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
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+v -riscv-v-vector-bits-min=256 | FileCheck %s --check-prefixes=CHECK,V
; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+f,+zve32f -riscv-v-vector-bits-min=256 | FileCheck %s --check-prefixes=CHECK,ZVE32F
; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+v,+no-optimized-zero-stride-load -riscv-v-vector-bits-min=256 | FileCheck %s --check-prefixes=CHECK,NOT-OPTIMIZED
; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+f,+zve32f,+no-optimized-zero-stride-load -riscv-v-vector-bits-min=256 | FileCheck %s --check-prefixes=CHECK,NOT-OPTIMIZED
%struct.foo = type { i32, i32, i32, i32 }
; void gather(signed char * __restrict A, signed char * __restrict B) {
; for (int i = 0; i != 1024; ++i)
; A[i] += B[i * 5];
; }
define void @gather(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; CHECK-LABEL: gather:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: li a2, 1024
; CHECK-NEXT: li a3, 32
; CHECK-NEXT: li a4, 5
; CHECK-NEXT: .LBB0_1: # %vector.body
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: vsetvli zero, a3, e8, m1, ta, ma
; CHECK-NEXT: vlse8.v v8, (a1), a4
; CHECK-NEXT: vle8.v v9, (a0)
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vse8.v v8, (a0)
; CHECK-NEXT: addi a2, a2, -32
; CHECK-NEXT: addi a0, a0, 32
; CHECK-NEXT: addi a1, a1, 160
; CHECK-NEXT: bnez a2, .LBB0_1
; CHECK-NEXT: # %bb.2: # %for.cond.cleanup
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <32 x i64> [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15, i64 16, i64 17, i64 18, i64 19, i64 20, i64 21, i64 22, i64 23, i64 24, i64 25, i64 26, i64 27, i64 28, i64 29, i64 30, i64 31>, %entry ], [ %vec.ind.next, %vector.body ]
%i = mul nuw nsw <32 x i64> %vec.ind, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i1 = getelementptr inbounds i8, ptr %B, <32 x i64> %i
%wide.masked.gather = call <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr> %i1, i32 1, <32 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <32 x i8> undef)
%i2 = getelementptr inbounds i8, ptr %A, i64 %index
%wide.load = load <32 x i8>, ptr %i2, align 1
%i4 = add <32 x i8> %wide.load, %wide.masked.gather
store <32 x i8> %i4, ptr %i2, align 1
%index.next = add nuw i64 %index, 32
%vec.ind.next = add <32 x i64> %vec.ind, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i6 = icmp eq i64 %index.next, 1024
br i1 %i6, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
define void @gather_masked(ptr noalias nocapture %A, ptr noalias nocapture readonly %B, <32 x i8> %maskedoff) {
; V-LABEL: gather_masked:
; V: # %bb.0: # %entry
; V-NEXT: li a2, 1024
; V-NEXT: lui a3, 983765
; V-NEXT: addiw a3, a3, 873
; V-NEXT: vsetivli zero, 1, e32, mf2, ta, ma
; V-NEXT: vmv.s.x v0, a3
; V-NEXT: li a3, 32
; V-NEXT: li a4, 5
; V-NEXT: .LBB1_1: # %vector.body
; V-NEXT: # =>This Inner Loop Header: Depth=1
; V-NEXT: vsetvli zero, a3, e8, m1, ta, mu
; V-NEXT: vmv1r.v v9, v8
; V-NEXT: vlse8.v v9, (a1), a4, v0.t
; V-NEXT: vle8.v v10, (a0)
; V-NEXT: vadd.vv v9, v10, v9
; V-NEXT: vse8.v v9, (a0)
; V-NEXT: addi a2, a2, -32
; V-NEXT: addi a0, a0, 32
; V-NEXT: addi a1, a1, 160
; V-NEXT: bnez a2, .LBB1_1
; V-NEXT: # %bb.2: # %for.cond.cleanup
; V-NEXT: ret
;
; ZVE32F-LABEL: gather_masked:
; ZVE32F: # %bb.0: # %entry
; ZVE32F-NEXT: li a2, 1024
; ZVE32F-NEXT: lui a3, 983765
; ZVE32F-NEXT: addiw a3, a3, 873
; ZVE32F-NEXT: vsetivli zero, 1, e32, m1, ta, ma
; ZVE32F-NEXT: vmv.s.x v0, a3
; ZVE32F-NEXT: li a3, 32
; ZVE32F-NEXT: li a4, 5
; ZVE32F-NEXT: .LBB1_1: # %vector.body
; ZVE32F-NEXT: # =>This Inner Loop Header: Depth=1
; ZVE32F-NEXT: vsetvli zero, a3, e8, m1, ta, mu
; ZVE32F-NEXT: vmv1r.v v9, v8
; ZVE32F-NEXT: vlse8.v v9, (a1), a4, v0.t
; ZVE32F-NEXT: vle8.v v10, (a0)
; ZVE32F-NEXT: vadd.vv v9, v10, v9
; ZVE32F-NEXT: vse8.v v9, (a0)
; ZVE32F-NEXT: addi a2, a2, -32
; ZVE32F-NEXT: addi a0, a0, 32
; ZVE32F-NEXT: addi a1, a1, 160
; ZVE32F-NEXT: bnez a2, .LBB1_1
; ZVE32F-NEXT: # %bb.2: # %for.cond.cleanup
; ZVE32F-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <32 x i64> [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15, i64 16, i64 17, i64 18, i64 19, i64 20, i64 21, i64 22, i64 23, i64 24, i64 25, i64 26, i64 27, i64 28, i64 29, i64 30, i64 31>, %entry ], [ %vec.ind.next, %vector.body ]
%i = mul nuw nsw <32 x i64> %vec.ind, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i1 = getelementptr inbounds i8, ptr %B, <32 x i64> %i
%wide.masked.gather = call <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr> %i1, i32 1, <32 x i1> <i1 true, i1 false, i1 false, i1 true, i1 false, i1 true, i1 true, i1 false, i1 true, i1 true, i1 false, i1 false, i1 true, i1 false, i1 true, i1 false, i1 true, i1 false, i1 true, i1 true, i1 false, i1 true, i1 false, i1 false, i1 false, i1 false, i1 false, i1 false, i1 true, i1 true, i1 true, i1 true>, <32 x i8> %maskedoff)
%i2 = getelementptr inbounds i8, ptr %A, i64 %index
%wide.load = load <32 x i8>, ptr %i2, align 1
%i4 = add <32 x i8> %wide.load, %wide.masked.gather
store <32 x i8> %i4, ptr %i2, align 1
%index.next = add nuw i64 %index, 32
%vec.ind.next = add <32 x i64> %vec.ind, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i6 = icmp eq i64 %index.next, 1024
br i1 %i6, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
define void @gather_negative_stride(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; CHECK-LABEL: gather_negative_stride:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: addi a1, a1, 155
