[compiler-rt] [llvm] [PGO][AMDGPU] Add basic HIP offload PGO support (PR #177665)

[Matt Arsenault via llvm-commits]

================
@@ -160,6 +164,14 @@ cl::opt<bool> SpeculativeCounterPromotionToLoop(
              " update can be further/iteratively promoted into an acyclic "
              " region."));
 
+static cl::opt<unsigned> OffloadPGOSampling(
----------------
arsenm wrote:

Can this be a pass parameter instead 

https://github.com/llvm/llvm-project/pull/177665