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
|
//===- VPlan.h - Represent A Vectorizer Plan --------------------*- C++ -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
/// \file
/// This file contains the declarations of the Vectorization Plan base classes:
/// 1. VPBasicBlock and VPRegionBlock that inherit from a common pure virtual
/// VPBlockBase, together implementing a Hierarchical CFG;
/// 2. Pure virtual VPRecipeBase serving as the base class for recipes contained
/// within VPBasicBlocks;
/// 3. Pure virtual VPSingleDefRecipe serving as a base class for recipes that
/// also inherit from VPValue.
/// 4. VPInstruction, a concrete Recipe and VPUser modeling a single planned
/// instruction;
/// 5. The VPlan class holding a candidate for vectorization;
/// 6. The VPlanPrinter class providing a way to print a plan in dot format;
/// These are documented in docs/VectorizationPlan.rst.
//
//===----------------------------------------------------------------------===//
#ifndef LLVM_TRANSFORMS_VECTORIZE_VPLAN_H
#define LLVM_TRANSFORMS_VECTORIZE_VPLAN_H
#include "VPlanAnalysis.h"
#include "VPlanValue.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/MapVector.h"
#include "llvm/ADT/SmallBitVector.h"
#include "llvm/ADT/SmallPtrSet.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Twine.h"
#include "llvm/ADT/ilist.h"
#include "llvm/ADT/ilist_node.h"
#include "llvm/Analysis/IVDescriptors.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/VectorUtils.h"
#include "llvm/IR/DebugLoc.h"
#include "llvm/IR/FMF.h"
#include "llvm/IR/Operator.h"
#include <algorithm>
#include <cassert>
#include <cstddef>
#include <string>
namespace llvm {
class BasicBlock;
class DominatorTree;
class InnerLoopVectorizer;
class IRBuilderBase;
class LoopInfo;
class raw_ostream;
class RecurrenceDescriptor;
class SCEV;
class Type;
class VPBasicBlock;
class VPRegionBlock;
class VPlan;
class VPReplicateRecipe;
class VPlanSlp;
class Value;
class LoopVersioning;
namespace Intrinsic {
typedef unsigned ID;
}
/// Returns a calculation for the total number of elements for a given \p VF.
/// For fixed width vectors this value is a constant, whereas for scalable
/// vectors it is an expression determined at runtime.
Value *getRuntimeVF(IRBuilderBase &B, Type *Ty, ElementCount VF);
/// Return a value for Step multiplied by VF.
Value *createStepForVF(IRBuilderBase &B, Type *Ty, ElementCount VF,
int64_t Step);
const SCEV *createTripCountSCEV(Type *IdxTy, PredicatedScalarEvolution &PSE,
Loop *CurLoop = nullptr);
/// A range of powers-of-2 vectorization factors with fixed start and
/// adjustable end. The range includes start and excludes end, e.g.,:
/// [1, 16) = {1, 2, 4, 8}
struct VFRange {
// A power of 2.
const ElementCount Start;
// A power of 2. If End <= Start range is empty.
ElementCount End;
bool isEmpty() const {
return End.getKnownMinValue() <= Start.getKnownMinValue();
}
VFRange(const ElementCount &Start, const ElementCount &End)
: Start(Start), End(End) {
assert(Start.isScalable() == End.isScalable() &&
"Both Start and End should have the same scalable flag");
assert(isPowerOf2_32(Start.getKnownMinValue()) &&
"Expected Start to be a power of 2");
assert(isPowerOf2_32(End.getKnownMinValue()) &&
"Expected End to be a power of 2");
}
/// Iterator to iterate over vectorization factors in a VFRange.
class iterator
: public iterator_facade_base<iterator, std::forward_iterator_tag,
ElementCount> {
ElementCount VF;
public:
iterator(ElementCount VF) : VF(VF) {}
bool operator==(const iterator &Other) const { return VF == Other.VF; }
ElementCount operator*() const { return VF; }
iterator &operator++() {
VF *= 2;
return *this;
}
};
iterator begin() { return iterator(Start); }
iterator end() {
assert(isPowerOf2_32(End.getKnownMinValue()));
return iterator(End);
}
};
using VPlanPtr = std::unique_ptr<VPlan>;
/// In what follows, the term "input IR" refers to code that is fed into the
/// vectorizer whereas the term "output IR" refers to code that is generated by
/// the vectorizer.
/// VPLane provides a way to access lanes in both fixed width and scalable
/// vectors, where for the latter the lane index sometimes needs calculating
/// as a runtime expression.
class VPLane {
public:
/// Kind describes how to interpret Lane.
enum class Kind : uint8_t {
/// For First, Lane is the index into the first N elements of a
/// fixed-vector <N x <ElTy>> or a scalable vector <vscale x N x <ElTy>>.
First,
/// For ScalableLast, Lane is the offset from the start of the last
/// N-element subvector in a scalable vector <vscale x N x <ElTy>>. For
/// example, a Lane of 0 corresponds to lane `(vscale - 1) * N`, a Lane of
/// 1 corresponds to `((vscale - 1) * N) + 1`, etc.
ScalableLast
};
private:
/// in [0..VF)
unsigned Lane;
/// Indicates how the Lane should be interpreted, as described above.
Kind LaneKind;
public:
VPLane(unsigned Lane, Kind LaneKind) : Lane(Lane), LaneKind(LaneKind) {}
static VPLane getFirstLane() { return VPLane(0, VPLane::Kind::First); }
static VPLane getLastLaneForVF(const ElementCount &VF) {
unsigned LaneOffset = VF.getKnownMinValue() - 1;
Kind LaneKind;
if (VF.isScalable())
// In this case 'LaneOffset' refers to the offset from the start of the
// last subvector with VF.getKnownMinValue() elements.
LaneKind = VPLane::Kind::ScalableLast;
else
LaneKind = VPLane::Kind::First;
return VPLane(LaneOffset, LaneKind);
}
/// Returns a compile-time known value for the lane index and asserts if the
/// lane can only be calculated at runtime.
unsigned getKnownLane() const {
assert(LaneKind == Kind::First);
return Lane;
}
/// Returns an expression describing the lane index that can be used at
/// runtime.
Value *getAsRuntimeExpr(IRBuilderBase &Builder, const ElementCount &VF) const;
/// Returns the Kind of lane offset.
Kind getKind() const { return LaneKind; }
/// Returns true if this is the first lane of the whole vector.
bool isFirstLane() const { return Lane == 0 && LaneKind == Kind::First; }
/// Maps the lane to a cache index based on \p VF.
unsigned mapToCacheIndex(const ElementCount &VF) const {
switch (LaneKind) {
case VPLane::Kind::ScalableLast:
assert(VF.isScalable() && Lane < VF.getKnownMinValue());
return VF.getKnownMinValue() + Lane;
default:
assert(Lane < VF.getKnownMinValue());
return Lane;
}
}
/// Returns the maxmimum number of lanes that we are able to consider
/// caching for \p VF.
static unsigned getNumCachedLanes(const ElementCount &VF) {
return VF.getKnownMinValue() * (VF.isScalable() ? 2 : 1);
}
};
/// VPIteration represents a single point in the iteration space of the output
/// (vectorized and/or unrolled) IR loop.
struct VPIteration {
/// in [0..UF)
unsigned Part;
VPLane Lane;
VPIteration(unsigned Part, unsigned Lane,
VPLane::Kind Kind = VPLane::Kind::First)
: Part(Part), Lane(Lane, Kind) {}
VPIteration(unsigned Part, const VPLane &Lane) : Part(Part), Lane(Lane) {}
bool isFirstIteration() const { return Part == 0 && Lane.isFirstLane(); }
};
/// VPTransformState holds information passed down when "executing" a VPlan,
/// needed for generating the output IR.
struct VPTransformState {
VPTransformState(ElementCount VF, unsigned UF, LoopInfo *LI,
DominatorTree *DT, IRBuilderBase &Builder,
InnerLoopVectorizer *ILV, VPlan *Plan, LLVMContext &Ctx);
/// The chosen Vectorization and Unroll Factors of the loop being vectorized.
ElementCount VF;
unsigned UF;
/// Hold the indices to generate specific scalar instructions. Null indicates
/// that all instances are to be generated, using either scalar or vector
/// instructions.
std::optional<VPIteration> Instance;
struct DataState {
/// A type for vectorized values in the new loop. Each value from the
/// original loop, when vectorized, is represented by UF vector values in
/// the new unrolled loop, where UF is the unroll factor.
typedef SmallVector<Value *, 2> PerPartValuesTy;
DenseMap<VPValue *, PerPartValuesTy> PerPartOutput;
using ScalarsPerPartValuesTy = SmallVector<SmallVector<Value *, 4>, 2>;
DenseMap<VPValue *, ScalarsPerPartValuesTy> PerPartScalars;
} Data;
/// Get the generated Value for the given VPValue \p Def and the given \p Part.
/// \see set.
Value *get(VPValue *Def, unsigned Part);
/// Get the generated Value for a given VPValue and given Part and Lane.
Value *get(VPValue *Def, const VPIteration &Instance);
bool hasVectorValue(VPValue *Def, unsigned Part) {
auto I = Data.PerPartOutput.find(Def);
return I != Data.PerPartOutput.end() && Part < I->second.size() &&
I->second[Part];
}
bool hasScalarValue(VPValue *Def, VPIteration Instance) {
auto I = Data.PerPartScalars.find(Def);
if (I == Data.PerPartScalars.end())
return false;
unsigned CacheIdx = Instance.Lane.mapToCacheIndex(VF);
return Instance.Part < I->second.size() &&
CacheIdx < I->second[Instance.Part].size() &&
I->second[Instance.Part][CacheIdx];
}
/// Set the generated Value for a given VPValue and a given Part.
void set(VPValue *Def, Value *V, unsigned Part) {
if (!Data.PerPartOutput.count(Def)) {
DataState::PerPartValuesTy Entry(UF);
Data.PerPartOutput[Def] = Entry;
}
Data.PerPartOutput[Def][Part] = V;
}
/// Reset an existing vector value for \p Def and a given \p Part.
void reset(VPValue *Def, Value *V, unsigned Part) {
auto Iter = Data.PerPartOutput.find(Def);
assert(Iter != Data.PerPartOutput.end() &&
"need to overwrite existing value");
Iter->second[Part] = V;
}
/// Set the generated scalar \p V for \p Def and the given \p Instance.
void set(VPValue *Def, Value *V, const VPIteration &Instance) {
auto Iter = Data.PerPartScalars.insert({Def, {}});
auto &PerPartVec = Iter.first->second;
while (PerPartVec.size() <= Instance.Part)
PerPartVec.emplace_back();
auto &Scalars = PerPartVec[Instance.Part];
unsigned CacheIdx = Instance.Lane.mapToCacheIndex(VF);
while (Scalars.size() <= CacheIdx)
Scalars.push_back(nullptr);
assert(!Scalars[CacheIdx] && "should overwrite existing value");
Scalars[CacheIdx] = V;
}
/// Reset an existing scalar value for \p Def and a given \p Instance.
void reset(VPValue *Def, Value *V, const VPIteration &Instance) {
auto Iter = Data.PerPartScalars.find(Def);
assert(Iter != Data.PerPartScalars.end() &&
"need to overwrite existing value");
assert(Instance.Part < Iter->second.size() &&
"need to overwrite existing value");
unsigned CacheIdx = Instance.Lane.mapToCacheIndex(VF);
assert(CacheIdx < Iter->second[Instance.Part].size() &&
"need to overwrite existing value");
Iter->second[Instance.Part][CacheIdx] = V;
}
/// Add additional metadata to \p To that was not present on \p Orig.
///
/// Currently this is used to add the noalias annotations based on the
/// inserted memchecks. Use this for instructions that are *cloned* into the
/// vector loop.
void addNewMetadata(Instruction *To, const Instruction *Orig);
/// Add metadata from one instruction to another.
///
/// This includes both the original MDs from \p From and additional ones (\see
/// addNewMetadata). Use this for *newly created* instructions in the vector
/// loop.
void addMetadata(Instruction *To, Instruction *From);
/// Similar to the previous function but it adds the metadata to a
/// vector of instructions.
void addMetadata(ArrayRef<Value *> To, Instruction *From);
/// Set the debug location in the builder using the debug location \p DL.
void setDebugLocFrom(DebugLoc DL);
/// Construct the vector value of a scalarized value \p V one lane at a time.
void packScalarIntoVectorValue(VPValue *Def, const VPIteration &Instance);
/// Hold state information used when constructing the CFG of the output IR,
/// traversing the VPBasicBlocks and generating corresponding IR BasicBlocks.
struct CFGState {
/// The previous VPBasicBlock visited. Initially set to null.
VPBasicBlock *PrevVPBB = nullptr;
/// The previous IR BasicBlock created or used. Initially set to the new
/// header BasicBlock.
BasicBlock *PrevBB = nullptr;
/// The last IR BasicBlock in the output IR. Set to the exit block of the
/// vector loop.
BasicBlock *ExitBB = nullptr;
/// A mapping of each VPBasicBlock to the corresponding BasicBlock. In case
/// of replication, maps the BasicBlock of the last replica created.
SmallDenseMap<VPBasicBlock *, BasicBlock *> VPBB2IRBB;
CFGState() = default;
/// Returns the BasicBlock* mapped to the pre-header of the loop region
/// containing \p R.
BasicBlock *getPreheaderBBFor(VPRecipeBase *R);
} CFG;
/// Hold a pointer to LoopInfo to register new basic blocks in the loop.
LoopInfo *LI;
/// Hold a pointer to Dominator Tree to register new basic blocks in the loop.
DominatorTree *DT;
/// Hold a reference to the IRBuilder used to generate output IR code.
IRBuilderBase &Builder;
VPValue2ValueTy VPValue2Value;
/// Hold the canonical scalar IV of the vector loop (start=0, step=VF*UF).
Value *CanonicalIV = nullptr;
/// Hold a pointer to InnerLoopVectorizer to reuse its IR generation methods.
InnerLoopVectorizer *ILV;
/// Pointer to the VPlan code is generated for.
VPlan *Plan;
/// The loop object for the current parent region, or nullptr.
Loop *CurrentVectorLoop = nullptr;
/// LoopVersioning. It's only set up (non-null) if memchecks were
/// used.
///
/// This is currently only used to add no-alias metadata based on the
/// memchecks. The actually versioning is performed manually.
LoopVersioning *LVer = nullptr;
/// Map SCEVs to their expanded values. Populated when executing
/// VPExpandSCEVRecipes.
DenseMap<const SCEV *, Value *> ExpandedSCEVs;
/// VPlan-based type analysis.
VPTypeAnalysis TypeAnalysis;
};
/// VPBlockBase is the building block of the Hierarchical Control-Flow Graph.
