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
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
|
//===-- SelectionDAGISel.cpp - Implement the SelectionDAGISel class -------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This implements the SelectionDAGISel class.
//
//===----------------------------------------------------------------------===//
#define DEBUG_TYPE "isel"
#include "ScheduleDAGSDNodes.h"
#include "SelectionDAGBuilder.h"
#include "llvm/CodeGen/FunctionLoweringInfo.h"
#include "llvm/CodeGen/SelectionDAGISel.h"
#include "llvm/Analysis/AliasAnalysis.h"
#include "llvm/Analysis/DebugInfo.h"
#include "llvm/Constants.h"
#include "llvm/Function.h"
#include "llvm/InlineAsm.h"
#include "llvm/Instructions.h"
#include "llvm/Intrinsics.h"
#include "llvm/IntrinsicInst.h"
#include "llvm/LLVMContext.h"
#include "llvm/Module.h"
#include "llvm/CodeGen/FastISel.h"
#include "llvm/CodeGen/GCStrategy.h"
#include "llvm/CodeGen/GCMetadata.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/MachineModuleInfo.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/CodeGen/ScheduleHazardRecognizer.h"
#include "llvm/CodeGen/SchedulerRegistry.h"
#include "llvm/CodeGen/SelectionDAG.h"
#include "llvm/Target/TargetRegisterInfo.h"
#include "llvm/Target/TargetIntrinsicInfo.h"
#include "llvm/Target/TargetInstrInfo.h"
#include "llvm/Target/TargetLowering.h"
#include "llvm/Target/TargetMachine.h"
#include "llvm/Target/TargetOptions.h"
#include "llvm/Transforms/Utils/BasicBlockUtils.h"
#include "llvm/Support/Compiler.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/Timer.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/ADT/PostOrderIterator.h"
#include "llvm/ADT/Statistic.h"
#include <algorithm>
using namespace llvm;
STATISTIC(NumFastIselFailures, "Number of instructions fast isel failed on");
STATISTIC(NumFastIselBlocks, "Number of blocks selected entirely by fast isel");
STATISTIC(NumDAGBlocks, "Number of blocks selected using DAG");
STATISTIC(NumDAGIselRetries,"Number of times dag isel has to try another path");
#ifndef NDEBUG
STATISTIC(NumBBWithOutOfOrderLineInfo,
"Number of blocks with out of order line number info");
STATISTIC(NumMBBWithOutOfOrderLineInfo,
"Number of machine blocks with out of order line number info");
#endif
static cl::opt<bool>
EnableFastISelVerbose("fast-isel-verbose", cl::Hidden,
cl::desc("Enable verbose messages in the \"fast\" "
"instruction selector"));
static cl::opt<bool>
EnableFastISelAbort("fast-isel-abort", cl::Hidden,
cl::desc("Enable abort calls when \"fast\" instruction fails"));
#ifndef NDEBUG
static cl::opt<bool>
ViewDAGCombine1("view-dag-combine1-dags", cl::Hidden,
cl::desc("Pop up a window to show dags before the first "
"dag combine pass"));
static cl::opt<bool>
ViewLegalizeTypesDAGs("view-legalize-types-dags", cl::Hidden,
cl::desc("Pop up a window to show dags before legalize types"));
static cl::opt<bool>
ViewLegalizeDAGs("view-legalize-dags", cl::Hidden,
cl::desc("Pop up a window to show dags before legalize"));
static cl::opt<bool>
ViewDAGCombine2("view-dag-combine2-dags", cl::Hidden,
cl::desc("Pop up a window to show dags before the second "
"dag combine pass"));
static cl::opt<bool>
ViewDAGCombineLT("view-dag-combine-lt-dags", cl::Hidden,
cl::desc("Pop up a window to show dags before the post legalize types"
" dag combine pass"));
static cl::opt<bool>
ViewISelDAGs("view-isel-dags", cl::Hidden,
cl::desc("Pop up a window to show isel dags as they are selected"));
static cl::opt<bool>
ViewSchedDAGs("view-sched-dags", cl::Hidden,
cl::desc("Pop up a window to show sched dags as they are processed"));
static cl::opt<bool>
ViewSUnitDAGs("view-sunit-dags", cl::Hidden,
cl::desc("Pop up a window to show SUnit dags after they are processed"));
#else
static const bool ViewDAGCombine1 = false,
ViewLegalizeTypesDAGs = false, ViewLegalizeDAGs = false,
ViewDAGCombine2 = false,
ViewDAGCombineLT = false,
ViewISelDAGs = false, ViewSchedDAGs = false,
ViewSUnitDAGs = false;
#endif
//===---------------------------------------------------------------------===//
///
/// RegisterScheduler class - Track the registration of instruction schedulers.
///
//===---------------------------------------------------------------------===//
MachinePassRegistry RegisterScheduler::Registry;
//===---------------------------------------------------------------------===//
///
/// ISHeuristic command line option for instruction schedulers.
///
//===---------------------------------------------------------------------===//
static cl::opt<RegisterScheduler::FunctionPassCtor, false,
RegisterPassParser<RegisterScheduler> >
ISHeuristic("pre-RA-sched",
cl::init(&createDefaultScheduler),
cl::desc("Instruction schedulers available (before register"
" allocation):"));
static RegisterScheduler
defaultListDAGScheduler("default", "Best scheduler for the target",
createDefaultScheduler);
namespace llvm {
//===--------------------------------------------------------------------===//
/// createDefaultScheduler - This creates an instruction scheduler appropriate
/// for the target.
ScheduleDAGSDNodes* createDefaultScheduler(SelectionDAGISel *IS,
CodeGenOpt::Level OptLevel) {
const TargetLowering &TLI = IS->getTargetLowering();
if (OptLevel == CodeGenOpt::None)
return createSourceListDAGScheduler(IS, OptLevel);
if (TLI.getSchedulingPreference() == Sched::Latency)
return createTDListDAGScheduler(IS, OptLevel);
if (TLI.getSchedulingPreference() == Sched::RegPressure)
return createBURRListDAGScheduler(IS, OptLevel);
if (TLI.getSchedulingPreference() == Sched::Hybrid)
return createHybridListDAGScheduler(IS, OptLevel);
assert(TLI.getSchedulingPreference() == Sched::ILP &&
"Unknown sched type!");
return createILPListDAGScheduler(IS, OptLevel);
}
}
// EmitInstrWithCustomInserter - This method should be implemented by targets
// that mark instructions with the 'usesCustomInserter' flag. These
// instructions are special in various ways, which require special support to
// insert. The specified MachineInstr is created but not inserted into any
// basic blocks, and this method is called to expand it into a sequence of
// instructions, potentially also creating new basic blocks and control flow.
// When new basic blocks are inserted and the edges from MBB to its successors
// are modified, the method should insert pairs of <OldSucc, NewSucc> into the
// DenseMap.
MachineBasicBlock *
TargetLowering::EmitInstrWithCustomInserter(MachineInstr *MI,
MachineBasicBlock *MBB) const {
#ifndef NDEBUG
dbgs() << "If a target marks an instruction with "
"'usesCustomInserter', it must implement "
"TargetLowering::EmitInstrWithCustomInserter!";
#endif
llvm_unreachable(0);
return 0;
}
//===----------------------------------------------------------------------===//
// SelectionDAGISel code
//===----------------------------------------------------------------------===//
SelectionDAGISel::SelectionDAGISel(const TargetMachine &tm,
CodeGenOpt::Level OL) :
MachineFunctionPass(ID), TM(tm), TLI(*tm.getTargetLowering()),
FuncInfo(new FunctionLoweringInfo(TLI)),
CurDAG(new SelectionDAG(tm)),
SDB(new SelectionDAGBuilder(*CurDAG, *FuncInfo, OL)),
GFI(),
OptLevel(OL),
DAGSize(0) {
initializeGCModuleInfoPass(*PassRegistry::getPassRegistry());
initializeAliasAnalysisAnalysisGroup(*PassRegistry::getPassRegistry());
}
SelectionDAGISel::~SelectionDAGISel() {
delete SDB;
delete CurDAG;
delete FuncInfo;
}
void SelectionDAGISel::getAnalysisUsage(AnalysisUsage &AU) const {
AU.addRequired<AliasAnalysis>();
AU.addPreserved<AliasAnalysis>();
AU.addRequired<GCModuleInfo>();
AU.addPreserved<GCModuleInfo>();
MachineFunctionPass::getAnalysisUsage(AU);
}
/// FunctionCallsSetJmp - Return true if the function has a call to setjmp or
/// other function that gcc recognizes as "returning twice". This is used to
/// limit code-gen optimizations on the machine function.
///
/// FIXME: Remove after <rdar://problem/8031714> is fixed.
static bool FunctionCallsSetJmp(const Function *F) {
const Module *M = F->getParent();
static const char *ReturnsTwiceFns[] = {
"_setjmp",
"setjmp",
"sigsetjmp",
"setjmp_syscall",
"savectx",
"qsetjmp",
"vfork",
"getcontext"
};
#define NUM_RETURNS_TWICE_FNS sizeof(ReturnsTwiceFns) / sizeof(const char *)
for (unsigned I = 0; I < NUM_RETURNS_TWICE_FNS; ++I)
if (const Function *Callee = M->getFunction(ReturnsTwiceFns[I])) {
if (!Callee->use_empty())
for (Value::const_use_iterator
I = Callee->use_begin(), E = Callee->use_end();
I != E; ++I)
if (const CallInst *CI = dyn_cast<CallInst>(*I))
if (CI->getParent()->getParent() == F)
return true;
}
return false;
#undef NUM_RETURNS_TWICE_FNS
}
/// SplitCriticalSideEffectEdges - Look for critical edges with a PHI value that
/// may trap on it. In this case we have to split the edge so that the path
/// through the predecessor block that doesn't go to the phi block doesn't
/// execute the possibly trapping instruction.
///
/// This is required for correctness, so it must be done at -O0.
///
static void SplitCriticalSideEffectEdges(Function &Fn, Pass *SDISel) {
// Loop for blocks with phi nodes.
for (Function::iterator BB = Fn.begin(), E = Fn.end(); BB != E; ++BB) {
PHINode *PN = dyn_cast<PHINode>(BB->begin());
if (PN == 0) continue;
ReprocessBlock:
// For each block with a PHI node, check to see if any of the input values
// are potentially trapping constant expressions. Constant expressions are
// the only potentially trapping value that can occur as the argument to a
// PHI.
for (BasicBlock::iterator I = BB->begin(); (PN = dyn_cast<PHINode>(I)); ++I)
for (unsigned i = 0, e = PN->getNumIncomingValues(); i != e; ++i) {
ConstantExpr *CE = dyn_cast<ConstantExpr>(PN->getIncomingValue(i));
if (CE == 0 || !CE->canTrap()) continue;
// The only case we have to worry about is when the edge is critical.
// Since this block has a PHI Node, we assume it has multiple input
// edges: check to see if the pred has multiple successors.
BasicBlock *Pred = PN->getIncomingBlock(i);
if (Pred->getTerminator()->getNumSuccessors() == 1)
continue;
// Okay, we have to split this edge.
SplitCriticalEdge(Pred->getTerminator(),
GetSuccessorNumber(Pred, BB), SDISel, true);
goto ReprocessBlock;
}
}
}
bool SelectionDAGISel::runOnMachineFunction(MachineFunction &mf) {
// Do some sanity-checking on the command-line options.
assert((!EnableFastISelVerbose || EnableFastISel) &&
"-fast-isel-verbose requires -fast-isel");
assert((!EnableFastISelAbort || EnableFastISel) &&
"-fast-isel-abort requires -fast-isel");
const Function &Fn = *mf.getFunction();
const TargetInstrInfo &TII = *TM.getInstrInfo();
const TargetRegisterInfo &TRI = *TM.getRegisterInfo();
MF = &mf;
RegInfo = &MF->getRegInfo();
AA = &getAnalysis<AliasAnalysis>();
GFI = Fn.hasGC() ? &getAnalysis<GCModuleInfo>().getFunctionInfo(Fn) : 0;
DEBUG(dbgs() << "\n\n\n=== " << Fn.getName() << "\n");
SplitCriticalSideEffectEdges(const_cast<Function&>(Fn), this);
CurDAG->init(*MF);
FuncInfo->set(Fn, *MF);
SDB->init(GFI, *AA);
SelectAllBasicBlocks(Fn);
// If the first basic block in the function has live ins that need to be
// copied into vregs, emit the copies into the top of the block before
// emitting the code for the block.
MachineBasicBlock *EntryMBB = MF->begin();
RegInfo->EmitLiveInCopies(EntryMBB, TRI, TII);
DenseMap<unsigned, unsigned> LiveInMap;
if (!FuncInfo->ArgDbgValues.empty())
for (MachineRegisterInfo::livein_iterator LI = RegInfo->livein_begin(),
E = RegInfo->livein_end(); LI != E; ++LI)
if (LI->second)
LiveInMap.insert(std::make_pair(LI->first, LI->second));
// Insert DBG_VALUE instructions for function arguments to the entry block.
for (unsigned i = 0, e = FuncInfo->ArgDbgValues.size(); i != e; ++i) {
MachineInstr *MI = FuncInfo->ArgDbgValues[e-i-1];
unsigned Reg = MI->getOperand(0).getReg();
if (TargetRegisterInfo::isPhysicalRegister(Reg))
EntryMBB->insert(EntryMBB->begin(), MI);
else {
MachineInstr *Def = RegInfo->getVRegDef(Reg);
MachineBasicBlock::iterator InsertPos = Def;
// FIXME: VR def may not be in entry block.
