llvm.org GIT mirror llvm / release_80 lib / Target / AMDGPU / SOPInstructions.td
release_80

Tree @release_80 (Download .tar.gz)

SOPInstructions.td @release_80raw · 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
 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
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
//===-- SOPInstructions.td - SOP Instruction Defintions -------------------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//

def GPRIdxModeMatchClass : AsmOperandClass {
  let Name = "GPRIdxMode";
  let PredicateMethod = "isGPRIdxMode";
  let RenderMethod = "addImmOperands";
}

def GPRIdxMode : Operand<i32> {
  let PrintMethod = "printVGPRIndexMode";
  let ParserMatchClass = GPRIdxModeMatchClass;
  let OperandType = "OPERAND_IMMEDIATE";
}

class SOP_Pseudo<string opName, dag outs, dag ins, string asmOps,
                  list<dag> pattern=[]> :
    InstSI<outs, ins, "", pattern>,
    SIMCInstr<opName, SIEncodingFamily.NONE> {

  let isPseudo = 1;
  let isCodeGenOnly = 1;
  let SubtargetPredicate = isGCN;

  string Mnemonic = opName;
  string AsmOperands = asmOps;

  bits<1> has_sdst = 0;
}

//===----------------------------------------------------------------------===//
// SOP1 Instructions
//===----------------------------------------------------------------------===//

class SOP1_Pseudo <string opName, dag outs, dag ins,
                   string asmOps, list<dag> pattern=[]> :
  SOP_Pseudo<opName, outs, ins, asmOps, pattern> {

  let mayLoad = 0;
  let mayStore = 0;
  let hasSideEffects = 0;
  let SALU = 1;
  let SOP1 = 1;
  let SchedRW = [WriteSALU];
  let Size = 4;
  let UseNamedOperandTable = 1;

  bits<1> has_src0 = 1;
  bits<1> has_sdst = 1;
}

class SOP1_Real<bits<8> op, SOP1_Pseudo ps> :
  InstSI <ps.OutOperandList, ps.InOperandList,
          ps.Mnemonic # " " # ps.AsmOperands, []>,
  Enc32 {

  let isPseudo = 0;
  let isCodeGenOnly = 0;
  let Size = 4;

  // copy relevant pseudo op flags
  let SubtargetPredicate = ps.SubtargetPredicate;
  let AsmMatchConverter  = ps.AsmMatchConverter;

  // encoding
  bits<7> sdst;
  bits<8> src0;

  let Inst{7-0} = !if(ps.has_src0, src0, ?);
  let Inst{15-8} = op;
  let Inst{22-16} = !if(ps.has_sdst, sdst, ?);
  let Inst{31-23} = 0x17d; //encoding;
}

class SOP1_32 <string opName, list<dag> pattern=[]> : SOP1_Pseudo <
  opName, (outs SReg_32:$sdst), (ins SSrc_b32:$src0),
  "$sdst, $src0", pattern
>;

// 32-bit input, no output.
class SOP1_0_32 <string opName, list<dag> pattern = []> : SOP1_Pseudo <
  opName, (outs), (ins SSrc_b32:$src0),
  "$src0", pattern> {
  let has_sdst = 0;
}

class SOP1_0_32R <string opName, list<dag> pattern = []> : SOP1_Pseudo <
  opName, (outs), (ins SReg_32:$src0),
  "$src0", pattern> {
  let has_sdst = 0;
}

class SOP1_64 <string opName, list<dag> pattern=[]> : SOP1_Pseudo <
  opName, (outs SReg_64:$sdst), (ins SSrc_b64:$src0),
  "$sdst, $src0", pattern
>;

// 64-bit input, 32-bit output.
class SOP1_32_64 <string opName, list<dag> pattern=[]> : SOP1_Pseudo <
  opName, (outs SReg_32:$sdst), (ins SSrc_b64:$src0),
  "$sdst, $src0", pattern
>;

// 32-bit input, 64-bit output.
class SOP1_64_32 <string opName, list<dag> pattern=[]> : SOP1_Pseudo <
  opName, (outs SReg_64:$sdst), (ins SSrc_b32:$src0),
  "$sdst, $src0", pattern
>;

// no input, 64-bit output.
class SOP1_64_0 <string opName, list<dag> pattern=[]> : SOP1_Pseudo <
  opName, (outs SReg_64:$sdst), (ins), "$sdst", pattern> {
  let has_src0 = 0;
}

// 64-bit input, no output
class SOP1_1 <string opName, list<dag> pattern=[]> : SOP1_Pseudo <
  opName, (outs), (ins SReg_64:$src0), "$src0", pattern> {
  let has_sdst = 0;
}


let isMoveImm = 1 in {
  let isReMaterializable = 1, isAsCheapAsAMove = 1 in {
    def S_MOV_B32 : SOP1_32 <"s_mov_b32">;
    def S_MOV_B64 : SOP1_64 <"s_mov_b64">;
  } // End isRematerializeable = 1

  let Uses = [SCC] in {
    def S_CMOV_B32 : SOP1_32 <"s_cmov_b32">;
    def S_CMOV_B64 : SOP1_64 <"s_cmov_b64">;
  } // End Uses = [SCC]
} // End isMoveImm = 1

let Defs = [SCC] in {
  def S_NOT_B32 : SOP1_32 <"s_not_b32",
    [(set i32:$sdst, (not i32:$src0))]
  >;

  def S_NOT_B64 : SOP1_64 <"s_not_b64",
    [(set i64:$sdst, (not i64:$src0))]
  >;
  def S_WQM_B32 : SOP1_32 <"s_wqm_b32">;
  def S_WQM_B64 : SOP1_64 <"s_wqm_b64",
    [(set i1:$sdst, (int_amdgcn_wqm_vote i1:$src0))]
  >;
} // End Defs = [SCC]


def S_BREV_B32 : SOP1_32 <"s_brev_b32",
  [(set i32:$sdst, (bitreverse i32:$src0))]
>;
def S_BREV_B64 : SOP1_64 <"s_brev_b64">;

let Defs = [SCC] in {
def S_BCNT0_I32_B32 : SOP1_32 <"s_bcnt0_i32_b32">;
def S_BCNT0_I32_B64 : SOP1_32_64 <"s_bcnt0_i32_b64">;
def S_BCNT1_I32_B32 : SOP1_32 <"s_bcnt1_i32_b32",
  [(set i32:$sdst, (ctpop i32:$src0))]
>;
def S_BCNT1_I32_B64 : SOP1_32_64 <"s_bcnt1_i32_b64">;
} // End Defs = [SCC]

def S_FF0_I32_B32 : SOP1_32 <"s_ff0_i32_b32">;
def S_FF0_I32_B64 : SOP1_32_64 <"s_ff0_i32_b64">;
def S_FF1_I32_B64 : SOP1_32_64 <"s_ff1_i32_b64">;

def S_FF1_I32_B32 : SOP1_32 <"s_ff1_i32_b32",
  [(set i32:$sdst, (AMDGPUffbl_b32 i32:$src0))]
>;

def S_FLBIT_I32_B32 : SOP1_32 <"s_flbit_i32_b32",
  [(set i32:$sdst, (AMDGPUffbh_u32 i32:$src0))]
>;

def S_FLBIT_I32_B64 : SOP1_32_64 <"s_flbit_i32_b64">;
def S_FLBIT_I32 : SOP1_32 <"s_flbit_i32",
  [(set i32:$sdst, (AMDGPUffbh_i32 i32:$src0))]
>;
def S_FLBIT_I32_I64 : SOP1_32_64 <"s_flbit_i32_i64">;
def S_SEXT_I32_I8 : SOP1_32 <"s_sext_i32_i8",
  [(set i32:$sdst, (sext_inreg i32:$src0, i8))]
>;
def S_SEXT_I32_I16 : SOP1_32 <"s_sext_i32_i16",
  [(set i32:$sdst, (sext_inreg i32:$src0, i16))]
>;

def S_BITSET0_B32 : SOP1_32    <"s_bitset0_b32">;
def S_BITSET0_B64 : SOP1_64_32 <"s_bitset0_b64">;
def S_BITSET1_B32 : SOP1_32    <"s_bitset1_b32">;
def S_BITSET1_B64 : SOP1_64_32 <"s_bitset1_b64">;
def S_GETPC_B64 : SOP1_64_0  <"s_getpc_b64",
  [(set i64:$sdst, (int_amdgcn_s_getpc))]
>;

let isTerminator = 1, isBarrier = 1, SchedRW = [WriteBranch] in {

let isBranch = 1, isIndirectBranch = 1 in {
def S_SETPC_B64 : SOP1_1  <"s_setpc_b64">;
} // End isBranch = 1, isIndirectBranch = 1

let isReturn = 1 in {
// Define variant marked as return rather than branch.
def S_SETPC_B64_return : SOP1_1<"", [(AMDGPUret_flag i64:$src0)]>;
}
} // End isTerminator = 1, isBarrier = 1

let isCall = 1 in {
def S_SWAPPC_B64 : SOP1_64 <"s_swappc_b64"
>;
}

def S_RFE_B64 : SOP1_1  <"s_rfe_b64">;

let hasSideEffects = 1, Uses = [EXEC], Defs = [EXEC, SCC] in {

def S_AND_SAVEEXEC_B64 : SOP1_64 <"s_and_saveexec_b64">;
def S_OR_SAVEEXEC_B64 : SOP1_64 <"s_or_saveexec_b64">;
def S_XOR_SAVEEXEC_B64 : SOP1_64 <"s_xor_saveexec_b64">;
def S_ANDN2_SAVEEXEC_B64 : SOP1_64 <"s_andn2_saveexec_b64">;
def S_ORN2_SAVEEXEC_B64 : SOP1_64 <"s_orn2_saveexec_b64">;
def S_NAND_SAVEEXEC_B64 : SOP1_64 <"s_nand_saveexec_b64">;
def S_NOR_SAVEEXEC_B64 : SOP1_64 <"s_nor_saveexec_b64">;
def S_XNOR_SAVEEXEC_B64 : SOP1_64 <"s_xnor_saveexec_b64">;

