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
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
|
//===-- SystemZInstrInfo.cpp - SystemZ instruction information ------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file contains the SystemZ implementation of the TargetInstrInfo class.
//
//===----------------------------------------------------------------------===//
#include "SystemZInstrInfo.h"
#include "MCTargetDesc/SystemZMCTargetDesc.h"
#include "SystemZ.h"
#include "SystemZInstrBuilder.h"
#include "SystemZSubtarget.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/CodeGen/LiveInterval.h"
#include "llvm/CodeGen/LiveIntervals.h"
#include "llvm/CodeGen/LiveVariables.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineInstr.h"
#include "llvm/CodeGen/MachineMemOperand.h"
#include "llvm/CodeGen/MachineOperand.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/CodeGen/SlotIndexes.h"
#include "llvm/CodeGen/TargetInstrInfo.h"
#include "llvm/CodeGen/TargetSubtargetInfo.h"
#include "llvm/MC/MCInstrDesc.h"
#include "llvm/MC/MCRegisterInfo.h"
#include "llvm/Support/BranchProbability.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/Target/TargetMachine.h"
#include <cassert>
#include <cstdint>
#include <iterator>
using namespace llvm;
#define GET_INSTRINFO_CTOR_DTOR
#define GET_INSTRMAP_INFO
#include "SystemZGenInstrInfo.inc"
#define DEBUG_TYPE "systemz-II"
STATISTIC(LOCRMuxJumps, "Number of LOCRMux jump-sequences (lower is better)");
// Return a mask with Count low bits set.
static uint64_t allOnes(unsigned int Count) {
return Count == 0 ? 0 : (uint64_t(1) << (Count - 1) << 1) - 1;
}
// Reg should be a 32-bit GPR. Return true if it is a high register rather
// than a low register.
static bool isHighReg(unsigned int Reg) {
if (SystemZ::GRH32BitRegClass.contains(Reg))
return true;
assert(SystemZ::GR32BitRegClass.contains(Reg) && "Invalid GRX32");
return false;
}
// Pin the vtable to this file.
void SystemZInstrInfo::anchor() {}
SystemZInstrInfo::SystemZInstrInfo(SystemZSubtarget &sti)
: SystemZGenInstrInfo(SystemZ::ADJCALLSTACKDOWN, SystemZ::ADJCALLSTACKUP),
RI(), STI(sti) {
}
// MI is a 128-bit load or store. Split it into two 64-bit loads or stores,
// each having the opcode given by NewOpcode.
void SystemZInstrInfo::splitMove(MachineBasicBlock::iterator MI,
unsigned NewOpcode) const {
MachineBasicBlock *MBB = MI->getParent();
MachineFunction &MF = *MBB->getParent();
// Get two load or store instructions. Use the original instruction for one
// of them (arbitrarily the second here) and create a clone for the other.
MachineInstr *EarlierMI = MF.CloneMachineInstr(&*MI);
MBB->insert(MI, EarlierMI);
// Set up the two 64-bit registers and remember super reg and its flags.
MachineOperand &HighRegOp = EarlierMI->getOperand(0);
MachineOperand &LowRegOp = MI->getOperand(0);
unsigned Reg128 = LowRegOp.getReg();
unsigned Reg128Killed = getKillRegState(LowRegOp.isKill());
unsigned Reg128Undef = getUndefRegState(LowRegOp.isUndef());
HighRegOp.setReg(RI.getSubReg(HighRegOp.getReg(), SystemZ::subreg_h64));
LowRegOp.setReg(RI.getSubReg(LowRegOp.getReg(), SystemZ::subreg_l64));
if (MI->mayStore()) {
// Add implicit uses of the super register in case one of the subregs is
// undefined. We could track liveness and skip storing an undefined
// subreg, but this is hopefully rare (discovered with llvm-stress).
// If Reg128 was killed, set kill flag on MI.
unsigned Reg128UndefImpl = (Reg128Undef | RegState::Implicit);
MachineInstrBuilder(MF, EarlierMI).addReg(Reg128, Reg128UndefImpl);
MachineInstrBuilder(MF, MI).addReg(Reg128, (Reg128UndefImpl | Reg128Killed));
}
// The address in the first (high) instruction is already correct.
// Adjust the offset in the second (low) instruction.
MachineOperand &HighOffsetOp = EarlierMI->getOperand(2);
MachineOperand &LowOffsetOp = MI->getOperand(2);
LowOffsetOp.setImm(LowOffsetOp.getImm() + 8);
// Clear the kill flags on the registers in the first instruction.
if (EarlierMI->getOperand(0).isReg() && EarlierMI->getOperand(0).isUse())
EarlierMI->getOperand(0).setIsKill(false);
EarlierMI->getOperand(1).setIsKill(false);
EarlierMI->getOperand(3).setIsKill(false);
// Set the opcodes.
unsigned HighOpcode = getOpcodeForOffset(NewOpcode, HighOffsetOp.getImm());
unsigned LowOpcode = getOpcodeForOffset(NewOpcode, LowOffsetOp.getImm());
assert(HighOpcode && LowOpcode && "Both offsets should be in range");
EarlierMI->setDesc(get(HighOpcode));
MI->setDesc(get(LowOpcode));
}
// Split ADJDYNALLOC instruction MI.
void SystemZInstrInfo::splitAdjDynAlloc(MachineBasicBlock::iterator MI) const {
MachineBasicBlock *MBB = MI->getParent();
MachineFunction &MF = *MBB->getParent();
MachineFrameInfo &MFFrame = MF.getFrameInfo();
MachineOperand &OffsetMO = MI->getOperand(2);
uint64_t Offset = (MFFrame.getMaxCallFrameSize() +
SystemZMC::CallFrameSize +
OffsetMO.getImm());
unsigned NewOpcode = getOpcodeForOffset(SystemZ::LA, Offset);
assert(NewOpcode && "No support for huge argument lists yet");
MI->setDesc(get(NewOpcode));
OffsetMO.setImm(Offset);
}
// MI is an RI-style pseudo instruction. Replace it with LowOpcode
// if the first operand is a low GR32 and HighOpcode if the first operand
// is a high GR32. ConvertHigh is true if LowOpcode takes a signed operand
// and HighOpcode takes an unsigned 32-bit operand. In those cases,
// MI has the same kind of operand as LowOpcode, so needs to be converted
// if HighOpcode is used.
void SystemZInstrInfo::expandRIPseudo(MachineInstr &MI, unsigned LowOpcode,
unsigned HighOpcode,
bool ConvertHigh) const {
unsigned Reg = MI.getOperand(0).getReg();
bool IsHigh = isHighReg(Reg);
MI.setDesc(get(IsHigh ? HighOpcode : LowOpcode));
if (IsHigh && ConvertHigh)
MI.getOperand(1).setImm(uint32_t(MI.getOperand(1).getImm()));
}
// MI is a three-operand RIE-style pseudo instruction. Replace it with
// LowOpcodeK if the registers are both low GR32s, otherwise use a move
// followed by HighOpcode or LowOpcode, depending on whether the target
// is a high or low GR32.
void SystemZInstrInfo::expandRIEPseudo(MachineInstr &MI, unsigned LowOpcode,
unsigned LowOpcodeK,
unsigned HighOpcode) const {
unsigned DestReg = MI.getOperand(0).getReg();
unsigned SrcReg = MI.getOperand(1).getReg();
bool DestIsHigh = isHighReg(DestReg);
bool SrcIsHigh = isHighReg(SrcReg);
if (!DestIsHigh && !SrcIsHigh)
MI.setDesc(get(LowOpcodeK));
else {
emitGRX32Move(*MI.getParent(), MI, MI.getDebugLoc(), DestReg, SrcReg,
SystemZ::LR, 32, MI.getOperand(1).isKill(),
MI.getOperand(1).isUndef());
MI.setDesc(get(DestIsHigh ? HighOpcode : LowOpcode));
MI.getOperand(1).setReg(DestReg);
MI.tieOperands(0, 1);
}
}
// MI is an RXY-style pseudo instruction. Replace it with LowOpcode
// if the first operand is a low GR32 and HighOpcode if the first operand
// is a high GR32.
void SystemZInstrInfo::expandRXYPseudo(MachineInstr &MI, unsigned LowOpcode,
unsigned HighOpcode) const {
unsigned Reg = MI.getOperand(0).getReg();
unsigned Opcode = getOpcodeForOffset(isHighReg(Reg) ? HighOpcode : LowOpcode,
MI.getOperand(2).getImm());
MI.setDesc(get(Opcode));
}
// MI is a load-on-condition pseudo instruction with a single register
// (source or destination) operand. Replace it with LowOpcode if the
// register is a low GR32 and HighOpcode if the register is a high GR32.
void SystemZInstrInfo::expandLOCPseudo(MachineInstr &MI, unsigned LowOpcode,
unsigned HighOpcode) const {
unsigned Reg = MI.getOperand(0).getReg();
unsigned Opcode = isHighReg(Reg) ? HighOpcode : LowOpcode;
MI.setDesc(get(Opcode));
}
// MI is a load-register-on-condition pseudo instruction. Replace it with
// LowOpcode if source and destination are both low GR32s and HighOpcode if
// source and destination are both high GR32s.
void SystemZInstrInfo::expandLOCRPseudo(MachineInstr &MI, unsigned LowOpcode,
unsigned HighOpcode) const {
unsigned DestReg = MI.getOperand(0).getReg();
unsigned SrcReg = MI.getOperand(2).getReg();
bool DestIsHigh = isHighReg(DestReg);
bool SrcIsHigh = isHighReg(SrcReg);
if (!DestIsHigh && !SrcIsHigh)
MI.setDesc(get(LowOpcode));
else if (DestIsHigh && SrcIsHigh)
MI.setDesc(get(HighOpcode));
else
LOCRMuxJumps++;
// If we were unable to implement the pseudo with a single instruction, we
// need to convert it back into a branch sequence. This cannot be done here
// since the caller of expandPostRAPseudo does not handle changes to the CFG
// correctly. This change is defered to the SystemZExpandPseudo pass.