Def->getParent()->insert(llvm::next(InsertPos), MI);
}
// If Reg is live-in then update debug info to track its copy in a vreg.
DenseMap<unsigned, unsigned>::iterator LDI = LiveInMap.find(Reg);
if (LDI != LiveInMap.end()) {
MachineInstr *Def = RegInfo->getVRegDef(LDI->second);
MachineBasicBlock::iterator InsertPos = Def;
const MDNode *Variable =
MI->getOperand(MI->getNumOperands()-1).getMetadata();
unsigned Offset = MI->getOperand(1).getImm();
// Def is never a terminator here, so it is ok to increment InsertPos.
BuildMI(*EntryMBB, ++InsertPos, MI->getDebugLoc(),
TII.get(TargetOpcode::DBG_VALUE))
.addReg(LDI->second, RegState::Debug)
.addImm(Offset).addMetadata(Variable);
// If this vreg is directly copied into an exported register then
// that COPY instructions also need DBG_VALUE, if it is the only
// user of LDI->second.
MachineInstr *CopyUseMI = NULL;
for (MachineRegisterInfo::use_iterator
UI = RegInfo->use_begin(LDI->second);
MachineInstr *UseMI = UI.skipInstruction();) {
if (UseMI->isDebugValue()) continue;
if (UseMI->isCopy() && !CopyUseMI && UseMI->getParent() == EntryMBB) {
CopyUseMI = UseMI; continue;
}
// Otherwise this is another use or second copy use.
CopyUseMI = NULL; break;
}
if (CopyUseMI) {
MachineInstr *NewMI =
BuildMI(*MF, CopyUseMI->getDebugLoc(),
TII.get(TargetOpcode::DBG_VALUE))
.addReg(CopyUseMI->getOperand(0).getReg(), RegState::Debug)
.addImm(Offset).addMetadata(Variable);
EntryMBB->insertAfter(CopyUseMI, NewMI);
}
}
}
// Determine if there are any calls in this machine function.
MachineFrameInfo *MFI = MF->getFrameInfo();
if (!MFI->hasCalls()) {
for (MachineFunction::const_iterator
I = MF->begin(), E = MF->end(); I != E; ++I) {
const MachineBasicBlock *MBB = I;
for (MachineBasicBlock::const_iterator
II = MBB->begin(), IE = MBB->end(); II != IE; ++II) {
const TargetInstrDesc &TID = TM.getInstrInfo()->get(II->getOpcode());
if ((TID.isCall() && !TID.isReturn()) ||
II->isStackAligningInlineAsm()) {
MFI->setHasCalls(true);
goto done;
}
}
}
done:;
}
// Determine if there is a call to setjmp in the machine function.
MF->setCallsSetJmp(FunctionCallsSetJmp(&Fn));
// Replace forward-declared registers with the registers containing
// the desired value.
MachineRegisterInfo &MRI = MF->getRegInfo();
for (DenseMap<unsigned, unsigned>::iterator
I = FuncInfo->RegFixups.begin(), E = FuncInfo->RegFixups.end();
I != E; ++I) {
unsigned From = I->first;
unsigned To = I->second;
// If To is also scheduled to be replaced, find what its ultimate
// replacement is.
for (;;) {
DenseMap<unsigned, unsigned>::iterator J =
FuncInfo->RegFixups.find(To);
if (J == E) break;
To = J->second;
}
// Replace it.
MRI.replaceRegWith(From, To);
}
// Release function-specific state. SDB and CurDAG are already cleared
// at this point.
FuncInfo->clear();
return true;
}
void
SelectionDAGISel::SelectBasicBlock(BasicBlock::const_iterator Begin,
BasicBlock::const_iterator End,
bool &HadTailCall) {
// Lower all of the non-terminator instructions. If a call is emitted
// as a tail call, cease emitting nodes for this block. Terminators
// are handled below.
for (BasicBlock::const_iterator I = Begin; I != End && !SDB->HasTailCall; ++I)
SDB->visit(*I);
// Make sure the root of the DAG is up-to-date.
CurDAG->setRoot(SDB->getControlRoot());
HadTailCall = SDB->HasTailCall;
SDB->clear();
// Final step, emit the lowered DAG as machine code.
CodeGenAndEmitDAG();
return;
}
void SelectionDAGISel::ComputeLiveOutVRegInfo() {
SmallPtrSet<SDNode*, 128> VisitedNodes;
SmallVector<SDNode*, 128> Worklist;
Worklist.push_back(CurDAG->getRoot().getNode());
APInt Mask;
APInt KnownZero;
APInt KnownOne;
do {
SDNode *N = Worklist.pop_back_val();
// If we've already seen this node, ignore it.
if (!VisitedNodes.insert(N))
continue;
// Otherwise, add all chain operands to the worklist.
for (unsigned i = 0, e = N->getNumOperands(); i != e; ++i)
if (N->getOperand(i).getValueType() == MVT::Other)
Worklist.push_back(N->getOperand(i).getNode());
// If this is a CopyToReg with a vreg dest, process it.
if (N->getOpcode() != ISD::CopyToReg)
continue;
unsigned DestReg = cast<RegisterSDNode>(N->getOperand(1))->getReg();
if (!TargetRegisterInfo::isVirtualRegister(DestReg))
continue;
// Ignore non-scalar or non-integer values.
SDValue Src = N->getOperand(2);
EVT SrcVT = Src.getValueType();
if (!SrcVT.isInteger() || SrcVT.isVector())
continue;
unsigned NumSignBits = CurDAG->ComputeNumSignBits(Src);
Mask = APInt::getAllOnesValue(SrcVT.getSizeInBits());
CurDAG->ComputeMaskedBits(Src, Mask, KnownZero, KnownOne);
FuncInfo->AddLiveOutRegInfo(DestReg, NumSignBits, KnownZero, KnownOne);
} while (!Worklist.empty());
}
void SelectionDAGISel::CodeGenAndEmitDAG() {
std::string GroupName;
if (TimePassesIsEnabled)
GroupName = "Instruction Selection and Scheduling";
std::string BlockName;
if (ViewDAGCombine1 || ViewLegalizeTypesDAGs || ViewLegalizeDAGs ||
ViewDAGCombine2 || ViewDAGCombineLT || ViewISelDAGs || ViewSchedDAGs ||
ViewSUnitDAGs)
BlockName = MF->getFunction()->getNameStr() + ":" +
FuncInfo->MBB->getBasicBlock()->getNameStr();
DEBUG(dbgs() << "Initial selection DAG:\n"; CurDAG->dump());
if (ViewDAGCombine1) CurDAG->viewGraph("dag-combine1 input for " + BlockName);
// Run the DAG combiner in pre-legalize mode.
{
NamedRegionTimer T("DAG Combining 1", GroupName, TimePassesIsEnabled);
CurDAG->Combine(Unrestricted, *AA, OptLevel);
}
DEBUG(dbgs() << "Optimized lowered selection DAG:\n"; CurDAG->dump());
// Second step, hack on the DAG until it only uses operations and types that
// the target supports.
if (ViewLegalizeTypesDAGs) CurDAG->viewGraph("legalize-types input for " +
BlockName);
bool Changed;
{
NamedRegionTimer T("Type Legalization", GroupName, TimePassesIsEnabled);
Changed = CurDAG->LegalizeTypes();
}
DEBUG(dbgs() << "Type-legalized selection DAG:\n"; CurDAG->dump());
if (Changed) {
if (ViewDAGCombineLT)
CurDAG->viewGraph("dag-combine-lt input for " + BlockName);
// Run the DAG combiner in post-type-legalize mode.
{
NamedRegionTimer T("DAG Combining after legalize types", GroupName,
TimePassesIsEnabled);
CurDAG->Combine(NoIllegalTypes, *AA, OptLevel);
}
DEBUG(dbgs() << "Optimized type-legalized selection DAG:\n";
CurDAG->dump());
}
{
NamedRegionTimer T("Vector Legalization", GroupName, TimePassesIsEnabled);
Changed = CurDAG->LegalizeVectors();
}
if (Changed) {
{
NamedRegionTimer T("Type Legalization 2", GroupName, TimePassesIsEnabled);
CurDAG->LegalizeTypes();
}
if (ViewDAGCombineLT)
CurDAG->viewGraph("dag-combine-lv input for " + BlockName);
// Run the DAG combiner in post-type-legalize mode.
{
NamedRegionTimer T("DAG Combining after legalize vectors", GroupName,
TimePassesIsEnabled);
CurDAG->Combine(NoIllegalOperations, *AA, OptLevel);
}
DEBUG(dbgs() << "Optimized vector-legalized selection DAG:\n";
CurDAG->dump());
}
if (ViewLegalizeDAGs) CurDAG->viewGraph("legalize input for " + BlockName);
{
NamedRegionTimer T("DAG Legalization", GroupName, TimePassesIsEnabled);
CurDAG->Legalize(OptLevel);
}
DEBUG(dbgs() << "Legalized selection DAG:\n"; CurDAG->dump());
if (ViewDAGCombine2) CurDAG->viewGraph("dag-combine2 input for " + BlockName);
// Run the DAG combiner in post-legalize mode.
{
NamedRegionTimer T("DAG Combining 2", GroupName, TimePassesIsEnabled);
CurDAG->Combine(NoIllegalOperations, *AA, OptLevel);
}
DEBUG(dbgs() << "Optimized legalized selection DAG:\n"; CurDAG->dump());
if (OptLevel != CodeGenOpt::None)
ComputeLiveOutVRegInfo();
if (ViewISelDAGs) CurDAG->viewGraph("isel input for " + BlockName);
// Third, instruction select all of the operations to machine code, adding the
// code to the MachineBasicBlock.
{
NamedRegionTimer T("Instruction Selection", GroupName, TimePassesIsEnabled);
DoInstructionSelection();
}
DEBUG(dbgs() << "Selected selection DAG:\n"; CurDAG->dump());
if (ViewSchedDAGs) CurDAG->viewGraph("scheduler input for " + BlockName);
// Schedule machine code.
ScheduleDAGSDNodes *Scheduler = CreateScheduler();
{
NamedRegionTimer T("Instruction Scheduling", GroupName,
TimePassesIsEnabled);
Scheduler->Run(CurDAG, FuncInfo->MBB, FuncInfo->InsertPt);
}
if (ViewSUnitDAGs) Scheduler->viewGraph();
// Emit machine code to BB. This can change 'BB' to the last block being
// inserted into.
MachineBasicBlock *FirstMBB = FuncInfo->MBB, *LastMBB;
{
NamedRegionTimer T("Instruction Creation", GroupName, TimePassesIsEnabled);
LastMBB = FuncInfo->MBB = Scheduler->EmitSchedule();
FuncInfo->InsertPt = Scheduler->InsertPos;
}
// If the block was split, make sure we update any references that are used to
// update PHI nodes later on.
if (FirstMBB != LastMBB)
SDB->UpdateSplitBlock(FirstMBB, LastMBB);
// Free the scheduler state.
{
NamedRegionTimer T("Instruction Scheduling Cleanup", GroupName,
TimePassesIsEnabled);
delete Scheduler;
}
// Free the SelectionDAG state, now that we're finished with it.
CurDAG->clear();
}
void SelectionDAGISel::DoInstructionSelection() {
DEBUG(errs() << "===== Instruction selection begins:\n");
PreprocessISelDAG();
// Select target instructions for the DAG.
{
// Number all nodes with a topological order and set DAGSize.
DAGSize = CurDAG->AssignTopologicalOrder();
// Create a dummy node (which is not added to allnodes), that adds
// a reference to the root node, preventing it from being deleted,
// and tracking any changes of the root.
HandleSDNode Dummy(CurDAG->getRoot());
ISelPosition = SelectionDAG::allnodes_iterator(CurDAG->getRoot().getNode());
++ISelPosition;
// The AllNodes list is now topological-sorted. Visit the
// nodes by starting at the end of the list (the root of the
// graph) and preceding back toward the beginning (the entry
// node).
while (ISelPosition != CurDAG->allnodes_begin()) {
SDNode *Node = --ISelPosition;
// Skip dead nodes. DAGCombiner is expected to eliminate all dead nodes,
// but there are currently some corner cases that it misses. Also, this
// makes it theoretically possible to disable the DAGCombiner.
if (Node->use_empty())
continue;
SDNode *ResNode = Select(Node);
// FIXME: This is pretty gross. 'Select' should be changed to not return
// anything at all and this code should be nuked with a tactical strike.
// If node should not be replaced, continue with the next one.
if (ResNode == Node || Node->getOpcode() == ISD::DELETED_NODE)
continue;
// Replace node.
if (ResNode)
ReplaceUses(Node, ResNode);
// If after the replacement this node is not used any more,
// remove this dead node.
if (Node->use_empty()) { // Don't delete EntryToken, etc.
ISelUpdater ISU(ISelPosition);
CurDAG->RemoveDeadNode(Node, &ISU);
}
}
CurDAG->setRoot(Dummy.getValue());
}
DEBUG(errs() << "===== Instruction selection ends:\n");
PostprocessISelDAG();
}
/// PrepareEHLandingPad - Emit an EH_LABEL, set up live-in registers, and
/// do other setup for EH landing-pad blocks.
void SelectionDAGISel::PrepareEHLandingPad() {
// Add a label to mark the beginning of the landing pad. Deletion of the
// landing pad can thus be detected via the MachineModuleInfo.
MCSymbol *Label = MF->getMMI().addLandingPad(FuncInfo->MBB);
const TargetInstrDesc &II = TM.getInstrInfo()->get(TargetOpcode::EH_LABEL);
BuildMI(*FuncInfo->MBB, FuncInfo->InsertPt, SDB->getCurDebugLoc(), II)
.addSym(Label);
// Mark exception register as live in.
unsigned Reg = TLI.getExceptionAddressRegister();
if (Reg) FuncInfo->MBB->addLiveIn(Reg);
// Mark exception selector register as live in.