} // End hasSideEffects = 1, Uses = [EXEC], Defs = [EXEC, SCC]

def S_QUADMASK_B32 : SOP1_32 <"s_quadmask_b32">;
def S_QUADMASK_B64 : SOP1_64 <"s_quadmask_b64">;

let Uses = [M0] in {
def S_MOVRELS_B32 : SOP1_32 <"s_movrels_b32">;
def S_MOVRELS_B64 : SOP1_64 <"s_movrels_b64">;
def S_MOVRELD_B32 : SOP1_32 <"s_movreld_b32">;
def S_MOVRELD_B64 : SOP1_64 <"s_movreld_b64">;
} // End Uses = [M0]

def S_CBRANCH_JOIN : SOP1_0_32R <"s_cbranch_join">;
def S_MOV_REGRD_B32 : SOP1_32 <"s_mov_regrd_b32">;
let Defs = [SCC] in {
def S_ABS_I32 : SOP1_32 <"s_abs_i32">;
} // End Defs = [SCC]
def S_MOV_FED_B32 : SOP1_32 <"s_mov_fed_b32">;

let SubtargetPredicate = HasVGPRIndexMode in {
def S_SET_GPR_IDX_IDX : SOP1_0_32<"s_set_gpr_idx_idx"> {
  let Uses = [M0];
  let Defs = [M0];
}
}

let SubtargetPredicate = isGFX9 in {
  let hasSideEffects = 1, Defs = [EXEC, SCC], Uses = [EXEC] in {
    def S_ANDN1_SAVEEXEC_B64 : SOP1_64<"s_andn1_saveexec_b64">;
    def S_ORN1_SAVEEXEC_B64  : SOP1_64<"s_orn1_saveexec_b64">;
    def S_ANDN1_WREXEC_B64   : SOP1_64<"s_andn1_wrexec_b64">;
    def S_ANDN2_WREXEC_B64   : SOP1_64<"s_andn2_wrexec_b64">;
  } // End hasSideEffects = 1, Defs = [EXEC, SCC], Uses = [EXEC]

  def S_BITREPLICATE_B64_B32 : SOP1_64_32<"s_bitreplicate_b64_b32">;
} // End SubtargetPredicate = isGFX9

//===----------------------------------------------------------------------===//
// SOP2 Instructions
//===----------------------------------------------------------------------===//

class SOP2_Pseudo<string opName, dag outs, dag ins,
                  string asmOps, list<dag> pattern=[]> :
  SOP_Pseudo<opName, outs, ins, asmOps, pattern> {

  let mayLoad = 0;
  let mayStore = 0;
  let hasSideEffects = 0;
  let SALU = 1;
  let SOP2 = 1;
  let SchedRW = [WriteSALU];
  let UseNamedOperandTable = 1;

  let has_sdst = 1;

  // Pseudo instructions have no encodings, but adding this field here allows
  // us to do:
  // let sdst = xxx in {
  // for multiclasses that include both real and pseudo instructions.
  // field bits<7> sdst = 0;
  // let Size = 4; // Do we need size here?
}

class SOP2_Real<bits<7> op, SOP_Pseudo ps> :
  InstSI <ps.OutOperandList, ps.InOperandList,
          ps.Mnemonic # " " # ps.AsmOperands, []>,
  Enc32 {
  let isPseudo = 0;
  let isCodeGenOnly = 0;

  // copy relevant pseudo op flags
  let SubtargetPredicate = ps.SubtargetPredicate;
  let AsmMatchConverter  = ps.AsmMatchConverter;

  // encoding
  bits<7> sdst;
  bits<8> src0;
  bits<8> src1;

  let Inst{7-0}   = src0;
  let Inst{15-8}  = src1;
  let Inst{22-16} = !if(ps.has_sdst, sdst, ?);
  let Inst{29-23} = op;
  let Inst{31-30} = 0x2; // encoding
}


class SOP2_32 <string opName, list<dag> pattern=[]> : SOP2_Pseudo <
  opName, (outs SReg_32:$sdst), (ins SSrc_b32:$src0, SSrc_b32:$src1),
  "$sdst, $src0, $src1", pattern
>;

class SOP2_64 <string opName, list<dag> pattern=[]> : SOP2_Pseudo <
  opName, (outs SReg_64:$sdst), (ins SSrc_b64:$src0, SSrc_b64:$src1),
  "$sdst, $src0, $src1", pattern
>;

class SOP2_64_32 <string opName, list<dag> pattern=[]> : SOP2_Pseudo <
  opName, (outs SReg_64:$sdst), (ins SSrc_b64:$src0, SSrc_b32:$src1),
  "$sdst, $src0, $src1", pattern
>;

class SOP2_64_32_32 <string opName, list<dag> pattern=[]> : SOP2_Pseudo <
  opName, (outs SReg_64:$sdst), (ins SSrc_b32:$src0, SSrc_b32:$src1),
  "$sdst, $src0, $src1", pattern
>;

class UniformUnaryFrag<SDPatternOperator Op> : PatFrag <
  (ops node:$src0),
  (Op $src0),
  [{ return !N->isDivergent(); }]
>;

class UniformBinFrag<SDPatternOperator Op> : PatFrag <
  (ops node:$src0, node:$src1),
  (Op $src0, $src1),
  [{ return !N->isDivergent(); }]
>;

let Defs = [SCC] in { // Carry out goes to SCC
let isCommutable = 1 in {
def S_ADD_U32 : SOP2_32 <"s_add_u32">;
def S_ADD_I32 : SOP2_32 <"s_add_i32",
  [(set i32:$sdst, (UniformBinFrag<add> SSrc_b32:$src0, SSrc_b32:$src1))]
>;
} // End isCommutable = 1

def S_SUB_U32 : SOP2_32 <"s_sub_u32">;
def S_SUB_I32 : SOP2_32 <"s_sub_i32",
  [(set i32:$sdst, (UniformBinFrag<sub> SSrc_b32:$src0, SSrc_b32:$src1))]
>;

let Uses = [SCC] in { // Carry in comes from SCC
let isCommutable = 1 in {
def S_ADDC_U32 : SOP2_32 <"s_addc_u32",
  [(set i32:$sdst, (UniformBinFrag<adde> (i32 SSrc_b32:$src0), (i32 SSrc_b32:$src1)))]>;
} // End isCommutable = 1

def S_SUBB_U32 : SOP2_32 <"s_subb_u32",
  [(set i32:$sdst, (UniformBinFrag<sube> (i32 SSrc_b32:$src0), (i32 SSrc_b32:$src1)))]>;
} // End Uses = [SCC]


let isCommutable = 1 in {
def S_MIN_I32 : SOP2_32 <"s_min_i32",
  [(set i32:$sdst, (UniformBinFrag<smin> i32:$src0, i32:$src1))]
>;
def S_MIN_U32 : SOP2_32 <"s_min_u32",
  [(set i32:$sdst, (UniformBinFrag<umin> i32:$src0, i32:$src1))]
>;
def S_MAX_I32 : SOP2_32 <"s_max_i32",
  [(set i32:$sdst, (UniformBinFrag<smax> i32:$src0, i32:$src1))]
>;
def S_MAX_U32 : SOP2_32 <"s_max_u32",
  [(set i32:$sdst, (UniformBinFrag<umax> i32:$src0, i32:$src1))]
>;
} // End isCommutable = 1
} // End Defs = [SCC]


let Uses = [SCC] in {
  def S_CSELECT_B32 : SOP2_32 <"s_cselect_b32">;
  def S_CSELECT_B64 : SOP2_64 <"s_cselect_b64">;
} // End Uses = [SCC]

let Defs = [SCC] in {
let isCommutable = 1 in {
def S_AND_B32 : SOP2_32 <"s_and_b32",
  [(set i32:$sdst, (UniformBinFrag<and> i32:$src0, i32:$src1))]
>;

def S_AND_B64 : SOP2_64 <"s_and_b64",
  [(set i64:$sdst, (UniformBinFrag<and> i64:$src0, i64:$src1))]
>;

def S_OR_B32 : SOP2_32 <"s_or_b32",
  [(set i32:$sdst, (UniformBinFrag<or> i32:$src0, i32:$src1))]
>;

def S_OR_B64 : SOP2_64 <"s_or_b64",
  [(set i64:$sdst, (UniformBinFrag<or> i64:$src0, i64:$src1))]
>;

def S_XOR_B32 : SOP2_32 <"s_xor_b32",
  [(set i32:$sdst, (UniformBinFrag<xor> i32:$src0, i32:$src1))]
>;

def S_XOR_B64 : SOP2_64 <"s_xor_b64",
  [(set i64:$sdst, (UniformBinFrag<xor> i64:$src0, i64:$src1))]
>;

def S_XNOR_B32 : SOP2_32 <"s_xnor_b32",
  [(set i32:$sdst, (not (xor_oneuse i32:$src0, i32:$src1)))]
>;

def S_XNOR_B64 : SOP2_64 <"s_xnor_b64",
  [(set i64:$sdst, (not (xor_oneuse i64:$src0, i64:$src1)))]
>;

def S_NAND_B32 : SOP2_32 <"s_nand_b32",
  [(set i32:$sdst, (not (and_oneuse i32:$src0, i32:$src1)))]
>;

def S_NAND_B64 : SOP2_64 <"s_nand_b64",
  [(set i64:$sdst, (not (and_oneuse i64:$src0, i64:$src1)))]
>;

def S_NOR_B32 : SOP2_32 <"s_nor_b32",
  [(set i32:$sdst, (not (or_oneuse i32:$src0, i32:$src1)))]
>;

def S_NOR_B64 : SOP2_64 <"s_nor_b64",
  [(set i64:$sdst, (not (or_oneuse i64:$src0, i64:$src1)))]
>;
} // End isCommutable = 1

def S_ANDN2_B32 : SOP2_32 <"s_andn2_b32",
  [(set i32:$sdst, (UniformBinFrag<and> i32:$src0, (UniformUnaryFrag<not> i32:$src1)))]
>;

def S_ANDN2_B64 : SOP2_64 <"s_andn2_b64",
  [(set i64:$sdst, (UniformBinFrag<and> i64:$src0, (UniformUnaryFrag<not> i64:$src1)))]
>;

def S_ORN2_B32 : SOP2_32 <"s_orn2_b32",
  [(set i32:$sdst, (UniformBinFrag<or> i32:$src0, (UniformUnaryFrag<not> i32:$src1)))]
>;

def S_ORN2_B64 : SOP2_64 <"s_orn2_b64",
  [(set i64:$sdst, (UniformBinFrag<or> i64:$src0, (UniformUnaryFrag<not> i64:$src1)))]
>;
} // End Defs = [SCC]

