llvm.org GIT mirror llvm / release_50 test / CodeGen / X86 / vector-shuffle-combining-ssse3.ll
release_50

Tree @release_50 (Download .tar.gz)

vector-shuffle-combining-ssse3.ll @release_50raw · history · blame

  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
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+ssse3 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSSE3
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx512f | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX512F
;
; Combine tests involving SSE3/SSSE3 target shuffles (MOVDDUP, MOVSHDUP, MOVSLDUP, PSHUFB)

declare <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8>, <16 x i8>)

define <16 x i8> @combine_vpshufb_as_zero(<16 x i8> %a0) {
; SSE-LABEL: combine_vpshufb_as_zero:
; SSE:       # BB#0:
; SSE-NEXT:    xorps %xmm0, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_vpshufb_as_zero:
; AVX:       # BB#0:
; AVX-NEXT:    vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 128, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>)
  %res1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %res0, <16 x i8> <i8 0, i8 128, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>)
  %res2 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %res1, <16 x i8> <i8 0, i8 1, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128>)
  ret <16 x i8> %res2
}

define <16 x i8> @combine_vpshufb_as_movq(<16 x i8> %a0) {
; SSE-LABEL: combine_vpshufb_as_movq:
; SSE:       # BB#0:
; SSE-NEXT:    movq {{.*#+}} xmm0 = xmm0[0],zero
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_vpshufb_as_movq:
; AVX:       # BB#0:
; AVX-NEXT:    vmovq {{.*#+}} xmm0 = xmm0[0],zero
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 0, i8 128, i8 1, i8 128, i8 2, i8 128, i8 3, i8 128, i8 4, i8 128, i8 5, i8 128, i8 6, i8 128, i8 7, i8 128>)
  %res1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %res0, <16 x i8> <i8 0, i8 2, i8 4, i8 6, i8 8, i8 10, i8 12, i8 14, i8 1, i8 3, i8 5, i8 7, i8 9, i8 11, i8 13, i8 15>)
  ret <16 x i8> %res1
}

define <2 x double> @combine_pshufb_as_movsd(<2 x double> %a0, <2 x double> %a1) {
; SSSE3-LABEL: combine_pshufb_as_movsd:
; SSSE3:       # BB#0:
; SSSE3-NEXT:    movsd {{.*#+}} xmm1 = xmm0[0],xmm1[1]
; SSSE3-NEXT:    movapd %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: combine_pshufb_as_movsd:
; SSE41:       # BB#0:
; SSE41-NEXT:    blendpd {{.*#+}} xmm0 = xmm0[0],xmm1[1]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: combine_pshufb_as_movsd:
; AVX1:       # BB#0:
; AVX1-NEXT:    vblendpd {{.*#+}} xmm0 = xmm0[0],xmm1[1]
; AVX1-NEXT:    retq
;
; AVX2-LABEL: combine_pshufb_as_movsd:
; AVX2:       # BB#0:
; AVX2-NEXT:    vblendpd {{.*#+}} xmm0 = xmm0[0],xmm1[1]
; AVX2-NEXT:    retq
;
; AVX512F-LABEL: combine_pshufb_as_movsd:
; AVX512F:       # BB#0:
; AVX512F-NEXT:    vmovsd {{.*#+}} xmm0 = xmm0[0],xmm1[1]
; AVX512F-NEXT:    retq
  %1 = shufflevector <2 x double> %a0, <2 x double> %a1, <2 x i32> <i32 3, i32 0>
  %2 = bitcast <2 x double> %1 to <16 x i8>
  %3 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %2, <16 x i8> <i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7>)
  %4 = bitcast <16 x i8> %3 to <2 x double>
  ret <2 x double> %4
}

define <4 x float> @combine_pshufb_as_movss(<4 x float> %a0, <4 x float> %a1) {
; SSSE3-LABEL: combine_pshufb_as_movss:
; SSSE3:       # BB#0:
; SSSE3-NEXT:    movss {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: combine_pshufb_as_movss:
; SSE41:       # BB#0:
; SSE41-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: combine_pshufb_as_movss:
; AVX1:       # BB#0:
; AVX1-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
; AVX1-NEXT:    retq
;
; AVX2-LABEL: combine_pshufb_as_movss:
; AVX2:       # BB#0:
; AVX2-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
; AVX2-NEXT:    retq
;
; AVX512F-LABEL: combine_pshufb_as_movss:
; AVX512F:       # BB#0:
; AVX512F-NEXT:    vmovss {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
; AVX512F-NEXT:    retq
  %1 = shufflevector <4 x float> %a0, <4 x float> %a1, <4 x i32> <i32 4, i32 3, i32 2, i32 1>
  %2 = bitcast <4 x float> %1 to <16 x i8>
  %3 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %2, <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 12, i8 13, i8 14, i8 15, i8 8, i8 9, i8 10, i8 11, i8 4, i8 5, i8 6, i8 7>)
  %4 = bitcast <16 x i8> %3 to <4 x float>
  ret <4 x float> %4
}