}
// MI is an RR-style pseudo instruction that zero-extends the low Size bits
// of one GRX32 into another. Replace it with LowOpcode if both operands
// are low registers, otherwise use RISB[LH]G.
void SystemZInstrInfo::expandZExtPseudo(MachineInstr &MI, unsigned LowOpcode,
unsigned Size) const {
MachineInstrBuilder MIB =
emitGRX32Move(*MI.getParent(), MI, MI.getDebugLoc(),
MI.getOperand(0).getReg(), MI.getOperand(1).getReg(), LowOpcode,
Size, MI.getOperand(1).isKill(), MI.getOperand(1).isUndef());
// Keep the remaining operands as-is.
for (unsigned I = 2; I < MI.getNumOperands(); ++I)
MIB.add(MI.getOperand(I));
MI.eraseFromParent();
}
void SystemZInstrInfo::expandLoadStackGuard(MachineInstr *MI) const {
MachineBasicBlock *MBB = MI->getParent();
MachineFunction &MF = *MBB->getParent();
const unsigned Reg64 = MI->getOperand(0).getReg();
const unsigned Reg32 = RI.getSubReg(Reg64, SystemZ::subreg_l32);
// EAR can only load the low subregister so us a shift for %a0 to produce
// the GR containing %a0 and %a1.
// ear <reg>, %a0
BuildMI(*MBB, MI, MI->getDebugLoc(), get(SystemZ::EAR), Reg32)
.addReg(SystemZ::A0)
.addReg(Reg64, RegState::ImplicitDefine);
// sllg <reg>, <reg>, 32
BuildMI(*MBB, MI, MI->getDebugLoc(), get(SystemZ::SLLG), Reg64)
.addReg(Reg64)
.addReg(0)
.addImm(32);
// ear <reg>, %a1
BuildMI(*MBB, MI, MI->getDebugLoc(), get(SystemZ::EAR), Reg32)
.addReg(SystemZ::A1);
// lg <reg>, 40(<reg>)
MI->setDesc(get(SystemZ::LG));
MachineInstrBuilder(MF, MI).addReg(Reg64).addImm(40).addReg(0);
}
// Emit a zero-extending move from 32-bit GPR SrcReg to 32-bit GPR
// DestReg before MBBI in MBB. Use LowLowOpcode when both DestReg and SrcReg
// are low registers, otherwise use RISB[LH]G. Size is the number of bits
// taken from the low end of SrcReg (8 for LLCR, 16 for LLHR and 32 for LR).
// KillSrc is true if this move is the last use of SrcReg.
MachineInstrBuilder
SystemZInstrInfo::emitGRX32Move(MachineBasicBlock &MBB,
MachineBasicBlock::iterator MBBI,
const DebugLoc &DL, unsigned DestReg,
unsigned SrcReg, unsigned LowLowOpcode,
unsigned Size, bool KillSrc,
bool UndefSrc) const {
unsigned Opcode;
bool DestIsHigh = isHighReg(DestReg);
bool SrcIsHigh = isHighReg(SrcReg);
if (DestIsHigh && SrcIsHigh)
Opcode = SystemZ::RISBHH;
else if (DestIsHigh && !SrcIsHigh)
Opcode = SystemZ::RISBHL;
else if (!DestIsHigh && SrcIsHigh)
Opcode = SystemZ::RISBLH;
else {
return BuildMI(MBB, MBBI, DL, get(LowLowOpcode), DestReg)
.addReg(SrcReg, getKillRegState(KillSrc) | getUndefRegState(UndefSrc));
}
unsigned Rotate = (DestIsHigh != SrcIsHigh ? 32 : 0);
return BuildMI(MBB, MBBI, DL, get(Opcode), DestReg)
.addReg(DestReg, RegState::Undef)
.addReg(SrcReg, getKillRegState(KillSrc) | getUndefRegState(UndefSrc))
.addImm(32 - Size).addImm(128 + 31).addImm(Rotate);
}
MachineInstr *SystemZInstrInfo::commuteInstructionImpl(MachineInstr &MI,
bool NewMI,
unsigned OpIdx1,
unsigned OpIdx2) const {
auto cloneIfNew = [NewMI](MachineInstr &MI) -> MachineInstr & {
if (NewMI)
return *MI.getParent()->getParent()->CloneMachineInstr(&MI);
return MI;
};
switch (MI.getOpcode()) {
case SystemZ::LOCRMux:
case SystemZ::LOCFHR:
case SystemZ::LOCR:
case SystemZ::LOCGR: {
auto &WorkingMI = cloneIfNew(MI);
// Invert condition.
unsigned CCValid = WorkingMI.getOperand(3).getImm();
unsigned CCMask = WorkingMI.getOperand(4).getImm();
WorkingMI.getOperand(4).setImm(CCMask ^ CCValid);
return TargetInstrInfo::commuteInstructionImpl(WorkingMI, /*NewMI=*/false,
OpIdx1, OpIdx2);
}
default:
return TargetInstrInfo::commuteInstructionImpl(MI, NewMI, OpIdx1, OpIdx2);
}
}
// If MI is a simple load or store for a frame object, return the register
// it loads or stores and set FrameIndex to the index of the frame object.
// Return 0 otherwise.
//
// Flag is SimpleBDXLoad for loads and SimpleBDXStore for stores.
static int isSimpleMove(const MachineInstr &MI, int &FrameIndex,
unsigned Flag) {
const MCInstrDesc &MCID = MI.getDesc();
if ((MCID.TSFlags & Flag) && MI.getOperand(1).isFI() &&
MI.getOperand(2).getImm() == 0 && MI.getOperand(3).getReg() == 0) {
FrameIndex = MI.getOperand(1).getIndex();
return MI.getOperand(0).getReg();
}
return 0;
}
unsigned SystemZInstrInfo::isLoadFromStackSlot(const MachineInstr &MI,
int &FrameIndex) const {
return isSimpleMove(MI, FrameIndex, SystemZII::SimpleBDXLoad);
}
unsigned SystemZInstrInfo::isStoreToStackSlot(const MachineInstr &MI,
int &FrameIndex) const {
return isSimpleMove(MI, FrameIndex, SystemZII::SimpleBDXStore);
}
bool SystemZInstrInfo::isStackSlotCopy(const MachineInstr &MI,
int &DestFrameIndex,
int &SrcFrameIndex) const {
// Check for MVC 0(Length,FI1),0(FI2)
const MachineFrameInfo &MFI = MI.getParent()->getParent()->getFrameInfo();
if (MI.getOpcode() != SystemZ::MVC || !MI.getOperand(0).isFI() ||
MI.getOperand(1).getImm() != 0 || !MI.getOperand(3).isFI() ||
MI.getOperand(4).getImm() != 0)
return false;
// Check that Length covers the full slots.
int64_t Length = MI.getOperand(2).getImm();
unsigned FI1 = MI.getOperand(0).getIndex();
unsigned FI2 = MI.getOperand(3).getIndex();
if (MFI.getObjectSize(FI1) != Length ||
MFI.getObjectSize(FI2) != Length)
return false;
DestFrameIndex = FI1;
SrcFrameIndex = FI2;
return true;
}
bool SystemZInstrInfo::analyzeBranch(MachineBasicBlock &MBB,
MachineBasicBlock *&TBB,
MachineBasicBlock *&FBB,
SmallVectorImpl<MachineOperand> &Cond,
bool AllowModify) const {
// Most of the code and comments here are boilerplate.
// Start from the bottom of the block and work up, examining the
// terminator instructions.
MachineBasicBlock::iterator I = MBB.end();
while (I != MBB.begin()) {
--I;
if (I->isDebugInstr())
continue;
// Working from the bottom, when we see a non-terminator instruction, we're
// done.
if (!isUnpredicatedTerminator(*I))
break;
// A terminator that isn't a branch can't easily be handled by this
// analysis.
if (!I->isBranch())
return true;
// Can't handle indirect branches.
SystemZII::Branch Branch(getBranchInfo(*I));
if (!Branch.Target->isMBB())
return true;
// Punt on compound branches.
if (Branch.Type != SystemZII::BranchNormal)
return true;
if (Branch.CCMask == SystemZ::CCMASK_ANY) {
// Handle unconditional branches.
if (!AllowModify) {
TBB = Branch.Target->getMBB();
continue;
}
// If the block has any instructions after a JMP, delete them.
while (std::next(I) != MBB.end())
std::next(I)->eraseFromParent();
Cond.clear();
FBB = nullptr;
// Delete the JMP if it's equivalent to a fall-through.
if (MBB.isLayoutSuccessor(Branch.Target->getMBB())) {
TBB = nullptr;
I->eraseFromParent();
I = MBB.end();
continue;
}
// TBB is used to indicate the unconditinal destination.