// Use added complexity so these patterns are preferred to the VALU patterns.
let AddedComplexity = 1 in {

let Defs = [SCC] in {
// TODO: b64 versions require VOP3 change since v_lshlrev_b64 is VOP3
def S_LSHL_B32 : SOP2_32 <"s_lshl_b32",
  [(set i32:$sdst, (UniformBinFrag<shl> i32:$src0, i32:$src1))]
>;
def S_LSHL_B64 : SOP2_64_32 <"s_lshl_b64",
  [(set i64:$sdst, (UniformBinFrag<shl> i64:$src0, i32:$src1))]
>;
def S_LSHR_B32 : SOP2_32 <"s_lshr_b32",
  [(set i32:$sdst, (UniformBinFrag<srl> i32:$src0, i32:$src1))]
>;
def S_LSHR_B64 : SOP2_64_32 <"s_lshr_b64",
  [(set i64:$sdst, (UniformBinFrag<srl> i64:$src0, i32:$src1))]
>;
def S_ASHR_I32 : SOP2_32 <"s_ashr_i32",
  [(set i32:$sdst, (UniformBinFrag<sra> i32:$src0, i32:$src1))]
>;
def S_ASHR_I64 : SOP2_64_32 <"s_ashr_i64",
  [(set i64:$sdst, (UniformBinFrag<sra> i64:$src0, i32:$src1))]
>;
} // End Defs = [SCC]

def S_BFM_B32 : SOP2_32 <"s_bfm_b32",
  [(set i32:$sdst, (UniformBinFrag<AMDGPUbfm> i32:$src0, i32:$src1))]>;
def S_BFM_B64 : SOP2_64_32_32 <"s_bfm_b64">;

// TODO: S_MUL_I32 require V_MUL_LO_I32 from VOP3 change
def S_MUL_I32 : SOP2_32 <"s_mul_i32",
  [(set i32:$sdst, (mul i32:$src0, i32:$src1))]> {
  let isCommutable = 1;
}

} // End AddedComplexity = 1

let Defs = [SCC] in {
def S_BFE_U32 : SOP2_32 <"s_bfe_u32">;
def S_BFE_I32 : SOP2_32 <"s_bfe_i32">;
def S_BFE_U64 : SOP2_64_32 <"s_bfe_u64">;
def S_BFE_I64 : SOP2_64_32 <"s_bfe_i64">;
} // End Defs = [SCC]

def S_CBRANCH_G_FORK : SOP2_Pseudo <
  "s_cbranch_g_fork", (outs),
  (ins SCSrc_b64:$src0, SCSrc_b64:$src1),
  "$src0, $src1"
> {
  let has_sdst = 0;
}

let Defs = [SCC] in {
def S_ABSDIFF_I32 : SOP2_32 <"s_absdiff_i32">;
} // End Defs = [SCC]

let SubtargetPredicate = isVI in {
  def S_RFE_RESTORE_B64 : SOP2_Pseudo <
    "s_rfe_restore_b64", (outs),
    (ins SSrc_b64:$src0, SSrc_b32:$src1),
    "$src0, $src1"
  > {
    let hasSideEffects = 1;
    let has_sdst = 0;
  }
}

let SubtargetPredicate = isGFX9 in {
  def S_PACK_LL_B32_B16 : SOP2_32<"s_pack_ll_b32_b16">;
  def S_PACK_LH_B32_B16 : SOP2_32<"s_pack_lh_b32_b16">;
  def S_PACK_HH_B32_B16 : SOP2_32<"s_pack_hh_b32_b16">;

  let Defs = [SCC] in {
    def S_LSHL1_ADD_U32 : SOP2_32<"s_lshl1_add_u32">;
    def S_LSHL2_ADD_U32 : SOP2_32<"s_lshl2_add_u32">;
    def S_LSHL3_ADD_U32 : SOP2_32<"s_lshl3_add_u32">;
    def S_LSHL4_ADD_U32 : SOP2_32<"s_lshl4_add_u32">;
  } // End Defs = [SCC]

  def S_MUL_HI_U32 : SOP2_32<"s_mul_hi_u32">;
  def S_MUL_HI_I32 : SOP2_32<"s_mul_hi_i32">;
}

//===----------------------------------------------------------------------===//
// SOPK Instructions
//===----------------------------------------------------------------------===//

class SOPK_Pseudo <string opName, dag outs, dag ins,
                   string asmOps, list<dag> pattern=[]> :
  InstSI <outs, ins, "", pattern>,
  SIMCInstr<opName, SIEncodingFamily.NONE> {
  let isPseudo = 1;
  let isCodeGenOnly = 1;
  let SubtargetPredicate = isGCN;
  let mayLoad = 0;
  let mayStore = 0;
  let hasSideEffects = 0;
  let SALU = 1;
  let SOPK = 1;
  let SchedRW = [WriteSALU];
  let UseNamedOperandTable = 1;
  string Mnemonic = opName;
  string AsmOperands = asmOps;

  bits<1> has_sdst = 1;
}

class SOPK_Real<bits<5> op, SOPK_Pseudo ps> :
  InstSI <ps.OutOperandList, ps.InOperandList,
          ps.Mnemonic # " " # ps.AsmOperands, []> {
  let isPseudo = 0;
  let isCodeGenOnly = 0;

  // copy relevant pseudo op flags
  let SubtargetPredicate = ps.SubtargetPredicate;
  let AsmMatchConverter  = ps.AsmMatchConverter;
  let DisableEncoding    = ps.DisableEncoding;
  let Constraints        = ps.Constraints;

  // encoding
  bits<7>  sdst;
  bits<16> simm16;
  bits<32> imm;
}

class SOPK_Real32<bits<5> op, SOPK_Pseudo ps> :
  SOPK_Real <op, ps>,
  Enc32 {
  let Inst{15-0}  = simm16;
  let Inst{22-16} = !if(ps.has_sdst, sdst, ?);
  let Inst{27-23} = op;
  let Inst{31-28} = 0xb; //encoding
}

class SOPK_Real64<bits<5> op, SOPK_Pseudo ps> :
  SOPK_Real<op, ps>,
  Enc64 {
  let Inst{15-0}  = simm16;
  let Inst{22-16} = !if(ps.has_sdst, sdst, ?);
  let Inst{27-23} = op;
  let Inst{31-28} = 0xb; //encoding
  let Inst{63-32} = imm;
}

class SOPKInstTable <bit is_sopk, string cmpOp = ""> {
  bit IsSOPK = is_sopk;
  string BaseCmpOp = cmpOp;
}

class SOPK_32 <string opName, list<dag> pattern=[]> : SOPK_Pseudo <
  opName,
  (outs SReg_32:$sdst),
  (ins s16imm:$simm16),
  "$sdst, $simm16",
  pattern>;

class SOPK_SCC <string opName, string base_op, bit isSignExt> : SOPK_Pseudo <
  opName,
  (outs),
  !if(isSignExt,
      (ins SReg_32:$sdst, s16imm:$simm16),
      (ins SReg_32:$sdst, u16imm:$simm16)),
  "$sdst, $simm16", []>,
  SOPKInstTable<1, base_op>{
  let Defs = [SCC];
}

class SOPK_32TIE <string opName, list<dag> pattern=[]> : SOPK_Pseudo <
  opName,
  (outs SReg_32:$sdst),
  (ins SReg_32:$src0, s16imm:$simm16),
  "$sdst, $simm16",
  pattern
>;

let isReMaterializable = 1, isMoveImm = 1 in {
def S_MOVK_I32 : SOPK_32 <"s_movk_i32">;
} // End isReMaterializable = 1
let Uses = [SCC] in {
def S_CMOVK_I32 : SOPK_32 <"s_cmovk_i32">;
}

let isCompare = 1 in {

// This instruction is disabled for now until we can figure out how to teach
// the instruction selector to correctly use the  S_CMP* vs V_CMP*
// instructions.
//
// When this instruction is enabled the code generator sometimes produces this
// invalid sequence:
//
// SCC = S_CMPK_EQ_I32 SGPR0, imm
// VCC = COPY SCC
// VGPR0 = V_CNDMASK VCC, VGPR0, VGPR1
//
// def S_CMPK_EQ_I32 : SOPK_SCC <"s_cmpk_eq_i32",
//   [(set i1:$dst, (setcc i32:$src0, imm:$src1, SETEQ))]
// >;