define <4 x i32> @combine_pshufb_as_zext(<16 x i8> %a0) {
; SSSE3-LABEL: combine_pshufb_as_zext:
; SSSE3:       # BB#0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: combine_pshufb_as_zext:
; SSE41:       # BB#0:
; SSE41-NEXT:    pmovzxbd {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
; SSE41-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_zext:
; AVX:       # BB#0:
; AVX-NEXT:    vpmovzxbd {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
; AVX-NEXT:    retq
  %1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 0, i8 -1, i8 -1, i8 -1, i8 1, i8 -1, i8 -1, i8 -1, i8 2, i8 -1, i8 -1, i8 -1, i8 3, i8 -1, i8 -1, i8 -1>)
  %2 = bitcast <16 x i8> %1 to <4 x i32>
  ret <4 x i32> %2
}

define <2 x double> @combine_pshufb_as_vzmovl_64(<2 x double> %a0) {
; SSE-LABEL: combine_pshufb_as_vzmovl_64:
; SSE:       # BB#0:
; SSE-NEXT:    movq {{.*#+}} xmm0 = xmm0[0],zero
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_vzmovl_64:
; AVX:       # BB#0:
; AVX-NEXT:    vmovq {{.*#+}} xmm0 = xmm0[0],zero
; AVX-NEXT:    retq
  %1 = bitcast <2 x double> %a0 to <16 x i8>
  %2 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>)
  %3 = bitcast <16 x i8> %2 to <2 x double>
  ret <2 x double> %3
}

define <4 x float> @combine_pshufb_as_vzmovl_32(<4 x float> %a0) {
; SSSE3-LABEL: combine_pshufb_as_vzmovl_32:
; SSSE3:       # BB#0:
; SSSE3-NEXT:    xorps %xmm1, %xmm1
; SSSE3-NEXT:    movss {{.*#+}} xmm1 = xmm0[0],xmm1[1,2,3]
; SSSE3-NEXT:    movaps %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: combine_pshufb_as_vzmovl_32:
; SSE41:       # BB#0:
; SSE41-NEXT:    xorps %xmm1, %xmm1
; SSE41-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: combine_pshufb_as_vzmovl_32:
; AVX1:       # BB#0:
; AVX1-NEXT:    vxorps %xmm1, %xmm1, %xmm1
; AVX1-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; AVX1-NEXT:    retq
;
; AVX2-LABEL: combine_pshufb_as_vzmovl_32:
; AVX2:       # BB#0:
; AVX2-NEXT:    vxorps %xmm1, %xmm1, %xmm1
; AVX2-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; AVX2-NEXT:    retq
;
; AVX512F-LABEL: combine_pshufb_as_vzmovl_32:
; AVX512F:       # BB#0:
; AVX512F-NEXT:    vxorps %xmm1, %xmm1, %xmm1
; AVX512F-NEXT:    vmovss {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; AVX512F-NEXT:    retq
  %1 = bitcast <4 x float> %a0 to <16 x i8>
  %2 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>)
  %3 = bitcast <16 x i8> %2 to <4 x float>
  ret <4 x float> %3
}

define <4 x float> @combine_pshufb_movddup(<4 x float> %a0) {
; SSE-LABEL: combine_pshufb_movddup:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[5,5,5,5,7,7,7,7,5,5,5,5,7,7,7,7]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_movddup:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[5,5,5,5,7,7,7,7,5,5,5,5,7,7,7,7]
; AVX-NEXT:    retq
  %1 = bitcast <4 x float> %a0 to <16 x i8>
  %2 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <i8 5, i8 5, i8 5, i8 5, i8 7, i8 7, i8 7, i8 7, i8 1, i8 1, i8 1, i8 1, i8 3, i8 3, i8 3, i8 3>)
  %3 = bitcast <16 x i8> %2 to <4 x float>
  %4 = shufflevector <4 x float> %3, <4 x float> undef, <4 x i32> <i32 0, i32 1, i32 0, i32 1>
  ret <4 x float> %4
}