; CHECK-NEXT: li a2, 1024
; CHECK-NEXT: li a3, 32
; CHECK-NEXT: li a4, -5
; CHECK-NEXT: .LBB2_1: # %vector.body
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: vsetvli zero, a3, e8, m1, ta, ma
; CHECK-NEXT: vlse8.v v8, (a1), a4
; CHECK-NEXT: vle8.v v9, (a0)
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vse8.v v8, (a0)
; CHECK-NEXT: addi a2, a2, -32
; CHECK-NEXT: addi a0, a0, 32
; CHECK-NEXT: addi a1, a1, 160
; CHECK-NEXT: bnez a2, .LBB2_1
; CHECK-NEXT: # %bb.2: # %for.cond.cleanup
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <32 x i64> [ <i64 31, i64 30, i64 29, i64 28, i64 27, i64 26, i64 25, i64 24, i64 23, i64 22, i64 21, i64 20, i64 19, i64 18, i64 17, i64 16, i64 15, i64 14, i64 13, i64 12, i64 11, i64 10, i64 9, i64 8, i64 7, i64 6, i64 5, i64 4, i64 3, i64 2, i64 1, i64 0>, %entry ], [ %vec.ind.next, %vector.body ]
%i = mul nuw nsw <32 x i64> %vec.ind, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i1 = getelementptr inbounds i8, ptr %B, <32 x i64> %i
%wide.masked.gather = call <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr> %i1, i32 1, <32 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <32 x i8> undef)
%i2 = getelementptr inbounds i8, ptr %A, i64 %index
%wide.load = load <32 x i8>, ptr %i2, align 1
%i4 = add <32 x i8> %wide.load, %wide.masked.gather
store <32 x i8> %i4, ptr %i2, align 1
%index.next = add nuw i64 %index, 32
%vec.ind.next = add <32 x i64> %vec.ind, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i6 = icmp eq i64 %index.next, 1024
br i1 %i6, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
define void @gather_zero_stride(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; CHECK-LABEL: gather_zero_stride:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: li a2, 1024
; CHECK-NEXT: li a3, 32
; CHECK-NEXT: .LBB3_1: # %vector.body
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: lbu a4, 0(a1)
; CHECK-NEXT: vsetvli zero, a3, e8, m1, ta, ma
; CHECK-NEXT: vle8.v v8, (a0)
; CHECK-NEXT: vadd.vx v8, v8, a4
; CHECK-NEXT: vse8.v v8, (a0)
; CHECK-NEXT: addi a2, a2, -32
; CHECK-NEXT: addi a0, a0, 32
; CHECK-NEXT: addi a1, a1, 160
; CHECK-NEXT: bnez a2, .LBB3_1
; CHECK-NEXT: # %bb.2: # %for.cond.cleanup
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <32 x i64> [ zeroinitializer, %entry ], [ %vec.ind.next, %vector.body ]
%i = mul nuw nsw <32 x i64> %vec.ind, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i1 = getelementptr inbounds i8, ptr %B, <32 x i64> %i
%wide.masked.gather = call <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr> %i1, i32 1, <32 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <32 x i8> undef)
%i2 = getelementptr inbounds i8, ptr %A, i64 %index
%wide.load = load <32 x i8>, ptr %i2, align 1
%i4 = add <32 x i8> %wide.load, %wide.masked.gather
store <32 x i8> %i4, ptr %i2, align 1
%index.next = add nuw i64 %index, 32
%vec.ind.next = add <32 x i64> %vec.ind, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i6 = icmp eq i64 %index.next, 1024
br i1 %i6, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
define void @gather_zero_stride_unfold(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; V-LABEL: gather_zero_stride_unfold:
; V: # %bb.0: # %entry
; V-NEXT: li a2, 1024
; V-NEXT: li a3, 32
; V-NEXT: .LBB4_1: # %vector.body
; V-NEXT: # =>This Inner Loop Header: Depth=1
; V-NEXT: vsetvli zero, a3, e8, m1, ta, ma
; V-NEXT: vlse8.v v8, (a1), zero
; V-NEXT: vle8.v v9, (a0)
; V-NEXT: vdivu.vv v8, v8, v9
; V-NEXT: vse8.v v8, (a0)
; V-NEXT: addi a2, a2, -32
; V-NEXT: addi a0, a0, 32
; V-NEXT: addi a1, a1, 160
; V-NEXT: bnez a2, .LBB4_1
; V-NEXT: # %bb.2: # %for.cond.cleanup
; V-NEXT: ret
;
; ZVE32F-LABEL: gather_zero_stride_unfold:
; ZVE32F: # %bb.0: # %entry
; ZVE32F-NEXT: li a2, 1024
; ZVE32F-NEXT: li a3, 32
; ZVE32F-NEXT: .LBB4_1: # %vector.body
; ZVE32F-NEXT: # =>This Inner Loop Header: Depth=1
; ZVE32F-NEXT: vsetvli zero, a3, e8, m1, ta, ma
; ZVE32F-NEXT: vlse8.v v8, (a1), zero
; ZVE32F-NEXT: vle8.v v9, (a0)
; ZVE32F-NEXT: vdivu.vv v8, v8, v9
; ZVE32F-NEXT: vse8.v v8, (a0)
; ZVE32F-NEXT: addi a2, a2, -32
; ZVE32F-NEXT: addi a0, a0, 32
; ZVE32F-NEXT: addi a1, a1, 160
; ZVE32F-NEXT: bnez a2, .LBB4_1
; ZVE32F-NEXT: # %bb.2: # %for.cond.cleanup
; ZVE32F-NEXT: ret
;
; NOT-OPTIMIZED-LABEL: gather_zero_stride_unfold:
; NOT-OPTIMIZED: # %bb.0: # %entry
; NOT-OPTIMIZED-NEXT: li a2, 1024
; NOT-OPTIMIZED-NEXT: li a3, 32
; NOT-OPTIMIZED-NEXT: .LBB4_1: # %vector.body
; NOT-OPTIMIZED-NEXT: # =>This Inner Loop Header: Depth=1
; NOT-OPTIMIZED-NEXT: lbu a4, 0(a1)
; NOT-OPTIMIZED-NEXT: vsetvli zero, a3, e8, m1, ta, ma
; NOT-OPTIMIZED-NEXT: vle8.v v8, (a0)
; NOT-OPTIMIZED-NEXT: vmv.v.x v9, a4
; NOT-OPTIMIZED-NEXT: vdivu.vv v8, v9, v8
; NOT-OPTIMIZED-NEXT: vse8.v v8, (a0)
; NOT-OPTIMIZED-NEXT: addi a2, a2, -32
; NOT-OPTIMIZED-NEXT: addi a0, a0, 32
; NOT-OPTIMIZED-NEXT: addi a1, a1, 160
; NOT-OPTIMIZED-NEXT: bnez a2, .LBB4_1
; NOT-OPTIMIZED-NEXT: # %bb.2: # %for.cond.cleanup
; NOT-OPTIMIZED-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <32 x i64> [ zeroinitializer, %entry ], [ %vec.ind.next, %vector.body ]
%i = mul nuw nsw <32 x i64> %vec.ind, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i1 = getelementptr inbounds i8, ptr %B, <32 x i64> %i
%wide.masked.gather = call <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr> %i1, i32 1, <32 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <32 x i8> undef)
%i2 = getelementptr inbounds i8, ptr %A, i64 %index
%wide.load = load <32 x i8>, ptr %i2, align 1
%i4 = udiv <32 x i8> %wide.masked.gather, %wide.load
