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
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
|
; RUN: llc < %s -verify-machineinstrs -mtriple=aarch64-none-linux-gnu -mattr=+neon -fp-contract=fast | FileCheck %s
declare <2 x float> @llvm.fma.v2f32(<2 x float>, <2 x float>, <2 x float>)
declare <4 x float> @llvm.fma.v4f32(<4 x float>, <4 x float>, <4 x float>)
declare <2 x i64> @llvm.arm.neon.vqdmull.v2i64(<2 x i32>, <2 x i32>)
declare <2 x i64> @llvm.arm.neon.vqsubs.v2i64(<2 x i64>, <2 x i64>)
declare <4 x i32> @llvm.arm.neon.vqdmull.v4i32(<4 x i16>, <4 x i16>)
declare <4 x i32> @llvm.arm.neon.vqsubs.v4i32(<4 x i32>, <4 x i32>)
declare <2 x i64> @llvm.arm.neon.vqadds.v2i64(<2 x i64>, <2 x i64>)
declare <4 x i32> @llvm.arm.neon.vqadds.v4i32(<4 x i32>, <4 x i32>)
declare <2 x i64> @llvm.arm.neon.vmullu.v2i64(<2 x i32>, <2 x i32>)
declare <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16>, <4 x i16>)
declare <2 x i64> @llvm.arm.neon.vmulls.v2i64(<2 x i32>, <2 x i32>)
declare <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16>, <4 x i16>)
define <4 x i32> @test_vmull_high_n_s16(<8 x i16> %a, i16 %b) {
; CHECK: test_vmull_high_n_s16:
; CHECK: smull2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[0]
entry:
%shuffle.i.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %b, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %b, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %b, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %b, i32 3
%vmull15.i.i = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
ret <4 x i32> %vmull15.i.i
}
define <2 x i64> @test_vmull_high_n_s32(<4 x i32> %a, i32 %b) {
; CHECK: test_vmull_high_n_s32:
; CHECK: smull2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[0]
entry:
%shuffle.i.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %b, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %b, i32 1
%vmull9.i.i = tail call <2 x i64> @llvm.arm.neon.vmulls.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
ret <2 x i64> %vmull9.i.i
}
define <4 x i32> @test_vmull_high_n_u16(<8 x i16> %a, i16 %b) {
; CHECK: test_vmull_high_n_u16:
; CHECK: umull2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[0]
entry:
%shuffle.i.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %b, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %b, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %b, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %b, i32 3
%vmull15.i.i = tail call <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
ret <4 x i32> %vmull15.i.i
}
define <2 x i64> @test_vmull_high_n_u32(<4 x i32> %a, i32 %b) {
; CHECK: test_vmull_high_n_u32:
; CHECK: umull2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[0]
entry:
%shuffle.i.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %b, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %b, i32 1
%vmull9.i.i = tail call <2 x i64> @llvm.arm.neon.vmullu.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
ret <2 x i64> %vmull9.i.i
}
define <4 x i32> @test_vqdmull_high_n_s16(<8 x i16> %a, i16 %b) {
; CHECK: test_vqdmull_high_n_s16:
; CHECK: sqdmull2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[0]
entry:
%shuffle.i.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %b, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %b, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %b, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %b, i32 3
