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
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=aarch64 -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-SD,CHECK-SD-NOFP16
; RUN: llc -mtriple=aarch64 -mattr=+fullfp16 -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-SD,CHECK-SD-FP16
; RUN: llc -mtriple=aarch64 -global-isel -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-GI,CHECK-GI-NOFP16
; RUN: llc -mtriple=aarch64 -mattr=+fullfp16 -global-isel -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-GI,CHECK-GI-FP16
define float @add_HalfS(<2 x float> %bin.rdx) {
; CHECK-LABEL: add_HalfS:
; CHECK: // %bb.0:
; CHECK-NEXT: faddp s0, v0.2s
; CHECK-NEXT: ret
%r = call fast float @llvm.vector.reduce.fadd.f32.v2f32(float -0.0, <2 x float> %bin.rdx)
ret float %r
}
define half @add_v2HalfH(<2 x half> %bin.rdx) {
; CHECK-SD-NOFP16-LABEL: add_v2HalfH:
; CHECK-SD-NOFP16: // %bb.0:
; CHECK-SD-NOFP16-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-NOFP16-NEXT: mov h1, v0.h[1]
; CHECK-SD-NOFP16-NEXT: fcvt s0, h0
; CHECK-SD-NOFP16-NEXT: fcvt s1, h1
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s1
; CHECK-SD-NOFP16-NEXT: fcvt h0, s0
; CHECK-SD-NOFP16-NEXT: ret
;
; CHECK-SD-FP16-LABEL: add_v2HalfH:
; CHECK-SD-FP16: // %bb.0:
; CHECK-SD-FP16-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-FP16-NEXT: faddp h0, v0.2h
; CHECK-SD-FP16-NEXT: ret
;
; CHECK-GI-NOFP16-LABEL: add_v2HalfH:
; CHECK-GI-NOFP16: // %bb.0:
; CHECK-GI-NOFP16-NEXT: fcvtl v0.4s, v0.4h
; CHECK-GI-NOFP16-NEXT: faddp s0, v0.2s
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: ret
;
; CHECK-GI-FP16-LABEL: add_v2HalfH:
; CHECK-GI-FP16: // %bb.0:
; CHECK-GI-FP16-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-GI-FP16-NEXT: mov h1, v0.h[1]
; CHECK-GI-FP16-NEXT: fadd h0, h0, h1
; CHECK-GI-FP16-NEXT: ret
%r = call fast half @llvm.vector.reduce.fadd.f16.v2f16(half -0.0, <2 x half> %bin.rdx)
ret half %r
}
define half @add_v3HalfH(<3 x half> %bin.rdx) {
; CHECK-SD-NOFP16-LABEL: add_v3HalfH:
; CHECK-SD-NOFP16: // %bb.0:
; CHECK-SD-NOFP16-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-NOFP16-NEXT: mov h1, v0.h[1]
; CHECK-SD-NOFP16-NEXT: fcvt s2, h0
; CHECK-SD-NOFP16-NEXT: mov h0, v0.h[2]
; CHECK-SD-NOFP16-NEXT: fcvt s1, h1
; CHECK-SD-NOFP16-NEXT: fcvt s0, h0
; CHECK-SD-NOFP16-NEXT: fadd s1, s2, s1
; CHECK-SD-NOFP16-NEXT: fadd s0, s1, s0
; CHECK-SD-NOFP16-NEXT: fcvt h0, s0
; CHECK-SD-NOFP16-NEXT: ret
;
; CHECK-SD-FP16-LABEL: add_v3HalfH:
; CHECK-SD-FP16: // %bb.0:
; CHECK-SD-FP16-NEXT: movi d1, #0000000000000000
; CHECK-SD-FP16-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-FP16-NEXT: mov v0.h[3], v1.h[0]
; CHECK-SD-FP16-NEXT: faddp v0.4h, v0.4h, v0.4h
; CHECK-SD-FP16-NEXT: faddp h0, v0.2h
; CHECK-SD-FP16-NEXT: ret
;
; CHECK-GI-NOFP16-LABEL: add_v3HalfH:
; CHECK-GI-NOFP16: // %bb.0:
; CHECK-GI-NOFP16-NEXT: movi v1.2s, #128, lsl #24
; CHECK-GI-NOFP16-NEXT: fcvtl v0.4s, v0.4h
; CHECK-GI-NOFP16-NEXT: mov v0.s[3], v1.s[0]
; CHECK-GI-NOFP16-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: faddp s0, v0.2s
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: ret
;
; CHECK-GI-FP16-LABEL: add_v3HalfH:
; CHECK-GI-FP16: // %bb.0:
; CHECK-GI-FP16-NEXT: adrp x8, .LCPI2_0
; CHECK-GI-FP16-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-GI-FP16-NEXT: ldr h1, [x8, :lo12:.LCPI2_0]
; CHECK-GI-FP16-NEXT: mov v0.h[3], v1.h[0]
; CHECK-GI-FP16-NEXT: faddp v0.4h, v0.4h, v0.4h
; CHECK-GI-FP16-NEXT: faddp h0, v0.2h
; CHECK-GI-FP16-NEXT: ret
%r = call fast half @llvm.vector.reduce.fadd.f16.v3f16(half -0.0, <3 x half> %bin.rdx)