Reg = TLI.getExceptionSelectorRegister();
if (Reg) FuncInfo->MBB->addLiveIn(Reg);
// FIXME: Hack around an exception handling flaw (PR1508): the personality
// function and list of typeids logically belong to the invoke (or, if you
// like, the basic block containing the invoke), and need to be associated
// with it in the dwarf exception handling tables. Currently however the
// information is provided by an intrinsic (eh.selector) that can be moved
// to unexpected places by the optimizers: if the unwind edge is critical,
// then breaking it can result in the intrinsics being in the successor of
// the landing pad, not the landing pad itself. This results
// in exceptions not being caught because no typeids are associated with
// the invoke. This may not be the only way things can go wrong, but it
// is the only way we try to work around for the moment.
const BasicBlock *LLVMBB = FuncInfo->MBB->getBasicBlock();
const BranchInst *Br = dyn_cast<BranchInst>(LLVMBB->getTerminator());
if (Br && Br->isUnconditional()) { // Critical edge?
BasicBlock::const_iterator I, E;
for (I = LLVMBB->begin(), E = --LLVMBB->end(); I != E; ++I)
if (isa<EHSelectorInst>(I))
break;
if (I == E)
// No catch info found - try to extract some from the successor.
CopyCatchInfo(Br->getSuccessor(0), LLVMBB, &MF->getMMI(), *FuncInfo);
}
}
bool SelectionDAGISel::TryToFoldFastISelLoad(const LoadInst *LI,
FastISel *FastIS) {
// Don't try to fold volatile loads. Target has to deal with alignment
// constraints.
if (LI->isVolatile()) return false;
// Figure out which vreg this is going into.
unsigned LoadReg = FastIS->getRegForValue(LI);
assert(LoadReg && "Load isn't already assigned a vreg? ");
// Check to see what the uses of this vreg are. If it has no uses, or more
// than one use (at the machine instr level) then we can't fold it.
MachineRegisterInfo::reg_iterator RI = RegInfo->reg_begin(LoadReg);
if (RI == RegInfo->reg_end())
return false;
// See if there is exactly one use of the vreg. If there are multiple uses,
// then the instruction got lowered to multiple machine instructions or the
// use of the loaded value ended up being multiple operands of the result, in
// either case, we can't fold this.
MachineRegisterInfo::reg_iterator PostRI = RI; ++PostRI;
if (PostRI != RegInfo->reg_end())
return false;
assert(RI.getOperand().isUse() &&
"The only use of the vreg must be a use, we haven't emitted the def!");
MachineInstr *User = &*RI;
// Set the insertion point properly. Folding the load can cause generation of
// other random instructions (like sign extends) for addressing modes, make
// sure they get inserted in a logical place before the new instruction.
FuncInfo->InsertPt = User;
FuncInfo->MBB = User->getParent();
// Ask the target to try folding the load.
return FastIS->TryToFoldLoad(User, RI.getOperandNo(), LI);
}
#ifndef NDEBUG
/// CheckLineNumbers - Check if basic block instructions follow source order
/// or not.
static void CheckLineNumbers(const BasicBlock *BB) {
unsigned Line = 0;
unsigned Col = 0;
for (BasicBlock::const_iterator BI = BB->begin(),
BE = BB->end(); BI != BE; ++BI) {
const DebugLoc DL = BI->getDebugLoc();
if (DL.isUnknown()) continue;
unsigned L = DL.getLine();
unsigned C = DL.getCol();
if (L < Line || (L == Line && C < Col)) {
++NumBBWithOutOfOrderLineInfo;
return;
}
Line = L;
Col = C;
}
}
/// CheckLineNumbers - Check if machine basic block instructions follow source
/// order or not.
static void CheckLineNumbers(const MachineBasicBlock *MBB) {
unsigned Line = 0;
unsigned Col = 0;
for (MachineBasicBlock::const_iterator MBI = MBB->begin(),
MBE = MBB->end(); MBI != MBE; ++MBI) {
const DebugLoc DL = MBI->getDebugLoc();
if (DL.isUnknown()) continue;
unsigned L = DL.getLine();
unsigned C = DL.getCol();
if (L < Line || (L == Line && C < Col)) {
++NumMBBWithOutOfOrderLineInfo;
return;
}
Line = L;
Col = C;
}
}
#endif
void SelectionDAGISel::SelectAllBasicBlocks(const Function &Fn) {
// Initialize the Fast-ISel state, if needed.
FastISel *FastIS = 0;
if (EnableFastISel)
FastIS = TLI.createFastISel(*FuncInfo);
// Iterate over all basic blocks in the function.
ReversePostOrderTraversal<const Function*> RPOT(&Fn);
for (ReversePostOrderTraversal<const Function*>::rpo_iterator
I = RPOT.begin(), E = RPOT.end(); I != E; ++I) {
const BasicBlock *LLVMBB = *I;
#ifndef NDEBUG
CheckLineNumbers(LLVMBB);
#endif
if (OptLevel != CodeGenOpt::None) {
bool AllPredsVisited = true;
for (const_pred_iterator PI = pred_begin(LLVMBB), PE = pred_end(LLVMBB);
PI != PE; ++PI) {
if (!FuncInfo->VisitedBBs.count(*PI)) {
AllPredsVisited = false;
break;
}
}
if (AllPredsVisited) {
for (BasicBlock::const_iterator I = LLVMBB->begin(), E = LLVMBB->end();
I != E && isa<PHINode>(I); ++I) {
FuncInfo->ComputePHILiveOutRegInfo(cast<PHINode>(I));
}
} else {
for (BasicBlock::const_iterator I = LLVMBB->begin(), E = LLVMBB->end();
I != E && isa<PHINode>(I); ++I) {
FuncInfo->InvalidatePHILiveOutRegInfo(cast<PHINode>(I));
}
}
FuncInfo->VisitedBBs.insert(LLVMBB);
}
FuncInfo->MBB = FuncInfo->MBBMap[LLVMBB];
FuncInfo->InsertPt = FuncInfo->MBB->getFirstNonPHI();
BasicBlock::const_iterator const Begin = LLVMBB->getFirstNonPHI();
BasicBlock::const_iterator const End = LLVMBB->end();
BasicBlock::const_iterator BI = End;
FuncInfo->InsertPt = FuncInfo->MBB->getFirstNonPHI();
// Setup an EH landing-pad block.
if (FuncInfo->MBB->isLandingPad())
PrepareEHLandingPad();
// Lower any arguments needed in this block if this is the entry block.
if (LLVMBB == &Fn.getEntryBlock())
LowerArguments(LLVMBB);
// Before doing SelectionDAG ISel, see if FastISel has been requested.
if (FastIS) {
FastIS->startNewBlock();
// Emit code for any incoming arguments. This must happen before
// beginning FastISel on the entry block.
if (LLVMBB == &Fn.getEntryBlock()) {
CurDAG->setRoot(SDB->getControlRoot());
SDB->clear();
CodeGenAndEmitDAG();
// If we inserted any instructions at the beginning, make a note of
// where they are, so we can be sure to emit subsequent instructions
// after them.
if (FuncInfo->InsertPt != FuncInfo->MBB->begin())
FastIS->setLastLocalValue(llvm::prior(FuncInfo->InsertPt));
else
FastIS->setLastLocalValue(0);
}
// Do FastISel on as many instructions as possible.
for (; BI != Begin; --BI) {
const Instruction *Inst = llvm::prior(BI);
// If we no longer require this instruction, skip it.
if (!Inst->mayWriteToMemory() &&
!isa<TerminatorInst>(Inst) &&
!isa<DbgInfoIntrinsic>(Inst) &&
!FuncInfo->isExportedInst(Inst))
continue;
// Bottom-up: reset the insert pos at the top, after any local-value
// instructions.
FastIS->recomputeInsertPt();
// Try to select the instruction with FastISel.
if (FastIS->SelectInstruction(Inst)) {
// If fast isel succeeded, check to see if there is a single-use
// non-volatile load right before the selected instruction, and see if
// the load is used by the instruction. If so, try to fold it.
const Instruction *BeforeInst = 0;
if (Inst != Begin)
BeforeInst = llvm::prior(llvm::prior(BI));
if (BeforeInst && isa<LoadInst>(BeforeInst) &&
BeforeInst->hasOneUse() && *BeforeInst->use_begin() == Inst &&
TryToFoldFastISelLoad(cast<LoadInst>(BeforeInst), FastIS))
--BI; // If we succeeded, don't re-select the load.
continue;
}
// Then handle certain instructions as single-LLVM-Instruction blocks.
if (isa<CallInst>(Inst)) {
++NumFastIselFailures;
if (EnableFastISelVerbose || EnableFastISelAbort) {
dbgs() << "FastISel missed call: ";
Inst->dump();
}
if (!Inst->getType()->isVoidTy() && !Inst->use_empty()) {
unsigned &R = FuncInfo->ValueMap[Inst];
if (!R)
R = FuncInfo->CreateRegs(Inst->getType());
}
bool HadTailCall = false;
SelectBasicBlock(Inst, BI, HadTailCall);
// If the call was emitted as a tail call, we're done with the block.
if (HadTailCall) {
--BI;
break;
}
continue;
}
// Otherwise, give up on FastISel for the rest of the block.
// For now, be a little lenient about non-branch terminators.
if (!isa<TerminatorInst>(Inst) || isa<BranchInst>(Inst)) {
++NumFastIselFailures;
if (EnableFastISelVerbose || EnableFastISelAbort) {
dbgs() << "FastISel miss: ";
Inst->dump();
}
if (EnableFastISelAbort)
// The "fast" selector couldn't handle something and bailed.
// For the purpose of debugging, just abort.
llvm_unreachable("FastISel didn't select the entire block");
}
break;
}
FastIS->recomputeInsertPt();
}
if (Begin != BI)
++NumDAGBlocks;
else
++NumFastIselBlocks;
// Run SelectionDAG instruction selection on the remainder of the block
// not handled by FastISel. If FastISel is not run, this is the entire
// block.
bool HadTailCall;
SelectBasicBlock(Begin, BI, HadTailCall);
FinishBasicBlock();
FuncInfo->PHINodesToUpdate.clear();
}
delete FastIS;
#ifndef NDEBUG
for (MachineFunction::const_iterator MBI = MF->begin(), MBE = MF->end();
MBI != MBE; ++MBI)
CheckLineNumbers(MBI);
#endif
}
void
SelectionDAGISel::FinishBasicBlock() {
DEBUG(dbgs() << "Total amount of phi nodes to update: "
<< FuncInfo->PHINodesToUpdate.size() << "\n";
for (unsigned i = 0, e = FuncInfo->PHINodesToUpdate.size(); i != e; ++i)
dbgs() << "Node " << i << " : ("
<< FuncInfo->PHINodesToUpdate[i].first
<< ", " << FuncInfo->PHINodesToUpdate[i].second << ")\n");
// Next, now that we know what the last MBB the LLVM BB expanded is, update
// PHI nodes in successors.
if (SDB->SwitchCases.empty() &&
SDB->JTCases.empty() &&
SDB->BitTestCases.empty()) {
for (unsigned i = 0, e = FuncInfo->PHINodesToUpdate.size(); i != e; ++i) {
MachineInstr *PHI = FuncInfo->PHINodesToUpdate[i].first;
assert(PHI->isPHI() &&
"This is not a machine PHI node that we are updating!");
if (!FuncInfo->MBB->isSuccessor(PHI->getParent()))
continue;
PHI->addOperand(
MachineOperand::CreateReg(FuncInfo->PHINodesToUpdate[i].second, false));
PHI->addOperand(MachineOperand::CreateMBB(FuncInfo->MBB));
}
return;
}
for (unsigned i = 0, e = SDB->BitTestCases.size(); i != e; ++i) {
// Lower header first, if it wasn't already lowered
if (!SDB->BitTestCases[i].Emitted) {
// Set the current basic block to the mbb we wish to insert the code into
FuncInfo->MBB = SDB->BitTestCases[i].Parent;
FuncInfo->InsertPt = FuncInfo->MBB->end();
// Emit the code
SDB->visitBitTestHeader(SDB->BitTestCases[i], FuncInfo->MBB);
CurDAG->setRoot(SDB->getRoot());
SDB->clear();
CodeGenAndEmitDAG();
}
for (unsigned j = 0, ej = SDB->BitTestCases[i].Cases.size(); j != ej; ++j) {
// Set the current basic block to the mbb we wish to insert the code into
FuncInfo->MBB = SDB->BitTestCases[i].Cases[j].ThisBB;
FuncInfo->InsertPt = FuncInfo->MBB->end();
// Emit the code
if (j+1 != ej)
SDB->visitBitTestCase(SDB->BitTestCases[i],
SDB->BitTestCases[i].Cases[j+1].ThisBB,
SDB->BitTestCases[i].Reg,
SDB->BitTestCases[i].Cases[j],
FuncInfo->MBB);
else
SDB->visitBitTestCase(SDB->BitTestCases[i],
SDB->BitTestCases[i].Default,
SDB->BitTestCases[i].Reg,
SDB->BitTestCases[i].Cases[j],
FuncInfo->MBB);
CurDAG->setRoot(SDB->getRoot());
SDB->clear();
CodeGenAndEmitDAG();
}
// Update PHI Nodes
for (unsigned pi = 0, pe = FuncInfo->PHINodesToUpdate.size();
pi != pe; ++pi) {
MachineInstr *PHI = FuncInfo->PHINodesToUpdate[pi].first;
MachineBasicBlock *PHIBB = PHI->getParent();
assert(PHI->isPHI() &&
"This is not a machine PHI node that we are updating!");
// This is "default" BB. We have two jumps to it. From "header" BB and
// from last "case" BB.
if (PHIBB == SDB->BitTestCases[i].Default) {
PHI->addOperand(MachineOperand::
CreateReg(FuncInfo->PHINodesToUpdate[pi].second,
false));
PHI->addOperand(MachineOperand::CreateMBB(SDB->BitTestCases[i].Parent));
PHI->addOperand(MachineOperand::
CreateReg(FuncInfo->PHINodesToUpdate[pi].second,
false));
PHI->addOperand(MachineOperand::CreateMBB(SDB->BitTestCases[i].Cases.
back().ThisBB));
}
// One of "cases" BB.
for (unsigned j = 0, ej = SDB->BitTestCases[i].Cases.size();
j != ej; ++j) {
MachineBasicBlock* cBB = SDB->BitTestCases[i].Cases[j].ThisBB;
if (cBB->isSuccessor(PHIBB)) {
PHI->addOperand(MachineOperand::
CreateReg(FuncInfo->PHINodesToUpdate[pi].second,
false));
PHI->addOperand(MachineOperand::CreateMBB(cBB));
}
}
}
}
SDB->BitTestCases.clear();
// If the JumpTable record is filled in, then we need to emit a jump table.