def S_CMPK_EQ_I32 : SOPK_SCC <"s_cmpk_eq_i32", "s_cmp_eq_i32", 1>;
def S_CMPK_LG_I32 : SOPK_SCC <"s_cmpk_lg_i32", "s_cmp_lg_i32", 1>;
def S_CMPK_GT_I32 : SOPK_SCC <"s_cmpk_gt_i32", "s_cmp_gt_i32", 1>;
def S_CMPK_GE_I32 : SOPK_SCC <"s_cmpk_ge_i32", "s_cmp_ge_i32", 1>;
def S_CMPK_LT_I32 : SOPK_SCC <"s_cmpk_lt_i32", "s_cmp_lt_i32", 1>;
def S_CMPK_LE_I32 : SOPK_SCC <"s_cmpk_le_i32", "s_cmp_le_i32", 1>;

let SOPKZext = 1 in {
def S_CMPK_EQ_U32 : SOPK_SCC <"s_cmpk_eq_u32", "s_cmp_eq_u32", 0>;
def S_CMPK_LG_U32 : SOPK_SCC <"s_cmpk_lg_u32", "s_cmp_lg_u32", 0>;
def S_CMPK_GT_U32 : SOPK_SCC <"s_cmpk_gt_u32", "s_cmp_gt_u32", 0>;
def S_CMPK_GE_U32 : SOPK_SCC <"s_cmpk_ge_u32", "s_cmp_ge_u32", 0>;
def S_CMPK_LT_U32 : SOPK_SCC <"s_cmpk_lt_u32", "s_cmp_lt_u32", 0>;
def S_CMPK_LE_U32 : SOPK_SCC <"s_cmpk_le_u32", "s_cmp_le_u32", 0>;
} // End SOPKZext = 1
} // End isCompare = 1

let Defs = [SCC], isCommutable = 1, DisableEncoding = "$src0",
    Constraints = "$sdst = $src0" in {
  def S_ADDK_I32 : SOPK_32TIE <"s_addk_i32">;
  def S_MULK_I32 : SOPK_32TIE <"s_mulk_i32">;
}

def S_CBRANCH_I_FORK : SOPK_Pseudo <
  "s_cbranch_i_fork",
  (outs), (ins SReg_64:$sdst, s16imm:$simm16),
  "$sdst, $simm16"
>;

let mayLoad = 1 in {
def S_GETREG_B32 : SOPK_Pseudo <
  "s_getreg_b32",
  (outs SReg_32:$sdst), (ins hwreg:$simm16),
  "$sdst, $simm16"
>;
}

let hasSideEffects = 1 in {

def S_SETREG_B32 : SOPK_Pseudo <
  "s_setreg_b32",
  (outs), (ins SReg_32:$sdst, hwreg:$simm16),
  "$simm16, $sdst",
  [(AMDGPUsetreg i32:$sdst, (i16 timm:$simm16))]
>;

// FIXME: Not on SI?
//def S_GETREG_REGRD_B32 : SOPK_32 <sopk<0x14, 0x13>, "s_getreg_regrd_b32">;

def S_SETREG_IMM32_B32 : SOPK_Pseudo <
  "s_setreg_imm32_b32",
  (outs), (ins i32imm:$imm, hwreg:$simm16),
  "$simm16, $imm"> {
  let Size = 8; // Unlike every other SOPK instruction.
  let has_sdst = 0;
}

} // End hasSideEffects = 1

let SubtargetPredicate = isGFX9 in {
  def S_CALL_B64 : SOPK_Pseudo<
      "s_call_b64",
      (outs SReg_64:$sdst),
      (ins s16imm:$simm16),
      "$sdst, $simm16"> {
    let isCall = 1;
  }
}

//===----------------------------------------------------------------------===//
// SOPC Instructions
//===----------------------------------------------------------------------===//

class SOPCe <bits<7> op> : Enc32 {
  bits<8> src0;
  bits<8> src1;

  let Inst{7-0} = src0;
  let Inst{15-8} = src1;
  let Inst{22-16} = op;
  let Inst{31-23} = 0x17e;
}

class SOPC <bits<7> op, dag outs, dag ins, string asm,
            list<dag> pattern = []> :
  InstSI<outs, ins, asm, pattern>, SOPCe <op> {
  let mayLoad = 0;
  let mayStore = 0;
  let hasSideEffects = 0;
  let SALU = 1;
  let SOPC = 1;
  let isCodeGenOnly = 0;
  let Defs = [SCC];
  let SchedRW = [WriteSALU];
  let UseNamedOperandTable = 1;
  let SubtargetPredicate = isGCN;
}

class SOPC_Base <bits<7> op, RegisterOperand rc0, RegisterOperand rc1,
                 string opName, list<dag> pattern = []> : SOPC <
  op, (outs), (ins rc0:$src0, rc1:$src1),
  opName#" $src0, $src1", pattern > {
  let Defs = [SCC];
}
class SOPC_Helper <bits<7> op, RegisterOperand rc, ValueType vt,
                    string opName, PatLeaf cond> : SOPC_Base <
  op, rc, rc, opName,
  [(set SCC, (si_setcc_uniform vt:$src0, vt:$src1, cond))] > {
}

class SOPC_CMP_32<bits<7> op, string opName,
                  PatLeaf cond = COND_NULL, string revOp = opName>
  : SOPC_Helper<op, SSrc_b32, i32, opName, cond>,
    Commutable_REV<revOp, !eq(revOp, opName)>,
    SOPKInstTable<0, opName> {
  let isCompare = 1;
  let isCommutable = 1;
}

class SOPC_CMP_64<bits<7> op, string opName,
                  PatLeaf cond = COND_NULL, string revOp = opName>
  : SOPC_Helper<op, SSrc_b64, i64, opName, cond>,
    Commutable_REV<revOp, !eq(revOp, opName)> {
  let isCompare = 1;
  let isCommutable = 1;
}

class SOPC_32<bits<7> op, string opName, list<dag> pattern = []>
  : SOPC_Base<op, SSrc_b32, SSrc_b32, opName, pattern>;

class SOPC_64_32<bits<7> op, string opName, list<dag> pattern = []>
  : SOPC_Base<op, SSrc_b64, SSrc_b32, opName, pattern>;

def S_CMP_EQ_I32 : SOPC_CMP_32 <0x00, "s_cmp_eq_i32">;
def S_CMP_LG_I32 : SOPC_CMP_32 <0x01, "s_cmp_lg_i32">;
def S_CMP_GT_I32 : SOPC_CMP_32 <0x02, "s_cmp_gt_i32", COND_SGT>;
def S_CMP_GE_I32 : SOPC_CMP_32 <0x03, "s_cmp_ge_i32", COND_SGE>;
def S_CMP_LT_I32 : SOPC_CMP_32 <0x04, "s_cmp_lt_i32", COND_SLT, "s_cmp_gt_i32">;
def S_CMP_LE_I32 : SOPC_CMP_32 <0x05, "s_cmp_le_i32", COND_SLE, "s_cmp_ge_i32">;
def S_CMP_EQ_U32 : SOPC_CMP_32 <0x06, "s_cmp_eq_u32", COND_EQ>;
def S_CMP_LG_U32 : SOPC_CMP_32 <0x07, "s_cmp_lg_u32", COND_NE>;
def S_CMP_GT_U32 : SOPC_CMP_32 <0x08, "s_cmp_gt_u32", COND_UGT>;
def S_CMP_GE_U32 : SOPC_CMP_32 <0x09, "s_cmp_ge_u32", COND_UGE>;
def S_CMP_LT_U32 : SOPC_CMP_32 <0x0a, "s_cmp_lt_u32", COND_ULT, "s_cmp_gt_u32">;
def S_CMP_LE_U32 : SOPC_CMP_32 <0x0b, "s_cmp_le_u32", COND_ULE, "s_cmp_ge_u32">;

def S_BITCMP0_B32 : SOPC_32 <0x0c, "s_bitcmp0_b32">;
def S_BITCMP1_B32 : SOPC_32 <0x0d, "s_bitcmp1_b32">;
def S_BITCMP0_B64 : SOPC_64_32 <0x0e, "s_bitcmp0_b64">;
def S_BITCMP1_B64 : SOPC_64_32 <0x0f, "s_bitcmp1_b64">;
def S_SETVSKIP : SOPC_32 <0x10, "s_setvskip">;

let SubtargetPredicate = isVI in {
def S_CMP_EQ_U64 : SOPC_CMP_64 <0x12, "s_cmp_eq_u64", COND_EQ>;
def S_CMP_LG_U64 : SOPC_CMP_64 <0x13, "s_cmp_lg_u64", COND_NE>;
}

let SubtargetPredicate = HasVGPRIndexMode in {
def S_SET_GPR_IDX_ON : SOPC <0x11,
  (outs),
  (ins SSrc_b32:$src0, GPRIdxMode:$src1),
  "s_set_gpr_idx_on $src0,$src1"> {
  let Defs = [M0]; // No scc def
  let Uses = [M0]; // Other bits of m0 unmodified.
  let hasSideEffects = 1; // Sets mode.gpr_idx_en
  let FixedSize = 1;
}
}