TBB = Branch.Target->getMBB();
continue;
}
// Working from the bottom, handle the first conditional branch.
if (Cond.empty()) {
// FIXME: add X86-style branch swap
FBB = TBB;
TBB = Branch.Target->getMBB();
Cond.push_back(MachineOperand::CreateImm(Branch.CCValid));
Cond.push_back(MachineOperand::CreateImm(Branch.CCMask));
continue;
}
// Handle subsequent conditional branches.
assert(Cond.size() == 2 && TBB && "Should have seen a conditional branch");
// Only handle the case where all conditional branches branch to the same
// destination.
if (TBB != Branch.Target->getMBB())
return true;
// If the conditions are the same, we can leave them alone.
unsigned OldCCValid = Cond[0].getImm();
unsigned OldCCMask = Cond[1].getImm();
if (OldCCValid == Branch.CCValid && OldCCMask == Branch.CCMask)
continue;
// FIXME: Try combining conditions like X86 does. Should be easy on Z!
return false;
}
return false;
}
unsigned SystemZInstrInfo::removeBranch(MachineBasicBlock &MBB,
int *BytesRemoved) const {
assert(!BytesRemoved && "code size not handled");
// Most of the code and comments here are boilerplate.
MachineBasicBlock::iterator I = MBB.end();
unsigned Count = 0;
while (I != MBB.begin()) {
--I;
if (I->isDebugInstr())
continue;
if (!I->isBranch())
break;
if (!getBranchInfo(*I).Target->isMBB())
break;
// Remove the branch.
I->eraseFromParent();
I = MBB.end();
++Count;
}
return Count;
}
bool SystemZInstrInfo::
reverseBranchCondition(SmallVectorImpl<MachineOperand> &Cond) const {
assert(Cond.size() == 2 && "Invalid condition");
Cond[1].setImm(Cond[1].getImm() ^ Cond[0].getImm());
return false;
}
unsigned SystemZInstrInfo::insertBranch(MachineBasicBlock &MBB,
MachineBasicBlock *TBB,
MachineBasicBlock *FBB,
ArrayRef<MachineOperand> Cond,
const DebugLoc &DL,
int *BytesAdded) const {
// In this function we output 32-bit branches, which should always
// have enough range. They can be shortened and relaxed by later code
// in the pipeline, if desired.
// Shouldn't be a fall through.
assert(TBB && "insertBranch must not be told to insert a fallthrough");
assert((Cond.size() == 2 || Cond.size() == 0) &&
"SystemZ branch conditions have one component!");
assert(!BytesAdded && "code size not handled");
if (Cond.empty()) {
// Unconditional branch?
assert(!FBB && "Unconditional branch with multiple successors!");
BuildMI(&MBB, DL, get(SystemZ::J)).addMBB(TBB);
return 1;
}
// Conditional branch.
unsigned Count = 0;
unsigned CCValid = Cond[0].getImm();
unsigned CCMask = Cond[1].getImm();
BuildMI(&MBB, DL, get(SystemZ::BRC))
.addImm(CCValid).addImm(CCMask).addMBB(TBB);
++Count;
if (FBB) {
// Two-way Conditional branch. Insert the second branch.
BuildMI(&MBB, DL, get(SystemZ::J)).addMBB(FBB);
++Count;
}
return Count;
}
bool SystemZInstrInfo::analyzeCompare(const MachineInstr &MI, unsigned &SrcReg,
unsigned &SrcReg2, int &Mask,
int &Value) const {
assert(MI.isCompare() && "Caller should have checked for a comparison");
if (MI.getNumExplicitOperands() == 2 && MI.getOperand(0).isReg() &&
MI.getOperand(1).isImm()) {
SrcReg = MI.getOperand(0).getReg();
SrcReg2 = 0;
Value = MI.getOperand(1).getImm();
Mask = ~0;
return true;
}
return false;
}
// If Reg is a virtual register, return its definition, otherwise return null.
static MachineInstr *getDef(unsigned Reg,
const MachineRegisterInfo *MRI) {
if (TargetRegisterInfo::isPhysicalRegister(Reg))
return nullptr;
return MRI->getUniqueVRegDef(Reg);
}
// Return true if MI is a shift of type Opcode by Imm bits.
static bool isShift(MachineInstr *MI, unsigned Opcode, int64_t Imm) {
return (MI->getOpcode() == Opcode &&
!MI->getOperand(2).getReg() &&
MI->getOperand(3).getImm() == Imm);
}
// If the destination of MI has no uses, delete it as dead.
static void eraseIfDead(MachineInstr *MI, const MachineRegisterInfo *MRI) {
if (MRI->use_nodbg_empty(MI->getOperand(0).getReg()))
MI->eraseFromParent();
}
// Compare compares SrcReg against zero. Check whether SrcReg contains
// the result of an IPM sequence whose input CC survives until Compare,
// and whether Compare is therefore redundant. Delete it and return
// true if so.
static bool removeIPMBasedCompare(MachineInstr &Compare, unsigned SrcReg,
const MachineRegisterInfo *MRI,
const TargetRegisterInfo *TRI) {
MachineInstr *LGFR = nullptr;
MachineInstr *RLL = getDef(SrcReg, MRI);
if (RLL && RLL->getOpcode() == SystemZ::LGFR) {
LGFR = RLL;
RLL = getDef(LGFR->getOperand(1).getReg(), MRI);
}
if (!RLL || !isShift(RLL, SystemZ::RLL, 31))
return false;
MachineInstr *SRL = getDef(RLL->getOperand(1).getReg(), MRI);
if (!SRL || !isShift(SRL, SystemZ::SRL, SystemZ::IPM_CC))
return false;
MachineInstr *IPM = getDef(SRL->getOperand(1).getReg(), MRI);
if (!IPM || IPM->getOpcode() != SystemZ::IPM)
return false;
// Check that there are no assignments to CC between the IPM and Compare,
if (IPM->getParent() != Compare.getParent())
return false;
MachineBasicBlock::iterator MBBI = IPM, MBBE = Compare.getIterator();
for (++MBBI; MBBI != MBBE; ++MBBI) {
MachineInstr &MI = *MBBI;
if (MI.modifiesRegister(SystemZ::CC, TRI))
return false;
}
Compare.eraseFromParent();
if (LGFR)
eraseIfDead(LGFR, MRI);
eraseIfDead(RLL, MRI);
eraseIfDead(SRL, MRI);
eraseIfDead(IPM, MRI);
return true;
}
bool SystemZInstrInfo::optimizeCompareInstr(
MachineInstr &Compare, unsigned SrcReg, unsigned SrcReg2, int Mask,
int Value, const MachineRegisterInfo *MRI) const {
assert(!SrcReg2 && "Only optimizing constant comparisons so far");
bool IsLogical = (Compare.getDesc().TSFlags & SystemZII::IsLogical) != 0;
return Value == 0 && !IsLogical &&
removeIPMBasedCompare(Compare, SrcReg, MRI, &RI);
}
bool SystemZInstrInfo::canInsertSelect(const MachineBasicBlock &MBB,
ArrayRef<MachineOperand> Pred,
unsigned TrueReg, unsigned FalseReg,
int &CondCycles, int &TrueCycles,
int &FalseCycles) const {
// Not all subtargets have LOCR instructions.
if (!STI.hasLoadStoreOnCond())
return false;
if (Pred.size() != 2)
return false;
// Check register classes.
const MachineRegisterInfo &MRI = MBB.getParent()->getRegInfo();
const TargetRegisterClass *RC =
RI.getCommonSubClass(MRI.getRegClass(TrueReg), MRI.getRegClass(FalseReg));
if (!RC)
return false;
// We have LOCR instructions for 32 and 64 bit general purpose registers.
if ((STI.hasLoadStoreOnCond2() &&
SystemZ::GRX32BitRegClass.hasSubClassEq(RC)) ||
SystemZ::GR32BitRegClass.hasSubClassEq(RC) ||
SystemZ::GR64BitRegClass.hasSubClassEq(RC)) {
CondCycles = 2;
TrueCycles = 2;
FalseCycles = 2;
return true;
}
// Can't do anything else.