// Updating the PHI nodes is tricky in this case, since we need to determine
// whether the PHI is a successor of the range check MBB or the jump table MBB
for (unsigned i = 0, e = SDB->JTCases.size(); i != e; ++i) {
// Lower header first, if it wasn't already lowered
if (!SDB->JTCases[i].first.Emitted) {
// Set the current basic block to the mbb we wish to insert the code into
FuncInfo->MBB = SDB->JTCases[i].first.HeaderBB;
FuncInfo->InsertPt = FuncInfo->MBB->end();
// Emit the code
SDB->visitJumpTableHeader(SDB->JTCases[i].second, SDB->JTCases[i].first,
FuncInfo->MBB);
CurDAG->setRoot(SDB->getRoot());
SDB->clear();
CodeGenAndEmitDAG();
}
// Set the current basic block to the mbb we wish to insert the code into
FuncInfo->MBB = SDB->JTCases[i].second.MBB;
FuncInfo->InsertPt = FuncInfo->MBB->end();
// Emit the code
SDB->visitJumpTable(SDB->JTCases[i].second);
CurDAG->setRoot(SDB->getRoot());
SDB->clear();
CodeGenAndEmitDAG();
// Update PHI Nodes
for (unsigned pi = 0, pe = FuncInfo->PHINodesToUpdate.size();
pi != pe; ++pi) {
MachineInstr *PHI = FuncInfo->PHINodesToUpdate[pi].first;
MachineBasicBlock *PHIBB = PHI->getParent();
assert(PHI->isPHI() &&
"This is not a machine PHI node that we are updating!");
// "default" BB. We can go there only from header BB.
if (PHIBB == SDB->JTCases[i].second.Default) {
PHI->addOperand
(MachineOperand::CreateReg(FuncInfo->PHINodesToUpdate[pi].second,
false));
PHI->addOperand
(MachineOperand::CreateMBB(SDB->JTCases[i].first.HeaderBB));
}
// JT BB. Just iterate over successors here
if (FuncInfo->MBB->isSuccessor(PHIBB)) {
PHI->addOperand
(MachineOperand::CreateReg(FuncInfo->PHINodesToUpdate[pi].second,
false));
PHI->addOperand(MachineOperand::CreateMBB(FuncInfo->MBB));
}
}
}
SDB->JTCases.clear();
// If the switch block involved a branch to one of the actual successors, we
// need to update PHI nodes in that block.
for (unsigned i = 0, e = FuncInfo->PHINodesToUpdate.size(); i != e; ++i) {
MachineInstr *PHI = FuncInfo->PHINodesToUpdate[i].first;
assert(PHI->isPHI() &&
"This is not a machine PHI node that we are updating!");
if (FuncInfo->MBB->isSuccessor(PHI->getParent())) {
PHI->addOperand(
MachineOperand::CreateReg(FuncInfo->PHINodesToUpdate[i].second, false));
PHI->addOperand(MachineOperand::CreateMBB(FuncInfo->MBB));
}
}
// If we generated any switch lowering information, build and codegen any
// additional DAGs necessary.
for (unsigned i = 0, e = SDB->SwitchCases.size(); i != e; ++i) {
// Set the current basic block to the mbb we wish to insert the code into
FuncInfo->MBB = SDB->SwitchCases[i].ThisBB;
FuncInfo->InsertPt = FuncInfo->MBB->end();
// Determine the unique successors.
SmallVector<MachineBasicBlock *, 2> Succs;
Succs.push_back(SDB->SwitchCases[i].TrueBB);
if (SDB->SwitchCases[i].TrueBB != SDB->SwitchCases[i].FalseBB)
Succs.push_back(SDB->SwitchCases[i].FalseBB);
// Emit the code. Note that this could result in FuncInfo->MBB being split.
SDB->visitSwitchCase(SDB->SwitchCases[i], FuncInfo->MBB);
CurDAG->setRoot(SDB->getRoot());
SDB->clear();
CodeGenAndEmitDAG();
// Remember the last block, now that any splitting is done, for use in
// populating PHI nodes in successors.
MachineBasicBlock *ThisBB = FuncInfo->MBB;
// Handle any PHI nodes in successors of this chunk, as if we were coming
// from the original BB before switch expansion. Note that PHI nodes can
// occur multiple times in PHINodesToUpdate. We have to be very careful to
// handle them the right number of times.
for (unsigned i = 0, e = Succs.size(); i != e; ++i) {
FuncInfo->MBB = Succs[i];
FuncInfo->InsertPt = FuncInfo->MBB->end();
// FuncInfo->MBB may have been removed from the CFG if a branch was
// constant folded.
if (ThisBB->isSuccessor(FuncInfo->MBB)) {
for (MachineBasicBlock::iterator Phi = FuncInfo->MBB->begin();
Phi != FuncInfo->MBB->end() && Phi->isPHI();
++Phi) {
// This value for this PHI node is recorded in PHINodesToUpdate.
for (unsigned pn = 0; ; ++pn) {
assert(pn != FuncInfo->PHINodesToUpdate.size() &&
"Didn't find PHI entry!");
if (FuncInfo->PHINodesToUpdate[pn].first == Phi) {
Phi->addOperand(MachineOperand::
CreateReg(FuncInfo->PHINodesToUpdate[pn].second,
false));
Phi->addOperand(MachineOperand::CreateMBB(ThisBB));
break;
}
}
}
}
}
}
SDB->SwitchCases.clear();
}
/// Create the scheduler. If a specific scheduler was specified
/// via the SchedulerRegistry, use it, otherwise select the
/// one preferred by the target.
///
ScheduleDAGSDNodes *SelectionDAGISel::CreateScheduler() {
RegisterScheduler::FunctionPassCtor Ctor = RegisterScheduler::getDefault();
if (!Ctor) {
Ctor = ISHeuristic;
RegisterScheduler::setDefault(Ctor);
}
return Ctor(this, OptLevel);
}
//===----------------------------------------------------------------------===//
// Helper functions used by the generated instruction selector.
//===----------------------------------------------------------------------===//
// Calls to these methods are generated by tblgen.
/// CheckAndMask - The isel is trying to match something like (and X, 255). If
/// the dag combiner simplified the 255, we still want to match. RHS is the
/// actual value in the DAG on the RHS of an AND, and DesiredMaskS is the value
/// specified in the .td file (e.g. 255).
bool SelectionDAGISel::CheckAndMask(SDValue LHS, ConstantSDNode *RHS,
int64_t DesiredMaskS) const {
const APInt &ActualMask = RHS->getAPIntValue();
const APInt &DesiredMask = APInt(LHS.getValueSizeInBits(), DesiredMaskS);
// If the actual mask exactly matches, success!
if (ActualMask == DesiredMask)
return true;
// If the actual AND mask is allowing unallowed bits, this doesn't match.
if (ActualMask.intersects(~DesiredMask))
return false;
// Otherwise, the DAG Combiner may have proven that the value coming in is
// either already zero or is not demanded. Check for known zero input bits.
APInt NeededMask = DesiredMask & ~ActualMask;
if (CurDAG->MaskedValueIsZero(LHS, NeededMask))
return true;
// TODO: check to see if missing bits are just not demanded.
// Otherwise, this pattern doesn't match.
return false;
}
/// CheckOrMask - The isel is trying to match something like (or X, 255). If
/// the dag combiner simplified the 255, we still want to match. RHS is the
/// actual value in the DAG on the RHS of an OR, and DesiredMaskS is the value
/// specified in the .td file (e.g. 255).
bool SelectionDAGISel::CheckOrMask(SDValue LHS, ConstantSDNode *RHS,
int64_t DesiredMaskS) const {
const APInt &ActualMask = RHS->getAPIntValue();
const APInt &DesiredMask = APInt(LHS.getValueSizeInBits(), DesiredMaskS);
// If the actual mask exactly matches, success!
if (ActualMask == DesiredMask)
return true;
// If the actual AND mask is allowing unallowed bits, this doesn't match.
if (ActualMask.intersects(~DesiredMask))
return false;
// Otherwise, the DAG Combiner may have proven that the value coming in is
// either already zero or is not demanded. Check for known zero input bits.
APInt NeededMask = DesiredMask & ~ActualMask;
APInt KnownZero, KnownOne;
CurDAG->ComputeMaskedBits(LHS, NeededMask, KnownZero, KnownOne);
// If all the missing bits in the or are already known to be set, match!
if ((NeededMask & KnownOne) == NeededMask)
return true;
// TODO: check to see if missing bits are just not demanded.
// Otherwise, this pattern doesn't match.
return false;
}
/// SelectInlineAsmMemoryOperands - Calls to this are automatically generated
/// by tblgen. Others should not call it.
void SelectionDAGISel::
SelectInlineAsmMemoryOperands(std::vector<SDValue> &Ops) {
std::vector<SDValue> InOps;
std::swap(InOps, Ops);
Ops.push_back(InOps[InlineAsm::Op_InputChain]); // 0
Ops.push_back(InOps[InlineAsm::Op_AsmString]); // 1
Ops.push_back(InOps[InlineAsm::Op_MDNode]); // 2, !srcloc
Ops.push_back(InOps[InlineAsm::Op_ExtraInfo]); // 3 (SideEffect, AlignStack)
unsigned i = InlineAsm::Op_FirstOperand, e = InOps.size();
if (InOps[e-1].getValueType() == MVT::Glue)
--e; // Don't process a glue operand if it is here.
while (i != e) {
unsigned Flags = cast<ConstantSDNode>(InOps[i])->getZExtValue();
if (!InlineAsm::isMemKind(Flags)) {
// Just skip over this operand, copying the operands verbatim.
Ops.insert(Ops.end(), InOps.begin()+i,
InOps.begin()+i+InlineAsm::getNumOperandRegisters(Flags) + 1);
i += InlineAsm::getNumOperandRegisters(Flags) + 1;
} else {
assert(InlineAsm::getNumOperandRegisters(Flags) == 1 &&
"Memory operand with multiple values?");
// Otherwise, this is a memory operand. Ask the target to select it.
std::vector<SDValue> SelOps;
if (SelectInlineAsmMemoryOperand(InOps[i+1], 'm', SelOps))
report_fatal_error("Could not match memory address. Inline asm"
" failure!");
// Add this to the output node.
unsigned NewFlags =
InlineAsm::getFlagWord(InlineAsm::Kind_Mem, SelOps.size());
Ops.push_back(CurDAG->getTargetConstant(NewFlags, MVT::i32));
Ops.insert(Ops.end(), SelOps.begin(), SelOps.end());
i += 2;
}
}
// Add the glue input back if present.
if (e != InOps.size())
Ops.push_back(InOps.back());
}
/// findGlueUse - Return use of MVT::Glue value produced by the specified
/// SDNode.
///
static SDNode *findGlueUse(SDNode *N) {
unsigned FlagResNo = N->getNumValues()-1;
for (SDNode::use_iterator I = N->use_begin(), E = N->use_end(); I != E; ++I) {
SDUse &Use = I.getUse();
if (Use.getResNo() == FlagResNo)
return Use.getUser();
}
return NULL;
}
/// findNonImmUse - Return true if "Use" is a non-immediate use of "Def".
/// This function recursively traverses up the operand chain, ignoring
/// certain nodes.
static bool findNonImmUse(SDNode *Use, SDNode* Def, SDNode *ImmedUse,
SDNode *Root, SmallPtrSet<SDNode*, 16> &Visited,
bool IgnoreChains) {
// The NodeID's are given uniques ID's where a node ID is guaranteed to be
// greater than all of its (recursive) operands. If we scan to a point where
// 'use' is smaller than the node we're scanning for, then we know we will
// never find it.