//===----------------------------------------------------------------------===//
// SOPP Instructions
//===----------------------------------------------------------------------===//

class SOPPe <bits<7> op> : Enc32 {
  bits <16> simm16;

  let Inst{15-0} = simm16;
  let Inst{22-16} = op;
  let Inst{31-23} = 0x17f; // encoding
}

class SOPP <bits<7> op, dag ins, string asm, list<dag> pattern = []> :
  InstSI <(outs), ins, asm, pattern >, SOPPe <op> {

  let mayLoad = 0;
  let mayStore = 0;
  let hasSideEffects = 0;
  let SALU = 1;
  let SOPP = 1;
  let Size = 4;
  let SchedRW = [WriteSALU];

  let UseNamedOperandTable = 1;
  let SubtargetPredicate = isGCN;
}


def S_NOP : SOPP <0x00000000, (ins i16imm:$simm16), "s_nop $simm16">;

let isTerminator = 1 in {

def S_ENDPGM : SOPP <0x00000001, (ins), "s_endpgm",
  [(AMDGPUendpgm)]> {
  let simm16 = 0;
  let isBarrier = 1;
  let isReturn = 1;
}

let SubtargetPredicate = isVI in {
def S_ENDPGM_SAVED : SOPP <0x0000001B, (ins), "s_endpgm_saved"> {
  let simm16 = 0;
  let isBarrier = 1;
  let isReturn = 1;
}
}

let SubtargetPredicate = isGFX9 in {
  let isBarrier = 1, isReturn = 1, simm16 = 0 in {
    def S_ENDPGM_ORDERED_PS_DONE :
      SOPP<0x01e, (ins), "s_endpgm_ordered_ps_done">;
  } // End isBarrier = 1, isReturn = 1, simm16 = 0
} // End SubtargetPredicate = isGFX9

let isBranch = 1, SchedRW = [WriteBranch] in {
def S_BRANCH : SOPP <
  0x00000002, (ins sopp_brtarget:$simm16), "s_branch $simm16",
  [(br bb:$simm16)]> {
  let isBarrier = 1;
}

let Uses = [SCC] in {
def S_CBRANCH_SCC0 : SOPP <
  0x00000004, (ins sopp_brtarget:$simm16),
  "s_cbranch_scc0 $simm16"
>;
def S_CBRANCH_SCC1 : SOPP <
  0x00000005, (ins sopp_brtarget:$simm16),
  "s_cbranch_scc1 $simm16"
>;
} // End Uses = [SCC]

let Uses = [VCC] in {
def S_CBRANCH_VCCZ : SOPP <
  0x00000006, (ins sopp_brtarget:$simm16),
  "s_cbranch_vccz $simm16"
>;
def S_CBRANCH_VCCNZ : SOPP <
  0x00000007, (ins sopp_brtarget:$simm16),
  "s_cbranch_vccnz $simm16"
>;
} // End Uses = [VCC]

let Uses = [EXEC] in {
def S_CBRANCH_EXECZ : SOPP <
  0x00000008, (ins sopp_brtarget:$simm16),
  "s_cbranch_execz $simm16"
>;
def S_CBRANCH_EXECNZ : SOPP <
  0x00000009, (ins sopp_brtarget:$simm16),
  "s_cbranch_execnz $simm16"
>;
} // End Uses = [EXEC]

def S_CBRANCH_CDBGSYS : SOPP <
  0x00000017, (ins sopp_brtarget:$simm16),
  "s_cbranch_cdbgsys $simm16"
>;

def S_CBRANCH_CDBGSYS_AND_USER : SOPP <
  0x0000001A, (ins sopp_brtarget:$simm16),
  "s_cbranch_cdbgsys_and_user $simm16"
>;

def S_CBRANCH_CDBGSYS_OR_USER : SOPP <
  0x00000019, (ins sopp_brtarget:$simm16),
  "s_cbranch_cdbgsys_or_user $simm16"
>;

def S_CBRANCH_CDBGUSER : SOPP <
  0x00000018, (ins sopp_brtarget:$simm16),
  "s_cbranch_cdbguser $simm16"
>;

} // End isBranch = 1
} // End isTerminator = 1

let hasSideEffects = 1 in {
def S_BARRIER : SOPP <0x0000000a, (ins), "s_barrier",
  [(int_amdgcn_s_barrier)]> {
  let SchedRW = [WriteBarrier];
  let simm16 = 0;
  let mayLoad = 1;
  let mayStore = 1;
  let isConvergent = 1;
}

let SubtargetPredicate = isVI in {
def S_WAKEUP : SOPP <0x00000003, (ins), "s_wakeup"> {
  let simm16 = 0;
  let mayLoad = 1;
  let mayStore = 1;
}
}

let mayLoad = 1, mayStore = 1, hasSideEffects = 1 in
def S_WAITCNT : SOPP <0x0000000c, (ins WAIT_FLAG:$simm16), "s_waitcnt $simm16">;
def S_SETHALT : SOPP <0x0000000d, (ins i16imm:$simm16), "s_sethalt $simm16">;
def S_SETKILL : SOPP <0x0000000b, (ins i16imm:$simm16), "s_setkill $simm16">;

// On SI the documentation says sleep for approximately 64 * low 2
// bits, consistent with the reported maximum of 448. On VI the
// maximum reported is 960 cycles, so 960 / 64 = 15 max, so is the
// maximum really 15 on VI?
def S_SLEEP : SOPP <0x0000000e, (ins i32imm:$simm16),
  "s_sleep $simm16", [(int_amdgcn_s_sleep SIMM16bit:$simm16)]> {
  let hasSideEffects = 1;
  let mayLoad = 1;
  let mayStore = 1;
}

def S_SETPRIO : SOPP <0x0000000f, (ins i16imm:$simm16), "s_setprio $simm16">;

let Uses = [EXEC, M0] in {
// FIXME: Should this be mayLoad+mayStore?
def S_SENDMSG : SOPP <0x00000010, (ins SendMsgImm:$simm16), "s_sendmsg $simm16",
  [(AMDGPUsendmsg (i32 imm:$simm16))]
>;

def S_SENDMSGHALT : SOPP <0x00000011, (ins SendMsgImm:$simm16), "s_sendmsghalt $simm16",
  [(AMDGPUsendmsghalt (i32 imm:$simm16))]
>;
} // End Uses = [EXEC, M0]

def S_TRAP : SOPP <0x00000012, (ins i16imm:$simm16), "s_trap $simm16">;
def S_ICACHE_INV : SOPP <0x00000013, (ins), "s_icache_inv"> {
  let simm16 = 0;
}
def S_INCPERFLEVEL : SOPP <0x00000014, (ins i32imm:$simm16), "s_incperflevel $simm16",
  [(int_amdgcn_s_incperflevel SIMM16bit:$simm16)]> {
  let hasSideEffects = 1;
  let mayLoad = 1;
  let mayStore = 1;
}
def S_DECPERFLEVEL : SOPP <0x00000015, (ins i32imm:$simm16), "s_decperflevel $simm16",
  [(int_amdgcn_s_decperflevel SIMM16bit:$simm16)]> {
  let hasSideEffects = 1;
  let mayLoad = 1;
  let mayStore = 1;
}
def S_TTRACEDATA : SOPP <0x00000016, (ins), "s_ttracedata"> {
  let simm16 = 0;
}

let SubtargetPredicate = HasVGPRIndexMode in {
def S_SET_GPR_IDX_OFF : SOPP<0x1c, (ins), "s_set_gpr_idx_off"> {
  let simm16 = 0;
}
}
} // End hasSideEffects

let SubtargetPredicate = HasVGPRIndexMode in {
def S_SET_GPR_IDX_MODE : SOPP<0x1d, (ins GPRIdxMode:$simm16),
  "s_set_gpr_idx_mode$simm16"> {
  let Defs = [M0];
}
}

//===----------------------------------------------------------------------===//
// S_GETREG_B32 Intrinsic Pattern.
//===----------------------------------------------------------------------===//
def : GCNPat <
  (int_amdgcn_s_getreg imm:$simm16),
  (S_GETREG_B32 (as_i16imm $simm16))
>;

//===----------------------------------------------------------------------===//
// SOP1 Patterns
//===----------------------------------------------------------------------===//

def : GCNPat <
  (i64 (ctpop i64:$src)),
    (i64 (REG_SEQUENCE SReg_64,
     (i32 (COPY_TO_REGCLASS (S_BCNT1_I32_B64 $src), SReg_32)), sub0,
     (S_MOV_B32 (i32 0)), sub1))
>;

def : GCNPat <
  (i32 (smax i32:$x, (i32 (ineg i32:$x)))),
  (S_ABS_I32 $x)
>;

def : GCNPat <
  (i16 imm:$imm),
  (S_MOV_B32 imm:$imm)
>;

// Same as a 32-bit inreg
def : GCNPat<
  (i32 (sext i16:$src)),
  (S_SEXT_I32_I16 $src)
>;


//===----------------------------------------------------------------------===//
// SOP2 Patterns
//===----------------------------------------------------------------------===//

// V_ADD_I32_e32/S_ADD_U32 produces carry in VCC/SCC. For the vector
// case, the sgpr-copies pass will fix this to use the vector version.
def : GCNPat <
  (i32 (addc i32:$src0, i32:$src1)),
  (S_ADD_U32 $src0, $src1)
>;

// FIXME: We need to use COPY_TO_REGCLASS to work-around the fact that
// REG_SEQUENCE patterns don't support instructions with multiple
// outputs.
def : GCNPat<
  (i64 (zext i16:$src)),
    (REG_SEQUENCE SReg_64,
      (i32 (COPY_TO_REGCLASS (S_AND_B32 $src, (S_MOV_B32 (i32 0xffff))), SGPR_32)), sub0,
      (S_MOV_B32 (i32 0)), sub1)
>;

def : GCNPat <
  (i64 (sext i16:$src)),
    (REG_SEQUENCE SReg_64, (i32 (S_SEXT_I32_I16 $src)), sub0,
    (i32 (COPY_TO_REGCLASS (S_ASHR_I32 (i32 (S_SEXT_I32_I16 $src)), (S_MOV_B32 (i32 31))), SGPR_32)), sub1)
>;

def : GCNPat<
  (i32 (zext i16:$src)),
  (S_AND_B32 (S_MOV_B32 (i32 0xffff)), $src)
>;



//===----------------------------------------------------------------------===//
// SOPP Patterns
//===----------------------------------------------------------------------===//

def : GCNPat <
  (int_amdgcn_s_waitcnt i32:$simm16),
  (S_WAITCNT (as_i16imm $simm16))
>;


