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
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme2 -force-streaming -verify-machineinstrs < %s | FileCheck %s
; SRSHL (Single, x2)
define { <vscale x 16 x i8>, <vscale x 16 x i8> } @multi_vec_rounding_shl_single_x2_s8(<vscale x 16 x i8> %dummy, <vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x2_s8:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: srshl { z4.b, z5.b }, { z4.b, z5.b }, z3.b
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.srshl.single.x2.nxv16i8(<vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zm)
ret { <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
define { <vscale x 8 x i16>, <vscale x 8 x i16> } @multi_vec_rounding_shl_single_x2_s16(<vscale x 8 x i16> %dummy, <vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x2_s16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: srshl { z4.h, z5.h }, { z4.h, z5.h }, z3.h
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.srshl.single.x2.nxv8i16(<vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zm)
ret { <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 4 x i32>, <vscale x 4 x i32> } @multi_vec_rounding_shl_single_x2_s32(<vscale x 4 x i32> %dummy, <vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x2_s32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: srshl { z4.s, z5.s }, { z4.s, z5.s }, z3.s
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.srshl.single.x2.nxv4i32(<vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zm)
ret { <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 2 x i64>, <vscale x 2 x i64> } @multi_vec_rounding_shl_single_x2_s64(<vscale x 2 x i64> %dummy, <vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x2_s64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: srshl { z4.d, z5.d }, { z4.d, z5.d }, z3.d
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.srshl.single.x2.nxv2i64(<vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zm)
ret { <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
; SRSHL (Single, x4)
define { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @multi_vec_rounding_shl_single_x4_s8(<vscale x 16 x i8> %dummy, <vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zdn3, <vscale x 16 x i8> %zdn4, <vscale x 16 x i8> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x4_s8:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: srshl { z24.b - z27.b }, { z24.b - z27.b }, z5.b
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> }
@llvm.aarch64.sve.srshl.single.x4.nxv16i8(<vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zdn3, <vscale x 16 x i8> %zdn4, <vscale x 16 x i8> %zm)
ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
define { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @multi_vec_rounding_shl_single_x4_s16(<vscale x 8 x i16> %dummy, <vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zdn3, <vscale x 8 x i16> %zdn4, <vscale x 8 x i16> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x4_s16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: srshl { z24.h - z27.h }, { z24.h - z27.h }, z5.h
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> }
@llvm.aarch64.sve.srshl.single.x4.nxv8i16(<vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zdn3, <vscale x 8 x i16> %zdn4, <vscale x 8 x i16> %zm)
ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @multi_vec_rounding_shl_single_x4_s32(<vscale x 4 x i32> %dummy, <vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zdn3, <vscale x 4 x i32> %zdn4, <vscale x 4 x i32> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x4_s32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: srshl { z24.s - z27.s }, { z24.s - z27.s }, z5.s
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> }
@llvm.aarch64.sve.srshl.single.x4.nxv4i32(<vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zdn3, <vscale x 4 x i32> %zdn4, <vscale x 4 x i32> %zm)
ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @multi_vec_rounding_shl_single_x4_s64(<vscale x 2 x i64> %dummy, <vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zdn3, <vscale x 2 x i64> %zdn4, <vscale x 2 x i64> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x4_s64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: srshl { z24.d - z27.d }, { z24.d - z27.d }, z5.d
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> }
@llvm.aarch64.sve.srshl.single.x4.nxv2i64(<vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zdn3, <vscale x 2 x i64> %zdn4, <vscale x 2 x i64> %zm)
ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
; URSHL (Single, x2)
define { <vscale x 16 x i8>, <vscale x 16 x i8> } @multi_vec_rounding_shl_single_x2_u8(<vscale x 16 x i8> %dummy, <vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x2_u8:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: urshl { z4.b, z5.b }, { z4.b, z5.b }, z3.b
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.urshl.single.x2.nxv16i8(<vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zm)
ret { <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
define { <vscale x 8 x i16>, <vscale x 8 x i16> } @multi_vec_rounding_shl_single_x2_u16(<vscale x 8 x i16> %dummy, <vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x2_u16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: urshl { z4.h, z5.h }, { z4.h, z5.h }, z3.h
