On 6/17/25 5:58 PM, Jeff Law wrote: > You could make the argument that instead of defaulting to rocket that we > should use a default generic tuning model. That would make much more > sense than deliberately choosing the wrong values for the rocket uarch > because it happens to be used as the default.
I think a generic tuning model would be a good idea. I can also see that we'd probably want a generic 32-bit tuning model and a generic 64-bit tuning model, since the 32-bit cores are probably targeted for smaller/simpler use cases and probably employ simpler uarches with different timings that 64-bit cores. Peter
