aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/llvm14/lib/CodeGen/MachineScheduler.cpp
blob: b043d4c1b0c1fb8c1c1e74050858fb172328f75b (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
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
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
//===- MachineScheduler.cpp - Machine Instruction Scheduler ---------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// MachineScheduler schedules machine instructions after phi elimination. It
// preserves LiveIntervals so it can be invoked before register allocation.
//
//===----------------------------------------------------------------------===//

#include "llvm/CodeGen/MachineScheduler.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/BitVector.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/PriorityQueue.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/ADT/iterator_range.h"
#include "llvm/Analysis/AliasAnalysis.h"
#include "llvm/CodeGen/LiveInterval.h"
#include "llvm/CodeGen/LiveIntervals.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineDominators.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineFunctionPass.h"
#include "llvm/CodeGen/MachineInstr.h"
#include "llvm/CodeGen/MachineLoopInfo.h"
#include "llvm/CodeGen/MachineOperand.h"
#include "llvm/CodeGen/MachinePassRegistry.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/CodeGen/Passes.h"
#include "llvm/CodeGen/RegisterClassInfo.h"
#include "llvm/CodeGen/RegisterPressure.h"
#include "llvm/CodeGen/ScheduleDAG.h"
#include "llvm/CodeGen/ScheduleDAGInstrs.h"
#include "llvm/CodeGen/ScheduleDAGMutation.h"
#include "llvm/CodeGen/ScheduleDFS.h"
#include "llvm/CodeGen/ScheduleHazardRecognizer.h"
#include "llvm/CodeGen/SlotIndexes.h"
#include "llvm/CodeGen/TargetFrameLowering.h"
#include "llvm/CodeGen/TargetInstrInfo.h"
#include "llvm/CodeGen/TargetLowering.h"
#include "llvm/CodeGen/TargetPassConfig.h"
#include "llvm/CodeGen/TargetRegisterInfo.h"
#include "llvm/CodeGen/TargetSchedule.h"
#include "llvm/CodeGen/TargetSubtargetInfo.h"
#include "llvm/Config/llvm-config.h"
#include "llvm/InitializePasses.h"
#include "llvm/MC/LaneBitmask.h"
#include "llvm/Pass.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Compiler.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/GraphWriter.h"
#include "llvm/Support/MachineValueType.h"
#include "llvm/Support/raw_ostream.h"
#include <algorithm>
#include <cassert>
#include <cstdint>
#include <iterator>
#include <limits>
#include <memory>
#include <string>
#include <tuple>
#include <utility>
#include <vector>

using namespace llvm;

#define DEBUG_TYPE "machine-scheduler"

STATISTIC(NumClustered, "Number of load/store pairs clustered");

namespace llvm {

cl::opt<bool> ForceTopDown("misched-topdown", cl::Hidden,
                           cl::desc("Force top-down list scheduling"));
cl::opt<bool> ForceBottomUp("misched-bottomup", cl::Hidden,
                            cl::desc("Force bottom-up list scheduling"));
cl::opt<bool>
DumpCriticalPathLength("misched-dcpl", cl::Hidden,
                       cl::desc("Print critical path length to stdout"));

cl::opt<bool> VerifyScheduling(
    "verify-misched", cl::Hidden,
    cl::desc("Verify machine instrs before and after machine scheduling"));

#ifndef NDEBUG
cl::opt<bool> ViewMISchedDAGs(
    "view-misched-dags", cl::Hidden,
    cl::desc("Pop up a window to show MISched dags after they are processed"));
#else
const bool ViewMISchedDAGs = false;
#endif // NDEBUG

} // end namespace llvm

#ifndef NDEBUG
/// In some situations a few uninteresting nodes depend on nearly all other
/// nodes in the graph, provide a cutoff to hide them.
static cl::opt<unsigned> ViewMISchedCutoff("view-misched-cutoff", cl::Hidden,
  cl::desc("Hide nodes with more predecessor/successor than cutoff"));

static cl::opt<unsigned> MISchedCutoff("misched-cutoff", cl::Hidden,
  cl::desc("Stop scheduling after N instructions"), cl::init(~0U));

static cl::opt<std::string> SchedOnlyFunc("misched-only-func", cl::Hidden,
  cl::desc("Only schedule this function"));
static cl::opt<unsigned> SchedOnlyBlock("misched-only-block", cl::Hidden,
                                        cl::desc("Only schedule this MBB#"));
static cl::opt<bool> PrintDAGs("misched-print-dags", cl::Hidden,
                              cl::desc("Print schedule DAGs"));
#else
static const bool PrintDAGs = false;
#endif // NDEBUG

/// Avoid quadratic complexity in unusually large basic blocks by limiting the
/// size of the ready lists.
static cl::opt<unsigned> ReadyListLimit("misched-limit", cl::Hidden,
  cl::desc("Limit ready list to N instructions"), cl::init(256));

static cl::opt<bool> EnableRegPressure("misched-regpressure", cl::Hidden,
  cl::desc("Enable register pressure scheduling."), cl::init(true));

static cl::opt<bool> EnableCyclicPath("misched-cyclicpath", cl::Hidden,
  cl::desc("Enable cyclic critical path analysis."), cl::init(true));

static cl::opt<bool> EnableMemOpCluster("misched-cluster", cl::Hidden,
                                        cl::desc("Enable memop clustering."),
                                        cl::init(true));
static cl::opt<bool>
    ForceFastCluster("force-fast-cluster", cl::Hidden,
                     cl::desc("Switch to fast cluster algorithm with the lost "
                              "of some fusion opportunities"),
                     cl::init(false));
static cl::opt<unsigned>
    FastClusterThreshold("fast-cluster-threshold", cl::Hidden,
                         cl::desc("The threshold for fast cluster"),
                         cl::init(1000));

// DAG subtrees must have at least this many nodes.
static const unsigned MinSubtreeSize = 8;

// Pin the vtables to this file.
void MachineSchedStrategy::anchor() {}

void ScheduleDAGMutation::anchor() {}

//===----------------------------------------------------------------------===//
// Machine Instruction Scheduling Pass and Registry
//===----------------------------------------------------------------------===//

MachineSchedContext::MachineSchedContext() {
  RegClassInfo = new RegisterClassInfo();
}

MachineSchedContext::~MachineSchedContext() {
  delete RegClassInfo;
}

namespace {

/// Base class for a machine scheduler class that can run at any point.
class MachineSchedulerBase : public MachineSchedContext,
                             public MachineFunctionPass {
public:
  MachineSchedulerBase(char &ID): MachineFunctionPass(ID) {}

  void print(raw_ostream &O, const Module* = nullptr) const override;

protected:
  void scheduleRegions(ScheduleDAGInstrs &Scheduler, bool FixKillFlags);
};

/// MachineScheduler runs after coalescing and before register allocation.
class MachineScheduler : public MachineSchedulerBase {
public:
  MachineScheduler();

  void getAnalysisUsage(AnalysisUsage &AU) const override;

  bool runOnMachineFunction(MachineFunction&) override;

  static char ID; // Class identification, replacement for typeinfo

protected:
  ScheduleDAGInstrs *createMachineScheduler();
};

/// PostMachineScheduler runs after shortly before code emission.
class PostMachineScheduler : public MachineSchedulerBase {
public:
  PostMachineScheduler();

  void getAnalysisUsage(AnalysisUsage &AU) const override;

  bool runOnMachineFunction(MachineFunction&) override;

  static char ID; // Class identification, replacement for typeinfo

protected:
  ScheduleDAGInstrs *createPostMachineScheduler();
};

} // end anonymous namespace

char MachineScheduler::ID = 0;

char &llvm::MachineSchedulerID = MachineScheduler::ID;

INITIALIZE_PASS_BEGIN(MachineScheduler, DEBUG_TYPE,
                      "Machine Instruction Scheduler", false, false)
INITIALIZE_PASS_DEPENDENCY(AAResultsWrapperPass)
INITIALIZE_PASS_DEPENDENCY(MachineDominatorTree)
INITIALIZE_PASS_DEPENDENCY(MachineLoopInfo)
INITIALIZE_PASS_DEPENDENCY(SlotIndexes)
INITIALIZE_PASS_DEPENDENCY(LiveIntervals)
INITIALIZE_PASS_END(MachineScheduler, DEBUG_TYPE,
                    "Machine Instruction Scheduler", false, false)

MachineScheduler::MachineScheduler() : MachineSchedulerBase(ID) {
  initializeMachineSchedulerPass(*PassRegistry::getPassRegistry());
}

void MachineScheduler::getAnalysisUsage(AnalysisUsage &AU) const {
  AU.setPreservesCFG();
  AU.addRequired<MachineDominatorTree>();
  AU.addRequired<MachineLoopInfo>();
  AU.addRequired<AAResultsWrapperPass>();
  AU.addRequired<TargetPassConfig>();
  AU.addRequired<SlotIndexes>();
  AU.addPreserved<SlotIndexes>();
  AU.addRequired<LiveIntervals>();
  AU.addPreserved<LiveIntervals>();
  MachineFunctionPass::getAnalysisUsage(AU);
}

char PostMachineScheduler::ID = 0;

char &llvm::PostMachineSchedulerID = PostMachineScheduler::ID;

INITIALIZE_PASS_BEGIN(PostMachineScheduler, "postmisched",
                      "PostRA Machine Instruction Scheduler", false, false)
INITIALIZE_PASS_DEPENDENCY(MachineDominatorTree)
INITIALIZE_PASS_DEPENDENCY(MachineLoopInfo)
INITIALIZE_PASS_DEPENDENCY(AAResultsWrapperPass)
INITIALIZE_PASS_END(PostMachineScheduler, "postmisched",
                    "PostRA Machine Instruction Scheduler", false, false)

PostMachineScheduler::PostMachineScheduler() : MachineSchedulerBase(ID) {
  initializePostMachineSchedulerPass(*PassRegistry::getPassRegistry());
}

void PostMachineScheduler::getAnalysisUsage(AnalysisUsage &AU) const {
  AU.setPreservesCFG();
  AU.addRequired<MachineDominatorTree>();
  AU.addRequired<MachineLoopInfo>();
  AU.addRequired<AAResultsWrapperPass>();
  AU.addRequired<TargetPassConfig>();
  MachineFunctionPass::getAnalysisUsage(AU);
}

MachinePassRegistry<MachineSchedRegistry::ScheduleDAGCtor>
    MachineSchedRegistry::Registry;

/// A dummy default scheduler factory indicates whether the scheduler
/// is overridden on the command line.
static ScheduleDAGInstrs *useDefaultMachineSched(MachineSchedContext *C) {
  return nullptr;
}

/// MachineSchedOpt allows command line selection of the scheduler.
static cl::opt<MachineSchedRegistry::ScheduleDAGCtor, false,
               RegisterPassParser<MachineSchedRegistry>>
MachineSchedOpt("misched",
                cl::init(&useDefaultMachineSched), cl::Hidden,
                cl::desc("Machine instruction scheduler to use"));

static MachineSchedRegistry
DefaultSchedRegistry("default", "Use the target's default scheduler choice.",
                     useDefaultMachineSched);

static cl::opt<bool> EnableMachineSched(
    "enable-misched",
    cl::desc("Enable the machine instruction scheduling pass."), cl::init(true),
    cl::Hidden);

static cl::opt<bool> EnablePostRAMachineSched(
    "enable-post-misched",
    cl::desc("Enable the post-ra machine instruction scheduling pass."),
    cl::init(true), cl::Hidden);

/// Decrement this iterator until reaching the top or a non-debug instr.
static MachineBasicBlock::const_iterator
priorNonDebug(MachineBasicBlock::const_iterator I,
              MachineBasicBlock::const_iterator Beg) {
  assert(I != Beg && "reached the top of the region, cannot decrement");
  while (--I != Beg) {
    if (!I->isDebugOrPseudoInstr())
      break;
  }
  return I;
}

/// Non-const version.
static MachineBasicBlock::iterator
priorNonDebug(MachineBasicBlock::iterator I,
              MachineBasicBlock::const_iterator Beg) {
  return priorNonDebug(MachineBasicBlock::const_iterator(I), Beg)
      .getNonConstIterator();
}

/// If this iterator is a debug value, increment until reaching the End or a
/// non-debug instruction.
static MachineBasicBlock::const_iterator
nextIfDebug(MachineBasicBlock::const_iterator I,
            MachineBasicBlock::const_iterator End) {
  for(; I != End; ++I) {
    if (!I->isDebugOrPseudoInstr())
      break;
  }
  return I;
}

/// Non-const version.
static MachineBasicBlock::iterator
nextIfDebug(MachineBasicBlock::iterator I,
            MachineBasicBlock::const_iterator End) {
  return nextIfDebug(MachineBasicBlock::const_iterator(I), End)
      .getNonConstIterator();
}

/// Instantiate a ScheduleDAGInstrs that will be owned by the caller.
ScheduleDAGInstrs *MachineScheduler::createMachineScheduler() {
  // Select the scheduler, or set the default.
  MachineSchedRegistry::ScheduleDAGCtor Ctor = MachineSchedOpt;
  if (Ctor != useDefaultMachineSched)
    return Ctor(this);

  // Get the default scheduler set by the target for this function.
  ScheduleDAGInstrs *Scheduler = PassConfig->createMachineScheduler(this);
  if (Scheduler)
    return Scheduler;

  // Default to GenericScheduler.
  return createGenericSchedLive(this);
}

/// Instantiate a ScheduleDAGInstrs for PostRA scheduling that will be owned by
/// the caller. We don't have a command line option to override the postRA
/// scheduler. The Target must configure it.
ScheduleDAGInstrs *PostMachineScheduler::createPostMachineScheduler() {
  // Get the postRA scheduler set by the target for this function.
  ScheduleDAGInstrs *Scheduler = PassConfig->createPostMachineScheduler(this);
  if (Scheduler)
    return Scheduler;

  // Default to GenericScheduler.
  return createGenericSchedPostRA(this);
}

/// Top-level MachineScheduler pass driver.
///
/// Visit blocks in function order. Divide each block into scheduling regions
/// and visit them bottom-up. Visiting regions bottom-up is not required, but is
/// consistent with the DAG builder, which traverses the interior of the
/// scheduling regions bottom-up.
///
/// This design avoids exposing scheduling boundaries to the DAG builder,
/// simplifying the DAG builder's support for "special" target instructions.
/// At the same time the design allows target schedulers to operate across
/// scheduling boundaries, for example to bundle the boundary instructions
/// without reordering them. This creates complexity, because the target
/// scheduler must update the RegionBegin and RegionEnd positions cached by
/// ScheduleDAGInstrs whenever adding or removing instructions. A much simpler
/// design would be to split blocks at scheduling boundaries, but LLVM has a
/// general bias against block splitting purely for implementation simplicity.
bool MachineScheduler::runOnMachineFunction(MachineFunction &mf) {
  if (skipFunction(mf.getFunction()))
    return false;

  if (EnableMachineSched.getNumOccurrences()) {
    if (!EnableMachineSched)
      return false;
  } else if (!mf.getSubtarget().enableMachineScheduler())
    return false;

  LLVM_DEBUG(dbgs() << "Before MISched:\n"; mf.print(dbgs()));

  // Initialize the context of the pass.
  MF = &mf;
  MLI = &getAnalysis<MachineLoopInfo>();
  MDT = &getAnalysis<MachineDominatorTree>();
  PassConfig = &getAnalysis<TargetPassConfig>();
  AA = &getAnalysis<AAResultsWrapperPass>().getAAResults();

  LIS = &getAnalysis<LiveIntervals>();

  if (VerifyScheduling) {
    LLVM_DEBUG(LIS->dump());
    MF->verify(this, "Before machine scheduling.");
  }
  RegClassInfo->runOnMachineFunction(*MF);

  // Instantiate the selected scheduler for this target, function, and
  // optimization level.
  std::unique_ptr<ScheduleDAGInstrs> Scheduler(createMachineScheduler());
  scheduleRegions(*Scheduler, false);

  LLVM_DEBUG(LIS->dump());
  if (VerifyScheduling)
    MF->verify(this, "After machine scheduling.");
  return true;
}

bool PostMachineScheduler::runOnMachineFunction(MachineFunction &mf) {
  if (skipFunction(mf.getFunction()))
    return false;

  if (EnablePostRAMachineSched.getNumOccurrences()) {
    if (!EnablePostRAMachineSched)
      return false;
  } else if (!mf.getSubtarget().enablePostRAMachineScheduler()) {
    LLVM_DEBUG(dbgs() << "Subtarget disables post-MI-sched.\n");
    return false;
  }
  LLVM_DEBUG(dbgs() << "Before post-MI-sched:\n"; mf.print(dbgs()));

  // Initialize the context of the pass.
  MF = &mf;
  MLI = &getAnalysis<MachineLoopInfo>();
  PassConfig = &getAnalysis<TargetPassConfig>();
  AA = &getAnalysis<AAResultsWrapperPass>().getAAResults();

  if (VerifyScheduling)
    MF->verify(this, "Before post machine scheduling.");

  // Instantiate the selected scheduler for this target, function, and
  // optimization level.
  std::unique_ptr<ScheduleDAGInstrs> Scheduler(createPostMachineScheduler());
  scheduleRegions(*Scheduler, true);

  if (VerifyScheduling)
    MF->verify(this, "After post machine scheduling.");
  return true;
}

/// Return true of the given instruction should not be included in a scheduling
/// region.
///
/// MachineScheduler does not currently support scheduling across calls. To
/// handle calls, the DAG builder needs to be modified to create register
/// anti/output dependencies on the registers clobbered by the call's regmask
/// operand. In PreRA scheduling, the stack pointer adjustment already prevents
/// scheduling across calls. In PostRA scheduling, we need the isCall to enforce
/// the boundary, but there would be no benefit to postRA scheduling across
/// calls this late anyway.
static bool isSchedBoundary(MachineBasicBlock::iterator MI,
                            MachineBasicBlock *MBB,
                            MachineFunction *MF,
                            const TargetInstrInfo *TII) {
  return MI->isCall() || TII->isSchedulingBoundary(*MI, MBB, *MF);
}

/// A region of an MBB for scheduling.
namespace {
struct SchedRegion {
  /// RegionBegin is the first instruction in the scheduling region, and
  /// RegionEnd is either MBB->end() or the scheduling boundary after the
  /// last instruction in the scheduling region. These iterators cannot refer
  /// to instructions outside of the identified scheduling region because
  /// those may be reordered before scheduling this region.
  MachineBasicBlock::iterator RegionBegin;
  MachineBasicBlock::iterator RegionEnd;
  unsigned NumRegionInstrs;

  SchedRegion(MachineBasicBlock::iterator B, MachineBasicBlock::iterator E,
              unsigned N) :
    RegionBegin(B), RegionEnd(E), NumRegionInstrs(N) {}
};
} // end anonymous namespace

using MBBRegionsVector = SmallVector<SchedRegion, 16>;

static void
getSchedRegions(MachineBasicBlock *MBB,
                MBBRegionsVector &Regions,
                bool RegionsTopDown) {
  MachineFunction *MF = MBB->getParent();
  const TargetInstrInfo *TII = MF->getSubtarget().getInstrInfo();

  MachineBasicBlock::iterator I = nullptr;
  for(MachineBasicBlock::iterator RegionEnd = MBB->end();
      RegionEnd != MBB->begin(); RegionEnd = I) {

    // Avoid decrementing RegionEnd for blocks with no terminator.
    if (RegionEnd != MBB->end() ||
        isSchedBoundary(&*std::prev(RegionEnd), &*MBB, MF, TII)) {
      --RegionEnd;
    }

    // The next region starts above the previous region. Look backward in the
    // instruction stream until we find the nearest boundary.
    unsigned NumRegionInstrs = 0;
    I = RegionEnd;
    for (;I != MBB->begin(); --I) {
      MachineInstr &MI = *std::prev(I);
      if (isSchedBoundary(&MI, &*MBB, MF, TII))
        break;
      if (!MI.isDebugOrPseudoInstr()) {
        // MBB::size() uses instr_iterator to count. Here we need a bundle to
        // count as a single instruction.
        ++NumRegionInstrs;
      }
    }

    // It's possible we found a scheduling region that only has debug
    // instructions. Don't bother scheduling these.
    if (NumRegionInstrs != 0)
      Regions.push_back(SchedRegion(I, RegionEnd, NumRegionInstrs));
  }

  if (RegionsTopDown)
    std::reverse(Regions.begin(), Regions.end());
}

/// Main driver for both MachineScheduler and PostMachineScheduler.
void MachineSchedulerBase::scheduleRegions(ScheduleDAGInstrs &Scheduler,
                                           bool FixKillFlags) {
  // Visit all machine basic blocks.
  //
  // TODO: Visit blocks in global postorder or postorder within the bottom-up
  // loop tree. Then we can optionally compute global RegPressure.
  for (MachineFunction::iterator MBB = MF->begin(), MBBEnd = MF->end();
       MBB != MBBEnd; ++MBB) {

    Scheduler.startBlock(&*MBB);

#ifndef NDEBUG
    if (SchedOnlyFunc.getNumOccurrences() && SchedOnlyFunc != MF->getName())
      continue;
    if (SchedOnlyBlock.getNumOccurrences()
        && (int)SchedOnlyBlock != MBB->getNumber())
      continue;
#endif

    // Break the block into scheduling regions [I, RegionEnd). RegionEnd
    // points to the scheduling boundary at the bottom of the region. The DAG
    // does not include RegionEnd, but the region does (i.e. the next
    // RegionEnd is above the previous RegionBegin). If the current block has
    // no terminator then RegionEnd == MBB->end() for the bottom region.
    //
    // All the regions of MBB are first found and stored in MBBRegions, which
    // will be processed (MBB) top-down if initialized with true.
    //
    // The Scheduler may insert instructions during either schedule() or
    // exitRegion(), even for empty regions. So the local iterators 'I' and
    // 'RegionEnd' are invalid across these calls. Instructions must not be
    // added to other regions than the current one without updating MBBRegions.

