[clang] [llvm] [clang][NVPTX] Add f16(x2) add/mul FTZ intrinsics (PR #178237)

[Srinivasa Ravi via cfe-commits]

================
@@ -1373,6 +1373,12 @@ let TargetPrefix = "nvvm" in {
       def int_nvvm_mul_rn # ftz # _sat_v2f16 : NVVMBuiltin,
         DefaultAttrsIntrinsic<[llvm_v2f16_ty], [llvm_v2f16_ty, llvm_v2f16_ty]>;
     } // ftz
+    
+    def int_nvvm_mul_rn_ftz_f16 : NVVMBuiltin,
+      DefaultAttrsIntrinsic<[llvm_half_ty], [llvm_half_ty, llvm_half_ty]>;
+
+    def int_nvvm_mul_rn_ftz_v2f16 : NVVMBuiltin,
+      DefaultAttrsIntrinsic<[llvm_v2f16_ty], [llvm_v2f16_ty, llvm_v2f16_ty]>;
----------------
Wolfram70 wrote:

Since this is not urgent, I think we can revisit this PR once https://github.com/llvm/llvm-project/pull/172442 lands, to avoid rewriting these again then.

https://github.com/llvm/llvm-project/pull/178237