Squashed commit of the following:

commit 8d41d93e3fceb3f3af77266f5a8388fc585150a5
Author: Pol Marcet Sardà <polmarcetsarda@gmail.com>
Date:   Sat Apr 20 12:19:49 2024 +0200

    Address some misc comments; added a diagnostic and expanded macros in
    testing.

commit 9493c0f290b558947d8b3ae8e1adf909b0fb9dcd
Author: Pol Marcet Sardà <polmarcetsarda@gmail.com>
Date:   Sun Mar 31 18:18:45 2024 +0200

    Following the review of sethp, I have made the following changes:

    -- Added diagnostic for the undefined shuffle of -1
    -- Validated support for _BitInt
    -- A bunch of other minnor tweaks here and there

commit 8273abc8d56ef8225cf4dba84f66a1e54a2ef036
Author: Pol Marcet Sardà <polmarcetsarda@gmail.com>
Date:   Thu Jan 4 12:31:08 2024 +0100

    Fix typo in file name

commit ff68f23921966c7d9605f91a47d6b481bf1d7a7b
Author: Pol Marcet Sardà <polmarcetsarda@gmail.com>
Date:   Thu Jan 4 11:26:08 2024 +0100

    Address suggestions from RKSimon

commit c14783de45687c754253c0cbf8a7834c7f986d80
Author: Pol Marcet Sardà <polmarcetsarda@gmail.com>
Date:   Sat Dec 30 13:59:00 2023 +0100

    [clang] Constexpr for __builtin_shufflevector and __builtin_convertvector

    Summary:

    This patch adds constexpr support for __builtin_shufflevector
    and __builtin_convertvector.

    A small oddity encountered was that the arg to the intrinsics may be an
    lvalue without any sort of implicit cast of any kind. I solved this
    through the EvaluateVectorOrLValue function, which treats the lvalue as
    if it was in an rvalue cast, which gets me the desired vector.

Co-Authored-By: Seth Pellegrino <seth@codecopse.net>

6 files changed, 863 insertions, 5 deletions

diff --git a/clang/docs/LanguageExtensions.rst b/clang/docs/LanguageExtensions.rst
index 87cb743..c2e90f4 100644
--- a/clang/docs/LanguageExtensions.rst
+++ b/clang/docs/LanguageExtensions.rst
@@ -2931,7 +2931,7 @@ Query for this feature with ``__has_builtin(__builtin_dump_struct)``
 ``__builtin_shufflevector`` is used to express generic vector
 permutation/shuffle/swizzle operations.  This builtin is also very important
 for the implementation of various target-specific header files like
-``<xmmintrin.h>``.
+``<xmmintrin.h>``. This builtin can be used within constant expressions.
 
 **Syntax**:
 
@@ -2958,7 +2958,7 @@ for the implementation of various target-specific header files like
   // Concatenate every other element of 8-element vectors V1 and V2.
   __builtin_shufflevector(V1, V2, 0, 2, 4, 6, 8, 10, 12, 14)
 
-  // Shuffle v1 with some elements being undefined
+  // Shuffle v1 with some elements being undefined. Not allowed in constexpr.
   __builtin_shufflevector(v1, v1, 3, -1, 1, -1)
 
 **Description**:
@@ -2971,6 +2971,7 @@ starting with the first vector, continuing into the second vector.  Thus, if
 ``vec1`` is a 4-element vector, index 5 would refer to the second element of
 ``vec2``. An index of -1 can be used to indicate that the corresponding element
 in the returned vector is a don't care and can be optimized by the backend.
+Values of -1 are not supported in constant expressions.
 
 The result of ``__builtin_shufflevector`` is a vector with the same element
 type as ``vec1``/``vec2`` but that has an element count equal to the number of
@@ -2985,7 +2986,8 @@ Query for this feature with ``__has_builtin(__builtin_shufflevector)``.
 
 ``__builtin_convertvector`` is used to express generic vector
 type-conversion operations. The input vector and the output vector
-type must have the same number of elements.
+type must have the same number of elements. This builtin can be used within
+constant expressions.
 
 **Syntax**:
 
diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index 4cb2462..2d0e663 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -233,6 +233,9 @@ Non-comprehensive list of changes in this release
 * ``-fdenormal-fp-math=preserve-sign`` is no longer implied by ``-ffast-math``
   on x86 systems.
 