define <4 x float> @combine_pshufb_movshdup(<4 x float> %a0) {
; SSE-LABEL: combine_pshufb_movshdup:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[7,7,7,7,7,7,7,7,3,3,3,3,3,3,3,3]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_movshdup:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[7,7,7,7,7,7,7,7,3,3,3,3,3,3,3,3]
; AVX-NEXT:    retq
  %1 = bitcast <4 x float> %a0 to <16 x i8>
  %2 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <i8 5, i8 5, i8 5, i8 5, i8 7, i8 7, i8 7, i8 7, i8 1, i8 1, i8 1, i8 1, i8 3, i8 3, i8 3, i8 3>)
  %3 = bitcast <16 x i8> %2 to <4 x float>
  %4 = shufflevector <4 x float> %3, <4 x float> undef, <4 x i32> <i32 1, i32 1, i32 3, i32 3>
  ret <4 x float> %4
}

define <4 x float> @combine_pshufb_movsldup(<4 x float> %a0) {
; SSE-LABEL: combine_pshufb_movsldup:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[5,5,5,5,5,5,5,5,1,1,1,1,1,1,1,1]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_movsldup:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[5,5,5,5,5,5,5,5,1,1,1,1,1,1,1,1]
; AVX-NEXT:    retq
  %1 = bitcast <4 x float> %a0 to <16 x i8>
  %2 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <i8 5, i8 5, i8 5, i8 5, i8 7, i8 7, i8 7, i8 7, i8 1, i8 1, i8 1, i8 1, i8 3, i8 3, i8 3, i8 3>)
  %3 = bitcast <16 x i8> %2 to <4 x float>
  %4 = shufflevector <4 x float> %3, <4 x float> undef, <4 x i32> <i32 0, i32 0, i32 2, i32 2>
  ret <4 x float> %4
}

define <16 x i8> @combine_pshufb_palignr(<16 x i8> %a0, <16 x i8> %a1) {
; SSE-LABEL: combine_pshufb_palignr:
; SSE:       # BB#0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_palignr:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; AVX-NEXT:    retq
  %1 = shufflevector <16 x i8> %a0, <16 x i8> %a1, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23>
  %2 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7>)
  ret <16 x i8> %2
}

define <16 x i8> @combine_pshufb_pslldq(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_pslldq:
; SSE:       # BB#0:
; SSE-NEXT:    xorps %xmm0, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_pslldq:
; AVX:       # BB#0:
; AVX-NEXT:    vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT:    retq
  %1 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7>)
  %2 = shufflevector <16 x i8> %1, <16 x i8> zeroinitializer, <16 x i32> <i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  ret <16 x i8> %2
}

define <16 x i8> @combine_pshufb_psrldq(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_psrldq:
; SSE:       # BB#0:
; SSE-NEXT:    xorps %xmm0, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_psrldq:
; AVX:       # BB#0:
; AVX-NEXT:    vxorps %xmm0, %xmm0, %xmm0
; AVX-NEXT:    retq
  %1 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128>)
  %2 = shufflevector <16 x i8> %1, <16 x i8> zeroinitializer, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16>
  ret <16 x i8> %2
}

define <16 x i8> @combine_and_pshufb(<16 x i8> %a0) {
; SSSE3-LABEL: combine_and_pshufb:
; SSSE3:       # BB#0:
; SSSE3-NEXT:    andps {{.*}}(%rip), %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: combine_and_pshufb:
; SSE41:       # BB#0:
; SSE41-NEXT:    pxor %xmm1, %xmm1
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm1[0,1,2,3],xmm0[4],xmm1[5,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: combine_and_pshufb:
; AVX:       # BB#0:
; AVX-NEXT:    vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm1[0,1,2,3],xmm0[4],xmm1[5,6,7]
; AVX-NEXT:    retq
  %1 = shufflevector <16 x i8> %a0, <16 x i8> zeroinitializer, <16 x i32> <i32 16, i32 16, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  %2 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <i8 0, i8 1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 8, i8 9, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>)
  ret <16 x i8> %2
}

