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
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
| ; 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-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+xop,+avx | FileCheck %s --check-prefix=ALL --check-prefix=XOP --check-prefix=XOPAVX1
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+xop,+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=XOP --check-prefix=XOPAVX2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512dq | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512DQ
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512BW
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512dq,+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=AVX512VL --check-prefix=AVX512DQVL
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=AVX512VL --check-prefix=AVX512BWVL
;
; Just one 32-bit run to make sure we do reasonable things for i64 shifts.
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=ALL --check-prefix=X32-SSE --check-prefix=X32-SSE2
;
; Variable Shifts
;
define <2 x i64> @var_shift_v2i64(<2 x i64> %a, <2 x i64> %b) nounwind {
; SSE2-LABEL: var_shift_v2i64:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: psrlq %xmm1, %xmm2
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,0,1]
; SSE2-NEXT: psrlq %xmm1, %xmm0
; SSE2-NEXT: movsd {{.*#+}} xmm0 = xmm2[0],xmm0[1]
; SSE2-NEXT: retq
;
; SSE41-LABEL: var_shift_v2i64:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm2
; SSE41-NEXT: psrlq %xmm1, %xmm2
; SSE41-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,0,1]
; SSE41-NEXT: psrlq %xmm1, %xmm0
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm2[0,1,2,3],xmm0[4,5,6,7]
; SSE41-NEXT: retq
;
; AVX1-LABEL: var_shift_v2i64:
; AVX1: # %bb.0:
; AVX1-NEXT: vpsrlq %xmm1, %xmm0, %xmm2
; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[2,3,0,1]
; AVX1-NEXT: vpsrlq %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm2[0,1,2,3],xmm0[4,5,6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: var_shift_v2i64:
; AVX2: # %bb.0:
; AVX2-NEXT: vpsrlvq %xmm1, %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOPAVX1-LABEL: var_shift_v2i64:
; XOPAVX1: # %bb.0:
; XOPAVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
; XOPAVX1-NEXT: vpsubq %xmm1, %xmm2, %xmm1
; XOPAVX1-NEXT: vpshlq %xmm1, %xmm0, %xmm0
; XOPAVX1-NEXT: retq
;
; XOPAVX2-LABEL: var_shift_v2i64:
; XOPAVX2: # %bb.0:
; XOPAVX2-NEXT: vpsrlvq %xmm1, %xmm0, %xmm0
; XOPAVX2-NEXT: retq
;
; AVX512-LABEL: var_shift_v2i64:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrlvq %xmm1, %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: var_shift_v2i64:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrlvq %xmm1, %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: var_shift_v2i64:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: movdqa %xmm0, %xmm2
; X32-SSE-NEXT: psrlq %xmm1, %xmm2
; X32-SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,0,1]
; X32-SSE-NEXT: psrlq %xmm1, %xmm0
; X32-SSE-NEXT: movsd {{.*#+}} xmm0 = xmm2[0],xmm0[1]
; X32-SSE-NEXT: retl
%shift = lshr <2 x i64> %a, %b
ret <2 x i64> %shift
}
define <4 x i32> @var_shift_v4i32(<4 x i32> %a, <4 x i32> %b) nounwind {
; SSE2-LABEL: var_shift_v4i32:
; 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[2,3,0,1]
; 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: var_shift_v4i32:
; 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[2,3,0,1]
; 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: var_shift_v4i32:
; 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: var_shift_v4i32:
; AVX2: # %bb.0:
; AVX2-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOPAVX1-LABEL: var_shift_v4i32:
; XOPAVX1: # %bb.0:
; XOPAVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
; XOPAVX1-NEXT: vpsubd %xmm1, %xmm2, %xmm1
; XOPAVX1-NEXT: vpshld %xmm1, %xmm0, %xmm0
; XOPAVX1-NEXT: retq
;
; XOPAVX2-LABEL: var_shift_v4i32:
; XOPAVX2: # %bb.0:
; XOPAVX2-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; XOPAVX2-NEXT: retq
;
; AVX512-LABEL: var_shift_v4i32:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: var_shift_v4i32:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrlvd %xmm1, %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: var_shift_v4i32:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[2,3,3,3,4,5,6,7]
; X32-SSE-NEXT: movdqa %xmm0, %xmm3
; X32-SSE-NEXT: psrld %xmm2, %xmm3
; X32-SSE-NEXT: pshuflw {{.*#+}} xmm4 = xmm1[0,1,1,1,4,5,6,7]
; X32-SSE-NEXT: movdqa %xmm0, %xmm2
; X32-SSE-NEXT: psrld %xmm4, %xmm2
; X32-SSE-NEXT: punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
; X32-SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,0,1]
; X32-SSE-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
; X32-SSE-NEXT: movdqa %xmm0, %xmm4
; X32-SSE-NEXT: psrld %xmm3, %xmm4
; X32-SSE-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,1,1,1,4,5,6,7]
; X32-SSE-NEXT: psrld %xmm1, %xmm0
; X32-SSE-NEXT: punpckhqdq {{.*#+}} xmm0 = xmm0[1],xmm4[1]
; X32-SSE-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,3],xmm0[0,3]
; X32-SSE-NEXT: movaps %xmm2, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <4 x i32> %a, %b
ret <4 x i32> %shift
}
define <8 x i16> @var_shift_v8i16(<8 x i16> %a, <8 x i16> %b) nounwind {
; SSE2-LABEL: var_shift_v8i16:
; SSE2: # %bb.0:
; SSE2-NEXT: psllw $12, %xmm1
; SSE2-NEXT: movdqa %xmm1, %xmm2
; SSE2-NEXT: psraw $15, %xmm2
; SSE2-NEXT: movdqa %xmm2, %xmm3
; SSE2-NEXT: pandn %xmm0, %xmm3
; SSE2-NEXT: psrlw $8, %xmm0
; SSE2-NEXT: pand %xmm2, %xmm0
; SSE2-NEXT: por %xmm3, %xmm0
; SSE2-NEXT: paddw %xmm1, %xmm1