    MBBRegionsVector MBBRegions;
    getSchedRegions(&*MBB, MBBRegions, Scheduler.doMBBSchedRegionsTopDown());
    for (const SchedRegion &R : MBBRegions) {
      MachineBasicBlock::iterator I = R.RegionBegin;
      MachineBasicBlock::iterator RegionEnd = R.RegionEnd;
      unsigned NumRegionInstrs = R.NumRegionInstrs;

      // Notify the scheduler of the region, even if we may skip scheduling
      // it. Perhaps it still needs to be bundled.
      Scheduler.enterRegion(&*MBB, I, RegionEnd, NumRegionInstrs);

      // Skip empty scheduling regions (0 or 1 schedulable instructions).
      if (I == RegionEnd || I == std::prev(RegionEnd)) {
        // Close the current region. Bundle the terminator if needed.
        // This invalidates 'RegionEnd' and 'I'.
        Scheduler.exitRegion();
        continue;
      }
      LLVM_DEBUG(dbgs() << "********** MI Scheduling **********\n");
      LLVM_DEBUG(dbgs() << MF->getName() << ":" << printMBBReference(*MBB)
                        << " " << MBB->getName() << "\n  From: " << *I
                        << "    To: ";
                 if (RegionEnd != MBB->end()) dbgs() << *RegionEnd;
                 else dbgs() << "End\n";
                 dbgs() << " RegionInstrs: " << NumRegionInstrs << '\n');
      if (DumpCriticalPathLength) {
        errs() << MF->getName();
        errs() << ":%bb. " << MBB->getNumber();
        errs() << " " << MBB->getName() << " \n";
      }

      // Schedule a region: possibly reorder instructions.
      // This invalidates the original region iterators.
      Scheduler.schedule();

      // Close the current region.
      Scheduler.exitRegion();
    }
    Scheduler.finishBlock();
    // FIXME: Ideally, no further passes should rely on kill flags. However,
    // thumb2 size reduction is currently an exception, so the PostMIScheduler
    // needs to do this.
    if (FixKillFlags)
      Scheduler.fixupKills(*MBB);
  }
  Scheduler.finalizeSchedule();
}

void MachineSchedulerBase::print(raw_ostream &O, const Module* m) const {
  // unimplemented
}

#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
LLVM_DUMP_METHOD void ReadyQueue::dump() const {
  dbgs() << "Queue " << Name << ": ";
  for (const SUnit *SU : Queue)
    dbgs() << SU->NodeNum << " ";
  dbgs() << "\n";
}
#endif

//===----------------------------------------------------------------------===//
// ScheduleDAGMI - Basic machine instruction scheduling. This is
// independent of PreRA/PostRA scheduling and involves no extra book-keeping for
// virtual registers.
// ===----------------------------------------------------------------------===/

// Provide a vtable anchor.
ScheduleDAGMI::~ScheduleDAGMI() = default;

/// ReleaseSucc - Decrement the NumPredsLeft count of a successor. When
/// NumPredsLeft reaches zero, release the successor node.
///
/// FIXME: Adjust SuccSU height based on MinLatency.
void ScheduleDAGMI::releaseSucc(SUnit *SU, SDep *SuccEdge) {
  SUnit *SuccSU = SuccEdge->getSUnit();

  if (SuccEdge->isWeak()) {
    --SuccSU->WeakPredsLeft;
    if (SuccEdge->isCluster())
      NextClusterSucc = SuccSU;
    return;
  }
#ifndef NDEBUG
  if (SuccSU->NumPredsLeft == 0) {
    dbgs() << "*** Scheduling failed! ***\n";
    dumpNode(*SuccSU);
    dbgs() << " has been released too many times!\n";
    llvm_unreachable(nullptr);
  }
#endif
  // SU->TopReadyCycle was set to CurrCycle when it was scheduled. However,
  // CurrCycle may have advanced since then.
  if (SuccSU->TopReadyCycle < SU->TopReadyCycle + SuccEdge->getLatency())
    SuccSU->TopReadyCycle = SU->TopReadyCycle + SuccEdge->getLatency();

  --SuccSU->NumPredsLeft;
  if (SuccSU->NumPredsLeft == 0 && SuccSU != &ExitSU)
    SchedImpl->releaseTopNode(SuccSU);
}

/// releaseSuccessors - Call releaseSucc on each of SU's successors.
void ScheduleDAGMI::releaseSuccessors(SUnit *SU) {
  for (SDep &Succ : SU->Succs)
    releaseSucc(SU, &Succ);
}

/// ReleasePred - Decrement the NumSuccsLeft count of a predecessor. When
/// NumSuccsLeft reaches zero, release the predecessor node.
///
/// FIXME: Adjust PredSU height based on MinLatency.
void ScheduleDAGMI::releasePred(SUnit *SU, SDep *PredEdge) {
  SUnit *PredSU = PredEdge->getSUnit();

  if (PredEdge->isWeak()) {
    --PredSU->WeakSuccsLeft;
    if (PredEdge->isCluster())
      NextClusterPred = PredSU;
    return;
  }
#ifndef NDEBUG
  if (PredSU->NumSuccsLeft == 0) {
    dbgs() << "*** Scheduling failed! ***\n";
    dumpNode(*PredSU);
    dbgs() << " has been released too many times!\n";
    llvm_unreachable(nullptr);
  }
#endif
  // SU->BotReadyCycle was set to CurrCycle when it was scheduled. However,
  // CurrCycle may have advanced since then.
  if (PredSU->BotReadyCycle < SU->BotReadyCycle + PredEdge->getLatency())
    PredSU->BotReadyCycle = SU->BotReadyCycle + PredEdge->getLatency();

  --PredSU->NumSuccsLeft;
  if (PredSU->NumSuccsLeft == 0 && PredSU != &EntrySU)
    SchedImpl->releaseBottomNode(PredSU);
}

/// releasePredecessors - Call releasePred on each of SU's predecessors.
void ScheduleDAGMI::releasePredecessors(SUnit *SU) {
  for (SDep &Pred : SU->Preds)
    releasePred(SU, &Pred);
}

void ScheduleDAGMI::startBlock(MachineBasicBlock *bb) {
  ScheduleDAGInstrs::startBlock(bb);
  SchedImpl->enterMBB(bb);
}

void ScheduleDAGMI::finishBlock() {
  SchedImpl->leaveMBB();
  ScheduleDAGInstrs::finishBlock();
}

/// enterRegion - Called back from MachineScheduler::runOnMachineFunction after
/// crossing a scheduling boundary. [begin, end) includes all instructions in
/// the region, including the boundary itself and single-instruction regions
/// that don't get scheduled.
void ScheduleDAGMI::enterRegion(MachineBasicBlock *bb,
                                     MachineBasicBlock::iterator begin,
                                     MachineBasicBlock::iterator end,
                                     unsigned regioninstrs)
{
  ScheduleDAGInstrs::enterRegion(bb, begin, end, regioninstrs);

  SchedImpl->initPolicy(begin, end, regioninstrs);
}

/// This is normally called from the main scheduler loop but may also be invoked
/// by the scheduling strategy to perform additional code motion.
void ScheduleDAGMI::moveInstruction(
  MachineInstr *MI, MachineBasicBlock::iterator InsertPos) {
  // Advance RegionBegin if the first instruction moves down.
  if (&*RegionBegin == MI)
    ++RegionBegin;

  // Update the instruction stream.
  BB->splice(InsertPos, BB, MI);

  // Update LiveIntervals
  if (LIS)
    LIS->handleMove(*MI, /*UpdateFlags=*/true);

  // Recede RegionBegin if an instruction moves above the first.
  if (RegionBegin == InsertPos)
    RegionBegin = MI;
}

bool ScheduleDAGMI::checkSchedLimit() {
#ifndef NDEBUG
  if (NumInstrsScheduled == MISchedCutoff && MISchedCutoff != ~0U) {
    CurrentTop = CurrentBottom;
    return false;
  }
  ++NumInstrsScheduled;
#endif
  return true;
}

/// Per-region scheduling driver, called back from
/// MachineScheduler::runOnMachineFunction. This is a simplified driver that
/// does not consider liveness or register pressure. It is useful for PostRA
/// scheduling and potentially other custom schedulers.
void ScheduleDAGMI::schedule() {
  LLVM_DEBUG(dbgs() << "ScheduleDAGMI::schedule starting\n");
  LLVM_DEBUG(SchedImpl->dumpPolicy());

  // Build the DAG.
  buildSchedGraph(AA);

  postprocessDAG();

  SmallVector<SUnit*, 8> TopRoots, BotRoots;
  findRootsAndBiasEdges(TopRoots, BotRoots);

  LLVM_DEBUG(dump());
  if (PrintDAGs) dump();
  if (ViewMISchedDAGs) viewGraph();

  // Initialize the strategy before modifying the DAG.
  // This may initialize a DFSResult to be used for queue priority.
  SchedImpl->initialize(this);

  // Initialize ready queues now that the DAG and priority data are finalized.
  initQueues(TopRoots, BotRoots);

  bool IsTopNode = false;
  while (true) {
    LLVM_DEBUG(dbgs() << "** ScheduleDAGMI::schedule picking next node\n");
    SUnit *SU = SchedImpl->pickNode(IsTopNode);
    if (!SU) break;

    assert(!SU->isScheduled && "Node already scheduled");
    if (!checkSchedLimit())
      break;

    MachineInstr *MI = SU->getInstr();
    if (IsTopNode) {
      assert(SU->isTopReady() && "node still has unscheduled dependencies");
      if (&*CurrentTop == MI)
        CurrentTop = nextIfDebug(++CurrentTop, CurrentBottom);
      else
        moveInstruction(MI, CurrentTop);
    } else {
      assert(SU->isBottomReady() && "node still has unscheduled dependencies");
      MachineBasicBlock::iterator priorII =
        priorNonDebug(CurrentBottom, CurrentTop);
      if (&*priorII == MI)
        CurrentBottom = priorII;
      else {
        if (&*CurrentTop == MI)
          CurrentTop = nextIfDebug(++CurrentTop, priorII);
        moveInstruction(MI, CurrentBottom);
        CurrentBottom = MI;
      }
    }
    // Notify the scheduling strategy before updating the DAG.
    // This sets the scheduled node's ReadyCycle to CurrCycle. When updateQueues
    // runs, it can then use the accurate ReadyCycle time to determine whether
    // newly released nodes can move to the readyQ.
    SchedImpl->schedNode(SU, IsTopNode);

    updateQueues(SU, IsTopNode);
  }
  assert(CurrentTop == CurrentBottom && "Nonempty unscheduled zone.");

  placeDebugValues();

  LLVM_DEBUG({
    dbgs() << "*** Final schedule for "
           << printMBBReference(*begin()->getParent()) << " ***\n";
    dumpSchedule();
    dbgs() << '\n';
  });
}

/// Apply each ScheduleDAGMutation step in order.
void ScheduleDAGMI::postprocessDAG() {
  for (auto &m : Mutations)
    m->apply(this);
}

void ScheduleDAGMI::
findRootsAndBiasEdges(SmallVectorImpl<SUnit*> &TopRoots,
                      SmallVectorImpl<SUnit*> &BotRoots) {
  for (SUnit &SU : SUnits) {
    assert(!SU.isBoundaryNode() && "Boundary node should not be in SUnits");

    // Order predecessors so DFSResult follows the critical path.
    SU.biasCriticalPath();

    // A SUnit is ready to top schedule if it has no predecessors.
    if (!SU.NumPredsLeft)
      TopRoots.push_back(&SU);
    // A SUnit is ready to bottom schedule if it has no successors.
    if (!SU.NumSuccsLeft)
      BotRoots.push_back(&SU);
  }
  ExitSU.biasCriticalPath();
}

/// Identify DAG roots and setup scheduler queues.
void ScheduleDAGMI::initQueues(ArrayRef<SUnit*> TopRoots,
                               ArrayRef<SUnit*> BotRoots) {
  NextClusterSucc = nullptr;
  NextClusterPred = nullptr;

  // Release all DAG roots for scheduling, not including EntrySU/ExitSU.
  //
  // Nodes with unreleased weak edges can still be roots.
  // Release top roots in forward order.
  for (SUnit *SU : TopRoots)
    SchedImpl->releaseTopNode(SU);

  // Release bottom roots in reverse order so the higher priority nodes appear
  // first. This is more natural and slightly more efficient.
  for (SmallVectorImpl<SUnit*>::const_reverse_iterator
         I = BotRoots.rbegin(), E = BotRoots.rend(); I != E; ++I) {
    SchedImpl->releaseBottomNode(*I);
  }

  releaseSuccessors(&EntrySU);
  releasePredecessors(&ExitSU);

  SchedImpl->registerRoots();

  // Advance past initial DebugValues.
  CurrentTop = nextIfDebug(RegionBegin, RegionEnd);
  CurrentBottom = RegionEnd;
}

/// Update scheduler queues after scheduling an instruction.
void ScheduleDAGMI::updateQueues(SUnit *SU, bool IsTopNode) {
  // Release dependent instructions for scheduling.
  if (IsTopNode)
    releaseSuccessors(SU);
  else
    releasePredecessors(SU);

  SU->isScheduled = true;
}

/// Reinsert any remaining debug_values, just like the PostRA scheduler.
void ScheduleDAGMI::placeDebugValues() {
  // If first instruction was a DBG_VALUE then put it back.
  if (FirstDbgValue) {
    BB->splice(RegionBegin, BB, FirstDbgValue);
    RegionBegin = FirstDbgValue;
  }

  for (std::vector<std::pair<MachineInstr *, MachineInstr *>>::iterator
         DI = DbgValues.end(), DE = DbgValues.begin(); DI != DE; --DI) {
    std::pair<MachineInstr *, MachineInstr *> P = *std::prev(DI);
    MachineInstr *DbgValue = P.first;
    MachineBasicBlock::iterator OrigPrevMI = P.second;
    if (&*RegionBegin == DbgValue)
      ++RegionBegin;
    BB->splice(++OrigPrevMI, BB, DbgValue);
    if (OrigPrevMI == std::prev(RegionEnd))
      RegionEnd = DbgValue;
  }
  DbgValues.clear();
  FirstDbgValue = nullptr;
}

#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
LLVM_DUMP_METHOD void ScheduleDAGMI::dumpSchedule() const {
  for (MachineInstr &MI : *this) {
    if (SUnit *SU = getSUnit(&MI))
      dumpNode(*SU);
    else
      dbgs() << "Missing SUnit\n";
  }
}
#endif

//===----------------------------------------------------------------------===//
// ScheduleDAGMILive - Base class for MachineInstr scheduling with LiveIntervals
// preservation.
//===----------------------------------------------------------------------===//

ScheduleDAGMILive::~ScheduleDAGMILive() {
  delete DFSResult;
}

void ScheduleDAGMILive::collectVRegUses(SUnit &SU) {
  const MachineInstr &MI = *SU.getInstr();
  for (const MachineOperand &MO : MI.operands()) {
    if (!MO.isReg())
      continue;
    if (!MO.readsReg())
      continue;
    if (TrackLaneMasks && !MO.isUse())
      continue;

    Register Reg = MO.getReg();
    if (!Register::isVirtualRegister(Reg))
      continue;

    // Ignore re-defs.
    if (TrackLaneMasks) {
      bool FoundDef = false;
      for (const MachineOperand &MO2 : MI.operands()) {
        if (MO2.isReg() && MO2.isDef() && MO2.getReg() == Reg && !MO2.isDead()) {
          FoundDef = true;
          break;
        }
      }
      if (FoundDef)
        continue;
    }

    // Record this local VReg use.
    VReg2SUnitMultiMap::iterator UI = VRegUses.find(Reg);
    for (; UI != VRegUses.end(); ++UI) {
      if (UI->SU == &SU)
        break;
    }
    if (UI == VRegUses.end())
      VRegUses.insert(VReg2SUnit(Reg, LaneBitmask::getNone(), &SU));
  }
}

/// enterRegion - Called back from MachineScheduler::runOnMachineFunction after
/// crossing a scheduling boundary. [begin, end) includes all instructions in
/// the region, including the boundary itself and single-instruction regions
/// that don't get scheduled.
void ScheduleDAGMILive::enterRegion(MachineBasicBlock *bb,
                                MachineBasicBlock::iterator begin,
                                MachineBasicBlock::iterator end,
                                unsigned regioninstrs)
{
  // ScheduleDAGMI initializes SchedImpl's per-region policy.
  ScheduleDAGMI::enterRegion(bb, begin, end, regioninstrs);

  // For convenience remember the end of the liveness region.
  LiveRegionEnd = (RegionEnd == bb->end()) ? RegionEnd : std::next(RegionEnd);

  SUPressureDiffs.clear();

  ShouldTrackPressure = SchedImpl->shouldTrackPressure();
  ShouldTrackLaneMasks = SchedImpl->shouldTrackLaneMasks();

  assert((!ShouldTrackLaneMasks || ShouldTrackPressure) &&
         "ShouldTrackLaneMasks requires ShouldTrackPressure");
}

// Setup the register pressure trackers for the top scheduled and bottom
// scheduled regions.
void ScheduleDAGMILive::initRegPressure() {
  VRegUses.clear();
  VRegUses.setUniverse(MRI.getNumVirtRegs());
  for (SUnit &SU : SUnits)
    collectVRegUses(SU);

  TopRPTracker.init(&MF, RegClassInfo, LIS, BB, RegionBegin,
                    ShouldTrackLaneMasks, false);
  BotRPTracker.init(&MF, RegClassInfo, LIS, BB, LiveRegionEnd,
                    ShouldTrackLaneMasks, false);

  // Close the RPTracker to finalize live ins.
  RPTracker.closeRegion();

  LLVM_DEBUG(RPTracker.dump());

  // Initialize the live ins and live outs.
  TopRPTracker.addLiveRegs(RPTracker.getPressure().LiveInRegs);
  BotRPTracker.addLiveRegs(RPTracker.getPressure().LiveOutRegs);

  // Close one end of the tracker so we can call
  // getMaxUpward/DownwardPressureDelta before advancing across any
  // instructions. This converts currently live regs into live ins/outs.
  TopRPTracker.closeTop();
  BotRPTracker.closeBottom();

  BotRPTracker.initLiveThru(RPTracker);
  if (!BotRPTracker.getLiveThru().empty()) {
    TopRPTracker.initLiveThru(BotRPTracker.getLiveThru());
    LLVM_DEBUG(dbgs() << "Live Thru: ";
               dumpRegSetPressure(BotRPTracker.getLiveThru(), TRI));
  };

  // For each live out vreg reduce the pressure change associated with other
  // uses of the same vreg below the live-out reaching def.
  updatePressureDiffs(RPTracker.getPressure().LiveOutRegs);

  // Account for liveness generated by the region boundary.
  if (LiveRegionEnd != RegionEnd) {
    SmallVector<RegisterMaskPair, 8> LiveUses;
    BotRPTracker.recede(&LiveUses);
    updatePressureDiffs(LiveUses);
  }