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.urshl.single.x2.nxv8i16(<vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zm)
ret { <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 4 x i32>, <vscale x 4 x i32> } @multi_vec_rounding_shl_single_x2_u32(<vscale x 4 x i32> %dummy, <vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x2_u32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: urshl { z4.s, z5.s }, { z4.s, z5.s }, z3.s
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.urshl.single.x2.nxv4i32(<vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zm)
ret { <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 2 x i64>, <vscale x 2 x i64> } @multi_vec_rounding_shl_single_x2_u64(<vscale x 2 x i64> %dummy, <vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x2_u64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: urshl { z4.d, z5.d }, { z4.d, z5.d }, z3.d
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.urshl.single.x2.nxv2i64(<vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zm)
ret { <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
; URSHL (Single, x4)
define { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @multi_vec_rounding_shl_single_x4_u8(<vscale x 16 x i8> %dummy, <vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zdn3, <vscale x 16 x i8> %zdn4, <vscale x 16 x i8> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x4_u8:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: urshl { z24.b - z27.b }, { z24.b - z27.b }, z5.b
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> }
@llvm.aarch64.sve.urshl.single.x4.nxv16i8(<vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zdn3, <vscale x 16 x i8> %zdn4, <vscale x 16 x i8> %zm)
ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
define { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @multi_vec_rounding_shl_single_x4_u16(<vscale x 8 x i16> %dummy, <vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zdn3, <vscale x 8 x i16> %zdn4, <vscale x 8 x i16> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x4_u16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: urshl { z24.h - z27.h }, { z24.h - z27.h }, z5.h
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> }
@llvm.aarch64.sve.urshl.single.x4.nxv8i16(<vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zdn3, <vscale x 8 x i16> %zdn4, <vscale x 8 x i16> %zm)
ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @multi_vec_rounding_shl_single_x4_u32(<vscale x 4 x i32> %dummy, <vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zdn3, <vscale x 4 x i32> %zdn4, <vscale x 4 x i32> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x4_u32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: urshl { z24.s - z27.s }, { z24.s - z27.s }, z5.s
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> }
@llvm.aarch64.sve.urshl.single.x4.nxv4i32(<vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zdn3, <vscale x 4 x i32> %zdn4, <vscale x 4 x i32> %zm)
ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @multi_vec_rounding_shl_single_x4_u64(<vscale x 2 x i64> %dummy, <vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zdn3, <vscale x 2 x i64> %zdn4, <vscale x 2 x i64> %zm) {
; CHECK-LABEL: multi_vec_rounding_shl_single_x4_u64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: urshl { z24.d - z27.d }, { z24.d - z27.d }, z5.d
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> }
@llvm.aarch64.sve.urshl.single.x4.nxv2i64(<vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zdn3, <vscale x 2 x i64> %zdn4, <vscale x 2 x i64> %zm)
ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
; SRSHL (Multi, x2)
define { <vscale x 16 x i8>, <vscale x 16 x i8> } @multi_vec_rounding_shl_x2_s8(<vscale x 16 x i8> %dummy, <vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zm1, <vscale x 16 x i8> %zm2) {
; CHECK-LABEL: multi_vec_rounding_shl_x2_s8:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z7.d, z4.d
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z6.d, z3.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: srshl { z4.b, z5.b }, { z4.b, z5.b }, { z6.b, z7.b }
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.srshl.x2.nxv16i8(<vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zm1, <vscale x 16 x i8> %zm2)
ret { <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
define { <vscale x 8 x i16>, <vscale x 8 x i16> } @multi_vec_rounding_shl_x2_s16(<vscale x 8 x i16> %dummy, <vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zm1, <vscale x 8 x i16> %zm2) {
; CHECK-LABEL: multi_vec_rounding_shl_x2_s16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z7.d, z4.d
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z6.d, z3.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: srshl { z4.h, z5.h }, { z4.h, z5.h }, { z6.h, z7.h }
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.srshl.x2.nxv8i16(<vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zm1, <vscale x 8 x i16> %zm2)