//
// The Use may be -1 (unassigned) if it is a newly allocated node. This can
// happen because we scan down to newly selected nodes in the case of glue
// uses.
if ((Use->getNodeId() < Def->getNodeId() && Use->getNodeId() != -1))
return false;
// Don't revisit nodes if we already scanned it and didn't fail, we know we
// won't fail if we scan it again.
if (!Visited.insert(Use))
return false;
for (unsigned i = 0, e = Use->getNumOperands(); i != e; ++i) {
// Ignore chain uses, they are validated by HandleMergeInputChains.
if (Use->getOperand(i).getValueType() == MVT::Other && IgnoreChains)
continue;
SDNode *N = Use->getOperand(i).getNode();
if (N == Def) {
if (Use == ImmedUse || Use == Root)
continue; // We are not looking for immediate use.
assert(N != Root);
return true;
}
// Traverse up the operand chain.
if (findNonImmUse(N, Def, ImmedUse, Root, Visited, IgnoreChains))
return true;
}
return false;
}
/// IsProfitableToFold - Returns true if it's profitable to fold the specific
/// operand node N of U during instruction selection that starts at Root.
bool SelectionDAGISel::IsProfitableToFold(SDValue N, SDNode *U,
SDNode *Root) const {
if (OptLevel == CodeGenOpt::None) return false;
return N.hasOneUse();
}
/// IsLegalToFold - Returns true if the specific operand node N of
/// U can be folded during instruction selection that starts at Root.
bool SelectionDAGISel::IsLegalToFold(SDValue N, SDNode *U, SDNode *Root,
CodeGenOpt::Level OptLevel,
bool IgnoreChains) {
if (OptLevel == CodeGenOpt::None) return false;
// If Root use can somehow reach N through a path that that doesn't contain
// U then folding N would create a cycle. e.g. In the following
// diagram, Root can reach N through X. If N is folded into into Root, then
// X is both a predecessor and a successor of U.
//
// [N*] //
// ^ ^ //
// / \ //
// [U*] [X]? //
// ^ ^ //
// \ / //
// \ / //
// [Root*] //
//
// * indicates nodes to be folded together.
//
// If Root produces glue, then it gets (even more) interesting. Since it
// will be "glued" together with its glue use in the scheduler, we need to
// check if it might reach N.
//
// [N*] //
// ^ ^ //
// / \ //
// [U*] [X]? //
// ^ ^ //
// \ \ //
// \ | //
// [Root*] | //
// ^ | //
// f | //
// | / //
// [Y] / //
// ^ / //
// f / //
// | / //
// [GU] //
//
// If GU (glue use) indirectly reaches N (the load), and Root folds N
// (call it Fold), then X is a predecessor of GU and a successor of
// Fold. But since Fold and GU are glued together, this will create
// a cycle in the scheduling graph.
// If the node has glue, walk down the graph to the "lowest" node in the
// glueged set.
EVT VT = Root->getValueType(Root->getNumValues()-1);
while (VT == MVT::Glue) {
SDNode *GU = findGlueUse(Root);
if (GU == NULL)
break;
Root = GU;
VT = Root->getValueType(Root->getNumValues()-1);
// If our query node has a glue result with a use, we've walked up it. If
// the user (which has already been selected) has a chain or indirectly uses
// the chain, our WalkChainUsers predicate will not consider it. Because of
// this, we cannot ignore chains in this predicate.
IgnoreChains = false;
}
SmallPtrSet<SDNode*, 16> Visited;
return !findNonImmUse(Root, N.getNode(), U, Root, Visited, IgnoreChains);
}
SDNode *SelectionDAGISel::Select_INLINEASM(SDNode *N) {
std::vector<SDValue> Ops(N->op_begin(), N->op_end());
SelectInlineAsmMemoryOperands(Ops);
std::vector<EVT> VTs;
VTs.push_back(MVT::Other);
VTs.push_back(MVT::Glue);
SDValue New = CurDAG->getNode(ISD::INLINEASM, N->getDebugLoc(),
VTs, &Ops[0], Ops.size());
New->setNodeId(-1);
return New.getNode();
}
SDNode *SelectionDAGISel::Select_UNDEF(SDNode *N) {
return CurDAG->SelectNodeTo(N, TargetOpcode::IMPLICIT_DEF,N->getValueType(0));
}
/// GetVBR - decode a vbr encoding whose top bit is set.
LLVM_ATTRIBUTE_ALWAYS_INLINE static uint64_t
GetVBR(uint64_t Val, const unsigned char *MatcherTable, unsigned &Idx) {
assert(Val >= 128 && "Not a VBR");
Val &= 127; // Remove first vbr bit.
unsigned Shift = 7;
uint64_t NextBits;
do {
NextBits = MatcherTable[Idx++];
Val |= (NextBits&127) << Shift;
Shift += 7;
} while (NextBits & 128);
return Val;
}
/// UpdateChainsAndGlue - When a match is complete, this method updates uses of
/// interior glue and chain results to use the new glue and chain results.
void SelectionDAGISel::
UpdateChainsAndGlue(SDNode *NodeToMatch, SDValue InputChain,
const SmallVectorImpl<SDNode*> &ChainNodesMatched,
SDValue InputGlue,
const SmallVectorImpl<SDNode*> &GlueResultNodesMatched,
bool isMorphNodeTo) {
SmallVector<SDNode*, 4> NowDeadNodes;
ISelUpdater ISU(ISelPosition);
// Now that all the normal results are replaced, we replace the chain and
// glue results if present.
if (!ChainNodesMatched.empty()) {
assert(InputChain.getNode() != 0 &&
"Matched input chains but didn't produce a chain");
// Loop over all of the nodes we matched that produced a chain result.
// Replace all the chain results with the final chain we ended up with.
for (unsigned i = 0, e = ChainNodesMatched.size(); i != e; ++i) {
SDNode *ChainNode = ChainNodesMatched[i];
// If this node was already deleted, don't look at it.
if (ChainNode->getOpcode() == ISD::DELETED_NODE)
continue;
// Don't replace the results of the root node if we're doing a
// MorphNodeTo.
if (ChainNode == NodeToMatch && isMorphNodeTo)
continue;
SDValue ChainVal = SDValue(ChainNode, ChainNode->getNumValues()-1);
if (ChainVal.getValueType() == MVT::Glue)
ChainVal = ChainVal.getValue(ChainVal->getNumValues()-2);
assert(ChainVal.getValueType() == MVT::Other && "Not a chain?");
CurDAG->ReplaceAllUsesOfValueWith(ChainVal, InputChain, &ISU);
// If the node became dead and we haven't already seen it, delete it.
if (ChainNode->use_empty() &&
!std::count(NowDeadNodes.begin(), NowDeadNodes.end(), ChainNode))
NowDeadNodes.push_back(ChainNode);
}
}
// If the result produces glue, update any glue results in the matched
// pattern with the glue result.
if (InputGlue.getNode() != 0) {
// Handle any interior nodes explicitly marked.
for (unsigned i = 0, e = GlueResultNodesMatched.size(); i != e; ++i) {
SDNode *FRN = GlueResultNodesMatched[i];
// If this node was already deleted, don't look at it.
if (FRN->getOpcode() == ISD::DELETED_NODE)
continue;
assert(FRN->getValueType(FRN->getNumValues()-1) == MVT::Glue &&
"Doesn't have a glue result");
CurDAG->ReplaceAllUsesOfValueWith(SDValue(FRN, FRN->getNumValues()-1),
InputGlue, &ISU);
// If the node became dead and we haven't already seen it, delete it.
if (FRN->use_empty() &&
!std::count(NowDeadNodes.begin(), NowDeadNodes.end(), FRN))
NowDeadNodes.push_back(FRN);
}
}
if (!NowDeadNodes.empty())
CurDAG->RemoveDeadNodes(NowDeadNodes, &ISU);
DEBUG(errs() << "ISEL: Match complete!\n");
}
enum ChainResult {
CR_Simple,
CR_InducesCycle,
CR_LeadsToInteriorNode
};
/// WalkChainUsers - Walk down the users of the specified chained node that is
/// part of the pattern we're matching, looking at all of the users we find.
/// This determines whether something is an interior node, whether we have a
/// non-pattern node in between two pattern nodes (which prevent folding because
/// it would induce a cycle) and whether we have a TokenFactor node sandwiched
/// between pattern nodes (in which case the TF becomes part of the pattern).
///
/// The walk we do here is guaranteed to be small because we quickly get down to
/// already selected nodes "below" us.
static ChainResult
WalkChainUsers(SDNode *ChainedNode,
SmallVectorImpl<SDNode*> &ChainedNodesInPattern,
SmallVectorImpl<SDNode*> &InteriorChainedNodes) {
ChainResult Result = CR_Simple;
for (SDNode::use_iterator UI = ChainedNode->use_begin(),
E = ChainedNode->use_end(); UI != E; ++UI) {
// Make sure the use is of the chain, not some other value we produce.
if (UI.getUse().getValueType() != MVT::Other) continue;
SDNode *User = *UI;
// If we see an already-selected machine node, then we've gone beyond the
// pattern that we're selecting down into the already selected chunk of the
// DAG.
if (User->isMachineOpcode() ||
User->getOpcode() == ISD::HANDLENODE) // Root of the graph.
continue;
if (User->getOpcode() == ISD::CopyToReg ||
User->getOpcode() == ISD::CopyFromReg ||
User->getOpcode() == ISD::INLINEASM ||
User->getOpcode() == ISD::EH_LABEL) {
// If their node ID got reset to -1 then they've already been selected.
// Treat them like a MachineOpcode.
if (User->getNodeId() == -1)
continue;
}
// If we have a TokenFactor, we handle it specially.
if (User->getOpcode() != ISD::TokenFactor) {
// If the node isn't a token factor and isn't part of our pattern, then it
// must be a random chained node in between two nodes we're selecting.
// This happens when we have something like:
// x = load ptr
// call
// y = x+4
// store y -> ptr
// Because we structurally match the load/store as a read/modify/write,
// but the call is chained between them. We cannot fold in this case
// because it would induce a cycle in the graph.
if (!std::count(ChainedNodesInPattern.begin(),
ChainedNodesInPattern.end(), User))
return CR_InducesCycle;
// Otherwise we found a node that is part of our pattern. For example in:
// x = load ptr
// y = x+4
// store y -> ptr
// This would happen when we're scanning down from the load and see the
// store as a user. Record that there is a use of ChainedNode that is
// part of the pattern and keep scanning uses.
Result = CR_LeadsToInteriorNode;
InteriorChainedNodes.push_back(User);
continue;
}
// If we found a TokenFactor, there are two cases to consider: first if the
// TokenFactor is just hanging "below" the pattern we're matching (i.e. no
// uses of the TF are in our pattern) we just want to ignore it. Second,
// the TokenFactor can be sandwiched in between two chained nodes, like so:
// [Load chain]
// ^
// |
// [Load]
// ^ ^
// | \ DAG's like cheese
// / \ do you?
// / |
// [TokenFactor] [Op]
// ^ ^
// | |
// \ /
// \ /
// [Store]
//
// In this case, the TokenFactor becomes part of our match and we rewrite it
// as a new TokenFactor.
//
// To distinguish these two cases, do a recursive walk down the uses.
switch (WalkChainUsers(User, ChainedNodesInPattern, InteriorChainedNodes)) {
case CR_Simple:
// If the uses of the TokenFactor are just already-selected nodes, ignore
// it, it is "below" our pattern.
continue;
case CR_InducesCycle:
// If the uses of the TokenFactor lead to nodes that are not part of our
// pattern that are not selected, folding would turn this into a cycle,
// bail out now.
return CR_InducesCycle;
case CR_LeadsToInteriorNode:
break; // Otherwise, keep processing.
}
// Okay, we know we're in the interesting interior case. The TokenFactor
// is now going to be considered part of the pattern so that we rewrite its
// uses (it may have uses that are not part of the pattern) with the
// ultimate chain result of the generated code. We will also add its chain
// inputs as inputs to the ultimate TokenFactor we create.
Result = CR_LeadsToInteriorNode;
ChainedNodesInPattern.push_back(User);
InteriorChainedNodes.push_back(User);
continue;
}
return Result;
}
/// HandleMergeInputChains - This implements the OPC_EmitMergeInputChains
/// operation for when the pattern matched at least one node with a chains. The
/// input vector contains a list of all of the chained nodes that we match. We
/// must determine if this is a valid thing to cover (i.e. matching it won't
/// induce cycles in the DAG) and if so, creating a TokenFactor node. that will
/// be used as the input node chain for the generated nodes.
static SDValue
HandleMergeInputChains(SmallVectorImpl<SDNode*> &ChainNodesMatched,
SelectionDAG *CurDAG) {
// Walk all of the chained nodes we've matched, recursively scanning down the
// users of the chain result. This adds any TokenFactor nodes that are caught
// in between chained nodes to the chained and interior nodes list.
SmallVector<SDNode*, 3> InteriorChainedNodes;
for (unsigned i = 0, e = ChainNodesMatched.size(); i != e; ++i) {
if (WalkChainUsers(ChainNodesMatched[i], ChainNodesMatched,
InteriorChainedNodes) == CR_InducesCycle)
return SDValue(); // Would induce a cycle.
}
// Okay, we have walked all the matched nodes and collected TokenFactor nodes
// that we are interested in. Form our input TokenFactor node.
SmallVector<SDValue, 3> InputChains;
for (unsigned i = 0, e = ChainNodesMatched.size(); i != e; ++i) {
// Add the input chain of this node to the InputChains list (which will be
// the operands of the generated TokenFactor) if it's not an interior node.
SDNode *N = ChainNodesMatched[i];
if (N->getOpcode() != ISD::TokenFactor) {
if (std::count(InteriorChainedNodes.begin(),InteriorChainedNodes.end(),N))
continue;
// Otherwise, add the input chain.