store <32 x i8> %i4, ptr %i2, align 1
%index.next = add nuw i64 %index, 32
%vec.ind.next = add <32 x i64> %vec.ind, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i6 = icmp eq i64 %index.next, 1024
br i1 %i6, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
;void scatter(signed char * __restrict A, signed char * __restrict B) {
; for (int i = 0; i < 1024; ++i)
; A[i * 5] += B[i];
;}
define void @scatter(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; CHECK-LABEL: scatter:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: li a2, 1024
; CHECK-NEXT: li a3, 32
; CHECK-NEXT: li a4, 5
; CHECK-NEXT: .LBB5_1: # %vector.body
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: vsetvli zero, a3, e8, m1, ta, ma
; CHECK-NEXT: vle8.v v8, (a1)
; CHECK-NEXT: vlse8.v v9, (a0), a4
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vsse8.v v8, (a0), a4
; CHECK-NEXT: addi a2, a2, -32
; CHECK-NEXT: addi a1, a1, 32
; CHECK-NEXT: addi a0, a0, 160
; CHECK-NEXT: bnez a2, .LBB5_1
; CHECK-NEXT: # %bb.2: # %for.cond.cleanup
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <32 x i64> [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15, i64 16, i64 17, i64 18, i64 19, i64 20, i64 21, i64 22, i64 23, i64 24, i64 25, i64 26, i64 27, i64 28, i64 29, i64 30, i64 31>, %entry ], [ %vec.ind.next, %vector.body ]
%i = getelementptr inbounds i8, ptr %B, i64 %index
%wide.load = load <32 x i8>, ptr %i, align 1
%i2 = mul nuw nsw <32 x i64> %vec.ind, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i3 = getelementptr inbounds i8, ptr %A, <32 x i64> %i2
%wide.masked.gather = call <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr> %i3, i32 1, <32 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <32 x i8> undef)
%i4 = add <32 x i8> %wide.masked.gather, %wide.load
call void @llvm.masked.scatter.v32i8.v32p0(<32 x i8> %i4, <32 x ptr> %i3, i32 1, <32 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
%index.next = add nuw i64 %index, 32
%vec.ind.next = add <32 x i64> %vec.ind, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i5 = icmp eq i64 %index.next, 1024
br i1 %i5, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
define void @scatter_masked(ptr noalias nocapture %A, ptr noalias nocapture readonly %B, <32 x i8> %maskedoff) {
; V-LABEL: scatter_masked:
; V: # %bb.0: # %entry
; V-NEXT: li a2, 1024
; V-NEXT: li a3, 32
; V-NEXT: lui a4, 983765
; V-NEXT: addiw a4, a4, 873
; V-NEXT: vsetivli zero, 1, e32, mf2, ta, ma
; V-NEXT: vmv.s.x v0, a4
; V-NEXT: li a4, 5
; V-NEXT: .LBB6_1: # %vector.body
; V-NEXT: # =>This Inner Loop Header: Depth=1
; V-NEXT: vsetvli zero, a3, e8, m1, ta, mu
; V-NEXT: vle8.v v9, (a1)
; V-NEXT: vmv1r.v v10, v8
; V-NEXT: vlse8.v v10, (a0), a4, v0.t
; V-NEXT: vadd.vv v9, v10, v9
; V-NEXT: vsse8.v v9, (a0), a4, v0.t
; V-NEXT: addi a2, a2, -32
; V-NEXT: addi a1, a1, 32
; V-NEXT: addi a0, a0, 160
; V-NEXT: bnez a2, .LBB6_1
; V-NEXT: # %bb.2: # %for.cond.cleanup
; V-NEXT: ret
;
; ZVE32F-LABEL: scatter_masked:
; ZVE32F: # %bb.0: # %entry
; ZVE32F-NEXT: li a2, 1024
; ZVE32F-NEXT: li a3, 32
; ZVE32F-NEXT: lui a4, 983765
; ZVE32F-NEXT: addiw a4, a4, 873
; ZVE32F-NEXT: vsetivli zero, 1, e32, m1, ta, ma
; ZVE32F-NEXT: vmv.s.x v0, a4
; ZVE32F-NEXT: li a4, 5
; ZVE32F-NEXT: .LBB6_1: # %vector.body
; ZVE32F-NEXT: # =>This Inner Loop Header: Depth=1
; ZVE32F-NEXT: vsetvli zero, a3, e8, m1, ta, mu
; ZVE32F-NEXT: vle8.v v9, (a1)
; ZVE32F-NEXT: vmv1r.v v10, v8
; ZVE32F-NEXT: vlse8.v v10, (a0), a4, v0.t
; ZVE32F-NEXT: vadd.vv v9, v10, v9
; ZVE32F-NEXT: vsse8.v v9, (a0), a4, v0.t
; ZVE32F-NEXT: addi a2, a2, -32
; ZVE32F-NEXT: addi a1, a1, 32
; ZVE32F-NEXT: addi a0, a0, 160
; ZVE32F-NEXT: bnez a2, .LBB6_1
; ZVE32F-NEXT: # %bb.2: # %for.cond.cleanup
; ZVE32F-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <32 x i64> [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15, i64 16, i64 17, i64 18, i64 19, i64 20, i64 21, i64 22, i64 23, i64 24, i64 25, i64 26, i64 27, i64 28, i64 29, i64 30, i64 31>, %entry ], [ %vec.ind.next, %vector.body ]
%i = getelementptr inbounds i8, ptr %B, i64 %index
%wide.load = load <32 x i8>, ptr %i, align 1
%i2 = mul nuw nsw <32 x i64> %vec.ind, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i3 = getelementptr inbounds i8, ptr %A, <32 x i64> %i2
%wide.masked.gather = call <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr> %i3, i32 1, <32 x i1> <i1 true, i1 false, i1 false, i1 true, i1 false, i1 true, i1 true, i1 false, i1 true, i1 true, i1 false, i1 false, i1 true, i1 false, i1 true, i1 false, i1 true, i1 false, i1 true, i1 true, i1 false, i1 true, i1 false, i1 false, i1 false, i1 false, i1 false, i1 false, i1 true, i1 true, i1 true, i1 true>, <32 x i8> %maskedoff)
%i4 = add <32 x i8> %wide.masked.gather, %wide.load
call void @llvm.masked.scatter.v32i8.v32p0(<32 x i8> %i4, <32 x ptr> %i3, i32 1, <32 x i1> <i1 true, i1 false, i1 false, i1 true, i1 false, i1 true, i1 true, i1 false, i1 true, i1 true, i1 false, i1 false, i1 true, i1 false, i1 true, i1 false, i1 true, i1 false, i1 true, i1 true, i1 false, i1 true, i1 false, i1 false, i1 false, i1 false, i1 false, i1 false, i1 true, i1 true, i1 true, i1 true>)
%index.next = add nuw i64 %index, 32
%vec.ind.next = add <32 x i64> %vec.ind, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i5 = icmp eq i64 %index.next, 1024
br i1 %i5, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
; void gather_pow2(signed char * __restrict A, signed char * __restrict B) {
; for (int i = 0; i != 1024; ++i)
; A[i] += B[i * 4];
; }