return false;
}
void SystemZInstrInfo::insertSelect(MachineBasicBlock &MBB,
MachineBasicBlock::iterator I,
const DebugLoc &DL, unsigned DstReg,
ArrayRef<MachineOperand> Pred,
unsigned TrueReg,
unsigned FalseReg) const {
MachineRegisterInfo &MRI = MBB.getParent()->getRegInfo();
const TargetRegisterClass *RC = MRI.getRegClass(DstReg);
assert(Pred.size() == 2 && "Invalid condition");
unsigned CCValid = Pred[0].getImm();
unsigned CCMask = Pred[1].getImm();
unsigned Opc;
if (SystemZ::GRX32BitRegClass.hasSubClassEq(RC)) {
if (STI.hasLoadStoreOnCond2())
Opc = SystemZ::LOCRMux;
else {
Opc = SystemZ::LOCR;
MRI.constrainRegClass(DstReg, &SystemZ::GR32BitRegClass);
unsigned TReg = MRI.createVirtualRegister(&SystemZ::GR32BitRegClass);
unsigned FReg = MRI.createVirtualRegister(&SystemZ::GR32BitRegClass);
BuildMI(MBB, I, DL, get(TargetOpcode::COPY), TReg).addReg(TrueReg);
BuildMI(MBB, I, DL, get(TargetOpcode::COPY), FReg).addReg(FalseReg);
TrueReg = TReg;
FalseReg = FReg;
}
} else if (SystemZ::GR64BitRegClass.hasSubClassEq(RC))
Opc = SystemZ::LOCGR;
else
llvm_unreachable("Invalid register class");
BuildMI(MBB, I, DL, get(Opc), DstReg)
.addReg(FalseReg).addReg(TrueReg)
.addImm(CCValid).addImm(CCMask);
}
bool SystemZInstrInfo::FoldImmediate(MachineInstr &UseMI, MachineInstr &DefMI,
unsigned Reg,
MachineRegisterInfo *MRI) const {
unsigned DefOpc = DefMI.getOpcode();
if (DefOpc != SystemZ::LHIMux && DefOpc != SystemZ::LHI &&
DefOpc != SystemZ::LGHI)
return false;
if (DefMI.getOperand(0).getReg() != Reg)
return false;
int32_t ImmVal = (int32_t)DefMI.getOperand(1).getImm();
unsigned UseOpc = UseMI.getOpcode();
unsigned NewUseOpc;
unsigned UseIdx;
int CommuteIdx = -1;
switch (UseOpc) {
case SystemZ::LOCRMux:
if (!STI.hasLoadStoreOnCond2())
return false;
NewUseOpc = SystemZ::LOCHIMux;
if (UseMI.getOperand(2).getReg() == Reg)
UseIdx = 2;
else if (UseMI.getOperand(1).getReg() == Reg)
UseIdx = 2, CommuteIdx = 1;
else
return false;
break;
case SystemZ::LOCGR:
if (!STI.hasLoadStoreOnCond2())
return false;
NewUseOpc = SystemZ::LOCGHI;
if (UseMI.getOperand(2).getReg() == Reg)
UseIdx = 2;
else if (UseMI.getOperand(1).getReg() == Reg)
UseIdx = 2, CommuteIdx = 1;
else
return false;
break;
default:
return false;
}
if (CommuteIdx != -1)
if (!commuteInstruction(UseMI, false, CommuteIdx, UseIdx))
return false;
bool DeleteDef = MRI->hasOneNonDBGUse(Reg);
UseMI.setDesc(get(NewUseOpc));
UseMI.getOperand(UseIdx).ChangeToImmediate(ImmVal);
if (DeleteDef)
DefMI.eraseFromParent();
return true;
}
bool SystemZInstrInfo::isPredicable(const MachineInstr &MI) const {
unsigned Opcode = MI.getOpcode();
if (Opcode == SystemZ::Return ||
Opcode == SystemZ::Trap ||
Opcode == SystemZ::CallJG ||
Opcode == SystemZ::CallBR)
return true;
return false;
}
bool SystemZInstrInfo::
isProfitableToIfCvt(MachineBasicBlock &MBB,
unsigned NumCycles, unsigned ExtraPredCycles,
BranchProbability Probability) const {
// Avoid using conditional returns at the end of a loop (since then
// we'd need to emit an unconditional branch to the beginning anyway,
// making the loop body longer). This doesn't apply for low-probability
// loops (eg. compare-and-swap retry), so just decide based on branch
// probability instead of looping structure.
// However, since Compare and Trap instructions cost the same as a regular
// Compare instruction, we should allow the if conversion to convert this
// into a Conditional Compare regardless of the branch probability.
if (MBB.getLastNonDebugInstr()->getOpcode() != SystemZ::Trap &&
MBB.succ_empty() && Probability < BranchProbability(1, 8))
return false;
// For now only convert single instructions.
return NumCycles == 1;
}
bool SystemZInstrInfo::
isProfitableToIfCvt(MachineBasicBlock &TMBB,
unsigned NumCyclesT, unsigned ExtraPredCyclesT,
MachineBasicBlock &FMBB,
unsigned NumCyclesF, unsigned ExtraPredCyclesF,
BranchProbability Probability) const {
// For now avoid converting mutually-exclusive cases.
return false;
}
bool SystemZInstrInfo::
isProfitableToDupForIfCvt(MachineBasicBlock &MBB, unsigned NumCycles,
BranchProbability Probability) const {
// For now only duplicate single instructions.
return NumCycles == 1;
}
bool SystemZInstrInfo::PredicateInstruction(
MachineInstr &MI, ArrayRef<MachineOperand> Pred) const {
assert(Pred.size() == 2 && "Invalid condition");
unsigned CCValid = Pred[0].getImm();
unsigned CCMask = Pred[1].getImm();
assert(CCMask > 0 && CCMask < 15 && "Invalid predicate");
unsigned Opcode = MI.getOpcode();
if (Opcode == SystemZ::Trap) {
MI.setDesc(get(SystemZ::CondTrap));
MachineInstrBuilder(*MI.getParent()->getParent(), MI)
.addImm(CCValid).addImm(CCMask)
.addReg(SystemZ::CC, RegState::Implicit);
return true;
}
if (Opcode == SystemZ::Return) {
MI.setDesc(get(SystemZ::CondReturn));
MachineInstrBuilder(*MI.getParent()->getParent(), MI)
.addImm(CCValid).addImm(CCMask)
.addReg(SystemZ::CC, RegState::Implicit);
return true;
}
if (Opcode == SystemZ::CallJG) {
MachineOperand FirstOp = MI.getOperand(0);
const uint32_t *RegMask = MI.getOperand(1).getRegMask();
MI.RemoveOperand(1);
MI.RemoveOperand(0);
MI.setDesc(get(SystemZ::CallBRCL));
MachineInstrBuilder(*MI.getParent()->getParent(), MI)
.addImm(CCValid)
.addImm(CCMask)
.add(FirstOp)
.addRegMask(RegMask)
.addReg(SystemZ::CC, RegState::Implicit);
return true;
}
if (Opcode == SystemZ::CallBR) {
const uint32_t *RegMask = MI.getOperand(0).getRegMask();
MI.RemoveOperand(0);
MI.setDesc(get(SystemZ::CallBCR));
MachineInstrBuilder(*MI.getParent()->getParent(), MI)
.addImm(CCValid).addImm(CCMask)
.addRegMask(RegMask)
.addReg(SystemZ::CC, RegState::Implicit);
return true;
}
return false;
}
void SystemZInstrInfo::copyPhysReg(MachineBasicBlock &MBB,
MachineBasicBlock::iterator MBBI,
const DebugLoc &DL, unsigned DestReg,
unsigned SrcReg, bool KillSrc) const {
// Split 128-bit GPR moves into two 64-bit moves. Add implicit uses of the
// super register in case one of the subregs is undefined.
// This handles ADDR128 too.
if (SystemZ::GR128BitRegClass.contains(DestReg, SrcReg)) {
copyPhysReg(MBB, MBBI, DL, RI.getSubReg(DestReg, SystemZ::subreg_h64),
RI.getSubReg(SrcReg, SystemZ::subreg_h64), KillSrc);
MachineInstrBuilder(*MBB.getParent(), std::prev(MBBI))
.addReg(SrcReg, RegState::Implicit);
copyPhysReg(MBB, MBBI, DL, RI.getSubReg(DestReg, SystemZ::subreg_l64),
RI.getSubReg(SrcReg, SystemZ::subreg_l64), KillSrc);
MachineInstrBuilder(*MBB.getParent(), std::prev(MBBI))
.addReg(SrcReg, (getKillRegState(KillSrc) | RegState::Implicit));
return;
}
if (SystemZ::GRX32BitRegClass.contains(DestReg, SrcReg)) {
emitGRX32Move(MBB, MBBI, DL, DestReg, SrcReg, SystemZ::LR, 32, KillSrc,
false);
return;
}
// Move 128-bit floating-point values between VR128 and FP128.
if (SystemZ::VR128BitRegClass.contains(DestReg) &&
SystemZ::FP128BitRegClass.contains(SrcReg)) {
unsigned SrcRegHi =
RI.getMatchingSuperReg(RI.getSubReg(SrcReg, SystemZ::subreg_h64),
SystemZ::subreg_h64, &SystemZ::VR128BitRegClass);
unsigned SrcRegLo =
RI.getMatchingSuperReg(RI.getSubReg(SrcReg, SystemZ::subreg_l64),
SystemZ::subreg_h64, &SystemZ::VR128BitRegClass);
BuildMI(MBB, MBBI, DL, get(SystemZ::VMRHG), DestReg)
.addReg(SrcRegHi, getKillRegState(KillSrc))
.addReg(SrcRegLo, getKillRegState(KillSrc));
return;
}
if (SystemZ::FP128BitRegClass.contains(DestReg) &&
SystemZ::VR128BitRegClass.contains(SrcReg)) {
unsigned DestRegHi =
RI.getMatchingSuperReg(RI.getSubReg(DestReg, SystemZ::subreg_h64),
SystemZ::subreg_h64, &SystemZ::VR128BitRegClass);
unsigned DestRegLo =
RI.getMatchingSuperReg(RI.getSubReg(DestReg, SystemZ::subreg_l64),
SystemZ::subreg_h64, &SystemZ::VR128BitRegClass);
if (DestRegHi != SrcReg)
copyPhysReg(MBB, MBBI, DL, DestRegHi, SrcReg, false);
BuildMI(MBB, MBBI, DL, get(SystemZ::VREPG), DestRegLo)
.addReg(SrcReg, getKillRegState(KillSrc)).addImm(1);
return;
}
// Move CC value from/to a GR32.
if (SrcReg == SystemZ::CC) {
auto MIB = BuildMI(MBB, MBBI, DL, get(SystemZ::IPM), DestReg);
if (KillSrc) {
const MachineFunction *MF = MBB.getParent();
const TargetRegisterInfo *TRI = MF->getSubtarget().getRegisterInfo();
MIB->addRegisterKilled(SrcReg, TRI);
}
return;
}
if (DestReg == SystemZ::CC) {
BuildMI(MBB, MBBI, DL, get(SystemZ::TMLH))
.addReg(SrcReg, getKillRegState(KillSrc))
.addImm(3 << (SystemZ::IPM_CC - 16));
return;
}
// Everything else needs only one instruction.
unsigned Opcode;
if (SystemZ::GR64BitRegClass.contains(DestReg, SrcReg))
Opcode = SystemZ::LGR;
else if (SystemZ::FP32BitRegClass.contains(DestReg, SrcReg))
// For z13 we prefer LDR over LER to avoid partial register dependencies.