/// A VPBlockBase can be either a VPBasicBlock or a VPRegionBlock.
class VPBlockBase {
friend class VPBlockUtils;
const unsigned char SubclassID; ///< Subclass identifier (for isa/dyn_cast).
/// An optional name for the block.
std::string Name;
/// The immediate VPRegionBlock which this VPBlockBase belongs to, or null if
/// it is a topmost VPBlockBase.
VPRegionBlock *Parent = nullptr;
/// List of predecessor blocks.
SmallVector<VPBlockBase *, 1> Predecessors;
/// List of successor blocks.
SmallVector<VPBlockBase *, 1> Successors;
/// VPlan containing the block. Can only be set on the entry block of the
/// plan.
VPlan *Plan = nullptr;
/// Add \p Successor as the last successor to this block.
void appendSuccessor(VPBlockBase *Successor) {
assert(Successor && "Cannot add nullptr successor!");
Successors.push_back(Successor);
}
/// Add \p Predecessor as the last predecessor to this block.
void appendPredecessor(VPBlockBase *Predecessor) {
assert(Predecessor && "Cannot add nullptr predecessor!");
Predecessors.push_back(Predecessor);
}
/// Remove \p Predecessor from the predecessors of this block.
void removePredecessor(VPBlockBase *Predecessor) {
auto Pos = find(Predecessors, Predecessor);
assert(Pos && "Predecessor does not exist");
Predecessors.erase(Pos);
}
/// Remove \p Successor from the successors of this block.
void removeSuccessor(VPBlockBase *Successor) {
auto Pos = find(Successors, Successor);
assert(Pos && "Successor does not exist");
Successors.erase(Pos);
}
protected:
VPBlockBase(const unsigned char SC, const std::string &N)
: SubclassID(SC), Name(N) {}
public:
/// An enumeration for keeping track of the concrete subclass of VPBlockBase
/// that are actually instantiated. Values of this enumeration are kept in the
/// SubclassID field of the VPBlockBase objects. They are used for concrete
/// type identification.
using VPBlockTy = enum { VPBasicBlockSC, VPRegionBlockSC };
using VPBlocksTy = SmallVectorImpl<VPBlockBase *>;
virtual ~VPBlockBase() = default;
const std::string &getName() const { return Name; }
void setName(const Twine &newName) { Name = newName.str(); }
/// \return an ID for the concrete type of this object.
/// This is used to implement the classof checks. This should not be used
/// for any other purpose, as the values may change as LLVM evolves.
unsigned getVPBlockID() const { return SubclassID; }
VPRegionBlock *getParent() { return Parent; }
const VPRegionBlock *getParent() const { return Parent; }
/// \return A pointer to the plan containing the current block.
VPlan *getPlan();
const VPlan *getPlan() const;
/// Sets the pointer of the plan containing the block. The block must be the
/// entry block into the VPlan.
void setPlan(VPlan *ParentPlan);
void setParent(VPRegionBlock *P) { Parent = P; }
/// \return the VPBasicBlock that is the entry of this VPBlockBase,
/// recursively, if the latter is a VPRegionBlock. Otherwise, if this
/// VPBlockBase is a VPBasicBlock, it is returned.
const VPBasicBlock *getEntryBasicBlock() const;
VPBasicBlock *getEntryBasicBlock();
/// \return the VPBasicBlock that is the exiting this VPBlockBase,
/// recursively, if the latter is a VPRegionBlock. Otherwise, if this
/// VPBlockBase is a VPBasicBlock, it is returned.
const VPBasicBlock *getExitingBasicBlock() const;
VPBasicBlock *getExitingBasicBlock();
const VPBlocksTy &getSuccessors() const { return Successors; }
VPBlocksTy &getSuccessors() { return Successors; }
iterator_range<VPBlockBase **> successors() { return Successors; }
const VPBlocksTy &getPredecessors() const { return Predecessors; }
VPBlocksTy &getPredecessors() { return Predecessors; }
/// \return the successor of this VPBlockBase if it has a single successor.
/// Otherwise return a null pointer.
VPBlockBase *getSingleSuccessor() const {
return (Successors.size() == 1 ? *Successors.begin() : nullptr);
}
/// \return the predecessor of this VPBlockBase if it has a single
/// predecessor. Otherwise return a null pointer.
VPBlockBase *getSinglePredecessor() const {
return (Predecessors.size() == 1 ? *Predecessors.begin() : nullptr);
}
size_t getNumSuccessors() const { return Successors.size(); }
size_t getNumPredecessors() const { return Predecessors.size(); }
/// An Enclosing Block of a block B is any block containing B, including B
/// itself. \return the closest enclosing block starting from "this", which
/// has successors. \return the root enclosing block if all enclosing blocks
/// have no successors.
VPBlockBase *getEnclosingBlockWithSuccessors();
/// \return the closest enclosing block starting from "this", which has
/// predecessors. \return the root enclosing block if all enclosing blocks
/// have no predecessors.
VPBlockBase *getEnclosingBlockWithPredecessors();
/// \return the successors either attached directly to this VPBlockBase or, if
/// this VPBlockBase is the exit block of a VPRegionBlock and has no
/// successors of its own, search recursively for the first enclosing
/// VPRegionBlock that has successors and return them. If no such
/// VPRegionBlock exists, return the (empty) successors of the topmost
/// VPBlockBase reached.
const VPBlocksTy &getHierarchicalSuccessors() {
return getEnclosingBlockWithSuccessors()->getSuccessors();
}
/// \return the hierarchical successor of this VPBlockBase if it has a single
/// hierarchical successor. Otherwise return a null pointer.
VPBlockBase *getSingleHierarchicalSuccessor() {
return getEnclosingBlockWithSuccessors()->getSingleSuccessor();
}
/// \return the predecessors either attached directly to this VPBlockBase or,
/// if this VPBlockBase is the entry block of a VPRegionBlock and has no
/// predecessors of its own, search recursively for the first enclosing
/// VPRegionBlock that has predecessors and return them. If no such
/// VPRegionBlock exists, return the (empty) predecessors of the topmost
/// VPBlockBase reached.
const VPBlocksTy &getHierarchicalPredecessors() {
return getEnclosingBlockWithPredecessors()->getPredecessors();
}
/// \return the hierarchical predecessor of this VPBlockBase if it has a
/// single hierarchical predecessor. Otherwise return a null pointer.
VPBlockBase *getSingleHierarchicalPredecessor() {
return getEnclosingBlockWithPredecessors()->getSinglePredecessor();
}
/// Set a given VPBlockBase \p Successor as the single successor of this
/// VPBlockBase. This VPBlockBase is not added as predecessor of \p Successor.
/// This VPBlockBase must have no successors.
void setOneSuccessor(VPBlockBase *Successor) {
assert(Successors.empty() && "Setting one successor when others exist.");
assert(Successor->getParent() == getParent() &&
"connected blocks must have the same parent");
appendSuccessor(Successor);
}
/// Set two given VPBlockBases \p IfTrue and \p IfFalse to be the two
/// successors of this VPBlockBase. This VPBlockBase is not added as
/// predecessor of \p IfTrue or \p IfFalse. This VPBlockBase must have no
/// successors.
void setTwoSuccessors(VPBlockBase *IfTrue, VPBlockBase *IfFalse) {
assert(Successors.empty() && "Setting two successors when others exist.");
appendSuccessor(IfTrue);
appendSuccessor(IfFalse);
}
/// Set each VPBasicBlock in \p NewPreds as predecessor of this VPBlockBase.
/// This VPBlockBase must have no predecessors. This VPBlockBase is not added
/// as successor of any VPBasicBlock in \p NewPreds.
void setPredecessors(ArrayRef<VPBlockBase *> NewPreds) {
assert(Predecessors.empty() && "Block predecessors already set.");
for (auto *Pred : NewPreds)
appendPredecessor(Pred);
}
/// Remove all the predecessor of this block.
void clearPredecessors() { Predecessors.clear(); }
/// Remove all the successors of this block.
void clearSuccessors() { Successors.clear(); }
/// The method which generates the output IR that correspond to this
/// VPBlockBase, thereby "executing" the VPlan.
virtual void execute(VPTransformState *State) = 0;
/// Delete all blocks reachable from a given VPBlockBase, inclusive.
static void deleteCFG(VPBlockBase *Entry);
/// Return true if it is legal to hoist instructions into this block.
bool isLegalToHoistInto() {
// There are currently no constraints that prevent an instruction to be
// hoisted into a VPBlockBase.
return true;
}
/// Replace all operands of VPUsers in the block with \p NewValue and also
/// replaces all uses of VPValues defined in the block with NewValue.
virtual void dropAllReferences(VPValue *NewValue) = 0;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
void printAsOperand(raw_ostream &OS, bool PrintType) const {
OS << getName();
}
/// Print plain-text dump of this VPBlockBase to \p O, prefixing all lines
/// with \p Indent. \p SlotTracker is used to print unnamed VPValue's using
/// consequtive numbers.
///
/// Note that the numbering is applied to the whole VPlan, so printing
/// individual blocks is consistent with the whole VPlan printing.
virtual void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const = 0;
/// Print plain-text dump of this VPlan to \p O.
void print(raw_ostream &O) const {
VPSlotTracker SlotTracker(getPlan());
print(O, "", SlotTracker);
}
/// Print the successors of this block to \p O, prefixing all lines with \p
/// Indent.
void printSuccessors(raw_ostream &O, const Twine &Indent) const;
/// Dump this VPBlockBase to dbgs().
LLVM_DUMP_METHOD void dump() const { print(dbgs()); }
#endif
/// Clone the current block and it's recipes without updating the operands of
/// the cloned recipes, including all blocks in the single-entry single-exit
/// region for VPRegionBlocks.
virtual VPBlockBase *clone() = 0;
};
/// A value that is used outside the VPlan. The operand of the user needs to be
/// added to the associated LCSSA phi node.
class VPLiveOut : public VPUser {
PHINode *Phi;
public:
VPLiveOut(PHINode *Phi, VPValue *Op)
: VPUser({Op}, VPUser::VPUserID::LiveOut), Phi(Phi) {}
static inline bool classof(const VPUser *U) {
return U->getVPUserID() == VPUser::VPUserID::LiveOut;
}
/// Fixup the wrapped LCSSA phi node in the unique exit block. This simply
/// means we need to add the appropriate incoming value from the middle
/// block as exiting edges from the scalar epilogue loop (if present) are
/// already in place, and we exit the vector loop exclusively to the middle
/// block.
void fixPhi(VPlan &Plan, VPTransformState &State);
/// Returns true if the VPLiveOut uses scalars of operand \p Op.
bool usesScalars(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
PHINode *getPhi() const { return Phi; }
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the VPLiveOut to \p O.
void print(raw_ostream &O, VPSlotTracker &SlotTracker) const;
#endif
};
/// VPRecipeBase is a base class modeling a sequence of one or more output IR
/// instructions. VPRecipeBase owns the VPValues it defines through VPDef
/// and is responsible for deleting its defined values. Single-value
/// recipes must inherit from VPSingleDef instead of inheriting from both
/// VPRecipeBase and VPValue separately.
class VPRecipeBase : public ilist_node_with_parent<VPRecipeBase, VPBasicBlock>,
public VPDef,
public VPUser {
friend VPBasicBlock;
friend class VPBlockUtils;
/// Each VPRecipe belongs to a single VPBasicBlock.
VPBasicBlock *Parent = nullptr;
/// The debug location for the recipe.
DebugLoc DL;
public:
VPRecipeBase(const unsigned char SC, ArrayRef<VPValue *> Operands,
DebugLoc DL = {})
: VPDef(SC), VPUser(Operands, VPUser::VPUserID::Recipe), DL(DL) {}
template <typename IterT>
VPRecipeBase(const unsigned char SC, iterator_range<IterT> Operands,
DebugLoc DL = {})
: VPDef(SC), VPUser(Operands, VPUser::VPUserID::Recipe), DL(DL) {}
virtual ~VPRecipeBase() = default;
/// Clone the current recipe.
virtual VPRecipeBase *clone() = 0;
/// \return the VPBasicBlock which this VPRecipe belongs to.
VPBasicBlock *getParent() { return Parent; }
const VPBasicBlock *getParent() const { return Parent; }
/// The method which generates the output IR instructions that correspond to
/// this VPRecipe, thereby "executing" the VPlan.
virtual void execute(VPTransformState &State) = 0;
/// Insert an unlinked recipe into a basic block immediately before
/// the specified recipe.
void insertBefore(VPRecipeBase *InsertPos);
/// Insert an unlinked recipe into \p BB immediately before the insertion
/// point \p IP;
void insertBefore(VPBasicBlock &BB, iplist<VPRecipeBase>::iterator IP);
/// Insert an unlinked Recipe into a basic block immediately after
/// the specified Recipe.
void insertAfter(VPRecipeBase *InsertPos);
/// Unlink this recipe from its current VPBasicBlock and insert it into
/// the VPBasicBlock that MovePos lives in, right after MovePos.
void moveAfter(VPRecipeBase *MovePos);
/// Unlink this recipe and insert into BB before I.
///
/// \pre I is a valid iterator into BB.
void moveBefore(VPBasicBlock &BB, iplist<VPRecipeBase>::iterator I);
/// This method unlinks 'this' from the containing basic block, but does not
/// delete it.
void removeFromParent();
/// This method unlinks 'this' from the containing basic block and deletes it.
///
/// \returns an iterator pointing to the element after the erased one
iplist<VPRecipeBase>::iterator eraseFromParent();
/// Method to support type inquiry through isa, cast, and dyn_cast.
static inline bool classof(const VPDef *D) {
// All VPDefs are also VPRecipeBases.
return true;
}
static inline bool classof(const VPUser *U) {
return U->getVPUserID() == VPUser::VPUserID::Recipe;
}
/// Returns true if the recipe may have side-effects.
bool mayHaveSideEffects() const;
/// Returns true for PHI-like recipes.
bool isPhi() const {
return getVPDefID() >= VPFirstPHISC && getVPDefID() <= VPLastPHISC;
}
/// Returns true if the recipe may read from memory.
bool mayReadFromMemory() const;
/// Returns true if the recipe may write to memory.
bool mayWriteToMemory() const;
/// Returns true if the recipe may read from or write to memory.
bool mayReadOrWriteMemory() const {
return mayReadFromMemory() || mayWriteToMemory();
}
/// Returns the debug location of the recipe.
DebugLoc getDebugLoc() const { return DL; }
};
// Helper macro to define common classof implementations for recipes.