//===----------------------------------------------------------------------===//
// Real target instructions, move this to the appropriate subtarget TD file
//===----------------------------------------------------------------------===//

class Select_si<string opName> :
  SIMCInstr<opName, SIEncodingFamily.SI> {
  list<Predicate> AssemblerPredicates = [isSICI];
  string DecoderNamespace = "SICI";
}

class SOP1_Real_si<bits<8> op, SOP1_Pseudo ps> :
  SOP1_Real<op, ps>,
  Select_si<ps.Mnemonic>;

class SOP2_Real_si<bits<7> op, SOP2_Pseudo ps> :
  SOP2_Real<op, ps>,
  Select_si<ps.Mnemonic>;

class SOPK_Real_si<bits<5> op, SOPK_Pseudo ps> :
  SOPK_Real32<op, ps>,
  Select_si<ps.Mnemonic>;

def S_MOV_B32_si           : SOP1_Real_si <0x03, S_MOV_B32>;
def S_MOV_B64_si           : SOP1_Real_si <0x04, S_MOV_B64>;
def S_CMOV_B32_si          : SOP1_Real_si <0x05, S_CMOV_B32>;
def S_CMOV_B64_si          : SOP1_Real_si <0x06, S_CMOV_B64>;
def S_NOT_B32_si           : SOP1_Real_si <0x07, S_NOT_B32>;
def S_NOT_B64_si           : SOP1_Real_si <0x08, S_NOT_B64>;
def S_WQM_B32_si           : SOP1_Real_si <0x09, S_WQM_B32>;
def S_WQM_B64_si           : SOP1_Real_si <0x0a, S_WQM_B64>;
def S_BREV_B32_si          : SOP1_Real_si <0x0b, S_BREV_B32>;
def S_BREV_B64_si          : SOP1_Real_si <0x0c, S_BREV_B64>;
def S_BCNT0_I32_B32_si     : SOP1_Real_si <0x0d, S_BCNT0_I32_B32>;
def S_BCNT0_I32_B64_si     : SOP1_Real_si <0x0e, S_BCNT0_I32_B64>;
def S_BCNT1_I32_B32_si     : SOP1_Real_si <0x0f, S_BCNT1_I32_B32>;
def S_BCNT1_I32_B64_si     : SOP1_Real_si <0x10, S_BCNT1_I32_B64>;
def S_FF0_I32_B32_si       : SOP1_Real_si <0x11, S_FF0_I32_B32>;
def S_FF0_I32_B64_si       : SOP1_Real_si <0x12, S_FF0_I32_B64>;
def S_FF1_I32_B32_si       : SOP1_Real_si <0x13, S_FF1_I32_B32>;
def S_FF1_I32_B64_si       : SOP1_Real_si <0x14, S_FF1_I32_B64>;
def S_FLBIT_I32_B32_si     : SOP1_Real_si <0x15, S_FLBIT_I32_B32>;
def S_FLBIT_I32_B64_si     : SOP1_Real_si <0x16, S_FLBIT_I32_B64>;
def S_FLBIT_I32_si         : SOP1_Real_si <0x17, S_FLBIT_I32>;
def S_FLBIT_I32_I64_si     : SOP1_Real_si <0x18, S_FLBIT_I32_I64>;
def S_SEXT_I32_I8_si       : SOP1_Real_si <0x19, S_SEXT_I32_I8>;
def S_SEXT_I32_I16_si      : SOP1_Real_si <0x1a, S_SEXT_I32_I16>;
def S_BITSET0_B32_si       : SOP1_Real_si <0x1b, S_BITSET0_B32>;
def S_BITSET0_B64_si       : SOP1_Real_si <0x1c, S_BITSET0_B64>;
def S_BITSET1_B32_si       : SOP1_Real_si <0x1d, S_BITSET1_B32>;
def S_BITSET1_B64_si       : SOP1_Real_si <0x1e, S_BITSET1_B64>;
def S_GETPC_B64_si         : SOP1_Real_si <0x1f, S_GETPC_B64>;
def S_SETPC_B64_si         : SOP1_Real_si <0x20, S_SETPC_B64>;
def S_SWAPPC_B64_si        : SOP1_Real_si <0x21, S_SWAPPC_B64>;
def S_RFE_B64_si           : SOP1_Real_si <0x22, S_RFE_B64>;
def S_AND_SAVEEXEC_B64_si  : SOP1_Real_si <0x24, S_AND_SAVEEXEC_B64>;
def S_OR_SAVEEXEC_B64_si   : SOP1_Real_si <0x25, S_OR_SAVEEXEC_B64>;
def S_XOR_SAVEEXEC_B64_si  : SOP1_Real_si <0x26, S_XOR_SAVEEXEC_B64>;
def S_ANDN2_SAVEEXEC_B64_si: SOP1_Real_si <0x27, S_ANDN2_SAVEEXEC_B64>;
def S_ORN2_SAVEEXEC_B64_si : SOP1_Real_si <0x28, S_ORN2_SAVEEXEC_B64>;
def S_NAND_SAVEEXEC_B64_si : SOP1_Real_si <0x29, S_NAND_SAVEEXEC_B64>;
def S_NOR_SAVEEXEC_B64_si  : SOP1_Real_si <0x2a, S_NOR_SAVEEXEC_B64>;
def S_XNOR_SAVEEXEC_B64_si : SOP1_Real_si <0x2b, S_XNOR_SAVEEXEC_B64>;
def S_QUADMASK_B32_si      : SOP1_Real_si <0x2c, S_QUADMASK_B32>;
def S_QUADMASK_B64_si      : SOP1_Real_si <0x2d, S_QUADMASK_B64>;
def S_MOVRELS_B32_si       : SOP1_Real_si <0x2e, S_MOVRELS_B32>;
def S_MOVRELS_B64_si       : SOP1_Real_si <0x2f, S_MOVRELS_B64>;
def S_MOVRELD_B32_si       : SOP1_Real_si <0x30, S_MOVRELD_B32>;
def S_MOVRELD_B64_si       : SOP1_Real_si <0x31, S_MOVRELD_B64>;
def S_CBRANCH_JOIN_si      : SOP1_Real_si <0x32, S_CBRANCH_JOIN>;
def S_MOV_REGRD_B32_si     : SOP1_Real_si <0x33, S_MOV_REGRD_B32>;
def S_ABS_I32_si           : SOP1_Real_si <0x34, S_ABS_I32>;
def S_MOV_FED_B32_si       : SOP1_Real_si <0x35, S_MOV_FED_B32>;

def S_ADD_U32_si           : SOP2_Real_si <0x00, S_ADD_U32>;
def S_ADD_I32_si           : SOP2_Real_si <0x02, S_ADD_I32>;
def S_SUB_U32_si           : SOP2_Real_si <0x01, S_SUB_U32>;
def S_SUB_I32_si           : SOP2_Real_si <0x03, S_SUB_I32>;
def S_ADDC_U32_si          : SOP2_Real_si <0x04, S_ADDC_U32>;
def S_SUBB_U32_si          : SOP2_Real_si <0x05, S_SUBB_U32>;
def S_MIN_I32_si           : SOP2_Real_si <0x06, S_MIN_I32>;
def S_MIN_U32_si           : SOP2_Real_si <0x07, S_MIN_U32>;
def S_MAX_I32_si           : SOP2_Real_si <0x08, S_MAX_I32>;
def S_MAX_U32_si           : SOP2_Real_si <0x09, S_MAX_U32>;
def S_CSELECT_B32_si       : SOP2_Real_si <0x0a, S_CSELECT_B32>;
def S_CSELECT_B64_si       : SOP2_Real_si <0x0b, S_CSELECT_B64>;
def S_AND_B32_si           : SOP2_Real_si <0x0e, S_AND_B32>;
def S_AND_B64_si           : SOP2_Real_si <0x0f, S_AND_B64>;
def S_OR_B32_si            : SOP2_Real_si <0x10, S_OR_B32>;
def S_OR_B64_si            : SOP2_Real_si <0x11, S_OR_B64>;
def S_XOR_B32_si           : SOP2_Real_si <0x12, S_XOR_B32>;
def S_XOR_B64_si           : SOP2_Real_si <0x13, S_XOR_B64>;
def S_ANDN2_B32_si         : SOP2_Real_si <0x14, S_ANDN2_B32>;
def S_ANDN2_B64_si         : SOP2_Real_si <0x15, S_ANDN2_B64>;
def S_ORN2_B32_si          : SOP2_Real_si <0x16, S_ORN2_B32>;
def S_ORN2_B64_si          : SOP2_Real_si <0x17, S_ORN2_B64>;
def S_NAND_B32_si          : SOP2_Real_si <0x18, S_NAND_B32>;
def S_NAND_B64_si          : SOP2_Real_si <0x19, S_NAND_B64>;
def S_NOR_B32_si           : SOP2_Real_si <0x1a, S_NOR_B32>;
def S_NOR_B64_si           : SOP2_Real_si <0x1b, S_NOR_B64>;
def S_XNOR_B32_si          : SOP2_Real_si <0x1c, S_XNOR_B32>;
def S_XNOR_B64_si          : SOP2_Real_si <0x1d, S_XNOR_B64>;
def S_LSHL_B32_si          : SOP2_Real_si <0x1e, S_LSHL_B32>;
def S_LSHL_B64_si          : SOP2_Real_si <0x1f, S_LSHL_B64>;
def S_LSHR_B32_si          : SOP2_Real_si <0x20, S_LSHR_B32>;
def S_LSHR_B64_si          : SOP2_Real_si <0x21, S_LSHR_B64>;
def S_ASHR_I32_si          : SOP2_Real_si <0x22, S_ASHR_I32>;
def S_ASHR_I64_si          : SOP2_Real_si <0x23, S_ASHR_I64>;
def S_BFM_B32_si           : SOP2_Real_si <0x24, S_BFM_B32>;
def S_BFM_B64_si           : SOP2_Real_si <0x25, S_BFM_B64>;
def S_MUL_I32_si           : SOP2_Real_si <0x26, S_MUL_I32>;
def S_BFE_U32_si           : SOP2_Real_si <0x27, S_BFE_U32>;
def S_BFE_I32_si           : SOP2_Real_si <0x28, S_BFE_I32>;
def S_BFE_U64_si           : SOP2_Real_si <0x29, S_BFE_U64>;
def S_BFE_I64_si           : SOP2_Real_si <0x2a, S_BFE_I64>;
def S_CBRANCH_G_FORK_si    : SOP2_Real_si <0x2b, S_CBRANCH_G_FORK>;
def S_ABSDIFF_I32_si       : SOP2_Real_si <0x2c, S_ABSDIFF_I32>;

