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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=CHECK,SSE,SSE2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefixes=CHECK,SSE,SSE41
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=CHECK,AVX,AVX1
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX,AVX2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+xop | FileCheck %s --check-prefixes=CHECK,XOP
; fold (udiv x, 1) -> x
define i32 @combine_udiv_by_one(i32 %x) {
; CHECK-LABEL: combine_udiv_by_one:
; CHECK: # %bb.0:
; CHECK-NEXT: movl %edi, %eax
; CHECK-NEXT: retq
%1 = udiv i32 %x, 1
ret i32 %1
}
define <4 x i32> @combine_vec_udiv_by_one(<4 x i32> %x) {
; CHECK-LABEL: combine_vec_udiv_by_one:
; CHECK: # %bb.0:
; CHECK-NEXT: retq
%1 = udiv <4 x i32> %x, <i32 1, i32 1, i32 1, i32 1>
ret <4 x i32> %1
}
; fold (udiv x, -1) -> select((icmp eq x, -1), 1, 0)
define i32 @combine_udiv_by_negone(i32 %x) {
; CHECK-LABEL: combine_udiv_by_negone:
; CHECK: # %bb.0:
; CHECK-NEXT: xorl %eax, %eax
; CHECK-NEXT: cmpl $-1, %edi
; CHECK-NEXT: sete %al
; CHECK-NEXT: retq
%1 = udiv i32 %x, -1
ret i32 %1
}
define <4 x i32> @combine_vec_udiv_by_negone(<4 x i32> %x) {
; SSE-LABEL: combine_vec_udiv_by_negone:
; SSE: # %bb.0:
; SSE-NEXT: pcmpeqd %xmm1, %xmm1
; SSE-NEXT: pcmpeqd %xmm1, %xmm0
; SSE-NEXT: psrld $31, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_by_negone:
; AVX: # %bb.0:
; AVX-NEXT: vpcmpeqd %xmm1, %xmm1, %xmm1
; AVX-NEXT: vpcmpeqd %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpsrld $31, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_by_negone:
; XOP: # %bb.0:
; XOP-NEXT: vpcmpeqd %xmm1, %xmm1, %xmm1
; XOP-NEXT: vpcomeqd %xmm1, %xmm0, %xmm0
; XOP-NEXT: vpsrld $31, %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <4 x i32> %x, <i32 -1, i32 -1, i32 -1, i32 -1>
ret <4 x i32> %1
}
; fold (udiv x, INT_MIN) -> (srl x, 31)
define i32 @combine_udiv_by_minsigned(i32 %x) {
; CHECK-LABEL: combine_udiv_by_minsigned:
; CHECK: # %bb.0:
; CHECK-NEXT: movl %edi, %eax
; CHECK-NEXT: shrl $31, %eax
; CHECK-NEXT: retq
%1 = udiv i32 %x, -2147483648
ret i32 %1
}
define <4 x i32> @combine_vec_udiv_by_minsigned(<4 x i32> %x) {
; SSE-LABEL: combine_vec_udiv_by_minsigned:
; SSE: # %bb.0:
; SSE-NEXT: psrld $31, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_by_minsigned:
; AVX: # %bb.0:
; AVX-NEXT: vpsrld $31, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_by_minsigned:
; XOP: # %bb.0:
; XOP-NEXT: vpsrld $31, %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <4 x i32> %x, <i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648>
ret <4 x i32> %1
}
; fold (udiv 0, x) -> 0
define i32 @combine_udiv_zero(i32 %x) {
; CHECK-LABEL: combine_udiv_zero:
; CHECK: # %bb.0:
; CHECK-NEXT: xorl %eax, %eax
; CHECK-NEXT: retq
%1 = udiv i32 0, %x
ret i32 %1
}
define <4 x i32> @combine_vec_udiv_zero(<4 x i32> %x) {
; SSE-LABEL: combine_vec_udiv_zero:
; SSE: # %bb.0:
; SSE-NEXT: xorps %xmm0, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_zero:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_zero:
; XOP: # %bb.0:
; XOP-NEXT: vxorps %xmm0, %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <4 x i32> zeroinitializer, %x
ret <4 x i32> %1
}
; fold (udiv x, x) -> 1
define i32 @combine_udiv_dupe(i32 %x) {
; CHECK-LABEL: combine_udiv_dupe:
; CHECK: # %bb.0:
; CHECK-NEXT: movl $1, %eax
; CHECK-NEXT: retq
%1 = udiv i32 %x, %x
ret i32 %1
}
define <4 x i32> @combine_vec_udiv_dupe(<4 x i32> %x) {
; SSE-LABEL: combine_vec_udiv_dupe:
; SSE: # %bb.0:
; SSE-NEXT: movaps {{.*#+}} xmm0 = [1,1,1,1]
; SSE-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_dupe:
; AVX: # %bb.0:
; AVX-NEXT: vbroadcastss {{.*#+}} xmm0 = [1,1,1,1]
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_dupe:
; XOP: # %bb.0:
; XOP-NEXT: vbroadcastss {{.*#+}} xmm0 = [1,1,1,1]
; XOP-NEXT: retq
%1 = udiv <4 x i32> %x, %x
ret <4 x i32> %1
}
; fold (udiv x, (1 << c)) -> x >>u c
define <4 x i32> @combine_vec_udiv_by_pow2a(<4 x i32> %x) {
; SSE-LABEL: combine_vec_udiv_by_pow2a:
; SSE: # %bb.0:
; SSE-NEXT: psrld $2, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_by_pow2a:
; AVX: # %bb.0:
; AVX-NEXT: vpsrld $2, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_by_pow2a:
; XOP: # %bb.0:
; XOP-NEXT: vpsrld $2, %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <4 x i32> %x, <i32 4, i32 4, i32 4, i32 4>
ret <4 x i32> %1
}
define <4 x i32> @combine_vec_udiv_by_pow2b(<4 x i32> %x) {
; SSE2-LABEL: combine_vec_udiv_by_pow2b:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm1
; SSE2-NEXT: psrld $4, %xmm1
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: psrld $3, %xmm2
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm2 = xmm2[1],xmm1[1]
; SSE2-NEXT: movdqa %xmm0, %xmm1
; SSE2-NEXT: psrld $2, %xmm1
; SSE2-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,3],xmm2[0,3]
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_udiv_by_pow2b:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: psrld $4, %xmm1
; SSE41-NEXT: movdqa %xmm0, %xmm2
; SSE41-NEXT: psrld $2, %xmm2
; SSE41-NEXT: pblendw {{.*#+}} xmm2 = xmm2[0,1,2,3],xmm1[4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: psrld $3, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm2[2,3],xmm0[4,5],xmm2[6,7]
; SSE41-NEXT: retq
;
; AVX1-LABEL: combine_vec_udiv_by_pow2b:
; AVX1: # %bb.0:
; AVX1-NEXT: vpsrld $4, %xmm0, %xmm1
; AVX1-NEXT: vpsrld $2, %xmm0, %xmm2
; AVX1-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0,1,2,3],xmm1[4,5,6,7]
; AVX1-NEXT: vpsrld $3, %xmm0, %xmm2
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm2[4,5,6,7]
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3],xmm0[4,5],xmm1[6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: combine_vec_udiv_by_pow2b:
; AVX2: # %bb.0:
; AVX2-NEXT: vpsrlvd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_by_pow2b:
; XOP: # %bb.0:
; XOP-NEXT: vpshld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <4 x i32> %x, <i32 1, i32 4, i32 8, i32 16>
ret <4 x i32> %1
}
define <4 x i32> @combine_vec_udiv_by_pow2c(<4 x i32> %x, <4 x i32> %y) {
; SSE2-LABEL: combine_vec_udiv_by_pow2c:
; SSE2: # %bb.0:
; SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm3
; SSE2-NEXT: psrld %xmm2, %xmm3
; SSE2-NEXT: pshuflw {{.*#+}} xmm4 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: psrld %xmm4, %xmm2
; SSE2-NEXT: punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm4
; SSE2-NEXT: psrld %xmm3, %xmm4
; SSE2-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: psrld %xmm1, %xmm0
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm0 = xmm0[1],xmm4[1]
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,3],xmm0[0,3]
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_udiv_by_pow2c:
; SSE41: # %bb.0:
; SSE41-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm3
; SSE41-NEXT: psrld %xmm2, %xmm3
; SSE41-NEXT: pshufd {{.*#+}} xmm2 = xmm1[2,3,2,3]
; SSE41-NEXT: pshuflw {{.*#+}} xmm4 = xmm2[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm5
; SSE41-NEXT: psrld %xmm4, %xmm5
; SSE41-NEXT: pblendw {{.*#+}} xmm5 = xmm3[0,1,2,3],xmm5[4,5,6,7]
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm3
; SSE41-NEXT: psrld %xmm1, %xmm3
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm2[0,1,1,1,4,5,6,7]
; SSE41-NEXT: psrld %xmm1, %xmm0