#define VP_CLASSOF_IMPL(VPDefID) \
static inline bool classof(const VPDef *D) { \
return D->getVPDefID() == VPDefID; \
} \
static inline bool classof(const VPValue *V) { \
auto *R = V->getDefiningRecipe(); \
return R && R->getVPDefID() == VPDefID; \
} \
static inline bool classof(const VPUser *U) { \
auto *R = dyn_cast<VPRecipeBase>(U); \
return R && R->getVPDefID() == VPDefID; \
} \
static inline bool classof(const VPRecipeBase *R) { \
return R->getVPDefID() == VPDefID; \
} \
static inline bool classof(const VPSingleDefRecipe *R) { \
return R->getVPDefID() == VPDefID; \
}
/// VPSingleDef is a base class for recipes for modeling a sequence of one or
/// more output IR that define a single result VPValue.
/// Note that VPRecipeBase must be inherited from before VPValue.
class VPSingleDefRecipe : public VPRecipeBase, public VPValue {
public:
template <typename IterT>
VPSingleDefRecipe(const unsigned char SC, IterT Operands, DebugLoc DL = {})
: VPRecipeBase(SC, Operands, DL), VPValue(this) {}
VPSingleDefRecipe(const unsigned char SC, ArrayRef<VPValue *> Operands,
DebugLoc DL = {})
: VPRecipeBase(SC, Operands, DL), VPValue(this) {}
template <typename IterT>
VPSingleDefRecipe(const unsigned char SC, IterT Operands, Value *UV,
DebugLoc DL = {})
: VPRecipeBase(SC, Operands, DL), VPValue(this, UV) {}
static inline bool classof(const VPRecipeBase *R) {
switch (R->getVPDefID()) {
case VPRecipeBase::VPDerivedIVSC:
case VPRecipeBase::VPExpandSCEVSC:
case VPRecipeBase::VPInstructionSC:
case VPRecipeBase::VPReductionSC:
case VPRecipeBase::VPReplicateSC:
case VPRecipeBase::VPScalarIVStepsSC:
case VPRecipeBase::VPVectorPointerSC:
case VPRecipeBase::VPWidenCallSC:
case VPRecipeBase::VPWidenCanonicalIVSC:
case VPRecipeBase::VPWidenCastSC:
case VPRecipeBase::VPWidenGEPSC:
case VPRecipeBase::VPWidenSC:
case VPRecipeBase::VPWidenSelectSC:
case VPRecipeBase::VPBlendSC:
case VPRecipeBase::VPPredInstPHISC:
case VPRecipeBase::VPCanonicalIVPHISC:
case VPRecipeBase::VPActiveLaneMaskPHISC:
case VPRecipeBase::VPFirstOrderRecurrencePHISC:
case VPRecipeBase::VPWidenPHISC:
case VPRecipeBase::VPWidenIntOrFpInductionSC:
case VPRecipeBase::VPWidenPointerInductionSC:
case VPRecipeBase::VPReductionPHISC:
case VPRecipeBase::VPScalarCastSC:
return true;
case VPRecipeBase::VPInterleaveSC:
case VPRecipeBase::VPBranchOnMaskSC:
case VPRecipeBase::VPWidenMemoryInstructionSC:
// TODO: Widened stores don't define a value, but widened loads do. Split
// the recipes to be able to make widened loads VPSingleDefRecipes.
return false;
}
llvm_unreachable("Unhandled VPDefID");
}
static inline bool classof(const VPUser *U) {
auto *R = dyn_cast<VPRecipeBase>(U);
return R && classof(R);
}
/// Returns the underlying instruction.
Instruction *getUnderlyingInstr() {
return cast<Instruction>(getUnderlyingValue());
}
const Instruction *getUnderlyingInstr() const {
return cast<Instruction>(getUnderlyingValue());
}
};
/// Class to record LLVM IR flag for a recipe along with it.
class VPRecipeWithIRFlags : public VPSingleDefRecipe {
enum class OperationType : unsigned char {
Cmp,
OverflowingBinOp,
DisjointOp,
PossiblyExactOp,
GEPOp,
FPMathOp,
NonNegOp,
Other
};
public:
struct WrapFlagsTy {
char HasNUW : 1;
char HasNSW : 1;
WrapFlagsTy(bool HasNUW, bool HasNSW) : HasNUW(HasNUW), HasNSW(HasNSW) {}
};
protected:
struct GEPFlagsTy {
char IsInBounds : 1;
GEPFlagsTy(bool IsInBounds) : IsInBounds(IsInBounds) {}
};
private:
struct DisjointFlagsTy {
char IsDisjoint : 1;
};
struct ExactFlagsTy {
char IsExact : 1;
};
struct NonNegFlagsTy {
char NonNeg : 1;
};
struct FastMathFlagsTy {
char AllowReassoc : 1;
char NoNaNs : 1;
char NoInfs : 1;
char NoSignedZeros : 1;
char AllowReciprocal : 1;
char AllowContract : 1;
char ApproxFunc : 1;
FastMathFlagsTy(const FastMathFlags &FMF);
};
OperationType OpType;
union {
CmpInst::Predicate CmpPredicate;
WrapFlagsTy WrapFlags;
DisjointFlagsTy DisjointFlags;
ExactFlagsTy ExactFlags;
GEPFlagsTy GEPFlags;
NonNegFlagsTy NonNegFlags;
FastMathFlagsTy FMFs;
unsigned AllFlags;
};
protected:
void transferFlags(VPRecipeWithIRFlags &Other) {
OpType = Other.OpType;
AllFlags = Other.AllFlags;
}
public:
template <typename IterT>
VPRecipeWithIRFlags(const unsigned char SC, IterT Operands, DebugLoc DL = {})
: VPSingleDefRecipe(SC, Operands, DL) {
OpType = OperationType::Other;
AllFlags = 0;
}
template <typename IterT>
VPRecipeWithIRFlags(const unsigned char SC, IterT Operands, Instruction &I)
: VPSingleDefRecipe(SC, Operands, &I, I.getDebugLoc()) {
if (auto *Op = dyn_cast<CmpInst>(&I)) {
OpType = OperationType::Cmp;
CmpPredicate = Op->getPredicate();
} else if (auto *Op = dyn_cast<PossiblyDisjointInst>(&I)) {
OpType = OperationType::DisjointOp;
DisjointFlags.IsDisjoint = Op->isDisjoint();
} else if (auto *Op = dyn_cast<OverflowingBinaryOperator>(&I)) {
OpType = OperationType::OverflowingBinOp;
WrapFlags = {Op->hasNoUnsignedWrap(), Op->hasNoSignedWrap()};
} else if (auto *Op = dyn_cast<PossiblyExactOperator>(&I)) {
OpType = OperationType::PossiblyExactOp;
ExactFlags.IsExact = Op->isExact();
} else if (auto *GEP = dyn_cast<GetElementPtrInst>(&I)) {
OpType = OperationType::GEPOp;
GEPFlags.IsInBounds = GEP->isInBounds();
} else if (auto *PNNI = dyn_cast<PossiblyNonNegInst>(&I)) {
OpType = OperationType::NonNegOp;
NonNegFlags.NonNeg = PNNI->hasNonNeg();
} else if (auto *Op = dyn_cast<FPMathOperator>(&I)) {
OpType = OperationType::FPMathOp;
FMFs = Op->getFastMathFlags();
} else {
OpType = OperationType::Other;
AllFlags = 0;
}
}
template <typename IterT>
VPRecipeWithIRFlags(const unsigned char SC, IterT Operands,
CmpInst::Predicate Pred, DebugLoc DL = {})
: VPSingleDefRecipe(SC, Operands, DL), OpType(OperationType::Cmp),
CmpPredicate(Pred) {}
template <typename IterT>
VPRecipeWithIRFlags(const unsigned char SC, IterT Operands,
WrapFlagsTy WrapFlags, DebugLoc DL = {})
: VPSingleDefRecipe(SC, Operands, DL),
OpType(OperationType::OverflowingBinOp), WrapFlags(WrapFlags) {}
template <typename IterT>
VPRecipeWithIRFlags(const unsigned char SC, IterT Operands,
FastMathFlags FMFs, DebugLoc DL = {})
: VPSingleDefRecipe(SC, Operands, DL), OpType(OperationType::FPMathOp),
FMFs(FMFs) {}
protected:
template <typename IterT>
VPRecipeWithIRFlags(const unsigned char SC, IterT Operands,
GEPFlagsTy GEPFlags, DebugLoc DL = {})
: VPSingleDefRecipe(SC, Operands, DL), OpType(OperationType::GEPOp),
GEPFlags(GEPFlags) {}
public:
static inline bool classof(const VPRecipeBase *R) {
return R->getVPDefID() == VPRecipeBase::VPInstructionSC ||
R->getVPDefID() == VPRecipeBase::VPWidenSC ||
R->getVPDefID() == VPRecipeBase::VPWidenGEPSC ||
R->getVPDefID() == VPRecipeBase::VPWidenCastSC ||
R->getVPDefID() == VPRecipeBase::VPReplicateSC ||
R->getVPDefID() == VPRecipeBase::VPVectorPointerSC;
}
/// Drop all poison-generating flags.
void dropPoisonGeneratingFlags() {
// NOTE: This needs to be kept in-sync with
// Instruction::dropPoisonGeneratingFlags.
switch (OpType) {
case OperationType::OverflowingBinOp:
WrapFlags.HasNUW = false;
WrapFlags.HasNSW = false;
break;
case OperationType::DisjointOp:
DisjointFlags.IsDisjoint = false;
break;
case OperationType::PossiblyExactOp:
ExactFlags.IsExact = false;
break;
case OperationType::GEPOp:
GEPFlags.IsInBounds = false;
break;
case OperationType::FPMathOp:
FMFs.NoNaNs = false;
FMFs.NoInfs = false;
break;
case OperationType::NonNegOp:
NonNegFlags.NonNeg = false;
break;
case OperationType::Cmp:
case OperationType::Other:
break;
}
}
/// Set the IR flags for \p I.
void setFlags(Instruction *I) const {
switch (OpType) {
case OperationType::OverflowingBinOp:
I->setHasNoUnsignedWrap(WrapFlags.HasNUW);
I->setHasNoSignedWrap(WrapFlags.HasNSW);
break;
case OperationType::DisjointOp:
cast<PossiblyDisjointInst>(I)->setIsDisjoint(DisjointFlags.IsDisjoint);
break;
case OperationType::PossiblyExactOp:
I->setIsExact(ExactFlags.IsExact);
break;
case OperationType::GEPOp:
cast<GetElementPtrInst>(I)->setIsInBounds(GEPFlags.IsInBounds);
break;
case OperationType::FPMathOp:
I->setHasAllowReassoc(FMFs.AllowReassoc);
I->setHasNoNaNs(FMFs.NoNaNs);
I->setHasNoInfs(FMFs.NoInfs);
I->setHasNoSignedZeros(FMFs.NoSignedZeros);
I->setHasAllowReciprocal(FMFs.AllowReciprocal);
I->setHasAllowContract(FMFs.AllowContract);
I->setHasApproxFunc(FMFs.ApproxFunc);
break;
case OperationType::NonNegOp:
I->setNonNeg(NonNegFlags.NonNeg);
break;
case OperationType::Cmp:
case OperationType::Other:
break;
}
}
CmpInst::Predicate getPredicate() const {
assert(OpType == OperationType::Cmp &&
"recipe doesn't have a compare predicate");
return CmpPredicate;
}
bool isInBounds() const {
assert(OpType == OperationType::GEPOp &&
"recipe doesn't have inbounds flag");
return GEPFlags.IsInBounds;
}
/// Returns true if the recipe has fast-math flags.
bool hasFastMathFlags() const { return OpType == OperationType::FPMathOp; }
FastMathFlags getFastMathFlags() const;
bool hasNoUnsignedWrap() const {
assert(OpType == OperationType::OverflowingBinOp &&
"recipe doesn't have a NUW flag");
return WrapFlags.HasNUW;
}
bool hasNoSignedWrap() const {
assert(OpType == OperationType::OverflowingBinOp &&
"recipe doesn't have a NSW flag");
return WrapFlags.HasNSW;
}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
void printFlags(raw_ostream &O) const;
#endif
};
/// This is a concrete Recipe that models a single VPlan-level instruction.
/// While as any Recipe it may generate a sequence of IR instructions when
/// executed, these instructions would always form a single-def expression as
/// the VPInstruction is also a single def-use vertex.
class VPInstruction : public VPRecipeWithIRFlags {
friend class VPlanSlp;
public:
/// VPlan opcodes, extending LLVM IR with idiomatics instructions.
enum {
FirstOrderRecurrenceSplice =
Instruction::OtherOpsEnd + 1, // Combines the incoming and previous
// values of a first-order recurrence.
Not,
SLPLoad,
SLPStore,
ActiveLaneMask,
CalculateTripCountMinusVF,
// Increment the canonical IV separately for each unrolled part.
CanonicalIVIncrementForPart,
BranchOnCount,
BranchOnCond,
ComputeReductionResult,
};
private:
typedef unsigned char OpcodeTy;
OpcodeTy Opcode;
/// An optional name that can be used for the generated IR instruction.
const std::string Name;
/// Utility method serving execute(): generates a single instance of the
/// modeled instruction. \returns the generated value for \p Part.
/// In some cases an existing value is returned rather than a generated
/// one.
Value *generateInstruction(VPTransformState &State, unsigned Part);
#if !defined(NDEBUG)
/// Return true if the VPInstruction is a floating point math operation, i.e.
/// has fast-math flags.
bool isFPMathOp() const;
#endif
public:
VPInstruction(unsigned Opcode, ArrayRef<VPValue *> Operands, DebugLoc DL,
const Twine &Name = "")
: VPRecipeWithIRFlags(VPDef::VPInstructionSC, Operands, DL),
Opcode(Opcode), Name(Name.str()) {}
VPInstruction(unsigned Opcode, std::initializer_list<VPValue *> Operands,
DebugLoc DL = {}, const Twine &Name = "")
: VPInstruction(Opcode, ArrayRef<VPValue *>(Operands), DL, Name) {}
VPInstruction(unsigned Opcode, CmpInst::Predicate Pred, VPValue *A,
VPValue *B, DebugLoc DL = {}, const Twine &Name = "");
VPInstruction(unsigned Opcode, std::initializer_list<VPValue *> Operands,
WrapFlagsTy WrapFlags, DebugLoc DL = {}, const Twine &Name = "")
: VPRecipeWithIRFlags(VPDef::VPInstructionSC, Operands, WrapFlags, DL),
Opcode(Opcode), Name(Name.str()) {}
VPInstruction(unsigned Opcode, std::initializer_list<VPValue *> Operands,
FastMathFlags FMFs, DebugLoc DL = {}, const Twine &Name = "");
VP_CLASSOF_IMPL(VPDef::VPInstructionSC)
VPRecipeBase *clone() override {
SmallVector<VPValue *, 2> Operands(operands());
auto *New = new VPInstruction(Opcode, Operands, getDebugLoc(), Name);
New->transferFlags(*this);
return New;
}
unsigned getOpcode() const { return Opcode; }
/// Generate the instruction.