+- Builtins ``__builtin_shufflevector()`` and ``__builtin_convertvector()`` may
+  now be used within constant expressions.
+
 New Compiler Flags
 ------------------
 - ``-fsanitize=implicit-bitfield-conversion`` checks implicit truncation and
diff --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td b/clang/include/clang/Basic/DiagnosticSemaKinds.td
index 8486fa3..502c771 100644
--- a/clang/include/clang/Basic/DiagnosticSemaKinds.td
+++ b/clang/include/clang/Basic/DiagnosticSemaKinds.td
@@ -10330,9 +10330,13 @@ def err_shufflevector_nonconstant_argument : Error<
 def err_shufflevector_argument_too_large : Error<
   "index for __builtin_shufflevector must be less than the total number "
   "of vector elements">;
+def err_shufflevector_minus_one_is_undefined_behavior_constexpr : Error<
+  "index for __builtin_shufflevector not within the bounds of the input vectors; index of -1 found at position %0 not permitted in a constexpr context.">;
 
 def err_convertvector_non_vector : Error<
   "first argument to __builtin_convertvector must be a vector">;
+def err_convertvector_constexpr_unsupported_vector_cast : Error<
+  "unsupported vector cast from %0 to %1 in a constant expression.">;
 def err_builtin_non_vector_type : Error<
   "%0 argument to %1 must be of vector type">;
 def err_convertvector_incompatible_vector : Error<
diff --git a/clang/lib/AST/ExprConstant.cpp b/clang/lib/AST/ExprConstant.cpp
index 71b5f91..f1aa19e 100644
--- a/clang/lib/AST/ExprConstant.cpp
+++ b/clang/lib/AST/ExprConstant.cpp
@@ -2706,7 +2706,11 @@ static bool checkFloatingPointResult(EvalInfo &Info, const Expr *E,
 static bool HandleFloatToFloatCast(EvalInfo &Info, const Expr *E,
                                    QualType SrcType, QualType DestType,
                                    APFloat &Result) {
-  assert(isa<CastExpr>(E) || isa<CompoundAssignOperator>(E));
+  assert((isa<CastExpr>(E) || isa<CompoundAssignOperator>(E) ||
+          isa<ConvertVectorExpr>(E)) &&
+         "HandleFloatToFloatCast has been checked with only CastExpr, "
+         "CompoundAssignOperator and ConvertVectorExpr. Please either validate "
+         "the new expression or address the root cause of this usage.");
   llvm::RoundingMode RM = getActiveRoundingMode(Info, E);
   APFloat::opStatus St;
   APFloat Value = Result;
@@ -10710,8 +10714,11 @@ namespace {
     bool VisitUnaryImag(const UnaryOperator *E);
     bool VisitBinaryOperator(const BinaryOperator *E);
     bool VisitUnaryOperator(const UnaryOperator *E);
+    bool VisitConvertVectorExpr(const ConvertVectorExpr *E);
+    bool VisitShuffleVectorExpr(const ShuffleVectorExpr *E);
+
     // FIXME: Missing: conditional operator (for GNU
-    //                 conditional select), shufflevector, ExtVectorElementExpr
+    //                 conditional select), ExtVectorElementExpr
   };
 } // end anonymous namespace
 
@@ -10962,6 +10969,122 @@ bool VectorExprEvaluator::VisitUnaryOperator(const UnaryOperator *E) {
   return Success(APValue(ResultElements.data(), ResultElements.size()), E);
 }
 