%vqdmull15.i.i = tail call <4 x i32> @llvm.arm.neon.vqdmull.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
ret <4 x i32> %vqdmull15.i.i
}
define <2 x i64> @test_vqdmull_high_n_s32(<4 x i32> %a, i32 %b) {
; CHECK: test_vqdmull_high_n_s32:
; CHECK: sqdmull2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[0]
entry:
%shuffle.i.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %b, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %b, i32 1
%vqdmull9.i.i = tail call <2 x i64> @llvm.arm.neon.vqdmull.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
ret <2 x i64> %vqdmull9.i.i
}
define <4 x i32> @test_vmlal_high_n_s16(<4 x i32> %a, <8 x i16> %b, i16 %c) {
; CHECK: test_vmlal_high_n_s16:
; CHECK: smlal2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %c, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %c, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %c, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %c, i32 3
%vmull2.i.i.i = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
%add.i.i = add <4 x i32> %vmull2.i.i.i, %a
ret <4 x i32> %add.i.i
}
define <2 x i64> @test_vmlal_high_n_s32(<2 x i64> %a, <4 x i32> %b, i32 %c) {
; CHECK: test_vmlal_high_n_s32:
; CHECK: smlal2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %c, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %c, i32 1
%vmull2.i.i.i = tail call <2 x i64> @llvm.arm.neon.vmulls.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
%add.i.i = add <2 x i64> %vmull2.i.i.i, %a
ret <2 x i64> %add.i.i
}
define <4 x i32> @test_vmlal_high_n_u16(<4 x i32> %a, <8 x i16> %b, i16 %c) {
; CHECK: test_vmlal_high_n_u16:
; CHECK: umlal2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %c, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %c, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %c, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %c, i32 3
%vmull2.i.i.i = tail call <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
%add.i.i = add <4 x i32> %vmull2.i.i.i, %a
ret <4 x i32> %add.i.i
}
define <2 x i64> @test_vmlal_high_n_u32(<2 x i64> %a, <4 x i32> %b, i32 %c) {
; CHECK: test_vmlal_high_n_u32:
; CHECK: umlal2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %c, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %c, i32 1
%vmull2.i.i.i = tail call <2 x i64> @llvm.arm.neon.vmullu.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
%add.i.i = add <2 x i64> %vmull2.i.i.i, %a
ret <2 x i64> %add.i.i
}
define <4 x i32> @test_vqdmlal_high_n_s16(<4 x i32> %a, <8 x i16> %b, i16 %c) {
; CHECK: test_vqdmlal_high_n_s16:
; CHECK: sqdmlal2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %c, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %c, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %c, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %c, i32 3
%vqdmlal15.i.i = tail call <4 x i32> @llvm.arm.neon.vqdmull.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
%vqdmlal17.i.i = tail call <4 x i32> @llvm.arm.neon.vqadds.v4i32(<4 x i32> %a, <4 x i32> %vqdmlal15.i.i)
ret <4 x i32> %vqdmlal17.i.i
}
define <2 x i64> @test_vqdmlal_high_n_s32(<2 x i64> %a, <4 x i32> %b, i32 %c) {
; CHECK: test_vqdmlal_high_n_s32:
; CHECK: sqdmlal2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %c, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %c, i32 1
%vqdmlal9.i.i = tail call <2 x i64> @llvm.arm.neon.vqdmull.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
