On Wed Feb 8 19:21:38 2023 +0000, Zebediah Figura wrote:
I'm not a fan of this for clarity. I'd rather guard out the offset_regset assignment by checking if the type is a struct (or array?) and then just leave it undefined otherwise. [I know this is kind of what you were trying to achieve with HLSL_REGSET_VOID, but I still am inclined to say that's not worth existing.] I can live with it on the grounds that this function is eventually going to go away, but I do want to push for actually moving forward on that. I don't want to litter the code with half-finished refactors; it's not the worst thing for maintenance but it's not great either.
Okay, I added the guard in a new patch before this one. I had already removed `HLSL_REGSET_VOID`.