Open Timmmm opened 7 months ago
In the concurrency model, x1
.. xN
are separate registers for intra-instruction dependencies (so instructions modifying different registers can be re-ordered), whereas xrf : vector(...)
would be a single register.
Ah very subtle but that makes sense. I'll leave this issue open so we add a comment to the code.
As far as I can see there's no reason to have this:
Instead of
This is done for all the X, F and V registers.