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
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
| //===-- SIRegisterInfo.cpp - SI Register Information ---------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
/// \file
/// SI implementation of the TargetRegisterInfo class.
//
//===----------------------------------------------------------------------===//
#include "SIRegisterInfo.h"
#include "AMDGPURegisterBankInfo.h"
#include "AMDGPUSubtarget.h"
#include "SIInstrInfo.h"
#include "SIMachineFunctionInfo.h"
#include "MCTargetDesc/AMDGPUInstPrinter.h"
#include "MCTargetDesc/AMDGPUMCTargetDesc.h"
#include "llvm/CodeGen/LiveIntervals.h"
#include "llvm/CodeGen/MachineDominators.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/RegisterScavenging.h"
#include "llvm/CodeGen/SlotIndexes.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/LLVMContext.h"
using namespace llvm;
static bool hasPressureSet(const int *PSets, unsigned PSetID) {
for (unsigned i = 0; PSets[i] != -1; ++i) {
if (PSets[i] == (int)PSetID)
return true;
}
return false;
}
void SIRegisterInfo::classifyPressureSet(unsigned PSetID, unsigned Reg,
BitVector &PressureSets) const {
for (MCRegUnitIterator U(Reg, this); U.isValid(); ++U) {
const int *PSets = getRegUnitPressureSets(*U);
if (hasPressureSet(PSets, PSetID)) {
PressureSets.set(PSetID);
break;
}
}
}
static cl::opt<bool> EnableSpillSGPRToVGPR(
"amdgpu-spill-sgpr-to-vgpr",
cl::desc("Enable spilling VGPRs to SGPRs"),
cl::ReallyHidden,
cl::init(true));
SIRegisterInfo::SIRegisterInfo(const GCNSubtarget &ST) :
AMDGPURegisterInfo(),
ST(ST),
SGPRPressureSets(getNumRegPressureSets()),
VGPRPressureSets(getNumRegPressureSets()),
AGPRPressureSets(getNumRegPressureSets()),
SpillSGPRToVGPR(EnableSpillSGPRToVGPR),
isWave32(ST.isWave32()) {
unsigned NumRegPressureSets = getNumRegPressureSets();
SGPRSetID = NumRegPressureSets;
VGPRSetID = NumRegPressureSets;
AGPRSetID = NumRegPressureSets;
for (unsigned i = 0; i < NumRegPressureSets; ++i) {
classifyPressureSet(i, AMDGPU::SGPR0, SGPRPressureSets);
classifyPressureSet(i, AMDGPU::VGPR0, VGPRPressureSets);
classifyPressureSet(i, AMDGPU::AGPR0, AGPRPressureSets);
}
// Determine the number of reg units for each pressure set.
std::vector<unsigned> PressureSetRegUnits(NumRegPressureSets, 0);
for (unsigned i = 0, e = getNumRegUnits(); i != e; ++i) {
const int *PSets = getRegUnitPressureSets(i);
for (unsigned j = 0; PSets[j] != -1; ++j) {
++PressureSetRegUnits[PSets[j]];
}
}
unsigned VGPRMax = 0, SGPRMax = 0, AGPRMax = 0;
for (unsigned i = 0; i < NumRegPressureSets; ++i) {
if (isVGPRPressureSet(i) && PressureSetRegUnits[i] > VGPRMax) {
VGPRSetID = i;
VGPRMax = PressureSetRegUnits[i];
continue;
}
if (isSGPRPressureSet(i) && PressureSetRegUnits[i] > SGPRMax) {
SGPRSetID = i;
SGPRMax = PressureSetRegUnits[i];
}
if (isAGPRPressureSet(i) && PressureSetRegUnits[i] > AGPRMax) {
AGPRSetID = i;
AGPRMax = PressureSetRegUnits[i];
continue;
}
}
assert(SGPRSetID < NumRegPressureSets &&
VGPRSetID < NumRegPressureSets &&
AGPRSetID < NumRegPressureSets);
}
unsigned SIRegisterInfo::reservedPrivateSegmentBufferReg(
const MachineFunction &MF) const {
unsigned BaseIdx = alignDown(ST.getMaxNumSGPRs(MF), 4) - 4;
unsigned BaseReg(AMDGPU::SGPR_32RegClass.getRegister(BaseIdx));
return getMatchingSuperReg(BaseReg, AMDGPU::sub0, &AMDGPU::SGPR_128RegClass);
}
static unsigned findPrivateSegmentWaveByteOffsetRegIndex(unsigned RegCount) {
unsigned Reg;
// Try to place it in a hole after PrivateSegmentBufferReg.
if (RegCount & 3) {
// We cannot put the segment buffer in (Idx - 4) ... (Idx - 1) due to
// alignment constraints, so we have a hole where can put the wave offset.
Reg = RegCount - 1;
} else {
// We can put the segment buffer in (Idx - 4) ... (Idx - 1) and put the
// wave offset before it.
Reg = RegCount - 5;
}
return Reg;
}
unsigned SIRegisterInfo::reservedPrivateSegmentWaveByteOffsetReg(
const MachineFunction &MF) const {
unsigned Reg = findPrivateSegmentWaveByteOffsetRegIndex(ST.getMaxNumSGPRs(MF));
return AMDGPU::SGPR_32RegClass.getRegister(Reg);
}
BitVector SIRegisterInfo::getReservedRegs(const MachineFunction &MF) const {
BitVector Reserved(getNumRegs());
// EXEC_LO and EXEC_HI could be allocated and used as regular register, but
// this seems likely to result in bugs, so I'm marking them as reserved.
reserveRegisterTuples(Reserved, AMDGPU::EXEC);
reserveRegisterTuples(Reserved, AMDGPU::FLAT_SCR);
// M0 has to be reserved so that llvm accepts it as a live-in into a block.
reserveRegisterTuples(Reserved, AMDGPU::M0);
// Reserve src_vccz, src_execz, src_scc.
reserveRegisterTuples(Reserved, AMDGPU::SRC_VCCZ);
reserveRegisterTuples(Reserved, AMDGPU::SRC_EXECZ);
reserveRegisterTuples(Reserved, AMDGPU::SRC_SCC);
// Reserve the memory aperture registers.
reserveRegisterTuples(Reserved, AMDGPU::SRC_SHARED_BASE);
reserveRegisterTuples(Reserved, AMDGPU::SRC_SHARED_LIMIT);
reserveRegisterTuples(Reserved, AMDGPU::SRC_PRIVATE_BASE);
reserveRegisterTuples(Reserved, AMDGPU::SRC_PRIVATE_LIMIT);
// Reserve src_pops_exiting_wave_id - support is not implemented in Codegen.
reserveRegisterTuples(Reserved, AMDGPU::SRC_POPS_EXITING_WAVE_ID);
// Reserve xnack_mask registers - support is not implemented in Codegen.
reserveRegisterTuples(Reserved, AMDGPU::XNACK_MASK);
// Reserve lds_direct register - support is not implemented in Codegen.
reserveRegisterTuples(Reserved, AMDGPU::LDS_DIRECT);
// Reserve Trap Handler registers - support is not implemented in Codegen.
reserveRegisterTuples(Reserved, AMDGPU::TBA);
reserveRegisterTuples(Reserved, AMDGPU::TMA);
reserveRegisterTuples(Reserved, AMDGPU::TTMP0_TTMP1);
reserveRegisterTuples(Reserved, AMDGPU::TTMP2_TTMP3);
reserveRegisterTuples(Reserved, AMDGPU::TTMP4_TTMP5);
reserveRegisterTuples(Reserved, AMDGPU::TTMP6_TTMP7);
reserveRegisterTuples(Reserved, AMDGPU::TTMP8_TTMP9);
reserveRegisterTuples(Reserved, AMDGPU::TTMP10_TTMP11);
reserveRegisterTuples(Reserved, AMDGPU::TTMP12_TTMP13);
reserveRegisterTuples(Reserved, AMDGPU::TTMP14_TTMP15);
// Reserve null register - it shall never be allocated
reserveRegisterTuples(Reserved, AMDGPU::SGPR_NULL);
// Disallow vcc_hi allocation in wave32. It may be allocated but most likely
// will result in bugs.
if (isWave32) {
Reserved.set(AMDGPU::VCC);
Reserved.set(AMDGPU::VCC_HI);
}
unsigned MaxNumSGPRs = ST.getMaxNumSGPRs(MF);
unsigned TotalNumSGPRs = AMDGPU::SGPR_32RegClass.getNumRegs();
for (unsigned i = MaxNumSGPRs; i < TotalNumSGPRs; ++i) {
unsigned Reg = AMDGPU::SGPR_32RegClass.getRegister(i);
reserveRegisterTuples(Reserved, Reg);
}
unsigned MaxNumVGPRs = ST.getMaxNumVGPRs(MF);
unsigned TotalNumVGPRs = AMDGPU::VGPR_32RegClass.getNumRegs();
for (unsigned i = MaxNumVGPRs; i < TotalNumVGPRs; ++i) {
unsigned Reg = AMDGPU::VGPR_32RegClass.getRegister(i);
reserveRegisterTuples(Reserved, Reg);
Reg = AMDGPU::AGPR_32RegClass.getRegister(i);
reserveRegisterTuples(Reserved, Reg);
}
// Reserve all the rest AGPRs if there are no instructions to use it.
if (!ST.hasMAIInsts()) {
for (unsigned i = 0; i < MaxNumVGPRs; ++i) {
unsigned Reg = AMDGPU::AGPR_32RegClass.getRegister(i);
reserveRegisterTuples(Reserved, Reg);
}
}
const SIMachineFunctionInfo *MFI = MF.getInfo<SIMachineFunctionInfo>();
unsigned ScratchWaveOffsetReg = MFI->getScratchWaveOffsetReg();
if (ScratchWaveOffsetReg != AMDGPU::NoRegister) {
// Reserve 1 SGPR for scratch wave offset in case we need to spill.
reserveRegisterTuples(Reserved, ScratchWaveOffsetReg);
}
unsigned ScratchRSrcReg = MFI->getScratchRSrcReg();
if (ScratchRSrcReg != AMDGPU::NoRegister) {
// Reserve 4 SGPRs for the scratch buffer resource descriptor in case we need
// to spill.