  LLVM_DEBUG(dbgs() << "Top Pressure:\n";
             dumpRegSetPressure(TopRPTracker.getRegSetPressureAtPos(), TRI);
             dbgs() << "Bottom Pressure:\n";
             dumpRegSetPressure(BotRPTracker.getRegSetPressureAtPos(), TRI););

  assert((BotRPTracker.getPos() == RegionEnd ||
          (RegionEnd->isDebugInstr() &&
           BotRPTracker.getPos() == priorNonDebug(RegionEnd, RegionBegin))) &&
         "Can't find the region bottom");

  // Cache the list of excess pressure sets in this region. This will also track
  // the max pressure in the scheduled code for these sets.
  RegionCriticalPSets.clear();
  const std::vector<unsigned> &RegionPressure =
    RPTracker.getPressure().MaxSetPressure;
  for (unsigned i = 0, e = RegionPressure.size(); i < e; ++i) {
    unsigned Limit = RegClassInfo->getRegPressureSetLimit(i);
    if (RegionPressure[i] > Limit) {
      LLVM_DEBUG(dbgs() << TRI->getRegPressureSetName(i) << " Limit " << Limit
                        << " Actual " << RegionPressure[i] << "\n");
      RegionCriticalPSets.push_back(PressureChange(i));
    }
  }
  LLVM_DEBUG(dbgs() << "Excess PSets: ";
             for (const PressureChange &RCPS
                  : RegionCriticalPSets) dbgs()
             << TRI->getRegPressureSetName(RCPS.getPSet()) << " ";
             dbgs() << "\n");
}

void ScheduleDAGMILive::
updateScheduledPressure(const SUnit *SU,
                        const std::vector<unsigned> &NewMaxPressure) {
  const PressureDiff &PDiff = getPressureDiff(SU);
  unsigned CritIdx = 0, CritEnd = RegionCriticalPSets.size();
  for (const PressureChange &PC : PDiff) {
    if (!PC.isValid())
      break;
    unsigned ID = PC.getPSet();
    while (CritIdx != CritEnd && RegionCriticalPSets[CritIdx].getPSet() < ID)
      ++CritIdx;
    if (CritIdx != CritEnd && RegionCriticalPSets[CritIdx].getPSet() == ID) {
      if ((int)NewMaxPressure[ID] > RegionCriticalPSets[CritIdx].getUnitInc()
          && NewMaxPressure[ID] <= (unsigned)std::numeric_limits<int16_t>::max())
        RegionCriticalPSets[CritIdx].setUnitInc(NewMaxPressure[ID]);
    }
    unsigned Limit = RegClassInfo->getRegPressureSetLimit(ID);
    if (NewMaxPressure[ID] >= Limit - 2) {
      LLVM_DEBUG(dbgs() << "  " << TRI->getRegPressureSetName(ID) << ": "
                        << NewMaxPressure[ID]
                        << ((NewMaxPressure[ID] > Limit) ? " > " : " <= ")
                        << Limit << "(+ " << BotRPTracker.getLiveThru()[ID]
                        << " livethru)\n");
    }
  }
}

/// Update the PressureDiff array for liveness after scheduling this
/// instruction.
void ScheduleDAGMILive::updatePressureDiffs(
    ArrayRef<RegisterMaskPair> LiveUses) {
  for (const RegisterMaskPair &P : LiveUses) {
    Register Reg = P.RegUnit;
    /// FIXME: Currently assuming single-use physregs.
    if (!Register::isVirtualRegister(Reg))
      continue;

    if (ShouldTrackLaneMasks) {
      // If the register has just become live then other uses won't change
      // this fact anymore => decrement pressure.
      // If the register has just become dead then other uses make it come
      // back to life => increment pressure.
      bool Decrement = P.LaneMask.any();

      for (const VReg2SUnit &V2SU
           : make_range(VRegUses.find(Reg), VRegUses.end())) {
        SUnit &SU = *V2SU.SU;
        if (SU.isScheduled || &SU == &ExitSU)
          continue;

        PressureDiff &PDiff = getPressureDiff(&SU);
        PDiff.addPressureChange(Reg, Decrement, &MRI);
        LLVM_DEBUG(dbgs() << "  UpdateRegP: SU(" << SU.NodeNum << ") "
                          << printReg(Reg, TRI) << ':'
                          << PrintLaneMask(P.LaneMask) << ' ' << *SU.getInstr();
                   dbgs() << "              to "; PDiff.dump(*TRI););
      }
    } else {
      assert(P.LaneMask.any());
      LLVM_DEBUG(dbgs() << "  LiveReg: " << printVRegOrUnit(Reg, TRI) << "\n");
      // This may be called before CurrentBottom has been initialized. However,
      // BotRPTracker must have a valid position. We want the value live into the
      // instruction or live out of the block, so ask for the previous
      // instruction's live-out.
      const LiveInterval &LI = LIS->getInterval(Reg);
      VNInfo *VNI;
      MachineBasicBlock::const_iterator I =
        nextIfDebug(BotRPTracker.getPos(), BB->end());
      if (I == BB->end())
        VNI = LI.getVNInfoBefore(LIS->getMBBEndIdx(BB));
      else {
        LiveQueryResult LRQ = LI.Query(LIS->getInstructionIndex(*I));
        VNI = LRQ.valueIn();
      }
      // RegisterPressureTracker guarantees that readsReg is true for LiveUses.
      assert(VNI && "No live value at use.");
      for (const VReg2SUnit &V2SU
           : make_range(VRegUses.find(Reg), VRegUses.end())) {
        SUnit *SU = V2SU.SU;
        // If this use comes before the reaching def, it cannot be a last use,
        // so decrease its pressure change.
        if (!SU->isScheduled && SU != &ExitSU) {
          LiveQueryResult LRQ =
              LI.Query(LIS->getInstructionIndex(*SU->getInstr()));
          if (LRQ.valueIn() == VNI) {
            PressureDiff &PDiff = getPressureDiff(SU);
            PDiff.addPressureChange(Reg, true, &MRI);
            LLVM_DEBUG(dbgs() << "  UpdateRegP: SU(" << SU->NodeNum << ") "
                              << *SU->getInstr();
                       dbgs() << "              to "; PDiff.dump(*TRI););
          }
        }
      }
    }
  }
}

void ScheduleDAGMILive::dump() const {
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
  if (EntrySU.getInstr() != nullptr)
    dumpNodeAll(EntrySU);
  for (const SUnit &SU : SUnits) {
    dumpNodeAll(SU);
    if (ShouldTrackPressure) {
      dbgs() << "  Pressure Diff      : ";
      getPressureDiff(&SU).dump(*TRI);
    }
    dbgs() << "  Single Issue       : ";
    if (SchedModel.mustBeginGroup(SU.getInstr()) &&
        SchedModel.mustEndGroup(SU.getInstr()))
      dbgs() << "true;";
    else
      dbgs() << "false;";
    dbgs() << '\n';
  }
  if (ExitSU.getInstr() != nullptr)
    dumpNodeAll(ExitSU);
#endif
}

/// schedule - Called back from MachineScheduler::runOnMachineFunction
/// after setting up the current scheduling region. [RegionBegin, RegionEnd)
/// only includes instructions that have DAG nodes, not scheduling boundaries.
///
/// This is a skeletal driver, with all the functionality pushed into helpers,
/// so that it can be easily extended by experimental schedulers. Generally,
/// implementing MachineSchedStrategy should be sufficient to implement a new
/// scheduling algorithm. However, if a scheduler further subclasses
/// ScheduleDAGMILive then it will want to override this virtual method in order
/// to update any specialized state.
void ScheduleDAGMILive::schedule() {
  LLVM_DEBUG(dbgs() << "ScheduleDAGMILive::schedule starting\n");
  LLVM_DEBUG(SchedImpl->dumpPolicy());
  buildDAGWithRegPressure();

  postprocessDAG();

  SmallVector<SUnit*, 8> TopRoots, BotRoots;
  findRootsAndBiasEdges(TopRoots, BotRoots);

  // Initialize the strategy before modifying the DAG.
  // This may initialize a DFSResult to be used for queue priority.
  SchedImpl->initialize(this);

  LLVM_DEBUG(dump());
  if (PrintDAGs) dump();
  if (ViewMISchedDAGs) viewGraph();

  // Initialize ready queues now that the DAG and priority data are finalized.
  initQueues(TopRoots, BotRoots);

  bool IsTopNode = false;
  while (true) {
    LLVM_DEBUG(dbgs() << "** ScheduleDAGMILive::schedule picking next node\n");
    SUnit *SU = SchedImpl->pickNode(IsTopNode);
    if (!SU) break;

    assert(!SU->isScheduled && "Node already scheduled");
    if (!checkSchedLimit())
      break;

    scheduleMI(SU, IsTopNode);

    if (DFSResult) {
      unsigned SubtreeID = DFSResult->getSubtreeID(SU);
      if (!ScheduledTrees.test(SubtreeID)) {
        ScheduledTrees.set(SubtreeID);
        DFSResult->scheduleTree(SubtreeID);
        SchedImpl->scheduleTree(SubtreeID);
      }
    }

    // Notify the scheduling strategy after updating the DAG.
    SchedImpl->schedNode(SU, IsTopNode);

    updateQueues(SU, IsTopNode);
  }
  assert(CurrentTop == CurrentBottom && "Nonempty unscheduled zone.");

  placeDebugValues();

  LLVM_DEBUG({
    dbgs() << "*** Final schedule for "
           << printMBBReference(*begin()->getParent()) << " ***\n";
    dumpSchedule();
    dbgs() << '\n';
  });
}

/// Build the DAG and setup three register pressure trackers.
void ScheduleDAGMILive::buildDAGWithRegPressure() {
  if (!ShouldTrackPressure) {
    RPTracker.reset();
    RegionCriticalPSets.clear();
    buildSchedGraph(AA);
    return;
  }

  // Initialize the register pressure tracker used by buildSchedGraph.
  RPTracker.init(&MF, RegClassInfo, LIS, BB, LiveRegionEnd,
                 ShouldTrackLaneMasks, /*TrackUntiedDefs=*/true);

  // Account for liveness generate by the region boundary.
  if (LiveRegionEnd != RegionEnd)
    RPTracker.recede();

  // Build the DAG, and compute current register pressure.
  buildSchedGraph(AA, &RPTracker, &SUPressureDiffs, LIS, ShouldTrackLaneMasks);

  // Initialize top/bottom trackers after computing region pressure.
  initRegPressure();
}

void ScheduleDAGMILive::computeDFSResult() {
  if (!DFSResult)
    DFSResult = new SchedDFSResult(/*BottomU*/true, MinSubtreeSize);
  DFSResult->clear();
  ScheduledTrees.clear();
  DFSResult->resize(SUnits.size());
  DFSResult->compute(SUnits);
  ScheduledTrees.resize(DFSResult->getNumSubtrees());
}

/// Compute the max cyclic critical path through the DAG. The scheduling DAG
/// only provides the critical path for single block loops. To handle loops that
/// span blocks, we could use the vreg path latencies provided by
/// MachineTraceMetrics instead. However, MachineTraceMetrics is not currently
/// available for use in the scheduler.
///
/// The cyclic path estimation identifies a def-use pair that crosses the back
/// edge and considers the depth and height of the nodes. For example, consider
/// the following instruction sequence where each instruction has unit latency
/// and defines an eponymous virtual register:
///
/// a->b(a,c)->c(b)->d(c)->exit
///
/// The cyclic critical path is a two cycles: b->c->b
/// The acyclic critical path is four cycles: a->b->c->d->exit
/// LiveOutHeight = height(c) = len(c->d->exit) = 2
/// LiveOutDepth = depth(c) + 1 = len(a->b->c) + 1 = 3
/// LiveInHeight = height(b) + 1 = len(b->c->d->exit) + 1 = 4
/// LiveInDepth = depth(b) = len(a->b) = 1
///
/// LiveOutDepth - LiveInDepth = 3 - 1 = 2
/// LiveInHeight - LiveOutHeight = 4 - 2 = 2
/// CyclicCriticalPath = min(2, 2) = 2
///
/// This could be relevant to PostRA scheduling, but is currently implemented
/// assuming LiveIntervals.
unsigned ScheduleDAGMILive::computeCyclicCriticalPath() {
  // This only applies to single block loop.
  if (!BB->isSuccessor(BB))
    return 0;

  unsigned MaxCyclicLatency = 0;
  // Visit each live out vreg def to find def/use pairs that cross iterations.
  for (const RegisterMaskPair &P : RPTracker.getPressure().LiveOutRegs) {
    Register Reg = P.RegUnit;
    if (!Register::isVirtualRegister(Reg))
      continue;
    const LiveInterval &LI = LIS->getInterval(Reg);
    const VNInfo *DefVNI = LI.getVNInfoBefore(LIS->getMBBEndIdx(BB));
    if (!DefVNI)
      continue;

    MachineInstr *DefMI = LIS->getInstructionFromIndex(DefVNI->def);
    const SUnit *DefSU = getSUnit(DefMI);
    if (!DefSU)
      continue;

    unsigned LiveOutHeight = DefSU->getHeight();
    unsigned LiveOutDepth = DefSU->getDepth() + DefSU->Latency;
    // Visit all local users of the vreg def.
    for (const VReg2SUnit &V2SU
         : make_range(VRegUses.find(Reg), VRegUses.end())) {
      SUnit *SU = V2SU.SU;
      if (SU == &ExitSU)
        continue;

      // Only consider uses of the phi.
      LiveQueryResult LRQ = LI.Query(LIS->getInstructionIndex(*SU->getInstr()));
      if (!LRQ.valueIn()->isPHIDef())
        continue;

      // Assume that a path spanning two iterations is a cycle, which could
      // overestimate in strange cases. This allows cyclic latency to be
      // estimated as the minimum slack of the vreg's depth or height.
      unsigned CyclicLatency = 0;
      if (LiveOutDepth > SU->getDepth())
        CyclicLatency = LiveOutDepth - SU->getDepth();

      unsigned LiveInHeight = SU->getHeight() + DefSU->Latency;
      if (LiveInHeight > LiveOutHeight) {
        if (LiveInHeight - LiveOutHeight < CyclicLatency)
          CyclicLatency = LiveInHeight - LiveOutHeight;
      } else
        CyclicLatency = 0;

      LLVM_DEBUG(dbgs() << "Cyclic Path: SU(" << DefSU->NodeNum << ") -> SU("
                        << SU->NodeNum << ") = " << CyclicLatency << "c\n");
      if (CyclicLatency > MaxCyclicLatency)
        MaxCyclicLatency = CyclicLatency;
    }
  }
  LLVM_DEBUG(dbgs() << "Cyclic Critical Path: " << MaxCyclicLatency << "c\n");
  return MaxCyclicLatency;
}

/// Release ExitSU predecessors and setup scheduler queues. Re-position
/// the Top RP tracker in case the region beginning has changed.
void ScheduleDAGMILive::initQueues(ArrayRef<SUnit*> TopRoots,
                                   ArrayRef<SUnit*> BotRoots) {
  ScheduleDAGMI::initQueues(TopRoots, BotRoots);
  if (ShouldTrackPressure) {
    assert(TopRPTracker.getPos() == RegionBegin && "bad initial Top tracker");
    TopRPTracker.setPos(CurrentTop);
  }
}

/// Move an instruction and update register pressure.
void ScheduleDAGMILive::scheduleMI(SUnit *SU, bool IsTopNode) {
  // Move the instruction to its new location in the instruction stream.
  MachineInstr *MI = SU->getInstr();

  if (IsTopNode) {
    assert(SU->isTopReady() && "node still has unscheduled dependencies");
    if (&*CurrentTop == MI)
      CurrentTop = nextIfDebug(++CurrentTop, CurrentBottom);
    else {
      moveInstruction(MI, CurrentTop);
      TopRPTracker.setPos(MI);
    }

    if (ShouldTrackPressure) {
      // Update top scheduled pressure.
      RegisterOperands RegOpers;
      RegOpers.collect(*MI, *TRI, MRI, ShouldTrackLaneMasks, false);
      if (ShouldTrackLaneMasks) {
        // Adjust liveness and add missing dead+read-undef flags.
        SlotIndex SlotIdx = LIS->getInstructionIndex(*MI).getRegSlot();
        RegOpers.adjustLaneLiveness(*LIS, MRI, SlotIdx, MI);
      } else {
        // Adjust for missing dead-def flags.
        RegOpers.detectDeadDefs(*MI, *LIS);
      }

      TopRPTracker.advance(RegOpers);
      assert(TopRPTracker.getPos() == CurrentTop && "out of sync");
      LLVM_DEBUG(dbgs() << "Top Pressure:\n"; dumpRegSetPressure(
                     TopRPTracker.getRegSetPressureAtPos(), TRI););

      updateScheduledPressure(SU, TopRPTracker.getPressure().MaxSetPressure);
    }
  } else {
    assert(SU->isBottomReady() && "node still has unscheduled dependencies");
    MachineBasicBlock::iterator priorII =
      priorNonDebug(CurrentBottom, CurrentTop);
    if (&*priorII == MI)
      CurrentBottom = priorII;
    else {
      if (&*CurrentTop == MI) {
        CurrentTop = nextIfDebug(++CurrentTop, priorII);
        TopRPTracker.setPos(CurrentTop);
      }
      moveInstruction(MI, CurrentBottom);
      CurrentBottom = MI;
      BotRPTracker.setPos(CurrentBottom);
    }
    if (ShouldTrackPressure) {
      RegisterOperands RegOpers;
      RegOpers.collect(*MI, *TRI, MRI, ShouldTrackLaneMasks, false);
      if (ShouldTrackLaneMasks) {
        // Adjust liveness and add missing dead+read-undef flags.
        SlotIndex SlotIdx = LIS->getInstructionIndex(*MI).getRegSlot();
        RegOpers.adjustLaneLiveness(*LIS, MRI, SlotIdx, MI);
      } else {
        // Adjust for missing dead-def flags.
        RegOpers.detectDeadDefs(*MI, *LIS);
      }

      if (BotRPTracker.getPos() != CurrentBottom)
        BotRPTracker.recedeSkipDebugValues();
      SmallVector<RegisterMaskPair, 8> LiveUses;
      BotRPTracker.recede(RegOpers, &LiveUses);
      assert(BotRPTracker.getPos() == CurrentBottom && "out of sync");
      LLVM_DEBUG(dbgs() << "Bottom Pressure:\n"; dumpRegSetPressure(
                     BotRPTracker.getRegSetPressureAtPos(), TRI););

      updateScheduledPressure(SU, BotRPTracker.getPressure().MaxSetPressure);
      updatePressureDiffs(LiveUses);
    }
  }
}

//===----------------------------------------------------------------------===//
// BaseMemOpClusterMutation - DAG post-processing to cluster loads or stores.
//===----------------------------------------------------------------------===//

namespace {

/// Post-process the DAG to create cluster edges between neighboring
/// loads or between neighboring stores.
class BaseMemOpClusterMutation : public ScheduleDAGMutation {
  struct MemOpInfo {
    SUnit *SU;
    SmallVector<const MachineOperand *, 4> BaseOps;
    int64_t Offset;
    unsigned Width;