define <16 x i8> @combine_pshufb_and(<16 x i8> %a0) {
; SSSE3-LABEL: combine_pshufb_and:
; SSSE3:       # BB#0:
; SSSE3-NEXT:    andps {{.*}}(%rip), %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: combine_pshufb_and:
; SSE41:       # BB#0:
; SSE41-NEXT:    pxor %xmm1, %xmm1
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm1[0,1,2,3],xmm0[4],xmm1[5,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_and:
; AVX:       # BB#0:
; AVX-NEXT:    vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm1[0,1,2,3],xmm0[4],xmm1[5,6,7]
; AVX-NEXT:    retq
  %1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 0, i8 1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 8, i8 9, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>)
  %2 = shufflevector <16 x i8> %1, <16 x i8> zeroinitializer, <16 x i32> <i32 16, i32 16, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  ret <16 x i8> %2
}

define <16 x i8> @combine_pshufb_as_palignr(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_palignr:
; SSE:       # BB#0:
; SSE-NEXT:    palignr {{.*#+}} xmm0 = xmm0[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,0]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_palignr:
; AVX:       # BB#0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,0]
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 undef, i8 undef, i8 0>)
  ret <16 x i8> %res0
}

define <16 x i8> @combine_pshufb_as_pslldq(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_pslldq:
; SSE:       # BB#0:
; SSE-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_pslldq:
; AVX:       # BB#0:
; AVX-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5>)
  ret <16 x i8> %res0
}

define <16 x i8> @combine_pshufb_as_psrldq(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_psrldq:
; SSE:       # BB#0:
; SSE-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_psrldq:
; AVX:       # BB#0:
; AVX-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 15, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128, i8 128>)
  ret <16 x i8> %res0
}

define <16 x i8> @combine_pshufb_as_psrlw(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_psrlw:
; SSE:       # BB#0:
; SSE-NEXT:    psrlw $8, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_psrlw:
; AVX:       # BB#0:
; AVX-NEXT:    vpsrlw $8, %xmm0, %xmm0
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 1, i8 128, i8 3, i8 128, i8 5, i8 128, i8 7, i8 128, i8 9, i8 128, i8 11, i8 128, i8 13, i8 128, i8 15, i8 128>)
  ret <16 x i8> %res0
}

define <16 x i8> @combine_pshufb_as_pslld(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_pslld:
; SSE:       # BB#0:
; SSE-NEXT:    pslld $24, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_pslld:
; AVX:       # BB#0:
; AVX-NEXT:    vpslld $24, %xmm0, %xmm0
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 128, i8 128, i8 128, i8 0, i8 128, i8 128, i8 128, i8 4, i8 128, i8 128, i8 128, i8 8, i8 128, i8 128, i8 128, i8 12>)
  ret <16 x i8> %res0
}

define <16 x i8> @combine_pshufb_as_psrlq(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_psrlq:
; SSE:       # BB#0:
; SSE-NEXT:    psrlq $40, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_psrlq:
; AVX:       # BB#0:
; AVX-NEXT:    vpsrlq $40, %xmm0, %xmm0
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 5, i8 6, i8 7, i8 128, i8 128, i8 128, i8 128, i8 128, i8 13, i8 14, i8 15, i8 128, i8 128, i8 128, i8 128, i8 128>)
  ret <16 x i8> %res0
}

define <16 x i8> @combine_pshufb_as_pshuflw(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_pshuflw:
; SSE:       # BB#0:
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_pshuflw:
; AVX:       # BB#0:
; AVX-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 2, i8 3, i8 0, i8 1, i8 6, i8 7, i8 4, i8 5, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>)
  ret <16 x i8> %res0
}

define <16 x i8> @combine_pshufb_as_pshufhw(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_pshufhw:
; SSE:       # BB#0:
; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_pshufhw:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 10, i8 11, i8 8, i8 9, i8 14, i8 15, i8 12, i8 13>)
  ret <16 x i8> %res0
}

define <16 x i8> @combine_pshufb_not_as_pshufw(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_not_as_pshufw:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,3,0,1,6,7,4,5,10,11,8,9,14,15,12,13]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_not_as_pshufw:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,0,1,6,7,4,5,10,11,8,9,14,15,12,13]
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 2, i8 3, i8 0, i8 1, i8 6, i8 7, i8 4, i8 5, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>)
  %res1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %res0, <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 10, i8 11, i8 8, i8 9, i8 14, i8 15, i8 12, i8 13>)
  ret <16 x i8> %res1
}