/// TODO: We currently execute only per-part unless a specific instance is
/// provided.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the VPInstruction to \p O.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
/// Print the VPInstruction to dbgs() (for debugging).
LLVM_DUMP_METHOD void dump() const;
#endif
/// Return true if this instruction may modify memory.
bool mayWriteToMemory() const {
// TODO: we can use attributes of the called function to rule out memory
// modifications.
return Opcode == Instruction::Store || Opcode == Instruction::Call ||
Opcode == Instruction::Invoke || Opcode == SLPStore;
}
bool hasResult() const {
// CallInst may or may not have a result, depending on the called function.
// Conservatively return calls have results for now.
switch (getOpcode()) {
case Instruction::Ret:
case Instruction::Br:
case Instruction::Store:
case Instruction::Switch:
case Instruction::IndirectBr:
case Instruction::Resume:
case Instruction::CatchRet:
case Instruction::Unreachable:
case Instruction::Fence:
case Instruction::AtomicRMW:
case VPInstruction::BranchOnCond:
case VPInstruction::BranchOnCount:
return false;
default:
return true;
}
}
/// Returns true if the recipe only uses the first lane of operand \p Op.
bool onlyFirstLaneUsed(const VPValue *Op) const override;
/// Returns true if the recipe only uses the first part of operand \p Op.
bool onlyFirstPartUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
if (getOperand(0) != Op)
return false;
switch (getOpcode()) {
default:
return false;
case VPInstruction::BranchOnCount:
return true;
};
llvm_unreachable("switch should return");
}
};
/// VPWidenRecipe is a recipe for producing a copy of vector type its
/// ingredient. This recipe covers most of the traditional vectorization cases
/// where each ingredient transforms into a vectorized version of itself.
class VPWidenRecipe : public VPRecipeWithIRFlags {
unsigned Opcode;
public:
template <typename IterT>
VPWidenRecipe(Instruction &I, iterator_range<IterT> Operands)
: VPRecipeWithIRFlags(VPDef::VPWidenSC, Operands, I),
Opcode(I.getOpcode()) {}
~VPWidenRecipe() override = default;
VPRecipeBase *clone() override {
auto *R = new VPWidenRecipe(*getUnderlyingInstr(), operands());
R->transferFlags(*this);
return R;
}
VP_CLASSOF_IMPL(VPDef::VPWidenSC)
/// Produce widened copies of all Ingredients.
void execute(VPTransformState &State) override;
unsigned getOpcode() const { return Opcode; }
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
};
/// VPWidenCastRecipe is a recipe to create vector cast instructions.
class VPWidenCastRecipe : public VPRecipeWithIRFlags {
/// Cast instruction opcode.
Instruction::CastOps Opcode;
/// Result type for the cast.
Type *ResultTy;
public:
VPWidenCastRecipe(Instruction::CastOps Opcode, VPValue *Op, Type *ResultTy,
CastInst &UI)
: VPRecipeWithIRFlags(VPDef::VPWidenCastSC, Op, UI), Opcode(Opcode),
ResultTy(ResultTy) {
assert(UI.getOpcode() == Opcode &&
"opcode of underlying cast doesn't match");
assert(UI.getType() == ResultTy &&
"result type of underlying cast doesn't match");
}
VPWidenCastRecipe(Instruction::CastOps Opcode, VPValue *Op, Type *ResultTy)
: VPRecipeWithIRFlags(VPDef::VPWidenCastSC, Op), Opcode(Opcode),
ResultTy(ResultTy) {}
~VPWidenCastRecipe() override = default;
VPRecipeBase *clone() override {
if (auto *UV = getUnderlyingValue())
return new VPWidenCastRecipe(Opcode, getOperand(0), ResultTy,
*cast<CastInst>(UV));
return new VPWidenCastRecipe(Opcode, getOperand(0), ResultTy);
}
VP_CLASSOF_IMPL(VPDef::VPWidenCastSC)
/// Produce widened copies of the cast.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
Instruction::CastOps getOpcode() const { return Opcode; }
/// Returns the result type of the cast.
Type *getResultType() const { return ResultTy; }
};
/// VPScalarCastRecipe is a recipe to create scalar cast instructions.
class VPScalarCastRecipe : public VPSingleDefRecipe {
Instruction::CastOps Opcode;
Type *ResultTy;
Value *generate(VPTransformState &State, unsigned Part);
public:
VPScalarCastRecipe(Instruction::CastOps Opcode, VPValue *Op, Type *ResultTy)
: VPSingleDefRecipe(VPDef::VPScalarCastSC, {Op}), Opcode(Opcode),
ResultTy(ResultTy) {}
~VPScalarCastRecipe() override = default;
VPRecipeBase *clone() override {
return new VPScalarCastRecipe(Opcode, getOperand(0), ResultTy);
}
VP_CLASSOF_IMPL(VPDef::VPScalarCastSC)
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
/// Returns the result type of the cast.
Type *getResultType() const { return ResultTy; }
};
/// A recipe for widening Call instructions.
class VPWidenCallRecipe : public VPSingleDefRecipe {
/// ID of the vector intrinsic to call when widening the call. If set the
/// Intrinsic::not_intrinsic, a library call will be used instead.
Intrinsic::ID VectorIntrinsicID;
/// If this recipe represents a library call, Variant stores a pointer to
/// the chosen function. There is a 1:1 mapping between a given VF and the
/// chosen vectorized variant, so there will be a different vplan for each
/// VF with a valid variant.
Function *Variant;
public:
template <typename IterT>
VPWidenCallRecipe(CallInst &I, iterator_range<IterT> CallArguments,
Intrinsic::ID VectorIntrinsicID, DebugLoc DL = {},
Function *Variant = nullptr)
: VPSingleDefRecipe(VPDef::VPWidenCallSC, CallArguments, &I, DL),
VectorIntrinsicID(VectorIntrinsicID), Variant(Variant) {}
~VPWidenCallRecipe() override = default;
VPRecipeBase *clone() override {
return new VPWidenCallRecipe(*cast<CallInst>(getUnderlyingInstr()),
operands(), VectorIntrinsicID, getDebugLoc(),
Variant);
}
VP_CLASSOF_IMPL(VPDef::VPWidenCallSC)
/// Produce a widened version of the call instruction.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
};
/// A recipe for widening select instructions.
struct VPWidenSelectRecipe : public VPSingleDefRecipe {
template <typename IterT>
VPWidenSelectRecipe(SelectInst &I, iterator_range<IterT> Operands)
: VPSingleDefRecipe(VPDef::VPWidenSelectSC, Operands, &I,
I.getDebugLoc()) {}
~VPWidenSelectRecipe() override = default;
VPRecipeBase *clone() override {
return new VPWidenSelectRecipe(*cast<SelectInst>(getUnderlyingInstr()),
operands());
}
VP_CLASSOF_IMPL(VPDef::VPWidenSelectSC)
/// Produce a widened version of the select instruction.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
VPValue *getCond() const {
return getOperand(0);
}
bool isInvariantCond() const {
return getCond()->isDefinedOutsideVectorRegions();
}
};
/// A recipe for handling GEP instructions.
class VPWidenGEPRecipe : public VPRecipeWithIRFlags {
bool isPointerLoopInvariant() const {
return getOperand(0)->isDefinedOutsideVectorRegions();
}
bool isIndexLoopInvariant(unsigned I) const {
return getOperand(I + 1)->isDefinedOutsideVectorRegions();
}
bool areAllOperandsInvariant() const {
return all_of(operands(), [](VPValue *Op) {
return Op->isDefinedOutsideVectorRegions();
});
}
public:
template <typename IterT>
VPWidenGEPRecipe(GetElementPtrInst *GEP, iterator_range<IterT> Operands)
: VPRecipeWithIRFlags(VPDef::VPWidenGEPSC, Operands, *GEP) {}
~VPWidenGEPRecipe() override = default;
VPRecipeBase *clone() override {
return new VPWidenGEPRecipe(cast<GetElementPtrInst>(getUnderlyingInstr()),
operands());
}
VP_CLASSOF_IMPL(VPDef::VPWidenGEPSC)
/// Generate the gep nodes.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
};
/// A recipe to compute the pointers for widened memory accesses of IndexTy for
/// all parts. If IsReverse is true, compute pointers for accessing the input in
/// reverse order per part.
class VPVectorPointerRecipe : public VPRecipeWithIRFlags {
Type *IndexedTy;
bool IsReverse;
public:
VPVectorPointerRecipe(VPValue *Ptr, Type *IndexedTy, bool IsReverse,
bool IsInBounds, DebugLoc DL)
: VPRecipeWithIRFlags(VPDef::VPVectorPointerSC, ArrayRef<VPValue *>(Ptr),
GEPFlagsTy(IsInBounds), DL),
IndexedTy(IndexedTy), IsReverse(IsReverse) {}
VP_CLASSOF_IMPL(VPDef::VPVectorPointerSC)
void execute(VPTransformState &State) override;
bool onlyFirstLaneUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
VPRecipeBase *clone() override {
return new VPVectorPointerRecipe(getOperand(0), IndexedTy, IsReverse,
isInBounds(), getDebugLoc());
}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
};
/// A pure virtual base class for all recipes modeling header phis, including
/// phis for first order recurrences, pointer inductions and reductions. The
/// start value is the first operand of the recipe and the incoming value from
/// the backedge is the second operand.
///
/// Inductions are modeled using the following sub-classes:
/// * VPCanonicalIVPHIRecipe: Canonical scalar induction of the vector loop,
/// starting at a specified value (zero for the main vector loop, the resume
/// value for the epilogue vector loop) and stepping by 1. The induction
/// controls exiting of the vector loop by comparing against the vector trip
/// count. Produces a single scalar PHI for the induction value per
/// iteration.
/// * VPWidenIntOrFpInductionRecipe: Generates vector values for integer and
/// floating point inductions with arbitrary start and step values. Produces
/// a vector PHI per-part.
/// * VPDerivedIVRecipe: Converts the canonical IV value to the corresponding
/// value of an IV with different start and step values. Produces a single
/// scalar value per iteration
/// * VPScalarIVStepsRecipe: Generates scalar values per-lane based on a
/// canonical or derived induction.
/// * VPWidenPointerInductionRecipe: Generate vector and scalar values for a
/// pointer induction. Produces either a vector PHI per-part or scalar values
/// per-lane based on the canonical induction.
class VPHeaderPHIRecipe : public VPSingleDefRecipe {
protected:
VPHeaderPHIRecipe(unsigned char VPDefID, Instruction *UnderlyingInstr,
VPValue *Start = nullptr, DebugLoc DL = {})
: VPSingleDefRecipe(VPDefID, ArrayRef<VPValue *>(), UnderlyingInstr, DL) {
if (Start)
addOperand(Start);
}
public:
~VPHeaderPHIRecipe() override = default;
/// Method to support type inquiry through isa, cast, and dyn_cast.
static inline bool classof(const VPRecipeBase *B) {
return B->getVPDefID() >= VPDef::VPFirstHeaderPHISC &&
B->getVPDefID() <= VPDef::VPLastHeaderPHISC;
}
static inline bool classof(const VPValue *V) {
auto *B = V->getDefiningRecipe();
return B && B->getVPDefID() >= VPRecipeBase::VPFirstHeaderPHISC &&
B->getVPDefID() <= VPRecipeBase::VPLastHeaderPHISC;
}
/// Generate the phi nodes.
void execute(VPTransformState &State) override = 0;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override = 0;
#endif
/// Returns the start value of the phi, if one is set.
VPValue *getStartValue() {
return getNumOperands() == 0 ? nullptr : getOperand(0);
}
VPValue *getStartValue() const {
return getNumOperands() == 0 ? nullptr : getOperand(0);
}
/// Update the start value of the recipe.
void setStartValue(VPValue *V) { setOperand(0, V); }
/// Returns the incoming value from the loop backedge.
virtual VPValue *getBackedgeValue() {
return getOperand(1);
}
/// Returns the backedge value as a recipe. The backedge value is guaranteed
/// to be a recipe.
virtual VPRecipeBase &getBackedgeRecipe() {
return *getBackedgeValue()->getDefiningRecipe();
}
};
/// A recipe for handling phi nodes of integer and floating-point inductions,
/// producing their vector values.
class VPWidenIntOrFpInductionRecipe : public VPHeaderPHIRecipe {
PHINode *IV;
TruncInst *Trunc;
const InductionDescriptor &IndDesc;
public:
VPWidenIntOrFpInductionRecipe(PHINode *IV, VPValue *Start, VPValue *Step,
const InductionDescriptor &IndDesc)
: VPHeaderPHIRecipe(VPDef::VPWidenIntOrFpInductionSC, IV, Start), IV(IV),
Trunc(nullptr), IndDesc(IndDesc) {
addOperand(Step);
}
VPWidenIntOrFpInductionRecipe(PHINode *IV, VPValue *Start, VPValue *Step,
const InductionDescriptor &IndDesc,
TruncInst *Trunc)
: VPHeaderPHIRecipe(VPDef::VPWidenIntOrFpInductionSC, Trunc, Start),
IV(IV), Trunc(Trunc), IndDesc(IndDesc) {
addOperand(Step);
}
~VPWidenIntOrFpInductionRecipe() override = default;
VPRecipeBase *clone() override {
return new VPWidenIntOrFpInductionRecipe(IV, getStartValue(),
getStepValue(), IndDesc, Trunc);
}
VP_CLASSOF_IMPL(VPDef::VPWidenIntOrFpInductionSC)
/// Generate the vectorized and scalarized versions of the phi node as
/// needed by their users.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
VPValue *getBackedgeValue() override {
// TODO: All operands of base recipe must exist and be at same index in
// derived recipe.
llvm_unreachable(
"VPWidenIntOrFpInductionRecipe generates its own backedge value");
}
VPRecipeBase &getBackedgeRecipe() override {
// TODO: All operands of base recipe must exist and be at same index in
// derived recipe.
llvm_unreachable(
"VPWidenIntOrFpInductionRecipe generates its own backedge value");
}
/// Returns the step value of the induction.
VPValue *getStepValue() { return getOperand(1); }
const VPValue *getStepValue() const { return getOperand(1); }
/// Returns the first defined value as TruncInst, if it is one or nullptr
/// otherwise.
TruncInst *getTruncInst() { return Trunc; }
const TruncInst *getTruncInst() const { return Trunc; }
PHINode *getPHINode() { return IV; }
/// Returns the induction descriptor for the recipe.
const InductionDescriptor &getInductionDescriptor() const { return IndDesc; }
/// Returns true if the induction is canonical, i.e. starting at 0 and
/// incremented by UF * VF (= the original IV is incremented by 1).
bool isCanonical() const;
/// Returns the scalar type of the induction.