// TODO: May need to reserve a VGPR if doing LDS spilling.
reserveRegisterTuples(Reserved, ScratchRSrcReg);
assert(!isSubRegister(ScratchRSrcReg, ScratchWaveOffsetReg));
}
// We have to assume the SP is needed in case there are calls in the function,
// which is detected after the function is lowered. If we aren't really going
// to need SP, don't bother reserving it.
unsigned StackPtrReg = MFI->getStackPtrOffsetReg();
if (StackPtrReg != AMDGPU::NoRegister) {
reserveRegisterTuples(Reserved, StackPtrReg);
assert(!isSubRegister(ScratchRSrcReg, StackPtrReg));
}
unsigned FrameReg = MFI->getFrameOffsetReg();
if (FrameReg != AMDGPU::NoRegister) {
reserveRegisterTuples(Reserved, FrameReg);
assert(!isSubRegister(ScratchRSrcReg, FrameReg));
}
for (unsigned Reg : MFI->WWMReservedRegs) {
reserveRegisterTuples(Reserved, Reg);
}
// FIXME: Stop using reserved registers for this.
for (MCPhysReg Reg : MFI->getAGPRSpillVGPRs())
reserveRegisterTuples(Reserved, Reg);
for (MCPhysReg Reg : MFI->getVGPRSpillAGPRs())
reserveRegisterTuples(Reserved, Reg);
return Reserved;
}
bool SIRegisterInfo::canRealignStack(const MachineFunction &MF) const {
const SIMachineFunctionInfo *Info = MF.getInfo<SIMachineFunctionInfo>();
// On entry, the base address is 0, so it can't possibly need any more
// alignment.
// FIXME: Should be able to specify the entry frame alignment per calling
// convention instead.
if (Info->isEntryFunction())
return false;
return TargetRegisterInfo::canRealignStack(MF);
}
bool SIRegisterInfo::requiresRegisterScavenging(const MachineFunction &Fn) const {
const SIMachineFunctionInfo *Info = Fn.getInfo<SIMachineFunctionInfo>();
if (Info->isEntryFunction()) {
const MachineFrameInfo &MFI = Fn.getFrameInfo();
return MFI.hasStackObjects() || MFI.hasCalls();
}
// May need scavenger for dealing with callee saved registers.
return true;
}
bool SIRegisterInfo::requiresFrameIndexScavenging(
const MachineFunction &MF) const {
// Do not use frame virtual registers. They used to be used for SGPRs, but
// once we reach PrologEpilogInserter, we can no longer spill SGPRs. If the
// scavenger fails, we can increment/decrement the necessary SGPRs to avoid a
// spill.
return false;
}
bool SIRegisterInfo::requiresFrameIndexReplacementScavenging(
const MachineFunction &MF) const {
const MachineFrameInfo &MFI = MF.getFrameInfo();
return MFI.hasStackObjects();
}
bool SIRegisterInfo::requiresVirtualBaseRegisters(
const MachineFunction &) const {
// There are no special dedicated stack or frame pointers.
return true;
}
bool SIRegisterInfo::trackLivenessAfterRegAlloc(const MachineFunction &MF) const {
// This helps catch bugs as verifier errors.
return true;
}
int64_t SIRegisterInfo::getMUBUFInstrOffset(const MachineInstr *MI) const {
assert(SIInstrInfo::isMUBUF(*MI));
int OffIdx = AMDGPU::getNamedOperandIdx(MI->getOpcode(),
AMDGPU::OpName::offset);
return MI->getOperand(OffIdx).getImm();
}
int64_t SIRegisterInfo::getFrameIndexInstrOffset(const MachineInstr *MI,
int Idx) const {
if (!SIInstrInfo::isMUBUF(*MI))
return 0;
assert(Idx == AMDGPU::getNamedOperandIdx(MI->getOpcode(),
AMDGPU::OpName::vaddr) &&
"Should never see frame index on non-address operand");
return getMUBUFInstrOffset(MI);
}
bool SIRegisterInfo::needsFrameBaseReg(MachineInstr *MI, int64_t Offset) const {
if (!MI->mayLoadOrStore())
return false;
int64_t FullOffset = Offset + getMUBUFInstrOffset(MI);
return !isUInt<12>(FullOffset);
}
void SIRegisterInfo::materializeFrameBaseRegister(MachineBasicBlock *MBB,
unsigned BaseReg,
int FrameIdx,
int64_t Offset) const {
MachineBasicBlock::iterator Ins = MBB->begin();
DebugLoc DL; // Defaults to "unknown"
if (Ins != MBB->end())
DL = Ins->getDebugLoc();
MachineFunction *MF = MBB->getParent();
const SIInstrInfo *TII = ST.getInstrInfo();
if (Offset == 0) {
BuildMI(*MBB, Ins, DL, TII->get(AMDGPU::V_MOV_B32_e32), BaseReg)
.addFrameIndex(FrameIdx);
return;
}
MachineRegisterInfo &MRI = MF->getRegInfo();
Register OffsetReg = MRI.createVirtualRegister(&AMDGPU::SReg_32_XM0RegClass);
Register FIReg = MRI.createVirtualRegister(&AMDGPU::VGPR_32RegClass);
BuildMI(*MBB, Ins, DL, TII->get(AMDGPU::S_MOV_B32), OffsetReg)
.addImm(Offset);
BuildMI(*MBB, Ins, DL, TII->get(AMDGPU::V_MOV_B32_e32), FIReg)
.addFrameIndex(FrameIdx);
TII->getAddNoCarry(*MBB, Ins, DL, BaseReg)
.addReg(OffsetReg, RegState::Kill)
.addReg(FIReg)
.addImm(0); // clamp bit
}
void SIRegisterInfo::resolveFrameIndex(MachineInstr &MI, unsigned BaseReg,
int64_t Offset) const {
const SIInstrInfo *TII = ST.getInstrInfo();
#ifndef NDEBUG
// FIXME: Is it possible to be storing a frame index to itself?
bool SeenFI = false;
for (const MachineOperand &MO: MI.operands()) {
if (MO.isFI()) {
if (SeenFI)
llvm_unreachable("should not see multiple frame indices");
SeenFI = true;
}
}
#endif
MachineOperand *FIOp = TII->getNamedOperand(MI, AMDGPU::OpName::vaddr);
#ifndef NDEBUG
MachineBasicBlock *MBB = MI.getParent();
MachineFunction *MF = MBB->getParent();
#endif
assert(FIOp && FIOp->isFI() && "frame index must be address operand");
assert(TII->isMUBUF(MI));
assert(TII->getNamedOperand(MI, AMDGPU::OpName::soffset)->getReg() ==
MF->getInfo<SIMachineFunctionInfo>()->getStackPtrOffsetReg() &&
"should only be seeing stack pointer offset relative FrameIndex");
MachineOperand *OffsetOp = TII->getNamedOperand(MI, AMDGPU::OpName::offset);
int64_t NewOffset = OffsetOp->getImm() + Offset;
assert(isUInt<12>(NewOffset) && "offset should be legal");
FIOp->ChangeToRegister(BaseReg, false);
OffsetOp->setImm(NewOffset);
}
bool SIRegisterInfo::isFrameOffsetLegal(const MachineInstr *MI,
unsigned BaseReg,
int64_t Offset) const {
if (!SIInstrInfo::isMUBUF(*MI))
return false;
int64_t NewOffset = Offset + getMUBUFInstrOffset(MI);
return isUInt<12>(NewOffset);
}
const TargetRegisterClass *SIRegisterInfo::getPointerRegClass(
const MachineFunction &MF, unsigned Kind) const {
// This is inaccurate. It depends on the instruction and address space. The
// only place where we should hit this is for dealing with frame indexes /
// private accesses, so this is correct in that case.