def S_MOVK_I32_si          : SOPK_Real_si <0x00, S_MOVK_I32>;
def S_CMOVK_I32_si         : SOPK_Real_si <0x02, S_CMOVK_I32>;
def S_CMPK_EQ_I32_si       : SOPK_Real_si <0x03, S_CMPK_EQ_I32>;
def S_CMPK_LG_I32_si       : SOPK_Real_si <0x04, S_CMPK_LG_I32>;
def S_CMPK_GT_I32_si       : SOPK_Real_si <0x05, S_CMPK_GT_I32>;
def S_CMPK_GE_I32_si       : SOPK_Real_si <0x06, S_CMPK_GE_I32>;
def S_CMPK_LT_I32_si       : SOPK_Real_si <0x07, S_CMPK_LT_I32>;
def S_CMPK_LE_I32_si       : SOPK_Real_si <0x08, S_CMPK_LE_I32>;
def S_CMPK_EQ_U32_si       : SOPK_Real_si <0x09, S_CMPK_EQ_U32>;
def S_CMPK_LG_U32_si       : SOPK_Real_si <0x0a, S_CMPK_LG_U32>;
def S_CMPK_GT_U32_si       : SOPK_Real_si <0x0b, S_CMPK_GT_U32>;
def S_CMPK_GE_U32_si       : SOPK_Real_si <0x0c, S_CMPK_GE_U32>;
def S_CMPK_LT_U32_si       : SOPK_Real_si <0x0d, S_CMPK_LT_U32>;
def S_CMPK_LE_U32_si       : SOPK_Real_si <0x0e, S_CMPK_LE_U32>;
def S_ADDK_I32_si          : SOPK_Real_si <0x0f, S_ADDK_I32>;
def S_MULK_I32_si          : SOPK_Real_si <0x10, S_MULK_I32>;
def S_CBRANCH_I_FORK_si    : SOPK_Real_si <0x11, S_CBRANCH_I_FORK>;
def S_GETREG_B32_si        : SOPK_Real_si <0x12, S_GETREG_B32>;
def S_SETREG_B32_si        : SOPK_Real_si <0x13, S_SETREG_B32>;
//def S_GETREG_REGRD_B32_si  : SOPK_Real_si <0x14, S_GETREG_REGRD_B32>; // see pseudo for comments
def S_SETREG_IMM32_B32_si  : SOPK_Real64<0x15, S_SETREG_IMM32_B32>,
                             Select_si<S_SETREG_IMM32_B32.Mnemonic>;


class Select_vi<string opName> :
  SIMCInstr<opName, SIEncodingFamily.VI> {
  list<Predicate> AssemblerPredicates = [isVI];
  string DecoderNamespace = "VI";
}

class SOP1_Real_vi<bits<8> op, SOP1_Pseudo ps> :
  SOP1_Real<op, ps>,
  Select_vi<ps.Mnemonic>;


class SOP2_Real_vi<bits<7> op, SOP2_Pseudo ps> :
  SOP2_Real<op, ps>,
  Select_vi<ps.Mnemonic>;

class SOPK_Real_vi<bits<5> op, SOPK_Pseudo ps> :
  SOPK_Real32<op, ps>,
  Select_vi<ps.Mnemonic>;

def S_MOV_B32_vi           : SOP1_Real_vi <0x00, S_MOV_B32>;
def S_MOV_B64_vi           : SOP1_Real_vi <0x01, S_MOV_B64>;
def S_CMOV_B32_vi          : SOP1_Real_vi <0x02, S_CMOV_B32>;
def S_CMOV_B64_vi          : SOP1_Real_vi <0x03, S_CMOV_B64>;
def S_NOT_B32_vi           : SOP1_Real_vi <0x04, S_NOT_B32>;
def S_NOT_B64_vi           : SOP1_Real_vi <0x05, S_NOT_B64>;
def S_WQM_B32_vi           : SOP1_Real_vi <0x06, S_WQM_B32>;
def S_WQM_B64_vi           : SOP1_Real_vi <0x07, S_WQM_B64>;
def S_BREV_B32_vi          : SOP1_Real_vi <0x08, S_BREV_B32>;
def S_BREV_B64_vi          : SOP1_Real_vi <0x09, S_BREV_B64>;
def S_BCNT0_I32_B32_vi     : SOP1_Real_vi <0x0a, S_BCNT0_I32_B32>;
def S_BCNT0_I32_B64_vi     : SOP1_Real_vi <0x0b, S_BCNT0_I32_B64>;
def S_BCNT1_I32_B32_vi     : SOP1_Real_vi <0x0c, S_BCNT1_I32_B32>;
def S_BCNT1_I32_B64_vi     : SOP1_Real_vi <0x0d, S_BCNT1_I32_B64>;
def S_FF0_I32_B32_vi       : SOP1_Real_vi <0x0e, S_FF0_I32_B32>;
def S_FF0_I32_B64_vi       : SOP1_Real_vi <0x0f, S_FF0_I32_B64>;
def S_FF1_I32_B32_vi       : SOP1_Real_vi <0x10, S_FF1_I32_B32>;
def S_FF1_I32_B64_vi       : SOP1_Real_vi <0x11, S_FF1_I32_B64>;
def S_FLBIT_I32_B32_vi     : SOP1_Real_vi <0x12, S_FLBIT_I32_B32>;
def S_FLBIT_I32_B64_vi     : SOP1_Real_vi <0x13, S_FLBIT_I32_B64>;
def S_FLBIT_I32_vi         : SOP1_Real_vi <0x14, S_FLBIT_I32>;
def S_FLBIT_I32_I64_vi     : SOP1_Real_vi <0x15, S_FLBIT_I32_I64>;
def S_SEXT_I32_I8_vi       : SOP1_Real_vi <0x16, S_SEXT_I32_I8>;
def S_SEXT_I32_I16_vi      : SOP1_Real_vi <0x17, S_SEXT_I32_I16>;
def S_BITSET0_B32_vi       : SOP1_Real_vi <0x18, S_BITSET0_B32>;
def S_BITSET0_B64_vi       : SOP1_Real_vi <0x19, S_BITSET0_B64>;
def S_BITSET1_B32_vi       : SOP1_Real_vi <0x1a, S_BITSET1_B32>;
def S_BITSET1_B64_vi       : SOP1_Real_vi <0x1b, S_BITSET1_B64>;
def S_GETPC_B64_vi         : SOP1_Real_vi <0x1c, S_GETPC_B64>;
def S_SETPC_B64_vi         : SOP1_Real_vi <0x1d, S_SETPC_B64>;
def S_SWAPPC_B64_vi        : SOP1_Real_vi <0x1e, S_SWAPPC_B64>;
def S_RFE_B64_vi           : SOP1_Real_vi <0x1f, S_RFE_B64>;
def S_AND_SAVEEXEC_B64_vi  : SOP1_Real_vi <0x20, S_AND_SAVEEXEC_B64>;
def S_OR_SAVEEXEC_B64_vi   : SOP1_Real_vi <0x21, S_OR_SAVEEXEC_B64>;
def S_XOR_SAVEEXEC_B64_vi  : SOP1_Real_vi <0x22, S_XOR_SAVEEXEC_B64>;
def S_ANDN2_SAVEEXEC_B64_vi: SOP1_Real_vi <0x23, S_ANDN2_SAVEEXEC_B64>;
def S_ORN2_SAVEEXEC_B64_vi : SOP1_Real_vi <0x24, S_ORN2_SAVEEXEC_B64>;
def S_NAND_SAVEEXEC_B64_vi : SOP1_Real_vi <0x25, S_NAND_SAVEEXEC_B64>;
def S_NOR_SAVEEXEC_B64_vi  : SOP1_Real_vi <0x26, S_NOR_SAVEEXEC_B64>;
def S_XNOR_SAVEEXEC_B64_vi : SOP1_Real_vi <0x27, S_XNOR_SAVEEXEC_B64>;
def S_QUADMASK_B32_vi      : SOP1_Real_vi <0x28, S_QUADMASK_B32>;
def S_QUADMASK_B64_vi      : SOP1_Real_vi <0x29, S_QUADMASK_B64>;
def S_MOVRELS_B32_vi       : SOP1_Real_vi <0x2a, S_MOVRELS_B32>;
def S_MOVRELS_B64_vi       : SOP1_Real_vi <0x2b, S_MOVRELS_B64>;
def S_MOVRELD_B32_vi       : SOP1_Real_vi <0x2c, S_MOVRELD_B32>;
def S_MOVRELD_B64_vi       : SOP1_Real_vi <0x2d, S_MOVRELD_B64>;
def S_CBRANCH_JOIN_vi      : SOP1_Real_vi <0x2e, S_CBRANCH_JOIN>;
def S_MOV_REGRD_B32_vi     : SOP1_Real_vi <0x2f, S_MOV_REGRD_B32>;
def S_ABS_I32_vi           : SOP1_Real_vi <0x30, S_ABS_I32>;
def S_MOV_FED_B32_vi       : SOP1_Real_vi <0x31, S_MOV_FED_B32>;
def S_SET_GPR_IDX_IDX_vi   : SOP1_Real_vi <0x32, S_SET_GPR_IDX_IDX>;