Type *getScalarType() const {
return Trunc ? Trunc->getType() : IV->getType();
}
};
class VPWidenPointerInductionRecipe : public VPHeaderPHIRecipe {
const InductionDescriptor &IndDesc;
bool IsScalarAfterVectorization;
public:
/// Create a new VPWidenPointerInductionRecipe for \p Phi with start value \p
/// Start.
VPWidenPointerInductionRecipe(PHINode *Phi, VPValue *Start, VPValue *Step,
const InductionDescriptor &IndDesc,
bool IsScalarAfterVectorization)
: VPHeaderPHIRecipe(VPDef::VPWidenPointerInductionSC, Phi),
IndDesc(IndDesc),
IsScalarAfterVectorization(IsScalarAfterVectorization) {
addOperand(Start);
addOperand(Step);
}
~VPWidenPointerInductionRecipe() override = default;
VPRecipeBase *clone() override {
return new VPWidenPointerInductionRecipe(
cast<PHINode>(getUnderlyingInstr()), getOperand(0), getOperand(1),
IndDesc, IsScalarAfterVectorization);
}
VP_CLASSOF_IMPL(VPDef::VPWidenPointerInductionSC)
/// Generate vector values for the pointer induction.
void execute(VPTransformState &State) override;
/// Returns true if only scalar values will be generated.
bool onlyScalarsGenerated(bool IsScalable);
/// Returns the induction descriptor for the recipe.
const InductionDescriptor &getInductionDescriptor() const { return IndDesc; }
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
};
/// A recipe for handling phis that are widened in the vector loop.
/// In the VPlan native path, all incoming VPValues & VPBasicBlock pairs are
/// managed in the recipe directly.
class VPWidenPHIRecipe : public VPSingleDefRecipe {
/// List of incoming blocks. Only used in the VPlan native path.
SmallVector<VPBasicBlock *, 2> IncomingBlocks;
public:
/// Create a new VPWidenPHIRecipe for \p Phi with start value \p Start.
VPWidenPHIRecipe(PHINode *Phi, VPValue *Start = nullptr)
: VPSingleDefRecipe(VPDef::VPWidenPHISC, ArrayRef<VPValue *>(), Phi) {
if (Start)
addOperand(Start);
}
VPRecipeBase *clone() override {
llvm_unreachable("cloning not implemented yet");
}
~VPWidenPHIRecipe() override = default;
VP_CLASSOF_IMPL(VPDef::VPWidenPHISC)
/// Generate the phi/select nodes.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
/// Adds a pair (\p IncomingV, \p IncomingBlock) to the phi.
void addIncoming(VPValue *IncomingV, VPBasicBlock *IncomingBlock) {
addOperand(IncomingV);
IncomingBlocks.push_back(IncomingBlock);
}
/// Returns the \p I th incoming VPBasicBlock.
VPBasicBlock *getIncomingBlock(unsigned I) { return IncomingBlocks[I]; }
/// Returns the \p I th incoming VPValue.
VPValue *getIncomingValue(unsigned I) { return getOperand(I); }
};
/// A recipe for handling first-order recurrence phis. The start value is the
/// first operand of the recipe and the incoming value from the backedge is the
/// second operand.
struct VPFirstOrderRecurrencePHIRecipe : public VPHeaderPHIRecipe {
VPFirstOrderRecurrencePHIRecipe(PHINode *Phi, VPValue &Start)
: VPHeaderPHIRecipe(VPDef::VPFirstOrderRecurrencePHISC, Phi, &Start) {}
VP_CLASSOF_IMPL(VPDef::VPFirstOrderRecurrencePHISC)
static inline bool classof(const VPHeaderPHIRecipe *R) {
return R->getVPDefID() == VPDef::VPFirstOrderRecurrencePHISC;
}
VPRecipeBase *clone() override {
return new VPFirstOrderRecurrencePHIRecipe(
cast<PHINode>(getUnderlyingInstr()), *getOperand(0));
}
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
};
/// A recipe for handling reduction phis. The start value is the first operand
/// of the recipe and the incoming value from the backedge is the second
/// operand.
class VPReductionPHIRecipe : public VPHeaderPHIRecipe {
/// Descriptor for the reduction.
const RecurrenceDescriptor &RdxDesc;
/// The phi is part of an in-loop reduction.
bool IsInLoop;
/// The phi is part of an ordered reduction. Requires IsInLoop to be true.
bool IsOrdered;
public:
/// Create a new VPReductionPHIRecipe for the reduction \p Phi described by \p
/// RdxDesc.
VPReductionPHIRecipe(PHINode *Phi, const RecurrenceDescriptor &RdxDesc,
VPValue &Start, bool IsInLoop = false,
bool IsOrdered = false)
: VPHeaderPHIRecipe(VPDef::VPReductionPHISC, Phi, &Start),
RdxDesc(RdxDesc), IsInLoop(IsInLoop), IsOrdered(IsOrdered) {
assert((!IsOrdered || IsInLoop) && "IsOrdered requires IsInLoop");
}
~VPReductionPHIRecipe() override = default;
VPRecipeBase *clone() override {
auto *R =
new VPReductionPHIRecipe(cast<PHINode>(getUnderlyingInstr()), RdxDesc,
*getOperand(0), IsInLoop, IsOrdered);
R->addOperand(getBackedgeValue());
return R;
}
VP_CLASSOF_IMPL(VPDef::VPReductionPHISC)
static inline bool classof(const VPHeaderPHIRecipe *R) {
return R->getVPDefID() == VPDef::VPReductionPHISC;
}
/// Generate the phi/select nodes.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
const RecurrenceDescriptor &getRecurrenceDescriptor() const {
return RdxDesc;
}
/// Returns true, if the phi is part of an ordered reduction.
bool isOrdered() const { return IsOrdered; }
/// Returns true, if the phi is part of an in-loop reduction.
bool isInLoop() const { return IsInLoop; }
};
/// A recipe for vectorizing a phi-node as a sequence of mask-based select
/// instructions.
class VPBlendRecipe : public VPSingleDefRecipe {
public:
/// The blend operation is a User of the incoming values and of their
/// respective masks, ordered [I0, M0, I1, M1, ...]. Note that a single value
/// might be incoming with a full mask for which there is no VPValue.
VPBlendRecipe(PHINode *Phi, ArrayRef<VPValue *> Operands)
: VPSingleDefRecipe(VPDef::VPBlendSC, Operands, Phi, Phi->getDebugLoc()) {
assert(Operands.size() > 0 &&
((Operands.size() == 1) || (Operands.size() % 2 == 0)) &&
"Expected either a single incoming value or a positive even number "
"of operands");
}
VPRecipeBase *clone() override {
SmallVector<VPValue *> Ops(operands());
return new VPBlendRecipe(cast<PHINode>(getUnderlyingValue()), Ops);
}
VP_CLASSOF_IMPL(VPDef::VPBlendSC)
/// Return the number of incoming values, taking into account that a single
/// incoming value has no mask.
unsigned getNumIncomingValues() const { return (getNumOperands() + 1) / 2; }
/// Return incoming value number \p Idx.
VPValue *getIncomingValue(unsigned Idx) const { return getOperand(Idx * 2); }
/// Return mask number \p Idx.
VPValue *getMask(unsigned Idx) const { return getOperand(Idx * 2 + 1); }
/// Generate the phi/select nodes.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
/// Returns true if the recipe only uses the first lane of operand \p Op.
bool onlyFirstLaneUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
// Recursing through Blend recipes only, must terminate at header phi's the
// latest.
return all_of(users(),
[this](VPUser *U) { return U->onlyFirstLaneUsed(this); });
}
};
/// VPInterleaveRecipe is a recipe for transforming an interleave group of load
/// or stores into one wide load/store and shuffles. The first operand of a
/// VPInterleave recipe is the address, followed by the stored values, followed
/// by an optional mask.
class VPInterleaveRecipe : public VPRecipeBase {
const InterleaveGroup<Instruction> *IG;
/// Indicates if the interleave group is in a conditional block and requires a
/// mask.
bool HasMask = false;
/// Indicates if gaps between members of the group need to be masked out or if
/// unusued gaps can be loaded speculatively.
bool NeedsMaskForGaps = false;
public:
VPInterleaveRecipe(const InterleaveGroup<Instruction> *IG, VPValue *Addr,
ArrayRef<VPValue *> StoredValues, VPValue *Mask,
bool NeedsMaskForGaps)
: VPRecipeBase(VPDef::VPInterleaveSC, {Addr}), IG(IG),
NeedsMaskForGaps(NeedsMaskForGaps) {
for (unsigned i = 0; i < IG->getFactor(); ++i)
if (Instruction *I = IG->getMember(i)) {
if (I->getType()->isVoidTy())
continue;
new VPValue(I, this);
}
for (auto *SV : StoredValues)
addOperand(SV);
if (Mask) {
HasMask = true;
addOperand(Mask);
}
}
~VPInterleaveRecipe() override = default;
VPRecipeBase *clone() override {
return new VPInterleaveRecipe(IG, getAddr(), getStoredValues(), getMask(),
NeedsMaskForGaps);
}
VP_CLASSOF_IMPL(VPDef::VPInterleaveSC)
/// Return the address accessed by this recipe.
VPValue *getAddr() const {
return getOperand(0); // Address is the 1st, mandatory operand.
}
/// Return the mask used by this recipe. Note that a full mask is represented
/// by a nullptr.
VPValue *getMask() const {
// Mask is optional and therefore the last, currently 2nd operand.
return HasMask ? getOperand(getNumOperands() - 1) : nullptr;
}
/// Return the VPValues stored by this interleave group. If it is a load
/// interleave group, return an empty ArrayRef.
ArrayRef<VPValue *> getStoredValues() const {
// The first operand is the address, followed by the stored values, followed
// by an optional mask.
return ArrayRef<VPValue *>(op_begin(), getNumOperands())
.slice(1, getNumStoreOperands());
}
/// Generate the wide load or store, and shuffles.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
const InterleaveGroup<Instruction> *getInterleaveGroup() { return IG; }
/// Returns the number of stored operands of this interleave group. Returns 0
/// for load interleave groups.
unsigned getNumStoreOperands() const {
return getNumOperands() - (HasMask ? 2 : 1);
}
/// The recipe only uses the first lane of the address.
bool onlyFirstLaneUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return Op == getAddr() && !llvm::is_contained(getStoredValues(), Op);
}
};
/// A recipe to represent inloop reduction operations, performing a reduction on
/// a vector operand into a scalar value, and adding the result to a chain.
/// The Operands are {ChainOp, VecOp, [Condition]}.
class VPReductionRecipe : public VPSingleDefRecipe {
/// The recurrence decriptor for the reduction in question.
const RecurrenceDescriptor &RdxDesc;
public:
VPReductionRecipe(const RecurrenceDescriptor &R, Instruction *I,
VPValue *ChainOp, VPValue *VecOp, VPValue *CondOp)
: VPSingleDefRecipe(VPDef::VPReductionSC,
ArrayRef<VPValue *>({ChainOp, VecOp}), I),
RdxDesc(R) {
if (CondOp)
addOperand(CondOp);
}
~VPReductionRecipe() override = default;
VPRecipeBase *clone() override {
return new VPReductionRecipe(RdxDesc, getUnderlyingInstr(), getChainOp(),
getVecOp(), getCondOp());
}
VP_CLASSOF_IMPL(VPDef::VPReductionSC)
/// Generate the reduction in the loop
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
/// The VPValue of the scalar Chain being accumulated.
VPValue *getChainOp() const { return getOperand(0); }
/// The VPValue of the vector value to be reduced.
VPValue *getVecOp() const { return getOperand(1); }
/// The VPValue of the condition for the block.
VPValue *getCondOp() const {
return getNumOperands() > 2 ? getOperand(2) : nullptr;
}
};
/// VPReplicateRecipe replicates a given instruction producing multiple scalar
/// copies of the original scalar type, one per lane, instead of producing a
/// single copy of widened type for all lanes. If the instruction is known to be
/// uniform only one copy, per lane zero, will be generated.
class VPReplicateRecipe : public VPRecipeWithIRFlags {
/// Indicator if only a single replica per lane is needed.
bool IsUniform;
/// Indicator if the replicas are also predicated.
bool IsPredicated;
public:
template <typename IterT>
VPReplicateRecipe(Instruction *I, iterator_range<IterT> Operands,
bool IsUniform, VPValue *Mask = nullptr)
: VPRecipeWithIRFlags(VPDef::VPReplicateSC, Operands, *I),
IsUniform(IsUniform), IsPredicated(Mask) {
if (Mask)
addOperand(Mask);
}
~VPReplicateRecipe() override = default;
VPRecipeBase *clone() override {
auto *Copy =
new VPReplicateRecipe(getUnderlyingInstr(), operands(), IsUniform,
isPredicated() ? getMask() : nullptr);
Copy->transferFlags(*this);
return Copy;
}
VP_CLASSOF_IMPL(VPDef::VPReplicateSC)
/// Generate replicas of the desired Ingredient. Replicas will be generated
/// for all parts and lanes unless a specific part and lane are specified in
/// the \p State.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
bool isUniform() const { return IsUniform; }
bool isPredicated() const { return IsPredicated; }
/// Returns true if the recipe only uses the first lane of operand \p Op.
bool onlyFirstLaneUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return isUniform();
}
/// Returns true if the recipe uses scalars of operand \p Op.
bool usesScalars(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
/// Returns true if the recipe is used by a widened recipe via an intervening
/// VPPredInstPHIRecipe. In this case, the scalar values should also be packed
/// in a vector.
bool shouldPack() const;
/// Return the mask of a predicated VPReplicateRecipe.
VPValue *getMask() {
assert(isPredicated() && "Trying to get the mask of a unpredicated recipe");
return getOperand(getNumOperands() - 1);
}
};
/// A recipe for generating conditional branches on the bits of a mask.
class VPBranchOnMaskRecipe : public VPRecipeBase {
public:
VPBranchOnMaskRecipe(VPValue *BlockInMask)
: VPRecipeBase(VPDef::VPBranchOnMaskSC, {}) {
if (BlockInMask) // nullptr means all-one mask.
addOperand(BlockInMask);
}
VPRecipeBase *clone() override {
return new VPBranchOnMaskRecipe(getOperand(0));
}
VP_CLASSOF_IMPL(VPDef::VPBranchOnMaskSC)
/// Generate the extraction of the appropriate bit from the block mask and the
/// conditional branch.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override {
O << Indent << "BRANCH-ON-MASK ";
if (VPValue *Mask = getMask())
Mask->printAsOperand(O, SlotTracker);
else
O << " All-One";
}
#endif
/// Return the mask used by this recipe. Note that a full mask is represented
/// by a nullptr.
VPValue *getMask() const {
assert(getNumOperands() <= 1 && "should have either 0 or 1 operands");
// Mask is optional.
return getNumOperands() == 1 ? getOperand(0) : nullptr;
}
/// Returns true if the recipe uses scalars of operand \p Op.
bool usesScalars(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
};
/// VPPredInstPHIRecipe is a recipe for generating the phi nodes needed when
/// control converges back from a Branch-on-Mask. The phi nodes are needed in
/// order to merge values that are set under such a branch and feed their uses.