return &AMDGPU::VGPR_32RegClass;
}
static unsigned getNumSubRegsForSpillOp(unsigned Op) {
switch (Op) {
case AMDGPU::SI_SPILL_S1024_SAVE:
case AMDGPU::SI_SPILL_S1024_RESTORE:
case AMDGPU::SI_SPILL_V1024_SAVE:
case AMDGPU::SI_SPILL_V1024_RESTORE:
case AMDGPU::SI_SPILL_A1024_SAVE:
case AMDGPU::SI_SPILL_A1024_RESTORE:
return 32;
case AMDGPU::SI_SPILL_S512_SAVE:
case AMDGPU::SI_SPILL_S512_RESTORE:
case AMDGPU::SI_SPILL_V512_SAVE:
case AMDGPU::SI_SPILL_V512_RESTORE:
case AMDGPU::SI_SPILL_A512_SAVE:
case AMDGPU::SI_SPILL_A512_RESTORE:
return 16;
case AMDGPU::SI_SPILL_S256_SAVE:
case AMDGPU::SI_SPILL_S256_RESTORE:
case AMDGPU::SI_SPILL_V256_SAVE:
case AMDGPU::SI_SPILL_V256_RESTORE:
return 8;
case AMDGPU::SI_SPILL_S160_SAVE:
case AMDGPU::SI_SPILL_S160_RESTORE:
case AMDGPU::SI_SPILL_V160_SAVE:
case AMDGPU::SI_SPILL_V160_RESTORE:
return 5;
case AMDGPU::SI_SPILL_S128_SAVE:
case AMDGPU::SI_SPILL_S128_RESTORE:
case AMDGPU::SI_SPILL_V128_SAVE:
case AMDGPU::SI_SPILL_V128_RESTORE:
case AMDGPU::SI_SPILL_A128_SAVE:
case AMDGPU::SI_SPILL_A128_RESTORE:
return 4;
case AMDGPU::SI_SPILL_S96_SAVE:
case AMDGPU::SI_SPILL_S96_RESTORE:
case AMDGPU::SI_SPILL_V96_SAVE:
case AMDGPU::SI_SPILL_V96_RESTORE:
return 3;
case AMDGPU::SI_SPILL_S64_SAVE:
case AMDGPU::SI_SPILL_S64_RESTORE:
case AMDGPU::SI_SPILL_V64_SAVE:
case AMDGPU::SI_SPILL_V64_RESTORE:
case AMDGPU::SI_SPILL_A64_SAVE:
case AMDGPU::SI_SPILL_A64_RESTORE:
return 2;
case AMDGPU::SI_SPILL_S32_SAVE:
case AMDGPU::SI_SPILL_S32_RESTORE:
case AMDGPU::SI_SPILL_V32_SAVE:
case AMDGPU::SI_SPILL_V32_RESTORE:
case AMDGPU::SI_SPILL_A32_SAVE:
case AMDGPU::SI_SPILL_A32_RESTORE:
return 1;
default: llvm_unreachable("Invalid spill opcode");
}
}
static int getOffsetMUBUFStore(unsigned Opc) {
switch (Opc) {
case AMDGPU::BUFFER_STORE_DWORD_OFFEN:
return AMDGPU::BUFFER_STORE_DWORD_OFFSET;
case AMDGPU::BUFFER_STORE_BYTE_OFFEN:
return AMDGPU::BUFFER_STORE_BYTE_OFFSET;
case AMDGPU::BUFFER_STORE_SHORT_OFFEN:
return AMDGPU::BUFFER_STORE_SHORT_OFFSET;
case AMDGPU::BUFFER_STORE_DWORDX2_OFFEN:
return AMDGPU::BUFFER_STORE_DWORDX2_OFFSET;
case AMDGPU::BUFFER_STORE_DWORDX4_OFFEN:
return AMDGPU::BUFFER_STORE_DWORDX4_OFFSET;
case AMDGPU::BUFFER_STORE_SHORT_D16_HI_OFFEN:
return AMDGPU::BUFFER_STORE_SHORT_D16_HI_OFFSET;
case AMDGPU::BUFFER_STORE_BYTE_D16_HI_OFFEN:
return AMDGPU::BUFFER_STORE_BYTE_D16_HI_OFFSET;
default:
return -1;
}
}
static int getOffsetMUBUFLoad(unsigned Opc) {
switch (Opc) {
case AMDGPU::BUFFER_LOAD_DWORD_OFFEN:
return AMDGPU::BUFFER_LOAD_DWORD_OFFSET;
case AMDGPU::BUFFER_LOAD_UBYTE_OFFEN:
return AMDGPU::BUFFER_LOAD_UBYTE_OFFSET;
case AMDGPU::BUFFER_LOAD_SBYTE_OFFEN:
return AMDGPU::BUFFER_LOAD_SBYTE_OFFSET;
case AMDGPU::BUFFER_LOAD_USHORT_OFFEN:
return AMDGPU::BUFFER_LOAD_USHORT_OFFSET;
case AMDGPU::BUFFER_LOAD_SSHORT_OFFEN:
return AMDGPU::BUFFER_LOAD_SSHORT_OFFSET;
case AMDGPU::BUFFER_LOAD_DWORDX2_OFFEN:
return AMDGPU::BUFFER_LOAD_DWORDX2_OFFSET;
case AMDGPU::BUFFER_LOAD_DWORDX4_OFFEN:
return AMDGPU::BUFFER_LOAD_DWORDX4_OFFSET;
case AMDGPU::BUFFER_LOAD_UBYTE_D16_OFFEN:
return AMDGPU::BUFFER_LOAD_UBYTE_D16_OFFSET;
case AMDGPU::BUFFER_LOAD_UBYTE_D16_HI_OFFEN:
return AMDGPU::BUFFER_LOAD_UBYTE_D16_HI_OFFSET;
case AMDGPU::BUFFER_LOAD_SBYTE_D16_OFFEN:
return AMDGPU::BUFFER_LOAD_SBYTE_D16_OFFSET;
case AMDGPU::BUFFER_LOAD_SBYTE_D16_HI_OFFEN:
return AMDGPU::BUFFER_LOAD_SBYTE_D16_HI_OFFSET;
case AMDGPU::BUFFER_LOAD_SHORT_D16_OFFEN:
return AMDGPU::BUFFER_LOAD_SHORT_D16_OFFSET;
case AMDGPU::BUFFER_LOAD_SHORT_D16_HI_OFFEN:
return AMDGPU::BUFFER_LOAD_SHORT_D16_HI_OFFSET;
default:
return -1;
}
}
static MachineInstrBuilder spillVGPRtoAGPR(const GCNSubtarget &ST,
MachineBasicBlock::iterator MI,
int Index,
unsigned Lane,
unsigned ValueReg,
bool IsKill) {
MachineBasicBlock *MBB = MI->getParent();
MachineFunction *MF = MI->getParent()->getParent();
SIMachineFunctionInfo *MFI = MF->getInfo<SIMachineFunctionInfo>();
const SIInstrInfo *TII = ST.getInstrInfo();
MCPhysReg Reg = MFI->getVGPRToAGPRSpill(Index, Lane);
if (Reg == AMDGPU::NoRegister)
return MachineInstrBuilder();
bool IsStore = MI->mayStore();
MachineRegisterInfo &MRI = MF->getRegInfo();
auto *TRI = static_cast<const SIRegisterInfo*>(MRI.getTargetRegisterInfo());
unsigned Dst = IsStore ? Reg : ValueReg;
unsigned Src = IsStore ? ValueReg : Reg;
unsigned Opc = (IsStore ^ TRI->isVGPR(MRI, Reg)) ? AMDGPU::V_ACCVGPR_WRITE_B32
: AMDGPU::V_ACCVGPR_READ_B32;
return BuildMI(*MBB, MI, MI->getDebugLoc(), TII->get(Opc), Dst)
.addReg(Src, getKillRegState(IsKill));
}
// This differs from buildSpillLoadStore by only scavenging a VGPR. It does not
// need to handle the case where an SGPR may need to be spilled while spilling.
static bool buildMUBUFOffsetLoadStore(const GCNSubtarget &ST,
MachineFrameInfo &MFI,
MachineBasicBlock::iterator MI,
int Index,
int64_t Offset) {
const SIInstrInfo *TII = ST.getInstrInfo();
MachineBasicBlock *MBB = MI->getParent();
const DebugLoc &DL = MI->getDebugLoc();
bool IsStore = MI->mayStore();
unsigned Opc = MI->getOpcode();
int LoadStoreOp = IsStore ?
getOffsetMUBUFStore(Opc) : getOffsetMUBUFLoad(Opc);
if (LoadStoreOp == -1)
return false;
const MachineOperand *Reg = TII->getNamedOperand(*MI, AMDGPU::OpName::vdata);
if (spillVGPRtoAGPR(ST, MI, Index, 0, Reg->getReg(), false).getInstr())
return true;
MachineInstrBuilder NewMI =
BuildMI(*MBB, MI, DL, TII->get(LoadStoreOp))
.add(*Reg)
.add(*TII->getNamedOperand(*MI, AMDGPU::OpName::srsrc))
.add(*TII->getNamedOperand(*MI, AMDGPU::OpName::soffset))
.addImm(Offset)
.addImm(0) // glc
.addImm(0) // slc
.addImm(0) // tfe
.addImm(0) // dlc
.addImm(0) // swz
.cloneMemRefs(*MI);
const MachineOperand *VDataIn = TII->getNamedOperand(*MI,
AMDGPU::OpName::vdata_in);
if (VDataIn)
NewMI.add(*VDataIn);
return true;
}
void SIRegisterInfo::buildSpillLoadStore(MachineBasicBlock::iterator MI,
unsigned LoadStoreOp,
int Index,
unsigned ValueReg,
bool IsKill,
unsigned ScratchRsrcReg,
unsigned ScratchOffsetReg,
int64_t InstOffset,
MachineMemOperand *MMO,
RegScavenger *RS) const {
MachineBasicBlock *MBB = MI->getParent();
MachineFunction *MF = MI->getParent()->getParent();
const SIInstrInfo *TII = ST.getInstrInfo();
const MachineFrameInfo &MFI = MF->getFrameInfo();
const MCInstrDesc &Desc = TII->get(LoadStoreOp);
const DebugLoc &DL = MI->getDebugLoc();
bool IsStore = Desc.mayStore();
bool Scavenged = false;
unsigned SOffset = ScratchOffsetReg;
const unsigned EltSize = 4;
const TargetRegisterClass *RC = getRegClassForReg(MF->getRegInfo(), ValueReg);
unsigned NumSubRegs = AMDGPU::getRegBitWidth(RC->getID()) / (EltSize * CHAR_BIT);
unsigned Size = NumSubRegs * EltSize;
int64_t Offset = InstOffset + MFI.getObjectOffset(Index);
int64_t ScratchOffsetRegDelta = 0;
unsigned Align = MFI.getObjectAlignment(Index);
const MachinePointerInfo &BasePtrInfo = MMO->getPointerInfo();
Register TmpReg =
hasAGPRs(RC) ? TII->getNamedOperand(*MI, AMDGPU::OpName::tmp)->getReg()
: Register();
assert((Offset % EltSize) == 0 && "unexpected VGPR spill offset");
if (!isUInt<12>(Offset + Size - EltSize)) {
SOffset = AMDGPU::NoRegister;
// We currently only support spilling VGPRs to EltSize boundaries, meaning
// we can simplify the adjustment of Offset here to just scale with
// WavefrontSize.