+static bool handleVectorElementCast(EvalInfo &Info, const FPOptions FPO,
+                                    const Expr *E, QualType SourceTy,
+                                    QualType DestTy, APValue const &Original,
+                                    APValue &Result) {
+  if (SourceTy->isIntegerType()) {
+    if (DestTy->isRealFloatingType()) {
+      Result = APValue(APFloat(0.0));
+      return HandleIntToFloatCast(Info, E, FPO, SourceTy, Original.getInt(),
+                                  DestTy, Result.getFloat());
+    }
+    if (DestTy->isIntegerType()) {
+      Result = APValue(
+          HandleIntToIntCast(Info, E, DestTy, SourceTy, Original.getInt()));
+      return true;
+    }
+  } else if (SourceTy->isRealFloatingType()) {
+    if (DestTy->isRealFloatingType()) {
+      Result = Original;
+      return HandleFloatToFloatCast(Info, E, SourceTy, DestTy,
+                                    Result.getFloat());
+    }
+    if (DestTy->isIntegerType()) {
+      Result = APValue(APSInt());
+      return HandleFloatToIntCast(Info, E, SourceTy, Original.getFloat(),
+                                  DestTy, Result.getInt());
+    }
+  }
+
+  Info.FFDiag(E, diag::err_convertvector_constexpr_unsupported_vector_cast)
+      << SourceTy << DestTy;
+  return false;
+}
+
+bool VectorExprEvaluator::VisitConvertVectorExpr(const ConvertVectorExpr *E) {
+  APValue Source;
+  QualType SourceVecType = E->getSrcExpr()->getType();
+  if (!EvaluateAsRValue(Info, E->getSrcExpr(), Source))
+    return false;
+
+  QualType DestTy = E->getType()->castAs<VectorType>()->getElementType();
+  QualType SourceTy = SourceVecType->castAs<VectorType>()->getElementType();
+
+  const FPOptions FPO = E->getFPFeaturesInEffect(Info.Ctx.getLangOpts());
+
+  auto SourceLen = Source.getVectorLength();
+  SmallVector<APValue, 4> ResultElements;
+  ResultElements.reserve(SourceLen);
+  for (unsigned EltNum = 0; EltNum < SourceLen; ++EltNum) {
+    APValue Elt;
+    if (!handleVectorElementCast(Info, FPO, E, SourceTy, DestTy,
+                                 Source.getVectorElt(EltNum), Elt))
+      return false;
+    ResultElements.push_back(std::move(Elt));
+  }
+
+  return Success(APValue(ResultElements.data(), ResultElements.size()), E);
+}
+
+static bool handleVectorShuffle(EvalInfo &Info, const ShuffleVectorExpr *E,
+                                QualType ElemType, APValue const &VecVal1,
+                                APValue const &VecVal2, unsigned EltNum,
+                                APValue &Result) {
+  unsigned const TotalElementsInInputVector1 = VecVal1.getVectorLength();
+  unsigned const TotalElementsInInputVector2 = VecVal2.getVectorLength();
+
+  APSInt IndexVal = E->getShuffleMaskIdx(Info.Ctx, EltNum);
+  int64_t index = IndexVal.getExtValue();
+  // The spec says that -1 should be treated as undef for optimizations,
+  // but in constexpr we'd have to produce an APValue::Indeterminate,
+  // which is prohibited from being a top-level constant value. Emit a
+  // diagnostic instead.
+  if (index == -1) {
+    Info.FFDiag(
+        E, diag::err_shufflevector_minus_one_is_undefined_behavior_constexpr)
+        << EltNum;
+    return false;
+  }
+
+  if (index < 0 ||
+      index >= TotalElementsInInputVector1 + TotalElementsInInputVector2)
+    llvm_unreachable("Out of bounds shuffle index");
+
+  if (index >= TotalElementsInInputVector1)
+    Result = VecVal2.getVectorElt(index - TotalElementsInInputVector1);
+  else
+    Result = VecVal1.getVectorElt(index);
+  return true;
+}
+
+bool VectorExprEvaluator::VisitShuffleVectorExpr(const ShuffleVectorExpr *E) {
+  APValue VecVal1;
+  const Expr *Vec1 = E->getExpr(0);
+  if (!EvaluateAsRValue(Info, Vec1, VecVal1))
+    return false;
+  APValue VecVal2;
+  const Expr *Vec2 = E->getExpr(1);
+  if (!EvaluateAsRValue(Info, Vec2, VecVal2))
+    return false;
+
+  VectorType const *DestVecTy = E->getType()->castAs<VectorType>();
+  QualType DestElTy = DestVecTy->getElementType();
+
+  auto TotalElementsInOutputVector = DestVecTy->getNumElements();
+
+  SmallVector<APValue, 4> ResultElements;
+  ResultElements.reserve(TotalElementsInOutputVector);
+  for (unsigned EltNum = 0; EltNum < TotalElementsInOutputVector; ++EltNum) {
+    APValue Elt;
+    if (!handleVectorShuffle(Info, E, DestElTy, VecVal1, VecVal2, EltNum, Elt))
+      return false;
+    ResultElements.push_back(std::move(Elt));
+  }
+
+  return Success(APValue(ResultElements.data(), ResultElements.size()), E);
+}
+
 //===----------------------------------------------------------------------===//
 // Array Evaluation
 //===----------------------------------------------------------------------===//
diff --git a/clang/test/Sema/constant_builtins_vector.cpp b/clang/test/Sema/constant_builtins_vector.cpp
new file mode 100644
index 0000000..68620d4
--- /dev/null
+++ b/clang/test/Sema/constant_builtins_vector.cpp
@@ -0,0 +1,723 @@
+// RUN: %clang_cc1 -verify -std=c++2a -fsyntax-only -Wno-bit-int-extension %s
+
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
+#define LITTLE_END 1
+#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+#define LITTLE_END 0
+#else
+#error "huh?"
+#endif
+
+// We also support _BitInt as long as it is >=8 and a power of 2.
+typedef _BitInt(8) BitInt8;
+typedef _BitInt(32) BitInt32;
+typedef _BitInt(128) BitInt128;
+
+typedef double vector4double __attribute__((__vector_size__(32)));
+typedef float vector4float __attribute__((__vector_size__(16)));
+typedef long long vector4long __attribute__((__vector_size__(32)));
+typedef int vector4int __attribute__((__vector_size__(16)));
+typedef short vector4short __attribute__((__vector_size__(8)));
+typedef char vector4char __attribute__((__vector_size__(4)));
+typedef BitInt8 vector4BitInt8 __attribute__((__vector_size__(4)));
+typedef BitInt32 vector4BitInt32 __attribute__((__vector_size__(16)));
+typedef BitInt128 vector4BitInt128 __attribute__((__vector_size__(64)));
+typedef double vector8double __attribute__((__vector_size__(64)));
