1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
|
; RUN: llc < %s -verify-machineinstrs -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s
declare <2 x double> @llvm.arm.neon.vbsl.v2f64(<2 x double>, <2 x double>, <2 x double>)
declare <8 x i16> @llvm.arm.neon.vbsl.v8i16(<8 x i16>, <8 x i16>, <8 x i16>)
declare <16 x i8> @llvm.arm.neon.vbsl.v16i8(<16 x i8>, <16 x i8>, <16 x i8>)
declare <4 x float> @llvm.arm.neon.vbsl.v4f32(<4 x float>, <4 x float>, <4 x float>)
declare <2 x i64> @llvm.arm.neon.vbsl.v2i64(<2 x i64>, <2 x i64>, <2 x i64>)
declare <4 x i32> @llvm.arm.neon.vbsl.v4i32(<4 x i32>, <4 x i32>, <4 x i32>)
declare <4 x i16> @llvm.arm.neon.vbsl.v4i16(<4 x i16>, <4 x i16>, <4 x i16>)
declare <8 x i8> @llvm.arm.neon.vbsl.v8i8(<8 x i8>, <8 x i8>, <8 x i8>)
declare <1 x double> @llvm.arm.neon.vbsl.v1f64(<1 x double>, <1 x double>, <1 x double>)
declare <2 x float> @llvm.arm.neon.vbsl.v2f32(<2 x float>, <2 x float>, <2 x float>)
declare <1 x i64> @llvm.arm.neon.vbsl.v1i64(<1 x i64>, <1 x i64>, <1 x i64>)
declare <2 x i32> @llvm.arm.neon.vbsl.v2i32(<2 x i32>, <2 x i32>, <2 x i32>)
define <8 x i8> @test_vbsl_s8(<8 x i8> %v1, <8 x i8> %v2, <8 x i8> %v3) {
; CHECK-LABEL: test_vbsl_s8:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl.i = tail call <8 x i8> @llvm.arm.neon.vbsl.v8i8(<8 x i8> %v1, <8 x i8> %v2, <8 x i8> %v3)
ret <8 x i8> %vbsl.i
}
define <8 x i8> @test_vbsl_s16(<4 x i16> %v1, <4 x i16> %v2, <4 x i16> %v3) {
; CHECK-LABEL: test_vbsl_s16:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl3.i = tail call <4 x i16> @llvm.arm.neon.vbsl.v4i16(<4 x i16> %v1, <4 x i16> %v2, <4 x i16> %v3)
%0 = bitcast <4 x i16> %vbsl3.i to <8 x i8>
ret <8 x i8> %0
}
define <2 x i32> @test_vbsl_s32(<2 x i32> %v1, <2 x i32> %v2, <2 x i32> %v3) {
; CHECK-LABEL: test_vbsl_s32:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl3.i = tail call <2 x i32> @llvm.arm.neon.vbsl.v2i32(<2 x i32> %v1, <2 x i32> %v2, <2 x i32> %v3)
ret <2 x i32> %vbsl3.i
}
define <1 x i64> @test_vbsl_s64(<1 x i64> %v1, <1 x i64> %v2, <1 x i64> %v3) {
; CHECK-LABEL: test_vbsl_s64:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl3.i = tail call <1 x i64> @llvm.arm.neon.vbsl.v1i64(<1 x i64> %v1, <1 x i64> %v2, <1 x i64> %v3)
ret <1 x i64> %vbsl3.i
}
define <8 x i8> @test_vbsl_u8(<8 x i8> %v1, <8 x i8> %v2, <8 x i8> %v3) {
; CHECK-LABEL: test_vbsl_u8:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl.i = tail call <8 x i8> @llvm.arm.neon.vbsl.v8i8(<8 x i8> %v1, <8 x i8> %v2, <8 x i8> %v3)
ret <8 x i8> %vbsl.i
}
define <4 x i16> @test_vbsl_u16(<4 x i16> %v1, <4 x i16> %v2, <4 x i16> %v3) {
; CHECK-LABEL: test_vbsl_u16:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl3.i = tail call <4 x i16> @llvm.arm.neon.vbsl.v4i16(<4 x i16> %v1, <4 x i16> %v2, <4 x i16> %v3)