ret { <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 4 x i32>, <vscale x 4 x i32> } @multi_vec_rounding_shl_x2_s32(<vscale x 4 x i32> %dummy, <vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2) {
; CHECK-LABEL: multi_vec_rounding_shl_x2_s32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z7.d, z4.d
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z6.d, z3.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: srshl { z4.s, z5.s }, { z4.s, z5.s }, { z6.s, z7.s }
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.srshl.x2.nxv4i32(<vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2)
ret { <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 2 x i64>, <vscale x 2 x i64> } @multi_vec_rounding_shl_x2_s64(<vscale x 2 x i64> %dummy, <vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2) {
; CHECK-LABEL: multi_vec_rounding_shl_x2_s64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z7.d, z4.d
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z6.d, z3.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: srshl { z4.d, z5.d }, { z4.d, z5.d }, { z6.d, z7.d }
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.srshl.x2.nxv2i64(<vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2)
ret { <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
; SRSHL (Multi, x4)
define { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @multi_vec_rounding_shl_x4_s8(<vscale x 16 x i8> %dummy, <vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zdn3, <vscale x 16 x i8> %zdn4, <vscale x 16 x i8> %zm1, <vscale x 16 x i8> %zm2, <vscale x 16 x i8> %zm3, <vscale x 16 x i8> %zm4) {
; CHECK-LABEL: multi_vec_rounding_shl_x4_s8:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z30.d, z7.d
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z29.d, z6.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z28.d, z5.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: ldr z31, [x0]
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: srshl { z24.b - z27.b }, { z24.b - z27.b }, { z28.b - z31.b }
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> }
@llvm.aarch64.sve.srshl.x4.nxv16i8(<vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zdn3, <vscale x 16 x i8> %zdn4,
<vscale x 16 x i8> %zm1, <vscale x 16 x i8> %zm2, <vscale x 16 x i8> %zm3, <vscale x 16 x i8> %zm4)
ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
define { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @multi_vec_rounding_shl_x4_s16(<vscale x 8 x i16> %dummy, <vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zdn3, <vscale x 8 x i16> %zdn4, <vscale x 8 x i16> %zm1, <vscale x 8 x i16> %zm2, <vscale x 8 x i16> %zm3, <vscale x 8 x i16> %zm4) {
; CHECK-LABEL: multi_vec_rounding_shl_x4_s16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z30.d, z7.d
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z29.d, z6.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z28.d, z5.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: ldr z31, [x0]
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: srshl { z24.h - z27.h }, { z24.h - z27.h }, { z28.h - z31.h }
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> }
@llvm.aarch64.sve.srshl.x4.nxv8i16(<vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zdn3, <vscale x 8 x i16> %zdn4,
<vscale x 8 x i16> %zm1, <vscale x 8 x i16> %zm2, <vscale x 8 x i16> %zm3, <vscale x 8 x i16> %zm4)
ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @multi_vec_rounding_shl_x4_s32(<vscale x 4 x i32> %dummy, <vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zdn3, <vscale x 4 x i32> %zdn4, <vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2, <vscale x 4 x i32> %zm3, <vscale x 4 x i32> %zm4) {
; CHECK-LABEL: multi_vec_rounding_shl_x4_s32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z30.d, z7.d
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z29.d, z6.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z28.d, z5.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: ldr z31, [x0]
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: srshl { z24.s - z27.s }, { z24.s - z27.s }, { z28.s - z31.s }
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> }
@llvm.aarch64.sve.srshl.x4.nxv4i32(<vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zdn3, <vscale x 4 x i32> %zdn4,
<vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2, <vscale x 4 x i32> %zm3, <vscale x 4 x i32> %zm4)
ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @multi_vec_rounding_shl_x4_s64(<vscale x 2 x i64> %dummy, <vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zdn3, <vscale x 2 x i64> %zdn4, <vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2, <vscale x 2 x i64> %zm3, <vscale x 2 x i64> %zm4) {
; CHECK-LABEL: multi_vec_rounding_shl_x4_s64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z30.d, z7.d
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z29.d, z6.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z28.d, z5.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: ldr z31, [x0]
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: srshl { z24.d - z27.d }, { z24.d - z27.d }, { z28.d - z31.d }
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> }
@llvm.aarch64.sve.srshl.x4.nxv2i64(<vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zdn3, <vscale x 2 x i64> %zdn4,
<vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2, <vscale x 2 x i64> %zm3, <vscale x 2 x i64> %zm4)
ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
; URSHL (Multi, x2)
define { <vscale x 16 x i8>, <vscale x 16 x i8> } @multi_vec_rounding_uhl_x2_u8(<vscale x 16 x i8> %dummy, <vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zm1, <vscale x 16 x i8> %zm2) {
; CHECK-LABEL: multi_vec_rounding_uhl_x2_u8:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z7.d, z4.d
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z6.d, z3.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: urshl { z4.b, z5.b }, { z4.b, z5.b }, { z6.b, z7.b }
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.urshl.x2.nxv16i8(<vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zm1, <vscale x 16 x i8> %zm2)
ret { <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
define { <vscale x 8 x i16>, <vscale x 8 x i16> } @multi_vec_rounding_uhl_x2_u16(<vscale x 8 x i16> %dummy, <vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zm1, <vscale x 8 x i16> %zm2) {
; CHECK-LABEL: multi_vec_rounding_uhl_x2_u16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z7.d, z4.d
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z6.d, z3.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: urshl { z4.h, z5.h }, { z4.h, z5.h }, { z6.h, z7.h }
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.urshl.x2.nxv8i16(<vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zm1, <vscale x 8 x i16> %zm2)
ret { <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 4 x i32>, <vscale x 4 x i32> } @multi_vec_rounding_uhl_x2_u32(<vscale x 4 x i32> %dummy, <vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2) {
; CHECK-LABEL: multi_vec_rounding_uhl_x2_u32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z7.d, z4.d
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z6.d, z3.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: urshl { z4.s, z5.s }, { z4.s, z5.s }, { z6.s, z7.s }
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.urshl.x2.nxv4i32(<vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2)
ret { <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 2 x i64>, <vscale x 2 x i64> } @multi_vec_rounding_uhl_x2_u64(<vscale x 2 x i64> %dummy, <vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2) {
; CHECK-LABEL: multi_vec_rounding_uhl_x2_u64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z7.d, z4.d
; CHECK-NEXT: mov z5.d, z2.d
; CHECK-NEXT: mov z6.d, z3.d
; CHECK-NEXT: mov z4.d, z1.d
; CHECK-NEXT: urshl { z4.d, z5.d }, { z4.d, z5.d }, { z6.d, z7.d }
; CHECK-NEXT: mov z0.d, z4.d
; CHECK-NEXT: mov z1.d, z5.d
; CHECK-NEXT: ret
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.urshl.x2.nxv2i64(<vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2)
ret { <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
; URSHL (Multi, x4)
define { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @multi_vec_rounding_shl_x4_u8(<vscale x 16 x i8> %dummy, <vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zdn3, <vscale x 16 x i8> %zdn4, <vscale x 16 x i8> %zm1, <vscale x 16 x i8> %zm2, <vscale x 16 x i8> %zm3, <vscale x 16 x i8> %zm4) {
; CHECK-LABEL: multi_vec_rounding_shl_x4_u8:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z30.d, z7.d
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z29.d, z6.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z28.d, z5.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: ldr z31, [x0]
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: urshl { z24.b - z27.b }, { z24.b - z27.b }, { z28.b - z31.b }
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> }
@llvm.aarch64.sve.urshl.x4.nxv16i8(<vscale x 16 x i8> %zdn1, <vscale x 16 x i8> %zdn2, <vscale x 16 x i8> %zdn3, <vscale x 16 x i8> %zdn4,
<vscale x 16 x i8> %zm1, <vscale x 16 x i8> %zm2, <vscale x 16 x i8> %zm3, <vscale x 16 x i8> %zm4)
ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
define { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @multi_vec_rounding_shl_x4_u16(<vscale x 8 x i16> %dummy, <vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zdn3, <vscale x 8 x i16> %zdn4, <vscale x 8 x i16> %zm1, <vscale x 8 x i16> %zm2, <vscale x 8 x i16> %zm3, <vscale x 8 x i16> %zm4) {
; CHECK-LABEL: multi_vec_rounding_shl_x4_u16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z30.d, z7.d
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z29.d, z6.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z28.d, z5.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: ldr z31, [x0]
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: urshl { z24.h - z27.h }, { z24.h - z27.h }, { z28.h - z31.h }
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> }
@llvm.aarch64.sve.urshl.x4.nxv8i16(<vscale x 8 x i16> %zdn1, <vscale x 8 x i16> %zdn2, <vscale x 8 x i16> %zdn3, <vscale x 8 x i16> %zdn4,
<vscale x 8 x i16> %zm1, <vscale x 8 x i16> %zm2, <vscale x 8 x i16> %zm3, <vscale x 8 x i16> %zm4)
ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @multi_vec_rounding_shl_x4_u32(<vscale x 4 x i32> %dummy, <vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zdn3, <vscale x 4 x i32> %zdn4, <vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2, <vscale x 4 x i32> %zm3, <vscale x 4 x i32> %zm4) {
