File: sve-intrinsic-muladdsub.ll

package info (click to toggle)
llvm-toolchain-17 1%3A17.0.6-22
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 1,799,624 kB
  • sloc: cpp: 6,428,607; ansic: 1,383,196; asm: 793,408; python: 223,504; objc: 75,364; f90: 60,502; lisp: 33,869; pascal: 15,282; sh: 9,684; perl: 7,453; ml: 4,937; awk: 3,523; makefile: 2,889; javascript: 2,149; xml: 888; fortran: 619; cs: 573
file content (443 lines) | stat: -rw-r--r-- 33,095 bytes parent folder | download | duplicates (10)
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
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -S -passes=instcombine < %s | FileCheck %s

target triple = "aarch64-unknown-linux-gnu"

; fadd(a, fmul(b, c)) -> fmla(a, b, c)
define <vscale x 8 x half> @combine_fmuladd_1(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmuladd_1(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmla.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

; fadd(a, fmul_u(b, c)) -> fmla(a, b, c)
define <vscale x 8 x half> @combine_fmuladd_2(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmuladd_2(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmla.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

; fadd_u(a, fmul_u(b, c)) -> fmla_u(a, b, c)
define <vscale x 8 x half> @combine_fmuladd_3(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmuladd_3(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmla.u.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

; TODO: Missing combine!
; fadd_u(a, fmul(b, c)) -> fmla_u(a, b, c)
define <vscale x 8 x half> @combine_fmuladd_4(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmuladd_4(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmla.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

; fadd(fmul(b, c), a) -> fmad(b, c, a)
define <vscale x 8 x half> @combine_fmuladd_5(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmuladd_5(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmad.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]], <vscale x 8 x half> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %1, <vscale x 8 x half> %a)
  ret <vscale x 8 x half> %2
}

; TODO: Missing combine!
; fadd(fmul_u(b, c), a) -> fmla_u(a, b, c)
define <vscale x 8 x half> @combine_fmuladd_6(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmuladd_6(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> [[P]], <vscale x 8 x half> [[TMP1]], <vscale x 8 x half> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP2]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %1, <vscale x 8 x half> %a)
  ret <vscale x 8 x half> %2
}

; TODO: Missing combine!
; fadd_u(fmul_u(b, c), a) -> fmla_u(a, b, c)
define <vscale x 8 x half> @combine_fmuladd_7(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmuladd_7(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.u.nxv8f16(<vscale x 8 x i1> [[P]], <vscale x 8 x half> [[TMP1]], <vscale x 8 x half> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP2]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %1, <vscale x 8 x half> %a)
  ret <vscale x 8 x half> %2
}

; TODO: Missing combine!
; fadd_u(fmul(b, c), a) -> fmla_u(a, b, c)
define <vscale x 8 x half> @combine_fmuladd_8(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmuladd_8(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmad.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]], <vscale x 8 x half> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %1, <vscale x 8 x half> %a)
  ret <vscale x 8 x half> %2
}

; fsub(a, fmul(b, c)) -> fmls(a, b, c)
define <vscale x 8 x half> @combine_fmulsub_1(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmulsub_1(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmls.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

define <vscale x 8 x half> @combine_fmulsub_2(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmulsub_2(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmls.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

; fsub_u(a, fmul_u(b, c)) -> fmls_u(a, b, c)
define <vscale x 8 x half> @combine_fmulsub_3(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmulsub_3(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmls.u.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

; TODO: Missing combine!
; fsub_u(a, fmul(b, c)) -> fmls_u(a, b, c)
define <vscale x 8 x half> @combine_fmulsub_4(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmulsub_4(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fmls.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

; fsub(mul(b, c), a) -> fnmsb(b, c, a)
define <vscale x 8 x half> @combine_fmulsub_5(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmulsub_5(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fnmsb.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]], <vscale x 8 x half> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %1, <vscale x 8 x half> %a)
  ret <vscale x 8 x half> %2
}