Offset *= ST.getWavefrontSize();
// We don't have access to the register scavenger if this function is called
// during PEI::scavengeFrameVirtualRegs().
if (RS)
SOffset = RS->scavengeRegister(&AMDGPU::SGPR_32RegClass, MI, 0, false);
if (SOffset == AMDGPU::NoRegister) {
// There are no free SGPRs, and since we are in the process of spilling
// VGPRs too. Since we need a VGPR in order to spill SGPRs (this is true
// on SI/CI and on VI it is true until we implement spilling using scalar
// stores), we have no way to free up an SGPR. Our solution here is to
// add the offset directly to the ScratchOffset register, and then
// subtract the offset after the spill to return ScratchOffset to it's
// original value.
SOffset = ScratchOffsetReg;
ScratchOffsetRegDelta = Offset;
} else {
Scavenged = true;
}
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_ADD_U32), SOffset)
.addReg(ScratchOffsetReg)
.addImm(Offset);
Offset = 0;
}
for (unsigned i = 0, e = NumSubRegs; i != e; ++i, Offset += EltSize) {
Register SubReg = NumSubRegs == 1
? Register(ValueReg)
: getSubReg(ValueReg, getSubRegFromChannel(i));
unsigned SOffsetRegState = 0;
unsigned SrcDstRegState = getDefRegState(!IsStore);
if (i + 1 == e) {
SOffsetRegState |= getKillRegState(Scavenged);
// The last implicit use carries the "Kill" flag.
SrcDstRegState |= getKillRegState(IsKill);
}
auto MIB = spillVGPRtoAGPR(ST, MI, Index, i, SubReg, IsKill);
if (!MIB.getInstr()) {
unsigned FinalReg = SubReg;
if (TmpReg != AMDGPU::NoRegister) {
if (IsStore)
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_ACCVGPR_READ_B32), TmpReg)
.addReg(SubReg, getKillRegState(IsKill));
SubReg = TmpReg;
}
MachinePointerInfo PInfo = BasePtrInfo.getWithOffset(EltSize * i);
MachineMemOperand *NewMMO
= MF->getMachineMemOperand(PInfo, MMO->getFlags(),
EltSize, MinAlign(Align, EltSize * i));
MIB = BuildMI(*MBB, MI, DL, Desc)
.addReg(SubReg, getDefRegState(!IsStore) | getKillRegState(IsKill))
.addReg(ScratchRsrcReg)
.addReg(SOffset, SOffsetRegState)
.addImm(Offset)
.addImm(0) // glc
.addImm(0) // slc
.addImm(0) // tfe
.addImm(0) // dlc
.addImm(0) // swz
.addMemOperand(NewMMO);
if (!IsStore && TmpReg != AMDGPU::NoRegister)
MIB = BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_ACCVGPR_WRITE_B32),
FinalReg)
.addReg(TmpReg, RegState::Kill);
}
if (NumSubRegs > 1)
MIB.addReg(ValueReg, RegState::Implicit | SrcDstRegState);
}
if (ScratchOffsetRegDelta != 0) {
// Subtract the offset we added to the ScratchOffset register.
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_SUB_U32), ScratchOffsetReg)
.addReg(ScratchOffsetReg)
.addImm(ScratchOffsetRegDelta);
}
}
bool SIRegisterInfo::spillSGPR(MachineBasicBlock::iterator MI,
int Index,
RegScavenger *RS,
bool OnlyToVGPR) const {
MachineBasicBlock *MBB = MI->getParent();
MachineFunction *MF = MBB->getParent();
SIMachineFunctionInfo *MFI = MF->getInfo<SIMachineFunctionInfo>();
DenseSet<unsigned> SGPRSpillVGPRDefinedSet;
ArrayRef<SIMachineFunctionInfo::SpilledReg> VGPRSpills
= MFI->getSGPRToVGPRSpills(Index);
bool SpillToVGPR = !VGPRSpills.empty();
if (OnlyToVGPR && !SpillToVGPR)
return false;
const SIInstrInfo *TII = ST.getInstrInfo();
Register SuperReg = MI->getOperand(0).getReg();
bool IsKill = MI->getOperand(0).isKill();
const DebugLoc &DL = MI->getDebugLoc();
MachineFrameInfo &FrameInfo = MF->getFrameInfo();
assert(SpillToVGPR || (SuperReg != MFI->getStackPtrOffsetReg() &&
SuperReg != MFI->getFrameOffsetReg() &&
SuperReg != MFI->getScratchWaveOffsetReg()));
assert(SuperReg != AMDGPU::M0 && "m0 should never spill");
unsigned M0CopyReg = AMDGPU::NoRegister;
unsigned EltSize = 4;
const TargetRegisterClass *RC = getPhysRegClass(SuperReg);
ArrayRef<int16_t> SplitParts = getRegSplitParts(RC, EltSize);
unsigned NumSubRegs = SplitParts.empty() ? 1 : SplitParts.size();
// Scavenged temporary VGPR to use. It must be scavenged once for any number
// of spilled subregs.
Register TmpVGPR;
// SubReg carries the "Kill" flag when SubReg == SuperReg.
unsigned SubKillState = getKillRegState((NumSubRegs == 1) && IsKill);
for (unsigned i = 0, e = NumSubRegs; i < e; ++i) {
Register SubReg =
NumSubRegs == 1 ? SuperReg : getSubReg(SuperReg, SplitParts[i]);
if (SpillToVGPR) {
SIMachineFunctionInfo::SpilledReg Spill = VGPRSpills[i];
// During SGPR spilling to VGPR, determine if the VGPR is defined. The
// only circumstance in which we say it is undefined is when it is the
// first spill to this VGPR in the first basic block.
bool VGPRDefined = true;
if (MBB == &MF->front())
VGPRDefined = !SGPRSpillVGPRDefinedSet.insert(Spill.VGPR).second;
// Mark the "old value of vgpr" input undef only if this is the first sgpr
// spill to this specific vgpr in the first basic block.
BuildMI(*MBB, MI, DL,
TII->getMCOpcodeFromPseudo(AMDGPU::V_WRITELANE_B32),
Spill.VGPR)
.addReg(SubReg, getKillRegState(IsKill))
.addImm(Spill.Lane)
.addReg(Spill.VGPR, VGPRDefined ? 0 : RegState::Undef);
// FIXME: Since this spills to another register instead of an actual
// frame index, we should delete the frame index when all references to
// it are fixed.
} else {
// XXX - Can to VGPR spill fail for some subregisters but not others?
if (OnlyToVGPR)
return false;
// Spill SGPR to a frame index.
if (!TmpVGPR.isValid())
TmpVGPR = RS->scavengeRegister(&AMDGPU::VGPR_32RegClass, MI, 0);
MachineInstrBuilder Mov
= BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_MOV_B32_e32), TmpVGPR)
.addReg(SubReg, SubKillState);
// There could be undef components of a spilled super register.
// TODO: Can we detect this and skip the spill?
if (NumSubRegs > 1) {
// The last implicit use of the SuperReg carries the "Kill" flag.
unsigned SuperKillState = 0;
if (i + 1 == e)
SuperKillState |= getKillRegState(IsKill);
Mov.addReg(SuperReg, RegState::Implicit | SuperKillState);
}
unsigned Align = FrameInfo.getObjectAlignment(Index);
MachinePointerInfo PtrInfo
= MachinePointerInfo::getFixedStack(*MF, Index, EltSize * i);
MachineMemOperand *MMO
= MF->getMachineMemOperand(PtrInfo, MachineMemOperand::MOStore,
EltSize, MinAlign(Align, EltSize * i));
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::SI_SPILL_V32_SAVE))
.addReg(TmpVGPR, RegState::Kill) // src
.addFrameIndex(Index) // vaddr
.addReg(MFI->getScratchRSrcReg()) // srrsrc
.addReg(MFI->getStackPtrOffsetReg()) // soffset
.addImm(i * 4) // offset
.addMemOperand(MMO);
}
}
if (M0CopyReg != AMDGPU::NoRegister) {
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::COPY), AMDGPU::M0)
.addReg(M0CopyReg, RegState::Kill);
}
MI->eraseFromParent();
MFI->addToSpilledSGPRs(NumSubRegs);
return true;
}
bool SIRegisterInfo::restoreSGPR(MachineBasicBlock::iterator MI,
int Index,
RegScavenger *RS,
bool OnlyToVGPR) const {
MachineFunction *MF = MI->getParent()->getParent();
MachineBasicBlock *MBB = MI->getParent();
SIMachineFunctionInfo *MFI = MF->getInfo<SIMachineFunctionInfo>();
ArrayRef<SIMachineFunctionInfo::SpilledReg> VGPRSpills
= MFI->getSGPRToVGPRSpills(Index);
bool SpillToVGPR = !VGPRSpills.empty();
if (OnlyToVGPR && !SpillToVGPR)
return false;
MachineFrameInfo &FrameInfo = MF->getFrameInfo();
const SIInstrInfo *TII = ST.getInstrInfo();
const DebugLoc &DL = MI->getDebugLoc();
Register SuperReg = MI->getOperand(0).getReg();
assert(SuperReg != AMDGPU::M0 && "m0 should never spill");
unsigned M0CopyReg = AMDGPU::NoRegister;
unsigned EltSize = 4;
const TargetRegisterClass *RC = getPhysRegClass(SuperReg);
ArrayRef<int16_t> SplitParts = getRegSplitParts(RC, EltSize);
unsigned NumSubRegs = SplitParts.empty() ? 1 : SplitParts.size();
Register TmpVGPR;
for (unsigned i = 0, e = NumSubRegs; i < e; ++i) {
Register SubReg =
NumSubRegs == 1 ? SuperReg : getSubReg(SuperReg, SplitParts[i]);
if (SpillToVGPR) {
SIMachineFunctionInfo::SpilledReg Spill = VGPRSpills[i];
auto MIB =
BuildMI(*MBB, MI, DL, TII->getMCOpcodeFromPseudo(AMDGPU::V_READLANE_B32),
SubReg)
.addReg(Spill.VGPR)
.addImm(Spill.Lane);
if (NumSubRegs > 1 && i == 0)
MIB.addReg(SuperReg, RegState::ImplicitDefine);
} else {
if (OnlyToVGPR)
return false;
// Restore SGPR from a stack slot.
