1 ; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
3 define void @vst2i8(i8* %A, <8 x i8>* %B) nounwind {
5 ;Check the alignment value. Max for this instruction is 128 bits:
6 ;CHECK: vst2.8 {d16, d17}, [r0, :64]
7 %tmp1 = load <8 x i8>* %B
8 call void @llvm.arm.neon.vst2.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
12 define void @vst2i16(i16* %A, <4 x i16>* %B) nounwind {
14 ;Check the alignment value. Max for this instruction is 128 bits:
15 ;CHECK: vst2.16 {d16, d17}, [r0, :128]
16 %tmp0 = bitcast i16* %A to i8*
17 %tmp1 = load <4 x i16>* %B
18 call void @llvm.arm.neon.vst2.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 32)
22 define void @vst2i32(i32* %A, <2 x i32>* %B) nounwind {
25 %tmp0 = bitcast i32* %A to i8*
26 %tmp1 = load <2 x i32>* %B
27 call void @llvm.arm.neon.vst2.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
31 define void @vst2f(float* %A, <2 x float>* %B) nounwind {
34 %tmp0 = bitcast float* %A to i8*
35 %tmp1 = load <2 x float>* %B
36 call void @llvm.arm.neon.vst2.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
40 define void @vst2i64(i64* %A, <1 x i64>* %B) nounwind {
42 ;Check the alignment value. Max for this instruction is 128 bits:
43 ;CHECK: vst1.64 {d16, d17}, [r0, :128]
44 %tmp0 = bitcast i64* %A to i8*
45 %tmp1 = load <1 x i64>* %B
46 call void @llvm.arm.neon.vst2.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 32)
50 define void @vst2Qi8(i8* %A, <16 x i8>* %B) nounwind {
52 ;Check the alignment value. Max for this instruction is 256 bits:
53 ;CHECK: vst2.8 {d16, d17, d18, d19}, [r0, :64]
54 %tmp1 = load <16 x i8>* %B
55 call void @llvm.arm.neon.vst2.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 8)
59 define void @vst2Qi16(i16* %A, <8 x i16>* %B) nounwind {
61 ;Check the alignment value. Max for this instruction is 256 bits:
62 ;CHECK: vst2.16 {d16, d17, d18, d19}, [r0, :128]
63 %tmp0 = bitcast i16* %A to i8*
64 %tmp1 = load <8 x i16>* %B
65 call void @llvm.arm.neon.vst2.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 16)
69 define void @vst2Qi32(i32* %A, <4 x i32>* %B) nounwind {
71 ;Check the alignment value. Max for this instruction is 256 bits:
72 ;CHECK: vst2.32 {d16, d17, d18, d19}, [r0, :256]
73 %tmp0 = bitcast i32* %A to i8*
74 %tmp1 = load <4 x i32>* %B
75 call void @llvm.arm.neon.vst2.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 64)
79 define void @vst2Qf(float* %A, <4 x float>* %B) nounwind {
82 %tmp0 = bitcast float* %A to i8*
83 %tmp1 = load <4 x float>* %B
84 call void @llvm.arm.neon.vst2.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
88 declare void @llvm.arm.neon.vst2.v8i8(i8*, <8 x i8>, <8 x i8>, i32) nounwind
89 declare void @llvm.arm.neon.vst2.v4i16(i8*, <4 x i16>, <4 x i16>, i32) nounwind
90 declare void @llvm.arm.neon.vst2.v2i32(i8*, <2 x i32>, <2 x i32>, i32) nounwind
91 declare void @llvm.arm.neon.vst2.v2f32(i8*, <2 x float>, <2 x float>, i32) nounwind
92 declare void @llvm.arm.neon.vst2.v1i64(i8*, <1 x i64>, <1 x i64>, i32) nounwind
94 declare void @llvm.arm.neon.vst2.v16i8(i8*, <16 x i8>, <16 x i8>, i32) nounwind
95 declare void @llvm.arm.neon.vst2.v8i16(i8*, <8 x i16>, <8 x i16>, i32) nounwind
96 declare void @llvm.arm.neon.vst2.v4i32(i8*, <4 x i32>, <4 x i32>, i32) nounwind
97 declare void @llvm.arm.neon.vst2.v4f32(i8*, <4 x float>, <4 x float>, i32) nounwind