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
|
//===- LoopVectorize.cpp - A Loop Vectorizer ------------------------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "LoopVectorize.h"
#include "llvm/ADT/SmallSet.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/Analysis/AliasAnalysis.h"
#include "llvm/Analysis/AliasSetTracker.h"
#include "llvm/Analysis/Dominators.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/LoopIterator.h"
#include "llvm/Analysis/LoopPass.h"
#include "llvm/Analysis/ScalarEvolutionExpander.h"
#include "llvm/Analysis/ScalarEvolutionExpressions.h"
#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/Analysis/ValueTracking.h"
#include "llvm/Analysis/Verifier.h"
#include "llvm/IR/Constants.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/DerivedTypes.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/Instructions.h"
#include "llvm/IR/IntrinsicInst.h"
#include "llvm/IR/LLVMContext.h"
#include "llvm/IR/Module.h"
#include "llvm/IR/Type.h"
#include "llvm/IR/Value.h"
#include "llvm/Pass.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/Transforms/Scalar.h"
#include "llvm/Transforms/Utils/BasicBlockUtils.h"
#include "llvm/Transforms/Utils/Local.h"
#include "llvm/Transforms/Vectorize.h"
static cl::opt<unsigned>
VectorizationFactor("force-vector-width", cl::init(0), cl::Hidden,
cl::desc("Sets the SIMD width. Zero is autoselect."));
static cl::opt<unsigned>
VectorizationUnroll("force-vector-unroll", cl::init(0), cl::Hidden,
cl::desc("Sets the vectorization unroll count. "
"Zero is autoselect."));
static cl::opt<bool>
EnableIfConversion("enable-if-conversion", cl::init(true), cl::Hidden,
cl::desc("Enable if-conversion during vectorization."));
namespace {
/// The LoopVectorize Pass.
struct LoopVectorize : public LoopPass {
/// Pass identification, replacement for typeid
static char ID;
explicit LoopVectorize() : LoopPass(ID) {
initializeLoopVectorizePass(*PassRegistry::getPassRegistry());
}
ScalarEvolution *SE;
DataLayout *DL;
LoopInfo *LI;
TargetTransformInfo *TTI;
DominatorTree *DT;
virtual bool runOnLoop(Loop *L, LPPassManager &LPM) {
// We only vectorize innermost loops.
if (!L->empty())
return false;
SE = &getAnalysis<ScalarEvolution>();
DL = getAnalysisIfAvailable<DataLayout>();
LI = &getAnalysis<LoopInfo>();
TTI = getAnalysisIfAvailable<TargetTransformInfo>();
DT = &getAnalysis<DominatorTree>();
DEBUG(dbgs() << "LV: Checking a loop in \"" <<
L->getHeader()->getParent()->getName() << "\"\n");
// Check if it is legal to vectorize the loop.
LoopVectorizationLegality LVL(L, SE, DL, DT);
if (!LVL.canVectorize()) {
DEBUG(dbgs() << "LV: Not vectorizing.\n");
return false;
}
// Use the cost model.
LoopVectorizationCostModel CM(L, SE, LI, &LVL, TTI);
// Check the function attribues to find out if this function should be
// optimized for size.
Function *F = L->getHeader()->getParent();
Attribute::AttrKind SzAttr = Attribute::OptimizeForSize;
Attribute::AttrKind FlAttr = Attribute::NoImplicitFloat;
unsigned FnIndex = AttributeSet::FunctionIndex;
bool OptForSize = F->getAttributes().hasAttribute(FnIndex, SzAttr);
bool NoFloat = F->getAttributes().hasAttribute(FnIndex, FlAttr);
if (NoFloat) {
DEBUG(dbgs() << "LV: Can't vectorize when the NoImplicitFloat"
"attribute is used.\n");
return false;
}
unsigned VF = CM.selectVectorizationFactor(OptForSize, VectorizationFactor);
unsigned UF = CM.selectUnrollFactor(OptForSize, VectorizationUnroll);
if (VF == 1) {
DEBUG(dbgs() << "LV: Vectorization is possible but not beneficial.\n");
return false;
}
DEBUG(dbgs() << "LV: Found a vectorizable loop ("<< VF << ") in "<<
F->getParent()->getModuleIdentifier()<<"\n");
DEBUG(dbgs() << "LV: Unroll Factor is " << UF << "\n");
// If we decided that it is *legal* to vectorizer the loop then do it.