; SSE2-NEXT: movdqa %xmm1, %xmm2
; SSE2-NEXT: psraw $15, %xmm2
; SSE2-NEXT: movdqa %xmm2, %xmm3
; SSE2-NEXT: pandn %xmm0, %xmm3
; SSE2-NEXT: psrlw $4, %xmm0
; SSE2-NEXT: pand %xmm2, %xmm0
; SSE2-NEXT: por %xmm3, %xmm0
; SSE2-NEXT: paddw %xmm1, %xmm1
; SSE2-NEXT: movdqa %xmm1, %xmm2
; SSE2-NEXT: psraw $15, %xmm2
; SSE2-NEXT: movdqa %xmm2, %xmm3
; SSE2-NEXT: pandn %xmm0, %xmm3
; SSE2-NEXT: psrlw $2, %xmm0
; SSE2-NEXT: pand %xmm2, %xmm0
; SSE2-NEXT: por %xmm3, %xmm0
; SSE2-NEXT: paddw %xmm1, %xmm1
; SSE2-NEXT: psraw $15, %xmm1
; SSE2-NEXT: movdqa %xmm1, %xmm2
; SSE2-NEXT: pandn %xmm0, %xmm2
; SSE2-NEXT: psrlw $1, %xmm0
; SSE2-NEXT: pand %xmm1, %xmm0
; SSE2-NEXT: por %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: var_shift_v8i16:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm1, %xmm2
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: movdqa %xmm2, %xmm0
; SSE41-NEXT: psllw $12, %xmm0
; SSE41-NEXT: psllw $4, %xmm2
; SSE41-NEXT: por %xmm0, %xmm2
; SSE41-NEXT: movdqa %xmm2, %xmm3
; SSE41-NEXT: paddw %xmm2, %xmm3
; SSE41-NEXT: movdqa %xmm1, %xmm4
; SSE41-NEXT: psrlw $8, %xmm4
; SSE41-NEXT: movdqa %xmm2, %xmm0
; SSE41-NEXT: pblendvb %xmm0, %xmm4, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm2
; SSE41-NEXT: psrlw $4, %xmm2
; SSE41-NEXT: movdqa %xmm3, %xmm0
; SSE41-NEXT: pblendvb %xmm0, %xmm2, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm2
; SSE41-NEXT: psrlw $2, %xmm2
; SSE41-NEXT: paddw %xmm3, %xmm3
; SSE41-NEXT: movdqa %xmm3, %xmm0
; SSE41-NEXT: pblendvb %xmm0, %xmm2, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm2
; SSE41-NEXT: psrlw $1, %xmm2
; SSE41-NEXT: paddw %xmm3, %xmm3
; SSE41-NEXT: movdqa %xmm3, %xmm0
; SSE41-NEXT: pblendvb %xmm0, %xmm2, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm0
; SSE41-NEXT: retq
;
; AVX1-LABEL: var_shift_v8i16:
; AVX1: # %bb.0:
; AVX1-NEXT: vpsllw $12, %xmm1, %xmm2
; AVX1-NEXT: vpsllw $4, %xmm1, %xmm1
; AVX1-NEXT: vpor %xmm2, %xmm1, %xmm1
; AVX1-NEXT: vpaddw %xmm1, %xmm1, %xmm2
; AVX1-NEXT: vpsrlw $8, %xmm0, %xmm3
; AVX1-NEXT: vpblendvb %xmm1, %xmm3, %xmm0, %xmm0
; AVX1-NEXT: vpsrlw $4, %xmm0, %xmm1
; AVX1-NEXT: vpblendvb %xmm2, %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vpsrlw $2, %xmm0, %xmm1
; AVX1-NEXT: vpaddw %xmm2, %xmm2, %xmm2
; AVX1-NEXT: vpblendvb %xmm2, %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vpsrlw $1, %xmm0, %xmm1
; AVX1-NEXT: vpaddw %xmm2, %xmm2, %xmm2
; AVX1-NEXT: vpblendvb %xmm2, %xmm1, %xmm0, %xmm0
; AVX1-NEXT: retq
;
; AVX2-LABEL: var_shift_v8i16:
; AVX2: # %bb.0:
; AVX2-NEXT: vpmovzxwd {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero
; AVX2-NEXT: vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
; AVX2-NEXT: vpsrlvd %ymm1, %ymm0, %ymm0
; AVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
; AVX2-NEXT: # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: retq
;
; XOP-LABEL: var_shift_v8i16:
; XOP: # %bb.0:
; XOP-NEXT: vpxor %xmm2, %xmm2, %xmm2
; XOP-NEXT: vpsubw %xmm1, %xmm2, %xmm1
; XOP-NEXT: vpshlw %xmm1, %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512DQ-LABEL: var_shift_v8i16:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vpmovzxwd {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero
; AVX512DQ-NEXT: vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
; AVX512DQ-NEXT: vpsrlvd %ymm1, %ymm0, %ymm0
; AVX512DQ-NEXT: vpmovdw %zmm0, %ymm0
; AVX512DQ-NEXT: # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX512DQ-NEXT: vzeroupper
; AVX512DQ-NEXT: retq
;
; AVX512BW-LABEL: var_shift_v8i16:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: # kill: def $xmm1 killed $xmm1 def $zmm1
; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
; AVX512BW-NEXT: vpsrlvw %zmm1, %zmm0, %zmm0
; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
; AVX512BW-NEXT: vzeroupper
; AVX512BW-NEXT: retq
;
; AVX512DQVL-LABEL: var_shift_v8i16:
; AVX512DQVL: # %bb.0:
; AVX512DQVL-NEXT: vpmovzxwd {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero
; AVX512DQVL-NEXT: vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
; AVX512DQVL-NEXT: vpsrlvd %ymm1, %ymm0, %ymm0
; AVX512DQVL-NEXT: vpmovdw %ymm0, %xmm0
; AVX512DQVL-NEXT: vzeroupper
; AVX512DQVL-NEXT: retq
;
; AVX512BWVL-LABEL: var_shift_v8i16:
; AVX512BWVL: # %bb.0:
; AVX512BWVL-NEXT: vpsrlvw %xmm1, %xmm0, %xmm0
; AVX512BWVL-NEXT: retq
;
; X32-SSE-LABEL: var_shift_v8i16:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: psllw $12, %xmm1
; X32-SSE-NEXT: movdqa %xmm1, %xmm2
; X32-SSE-NEXT: psraw $15, %xmm2
; X32-SSE-NEXT: movdqa %xmm2, %xmm3
; X32-SSE-NEXT: pandn %xmm0, %xmm3
; X32-SSE-NEXT: psrlw $8, %xmm0
; X32-SSE-NEXT: pand %xmm2, %xmm0
; X32-SSE-NEXT: por %xmm3, %xmm0
; X32-SSE-NEXT: paddw %xmm1, %xmm1
; X32-SSE-NEXT: movdqa %xmm1, %xmm2
; X32-SSE-NEXT: psraw $15, %xmm2
; X32-SSE-NEXT: movdqa %xmm2, %xmm3
; X32-SSE-NEXT: pandn %xmm0, %xmm3
; X32-SSE-NEXT: psrlw $4, %xmm0
; X32-SSE-NEXT: pand %xmm2, %xmm0
; X32-SSE-NEXT: por %xmm3, %xmm0
; X32-SSE-NEXT: paddw %xmm1, %xmm1
; X32-SSE-NEXT: movdqa %xmm1, %xmm2
; X32-SSE-NEXT: psraw $15, %xmm2
; X32-SSE-NEXT: movdqa %xmm2, %xmm3
; X32-SSE-NEXT: pandn %xmm0, %xmm3
; X32-SSE-NEXT: psrlw $2, %xmm0
; X32-SSE-NEXT: pand %xmm2, %xmm0
; X32-SSE-NEXT: por %xmm3, %xmm0
; X32-SSE-NEXT: paddw %xmm1, %xmm1
; X32-SSE-NEXT: psraw $15, %xmm1
; X32-SSE-NEXT: movdqa %xmm1, %xmm2
; X32-SSE-NEXT: pandn %xmm0, %xmm2
; X32-SSE-NEXT: psrlw $1, %xmm0
; X32-SSE-NEXT: pand %xmm1, %xmm0
; X32-SSE-NEXT: por %xmm2, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <8 x i16> %a, %b
ret <8 x i16> %shift
}
define <16 x i8> @var_shift_v16i8(<16 x i8> %a, <16 x i8> %b) nounwind {