Opcode = STI.hasVector() ? SystemZ::LDR32 : SystemZ::LER;
else if (SystemZ::FP64BitRegClass.contains(DestReg, SrcReg))
Opcode = SystemZ::LDR;
else if (SystemZ::FP128BitRegClass.contains(DestReg, SrcReg))
Opcode = SystemZ::LXR;
else if (SystemZ::VR32BitRegClass.contains(DestReg, SrcReg))
Opcode = SystemZ::VLR32;
else if (SystemZ::VR64BitRegClass.contains(DestReg, SrcReg))
Opcode = SystemZ::VLR64;
else if (SystemZ::VR128BitRegClass.contains(DestReg, SrcReg))
Opcode = SystemZ::VLR;
else if (SystemZ::AR32BitRegClass.contains(DestReg, SrcReg))
Opcode = SystemZ::CPYA;
else if (SystemZ::AR32BitRegClass.contains(DestReg) &&
SystemZ::GR32BitRegClass.contains(SrcReg))
Opcode = SystemZ::SAR;
else if (SystemZ::GR32BitRegClass.contains(DestReg) &&
SystemZ::AR32BitRegClass.contains(SrcReg))
Opcode = SystemZ::EAR;
else
llvm_unreachable("Impossible reg-to-reg copy");
BuildMI(MBB, MBBI, DL, get(Opcode), DestReg)
.addReg(SrcReg, getKillRegState(KillSrc));
}
void SystemZInstrInfo::storeRegToStackSlot(
MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI, unsigned SrcReg,
bool isKill, int FrameIdx, const TargetRegisterClass *RC,
const TargetRegisterInfo *TRI) const {
DebugLoc DL = MBBI != MBB.end() ? MBBI->getDebugLoc() : DebugLoc();
// Callers may expect a single instruction, so keep 128-bit moves
// together for now and lower them after register allocation.
unsigned LoadOpcode, StoreOpcode;
getLoadStoreOpcodes(RC, LoadOpcode, StoreOpcode);
addFrameReference(BuildMI(MBB, MBBI, DL, get(StoreOpcode))
.addReg(SrcReg, getKillRegState(isKill)),
FrameIdx);
}
void SystemZInstrInfo::loadRegFromStackSlot(
MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI, unsigned DestReg,
int FrameIdx, const TargetRegisterClass *RC,
const TargetRegisterInfo *TRI) const {
DebugLoc DL = MBBI != MBB.end() ? MBBI->getDebugLoc() : DebugLoc();
// Callers may expect a single instruction, so keep 128-bit moves
// together for now and lower them after register allocation.
unsigned LoadOpcode, StoreOpcode;
getLoadStoreOpcodes(RC, LoadOpcode, StoreOpcode);
addFrameReference(BuildMI(MBB, MBBI, DL, get(LoadOpcode), DestReg),
FrameIdx);
}
// Return true if MI is a simple load or store with a 12-bit displacement
// and no index. Flag is SimpleBDXLoad for loads and SimpleBDXStore for stores.
static bool isSimpleBD12Move(const MachineInstr *MI, unsigned Flag) {
const MCInstrDesc &MCID = MI->getDesc();
return ((MCID.TSFlags & Flag) &&
isUInt<12>(MI->getOperand(2).getImm()) &&
MI->getOperand(3).getReg() == 0);
}
namespace {
struct LogicOp {
LogicOp() = default;
LogicOp(unsigned regSize, unsigned immLSB, unsigned immSize)
: RegSize(regSize), ImmLSB(immLSB), ImmSize(immSize) {}
explicit operator bool() const { return RegSize; }
unsigned RegSize = 0;
unsigned ImmLSB = 0;
unsigned ImmSize = 0;
};
} // end anonymous namespace
static LogicOp interpretAndImmediate(unsigned Opcode) {
switch (Opcode) {
case SystemZ::NILMux: return LogicOp(32, 0, 16);
case SystemZ::NIHMux: return LogicOp(32, 16, 16);
case SystemZ::NILL64: return LogicOp(64, 0, 16);
case SystemZ::NILH64: return LogicOp(64, 16, 16);
case SystemZ::NIHL64: return LogicOp(64, 32, 16);
case SystemZ::NIHH64: return LogicOp(64, 48, 16);
case SystemZ::NIFMux: return LogicOp(32, 0, 32);
case SystemZ::NILF64: return LogicOp(64, 0, 32);
case SystemZ::NIHF64: return LogicOp(64, 32, 32);
default: return LogicOp();
}
}
static void transferDeadCC(MachineInstr *OldMI, MachineInstr *NewMI) {
if (OldMI->registerDefIsDead(SystemZ::CC)) {
MachineOperand *CCDef = NewMI->findRegisterDefOperand(SystemZ::CC);
if (CCDef != nullptr)
CCDef->setIsDead(true);
}
}
// Used to return from convertToThreeAddress after replacing two-address
// instruction OldMI with three-address instruction NewMI.
static MachineInstr *finishConvertToThreeAddress(MachineInstr *OldMI,
MachineInstr *NewMI,
LiveVariables *LV) {
if (LV) {
unsigned NumOps = OldMI->getNumOperands();
for (unsigned I = 1; I < NumOps; ++I) {
MachineOperand &Op = OldMI->getOperand(I);
if (Op.isReg() && Op.isKill())
LV->replaceKillInstruction(Op.getReg(), *OldMI, *NewMI);
}
}
transferDeadCC(OldMI, NewMI);
return NewMI;
}
MachineInstr *SystemZInstrInfo::convertToThreeAddress(
MachineFunction::iterator &MFI, MachineInstr &MI, LiveVariables *LV) const {
MachineBasicBlock *MBB = MI.getParent();
MachineFunction *MF = MBB->getParent();
MachineRegisterInfo &MRI = MF->getRegInfo();
unsigned Opcode = MI.getOpcode();
unsigned NumOps = MI.getNumOperands();
// Try to convert something like SLL into SLLK, if supported.
// We prefer to keep the two-operand form where possible both
// because it tends to be shorter and because some instructions
// have memory forms that can be used during spilling.
if (STI.hasDistinctOps()) {
MachineOperand &Dest = MI.getOperand(0);
MachineOperand &Src = MI.getOperand(1);
unsigned DestReg = Dest.getReg();
unsigned SrcReg = Src.getReg();
// AHIMux is only really a three-operand instruction when both operands
// are low registers. Try to constrain both operands to be low if
// possible.
if (Opcode == SystemZ::AHIMux &&
TargetRegisterInfo::isVirtualRegister(DestReg) &&
TargetRegisterInfo::isVirtualRegister(SrcReg) &&
MRI.getRegClass(DestReg)->contains(SystemZ::R1L) &&
MRI.getRegClass(SrcReg)->contains(SystemZ::R1L)) {
MRI.constrainRegClass(DestReg, &SystemZ::GR32BitRegClass);
MRI.constrainRegClass(SrcReg, &SystemZ::GR32BitRegClass);
}
int ThreeOperandOpcode = SystemZ::getThreeOperandOpcode(Opcode);
if (ThreeOperandOpcode >= 0) {
// Create three address instruction without adding the implicit
// operands. Those will instead be copied over from the original
// instruction by the loop below.
MachineInstrBuilder MIB(
*MF, MF->CreateMachineInstr(get(ThreeOperandOpcode), MI.getDebugLoc(),
/*NoImplicit=*/true));
MIB.add(Dest);
// Keep the kill state, but drop the tied flag.
MIB.addReg(Src.getReg(), getKillRegState(Src.isKill()), Src.getSubReg());
// Keep the remaining operands as-is.
for (unsigned I = 2; I < NumOps; ++I)
MIB.add(MI.getOperand(I));
MBB->insert(MI, MIB);
return finishConvertToThreeAddress(&MI, MIB, LV);
}
}
// Try to convert an AND into an RISBG-type instruction.
if (LogicOp And = interpretAndImmediate(Opcode)) {
uint64_t Imm = MI.getOperand(2).getImm() << And.ImmLSB;
// AND IMMEDIATE leaves the other bits of the register unchanged.
Imm |= allOnes(And.RegSize) & ~(allOnes(And.ImmSize) << And.ImmLSB);
unsigned Start, End;
if (isRxSBGMask(Imm, And.RegSize, Start, End)) {
unsigned NewOpcode;
if (And.RegSize == 64) {
NewOpcode = SystemZ::RISBG;
// Prefer RISBGN if available, since it does not clobber CC.
if (STI.hasMiscellaneousExtensions())
NewOpcode = SystemZ::RISBGN;
} else {
NewOpcode = SystemZ::RISBMux;
Start &= 31;
End &= 31;
}
MachineOperand &Dest = MI.getOperand(0);
MachineOperand &Src = MI.getOperand(1);
MachineInstrBuilder MIB =
BuildMI(*MBB, MI, MI.getDebugLoc(), get(NewOpcode))
.add(Dest)
.addReg(0)
.addReg(Src.getReg(), getKillRegState(Src.isKill()),
Src.getSubReg())
.addImm(Start)
.addImm(End + 128)
.addImm(0);
return finishConvertToThreeAddress(&MI, MIB, LV);
}
}
return nullptr;
}
MachineInstr *SystemZInstrInfo::foldMemoryOperandImpl(
MachineFunction &MF, MachineInstr &MI, ArrayRef<unsigned> Ops,
MachineBasicBlock::iterator InsertPt, int FrameIndex,
LiveIntervals *LIS) const {
const TargetRegisterInfo *TRI = MF.getSubtarget().getRegisterInfo();
const MachineFrameInfo &MFI = MF.getFrameInfo();
unsigned Size = MFI.getObjectSize(FrameIndex);
unsigned Opcode = MI.getOpcode();
if (Ops.size() == 2 && Ops[0] == 0 && Ops[1] == 1) {
if (LIS != nullptr && (Opcode == SystemZ::LA || Opcode == SystemZ::LAY) &&
isInt<8>(MI.getOperand(2).getImm()) && !MI.getOperand(3).getReg()) {
// Check CC liveness, since new instruction introduces a dead
// def of CC.