SDValue InChain = ChainNodesMatched[i]->getOperand(0);
assert(InChain.getValueType() == MVT::Other && "Not a chain");
InputChains.push_back(InChain);
continue;
}
// If we have a token factor, we want to add all inputs of the token factor
// that are not part of the pattern we're matching.
for (unsigned op = 0, e = N->getNumOperands(); op != e; ++op) {
if (!std::count(ChainNodesMatched.begin(), ChainNodesMatched.end(),
N->getOperand(op).getNode()))
InputChains.push_back(N->getOperand(op));
}
}
SDValue Res;
if (InputChains.size() == 1)
return InputChains[0];
return CurDAG->getNode(ISD::TokenFactor, ChainNodesMatched[0]->getDebugLoc(),
MVT::Other, &InputChains[0], InputChains.size());
}
/// MorphNode - Handle morphing a node in place for the selector.
SDNode *SelectionDAGISel::
MorphNode(SDNode *Node, unsigned TargetOpc, SDVTList VTList,
const SDValue *Ops, unsigned NumOps, unsigned EmitNodeInfo) {
// It is possible we're using MorphNodeTo to replace a node with no
// normal results with one that has a normal result (or we could be
// adding a chain) and the input could have glue and chains as well.
// In this case we need to shift the operands down.
// FIXME: This is a horrible hack and broken in obscure cases, no worse
// than the old isel though.
int OldGlueResultNo = -1, OldChainResultNo = -1;
unsigned NTMNumResults = Node->getNumValues();
if (Node->getValueType(NTMNumResults-1) == MVT::Glue) {
OldGlueResultNo = NTMNumResults-1;
if (NTMNumResults != 1 &&
Node->getValueType(NTMNumResults-2) == MVT::Other)
OldChainResultNo = NTMNumResults-2;
} else if (Node->getValueType(NTMNumResults-1) == MVT::Other)
OldChainResultNo = NTMNumResults-1;
// Call the underlying SelectionDAG routine to do the transmogrification. Note
// that this deletes operands of the old node that become dead.
SDNode *Res = CurDAG->MorphNodeTo(Node, ~TargetOpc, VTList, Ops, NumOps);
// MorphNodeTo can operate in two ways: if an existing node with the
// specified operands exists, it can just return it. Otherwise, it
// updates the node in place to have the requested operands.
if (Res == Node) {
// If we updated the node in place, reset the node ID. To the isel,
// this should be just like a newly allocated machine node.
Res->setNodeId(-1);
}
unsigned ResNumResults = Res->getNumValues();
// Move the glue if needed.
if ((EmitNodeInfo & OPFL_GlueOutput) && OldGlueResultNo != -1 &&
(unsigned)OldGlueResultNo != ResNumResults-1)
CurDAG->ReplaceAllUsesOfValueWith(SDValue(Node, OldGlueResultNo),
SDValue(Res, ResNumResults-1));
if ((EmitNodeInfo & OPFL_GlueOutput) != 0)
--ResNumResults;
// Move the chain reference if needed.
if ((EmitNodeInfo & OPFL_Chain) && OldChainResultNo != -1 &&
(unsigned)OldChainResultNo != ResNumResults-1)
CurDAG->ReplaceAllUsesOfValueWith(SDValue(Node, OldChainResultNo),
SDValue(Res, ResNumResults-1));
// Otherwise, no replacement happened because the node already exists. Replace
// Uses of the old node with the new one.
if (Res != Node)
CurDAG->ReplaceAllUsesWith(Node, Res);
return Res;
}
/// CheckPatternPredicate - Implements OP_CheckPatternPredicate.
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckSame(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDValue N,
const SmallVectorImpl<std::pair<SDValue, SDNode*> > &RecordedNodes) {
// Accept if it is exactly the same as a previously recorded node.
unsigned RecNo = MatcherTable[MatcherIndex++];
assert(RecNo < RecordedNodes.size() && "Invalid CheckSame");
return N == RecordedNodes[RecNo].first;
}
/// CheckPatternPredicate - Implements OP_CheckPatternPredicate.
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckPatternPredicate(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SelectionDAGISel &SDISel) {
return SDISel.CheckPatternPredicate(MatcherTable[MatcherIndex++]);
}
/// CheckNodePredicate - Implements OP_CheckNodePredicate.
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckNodePredicate(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SelectionDAGISel &SDISel, SDNode *N) {
return SDISel.CheckNodePredicate(N, MatcherTable[MatcherIndex++]);
}
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckOpcode(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDNode *N) {
uint16_t Opc = MatcherTable[MatcherIndex++];
Opc |= (unsigned short)MatcherTable[MatcherIndex++] << 8;
return N->getOpcode() == Opc;
}
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckType(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDValue N, const TargetLowering &TLI) {
MVT::SimpleValueType VT = (MVT::SimpleValueType)MatcherTable[MatcherIndex++];
if (N.getValueType() == VT) return true;
// Handle the case when VT is iPTR.
return VT == MVT::iPTR && N.getValueType() == TLI.getPointerTy();
}
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckChildType(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDValue N, const TargetLowering &TLI,
unsigned ChildNo) {
if (ChildNo >= N.getNumOperands())
return false; // Match fails if out of range child #.
return ::CheckType(MatcherTable, MatcherIndex, N.getOperand(ChildNo), TLI);
}
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckCondCode(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDValue N) {
return cast<CondCodeSDNode>(N)->get() ==
(ISD::CondCode)MatcherTable[MatcherIndex++];
}
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckValueType(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDValue N, const TargetLowering &TLI) {
MVT::SimpleValueType VT = (MVT::SimpleValueType)MatcherTable[MatcherIndex++];
if (cast<VTSDNode>(N)->getVT() == VT)
return true;
// Handle the case when VT is iPTR.
return VT == MVT::iPTR && cast<VTSDNode>(N)->getVT() == TLI.getPointerTy();
}
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckInteger(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDValue N) {
int64_t Val = MatcherTable[MatcherIndex++];
if (Val & 128)
Val = GetVBR(Val, MatcherTable, MatcherIndex);
ConstantSDNode *C = dyn_cast<ConstantSDNode>(N);
return C != 0 && C->getSExtValue() == Val;
}
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckAndImm(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDValue N, SelectionDAGISel &SDISel) {
int64_t Val = MatcherTable[MatcherIndex++];
if (Val & 128)
Val = GetVBR(Val, MatcherTable, MatcherIndex);
if (N->getOpcode() != ISD::AND) return false;
ConstantSDNode *C = dyn_cast<ConstantSDNode>(N->getOperand(1));
return C != 0 && SDISel.CheckAndMask(N.getOperand(0), C, Val);
}
LLVM_ATTRIBUTE_ALWAYS_INLINE static bool
CheckOrImm(const unsigned char *MatcherTable, unsigned &MatcherIndex,
SDValue N, SelectionDAGISel &SDISel) {
int64_t Val = MatcherTable[MatcherIndex++];
if (Val & 128)
Val = GetVBR(Val, MatcherTable, MatcherIndex);
if (N->getOpcode() != ISD::OR) return false;
ConstantSDNode *C = dyn_cast<ConstantSDNode>(N->getOperand(1));
return C != 0 && SDISel.CheckOrMask(N.getOperand(0), C, Val);
}
/// IsPredicateKnownToFail - If we know how and can do so without pushing a
/// scope, evaluate the current node. If the current predicate is known to
/// fail, set Result=true and return anything. If the current predicate is
/// known to pass, set Result=false and return the MatcherIndex to continue
/// with. If the current predicate is unknown, set Result=false and return the
/// MatcherIndex to continue with.
static unsigned IsPredicateKnownToFail(const unsigned char *Table,
unsigned Index, SDValue N,
bool &Result, SelectionDAGISel &SDISel,
SmallVectorImpl<std::pair<SDValue, SDNode*> > &RecordedNodes) {
switch (Table[Index++]) {
default:
Result = false;
return Index-1; // Could not evaluate this predicate.
case SelectionDAGISel::OPC_CheckSame:
Result = !::CheckSame(Table, Index, N, RecordedNodes);
return Index;
case SelectionDAGISel::OPC_CheckPatternPredicate:
Result = !::CheckPatternPredicate(Table, Index, SDISel);
return Index;
case SelectionDAGISel::OPC_CheckPredicate:
Result = !::CheckNodePredicate(Table, Index, SDISel, N.getNode());
return Index;
case SelectionDAGISel::OPC_CheckOpcode:
Result = !::CheckOpcode(Table, Index, N.getNode());
return Index;
case SelectionDAGISel::OPC_CheckType:
Result = !::CheckType(Table, Index, N, SDISel.TLI);
return Index;
case SelectionDAGISel::OPC_CheckChild0Type:
case SelectionDAGISel::OPC_CheckChild1Type:
case SelectionDAGISel::OPC_CheckChild2Type:
case SelectionDAGISel::OPC_CheckChild3Type:
case SelectionDAGISel::OPC_CheckChild4Type:
case SelectionDAGISel::OPC_CheckChild5Type:
case SelectionDAGISel::OPC_CheckChild6Type:
case SelectionDAGISel::OPC_CheckChild7Type:
Result = !::CheckChildType(Table, Index, N, SDISel.TLI,
Table[Index-1] - SelectionDAGISel::OPC_CheckChild0Type);
return Index;
case SelectionDAGISel::OPC_CheckCondCode:
Result = !::CheckCondCode(Table, Index, N);
return Index;
case SelectionDAGISel::OPC_CheckValueType:
Result = !::CheckValueType(Table, Index, N, SDISel.TLI);
return Index;
case SelectionDAGISel::OPC_CheckInteger:
Result = !::CheckInteger(Table, Index, N);
return Index;
case SelectionDAGISel::OPC_CheckAndImm:
Result = !::CheckAndImm(Table, Index, N, SDISel);
return Index;
case SelectionDAGISel::OPC_CheckOrImm:
Result = !::CheckOrImm(Table, Index, N, SDISel);
return Index;
}
}
namespace {
struct MatchScope {
/// FailIndex - If this match fails, this is the index to continue with.
unsigned FailIndex;
/// NodeStack - The node stack when the scope was formed.
SmallVector<SDValue, 4> NodeStack;
/// NumRecordedNodes - The number of recorded nodes when the scope was formed.
unsigned NumRecordedNodes;
/// NumMatchedMemRefs - The number of matched memref entries.
unsigned NumMatchedMemRefs;
/// InputChain/InputGlue - The current chain/glue
SDValue InputChain, InputGlue;
/// HasChainNodesMatched - True if the ChainNodesMatched list is non-empty.
bool HasChainNodesMatched, HasGlueResultNodesMatched;
};
}
SDNode *SelectionDAGISel::
SelectCodeCommon(SDNode *NodeToMatch, const unsigned char *MatcherTable,
unsigned TableSize) {
// FIXME: Should these even be selected? Handle these cases in the caller?
switch (NodeToMatch->getOpcode()) {
default:
break;
case ISD::EntryToken: // These nodes remain the same.
case ISD::BasicBlock:
case ISD::Register:
//case ISD::VALUETYPE:
//case ISD::CONDCODE:
case ISD::HANDLENODE:
case ISD::MDNODE_SDNODE:
case ISD::TargetConstant:
case ISD::TargetConstantFP:
case ISD::TargetConstantPool:
case ISD::TargetFrameIndex:
case ISD::TargetExternalSymbol:
case ISD::TargetBlockAddress:
case ISD::TargetJumpTable:
case ISD::TargetGlobalTLSAddress:
case ISD::TargetGlobalAddress:
case ISD::TokenFactor:
case ISD::CopyFromReg:
case ISD::CopyToReg:
case ISD::EH_LABEL:
NodeToMatch->setNodeId(-1); // Mark selected.
return 0;
case ISD::AssertSext:
case ISD::AssertZext:
CurDAG->ReplaceAllUsesOfValueWith(SDValue(NodeToMatch, 0),
NodeToMatch->getOperand(0));
return 0;
case ISD::INLINEASM: return Select_INLINEASM(NodeToMatch);
case ISD::UNDEF: return Select_UNDEF(NodeToMatch);
}
assert(!NodeToMatch->isMachineOpcode() && "Node already selected!");
// Set up the node stack with NodeToMatch as the only node on the stack.
SmallVector<SDValue, 8> NodeStack;
SDValue N = SDValue(NodeToMatch, 0);
NodeStack.push_back(N);
// MatchScopes - Scopes used when matching, if a match failure happens, this
// indicates where to continue checking.
SmallVector<MatchScope, 8> MatchScopes;
// RecordedNodes - This is the set of nodes that have been recorded by the
// state machine. The second value is the parent of the node, or null if the
// root is recorded.
SmallVector<std::pair<SDValue, SDNode*>, 8> RecordedNodes;
// MatchedMemRefs - This is the set of MemRef's we've seen in the input
// pattern.
SmallVector<MachineMemOperand*, 2> MatchedMemRefs;
// These are the current input chain and glue for use when generating nodes.
// Various Emit operations change these. For example, emitting a copytoreg
// uses and updates these.
SDValue InputChain, InputGlue;
// ChainNodesMatched - If a pattern matches nodes that have input/output
// chains, the OPC_EmitMergeInputChains operation is emitted which indicates
// which ones they are. The result is captured into this list so that we can
// update the chain results when the pattern is complete.