/// The phi nodes can be scalar or vector depending on the users of the value.
/// This recipe works in concert with VPBranchOnMaskRecipe.
class VPPredInstPHIRecipe : public VPSingleDefRecipe {
public:
/// Construct a VPPredInstPHIRecipe given \p PredInst whose value needs a phi
/// nodes after merging back from a Branch-on-Mask.
VPPredInstPHIRecipe(VPValue *PredV)
: VPSingleDefRecipe(VPDef::VPPredInstPHISC, PredV) {}
~VPPredInstPHIRecipe() override = default;
VPRecipeBase *clone() override {
return new VPPredInstPHIRecipe(getOperand(0));
}
VP_CLASSOF_IMPL(VPDef::VPPredInstPHISC)
/// Generates phi nodes for live-outs as needed to retain SSA form.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
/// Returns true if the recipe uses scalars of operand \p Op.
bool usesScalars(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
};
/// A Recipe for widening load/store operations.
/// The recipe uses the following VPValues:
/// - For load: Address, optional mask
/// - For store: Address, stored value, optional mask
/// TODO: We currently execute only per-part unless a specific instance is
/// provided.
class VPWidenMemoryInstructionRecipe : public VPRecipeBase {
Instruction &Ingredient;
// Whether the loaded-from / stored-to addresses are consecutive.
bool Consecutive;
// Whether the consecutive loaded/stored addresses are in reverse order.
bool Reverse;
void setMask(VPValue *Mask) {
if (!Mask)
return;
addOperand(Mask);
}
bool isMasked() const {
return isStore() ? getNumOperands() == 3 : getNumOperands() == 2;
}
public:
VPWidenMemoryInstructionRecipe(LoadInst &Load, VPValue *Addr, VPValue *Mask,
bool Consecutive, bool Reverse)
: VPRecipeBase(VPDef::VPWidenMemoryInstructionSC, {Addr}),
Ingredient(Load), Consecutive(Consecutive), Reverse(Reverse) {
assert((Consecutive || !Reverse) && "Reverse implies consecutive");
new VPValue(this, &Load);
setMask(Mask);
}
VPWidenMemoryInstructionRecipe(StoreInst &Store, VPValue *Addr,
VPValue *StoredValue, VPValue *Mask,
bool Consecutive, bool Reverse)
: VPRecipeBase(VPDef::VPWidenMemoryInstructionSC, {Addr, StoredValue}),
Ingredient(Store), Consecutive(Consecutive), Reverse(Reverse) {
assert((Consecutive || !Reverse) && "Reverse implies consecutive");
setMask(Mask);
}
VPRecipeBase *clone() override {
if (isStore())
return new VPWidenMemoryInstructionRecipe(
cast<StoreInst>(Ingredient), getAddr(), getStoredValue(), getMask(),
Consecutive, Reverse);
return new VPWidenMemoryInstructionRecipe(
cast<LoadInst>(Ingredient), getAddr(), getMask(), Consecutive, Reverse);
}
VP_CLASSOF_IMPL(VPDef::VPWidenMemoryInstructionSC)
/// Return the address accessed by this recipe.
VPValue *getAddr() const {
return getOperand(0); // Address is the 1st, mandatory operand.
}
/// Return the mask used by this recipe. Note that a full mask is represented
/// by a nullptr.
VPValue *getMask() const {
// Mask is optional and therefore the last operand.
return isMasked() ? getOperand(getNumOperands() - 1) : nullptr;
}
/// Returns true if this recipe is a store.
bool isStore() const { return isa<StoreInst>(Ingredient); }
/// Return the address accessed by this recipe.
VPValue *getStoredValue() const {
assert(isStore() && "Stored value only available for store instructions");
return getOperand(1); // Stored value is the 2nd, mandatory operand.
}
// Return whether the loaded-from / stored-to addresses are consecutive.
bool isConsecutive() const { return Consecutive; }
// Return whether the consecutive loaded/stored addresses are in reverse
// order.
bool isReverse() const { return Reverse; }
/// Generate the wide load/store.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
/// Returns true if the recipe only uses the first lane of operand \p Op.
bool onlyFirstLaneUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
// Widened, consecutive memory operations only demand the first lane of
// their address, unless the same operand is also stored. That latter can
// happen with opaque pointers.
return Op == getAddr() && isConsecutive() &&
(!isStore() || Op != getStoredValue());
}
Instruction &getIngredient() const { return Ingredient; }
};
/// Recipe to expand a SCEV expression.
class VPExpandSCEVRecipe : public VPSingleDefRecipe {
const SCEV *Expr;
ScalarEvolution &SE;
public:
VPExpandSCEVRecipe(const SCEV *Expr, ScalarEvolution &SE)
: VPSingleDefRecipe(VPDef::VPExpandSCEVSC, {}), Expr(Expr), SE(SE) {}
~VPExpandSCEVRecipe() override = default;
VPRecipeBase *clone() override { return new VPExpandSCEVRecipe(Expr, SE); }
VP_CLASSOF_IMPL(VPDef::VPExpandSCEVSC)
/// Generate a canonical vector induction variable of the vector loop, with
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
const SCEV *getSCEV() const { return Expr; }
};
/// Canonical scalar induction phi of the vector loop. Starting at the specified
/// start value (either 0 or the resume value when vectorizing the epilogue
/// loop). VPWidenCanonicalIVRecipe represents the vector version of the
/// canonical induction variable.
class VPCanonicalIVPHIRecipe : public VPHeaderPHIRecipe {
public:
VPCanonicalIVPHIRecipe(VPValue *StartV, DebugLoc DL)
: VPHeaderPHIRecipe(VPDef::VPCanonicalIVPHISC, nullptr, StartV, DL) {}
~VPCanonicalIVPHIRecipe() override = default;
VPRecipeBase *clone() override {
auto *R = new VPCanonicalIVPHIRecipe(getOperand(0), getDebugLoc());
R->addOperand(getBackedgeValue());
return R;
}
VP_CLASSOF_IMPL(VPDef::VPCanonicalIVPHISC)
static inline bool classof(const VPHeaderPHIRecipe *D) {
return D->getVPDefID() == VPDef::VPCanonicalIVPHISC;
}
/// Generate the canonical scalar induction phi of the vector loop.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
/// Returns the scalar type of the induction.
Type *getScalarType() const {
return getStartValue()->getLiveInIRValue()->getType();
}
/// Returns true if the recipe only uses the first lane of operand \p Op.
bool onlyFirstLaneUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
/// Returns true if the recipe only uses the first part of operand \p Op.
bool onlyFirstPartUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
/// Check if the induction described by \p Kind, /p Start and \p Step is
/// canonical, i.e. has the same start and step (of 1) as the canonical IV.
bool isCanonical(InductionDescriptor::InductionKind Kind, VPValue *Start,
VPValue *Step) const;
};
/// A recipe for generating the active lane mask for the vector loop that is
/// used to predicate the vector operations.
/// TODO: It would be good to use the existing VPWidenPHIRecipe instead and
/// remove VPActiveLaneMaskPHIRecipe.
class VPActiveLaneMaskPHIRecipe : public VPHeaderPHIRecipe {
public:
VPActiveLaneMaskPHIRecipe(VPValue *StartMask, DebugLoc DL)
: VPHeaderPHIRecipe(VPDef::VPActiveLaneMaskPHISC, nullptr, StartMask,
DL) {}
~VPActiveLaneMaskPHIRecipe() override = default;
VPRecipeBase *clone() override {
return new VPActiveLaneMaskPHIRecipe(getOperand(0), getDebugLoc());
}
VP_CLASSOF_IMPL(VPDef::VPActiveLaneMaskPHISC)
static inline bool classof(const VPHeaderPHIRecipe *D) {
return D->getVPDefID() == VPDef::VPActiveLaneMaskPHISC;
}
/// Generate the active lane mask phi of the vector loop.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
};
/// A Recipe for widening the canonical induction variable of the vector loop.
class VPWidenCanonicalIVRecipe : public VPSingleDefRecipe {
public:
VPWidenCanonicalIVRecipe(VPCanonicalIVPHIRecipe *CanonicalIV)
: VPSingleDefRecipe(VPDef::VPWidenCanonicalIVSC, {CanonicalIV}) {}
~VPWidenCanonicalIVRecipe() override = default;
VPRecipeBase *clone() override {
return new VPWidenCanonicalIVRecipe(
cast<VPCanonicalIVPHIRecipe>(getOperand(0)));
}
VP_CLASSOF_IMPL(VPDef::VPWidenCanonicalIVSC)
/// Generate a canonical vector induction variable of the vector loop, with
/// start = {<Part*VF, Part*VF+1, ..., Part*VF+VF-1> for 0 <= Part < UF}, and
/// step = <VF*UF, VF*UF, ..., VF*UF>.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
/// Returns the scalar type of the induction.
const Type *getScalarType() const {
return cast<VPCanonicalIVPHIRecipe>(getOperand(0)->getDefiningRecipe())
->getScalarType();
}
};
/// A recipe for converting the canonical IV value to the corresponding value of
/// an IV with different start and step values, using Start + CanonicalIV *
/// Step.
class VPDerivedIVRecipe : public VPSingleDefRecipe {
/// Kind of the induction.
const InductionDescriptor::InductionKind Kind;
/// If not nullptr, the floating point induction binary operator. Must be set
/// for floating point inductions.
const FPMathOperator *FPBinOp;
VPDerivedIVRecipe(InductionDescriptor::InductionKind Kind,
const FPMathOperator *FPBinOp, VPValue *Start,
VPCanonicalIVPHIRecipe *CanonicalIV, VPValue *Step)
: VPSingleDefRecipe(VPDef::VPDerivedIVSC, {Start, CanonicalIV, Step}),
Kind(Kind), FPBinOp(FPBinOp) {}
public:
VPDerivedIVRecipe(const InductionDescriptor &IndDesc, VPValue *Start,
VPCanonicalIVPHIRecipe *CanonicalIV, VPValue *Step)
: VPDerivedIVRecipe(
IndDesc.getKind(),
dyn_cast_or_null<FPMathOperator>(IndDesc.getInductionBinOp()),
Start, CanonicalIV, Step) {}
~VPDerivedIVRecipe() override = default;
VPRecipeBase *clone() override {
return new VPDerivedIVRecipe(Kind, FPBinOp, getStartValue(),
getCanonicalIV(), getStepValue());
}
VP_CLASSOF_IMPL(VPDef::VPDerivedIVSC)
/// Generate the transformed value of the induction at offset StartValue (1.
/// operand) + IV (2. operand) * StepValue (3, operand).
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
Type *getScalarType() const {
return getStartValue()->getLiveInIRValue()->getType();
}
VPValue *getStartValue() const { return getOperand(0); }
VPCanonicalIVPHIRecipe *getCanonicalIV() const {
return cast<VPCanonicalIVPHIRecipe>(getOperand(1));
}
VPValue *getStepValue() const { return getOperand(2); }
/// Returns true if the recipe only uses the first lane of operand \p Op.
bool onlyFirstLaneUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
};
/// A recipe for handling phi nodes of integer and floating-point inductions,
/// producing their scalar values.
class VPScalarIVStepsRecipe : public VPRecipeWithIRFlags {
Instruction::BinaryOps InductionOpcode;
public:
VPScalarIVStepsRecipe(VPValue *IV, VPValue *Step,
Instruction::BinaryOps Opcode, FastMathFlags FMFs)
: VPRecipeWithIRFlags(VPDef::VPScalarIVStepsSC,
ArrayRef<VPValue *>({IV, Step}), FMFs),
InductionOpcode(Opcode) {}
VPScalarIVStepsRecipe(const InductionDescriptor &IndDesc, VPValue *IV,
VPValue *Step)
: VPScalarIVStepsRecipe(
IV, Step, IndDesc.getInductionOpcode(),
dyn_cast_or_null<FPMathOperator>(IndDesc.getInductionBinOp())
? IndDesc.getInductionBinOp()->getFastMathFlags()
: FastMathFlags()) {}
~VPScalarIVStepsRecipe() override = default;
VPRecipeBase *clone() override {
return new VPScalarIVStepsRecipe(
getOperand(0), getOperand(1), InductionOpcode,
hasFastMathFlags() ? getFastMathFlags() : FastMathFlags());
}
VP_CLASSOF_IMPL(VPDef::VPScalarIVStepsSC)
/// Generate the scalarized versions of the phi node as needed by their users.
void execute(VPTransformState &State) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the recipe.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
#endif
VPValue *getStepValue() const { return getOperand(1); }
/// Returns true if the recipe only uses the first lane of operand \p Op.
bool onlyFirstLaneUsed(const VPValue *Op) const override {
assert(is_contained(operands(), Op) &&
"Op must be an operand of the recipe");
return true;
}
};
/// VPBasicBlock serves as the leaf of the Hierarchical Control-Flow Graph. It
/// holds a sequence of zero or more VPRecipe's each representing a sequence of
/// output IR instructions. All PHI-like recipes must come before any non-PHI recipes.
class VPBasicBlock : public VPBlockBase {
public:
using RecipeListTy = iplist<VPRecipeBase>;
private:
/// The VPRecipes held in the order of output instructions to generate.
RecipeListTy Recipes;
public:
VPBasicBlock(const Twine &Name = "", VPRecipeBase *Recipe = nullptr)
: VPBlockBase(VPBasicBlockSC, Name.str()) {
if (Recipe)
appendRecipe(Recipe);
}
~VPBasicBlock() override {
while (!Recipes.empty())
Recipes.pop_back();
}
/// Instruction iterators...
using iterator = RecipeListTy::iterator;
using const_iterator = RecipeListTy::const_iterator;
using reverse_iterator = RecipeListTy::reverse_iterator;
using const_reverse_iterator = RecipeListTy::const_reverse_iterator;
//===--------------------------------------------------------------------===//
/// Recipe iterator methods
///
inline iterator begin() { return Recipes.begin(); }
inline const_iterator begin() const { return Recipes.begin(); }
inline iterator end() { return Recipes.end(); }
inline const_iterator end() const { return Recipes.end(); }
inline reverse_iterator rbegin() { return Recipes.rbegin(); }
inline const_reverse_iterator rbegin() const { return Recipes.rbegin(); }
inline reverse_iterator rend() { return Recipes.rend(); }
inline const_reverse_iterator rend() const { return Recipes.rend(); }
inline size_t size() const { return Recipes.size(); }
inline bool empty() const { return Recipes.empty(); }
inline const VPRecipeBase &front() const { return Recipes.front(); }
inline VPRecipeBase &front() { return Recipes.front(); }
inline const VPRecipeBase &back() const { return Recipes.back(); }
inline VPRecipeBase &back() { return Recipes.back(); }
/// Returns a reference to the list of recipes.