MCRegUnitIterator CCUnit(SystemZ::CC, TRI);
LiveRange &CCLiveRange = LIS->getRegUnit(*CCUnit);
++CCUnit;
assert(!CCUnit.isValid() && "CC only has one reg unit.");
SlotIndex MISlot =
LIS->getSlotIndexes()->getInstructionIndex(MI).getRegSlot();
if (!CCLiveRange.liveAt(MISlot)) {
// LA(Y) %reg, CONST(%reg) -> AGSI %mem, CONST
MachineInstr *BuiltMI = BuildMI(*InsertPt->getParent(), InsertPt,
MI.getDebugLoc(), get(SystemZ::AGSI))
.addFrameIndex(FrameIndex)
.addImm(0)
.addImm(MI.getOperand(2).getImm());
BuiltMI->findRegisterDefOperand(SystemZ::CC)->setIsDead(true);
CCLiveRange.createDeadDef(MISlot, LIS->getVNInfoAllocator());
return BuiltMI;
}
}
return nullptr;
}
// All other cases require a single operand.
if (Ops.size() != 1)
return nullptr;
unsigned OpNum = Ops[0];
assert(Size * 8 ==
TRI->getRegSizeInBits(*MF.getRegInfo()
.getRegClass(MI.getOperand(OpNum).getReg())) &&
"Invalid size combination");
if ((Opcode == SystemZ::AHI || Opcode == SystemZ::AGHI) && OpNum == 0 &&
isInt<8>(MI.getOperand(2).getImm())) {
// A(G)HI %reg, CONST -> A(G)SI %mem, CONST
Opcode = (Opcode == SystemZ::AHI ? SystemZ::ASI : SystemZ::AGSI);
MachineInstr *BuiltMI =
BuildMI(*InsertPt->getParent(), InsertPt, MI.getDebugLoc(), get(Opcode))
.addFrameIndex(FrameIndex)
.addImm(0)
.addImm(MI.getOperand(2).getImm());
transferDeadCC(&MI, BuiltMI);
return BuiltMI;
}
if ((Opcode == SystemZ::ALFI && OpNum == 0 &&
isInt<8>((int32_t)MI.getOperand(2).getImm())) ||
(Opcode == SystemZ::ALGFI && OpNum == 0 &&
isInt<8>((int64_t)MI.getOperand(2).getImm()))) {
// AL(G)FI %reg, CONST -> AL(G)SI %mem, CONST
Opcode = (Opcode == SystemZ::ALFI ? SystemZ::ALSI : SystemZ::ALGSI);
MachineInstr *BuiltMI =
BuildMI(*InsertPt->getParent(), InsertPt, MI.getDebugLoc(), get(Opcode))
.addFrameIndex(FrameIndex)
.addImm(0)
.addImm((int8_t)MI.getOperand(2).getImm());
transferDeadCC(&MI, BuiltMI);
return BuiltMI;
}
if ((Opcode == SystemZ::SLFI && OpNum == 0 &&
isInt<8>((int32_t)-MI.getOperand(2).getImm())) ||
(Opcode == SystemZ::SLGFI && OpNum == 0 &&
isInt<8>((int64_t)-MI.getOperand(2).getImm()))) {
// SL(G)FI %reg, CONST -> AL(G)SI %mem, -CONST
Opcode = (Opcode == SystemZ::SLFI ? SystemZ::ALSI : SystemZ::ALGSI);
MachineInstr *BuiltMI =
BuildMI(*InsertPt->getParent(), InsertPt, MI.getDebugLoc(), get(Opcode))
.addFrameIndex(FrameIndex)
.addImm(0)
.addImm((int8_t)-MI.getOperand(2).getImm());
transferDeadCC(&MI, BuiltMI);
return BuiltMI;
}
if (Opcode == SystemZ::LGDR || Opcode == SystemZ::LDGR) {
bool Op0IsGPR = (Opcode == SystemZ::LGDR);
bool Op1IsGPR = (Opcode == SystemZ::LDGR);
// If we're spilling the destination of an LDGR or LGDR, store the
// source register instead.
if (OpNum == 0) {
unsigned StoreOpcode = Op1IsGPR ? SystemZ::STG : SystemZ::STD;
return BuildMI(*InsertPt->getParent(), InsertPt, MI.getDebugLoc(),
get(StoreOpcode))
.add(MI.getOperand(1))
.addFrameIndex(FrameIndex)
.addImm(0)
.addReg(0);
}
// If we're spilling the source of an LDGR or LGDR, load the
// destination register instead.
if (OpNum == 1) {
unsigned LoadOpcode = Op0IsGPR ? SystemZ::LG : SystemZ::LD;
return BuildMI(*InsertPt->getParent(), InsertPt, MI.getDebugLoc(),
get(LoadOpcode))
.add(MI.getOperand(0))
.addFrameIndex(FrameIndex)
.addImm(0)
.addReg(0);
}
}
// Look for cases where the source of a simple store or the destination
// of a simple load is being spilled. Try to use MVC instead.
//
// Although MVC is in practice a fast choice in these cases, it is still
// logically a bytewise copy. This means that we cannot use it if the
// load or store is volatile. We also wouldn't be able to use MVC if
// the two memories partially overlap, but that case cannot occur here,
// because we know that one of the memories is a full frame index.
//
// For performance reasons, we also want to avoid using MVC if the addresses
// might be equal. We don't worry about that case here, because spill slot
// coloring happens later, and because we have special code to remove
// MVCs that turn out to be redundant.
if (OpNum == 0 && MI.hasOneMemOperand()) {
MachineMemOperand *MMO = *MI.memoperands_begin();
if (MMO->getSize() == Size && !MMO->isVolatile()) {
// Handle conversion of loads.
if (isSimpleBD12Move(&MI, SystemZII::SimpleBDXLoad)) {
return BuildMI(*InsertPt->getParent(), InsertPt, MI.getDebugLoc(),
get(SystemZ::MVC))
.addFrameIndex(FrameIndex)
.addImm(0)
.addImm(Size)
.add(MI.getOperand(1))
.addImm(MI.getOperand(2).getImm())
.addMemOperand(MMO);
}
// Handle conversion of stores.
if (isSimpleBD12Move(&MI, SystemZII::SimpleBDXStore)) {
return BuildMI(*InsertPt->getParent(), InsertPt, MI.getDebugLoc(),
get(SystemZ::MVC))
.add(MI.getOperand(1))
.addImm(MI.getOperand(2).getImm())
.addImm(Size)
.addFrameIndex(FrameIndex)
.addImm(0)
.addMemOperand(MMO);
}
}
}
// If the spilled operand is the final one, try to change <INSN>R
// into <INSN>.