def S_ADD_U32_vi           : SOP2_Real_vi <0x00, S_ADD_U32>;
def S_ADD_I32_vi           : SOP2_Real_vi <0x02, S_ADD_I32>;
def S_SUB_U32_vi           : SOP2_Real_vi <0x01, S_SUB_U32>;
def S_SUB_I32_vi           : SOP2_Real_vi <0x03, S_SUB_I32>;
def S_ADDC_U32_vi          : SOP2_Real_vi <0x04, S_ADDC_U32>;
def S_SUBB_U32_vi          : SOP2_Real_vi <0x05, S_SUBB_U32>;
def S_MIN_I32_vi           : SOP2_Real_vi <0x06, S_MIN_I32>;
def S_MIN_U32_vi           : SOP2_Real_vi <0x07, S_MIN_U32>;
def S_MAX_I32_vi           : SOP2_Real_vi <0x08, S_MAX_I32>;
def S_MAX_U32_vi           : SOP2_Real_vi <0x09, S_MAX_U32>;
def S_CSELECT_B32_vi       : SOP2_Real_vi <0x0a, S_CSELECT_B32>;
def S_CSELECT_B64_vi       : SOP2_Real_vi <0x0b, S_CSELECT_B64>;
def S_AND_B32_vi           : SOP2_Real_vi <0x0c, S_AND_B32>;
def S_AND_B64_vi           : SOP2_Real_vi <0x0d, S_AND_B64>;
def S_OR_B32_vi            : SOP2_Real_vi <0x0e, S_OR_B32>;
def S_OR_B64_vi            : SOP2_Real_vi <0x0f, S_OR_B64>;
def S_XOR_B32_vi           : SOP2_Real_vi <0x10, S_XOR_B32>;
def S_XOR_B64_vi           : SOP2_Real_vi <0x11, S_XOR_B64>;
def S_ANDN2_B32_vi         : SOP2_Real_vi <0x12, S_ANDN2_B32>;
def S_ANDN2_B64_vi         : SOP2_Real_vi <0x13, S_ANDN2_B64>;
def S_ORN2_B32_vi          : SOP2_Real_vi <0x14, S_ORN2_B32>;
def S_ORN2_B64_vi          : SOP2_Real_vi <0x15, S_ORN2_B64>;
def S_NAND_B32_vi          : SOP2_Real_vi <0x16, S_NAND_B32>;
def S_NAND_B64_vi          : SOP2_Real_vi <0x17, S_NAND_B64>;
def S_NOR_B32_vi           : SOP2_Real_vi <0x18, S_NOR_B32>;
def S_NOR_B64_vi           : SOP2_Real_vi <0x19, S_NOR_B64>;
def S_XNOR_B32_vi          : SOP2_Real_vi <0x1a, S_XNOR_B32>;
def S_XNOR_B64_vi          : SOP2_Real_vi <0x1b, S_XNOR_B64>;
def S_LSHL_B32_vi          : SOP2_Real_vi <0x1c, S_LSHL_B32>;
def S_LSHL_B64_vi          : SOP2_Real_vi <0x1d, S_LSHL_B64>;
def S_LSHR_B32_vi          : SOP2_Real_vi <0x1e, S_LSHR_B32>;
def S_LSHR_B64_vi          : SOP2_Real_vi <0x1f, S_LSHR_B64>;
def S_ASHR_I32_vi          : SOP2_Real_vi <0x20, S_ASHR_I32>;
def S_ASHR_I64_vi          : SOP2_Real_vi <0x21, S_ASHR_I64>;
def S_BFM_B32_vi           : SOP2_Real_vi <0x22, S_BFM_B32>;
def S_BFM_B64_vi           : SOP2_Real_vi <0x23, S_BFM_B64>;
def S_MUL_I32_vi           : SOP2_Real_vi <0x24, S_MUL_I32>;
def S_BFE_U32_vi           : SOP2_Real_vi <0x25, S_BFE_U32>;
def S_BFE_I32_vi           : SOP2_Real_vi <0x26, S_BFE_I32>;
def S_BFE_U64_vi           : SOP2_Real_vi <0x27, S_BFE_U64>;
def S_BFE_I64_vi           : SOP2_Real_vi <0x28, S_BFE_I64>;
def S_CBRANCH_G_FORK_vi    : SOP2_Real_vi <0x29, S_CBRANCH_G_FORK>;
def S_ABSDIFF_I32_vi       : SOP2_Real_vi <0x2a, S_ABSDIFF_I32>;
def S_PACK_LL_B32_B16_vi   : SOP2_Real_vi <0x32, S_PACK_LL_B32_B16>;
def S_PACK_LH_B32_B16_vi   : SOP2_Real_vi <0x33, S_PACK_LH_B32_B16>;
def S_PACK_HH_B32_B16_vi   : SOP2_Real_vi <0x34, S_PACK_HH_B32_B16>;
def S_RFE_RESTORE_B64_vi   : SOP2_Real_vi <0x2b, S_RFE_RESTORE_B64>;

def S_MOVK_I32_vi          : SOPK_Real_vi <0x00, S_MOVK_I32>;
def S_CMOVK_I32_vi         : SOPK_Real_vi <0x01, S_CMOVK_I32>;
def S_CMPK_EQ_I32_vi       : SOPK_Real_vi <0x02, S_CMPK_EQ_I32>;
def S_CMPK_LG_I32_vi       : SOPK_Real_vi <0x03, S_CMPK_LG_I32>;
def S_CMPK_GT_I32_vi       : SOPK_Real_vi <0x04, S_CMPK_GT_I32>;
def S_CMPK_GE_I32_vi       : SOPK_Real_vi <0x05, S_CMPK_GE_I32>;
def S_CMPK_LT_I32_vi       : SOPK_Real_vi <0x06, S_CMPK_LT_I32>;
def S_CMPK_LE_I32_vi       : SOPK_Real_vi <0x07, S_CMPK_LE_I32>;
def S_CMPK_EQ_U32_vi       : SOPK_Real_vi <0x08, S_CMPK_EQ_U32>;
def S_CMPK_LG_U32_vi       : SOPK_Real_vi <0x09, S_CMPK_LG_U32>;
def S_CMPK_GT_U32_vi       : SOPK_Real_vi <0x0A, S_CMPK_GT_U32>;
def S_CMPK_GE_U32_vi       : SOPK_Real_vi <0x0B, S_CMPK_GE_U32>;
def S_CMPK_LT_U32_vi       : SOPK_Real_vi <0x0C, S_CMPK_LT_U32>;
def S_CMPK_LE_U32_vi       : SOPK_Real_vi <0x0D, S_CMPK_LE_U32>;
def S_ADDK_I32_vi          : SOPK_Real_vi <0x0E, S_ADDK_I32>;
def S_MULK_I32_vi          : SOPK_Real_vi <0x0F, S_MULK_I32>;
def S_CBRANCH_I_FORK_vi    : SOPK_Real_vi <0x10, S_CBRANCH_I_FORK>;
def S_GETREG_B32_vi        : SOPK_Real_vi <0x11, S_GETREG_B32>;
def S_SETREG_B32_vi        : SOPK_Real_vi <0x12, S_SETREG_B32>;
//def S_GETREG_REGRD_B32_vi  : SOPK_Real_vi <0x13, S_GETREG_REGRD_B32>; // see pseudo for comments
def S_SETREG_IMM32_B32_vi  : SOPK_Real64<0x14, S_SETREG_IMM32_B32>,
                             Select_vi<S_SETREG_IMM32_B32.Mnemonic>;

def S_CALL_B64_vi          : SOPK_Real_vi <0x15, S_CALL_B64>;

//===----------------------------------------------------------------------===//
// SOP1 - GFX9.
//===----------------------------------------------------------------------===//

def S_ANDN1_SAVEEXEC_B64_vi   : SOP1_Real_vi<0x33, S_ANDN1_SAVEEXEC_B64>;
def S_ORN1_SAVEEXEC_B64_vi    : SOP1_Real_vi<0x34, S_ORN1_SAVEEXEC_B64>;
def S_ANDN1_WREXEC_B64_vi     : SOP1_Real_vi<0x35, S_ANDN1_WREXEC_B64>;
def S_ANDN2_WREXEC_B64_vi     : SOP1_Real_vi<0x36, S_ANDN2_WREXEC_B64>;
def S_BITREPLICATE_B64_B32_vi : SOP1_Real_vi<0x37, S_BITREPLICATE_B64_B32>;

//===----------------------------------------------------------------------===//
// SOP2 - GFX9.
//===----------------------------------------------------------------------===//

def S_LSHL1_ADD_U32_vi   : SOP2_Real_vi<0x2e, S_LSHL1_ADD_U32>;
def S_LSHL2_ADD_U32_vi   : SOP2_Real_vi<0x2f, S_LSHL2_ADD_U32>;
def S_LSHL3_ADD_U32_vi   : SOP2_Real_vi<0x30, S_LSHL3_ADD_U32>;
def S_LSHL4_ADD_U32_vi   : SOP2_Real_vi<0x31, S_LSHL4_ADD_U32>;
def S_MUL_HI_U32_vi      : SOP2_Real_vi<0x2c, S_MUL_HI_U32>;
def S_MUL_HI_I32_vi      : SOP2_Real_vi<0x2d, S_MUL_HI_I32>;