1 ; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
3 define void @vst1i8(i8* %A, <8 x i8>* %B) nounwind {
5 ;Check the alignment value. Max for this instruction is 64 bits:
6 ;CHECK: vst1.8 {d16}, [r0, :64]
7 %tmp1 = load <8 x i8>* %B
8 call void @llvm.arm.neon.vst1.v8i8(i8* %A, <8 x i8> %tmp1, i32 16)
12 define void @vst1i16(i16* %A, <4 x i16>* %B) nounwind {
15 %tmp0 = bitcast i16* %A to i8*
16 %tmp1 = load <4 x i16>* %B
17 call void @llvm.arm.neon.vst1.v4i16(i8* %tmp0, <4 x i16> %tmp1, i32 1)
21 define void @vst1i32(i32* %A, <2 x i32>* %B) nounwind {
24 %tmp0 = bitcast i32* %A to i8*
25 %tmp1 = load <2 x i32>* %B
26 call void @llvm.arm.neon.vst1.v2i32(i8* %tmp0, <2 x i32> %tmp1, i32 1)
30 define void @vst1f(float* %A, <2 x float>* %B) nounwind {
33 %tmp0 = bitcast float* %A to i8*
34 %tmp1 = load <2 x float>* %B
35 call void @llvm.arm.neon.vst1.v2f32(i8* %tmp0, <2 x float> %tmp1, i32 1)
39 define void @vst1i64(i64* %A, <1 x i64>* %B) nounwind {
42 %tmp0 = bitcast i64* %A to i8*
43 %tmp1 = load <1 x i64>* %B
44 call void @llvm.arm.neon.vst1.v1i64(i8* %tmp0, <1 x i64> %tmp1, i32 1)
48 define void @vst1Qi8(i8* %A, <16 x i8>* %B) nounwind {
50 ;Check the alignment value. Max for this instruction is 128 bits:
51 ;CHECK: vst1.8 {d16, d17}, [r0, :64]
52 %tmp1 = load <16 x i8>* %B
53 call void @llvm.arm.neon.vst1.v16i8(i8* %A, <16 x i8> %tmp1, i32 8)
57 define void @vst1Qi16(i16* %A, <8 x i16>* %B) nounwind {
59 ;Check the alignment value. Max for this instruction is 128 bits:
60 ;CHECK: vst1.16 {d16, d17}, [r0, :128]
61 %tmp0 = bitcast i16* %A to i8*
62 %tmp1 = load <8 x i16>* %B
63 call void @llvm.arm.neon.vst1.v8i16(i8* %tmp0, <8 x i16> %tmp1, i32 32)
67 define void @vst1Qi32(i32* %A, <4 x i32>* %B) nounwind {
70 %tmp0 = bitcast i32* %A to i8*
71 %tmp1 = load <4 x i32>* %B
72 call void @llvm.arm.neon.vst1.v4i32(i8* %tmp0, <4 x i32> %tmp1, i32 1)
76 define void @vst1Qf(float* %A, <4 x float>* %B) nounwind {
79 %tmp0 = bitcast float* %A to i8*
80 %tmp1 = load <4 x float>* %B
81 call void @llvm.arm.neon.vst1.v4f32(i8* %tmp0, <4 x float> %tmp1, i32 1)
85 define void @vst1Qi64(i64* %A, <2 x i64>* %B) nounwind {
88 %tmp0 = bitcast i64* %A to i8*
89 %tmp1 = load <2 x i64>* %B
90 call void @llvm.arm.neon.vst1.v2i64(i8* %tmp0, <2 x i64> %tmp1, i32 1)
94 declare void @llvm.arm.neon.vst1.v8i8(i8*, <8 x i8>, i32) nounwind
95 declare void @llvm.arm.neon.vst1.v4i16(i8*, <4 x i16>, i32) nounwind
96 declare void @llvm.arm.neon.vst1.v2i32(i8*, <2 x i32>, i32) nounwind
97 declare void @llvm.arm.neon.vst1.v2f32(i8*, <2 x float>, i32) nounwind
98 declare void @llvm.arm.neon.vst1.v1i64(i8*, <1 x i64>, i32) nounwind
100 declare void @llvm.arm.neon.vst1.v16i8(i8*, <16 x i8>, i32) nounwind
101 declare void @llvm.arm.neon.vst1.v8i16(i8*, <8 x i16>, i32) nounwind
102 declare void @llvm.arm.neon.vst1.v4i32(i8*, <4 x i32>, i32) nounwind
103 declare void @llvm.arm.neon.vst1.v4f32(i8*, <4 x float>, i32) nounwind
104 declare void @llvm.arm.neon.vst1.v2i64(i8*, <2 x i64>, i32) nounwind