RecipeListTy &getRecipeList() { return Recipes; }
/// Returns a pointer to a member of the recipe list.
static RecipeListTy VPBasicBlock::*getSublistAccess(VPRecipeBase *) {
return &VPBasicBlock::Recipes;
}
/// Method to support type inquiry through isa, cast, and dyn_cast.
static inline bool classof(const VPBlockBase *V) {
return V->getVPBlockID() == VPBlockBase::VPBasicBlockSC;
}
void insert(VPRecipeBase *Recipe, iterator InsertPt) {
assert(Recipe && "No recipe to append.");
assert(!Recipe->Parent && "Recipe already in VPlan");
Recipe->Parent = this;
Recipes.insert(InsertPt, Recipe);
}
/// Augment the existing recipes of a VPBasicBlock with an additional
/// \p Recipe as the last recipe.
void appendRecipe(VPRecipeBase *Recipe) { insert(Recipe, end()); }
/// The method which generates the output IR instructions that correspond to
/// this VPBasicBlock, thereby "executing" the VPlan.
void execute(VPTransformState *State) override;
/// Return the position of the first non-phi node recipe in the block.
iterator getFirstNonPhi();
/// Returns an iterator range over the PHI-like recipes in the block.
iterator_range<iterator> phis() {
return make_range(begin(), getFirstNonPhi());
}
void dropAllReferences(VPValue *NewValue) override;
/// Split current block at \p SplitAt by inserting a new block between the
/// current block and its successors and moving all recipes starting at
/// SplitAt to the new block. Returns the new block.
VPBasicBlock *splitAt(iterator SplitAt);
VPRegionBlock *getEnclosingLoopRegion();
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print this VPBsicBlock to \p O, prefixing all lines with \p Indent. \p
/// SlotTracker is used to print unnamed VPValue's using consequtive numbers.
///
/// Note that the numbering is applied to the whole VPlan, so printing
/// individual blocks is consistent with the whole VPlan printing.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
using VPBlockBase::print; // Get the print(raw_stream &O) version.
#endif
/// If the block has multiple successors, return the branch recipe terminating
/// the block. If there are no or only a single successor, return nullptr;
VPRecipeBase *getTerminator();
const VPRecipeBase *getTerminator() const;
/// Returns true if the block is exiting it's parent region.
bool isExiting() const;
/// Clone the current block and it's recipes, without updating the operands of
/// the cloned recipes.
VPBasicBlock *clone() override {
auto *NewBlock = new VPBasicBlock(getName());
for (VPRecipeBase &R : *this)
NewBlock->appendRecipe(R.clone());
return NewBlock;
}
private:
/// Create an IR BasicBlock to hold the output instructions generated by this
/// VPBasicBlock, and return it. Update the CFGState accordingly.
BasicBlock *createEmptyBasicBlock(VPTransformState::CFGState &CFG);
};
/// VPRegionBlock represents a collection of VPBasicBlocks and VPRegionBlocks
/// which form a Single-Entry-Single-Exiting subgraph of the output IR CFG.
/// A VPRegionBlock may indicate that its contents are to be replicated several
/// times. This is designed to support predicated scalarization, in which a
/// scalar if-then code structure needs to be generated VF * UF times. Having
/// this replication indicator helps to keep a single model for multiple
/// candidate VF's. The actual replication takes place only once the desired VF
/// and UF have been determined.
class VPRegionBlock : public VPBlockBase {
/// Hold the Single Entry of the SESE region modelled by the VPRegionBlock.
VPBlockBase *Entry;
/// Hold the Single Exiting block of the SESE region modelled by the
/// VPRegionBlock.
VPBlockBase *Exiting;
/// An indicator whether this region is to generate multiple replicated
/// instances of output IR corresponding to its VPBlockBases.
bool IsReplicator;
public:
VPRegionBlock(VPBlockBase *Entry, VPBlockBase *Exiting,
const std::string &Name = "", bool IsReplicator = false)
: VPBlockBase(VPRegionBlockSC, Name), Entry(Entry), Exiting(Exiting),
IsReplicator(IsReplicator) {
assert(Entry->getPredecessors().empty() && "Entry block has predecessors.");
assert(Exiting->getSuccessors().empty() && "Exit block has successors.");
Entry->setParent(this);
Exiting->setParent(this);
}
VPRegionBlock(const std::string &Name = "", bool IsReplicator = false)
: VPBlockBase(VPRegionBlockSC, Name), Entry(nullptr), Exiting(nullptr),
IsReplicator(IsReplicator) {}
~VPRegionBlock() override {
if (Entry) {
VPValue DummyValue;
Entry->dropAllReferences(&DummyValue);
deleteCFG(Entry);
}
}
/// Method to support type inquiry through isa, cast, and dyn_cast.
static inline bool classof(const VPBlockBase *V) {
return V->getVPBlockID() == VPBlockBase::VPRegionBlockSC;
}
const VPBlockBase *getEntry() const { return Entry; }
VPBlockBase *getEntry() { return Entry; }
/// Set \p EntryBlock as the entry VPBlockBase of this VPRegionBlock. \p
/// EntryBlock must have no predecessors.
void setEntry(VPBlockBase *EntryBlock) {
assert(EntryBlock->getPredecessors().empty() &&
"Entry block cannot have predecessors.");
Entry = EntryBlock;
EntryBlock->setParent(this);
}
const VPBlockBase *getExiting() const { return Exiting; }
VPBlockBase *getExiting() { return Exiting; }
/// Set \p ExitingBlock as the exiting VPBlockBase of this VPRegionBlock. \p
/// ExitingBlock must have no successors.
void setExiting(VPBlockBase *ExitingBlock) {
assert(ExitingBlock->getSuccessors().empty() &&
"Exit block cannot have successors.");
Exiting = ExitingBlock;
ExitingBlock->setParent(this);
}
/// Returns the pre-header VPBasicBlock of the loop region.
VPBasicBlock *getPreheaderVPBB() {
assert(!isReplicator() && "should only get pre-header of loop regions");
return getSinglePredecessor()->getExitingBasicBlock();
}
/// An indicator whether this region is to generate multiple replicated
/// instances of output IR corresponding to its VPBlockBases.
bool isReplicator() const { return IsReplicator; }
/// The method which generates the output IR instructions that correspond to
/// this VPRegionBlock, thereby "executing" the VPlan.
void execute(VPTransformState *State) override;
void dropAllReferences(VPValue *NewValue) override;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print this VPRegionBlock to \p O (recursively), prefixing all lines with
/// \p Indent. \p SlotTracker is used to print unnamed VPValue's using
/// consequtive numbers.
///
/// Note that the numbering is applied to the whole VPlan, so printing
/// individual regions is consistent with the whole VPlan printing.
void print(raw_ostream &O, const Twine &Indent,
VPSlotTracker &SlotTracker) const override;
using VPBlockBase::print; // Get the print(raw_stream &O) version.
#endif
/// Clone all blocks in the single-entry single-exit region of the block and
/// their recipes without updating the operands of the cloned recipes.
VPRegionBlock *clone() override;
};
/// VPlan models a candidate for vectorization, encoding various decisions take
/// to produce efficient output IR, including which branches, basic-blocks and
/// output IR instructions to generate, and their cost. VPlan holds a
/// Hierarchical-CFG of VPBasicBlocks and VPRegionBlocks rooted at an Entry
/// VPBasicBlock.
class VPlan {
friend class VPlanPrinter;
friend class VPSlotTracker;
/// Hold the single entry to the Hierarchical CFG of the VPlan, i.e. the
/// preheader of the vector loop.
VPBasicBlock *Entry;
/// VPBasicBlock corresponding to the original preheader. Used to place
/// VPExpandSCEV recipes for expressions used during skeleton creation and the
/// rest of VPlan execution.
VPBasicBlock *Preheader;
/// Holds the VFs applicable to this VPlan.
SmallSetVector<ElementCount, 2> VFs;
/// Holds the UFs applicable to this VPlan. If empty, the VPlan is valid for
/// any UF.
SmallSetVector<unsigned, 2> UFs;
/// Holds the name of the VPlan, for printing.
std::string Name;
/// Represents the trip count of the original loop, for folding
/// the tail.
VPValue *TripCount = nullptr;
/// Represents the backedge taken count of the original loop, for folding
/// the tail. It equals TripCount - 1.
VPValue *BackedgeTakenCount = nullptr;
/// Represents the vector trip count.
VPValue VectorTripCount;
/// Represents the loop-invariant VF * UF of the vector loop region.
VPValue VFxUF;
/// Holds a mapping between Values and their corresponding VPValue inside
/// VPlan.
Value2VPValueTy Value2VPValue;
/// Contains all the external definitions created for this VPlan. External
/// definitions are VPValues that hold a pointer to their underlying IR.
SmallVector<VPValue *, 16> VPLiveInsToFree;
/// Indicates whether it is safe use the Value2VPValue mapping or if the
/// mapping cannot be used any longer, because it is stale.
bool Value2VPValueEnabled = true;
/// Values used outside the plan.
MapVector<PHINode *, VPLiveOut *> LiveOuts;
/// Mapping from SCEVs to the VPValues representing their expansions.
/// NOTE: This mapping is temporary and will be removed once all users have
/// been modeled in VPlan directly.
DenseMap<const SCEV *, VPValue *> SCEVToExpansion;
public:
/// Construct a VPlan with original preheader \p Preheader, trip count \p TC
/// and \p Entry to the plan. At the moment, \p Preheader and \p Entry need to
/// be disconnected, as the bypass blocks between them are not yet modeled in
/// VPlan.
VPlan(VPBasicBlock *Preheader, VPValue *TC, VPBasicBlock *Entry)
: VPlan(Preheader, Entry) {
TripCount = TC;
}
/// Construct a VPlan with original preheader \p Preheader and \p Entry to
/// the plan. At the moment, \p Preheader and \p Entry need to be
/// disconnected, as the bypass blocks between them are not yet modeled in
/// VPlan.
VPlan(VPBasicBlock *Preheader, VPBasicBlock *Entry)
: Entry(Entry), Preheader(Preheader) {
Entry->setPlan(this);
Preheader->setPlan(this);
assert(Preheader->getNumSuccessors() == 0 &&
Preheader->getNumPredecessors() == 0 &&
"preheader must be disconnected");
}
~VPlan();
/// Create initial VPlan skeleton, having an "entry" VPBasicBlock (wrapping
/// original scalar pre-header) which contains SCEV expansions that need to
/// happen before the CFG is modified; a VPBasicBlock for the vector
/// pre-header, followed by a region for the vector loop, followed by the
/// middle VPBasicBlock.
static VPlanPtr createInitialVPlan(const SCEV *TripCount,
ScalarEvolution &PSE);
/// Prepare the plan for execution, setting up the required live-in values.
void prepareToExecute(Value *TripCount, Value *VectorTripCount,
Value *CanonicalIVStartValue, VPTransformState &State);
/// Generate the IR code for this VPlan.
void execute(VPTransformState *State);
VPBasicBlock *getEntry() { return Entry; }
const VPBasicBlock *getEntry() const { return Entry; }
/// The trip count of the original loop.
VPValue *getTripCount() const {
assert(TripCount && "trip count needs to be set before accessing it");
return TripCount;
}
/// The backedge taken count of the original loop.
VPValue *getOrCreateBackedgeTakenCount() {
if (!BackedgeTakenCount)
BackedgeTakenCount = new VPValue();
return BackedgeTakenCount;
}
/// The vector trip count.
VPValue &getVectorTripCount() { return VectorTripCount; }
/// Returns VF * UF of the vector loop region.
VPValue &getVFxUF() { return VFxUF; }
/// Mark the plan to indicate that using Value2VPValue is not safe any
/// longer, because it may be stale.
void disableValue2VPValue() { Value2VPValueEnabled = false; }
void addVF(ElementCount VF) { VFs.insert(VF); }
void setVF(ElementCount VF) {
assert(hasVF(VF) && "Cannot set VF not already in plan");
VFs.clear();
VFs.insert(VF);
}
bool hasVF(ElementCount VF) { return VFs.count(VF); }
bool hasScalableVF() {
return any_of(VFs, [](ElementCount VF) { return VF.isScalable(); });
}
bool hasScalarVFOnly() const { return VFs.size() == 1 && VFs[0].isScalar(); }
bool hasUF(unsigned UF) const { return UFs.empty() || UFs.contains(UF); }
void setUF(unsigned UF) {
assert(hasUF(UF) && "Cannot set the UF not already in plan");
UFs.clear();
UFs.insert(UF);
}
/// Return a string with the name of the plan and the applicable VFs and UFs.
std::string getName() const;
void setName(const Twine &newName) { Name = newName.str(); }
void addVPValue(Value *V, VPValue *VPV) {
assert((Value2VPValueEnabled || VPV->isLiveIn()) &&
"Value2VPValue mapping may be out of date!");
assert(V && "Trying to add a null Value to VPlan");
assert(!Value2VPValue.count(V) && "Value already exists in VPlan");
Value2VPValue[V] = VPV;
}
/// Returns the VPValue for \p V. \p OverrideAllowed can be used to disable
/// /// checking whether it is safe to query VPValues using IR Values.
VPValue *getVPValue(Value *V, bool OverrideAllowed = false) {
assert(V && "Trying to get the VPValue of a null Value");
assert(Value2VPValue.count(V) && "Value does not exist in VPlan");
assert((Value2VPValueEnabled || OverrideAllowed ||
Value2VPValue[V]->isLiveIn()) &&
"Value2VPValue mapping may be out of date!");
return Value2VPValue[V];
}
/// Gets the VPValue for \p V or adds a new live-in (if none exists yet) for
/// \p V.
VPValue *getVPValueOrAddLiveIn(Value *V) {
assert(V && "Trying to get or add the VPValue of a null Value");
if (!Value2VPValue.count(V)) {
VPValue *VPV = new VPValue(V);
VPLiveInsToFree.push_back(VPV);
addVPValue(V, VPV);
}
return getVPValue(V);
}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print the live-ins of this VPlan to \p O.
void printLiveIns(raw_ostream &O) const;
/// Print this VPlan to \p O.
void print(raw_ostream &O) const;
/// Print this VPlan in DOT format to \p O.
void printDOT(raw_ostream &O) const;
/// Dump the plan to stderr (for debugging).
LLVM_DUMP_METHOD void dump() const;
#endif
/// Returns a range mapping the values the range \p Operands to their
/// corresponding VPValues.
iterator_range<mapped_iterator<Use *, std::function<VPValue *(Value *)>>>
mapToVPValues(User::op_range Operands) {
std::function<VPValue *(Value *)> Fn = [this](Value *Op) {
return getVPValueOrAddLiveIn(Op);
};
return map_range(Operands, Fn);
}
/// Returns the VPRegionBlock of the vector loop.