ret half %r
}
define half @add_HalfH(<4 x half> %bin.rdx) {
; CHECK-SD-NOFP16-LABEL: add_HalfH:
; CHECK-SD-NOFP16: // %bb.0:
; CHECK-SD-NOFP16-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-NOFP16-NEXT: mov h1, v0.h[1]
; CHECK-SD-NOFP16-NEXT: mov h2, v0.h[3]
; CHECK-SD-NOFP16-NEXT: mov h3, v0.h[2]
; CHECK-SD-NOFP16-NEXT: fcvt s0, h0
; CHECK-SD-NOFP16-NEXT: fcvt s1, h1
; CHECK-SD-NOFP16-NEXT: fcvt s3, h3
; CHECK-SD-NOFP16-NEXT: fcvt s2, h2
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s1
; CHECK-SD-NOFP16-NEXT: fadd s1, s3, s2
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s1
; CHECK-SD-NOFP16-NEXT: fcvt h0, s0
; CHECK-SD-NOFP16-NEXT: ret
;
; CHECK-SD-FP16-LABEL: add_HalfH:
; CHECK-SD-FP16: // %bb.0:
; CHECK-SD-FP16-NEXT: faddp v0.4h, v0.4h, v0.4h
; CHECK-SD-FP16-NEXT: faddp h0, v0.2h
; CHECK-SD-FP16-NEXT: ret
;
; CHECK-GI-NOFP16-LABEL: add_HalfH:
; CHECK-GI-NOFP16: // %bb.0:
; CHECK-GI-NOFP16-NEXT: fcvtl v0.4s, v0.4h
; CHECK-GI-NOFP16-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: faddp s0, v0.2s
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: ret
;
; CHECK-GI-FP16-LABEL: add_HalfH:
; CHECK-GI-FP16: // %bb.0:
; CHECK-GI-FP16-NEXT: faddp v0.4h, v0.4h, v0.4h
; CHECK-GI-FP16-NEXT: faddp h0, v0.2h
; CHECK-GI-FP16-NEXT: ret
%r = call fast half @llvm.vector.reduce.fadd.f16.v4f16(half -0.0, <4 x half> %bin.rdx)
ret half %r
}
define half @add_H(<8 x half> %bin.rdx) {
; CHECK-SD-NOFP16-LABEL: add_H:
; CHECK-SD-NOFP16: // %bb.0:
; CHECK-SD-NOFP16-NEXT: mov h1, v0.h[2]
; CHECK-SD-NOFP16-NEXT: mov h2, v0.h[1]
; CHECK-SD-NOFP16-NEXT: mov h3, v0.h[5]
; CHECK-SD-NOFP16-NEXT: mov h4, v0.h[4]
; CHECK-SD-NOFP16-NEXT: mov h5, v0.h[3]
; CHECK-SD-NOFP16-NEXT: mov h6, v0.h[6]
; CHECK-SD-NOFP16-NEXT: fcvt s7, h0
; CHECK-SD-NOFP16-NEXT: mov h0, v0.h[7]
; CHECK-SD-NOFP16-NEXT: fcvt s2, h2
; CHECK-SD-NOFP16-NEXT: fcvt s1, h1
; CHECK-SD-NOFP16-NEXT: fcvt s3, h3
; CHECK-SD-NOFP16-NEXT: fcvt s5, h5
; CHECK-SD-NOFP16-NEXT: fcvt s4, h4
; CHECK-SD-NOFP16-NEXT: fcvt s0, h0
; CHECK-SD-NOFP16-NEXT: fadd s2, s7, s2
; CHECK-SD-NOFP16-NEXT: fadd s1, s1, s5
; CHECK-SD-NOFP16-NEXT: fadd s3, s4, s3
; CHECK-SD-NOFP16-NEXT: fcvt s4, h6
; CHECK-SD-NOFP16-NEXT: fadd s1, s2, s1
; CHECK-SD-NOFP16-NEXT: fadd s2, s3, s4
; CHECK-SD-NOFP16-NEXT: fadd s1, s1, s2
; CHECK-SD-NOFP16-NEXT: fadd s0, s1, s0
; CHECK-SD-NOFP16-NEXT: fcvt h0, s0
; CHECK-SD-NOFP16-NEXT: ret
;
; CHECK-SD-FP16-LABEL: add_H:
; CHECK-SD-FP16: // %bb.0:
; CHECK-SD-FP16-NEXT: faddp v1.8h, v0.8h, v0.8h
; CHECK-SD-FP16-NEXT: faddp v0.8h, v1.8h, v0.8h
; CHECK-SD-FP16-NEXT: faddp h0, v0.2h
; CHECK-SD-FP16-NEXT: ret
;
; CHECK-GI-NOFP16-LABEL: add_H:
; CHECK-GI-NOFP16: // %bb.0:
; CHECK-GI-NOFP16-NEXT: fcvtl v1.4s, v0.4h
; CHECK-GI-NOFP16-NEXT: fcvtl2 v0.4s, v0.8h
; CHECK-GI-NOFP16-NEXT: fadd v0.4s, v1.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: faddp s0, v0.2s
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: ret
;
; CHECK-GI-FP16-LABEL: add_H:
; CHECK-GI-FP16: // %bb.0:
; CHECK-GI-FP16-NEXT: faddp v1.8h, v0.8h, v0.8h
; CHECK-GI-FP16-NEXT: faddp v0.8h, v1.8h, v0.8h
; CHECK-GI-FP16-NEXT: faddp h0, v0.2h
; CHECK-GI-FP16-NEXT: ret
%r = call fast half @llvm.vector.reduce.fadd.f16.v8f16(half -0.0, <8 x half> %bin.rdx)
ret half %r
}
define float @add_S(<4 x float> %bin.rdx) {
; CHECK-LABEL: add_S:
; CHECK: // %bb.0:
; CHECK-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-NEXT: faddp s0, v0.2s
; CHECK-NEXT: ret
%r = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %bin.rdx)
ret float %r