+typedef float vector8float __attribute__((__vector_size__(32)));
+typedef long long vector8long __attribute__((__vector_size__(64)));
+typedef int vector8int __attribute__((__vector_size__(32)));
+typedef short vector8short __attribute__((__vector_size__(16)));
+typedef char vector8char __attribute__((__vector_size__(8)));
+typedef BitInt8 vector8BitInt8 __attribute__((__vector_size__(8)));
+typedef BitInt32 vector8BitInt32 __attribute__((__vector_size__(32)));
+typedef BitInt128 vector8BitInt128 __attribute__((__vector_size__(128)));
+
+#define CHECK_NUM(__size, __typeFrom, __typeTo, ...)                            \
+  constexpr vector##__size##__typeTo                                            \
+      from_##vector##__size##__typeFrom##_to_##vector##__size##__typeTo##_var = \
+          __builtin_convertvector((vector##__size##__typeFrom){__VA_ARGS__},    \
+                                  vector##__size##__typeTo);
+#define CHECK_TO_ALL_TYPES(__size, __typeFrom, ...)                            \
+  CHECK_NUM(__size, __typeFrom, double, __VA_ARGS__)                           \
+  CHECK_NUM(__size, __typeFrom, float, __VA_ARGS__)                            \
+  CHECK_NUM(__size, __typeFrom, long, __VA_ARGS__)                             \
+  CHECK_NUM(__size, __typeFrom, int, __VA_ARGS__)                              \
+  CHECK_NUM(__size, __typeFrom, short, __VA_ARGS__)                            \
+  CHECK_NUM(__size, __typeFrom, char, __VA_ARGS__)                             \
+  CHECK_NUM(__size, __typeFrom, BitInt8, __VA_ARGS__)                          \
+  CHECK_NUM(__size, __typeFrom, BitInt32, __VA_ARGS__)                         \
+  CHECK_NUM(__size, __typeFrom, BitInt128, __VA_ARGS__)                        \
+  static_assert(                                                               \
+      __builtin_bit_cast(                                                      \
+          unsigned,                                                            \
+          __builtin_shufflevector(                                             \
+              from_vector##__size##__typeFrom##_to_vector##__size##char_var,   \
+              from_vector##__size##__typeFrom##_to_vector##__size##char_var,   \
+              0, 1, 2, 3)) == (LITTLE_END ? 0x03020100 : 0x00010203));         \
+  static_assert(                                                               \
+      __builtin_bit_cast(                                                      \
+          unsigned long long,                                                  \
+          __builtin_shufflevector(                                             \
+              from_vector##__size##__typeFrom##_to_vector##__size##short_var,  \
+              from_vector##__size##__typeFrom##_to_vector##__size##short_var,  \
+              0, 1, 2, 3)) ==                                                  \
+      (LITTLE_END ? 0x0003000200010000 : 0x0000000100020003));
+
+#define CHECK_ALL_COMBINATIONS(__size, ...)                                    \
+  CHECK_TO_ALL_TYPES(__size, double, __VA_ARGS__)                              \
+  CHECK_TO_ALL_TYPES(__size, float, __VA_ARGS__)                               \
+  CHECK_TO_ALL_TYPES(__size, long, __VA_ARGS__)                                \
+  CHECK_TO_ALL_TYPES(__size, int, __VA_ARGS__)                                 \
+  CHECK_TO_ALL_TYPES(__size, short, __VA_ARGS__)                               \
+  CHECK_TO_ALL_TYPES(__size, char, __VA_ARGS__)                                \
+  CHECK_TO_ALL_TYPES(__size, BitInt8, __VA_ARGS__)                             \
+  CHECK_TO_ALL_TYPES(__size, BitInt32, __VA_ARGS__)                            \
+  CHECK_TO_ALL_TYPES(__size, BitInt128, __VA_ARGS__)
+
+// The result below is expanded from these macros. Use them to autogenerate the
+// test cases below.
+// CHECK_ALL_COMBINATIONS(4, 0, 1, 2, 3);
+// CHECK_ALL_COMBINATIONS(8, 0, 1, 2, 3, 4, 5, 6, 7);
+
+constexpr vector4double from_vector4double_to_vector4double_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4double_to_vector4float_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4double_to_vector4long_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4double_to_vector4int_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4double_to_vector4short_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4double_to_vector4char_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4double_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4double_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4double_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(
+                  unsigned,
+                  __builtin_shufflevector(from_vector4double_to_vector4char_var,
+                                          from_vector4double_to_vector4char_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector4double_to_vector4short_var,
+                                     from_vector4double_to_vector4short_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4float_to_vector4double_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4float_to_vector4float_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4float_to_vector4long_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4float_to_vector4int_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4float_to_vector4short_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4float_to_vector4char_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4float_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4float_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4float_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4float_to_vector4char_var,
+                                     from_vector4float_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4float_to_vector4short_var,