; SSE2-LABEL: var_shift_v16i8:
; SSE2: # %bb.0:
; SSE2-NEXT: psllw $5, %xmm1
; SSE2-NEXT: pxor %xmm2, %xmm2
; SSE2-NEXT: pxor %xmm3, %xmm3
; SSE2-NEXT: pcmpgtb %xmm1, %xmm3
; SSE2-NEXT: movdqa %xmm3, %xmm4
; SSE2-NEXT: pandn %xmm0, %xmm4
; SSE2-NEXT: psrlw $4, %xmm0
; SSE2-NEXT: pand %xmm3, %xmm0
; SSE2-NEXT: pand {{.*}}(%rip), %xmm0
; SSE2-NEXT: por %xmm4, %xmm0
; SSE2-NEXT: paddb %xmm1, %xmm1
; SSE2-NEXT: pxor %xmm3, %xmm3
; SSE2-NEXT: pcmpgtb %xmm1, %xmm3
; SSE2-NEXT: movdqa %xmm3, %xmm4
; SSE2-NEXT: pandn %xmm0, %xmm4
; SSE2-NEXT: psrlw $2, %xmm0
; SSE2-NEXT: pand %xmm3, %xmm0
; SSE2-NEXT: pand {{.*}}(%rip), %xmm0
; SSE2-NEXT: por %xmm4, %xmm0
; SSE2-NEXT: paddb %xmm1, %xmm1
; SSE2-NEXT: pcmpgtb %xmm1, %xmm2
; SSE2-NEXT: movdqa %xmm2, %xmm1
; SSE2-NEXT: pandn %xmm0, %xmm1
; SSE2-NEXT: psrlw $1, %xmm0
; SSE2-NEXT: pand %xmm2, %xmm0
; SSE2-NEXT: pand {{.*}}(%rip), %xmm0
; SSE2-NEXT: por %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: var_shift_v16i8:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm2
; SSE41-NEXT: psllw $5, %xmm1
; SSE41-NEXT: movdqa %xmm0, %xmm3
; SSE41-NEXT: psrlw $4, %xmm3
; SSE41-NEXT: pand {{.*}}(%rip), %xmm3
; SSE41-NEXT: movdqa %xmm1, %xmm0
; SSE41-NEXT: pblendvb %xmm0, %xmm3, %xmm2
; SSE41-NEXT: movdqa %xmm2, %xmm3
; SSE41-NEXT: psrlw $2, %xmm3
; SSE41-NEXT: pand {{.*}}(%rip), %xmm3
; SSE41-NEXT: paddb %xmm1, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm0
; SSE41-NEXT: pblendvb %xmm0, %xmm3, %xmm2
; SSE41-NEXT: movdqa %xmm2, %xmm3
; SSE41-NEXT: psrlw $1, %xmm3
; SSE41-NEXT: pand {{.*}}(%rip), %xmm3
; SSE41-NEXT: paddb %xmm1, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm0
; SSE41-NEXT: pblendvb %xmm0, %xmm3, %xmm2
; SSE41-NEXT: movdqa %xmm2, %xmm0
; SSE41-NEXT: retq
;
; AVX-LABEL: var_shift_v16i8:
; AVX: # %bb.0:
; AVX-NEXT: vpsllw $5, %xmm1, %xmm1
; AVX-NEXT: vpsrlw $4, %xmm0, %xmm2
; AVX-NEXT: vpand {{.*}}(%rip), %xmm2, %xmm2
; AVX-NEXT: vpblendvb %xmm1, %xmm2, %xmm0, %xmm0
; AVX-NEXT: vpsrlw $2, %xmm0, %xmm2
; AVX-NEXT: vpand {{.*}}(%rip), %xmm2, %xmm2
; AVX-NEXT: vpaddb %xmm1, %xmm1, %xmm1
; AVX-NEXT: vpblendvb %xmm1, %xmm2, %xmm0, %xmm0
; AVX-NEXT: vpsrlw $1, %xmm0, %xmm2
; AVX-NEXT: vpand {{.*}}(%rip), %xmm2, %xmm2
; AVX-NEXT: vpaddb %xmm1, %xmm1, %xmm1
; AVX-NEXT: vpblendvb %xmm1, %xmm2, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: var_shift_v16i8:
; XOP: # %bb.0:
; XOP-NEXT: vpxor %xmm2, %xmm2, %xmm2
; XOP-NEXT: vpsubb %xmm1, %xmm2, %xmm1
; XOP-NEXT: vpshlb %xmm1, %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512DQ-LABEL: var_shift_v16i8:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
; AVX512DQ-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
; AVX512DQ-NEXT: vpsrlvd %zmm1, %zmm0, %zmm0
; AVX512DQ-NEXT: vpmovdb %zmm0, %xmm0
; AVX512DQ-NEXT: vzeroupper
; AVX512DQ-NEXT: retq
;
; AVX512BW-LABEL: var_shift_v16i8:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
; AVX512BW-NEXT: vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
; AVX512BW-NEXT: vpsrlvw %zmm1, %zmm0, %zmm0
; AVX512BW-NEXT: vpmovwb %zmm0, %ymm0
; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX512BW-NEXT: vzeroupper
; AVX512BW-NEXT: retq
;
; AVX512DQVL-LABEL: var_shift_v16i8:
; AVX512DQVL: # %bb.0:
; AVX512DQVL-NEXT: vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
; AVX512DQVL-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
; AVX512DQVL-NEXT: vpsrlvd %zmm1, %zmm0, %zmm0
; AVX512DQVL-NEXT: vpmovdb %zmm0, %xmm0
; AVX512DQVL-NEXT: vzeroupper
; AVX512DQVL-NEXT: retq
;
; AVX512BWVL-LABEL: var_shift_v16i8:
; AVX512BWVL: # %bb.0:
; AVX512BWVL-NEXT: vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
; AVX512BWVL-NEXT: vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
; AVX512BWVL-NEXT: vpsrlvw %ymm1, %ymm0, %ymm0
; AVX512BWVL-NEXT: vpmovwb %ymm0, %xmm0
; AVX512BWVL-NEXT: vzeroupper
; AVX512BWVL-NEXT: retq
;
; X32-SSE-LABEL: var_shift_v16i8:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: psllw $5, %xmm1
; X32-SSE-NEXT: pxor %xmm2, %xmm2
; X32-SSE-NEXT: pxor %xmm3, %xmm3
; X32-SSE-NEXT: pcmpgtb %xmm1, %xmm3
; X32-SSE-NEXT: movdqa %xmm3, %xmm4
; X32-SSE-NEXT: pandn %xmm0, %xmm4
; X32-SSE-NEXT: psrlw $4, %xmm0
; X32-SSE-NEXT: pand %xmm3, %xmm0
; X32-SSE-NEXT: pand {{\.LCPI.*}}, %xmm0
; X32-SSE-NEXT: por %xmm4, %xmm0
; X32-SSE-NEXT: paddb %xmm1, %xmm1
; X32-SSE-NEXT: pxor %xmm3, %xmm3
; X32-SSE-NEXT: pcmpgtb %xmm1, %xmm3
; X32-SSE-NEXT: movdqa %xmm3, %xmm4
; X32-SSE-NEXT: pandn %xmm0, %xmm4
; X32-SSE-NEXT: psrlw $2, %xmm0
; X32-SSE-NEXT: pand %xmm3, %xmm0
; X32-SSE-NEXT: pand {{\.LCPI.*}}, %xmm0
; X32-SSE-NEXT: por %xmm4, %xmm0
; X32-SSE-NEXT: paddb %xmm1, %xmm1
; X32-SSE-NEXT: pcmpgtb %xmm1, %xmm2
; X32-SSE-NEXT: movdqa %xmm2, %xmm1
; X32-SSE-NEXT: pandn %xmm0, %xmm1
; X32-SSE-NEXT: psrlw $1, %xmm0
; X32-SSE-NEXT: pand %xmm2, %xmm0
; X32-SSE-NEXT: pand {{\.LCPI.*}}, %xmm0
; X32-SSE-NEXT: por %xmm1, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <16 x i8> %a, %b
ret <16 x i8> %shift
}
;
; Uniform Variable Shifts
;
define <2 x i64> @splatvar_shift_v2i64(<2 x i64> %a, <2 x i64> %b) nounwind {
; SSE-LABEL: splatvar_shift_v2i64:
; SSE: # %bb.0:
; SSE-NEXT: psrlq %xmm1, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: splatvar_shift_v2i64:
; AVX: # %bb.0:
; AVX-NEXT: vpsrlq %xmm1, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: splatvar_shift_v2i64:
; XOP: # %bb.0:
; XOP-NEXT: vpsrlq %xmm1, %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512-LABEL: splatvar_shift_v2i64:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrlq %xmm1, %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: splatvar_shift_v2i64:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrlq %xmm1, %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: splatvar_shift_v2i64:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: psrlq %xmm1, %xmm0
; X32-SSE-NEXT: retl
%splat = shufflevector <2 x i64> %b, <2 x i64> undef, <2 x i32> zeroinitializer
%shift = lshr <2 x i64> %a, %splat
ret <2 x i64> %shift
}
define <4 x i32> @splatvar_shift_v4i32(<4 x i32> %a, <4 x i32> %b) nounwind {
; SSE2-LABEL: splatvar_shift_v4i32:
; SSE2: # %bb.0:
; SSE2-NEXT: xorps %xmm2, %xmm2
; SSE2-NEXT: movss {{.*#+}} xmm2 = xmm1[0],xmm2[1,2,3]
; SSE2-NEXT: psrld %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: splatvar_shift_v4i32:
; SSE41: # %bb.0:
; SSE41-NEXT: pmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; SSE41-NEXT: psrld %xmm1, %xmm0
; SSE41-NEXT: retq
;
; AVX-LABEL: splatvar_shift_v4i32:
; AVX: # %bb.0:
; AVX-NEXT: vpmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; AVX-NEXT: vpsrld %xmm1, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: splatvar_shift_v4i32:
; XOP: # %bb.0:
; XOP-NEXT: vpmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; XOP-NEXT: vpsrld %xmm1, %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512-LABEL: splatvar_shift_v4i32:
; AVX512: # %bb.0:
; AVX512-NEXT: vpmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; AVX512-NEXT: vpsrld %xmm1, %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: splatvar_shift_v4i32:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; AVX512VL-NEXT: vpsrld %xmm1, %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: splatvar_shift_v4i32:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: xorps %xmm2, %xmm2
; X32-SSE-NEXT: movss {{.*#+}} xmm2 = xmm1[0],xmm2[1,2,3]
; X32-SSE-NEXT: psrld %xmm2, %xmm0
; X32-SSE-NEXT: retl
%splat = shufflevector <4 x i32> %b, <4 x i32> undef, <4 x i32> zeroinitializer
%shift = lshr <4 x i32> %a, %splat
ret <4 x i32> %shift
}
define <8 x i16> @splatvar_shift_v8i16(<8 x i16> %a, <8 x i16> %b) nounwind {
; SSE2-LABEL: splatvar_shift_v8i16:
; SSE2: # %bb.0:
; SSE2-NEXT: pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0,1]
; SSE2-NEXT: psrldq {{.*#+}} xmm1 = xmm1[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE2-NEXT: psrlw %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: splatvar_shift_v8i16:
; SSE41: # %bb.0:
; SSE41-NEXT: pmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
; SSE41-NEXT: psrlw %xmm1, %xmm0
; SSE41-NEXT: retq
;
; AVX-LABEL: splatvar_shift_v8i16:
; AVX: # %bb.0:
; AVX-NEXT: vpmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
; AVX-NEXT: vpsrlw %xmm1, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: splatvar_shift_v8i16:
; XOP: # %bb.0:
; XOP-NEXT: vpmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
; XOP-NEXT: vpsrlw %xmm1, %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512-LABEL: splatvar_shift_v8i16:
; AVX512: # %bb.0:
; AVX512-NEXT: vpmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
; AVX512-NEXT: vpsrlw %xmm1, %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: splatvar_shift_v8i16:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
; AVX512VL-NEXT: vpsrlw %xmm1, %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: splatvar_shift_v8i16:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0,1]
; X32-SSE-NEXT: psrldq {{.*#+}} xmm1 = xmm1[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; X32-SSE-NEXT: psrlw %xmm1, %xmm0
; X32-SSE-NEXT: retl
%splat = shufflevector <8 x i16> %b, <8 x i16> undef, <8 x i32> zeroinitializer
%shift = lshr <8 x i16> %a, %splat
ret <8 x i16> %shift
}
define <16 x i8> @splatvar_shift_v16i8(<16 x i8> %a, <16 x i8> %b) nounwind {
; SSE2-LABEL: splatvar_shift_v16i8:
; SSE2: # %bb.0:
; SSE2-NEXT: pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0]
; SSE2-NEXT: psrldq {{.*#+}} xmm1 = xmm1[15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE2-NEXT: psrlw %xmm1, %xmm0
; SSE2-NEXT: pcmpeqd %xmm2, %xmm2
; SSE2-NEXT: psrlw %xmm1, %xmm2
; SSE2-NEXT: psrlw $8, %xmm2
; SSE2-NEXT: punpcklbw {{.*#+}} xmm2 = xmm2[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; SSE2-NEXT: pshuflw {{.*#+}} xmm1 = xmm2[0,0,2,3,4,5,6,7]
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[0,0,0,0]
; SSE2-NEXT: pand %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: splatvar_shift_v16i8:
; SSE41: # %bb.0:
; SSE41-NEXT: pmovzxbq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,zero,zero,zero,zero,xmm1[1],zero,zero,zero,zero,zero,zero,zero
; SSE41-NEXT: psrlw %xmm1, %xmm0
; SSE41-NEXT: pcmpeqd %xmm2, %xmm2
; SSE41-NEXT: psrlw %xmm1, %xmm2
; SSE41-NEXT: pshufb {{.*#+}} xmm2 = xmm2[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
; SSE41-NEXT: pand %xmm2, %xmm0
; SSE41-NEXT: retq
;
; AVX1-LABEL: splatvar_shift_v16i8:
; AVX1: # %bb.0:
; AVX1-NEXT: vpmovzxbq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,zero,zero,zero,zero,xmm1[1],zero,zero,zero,zero,zero,zero,zero
; AVX1-NEXT: vpsrlw %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vpcmpeqd %xmm2, %xmm2, %xmm2
; AVX1-NEXT: vpsrlw %xmm1, %xmm2, %xmm1
; AVX1-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
; AVX1-NEXT: vpand %xmm1, %xmm0, %xmm0
; AVX1-NEXT: retq
;
; AVX2-LABEL: splatvar_shift_v16i8:
; AVX2: # %bb.0:
; AVX2-NEXT: vpmovzxbq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,zero,zero,zero,zero,xmm1[1],zero,zero,zero,zero,zero,zero,zero
; AVX2-NEXT: vpsrlw %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vpcmpeqd %xmm2, %xmm2, %xmm2
; AVX2-NEXT: vpsrlw %xmm1, %xmm2, %xmm1
; AVX2-NEXT: vpsrlw $8, %xmm1, %xmm1