}
define double @add_D(<2 x double> %bin.rdx) {
; CHECK-LABEL: add_D:
; CHECK: // %bb.0:
; CHECK-NEXT: faddp d0, v0.2d
; CHECK-NEXT: ret
%r = call fast double @llvm.vector.reduce.fadd.f64.v2f64(double -0.0, <2 x double> %bin.rdx)
ret double %r
}
define half @add_2H(<16 x half> %bin.rdx) {
; CHECK-SD-NOFP16-LABEL: add_2H:
; CHECK-SD-NOFP16: // %bb.0:
; CHECK-SD-NOFP16-NEXT: fcvtl v2.4s, v1.4h
; CHECK-SD-NOFP16-NEXT: fcvtl v3.4s, v0.4h
; CHECK-SD-NOFP16-NEXT: fcvtl2 v1.4s, v1.8h
; CHECK-SD-NOFP16-NEXT: fcvtl2 v0.4s, v0.8h
; CHECK-SD-NOFP16-NEXT: fadd v2.4s, v3.4s, v2.4s
; CHECK-SD-NOFP16-NEXT: fadd v0.4s, v0.4s, v1.4s
; CHECK-SD-NOFP16-NEXT: fcvtn v1.4h, v2.4s
; CHECK-SD-NOFP16-NEXT: fcvtn2 v1.8h, v0.4s
; CHECK-SD-NOFP16-NEXT: mov h0, v1.h[1]
; CHECK-SD-NOFP16-NEXT: mov h2, v1.h[2]
; CHECK-SD-NOFP16-NEXT: mov h3, v1.h[3]
; CHECK-SD-NOFP16-NEXT: mov h4, v1.h[4]
; CHECK-SD-NOFP16-NEXT: mov h5, v1.h[5]
; CHECK-SD-NOFP16-NEXT: fcvt s6, h1
; CHECK-SD-NOFP16-NEXT: mov h7, v1.h[6]
; CHECK-SD-NOFP16-NEXT: mov h1, v1.h[7]
; CHECK-SD-NOFP16-NEXT: fcvt s0, h0
; CHECK-SD-NOFP16-NEXT: fcvt s2, h2
; CHECK-SD-NOFP16-NEXT: fcvt s3, h3
; CHECK-SD-NOFP16-NEXT: fcvt s4, h4
; CHECK-SD-NOFP16-NEXT: fcvt s5, h5
; CHECK-SD-NOFP16-NEXT: fcvt s1, h1
; CHECK-SD-NOFP16-NEXT: fadd s0, s6, s0
; CHECK-SD-NOFP16-NEXT: fadd s2, s2, s3
; CHECK-SD-NOFP16-NEXT: fadd s3, s4, s5
; CHECK-SD-NOFP16-NEXT: fcvt s4, h7
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s2
; CHECK-SD-NOFP16-NEXT: fadd s2, s3, s4
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s2
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s1
; CHECK-SD-NOFP16-NEXT: fcvt h0, s0
; CHECK-SD-NOFP16-NEXT: ret
;
; CHECK-SD-FP16-LABEL: add_2H:
; CHECK-SD-FP16: // %bb.0:
; CHECK-SD-FP16-NEXT: fadd v0.8h, v0.8h, v1.8h
; CHECK-SD-FP16-NEXT: faddp v1.8h, v0.8h, v0.8h
; CHECK-SD-FP16-NEXT: faddp v0.8h, v1.8h, v0.8h
; CHECK-SD-FP16-NEXT: faddp h0, v0.2h
; CHECK-SD-FP16-NEXT: ret
;
; CHECK-GI-NOFP16-LABEL: add_2H:
; CHECK-GI-NOFP16: // %bb.0:
; CHECK-GI-NOFP16-NEXT: fcvtl v2.4s, v0.4h
; CHECK-GI-NOFP16-NEXT: fcvtl2 v0.4s, v0.8h
; CHECK-GI-NOFP16-NEXT: fcvtl v3.4s, v1.4h
; CHECK-GI-NOFP16-NEXT: fcvtl2 v1.4s, v1.8h
; CHECK-GI-NOFP16-NEXT: fadd v0.4s, v2.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: fadd v1.4s, v3.4s, v1.4s
; CHECK-GI-NOFP16-NEXT: fadd v0.4s, v0.4s, v1.4s
; CHECK-GI-NOFP16-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: faddp s0, v0.2s
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: ret
;
; CHECK-GI-FP16-LABEL: add_2H:
; CHECK-GI-FP16: // %bb.0:
; CHECK-GI-FP16-NEXT: fadd v0.8h, v0.8h, v1.8h
; CHECK-GI-FP16-NEXT: faddp v1.8h, v0.8h, v0.8h
; CHECK-GI-FP16-NEXT: faddp v0.8h, v1.8h, v0.8h
; CHECK-GI-FP16-NEXT: faddp h0, v0.2h
; CHECK-GI-FP16-NEXT: ret
%r = call fast half @llvm.vector.reduce.fadd.f16.v16f16(half -0.0, <16 x half> %bin.rdx)
ret half %r
}
define float @add_2S(<8 x float> %bin.rdx) {
; CHECK-LABEL: add_2S:
; CHECK: // %bb.0:
; CHECK-NEXT: fadd v0.4s, v0.4s, v1.4s
; CHECK-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-NEXT: faddp s0, v0.2s
; CHECK-NEXT: ret
%r = call fast float @llvm.vector.reduce.fadd.f32.v8f32(float -0.0, <8 x float> %bin.rdx)
ret float %r
}
define double @add_2D(<4 x double> %bin.rdx) {
; CHECK-LABEL: add_2D:
; CHECK: // %bb.0:
; CHECK-NEXT: fadd v0.2d, v0.2d, v1.2d
; CHECK-NEXT: faddp d0, v0.2d
; CHECK-NEXT: ret
%r = call fast double @llvm.vector.reduce.fadd.f64.v4f64(double -0.0, <4 x double> %bin.rdx)
ret double %r
}
; Added at least one test where the start value is not -0.0.