; TODO: Missing combine!
; fsub(fmul_u(b, c), a) -> fnmls_u(a, b, c)
define <vscale x 8 x half> @combine_fmulsub_6(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmulsub_6(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.nxv8f16(<vscale x 8 x i1> [[P]], <vscale x 8 x half> [[TMP1]], <vscale x 8 x half> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP2]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %1, <vscale x 8 x half> %a)
  ret <vscale x 8 x half> %2
}

; TODO: Missing combine!
; fsub_u(fmul_u(b, c), a) -> fnmls_u(a, b, c)
define <vscale x 8 x half> @combine_fmulsub_7(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmulsub_7(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.u.nxv8f16(<vscale x 8 x i1> [[P]], <vscale x 8 x half> [[TMP1]], <vscale x 8 x half> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP2]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %1, <vscale x 8 x half> %a)
  ret <vscale x 8 x half> %2
}

; TODO: Missing combine!
; fsub_u(fmul(b, c), a) -> fnmls_u(a, b, c)
define <vscale x 8 x half> @combine_fmulsub_8(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @combine_fmulsub_8(
; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 8 x half> @llvm.aarch64.sve.fnmsb.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]], <vscale x 8 x half> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP1]]
;
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fsub.u.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %1, <vscale x 8 x half> %a)
  ret <vscale x 8 x half> %2
}

; add(a, mul(b, c)) -> mla(a, b, c)
define <vscale x 16 x i8> @combine_muladd_1(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_muladd_1(
; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mla.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %1)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; add(a, mul_u(b, c)) -> mla(a, b, c)
define <vscale x 16 x i8> @combine_muladd_2(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_muladd_2(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[TMP1]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %1)
  ret <vscale x 16 x i8> %2
}

; add_u(a, mul_u(b, c)) -> mla_u(a, b, c)
define <vscale x 16 x i8> @combine_muladd_3(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_muladd_3(
; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mla.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %1)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; add_u(a, mul(b, c)) -> mla_u(a, b, c)
define <vscale x 16 x i8> @combine_muladd_4(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_muladd_4(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[TMP1]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %1)
  ret <vscale x 16 x i8> %2
}

; add(mul(b, c), a) -> mad(b, c, c)
define <vscale x 16 x i8> @combine_muladd_5(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_muladd_5(
; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mad.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]], <vscale x 16 x i8> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %1, <vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; add(mul_u(b, c), a) -> mla_u(a, b, c)
define <vscale x 16 x i8> @combine_muladd_6(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_muladd_6(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[TMP1]], <vscale x 16 x i8> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %1, <vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; add_u(mul_u(b, c), a) -> mla_u(a, b, c)
define <vscale x 16 x i8> @combine_muladd_7(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_muladd_7(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[TMP1]], <vscale x 16 x i8> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %1, <vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; add_u(mul(b, c), a) -> mla_u(a, b, c)
define <vscale x 16 x i8> @combine_muladd_8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_muladd_8(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[TMP1]], <vscale x 16 x i8> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %1, <vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %2
}

; sub(a, mul(b, c)) -> mls(a, b, c)
define <vscale x 16 x i8> @combine_mulsub_1(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_mulsub_1(
; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mls.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %1)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; sub(a, mul_u(b, c)) -> mls(a, b, c)
define <vscale x 16 x i8> @combine_mulsub_2(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_mulsub_2(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[TMP1]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %1)
  ret <vscale x 16 x i8> %2
}

; sub_u(a, mul_u(b, c)) -> mls_u(a, b, c)
define <vscale x 16 x i8> @combine_mulsub_3(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_mulsub_3(
; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x i8> @llvm.aarch64.sve.mls.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP1]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %1)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; sub_u(a, mul(b, c)) -> mls_u(a, b, c)
define <vscale x 16 x i8> @combine_mulsub_4(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_mulsub_4(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[A:%.*]], <vscale x 16 x i8> [[TMP1]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %1)
  ret <vscale x 16 x i8> %2
}

; sub(mul(b, c), a) -> sub(mul(b, c), a)
define <vscale x 16 x i8> @combine_mulsub_5(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_mulsub_5(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[TMP1]], <vscale x 16 x i8> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %1, <vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; sub(mul_u(b, c), a) -> sub_u(mul_u(b, c), a)
define <vscale x 16 x i8> @combine_mulsub_6(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_mulsub_6(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[TMP1]], <vscale x 16 x i8> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %1, <vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %2
}