SmallVector<SDNode*, 3> ChainNodesMatched;
SmallVector<SDNode*, 3> GlueResultNodesMatched;
DEBUG(errs() << "ISEL: Starting pattern match on root node: ";
NodeToMatch->dump(CurDAG);
errs() << '\n');
// Determine where to start the interpreter. Normally we start at opcode #0,
// but if the state machine starts with an OPC_SwitchOpcode, then we
// accelerate the first lookup (which is guaranteed to be hot) with the
// OpcodeOffset table.
unsigned MatcherIndex = 0;
if (!OpcodeOffset.empty()) {
// Already computed the OpcodeOffset table, just index into it.
if (N.getOpcode() < OpcodeOffset.size())
MatcherIndex = OpcodeOffset[N.getOpcode()];
DEBUG(errs() << " Initial Opcode index to " << MatcherIndex << "\n");
} else if (MatcherTable[0] == OPC_SwitchOpcode) {
// Otherwise, the table isn't computed, but the state machine does start
// with an OPC_SwitchOpcode instruction. Populate the table now, since this
// is the first time we're selecting an instruction.
unsigned Idx = 1;
while (1) {
// Get the size of this case.
unsigned CaseSize = MatcherTable[Idx++];
if (CaseSize & 128)
CaseSize = GetVBR(CaseSize, MatcherTable, Idx);
if (CaseSize == 0) break;
// Get the opcode, add the index to the table.
uint16_t Opc = MatcherTable[Idx++];
Opc |= (unsigned short)MatcherTable[Idx++] << 8;
if (Opc >= OpcodeOffset.size())
OpcodeOffset.resize((Opc+1)*2);
OpcodeOffset[Opc] = Idx;
Idx += CaseSize;
}
// Okay, do the lookup for the first opcode.
if (N.getOpcode() < OpcodeOffset.size())
MatcherIndex = OpcodeOffset[N.getOpcode()];
}
while (1) {
assert(MatcherIndex < TableSize && "Invalid index");
#ifndef NDEBUG
unsigned CurrentOpcodeIndex = MatcherIndex;
#endif
BuiltinOpcodes Opcode = (BuiltinOpcodes)MatcherTable[MatcherIndex++];
switch (Opcode) {
case OPC_Scope: {
// Okay, the semantics of this operation are that we should push a scope
// then evaluate the first child. However, pushing a scope only to have
// the first check fail (which then pops it) is inefficient. If we can
// determine immediately that the first check (or first several) will
// immediately fail, don't even bother pushing a scope for them.
unsigned FailIndex;
while (1) {
unsigned NumToSkip = MatcherTable[MatcherIndex++];
if (NumToSkip & 128)
NumToSkip = GetVBR(NumToSkip, MatcherTable, MatcherIndex);
// Found the end of the scope with no match.
if (NumToSkip == 0) {
FailIndex = 0;
break;
}
FailIndex = MatcherIndex+NumToSkip;
unsigned MatcherIndexOfPredicate = MatcherIndex;
(void)MatcherIndexOfPredicate; // silence warning.
// If we can't evaluate this predicate without pushing a scope (e.g. if
// it is a 'MoveParent') or if the predicate succeeds on this node, we
// push the scope and evaluate the full predicate chain.
bool Result;
MatcherIndex = IsPredicateKnownToFail(MatcherTable, MatcherIndex, N,
Result, *this, RecordedNodes);
if (!Result)
break;
DEBUG(errs() << " Skipped scope entry (due to false predicate) at "
<< "index " << MatcherIndexOfPredicate
<< ", continuing at " << FailIndex << "\n");
++NumDAGIselRetries;
// Otherwise, we know that this case of the Scope is guaranteed to fail,
// move to the next case.
MatcherIndex = FailIndex;
}
// If the whole scope failed to match, bail.
if (FailIndex == 0) break;
// Push a MatchScope which indicates where to go if the first child fails
// to match.
MatchScope NewEntry;
NewEntry.FailIndex = FailIndex;
NewEntry.NodeStack.append(NodeStack.begin(), NodeStack.end());
NewEntry.NumRecordedNodes = RecordedNodes.size();
NewEntry.NumMatchedMemRefs = MatchedMemRefs.size();
NewEntry.InputChain = InputChain;
NewEntry.InputGlue = InputGlue;
NewEntry.HasChainNodesMatched = !ChainNodesMatched.empty();
NewEntry.HasGlueResultNodesMatched = !GlueResultNodesMatched.empty();
MatchScopes.push_back(NewEntry);
continue;
}
case OPC_RecordNode: {
// Remember this node, it may end up being an operand in the pattern.
SDNode *Parent = 0;
if (NodeStack.size() > 1)
Parent = NodeStack[NodeStack.size()-2].getNode();
RecordedNodes.push_back(std::make_pair(N, Parent));
continue;
}
case OPC_RecordChild0: case OPC_RecordChild1:
case OPC_RecordChild2: case OPC_RecordChild3:
case OPC_RecordChild4: case OPC_RecordChild5:
case OPC_RecordChild6: case OPC_RecordChild7: {
unsigned ChildNo = Opcode-OPC_RecordChild0;
if (ChildNo >= N.getNumOperands())
break; // Match fails if out of range child #.
RecordedNodes.push_back(std::make_pair(N->getOperand(ChildNo),
N.getNode()));
continue;
}
case OPC_RecordMemRef:
MatchedMemRefs.push_back(cast<MemSDNode>(N)->getMemOperand());
continue;
case OPC_CaptureGlueInput:
// If the current node has an input glue, capture it in InputGlue.
if (N->getNumOperands() != 0 &&
N->getOperand(N->getNumOperands()-1).getValueType() == MVT::Glue)
InputGlue = N->getOperand(N->getNumOperands()-1);
continue;
case OPC_MoveChild: {
unsigned ChildNo = MatcherTable[MatcherIndex++];
if (ChildNo >= N.getNumOperands())
break; // Match fails if out of range child #.
N = N.getOperand(ChildNo);
NodeStack.push_back(N);
continue;
}
case OPC_MoveParent:
// Pop the current node off the NodeStack.
NodeStack.pop_back();
assert(!NodeStack.empty() && "Node stack imbalance!");
N = NodeStack.back();
continue;
case OPC_CheckSame:
if (!::CheckSame(MatcherTable, MatcherIndex, N, RecordedNodes)) break;
continue;
case OPC_CheckPatternPredicate:
if (!::CheckPatternPredicate(MatcherTable, MatcherIndex, *this)) break;
continue;
case OPC_CheckPredicate:
if (!::CheckNodePredicate(MatcherTable, MatcherIndex, *this,
N.getNode()))
break;
continue;
case OPC_CheckComplexPat: {
unsigned CPNum = MatcherTable[MatcherIndex++];
unsigned RecNo = MatcherTable[MatcherIndex++];
assert(RecNo < RecordedNodes.size() && "Invalid CheckComplexPat");
if (!CheckComplexPattern(NodeToMatch, RecordedNodes[RecNo].second,
RecordedNodes[RecNo].first, CPNum,
RecordedNodes))
break;
continue;
}
case OPC_CheckOpcode:
if (!::CheckOpcode(MatcherTable, MatcherIndex, N.getNode())) break;
continue;
case OPC_CheckType:
if (!::CheckType(MatcherTable, MatcherIndex, N, TLI)) break;
continue;
case OPC_SwitchOpcode: {
unsigned CurNodeOpcode = N.getOpcode();
unsigned SwitchStart = MatcherIndex-1; (void)SwitchStart;
unsigned CaseSize;
while (1) {
// Get the size of this case.
CaseSize = MatcherTable[MatcherIndex++];
if (CaseSize & 128)
CaseSize = GetVBR(CaseSize, MatcherTable, MatcherIndex);
if (CaseSize == 0) break;
uint16_t Opc = MatcherTable[MatcherIndex++];
Opc |= (unsigned short)MatcherTable[MatcherIndex++] << 8;
// If the opcode matches, then we will execute this case.
if (CurNodeOpcode == Opc)
break;
// Otherwise, skip over this case.
MatcherIndex += CaseSize;
}
// If no cases matched, bail out.
if (CaseSize == 0) break;
// Otherwise, execute the case we found.
DEBUG(errs() << " OpcodeSwitch from " << SwitchStart
<< " to " << MatcherIndex << "\n");
continue;
}
case OPC_SwitchType: {
MVT CurNodeVT = N.getValueType().getSimpleVT();
unsigned SwitchStart = MatcherIndex-1; (void)SwitchStart;
unsigned CaseSize;
while (1) {
// Get the size of this case.
CaseSize = MatcherTable[MatcherIndex++];
if (CaseSize & 128)
CaseSize = GetVBR(CaseSize, MatcherTable, MatcherIndex);
if (CaseSize == 0) break;
MVT CaseVT = (MVT::SimpleValueType)MatcherTable[MatcherIndex++];
if (CaseVT == MVT::iPTR)
CaseVT = TLI.getPointerTy();
// If the VT matches, then we will execute this case.
if (CurNodeVT == CaseVT)
break;
// Otherwise, skip over this case.
MatcherIndex += CaseSize;
}
// If no cases matched, bail out.
if (CaseSize == 0) break;
// Otherwise, execute the case we found.
DEBUG(errs() << " TypeSwitch[" << EVT(CurNodeVT).getEVTString()
<< "] from " << SwitchStart << " to " << MatcherIndex<<'\n');
continue;
}
case OPC_CheckChild0Type: case OPC_CheckChild1Type:
case OPC_CheckChild2Type: case OPC_CheckChild3Type:
case OPC_CheckChild4Type: case OPC_CheckChild5Type:
case OPC_CheckChild6Type: case OPC_CheckChild7Type:
if (!::CheckChildType(MatcherTable, MatcherIndex, N, TLI,
Opcode-OPC_CheckChild0Type))
break;
continue;
case OPC_CheckCondCode:
if (!::CheckCondCode(MatcherTable, MatcherIndex, N)) break;
continue;
case OPC_CheckValueType:
if (!::CheckValueType(MatcherTable, MatcherIndex, N, TLI)) break;
continue;
case OPC_CheckInteger:
if (!::CheckInteger(MatcherTable, MatcherIndex, N)) break;
continue;
case OPC_CheckAndImm:
if (!::CheckAndImm(MatcherTable, MatcherIndex, N, *this)) break;
continue;
case OPC_CheckOrImm:
if (!::CheckOrImm(MatcherTable, MatcherIndex, N, *this)) break;
continue;
case OPC_CheckFoldableChainNode: {
assert(NodeStack.size() != 1 && "No parent node");
// Verify that all intermediate nodes between the root and this one have
// a single use.
bool HasMultipleUses = false;
for (unsigned i = 1, e = NodeStack.size()-1; i != e; ++i)
if (!NodeStack[i].hasOneUse()) {
HasMultipleUses = true;
break;
}
if (HasMultipleUses) break;
// Check to see that the target thinks this is profitable to fold and that
// we can fold it without inducing cycles in the graph.
if (!IsProfitableToFold(N, NodeStack[NodeStack.size()-2].getNode(),
NodeToMatch) ||
!IsLegalToFold(N, NodeStack[NodeStack.size()-2].getNode(),
NodeToMatch, OptLevel,
true/*We validate our own chains*/))
break;
continue;
}
case OPC_EmitInteger: {
MVT::SimpleValueType VT =
(MVT::SimpleValueType)MatcherTable[MatcherIndex++];
int64_t Val = MatcherTable[MatcherIndex++];
if (Val & 128)
Val = GetVBR(Val, MatcherTable, MatcherIndex);
RecordedNodes.push_back(std::pair<SDValue, SDNode*>(
CurDAG->getTargetConstant(Val, VT), (SDNode*)0));
continue;
}
case OPC_EmitRegister: {
MVT::SimpleValueType VT =
(MVT::SimpleValueType)MatcherTable[MatcherIndex++];
unsigned RegNo = MatcherTable[MatcherIndex++];
RecordedNodes.push_back(std::pair<SDValue, SDNode*>(
CurDAG->getRegister(RegNo, VT), (SDNode*)0));
continue;
}
case OPC_EmitRegister2: {
// For targets w/ more than 256 register names, the register enum
// values are stored in two bytes in the matcher table (just like
// opcodes).
MVT::SimpleValueType VT =
(MVT::SimpleValueType)MatcherTable[MatcherIndex++];
unsigned RegNo = MatcherTable[MatcherIndex++];
RegNo |= MatcherTable[MatcherIndex++] << 8;
RecordedNodes.push_back(std::pair<SDValue, SDNode*>(
CurDAG->getRegister(RegNo, VT), (SDNode*)0));
continue;
}
case OPC_EmitConvertToTarget: {
// Convert from IMM/FPIMM to target version.
unsigned RecNo = MatcherTable[MatcherIndex++];
assert(RecNo < RecordedNodes.size() && "Invalid CheckSame");
SDValue Imm = RecordedNodes[RecNo].first;
if (Imm->getOpcode() == ISD::Constant) {
int64_t Val = cast<ConstantSDNode>(Imm)->getZExtValue();
Imm = CurDAG->getTargetConstant(Val, Imm.getValueType());
} else if (Imm->getOpcode() == ISD::ConstantFP) {
const ConstantFP *Val=cast<ConstantFPSDNode>(Imm)->getConstantFPValue();
Imm = CurDAG->getTargetConstantFP(*Val, Imm.getValueType());
}
RecordedNodes.push_back(std::make_pair(Imm, RecordedNodes[RecNo].second));
continue;
}
case OPC_EmitMergeInputChains1_0: // OPC_EmitMergeInputChains, 1, 0
case OPC_EmitMergeInputChains1_1: { // OPC_EmitMergeInputChains, 1, 1
// These are space-optimized forms of OPC_EmitMergeInputChains.
assert(InputChain.getNode() == 0 &&
"EmitMergeInputChains should be the first chain producing node");
assert(ChainNodesMatched.empty() &&
"Should only have one EmitMergeInputChains per match");
// Read all of the chained nodes.
unsigned RecNo = Opcode == OPC_EmitMergeInputChains1_1;
assert(RecNo < RecordedNodes.size() && "Invalid CheckSame");
ChainNodesMatched.push_back(RecordedNodes[RecNo].first.getNode());
// FIXME: What if other value results of the node have uses not matched
// by this pattern?
if (ChainNodesMatched.back() != NodeToMatch &&
!RecordedNodes[RecNo].first.hasOneUse()) {
ChainNodesMatched.clear();
break;
}
// Merge the input chains if they are not intra-pattern references.