    MemOpInfo(SUnit *SU, ArrayRef<const MachineOperand *> BaseOps,
              int64_t Offset, unsigned Width)
        : SU(SU), BaseOps(BaseOps.begin(), BaseOps.end()), Offset(Offset),
          Width(Width) {}

    static bool Compare(const MachineOperand *const &A,
                        const MachineOperand *const &B) {
      if (A->getType() != B->getType())
        return A->getType() < B->getType();
      if (A->isReg())
        return A->getReg() < B->getReg();
      if (A->isFI()) {
        const MachineFunction &MF = *A->getParent()->getParent()->getParent();
        const TargetFrameLowering &TFI = *MF.getSubtarget().getFrameLowering();
        bool StackGrowsDown = TFI.getStackGrowthDirection() ==
                              TargetFrameLowering::StackGrowsDown;
        return StackGrowsDown ? A->getIndex() > B->getIndex()
                              : A->getIndex() < B->getIndex();
      }

      llvm_unreachable("MemOpClusterMutation only supports register or frame "
                       "index bases.");
    }

    bool operator<(const MemOpInfo &RHS) const {
      // FIXME: Don't compare everything twice. Maybe use C++20 three way
      // comparison instead when it's available.
      if (std::lexicographical_compare(BaseOps.begin(), BaseOps.end(),
                                       RHS.BaseOps.begin(), RHS.BaseOps.end(),
                                       Compare))
        return true;
      if (std::lexicographical_compare(RHS.BaseOps.begin(), RHS.BaseOps.end(),
                                       BaseOps.begin(), BaseOps.end(), Compare))
        return false;
      if (Offset != RHS.Offset)
        return Offset < RHS.Offset;
      return SU->NodeNum < RHS.SU->NodeNum;
    }
  };

  const TargetInstrInfo *TII;
  const TargetRegisterInfo *TRI;
  bool IsLoad;

public:
  BaseMemOpClusterMutation(const TargetInstrInfo *tii,
                           const TargetRegisterInfo *tri, bool IsLoad)
      : TII(tii), TRI(tri), IsLoad(IsLoad) {}

  void apply(ScheduleDAGInstrs *DAGInstrs) override;

protected:
  void clusterNeighboringMemOps(ArrayRef<MemOpInfo> MemOps, bool FastCluster,
                                ScheduleDAGInstrs *DAG);
  void collectMemOpRecords(std::vector<SUnit> &SUnits,
                           SmallVectorImpl<MemOpInfo> &MemOpRecords);
  bool groupMemOps(ArrayRef<MemOpInfo> MemOps, ScheduleDAGInstrs *DAG,
                   DenseMap<unsigned, SmallVector<MemOpInfo, 32>> &Groups);
};

class StoreClusterMutation : public BaseMemOpClusterMutation {
public:
  StoreClusterMutation(const TargetInstrInfo *tii,
                       const TargetRegisterInfo *tri)
      : BaseMemOpClusterMutation(tii, tri, false) {}
};

class LoadClusterMutation : public BaseMemOpClusterMutation {
public:
  LoadClusterMutation(const TargetInstrInfo *tii, const TargetRegisterInfo *tri)
      : BaseMemOpClusterMutation(tii, tri, true) {}
};

} // end anonymous namespace

namespace llvm {

std::unique_ptr<ScheduleDAGMutation>
createLoadClusterDAGMutation(const TargetInstrInfo *TII,
                             const TargetRegisterInfo *TRI) {
  return EnableMemOpCluster ? std::make_unique<LoadClusterMutation>(TII, TRI)
                            : nullptr;
}

std::unique_ptr<ScheduleDAGMutation>
createStoreClusterDAGMutation(const TargetInstrInfo *TII,
                              const TargetRegisterInfo *TRI) {
  return EnableMemOpCluster ? std::make_unique<StoreClusterMutation>(TII, TRI)
                            : nullptr;
}

} // end namespace llvm

// Sorting all the loads/stores first, then for each load/store, checking the
// following load/store one by one, until reach the first non-dependent one and
// call target hook to see if they can cluster.
// If FastCluster is enabled, we assume that, all the loads/stores have been
// preprocessed and now, they didn't have dependencies on each other.
void BaseMemOpClusterMutation::clusterNeighboringMemOps(
    ArrayRef<MemOpInfo> MemOpRecords, bool FastCluster,
    ScheduleDAGInstrs *DAG) {
  // Keep track of the current cluster length and bytes for each SUnit.
  DenseMap<unsigned, std::pair<unsigned, unsigned>> SUnit2ClusterInfo;

  // At this point, `MemOpRecords` array must hold atleast two mem ops. Try to
  // cluster mem ops collected within `MemOpRecords` array.
  for (unsigned Idx = 0, End = MemOpRecords.size(); Idx < (End - 1); ++Idx) {
    // Decision to cluster mem ops is taken based on target dependent logic
    auto MemOpa = MemOpRecords[Idx];

    // Seek for the next load/store to do the cluster.
    unsigned NextIdx = Idx + 1;
    for (; NextIdx < End; ++NextIdx)
      // Skip if MemOpb has been clustered already or has dependency with
      // MemOpa.
      if (!SUnit2ClusterInfo.count(MemOpRecords[NextIdx].SU->NodeNum) &&
          (FastCluster ||
           (!DAG->IsReachable(MemOpRecords[NextIdx].SU, MemOpa.SU) &&
            !DAG->IsReachable(MemOpa.SU, MemOpRecords[NextIdx].SU))))
        break;
    if (NextIdx == End)
      continue;

    auto MemOpb = MemOpRecords[NextIdx];
    unsigned ClusterLength = 2;
    unsigned CurrentClusterBytes = MemOpa.Width + MemOpb.Width;
    if (SUnit2ClusterInfo.count(MemOpa.SU->NodeNum)) {
      ClusterLength = SUnit2ClusterInfo[MemOpa.SU->NodeNum].first + 1;
      CurrentClusterBytes =
          SUnit2ClusterInfo[MemOpa.SU->NodeNum].second + MemOpb.Width;
    }

    if (!TII->shouldClusterMemOps(MemOpa.BaseOps, MemOpb.BaseOps, ClusterLength,
                                  CurrentClusterBytes))
      continue;

    SUnit *SUa = MemOpa.SU;
    SUnit *SUb = MemOpb.SU;
    if (SUa->NodeNum > SUb->NodeNum)
      std::swap(SUa, SUb);

    // FIXME: Is this check really required?
    if (!DAG->addEdge(SUb, SDep(SUa, SDep::Cluster)))
      continue;

    LLVM_DEBUG(dbgs() << "Cluster ld/st SU(" << SUa->NodeNum << ") - SU("
                      << SUb->NodeNum << ")\n");
    ++NumClustered;

    if (IsLoad) {
      // Copy successor edges from SUa to SUb. Interleaving computation
      // dependent on SUa can prevent load combining due to register reuse.
      // Predecessor edges do not need to be copied from SUb to SUa since
      // nearby loads should have effectively the same inputs.
      for (const SDep &Succ : SUa->Succs) {
        if (Succ.getSUnit() == SUb)
          continue;
        LLVM_DEBUG(dbgs() << "  Copy Succ SU(" << Succ.getSUnit()->NodeNum
                          << ")\n");
        DAG->addEdge(Succ.getSUnit(), SDep(SUb, SDep::Artificial));
      }
    } else {
      // Copy predecessor edges from SUb to SUa to avoid the SUnits that
      // SUb dependent on scheduled in-between SUb and SUa. Successor edges
      // do not need to be copied from SUa to SUb since no one will depend
      // on stores.
      // Notice that, we don't need to care about the memory dependency as
      // we won't try to cluster them if they have any memory dependency.
      for (const SDep &Pred : SUb->Preds) {
        if (Pred.getSUnit() == SUa)
          continue;
        LLVM_DEBUG(dbgs() << "  Copy Pred SU(" << Pred.getSUnit()->NodeNum
                          << ")\n");
        DAG->addEdge(SUa, SDep(Pred.getSUnit(), SDep::Artificial));
      }
    }

    SUnit2ClusterInfo[MemOpb.SU->NodeNum] = {ClusterLength,
                                             CurrentClusterBytes};

    LLVM_DEBUG(dbgs() << "  Curr cluster length: " << ClusterLength
                      << ", Curr cluster bytes: " << CurrentClusterBytes
                      << "\n");
  }
}

void BaseMemOpClusterMutation::collectMemOpRecords(
    std::vector<SUnit> &SUnits, SmallVectorImpl<MemOpInfo> &MemOpRecords) {
  for (auto &SU : SUnits) {
    if ((IsLoad && !SU.getInstr()->mayLoad()) ||
        (!IsLoad && !SU.getInstr()->mayStore()))
      continue;

    const MachineInstr &MI = *SU.getInstr();
    SmallVector<const MachineOperand *, 4> BaseOps;
    int64_t Offset;
    bool OffsetIsScalable;
    unsigned Width;
    if (TII->getMemOperandsWithOffsetWidth(MI, BaseOps, Offset,
                                           OffsetIsScalable, Width, TRI)) {
      MemOpRecords.push_back(MemOpInfo(&SU, BaseOps, Offset, Width));

      LLVM_DEBUG(dbgs() << "Num BaseOps: " << BaseOps.size() << ", Offset: "
                        << Offset << ", OffsetIsScalable: " << OffsetIsScalable
                        << ", Width: " << Width << "\n");
    }
#ifndef NDEBUG
    for (auto *Op : BaseOps)
      assert(Op);
#endif
  }
}

bool BaseMemOpClusterMutation::groupMemOps(
    ArrayRef<MemOpInfo> MemOps, ScheduleDAGInstrs *DAG,
    DenseMap<unsigned, SmallVector<MemOpInfo, 32>> &Groups) {
  bool FastCluster =
      ForceFastCluster ||
      MemOps.size() * DAG->SUnits.size() / 1000 > FastClusterThreshold;

  for (const auto &MemOp : MemOps) {
    unsigned ChainPredID = DAG->SUnits.size();
    if (FastCluster) {
      for (const SDep &Pred : MemOp.SU->Preds) {
        // We only want to cluster the mem ops that have the same ctrl(non-data)
        // pred so that they didn't have ctrl dependency for each other. But for
        // store instrs, we can still cluster them if the pred is load instr.
        if ((Pred.isCtrl() &&
             (IsLoad ||
              (Pred.getSUnit() && Pred.getSUnit()->getInstr()->mayStore()))) &&
            !Pred.isArtificial()) {
          ChainPredID = Pred.getSUnit()->NodeNum;
          break;
        }
      }
    } else
      ChainPredID = 0;

    Groups[ChainPredID].push_back(MemOp);
  }
  return FastCluster;
}

/// Callback from DAG postProcessing to create cluster edges for loads/stores.
void BaseMemOpClusterMutation::apply(ScheduleDAGInstrs *DAG) {
  // Collect all the clusterable loads/stores
  SmallVector<MemOpInfo, 32> MemOpRecords;
  collectMemOpRecords(DAG->SUnits, MemOpRecords);

  if (MemOpRecords.size() < 2)
    return;

  // Put the loads/stores without dependency into the same group with some
  // heuristic if the DAG is too complex to avoid compiling time blow up.
  // Notice that, some fusion pair could be lost with this.
  DenseMap<unsigned, SmallVector<MemOpInfo, 32>> Groups;
  bool FastCluster = groupMemOps(MemOpRecords, DAG, Groups);

  for (auto &Group : Groups) {
    // Sorting the loads/stores, so that, we can stop the cluster as early as
    // possible.
    llvm::sort(Group.second);

    // Trying to cluster all the neighboring loads/stores.
    clusterNeighboringMemOps(Group.second, FastCluster, DAG);
  }
}

//===----------------------------------------------------------------------===//
// CopyConstrain - DAG post-processing to encourage copy elimination.
//===----------------------------------------------------------------------===//

namespace {

/// Post-process the DAG to create weak edges from all uses of a copy to
/// the one use that defines the copy's source vreg, most likely an induction
/// variable increment.
class CopyConstrain : public ScheduleDAGMutation {
  // Transient state.
  SlotIndex RegionBeginIdx;

  // RegionEndIdx is the slot index of the last non-debug instruction in the
  // scheduling region. So we may have RegionBeginIdx == RegionEndIdx.
  SlotIndex RegionEndIdx;

public:
  CopyConstrain(const TargetInstrInfo *, const TargetRegisterInfo *) {}

  void apply(ScheduleDAGInstrs *DAGInstrs) override;

protected:
  void constrainLocalCopy(SUnit *CopySU, ScheduleDAGMILive *DAG);
};

} // end anonymous namespace

namespace llvm {

std::unique_ptr<ScheduleDAGMutation>
createCopyConstrainDAGMutation(const TargetInstrInfo *TII,
                               const TargetRegisterInfo *TRI) {
  return std::make_unique<CopyConstrain>(TII, TRI);
}

} // end namespace llvm

/// constrainLocalCopy handles two possibilities:
/// 1) Local src:
/// I0:     = dst
/// I1: src = ...
/// I2:     = dst
/// I3: dst = src (copy)
/// (create pred->succ edges I0->I1, I2->I1)
///
/// 2) Local copy:
/// I0: dst = src (copy)
/// I1:     = dst
/// I2: src = ...
/// I3:     = dst
/// (create pred->succ edges I1->I2, I3->I2)
///
/// Although the MachineScheduler is currently constrained to single blocks,
/// this algorithm should handle extended blocks. An EBB is a set of
/// contiguously numbered blocks such that the previous block in the EBB is
/// always the single predecessor.
void CopyConstrain::constrainLocalCopy(SUnit *CopySU, ScheduleDAGMILive *DAG) {
  LiveIntervals *LIS = DAG->getLIS();
  MachineInstr *Copy = CopySU->getInstr();

  // Check for pure vreg copies.
  const MachineOperand &SrcOp = Copy->getOperand(1);
  Register SrcReg = SrcOp.getReg();
  if (!Register::isVirtualRegister(SrcReg) || !SrcOp.readsReg())
    return;

  const MachineOperand &DstOp = Copy->getOperand(0);
  Register DstReg = DstOp.getReg();
  if (!Register::isVirtualRegister(DstReg) || DstOp.isDead())
    return;

  // Check if either the dest or source is local. If it's live across a back
  // edge, it's not local. Note that if both vregs are live across the back
  // edge, we cannot successfully contrain the copy without cyclic scheduling.
  // If both the copy's source and dest are local live intervals, then we
  // should treat the dest as the global for the purpose of adding
  // constraints. This adds edges from source's other uses to the copy.
  unsigned LocalReg = SrcReg;
  unsigned GlobalReg = DstReg;
  LiveInterval *LocalLI = &LIS->getInterval(LocalReg);
  if (!LocalLI->isLocal(RegionBeginIdx, RegionEndIdx)) {
    LocalReg = DstReg;
    GlobalReg = SrcReg;
    LocalLI = &LIS->getInterval(LocalReg);
    if (!LocalLI->isLocal(RegionBeginIdx, RegionEndIdx))
      return;
  }
  LiveInterval *GlobalLI = &LIS->getInterval(GlobalReg);

  // Find the global segment after the start of the local LI.
  LiveInterval::iterator GlobalSegment = GlobalLI->find(LocalLI->beginIndex());
  // If GlobalLI does not overlap LocalLI->start, then a copy directly feeds a
  // local live range. We could create edges from other global uses to the local
  // start, but the coalescer should have already eliminated these cases, so
  // don't bother dealing with it.
  if (GlobalSegment == GlobalLI->end())
    return;

  // If GlobalSegment is killed at the LocalLI->start, the call to find()
  // returned the next global segment. But if GlobalSegment overlaps with
  // LocalLI->start, then advance to the next segment. If a hole in GlobalLI
  // exists in LocalLI's vicinity, GlobalSegment will be the end of the hole.
  if (GlobalSegment->contains(LocalLI->beginIndex()))
    ++GlobalSegment;

  if (GlobalSegment == GlobalLI->end())
    return;

  // Check if GlobalLI contains a hole in the vicinity of LocalLI.
  if (GlobalSegment != GlobalLI->begin()) {
    // Two address defs have no hole.
    if (SlotIndex::isSameInstr(std::prev(GlobalSegment)->end,
                               GlobalSegment->start)) {
      return;
    }
    // If the prior global segment may be defined by the same two-address
    // instruction that also defines LocalLI, then can't make a hole here.
    if (SlotIndex::isSameInstr(std::prev(GlobalSegment)->start,
                               LocalLI->beginIndex())) {
      return;
    }
    // If GlobalLI has a prior segment, it must be live into the EBB. Otherwise
    // it would be a disconnected component in the live range.
    assert(std::prev(GlobalSegment)->start < LocalLI->beginIndex() &&
           "Disconnected LRG within the scheduling region.");
  }
  MachineInstr *GlobalDef = LIS->getInstructionFromIndex(GlobalSegment->start);
  if (!GlobalDef)
    return;

  SUnit *GlobalSU = DAG->getSUnit(GlobalDef);
  if (!GlobalSU)
    return;

  // GlobalDef is the bottom of the GlobalLI hole. Open the hole by
  // constraining the uses of the last local def to precede GlobalDef.
  SmallVector<SUnit*,8> LocalUses;
  const VNInfo *LastLocalVN = LocalLI->getVNInfoBefore(LocalLI->endIndex());
  MachineInstr *LastLocalDef = LIS->getInstructionFromIndex(LastLocalVN->def);
  SUnit *LastLocalSU = DAG->getSUnit(LastLocalDef);
  for (const SDep &Succ : LastLocalSU->Succs) {
    if (Succ.getKind() != SDep::Data || Succ.getReg() != LocalReg)
      continue;
    if (Succ.getSUnit() == GlobalSU)
      continue;
    if (!DAG->canAddEdge(GlobalSU, Succ.getSUnit()))
      return;
    LocalUses.push_back(Succ.getSUnit());
  }
  // Open the top of the GlobalLI hole by constraining any earlier global uses
  // to precede the start of LocalLI.
  SmallVector<SUnit*,8> GlobalUses;
  MachineInstr *FirstLocalDef =
    LIS->getInstructionFromIndex(LocalLI->beginIndex());
  SUnit *FirstLocalSU = DAG->getSUnit(FirstLocalDef);
  for (const SDep &Pred : GlobalSU->Preds) {
    if (Pred.getKind() != SDep::Anti || Pred.getReg() != GlobalReg)
      continue;
    if (Pred.getSUnit() == FirstLocalSU)
      continue;
    if (!DAG->canAddEdge(FirstLocalSU, Pred.getSUnit()))
      return;
    GlobalUses.push_back(Pred.getSUnit());
  }
  LLVM_DEBUG(dbgs() << "Constraining copy SU(" << CopySU->NodeNum << ")\n");
  // Add the weak edges.
  for (SUnit *LU : LocalUses) {
    LLVM_DEBUG(dbgs() << "  Local use SU(" << LU->NodeNum << ") -> SU("
                      << GlobalSU->NodeNum << ")\n");
    DAG->addEdge(GlobalSU, SDep(LU, SDep::Weak));
  }
  for (SUnit *GU : GlobalUses) {
    LLVM_DEBUG(dbgs() << "  Global use SU(" << GU->NodeNum << ") -> SU("
                      << FirstLocalSU->NodeNum << ")\n");
    DAG->addEdge(FirstLocalSU, SDep(GU, SDep::Weak));
  }
}

/// Callback from DAG postProcessing to create weak edges to encourage
/// copy elimination.
void CopyConstrain::apply(ScheduleDAGInstrs *DAGInstrs) {
  ScheduleDAGMI *DAG = static_cast<ScheduleDAGMI*>(DAGInstrs);
  assert(DAG->hasVRegLiveness() && "Expect VRegs with LiveIntervals");

  MachineBasicBlock::iterator FirstPos = nextIfDebug(DAG->begin(), DAG->end());
  if (FirstPos == DAG->end())
    return;
  RegionBeginIdx = DAG->getLIS()->getInstructionIndex(*FirstPos);
  RegionEndIdx = DAG->getLIS()->getInstructionIndex(
      *priorNonDebug(DAG->end(), DAG->begin()));

  for (SUnit &SU : DAG->SUnits) {
    if (!SU.getInstr()->isCopy())
      continue;

    constrainLocalCopy(&SU, static_cast<ScheduleDAGMILive*>(DAG));
  }
}

//===----------------------------------------------------------------------===//
// MachineSchedStrategy helpers used by GenericScheduler, GenericPostScheduler
// and possibly other custom schedulers.
//===----------------------------------------------------------------------===//

static const unsigned InvalidCycle = ~0U;