%vqdmlal11.i.i = tail call <2 x i64> @llvm.arm.neon.vqadds.v2i64(<2 x i64> %a, <2 x i64> %vqdmlal9.i.i)
ret <2 x i64> %vqdmlal11.i.i
}
define <4 x i32> @test_vmlsl_high_n_s16(<4 x i32> %a, <8 x i16> %b, i16 %c) {
; CHECK: test_vmlsl_high_n_s16:
; CHECK: smlsl2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %c, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %c, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %c, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %c, i32 3
%vmull2.i.i.i = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
%sub.i.i = sub <4 x i32> %a, %vmull2.i.i.i
ret <4 x i32> %sub.i.i
}
define <2 x i64> @test_vmlsl_high_n_s32(<2 x i64> %a, <4 x i32> %b, i32 %c) {
; CHECK: test_vmlsl_high_n_s32:
; CHECK: smlsl2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %c, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %c, i32 1
%vmull2.i.i.i = tail call <2 x i64> @llvm.arm.neon.vmulls.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
%sub.i.i = sub <2 x i64> %a, %vmull2.i.i.i
ret <2 x i64> %sub.i.i
}
define <4 x i32> @test_vmlsl_high_n_u16(<4 x i32> %a, <8 x i16> %b, i16 %c) {
; CHECK: test_vmlsl_high_n_u16:
; CHECK: umlsl2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %c, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %c, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %c, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %c, i32 3
%vmull2.i.i.i = tail call <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
%sub.i.i = sub <4 x i32> %a, %vmull2.i.i.i
ret <4 x i32> %sub.i.i
}
define <2 x i64> @test_vmlsl_high_n_u32(<2 x i64> %a, <4 x i32> %b, i32 %c) {
; CHECK: test_vmlsl_high_n_u32:
; CHECK: umlsl2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %c, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %c, i32 1
%vmull2.i.i.i = tail call <2 x i64> @llvm.arm.neon.vmullu.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
%sub.i.i = sub <2 x i64> %a, %vmull2.i.i.i
ret <2 x i64> %sub.i.i
}
define <4 x i32> @test_vqdmlsl_high_n_s16(<4 x i32> %a, <8 x i16> %b, i16 %c) {
; CHECK: test_vqdmlsl_high_n_s16:
; CHECK: sqdmlsl2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, {{v[0-9]+}}.h[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <8 x i16> %b, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%vecinit.i.i = insertelement <4 x i16> undef, i16 %c, i32 0
%vecinit1.i.i = insertelement <4 x i16> %vecinit.i.i, i16 %c, i32 1
%vecinit2.i.i = insertelement <4 x i16> %vecinit1.i.i, i16 %c, i32 2
%vecinit3.i.i = insertelement <4 x i16> %vecinit2.i.i, i16 %c, i32 3
%vqdmlsl15.i.i = tail call <4 x i32> @llvm.arm.neon.vqdmull.v4i32(<4 x i16> %shuffle.i.i, <4 x i16> %vecinit3.i.i)
%vqdmlsl17.i.i = tail call <4 x i32> @llvm.arm.neon.vqsubs.v4i32(<4 x i32> %a, <4 x i32> %vqdmlsl15.i.i)
ret <4 x i32> %vqdmlsl17.i.i
}
define <2 x i64> @test_vqdmlsl_high_n_s32(<2 x i64> %a, <4 x i32> %b, i32 %c) {
; CHECK: test_vqdmlsl_high_n_s32:
; CHECK: sqdmlsl2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%shuffle.i.i = shufflevector <4 x i32> %b, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
%vecinit.i.i = insertelement <2 x i32> undef, i32 %c, i32 0
%vecinit1.i.i = insertelement <2 x i32> %vecinit.i.i, i32 %c, i32 1
%vqdmlsl9.i.i = tail call <2 x i64> @llvm.arm.neon.vqdmull.v2i64(<2 x i32> %shuffle.i.i, <2 x i32> %vecinit1.i.i)
%vqdmlsl11.i.i = tail call <2 x i64> @llvm.arm.neon.vqsubs.v2i64(<2 x i64> %a, <2 x i64> %vqdmlsl9.i.i)
ret <2 x i64> %vqdmlsl11.i.i
}
