define <8 x i8> @vld4i8(i8* %A) nounwind {
;CHECK: vld4i8:
-;CHECK: vld4.8
- %tmp1 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8(i8* %A)
+;Check the alignment value. Max for this instruction is 256 bits:
+;CHECK: vld4.8 {d0, d1, d2, d3}, [r0, :64]
+ %tmp1 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8(i8* %A, i32 8)
%tmp2 = extractvalue %struct.__neon_int8x8x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_int8x8x4_t %tmp1, 2
%tmp4 = add <8 x i8> %tmp2, %tmp3
define <4 x i16> @vld4i16(i16* %A) nounwind {
;CHECK: vld4i16:
-;CHECK: vld4.16
- %tmp1 = call %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4.v4i16(i16* %A)
+;Check the alignment value. Max for this instruction is 256 bits:
+;CHECK: vld4.16 {d0, d1, d2, d3}, [r0, :128]
+ %tmp0 = bitcast i16* %A to i8*
+ %tmp1 = call %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4.v4i16(i8* %tmp0, i32 16)
%tmp2 = extractvalue %struct.__neon_int16x4x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_int16x4x4_t %tmp1, 2
%tmp4 = add <4 x i16> %tmp2, %tmp3
define <2 x i32> @vld4i32(i32* %A) nounwind {
;CHECK: vld4i32:
-;CHECK: vld4.32
- %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i32* %A)
+;Check the alignment value. Max for this instruction is 256 bits:
+;CHECK: vld4.32 {d0, d1, d2, d3}, [r0, :256]
+ %tmp0 = bitcast i32* %A to i8*
+ %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8* %tmp0, i32 32)
%tmp2 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 2
%tmp4 = add <2 x i32> %tmp2, %tmp3
define <2 x float> @vld4f(float* %A) nounwind {
;CHECK: vld4f:
;CHECK: vld4.32
- %tmp1 = call %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4.v2f32(float* %A)
+ %tmp0 = bitcast float* %A to i8*
+ %tmp1 = call %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4.v2f32(i8* %tmp0, i32 1)
%tmp2 = extractvalue %struct.__neon_float32x2x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_float32x2x4_t %tmp1, 2
- %tmp4 = add <2 x float> %tmp2, %tmp3
+ %tmp4 = fadd <2 x float> %tmp2, %tmp3
ret <2 x float> %tmp4
}
define <1 x i64> @vld4i64(i64* %A) nounwind {
;CHECK: vld4i64:
-;CHECK: vld1.64
- %tmp1 = call %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64(i64* %A)
+;Check the alignment value. Max for this instruction is 256 bits:
+;CHECK: vld1.64 {d0, d1, d2, d3}, [r0, :256]
+ %tmp0 = bitcast i64* %A to i8*
+ %tmp1 = call %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64(i8* %tmp0, i32 64)
%tmp2 = extractvalue %struct.__neon_int64x1x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_int64x1x4_t %tmp1, 2
%tmp4 = add <1 x i64> %tmp2, %tmp3
define <16 x i8> @vld4Qi8(i8* %A) nounwind {
;CHECK: vld4Qi8:
-;CHECK: vld4.8
-;CHECK: vld4.8
- %tmp1 = call %struct.__neon_int8x16x4_t @llvm.arm.neon.vld4.v16i8(i8* %A)
+;Check the alignment value. Max for this instruction is 256 bits:
+;CHECK: vld4.8 {d0, d2, d4, d6}, [r0, :256]!
+;CHECK: vld4.8 {d1, d3, d5, d7}, [r0, :256]
+ %tmp1 = call %struct.__neon_int8x16x4_t @llvm.arm.neon.vld4.v16i8(i8* %A, i32 64)
%tmp2 = extractvalue %struct.__neon_int8x16x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_int8x16x4_t %tmp1, 2
%tmp4 = add <16 x i8> %tmp2, %tmp3
define <8 x i16> @vld4Qi16(i16* %A) nounwind {
;CHECK: vld4Qi16:
-;CHECK: vld4.16
-;CHECK: vld4.16
- %tmp1 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16(i16* %A)
+;Check for no alignment specifier.
+;CHECK: vld4.16 {d0, d2, d4, d6}, [r0]!
+;CHECK: vld4.16 {d1, d3, d5, d7}, [r0]
+ %tmp0 = bitcast i16* %A to i8*
+ %tmp1 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16(i8* %tmp0, i32 1)
%tmp2 = extractvalue %struct.__neon_int16x8x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_int16x8x4_t %tmp1, 2
%tmp4 = add <8 x i16> %tmp2, %tmp3
;CHECK: vld4Qi32:
;CHECK: vld4.32
;CHECK: vld4.32
- %tmp1 = call %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4.v4i32(i32* %A)
+ %tmp0 = bitcast i32* %A to i8*
+ %tmp1 = call %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4.v4i32(i8* %tmp0, i32 1)
%tmp2 = extractvalue %struct.__neon_int32x4x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_int32x4x4_t %tmp1, 2
%tmp4 = add <4 x i32> %tmp2, %tmp3
;CHECK: vld4Qf:
;CHECK: vld4.32
;CHECK: vld4.32
- %tmp1 = call %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4.v4f32(float* %A)
+ %tmp0 = bitcast float* %A to i8*
+ %tmp1 = call %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4.v4f32(i8* %tmp0, i32 1)
%tmp2 = extractvalue %struct.__neon_float32x4x4_t %tmp1, 0
%tmp3 = extractvalue %struct.__neon_float32x4x4_t %tmp1, 2
- %tmp4 = add <4 x float> %tmp2, %tmp3
+ %tmp4 = fadd <4 x float> %tmp2, %tmp3
ret <4 x float> %tmp4
}
-declare %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8(i8*) nounwind readonly
-declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4.v4i16(i8*) nounwind readonly
-declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8*) nounwind readonly
-declare %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4.v2f32(i8*) nounwind readonly
-declare %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64(i8*) nounwind readonly
+declare %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4.v4i16(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8*, i32) nounwind readonly
+declare %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4.v2f32(i8*, i32) nounwind readonly
+declare %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64(i8*, i32) nounwind readonly
-declare %struct.__neon_int8x16x4_t @llvm.arm.neon.vld4.v16i8(i8*) nounwind readonly
-declare %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16(i8*) nounwind readonly
-declare %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4.v4i32(i8*) nounwind readonly
-declare %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4.v4f32(i8*) nounwind readonly
+declare %struct.__neon_int8x16x4_t @llvm.arm.neon.vld4.v16i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4.v4i32(i8*, i32) nounwind readonly
+declare %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4.v4f32(i8*, i32) nounwind readonly