+                                          from_vector4float_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4long_to_vector4double_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4long_to_vector4float_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4long_to_vector4long_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4long_to_vector4int_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4long_to_vector4short_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4long_to_vector4char_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4long_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4long_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4long_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4long_to_vector4char_var,
+                                     from_vector4long_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4long_to_vector4short_var,
+                                          from_vector4long_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4int_to_vector4double_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4int_to_vector4float_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4int_to_vector4long_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4int_to_vector4int_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4int_to_vector4short_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4int_to_vector4char_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4int_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4int_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4int_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4int_to_vector4char_var,
+                                     from_vector4int_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4int_to_vector4short_var,
+                                          from_vector4int_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4short_to_vector4double_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4short_to_vector4float_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4short_to_vector4long_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4short_to_vector4int_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4short_to_vector4short_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4short_to_vector4char_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4short_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4short_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4short_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4short_to_vector4char_var,
+                                     from_vector4short_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4short_to_vector4short_var,
+                                          from_vector4short_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4char_to_vector4double_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4char_to_vector4float_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4char_to_vector4long_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4char_to_vector4int_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4char_to_vector4short_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4char_to_vector4char_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4char_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4char_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4char_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4char_to_vector4char_var,
+                                     from_vector4char_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4char_to_vector4short_var,
+                                          from_vector4char_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4BitInt8_to_vector4double_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4BitInt8_to_vector4float_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4BitInt8_to_vector4long_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4BitInt8_to_vector4int_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4BitInt8_to_vector4short_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4BitInt8_to_vector4char_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4BitInt8_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4BitInt8_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4BitInt8_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt8_to_vector4char_var,
+                                     from_vector4BitInt8_to_vector4char_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt8_to_vector4short_var,
+                                     from_vector4BitInt8_to_vector4short_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4BitInt32_to_vector4double_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4BitInt32_to_vector4float_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4BitInt32_to_vector4long_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4BitInt32_to_vector4int_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4BitInt32_to_vector4short_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4BitInt32_to_vector4char_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4BitInt32_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4BitInt32_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4BitInt32_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt32_to_vector4char_var,
+                                     from_vector4BitInt32_to_vector4char_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt32_to_vector4short_var,
+                                     from_vector4BitInt32_to_vector4short_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4BitInt128_to_vector4double_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4BitInt128_to_vector4float_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4BitInt128_to_vector4long_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4BitInt128_to_vector4int_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4BitInt128_to_vector4short_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4BitInt128_to_vector4char_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4BitInt128_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4BitInt128_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4BitInt128_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt128_to_vector4char_var,