// FIXME: We should use S_LOAD_DWORD here for VI.
if (!TmpVGPR.isValid())
TmpVGPR = RS->scavengeRegister(&AMDGPU::VGPR_32RegClass, MI, 0);
unsigned Align = FrameInfo.getObjectAlignment(Index);
MachinePointerInfo PtrInfo
= MachinePointerInfo::getFixedStack(*MF, Index, EltSize * i);
MachineMemOperand *MMO = MF->getMachineMemOperand(PtrInfo,
MachineMemOperand::MOLoad, EltSize,
MinAlign(Align, EltSize * i));
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::SI_SPILL_V32_RESTORE), TmpVGPR)
.addFrameIndex(Index) // vaddr
.addReg(MFI->getScratchRSrcReg()) // srsrc
.addReg(MFI->getStackPtrOffsetReg()) // soffset
.addImm(i * 4) // offset
.addMemOperand(MMO);
auto MIB =
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_READFIRSTLANE_B32), SubReg)
.addReg(TmpVGPR, RegState::Kill);
if (NumSubRegs > 1)
MIB.addReg(MI->getOperand(0).getReg(), RegState::ImplicitDefine);
}
}
if (M0CopyReg != AMDGPU::NoRegister) {
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::COPY), AMDGPU::M0)
.addReg(M0CopyReg, RegState::Kill);
}
MI->eraseFromParent();
return true;
}
/// Special case of eliminateFrameIndex. Returns true if the SGPR was spilled to
/// a VGPR and the stack slot can be safely eliminated when all other users are
/// handled.
bool SIRegisterInfo::eliminateSGPRToVGPRSpillFrameIndex(
MachineBasicBlock::iterator MI,
int FI,
RegScavenger *RS) const {
switch (MI->getOpcode()) {
case AMDGPU::SI_SPILL_S1024_SAVE:
case AMDGPU::SI_SPILL_S512_SAVE:
case AMDGPU::SI_SPILL_S256_SAVE:
case AMDGPU::SI_SPILL_S160_SAVE:
case AMDGPU::SI_SPILL_S128_SAVE:
case AMDGPU::SI_SPILL_S96_SAVE:
case AMDGPU::SI_SPILL_S64_SAVE:
case AMDGPU::SI_SPILL_S32_SAVE:
return spillSGPR(MI, FI, RS, true);
case AMDGPU::SI_SPILL_S1024_RESTORE:
case AMDGPU::SI_SPILL_S512_RESTORE:
case AMDGPU::SI_SPILL_S256_RESTORE:
case AMDGPU::SI_SPILL_S160_RESTORE:
case AMDGPU::SI_SPILL_S128_RESTORE:
case AMDGPU::SI_SPILL_S96_RESTORE:
case AMDGPU::SI_SPILL_S64_RESTORE:
case AMDGPU::SI_SPILL_S32_RESTORE:
return restoreSGPR(MI, FI, RS, true);
default:
llvm_unreachable("not an SGPR spill instruction");
}
}
void SIRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator MI,
int SPAdj, unsigned FIOperandNum,
RegScavenger *RS) const {
MachineFunction *MF = MI->getParent()->getParent();
MachineBasicBlock *MBB = MI->getParent();
SIMachineFunctionInfo *MFI = MF->getInfo<SIMachineFunctionInfo>();
MachineFrameInfo &FrameInfo = MF->getFrameInfo();
const SIInstrInfo *TII = ST.getInstrInfo();
DebugLoc DL = MI->getDebugLoc();
assert(SPAdj == 0 && "unhandled SP adjustment in call sequence?");
MachineOperand &FIOp = MI->getOperand(FIOperandNum);
int Index = MI->getOperand(FIOperandNum).getIndex();
Register FrameReg = getFrameRegister(*MF);
switch (MI->getOpcode()) {
// SGPR register spill
case AMDGPU::SI_SPILL_S1024_SAVE:
case AMDGPU::SI_SPILL_S512_SAVE:
case AMDGPU::SI_SPILL_S256_SAVE:
case AMDGPU::SI_SPILL_S160_SAVE:
case AMDGPU::SI_SPILL_S128_SAVE:
case AMDGPU::SI_SPILL_S96_SAVE:
case AMDGPU::SI_SPILL_S64_SAVE:
case AMDGPU::SI_SPILL_S32_SAVE: {
spillSGPR(MI, Index, RS);
break;
}
// SGPR register restore
case AMDGPU::SI_SPILL_S1024_RESTORE:
case AMDGPU::SI_SPILL_S512_RESTORE:
case AMDGPU::SI_SPILL_S256_RESTORE:
case AMDGPU::SI_SPILL_S160_RESTORE:
case AMDGPU::SI_SPILL_S128_RESTORE:
case AMDGPU::SI_SPILL_S96_RESTORE:
case AMDGPU::SI_SPILL_S64_RESTORE:
case AMDGPU::SI_SPILL_S32_RESTORE: {
restoreSGPR(MI, Index, RS);
break;
}
// VGPR register spill
case AMDGPU::SI_SPILL_V1024_SAVE:
case AMDGPU::SI_SPILL_V512_SAVE:
case AMDGPU::SI_SPILL_V256_SAVE:
case AMDGPU::SI_SPILL_V160_SAVE:
case AMDGPU::SI_SPILL_V128_SAVE:
case AMDGPU::SI_SPILL_V96_SAVE:
case AMDGPU::SI_SPILL_V64_SAVE:
case AMDGPU::SI_SPILL_V32_SAVE:
case AMDGPU::SI_SPILL_A1024_SAVE:
case AMDGPU::SI_SPILL_A512_SAVE:
case AMDGPU::SI_SPILL_A128_SAVE:
case AMDGPU::SI_SPILL_A64_SAVE:
case AMDGPU::SI_SPILL_A32_SAVE: {
const MachineOperand *VData = TII->getNamedOperand(*MI,
AMDGPU::OpName::vdata);
assert(TII->getNamedOperand(*MI, AMDGPU::OpName::soffset)->getReg() ==
MFI->getStackPtrOffsetReg());
buildSpillLoadStore(MI, AMDGPU::BUFFER_STORE_DWORD_OFFSET,
Index,
VData->getReg(), VData->isKill(),
TII->getNamedOperand(*MI, AMDGPU::OpName::srsrc)->getReg(),
FrameReg,
TII->getNamedOperand(*MI, AMDGPU::OpName::offset)->getImm(),
*MI->memoperands_begin(),
RS);
MFI->addToSpilledVGPRs(getNumSubRegsForSpillOp(MI->getOpcode()));
MI->eraseFromParent();
break;
}
case AMDGPU::SI_SPILL_V32_RESTORE:
case AMDGPU::SI_SPILL_V64_RESTORE:
case AMDGPU::SI_SPILL_V96_RESTORE:
case AMDGPU::SI_SPILL_V128_RESTORE:
case AMDGPU::SI_SPILL_V160_RESTORE:
case AMDGPU::SI_SPILL_V256_RESTORE:
case AMDGPU::SI_SPILL_V512_RESTORE:
case AMDGPU::SI_SPILL_V1024_RESTORE:
case AMDGPU::SI_SPILL_A32_RESTORE:
case AMDGPU::SI_SPILL_A64_RESTORE:
case AMDGPU::SI_SPILL_A128_RESTORE:
case AMDGPU::SI_SPILL_A512_RESTORE:
case AMDGPU::SI_SPILL_A1024_RESTORE: {
const MachineOperand *VData = TII->getNamedOperand(*MI,
AMDGPU::OpName::vdata);
assert(TII->getNamedOperand(*MI, AMDGPU::OpName::soffset)->getReg() ==
MFI->getStackPtrOffsetReg());
buildSpillLoadStore(MI, AMDGPU::BUFFER_LOAD_DWORD_OFFSET,
Index,
VData->getReg(), VData->isKill(),
TII->getNamedOperand(*MI, AMDGPU::OpName::srsrc)->getReg(),
FrameReg,
TII->getNamedOperand(*MI, AMDGPU::OpName::offset)->getImm(),
*MI->memoperands_begin(),
RS);
MI->eraseFromParent();
break;
}
default: {
const DebugLoc &DL = MI->getDebugLoc();
bool IsMUBUF = TII->isMUBUF(*MI);
if (!IsMUBUF && !MFI->isEntryFunction()) {
// Convert to an absolute stack address by finding the offset from the
// scratch wave base and scaling by the wave size.
//
// In an entry function/kernel the offset is already the absolute
// address relative to the frame register.
Register TmpDiffReg =
RS->scavengeRegister(&AMDGPU::SReg_32_XM0RegClass, MI, 0, false);
// If there's no free SGPR, in-place modify the FP
Register DiffReg = TmpDiffReg.isValid() ? TmpDiffReg : FrameReg;
bool IsCopy = MI->getOpcode() == AMDGPU::V_MOV_B32_e32;
Register ResultReg = IsCopy ?
MI->getOperand(0).getReg() :
RS->scavengeRegister(&AMDGPU::VGPR_32RegClass, MI, 0);
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_SUB_U32), DiffReg)
.addReg(FrameReg)
.addReg(MFI->getScratchWaveOffsetReg());
int64_t Offset = FrameInfo.getObjectOffset(Index);
if (Offset == 0) {
// XXX - This never happens because of emergency scavenging slot at 0?