define float @add_S_init_42(<4 x float> %bin.rdx) {
; CHECK-LABEL: add_S_init_42:
; CHECK: // %bb.0:
; CHECK-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-NEXT: mov w8, #1109917696 // =0x42280000
; CHECK-NEXT: fmov s1, w8
; CHECK-NEXT: faddp s0, v0.2s
; CHECK-NEXT: fadd s0, s0, s1
; CHECK-NEXT: ret
%r = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float 42.0, <4 x float> %bin.rdx)
ret float %r
}
; The faddp.4s in the loop should not use v0.4s as second operand,
; because this introduces an unnecessary cross-iteration dependency.
define float @fadd_reduction_v4f32_in_loop(ptr %ptr.start) {
; CHECK-LABEL: fadd_reduction_v4f32_in_loop:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: movi d0, #0000000000000000
; CHECK-NEXT: mov x8, xzr
; CHECK-NEXT: .LBB11_1: // %loop
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: ldr q1, [x0, x8]
; CHECK-NEXT: add x8, x8, #16
; CHECK-NEXT: cmp w8, #112
; CHECK-NEXT: faddp v1.4s, v1.4s, v1.4s
; CHECK-NEXT: faddp s1, v1.2s
; CHECK-NEXT: fadd s0, s1, s0
; CHECK-NEXT: b.ne .LBB11_1
; CHECK-NEXT: // %bb.2: // %exit
; CHECK-NEXT: ret
entry:
br label %loop
loop:
%iv = phi i32 [ 1, %entry ], [ %iv.next, %loop ]
%ptr = phi ptr [ %ptr.start, %entry ], [ %ptr.next, %loop ]
%red = phi float [ 0.000000e+00, %entry ], [ %red.next, %loop ]
%lv = load <4 x float>, ptr %ptr, align 4
%r = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %lv)
%red.next = fadd fast float %r, %red
%ec = icmp eq i32 %iv, 7
%ptr.next = getelementptr inbounds float, ptr %ptr, i64 4
%iv.next= add nuw nsw i32 %iv, 1
br i1 %ec, label %exit, label %loop
exit:
ret float %red.next
}
; The faddp.4h in the loop should not use v0.4h as second operand,
; because this introduces an unnecessary cross-iteration dependency.
define half @fadd_reduction_v4f16_in_loop(ptr %ptr.start) {
; CHECK-SD-NOFP16-LABEL: fadd_reduction_v4f16_in_loop:
; CHECK-SD-NOFP16: // %bb.0: // %entry
; CHECK-SD-NOFP16-NEXT: movi d0, #0000000000000000
; CHECK-SD-NOFP16-NEXT: mov x8, xzr
; CHECK-SD-NOFP16-NEXT: .LBB12_1: // %loop
; CHECK-SD-NOFP16-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-SD-NOFP16-NEXT: ldr d1, [x0, x8]
; CHECK-SD-NOFP16-NEXT: fcvt s0, h0
; CHECK-SD-NOFP16-NEXT: add x8, x8, #8
; CHECK-SD-NOFP16-NEXT: cmp w8, #56
; CHECK-SD-NOFP16-NEXT: mov h2, v1.h[1]
; CHECK-SD-NOFP16-NEXT: mov h3, v1.h[3]
; CHECK-SD-NOFP16-NEXT: mov h4, v1.h[2]
; CHECK-SD-NOFP16-NEXT: fcvt s1, h1
; CHECK-SD-NOFP16-NEXT: fcvt s2, h2
; CHECK-SD-NOFP16-NEXT: fcvt s4, h4
; CHECK-SD-NOFP16-NEXT: fcvt s3, h3
; CHECK-SD-NOFP16-NEXT: fadd s1, s1, s2
; CHECK-SD-NOFP16-NEXT: fadd s2, s4, s3
; CHECK-SD-NOFP16-NEXT: fadd s1, s1, s2
; CHECK-SD-NOFP16-NEXT: fadd s0, s1, s0
; CHECK-SD-NOFP16-NEXT: fcvt h0, s0
; CHECK-SD-NOFP16-NEXT: b.ne .LBB12_1
; CHECK-SD-NOFP16-NEXT: // %bb.2: // %exit
; CHECK-SD-NOFP16-NEXT: ret
;
; CHECK-SD-FP16-LABEL: fadd_reduction_v4f16_in_loop:
; CHECK-SD-FP16: // %bb.0: // %entry
; CHECK-SD-FP16-NEXT: movi d0, #0000000000000000
; CHECK-SD-FP16-NEXT: mov x8, xzr
; CHECK-SD-FP16-NEXT: .LBB12_1: // %loop
; CHECK-SD-FP16-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-SD-FP16-NEXT: ldr d1, [x0, x8]
; CHECK-SD-FP16-NEXT: add x8, x8, #8
; CHECK-SD-FP16-NEXT: cmp w8, #56
; CHECK-SD-FP16-NEXT: faddp v1.4h, v1.4h, v1.4h
; CHECK-SD-FP16-NEXT: faddp h1, v1.2h
; CHECK-SD-FP16-NEXT: fadd h0, h1, h0
; CHECK-SD-FP16-NEXT: b.ne .LBB12_1
; CHECK-SD-FP16-NEXT: // %bb.2: // %exit
; CHECK-SD-FP16-NEXT: ret
;
; CHECK-GI-NOFP16-LABEL: fadd_reduction_v4f16_in_loop:
; CHECK-GI-NOFP16: // %bb.0: // %entry
; CHECK-GI-NOFP16-NEXT: mov x8, xzr
; CHECK-GI-NOFP16-NEXT: mov w9, #0 // =0x0
; CHECK-GI-NOFP16-NEXT: .LBB12_1: // %loop
