Index: clang/lib/CodeGen/CGExprScalar.cpp
===================================================================
--- clang/lib/CodeGen/CGExprScalar.cpp
+++ clang/lib/CodeGen/CGExprScalar.cpp
@@ -4796,12 +4796,10 @@
   // to vec4 if the original type is not vec4, then a shuffle vector to
   // get a vec3.
   if (NumElementsSrc != 3 && NumElementsDst == 3) {
-    if (!CGF.CGM.getCodeGenOpts().PreserveVec3Type) {
-      auto *Vec4Ty = llvm::FixedVectorType::get(
-          cast<llvm::VectorType>(DstTy)->getElementType(), 4);
-      Src = createCastsForTypeOfSameSize(Builder, CGF.CGM.getDataLayout(), Src,
-                                         Vec4Ty);
-    }
+    auto *Vec4Ty = llvm::FixedVectorType::get(
+        cast<llvm::VectorType>(DstTy)->getElementType(), 4);
+    Src = createCastsForTypeOfSameSize(Builder, CGF.CGM.getDataLayout(), Src,
+                                       Vec4Ty);
 
     Src = ConvertVec3AndVec4(Builder, CGF, Src, 3);
     Src->setName("astype");
Index: clang/test/CodeGenOpenCL/preserve_vec3.cl
===================================================================
--- clang/test/CodeGenOpenCL/preserve_vec3.cl
+++ clang/test/CodeGenOpenCL/preserve_vec3.cl
@@ -1,6 +1,7 @@
 // RUN: %clang_cc1 %s -emit-llvm -o - -triple spir-unknown-unknown -fpreserve-vec3-type | FileCheck %s
 
 typedef char char3 __attribute__((ext_vector_type(3)));
+typedef char char8 __attribute__((ext_vector_type(8)));
 typedef short short3 __attribute__((ext_vector_type(3)));
 typedef double double2 __attribute__((ext_vector_type(2)));
 typedef float float3 __attribute__((ext_vector_type(3)));
@@ -38,6 +39,15 @@
   *b = __builtin_astype(*a, double2);
 }
 
+void kernel char8_to_short3(global short3 *a, global char8 *b) {
+  // CHECK-LABEL: spir_kernel void @char8_to_short3
+  // CHECK: %[[IN_BC:.*]] = bitcast <8 x i8> addrspace(1)* %b to <4 x i16> addrspace(1)*
+  // CHECK: %[[LOAD_B:.*]] = load <4 x i16>, <4 x i16> addrspace(1)* %[[IN_BC]]
+  // CHECK: %[[ASTYPE:.*]] = shufflevector <4 x i16> %[[LOAD_B]], <4 x i16> poison, <3 x i32> <i32 0, i32 1, i32 2>
+  // CHECK: store <3 x i16> %[[ASTYPE]], <3 x i16> addrspace(1)* %a, align 8
+  *a = __builtin_astype(*b, short3);
+}
+
 void from_char3(char3 a, global int *out) {
   // CHECK-LABEL: void @from_char3
   // CHECK: %[[ASTYPE:.*]] = shufflevector <3 x i8> %a, <3 x i8> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 undef>
@@ -53,3 +63,19 @@
   // CHECK: store <4 x i16> %[[ASTYPE]], <4 x i16> addrspace(1)* %[[OUT_BC]]
   *out = __builtin_astype(a, long);
 }
+
+void scalar_to_char3(int a, global char3 *out) {
+  // CHECK-LABEL: void @scalar_to_char3
+  // CHECK: %[[IN_BC:.*]] = bitcast i32 %a to <4 x i8>
+  // CHECK: %[[ASTYPE:.*]] = shufflevector <4 x i8> %[[IN_BC]], <4 x i8> poison, <3 x i32> <i32 0, i32 1, i32 2>
+  // CHECK: store <3 x i8> %[[ASTYPE]], <3 x i8> addrspace(1)* %out
+  *out = __builtin_astype(a, char3);
+}
+
+void scalar_to_short3(long a, global short3 *out) {
+  // CHECK-LABEL: void @scalar_to_short3
+  // CHECK: %[[IN_BC:.*]] = bitcast i64 %a to <4 x i16>
+  // CHECK: %[[ASTYPE:.*]] = shufflevector <4 x i16> %[[IN_BC]], <4 x i16> poison, <3 x i32> <i32 0, i32 1, i32 2>
+  // CHECK: store <3 x i16> %[[ASTYPE]], <3 x i16> addrspace(1)* %out
+  *out = __builtin_astype(a, short3);
+}