[PATCH] D114957: [AMDGPU] Change llvm.amdgcn.image.bvh.intersect.ray to take vec3 args

Jay Foad via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Fri Dec 3 02:00:50 PST 2021

foad added a comment.

In D114957#3167700 <https://reviews.llvm.org/D114957#3167700>, @arsenm wrote:

> I think this macro is purely terrible and should not be added (and at least should be all caps?). If we can't just hard break users, I would rather just leave the builtin signatures broken

OK, how about D115032 <https://reviews.llvm.org/D115032>?

Personally I have no opinion about what's best to do with the OpenCL builtins, but I would like to make progress with changing the intrinsics. So I have a slight preference for D115032 <https://reviews.llvm.org/D115032> because it gives me a way forward without changing OpenCL behaviour. The OpenCL team can then decide whether or not to update the builtins at their leisure.

  rG LLVM Github Monorepo



More information about the cfe-commits mailing list