to
place the RISC-V worid_id.
We can use this extra union to as SoC-specific signals in the bus,
like AXI AxUSER signal.
Do you think it is suitable?


Thanks,

Jim

Reply via email to