int MemOpcode = SystemZ::getMemOpcode(Opcode);
if (MemOpcode >= 0) {
unsigned NumOps = MI.getNumExplicitOperands();
if (OpNum == NumOps - 1) {
const MCInstrDesc &MemDesc = get(MemOpcode);
uint64_t AccessBytes = SystemZII::getAccessSize(MemDesc.TSFlags);
assert(AccessBytes != 0 && "Size of access should be known");
assert(AccessBytes <= Size && "Access outside the frame index");
uint64_t Offset = Size - AccessBytes;
MachineInstrBuilder MIB = BuildMI(*InsertPt->getParent(), InsertPt,
MI.getDebugLoc(), get(MemOpcode));
for (unsigned I = 0; I < OpNum; ++I)
MIB.add(MI.getOperand(I));
MIB.addFrameIndex(FrameIndex).addImm(Offset);
if (MemDesc.TSFlags & SystemZII::HasIndex)
MIB.addReg(0);
transferDeadCC(&MI, MIB);
return MIB;
}
}
return nullptr;
}
MachineInstr *SystemZInstrInfo::foldMemoryOperandImpl(
MachineFunction &MF, MachineInstr &MI, ArrayRef<unsigned> Ops,
MachineBasicBlock::iterator InsertPt, MachineInstr &LoadMI,
LiveIntervals *LIS) const {
return nullptr;
}
bool SystemZInstrInfo::expandPostRAPseudo(MachineInstr &MI) const {
switch (MI.getOpcode()) {
case SystemZ::L128:
splitMove(MI, SystemZ::LG);
return true;
case SystemZ::ST128:
splitMove(MI, SystemZ::STG);
return true;
case SystemZ::LX:
splitMove(MI, SystemZ::LD);
return true;
case SystemZ::STX:
splitMove(MI, SystemZ::STD);
return true;
case SystemZ::LBMux:
expandRXYPseudo(MI, SystemZ::LB, SystemZ::LBH);
return true;
case SystemZ::LHMux:
expandRXYPseudo(MI, SystemZ::LH, SystemZ::LHH);
return true;
case SystemZ::LLCRMux:
expandZExtPseudo(MI, SystemZ::LLCR, 8);
return true;
case SystemZ::LLHRMux:
expandZExtPseudo(MI, SystemZ::LLHR, 16);
return true;
case SystemZ::LLCMux:
expandRXYPseudo(MI, SystemZ::LLC, SystemZ::LLCH);
return true;
case SystemZ::LLHMux:
expandRXYPseudo(MI, SystemZ::LLH, SystemZ::LLHH);
return true;
case SystemZ::LMux:
expandRXYPseudo(MI, SystemZ::L, SystemZ::LFH);
return true;
case SystemZ::LOCMux:
expandLOCPseudo(MI, SystemZ::LOC, SystemZ::LOCFH);
return true;
case SystemZ::LOCHIMux:
expandLOCPseudo(MI, SystemZ::LOCHI, SystemZ::LOCHHI);
return true;
case SystemZ::LOCRMux:
expandLOCRPseudo(MI, SystemZ::LOCR, SystemZ::LOCFHR);
return true;
case SystemZ::STCMux:
expandRXYPseudo(MI, SystemZ::STC, SystemZ::STCH);
return true;
case SystemZ::STHMux:
expandRXYPseudo(MI, SystemZ::STH, SystemZ::STHH);
return true;
case SystemZ::STMux:
expandRXYPseudo(MI, SystemZ::ST, SystemZ::STFH);
return true;
case SystemZ::STOCMux:
expandLOCPseudo(MI, SystemZ::STOC, SystemZ::STOCFH);
return true;
case SystemZ::LHIMux:
expandRIPseudo(MI, SystemZ::LHI, SystemZ::IIHF, true);
return true;
case SystemZ::IIFMux:
expandRIPseudo(MI, SystemZ::IILF, SystemZ::IIHF, false);
return true;
case SystemZ::IILMux:
expandRIPseudo(MI, SystemZ::IILL, SystemZ::IIHL, false);
return true;
case SystemZ::IIHMux:
expandRIPseudo(MI, SystemZ::IILH, SystemZ::IIHH, false);
return true;
case SystemZ::NIFMux:
expandRIPseudo(MI, SystemZ::NILF, SystemZ::NIHF, false);
return true;
case SystemZ::NILMux:
expandRIPseudo(MI, SystemZ::NILL, SystemZ::NIHL, false);
return true;
case SystemZ::NIHMux:
expandRIPseudo(MI, SystemZ::NILH, SystemZ::NIHH, false);
return true;
case SystemZ::OIFMux:
expandRIPseudo(MI, SystemZ::OILF, SystemZ::OIHF, false);
return true;
case SystemZ::OILMux:
expandRIPseudo(MI, SystemZ::OILL, SystemZ::OIHL, false);
return true;
case SystemZ::OIHMux:
expandRIPseudo(MI, SystemZ::OILH, SystemZ::OIHH, false);
return true;
case SystemZ::XIFMux:
expandRIPseudo(MI, SystemZ::XILF, SystemZ::XIHF, false);
return true;
case SystemZ::TMLMux:
expandRIPseudo(MI, SystemZ::TMLL, SystemZ::TMHL, false);
return true;
case SystemZ::TMHMux:
expandRIPseudo(MI, SystemZ::TMLH, SystemZ::TMHH, false);
return true;
case SystemZ::AHIMux:
expandRIPseudo(MI, SystemZ::AHI, SystemZ::AIH, false);
return true;
case SystemZ::AHIMuxK:
expandRIEPseudo(MI, SystemZ::AHI, SystemZ::AHIK, SystemZ::AIH);
return true;
case SystemZ::AFIMux:
expandRIPseudo(MI, SystemZ::AFI, SystemZ::AIH, false);
return true;
case SystemZ::CHIMux:
expandRIPseudo(MI, SystemZ::CHI, SystemZ::CIH, false);
return true;
case SystemZ::CFIMux:
expandRIPseudo(MI, SystemZ::CFI, SystemZ::CIH, false);
return true;
case SystemZ::CLFIMux:
expandRIPseudo(MI, SystemZ::CLFI, SystemZ::CLIH, false);
return true;
case SystemZ::CMux:
expandRXYPseudo(MI, SystemZ::C, SystemZ::CHF);
return true;
case SystemZ::CLMux:
expandRXYPseudo(MI, SystemZ::CL, SystemZ::CLHF);
return true;
case SystemZ::RISBMux: {
bool DestIsHigh = isHighReg(MI.getOperand(0).getReg());
bool SrcIsHigh = isHighReg(MI.getOperand(2).getReg());
if (SrcIsHigh == DestIsHigh)
MI.setDesc(get(DestIsHigh ? SystemZ::RISBHH : SystemZ::RISBLL));
else {
MI.setDesc(get(DestIsHigh ? SystemZ::RISBHL : SystemZ::RISBLH));
MI.getOperand(5).setImm(MI.getOperand(5).getImm() ^ 32);
}
return true;
}
case SystemZ::ADJDYNALLOC:
splitAdjDynAlloc(MI);
return true;
case TargetOpcode::LOAD_STACK_GUARD:
expandLoadStackGuard(&MI);
return true;
default:
return false;
}
}
unsigned SystemZInstrInfo::getInstSizeInBytes(const MachineInstr &MI) const {
if (MI.getOpcode() == TargetOpcode::INLINEASM) {
const MachineFunction *MF = MI.getParent()->getParent();
const char *AsmStr = MI.getOperand(0).getSymbolName();
return getInlineAsmLength(AsmStr, *MF->getTarget().getMCAsmInfo());
}
return MI.getDesc().getSize();
}
SystemZII::Branch
SystemZInstrInfo::getBranchInfo(const MachineInstr &MI) const {
switch (MI.getOpcode()) {
case SystemZ::BR:
case SystemZ::BI:
case SystemZ::J:
case SystemZ::JG:
return SystemZII::Branch(SystemZII::BranchNormal, SystemZ::CCMASK_ANY,
SystemZ::CCMASK_ANY, &MI.getOperand(0));
case SystemZ::BRC:
case SystemZ::BRCL:
return SystemZII::Branch(SystemZII::BranchNormal, MI.getOperand(0).getImm(),
MI.getOperand(1).getImm(), &MI.getOperand(2));
case SystemZ::BRCT:
case SystemZ::BRCTH:
return SystemZII::Branch(SystemZII::BranchCT, SystemZ::CCMASK_ICMP,
SystemZ::CCMASK_CMP_NE, &MI.getOperand(2));
case SystemZ::BRCTG:
return SystemZII::Branch(SystemZII::BranchCTG, SystemZ::CCMASK_ICMP,
SystemZ::CCMASK_CMP_NE, &MI.getOperand(2));
case SystemZ::CIJ:
case SystemZ::CRJ:
return SystemZII::Branch(SystemZII::BranchC, SystemZ::CCMASK_ICMP,
MI.getOperand(2).getImm(), &MI.getOperand(3));
case SystemZ::CLIJ:
case SystemZ::CLRJ:
return SystemZII::Branch(SystemZII::BranchCL, SystemZ::CCMASK_ICMP,
MI.getOperand(2).getImm(), &MI.getOperand(3));
case SystemZ::CGIJ:
case SystemZ::CGRJ:
return SystemZII::Branch(SystemZII::BranchCG, SystemZ::CCMASK_ICMP,
MI.getOperand(2).getImm(), &MI.getOperand(3));
case SystemZ::CLGIJ:
case SystemZ::CLGRJ:
return SystemZII::Branch(SystemZII::BranchCLG, SystemZ::CCMASK_ICMP,
MI.getOperand(2).getImm(), &MI.getOperand(3));
default:
llvm_unreachable("Unrecognized branch opcode");
}
}
void SystemZInstrInfo::getLoadStoreOpcodes(const TargetRegisterClass *RC,
unsigned &LoadOpcode,
unsigned &StoreOpcode) const {
if (RC == &SystemZ::GR32BitRegClass || RC == &SystemZ::ADDR32BitRegClass) {
LoadOpcode = SystemZ::L;
StoreOpcode = SystemZ::ST;
} else if (RC == &SystemZ::GRH32BitRegClass) {
LoadOpcode = SystemZ::LFH;
StoreOpcode = SystemZ::STFH;
} else if (RC == &SystemZ::GRX32BitRegClass) {
LoadOpcode = SystemZ::LMux;
StoreOpcode = SystemZ::STMux;
} else if (RC == &SystemZ::GR64BitRegClass ||
RC == &SystemZ::ADDR64BitRegClass) {
LoadOpcode = SystemZ::LG;
StoreOpcode = SystemZ::STG;
} else if (RC == &SystemZ::GR128BitRegClass ||
RC == &SystemZ::ADDR128BitRegClass) {
LoadOpcode = SystemZ::L128;
StoreOpcode = SystemZ::ST128;
} else if (RC == &SystemZ::FP32BitRegClass) {
LoadOpcode = SystemZ::LE;
StoreOpcode = SystemZ::STE;
} else if (RC == &SystemZ::FP64BitRegClass) {
LoadOpcode = SystemZ::LD;
StoreOpcode = SystemZ::STD;
} else if (RC == &SystemZ::FP128BitRegClass) {
LoadOpcode = SystemZ::LX;
StoreOpcode = SystemZ::STX;
} else if (RC == &SystemZ::VR32BitRegClass) {
LoadOpcode = SystemZ::VL32;
StoreOpcode = SystemZ::VST32;
} else if (RC == &SystemZ::VR64BitRegClass) {
LoadOpcode = SystemZ::VL64;
StoreOpcode = SystemZ::VST64;
} else if (RC == &SystemZ::VF128BitRegClass ||
RC == &SystemZ::VR128BitRegClass) {
LoadOpcode = SystemZ::VL;
StoreOpcode = SystemZ::VST;
} else
llvm_unreachable("Unsupported regclass to load or store");
}
unsigned SystemZInstrInfo::getOpcodeForOffset(unsigned Opcode,
int64_t Offset) const {
const MCInstrDesc &MCID = get(Opcode);
int64_t Offset2 = (MCID.TSFlags & SystemZII::Is128Bit ? Offset + 8 : Offset);
if (isUInt<12>(Offset) && isUInt<12>(Offset2)) {
// Get the instruction to use for unsigned 12-bit displacements.
int Disp12Opcode = SystemZ::getDisp12Opcode(Opcode);
if (Disp12Opcode >= 0)
return Disp12Opcode;
// All address-related instructions can use unsigned 12-bit
// displacements.