; CHECK-GI-NOFP16-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-GI-NOFP16-NEXT: ldr d0, [x0, x8]
; CHECK-GI-NOFP16-NEXT: fmov s1, w9
; CHECK-GI-NOFP16-NEXT: add x8, x8, #8
; CHECK-GI-NOFP16-NEXT: cmp w8, #56
; CHECK-GI-NOFP16-NEXT: fcvtl v0.4s, v0.4h
; CHECK-GI-NOFP16-NEXT: fcvt s1, h1
; CHECK-GI-NOFP16-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: faddp s0, v0.2s
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: fcvt s0, h0
; CHECK-GI-NOFP16-NEXT: fadd s0, s0, s1
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: fmov w9, s0
; CHECK-GI-NOFP16-NEXT: b.ne .LBB12_1
; CHECK-GI-NOFP16-NEXT: // %bb.2: // %exit
; CHECK-GI-NOFP16-NEXT: // kill: def $h0 killed $h0 killed $s0
; CHECK-GI-NOFP16-NEXT: ret
;
; CHECK-GI-FP16-LABEL: fadd_reduction_v4f16_in_loop:
; CHECK-GI-FP16: // %bb.0: // %entry
; CHECK-GI-FP16-NEXT: movi d0, #0000000000000000
; CHECK-GI-FP16-NEXT: mov x8, xzr
; CHECK-GI-FP16-NEXT: .LBB12_1: // %loop
; CHECK-GI-FP16-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-GI-FP16-NEXT: ldr d1, [x0, x8]
; CHECK-GI-FP16-NEXT: add x8, x8, #8
; CHECK-GI-FP16-NEXT: cmp w8, #56
; CHECK-GI-FP16-NEXT: faddp v1.4h, v1.4h, v1.4h
; CHECK-GI-FP16-NEXT: faddp h1, v1.2h
; CHECK-GI-FP16-NEXT: fadd h0, h1, h0
; CHECK-GI-FP16-NEXT: b.ne .LBB12_1
; CHECK-GI-FP16-NEXT: // %bb.2: // %exit
; CHECK-GI-FP16-NEXT: ret
entry:
br label %loop
loop:
%iv = phi i32 [ 1, %entry ], [ %iv.next, %loop ]
%ptr = phi ptr [ %ptr.start, %entry ], [ %ptr.next, %loop ]
%red = phi half [ 0.000000e+00, %entry ], [ %red.next, %loop ]
%lv = load <4 x half>, ptr %ptr, align 4
%r = call fast half @llvm.vector.reduce.fadd.f16.v4f16(half -0.0, <4 x half> %lv)
%red.next = fadd fast half %r, %red
%ec = icmp eq i32 %iv, 7
%ptr.next = getelementptr inbounds half, ptr %ptr, i64 4
%iv.next= add nuw nsw i32 %iv, 1
br i1 %ec, label %exit, label %loop
exit:
ret half %red.next
}
; The faddp.8h in the loop should not use v0.8h as second operand,
; because this introduces an unnecessary cross-iteration dependency.
define half @fadd_reduction_v8f16_in_loop(ptr %ptr.start) {
; CHECK-SD-NOFP16-LABEL: fadd_reduction_v8f16_in_loop:
; CHECK-SD-NOFP16: // %bb.0: // %entry
; CHECK-SD-NOFP16-NEXT: movi d0, #0000000000000000
; CHECK-SD-NOFP16-NEXT: mov x8, xzr
; CHECK-SD-NOFP16-NEXT: .LBB13_1: // %loop
; CHECK-SD-NOFP16-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-SD-NOFP16-NEXT: ldr q1, [x0, x8]
; CHECK-SD-NOFP16-NEXT: fcvt s0, h0
; CHECK-SD-NOFP16-NEXT: add x8, x8, #8
; CHECK-SD-NOFP16-NEXT: cmp w8, #56
; CHECK-SD-NOFP16-NEXT: mov h2, v1.h[3]
; CHECK-SD-NOFP16-NEXT: mov h3, v1.h[2]
; CHECK-SD-NOFP16-NEXT: mov h4, v1.h[1]
; CHECK-SD-NOFP16-NEXT: mov h5, v1.h[5]
; CHECK-SD-NOFP16-NEXT: mov h6, v1.h[4]
; CHECK-SD-NOFP16-NEXT: fcvt s7, h1
; CHECK-SD-NOFP16-NEXT: mov h16, v1.h[6]
; CHECK-SD-NOFP16-NEXT: mov h1, v1.h[7]
; CHECK-SD-NOFP16-NEXT: fcvt s4, h4
; CHECK-SD-NOFP16-NEXT: fcvt s3, h3
; CHECK-SD-NOFP16-NEXT: fcvt s2, h2
; CHECK-SD-NOFP16-NEXT: fcvt s6, h6
; CHECK-SD-NOFP16-NEXT: fcvt s5, h5
; CHECK-SD-NOFP16-NEXT: fcvt s1, h1
; CHECK-SD-NOFP16-NEXT: fadd s4, s7, s4
; CHECK-SD-NOFP16-NEXT: fadd s2, s3, s2
; CHECK-SD-NOFP16-NEXT: fcvt s3, h16
; CHECK-SD-NOFP16-NEXT: fadd s5, s6, s5
; CHECK-SD-NOFP16-NEXT: fadd s0, s1, s0
; CHECK-SD-NOFP16-NEXT: fadd s2, s4, s2
; CHECK-SD-NOFP16-NEXT: fadd s3, s5, s3
; CHECK-SD-NOFP16-NEXT: fadd s2, s2, s3
; CHECK-SD-NOFP16-NEXT: fadd s0, s2, s0
; CHECK-SD-NOFP16-NEXT: fcvt h0, s0
; CHECK-SD-NOFP16-NEXT: b.ne .LBB13_1
; CHECK-SD-NOFP16-NEXT: // %bb.2: // %exit
; CHECK-SD-NOFP16-NEXT: ret
;
; CHECK-SD-FP16-LABEL: fadd_reduction_v8f16_in_loop:
; CHECK-SD-FP16: // %bb.0: // %entry