define void @gather_pow2(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; CHECK-LABEL: gather_pow2:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: li a2, 1024
; CHECK-NEXT: li a3, 16
; CHECK-NEXT: li a4, 32
; CHECK-NEXT: .LBB7_1: # %vector.body
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: vsetivli zero, 8, e32, m1, ta, ma
; CHECK-NEXT: vlse32.v v8, (a1), a3
; CHECK-NEXT: vsetvli zero, a4, e8, m1, ta, ma
; CHECK-NEXT: vle8.v v9, (a0)
; CHECK-NEXT: vsetivli zero, 8, e32, m1, ta, ma
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vsetvli zero, a4, e8, m1, ta, ma
; CHECK-NEXT: vse8.v v8, (a0)
; CHECK-NEXT: addi a2, a2, -8
; CHECK-NEXT: addi a0, a0, 32
; CHECK-NEXT: addi a1, a1, 128
; CHECK-NEXT: bnez a2, .LBB7_1
; CHECK-NEXT: # %bb.2: # %for.cond.cleanup
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <8 x i64> [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7>, %entry ], [ %vec.ind.next, %vector.body ]
%i = shl nsw <8 x i64> %vec.ind, <i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2>
%i1 = getelementptr inbounds i32, ptr %B, <8 x i64> %i
%wide.masked.gather = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i1, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i2 = getelementptr inbounds i32, ptr %A, i64 %index
%wide.load = load <8 x i32>, ptr %i2, align 1
%i4 = add <8 x i32> %wide.load, %wide.masked.gather
store <8 x i32> %i4, ptr %i2, align 1
%index.next = add nuw i64 %index, 8
%vec.ind.next = add <8 x i64> %vec.ind, <i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8>
%i6 = icmp eq i64 %index.next, 1024
br i1 %i6, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
;void scatter_pow2(signed char * __restrict A, signed char * __restrict B) {
; for (int i = 0; i < 1024; ++i)
; A[i * 4] += B[i];
;}
define void @scatter_pow2(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; CHECK-LABEL: scatter_pow2:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: li a2, 1024
; CHECK-NEXT: li a3, 32
; CHECK-NEXT: li a4, 16
; CHECK-NEXT: .LBB8_1: # %vector.body
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: vsetvli zero, a3, e8, m1, ta, ma
; CHECK-NEXT: vle8.v v8, (a1)
; CHECK-NEXT: vsetivli zero, 8, e32, m1, ta, ma
; CHECK-NEXT: vlse32.v v9, (a0), a4
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vsse32.v v8, (a0), a4
; CHECK-NEXT: addi a2, a2, -8
; CHECK-NEXT: addi a1, a1, 32
; CHECK-NEXT: addi a0, a0, 128
; CHECK-NEXT: bnez a2, .LBB8_1
; CHECK-NEXT: # %bb.2: # %for.cond.cleanup
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <8 x i64> [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7>, %entry ], [ %vec.ind.next, %vector.body ]
%i = getelementptr inbounds i32, ptr %B, i64 %index
%wide.load = load <8 x i32>, ptr %i, align 1
%i2 = shl nuw nsw <8 x i64> %vec.ind, <i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2>
%i3 = getelementptr inbounds i32, ptr %A, <8 x i64> %i2
%wide.masked.gather = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i3, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i4 = add <8 x i32> %wide.masked.gather, %wide.load
call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> %i4, <8 x ptr> %i3, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
%index.next = add nuw i64 %index, 8
%vec.ind.next = add <8 x i64> %vec.ind, <i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8>
%i5 = icmp eq i64 %index.next, 1024
br i1 %i5, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
;struct foo {
; int a, b, c, d;
;};
;
;void struct_gather(int * __restrict A, struct foo * __restrict B) {
; for (int i = 0; i < 1024; ++i)
; A[i] += B[i].b;
;}
define void @struct_gather(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; CHECK-LABEL: struct_gather:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: addi a1, a1, 132
; CHECK-NEXT: li a2, 1024
; CHECK-NEXT: li a3, 16
; CHECK-NEXT: vsetivli zero, 8, e32, m1, ta, ma
; CHECK-NEXT: .LBB9_1: # %vector.body
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: addi a4, a0, 32
; CHECK-NEXT: addi a5, a1, -128
; CHECK-NEXT: vlse32.v v8, (a5), a3
; CHECK-NEXT: vlse32.v v9, (a1), a3
; CHECK-NEXT: vle32.v v10, (a0)
; CHECK-NEXT: vle32.v v11, (a4)
; CHECK-NEXT: vadd.vv v8, v10, v8
; CHECK-NEXT: vadd.vv v9, v11, v9
; CHECK-NEXT: vse32.v v8, (a0)
; CHECK-NEXT: vse32.v v9, (a4)
; CHECK-NEXT: addi a2, a2, -16
; CHECK-NEXT: addi a0, a0, 64
; CHECK-NEXT: addi a1, a1, 256
; CHECK-NEXT: bnez a2, .LBB9_1
; CHECK-NEXT: # %bb.2: # %for.cond.cleanup
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <8 x i64> [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7>, %entry ], [ %vec.ind.next, %vector.body ]
%step.add = add <8 x i64> %vec.ind, <i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8>
%i = getelementptr inbounds %struct.foo, ptr %B, <8 x i64> %vec.ind, i32 1
%i1 = getelementptr inbounds %struct.foo, ptr %B, <8 x i64> %step.add, i32 1
%wide.masked.gather = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%wide.masked.gather9 = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i1, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i2 = getelementptr inbounds i32, ptr %A, i64 %index
%wide.load = load <8 x i32>, ptr %i2, align 4
%i4 = getelementptr inbounds i32, ptr %i2, i64 8
%wide.load10 = load <8 x i32>, ptr %i4, align 4
%i6 = add nsw <8 x i32> %wide.load, %wide.masked.gather
%i7 = add nsw <8 x i32> %wide.load10, %wide.masked.gather9
store <8 x i32> %i6, ptr %i2, align 4
store <8 x i32> %i7, ptr %i4, align 4
%index.next = add nuw i64 %index, 16