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_LSHRREV_B32_e64), ResultReg)
.addImm(ST.getWavefrontSizeLog2())
.addReg(DiffReg);
} else {
if (auto MIB = TII->getAddNoCarry(*MBB, MI, DL, ResultReg, *RS)) {
Register ScaledReg =
RS->scavengeRegister(&AMDGPU::VGPR_32RegClass, MIB, 0);
BuildMI(*MBB, *MIB, DL, TII->get(AMDGPU::V_LSHRREV_B32_e64),
ScaledReg)
.addImm(ST.getWavefrontSizeLog2())
.addReg(DiffReg, RegState::Kill);
const bool IsVOP2 = MIB->getOpcode() == AMDGPU::V_ADD_U32_e32;
// TODO: Fold if use instruction is another add of a constant.
if (IsVOP2 || AMDGPU::isInlinableLiteral32(Offset, ST.hasInv2PiInlineImm())) {
// FIXME: This can fail
MIB.addImm(Offset);
MIB.addReg(ScaledReg, RegState::Kill);
if (!IsVOP2)
MIB.addImm(0); // clamp bit
} else {
Register ConstOffsetReg =
RS->scavengeRegister(&AMDGPU::SReg_32_XM0RegClass, MIB, 0, false);
// This should always be able to use the unused carry out.
assert(ConstOffsetReg && "this scavenge should not be able to fail");
BuildMI(*MBB, *MIB, DL, TII->get(AMDGPU::S_MOV_B32), ConstOffsetReg)
.addImm(Offset);
MIB.addReg(ConstOffsetReg, RegState::Kill);
MIB.addReg(ScaledReg, RegState::Kill);
MIB.addImm(0); // clamp bit
}
} else {
// We have to produce a carry out, and we there isn't a free SGPR
// pair for it. We can keep the whole computation on the SALU to
// avoid clobbering an additional register at the cost of an extra
// mov.
// We may have 1 free scratch SGPR even though a carry out is
// unavailable. Only one additional mov is needed.
Register TmpScaledReg =
RS->scavengeRegister(&AMDGPU::SReg_32_XM0RegClass, MI, 0, false);
Register ScaledReg = TmpScaledReg.isValid() ? TmpScaledReg : DiffReg;
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_LSHR_B32), ScaledReg)
.addReg(DiffReg, RegState::Kill)
.addImm(ST.getWavefrontSizeLog2());
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_ADD_U32), ScaledReg)
.addReg(ScaledReg, RegState::Kill)
.addImm(Offset);
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::COPY), ResultReg)
.addReg(ScaledReg, RegState::Kill);
// If there were truly no free SGPRs, we need to undo everything.
if (!TmpScaledReg.isValid()) {
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_SUB_U32), ScaledReg)
.addReg(ScaledReg, RegState::Kill)
.addImm(Offset);
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_LSHL_B32), ScaledReg)
.addReg(DiffReg, RegState::Kill)
.addImm(ST.getWavefrontSizeLog2());
}
}
}
if (!TmpDiffReg.isValid()) {
// Restore the FP.
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_ADD_U32), FrameReg)
.addReg(FrameReg)
.addReg(MFI->getScratchWaveOffsetReg());
}
// Don't introduce an extra copy if we're just materializing in a mov.
if (IsCopy)
MI->eraseFromParent();
else
FIOp.ChangeToRegister(ResultReg, false, false, true);
return;
}
if (IsMUBUF) {
// Disable offen so we don't need a 0 vgpr base.
assert(static_cast<int>(FIOperandNum) ==
AMDGPU::getNamedOperandIdx(MI->getOpcode(),
AMDGPU::OpName::vaddr));
assert(TII->getNamedOperand(*MI, AMDGPU::OpName::soffset)->getReg() ==
MFI->getStackPtrOffsetReg());
TII->getNamedOperand(*MI, AMDGPU::OpName::soffset)->setReg(FrameReg);
int64_t Offset = FrameInfo.getObjectOffset(Index);
int64_t OldImm
= TII->getNamedOperand(*MI, AMDGPU::OpName::offset)->getImm();
int64_t NewOffset = OldImm + Offset;
if (isUInt<12>(NewOffset) &&
buildMUBUFOffsetLoadStore(ST, FrameInfo, MI, Index, NewOffset)) {
MI->eraseFromParent();
return;
}
}
// If the offset is simply too big, don't convert to a scratch wave offset
// relative index.
int64_t Offset = FrameInfo.getObjectOffset(Index);
FIOp.ChangeToImmediate(Offset);
if (!TII->isImmOperandLegal(*MI, FIOperandNum, FIOp)) {
Register TmpReg = RS->scavengeRegister(&AMDGPU::VGPR_32RegClass, MI, 0);
BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_MOV_B32_e32), TmpReg)
.addImm(Offset);
FIOp.ChangeToRegister(TmpReg, false, false, true);
}
}
}
}
StringRef SIRegisterInfo::getRegAsmName(unsigned Reg) const {
return AMDGPUInstPrinter::getRegisterName(Reg);
}
// FIXME: This is very slow. It might be worth creating a map from physreg to
// register class.
const TargetRegisterClass *SIRegisterInfo::getPhysRegClass(unsigned Reg) const {
assert(!Register::isVirtualRegister(Reg));
static const TargetRegisterClass *const BaseClasses[] = {
&AMDGPU::VGPR_32RegClass,
&AMDGPU::SReg_32RegClass,
&AMDGPU::AGPR_32RegClass,
&AMDGPU::VReg_64RegClass,
&AMDGPU::SReg_64RegClass,
&AMDGPU::AReg_64RegClass,
&AMDGPU::VReg_96RegClass,
&AMDGPU::SReg_96RegClass,
&AMDGPU::VReg_128RegClass,
&AMDGPU::SReg_128RegClass,
&AMDGPU::AReg_128RegClass,
&AMDGPU::VReg_160RegClass,
&AMDGPU::SReg_160RegClass,
&AMDGPU::VReg_256RegClass,
&AMDGPU::SReg_256RegClass,
&AMDGPU::VReg_512RegClass,
&AMDGPU::SReg_512RegClass,
&AMDGPU::AReg_512RegClass,
&AMDGPU::SReg_1024RegClass,
&AMDGPU::VReg_1024RegClass,
&AMDGPU::AReg_1024RegClass,
&AMDGPU::SCC_CLASSRegClass,
&AMDGPU::Pseudo_SReg_32RegClass,
&AMDGPU::Pseudo_SReg_128RegClass,
};
for (const TargetRegisterClass *BaseClass : BaseClasses) {
if (BaseClass->contains(Reg)) {
return BaseClass;
}
}
return nullptr;
}
// TODO: It might be helpful to have some target specific flags in
// TargetRegisterClass to mark which classes are VGPRs to make this trivial.
bool SIRegisterInfo::hasVGPRs(const TargetRegisterClass *RC) const {
unsigned Size = getRegSizeInBits(*RC);
switch (Size) {
case 32:
return getCommonSubClass(&AMDGPU::VGPR_32RegClass, RC) != nullptr;
case 64:
return getCommonSubClass(&AMDGPU::VReg_64RegClass, RC) != nullptr;
case 96:
return getCommonSubClass(&AMDGPU::VReg_96RegClass, RC) != nullptr;
case 128:
return getCommonSubClass(&AMDGPU::VReg_128RegClass, RC) != nullptr;
case 160:
return getCommonSubClass(&AMDGPU::VReg_160RegClass, RC) != nullptr;
case 256:
return getCommonSubClass(&AMDGPU::VReg_256RegClass, RC) != nullptr;
case 512:
return getCommonSubClass(&AMDGPU::VReg_512RegClass, RC) != nullptr;
case 1024:
return getCommonSubClass(&AMDGPU::VReg_1024RegClass, RC) != nullptr;
case 1:
return getCommonSubClass(&AMDGPU::VReg_1RegClass, RC) != nullptr;
default:
assert(Size < 32 && "Invalid register class size");
return false;
}
}
bool SIRegisterInfo::hasAGPRs(const TargetRegisterClass *RC) const {
unsigned Size = getRegSizeInBits(*RC);
if (Size < 32)
return false;
switch (Size) {
case 32:
return getCommonSubClass(&AMDGPU::AGPR_32RegClass, RC) != nullptr;
case 64:
return getCommonSubClass(&AMDGPU::AReg_64RegClass, RC) != nullptr;
case 96:
return false;
case 128:
return getCommonSubClass(&AMDGPU::AReg_128RegClass, RC) != nullptr;
case 160:
case 256:
return false;
case 512:
return getCommonSubClass(&AMDGPU::AReg_512RegClass, RC) != nullptr;
case 1024:
return getCommonSubClass(&AMDGPU::AReg_1024RegClass, RC) != nullptr;
default:
llvm_unreachable("Invalid register class size");
}
}
const TargetRegisterClass *SIRegisterInfo::getEquivalentVGPRClass(
const TargetRegisterClass *SRC) const {
switch (getRegSizeInBits(*SRC)) {
case 32:
return &AMDGPU::VGPR_32RegClass;
case 64:
return &AMDGPU::VReg_64RegClass;
case 96:
return &AMDGPU::VReg_96RegClass;
case 128:
return &AMDGPU::VReg_128RegClass;
case 160:
return &AMDGPU::VReg_160RegClass;
case 256:
return &AMDGPU::VReg_256RegClass;
case 512:
return &AMDGPU::VReg_512RegClass;
case 1024:
return &AMDGPU::VReg_1024RegClass;
case 1:
return &AMDGPU::VReg_1RegClass;
default:
llvm_unreachable("Invalid register class size");
}
}
const TargetRegisterClass *SIRegisterInfo::getEquivalentAGPRClass(
const TargetRegisterClass *SRC) const {
switch (getRegSizeInBits(*SRC)) {
case 32:
return &AMDGPU::AGPR_32RegClass;
case 64:
return &AMDGPU::AReg_64RegClass;
case 128:
return &AMDGPU::AReg_128RegClass;
case 512:
return &AMDGPU::AReg_512RegClass;
case 1024:
return &AMDGPU::AReg_1024RegClass;
default:
llvm_unreachable("Invalid register class size");
}
}
const TargetRegisterClass *SIRegisterInfo::getEquivalentSGPRClass(
const TargetRegisterClass *VRC) const {
switch (getRegSizeInBits(*VRC)) {
case 32:
return &AMDGPU::SGPR_32RegClass;
case 64:
return &AMDGPU::SReg_64RegClass;
case 96:
return &AMDGPU::SReg_96RegClass;
case 128:
return &AMDGPU::SGPR_128RegClass;
case 160:
return &AMDGPU::SReg_160RegClass;
case 256:
return &AMDGPU::SReg_256RegClass;
case 512:
return &AMDGPU::SReg_512RegClass;
case 1024:
return &AMDGPU::SReg_1024RegClass;
default:
llvm_unreachable("Invalid register class size");
}
}
const TargetRegisterClass *SIRegisterInfo::getSubRegClass(
const TargetRegisterClass *RC, unsigned SubIdx) const {
if (SubIdx == AMDGPU::NoSubRegister)
return RC;
// We can assume that each lane corresponds to one 32-bit register.
