[clang] [AMDGPU][clang] provide device implementation for __builtin_logb and … (PR #129347)

via cfe-commits cfe-commits at lists.llvm.org
Fri Mar 14 13:16:27 PDT 2025


choikwa wrote:

> Are you trying to do an IR lowering? It'd probably make more sense to handle that in LLVM than clang, but I guess Matt would be the expert there.

I had considered it but it has the issue of name aliasing since it just appears as an external call in LLVM after it's emitted by clang. 

https://github.com/llvm/llvm-project/pull/129347


More information about the cfe-commits mailing list