; CHECK-LABEL: multi_vec_rounding_shl_x4_u32:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z30.d, z7.d
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z29.d, z6.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z28.d, z5.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: ldr z31, [x0]
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: urshl { z24.s - z27.s }, { z24.s - z27.s }, { z28.s - z31.s }
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> }
@llvm.aarch64.sve.urshl.x4.nxv4i32(<vscale x 4 x i32> %zdn1, <vscale x 4 x i32> %zdn2, <vscale x 4 x i32> %zdn3, <vscale x 4 x i32> %zdn4,
<vscale x 4 x i32> %zm1, <vscale x 4 x i32> %zm2, <vscale x 4 x i32> %zm3, <vscale x 4 x i32> %zm4)
ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @multi_vec_rounding_shl_x4_u64(<vscale x 2 x i64> %dummy, <vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zdn3, <vscale x 2 x i64> %zdn4, <vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2, <vscale x 2 x i64> %zm3, <vscale x 2 x i64> %zm4) {
; CHECK-LABEL: multi_vec_rounding_shl_x4_u64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z30.d, z7.d
; CHECK-NEXT: mov z27.d, z4.d
; CHECK-NEXT: mov z29.d, z6.d
; CHECK-NEXT: mov z26.d, z3.d
; CHECK-NEXT: mov z28.d, z5.d
; CHECK-NEXT: mov z25.d, z2.d
; CHECK-NEXT: ldr z31, [x0]
; CHECK-NEXT: mov z24.d, z1.d
; CHECK-NEXT: urshl { z24.d - z27.d }, { z24.d - z27.d }, { z28.d - z31.d }
; CHECK-NEXT: mov z0.d, z24.d
; CHECK-NEXT: mov z1.d, z25.d
; CHECK-NEXT: mov z2.d, z26.d
; CHECK-NEXT: mov z3.d, z27.d
; CHECK-NEXT: ret
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> }
@llvm.aarch64.sve.urshl.x4.nxv2i64(<vscale x 2 x i64> %zdn1, <vscale x 2 x i64> %zdn2, <vscale x 2 x i64> %zdn3, <vscale x 2 x i64> %zdn4,
<vscale x 2 x i64> %zm1, <vscale x 2 x i64> %zm2, <vscale x 2 x i64> %zm3, <vscale x 2 x i64> %zm4)
ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.srshl.single.x2.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.srshl.single.x2.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.srshl.single.x2.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.srshl.single.x2.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
declare { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.srshl.single.x4.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.srshl.single.x4.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
declare { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.srshl.single.x4.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
declare { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.srshl.single.x4.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.urshl.single.x2.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.urshl.single.x2.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.urshl.single.x2.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.urshl.single.x2.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
declare { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.urshl.single.x4.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.urshl.single.x4.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
declare { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.urshl.single.x4.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
declare { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.urshl.single.x4.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.srshl.x2.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.srshl.x2.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.srshl.x2.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.srshl.x2.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
declare { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> }
@llvm.aarch64.sve.srshl.x4.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> }
@llvm.aarch64.sve.srshl.x4.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
declare { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> }
@llvm.aarch64.sve.srshl.x4.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
declare { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> }
@llvm.aarch64.sve.srshl.x4.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> )
declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.urshl.x2.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.urshl.x2.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.urshl.x2.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.urshl.x2.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
declare { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> }
@llvm.aarch64.sve.urshl.x4.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>)
declare { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> }
@llvm.aarch64.sve.urshl.x4.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>)
declare { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> }
@llvm.aarch64.sve.urshl.x4.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>)
declare { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> }
@llvm.aarch64.sve.urshl.x4.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>)
|