SchedBoundary::~SchedBoundary() { delete HazardRec; }

/// Given a Count of resource usage and a Latency value, return true if a
/// SchedBoundary becomes resource limited.
/// If we are checking after scheduling a node, we should return true when
/// we just reach the resource limit.
static bool checkResourceLimit(unsigned LFactor, unsigned Count,
                               unsigned Latency, bool AfterSchedNode) {
  int ResCntFactor = (int)(Count - (Latency * LFactor));
  if (AfterSchedNode)
    return ResCntFactor >= (int)LFactor;
  else
    return ResCntFactor > (int)LFactor;
}

void SchedBoundary::reset() {
  // A new HazardRec is created for each DAG and owned by SchedBoundary.
  // Destroying and reconstructing it is very expensive though. So keep
  // invalid, placeholder HazardRecs.
  if (HazardRec && HazardRec->isEnabled()) {
    delete HazardRec;
    HazardRec = nullptr;
  }
  Available.clear();
  Pending.clear();
  CheckPending = false;
  CurrCycle = 0;
  CurrMOps = 0;
  MinReadyCycle = std::numeric_limits<unsigned>::max();
  ExpectedLatency = 0;
  DependentLatency = 0;
  RetiredMOps = 0;
  MaxExecutedResCount = 0;
  ZoneCritResIdx = 0;
  IsResourceLimited = false;
  ReservedCycles.clear();
  ReservedCyclesIndex.clear();
  ResourceGroupSubUnitMasks.clear();
#ifndef NDEBUG
  // Track the maximum number of stall cycles that could arise either from the
  // latency of a DAG edge or the number of cycles that a processor resource is
  // reserved (SchedBoundary::ReservedCycles).
  MaxObservedStall = 0;
#endif
  // Reserve a zero-count for invalid CritResIdx.
  ExecutedResCounts.resize(1);
  assert(!ExecutedResCounts[0] && "nonzero count for bad resource");
}

void SchedRemainder::
init(ScheduleDAGMI *DAG, const TargetSchedModel *SchedModel) {
  reset();
  if (!SchedModel->hasInstrSchedModel())
    return;
  RemainingCounts.resize(SchedModel->getNumProcResourceKinds());
  for (SUnit &SU : DAG->SUnits) {
    const MCSchedClassDesc *SC = DAG->getSchedClass(&SU);
    RemIssueCount += SchedModel->getNumMicroOps(SU.getInstr(), SC)
      * SchedModel->getMicroOpFactor();
    for (TargetSchedModel::ProcResIter
           PI = SchedModel->getWriteProcResBegin(SC),
           PE = SchedModel->getWriteProcResEnd(SC); PI != PE; ++PI) {
      unsigned PIdx = PI->ProcResourceIdx;
      unsigned Factor = SchedModel->getResourceFactor(PIdx);
      RemainingCounts[PIdx] += (Factor * PI->Cycles);
    }
  }
}

void SchedBoundary::
init(ScheduleDAGMI *dag, const TargetSchedModel *smodel, SchedRemainder *rem) {
  reset();
  DAG = dag;
  SchedModel = smodel;
  Rem = rem;
  if (SchedModel->hasInstrSchedModel()) {
    unsigned ResourceCount = SchedModel->getNumProcResourceKinds();
    ReservedCyclesIndex.resize(ResourceCount);
    ExecutedResCounts.resize(ResourceCount);
    ResourceGroupSubUnitMasks.resize(ResourceCount, APInt(ResourceCount, 0));
    unsigned NumUnits = 0;

    for (unsigned i = 0; i < ResourceCount; ++i) {
      ReservedCyclesIndex[i] = NumUnits;
      NumUnits += SchedModel->getProcResource(i)->NumUnits;
      if (isUnbufferedGroup(i)) {
        auto SubUnits = SchedModel->getProcResource(i)->SubUnitsIdxBegin;
        for (unsigned U = 0, UE = SchedModel->getProcResource(i)->NumUnits;
             U != UE; ++U)
          ResourceGroupSubUnitMasks[i].setBit(SubUnits[U]);
      }
    }

    ReservedCycles.resize(NumUnits, InvalidCycle);
  }
}

/// Compute the stall cycles based on this SUnit's ready time. Heuristics treat
/// these "soft stalls" differently than the hard stall cycles based on CPU
/// resources and computed by checkHazard(). A fully in-order model
/// (MicroOpBufferSize==0) will not make use of this since instructions are not
/// available for scheduling until they are ready. However, a weaker in-order
/// model may use this for heuristics. For example, if a processor has in-order
/// behavior when reading certain resources, this may come into play.
unsigned SchedBoundary::getLatencyStallCycles(SUnit *SU) {
  if (!SU->isUnbuffered)
    return 0;

  unsigned ReadyCycle = (isTop() ? SU->TopReadyCycle : SU->BotReadyCycle);
  if (ReadyCycle > CurrCycle)
    return ReadyCycle - CurrCycle;
  return 0;
}

/// Compute the next cycle at which the given processor resource unit
/// can be scheduled.
unsigned SchedBoundary::getNextResourceCycleByInstance(unsigned InstanceIdx,
                                                       unsigned Cycles) {
  unsigned NextUnreserved = ReservedCycles[InstanceIdx];
  // If this resource has never been used, always return cycle zero.
  if (NextUnreserved == InvalidCycle)
    return 0;
  // For bottom-up scheduling add the cycles needed for the current operation.
  if (!isTop())
    NextUnreserved += Cycles;
  return NextUnreserved;
}

/// Compute the next cycle at which the given processor resource can be
/// scheduled.  Returns the next cycle and the index of the processor resource
/// instance in the reserved cycles vector.
std::pair<unsigned, unsigned>
SchedBoundary::getNextResourceCycle(const MCSchedClassDesc *SC, unsigned PIdx,
                                    unsigned Cycles) {

  unsigned MinNextUnreserved = InvalidCycle;
  unsigned InstanceIdx = 0;
  unsigned StartIndex = ReservedCyclesIndex[PIdx];
  unsigned NumberOfInstances = SchedModel->getProcResource(PIdx)->NumUnits;
  assert(NumberOfInstances > 0 &&
         "Cannot have zero instances of a ProcResource");

  if (isUnbufferedGroup(PIdx)) {
    // If any subunits are used by the instruction, report that the resource
    // group is available at 0, effectively removing the group record from
    // hazarding and basing the hazarding decisions on the subunit records.
    // Otherwise, choose the first available instance from among the subunits.
    // Specifications which assign cycles to both the subunits and the group or
    // which use an unbuffered group with buffered subunits will appear to
    // schedule strangely. In the first case, the additional cycles for the
    // group will be ignored.  In the second, the group will be ignored
    // entirely.
    for (const MCWriteProcResEntry &PE :
         make_range(SchedModel->getWriteProcResBegin(SC),
                    SchedModel->getWriteProcResEnd(SC)))
      if (ResourceGroupSubUnitMasks[PIdx][PE.ProcResourceIdx])
        return std::make_pair(0u, StartIndex);

    auto SubUnits = SchedModel->getProcResource(PIdx)->SubUnitsIdxBegin;
    for (unsigned I = 0, End = NumberOfInstances; I < End; ++I) {
      unsigned NextUnreserved, NextInstanceIdx;
      std::tie(NextUnreserved, NextInstanceIdx) =
          getNextResourceCycle(SC, SubUnits[I], Cycles);
      if (MinNextUnreserved > NextUnreserved) {
        InstanceIdx = NextInstanceIdx;
        MinNextUnreserved = NextUnreserved;
      }
    }
    return std::make_pair(MinNextUnreserved, InstanceIdx);
  }

  for (unsigned I = StartIndex, End = StartIndex + NumberOfInstances; I < End;
       ++I) {
    unsigned NextUnreserved = getNextResourceCycleByInstance(I, Cycles);
    if (MinNextUnreserved > NextUnreserved) {
      InstanceIdx = I;
      MinNextUnreserved = NextUnreserved;
    }
  }
  return std::make_pair(MinNextUnreserved, InstanceIdx);
}

/// Does this SU have a hazard within the current instruction group.
///
/// The scheduler supports two modes of hazard recognition. The first is the
/// ScheduleHazardRecognizer API. It is a fully general hazard recognizer that
/// supports highly complicated in-order reservation tables
/// (ScoreboardHazardRecognizer) and arbitrary target-specific logic.
///
/// The second is a streamlined mechanism that checks for hazards based on
/// simple counters that the scheduler itself maintains. It explicitly checks
/// for instruction dispatch limitations, including the number of micro-ops that
/// can dispatch per cycle.
///
/// TODO: Also check whether the SU must start a new group.
bool SchedBoundary::checkHazard(SUnit *SU) {
  if (HazardRec->isEnabled()
      && HazardRec->getHazardType(SU) != ScheduleHazardRecognizer::NoHazard) {
    return true;
  }

  unsigned uops = SchedModel->getNumMicroOps(SU->getInstr());
  if ((CurrMOps > 0) && (CurrMOps + uops > SchedModel->getIssueWidth())) {
    LLVM_DEBUG(dbgs() << "  SU(" << SU->NodeNum << ") uops="
                      << SchedModel->getNumMicroOps(SU->getInstr()) << '\n');
    return true;
  }

  if (CurrMOps > 0 &&
      ((isTop() && SchedModel->mustBeginGroup(SU->getInstr())) ||
       (!isTop() && SchedModel->mustEndGroup(SU->getInstr())))) {
    LLVM_DEBUG(dbgs() << "  hazard: SU(" << SU->NodeNum << ") must "
                      << (isTop() ? "begin" : "end") << " group\n");
    return true;
  }

  if (SchedModel->hasInstrSchedModel() && SU->hasReservedResource) {
    const MCSchedClassDesc *SC = DAG->getSchedClass(SU);
    for (const MCWriteProcResEntry &PE :
          make_range(SchedModel->getWriteProcResBegin(SC),
                     SchedModel->getWriteProcResEnd(SC))) {
      unsigned ResIdx = PE.ProcResourceIdx;
      unsigned Cycles = PE.Cycles;
      unsigned NRCycle, InstanceIdx;
      std::tie(NRCycle, InstanceIdx) = getNextResourceCycle(SC, ResIdx, Cycles);
      if (NRCycle > CurrCycle) {
#ifndef NDEBUG
        MaxObservedStall = std::max(Cycles, MaxObservedStall);
#endif
        LLVM_DEBUG(dbgs() << "  SU(" << SU->NodeNum << ") "
                          << SchedModel->getResourceName(ResIdx)
                          << '[' << InstanceIdx - ReservedCyclesIndex[ResIdx]  << ']'
                          << "=" << NRCycle << "c\n");
        return true;
      }
    }
  }
  return false;
}

// Find the unscheduled node in ReadySUs with the highest latency.
unsigned SchedBoundary::
findMaxLatency(ArrayRef<SUnit*> ReadySUs) {
  SUnit *LateSU = nullptr;
  unsigned RemLatency = 0;
  for (SUnit *SU : ReadySUs) {
    unsigned L = getUnscheduledLatency(SU);
    if (L > RemLatency) {
      RemLatency = L;
      LateSU = SU;
    }
  }
  if (LateSU) {
    LLVM_DEBUG(dbgs() << Available.getName() << " RemLatency SU("
                      << LateSU->NodeNum << ") " << RemLatency << "c\n");
  }
  return RemLatency;
}

// Count resources in this zone and the remaining unscheduled
// instruction. Return the max count, scaled. Set OtherCritIdx to the critical
// resource index, or zero if the zone is issue limited.
unsigned SchedBoundary::
getOtherResourceCount(unsigned &OtherCritIdx) {
  OtherCritIdx = 0;
  if (!SchedModel->hasInstrSchedModel())
    return 0;

  unsigned OtherCritCount = Rem->RemIssueCount
    + (RetiredMOps * SchedModel->getMicroOpFactor());
  LLVM_DEBUG(dbgs() << "  " << Available.getName() << " + Remain MOps: "
                    << OtherCritCount / SchedModel->getMicroOpFactor() << '\n');
  for (unsigned PIdx = 1, PEnd = SchedModel->getNumProcResourceKinds();
       PIdx != PEnd; ++PIdx) {
    unsigned OtherCount = getResourceCount(PIdx) + Rem->RemainingCounts[PIdx];
    if (OtherCount > OtherCritCount) {
      OtherCritCount = OtherCount;
      OtherCritIdx = PIdx;
    }
  }
  if (OtherCritIdx) {
    LLVM_DEBUG(
        dbgs() << "  " << Available.getName() << " + Remain CritRes: "
               << OtherCritCount / SchedModel->getResourceFactor(OtherCritIdx)
               << " " << SchedModel->getResourceName(OtherCritIdx) << "\n");
  }
  return OtherCritCount;
}

void SchedBoundary::releaseNode(SUnit *SU, unsigned ReadyCycle, bool InPQueue,
                                unsigned Idx) {
  assert(SU->getInstr() && "Scheduled SUnit must have instr");

#ifndef NDEBUG
  // ReadyCycle was been bumped up to the CurrCycle when this node was
  // scheduled, but CurrCycle may have been eagerly advanced immediately after
  // scheduling, so may now be greater than ReadyCycle.
  if (ReadyCycle > CurrCycle)
    MaxObservedStall = std::max(ReadyCycle - CurrCycle, MaxObservedStall);
#endif

  if (ReadyCycle < MinReadyCycle)
    MinReadyCycle = ReadyCycle;

  // Check for interlocks first. For the purpose of other heuristics, an
  // instruction that cannot issue appears as if it's not in the ReadyQueue.
  bool IsBuffered = SchedModel->getMicroOpBufferSize() != 0;
  bool HazardDetected = (!IsBuffered && ReadyCycle > CurrCycle) ||
                        checkHazard(SU) || (Available.size() >= ReadyListLimit);

  if (!HazardDetected) {
    Available.push(SU);

    if (InPQueue)
      Pending.remove(Pending.begin() + Idx);
    return;
  }

  if (!InPQueue)
    Pending.push(SU);
}

/// Move the boundary of scheduled code by one cycle.
void SchedBoundary::bumpCycle(unsigned NextCycle) {
  if (SchedModel->getMicroOpBufferSize() == 0) {
    assert(MinReadyCycle < std::numeric_limits<unsigned>::max() &&
           "MinReadyCycle uninitialized");
    if (MinReadyCycle > NextCycle)
      NextCycle = MinReadyCycle;
  }
  // Update the current micro-ops, which will issue in the next cycle.
  unsigned DecMOps = SchedModel->getIssueWidth() * (NextCycle - CurrCycle);
  CurrMOps = (CurrMOps <= DecMOps) ? 0 : CurrMOps - DecMOps;

  // Decrement DependentLatency based on the next cycle.
  if ((NextCycle - CurrCycle) > DependentLatency)
    DependentLatency = 0;
  else
    DependentLatency -= (NextCycle - CurrCycle);

  if (!HazardRec->isEnabled()) {
    // Bypass HazardRec virtual calls.
    CurrCycle = NextCycle;
  } else {
    // Bypass getHazardType calls in case of long latency.
    for (; CurrCycle != NextCycle; ++CurrCycle) {
      if (isTop())
        HazardRec->AdvanceCycle();
      else
        HazardRec->RecedeCycle();
    }
  }
  CheckPending = true;
  IsResourceLimited =
      checkResourceLimit(SchedModel->getLatencyFactor(), getCriticalCount(),
                         getScheduledLatency(), true);

  LLVM_DEBUG(dbgs() << "Cycle: " << CurrCycle << ' ' << Available.getName()
                    << '\n');
}

void SchedBoundary::incExecutedResources(unsigned PIdx, unsigned Count) {
  ExecutedResCounts[PIdx] += Count;
  if (ExecutedResCounts[PIdx] > MaxExecutedResCount)
    MaxExecutedResCount = ExecutedResCounts[PIdx];
}

/// Add the given processor resource to this scheduled zone.
///
/// \param Cycles indicates the number of consecutive (non-pipelined) cycles
/// during which this resource is consumed.
///
/// \return the next cycle at which the instruction may execute without
/// oversubscribing resources.
unsigned SchedBoundary::countResource(const MCSchedClassDesc *SC, unsigned PIdx,
                                      unsigned Cycles, unsigned NextCycle) {
  unsigned Factor = SchedModel->getResourceFactor(PIdx);
  unsigned Count = Factor * Cycles;
  LLVM_DEBUG(dbgs() << "  " << SchedModel->getResourceName(PIdx) << " +"
                    << Cycles << "x" << Factor << "u\n");

  // Update Executed resources counts.
  incExecutedResources(PIdx, Count);
  assert(Rem->RemainingCounts[PIdx] >= Count && "resource double counted");
  Rem->RemainingCounts[PIdx] -= Count;

  // Check if this resource exceeds the current critical resource. If so, it
  // becomes the critical resource.
  if (ZoneCritResIdx != PIdx && (getResourceCount(PIdx) > getCriticalCount())) {
    ZoneCritResIdx = PIdx;
    LLVM_DEBUG(dbgs() << "  *** Critical resource "
                      << SchedModel->getResourceName(PIdx) << ": "
                      << getResourceCount(PIdx) / SchedModel->getLatencyFactor()
                      << "c\n");
  }
  // For reserved resources, record the highest cycle using the resource.
  unsigned NextAvailable, InstanceIdx;
  std::tie(NextAvailable, InstanceIdx) = getNextResourceCycle(SC, PIdx, Cycles);
  if (NextAvailable > CurrCycle) {
    LLVM_DEBUG(dbgs() << "  Resource conflict: "
                      << SchedModel->getResourceName(PIdx)
                      << '[' << InstanceIdx - ReservedCyclesIndex[PIdx]  << ']'
                      << " reserved until @" << NextAvailable << "\n");
  }
  return NextAvailable;
}

/// Move the boundary of scheduled code by one SUnit.
void SchedBoundary::bumpNode(SUnit *SU) {
  // Update the reservation table.
  if (HazardRec->isEnabled()) {
    if (!isTop() && SU->isCall) {
      // Calls are scheduled with their preceding instructions. For bottom-up
      // scheduling, clear the pipeline state before emitting.
      HazardRec->Reset();
    }
    HazardRec->EmitInstruction(SU);
    // Scheduling an instruction may have made pending instructions available.
    CheckPending = true;
  }
  // checkHazard should prevent scheduling multiple instructions per cycle that
  // exceed the issue width.
  const MCSchedClassDesc *SC = DAG->getSchedClass(SU);
  unsigned IncMOps = SchedModel->getNumMicroOps(SU->getInstr());
  assert(
      (CurrMOps == 0 || (CurrMOps + IncMOps) <= SchedModel->getIssueWidth()) &&
      "Cannot schedule this instruction's MicroOps in the current cycle.");

  unsigned ReadyCycle = (isTop() ? SU->TopReadyCycle : SU->BotReadyCycle);
  LLVM_DEBUG(dbgs() << "  Ready @" << ReadyCycle << "c\n");

  unsigned NextCycle = CurrCycle;
  switch (SchedModel->getMicroOpBufferSize()) {
  case 0:
    assert(ReadyCycle <= CurrCycle && "Broken PendingQueue");
    break;
  case 1:
    if (ReadyCycle > NextCycle) {
      NextCycle = ReadyCycle;
      LLVM_DEBUG(dbgs() << "  *** Stall until: " << ReadyCycle << "\n");
    }
    break;
  default:
    // We don't currently model the OOO reorder buffer, so consider all
    // scheduled MOps to be "retired". We do loosely model in-order resource
    // latency. If this instruction uses an in-order resource, account for any
    // likely stall cycles.
    if (SU->isUnbuffered && ReadyCycle > NextCycle)
      NextCycle = ReadyCycle;
    break;
  }
  RetiredMOps += IncMOps;

  // Update resource counts and critical resource.
  if (SchedModel->hasInstrSchedModel()) {
    unsigned DecRemIssue = IncMOps * SchedModel->getMicroOpFactor();
    assert(Rem->RemIssueCount >= DecRemIssue && "MOps double counted");
    Rem->RemIssueCount -= DecRemIssue;
    if (ZoneCritResIdx) {
      // Scale scheduled micro-ops for comparing with the critical resource.
      unsigned ScaledMOps =
        RetiredMOps * SchedModel->getMicroOpFactor();