%vec.ind.next = add <8 x i64> %vec.ind, <i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16>
%i10 = icmp eq i64 %index.next, 1024
br i1 %i10, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
;void gather_unroll(int * __restrict A, int * __restrict B) {
; for (int i = 0; i < 1024; i+= 4 ) {
; A[i] += B[i * 4];
; A[i+1] += B[(i+1) * 4];
; A[i+2] += B[(i+2) * 4];
; A[i+3] += B[(i+3) * 4];
; }
;}
define void @gather_unroll(ptr noalias nocapture %A, ptr noalias nocapture readonly %B) {
; CHECK-LABEL: gather_unroll:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: li a2, 256
; CHECK-NEXT: li a3, 64
; CHECK-NEXT: li a4, 16
; CHECK-NEXT: vsetivli zero, 8, e32, m1, ta, ma
; CHECK-NEXT: .LBB10_1: # %vector.body
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: vlse32.v v8, (a1), a3
; CHECK-NEXT: vlse32.v v9, (a0), a4
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vsse32.v v8, (a0), a4
; CHECK-NEXT: addi a5, a1, 16
; CHECK-NEXT: vlse32.v v8, (a5), a3
; CHECK-NEXT: addi a5, a0, 4
; CHECK-NEXT: vlse32.v v9, (a5), a4
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vsse32.v v8, (a5), a4
; CHECK-NEXT: addi a5, a1, 32
; CHECK-NEXT: vlse32.v v8, (a5), a3
; CHECK-NEXT: addi a5, a0, 8
; CHECK-NEXT: vlse32.v v9, (a5), a4
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vsse32.v v8, (a5), a4
; CHECK-NEXT: addi a5, a1, 48
; CHECK-NEXT: vlse32.v v8, (a5), a3
; CHECK-NEXT: addi a5, a0, 12
; CHECK-NEXT: vlse32.v v9, (a5), a4
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vsse32.v v8, (a5), a4
; CHECK-NEXT: addi a2, a2, -8
; CHECK-NEXT: addi a1, a1, 512
; CHECK-NEXT: addi a0, a0, 128
; CHECK-NEXT: bnez a2, .LBB10_1
; CHECK-NEXT: # %bb.2: # %for.cond.cleanup
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%vec.ind = phi <8 x i64> [ <i64 0, i64 4, i64 8, i64 12, i64 16, i64 20, i64 24, i64 28>, %entry ], [ %vec.ind.next, %vector.body ]
%i = shl nuw nsw <8 x i64> %vec.ind, <i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2>
%i1 = getelementptr inbounds i32, ptr %B, <8 x i64> %i
%wide.masked.gather = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i1, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i2 = getelementptr inbounds i32, ptr %A, <8 x i64> %vec.ind
%wide.masked.gather52 = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i2, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i3 = add nsw <8 x i32> %wide.masked.gather52, %wide.masked.gather
call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> %i3, <8 x ptr> %i2, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
%i4 = or <8 x i64> %vec.ind, <i64 1, i64 1, i64 1, i64 1, i64 1, i64 1, i64 1, i64 1>
%i5 = shl nsw <8 x i64> %i4, <i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2>
%i6 = getelementptr inbounds i32, ptr %B, <8 x i64> %i5
%wide.masked.gather53 = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i6, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i7 = getelementptr inbounds i32, ptr %A, <8 x i64> %i4
%wide.masked.gather54 = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i7, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i8 = add nsw <8 x i32> %wide.masked.gather54, %wide.masked.gather53
call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> %i8, <8 x ptr> %i7, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
%i9 = or <8 x i64> %vec.ind, <i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2>
%i10 = shl nsw <8 x i64> %i9, <i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2>
%i11 = getelementptr inbounds i32, ptr %B, <8 x i64> %i10
%wide.masked.gather55 = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i11, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i12 = getelementptr inbounds i32, ptr %A, <8 x i64> %i9
%wide.masked.gather56 = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i12, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i13 = add nsw <8 x i32> %wide.masked.gather56, %wide.masked.gather55
call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> %i13, <8 x ptr> %i12, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
%i14 = or <8 x i64> %vec.ind, <i64 3, i64 3, i64 3, i64 3, i64 3, i64 3, i64 3, i64 3>
%i15 = shl nsw <8 x i64> %i14, <i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2, i64 2>
%i16 = getelementptr inbounds i32, ptr %B, <8 x i64> %i15
%wide.masked.gather57 = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i16, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i17 = getelementptr inbounds i32, ptr %A, <8 x i64> %i14
%wide.masked.gather58 = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> %i17, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <8 x i32> undef)
%i18 = add nsw <8 x i32> %wide.masked.gather58, %wide.masked.gather57
call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> %i18, <8 x ptr> %i17, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
%index.next = add nuw i64 %index, 8
%vec.ind.next = add <8 x i64> %vec.ind, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i19 = icmp eq i64 %index.next, 256
br i1 %i19, label %for.cond.cleanup, label %vector.body
for.cond.cleanup: ; preds = %vector.body
ret void
}
declare <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr>, i32 immarg, <32 x i1>, <32 x i8>)
declare <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr>, i32 immarg, <8 x i1>, <8 x i32>)
declare void @llvm.masked.scatter.v32i8.v32p0(<32 x i8>, <32 x ptr>, i32 immarg, <32 x i1>)
declare void @llvm.masked.scatter.v8i32.v8p0(<8 x i32>, <8 x ptr>, i32 immarg, <8 x i1>)
; Make sure we don't crash in getTgtMemIntrinsic for a vector of pointers.
