[libclc] libclc: Use frexp builtins to implement frexp for amdgpu (PR #185637)

[Matt Arsenault via cfe-commits]

================
@@ -0,0 +1,47 @@
+//===----------------------------------------------------------------------===//
----------------
arsenm wrote:

That's a different notion of implementation though. That's the mechanics for the template instantiation, not the actual algorithmic part 

https://github.com/llvm/llvm-project/pull/185637