InputChain = HandleMergeInputChains(ChainNodesMatched, CurDAG);
if (InputChain.getNode() == 0)
break; // Failed to merge.
continue;
}
case OPC_EmitMergeInputChains: {
assert(InputChain.getNode() == 0 &&
"EmitMergeInputChains should be the first chain producing node");
// This node gets a list of nodes we matched in the input that have
// chains. We want to token factor all of the input chains to these nodes
// together. However, if any of the input chains is actually one of the
// nodes matched in this pattern, then we have an intra-match reference.
// Ignore these because the newly token factored chain should not refer to
// the old nodes.
unsigned NumChains = MatcherTable[MatcherIndex++];
assert(NumChains != 0 && "Can't TF zero chains");
assert(ChainNodesMatched.empty() &&
"Should only have one EmitMergeInputChains per match");
// Read all of the chained nodes.
for (unsigned i = 0; i != NumChains; ++i) {
unsigned RecNo = MatcherTable[MatcherIndex++];
assert(RecNo < RecordedNodes.size() && "Invalid CheckSame");
ChainNodesMatched.push_back(RecordedNodes[RecNo].first.getNode());
// FIXME: What if other value results of the node have uses not matched
// by this pattern?
if (ChainNodesMatched.back() != NodeToMatch &&
!RecordedNodes[RecNo].first.hasOneUse()) {
ChainNodesMatched.clear();
break;
}
}
// If the inner loop broke out, the match fails.
if (ChainNodesMatched.empty())
break;
// Merge the input chains if they are not intra-pattern references.
InputChain = HandleMergeInputChains(ChainNodesMatched, CurDAG);
if (InputChain.getNode() == 0)
break; // Failed to merge.
continue;
}
case OPC_EmitCopyToReg: {
unsigned RecNo = MatcherTable[MatcherIndex++];
assert(RecNo < RecordedNodes.size() && "Invalid CheckSame");
unsigned DestPhysReg = MatcherTable[MatcherIndex++];
if (InputChain.getNode() == 0)
InputChain = CurDAG->getEntryNode();
InputChain = CurDAG->getCopyToReg(InputChain, NodeToMatch->getDebugLoc(),
DestPhysReg, RecordedNodes[RecNo].first,
InputGlue);
InputGlue = InputChain.getValue(1);
continue;
}
case OPC_EmitNodeXForm: {
unsigned XFormNo = MatcherTable[MatcherIndex++];
unsigned RecNo = MatcherTable[MatcherIndex++];
assert(RecNo < RecordedNodes.size() && "Invalid CheckSame");
SDValue Res = RunSDNodeXForm(RecordedNodes[RecNo].first, XFormNo);
RecordedNodes.push_back(std::pair<SDValue,SDNode*>(Res, (SDNode*) 0));
continue;
}
case OPC_EmitNode:
case OPC_MorphNodeTo: {
uint16_t TargetOpc = MatcherTable[MatcherIndex++];
TargetOpc |= (unsigned short)MatcherTable[MatcherIndex++] << 8;
unsigned EmitNodeInfo = MatcherTable[MatcherIndex++];
// Get the result VT list.
unsigned NumVTs = MatcherTable[MatcherIndex++];
SmallVector<EVT, 4> VTs;
for (unsigned i = 0; i != NumVTs; ++i) {
MVT::SimpleValueType VT =
(MVT::SimpleValueType)MatcherTable[MatcherIndex++];
if (VT == MVT::iPTR) VT = TLI.getPointerTy().SimpleTy;
VTs.push_back(VT);
}
if (EmitNodeInfo & OPFL_Chain)
VTs.push_back(MVT::Other);
if (EmitNodeInfo & OPFL_GlueOutput)
VTs.push_back(MVT::Glue);
// This is hot code, so optimize the two most common cases of 1 and 2
// results.
SDVTList VTList;
if (VTs.size() == 1)
VTList = CurDAG->getVTList(VTs[0]);
else if (VTs.size() == 2)
VTList = CurDAG->getVTList(VTs[0], VTs[1]);
else
VTList = CurDAG->getVTList(VTs.data(), VTs.size());
// Get the operand list.
unsigned NumOps = MatcherTable[MatcherIndex++];
SmallVector<SDValue, 8> Ops;
for (unsigned i = 0; i != NumOps; ++i) {
unsigned RecNo = MatcherTable[MatcherIndex++];
if (RecNo & 128)
RecNo = GetVBR(RecNo, MatcherTable, MatcherIndex);
assert(RecNo < RecordedNodes.size() && "Invalid EmitNode");
Ops.push_back(RecordedNodes[RecNo].first);
}
// If there are variadic operands to add, handle them now.
if (EmitNodeInfo & OPFL_VariadicInfo) {
// Determine the start index to copy from.
unsigned FirstOpToCopy = getNumFixedFromVariadicInfo(EmitNodeInfo);
FirstOpToCopy += (EmitNodeInfo & OPFL_Chain) ? 1 : 0;
assert(NodeToMatch->getNumOperands() >= FirstOpToCopy &&
"Invalid variadic node");
// Copy all of the variadic operands, not including a potential glue
// input.
for (unsigned i = FirstOpToCopy, e = NodeToMatch->getNumOperands();
i != e; ++i) {
SDValue V = NodeToMatch->getOperand(i);
if (V.getValueType() == MVT::Glue) break;
Ops.push_back(V);
}
}
// If this has chain/glue inputs, add them.
if (EmitNodeInfo & OPFL_Chain)
Ops.push_back(InputChain);
if ((EmitNodeInfo & OPFL_GlueInput) && InputGlue.getNode() != 0)
Ops.push_back(InputGlue);
// Create the node.
SDNode *Res = 0;
if (Opcode != OPC_MorphNodeTo) {
// If this is a normal EmitNode command, just create the new node and
// add the results to the RecordedNodes list.
Res = CurDAG->getMachineNode(TargetOpc, NodeToMatch->getDebugLoc(),
VTList, Ops.data(), Ops.size());
// Add all the non-glue/non-chain results to the RecordedNodes list.
for (unsigned i = 0, e = VTs.size(); i != e; ++i) {
if (VTs[i] == MVT::Other || VTs[i] == MVT::Glue) break;
RecordedNodes.push_back(std::pair<SDValue,SDNode*>(SDValue(Res, i),
(SDNode*) 0));
}
} else {
Res = MorphNode(NodeToMatch, TargetOpc, VTList, Ops.data(), Ops.size(),
EmitNodeInfo);
}
// If the node had chain/glue results, update our notion of the current
// chain and glue.
if (EmitNodeInfo & OPFL_GlueOutput) {
InputGlue = SDValue(Res, VTs.size()-1);
if (EmitNodeInfo & OPFL_Chain)
InputChain = SDValue(Res, VTs.size()-2);
} else if (EmitNodeInfo & OPFL_Chain)
InputChain = SDValue(Res, VTs.size()-1);
// If the OPFL_MemRefs glue is set on this node, slap all of the
// accumulated memrefs onto it.
//
// FIXME: This is vastly incorrect for patterns with multiple outputs
// instructions that access memory and for ComplexPatterns that match
// loads.
if (EmitNodeInfo & OPFL_MemRefs) {
MachineSDNode::mmo_iterator MemRefs =
MF->allocateMemRefsArray(MatchedMemRefs.size());
std::copy(MatchedMemRefs.begin(), MatchedMemRefs.end(), MemRefs);
cast<MachineSDNode>(Res)
->setMemRefs(MemRefs, MemRefs + MatchedMemRefs.size());
}
DEBUG(errs() << " "
<< (Opcode == OPC_MorphNodeTo ? "Morphed" : "Created")
<< " node: "; Res->dump(CurDAG); errs() << "\n");
// If this was a MorphNodeTo then we're completely done!
if (Opcode == OPC_MorphNodeTo) {
// Update chain and glue uses.
UpdateChainsAndGlue(NodeToMatch, InputChain, ChainNodesMatched,
InputGlue, GlueResultNodesMatched, true);
return Res;
}
continue;
}
case OPC_MarkGlueResults: {
unsigned NumNodes = MatcherTable[MatcherIndex++];
// Read and remember all the glue-result nodes.
for (unsigned i = 0; i != NumNodes; ++i) {
unsigned RecNo = MatcherTable[MatcherIndex++];
if (RecNo & 128)
RecNo = GetVBR(RecNo, MatcherTable, MatcherIndex);
assert(RecNo < RecordedNodes.size() && "Invalid CheckSame");
GlueResultNodesMatched.push_back(RecordedNodes[RecNo].first.getNode());
}
continue;
}
case OPC_CompleteMatch: {
// The match has been completed, and any new nodes (if any) have been
// created. Patch up references to the matched dag to use the newly
// created nodes.
unsigned NumResults = MatcherTable[MatcherIndex++];
for (unsigned i = 0; i != NumResults; ++i) {
unsigned ResSlot = MatcherTable[MatcherIndex++];
if (ResSlot & 128)
ResSlot = GetVBR(ResSlot, MatcherTable, MatcherIndex);
assert(ResSlot < RecordedNodes.size() && "Invalid CheckSame");
SDValue Res = RecordedNodes[ResSlot].first;
assert(i < NodeToMatch->getNumValues() &&
NodeToMatch->getValueType(i) != MVT::Other &&
NodeToMatch->getValueType(i) != MVT::Glue &&
"Invalid number of results to complete!");
assert((NodeToMatch->getValueType(i) == Res.getValueType() ||
NodeToMatch->getValueType(i) == MVT::iPTR ||
Res.getValueType() == MVT::iPTR ||
NodeToMatch->getValueType(i).getSizeInBits() ==
Res.getValueType().getSizeInBits()) &&
"invalid replacement");
CurDAG->ReplaceAllUsesOfValueWith(SDValue(NodeToMatch, i), Res);
}
// If the root node defines glue, add it to the glue nodes to update list.
if (NodeToMatch->getValueType(NodeToMatch->getNumValues()-1) == MVT::Glue)
GlueResultNodesMatched.push_back(NodeToMatch);
// Update chain and glue uses.
UpdateChainsAndGlue(NodeToMatch, InputChain, ChainNodesMatched,
InputGlue, GlueResultNodesMatched, false);
assert(NodeToMatch->use_empty() &&
"Didn't replace all uses of the node?");
// FIXME: We just return here, which interacts correctly with SelectRoot
// above. We should fix this to not return an SDNode* anymore.
return 0;
}
}
// If the code reached this point, then the match failed. See if there is
// another child to try in the current 'Scope', otherwise pop it until we
// find a case to check.
DEBUG(errs() << " Match failed at index " << CurrentOpcodeIndex << "\n");
++NumDAGIselRetries;
while (1) {
if (MatchScopes.empty()) {
CannotYetSelect(NodeToMatch);
return 0;
}
// Restore the interpreter state back to the point where the scope was
// formed.
MatchScope &LastScope = MatchScopes.back();
RecordedNodes.resize(LastScope.NumRecordedNodes);
NodeStack.clear();
NodeStack.append(LastScope.NodeStack.begin(), LastScope.NodeStack.end());
N = NodeStack.back();
if (LastScope.NumMatchedMemRefs != MatchedMemRefs.size())
MatchedMemRefs.resize(LastScope.NumMatchedMemRefs);
MatcherIndex = LastScope.FailIndex;
DEBUG(errs() << " Continuing at " << MatcherIndex << "\n");
InputChain = LastScope.InputChain;
InputGlue = LastScope.InputGlue;
if (!LastScope.HasChainNodesMatched)
ChainNodesMatched.clear();
if (!LastScope.HasGlueResultNodesMatched)
GlueResultNodesMatched.clear();
// Check to see what the offset is at the new MatcherIndex. If it is zero
// we have reached the end of this scope, otherwise we have another child
// in the current scope to try.
unsigned NumToSkip = MatcherTable[MatcherIndex++];
if (NumToSkip & 128)
NumToSkip = GetVBR(NumToSkip, MatcherTable, MatcherIndex);
// If we have another child in this scope to match, update FailIndex and
// try it.
if (NumToSkip != 0) {
LastScope.FailIndex = MatcherIndex+NumToSkip;
break;
}
// End of this scope, pop it and try the next child in the containing
// scope.
MatchScopes.pop_back();
}
}
}
void SelectionDAGISel::CannotYetSelect(SDNode *N) {
std::string msg;
raw_string_ostream Msg(msg);
Msg << "Cannot select: ";
if (N->getOpcode() != ISD::INTRINSIC_W_CHAIN &&
N->getOpcode() != ISD::INTRINSIC_WO_CHAIN &&
N->getOpcode() != ISD::INTRINSIC_VOID) {
N->printrFull(Msg, CurDAG);
} else {
bool HasInputChain = N->getOperand(0).getValueType() == MVT::Other;
unsigned iid =
cast<ConstantSDNode>(N->getOperand(HasInputChain))->getZExtValue();
if (iid < Intrinsic::num_intrinsics)
Msg << "intrinsic %" << Intrinsic::getName((Intrinsic::ID)iid);
else if (const TargetIntrinsicInfo *TII = TM.getIntrinsicInfo())
Msg << "target intrinsic %" << TII->getName(iid);
else
Msg << "unknown intrinsic #" << iid;
}
report_fatal_error(Msg.str());
}
char SelectionDAGISel::ID = 0;
|