define <16 x i8> @combine_vpshufb_as_pshuflw_not_pslld(<16 x i8> *%a0) {
; SSE-LABEL: combine_vpshufb_as_pshuflw_not_pslld:
; SSE:       # BB#0:
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = mem[0,0,2,2,4,5,6,7]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_vpshufb_as_pshuflw_not_pslld:
; AVX:       # BB#0:
; AVX-NEXT:    vpshuflw {{.*#+}} xmm0 = mem[0,0,2,2,4,5,6,7]
; AVX-NEXT:    retq
  %res0 = load <16 x i8>, <16 x i8> *%a0, align 16
  %res1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %res0, <16 x i8> <i8 undef, i8 undef, i8 0, i8 1, i8 undef, i8 undef, i8 4, i8 5, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef>)
  ret <16 x i8> %res1
}

define <16 x i8> @combine_pshufb_as_unary_unpcklbw(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_unary_unpcklbw:
; SSE:       # BB#0:
; SSE-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_unary_unpcklbw:
; AVX:       # BB#0:
; AVX-NEXT:    vpunpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; AVX-NEXT:    retq
  %1 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 0, i8 undef, i8 undef, i8 1, i8 2, i8 2, i8 3, i8 3, i8 4, i8 4, i8 5, i8 5, i8 6, i8 6, i8 7, i8 7>)
  ret <16 x i8> %1
}

define <16 x i8> @combine_pshufb_as_unary_unpckhwd(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_unary_unpckhwd:
; SSE:       # BB#0:
; SSE-NEXT:    punpckhwd {{.*#+}} xmm0 = xmm0[4,4,5,5,6,6,7,7]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_unary_unpckhwd:
; AVX:       # BB#0:
; AVX-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm0[4,4,5,5,6,6,7,7]
; AVX-NEXT:    retq
  %1 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 8, i8 9, i8 8, i8 9, i8 10, i8 11, i8 10, i8 11, i8 12, i8 13, i8 12, i8 13, i8 14, i8 15, i8 undef, i8 undef>)
  ret <16 x i8> %1
}

define <8 x i16> @combine_pshufb_as_unpacklo_undef(<16 x i8> %a0) {
; ALL-LABEL: combine_pshufb_as_unpacklo_undef:
; ALL:       # BB#0:
; ALL-NEXT:    retq
  %1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 undef, i8 undef, i8 0, i8 1, i8 undef, i8 undef, i8 2, i8 3, i8 undef, i8 undef, i8 4, i8 5, i8 undef, i8 undef, i8 6, i8 7>)
  %2 = bitcast <16 x i8> %1 to <8 x i16>
  %3 = shufflevector <8 x i16> %2, <8 x i16> undef, <8 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6>
  ret <8 x i16> %3
}

define <16 x i8> @combine_pshufb_as_unpackhi_undef(<16 x i8> %a0) {
; ALL-LABEL: combine_pshufb_as_unpackhi_undef:
; ALL:       # BB#0:
; ALL-NEXT:    retq
  %1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 8, i8 undef, i8 9, i8 undef, i8 10, i8 undef, i8 11, i8 undef, i8 12, i8 undef, i8 13, i8 undef, i8 14, i8 undef, i8 15, i8 undef>)
  %2 = shufflevector <16 x i8> %1, <16 x i8> undef, <16 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7, i32 9, i32 9, i32 11, i32 11, i32 13, i32 13, i32 15, i32 15>
  ret <16 x i8> %2
}

define <16 x i8> @combine_pshufb_as_unpacklo_zero(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_unpacklo_zero:
; SSE:       # BB#0:
; SSE-NEXT:    pxor %xmm1, %xmm1
; SSE-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1]
; SSE-NEXT:    movdqa %xmm1, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_unpacklo_zero:
; AVX:       # BB#0:
; AVX-NEXT:    vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT:    vpunpckldq {{.*#+}} xmm0 = xmm1[0],xmm0[0],xmm1[1],xmm0[1]
; AVX-NEXT:    retq
  %1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 -1, i8 -1, i8 -1, i8 -1, i8 0, i8 1, i8 2, i8 3, i8 -1, i8 -1, i8 -1, i8 -1, i8 4, i8 5, i8 6, i8 7>)
  ret <16 x i8> %1
}