      // If scaled micro-ops are now more than the previous critical resource by
      // a full cycle, then micro-ops issue becomes critical.
      if ((int)(ScaledMOps - getResourceCount(ZoneCritResIdx))
          >= (int)SchedModel->getLatencyFactor()) {
        ZoneCritResIdx = 0;
        LLVM_DEBUG(dbgs() << "  *** Critical resource NumMicroOps: "
                          << ScaledMOps / SchedModel->getLatencyFactor()
                          << "c\n");
      }
    }
    for (TargetSchedModel::ProcResIter
           PI = SchedModel->getWriteProcResBegin(SC),
           PE = SchedModel->getWriteProcResEnd(SC); PI != PE; ++PI) {
      unsigned RCycle =
        countResource(SC, PI->ProcResourceIdx, PI->Cycles, NextCycle);
      if (RCycle > NextCycle)
        NextCycle = RCycle;
    }
    if (SU->hasReservedResource) {
      // For reserved resources, record the highest cycle using the resource.
      // For top-down scheduling, this is the cycle in which we schedule this
      // instruction plus the number of cycles the operations reserves the
      // resource. For bottom-up is it simply the instruction's cycle.
      for (TargetSchedModel::ProcResIter
             PI = SchedModel->getWriteProcResBegin(SC),
             PE = SchedModel->getWriteProcResEnd(SC); PI != PE; ++PI) {
        unsigned PIdx = PI->ProcResourceIdx;
        if (SchedModel->getProcResource(PIdx)->BufferSize == 0) {
          unsigned ReservedUntil, InstanceIdx;
          std::tie(ReservedUntil, InstanceIdx) =
              getNextResourceCycle(SC, PIdx, 0);
          if (isTop()) {
            ReservedCycles[InstanceIdx] =
                std::max(ReservedUntil, NextCycle + PI->Cycles);
          } else
            ReservedCycles[InstanceIdx] = NextCycle;
        }
      }
    }
  }
  // Update ExpectedLatency and DependentLatency.
  unsigned &TopLatency = isTop() ? ExpectedLatency : DependentLatency;
  unsigned &BotLatency = isTop() ? DependentLatency : ExpectedLatency;
  if (SU->getDepth() > TopLatency) {
    TopLatency = SU->getDepth();
    LLVM_DEBUG(dbgs() << "  " << Available.getName() << " TopLatency SU("
                      << SU->NodeNum << ") " << TopLatency << "c\n");
  }
  if (SU->getHeight() > BotLatency) {
    BotLatency = SU->getHeight();
    LLVM_DEBUG(dbgs() << "  " << Available.getName() << " BotLatency SU("
                      << SU->NodeNum << ") " << BotLatency << "c\n");
  }
  // If we stall for any reason, bump the cycle.
  if (NextCycle > CurrCycle)
    bumpCycle(NextCycle);
  else
    // After updating ZoneCritResIdx and ExpectedLatency, check if we're
    // resource limited. If a stall occurred, bumpCycle does this.
    IsResourceLimited =
        checkResourceLimit(SchedModel->getLatencyFactor(), getCriticalCount(),
                           getScheduledLatency(), true);

  // Update CurrMOps after calling bumpCycle to handle stalls, since bumpCycle
  // resets CurrMOps. Loop to handle instructions with more MOps than issue in
  // one cycle.  Since we commonly reach the max MOps here, opportunistically
  // bump the cycle to avoid uselessly checking everything in the readyQ.
  CurrMOps += IncMOps;

  // Bump the cycle count for issue group constraints.
  // This must be done after NextCycle has been adjust for all other stalls.
  // Calling bumpCycle(X) will reduce CurrMOps by one issue group and set
  // currCycle to X.
  if ((isTop() &&  SchedModel->mustEndGroup(SU->getInstr())) ||
      (!isTop() && SchedModel->mustBeginGroup(SU->getInstr()))) {
    LLVM_DEBUG(dbgs() << "  Bump cycle to " << (isTop() ? "end" : "begin")
                      << " group\n");
    bumpCycle(++NextCycle);
  }

  while (CurrMOps >= SchedModel->getIssueWidth()) {
    LLVM_DEBUG(dbgs() << "  *** Max MOps " << CurrMOps << " at cycle "
                      << CurrCycle << '\n');
    bumpCycle(++NextCycle);
  }
  LLVM_DEBUG(dumpScheduledState());
}

/// Release pending ready nodes in to the available queue. This makes them
/// visible to heuristics.
void SchedBoundary::releasePending() {
  // If the available queue is empty, it is safe to reset MinReadyCycle.
  if (Available.empty())
    MinReadyCycle = std::numeric_limits<unsigned>::max();

  // Check to see if any of the pending instructions are ready to issue.  If
  // so, add them to the available queue.
  for (unsigned I = 0, E = Pending.size(); I < E; ++I) {
    SUnit *SU = *(Pending.begin() + I);
    unsigned ReadyCycle = isTop() ? SU->TopReadyCycle : SU->BotReadyCycle;

    if (ReadyCycle < MinReadyCycle)
      MinReadyCycle = ReadyCycle;

    if (Available.size() >= ReadyListLimit)
      break;

    releaseNode(SU, ReadyCycle, true, I);
    if (E != Pending.size()) {
      --I;
      --E;
    }
  }
  CheckPending = false;
}

/// Remove SU from the ready set for this boundary.
void SchedBoundary::removeReady(SUnit *SU) {
  if (Available.isInQueue(SU))
    Available.remove(Available.find(SU));
  else {
    assert(Pending.isInQueue(SU) && "bad ready count");
    Pending.remove(Pending.find(SU));
  }
}

/// If this queue only has one ready candidate, return it. As a side effect,
/// defer any nodes that now hit a hazard, and advance the cycle until at least
/// one node is ready. If multiple instructions are ready, return NULL.
SUnit *SchedBoundary::pickOnlyChoice() {
  if (CheckPending)
    releasePending();

  // Defer any ready instrs that now have a hazard.
  for (ReadyQueue::iterator I = Available.begin(); I != Available.end();) {
    if (checkHazard(*I)) {
      Pending.push(*I);
      I = Available.remove(I);
      continue;
    }
    ++I;
  }
  for (unsigned i = 0; Available.empty(); ++i) {
//  FIXME: Re-enable assert once PR20057 is resolved.
//    assert(i <= (HazardRec->getMaxLookAhead() + MaxObservedStall) &&
//           "permanent hazard");
    (void)i;
    bumpCycle(CurrCycle + 1);
    releasePending();
  }

  LLVM_DEBUG(Pending.dump());
  LLVM_DEBUG(Available.dump());

  if (Available.size() == 1)
    return *Available.begin();
  return nullptr;
}

#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
// This is useful information to dump after bumpNode.
// Note that the Queue contents are more useful before pickNodeFromQueue.
LLVM_DUMP_METHOD void SchedBoundary::dumpScheduledState() const {
  unsigned ResFactor;
  unsigned ResCount;
  if (ZoneCritResIdx) {
    ResFactor = SchedModel->getResourceFactor(ZoneCritResIdx);
    ResCount = getResourceCount(ZoneCritResIdx);
  } else {
    ResFactor = SchedModel->getMicroOpFactor();
    ResCount = RetiredMOps * ResFactor;
  }
  unsigned LFactor = SchedModel->getLatencyFactor();
  dbgs() << Available.getName() << " @" << CurrCycle << "c\n"
         << "  Retired: " << RetiredMOps;
  dbgs() << "\n  Executed: " << getExecutedCount() / LFactor << "c";
  dbgs() << "\n  Critical: " << ResCount / LFactor << "c, "
         << ResCount / ResFactor << " "
         << SchedModel->getResourceName(ZoneCritResIdx)
         << "\n  ExpectedLatency: " << ExpectedLatency << "c\n"
         << (IsResourceLimited ? "  - Resource" : "  - Latency")
         << " limited.\n";
}
#endif

//===----------------------------------------------------------------------===//
// GenericScheduler - Generic implementation of MachineSchedStrategy.
//===----------------------------------------------------------------------===//

void GenericSchedulerBase::SchedCandidate::
initResourceDelta(const ScheduleDAGMI *DAG,
                  const TargetSchedModel *SchedModel) {
  if (!Policy.ReduceResIdx && !Policy.DemandResIdx)
    return;

  const MCSchedClassDesc *SC = DAG->getSchedClass(SU);
  for (TargetSchedModel::ProcResIter
         PI = SchedModel->getWriteProcResBegin(SC),
         PE = SchedModel->getWriteProcResEnd(SC); PI != PE; ++PI) {
    if (PI->ProcResourceIdx == Policy.ReduceResIdx)
      ResDelta.CritResources += PI->Cycles;
    if (PI->ProcResourceIdx == Policy.DemandResIdx)
      ResDelta.DemandedResources += PI->Cycles;
  }
}

/// Compute remaining latency. We need this both to determine whether the
/// overall schedule has become latency-limited and whether the instructions
/// outside this zone are resource or latency limited.
///
/// The "dependent" latency is updated incrementally during scheduling as the
/// max height/depth of scheduled nodes minus the cycles since it was
/// scheduled:
///   DLat = max (N.depth - (CurrCycle - N.ReadyCycle) for N in Zone
///
/// The "independent" latency is the max ready queue depth:
///   ILat = max N.depth for N in Available|Pending
///
/// RemainingLatency is the greater of independent and dependent latency.
///
/// These computations are expensive, especially in DAGs with many edges, so
/// only do them if necessary.
static unsigned computeRemLatency(SchedBoundary &CurrZone) {
  unsigned RemLatency = CurrZone.getDependentLatency();
  RemLatency = std::max(RemLatency,
                        CurrZone.findMaxLatency(CurrZone.Available.elements()));
  RemLatency = std::max(RemLatency,
                        CurrZone.findMaxLatency(CurrZone.Pending.elements()));
  return RemLatency;
}

/// Returns true if the current cycle plus remaning latency is greater than
/// the critical path in the scheduling region.
bool GenericSchedulerBase::shouldReduceLatency(const CandPolicy &Policy,
                                               SchedBoundary &CurrZone,
                                               bool ComputeRemLatency,
                                               unsigned &RemLatency) const {
  // The current cycle is already greater than the critical path, so we are
  // already latency limited and don't need to compute the remaining latency.
  if (CurrZone.getCurrCycle() > Rem.CriticalPath)
    return true;

  // If we haven't scheduled anything yet, then we aren't latency limited.
  if (CurrZone.getCurrCycle() == 0)
    return false;

  if (ComputeRemLatency)
    RemLatency = computeRemLatency(CurrZone);

  return RemLatency + CurrZone.getCurrCycle() > Rem.CriticalPath;
}

/// Set the CandPolicy given a scheduling zone given the current resources and
/// latencies inside and outside the zone.
void GenericSchedulerBase::setPolicy(CandPolicy &Policy, bool IsPostRA,
                                     SchedBoundary &CurrZone,
                                     SchedBoundary *OtherZone) {
  // Apply preemptive heuristics based on the total latency and resources
  // inside and outside this zone. Potential stalls should be considered before
  // following this policy.

  // Compute the critical resource outside the zone.
  unsigned OtherCritIdx = 0;
  unsigned OtherCount =
    OtherZone ? OtherZone->getOtherResourceCount(OtherCritIdx) : 0;

  bool OtherResLimited = false;
  unsigned RemLatency = 0;
  bool RemLatencyComputed = false;
  if (SchedModel->hasInstrSchedModel() && OtherCount != 0) {
    RemLatency = computeRemLatency(CurrZone);
    RemLatencyComputed = true;
    OtherResLimited = checkResourceLimit(SchedModel->getLatencyFactor(),
                                         OtherCount, RemLatency, false);
  }

  // Schedule aggressively for latency in PostRA mode. We don't check for
  // acyclic latency during PostRA, and highly out-of-order processors will
  // skip PostRA scheduling.
  if (!OtherResLimited &&
      (IsPostRA || shouldReduceLatency(Policy, CurrZone, !RemLatencyComputed,
                                       RemLatency))) {
    Policy.ReduceLatency |= true;
    LLVM_DEBUG(dbgs() << "  " << CurrZone.Available.getName()
                      << " RemainingLatency " << RemLatency << " + "
                      << CurrZone.getCurrCycle() << "c > CritPath "
                      << Rem.CriticalPath << "\n");
  }
  // If the same resource is limiting inside and outside the zone, do nothing.
  if (CurrZone.getZoneCritResIdx() == OtherCritIdx)
    return;

  LLVM_DEBUG(if (CurrZone.isResourceLimited()) {
    dbgs() << "  " << CurrZone.Available.getName() << " ResourceLimited: "
           << SchedModel->getResourceName(CurrZone.getZoneCritResIdx()) << "\n";
  } if (OtherResLimited) dbgs()
                 << "  RemainingLimit: "
                 << SchedModel->getResourceName(OtherCritIdx) << "\n";
             if (!CurrZone.isResourceLimited() && !OtherResLimited) dbgs()
             << "  Latency limited both directions.\n");

  if (CurrZone.isResourceLimited() && !Policy.ReduceResIdx)
    Policy.ReduceResIdx = CurrZone.getZoneCritResIdx();

  if (OtherResLimited)
    Policy.DemandResIdx = OtherCritIdx;
}

#ifndef NDEBUG
const char *GenericSchedulerBase::getReasonStr(
  GenericSchedulerBase::CandReason Reason) {
  switch (Reason) {
  case NoCand:         return "NOCAND    ";
  case Only1:          return "ONLY1     ";
  case PhysReg:        return "PHYS-REG  ";
  case RegExcess:      return "REG-EXCESS";
  case RegCritical:    return "REG-CRIT  ";
  case Stall:          return "STALL     ";
  case Cluster:        return "CLUSTER   ";
  case Weak:           return "WEAK      ";
  case RegMax:         return "REG-MAX   ";
  case ResourceReduce: return "RES-REDUCE";
  case ResourceDemand: return "RES-DEMAND";
  case TopDepthReduce: return "TOP-DEPTH ";
  case TopPathReduce:  return "TOP-PATH  ";
  case BotHeightReduce:return "BOT-HEIGHT";
  case BotPathReduce:  return "BOT-PATH  ";
  case NextDefUse:     return "DEF-USE   ";
  case NodeOrder:      return "ORDER     ";
  };
  llvm_unreachable("Unknown reason!");
}

void GenericSchedulerBase::traceCandidate(const SchedCandidate &Cand) {
  PressureChange P;
  unsigned ResIdx = 0;
  unsigned Latency = 0;
  switch (Cand.Reason) {
  default:
    break;
  case RegExcess:
    P = Cand.RPDelta.Excess;
    break;
  case RegCritical:
    P = Cand.RPDelta.CriticalMax;
    break;
  case RegMax:
    P = Cand.RPDelta.CurrentMax;
    break;
  case ResourceReduce:
    ResIdx = Cand.Policy.ReduceResIdx;
    break;
  case ResourceDemand:
    ResIdx = Cand.Policy.DemandResIdx;
    break;
  case TopDepthReduce:
    Latency = Cand.SU->getDepth();
    break;
  case TopPathReduce:
    Latency = Cand.SU->getHeight();
    break;
  case BotHeightReduce:
    Latency = Cand.SU->getHeight();
    break;
  case BotPathReduce:
    Latency = Cand.SU->getDepth();
    break;
  }
  dbgs() << "  Cand SU(" << Cand.SU->NodeNum << ") " << getReasonStr(Cand.Reason);
  if (P.isValid())
    dbgs() << " " << TRI->getRegPressureSetName(P.getPSet())
           << ":" << P.getUnitInc() << " ";
  else
    dbgs() << "      ";
  if (ResIdx)
    dbgs() << " " << SchedModel->getProcResource(ResIdx)->Name << " ";
  else
    dbgs() << "         ";
  if (Latency)
    dbgs() << " " << Latency << " cycles ";
  else
    dbgs() << "          ";
  dbgs() << '\n';
}
#endif

namespace llvm {
/// Return true if this heuristic determines order.
/// TODO: Consider refactor return type of these functions as integer or enum,
/// as we may need to differentiate whether TryCand is better than Cand.
bool tryLess(int TryVal, int CandVal,
             GenericSchedulerBase::SchedCandidate &TryCand,
             GenericSchedulerBase::SchedCandidate &Cand,
             GenericSchedulerBase::CandReason Reason) {
  if (TryVal < CandVal) {
    TryCand.Reason = Reason;
    return true;
  }
  if (TryVal > CandVal) {
    if (Cand.Reason > Reason)
      Cand.Reason = Reason;
    return true;
  }
  return false;
}

bool tryGreater(int TryVal, int CandVal,
                GenericSchedulerBase::SchedCandidate &TryCand,
                GenericSchedulerBase::SchedCandidate &Cand,
                GenericSchedulerBase::CandReason Reason) {
  if (TryVal > CandVal) {
    TryCand.Reason = Reason;
    return true;
  }
  if (TryVal < CandVal) {
    if (Cand.Reason > Reason)
      Cand.Reason = Reason;
    return true;
  }
  return false;
}

bool tryLatency(GenericSchedulerBase::SchedCandidate &TryCand,
                GenericSchedulerBase::SchedCandidate &Cand,
                SchedBoundary &Zone) {
  if (Zone.isTop()) {
    // Prefer the candidate with the lesser depth, but only if one of them has
    // depth greater than the total latency scheduled so far, otherwise either
    // of them could be scheduled now with no stall.
    if (std::max(TryCand.SU->getDepth(), Cand.SU->getDepth()) >
        Zone.getScheduledLatency()) {
      if (tryLess(TryCand.SU->getDepth(), Cand.SU->getDepth(),
                  TryCand, Cand, GenericSchedulerBase::TopDepthReduce))
        return true;
    }
    if (tryGreater(TryCand.SU->getHeight(), Cand.SU->getHeight(),
                   TryCand, Cand, GenericSchedulerBase::TopPathReduce))
      return true;
  } else {
    // Prefer the candidate with the lesser height, but only if one of them has
    // height greater than the total latency scheduled so far, otherwise either
    // of them could be scheduled now with no stall.
    if (std::max(TryCand.SU->getHeight(), Cand.SU->getHeight()) >
        Zone.getScheduledLatency()) {
      if (tryLess(TryCand.SU->getHeight(), Cand.SU->getHeight(),
                  TryCand, Cand, GenericSchedulerBase::BotHeightReduce))
        return true;
    }
    if (tryGreater(TryCand.SU->getDepth(), Cand.SU->getDepth(),
                   TryCand, Cand, GenericSchedulerBase::BotPathReduce))
      return true;
  }
  return false;
}
} // end namespace llvm

static void tracePick(GenericSchedulerBase::CandReason Reason, bool IsTop) {
  LLVM_DEBUG(dbgs() << "Pick " << (IsTop ? "Top " : "Bot ")
                    << GenericSchedulerBase::getReasonStr(Reason) << '\n');
}

static void tracePick(const GenericSchedulerBase::SchedCandidate &Cand) {
  tracePick(Cand.Reason, Cand.AtTop);
}

void GenericScheduler::initialize(ScheduleDAGMI *dag) {
  assert(dag->hasVRegLiveness() &&
         "(PreRA)GenericScheduler needs vreg liveness");
  DAG = static_cast<ScheduleDAGMILive*>(dag);
  SchedModel = DAG->getSchedModel();
  TRI = DAG->TRI;

  if (RegionPolicy.ComputeDFSResult)
    DAG->computeDFSResult();

  Rem.init(DAG, SchedModel);
  Top.init(DAG, SchedModel, &Rem);
  Bot.init(DAG, SchedModel, &Rem);

  // Initialize resource counts.