return Opcode;
}
if (isInt<20>(Offset) && isInt<20>(Offset2)) {
// Get the instruction to use for signed 20-bit displacements.
int Disp20Opcode = SystemZ::getDisp20Opcode(Opcode);
if (Disp20Opcode >= 0)
return Disp20Opcode;
// Check whether Opcode allows signed 20-bit displacements.
if (MCID.TSFlags & SystemZII::Has20BitOffset)
return Opcode;
}
return 0;
}
unsigned SystemZInstrInfo::getLoadAndTest(unsigned Opcode) const {
switch (Opcode) {
case SystemZ::L: return SystemZ::LT;
case SystemZ::LY: return SystemZ::LT;
case SystemZ::LG: return SystemZ::LTG;
case SystemZ::LGF: return SystemZ::LTGF;
case SystemZ::LR: return SystemZ::LTR;
case SystemZ::LGFR: return SystemZ::LTGFR;
case SystemZ::LGR: return SystemZ::LTGR;
case SystemZ::LER: return SystemZ::LTEBR;
case SystemZ::LDR: return SystemZ::LTDBR;
case SystemZ::LXR: return SystemZ::LTXBR;
case SystemZ::LCDFR: return SystemZ::LCDBR;
case SystemZ::LPDFR: return SystemZ::LPDBR;
case SystemZ::LNDFR: return SystemZ::LNDBR;
case SystemZ::LCDFR_32: return SystemZ::LCEBR;
case SystemZ::LPDFR_32: return SystemZ::LPEBR;
case SystemZ::LNDFR_32: return SystemZ::LNEBR;
// On zEC12 we prefer to use RISBGN. But if there is a chance to
// actually use the condition code, we may turn it back into RISGB.
// Note that RISBG is not really a "load-and-test" instruction,
// but sets the same condition code values, so is OK to use here.
case SystemZ::RISBGN: return SystemZ::RISBG;
default: return 0;
}
}
// Return true if Mask matches the regexp 0*1+0*, given that zero masks
// have already been filtered out. Store the first set bit in LSB and
// the number of set bits in Length if so.
static bool isStringOfOnes(uint64_t Mask, unsigned &LSB, unsigned &Length) {
unsigned First = findFirstSet(Mask);
uint64_t Top = (Mask >> First) + 1;
if ((Top & -Top) == Top) {
LSB = First;
Length = findFirstSet(Top);
return true;
}
return false;
}
bool SystemZInstrInfo::isRxSBGMask(uint64_t Mask, unsigned BitSize,
unsigned &Start, unsigned &End) const {
// Reject trivial all-zero masks.
Mask &= allOnes(BitSize);
if (Mask == 0)
return false;
// Handle the 1+0+ or 0+1+0* cases. Start then specifies the index of
// the msb and End specifies the index of the lsb.
unsigned LSB, Length;
if (isStringOfOnes(Mask, LSB, Length)) {
Start = 63 - (LSB + Length - 1);
End = 63 - LSB;
return true;
}
// Handle the wrap-around 1+0+1+ cases. Start then specifies the msb
// of the low 1s and End specifies the lsb of the high 1s.
if (isStringOfOnes(Mask ^ allOnes(BitSize), LSB, Length)) {
assert(LSB > 0 && "Bottom bit must be set");
assert(LSB + Length < BitSize && "Top bit must be set");
Start = 63 - (LSB - 1);
End = 63 - (LSB + Length);
return true;
}
return false;
}
unsigned SystemZInstrInfo::getFusedCompare(unsigned Opcode,
SystemZII::FusedCompareType Type,
const MachineInstr *MI) const {
switch (Opcode) {
case SystemZ::CHI:
case SystemZ::CGHI:
if (!(MI && isInt<8>(MI->getOperand(1).getImm())))
return 0;
break;
case SystemZ::CLFI:
case SystemZ::CLGFI:
if (!(MI && isUInt<8>(MI->getOperand(1).getImm())))
return 0;
break;
case SystemZ::CL:
case SystemZ::CLG:
if (!STI.hasMiscellaneousExtensions())
return 0;
if (!(MI && MI->getOperand(3).getReg() == 0))
return 0;
break;
}
switch (Type) {
case SystemZII::CompareAndBranch:
switch (Opcode) {
case SystemZ::CR:
return SystemZ::CRJ;
case SystemZ::CGR:
return SystemZ::CGRJ;
case SystemZ::CHI:
return SystemZ::CIJ;
case SystemZ::CGHI:
return SystemZ::CGIJ;
case SystemZ::CLR:
return SystemZ::CLRJ;
case SystemZ::CLGR:
return SystemZ::CLGRJ;
case SystemZ::CLFI:
return SystemZ::CLIJ;
case SystemZ::CLGFI:
return SystemZ::CLGIJ;
default:
return 0;
}
case SystemZII::CompareAndReturn:
switch (Opcode) {
case SystemZ::CR:
return SystemZ::CRBReturn;
case SystemZ::CGR:
return SystemZ::CGRBReturn;
case SystemZ::CHI:
return SystemZ::CIBReturn;
case SystemZ::CGHI:
return SystemZ::CGIBReturn;
case SystemZ::CLR:
return SystemZ::CLRBReturn;
case SystemZ::CLGR:
return SystemZ::CLGRBReturn;
case SystemZ::CLFI:
return SystemZ::CLIBReturn;
case SystemZ::CLGFI:
return SystemZ::CLGIBReturn;
default:
return 0;
}
case SystemZII::CompareAndSibcall:
switch (Opcode) {
case SystemZ::CR:
return SystemZ::CRBCall;
case SystemZ::CGR:
return SystemZ::CGRBCall;
case SystemZ::CHI:
return SystemZ::CIBCall;
case SystemZ::CGHI:
return SystemZ::CGIBCall;
case SystemZ::CLR:
return SystemZ::CLRBCall;
case SystemZ::CLGR:
return SystemZ::CLGRBCall;
case SystemZ::CLFI:
return SystemZ::CLIBCall;
case SystemZ::CLGFI:
return SystemZ::CLGIBCall;
default:
return 0;
}
case SystemZII::CompareAndTrap:
switch (Opcode) {
case SystemZ::CR:
return SystemZ::CRT;
case SystemZ::CGR:
return SystemZ::CGRT;
case SystemZ::CHI:
return SystemZ::CIT;
case SystemZ::CGHI:
return SystemZ::CGIT;
case SystemZ::CLR:
return SystemZ::CLRT;
case SystemZ::CLGR:
return SystemZ::CLGRT;
case SystemZ::CLFI:
return SystemZ::CLFIT;
case SystemZ::CLGFI:
return SystemZ::CLGIT;
case SystemZ::CL:
return SystemZ::CLT;
case SystemZ::CLG:
return SystemZ::CLGT;
default:
return 0;
}
}
return 0;
}
unsigned SystemZInstrInfo::getLoadAndTrap(unsigned Opcode) const {
if (!STI.hasLoadAndTrap())
return 0;
switch (Opcode) {
case SystemZ::L:
case SystemZ::LY:
return SystemZ::LAT;
case SystemZ::LG:
return SystemZ::LGAT;
case SystemZ::LFH:
return SystemZ::LFHAT;
case SystemZ::LLGF:
return SystemZ::LLGFAT;
case SystemZ::LLGT:
return SystemZ::LLGTAT;
}
return 0;
}
void SystemZInstrInfo::loadImmediate(MachineBasicBlock &MBB,
MachineBasicBlock::iterator MBBI,
unsigned Reg, uint64_t Value) const {
DebugLoc DL = MBBI != MBB.end() ? MBBI->getDebugLoc() : DebugLoc();
unsigned Opcode;
if (isInt<16>(Value))
Opcode = SystemZ::LGHI;
else if (SystemZ::isImmLL(Value))
Opcode = SystemZ::LLILL;
else if (SystemZ::isImmLH(Value)) {
Opcode = SystemZ::LLILH;
Value >>= 16;
} else {
assert(isInt<32>(Value) && "Huge values not handled yet");
Opcode = SystemZ::LGFI;
}
BuildMI(MBB, MBBI, DL, get(Opcode), Reg).addImm(Value);
}
bool SystemZInstrInfo::
areMemAccessesTriviallyDisjoint(MachineInstr &MIa, MachineInstr &MIb,
AliasAnalysis *AA) const {
if (!MIa.hasOneMemOperand() || !MIb.hasOneMemOperand())
return false;
// If mem-operands show that the same address Value is used by both
// instructions, check for non-overlapping offsets and widths. Not
// sure if a register based analysis would be an improvement...
MachineMemOperand *MMOa = *MIa.memoperands_begin();
MachineMemOperand *MMOb = *MIb.memoperands_begin();
const Value *VALa = MMOa->getValue();
const Value *VALb = MMOb->getValue();
bool SameVal = (VALa && VALb && (VALa == VALb));
if (!SameVal) {
const PseudoSourceValue *PSVa = MMOa->getPseudoValue();
const PseudoSourceValue *PSVb = MMOb->getPseudoValue();
if (PSVa && PSVb && (PSVa == PSVb))
SameVal = true;
}
if (SameVal) {
int OffsetA = MMOa->getOffset(), OffsetB = MMOb->getOffset();
int WidthA = MMOa->getSize(), WidthB = MMOb->getSize();
int LowOffset = OffsetA < OffsetB ? OffsetA : OffsetB;
int HighOffset = OffsetA < OffsetB ? OffsetB : OffsetA;
int LowWidth = (LowOffset == OffsetA) ? WidthA : WidthB;
if (LowOffset + LowWidth <= HighOffset)
return true;
}
return false;
}
|