unsigned Count = getSubRegIndexLaneMask(SubIdx).getNumLanes();
if (isSGPRClass(RC)) {
switch (Count) {
case 1:
return &AMDGPU::SGPR_32RegClass;
case 2:
return &AMDGPU::SReg_64RegClass;
case 3:
return &AMDGPU::SReg_96RegClass;
case 4:
return &AMDGPU::SGPR_128RegClass;
case 5:
return &AMDGPU::SReg_160RegClass;
case 8:
return &AMDGPU::SReg_256RegClass;
case 16:
return &AMDGPU::SReg_512RegClass;
case 32: /* fall-through */
default:
llvm_unreachable("Invalid sub-register class size");
}
} else if (hasAGPRs(RC)) {
switch (Count) {
case 1:
return &AMDGPU::AGPR_32RegClass;
case 2:
return &AMDGPU::AReg_64RegClass;
case 4:
return &AMDGPU::AReg_128RegClass;
case 16:
return &AMDGPU::AReg_512RegClass;
case 32: /* fall-through */
default:
llvm_unreachable("Invalid sub-register class size");
}
} else {
switch (Count) {
case 1:
return &AMDGPU::VGPR_32RegClass;
case 2:
return &AMDGPU::VReg_64RegClass;
case 3:
return &AMDGPU::VReg_96RegClass;
case 4:
return &AMDGPU::VReg_128RegClass;
case 5:
return &AMDGPU::VReg_160RegClass;
case 8:
return &AMDGPU::VReg_256RegClass;
case 16:
return &AMDGPU::VReg_512RegClass;
case 32: /* fall-through */
default:
llvm_unreachable("Invalid sub-register class size");
}
}
}
bool SIRegisterInfo::opCanUseInlineConstant(unsigned OpType) const {
if (OpType >= AMDGPU::OPERAND_REG_INLINE_AC_FIRST &&
OpType <= AMDGPU::OPERAND_REG_INLINE_AC_LAST)
return !ST.hasMFMAInlineLiteralBug();
return OpType >= AMDGPU::OPERAND_SRC_FIRST &&
OpType <= AMDGPU::OPERAND_SRC_LAST;
}
bool SIRegisterInfo::shouldRewriteCopySrc(
const TargetRegisterClass *DefRC,
unsigned DefSubReg,
const TargetRegisterClass *SrcRC,
unsigned SrcSubReg) const {
// We want to prefer the smallest register class possible, so we don't want to
// stop and rewrite on anything that looks like a subregister
// extract. Operations mostly don't care about the super register class, so we
// only want to stop on the most basic of copies between the same register
// class.
//
// e.g. if we have something like
// %0 = ...
// %1 = ...
// %2 = REG_SEQUENCE %0, sub0, %1, sub1, %2, sub2
// %3 = COPY %2, sub0
//
// We want to look through the COPY to find:
// => %3 = COPY %0
// Plain copy.
return getCommonSubClass(DefRC, SrcRC) != nullptr;
}
/// Returns a register that is not used at any point in the function.
/// If all registers are used, then this function will return
// AMDGPU::NoRegister.
unsigned
SIRegisterInfo::findUnusedRegister(const MachineRegisterInfo &MRI,
const TargetRegisterClass *RC,
const MachineFunction &MF) const {
for (unsigned Reg : *RC)
if (MRI.isAllocatable(Reg) && !MRI.isPhysRegUsed(Reg))
return Reg;
return AMDGPU::NoRegister;
}
ArrayRef<int16_t> SIRegisterInfo::getRegSplitParts(const TargetRegisterClass *RC,
unsigned EltSize) const {
if (EltSize == 4) {
static const int16_t Sub0_31[] = {
AMDGPU::sub0, AMDGPU::sub1, AMDGPU::sub2, AMDGPU::sub3,
AMDGPU::sub4, AMDGPU::sub5, AMDGPU::sub6, AMDGPU::sub7,
AMDGPU::sub8, AMDGPU::sub9, AMDGPU::sub10, AMDGPU::sub11,
AMDGPU::sub12, AMDGPU::sub13, AMDGPU::sub14, AMDGPU::sub15,
AMDGPU::sub16, AMDGPU::sub17, AMDGPU::sub18, AMDGPU::sub19,
AMDGPU::sub20, AMDGPU::sub21, AMDGPU::sub22, AMDGPU::sub23,
AMDGPU::sub24, AMDGPU::sub25, AMDGPU::sub26, AMDGPU::sub27,
AMDGPU::sub28, AMDGPU::sub29, AMDGPU::sub30, AMDGPU::sub31,
};
static const int16_t Sub0_15[] = {
AMDGPU::sub0, AMDGPU::sub1, AMDGPU::sub2, AMDGPU::sub3,
AMDGPU::sub4, AMDGPU::sub5, AMDGPU::sub6, AMDGPU::sub7,
AMDGPU::sub8, AMDGPU::sub9, AMDGPU::sub10, AMDGPU::sub11,
AMDGPU::sub12, AMDGPU::sub13, AMDGPU::sub14, AMDGPU::sub15,
};
static const int16_t Sub0_7[] = {
AMDGPU::sub0, AMDGPU::sub1, AMDGPU::sub2, AMDGPU::sub3,
AMDGPU::sub4, AMDGPU::sub5, AMDGPU::sub6, AMDGPU::sub7,
};
static const int16_t Sub0_4[] = {
AMDGPU::sub0, AMDGPU::sub1, AMDGPU::sub2, AMDGPU::sub3, AMDGPU::sub4,
};
static const int16_t Sub0_3[] = {
AMDGPU::sub0, AMDGPU::sub1, AMDGPU::sub2, AMDGPU::sub3,
};
static const int16_t Sub0_2[] = {
AMDGPU::sub0, AMDGPU::sub1, AMDGPU::sub2,
};
static const int16_t Sub0_1[] = {
AMDGPU::sub0, AMDGPU::sub1,
};
switch (AMDGPU::getRegBitWidth(*RC->MC)) {
case 32:
return {};
case 64:
return makeArrayRef(Sub0_1);
case 96:
return makeArrayRef(Sub0_2);
case 128:
return makeArrayRef(Sub0_3);
case 160:
return makeArrayRef(Sub0_4);
case 256:
return makeArrayRef(Sub0_7);
case 512:
return makeArrayRef(Sub0_15);
case 1024:
return makeArrayRef(Sub0_31);
default:
llvm_unreachable("unhandled register size");
}
}
if (EltSize == 8) {
static const int16_t Sub0_31_64[] = {
AMDGPU::sub0_sub1, AMDGPU::sub2_sub3,
AMDGPU::sub4_sub5, AMDGPU::sub6_sub7,
AMDGPU::sub8_sub9, AMDGPU::sub10_sub11,
AMDGPU::sub12_sub13, AMDGPU::sub14_sub15,
AMDGPU::sub16_sub17, AMDGPU::sub18_sub19,
AMDGPU::sub20_sub21, AMDGPU::sub22_sub23,
AMDGPU::sub24_sub25, AMDGPU::sub26_sub27,
AMDGPU::sub28_sub29, AMDGPU::sub30_sub31
};
static const int16_t Sub0_15_64[] = {
AMDGPU::sub0_sub1, AMDGPU::sub2_sub3,
AMDGPU::sub4_sub5, AMDGPU::sub6_sub7,
AMDGPU::sub8_sub9, AMDGPU::sub10_sub11,
AMDGPU::sub12_sub13, AMDGPU::sub14_sub15
};
static const int16_t Sub0_7_64[] = {
AMDGPU::sub0_sub1, AMDGPU::sub2_sub3,
AMDGPU::sub4_sub5, AMDGPU::sub6_sub7
};
static const int16_t Sub0_3_64[] = {
AMDGPU::sub0_sub1, AMDGPU::sub2_sub3
};
switch (AMDGPU::getRegBitWidth(*RC->MC)) {
case 64:
return {};
case 128:
return makeArrayRef(Sub0_3_64);
case 256:
return makeArrayRef(Sub0_7_64);
case 512:
return makeArrayRef(Sub0_15_64);
case 1024:
return makeArrayRef(Sub0_31_64);
default:
llvm_unreachable("unhandled register size");
}
}
if (EltSize == 16) {
static const int16_t Sub0_31_128[] = {
AMDGPU::sub0_sub1_sub2_sub3,
AMDGPU::sub4_sub5_sub6_sub7,
AMDGPU::sub8_sub9_sub10_sub11,
AMDGPU::sub12_sub13_sub14_sub15,
AMDGPU::sub16_sub17_sub18_sub19,
AMDGPU::sub20_sub21_sub22_sub23,
AMDGPU::sub24_sub25_sub26_sub27,
AMDGPU::sub28_sub29_sub30_sub31
};
static const int16_t Sub0_15_128[] = {
AMDGPU::sub0_sub1_sub2_sub3,
AMDGPU::sub4_sub5_sub6_sub7,
AMDGPU::sub8_sub9_sub10_sub11,
AMDGPU::sub12_sub13_sub14_sub15
};
static const int16_t Sub0_7_128[] = {
AMDGPU::sub0_sub1_sub2_sub3,
AMDGPU::sub4_sub5_sub6_sub7
};
switch (AMDGPU::getRegBitWidth(*RC->MC)) {
case 128:
return {};
case 256:
return makeArrayRef(Sub0_7_128);
case 512:
return makeArrayRef(Sub0_15_128);
case 1024:
return makeArrayRef(Sub0_31_128);
default:
llvm_unreachable("unhandled register size");
}
}
assert(EltSize == 32 && "unhandled elt size");
static const int16_t Sub0_31_256[] = {
AMDGPU::sub0_sub1_sub2_sub3_sub4_sub5_sub6_sub7,
AMDGPU::sub8_sub9_sub10_sub11_sub12_sub13_sub14_sub15,
AMDGPU::sub16_sub17_sub18_sub19_sub20_sub21_sub22_sub23,
AMDGPU::sub24_sub25_sub26_sub27_sub28_sub29_sub30_sub31
};
static const int16_t Sub0_15_256[] = {
AMDGPU::sub0_sub1_sub2_sub3_sub4_sub5_sub6_sub7,
AMDGPU::sub8_sub9_sub10_sub11_sub12_sub13_sub14_sub15
};
switch (AMDGPU::getRegBitWidth(*RC->MC)) {
case 256:
return {};
case 512:
return makeArrayRef(Sub0_15_256);
case 1024:
return makeArrayRef(Sub0_31_256);
default:
llvm_unreachable("unhandled register size");
}
}
const TargetRegisterClass*
SIRegisterInfo::getRegClassForReg(const MachineRegisterInfo &MRI,
unsigned Reg) const {
if (Register::isVirtualRegister(Reg))
return MRI.getRegClass(Reg);
return getPhysRegClass(Reg);
}
bool SIRegisterInfo::isVGPR(const MachineRegisterInfo &MRI,
unsigned Reg) const {
const TargetRegisterClass * RC = getRegClassForReg(MRI, Reg);
assert(RC && "Register class for the reg not found");
return hasVGPRs(RC);
}
bool SIRegisterInfo::isAGPR(const MachineRegisterInfo &MRI,
unsigned Reg) const {
const TargetRegisterClass * RC = getRegClassForReg(MRI, Reg);
assert(RC && "Register class for the reg not found");
return hasAGPRs(RC);
}
bool SIRegisterInfo::shouldCoalesce(MachineInstr *MI,
const TargetRegisterClass *SrcRC,
unsigned SubReg,
const TargetRegisterClass *DstRC,
unsigned DstSubReg,
const TargetRegisterClass *NewRC,
LiveIntervals &LIS) const {
unsigned SrcSize = getRegSizeInBits(*SrcRC);
unsigned DstSize = getRegSizeInBits(*DstRC);
unsigned NewSize = getRegSizeInBits(*NewRC);
// Do not increase size of registers beyond dword, we would need to allocate
// adjacent registers and constraint regalloc more than needed.