; AVX2-NEXT: vpbroadcastb %xmm1, %xmm1
; AVX2-NEXT: vpand %xmm1, %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOPAVX1-LABEL: splatvar_shift_v16i8:
; XOPAVX1: # %bb.0:
; XOPAVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
; XOPAVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
; XOPAVX1-NEXT: vpsubb %xmm1, %xmm2, %xmm1
; XOPAVX1-NEXT: vpshlb %xmm1, %xmm0, %xmm0
; XOPAVX1-NEXT: retq
;
; XOPAVX2-LABEL: splatvar_shift_v16i8:
; XOPAVX2: # %bb.0:
; XOPAVX2-NEXT: vpbroadcastb %xmm1, %xmm1
; XOPAVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
; XOPAVX2-NEXT: vpsubb %xmm1, %xmm2, %xmm1
; XOPAVX2-NEXT: vpshlb %xmm1, %xmm0, %xmm0
; XOPAVX2-NEXT: retq
;
; AVX512DQ-LABEL: splatvar_shift_v16i8:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vpbroadcastb %xmm1, %xmm1
; AVX512DQ-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
; AVX512DQ-NEXT: vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
; AVX512DQ-NEXT: vpsrlvd %zmm1, %zmm0, %zmm0
; AVX512DQ-NEXT: vpmovdb %zmm0, %xmm0
; AVX512DQ-NEXT: vzeroupper
; AVX512DQ-NEXT: retq
;
; AVX512BW-LABEL: splatvar_shift_v16i8:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vpbroadcastb %xmm1, %xmm1
; AVX512BW-NEXT: vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
; AVX512BW-NEXT: vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
; AVX512BW-NEXT: vpsrlvw %zmm1, %zmm0, %zmm0
; AVX512BW-NEXT: vpmovwb %zmm0, %ymm0
; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX512BW-NEXT: vzeroupper
; AVX512BW-NEXT: retq
;
; AVX512DQVL-LABEL: splatvar_shift_v16i8:
; AVX512DQVL: # %bb.0:
; AVX512DQVL-NEXT: vpbroadcastb %xmm1, %xmm1
; AVX512DQVL-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
; AVX512DQVL-NEXT: vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
; AVX512DQVL-NEXT: vpsrlvd %zmm1, %zmm0, %zmm0
; AVX512DQVL-NEXT: vpmovdb %zmm0, %xmm0
; AVX512DQVL-NEXT: vzeroupper
; AVX512DQVL-NEXT: retq
;
; AVX512BWVL-LABEL: splatvar_shift_v16i8:
; AVX512BWVL: # %bb.0:
; AVX512BWVL-NEXT: vpbroadcastb %xmm1, %xmm1
; AVX512BWVL-NEXT: vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
; AVX512BWVL-NEXT: vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
; AVX512BWVL-NEXT: vpsrlvw %ymm1, %ymm0, %ymm0
; AVX512BWVL-NEXT: vpmovwb %ymm0, %xmm0
; AVX512BWVL-NEXT: vzeroupper
; AVX512BWVL-NEXT: retq
;
; X32-SSE-LABEL: splatvar_shift_v16i8:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0]
; X32-SSE-NEXT: psrldq {{.*#+}} xmm1 = xmm1[15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; X32-SSE-NEXT: psrlw %xmm1, %xmm0
; X32-SSE-NEXT: pcmpeqd %xmm2, %xmm2
; X32-SSE-NEXT: psrlw %xmm1, %xmm2
; X32-SSE-NEXT: psrlw $8, %xmm2
; X32-SSE-NEXT: punpcklbw {{.*#+}} xmm2 = xmm2[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; X32-SSE-NEXT: pshuflw {{.*#+}} xmm1 = xmm2[0,0,2,3,4,5,6,7]
; X32-SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[0,0,0,0]
; X32-SSE-NEXT: pand %xmm1, %xmm0
; X32-SSE-NEXT: retl
%splat = shufflevector <16 x i8> %b, <16 x i8> undef, <16 x i32> zeroinitializer
%shift = lshr <16 x i8> %a, %splat
ret <16 x i8> %shift
}
;
; Constant Shifts
;
define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) nounwind {
; SSE2-LABEL: constant_shift_v2i64:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm1
; SSE2-NEXT: psrlq $1, %xmm1
; SSE2-NEXT: psrlq $7, %xmm0
; SSE2-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; SSE2-NEXT: retq
;
; SSE41-LABEL: constant_shift_v2i64:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: psrlq $7, %xmm1
; SSE41-NEXT: psrlq $1, %xmm0
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; SSE41-NEXT: retq
;
; AVX1-LABEL: constant_shift_v2i64:
; AVX1: # %bb.0:
; AVX1-NEXT: vpsrlq $7, %xmm0, %xmm1
; AVX1-NEXT: vpsrlq $1, %xmm0, %xmm0
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; AVX1-NEXT: retq
;
; AVX2-LABEL: constant_shift_v2i64:
; AVX2: # %bb.0:
; AVX2-NEXT: vpsrlvq {{.*}}(%rip), %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOPAVX1-LABEL: constant_shift_v2i64:
; XOPAVX1: # %bb.0:
; XOPAVX1-NEXT: vpshlq {{.*}}(%rip), %xmm0, %xmm0
; XOPAVX1-NEXT: retq
;
; XOPAVX2-LABEL: constant_shift_v2i64:
; XOPAVX2: # %bb.0:
; XOPAVX2-NEXT: vpsrlvq {{.*}}(%rip), %xmm0, %xmm0
; XOPAVX2-NEXT: retq
;
; AVX512-LABEL: constant_shift_v2i64:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrlvq {{.*}}(%rip), %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: constant_shift_v2i64:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrlvq {{.*}}(%rip), %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: constant_shift_v2i64:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: movdqa %xmm0, %xmm1
; X32-SSE-NEXT: psrlq $1, %xmm1
; X32-SSE-NEXT: psrlq $7, %xmm0
; X32-SSE-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; X32-SSE-NEXT: retl
%shift = lshr <2 x i64> %a, <i64 1, i64 7>
ret <2 x i64> %shift
}
define <4 x i32> @constant_shift_v4i32(<4 x i32> %a) nounwind {
; SSE2-LABEL: constant_shift_v4i32:
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm1
; SSE2-NEXT: psrld $7, %xmm1
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: psrld $6, %xmm2
; SSE2-NEXT: punpckhqdq {{.*#+}} xmm2 = xmm2[1],xmm1[1]
; SSE2-NEXT: movdqa %xmm0, %xmm1
; SSE2-NEXT: psrld $5, %xmm1
; SSE2-NEXT: psrld $4, %xmm0
; SSE2-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,3],xmm2[0,3]
; SSE2-NEXT: retq
;
; SSE41-LABEL: constant_shift_v4i32:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: psrld $7, %xmm1
; SSE41-NEXT: movdqa %xmm0, %xmm2
; SSE41-NEXT: psrld $5, %xmm2
; SSE41-NEXT: pblendw {{.*#+}} xmm2 = xmm2[0,1,2,3],xmm1[4,5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm1
