[PATCH] D37985: [AMDGPU] add LDS f32 intrinsics

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

Since it's atomic it needs to be added to isIntrinsicSourceOfDivergence. Also getTgtMemIntrinsic, and getAddrModeArguments (all with associated tests)



================
Comment at: include/llvm/IR/IntrinsicsAMDGPU.td:303
+class AMDGPUAtomicF32IntrinNORET : Intrinsic<[],
+    [LLVMQualPointerType<llvm_float_ty, 3>, llvm_float_ty],
+    [IntrArgMemOnly, NoCapture<0>, IntrNoReturn]
----------------
Should this have an operand added for the ordering?


================
Comment at: include/llvm/IR/IntrinsicsAMDGPU.td:304
+    [LLVMQualPointerType<llvm_float_ty, 3>, llvm_float_ty],
+    [IntrArgMemOnly, NoCapture<0>, IntrNoReturn]
+>;
----------------
This is certainly not IntrNoReturn


https://reviews.llvm.org/D37985