// Always allow dword coalescing.
if (SrcSize <= 32 || DstSize <= 32)
return true;
return NewSize <= DstSize || NewSize <= SrcSize;
}
unsigned SIRegisterInfo::getRegPressureLimit(const TargetRegisterClass *RC,
MachineFunction &MF) const {
const SIMachineFunctionInfo *MFI = MF.getInfo<SIMachineFunctionInfo>();
unsigned Occupancy = ST.getOccupancyWithLocalMemSize(MFI->getLDSSize(),
MF.getFunction());
switch (RC->getID()) {
default:
return AMDGPURegisterInfo::getRegPressureLimit(RC, MF);
case AMDGPU::VGPR_32RegClassID:
return std::min(ST.getMaxNumVGPRs(Occupancy), ST.getMaxNumVGPRs(MF));
case AMDGPU::SGPR_32RegClassID:
return std::min(ST.getMaxNumSGPRs(Occupancy, true), ST.getMaxNumSGPRs(MF));
}
}
unsigned SIRegisterInfo::getRegPressureSetLimit(const MachineFunction &MF,
unsigned Idx) const {
if (Idx == getVGPRPressureSet() || Idx == getAGPRPressureSet())
return getRegPressureLimit(&AMDGPU::VGPR_32RegClass,
const_cast<MachineFunction &>(MF));
if (Idx == getSGPRPressureSet())
return getRegPressureLimit(&AMDGPU::SGPR_32RegClass,
const_cast<MachineFunction &>(MF));
return AMDGPURegisterInfo::getRegPressureSetLimit(MF, Idx);
}
const int *SIRegisterInfo::getRegUnitPressureSets(unsigned RegUnit) const {
static const int Empty[] = { -1 };
if (hasRegUnit(AMDGPU::M0, RegUnit))
return Empty;
return AMDGPURegisterInfo::getRegUnitPressureSets(RegUnit);
}
unsigned SIRegisterInfo::getReturnAddressReg(const MachineFunction &MF) const {
// Not a callee saved register.
return AMDGPU::SGPR30_SGPR31;
}
const TargetRegisterClass *
SIRegisterInfo::getRegClassForSizeOnBank(unsigned Size,
const RegisterBank &RB,
const MachineRegisterInfo &MRI) const {
switch (Size) {
case 1: {
switch (RB.getID()) {
case AMDGPU::VGPRRegBankID:
return &AMDGPU::VGPR_32RegClass;
case AMDGPU::VCCRegBankID:
return isWave32 ?
&AMDGPU::SReg_32_XM0_XEXECRegClass : &AMDGPU::SReg_64_XEXECRegClass;
case AMDGPU::SGPRRegBankID:
return &AMDGPU::SReg_32RegClass;
case AMDGPU::SCCRegBankID:
// This needs to return an allocatable class, so don't bother returning
// the dummy SCC class.
//
// FIXME: This is a grotesque hack. We use SGPR_32 as an indication this
// was not an VCC bank value since we use the larger class SReg_32 for
// other values. These should all use SReg_32.
return &AMDGPU::SGPR_32RegClass;
default:
llvm_unreachable("unknown register bank");
}
}
case 32:
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VGPR_32RegClass :
&AMDGPU::SReg_32RegClass;
case 64:
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VReg_64RegClass :
&AMDGPU::SReg_64_XEXECRegClass;
case 96:
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VReg_96RegClass :
&AMDGPU::SReg_96RegClass;
case 128:
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VReg_128RegClass :
&AMDGPU::SGPR_128RegClass;
case 160:
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VReg_160RegClass :
&AMDGPU::SReg_160RegClass;
case 256:
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VReg_256RegClass :
&AMDGPU::SReg_256RegClass;
case 512:
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VReg_512RegClass :
&AMDGPU::SReg_512RegClass;
case 1024:
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VReg_1024RegClass :
&AMDGPU::SReg_1024RegClass;
default:
if (Size < 32)
return RB.getID() == AMDGPU::VGPRRegBankID ? &AMDGPU::VGPR_32RegClass :
&AMDGPU::SReg_32RegClass;
return nullptr;
}
}
const TargetRegisterClass *
SIRegisterInfo::getConstrainedRegClassForOperand(const MachineOperand &MO,
const MachineRegisterInfo &MRI) const {
const RegClassOrRegBank &RCOrRB = MRI.getRegClassOrRegBank(MO.getReg());
if (const RegisterBank *RB = RCOrRB.dyn_cast<const RegisterBank*>())
return getRegClassForTypeOnBank(MRI.getType(MO.getReg()), *RB, MRI);
const TargetRegisterClass *RC = RCOrRB.get<const TargetRegisterClass*>();
return getAllocatableClass(RC);
}
unsigned SIRegisterInfo::getVCC() const {
return isWave32 ? AMDGPU::VCC_LO : AMDGPU::VCC;
}
const TargetRegisterClass *
SIRegisterInfo::getRegClass(unsigned RCID) const {
switch ((int)RCID) {
case AMDGPU::SReg_1RegClassID:
return getBoolRC();
case AMDGPU::SReg_1_XEXECRegClassID:
return isWave32 ? &AMDGPU::SReg_32_XM0_XEXECRegClass
: &AMDGPU::SReg_64_XEXECRegClass;
case -1:
return nullptr;
default:
return AMDGPURegisterInfo::getRegClass(RCID);
}
}
// Find reaching register definition
MachineInstr *SIRegisterInfo::findReachingDef(unsigned Reg, unsigned SubReg,
MachineInstr &Use,
MachineRegisterInfo &MRI,
LiveIntervals *LIS) const {
auto &MDT = LIS->getAnalysis<MachineDominatorTree>();
SlotIndex UseIdx = LIS->getInstructionIndex(Use);
SlotIndex DefIdx;
if (Register::isVirtualRegister(Reg)) {
if (!LIS->hasInterval(Reg))
return nullptr;
LiveInterval &LI = LIS->getInterval(Reg);
LaneBitmask SubLanes = SubReg ? getSubRegIndexLaneMask(SubReg)
: MRI.getMaxLaneMaskForVReg(Reg);
VNInfo *V = nullptr;
if (LI.hasSubRanges()) {
for (auto &S : LI.subranges()) {
if ((S.LaneMask & SubLanes) == SubLanes) {
V = S.getVNInfoAt(UseIdx);
break;
}
}
} else {
V = LI.getVNInfoAt(UseIdx);
}
if (!V)
return nullptr;
DefIdx = V->def;
} else {
// Find last def.
for (MCRegUnitIterator Units(Reg, this); Units.isValid(); ++Units) {
LiveRange &LR = LIS->getRegUnit(*Units);
if (VNInfo *V = LR.getVNInfoAt(UseIdx)) {
if (!DefIdx.isValid() ||
MDT.dominates(LIS->getInstructionFromIndex(DefIdx),
LIS->getInstructionFromIndex(V->def)))
DefIdx = V->def;
} else {
return nullptr;
}
}
}
MachineInstr *Def = LIS->getInstructionFromIndex(DefIdx);
if (!Def || !MDT.dominates(Def, &Use))
return nullptr;
assert(Def->modifiesRegister(Reg, this));
return Def;
}
|