; CHECK-SD-FP16-NEXT: movi d0, #0000000000000000
; CHECK-SD-FP16-NEXT: mov x8, xzr
; CHECK-SD-FP16-NEXT: .LBB13_1: // %loop
; CHECK-SD-FP16-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-SD-FP16-NEXT: ldr q1, [x0, x8]
; CHECK-SD-FP16-NEXT: add x8, x8, #8
; CHECK-SD-FP16-NEXT: cmp w8, #56
; CHECK-SD-FP16-NEXT: faddp v2.8h, v1.8h, v1.8h
; CHECK-SD-FP16-NEXT: faddp v1.8h, v2.8h, v1.8h
; CHECK-SD-FP16-NEXT: faddp h1, v1.2h
; CHECK-SD-FP16-NEXT: fadd h0, h1, h0
; CHECK-SD-FP16-NEXT: b.ne .LBB13_1
; CHECK-SD-FP16-NEXT: // %bb.2: // %exit
; CHECK-SD-FP16-NEXT: ret
;
; CHECK-GI-NOFP16-LABEL: fadd_reduction_v8f16_in_loop:
; CHECK-GI-NOFP16: // %bb.0: // %entry
; CHECK-GI-NOFP16-NEXT: mov x8, xzr
; CHECK-GI-NOFP16-NEXT: mov w9, #0 // =0x0
; CHECK-GI-NOFP16-NEXT: .LBB13_1: // %loop
; CHECK-GI-NOFP16-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-GI-NOFP16-NEXT: ldr q0, [x0, x8]
; CHECK-GI-NOFP16-NEXT: add x8, x8, #8
; CHECK-GI-NOFP16-NEXT: cmp w8, #56
; CHECK-GI-NOFP16-NEXT: fcvtl v1.4s, v0.4h
; CHECK-GI-NOFP16-NEXT: fcvtl2 v0.4s, v0.8h
; CHECK-GI-NOFP16-NEXT: fadd v0.4s, v1.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: fmov s1, w9
; CHECK-GI-NOFP16-NEXT: fcvt s1, h1
; CHECK-GI-NOFP16-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: faddp s0, v0.2s
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: fcvt s0, h0
; CHECK-GI-NOFP16-NEXT: fadd s0, s0, s1
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: fmov w9, s0
; CHECK-GI-NOFP16-NEXT: b.ne .LBB13_1
; CHECK-GI-NOFP16-NEXT: // %bb.2: // %exit
; CHECK-GI-NOFP16-NEXT: // kill: def $h0 killed $h0 killed $s0
; CHECK-GI-NOFP16-NEXT: ret
;
; CHECK-GI-FP16-LABEL: fadd_reduction_v8f16_in_loop:
; CHECK-GI-FP16: // %bb.0: // %entry
; CHECK-GI-FP16-NEXT: movi d0, #0000000000000000
; CHECK-GI-FP16-NEXT: mov x8, xzr
; CHECK-GI-FP16-NEXT: .LBB13_1: // %loop
; CHECK-GI-FP16-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-GI-FP16-NEXT: ldr q1, [x0, x8]
; CHECK-GI-FP16-NEXT: add x8, x8, #8
; CHECK-GI-FP16-NEXT: cmp w8, #56
; CHECK-GI-FP16-NEXT: faddp v2.8h, v1.8h, v1.8h
; CHECK-GI-FP16-NEXT: faddp v1.8h, v2.8h, v1.8h
; CHECK-GI-FP16-NEXT: faddp h1, v1.2h
; CHECK-GI-FP16-NEXT: fadd h0, h1, h0
; CHECK-GI-FP16-NEXT: b.ne .LBB13_1
; CHECK-GI-FP16-NEXT: // %bb.2: // %exit
; CHECK-GI-FP16-NEXT: ret
entry:
br label %loop
loop:
%iv = phi i32 [ 1, %entry ], [ %iv.next, %loop ]
%ptr = phi ptr [ %ptr.start, %entry ], [ %ptr.next, %loop ]
%red = phi half [ 0.000000e+00, %entry ], [ %red.next, %loop ]
%lv = load <8 x half>, ptr %ptr, align 4
%r = call fast half @llvm.vector.reduce.fadd.f16.v8f16(half -0.0, <8 x half> %lv)
%red.next = fadd fast half %r, %red
%ec = icmp eq i32 %iv, 7
%ptr.next = getelementptr inbounds half, ptr %ptr, i64 4
%iv.next= add nuw nsw i32 %iv, 1
br i1 %ec, label %exit, label %loop
exit:
ret half %red.next
}
define half @fadd_reduct_reassoc_v8f16(<8 x half> %a, <8 x half> %b) {
; CHECK-SD-NOFP16-LABEL: fadd_reduct_reassoc_v8f16:
; CHECK-SD-NOFP16: // %bb.0:
; CHECK-SD-NOFP16-NEXT: fcvtl v2.4s, v1.4h
; CHECK-SD-NOFP16-NEXT: fcvtl v3.4s, v0.4h
; CHECK-SD-NOFP16-NEXT: fcvtl2 v1.4s, v1.8h
; CHECK-SD-NOFP16-NEXT: fcvtl2 v0.4s, v0.8h
; CHECK-SD-NOFP16-NEXT: fadd v2.4s, v3.4s, v2.4s
; CHECK-SD-NOFP16-NEXT: fadd v0.4s, v0.4s, v1.4s
; CHECK-SD-NOFP16-NEXT: fcvtn v1.4h, v2.4s
; CHECK-SD-NOFP16-NEXT: fcvtn2 v1.8h, v0.4s
; CHECK-SD-NOFP16-NEXT: mov h0, v1.h[1]
; CHECK-SD-NOFP16-NEXT: mov h2, v1.h[2]
; CHECK-SD-NOFP16-NEXT: mov h3, v1.h[3]
; CHECK-SD-NOFP16-NEXT: mov h4, v1.h[4]
; CHECK-SD-NOFP16-NEXT: mov h5, v1.h[5]
; CHECK-SD-NOFP16-NEXT: fcvt s6, h1
; CHECK-SD-NOFP16-NEXT: mov h7, v1.h[6]
; CHECK-SD-NOFP16-NEXT: mov h1, v1.h[7]
