1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon -fast-isel=0 -O0 %s -o - | FileCheck %s
3 define void @vst3i8(i8* %A, <8 x i8>* %B) nounwind {
5 ;Check the alignment value. Max for this instruction is 64 bits:
6 ;This test runs at -O0 so do not check for specific register numbers.
7 ;CHECK: vst3.8 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}:64]
8 %tmp1 = load <8 x i8>* %B
9 call void @llvm.arm.neon.vst3.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 32)
13 define void @vst3i16(i16* %A, <4 x i16>* %B) nounwind {
14 ;CHECK-LABEL: vst3i16:
16 %tmp0 = bitcast i16* %A to i8*
17 %tmp1 = load <4 x i16>* %B
18 call void @llvm.arm.neon.vst3.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1)
22 define void @vst3i32(i32* %A, <2 x i32>* %B) nounwind {
23 ;CHECK-LABEL: vst3i32:
25 %tmp0 = bitcast i32* %A to i8*
26 %tmp1 = load <2 x i32>* %B
27 call void @llvm.arm.neon.vst3.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
31 ;Check for a post-increment updating store.
32 define void @vst3i32_update(i32** %ptr, <2 x i32>* %B) nounwind {
33 ;CHECK-LABEL: vst3i32_update:
34 ;CHECK: vst3.32 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}]!
36 %tmp0 = bitcast i32* %A to i8*
37 %tmp1 = load <2 x i32>* %B
38 call void @llvm.arm.neon.vst3.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
39 %tmp2 = getelementptr i32* %A, i32 6
40 store i32* %tmp2, i32** %ptr
44 define void @vst3f(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.vst3.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
53 define void @vst3i64(i64* %A, <1 x i64>* %B) nounwind {
54 ;CHECK-LABEL: vst3i64:
55 ;Check the alignment value. Max for this instruction is 64 bits:
56 ;This test runs at -O0 so do not check for specific register numbers.
57 ;CHECK: vst1.64 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}:64]
58 %tmp0 = bitcast i64* %A to i8*
59 %tmp1 = load <1 x i64>* %B
60 call void @llvm.arm.neon.vst3.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 16)
64 define void @vst3i64_update(i64** %ptr, <1 x i64>* %B) nounwind {
65 ;CHECK-LABEL: vst3i64_update
66 ;CHECK: vst1.64 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}]!
68 %tmp0 = bitcast i64* %A to i8*
69 %tmp1 = load <1 x i64>* %B
70 call void @llvm.arm.neon.vst3.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 1)
71 %tmp2 = getelementptr i64* %A, i32 3
72 store i64* %tmp2, i64** %ptr
76 define void @vst3Qi8(i8* %A, <16 x i8>* %B) nounwind {
77 ;CHECK-LABEL: vst3Qi8:
78 ;Check the alignment value. Max for this instruction is 64 bits:
79 ;This test runs at -O0 so do not check for specific register numbers.
80 ;CHECK: vst3.8 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}:64]!
81 ;CHECK: vst3.8 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}:64]
82 %tmp1 = load <16 x i8>* %B
83 call void @llvm.arm.neon.vst3.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 32)
87 define void @vst3Qi16(i16* %A, <8 x i16>* %B) nounwind {
88 ;CHECK-LABEL: vst3Qi16:
91 %tmp0 = bitcast i16* %A to i8*
92 %tmp1 = load <8 x i16>* %B
93 call void @llvm.arm.neon.vst3.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
97 ;Check for a post-increment updating store.
98 define void @vst3Qi16_update(i16** %ptr, <8 x i16>* %B) nounwind {
99 ;CHECK-LABEL: vst3Qi16_update:
100 ;CHECK: vst3.16 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}]!
101 ;CHECK: vst3.16 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}]!
103 %tmp0 = bitcast i16* %A to i8*
104 %tmp1 = load <8 x i16>* %B
105 call void @llvm.arm.neon.vst3.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
106 %tmp2 = getelementptr i16* %A, i32 24
107 store i16* %tmp2, i16** %ptr
111 define void @vst3Qi32(i32* %A, <4 x i32>* %B) nounwind {
112 ;CHECK-LABEL: vst3Qi32:
115 %tmp0 = bitcast i32* %A to i8*
116 %tmp1 = load <4 x i32>* %B
117 call void @llvm.arm.neon.vst3.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 1)
121 define void @vst3Qf(float* %A, <4 x float>* %B) nounwind {
122 ;CHECK-LABEL: vst3Qf:
125 %tmp0 = bitcast float* %A to i8*
126 %tmp1 = load <4 x float>* %B
127 call void @llvm.arm.neon.vst3.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
131 declare void @llvm.arm.neon.vst3.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
132 declare void @llvm.arm.neon.vst3.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind
133 declare void @llvm.arm.neon.vst3.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
134 declare void @llvm.arm.neon.vst3.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, i32) nounwind
135 declare void @llvm.arm.neon.vst3.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, i32) nounwind
137 declare void @llvm.arm.neon.vst3.v16i8(i8*, <16 x i8>, <16 x i8>, <16 x i8>, i32) nounwind
138 declare void @llvm.arm.neon.vst3.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind
139 declare void @llvm.arm.neon.vst3.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
140 declare void @llvm.arm.neon.vst3.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, i32) nounwind