+
+define arm_aapcs_vfpcc void @test_vget_laneu16() nounwind {
+entry:
+; CHECK: vmov.u16 r0, d0[1]
+ %arg0_uint16x4_t = alloca <4 x i16> ; <<4 x i16>*> [#uses=1]
+ %out_uint16_t = alloca i16 ; <i16*> [#uses=1]
+ %"alloca point" = bitcast i32 0 to i32 ; <i32> [#uses=0]
+ %0 = load <4 x i16>* %arg0_uint16x4_t, align 8 ; <<4 x i16>> [#uses=1]
+ %1 = extractelement <4 x i16> %0, i32 1 ; <i16> [#uses=1]
+ store i16 %1, i16* %out_uint16_t, align 2
+ br label %return
+
+return: ; preds = %entry
+ ret void
+}
+
+define arm_aapcs_vfpcc void @test_vget_laneu8() nounwind {
+entry:
+; CHECK: vmov.u8 r0, d0[1]
+ %arg0_uint8x8_t = alloca <8 x i8> ; <<8 x i8>*> [#uses=1]
+ %out_uint8_t = alloca i8 ; <i8*> [#uses=1]
+ %"alloca point" = bitcast i32 0 to i32 ; <i32> [#uses=0]
+ %0 = load <8 x i8>* %arg0_uint8x8_t, align 8 ; <<8 x i8>> [#uses=1]
+ %1 = extractelement <8 x i8> %0, i32 1 ; <i8> [#uses=1]
+ store i8 %1, i8* %out_uint8_t, align 1
+ br label %return
+
+return: ; preds = %entry
+ ret void
+}
+
+define arm_aapcs_vfpcc void @test_vgetQ_laneu16() nounwind {
+entry:
+; CHECK: vmov.u16 r0, d0[1]
+ %arg0_uint16x8_t = alloca <8 x i16> ; <<8 x i16>*> [#uses=1]
+ %out_uint16_t = alloca i16 ; <i16*> [#uses=1]
+ %"alloca point" = bitcast i32 0 to i32 ; <i32> [#uses=0]
+ %0 = load <8 x i16>* %arg0_uint16x8_t, align 16 ; <<8 x i16>> [#uses=1]
+ %1 = extractelement <8 x i16> %0, i32 1 ; <i16> [#uses=1]
+ store i16 %1, i16* %out_uint16_t, align 2
+ br label %return
+
+return: ; preds = %entry
+ ret void
+}
+
+define arm_aapcs_vfpcc void @test_vgetQ_laneu8() nounwind {
+entry:
+; CHECK: vmov.u8 r0, d0[1]
+ %arg0_uint8x16_t = alloca <16 x i8> ; <<16 x i8>*> [#uses=1]
+ %out_uint8_t = alloca i8 ; <i8*> [#uses=1]
+ %"alloca point" = bitcast i32 0 to i32 ; <i32> [#uses=0]
+ %0 = load <16 x i8>* %arg0_uint8x16_t, align 16 ; <<16 x i8>> [#uses=1]
+ %1 = extractelement <16 x i8> %0, i32 1 ; <i8> [#uses=1]
+ store i8 %1, i8* %out_uint8_t, align 1
+ br label %return
+
+return: ; preds = %entry
+ ret void
+}
+
+define <8 x i8> @vset_lane8(<8 x i8>* %A, i8 %B) nounwind {
+;CHECK: vset_lane8:
+;CHECK: vmov.8
+ %tmp1 = load <8 x i8>* %A
+ %tmp2 = insertelement <8 x i8> %tmp1, i8 %B, i32 1
+ ret <8 x i8> %tmp2
+}
+
+define <4 x i16> @vset_lane16(<4 x i16>* %A, i16 %B) nounwind {
+;CHECK: vset_lane16:
+;CHECK: vmov.16
+ %tmp1 = load <4 x i16>* %A
+ %tmp2 = insertelement <4 x i16> %tmp1, i16 %B, i32 1
+ ret <4 x i16> %tmp2
+}
+
+define <2 x i32> @vset_lane32(<2 x i32>* %A, i32 %B) nounwind {
+;CHECK: vset_lane32:
+;CHECK: vmov.32
+ %tmp1 = load <2 x i32>* %A
+ %tmp2 = insertelement <2 x i32> %tmp1, i32 %B, i32 1
+ ret <2 x i32> %tmp2
+}
+
+define <16 x i8> @vsetQ_lane8(<16 x i8>* %A, i8 %B) nounwind {
+;CHECK: vsetQ_lane8:
+;CHECK: vmov.8
+ %tmp1 = load <16 x i8>* %A
+ %tmp2 = insertelement <16 x i8> %tmp1, i8 %B, i32 1
+ ret <16 x i8> %tmp2
+}
+
+define <8 x i16> @vsetQ_lane16(<8 x i16>* %A, i16 %B) nounwind {
+;CHECK: vsetQ_lane16:
+;CHECK: vmov.16
+ %tmp1 = load <8 x i16>* %A
+ %tmp2 = insertelement <8 x i16> %tmp1, i16 %B, i32 1
+ ret <8 x i16> %tmp2
+}
+
+define <4 x i32> @vsetQ_lane32(<4 x i32>* %A, i32 %B) nounwind {
+;CHECK: vsetQ_lane32:
+;CHECK: vmov.32
+ %tmp1 = load <4 x i32>* %A
+ %tmp2 = insertelement <4 x i32> %tmp1, i32 %B, i32 1
+ ret <4 x i32> %tmp2
+}
+
+define arm_aapcs_vfpcc <2 x float> @test_vset_lanef32(float %arg0_float32_t, <2 x float> %arg1_float32x2_t) nounwind {
+;CHECK: test_vset_lanef32:
+;CHECK: vmov.f32
+;CHECK: vmov.f32
+entry:
+ %0 = insertelement <2 x float> %arg1_float32x2_t, float %arg0_float32_t, i32 1 ; <<2 x float>> [#uses=1]
+ ret <2 x float> %0
+}