diff --git a/clang/include/clang/Basic/Builtins.def b/clang/include/clang/Basic/Builtins.def
--- a/clang/include/clang/Basic/Builtins.def
+++ b/clang/include/clang/Basic/Builtins.def
@@ -648,6 +648,7 @@
 BUILTIN(__builtin_elementwise_min, "v.", "nct")
 BUILTIN(__builtin_reduce_max, "v.", "nct")
 BUILTIN(__builtin_reduce_min, "v.", "nct")
+BUILTIN(__builtin_reduce_xor, "v.", "nct")
 
 BUILTIN(__builtin_matrix_transpose, "v.", "nFt")
 BUILTIN(__builtin_matrix_column_major_load, "v.", "nFt")
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3204,6 +3204,13 @@
     return RValue::get(Result);
   }
 
+  case Builtin::BI__builtin_reduce_xor: {
+    Value *Op0 = EmitScalarExpr(E->getArg(0));
+    Value *Result = Builder.CreateUnaryIntrinsic(
+        llvm::Intrinsic::vector_reduce_xor, Op0, nullptr, "rdx.xor");
+    return RValue::get(Result);
+  }
+
   case Builtin::BI__builtin_matrix_transpose: {
     const auto *MatrixTy = E->getArg(0)->getType()->getAs<ConstantMatrixType>();
     Value *MatValue = EmitScalarExpr(E->getArg(0));
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -2109,6 +2109,7 @@
     break;
   case Builtin::BI__builtin_reduce_max:
   case Builtin::BI__builtin_reduce_min:
+  case Builtin::BI__builtin_reduce_xor:
     if (SemaBuiltinReduceMath(TheCall))
       return ExprError();
     break;
diff --git a/clang/test/CodeGen/builtins-reduction-math.c b/clang/test/CodeGen/builtins-reduction-math.c
--- a/clang/test/CodeGen/builtins-reduction-math.c
+++ b/clang/test/CodeGen/builtins-reduction-math.c
@@ -57,3 +57,14 @@
   const si8 cvi1 = vi1;
   unsigned long long r5 = __builtin_reduce_min(cvi1);
 }
+
+void test_builtin_reduce_xor(float4 vf1, si8 vi1, u4 vu1) {
+
+  // CHECK:      [[VI1:%.+]] = load <8 x i16>, <8 x i16>* %vi1.addr, align 16
+  // CHECK-NEXT: call i16 @llvm.vector.reduce.xor.v8i16(<8 x i16> [[VI1]])
+  short r2 = __builtin_reduce_xor(vi1);
+
+  // CHECK:      [[VU1:%.+]] = load <4 x i32>, <4 x i32>* %vu1.addr, align 16
+  // CHECK-NEXT: call i32 @llvm.vector.reduce.xor.v4i32(<4 x i32> [[VU1]])
+  unsigned r3 = __builtin_reduce_xor(vu1);
+}
diff --git a/clang/test/Sema/builtins-reduction-math.c b/clang/test/Sema/builtins-reduction-math.c
--- a/clang/test/Sema/builtins-reduction-math.c
+++ b/clang/test/Sema/builtins-reduction-math.c
@@ -35,3 +35,17 @@
   i = __builtin_reduce_min(i);
   // expected-error@-1 {{1st argument must be a vector type (was 'int')}}
 }
+
+void test_builtin_reduce_xor(int i, float4 v, int3 iv) {
+  struct Foo s = __builtin_reduce_xor(iv);
+  // expected-error@-1 {{initializing 'struct Foo' with an expression of incompatible type 'int'}}
+
+  i = __builtin_reduce_xor(v, v);
+  // expected-error@-1 {{too many arguments to function call, expected 1, have 2}}
+
+  i = __builtin_reduce_xor();
+  // expected-error@-1 {{too few arguments to function call, expected 1, have 0}}
+
+  i = __builtin_reduce_xor(i);
+  // expected-error@-1 {{1st argument must be a vector type (was 'int')}}
+}