VPRegionBlock *getVectorLoopRegion() {
return cast<VPRegionBlock>(getEntry()->getSingleSuccessor());
}
const VPRegionBlock *getVectorLoopRegion() const {
return cast<VPRegionBlock>(getEntry()->getSingleSuccessor());
}
/// Returns the canonical induction recipe of the vector loop.
VPCanonicalIVPHIRecipe *getCanonicalIV() {
VPBasicBlock *EntryVPBB = getVectorLoopRegion()->getEntryBasicBlock();
if (EntryVPBB->empty()) {
// VPlan native path.
EntryVPBB = cast<VPBasicBlock>(EntryVPBB->getSingleSuccessor());
}
return cast<VPCanonicalIVPHIRecipe>(&*EntryVPBB->begin());
}
void addLiveOut(PHINode *PN, VPValue *V);
void removeLiveOut(PHINode *PN) {
delete LiveOuts[PN];
LiveOuts.erase(PN);
}
const MapVector<PHINode *, VPLiveOut *> &getLiveOuts() const {
return LiveOuts;
}
VPValue *getSCEVExpansion(const SCEV *S) const {
return SCEVToExpansion.lookup(S);
}
void addSCEVExpansion(const SCEV *S, VPValue *V) {
assert(!SCEVToExpansion.contains(S) && "SCEV already expanded");
SCEVToExpansion[S] = V;
}
/// \return The block corresponding to the original preheader.
VPBasicBlock *getPreheader() { return Preheader; }
const VPBasicBlock *getPreheader() const { return Preheader; }
/// Clone the current VPlan, update all VPValues of the new VPlan and cloned
/// recipes to refer to the clones, and return it.
VPlan *duplicate();
private:
/// Add to the given dominator tree the header block and every new basic block
/// that was created between it and the latch block, inclusive.
static void updateDominatorTree(DominatorTree *DT, BasicBlock *LoopLatchBB,
BasicBlock *LoopPreHeaderBB,
BasicBlock *LoopExitBB);
};
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// VPlanPrinter prints a given VPlan to a given output stream. The printing is
/// indented and follows the dot format.
class VPlanPrinter {
raw_ostream &OS;
const VPlan &Plan;
unsigned Depth = 0;
unsigned TabWidth = 2;
std::string Indent;
unsigned BID = 0;
SmallDenseMap<const VPBlockBase *, unsigned> BlockID;
VPSlotTracker SlotTracker;
/// Handle indentation.
void bumpIndent(int b) { Indent = std::string((Depth += b) * TabWidth, ' '); }
/// Print a given \p Block of the Plan.
void dumpBlock(const VPBlockBase *Block);
/// Print the information related to the CFG edges going out of a given
/// \p Block, followed by printing the successor blocks themselves.
void dumpEdges(const VPBlockBase *Block);
/// Print a given \p BasicBlock, including its VPRecipes, followed by printing
/// its successor blocks.
void dumpBasicBlock(const VPBasicBlock *BasicBlock);
/// Print a given \p Region of the Plan.
void dumpRegion(const VPRegionBlock *Region);
unsigned getOrCreateBID(const VPBlockBase *Block) {
return BlockID.count(Block) ? BlockID[Block] : BlockID[Block] = BID++;
}
Twine getOrCreateName(const VPBlockBase *Block);
Twine getUID(const VPBlockBase *Block);
/// Print the information related to a CFG edge between two VPBlockBases.
void drawEdge(const VPBlockBase *From, const VPBlockBase *To, bool Hidden,
const Twine &Label);
public:
VPlanPrinter(raw_ostream &O, const VPlan &P)
: OS(O), Plan(P), SlotTracker(&P) {}
LLVM_DUMP_METHOD void dump();
};
struct VPlanIngredient {
const Value *V;
VPlanIngredient(const Value *V) : V(V) {}
void print(raw_ostream &O) const;
};
inline raw_ostream &operator<<(raw_ostream &OS, const VPlanIngredient &I) {
I.print(OS);
return OS;
}
inline raw_ostream &operator<<(raw_ostream &OS, const VPlan &Plan) {
Plan.print(OS);
return OS;
}
#endif
//===----------------------------------------------------------------------===//
// VPlan Utilities
//===----------------------------------------------------------------------===//
/// Class that provides utilities for VPBlockBases in VPlan.
class VPBlockUtils {
public:
VPBlockUtils() = delete;
/// Insert disconnected VPBlockBase \p NewBlock after \p BlockPtr. Add \p
/// NewBlock as successor of \p BlockPtr and \p BlockPtr as predecessor of \p
/// NewBlock, and propagate \p BlockPtr parent to \p NewBlock. \p BlockPtr's
/// successors are moved from \p BlockPtr to \p NewBlock. \p NewBlock must
/// have neither successors nor predecessors.
static void insertBlockAfter(VPBlockBase *NewBlock, VPBlockBase *BlockPtr) {
assert(NewBlock->getSuccessors().empty() &&
NewBlock->getPredecessors().empty() &&
"Can't insert new block with predecessors or successors.");
NewBlock->setParent(BlockPtr->getParent());
SmallVector<VPBlockBase *> Succs(BlockPtr->successors());
for (VPBlockBase *Succ : Succs) {
disconnectBlocks(BlockPtr, Succ);
connectBlocks(NewBlock, Succ);
}
connectBlocks(BlockPtr, NewBlock);
}
/// Insert disconnected VPBlockBases \p IfTrue and \p IfFalse after \p
/// BlockPtr. Add \p IfTrue and \p IfFalse as succesors of \p BlockPtr and \p
/// BlockPtr as predecessor of \p IfTrue and \p IfFalse. Propagate \p BlockPtr
/// parent to \p IfTrue and \p IfFalse. \p BlockPtr must have no successors
/// and \p IfTrue and \p IfFalse must have neither successors nor
/// predecessors.
static void insertTwoBlocksAfter(VPBlockBase *IfTrue, VPBlockBase *IfFalse,
VPBlockBase *BlockPtr) {
assert(IfTrue->getSuccessors().empty() &&
"Can't insert IfTrue with successors.");
assert(IfFalse->getSuccessors().empty() &&
"Can't insert IfFalse with successors.");
BlockPtr->setTwoSuccessors(IfTrue, IfFalse);
IfTrue->setPredecessors({BlockPtr});
IfFalse->setPredecessors({BlockPtr});
IfTrue->setParent(BlockPtr->getParent());
IfFalse->setParent(BlockPtr->getParent());
}
/// Connect VPBlockBases \p From and \p To bi-directionally. Append \p To to
/// the successors of \p From and \p From to the predecessors of \p To. Both
/// VPBlockBases must have the same parent, which can be null. Both
/// VPBlockBases can be already connected to other VPBlockBases.
static void connectBlocks(VPBlockBase *From, VPBlockBase *To) {
assert((From->getParent() == To->getParent()) &&
"Can't connect two block with different parents");
assert(From->getNumSuccessors() < 2 &&
"Blocks can't have more than two successors.");
From->appendSuccessor(To);
To->appendPredecessor(From);
}
/// Disconnect VPBlockBases \p From and \p To bi-directionally. Remove \p To
/// from the successors of \p From and \p From from the predecessors of \p To.
static void disconnectBlocks(VPBlockBase *From, VPBlockBase *To) {
assert(To && "Successor to disconnect is null.");
From->removeSuccessor(To);
To->removePredecessor(From);
}
/// Return an iterator range over \p Range which only includes \p BlockTy
/// blocks. The accesses are casted to \p BlockTy.
template <typename BlockTy, typename T>
static auto blocksOnly(const T &Range) {
// Create BaseTy with correct const-ness based on BlockTy.
using BaseTy = std::conditional_t<std::is_const<BlockTy>::value,
const VPBlockBase, VPBlockBase>;
// We need to first create an iterator range over (const) BlocktTy & instead
// of (const) BlockTy * for filter_range to work properly.
auto Mapped =
map_range(Range, [](BaseTy *Block) -> BaseTy & { return *Block; });
auto Filter = make_filter_range(
Mapped, [](BaseTy &Block) { return isa<BlockTy>(&Block); });
return map_range(Filter, [](BaseTy &Block) -> BlockTy * {
return cast<BlockTy>(&Block);
});
}
};
class VPInterleavedAccessInfo {
DenseMap<VPInstruction *, InterleaveGroup<VPInstruction> *>
InterleaveGroupMap;
/// Type for mapping of instruction based interleave groups to VPInstruction
/// interleave groups
using Old2NewTy = DenseMap<InterleaveGroup<Instruction> *,
InterleaveGroup<VPInstruction> *>;
/// Recursively \p Region and populate VPlan based interleave groups based on
/// \p IAI.
void visitRegion(VPRegionBlock *Region, Old2NewTy &Old2New,
InterleavedAccessInfo &IAI);
/// Recursively traverse \p Block and populate VPlan based interleave groups
/// based on \p IAI.
void visitBlock(VPBlockBase *Block, Old2NewTy &Old2New,
InterleavedAccessInfo &IAI);
public:
VPInterleavedAccessInfo(VPlan &Plan, InterleavedAccessInfo &IAI);
~VPInterleavedAccessInfo() {
SmallPtrSet<InterleaveGroup<VPInstruction> *, 4> DelSet;
// Avoid releasing a pointer twice.
for (auto &I : InterleaveGroupMap)
DelSet.insert(I.second);
for (auto *Ptr : DelSet)
delete Ptr;
}
/// Get the interleave group that \p Instr belongs to.
///
/// \returns nullptr if doesn't have such group.
InterleaveGroup<VPInstruction> *
getInterleaveGroup(VPInstruction *Instr) const {
return InterleaveGroupMap.lookup(Instr);
}
};
/// Class that maps (parts of) an existing VPlan to trees of combined
/// VPInstructions.
class VPlanSlp {
enum class OpMode { Failed, Load, Opcode };
/// A DenseMapInfo implementation for using SmallVector<VPValue *, 4> as
/// DenseMap keys.
struct BundleDenseMapInfo {
static SmallVector<VPValue *, 4> getEmptyKey() {
return {reinterpret_cast<VPValue *>(-1)};
}
static SmallVector<VPValue *, 4> getTombstoneKey() {
return {reinterpret_cast<VPValue *>(-2)};
}
static unsigned getHashValue(const SmallVector<VPValue *, 4> &V) {
return static_cast<unsigned>(hash_combine_range(V.begin(), V.end()));
}
static bool isEqual(const SmallVector<VPValue *, 4> &LHS,
const SmallVector<VPValue *, 4> &RHS) {
return LHS == RHS;
}
};
/// Mapping of values in the original VPlan to a combined VPInstruction.
DenseMap<SmallVector<VPValue *, 4>, VPInstruction *, BundleDenseMapInfo>
BundleToCombined;
VPInterleavedAccessInfo &IAI;
/// Basic block to operate on. For now, only instructions in a single BB are
/// considered.
const VPBasicBlock &BB;
/// Indicates whether we managed to combine all visited instructions or not.
bool CompletelySLP = true;
/// Width of the widest combined bundle in bits.
unsigned WidestBundleBits = 0;
using MultiNodeOpTy =
typename std::pair<VPInstruction *, SmallVector<VPValue *, 4>>;
// Input operand bundles for the current multi node. Each multi node operand
// bundle contains values not matching the multi node's opcode. They will
// be reordered in reorderMultiNodeOps, once we completed building a
// multi node.
SmallVector<MultiNodeOpTy, 4> MultiNodeOps;
/// Indicates whether we are building a multi node currently.
bool MultiNodeActive = false;
/// Check if we can vectorize Operands together.
bool areVectorizable(ArrayRef<VPValue *> Operands) const;
/// Add combined instruction \p New for the bundle \p Operands.
void addCombined(ArrayRef<VPValue *> Operands, VPInstruction *New);
/// Indicate we hit a bundle we failed to combine. Returns nullptr for now.
VPInstruction *markFailed();
/// Reorder operands in the multi node to maximize sequential memory access
/// and commutative operations.
SmallVector<MultiNodeOpTy, 4> reorderMultiNodeOps();
/// Choose the best candidate to use for the lane after \p Last. The set of
/// candidates to choose from are values with an opcode matching \p Last's
/// or loads consecutive to \p Last.
std::pair<OpMode, VPValue *> getBest(OpMode Mode, VPValue *Last,
SmallPtrSetImpl<VPValue *> &Candidates,
VPInterleavedAccessInfo &IAI);
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
/// Print bundle \p Values to dbgs().
void dumpBundle(ArrayRef<VPValue *> Values);
#endif
public:
VPlanSlp(VPInterleavedAccessInfo &IAI, VPBasicBlock &BB) : IAI(IAI), BB(BB) {}
~VPlanSlp() = default;
/// Tries to build an SLP tree rooted at \p Operands and returns a
/// VPInstruction combining \p Operands, if they can be combined.
VPInstruction *buildGraph(ArrayRef<VPValue *> Operands);
/// Return the width of the widest combined bundle in bits.
unsigned getWidestBundleBits() const { return WidestBundleBits; }
/// Return true if all visited instruction can be combined.
bool isCompletelySLP() const { return CompletelySLP; }
};
namespace vputils {
/// Returns true if only the first lane of \p Def is used.
bool onlyFirstLaneUsed(const VPValue *Def);
/// Returns true if only the first part of \p Def is used.
bool onlyFirstPartUsed(const VPValue *Def);
/// Get or create a VPValue that corresponds to the expansion of \p Expr. If \p
/// Expr is a SCEVConstant or SCEVUnknown, return a VPValue wrapping the live-in
/// value. Otherwise return a VPExpandSCEVRecipe to expand \p Expr. If \p Plan's
/// pre-header already contains a recipe expanding \p Expr, return it. If not,
/// create a new one.
VPValue *getOrCreateVPValueForSCEVExpr(VPlan &Plan, const SCEV *Expr,
ScalarEvolution &SE);
/// Returns true if \p VPV is uniform after vectorization.
inline bool isUniformAfterVectorization(VPValue *VPV) {
// A value defined outside the vector region must be uniform after
// vectorization inside a vector region.
if (VPV->isDefinedOutsideVectorRegions())
return true;
VPRecipeBase *Def = VPV->getDefiningRecipe();
assert(Def && "Must have definition for value defined inside vector region");
if (auto Rep = dyn_cast<VPReplicateRecipe>(Def))
return Rep->isUniform();
if (auto *GEP = dyn_cast<VPWidenGEPRecipe>(Def))
return all_of(GEP->operands(), isUniformAfterVectorization);
if (auto *VPI = dyn_cast<VPInstruction>(Def))
return VPI->getOpcode() == VPInstruction::ComputeReductionResult;
return false;
}
} // end namespace vputils
} // end namespace llvm
#endif // LLVM_TRANSFORMS_VECTORIZE_VPLAN_H
|