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm3[0,1,2,3],xmm0[4,5,6,7]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm5[2,3],xmm0[4,5],xmm5[6,7]
; SSE41-NEXT: retq
;
; AVX1-LABEL: combine_vec_udiv_by_pow2c:
; AVX1: # %bb.0:
; AVX1-NEXT: vpsrldq {{.*#+}} xmm2 = xmm1[12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX1-NEXT: vpsrld %xmm2, %xmm0, %xmm2
; AVX1-NEXT: vpsrlq $32, %xmm1, %xmm3
; AVX1-NEXT: vpsrld %xmm3, %xmm0, %xmm3
; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm3[0,1,2,3],xmm2[4,5,6,7]
; AVX1-NEXT: vpxor %xmm3, %xmm3, %xmm3
; AVX1-NEXT: vpunpckhdq {{.*#+}} xmm3 = xmm1[2],xmm3[2],xmm1[3],xmm3[3]
; AVX1-NEXT: vpsrld %xmm3, %xmm0, %xmm3
; AVX1-NEXT: vpmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; AVX1-NEXT: vpsrld %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm3[4,5,6,7]
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm2[2,3],xmm0[4,5],xmm2[6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: combine_vec_udiv_by_pow2c:
; AVX2: # %bb.0:
; AVX2-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_by_pow2c:
; XOP: # %bb.0:
; XOP-NEXT: vpxor %xmm2, %xmm2, %xmm2
; XOP-NEXT: vpsubd %xmm1, %xmm2, %xmm1
; XOP-NEXT: vpshld %xmm1, %xmm0, %xmm0
; XOP-NEXT: retq
%1 = shl <4 x i32> <i32 1, i32 1, i32 1, i32 1>, %y
%2 = udiv <4 x i32> %x, %1
ret <4 x i32> %2
}
; fold (udiv x, (shl c, y)) -> x >>u (log2(c)+y) iff c is power of 2
define <4 x i32> @combine_vec_udiv_by_shl_pow2a(<4 x i32> %x, <4 x i32> %y) {
; SSE2-LABEL: combine_vec_udiv_by_shl_pow2a:
; SSE2: # %bb.0:
; SSE2-NEXT: paddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
; SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm3
; SSE2-NEXT: psrld %xmm2, %xmm3
; SSE2-NEXT: pshuflw {{.*#+}} xmm4 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: psrld %xmm4, %xmm2
; SSE2-NEXT: punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm4
; SSE2-NEXT: psrld %xmm3, %xmm4
; SSE2-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: psrld %xmm1, %xmm0
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm0 = xmm0[1],xmm4[1]
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,3],xmm0[0,3]
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_udiv_by_shl_pow2a:
; SSE41: # %bb.0:
; SSE41-NEXT: paddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
; SSE41-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm3
; SSE41-NEXT: psrld %xmm2, %xmm3
; SSE41-NEXT: pshufd {{.*#+}} xmm2 = xmm1[2,3,2,3]
; SSE41-NEXT: pshuflw {{.*#+}} xmm4 = xmm2[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm5
; SSE41-NEXT: psrld %xmm4, %xmm5
; SSE41-NEXT: pblendw {{.*#+}} xmm5 = xmm3[0,1,2,3],xmm5[4,5,6,7]
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm3
; SSE41-NEXT: psrld %xmm1, %xmm3
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm2[0,1,1,1,4,5,6,7]
; SSE41-NEXT: psrld %xmm1, %xmm0
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm3[0,1,2,3],xmm0[4,5,6,7]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm5[2,3],xmm0[4,5],xmm5[6,7]
; SSE41-NEXT: retq
;
; AVX1-LABEL: combine_vec_udiv_by_shl_pow2a:
; AVX1: # %bb.0:
; AVX1-NEXT: vpaddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
; AVX1-NEXT: vpsrldq {{.*#+}} xmm2 = xmm1[12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX1-NEXT: vpsrld %xmm2, %xmm0, %xmm2
; AVX1-NEXT: vpsrlq $32, %xmm1, %xmm3
; AVX1-NEXT: vpsrld %xmm3, %xmm0, %xmm3
; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm3[0,1,2,3],xmm2[4,5,6,7]
; AVX1-NEXT: vpxor %xmm3, %xmm3, %xmm3
; AVX1-NEXT: vpunpckhdq {{.*#+}} xmm3 = xmm1[2],xmm3[2],xmm1[3],xmm3[3]
; AVX1-NEXT: vpsrld %xmm3, %xmm0, %xmm3
; AVX1-NEXT: vpmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; AVX1-NEXT: vpsrld %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm3[4,5,6,7]
