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

[Simon Pilgrim via cfe-commits]

================
@@ -3,50 +3,250 @@
 // RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +avx512ifma -emit-llvm -o - -Wall -Werror | FileCheck %s
 // RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +avx512ifma -emit-llvm -o - -Wall -Werror | FileCheck %s
 
-// RUN: %clang_cc1 -x c -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +avx512ifma -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s
-// RUN: %clang_cc1 -x c -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +avx512ifma -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s
-// RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +avx512ifma -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s
-// RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +avx512ifma -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s
-
-
 #include <immintrin.h>
+#include "builtin_test_helpers.h"
 
 __m512i test_mm512_madd52hi_epu64(__m512i __X, __m512i __Y, __m512i __Z) {
   // CHECK-LABEL: test_mm512_madd52hi_epu64
   // CHECK: call {{.*}}<8 x i64> @llvm.x86.avx512.vpmadd52h.uq.512(<8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i64> %{{.*}})
-  return _mm512_madd52hi_epu64(__X, __Y, __Z); 
+  return _mm512_madd52hi_epu64(__X, __Y, __Z);
 }
 
+TEST_CONSTEXPR(match_v8di(_mm512_madd52hi_epu64(
+                              (__m512i)(__v8du){100, 0, 0, 0, 0, 0, 0, 0},
+                              (__m512i)(__v8du){10, 0, 0, 0, 0, 0, 0, 0},
+                              (__m512i)(__v8du){5, 0, 0, 0, 0, 0, 0, 0}),
+                          100, 0, 0, 0, 0, 0, 0, 0),
+               "mm512_madd52hi_epu64: high bits remain in accumulator");
----------------
RKSimon wrote:

drop the static_assert messages - its unnecessary inaccurate

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