[libclc] [libclc] Update default triple support for AMD and deprecate 32-bit PTX (PR #174415)

[Wenju He via cfe-commits]

wenju-he wrote:

> Yep, they're supposed to be generic wrappers around a lot of common GPU intrinsics, similar to what OpenCL would provide but done through a header. Like `__gpu_thread_id(__GPU_DIM_X)` which correctly dispatches to the intrinsic for AMDGCN / PTX. This is already used in the libc / openmp libraries.

thanks, it LGTM that common code is unified at a place.


https://github.com/llvm/llvm-project/pull/174415