; CHECK-SD-NOFP16-NEXT: fcvt s0, h0
; CHECK-SD-NOFP16-NEXT: fcvt s2, h2
; CHECK-SD-NOFP16-NEXT: fcvt s3, h3
; CHECK-SD-NOFP16-NEXT: fcvt s4, h4
; CHECK-SD-NOFP16-NEXT: fcvt s5, h5
; CHECK-SD-NOFP16-NEXT: fcvt s1, h1
; CHECK-SD-NOFP16-NEXT: fadd s0, s6, s0
; CHECK-SD-NOFP16-NEXT: fadd s2, s2, s3
; CHECK-SD-NOFP16-NEXT: fadd s3, s4, s5
; CHECK-SD-NOFP16-NEXT: fcvt s4, h7
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s2
; CHECK-SD-NOFP16-NEXT: fadd s2, s3, s4
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s2
; CHECK-SD-NOFP16-NEXT: fadd s0, s0, s1
; CHECK-SD-NOFP16-NEXT: fcvt h0, s0
; CHECK-SD-NOFP16-NEXT: ret
;
; CHECK-SD-FP16-LABEL: fadd_reduct_reassoc_v8f16:
; CHECK-SD-FP16: // %bb.0:
; CHECK-SD-FP16-NEXT: fadd v0.8h, v0.8h, v1.8h
; CHECK-SD-FP16-NEXT: faddp v1.8h, v0.8h, v0.8h
; CHECK-SD-FP16-NEXT: faddp v0.8h, v1.8h, v0.8h
; CHECK-SD-FP16-NEXT: faddp h0, v0.2h
; CHECK-SD-FP16-NEXT: ret
;
; CHECK-GI-NOFP16-LABEL: fadd_reduct_reassoc_v8f16:
; CHECK-GI-NOFP16: // %bb.0:
; CHECK-GI-NOFP16-NEXT: fcvtl v2.4s, v0.4h
; CHECK-GI-NOFP16-NEXT: fcvtl2 v0.4s, v0.8h
; CHECK-GI-NOFP16-NEXT: fcvtl v3.4s, v1.4h
; CHECK-GI-NOFP16-NEXT: fcvtl2 v1.4s, v1.8h
; CHECK-GI-NOFP16-NEXT: fadd v0.4s, v2.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: fadd v1.4s, v3.4s, v1.4s
; CHECK-GI-NOFP16-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NOFP16-NEXT: faddp v1.4s, v1.4s, v1.4s
; CHECK-GI-NOFP16-NEXT: faddp s0, v0.2s
; CHECK-GI-NOFP16-NEXT: faddp s1, v1.2s
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: fcvt h1, s1
; CHECK-GI-NOFP16-NEXT: fcvt s0, h0
; CHECK-GI-NOFP16-NEXT: fcvt s1, h1
; CHECK-GI-NOFP16-NEXT: fadd s0, s0, s1
; CHECK-GI-NOFP16-NEXT: fcvt h0, s0
; CHECK-GI-NOFP16-NEXT: ret
;
; CHECK-GI-FP16-LABEL: fadd_reduct_reassoc_v8f16:
; CHECK-GI-FP16: // %bb.0:
; CHECK-GI-FP16-NEXT: faddp v2.8h, v0.8h, v0.8h
; CHECK-GI-FP16-NEXT: faddp v3.8h, v1.8h, v1.8h
; CHECK-GI-FP16-NEXT: faddp v0.8h, v2.8h, v0.8h
; CHECK-GI-FP16-NEXT: faddp v1.8h, v3.8h, v1.8h
; CHECK-GI-FP16-NEXT: faddp h0, v0.2h
; CHECK-GI-FP16-NEXT: faddp h1, v1.2h
; CHECK-GI-FP16-NEXT: fadd h0, h0, h1
; CHECK-GI-FP16-NEXT: ret
%r1 = call fast half @llvm.vector.reduce.fadd.f16.v8f16(half -0.0, <8 x half> %a)
%r2 = call fast half @llvm.vector.reduce.fadd.f16.v8f16(half -0.0, <8 x half> %b)
%r = fadd fast half %r1, %r2
ret half %r
}
define float @fadd_reduct_reassoc_v8f32(<8 x float> %a, <8 x float> %b) {
; CHECK-SD-LABEL: fadd_reduct_reassoc_v8f32:
; CHECK-SD: // %bb.0:
; CHECK-SD-NEXT: fadd v2.4s, v2.4s, v3.4s
; CHECK-SD-NEXT: fadd v0.4s, v0.4s, v1.4s
; CHECK-SD-NEXT: fadd v0.4s, v0.4s, v2.4s
; CHECK-SD-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-SD-NEXT: faddp s0, v0.2s
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fadd_reduct_reassoc_v8f32:
; CHECK-GI: // %bb.0:
; CHECK-GI-NEXT: fadd v0.4s, v0.4s, v1.4s
; CHECK-GI-NEXT: fadd v1.4s, v2.4s, v3.4s
; CHECK-GI-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NEXT: faddp v1.4s, v1.4s, v1.4s
; CHECK-GI-NEXT: faddp s0, v0.2s
; CHECK-GI-NEXT: faddp s1, v1.2s
; CHECK-GI-NEXT: fadd s0, s0, s1
; CHECK-GI-NEXT: ret
%r1 = call fast float @llvm.vector.reduce.fadd.f32.v8f32(float -0.0, <8 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fadd.f32.v8f32(float -0.0, <8 x float> %b)
%r = fadd fast float %r1, %r2
ret float %r
}
define float @fadd_reduct_reassoc_v4f32(<4 x float> %a, <4 x float> %b) {
; CHECK-SD-LABEL: fadd_reduct_reassoc_v4f32:
; CHECK-SD: // %bb.0:
; CHECK-SD-NEXT: fadd v0.4s, v0.4s, v1.4s
; CHECK-SD-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-SD-NEXT: faddp s0, v0.2s
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fadd_reduct_reassoc_v4f32:
