[PATCH] D130096: [Clang][AMDGPU] Emit AMDGPU library control constants in clang

[Matt Arsenault via Phabricator via cfe-commits]

arsenm added inline comments.


================
Comment at: clang/lib/CodeGen/TargetInfo.cpp:9461-9463
+  bool CorrectSqrt = CGM.getLangOpts().OpenCL
+                         ? CGM.getCodeGenOpts().OpenCLCorrectlyRoundedDivSqrt
+                         : CGM.getCodeGenOpts().HIPCorrectlyRoundedDivSqrt;
----------------
Can we move this into something more proper in LangOpts?


================
Comment at: clang/lib/CodeGen/TargetInfo.cpp:9467
+  AddGlobal("__oclc_daz_opt", DenormAreZero, /*Size=*/8);
+  AddGlobal("__oclc_finite_only_opt", FiniteOnly || RelaxedMath, /*Size=*/8);
+  AddGlobal("__oclc_unsafe_math_opt", UnsafeMath || RelaxedMath, /*Size=*/8);
----------------
I'd hope you don't have to check relaxed math, finite only should suffice


================
Comment at: clang/lib/CodeGen/TargetInfo.cpp:9476
+            llvm::GlobalValue::LinkOnceODRLinkage);
+  AddGlobal("__oclc_ABI_version",
+            CGM.getTarget().getTargetOpts().CodeObjectVersion, /*Size=*/32,
----------------
This should probably get an __llvm_amdgcn prefix and be renamed


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D130096/new/

https://reviews.llvm.org/D130096