1 ; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
3 define void @vst4i8(i8* %A, <8 x i8>* %B) nounwind {
5 ;Check the alignment value. Max for this instruction is 256 bits:
6 ;CHECK: vst4.8 {d16, d17, d18, d19}, [r0, :64]
7 %tmp1 = load <8 x i8>* %B
8 call void @llvm.arm.neon.vst4.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
12 ;Check for a post-increment updating store with register increment.
13 define void @vst4i8_update(i8** %ptr, <8 x i8>* %B, i32 %inc) nounwind {
14 ;CHECK: vst4i8_update:
15 ;CHECK: vst4.8 {d16, d17, d18, d19}, [r1, :128], r2
17 %tmp1 = load <8 x i8>* %B
18 call void @llvm.arm.neon.vst4.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 16)
19 %tmp2 = getelementptr i8* %A, i32 %inc
20 store i8* %tmp2, i8** %ptr
24 define void @vst4i16(i16* %A, <4 x i16>* %B) nounwind {
26 ;Check the alignment value. Max for this instruction is 256 bits:
27 ;CHECK: vst4.16 {d16, d17, d18, d19}, [r0, :128]
28 %tmp0 = bitcast i16* %A to i8*
29 %tmp1 = load <4 x i16>* %B
30 call void @llvm.arm.neon.vst4.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 16)
34 define void @vst4i32(i32* %A, <2 x i32>* %B) nounwind {
36 ;Check the alignment value. Max for this instruction is 256 bits:
37 ;CHECK: vst4.32 {d16, d17, d18, d19}, [r0, :256]
38 %tmp0 = bitcast i32* %A to i8*
39 %tmp1 = load <2 x i32>* %B
40 call void @llvm.arm.neon.vst4.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 32)
44 define void @vst4f(float* %A, <2 x float>* %B) nounwind {
47 %tmp0 = bitcast float* %A to i8*
48 %tmp1 = load <2 x float>* %B
49 call void @llvm.arm.neon.vst4.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
53 define void @vst4i64(i64* %A, <1 x i64>* %B) nounwind {
55 ;Check the alignment value. Max for this instruction is 256 bits:
56 ;CHECK: vst1.64 {d16, d17, d18, d19}, [r0, :256]
57 %tmp0 = bitcast i64* %A to i8*
58 %tmp1 = load <1 x i64>* %B
59 call void @llvm.arm.neon.vst4.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 64)
63 define void @vst4Qi8(i8* %A, <16 x i8>* %B) nounwind {
65 ;Check the alignment value. Max for this instruction is 256 bits:
66 ;CHECK: vst4.8 {d16, d18, d20, d22}, [r0, :256]!
67 ;CHECK: vst4.8 {d17, d19, d21, d23}, [r0, :256]
68 %tmp1 = load <16 x i8>* %B
69 call void @llvm.arm.neon.vst4.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 64)
73 define void @vst4Qi16(i16* %A, <8 x i16>* %B) nounwind {
75 ;Check for no alignment specifier.
76 ;CHECK: vst4.16 {d16, d18, d20, d22}, [r0]!
77 ;CHECK: vst4.16 {d17, d19, d21, d23}, [r0]
78 %tmp0 = bitcast i16* %A to i8*
79 %tmp1 = load <8 x i16>* %B
80 call void @llvm.arm.neon.vst4.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
84 define void @vst4Qi32(i32* %A, <4 x i32>* %B) nounwind {
88 %tmp0 = bitcast i32* %A to i8*
89 %tmp1 = load <4 x i32>* %B
90 call void @llvm.arm.neon.vst4.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 1)
94 define void @vst4Qf(float* %A, <4 x float>* %B) nounwind {
98 %tmp0 = bitcast float* %A to i8*
99 %tmp1 = load <4 x float>* %B
100 call void @llvm.arm.neon.vst4.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
104 ;Check for a post-increment updating store.
105 define void @vst4Qf_update(float** %ptr, <4 x float>* %B) nounwind {
106 ;CHECK: vst4Qf_update:
107 ;CHECK: vst4.32 {d16, d18, d20, d22}, [r1]!
108 ;CHECK: vst4.32 {d17, d19, d21, d23}, [r1]!
109 %A = load float** %ptr
110 %tmp0 = bitcast float* %A to i8*
111 %tmp1 = load <4 x float>* %B
112 call void @llvm.arm.neon.vst4.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
113 %tmp2 = getelementptr float* %A, i32 16
114 store float* %tmp2, float** %ptr
118 declare void @llvm.arm.neon.vst4.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
119 declare void @llvm.arm.neon.vst4.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind
120 declare void @llvm.arm.neon.vst4.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
121 declare void @llvm.arm.neon.vst4.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32) nounwind
122 declare void @llvm.arm.neon.vst4.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64>, i32) nounwind
124 declare void @llvm.arm.neon.vst4.v16i8(i8*, <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8>, i32) nounwind
125 declare void @llvm.arm.neon.vst4.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind
126 declare void @llvm.arm.neon.vst4.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
127 declare void @llvm.arm.neon.vst4.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32) nounwind