  // Initialize the HazardRecognizers. If itineraries don't exist, are empty, or
  // are disabled, then these HazardRecs will be disabled.
  const InstrItineraryData *Itin = SchedModel->getInstrItineraries();
  if (!Top.HazardRec) {
    Top.HazardRec =
        DAG->MF.getSubtarget().getInstrInfo()->CreateTargetMIHazardRecognizer(
            Itin, DAG);
  }
  if (!Bot.HazardRec) {
    Bot.HazardRec =
        DAG->MF.getSubtarget().getInstrInfo()->CreateTargetMIHazardRecognizer(
            Itin, DAG);
  }
  TopCand.SU = nullptr;
  BotCand.SU = nullptr;
}

/// Initialize the per-region scheduling policy.
void GenericScheduler::initPolicy(MachineBasicBlock::iterator Begin,
                                  MachineBasicBlock::iterator End,
                                  unsigned NumRegionInstrs) {
  const MachineFunction &MF = *Begin->getMF();
  const TargetLowering *TLI = MF.getSubtarget().getTargetLowering();

  // Avoid setting up the register pressure tracker for small regions to save
  // compile time. As a rough heuristic, only track pressure when the number of
  // schedulable instructions exceeds half the integer register file.
  RegionPolicy.ShouldTrackPressure = true;
  for (unsigned VT = MVT::i32; VT > (unsigned)MVT::i1; --VT) {
    MVT::SimpleValueType LegalIntVT = (MVT::SimpleValueType)VT;
    if (TLI->isTypeLegal(LegalIntVT)) {
      unsigned NIntRegs = Context->RegClassInfo->getNumAllocatableRegs(
        TLI->getRegClassFor(LegalIntVT));
      RegionPolicy.ShouldTrackPressure = NumRegionInstrs > (NIntRegs / 2);
    }
  }

  // For generic targets, we default to bottom-up, because it's simpler and more
  // compile-time optimizations have been implemented in that direction.
  RegionPolicy.OnlyBottomUp = true;

  // Allow the subtarget to override default policy.
  MF.getSubtarget().overrideSchedPolicy(RegionPolicy, NumRegionInstrs);

  // After subtarget overrides, apply command line options.
  if (!EnableRegPressure) {
    RegionPolicy.ShouldTrackPressure = false;
    RegionPolicy.ShouldTrackLaneMasks = false;
  }

  // Check -misched-topdown/bottomup can force or unforce scheduling direction.
  // e.g. -misched-bottomup=false allows scheduling in both directions.
  assert((!ForceTopDown || !ForceBottomUp) &&
         "-misched-topdown incompatible with -misched-bottomup");
  if (ForceBottomUp.getNumOccurrences() > 0) {
    RegionPolicy.OnlyBottomUp = ForceBottomUp;
    if (RegionPolicy.OnlyBottomUp)
      RegionPolicy.OnlyTopDown = false;
  }
  if (ForceTopDown.getNumOccurrences() > 0) {
    RegionPolicy.OnlyTopDown = ForceTopDown;
    if (RegionPolicy.OnlyTopDown)
      RegionPolicy.OnlyBottomUp = false;
  }
}

void GenericScheduler::dumpPolicy() const {
  // Cannot completely remove virtual function even in release mode.
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
  dbgs() << "GenericScheduler RegionPolicy: "
         << " ShouldTrackPressure=" << RegionPolicy.ShouldTrackPressure
         << " OnlyTopDown=" << RegionPolicy.OnlyTopDown
         << " OnlyBottomUp=" << RegionPolicy.OnlyBottomUp
         << "\n";
#endif
}

/// Set IsAcyclicLatencyLimited if the acyclic path is longer than the cyclic
/// critical path by more cycles than it takes to drain the instruction buffer.
/// We estimate an upper bounds on in-flight instructions as:
///
/// CyclesPerIteration = max( CyclicPath, Loop-Resource-Height )
/// InFlightIterations = AcyclicPath / CyclesPerIteration
/// InFlightResources = InFlightIterations * LoopResources
///
/// TODO: Check execution resources in addition to IssueCount.
void GenericScheduler::checkAcyclicLatency() {
  if (Rem.CyclicCritPath == 0 || Rem.CyclicCritPath >= Rem.CriticalPath)
    return;

  // Scaled number of cycles per loop iteration.
  unsigned IterCount =
    std::max(Rem.CyclicCritPath * SchedModel->getLatencyFactor(),
             Rem.RemIssueCount);
  // Scaled acyclic critical path.
  unsigned AcyclicCount = Rem.CriticalPath * SchedModel->getLatencyFactor();
  // InFlightCount = (AcyclicPath / IterCycles) * InstrPerLoop
  unsigned InFlightCount =
    (AcyclicCount * Rem.RemIssueCount + IterCount-1) / IterCount;
  unsigned BufferLimit =
    SchedModel->getMicroOpBufferSize() * SchedModel->getMicroOpFactor();

  Rem.IsAcyclicLatencyLimited = InFlightCount > BufferLimit;

  LLVM_DEBUG(
      dbgs() << "IssueCycles="
             << Rem.RemIssueCount / SchedModel->getLatencyFactor() << "c "
             << "IterCycles=" << IterCount / SchedModel->getLatencyFactor()
             << "c NumIters=" << (AcyclicCount + IterCount - 1) / IterCount
             << " InFlight=" << InFlightCount / SchedModel->getMicroOpFactor()
             << "m BufferLim=" << SchedModel->getMicroOpBufferSize() << "m\n";
      if (Rem.IsAcyclicLatencyLimited) dbgs() << "  ACYCLIC LATENCY LIMIT\n");
}

void GenericScheduler::registerRoots() {
  Rem.CriticalPath = DAG->ExitSU.getDepth();

  // Some roots may not feed into ExitSU. Check all of them in case.
  for (const SUnit *SU : Bot.Available) {
    if (SU->getDepth() > Rem.CriticalPath)
      Rem.CriticalPath = SU->getDepth();
  }
  LLVM_DEBUG(dbgs() << "Critical Path(GS-RR ): " << Rem.CriticalPath << '\n');
  if (DumpCriticalPathLength) {
    errs() << "Critical Path(GS-RR ): " << Rem.CriticalPath << " \n";
  }

  if (EnableCyclicPath && SchedModel->getMicroOpBufferSize() > 0) {
    Rem.CyclicCritPath = DAG->computeCyclicCriticalPath();
    checkAcyclicLatency();
  }
}

namespace llvm {
bool tryPressure(const PressureChange &TryP,
                 const PressureChange &CandP,
                 GenericSchedulerBase::SchedCandidate &TryCand,
                 GenericSchedulerBase::SchedCandidate &Cand,
                 GenericSchedulerBase::CandReason Reason,
                 const TargetRegisterInfo *TRI,
                 const MachineFunction &MF) {
  // If one candidate decreases and the other increases, go with it.
  // Invalid candidates have UnitInc==0.
  if (tryGreater(TryP.getUnitInc() < 0, CandP.getUnitInc() < 0, TryCand, Cand,
                 Reason)) {
    return true;
  }
  // Do not compare the magnitude of pressure changes between top and bottom
  // boundary.
  if (Cand.AtTop != TryCand.AtTop)
    return false;

  // If both candidates affect the same set in the same boundary, go with the
  // smallest increase.
  unsigned TryPSet = TryP.getPSetOrMax();
  unsigned CandPSet = CandP.getPSetOrMax();
  if (TryPSet == CandPSet) {
    return tryLess(TryP.getUnitInc(), CandP.getUnitInc(), TryCand, Cand,
                   Reason);
  }

  int TryRank = TryP.isValid() ? TRI->getRegPressureSetScore(MF, TryPSet) :
                                 std::numeric_limits<int>::max();

  int CandRank = CandP.isValid() ? TRI->getRegPressureSetScore(MF, CandPSet) :
                                   std::numeric_limits<int>::max();

  // If the candidates are decreasing pressure, reverse priority.
  if (TryP.getUnitInc() < 0)
    std::swap(TryRank, CandRank);
  return tryGreater(TryRank, CandRank, TryCand, Cand, Reason);
}

unsigned getWeakLeft(const SUnit *SU, bool isTop) {
  return (isTop) ? SU->WeakPredsLeft : SU->WeakSuccsLeft;
}

/// Minimize physical register live ranges. Regalloc wants them adjacent to
/// their physreg def/use.
///
/// FIXME: This is an unnecessary check on the critical path. Most are root/leaf
/// copies which can be prescheduled. The rest (e.g. x86 MUL) could be bundled
/// with the operation that produces or consumes the physreg. We'll do this when
/// regalloc has support for parallel copies.
int biasPhysReg(const SUnit *SU, bool isTop) {
  const MachineInstr *MI = SU->getInstr();

  if (MI->isCopy()) {
    unsigned ScheduledOper = isTop ? 1 : 0;
    unsigned UnscheduledOper = isTop ? 0 : 1;
    // If we have already scheduled the physreg produce/consumer, immediately
    // schedule the copy.
    if (Register::isPhysicalRegister(MI->getOperand(ScheduledOper).getReg()))
      return 1;
    // If the physreg is at the boundary, defer it. Otherwise schedule it
    // immediately to free the dependent. We can hoist the copy later.
    bool AtBoundary = isTop ? !SU->NumSuccsLeft : !SU->NumPredsLeft;
    if (Register::isPhysicalRegister(MI->getOperand(UnscheduledOper).getReg()))
      return AtBoundary ? -1 : 1;
  }

  if (MI->isMoveImmediate()) {
    // If we have a move immediate and all successors have been assigned, bias
    // towards scheduling this later. Make sure all register defs are to
    // physical registers.
    bool DoBias = true;
    for (const MachineOperand &Op : MI->defs()) {
      if (Op.isReg() && !Register::isPhysicalRegister(Op.getReg())) {
        DoBias = false;
        break;
      }
    }

    if (DoBias)
      return isTop ? -1 : 1;
  }

  return 0;
}
} // end namespace llvm

void GenericScheduler::initCandidate(SchedCandidate &Cand, SUnit *SU,
                                     bool AtTop,
                                     const RegPressureTracker &RPTracker,
                                     RegPressureTracker &TempTracker) {
  Cand.SU = SU;
  Cand.AtTop = AtTop;
  if (DAG->isTrackingPressure()) {
    if (AtTop) {
      TempTracker.getMaxDownwardPressureDelta(
        Cand.SU->getInstr(),
        Cand.RPDelta,
        DAG->getRegionCriticalPSets(),
        DAG->getRegPressure().MaxSetPressure);
    } else {
      if (VerifyScheduling) {
        TempTracker.getMaxUpwardPressureDelta(
          Cand.SU->getInstr(),
          &DAG->getPressureDiff(Cand.SU),
          Cand.RPDelta,
          DAG->getRegionCriticalPSets(),
          DAG->getRegPressure().MaxSetPressure);
      } else {
        RPTracker.getUpwardPressureDelta(
          Cand.SU->getInstr(),
          DAG->getPressureDiff(Cand.SU),
          Cand.RPDelta,
          DAG->getRegionCriticalPSets(),
          DAG->getRegPressure().MaxSetPressure);
      }
    }
  }
  LLVM_DEBUG(if (Cand.RPDelta.Excess.isValid()) dbgs()
             << "  Try  SU(" << Cand.SU->NodeNum << ") "
             << TRI->getRegPressureSetName(Cand.RPDelta.Excess.getPSet()) << ":"
             << Cand.RPDelta.Excess.getUnitInc() << "\n");
}

/// Apply a set of heuristics to a new candidate. Heuristics are currently
/// hierarchical. This may be more efficient than a graduated cost model because
/// we don't need to evaluate all aspects of the model for each node in the
/// queue. But it's really done to make the heuristics easier to debug and
/// statistically analyze.
///
/// \param Cand provides the policy and current best candidate.
/// \param TryCand refers to the next SUnit candidate, otherwise uninitialized.
/// \param Zone describes the scheduled zone that we are extending, or nullptr
///             if Cand is from a different zone than TryCand.
/// \return \c true if TryCand is better than Cand (Reason is NOT NoCand)
bool GenericScheduler::tryCandidate(SchedCandidate &Cand,
                                    SchedCandidate &TryCand,
                                    SchedBoundary *Zone) const {
  // Initialize the candidate if needed.
  if (!Cand.isValid()) {
    TryCand.Reason = NodeOrder;
    return true;
  }

  // Bias PhysReg Defs and copies to their uses and defined respectively.
  if (tryGreater(biasPhysReg(TryCand.SU, TryCand.AtTop),
                 biasPhysReg(Cand.SU, Cand.AtTop), TryCand, Cand, PhysReg))
    return TryCand.Reason != NoCand;

  // Avoid exceeding the target's limit.
  if (DAG->isTrackingPressure() && tryPressure(TryCand.RPDelta.Excess,
                                               Cand.RPDelta.Excess,
                                               TryCand, Cand, RegExcess, TRI,
                                               DAG->MF))
    return TryCand.Reason != NoCand;

  // Avoid increasing the max critical pressure in the scheduled region.
  if (DAG->isTrackingPressure() && tryPressure(TryCand.RPDelta.CriticalMax,
                                               Cand.RPDelta.CriticalMax,
                                               TryCand, Cand, RegCritical, TRI,
                                               DAG->MF))
    return TryCand.Reason != NoCand;

  // We only compare a subset of features when comparing nodes between
  // Top and Bottom boundary. Some properties are simply incomparable, in many
  // other instances we should only override the other boundary if something
  // is a clear good pick on one boundary. Skip heuristics that are more
  // "tie-breaking" in nature.
  bool SameBoundary = Zone != nullptr;
  if (SameBoundary) {
    // For loops that are acyclic path limited, aggressively schedule for
    // latency. Within an single cycle, whenever CurrMOps > 0, allow normal
    // heuristics to take precedence.
    if (Rem.IsAcyclicLatencyLimited && !Zone->getCurrMOps() &&
        tryLatency(TryCand, Cand, *Zone))
      return TryCand.Reason != NoCand;

    // Prioritize instructions that read unbuffered resources by stall cycles.
    if (tryLess(Zone->getLatencyStallCycles(TryCand.SU),
                Zone->getLatencyStallCycles(Cand.SU), TryCand, Cand, Stall))
      return TryCand.Reason != NoCand;
  }

  // Keep clustered nodes together to encourage downstream peephole
  // optimizations which may reduce resource requirements.
  //
  // This is a best effort to set things up for a post-RA pass. Optimizations
  // like generating loads of multiple registers should ideally be done within
  // the scheduler pass by combining the loads during DAG postprocessing.
  const SUnit *CandNextClusterSU =
    Cand.AtTop ? DAG->getNextClusterSucc() : DAG->getNextClusterPred();
  const SUnit *TryCandNextClusterSU =
    TryCand.AtTop ? DAG->getNextClusterSucc() : DAG->getNextClusterPred();
  if (tryGreater(TryCand.SU == TryCandNextClusterSU,
                 Cand.SU == CandNextClusterSU,
                 TryCand, Cand, Cluster))
    return TryCand.Reason != NoCand;

  if (SameBoundary) {
    // Weak edges are for clustering and other constraints.
    if (tryLess(getWeakLeft(TryCand.SU, TryCand.AtTop),
                getWeakLeft(Cand.SU, Cand.AtTop),
                TryCand, Cand, Weak))
      return TryCand.Reason != NoCand;
  }

  // Avoid increasing the max pressure of the entire region.
  if (DAG->isTrackingPressure() && tryPressure(TryCand.RPDelta.CurrentMax,
                                               Cand.RPDelta.CurrentMax,
                                               TryCand, Cand, RegMax, TRI,
                                               DAG->MF))
    return TryCand.Reason != NoCand;

  if (SameBoundary) {
    // Avoid critical resource consumption and balance the schedule.
    TryCand.initResourceDelta(DAG, SchedModel);
    if (tryLess(TryCand.ResDelta.CritResources, Cand.ResDelta.CritResources,
                TryCand, Cand, ResourceReduce))
      return TryCand.Reason != NoCand;
    if (tryGreater(TryCand.ResDelta.DemandedResources,
                   Cand.ResDelta.DemandedResources,
                   TryCand, Cand, ResourceDemand))
      return TryCand.Reason != NoCand;

    // Avoid serializing long latency dependence chains.
    // For acyclic path limited loops, latency was already checked above.
    if (!RegionPolicy.DisableLatencyHeuristic && TryCand.Policy.ReduceLatency &&
        !Rem.IsAcyclicLatencyLimited && tryLatency(TryCand, Cand, *Zone))
      return TryCand.Reason != NoCand;

    // Fall through to original instruction order.
    if ((Zone->isTop() && TryCand.SU->NodeNum < Cand.SU->NodeNum)
        || (!Zone->isTop() && TryCand.SU->NodeNum > Cand.SU->NodeNum)) {
      TryCand.Reason = NodeOrder;
      return true;
    }
  }

  return false;
}

/// Pick the best candidate from the queue.
///
/// TODO: getMaxPressureDelta results can be mostly cached for each SUnit during
/// DAG building. To adjust for the current scheduling location we need to
/// maintain the number of vreg uses remaining to be top-scheduled.
void GenericScheduler::pickNodeFromQueue(SchedBoundary &Zone,
                                         const CandPolicy &ZonePolicy,
                                         const RegPressureTracker &RPTracker,
                                         SchedCandidate &Cand) {
  // getMaxPressureDelta temporarily modifies the tracker.
  RegPressureTracker &TempTracker = const_cast<RegPressureTracker&>(RPTracker);

  ReadyQueue &Q = Zone.Available;
  for (SUnit *SU : Q) {

    SchedCandidate TryCand(ZonePolicy);
    initCandidate(TryCand, SU, Zone.isTop(), RPTracker, TempTracker);
    // Pass SchedBoundary only when comparing nodes from the same boundary.
    SchedBoundary *ZoneArg = Cand.AtTop == TryCand.AtTop ? &Zone : nullptr;
    if (tryCandidate(Cand, TryCand, ZoneArg)) {
      // Initialize resource delta if needed in case future heuristics query it.
      if (TryCand.ResDelta == SchedResourceDelta())
        TryCand.initResourceDelta(DAG, SchedModel);
      Cand.setBest(TryCand);
      LLVM_DEBUG(traceCandidate(Cand));
    }
  }
}

/// Pick the best candidate node from either the top or bottom queue.
SUnit *GenericScheduler::pickNodeBidirectional(bool &IsTopNode) {
  // Schedule as far as possible in the direction of no choice. This is most
  // efficient, but also provides the best heuristics for CriticalPSets.
  if (SUnit *SU = Bot.pickOnlyChoice()) {
    IsTopNode = false;
    tracePick(Only1, false);
    return SU;
  }
  if (SUnit *SU = Top.pickOnlyChoice()) {
    IsTopNode = true;
    tracePick(Only1, true);
    return SU;
  }
  // Set the bottom-up policy based on the state of the current bottom zone and
  // the instructions outside the zone, including the top zone.
  CandPolicy BotPolicy;
  setPolicy(BotPolicy, /*IsPostRA=*/false, Bot, &Top);
  // Set the top-down policy based on the state of the current top zone and
  // the instructions outside the zone, including the bottom zone.
  CandPolicy TopPolicy;
  setPolicy(TopPolicy, /*IsPostRA=*/false, Top, &Bot);

  // See if BotCand is still valid (because we previously scheduled from Top).
  LLVM_DEBUG(dbgs() << "Picking from Bot:\n");
  if (!BotCand.isValid() || BotCand.SU->isScheduled ||
      BotCand.Policy != BotPolicy) {
    BotCand.reset(CandPolicy());
    pickNodeFromQueue(Bot, BotPolicy, DAG->getBotRPTracker(), BotCand);
    assert(BotCand.Reason != NoCand && "failed to find the first candidate");
  } else {
    LLVM_DEBUG(traceCandidate(BotCand));
#ifndef NDEBUG
    if (VerifyScheduling) {
      SchedCandidate TCand;
      TCand.reset(CandPolicy());
      pickNodeFromQueue(Bot, BotPolicy, DAG->getBotRPTracker(), TCand);
      assert(TCand.SU == BotCand.SU &&
             "Last pick result should correspond to re-picking right now");
    }
#endif
  }

  // Check if the top Q has a better candidate.
  LLVM_DEBUG(dbgs() << "Picking from Top:\n");
  if (!TopCand.isValid() || TopCand.SU->isScheduled ||
      TopCand.Policy != TopPolicy) {
    TopCand.reset(CandPolicy());
    pickNodeFromQueue(Top, TopPolicy, DAG->getTopRPTracker(), TopCand);
    assert(TopCand.Reason != NoCand && "failed to find the first candidate");
  } else {
    LLVM_DEBUG(traceCandidate(TopCand));
#ifndef NDEBUG
    if (VerifyScheduling) {
      SchedCandidate TCand;
      TCand.reset(CandPolicy());
      pickNodeFromQueue(Top, TopPolicy, DAG->getTopRPTracker(), TCand);
      assert(TCand.SU == TopCand.SU &&
           "Last pick result should correspond to re-picking right now");
    }
#endif
  }

  // Pick best from BotCand and TopCand.
  assert(BotCand.isValid());
  assert(TopCand.isValid());
  SchedCandidate Cand = BotCand;
  TopCand.Reason = NoCand;
  if (tryCandidate(Cand, TopCand, nullptr)) {
    Cand.setBest(TopCand);
    LLVM_DEBUG(traceCandidate(Cand));
  }

  IsTopNode = Cand.AtTop;
  tracePick(Cand);
  return Cand.SU;
}

/// Pick the best node to balance the schedule. Implements MachineSchedStrategy.
SUnit *GenericScheduler::pickNode(bool &IsTopNode) {
  if (DAG->top() == DAG->bottom()) {
    assert(Top.Available.empty() && Top.Pending.empty() &&
           Bot.Available.empty() && Bot.Pending.empty() && "ReadyQ garbage");
    return nullptr;
  }
  SUnit *SU;
  do {
    if (RegionPolicy.OnlyTopDown) {
      SU = Top.pickOnlyChoice();
      if (!SU) {
        CandPolicy NoPolicy;
        TopCand.reset(NoPolicy);
        pickNodeFromQueue(Top, NoPolicy, DAG->getTopRPTracker(), TopCand);
        assert(TopCand.Reason != NoCand && "failed to find a candidate");
        tracePick(TopCand);
        SU = TopCand.SU;
      }
      IsTopNode = true;
    } else if (RegionPolicy.OnlyBottomUp) {
      SU = Bot.pickOnlyChoice();
      if (!SU) {
        CandPolicy NoPolicy;
        BotCand.reset(NoPolicy);
        pickNodeFromQueue(Bot, NoPolicy, DAG->getBotRPTracker(), BotCand);
        assert(BotCand.Reason != NoCand && "failed to find a candidate");
        tracePick(BotCand);
        SU = BotCand.SU;
      }
      IsTopNode = false;
    } else {
      SU = pickNodeBidirectional(IsTopNode);
    }
  } while (SU->isScheduled);

  if (SU->isTopReady())
    Top.removeReady(SU);
  if (SU->isBottomReady())
    Bot.removeReady(SU);