ret <4 x i16> %vbsl3.i
}
define <2 x i32> @test_vbsl_u32(<2 x i32> %v1, <2 x i32> %v2, <2 x i32> %v3) {
; CHECK-LABEL: test_vbsl_u32:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl3.i = tail call <2 x i32> @llvm.arm.neon.vbsl.v2i32(<2 x i32> %v1, <2 x i32> %v2, <2 x i32> %v3)
ret <2 x i32> %vbsl3.i
}
define <1 x i64> @test_vbsl_u64(<1 x i64> %v1, <1 x i64> %v2, <1 x i64> %v3) {
; CHECK-LABEL: test_vbsl_u64:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl3.i = tail call <1 x i64> @llvm.arm.neon.vbsl.v1i64(<1 x i64> %v1, <1 x i64> %v2, <1 x i64> %v3)
ret <1 x i64> %vbsl3.i
}
define <2 x float> @test_vbsl_f32(<2 x float> %v1, <2 x float> %v2, <2 x float> %v3) {
; CHECK-LABEL: test_vbsl_f32:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl3.i = tail call <2 x float> @llvm.arm.neon.vbsl.v2f32(<2 x float> %v1, <2 x float> %v2, <2 x float> %v3)
ret <2 x float> %vbsl3.i
}
define <1 x double> @test_vbsl_f64(<1 x i64> %v1, <1 x double> %v2, <1 x double> %v3) {
; CHECK-LABEL: test_vbsl_f64:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl.i = bitcast <1 x i64> %v1 to <1 x double>
%vbsl3.i = tail call <1 x double> @llvm.arm.neon.vbsl.v1f64(<1 x double> %vbsl.i, <1 x double> %v2, <1 x double> %v3)
ret <1 x double> %vbsl3.i
}
define <8 x i8> @test_vbsl_p8(<8 x i8> %v1, <8 x i8> %v2, <8 x i8> %v3) {
; CHECK-LABEL: test_vbsl_p8:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl.i = tail call <8 x i8> @llvm.arm.neon.vbsl.v8i8(<8 x i8> %v1, <8 x i8> %v2, <8 x i8> %v3)
ret <8 x i8> %vbsl.i
}
define <4 x i16> @test_vbsl_p16(<4 x i16> %v1, <4 x i16> %v2, <4 x i16> %v3) {
; CHECK-LABEL: test_vbsl_p16:
; CHECK: bsl {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
entry:
%vbsl3.i = tail call <4 x i16> @llvm.arm.neon.vbsl.v4i16(<4 x i16> %v1, <4 x i16> %v2, <4 x i16> %v3)
ret <4 x i16> %vbsl3.i
}
define <16 x i8> @test_vbslq_s8(<16 x i8> %v1, <16 x i8> %v2, <16 x i8> %v3) {
; CHECK-LABEL: test_vbslq_s8:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl.i = tail call <16 x i8> @llvm.arm.neon.vbsl.v16i8(<16 x i8> %v1, <16 x i8> %v2, <16 x i8> %v3)
ret <16 x i8> %vbsl.i
}
define <8 x i16> @test_vbslq_s16(<8 x i16> %v1, <8 x i16> %v2, <8 x i16> %v3) {
; CHECK-LABEL: test_vbslq_s16:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl3.i = tail call <8 x i16> @llvm.arm.neon.vbsl.v8i16(<8 x i16> %v1, <8 x i16> %v2, <8 x i16> %v3)
ret <8 x i16> %vbsl3.i
}
define <4 x i32> @test_vbslq_s32(<4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3) {
; CHECK-LABEL: test_vbslq_s32:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl3.i = tail call <4 x i32> @llvm.arm.neon.vbsl.v4i32(<4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3)
ret <4 x i32> %vbsl3.i
}