+                                     from_vector4BitInt128_to_vector4char_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt128_to_vector4short_var,
+                                     from_vector4BitInt128_to_vector4short_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+;
+constexpr vector8double from_vector8double_to_vector8double_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8double_to_vector8float_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8double_to_vector8long_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8double_to_vector8int_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8int);
+constexpr vector8short from_vector8double_to_vector8short_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8double_to_vector8char_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8double_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8double_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8double_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(
+                  unsigned,
+                  __builtin_shufflevector(from_vector8double_to_vector8char_var,
+                                          from_vector8double_to_vector8char_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector8double_to_vector8short_var,
+                                     from_vector8double_to_vector8short_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8float_to_vector8double_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8float_to_vector8float_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8float_to_vector8long_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8float_to_vector8int_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8float_to_vector8short_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8float_to_vector8char_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8float_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8float_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8float_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8float_to_vector8char_var,
+                                     from_vector8float_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8float_to_vector8short_var,
+                                          from_vector8float_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8long_to_vector8double_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8long_to_vector8float_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8long_to_vector8long_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7}, vector8long);
+constexpr vector8int from_vector8long_to_vector8int_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8long_to_vector8short_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8long_to_vector8char_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7}, vector8char);
+constexpr vector8BitInt8 from_vector8long_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8long_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8long_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8long_to_vector8char_var,
+                                     from_vector8long_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8long_to_vector8short_var,
+                                          from_vector8long_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8int_to_vector8double_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8int_to_vector8float_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8float);
+constexpr vector8long from_vector8int_to_vector8long_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8long);
+constexpr vector8int from_vector8int_to_vector8int_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8int_to_vector8short_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8short);
+constexpr vector8char from_vector8int_to_vector8char_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8char);
+constexpr vector8BitInt8 from_vector8int_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8int_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8int_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8int_to_vector8char_var,
+                                     from_vector8int_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8int_to_vector8short_var,
+                                          from_vector8int_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8short_to_vector8double_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8short_to_vector8float_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8short_to_vector8long_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8short_to_vector8int_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8short_to_vector8short_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8short_to_vector8char_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8short_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8short_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8short_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8short_to_vector8char_var,
+                                     from_vector8short_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8short_to_vector8short_var,
+                                          from_vector8short_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8char_to_vector8double_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8char_to_vector8float_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8char_to_vector8long_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7}, vector8long);
+constexpr vector8int from_vector8char_to_vector8int_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8char_to_vector8short_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8char_to_vector8char_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7}, vector8char);
+constexpr vector8BitInt8 from_vector8char_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8char_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8char_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8char_to_vector8char_var,