  LLVM_DEBUG(dbgs() << "Scheduling SU(" << SU->NodeNum << ") "
                    << *SU->getInstr());
  return SU;
}

void GenericScheduler::reschedulePhysReg(SUnit *SU, bool isTop) {
  MachineBasicBlock::iterator InsertPos = SU->getInstr();
  if (!isTop)
    ++InsertPos;
  SmallVectorImpl<SDep> &Deps = isTop ? SU->Preds : SU->Succs;

  // Find already scheduled copies with a single physreg dependence and move
  // them just above the scheduled instruction.
  for (SDep &Dep : Deps) {
    if (Dep.getKind() != SDep::Data ||
        !Register::isPhysicalRegister(Dep.getReg()))
      continue;
    SUnit *DepSU = Dep.getSUnit();
    if (isTop ? DepSU->Succs.size() > 1 : DepSU->Preds.size() > 1)
      continue;
    MachineInstr *Copy = DepSU->getInstr();
    if (!Copy->isCopy() && !Copy->isMoveImmediate())
      continue;
    LLVM_DEBUG(dbgs() << "  Rescheduling physreg copy ";
               DAG->dumpNode(*Dep.getSUnit()));
    DAG->moveInstruction(Copy, InsertPos);
  }
}

/// Update the scheduler's state after scheduling a node. This is the same node
/// that was just returned by pickNode(). However, ScheduleDAGMILive needs to
/// update it's state based on the current cycle before MachineSchedStrategy
/// does.
///
/// FIXME: Eventually, we may bundle physreg copies rather than rescheduling
/// them here. See comments in biasPhysReg.
void GenericScheduler::schedNode(SUnit *SU, bool IsTopNode) {
  if (IsTopNode) {
    SU->TopReadyCycle = std::max(SU->TopReadyCycle, Top.getCurrCycle());
    Top.bumpNode(SU);
    if (SU->hasPhysRegUses)
      reschedulePhysReg(SU, true);
  } else {
    SU->BotReadyCycle = std::max(SU->BotReadyCycle, Bot.getCurrCycle());
    Bot.bumpNode(SU);
    if (SU->hasPhysRegDefs)
      reschedulePhysReg(SU, false);
  }
}

/// Create the standard converging machine scheduler. This will be used as the
/// default scheduler if the target does not set a default.
ScheduleDAGMILive *llvm::createGenericSchedLive(MachineSchedContext *C) {
  ScheduleDAGMILive *DAG =
      new ScheduleDAGMILive(C, std::make_unique<GenericScheduler>(C));
  // Register DAG post-processors.
  //
  // FIXME: extend the mutation API to allow earlier mutations to instantiate
  // data and pass it to later mutations. Have a single mutation that gathers
  // the interesting nodes in one pass.
  DAG->addMutation(createCopyConstrainDAGMutation(DAG->TII, DAG->TRI));
  return DAG;
}

static ScheduleDAGInstrs *createConvergingSched(MachineSchedContext *C) {
  return createGenericSchedLive(C);
}

static MachineSchedRegistry
GenericSchedRegistry("converge", "Standard converging scheduler.",
                     createConvergingSched);

//===----------------------------------------------------------------------===//
// PostGenericScheduler - Generic PostRA implementation of MachineSchedStrategy.
//===----------------------------------------------------------------------===//

void PostGenericScheduler::initialize(ScheduleDAGMI *Dag) {
  DAG = Dag;
  SchedModel = DAG->getSchedModel();
  TRI = DAG->TRI;

  Rem.init(DAG, SchedModel);
  Top.init(DAG, SchedModel, &Rem);
  BotRoots.clear();

  // Initialize the HazardRecognizers. If itineraries don't exist, are empty,
  // or are disabled, then these HazardRecs will be disabled.
  const InstrItineraryData *Itin = SchedModel->getInstrItineraries();
  if (!Top.HazardRec) {
    Top.HazardRec =
        DAG->MF.getSubtarget().getInstrInfo()->CreateTargetMIHazardRecognizer(
            Itin, DAG);
  }
}

void PostGenericScheduler::registerRoots() {
  Rem.CriticalPath = DAG->ExitSU.getDepth();

  // Some roots may not feed into ExitSU. Check all of them in case.
  for (const SUnit *SU : BotRoots) {
    if (SU->getDepth() > Rem.CriticalPath)
      Rem.CriticalPath = SU->getDepth();
  }
  LLVM_DEBUG(dbgs() << "Critical Path: (PGS-RR) " << Rem.CriticalPath << '\n');
  if (DumpCriticalPathLength) {
    errs() << "Critical Path(PGS-RR ): " << Rem.CriticalPath << " \n";
  }
}

/// Apply a set of heuristics to a new candidate for PostRA scheduling.
///
/// \param Cand provides the policy and current best candidate.
/// \param TryCand refers to the next SUnit candidate, otherwise uninitialized.
/// \return \c true if TryCand is better than Cand (Reason is NOT NoCand)
bool PostGenericScheduler::tryCandidate(SchedCandidate &Cand,
                                        SchedCandidate &TryCand) {
  // Initialize the candidate if needed.
  if (!Cand.isValid()) {
    TryCand.Reason = NodeOrder;
    return true;
  }

  // Prioritize instructions that read unbuffered resources by stall cycles.
  if (tryLess(Top.getLatencyStallCycles(TryCand.SU),
              Top.getLatencyStallCycles(Cand.SU), TryCand, Cand, Stall))
    return TryCand.Reason != NoCand;

  // Keep clustered nodes together.
  if (tryGreater(TryCand.SU == DAG->getNextClusterSucc(),
                 Cand.SU == DAG->getNextClusterSucc(),
                 TryCand, Cand, Cluster))
    return TryCand.Reason != NoCand;

  // Avoid critical resource consumption and balance the schedule.
  if (tryLess(TryCand.ResDelta.CritResources, Cand.ResDelta.CritResources,
              TryCand, Cand, ResourceReduce))
    return TryCand.Reason != NoCand;
  if (tryGreater(TryCand.ResDelta.DemandedResources,
                 Cand.ResDelta.DemandedResources,
                 TryCand, Cand, ResourceDemand))
    return TryCand.Reason != NoCand;

  // Avoid serializing long latency dependence chains.
  if (Cand.Policy.ReduceLatency && tryLatency(TryCand, Cand, Top)) {
    return TryCand.Reason != NoCand;
  }

  // Fall through to original instruction order.
  if (TryCand.SU->NodeNum < Cand.SU->NodeNum) {
    TryCand.Reason = NodeOrder;
    return true;
  }

  return false;
}

void PostGenericScheduler::pickNodeFromQueue(SchedCandidate &Cand) {
  ReadyQueue &Q = Top.Available;
  for (SUnit *SU : Q) {
    SchedCandidate TryCand(Cand.Policy);
    TryCand.SU = SU;
    TryCand.AtTop = true;
    TryCand.initResourceDelta(DAG, SchedModel);
    if (tryCandidate(Cand, TryCand)) {
      Cand.setBest(TryCand);
      LLVM_DEBUG(traceCandidate(Cand));
    }
  }
}

/// Pick the next node to schedule.
SUnit *PostGenericScheduler::pickNode(bool &IsTopNode) {
  if (DAG->top() == DAG->bottom()) {
    assert(Top.Available.empty() && Top.Pending.empty() && "ReadyQ garbage");
    return nullptr;
  }
  SUnit *SU;
  do {
    SU = Top.pickOnlyChoice();
    if (SU) {
      tracePick(Only1, true);
    } else {
      CandPolicy NoPolicy;
      SchedCandidate TopCand(NoPolicy);
      // Set the top-down policy based on the state of the current top zone and
      // the instructions outside the zone, including the bottom zone.
      setPolicy(TopCand.Policy, /*IsPostRA=*/true, Top, nullptr);
      pickNodeFromQueue(TopCand);
      assert(TopCand.Reason != NoCand && "failed to find a candidate");
      tracePick(TopCand);
      SU = TopCand.SU;
    }
  } while (SU->isScheduled);

  IsTopNode = true;
  Top.removeReady(SU);

  LLVM_DEBUG(dbgs() << "Scheduling SU(" << SU->NodeNum << ") "
                    << *SU->getInstr());
  return SU;
}

/// Called after ScheduleDAGMI has scheduled an instruction and updated
/// scheduled/remaining flags in the DAG nodes.
void PostGenericScheduler::schedNode(SUnit *SU, bool IsTopNode) {
  SU->TopReadyCycle = std::max(SU->TopReadyCycle, Top.getCurrCycle());
  Top.bumpNode(SU);
}

ScheduleDAGMI *llvm::createGenericSchedPostRA(MachineSchedContext *C) {
  return new ScheduleDAGMI(C, std::make_unique<PostGenericScheduler>(C),
                           /*RemoveKillFlags=*/true);
}

//===----------------------------------------------------------------------===//
// ILP Scheduler. Currently for experimental analysis of heuristics.
//===----------------------------------------------------------------------===//

namespace {

/// Order nodes by the ILP metric.
struct ILPOrder {
  const SchedDFSResult *DFSResult = nullptr;
  const BitVector *ScheduledTrees = nullptr;
  bool MaximizeILP;

  ILPOrder(bool MaxILP) : MaximizeILP(MaxILP) {}

  /// Apply a less-than relation on node priority.
  ///
  /// (Return true if A comes after B in the Q.)
  bool operator()(const SUnit *A, const SUnit *B) const {
    unsigned SchedTreeA = DFSResult->getSubtreeID(A);
    unsigned SchedTreeB = DFSResult->getSubtreeID(B);
    if (SchedTreeA != SchedTreeB) {
      // Unscheduled trees have lower priority.
      if (ScheduledTrees->test(SchedTreeA) != ScheduledTrees->test(SchedTreeB))
        return ScheduledTrees->test(SchedTreeB);

      // Trees with shallower connections have have lower priority.
      if (DFSResult->getSubtreeLevel(SchedTreeA)
          != DFSResult->getSubtreeLevel(SchedTreeB)) {
        return DFSResult->getSubtreeLevel(SchedTreeA)
          < DFSResult->getSubtreeLevel(SchedTreeB);
      }
    }
    if (MaximizeILP)
      return DFSResult->getILP(A) < DFSResult->getILP(B);
    else
      return DFSResult->getILP(A) > DFSResult->getILP(B);
  }
};

/// Schedule based on the ILP metric.
class ILPScheduler : public MachineSchedStrategy {
  ScheduleDAGMILive *DAG = nullptr;
  ILPOrder Cmp;

  std::vector<SUnit*> ReadyQ;

public:
  ILPScheduler(bool MaximizeILP) : Cmp(MaximizeILP) {}

  void initialize(ScheduleDAGMI *dag) override {
    assert(dag->hasVRegLiveness() && "ILPScheduler needs vreg liveness");
    DAG = static_cast<ScheduleDAGMILive*>(dag);
    DAG->computeDFSResult();
    Cmp.DFSResult = DAG->getDFSResult();
    Cmp.ScheduledTrees = &DAG->getScheduledTrees();
    ReadyQ.clear();
  }

  void registerRoots() override {
    // Restore the heap in ReadyQ with the updated DFS results.
    std::make_heap(ReadyQ.begin(), ReadyQ.end(), Cmp);
  }

  /// Implement MachineSchedStrategy interface.
  /// -----------------------------------------

  /// Callback to select the highest priority node from the ready Q.
  SUnit *pickNode(bool &IsTopNode) override {
    if (ReadyQ.empty()) return nullptr;
    std::pop_heap(ReadyQ.begin(), ReadyQ.end(), Cmp);
    SUnit *SU = ReadyQ.back();
    ReadyQ.pop_back();
    IsTopNode = false;
    LLVM_DEBUG(dbgs() << "Pick node "
                      << "SU(" << SU->NodeNum << ") "
                      << " ILP: " << DAG->getDFSResult()->getILP(SU)
                      << " Tree: " << DAG->getDFSResult()->getSubtreeID(SU)
                      << " @"
                      << DAG->getDFSResult()->getSubtreeLevel(
                             DAG->getDFSResult()->getSubtreeID(SU))
                      << '\n'
                      << "Scheduling " << *SU->getInstr());
    return SU;
  }

  /// Scheduler callback to notify that a new subtree is scheduled.
  void scheduleTree(unsigned SubtreeID) override {
    std::make_heap(ReadyQ.begin(), ReadyQ.end(), Cmp);
  }

  /// Callback after a node is scheduled. Mark a newly scheduled tree, notify
  /// DFSResults, and resort the priority Q.
  void schedNode(SUnit *SU, bool IsTopNode) override {
    assert(!IsTopNode && "SchedDFSResult needs bottom-up");
  }

  void releaseTopNode(SUnit *) override { /*only called for top roots*/ }

  void releaseBottomNode(SUnit *SU) override {
    ReadyQ.push_back(SU);
    std::push_heap(ReadyQ.begin(), ReadyQ.end(), Cmp);
  }
};

} // end anonymous namespace

static ScheduleDAGInstrs *createILPMaxScheduler(MachineSchedContext *C) {
  return new ScheduleDAGMILive(C, std::make_unique<ILPScheduler>(true));
}
static ScheduleDAGInstrs *createILPMinScheduler(MachineSchedContext *C) {
  return new ScheduleDAGMILive(C, std::make_unique<ILPScheduler>(false));
}

static MachineSchedRegistry ILPMaxRegistry(
  "ilpmax", "Schedule bottom-up for max ILP", createILPMaxScheduler);
static MachineSchedRegistry ILPMinRegistry(
  "ilpmin", "Schedule bottom-up for min ILP", createILPMinScheduler);

//===----------------------------------------------------------------------===//
// Machine Instruction Shuffler for Correctness Testing
//===----------------------------------------------------------------------===//

#ifndef NDEBUG
namespace {

/// Apply a less-than relation on the node order, which corresponds to the
/// instruction order prior to scheduling. IsReverse implements greater-than.
template<bool IsReverse>
struct SUnitOrder {
  bool operator()(SUnit *A, SUnit *B) const {
    if (IsReverse)
      return A->NodeNum > B->NodeNum;
    else
      return A->NodeNum < B->NodeNum;
  }
};

/// Reorder instructions as much as possible.
class InstructionShuffler : public MachineSchedStrategy {
  bool IsAlternating;
  bool IsTopDown;

  // Using a less-than relation (SUnitOrder<false>) for the TopQ priority
  // gives nodes with a higher number higher priority causing the latest
  // instructions to be scheduled first.
  PriorityQueue<SUnit*, std::vector<SUnit*>, SUnitOrder<false>>
    TopQ;

  // When scheduling bottom-up, use greater-than as the queue priority.
  PriorityQueue<SUnit*, std::vector<SUnit*>, SUnitOrder<true>>
    BottomQ;

public:
  InstructionShuffler(bool alternate, bool topdown)
    : IsAlternating(alternate), IsTopDown(topdown) {}

  void initialize(ScheduleDAGMI*) override {
    TopQ.clear();
    BottomQ.clear();
  }

  /// Implement MachineSchedStrategy interface.
  /// -----------------------------------------

  SUnit *pickNode(bool &IsTopNode) override {
    SUnit *SU;
    if (IsTopDown) {
      do {
        if (TopQ.empty()) return nullptr;
        SU = TopQ.top();
        TopQ.pop();
      } while (SU->isScheduled);
      IsTopNode = true;
    } else {
      do {
        if (BottomQ.empty()) return nullptr;
        SU = BottomQ.top();
        BottomQ.pop();
      } while (SU->isScheduled);
      IsTopNode = false;
    }
    if (IsAlternating)
      IsTopDown = !IsTopDown;
    return SU;
  }

  void schedNode(SUnit *SU, bool IsTopNode) override {}

  void releaseTopNode(SUnit *SU) override {
    TopQ.push(SU);
  }
  void releaseBottomNode(SUnit *SU) override {
    BottomQ.push(SU);
  }
};

} // end anonymous namespace

static ScheduleDAGInstrs *createInstructionShuffler(MachineSchedContext *C) {
  bool Alternate = !ForceTopDown && !ForceBottomUp;
  bool TopDown = !ForceBottomUp;
  assert((TopDown || !ForceTopDown) &&
         "-misched-topdown incompatible with -misched-bottomup");
  return new ScheduleDAGMILive(
      C, std::make_unique<InstructionShuffler>(Alternate, TopDown));
}

static MachineSchedRegistry ShufflerRegistry(
  "shuffle", "Shuffle machine instructions alternating directions",
  createInstructionShuffler);
#endif // !NDEBUG

//===----------------------------------------------------------------------===//
// GraphWriter support for ScheduleDAGMILive.
//===----------------------------------------------------------------------===//

#ifndef NDEBUG
namespace llvm {

template<> struct GraphTraits<
  ScheduleDAGMI*> : public GraphTraits<ScheduleDAG*> {};

template<>
struct DOTGraphTraits<ScheduleDAGMI*> : public DefaultDOTGraphTraits {
  DOTGraphTraits(bool isSimple = false) : DefaultDOTGraphTraits(isSimple) {}

  static std::string getGraphName(const ScheduleDAG *G) {
    return std::string(G->MF.getName());
  }

  static bool renderGraphFromBottomUp() {
    return true;
  }

  static bool isNodeHidden(const SUnit *Node, const ScheduleDAG *G) {
    if (ViewMISchedCutoff == 0)
      return false;
    return (Node->Preds.size() > ViewMISchedCutoff
         || Node->Succs.size() > ViewMISchedCutoff);
  }

  /// If you want to override the dot attributes printed for a particular
  /// edge, override this method.
  static std::string getEdgeAttributes(const SUnit *Node,
                                       SUnitIterator EI,
                                       const ScheduleDAG *Graph) {
    if (EI.isArtificialDep())
      return "color=cyan,style=dashed";
    if (EI.isCtrlDep())
      return "color=blue,style=dashed";
    return "";
  }

  static std::string getNodeLabel(const SUnit *SU, const ScheduleDAG *G) {
    std::string Str;
    raw_string_ostream SS(Str);
    const ScheduleDAGMI *DAG = static_cast<const ScheduleDAGMI*>(G);
    const SchedDFSResult *DFS = DAG->hasVRegLiveness() ?
      static_cast<const ScheduleDAGMILive*>(G)->getDFSResult() : nullptr;
    SS << "SU:" << SU->NodeNum;
    if (DFS)
      SS << " I:" << DFS->getNumInstrs(SU);
    return SS.str();
  }

  static std::string getNodeDescription(const SUnit *SU, const ScheduleDAG *G) {
    return G->getGraphNodeLabel(SU);
  }

  static std::string getNodeAttributes(const SUnit *N, const ScheduleDAG *G) {
    std::string Str("shape=Mrecord");
    const ScheduleDAGMI *DAG = static_cast<const ScheduleDAGMI*>(G);
    const SchedDFSResult *DFS = DAG->hasVRegLiveness() ?
      static_cast<const ScheduleDAGMILive*>(G)->getDFSResult() : nullptr;
    if (DFS) {
      Str += ",style=filled,fillcolor=\"#";
      Str += DOT::getColorString(DFS->getSubtreeID(N));
      Str += '"';
    }
    return Str;
  }
};

} // end namespace llvm
#endif // NDEBUG

/// viewGraph - Pop up a ghostview window with the reachable parts of the DAG
/// rendered using 'dot'.
void ScheduleDAGMI::viewGraph(const Twine &Name, const Twine &Title) {
#ifndef NDEBUG
  ViewGraph(this, Name, false, Title);
#else
  errs() << "ScheduleDAGMI::viewGraph is only available in debug builds on "
         << "systems with Graphviz or gv!\n";
#endif  // NDEBUG
}

/// Out-of-line implementation with no arguments is handy for gdb.
void ScheduleDAGMI::viewGraph() {
  viewGraph(getDAGName(), "Scheduling-Units Graph for " + getDAGName());
}