define <16 x i8> @combine_pshufb_as_unpackhi_zero(<16 x i8> %a0) {
; SSE-LABEL: combine_pshufb_as_unpackhi_zero:
; SSE:       # BB#0:
; SSE-NEXT:    pxor %xmm1, %xmm1
; SSE-NEXT:    punpckhbw {{.*#+}} xmm0 = xmm0[8],xmm1[8],xmm0[9],xmm1[9],xmm0[10],xmm1[10],xmm0[11],xmm1[11],xmm0[12],xmm1[12],xmm0[13],xmm1[13],xmm0[14],xmm1[14],xmm0[15],xmm1[15]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pshufb_as_unpackhi_zero:
; AVX:       # BB#0:
; AVX-NEXT:    vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT:    vpunpckhbw {{.*#+}} xmm0 = xmm0[8],xmm1[8],xmm0[9],xmm1[9],xmm0[10],xmm1[10],xmm0[11],xmm1[11],xmm0[12],xmm1[12],xmm0[13],xmm1[13],xmm0[14],xmm1[14],xmm0[15],xmm1[15]
; AVX-NEXT:    retq
  %1 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 8, i8 -1, i8 9, i8 -1, i8 10, i8 -1, i8 11, i8 -1, i8 12, i8 -1, i8 13, i8 -1, i8 14, i8 -1, i8 15, i8 -1>)
  ret <16 x i8> %1
}

define <16 x i8> @combine_psrlw_pshufb(<8 x i16> %a0) {
; SSE-LABEL: combine_psrlw_pshufb:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[1],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[1],zero,zero,zero
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_psrlw_pshufb:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[1],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[1],zero,zero,zero
; AVX-NEXT:    retq
  %1 = lshr <8 x i16> %a0, <i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8>
  %2 = bitcast <8 x i16> %1 to <16 x i8>
  %3 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %2, <16 x i8> <i8 0, i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1>)
  ret <16 x i8> %3
}

define <16 x i8> @combine_pslld_pshufb(<4 x i32> %a0) {
; SSE-LABEL: combine_pslld_pshufb:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,1,0],zero,xmm0[6,5,4],zero,xmm0[10,9,8],zero,xmm0[14,13,12],zero
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_pslld_pshufb:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,1,0],zero,xmm0[6,5,4],zero,xmm0[10,9,8],zero,xmm0[14,13,12],zero
; AVX-NEXT:    retq
  %1 = shl <4 x i32> %a0, <i32 8, i32 8, i32 8, i32 8>
  %2 = bitcast <4 x i32> %1 to <16 x i8>
  %3 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %2, <16 x i8> <i8 3, i8 2, i8 1, i8 0, i8 7, i8 6, i8 5, i8 4, i8 11, i8 10, i8 9, i8 8, i8 15, i8 14, i8 13, i8 12>)
  ret <16 x i8> %3
}

define <16 x i8> @combine_psrlq_pshufb(<2 x i64> %a0) {
; SSE-LABEL: combine_psrlq_pshufb:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[7,6],zero,zero,zero,zero,zero,zero,xmm0[15,14]
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_psrlq_pshufb:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[7,6],zero,zero,zero,zero,zero,zero,xmm0[15,14]
; AVX-NEXT:    retq
  %1 = lshr <2 x i64> %a0, <i64 48, i64 48>
  %2 = bitcast <2 x i64> %1 to <16 x i8>
  %3 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %2, <16 x i8> <i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 15, i8 14, i8 13, i8 12, i8 11, i8 10, i8 9, i8 8>)
  ret <16 x i8> %3
}

define <16 x i8> @combine_unpckl_arg0_pshufb(<16 x i8> %a0, <16 x i8> %a1) {
; SSE-LABEL: combine_unpckl_arg0_pshufb:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[0],zero,zero,zero,xmm0[0],zero,zero,zero,xmm0[0],zero,zero,zero
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_unpckl_arg0_pshufb:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[0],zero,zero,zero,xmm0[0],zero,zero,zero,xmm0[0],zero,zero,zero
; AVX-NEXT:    retq
  %1 = shufflevector <16 x i8> %a0, <16 x i8> %a1, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23>
  %2 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <i8 0, i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1>)
  ret <16 x i8> %2
}