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm2[2,3],xmm0[4,5],xmm2[6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: combine_vec_udiv_by_shl_pow2a:
; AVX2: # %bb.0:
; AVX2-NEXT: vpbroadcastd {{.*#+}} xmm2 = [2,2,2,2]
; AVX2-NEXT: vpaddd %xmm2, %xmm1, %xmm1
; AVX2-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_by_shl_pow2a:
; XOP: # %bb.0:
; XOP-NEXT: vbroadcastss {{.*#+}} xmm2 = [4294967294,4294967294,4294967294,4294967294]
; XOP-NEXT: vpsubd %xmm1, %xmm2, %xmm1
; XOP-NEXT: vpshld %xmm1, %xmm0, %xmm0
; XOP-NEXT: retq
%1 = shl <4 x i32> <i32 4, i32 4, i32 4, i32 4>, %y
%2 = udiv <4 x i32> %x, %1
ret <4 x i32> %2
}
define <4 x i32> @combine_vec_udiv_by_shl_pow2b(<4 x i32> %x, <4 x i32> %y) {
; SSE2-LABEL: combine_vec_udiv_by_shl_pow2b:
; SSE2: # %bb.0:
; SSE2-NEXT: paddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
; SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm3
; SSE2-NEXT: psrld %xmm2, %xmm3
; SSE2-NEXT: pshuflw {{.*#+}} xmm4 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: psrld %xmm4, %xmm2
; SSE2-NEXT: punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
; SSE2-NEXT: movdqa %xmm0, %xmm4
; SSE2-NEXT: psrld %xmm3, %xmm4
; SSE2-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE2-NEXT: psrld %xmm1, %xmm0
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm0 = xmm0[1],xmm4[1]
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,3],xmm0[0,3]
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_udiv_by_shl_pow2b:
; SSE41: # %bb.0:
; SSE41-NEXT: paddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
; SSE41-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm3
; SSE41-NEXT: psrld %xmm2, %xmm3
; SSE41-NEXT: pshufd {{.*#+}} xmm2 = xmm1[2,3,2,3]
; SSE41-NEXT: pshuflw {{.*#+}} xmm4 = xmm2[2,3,3,3,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm5
; SSE41-NEXT: psrld %xmm4, %xmm5
; SSE41-NEXT: pblendw {{.*#+}} xmm5 = xmm3[0,1,2,3],xmm5[4,5,6,7]
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm3
; SSE41-NEXT: psrld %xmm1, %xmm3
; SSE41-NEXT: pshuflw {{.*#+}} xmm1 = xmm2[0,1,1,1,4,5,6,7]
; SSE41-NEXT: psrld %xmm1, %xmm0
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm3[0,1,2,3],xmm0[4,5,6,7]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm5[2,3],xmm0[4,5],xmm5[6,7]
; SSE41-NEXT: retq
;
; AVX1-LABEL: combine_vec_udiv_by_shl_pow2b:
; AVX1: # %bb.0:
; AVX1-NEXT: vpaddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm2
; AVX1-NEXT: vpsrldq {{.*#+}} xmm3 = xmm2[12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX1-NEXT: vpsrld %xmm3, %xmm0, %xmm3
; AVX1-NEXT: vpsrlq $32, %xmm2, %xmm4
; AVX1-NEXT: vpsrld %xmm4, %xmm0, %xmm4
; AVX1-NEXT: vpblendw {{.*#+}} xmm3 = xmm4[0,1,2,3],xmm3[4,5,6,7]
; AVX1-NEXT: vpmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; AVX1-NEXT: vpsrld %xmm1, %xmm0, %xmm1
; AVX1-NEXT: vpxor %xmm4, %xmm4, %xmm4
; AVX1-NEXT: vpunpckhdq {{.*#+}} xmm2 = xmm2[2],xmm4[2],xmm2[3],xmm4[3]
; AVX1-NEXT: vpsrld %xmm2, %xmm0, %xmm0
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0,1,2,3],xmm0[4,5,6,7]
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm3[2,3],xmm0[4,5],xmm3[6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: combine_vec_udiv_by_shl_pow2b:
; AVX2: # %bb.0:
; AVX2-NEXT: vpaddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
; AVX2-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_by_shl_pow2b:
; XOP: # %bb.0:
; XOP-NEXT: vpmovsxbd {{.*#+}} xmm2 = [0,4294967294,4294967293,4294967292]
; XOP-NEXT: vpsubd %xmm1, %xmm2, %xmm1
; XOP-NEXT: vpshld %xmm1, %xmm0, %xmm0
; XOP-NEXT: retq
%1 = shl <4 x i32> <i32 1, i32 4, i32 8, i32 16>, %y