define void @gather_of_pointers(ptr noalias nocapture %arg, ptr noalias nocapture readonly %arg1) {
; V-LABEL: gather_of_pointers:
; V: # %bb.0: # %bb
; V-NEXT: li a2, 1024
; V-NEXT: li a3, 40
; V-NEXT: vsetivli zero, 2, e64, m1, ta, ma
; V-NEXT: .LBB11_1: # %bb2
; V-NEXT: # =>This Inner Loop Header: Depth=1
; V-NEXT: addi a4, a1, 80
; V-NEXT: vlse64.v v8, (a1), a3
; V-NEXT: vlse64.v v9, (a4), a3
; V-NEXT: addi a4, a0, 16
; V-NEXT: vse64.v v8, (a0)
; V-NEXT: vse64.v v9, (a4)
; V-NEXT: addi a2, a2, -4
; V-NEXT: addi a0, a0, 32
; V-NEXT: addi a1, a1, 160
; V-NEXT: bnez a2, .LBB11_1
; V-NEXT: # %bb.2: # %bb18
; V-NEXT: ret
;
; ZVE32F-LABEL: gather_of_pointers:
; ZVE32F: # %bb.0: # %bb
; ZVE32F-NEXT: li a2, 0
; ZVE32F-NEXT: li a3, 1
; ZVE32F-NEXT: li a4, 1024
; ZVE32F-NEXT: li a5, 40
; ZVE32F-NEXT: .LBB11_1: # %bb2
; ZVE32F-NEXT: # =>This Inner Loop Header: Depth=1
; ZVE32F-NEXT: mul a6, a3, a5
; ZVE32F-NEXT: add a6, a1, a6
; ZVE32F-NEXT: mul a7, a2, a5
; ZVE32F-NEXT: add a7, a1, a7
; ZVE32F-NEXT: ld t0, 0(a6)
; ZVE32F-NEXT: ld t1, 0(a7)
; ZVE32F-NEXT: ld a6, 80(a6)
; ZVE32F-NEXT: ld a7, 80(a7)
; ZVE32F-NEXT: sd t0, 8(a0)
; ZVE32F-NEXT: sd t1, 0(a0)
; ZVE32F-NEXT: sd a6, 24(a0)
; ZVE32F-NEXT: sd a7, 16(a0)
; ZVE32F-NEXT: addi a2, a2, 4
; ZVE32F-NEXT: addi a3, a3, 4
; ZVE32F-NEXT: addi a4, a4, -4
; ZVE32F-NEXT: addi a0, a0, 32
; ZVE32F-NEXT: bnez a4, .LBB11_1
; ZVE32F-NEXT: # %bb.2: # %bb18
; ZVE32F-NEXT: ret
bb:
br label %bb2
bb2: ; preds = %bb2, %bb
%i = phi i64 [ 0, %bb ], [ %i15, %bb2 ]
%i3 = phi <2 x i64> [ <i64 0, i64 1>, %bb ], [ %i16, %bb2 ]
%i4 = mul nuw nsw <2 x i64> %i3, <i64 5, i64 5>
%i5 = mul <2 x i64> %i3, <i64 5, i64 5>
%i6 = add <2 x i64> %i5, <i64 10, i64 10>
%i7 = getelementptr inbounds ptr, ptr %arg1, <2 x i64> %i4
%i8 = getelementptr inbounds ptr, ptr %arg1, <2 x i64> %i6
%i9 = call <2 x ptr> @llvm.masked.gather.v2p0.v2p0(<2 x ptr> %i7, i32 8, <2 x i1> <i1 true, i1 true>, <2 x ptr> undef)
%i10 = call <2 x ptr> @llvm.masked.gather.v2p0.v2p0(<2 x ptr> %i8, i32 8, <2 x i1> <i1 true, i1 true>, <2 x ptr> undef)
%i11 = getelementptr inbounds ptr, ptr %arg, i64 %i
store <2 x ptr> %i9, ptr %i11, align 8
%i13 = getelementptr inbounds ptr, ptr %i11, i64 2
store <2 x ptr> %i10, ptr %i13, align 8
%i15 = add nuw i64 %i, 4
%i16 = add <2 x i64> %i3, <i64 4, i64 4>
%i17 = icmp eq i64 %i15, 1024
br i1 %i17, label %bb18, label %bb2
bb18: ; preds = %bb2
ret void
}
declare <2 x ptr> @llvm.masked.gather.v2p0.v2p0(<2 x ptr>, i32 immarg, <2 x i1>, <2 x ptr>)
; Make sure we don't crash in getTgtMemIntrinsic for a vector of pointers.
