[clang] [llvm] [NVPTX] Add intrinsics and clang builtins for conversions of f4x2 type (PR #139244)

Durgadoss R via llvm-commits llvm-commits at lists.llvm.org
Fri May 9 08:56:53 PDT 2025


================
@@ -1663,6 +1663,13 @@ let TargetPrefix = "nvvm" in {
       def int_nvvm_ # type # _to_f16x2 # suffix : CVT_I16_TO_F16X2<type, suffix>;
     }
   }
+
+  // FP4 conversions.
+  foreach relu = ["", "_relu"] in {
+    defvar suffix = !strconcat("_rn", relu);
----------------
durga4github wrote:

nit: since it is only two variables, may be using # is simpler.

https://github.com/llvm/llvm-project/pull/139244


More information about the llvm-commits mailing list