define <2 x i64> @test_vbslq_s64(<2 x i64> %v1, <2 x i64> %v2, <2 x i64> %v3) {
; CHECK-LABEL: test_vbslq_s64:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl3.i = tail call <2 x i64> @llvm.arm.neon.vbsl.v2i64(<2 x i64> %v1, <2 x i64> %v2, <2 x i64> %v3)
ret <2 x i64> %vbsl3.i
}
define <16 x i8> @test_vbslq_u8(<16 x i8> %v1, <16 x i8> %v2, <16 x i8> %v3) {
; CHECK-LABEL: test_vbslq_u8:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl.i = tail call <16 x i8> @llvm.arm.neon.vbsl.v16i8(<16 x i8> %v1, <16 x i8> %v2, <16 x i8> %v3)
ret <16 x i8> %vbsl.i
}
define <8 x i16> @test_vbslq_u16(<8 x i16> %v1, <8 x i16> %v2, <8 x i16> %v3) {
; CHECK-LABEL: test_vbslq_u16:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl3.i = tail call <8 x i16> @llvm.arm.neon.vbsl.v8i16(<8 x i16> %v1, <8 x i16> %v2, <8 x i16> %v3)
ret <8 x i16> %vbsl3.i
}
define <4 x i32> @test_vbslq_u32(<4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3) {
; CHECK-LABEL: test_vbslq_u32:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl3.i = tail call <4 x i32> @llvm.arm.neon.vbsl.v4i32(<4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3)
ret <4 x i32> %vbsl3.i
}
define <2 x i64> @test_vbslq_u64(<2 x i64> %v1, <2 x i64> %v2, <2 x i64> %v3) {
; CHECK-LABEL: test_vbslq_u64:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl3.i = tail call <2 x i64> @llvm.arm.neon.vbsl.v2i64(<2 x i64> %v1, <2 x i64> %v2, <2 x i64> %v3)
ret <2 x i64> %vbsl3.i
}
define <4 x float> @test_vbslq_f32(<4 x i32> %v1, <4 x float> %v2, <4 x float> %v3) {
; CHECK-LABEL: test_vbslq_f32:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl.i = bitcast <4 x i32> %v1 to <4 x float>
%vbsl3.i = tail call <4 x float> @llvm.arm.neon.vbsl.v4f32(<4 x float> %vbsl.i, <4 x float> %v2, <4 x float> %v3)
ret <4 x float> %vbsl3.i
}
define <16 x i8> @test_vbslq_p8(<16 x i8> %v1, <16 x i8> %v2, <16 x i8> %v3) {
; CHECK-LABEL: test_vbslq_p8:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl.i = tail call <16 x i8> @llvm.arm.neon.vbsl.v16i8(<16 x i8> %v1, <16 x i8> %v2, <16 x i8> %v3)
ret <16 x i8> %vbsl.i
}
define <8 x i16> @test_vbslq_p16(<8 x i16> %v1, <8 x i16> %v2, <8 x i16> %v3) {
; CHECK-LABEL: test_vbslq_p16:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl3.i = tail call <8 x i16> @llvm.arm.neon.vbsl.v8i16(<8 x i16> %v1, <8 x i16> %v2, <8 x i16> %v3)
ret <8 x i16> %vbsl3.i
}
define <2 x double> @test_vbslq_f64(<2 x i64> %v1, <2 x double> %v2, <2 x double> %v3) {
; CHECK-LABEL: test_vbslq_f64:
; CHECK: bsl {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
entry:
%vbsl.i = bitcast <2 x i64> %v1 to <2 x double>
%vbsl3.i = tail call <2 x double> @llvm.arm.neon.vbsl.v2f64(<2 x double> %vbsl.i, <2 x double> %v2, <2 x double> %v3)
ret <2 x double> %vbsl3.i
}
|