define <2 x float> @test_vmul_n_f32(<2 x float> %a, float %b) {
; CHECK: test_vmul_n_f32:
; CHECK: fmul {{v[0-9]+}}.2s, {{v[0-9]+}}.2s, {{v[0-9]+}}.s[0]
entry:
%vecinit.i = insertelement <2 x float> undef, float %b, i32 0
%vecinit1.i = insertelement <2 x float> %vecinit.i, float %b, i32 1
%mul.i = fmul <2 x float> %vecinit1.i, %a
ret <2 x float> %mul.i
}
define <4 x float> @test_vmulq_n_f32(<4 x float> %a, float %b) {
; CHECK: test_vmulq_n_f32:
; CHECK: fmul {{v[0-9]+}}.4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[0]
entry:
%vecinit.i = insertelement <4 x float> undef, float %b, i32 0
%vecinit1.i = insertelement <4 x float> %vecinit.i, float %b, i32 1
%vecinit2.i = insertelement <4 x float> %vecinit1.i, float %b, i32 2
%vecinit3.i = insertelement <4 x float> %vecinit2.i, float %b, i32 3
%mul.i = fmul <4 x float> %vecinit3.i, %a
ret <4 x float> %mul.i
}
define <2 x double> @test_vmulq_n_f64(<2 x double> %a, double %b) {
; CHECK: test_vmulq_n_f64:
; CHECK: fmul {{v[0-9]+}}.2d, {{v[0-9]+}}.2d, {{v[0-9]+}}.d[0]
entry:
%vecinit.i = insertelement <2 x double> undef, double %b, i32 0
%vecinit1.i = insertelement <2 x double> %vecinit.i, double %b, i32 1
%mul.i = fmul <2 x double> %vecinit1.i, %a
ret <2 x double> %mul.i
}
define <2 x float> @test_vfma_n_f32(<2 x float> %a, <2 x float> %b, float %n) {
; CHECK: test_vfma_n_f32:
; CHECK: fmla {{v[0-9]+}}.2s, {{v[0-9]+}}.2s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%vecinit.i = insertelement <2 x float> undef, float %n, i32 0
%vecinit1.i = insertelement <2 x float> %vecinit.i, float %n, i32 1
%0 = tail call <2 x float> @llvm.fma.v2f32(<2 x float> %b, <2 x float> %vecinit1.i, <2 x float> %a)
ret <2 x float> %0
}
define <4 x float> @test_vfmaq_n_f32(<4 x float> %a, <4 x float> %b, float %n) {
; CHECK: test_vfmaq_n_f32:
; CHECK: fmla {{v[0-9]+}}.4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%vecinit.i = insertelement <4 x float> undef, float %n, i32 0
%vecinit1.i = insertelement <4 x float> %vecinit.i, float %n, i32 1
%vecinit2.i = insertelement <4 x float> %vecinit1.i, float %n, i32 2
%vecinit3.i = insertelement <4 x float> %vecinit2.i, float %n, i32 3
%0 = tail call <4 x float> @llvm.fma.v4f32(<4 x float> %b, <4 x float> %vecinit3.i, <4 x float> %a)
ret <4 x float> %0
}
define <2 x float> @test_vfms_n_f32(<2 x float> %a, <2 x float> %b, float %n) {
; CHECK: test_vfms_n_f32:
; CHECK: fmls {{v[0-9]+}}.2s, {{v[0-9]+}}.2s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%vecinit.i = insertelement <2 x float> undef, float %n, i32 0
%vecinit1.i = insertelement <2 x float> %vecinit.i, float %n, i32 1
%0 = fsub <2 x float> <float -0.000000e+00, float -0.000000e+00>, %b
%1 = tail call <2 x float> @llvm.fma.v2f32(<2 x float> %0, <2 x float> %vecinit1.i, <2 x float> %a)
ret <2 x float> %1
}
define <4 x float> @test_vfmsq_n_f32(<4 x float> %a, <4 x float> %b, float %n) {
; CHECK: test_vfmsq_n_f32:
; CHECK: fmls {{v[0-9]+}}.4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.s[{{[0-9]+}}]
entry:
%vecinit.i = insertelement <4 x float> undef, float %n, i32 0
%vecinit1.i = insertelement <4 x float> %vecinit.i, float %n, i32 1
%vecinit2.i = insertelement <4 x float> %vecinit1.i, float %n, i32 2
%vecinit3.i = insertelement <4 x float> %vecinit2.i, float %n, i32 3
%0 = fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %b
%1 = tail call <4 x float> @llvm.fma.v4f32(<4 x float> %0, <4 x float> %vecinit3.i, <4 x float> %a)
ret <4 x float> %1
}
|