define void @scatter_of_pointers(ptr noalias nocapture %arg, ptr noalias nocapture readonly %arg1) {
; V-LABEL: scatter_of_pointers:
; V: # %bb.0: # %bb
; V-NEXT: li a2, 1024
; V-NEXT: li a3, 40
; V-NEXT: vsetivli zero, 2, e64, m1, ta, ma
; V-NEXT: .LBB12_1: # %bb2
; V-NEXT: # =>This Inner Loop Header: Depth=1
; V-NEXT: addi a4, a1, 16
; V-NEXT: vle64.v v8, (a1)
; V-NEXT: vle64.v v9, (a4)
; V-NEXT: addi a4, a0, 80
; V-NEXT: vsse64.v v8, (a0), a3
; V-NEXT: vsse64.v v9, (a4), a3
; V-NEXT: addi a2, a2, -4
; V-NEXT: addi a1, a1, 32
; V-NEXT: addi a0, a0, 160
; V-NEXT: bnez a2, .LBB12_1
; V-NEXT: # %bb.2: # %bb18
; V-NEXT: ret
;
; ZVE32F-LABEL: scatter_of_pointers:
; ZVE32F: # %bb.0: # %bb
; ZVE32F-NEXT: li a2, 0
; ZVE32F-NEXT: li a3, 1
; ZVE32F-NEXT: li a4, 1024
; ZVE32F-NEXT: li a5, 40
; ZVE32F-NEXT: .LBB12_1: # %bb2
; ZVE32F-NEXT: # =>This Inner Loop Header: Depth=1
; ZVE32F-NEXT: ld a6, 8(a1)
; ZVE32F-NEXT: ld a7, 0(a1)
; ZVE32F-NEXT: ld t0, 24(a1)
; ZVE32F-NEXT: ld t1, 16(a1)
; ZVE32F-NEXT: mul t2, a3, a5
; ZVE32F-NEXT: add t2, a0, t2
; ZVE32F-NEXT: mul t3, a2, a5
; ZVE32F-NEXT: add t3, a0, t3
; ZVE32F-NEXT: sd a7, 0(t3)
; ZVE32F-NEXT: sd a6, 0(t2)
; ZVE32F-NEXT: sd t1, 80(t3)
; ZVE32F-NEXT: sd t0, 80(t2)
; ZVE32F-NEXT: addi a2, a2, 4
; ZVE32F-NEXT: addi a3, a3, 4
; ZVE32F-NEXT: addi a4, a4, -4
; ZVE32F-NEXT: addi a1, a1, 32
; ZVE32F-NEXT: bnez a4, .LBB12_1
; ZVE32F-NEXT: # %bb.2: # %bb18
; ZVE32F-NEXT: ret
bb:
br label %bb2
bb2: ; preds = %bb2, %bb
%i = phi i64 [ 0, %bb ], [ %i15, %bb2 ]
%i3 = phi <2 x i64> [ <i64 0, i64 1>, %bb ], [ %i16, %bb2 ]
%i4 = getelementptr inbounds ptr, ptr %arg1, i64 %i
%i6 = load <2 x ptr>, ptr %i4, align 8
%i7 = getelementptr inbounds ptr, ptr %i4, i64 2
%i9 = load <2 x ptr>, ptr %i7, align 8
%i10 = mul nuw nsw <2 x i64> %i3, <i64 5, i64 5>
%i11 = mul <2 x i64> %i3, <i64 5, i64 5>
%i12 = add <2 x i64> %i11, <i64 10, i64 10>
%i13 = getelementptr inbounds ptr, ptr %arg, <2 x i64> %i10
%i14 = getelementptr inbounds ptr, ptr %arg, <2 x i64> %i12
call void @llvm.masked.scatter.v2p0.v2p0(<2 x ptr> %i6, <2 x ptr> %i13, i32 8, <2 x i1> <i1 true, i1 true>)
call void @llvm.masked.scatter.v2p0.v2p0(<2 x ptr> %i9, <2 x ptr> %i14, i32 8, <2 x i1> <i1 true, i1 true>)
%i15 = add nuw i64 %i, 4
%i16 = add <2 x i64> %i3, <i64 4, i64 4>
%i17 = icmp eq i64 %i15, 1024
br i1 %i17, label %bb18, label %bb2
bb18: ; preds = %bb2
ret void
}
declare void @llvm.masked.scatter.v2p0.v2p0(<2 x ptr>, <2 x ptr>, i32 immarg, <2 x i1>)
define void @strided_load_startval_add_with_splat(ptr noalias nocapture %arg, ptr noalias nocapture readonly %arg1, i32 signext %arg2) {
; CHECK-LABEL: strided_load_startval_add_with_splat:
; CHECK: # %bb.0: # %bb
; CHECK-NEXT: li a3, 1024
; CHECK-NEXT: beq a2, a3, .LBB13_7
; CHECK-NEXT: # %bb.1: # %bb3
; CHECK-NEXT: li a4, 1023
; CHECK-NEXT: subw a4, a4, a2
; CHECK-NEXT: li a5, 31
; CHECK-NEXT: mv a3, a2
; CHECK-NEXT: bltu a4, a5, .LBB13_5
; CHECK-NEXT: # %bb.2: # %bb9
; CHECK-NEXT: slli a4, a4, 32
; CHECK-NEXT: srli a4, a4, 32
; CHECK-NEXT: addi a4, a4, 1
; CHECK-NEXT: andi a5, a4, -32
; CHECK-NEXT: add a3, a5, a2
; CHECK-NEXT: slli a7, a2, 2
; CHECK-NEXT: add a6, a0, a2
; CHECK-NEXT: add a2, a1, a2
; CHECK-NEXT: add a2, a2, a7
; CHECK-NEXT: li a7, 32
; CHECK-NEXT: li t0, 5
; CHECK-NEXT: mv t1, a5
; CHECK-NEXT: .LBB13_3: # %bb15
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: vsetvli zero, a7, e8, m1, ta, ma
; CHECK-NEXT: vlse8.v v8, (a2), t0
; CHECK-NEXT: vle8.v v9, (a6)
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vse8.v v8, (a6)
; CHECK-NEXT: addi t1, t1, -32
; CHECK-NEXT: addi a6, a6, 32
; CHECK-NEXT: addi a2, a2, 160
; CHECK-NEXT: bnez t1, .LBB13_3
; CHECK-NEXT: # %bb.4: # %bb30
; CHECK-NEXT: beq a4, a5, .LBB13_7
; CHECK-NEXT: .LBB13_5: # %bb32
; CHECK-NEXT: addiw a2, a3, -1024
; CHECK-NEXT: add a0, a0, a3
; CHECK-NEXT: slli a4, a3, 2
; CHECK-NEXT: add a1, a1, a3
; CHECK-NEXT: add a1, a1, a4
; CHECK-NEXT: .LBB13_6: # %bb35
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: lb a3, 0(a1)
; CHECK-NEXT: lb a4, 0(a0)
; CHECK-NEXT: add a3, a4, a3
; CHECK-NEXT: sb a3, 0(a0)
; CHECK-NEXT: addiw a2, a2, 1
; CHECK-NEXT: addi a0, a0, 1
; CHECK-NEXT: addi a1, a1, 5
