[clang] [Headers][X86] Allow MMX/SSE/AVX MOVMSK intrinsics to be used in constexpr (PR #161914)

[Simon Pilgrim via cfe-commits]

================
@@ -14795,6 +14794,44 @@ bool IntExprEvaluator::VisitBuiltinCallExpr(const CallExpr *E,
     return Success(CarryOut, E);
   }
 
+  case clang::X86::BI__builtin_ia32_movmskps:
+  case clang::X86::BI__builtin_ia32_movmskpd:
+  case clang::X86::BI__builtin_ia32_pmovmskb128:
+  case clang::X86::BI__builtin_ia32_pmovmskb256:
+  case clang::X86::BI__builtin_ia32_movmskps256:
+  case clang::X86::BI__builtin_ia32_movmskpd256: {
+    APValue Source;
+    if (!Evaluate(Source, Info, E->getArg(0)))
+      return false;
+    unsigned SourceLen = Source.getVectorLength();
+    const VectorType *VT = E->getArg(0)->getType()->castAs<VectorType>();
+    const QualType ElemQT = VT->getElementType();
+    unsigned LaneWidth = Info.Ctx.getTypeSize(ElemQT);
+
+    if (ElemQT->isIntegerType()) { // Get MSB of each byte of every lane
+      unsigned Byte = 8;
+      unsigned ResultLen = (LaneWidth * SourceLen) / Byte;
----------------
RKSimon wrote:

Determine ResultLen from Destination type

https://github.com/llvm/llvm-project/pull/161914