; CHECK-GI: // %bb.0:
; CHECK-GI-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NEXT: faddp v1.4s, v1.4s, v1.4s
; CHECK-GI-NEXT: faddp s0, v0.2s
; CHECK-GI-NEXT: faddp s1, v1.2s
; CHECK-GI-NEXT: fadd s0, s0, s1
; CHECK-GI-NEXT: ret
%r1 = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %b)
%r = fadd fast float %r1, %r2
ret float %r
}
define float @fadd_reduct_reassoc_v4f32_init(float %i, <4 x float> %a, <4 x float> %b) {
; CHECK-LABEL: fadd_reduct_reassoc_v4f32_init:
; CHECK: // %bb.0:
; CHECK-NEXT: faddp v1.4s, v1.4s, v1.4s
; CHECK-NEXT: faddp v2.4s, v2.4s, v2.4s
; CHECK-NEXT: faddp s1, v1.2s
; CHECK-NEXT: fadd s0, s0, s1
; CHECK-NEXT: faddp s1, v2.2s
; CHECK-NEXT: fadd s0, s0, s1
; CHECK-NEXT: ret
%r1 = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float %i, <4 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %b)
%r = fadd fast float %r1, %r2
ret float %r
}
define float @fadd_reduct_reassoc_v4v8f32(<4 x float> %a, <8 x float> %b) {
; CHECK-SD-LABEL: fadd_reduct_reassoc_v4v8f32:
; CHECK-SD: // %bb.0:
; CHECK-SD-NEXT: fadd v1.4s, v1.4s, v2.4s
; CHECK-SD-NEXT: fadd v0.4s, v0.4s, v1.4s
; CHECK-SD-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-SD-NEXT: faddp s0, v0.2s
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fadd_reduct_reassoc_v4v8f32:
; CHECK-GI: // %bb.0:
; CHECK-GI-NEXT: fadd v1.4s, v1.4s, v2.4s
; CHECK-GI-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-GI-NEXT: faddp v1.4s, v1.4s, v1.4s
; CHECK-GI-NEXT: faddp s0, v0.2s
; CHECK-GI-NEXT: faddp s1, v1.2s
; CHECK-GI-NEXT: fadd s0, s0, s1
; CHECK-GI-NEXT: ret
%r1 = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fadd.f32.v8f32(float -0.0, <8 x float> %b)
%r = fadd fast float %r1, %r2
ret float %r
}
define double @fadd_reduct_reassoc_v4f64(<4 x double> %a, <4 x double> %b) {
; CHECK-SD-LABEL: fadd_reduct_reassoc_v4f64:
; CHECK-SD: // %bb.0:
; CHECK-SD-NEXT: fadd v2.2d, v2.2d, v3.2d
; CHECK-SD-NEXT: fadd v0.2d, v0.2d, v1.2d
; CHECK-SD-NEXT: fadd v0.2d, v0.2d, v2.2d
; CHECK-SD-NEXT: faddp d0, v0.2d
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fadd_reduct_reassoc_v4f64:
; CHECK-GI: // %bb.0:
; CHECK-GI-NEXT: fadd v0.2d, v0.2d, v1.2d
; CHECK-GI-NEXT: fadd v1.2d, v2.2d, v3.2d
; CHECK-GI-NEXT: faddp d0, v0.2d
; CHECK-GI-NEXT: faddp d1, v1.2d
; CHECK-GI-NEXT: fadd d0, d0, d1
; CHECK-GI-NEXT: ret
%r1 = call fast double @llvm.vector.reduce.fadd.f64.v4f64(double -0.0, <4 x double> %a)
%r2 = call fast double @llvm.vector.reduce.fadd.f64.v4f64(double -0.0, <4 x double> %b)
%r = fadd fast double %r1, %r2
ret double %r
}
define float @fadd_reduct_reassoc_v4f32_extrause(<4 x float> %a, <4 x float> %b) {
; CHECK-LABEL: fadd_reduct_reassoc_v4f32_extrause:
; CHECK: // %bb.0:
; CHECK-NEXT: faddp v0.4s, v0.4s, v0.4s
; CHECK-NEXT: faddp v1.4s, v1.4s, v1.4s
; CHECK-NEXT: faddp s0, v0.2s
; CHECK-NEXT: faddp s1, v1.2s
; CHECK-NEXT: fadd s1, s0, s1
; CHECK-NEXT: fmul s0, s1, s0
; CHECK-NEXT: ret
%r1 = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %b)
%r = fadd fast float %r1, %r2
%p = fmul float %r, %r1
ret float %p
}
; Function Attrs: nounwind readnone
declare half @llvm.vector.reduce.fadd.f16.v4f16(half, <4 x half>)
declare half @llvm.vector.reduce.fadd.f16.v8f16(half, <8 x half>)
declare half @llvm.vector.reduce.fadd.f16.v16f16(half, <16 x half>)
declare float @llvm.vector.reduce.fadd.f32.v2f32(float, <2 x float>)
declare float @llvm.vector.reduce.fadd.f32.v4f32(float, <4 x float>)
declare float @llvm.vector.reduce.fadd.f32.v8f32(float, <8 x float>)
declare double @llvm.vector.reduce.fadd.f64.v2f64(double, <2 x double>)
declare double @llvm.vector.reduce.fadd.f64.v4f64(double, <4 x double>)
|