; sub_u(mul_u(b, c), a) -> sub_u(mul_u(b, c), a)
define <vscale x 16 x i8> @combine_mulsub_7(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_mulsub_7(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[TMP1]], <vscale x 16 x i8> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %1, <vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %2
}

; TODO: Missing combine!
; sub_u(mul(b, c), a) -> sub_u(mul_u(b, c), a)
define <vscale x 16 x i8> @combine_mulsub_8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c) #0 {
; CHECK-LABEL: @combine_mulsub_8(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> [[P:%.*]], <vscale x 16 x i8> [[B:%.*]], <vscale x 16 x i8> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1> [[P]], <vscale x 16 x i8> [[TMP1]], <vscale x 16 x i8> [[A:%.*]])
; CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP2]]
;
  %1 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %b, <vscale x 16 x i8> %c)
  %2 = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1> %p, <vscale x 16 x i8> %1, <vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %2
}

define <vscale x 8 x half> @neg_combine_fmuladd_neq_pred(<vscale x 8 x i1> %p1, <vscale x 8 x i1> %p2, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @neg_combine_fmuladd_neq_pred(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> [[P1:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> [[P2:%.*]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[TMP1]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP2]]
;
; ret <vscale x 8 x half> %9
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p1, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> %p2, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

define <vscale x 8 x half> @neg_combine_fmuladd_two_fmul_uses(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @neg_combine_fmuladd_two_fmul_uses(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> [[P]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[TMP1]])
; CHECK-NEXT:    [[TMP3:%.*]] = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> [[P]], <vscale x 8 x half> [[TMP2]], <vscale x 8 x half> [[TMP1]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP3]]
;
; ret <vscale x 8 x half> %8
  %1 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  %3 = tail call fast <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %2, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %3
}

define <vscale x 8 x half> @neg_combine_fmuladd_neq_flags(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) #0 {
; CHECK-LABEL: @neg_combine_fmuladd_neq_flags(
; CHECK-NEXT:    [[TMP1:%.*]] = tail call reassoc nnan contract <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> [[P:%.*]], <vscale x 8 x half> [[B:%.*]], <vscale x 8 x half> [[C:%.*]])
; CHECK-NEXT:    [[TMP2:%.*]] = tail call reassoc contract <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> [[P]], <vscale x 8 x half> [[A:%.*]], <vscale x 8 x half> [[TMP1]])
; CHECK-NEXT:    ret <vscale x 8 x half> [[TMP2]]
;
; ret <vscale x 8 x half> %7
  %1 = tail call reassoc nnan contract <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %b, <vscale x 8 x half> %c)
  %2 = tail call reassoc contract <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1> %p, <vscale x 8 x half> %a, <vscale x 8 x half> %1)
  ret <vscale x 8 x half> %2
}

declare <vscale x 8 x half> @llvm.aarch64.sve.fmul.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 8 x half> @llvm.aarch64.sve.fadd.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 8 x half> @llvm.aarch64.sve.fsub.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)

declare <vscale x 8 x half> @llvm.aarch64.sve.fmul.u.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 8 x half> @llvm.aarch64.sve.fadd.u.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 8 x half> @llvm.aarch64.sve.fsub.u.nxv8f16(<vscale x 8 x i1>, <vscale x 8 x half>, <vscale x 8 x half>)

declare <vscale x 16 x i8> @llvm.aarch64.sve.mul.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare <vscale x 16 x i8> @llvm.aarch64.sve.add.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare <vscale x 16 x i8> @llvm.aarch64.sve.sub.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)

declare <vscale x 16 x i8> @llvm.aarch64.sve.mul.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare <vscale x 16 x i8> @llvm.aarch64.sve.add.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare <vscale x 16 x i8> @llvm.aarch64.sve.sub.u.nxv16i8(<vscale x 16 x i1>, <vscale x 16 x i8>, <vscale x 16 x i8>)

attributes #0 = { "target-features"="+sve" }