Re: [PATCH v6 0/2] MR364: vkd3d-shader/hlsl: Implement inverse trigonometry.

[Henri Verbeet (＠hverbeet)]

> If the intrinsics aren't implemented upstream, is this something we can fix at all? I guess I'm still not clear what the problem is. The point is that *this* MR implements the intrinsics, right?

DXIL has its own intrinsics for acos/asin/atan. I.e., a HLSL acos() will get compiled to a DXIL DX_ACOS by dxcompiler, which our DXIL parser then needs to handle, most likely by translating it to a ACOS vsir instruction. I believe Conor has already written that code, although it's not quite upstream yet.

For this MR, all you'll likely need to do is to add "todo(sm>=6)" in the appropriate places.