; SSE41-NEXT: psrld $6, %xmm1
; SSE41-NEXT: psrld $4, %xmm0
; 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: constant_shift_v4i32:
; AVX1: # %bb.0:
; AVX1-NEXT: vpsrld $7, %xmm0, %xmm1
; AVX1-NEXT: vpsrld $5, %xmm0, %xmm2
; AVX1-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0,1,2,3],xmm1[4,5,6,7]
; AVX1-NEXT: vpsrld $6, %xmm0, %xmm2
; AVX1-NEXT: vpsrld $4, %xmm0, %xmm0
; 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: constant_shift_v4i32:
; AVX2: # %bb.0:
; AVX2-NEXT: vpsrlvd {{.*}}(%rip), %xmm0, %xmm0
; AVX2-NEXT: retq
;
; XOPAVX1-LABEL: constant_shift_v4i32:
; XOPAVX1: # %bb.0:
; XOPAVX1-NEXT: vpshld {{.*}}(%rip), %xmm0, %xmm0
; XOPAVX1-NEXT: retq
;
; XOPAVX2-LABEL: constant_shift_v4i32:
; XOPAVX2: # %bb.0:
; XOPAVX2-NEXT: vpsrlvd {{.*}}(%rip), %xmm0, %xmm0
; XOPAVX2-NEXT: retq
;
; AVX512-LABEL: constant_shift_v4i32:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrlvd {{.*}}(%rip), %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: constant_shift_v4i32:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrlvd {{.*}}(%rip), %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: constant_shift_v4i32:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: movdqa %xmm0, %xmm1
; X32-SSE-NEXT: psrld $7, %xmm1
; X32-SSE-NEXT: movdqa %xmm0, %xmm2
; X32-SSE-NEXT: psrld $6, %xmm2
; X32-SSE-NEXT: punpckhqdq {{.*#+}} xmm2 = xmm2[1],xmm1[1]
; X32-SSE-NEXT: movdqa %xmm0, %xmm1
; X32-SSE-NEXT: psrld $5, %xmm1
; X32-SSE-NEXT: psrld $4, %xmm0
; X32-SSE-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; X32-SSE-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,3],xmm2[0,3]
; X32-SSE-NEXT: retl
%shift = lshr <4 x i32> %a, <i32 4, i32 5, i32 6, i32 7>
ret <4 x i32> %shift
}
define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) nounwind {
; SSE2-LABEL: constant_shift_v8i16:
; 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: pmulhuw {{.*}}(%rip), %xmm0
; SSE2-NEXT: pand %xmm1, %xmm0
; SSE2-NEXT: por %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: constant_shift_v8i16:
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <u,32768,16384,8192,4096,2048,1024,512>
; SSE41-NEXT: pmulhuw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; SSE41-NEXT: retq
;
; AVX-LABEL: constant_shift_v8i16:
; AVX: # %bb.0:
; AVX-NEXT: vpmulhuw {{.*}}(%rip), %xmm0, %xmm1
; AVX-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; AVX-NEXT: retq
;
; XOP-LABEL: constant_shift_v8i16:
; XOP: # %bb.0:
; XOP-NEXT: vpshlw {{.*}}(%rip), %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512DQ-LABEL: constant_shift_v8i16:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vpmulhuw {{.*}}(%rip), %xmm0, %xmm1
; AVX512DQ-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; AVX512DQ-NEXT: retq
;
; AVX512BW-LABEL: constant_shift_v8i16:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = [0,1,2,3,4,5,6,7]
; AVX512BW-NEXT: vpsrlvw %zmm1, %zmm0, %zmm0
; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
; AVX512BW-NEXT: vzeroupper
; AVX512BW-NEXT: retq
;
; AVX512DQVL-LABEL: constant_shift_v8i16:
; AVX512DQVL: # %bb.0:
; AVX512DQVL-NEXT: vpmulhuw {{.*}}(%rip), %xmm0, %xmm1
; AVX512DQVL-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; AVX512DQVL-NEXT: retq
;
; AVX512BWVL-LABEL: constant_shift_v8i16:
; AVX512BWVL: # %bb.0:
; AVX512BWVL-NEXT: vpsrlvw {{.*}}(%rip), %xmm0, %xmm0
; AVX512BWVL-NEXT: retq
;
; X32-SSE-LABEL: constant_shift_v8i16:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,65535,65535,65535,65535,65535,65535]
; X32-SSE-NEXT: movdqa %xmm1, %xmm2
; X32-SSE-NEXT: pandn %xmm0, %xmm2
; X32-SSE-NEXT: pmulhuw {{\.LCPI.*}}, %xmm0
; X32-SSE-NEXT: pand %xmm1, %xmm0
; X32-SSE-NEXT: por %xmm2, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
ret <8 x i16> %shift
}
define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) nounwind {
; SSE2-LABEL: constant_shift_v16i8:
; SSE2: # %bb.0:
; SSE2-NEXT: pxor %xmm1, %xmm1
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1[12],xmm2[13],xmm1[13],xmm2[14],xmm1[14],xmm2[15],xmm1[15]
; SSE2-NEXT: pmullw {{.*}}(%rip), %xmm2
; SSE2-NEXT: psrlw $8, %xmm2
; SSE2-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; SSE2-NEXT: pmullw {{.*}}(%rip), %xmm0
; SSE2-NEXT: psrlw $8, %xmm0
; SSE2-NEXT: packuswb %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: constant_shift_v16i8:
; SSE41: # %bb.0:
; SSE41-NEXT: pxor %xmm2, %xmm2
; SSE41-NEXT: pmovzxbw {{.*#+}} 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
; SSE41-NEXT: punpckhbw {{.*#+}} xmm0 = xmm0[8],xmm2[8],xmm0[9],xmm2[9],xmm0[10],xmm2[10],xmm0[11],xmm2[11],xmm0[12],xmm2[12],xmm0[13],xmm2[13],xmm0[14],xmm2[14],xmm0[15],xmm2[15]
; SSE41-NEXT: pmullw {{.*}}(%rip), %xmm0
; SSE41-NEXT: psrlw $8, %xmm0
; SSE41-NEXT: pmullw {{.*}}(%rip), %xmm1
; SSE41-NEXT: psrlw $8, %xmm1
; SSE41-NEXT: packuswb %xmm0, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm0
; SSE41-NEXT: retq
;
; AVX1-LABEL: constant_shift_v16i8:
; AVX1: # %bb.0:
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX1-NEXT: vpunpckhbw {{.*#+}} xmm1 = 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]
; AVX1-NEXT: vpmullw {{.*}}(%rip), %xmm1, %xmm1
; AVX1-NEXT: vpsrlw $8, %xmm1, %xmm1
; AVX1-NEXT: vpmovzxbw {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
; AVX1-NEXT: vpmullw {{.*}}(%rip), %xmm0, %xmm0
; AVX1-NEXT: vpsrlw $8, %xmm0, %xmm0
; AVX1-NEXT: vpackuswb %xmm1, %xmm0, %xmm0
; AVX1-NEXT: retq
;
; AVX2-LABEL: constant_shift_v16i8:
; AVX2: # %bb.0:
; AVX2-NEXT: vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
; AVX2-NEXT: vpmullw {{.*}}(%rip), %ymm0, %ymm0