define <16 x i8> @combine_unpckl_arg1_pshufb(<16 x i8> %a0, <16 x i8> %a1) {
; SSE-LABEL: combine_unpckl_arg1_pshufb:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[0],zero,zero,zero,xmm1[0],zero,zero,zero,xmm1[0],zero,zero,zero
; SSE-NEXT:    movdqa %xmm1, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_unpckl_arg1_pshufb:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm1[0],zero,zero,zero,xmm1[0],zero,zero,zero,xmm1[0],zero,zero,zero,xmm1[0],zero,zero,zero
; AVX-NEXT:    retq
  %1 = shufflevector <16 x i8> %a0, <16 x i8> %a1, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23>
  %2 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %1, <16 x i8> <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, i8 -1>)
  ret <16 x i8> %2
}

define <8 x i16> @shuffle_combine_unpack_insert(<8 x i16> %a0) {
; SSE-LABEL: shuffle_combine_unpack_insert:
; SSE:       # BB#0:
; SSE-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[4,5,4,5,4,5,8,9,8,9,8,9,10,11,10,11]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_combine_unpack_insert:
; AVX:       # BB#0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,4,5,4,5,8,9,8,9,8,9,10,11,10,11]
; AVX-NEXT:    retq
  %1 = extractelement <8 x i16> %a0, i32 2
  %2 = extractelement <8 x i16> %a0, i32 4
  %3 = insertelement <8 x i16> %a0, i16 %1, i32 4
  %4 = insertelement <8 x i16> %a0, i16 %2, i32 2
  %5 = shufflevector <8 x i16> %3, <8 x i16> %4, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
  %6 = shufflevector <8 x i16> %5, <8 x i16> %3, <8 x i32> <i32 4, i32 12, i32 5, i32 13, i32 undef, i32 undef, i32 undef, i32 undef>
  %7 = shufflevector <8 x i16> %5, <8 x i16> %a0, <8 x i32> <i32 4, i32 12, i32 5, i32 13, i32 undef, i32 undef, i32 undef, i32 undef>
  %8 = shufflevector <8 x i16> %6, <8 x i16> %7, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
  ret <8 x i16> %8
}

define <16 x i8> @constant_fold_pshufb() {
; SSE-LABEL: constant_fold_pshufb:
; SSE:       # BB#0:
; SSE-NEXT:    movaps {{.*#+}} xmm0 = <14,0,0,0,u,u,0,0,0,0,0,0,0,0,8,9>
; SSE-NEXT:    retq
;
; AVX-LABEL: constant_fold_pshufb:
; AVX:       # BB#0:
; AVX-NEXT:    vmovaps {{.*#+}} xmm0 = <14,0,0,0,u,u,0,0,0,0,0,0,0,0,8,9>
; AVX-NEXT:    retq
  %1 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> <i8 15, i8 14, i8 13, i8 12, i8 11, i8 10, i8 9, i8 8, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>, <16 x i8> <i8 1, i8 -1, i8 -1, i8 -1, i8 undef, i8 undef, i8 -1, i8 -1, i8 15, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 7, i8 6>)
  ret <16 x i8> %1
}

; FIXME - unnecessary pshufb/broadcast being used - pshufb mask only needs lowest byte.
define <16 x i8> @constant_fold_pshufb_2() {
; SSE-LABEL: constant_fold_pshufb_2:
; SSE:       # BB#0:
; SSE-NEXT:    movl $2, %eax
; SSE-NEXT:    movd %eax, %xmm0
; SSE-NEXT:    pxor %xmm1, %xmm1
; SSE-NEXT:    pshufb %xmm1, %xmm0
; SSE-NEXT:    retq
;
; AVX1-LABEL: constant_fold_pshufb_2:
; AVX1:       # BB#0:
; AVX1-NEXT:    movl $2, %eax
; AVX1-NEXT:    vmovd %eax, %xmm0
; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
; AVX1-NEXT:    retq
;
; AVX2-LABEL: constant_fold_pshufb_2:
; AVX2:       # BB#0:
; AVX2-NEXT:    movl $2, %eax
; AVX2-NEXT:    vmovd %eax, %xmm0
; AVX2-NEXT:    vpbroadcastb %xmm0, %xmm0
; AVX2-NEXT:    retq
;
; AVX512F-LABEL: constant_fold_pshufb_2:
; AVX512F:       # BB#0:
; AVX512F-NEXT:    movl $2, %eax
; AVX512F-NEXT:    vmovd %eax, %xmm0
; AVX512F-NEXT:    vpbroadcastb %xmm0, %xmm0
; AVX512F-NEXT:    retq
  %1 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> <i8 2, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>, <16 x i8> <i8 0, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef>)
  ret <16 x i8> %1
}