%2 = udiv <4 x i32> %x, %1
ret <4 x i32> %2
}
; fold (udiv x, c1)
define i32 @combine_udiv_uniform(i32 %x) {
; CHECK-LABEL: combine_udiv_uniform:
; CHECK: # %bb.0:
; CHECK-NEXT: movl %edi, %ecx
; CHECK-NEXT: movl $2987803337, %eax # imm = 0xB21642C9
; CHECK-NEXT: imulq %rcx, %rax
; CHECK-NEXT: shrq $36, %rax
; CHECK-NEXT: # kill: def $eax killed $eax killed $rax
; CHECK-NEXT: retq
%1 = udiv i32 %x, 23
ret i32 %1
}
define <8 x i16> @combine_vec_udiv_uniform(<8 x i16> %x) {
; SSE-LABEL: combine_vec_udiv_uniform:
; SSE: # %bb.0:
; SSE-NEXT: movdqa {{.*#+}} xmm1 = [25645,25645,25645,25645,25645,25645,25645,25645]
; SSE-NEXT: pmulhuw %xmm0, %xmm1
; SSE-NEXT: psubw %xmm1, %xmm0
; SSE-NEXT: psrlw $1, %xmm0
; SSE-NEXT: paddw %xmm1, %xmm0
; SSE-NEXT: psrlw $4, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_uniform:
; AVX: # %bb.0:
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 # [25645,25645,25645,25645,25645,25645,25645,25645]
; AVX-NEXT: vpsubw %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpsrlw $1, %xmm0, %xmm0
; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpsrlw $4, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_uniform:
; XOP: # %bb.0:
; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 # [25645,25645,25645,25645,25645,25645,25645,25645]
; XOP-NEXT: vpsubw %xmm1, %xmm0, %xmm0
; XOP-NEXT: vpsrlw $1, %xmm0, %xmm0
; XOP-NEXT: vpaddw %xmm1, %xmm0, %xmm0
; XOP-NEXT: vpsrlw $4, %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <8 x i16> %x, <i16 23, i16 23, i16 23, i16 23, i16 23, i16 23, i16 23, i16 23>
ret <8 x i16> %1
}
define <8 x i16> @combine_vec_udiv_nonuniform(<8 x i16> %x) {
; SSE2-LABEL: combine_vec_udiv_nonuniform:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [65535,65535,65535,0,65535,65535,65535,65535]
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: pand %xmm1, %xmm2
; SSE2-NEXT: movdqa %xmm0, %xmm3
; SSE2-NEXT: psrlw $3, %xmm3
; SSE2-NEXT: pandn %xmm3, %xmm1
; SSE2-NEXT: por %xmm2, %xmm1
; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1 # [25645,61681,8195,9363,512,32769,32897,2]
; SSE2-NEXT: psubw %xmm1, %xmm0
; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: paddw %xmm1, %xmm0
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [65535,65535,65535,0,0,65535,65535,0]
; SSE2-NEXT: pandn %xmm0, %xmm1
; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: por %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_udiv_nonuniform:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: psrlw $3, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm0[0,1,2],xmm1[3],xmm0[4,5,6,7]
; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1 # [25645,61681,8195,9363,512,32769,32897,2]
; SSE41-NEXT: psubw %xmm1, %xmm0
; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE41-NEXT: paddw %xmm1, %xmm0
; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [4096,2048,8,u,u,2,2,u]
; SSE41-NEXT: pmulhuw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3,4],xmm1[5,6],xmm0[7]
; SSE41-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_nonuniform:
; AVX: # %bb.0:
; AVX-NEXT: vpsrlw $3, %xmm0, %xmm1
; AVX-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0,1,2],xmm1[3],xmm0[4,5,6,7]
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 # [25645,61681,8195,9363,512,32769,32897,2]
; AVX-NEXT: vpsubw %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 # [4096,2048,8,u,u,2,2,u]
; AVX-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3,4],xmm1[5,6],xmm0[7]
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_nonuniform:
; XOP: # %bb.0:
; XOP-NEXT: vpshlw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1
; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 # [25645,61681,8195,9363,512,32769,32897,2]
; XOP-NEXT: vpsubw %xmm1, %xmm0, %xmm0
; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; XOP-NEXT: vpaddw %xmm1, %xmm0, %xmm0
; XOP-NEXT: vpshlw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <8 x i16> %x, <i16 23, i16 34, i16 -23, i16 56, i16 128, i16 -1, i16 -256, i16 -32768>
ret <8 x i16> %1
}
define <8 x i16> @combine_vec_udiv_nonuniform2(<8 x i16> %x) {
; SSE2-LABEL: combine_vec_udiv_nonuniform2:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,65535,65535,65535,65535,65535,65535]
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: pand %xmm1, %xmm2
; SSE2-NEXT: psrlw $1, %xmm0
; SSE2-NEXT: pandn %xmm0, %xmm1
; SSE2-NEXT: por %xmm2, %xmm1
; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1 # [16393,59919,58255,32787,55189,8197,52429,32789]
; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1 # [8,2048,2048,2,2048,8,2048,2]
; SSE2-NEXT: movdqa %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_udiv_nonuniform2:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: psrlw $1, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3,4,5,6,7]
; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [16393,59919,58255,32787,55189,8197,52429,32789]
; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [8,2048,2048,2,2048,8,2048,2]
; SSE41-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_nonuniform2:
; AVX: # %bb.0:
; AVX-NEXT: vpsrlw $1, %xmm0, %xmm1
; AVX-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3,4,5,6,7]
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 # [16393,59919,58255,32787,55189,8197,52429,32789]
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 # [8,2048,2048,2,2048,8,2048,2]
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_nonuniform2:
; XOP: # %bb.0:
; XOP-NEXT: vpshlw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 # [16393,59919,58255,32787,55189,8197,52429,32789]
; XOP-NEXT: vpshlw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <8 x i16> %x, <i16 -34, i16 35, i16 36, i16 -37, i16 38, i16 -39, i16 40, i16 -41>
ret <8 x i16> %1
}
define <8 x i16> @combine_vec_udiv_nonuniform3(<8 x i16> %x) {
; SSE-LABEL: combine_vec_udiv_nonuniform3:
; SSE: # %bb.0:
; SSE-NEXT: movdqa {{.*#+}} xmm1 = [9363,25645,18351,12137,2115,23705,1041,517]
; SSE-NEXT: pmulhuw %xmm0, %xmm1
; SSE-NEXT: psubw %xmm1, %xmm0
; SSE-NEXT: psrlw $1, %xmm0
; SSE-NEXT: paddw %xmm1, %xmm0
; SSE-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [16384,4096,4096,4096,4096,2048,2048,1024]
; SSE-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_nonuniform3:
; AVX: # %bb.0:
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 # [9363,25645,18351,12137,2115,23705,1041,517]
; AVX-NEXT: vpsubw %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpsrlw $1, %xmm0, %xmm0
; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 # [16384,4096,4096,4096,4096,2048,2048,1024]
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_nonuniform3:
; XOP: # %bb.0:
; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 # [9363,25645,18351,12137,2115,23705,1041,517]
; XOP-NEXT: vpsubw %xmm1, %xmm0, %xmm0
; XOP-NEXT: vpsrlw $1, %xmm0, %xmm0
; XOP-NEXT: vpaddw %xmm1, %xmm0, %xmm0
; XOP-NEXT: vpshlw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; XOP-NEXT: retq
%1 = udiv <8 x i16> %x, <i16 7, i16 23, i16 25, i16 27, i16 31, i16 47, i16 63, i16 127>
ret <8 x i16> %1
}
define <16 x i8> @combine_vec_udiv_nonuniform4(<16 x i8> %x) {
; SSE2-LABEL: combine_vec_udiv_nonuniform4:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: pand %xmm1, %xmm2
; SSE2-NEXT: pxor %xmm3, %xmm3