; AVX2-NEXT: vpsrlw $8, %ymm0, %ymm0
; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-NEXT: vpackuswb %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: retq
;
; XOP-LABEL: constant_shift_v16i8:
; XOP: # %bb.0:
; XOP-NEXT: vpshlb {{.*}}(%rip), %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512DQ-LABEL: constant_shift_v16i8:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
; AVX512DQ-NEXT: vpsrlvd {{.*}}(%rip), %zmm0, %zmm0
; AVX512DQ-NEXT: vpmovdb %zmm0, %xmm0
; AVX512DQ-NEXT: vzeroupper
; AVX512DQ-NEXT: retq
;
; AVX512BW-LABEL: constant_shift_v16i8:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,4,5,6,7,7,6,5,4,3,2,1,0]
; AVX512BW-NEXT: vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
; AVX512BW-NEXT: vpsrlvw %zmm1, %zmm0, %zmm0
; AVX512BW-NEXT: vpmovwb %zmm0, %ymm0
; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 killed $ymm0
; AVX512BW-NEXT: vzeroupper
; AVX512BW-NEXT: retq
;
; AVX512DQVL-LABEL: constant_shift_v16i8:
; AVX512DQVL: # %bb.0:
; AVX512DQVL-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
; AVX512DQVL-NEXT: vpsrlvd {{.*}}(%rip), %zmm0, %zmm0
; AVX512DQVL-NEXT: vpmovdb %zmm0, %xmm0
; AVX512DQVL-NEXT: vzeroupper
; AVX512DQVL-NEXT: retq
;
; AVX512BWVL-LABEL: constant_shift_v16i8:
; AVX512BWVL: # %bb.0:
; AVX512BWVL-NEXT: vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
; AVX512BWVL-NEXT: vpsrlvw {{.*}}(%rip), %ymm0, %ymm0
; AVX512BWVL-NEXT: vpmovwb %ymm0, %xmm0
; AVX512BWVL-NEXT: vzeroupper
; AVX512BWVL-NEXT: retq
;
; X32-SSE-LABEL: constant_shift_v16i8:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: pxor %xmm1, %xmm1
; X32-SSE-NEXT: movdqa %xmm0, %xmm2
; X32-SSE-NEXT: punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1[12],xmm2[13],xmm1[13],xmm2[14],xmm1[14],xmm2[15],xmm1[15]
; X32-SSE-NEXT: pmullw {{\.LCPI.*}}, %xmm2
; X32-SSE-NEXT: psrlw $8, %xmm2
; X32-SSE-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; X32-SSE-NEXT: pmullw {{\.LCPI.*}}, %xmm0
; X32-SSE-NEXT: psrlw $8, %xmm0
; X32-SSE-NEXT: packuswb %xmm2, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <16 x i8> %a, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>
ret <16 x i8> %shift
}
;
; Uniform Constant Shifts
;
define <2 x i64> @splatconstant_shift_v2i64(<2 x i64> %a) nounwind {
; SSE-LABEL: splatconstant_shift_v2i64:
; SSE: # %bb.0:
; SSE-NEXT: psrlq $7, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: splatconstant_shift_v2i64:
; AVX: # %bb.0:
; AVX-NEXT: vpsrlq $7, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: splatconstant_shift_v2i64:
; XOP: # %bb.0:
; XOP-NEXT: vpsrlq $7, %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512-LABEL: splatconstant_shift_v2i64:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrlq $7, %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: splatconstant_shift_v2i64:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrlq $7, %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: splatconstant_shift_v2i64:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: psrlq $7, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <2 x i64> %a, <i64 7, i64 7>
ret <2 x i64> %shift
}
define <4 x i32> @splatconstant_shift_v4i32(<4 x i32> %a) nounwind {
; SSE-LABEL: splatconstant_shift_v4i32:
; SSE: # %bb.0:
; SSE-NEXT: psrld $5, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: splatconstant_shift_v4i32:
; AVX: # %bb.0:
; AVX-NEXT: vpsrld $5, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: splatconstant_shift_v4i32:
; XOP: # %bb.0:
; XOP-NEXT: vpsrld $5, %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512-LABEL: splatconstant_shift_v4i32:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrld $5, %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: splatconstant_shift_v4i32:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrld $5, %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: splatconstant_shift_v4i32:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: psrld $5, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <4 x i32> %a, <i32 5, i32 5, i32 5, i32 5>
ret <4 x i32> %shift
}
define <8 x i16> @splatconstant_shift_v8i16(<8 x i16> %a) nounwind {
; SSE-LABEL: splatconstant_shift_v8i16:
; SSE: # %bb.0:
; SSE-NEXT: psrlw $3, %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: splatconstant_shift_v8i16:
; AVX: # %bb.0:
; AVX-NEXT: vpsrlw $3, %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: splatconstant_shift_v8i16:
; XOP: # %bb.0:
; XOP-NEXT: vpsrlw $3, %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512-LABEL: splatconstant_shift_v8i16:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrlw $3, %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: splatconstant_shift_v8i16:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrlw $3, %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: splatconstant_shift_v8i16:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: psrlw $3, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <8 x i16> %a, <i16 3, i16 3, i16 3, i16 3, i16 3, i16 3, i16 3, i16 3>
ret <8 x i16> %shift
}
define <16 x i8> @splatconstant_shift_v16i8(<16 x i8> %a) nounwind {
; SSE-LABEL: splatconstant_shift_v16i8:
; SSE: # %bb.0:
; SSE-NEXT: psrlw $3, %xmm0
; SSE-NEXT: pand {{.*}}(%rip), %xmm0
; SSE-NEXT: retq
;
; AVX-LABEL: splatconstant_shift_v16i8:
; AVX: # %bb.0:
; AVX-NEXT: vpsrlw $3, %xmm0, %xmm0
; AVX-NEXT: vpand {{.*}}(%rip), %xmm0, %xmm0
; AVX-NEXT: retq
;
; XOP-LABEL: splatconstant_shift_v16i8:
; XOP: # %bb.0:
; XOP-NEXT: vpshlb {{.*}}(%rip), %xmm0, %xmm0
; XOP-NEXT: retq
;
; AVX512-LABEL: splatconstant_shift_v16i8:
; AVX512: # %bb.0:
; AVX512-NEXT: vpsrlw $3, %xmm0, %xmm0
; AVX512-NEXT: vpand {{.*}}(%rip), %xmm0, %xmm0
; AVX512-NEXT: retq
;
; AVX512VL-LABEL: splatconstant_shift_v16i8:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vpsrlw $3, %xmm0, %xmm0
; AVX512VL-NEXT: vpand {{.*}}(%rip), %xmm0, %xmm0
; AVX512VL-NEXT: retq
;
; X32-SSE-LABEL: splatconstant_shift_v16i8:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: psrlw $3, %xmm0
; X32-SSE-NEXT: pand {{\.LCPI.*}}, %xmm0
; X32-SSE-NEXT: retl
%shift = lshr <16 x i8> %a, <i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3, i8 3>
ret <16 x i8> %shift
}
define <4 x i32> @vector_variable_shift_right(<4 x i1> %cond, <4 x i32> %x, <4 x i32> %y, <4 x i32> %z) nounwind {
; SSE2-LABEL: vector_variable_shift_right:
; SSE2: # %bb.0:
; SSE2-NEXT: xorps %xmm4, %xmm4
; SSE2-NEXT: xorps %xmm5, %xmm5
; SSE2-NEXT: movss {{.*#+}} xmm5 = xmm2[0],xmm5[1,2,3]
; SSE2-NEXT: movss {{.*#+}} xmm4 = xmm1[0],xmm4[1,2,3]
; SSE2-NEXT: pslld $31, %xmm0
; SSE2-NEXT: psrad $31, %xmm0
; SSE2-NEXT: movdqa %xmm3, %xmm1
; SSE2-NEXT: psrld %xmm4, %xmm1
; SSE2-NEXT: psrld %xmm5, %xmm3
; SSE2-NEXT: pand %xmm0, %xmm1
; SSE2-NEXT: pandn %xmm3, %xmm0
; SSE2-NEXT: por %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSE41-LABEL: vector_variable_shift_right:
; SSE41: # %bb.0:
; SSE41-NEXT: pslld $31, %xmm0
; SSE41-NEXT: pmovzxdq {{.*#+}} xmm2 = xmm2[0],zero,xmm2[1],zero
; SSE41-NEXT: pmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; SSE41-NEXT: movdqa %xmm3, %xmm4
; SSE41-NEXT: psrld %xmm1, %xmm4
; SSE41-NEXT: psrld %xmm2, %xmm3
; SSE41-NEXT: blendvps %xmm0, %xmm4, %xmm3
; SSE41-NEXT: movaps %xmm3, %xmm0
; SSE41-NEXT: retq
;
; AVX1-LABEL: vector_variable_shift_right:
; AVX1: # %bb.0:
; AVX1-NEXT: vpslld $31, %xmm0, %xmm0
; AVX1-NEXT: vpmovzxdq {{.*#+}} xmm2 = xmm2[0],zero,xmm2[1],zero
; AVX1-NEXT: vpmovzxdq {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero
; AVX1-NEXT: vpsrld %xmm1, %xmm3, %xmm1
; AVX1-NEXT: vpsrld %xmm2, %xmm3, %xmm2
; AVX1-NEXT: vblendvps %xmm0, %xmm1, %xmm2, %xmm0
; AVX1-NEXT: retq
;
; AVX2-LABEL: vector_variable_shift_right:
; AVX2: # %bb.0:
; AVX2-NEXT: vpslld $31, %xmm0, %xmm0
; AVX2-NEXT: vbroadcastss %xmm1, %xmm1
; AVX2-NEXT: vbroadcastss %xmm2, %xmm2
; AVX2-NEXT: vblendvps %xmm0, %xmm1, %xmm2, %xmm0
; AVX2-NEXT: vpsrlvd %xmm0, %xmm3, %xmm0
; AVX2-NEXT: retq
;
; XOPAVX1-LABEL: vector_variable_shift_right:
; XOPAVX1: # %bb.0:
; XOPAVX1-NEXT: vpslld $31, %xmm0, %xmm0
; XOPAVX1-NEXT: vpermilps {{.*#+}} xmm1 = xmm1[0,0,0,0]
; XOPAVX1-NEXT: vpermilps {{.*#+}} xmm2 = xmm2[0,0,0,0]
; XOPAVX1-NEXT: vblendvps %xmm0, %xmm1, %xmm2, %xmm0
; XOPAVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
; XOPAVX1-NEXT: vpsubd %xmm0, %xmm1, %xmm0
; XOPAVX1-NEXT: vpshld %xmm0, %xmm3, %xmm0
; XOPAVX1-NEXT: retq
;
; XOPAVX2-LABEL: vector_variable_shift_right:
; XOPAVX2: # %bb.0:
; XOPAVX2-NEXT: vpslld $31, %xmm0, %xmm0
; XOPAVX2-NEXT: vbroadcastss %xmm1, %xmm1
; XOPAVX2-NEXT: vbroadcastss %xmm2, %xmm2
; XOPAVX2-NEXT: vblendvps %xmm0, %xmm1, %xmm2, %xmm0
; XOPAVX2-NEXT: vpsrlvd %xmm0, %xmm3, %xmm0
; XOPAVX2-NEXT: retq
;
; AVX512DQ-LABEL: vector_variable_shift_right:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vpslld $31, %xmm0, %xmm0
; AVX512DQ-NEXT: vpmovd2m %zmm0, %k1
; AVX512DQ-NEXT: vpbroadcastd %xmm1, %xmm0
; AVX512DQ-NEXT: vpbroadcastd %xmm2, %xmm1
; AVX512DQ-NEXT: vmovdqa32 %zmm0, %zmm1 {%k1}
; AVX512DQ-NEXT: vpsrlvd %xmm1, %xmm3, %xmm0
; AVX512DQ-NEXT: vzeroupper
; AVX512DQ-NEXT: retq
;
; AVX512BW-LABEL: vector_variable_shift_right:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vpslld $31, %xmm0, %xmm0
; AVX512BW-NEXT: vptestmd %zmm0, %zmm0, %k1
; AVX512BW-NEXT: vpbroadcastd %xmm1, %xmm0
; AVX512BW-NEXT: vpbroadcastd %xmm2, %xmm1
; AVX512BW-NEXT: vmovdqa32 %zmm0, %zmm1 {%k1}
; AVX512BW-NEXT: vpsrlvd %xmm1, %xmm3, %xmm0
; AVX512BW-NEXT: vzeroupper
; AVX512BW-NEXT: retq
;
; AVX512DQVL-LABEL: vector_variable_shift_right:
; AVX512DQVL: # %bb.0:
; AVX512DQVL-NEXT: vpslld $31, %xmm0, %xmm0
; AVX512DQVL-NEXT: vpmovd2m %xmm0, %k1
; AVX512DQVL-NEXT: vpbroadcastd %xmm2, %xmm0
; AVX512DQVL-NEXT: vpbroadcastd %xmm1, %xmm0 {%k1}
; AVX512DQVL-NEXT: vpsrlvd %xmm0, %xmm3, %xmm0
; AVX512DQVL-NEXT: retq
;
; AVX512BWVL-LABEL: vector_variable_shift_right:
; AVX512BWVL: # %bb.0:
; AVX512BWVL-NEXT: vpslld $31, %xmm0, %xmm0
; AVX512BWVL-NEXT: vptestmd %xmm0, %xmm0, %k1
; AVX512BWVL-NEXT: vpbroadcastd %xmm2, %xmm0
; AVX512BWVL-NEXT: vpbroadcastd %xmm1, %xmm0 {%k1}
; AVX512BWVL-NEXT: vpsrlvd %xmm0, %xmm3, %xmm0
; AVX512BWVL-NEXT: retq
;
; X32-SSE-LABEL: vector_variable_shift_right:
; X32-SSE: # %bb.0:
; X32-SSE-NEXT: pushl %ebp
; X32-SSE-NEXT: movl %esp, %ebp
; X32-SSE-NEXT: andl $-16, %esp
; X32-SSE-NEXT: subl $16, %esp
; X32-SSE-NEXT: xorps %xmm3, %xmm3
; X32-SSE-NEXT: xorps %xmm4, %xmm4
; X32-SSE-NEXT: movss {{.*#+}} xmm4 = xmm2[0],xmm4[1,2,3]
; X32-SSE-NEXT: movss {{.*#+}} xmm3 = xmm1[0],xmm3[1,2,3]
; X32-SSE-NEXT: pslld $31, %xmm0
; X32-SSE-NEXT: psrad $31, %xmm0
; X32-SSE-NEXT: movdqa 8(%ebp), %xmm1
; X32-SSE-NEXT: movdqa %xmm1, %xmm2
; X32-SSE-NEXT: psrld %xmm3, %xmm2
; X32-SSE-NEXT: psrld %xmm4, %xmm1
; X32-SSE-NEXT: pand %xmm0, %xmm2
; X32-SSE-NEXT: pandn %xmm1, %xmm0
; X32-SSE-NEXT: por %xmm2, %xmm0
; X32-SSE-NEXT: movl %ebp, %esp
; X32-SSE-NEXT: popl %ebp
; X32-SSE-NEXT: retl
%splat1 = shufflevector <4 x i32> %x, <4 x i32> undef, <4 x i32> zeroinitializer
%splat2 = shufflevector <4 x i32> %y, <4 x i32> undef, <4 x i32> zeroinitializer
%sel = select <4 x i1> %cond, <4 x i32> %splat1, <4 x i32> %splat2
%sh = lshr <4 x i32> %z, %sel
ret <4 x i32> %sh
}
|