>
> The main case where the premise doesn't hold is non-Ovlt. But we still
> don't want spills so we should always compare costs as long as there's
> no clear ceiling like spills ("dynamic") or the type ratio
> ("conv-dynamic").
Yes, and we mainly focus on non-Ovlt performance as so far, AFAIK all
our existing cores have been non-Ovlt.
So basically a dynamic max LMUL as it is now should work on Ovlt cores
quite well. Maybe we can guard the cost comparison with non-Ovlt
related tune conditionals?
Regards,
Bohan
