[clang] [X86][Clang] VectorExprEvaluator::VisitCallExpr / InterpretBuiltin - Allow AVX/AVX512 IFMA madd52 intrinsics to be used in constexpr (PR #161056)

[Simon Pilgrim via cfe-commits]

================
@@ -22,6 +22,14 @@
   __attribute__((__always_inline__, __nodebug__, __target__("avxifma"),        \
                  __min_vector_width__(256)))
 
+#if defined(__cplusplus) && (__cplusplus >= 201103L)
+#define __DEFAULT_FN_ATTRS256_CONSTEXPR constexpr __DEFAULT_FN_ATTRS256
+#define __DEFAULT_FN_ATTRS128_CONSTEXPR constexpr __DEFAULT_FN_ATTRS128
+#else
+#define __DEFAULT_FN_ATTRS256_CONSTEXPR __DEFAULT_FN_ATTRS256
+#define __DEFAULT_FN_ATTRS128_CONSTEXPR __DEFAULT_FN_ATTRS128
+#endif
----------------
RKSimon wrote:

unnecessary _CONSTEXPR variants - like the other headers all of these are now constexpr

https://github.com/llvm/llvm-project/pull/161056