+                                     from_vector8char_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8char_to_vector8short_var,
+                                          from_vector8char_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8BitInt8_to_vector8double_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8BitInt8_to_vector8float_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8BitInt8_to_vector8long_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8BitInt8_to_vector8int_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8int);
+constexpr vector8short from_vector8BitInt8_to_vector8short_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8BitInt8_to_vector8char_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8BitInt8_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8BitInt8_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8BitInt8_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt8_to_vector8char_var,
+                                     from_vector8BitInt8_to_vector8char_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt8_to_vector8short_var,
+                                     from_vector8BitInt8_to_vector8short_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8BitInt32_to_vector8double_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8BitInt32_to_vector8float_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8BitInt32_to_vector8long_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8BitInt32_to_vector8int_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8int);
+constexpr vector8short from_vector8BitInt32_to_vector8short_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8BitInt32_to_vector8char_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8BitInt32_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8BitInt32_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8BitInt32_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt32_to_vector8char_var,
+                                     from_vector8BitInt32_to_vector8char_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt32_to_vector8short_var,
+                                     from_vector8BitInt32_to_vector8short_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8BitInt128_to_vector8double_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8BitInt128_to_vector8float_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8BitInt128_to_vector8long_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8BitInt128_to_vector8int_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8int);
+constexpr vector8short from_vector8BitInt128_to_vector8short_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8BitInt128_to_vector8char_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8BitInt128_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8BitInt128_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8BitInt128_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt128_to_vector8char_var,
+                                     from_vector8BitInt128_to_vector8char_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt128_to_vector8short_var,
+                                     from_vector8BitInt128_to_vector8short_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+;
+#undef CHECK_ALL_COMBINATIONS
+#undef CHECK_TO_ALL_TYPES
+#undef CHECK_NUM
+
+// Shuffle vector
+constexpr vector4char vector4charConst1 = {0, 1, 2, 3};
+constexpr vector4char vector4charConst2 = {4, 5, 6, 7};
+constexpr vector8char vector8intConst = {8, 9, 10, 11, 12, 13, 14, 15};
+
+constexpr vector4char vectorShuffle1 =
+    __builtin_shufflevector(vector4charConst1, vector4charConst2, 0, 1, 2, 3);
+static_assert(__builtin_bit_cast(unsigned, vectorShuffle1) ==
+              (LITTLE_END ? 0x03020100 : 0x00010203));
+constexpr vector4char vectorShuffle2 =
+    __builtin_shufflevector(vector4charConst1, vector4charConst2, 4, 5, 6, 7);
+static_assert(__builtin_bit_cast(unsigned, vectorShuffle2) ==
+              (LITTLE_END ? 0x07060504 : 0x04050607));
+constexpr vector4char vectorShuffle3 =
+    __builtin_shufflevector(vector4charConst1, vector4charConst2, 0, 2, 4, 6);
+static_assert(__builtin_bit_cast(unsigned, vectorShuffle3) ==
+              (LITTLE_END ? 0x06040200 : 0x00020406));
+constexpr vector8char vectorShuffle4 = __builtin_shufflevector(
+    vector8intConst, vector8intConst, 0, 2, 4, 6, 8, 10, 12, 14);
+static_assert(__builtin_bit_cast(unsigned long long, vectorShuffle4) ==
+              (LITTLE_END ? 0x0E0C0A080E0C0A08 : 0x080A0C0E080A0C0E));
+constexpr vector4char vectorShuffle5 =
+    __builtin_shufflevector(vector8intConst, vector8intConst, 0, 2, 4, 6);
+static_assert(__builtin_bit_cast(unsigned, vectorShuffle5) ==
+              (LITTLE_END ? 0x0E0C0A08 : 0x080A0C0E));
+constexpr vector8char vectorShuffle6 = __builtin_shufflevector(
+    vector4charConst1, vector4charConst2, 0, 2, 4, 6, 1, 3, 5, 7);
+static_assert(__builtin_bit_cast(unsigned long long, vectorShuffle6) ==
+              (LITTLE_END ? 0x0705030106040200 : 0x0002040601030507));
+
+constexpr vector4char
+    vectorShuffleFail1 = // expected-error {{constexpr variable 'vectorShuffleFail1'\
+ must be initialized by a constant expression}}
+    __builtin_shufflevector( // expected-error {{index for __builtin_shufflevector \
+not within the bounds of the input vectors; index of -1 found at position 0 not \
+permitted in a constexpr context.}}
+        vector4charConst1,
+        vector4charConst2, -1, -1, -1, -1);
diff --git a/clang/test/Sema/convertvector.c b/clang/test/Sema/convertvector.c
index 8ae43c3..1ff04af 100644
--- a/clang/test/Sema/convertvector.c
+++ b/clang/test/Sema/convertvector.c
@@ -15,3 +15,6 @@ vector8float foo3(double x) {
   return __builtin_convertvector(x, vector8float);  // expected-error {{must be a vector}}
 }
 
+float foo4(float x) {
+  return __builtin_convertvector(x, float); // expected-error {{first argument to __builtin_convertvector must be a vector}}
+}