; SSE2-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1],xmm0[2],xmm3[2],xmm0[3],xmm3[3],xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7]
; SSE2-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: psrlw $15, %xmm0
; SSE2-NEXT: pandn %xmm0, %xmm1
; SSE2-NEXT: por %xmm2, %xmm1
; SSE2-NEXT: movdqa %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: combine_vec_udiv_nonuniform4:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: pmovzxbw {{.*#+}} xmm2 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
; SSE41-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2
; SSE41-NEXT: psrlw $8, %xmm2
; SSE41-NEXT: packuswb %xmm2, %xmm2
; SSE41-NEXT: psrlw $7, %xmm2
; SSE41-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2
; SSE41-NEXT: movaps {{.*#+}} xmm0 = [0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
; SSE41-NEXT: pblendvb %xmm0, %xmm1, %xmm2
; SSE41-NEXT: movdqa %xmm2, %xmm0
; SSE41-NEXT: retq
;
; AVX-LABEL: combine_vec_udiv_nonuniform4:
; AVX: # %bb.0:
; AVX-NEXT: vpmovzxbw {{.*#+}} xmm1 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
; AVX-NEXT: vpmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
; AVX-NEXT: vpsrlw $8, %xmm1, %xmm1
; AVX-NEXT: vpackuswb %xmm1, %xmm1, %xmm1
; AVX-NEXT: vpsrlw $7, %xmm1, %xmm1
; AVX-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
; AVX-NEXT: vpmovsxwq {{.*#+}} xmm2 = [18446744073709551360,18446744073709551615]
; AVX-NEXT: vpblendvb %xmm2, %xmm0, %xmm1, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: combine_vec_udiv_nonuniform4:
; XOP: # %bb.0:
; XOP-NEXT: movl $171, %eax
; XOP-NEXT: vmovd %eax, %xmm1
; XOP-NEXT: vpmovzxbw {{.*#+}} xmm2 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
; XOP-NEXT: vpmullw %xmm1, %xmm2, %xmm1
; XOP-NEXT: vpsrlw $8, %xmm1, %xmm1
; XOP-NEXT: movl $249, %eax
; XOP-NEXT: vmovd %eax, %xmm2
; XOP-NEXT: vpshlb %xmm2, %xmm1, %xmm1
; XOP-NEXT: vpmovsxwq {{.*#+}} xmm2 = [18446744073709551360,18446744073709551615]
; XOP-NEXT: vpblendvb %xmm2, %xmm0, %xmm1, %xmm0
; XOP-NEXT: retq
%div = udiv <16 x i8> %x, <i8 -64, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
ret <16 x i8> %div
}
define <8 x i16> @pr38477(<8 x i16> %a0) {
; SSE2-LABEL: pr38477:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,65535,65535,65535,65535,65535,65535]
; SSE2-NEXT: movdqa %xmm1, %xmm2
; SSE2-NEXT: pandn %xmm0, %xmm2
; SSE2-NEXT: movdqa {{.*#+}} xmm3 = [u,4957,57457,4103,16385,35545,2048,2115]
; SSE2-NEXT: pmulhuw %xmm0, %xmm3
; SSE2-NEXT: psubw %xmm3, %xmm0
; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [u,32768,0,0,0,0,0,32768]
; SSE2-NEXT: paddw %xmm3, %xmm0
; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm3
; SSE2-NEXT: por %xmm3, %xmm0
; SSE2-NEXT: pand %xmm1, %xmm0
; SSE2-NEXT: por %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: pr38477:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,4957,57457,4103,16385,35545,2048,2115]
; SSE41-NEXT: pmulhuw %xmm0, %xmm1
; SSE41-NEXT: movdqa %xmm0, %xmm2
; SSE41-NEXT: psubw %xmm1, %xmm2
; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2 # [u,32768,0,0,0,0,0,32768]
; SSE41-NEXT: paddw %xmm1, %xmm2
; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,1024,1024,16,4,1024,u,4096]
; SSE41-NEXT: pmulhuw %xmm2, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5],xmm2[6],xmm1[7]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; SSE41-NEXT: retq
;
; AVX-LABEL: pr38477:
; AVX: # %bb.0:
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 # [u,4957,57457,4103,16385,35545,2048,2115]
; AVX-NEXT: vpsubw %xmm1, %xmm0, %xmm2
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2, %xmm2 # [u,32768,0,0,0,0,0,32768]
; AVX-NEXT: vpaddw %xmm1, %xmm2, %xmm1
; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm2 # [u,1024,1024,16,4,1024,u,4096]
; AVX-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0,1,2,3,4,5],xmm1[6],xmm2[7]
; AVX-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; AVX-NEXT: retq
;
; XOP-LABEL: pr38477:
; XOP: # %bb.0:
; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 # [u,4957,57457,4103,16385,35545,2048,2115]
; XOP-NEXT: vpsubw %xmm1, %xmm0, %xmm2
; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2, %xmm2 # [u,32768,0,0,0,0,0,32768]
; XOP-NEXT: vpaddw %xmm1, %xmm2, %xmm1
; XOP-NEXT: vpshlw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
; XOP-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; XOP-NEXT: retq
%1 = udiv <8 x i16> %a0, <i16 1, i16 119, i16 73, i16 -111, i16 -3, i16 118, i16 32, i16 31>
ret <8 x i16> %1
}
define i1 @bool_udiv(i1 %x, i1 %y) {
; CHECK-LABEL: bool_udiv:
; CHECK: # %bb.0:
; CHECK-NEXT: movl %edi, %eax
; CHECK-NEXT: # kill: def $al killed $al killed $eax
; CHECK-NEXT: retq
%r = udiv i1 %x, %y
ret i1 %r
}
define <4 x i1> @boolvec_udiv(<4 x i1> %x, <4 x i1> %y) {
; CHECK-LABEL: boolvec_udiv:
; CHECK: # %bb.0:
; CHECK-NEXT: retq
%r = udiv <4 x i1> %x, %y
ret <4 x i1> %r
}
define <4 x i32> @vector_div_leading_zeros(<4 x i32> %x) {
; SSE2-LABEL: vector_div_leading_zeros:
; SSE2: # %bb.0:
; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [613566757,613566757,613566757,613566757]
; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm0[1,1,3,3]
; SSE2-NEXT: pmuludq %xmm1, %xmm0
; SSE2-NEXT: pshufd {{.*#+}} xmm0 = xmm0[1,3,2,3]
; SSE2-NEXT: pmuludq %xmm1, %xmm2
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm2[1,3,2,3]
; SSE2-NEXT: punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
; SSE2-NEXT: retq
;
; SSE41-LABEL: vector_div_leading_zeros:
; SSE41: # %bb.0:
; SSE41-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE41-NEXT: pshufd {{.*#+}} xmm1 = xmm0[1,1,3,3]
; SSE41-NEXT: movdqa {{.*#+}} xmm2 = [613566757,613566757,613566757,613566757]
; SSE41-NEXT: pmuludq %xmm2, %xmm1
; SSE41-NEXT: pmuludq %xmm2, %xmm0
; SSE41-NEXT: pshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3],xmm0[4,5],xmm1[6,7]
; SSE41-NEXT: retq
;
; AVX1-LABEL: vector_div_leading_zeros:
; AVX1: # %bb.0:
; AVX1-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX1-NEXT: vbroadcastss {{.*#+}} xmm2 = [613566757,613566757,613566757,613566757]
; AVX1-NEXT: vpmuludq %xmm2, %xmm1, %xmm1
; AVX1-NEXT: vpmuludq %xmm2, %xmm0, %xmm0
; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3],xmm0[4,5],xmm1[6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: vector_div_leading_zeros:
; AVX2: # %bb.0:
; AVX2-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX2-NEXT: vpbroadcastd {{.*#+}} xmm2 = [613566757,613566757,613566757,613566757]
; AVX2-NEXT: vpmuludq %xmm2, %xmm1, %xmm1
; AVX2-NEXT: vpmuludq %xmm2, %xmm0, %xmm0
; AVX2-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
; AVX2-NEXT: vpblendd {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2],xmm1[3]
; AVX2-NEXT: retq
;
; XOP-LABEL: vector_div_leading_zeros:
; XOP: # %bb.0:
; XOP-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; XOP-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,3,3]
; XOP-NEXT: vbroadcastss {{.*#+}} xmm2 = [613566757,613566757,613566757,613566757]
; XOP-NEXT: vpmuludq %xmm2, %xmm1, %xmm1
; XOP-NEXT: vpmuludq %xmm2, %xmm0, %xmm0
; XOP-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
; XOP-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3],xmm0[4,5],xmm1[6,7]
; XOP-NEXT: retq
%a = and <4 x i32> %x, <i32 255, i32 255, i32 255, i32 255>
%b = udiv <4 x i32> %a, <i32 7, i32 7, i32 7, i32 7>
ret <4 x i32> %b
}
|