; CHECK-NEXT: bnez a2, .LBB13_6
; CHECK-NEXT: .LBB13_7: # %bb34
; CHECK-NEXT: ret
bb:
%i = icmp eq i32 %arg2, 1024
br i1 %i, label %bb34, label %bb3
bb3: ; preds = %bb
%i4 = sext i32 %arg2 to i64
%i5 = sub i32 1023, %arg2
%i6 = zext i32 %i5 to i64
%i7 = add nuw nsw i64 %i6, 1
%i8 = icmp ult i32 %i5, 31
br i1 %i8, label %bb32, label %bb9
bb9: ; preds = %bb3
%i10 = and i64 %i7, 8589934560
%i11 = add nsw i64 %i10, %i4
%i12 = insertelement <32 x i64> poison, i64 %i4, i64 0
%i13 = shufflevector <32 x i64> %i12, <32 x i64> poison, <32 x i32> zeroinitializer
%i14 = add <32 x i64> %i13, <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15, i64 16, i64 17, i64 18, i64 19, i64 20, i64 21, i64 22, i64 23, i64 24, i64 25, i64 26, i64 27, i64 28, i64 29, i64 30, i64 31>
br label %bb15
bb15: ; preds = %bb15, %bb9
%i16 = phi i64 [ 0, %bb9 ], [ %i27, %bb15 ]
%i17 = phi <32 x i64> [ %i14, %bb9 ], [ %i28, %bb15 ]
%i18 = add i64 %i16, %i4
%i19 = mul nsw <32 x i64> %i17, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i20 = getelementptr inbounds i8, ptr %arg1, <32 x i64> %i19
%i21 = call <32 x i8> @llvm.masked.gather.v32i8.v32p0(<32 x ptr> %i20, i32 1, <32 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <32 x i8> undef)
%i22 = getelementptr inbounds i8, ptr %arg, i64 %i18
%i24 = load <32 x i8>, ptr %i22, align 1
%i25 = add <32 x i8> %i24, %i21
store <32 x i8> %i25, ptr %i22, align 1
%i27 = add nuw i64 %i16, 32
%i28 = add <32 x i64> %i17, <i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32, i64 32>
%i29 = icmp eq i64 %i27, %i10
br i1 %i29, label %bb30, label %bb15
bb30: ; preds = %bb15
%i31 = icmp eq i64 %i7, %i10
br i1 %i31, label %bb34, label %bb32
bb32: ; preds = %bb30, %bb3
%i33 = phi i64 [ %i4, %bb3 ], [ %i11, %bb30 ]
br label %bb35
bb34: ; preds = %bb35, %bb30, %bb
ret void
bb35: ; preds = %bb35, %bb32
%i36 = phi i64 [ %i43, %bb35 ], [ %i33, %bb32 ]
%i37 = mul nsw i64 %i36, 5
%i38 = getelementptr inbounds i8, ptr %arg1, i64 %i37
%i39 = load i8, ptr %i38, align 1
%i40 = getelementptr inbounds i8, ptr %arg, i64 %i36
%i41 = load i8, ptr %i40, align 1
%i42 = add i8 %i41, %i39
store i8 %i42, ptr %i40, align 1
%i43 = add nsw i64 %i36, 1
%i44 = trunc i64 %i43 to i32
%i45 = icmp eq i32 %i44, 1024
br i1 %i45, label %bb34, label %bb35
}
declare <16 x i8> @llvm.masked.gather.v16i8.v16p0(<16 x ptr>, i32 immarg, <16 x i1>, <16 x i8>)
declare void @llvm.masked.scatter.v16i8.v16p0(<16 x i8>, <16 x ptr>, i32 immarg, <16 x i1>)
define void @gather_no_scalar_remainder(ptr noalias nocapture noundef %arg, ptr noalias nocapture noundef readonly %arg1, i64 noundef %arg2) {
; CHECK-LABEL: gather_no_scalar_remainder:
; CHECK: # %bb.0: # %bb
; CHECK-NEXT: slli a2, a2, 4
; CHECK-NEXT: beqz a2, .LBB14_3
; CHECK-NEXT: # %bb.1: # %bb2
; CHECK-NEXT: li a3, 5
; CHECK-NEXT: vsetivli zero, 16, e8, mf2, ta, ma
; CHECK-NEXT: .LBB14_2: # %bb4
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
; CHECK-NEXT: vlse8.v v8, (a1), a3
; CHECK-NEXT: vle8.v v9, (a0)
; CHECK-NEXT: vadd.vv v8, v9, v8
; CHECK-NEXT: vse8.v v8, (a0)
; CHECK-NEXT: addi a2, a2, -16
; CHECK-NEXT: addi a0, a0, 16
; CHECK-NEXT: addi a1, a1, 80
; CHECK-NEXT: bnez a2, .LBB14_2
; CHECK-NEXT: .LBB14_3: # %bb16
; CHECK-NEXT: ret
bb:
%i = shl i64 %arg2, 4
%i3 = icmp eq i64 %i, 0
br i1 %i3, label %bb16, label %bb2
bb2: ; preds = %bb
br label %bb4
bb4: ; preds = %bb4, %bb2
%i5 = phi i64 [ %i13, %bb4 ], [ 0, %bb2 ]
%i6 = phi <16 x i64> [ %i14, %bb4 ], [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15>, %bb2 ]
%i7 = mul <16 x i64> %i6, <i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5, i64 5>
%i8 = getelementptr inbounds i8, ptr %arg1, <16 x i64> %i7
%i9 = call <16 x i8> @llvm.masked.gather.v16i8.v16p0(<16 x ptr> %i8, i32 1, <16 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <16 x i8> undef)
%i10 = getelementptr inbounds i8, ptr %arg, i64 %i5
%i11 = load <16 x i8>, ptr %i10, align 1
%i12 = add <16 x i8> %i11, %i9
store <16 x i8> %i12, ptr %i10, align 1
%i13 = add nuw i64 %i5, 16
%i14 = add <16 x i64> %i6, <i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16, i64 16>
%i15 = icmp eq i64 %i13, %i
br i1 %i15, label %bb16, label %bb4
bb16: ; preds = %bb4, %bb
ret void
}
|
