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

[Simon Pilgrim via cfe-commits]

================
@@ -2,6 +2,10 @@
 // RUN: %clang_cc1 -x c -ffreestanding %s -triple=i386-apple-darwin -target-feature +avxifma -emit-llvm -o - -Wall -Werror | FileCheck %s
 // RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +avxifma -emit-llvm -o - -Wall -Werror | FileCheck %s
 // RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=i386-apple-darwin -target-feature +avxifma -emit-llvm -o - -Wall -Werror | FileCheck %s
+// RUN: %clang_cc1 -x c++ -std=c++14 -fsyntax-only %s -triple=x86_64-unknown-unknown -target-feature +avxifma -ffreestanding -verify
+// RUN: %clang_cc1 -x c++ -std=c++14 -fsyntax-only %s -triple=x86_64-unknown-unknown -target-feature +avxifma -ffreestanding -fexperimental-new-constant-interpreter -verify
----------------
RKSimon wrote:

when you update you should now be able to remove these lines and reply on the new existing fexperimental-new-constant-interpreter RUNs

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