InnerLoopVectorizer LB(L, SE, LI, DT, DL, VF, UF);
LB.vectorize(&LVL);
DEBUG(verifyFunction(*L->getHeader()->getParent()));
return true;
}
virtual void getAnalysisUsage(AnalysisUsage &AU) const {
LoopPass::getAnalysisUsage(AU);
AU.addRequiredID(LoopSimplifyID);
AU.addRequiredID(LCSSAID);
AU.addRequired<LoopInfo>();
AU.addRequired<ScalarEvolution>();
AU.addRequired<DominatorTree>();
AU.addPreserved<LoopInfo>();
AU.addPreserved<DominatorTree>();
}
};
}// namespace
//===----------------------------------------------------------------------===//
// Implementation of LoopVectorizationLegality, InnerLoopVectorizer and
// LoopVectorizationCostModel.
//===----------------------------------------------------------------------===//
void
LoopVectorizationLegality::RuntimePointerCheck::insert(ScalarEvolution *SE,
Loop *Lp, Value *Ptr) {
const SCEV *Sc = SE->getSCEV(Ptr);
const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(Sc);
assert(AR && "Invalid addrec expression");
const SCEV *Ex = SE->getExitCount(Lp, Lp->getLoopLatch());
const SCEV *ScEnd = AR->evaluateAtIteration(Ex, *SE);
Pointers.push_back(Ptr);
Starts.push_back(AR->getStart());
Ends.push_back(ScEnd);
}
Value *InnerLoopVectorizer::getBroadcastInstrs(Value *V) {
// Save the current insertion location.
Instruction *Loc = Builder.GetInsertPoint();
// We need to place the broadcast of invariant variables outside the loop.
Instruction *Instr = dyn_cast<Instruction>(V);
bool NewInstr = (Instr && Instr->getParent() == LoopVectorBody);
bool Invariant = OrigLoop->isLoopInvariant(V) && !NewInstr;
// Place the code for broadcasting invariant variables in the new preheader.
if (Invariant)
Builder.SetInsertPoint(LoopVectorPreHeader->getTerminator());
// Broadcast the scalar into all locations in the vector.
Value *Shuf = Builder.CreateVectorSplat(VF, V, "broadcast");
// Restore the builder insertion point.
if (Invariant)
Builder.SetInsertPoint(Loc);
return Shuf;
}
Value *InnerLoopVectorizer::getConsecutiveVector(Value* Val, unsigned StartIdx,
bool Negate) {
assert(Val->getType()->isVectorTy() && "Must be a vector");
assert(Val->getType()->getScalarType()->isIntegerTy() &&
"Elem must be an integer");
// Create the types.
Type *ITy = Val->getType()->getScalarType();
VectorType *Ty = cast<VectorType>(Val->getType());
int VLen = Ty->getNumElements();
SmallVector<Constant*, 8> Indices;
// Create a vector of consecutive numbers from zero to VF.
for (int i = 0; i < VLen; ++i) {
int Idx = Negate ? (-i): i;
Indices.push_back(ConstantInt::get(ITy, StartIdx + Idx));
}
// Add the consecutive indices to the vector value.
Constant *Cv = ConstantVector::get(Indices);
assert(Cv->getType() == Val->getType() && "Invalid consecutive vec");
return Builder.CreateAdd(Val, Cv, "induction");
}
int LoopVectorizationLegality::isConsecutivePtr(Value *Ptr) {
assert(Ptr->getType()->isPointerTy() && "Unexpected non ptr");
// If this value is a pointer induction variable we know it is consecutive.
PHINode *Phi = dyn_cast_or_null<PHINode>(Ptr);
if (Phi && Inductions.count(Phi)) {
InductionInfo II = Inductions[Phi];
if (PtrInduction == II.IK)
return 1;
}
GetElementPtrInst *Gep = dyn_cast_or_null<GetElementPtrInst>(Ptr);
if (!Gep)
return 0;
unsigned NumOperands = Gep->getNumOperands();
Value *LastIndex = Gep->getOperand(NumOperands - 1);
// Check that all of the gep indices are uniform except for the last.
for (unsigned i = 0; i < NumOperands - 1; ++i)
if (!SE->isLoopInvariant(SE->getSCEV(Gep->getOperand(i)), TheLoop))
return 0;
// We can emit wide load/stores only if the last index is the induction
// variable.
const SCEV *Last = SE->getSCEV(LastIndex);
if (const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(Last)) {
const SCEV *Step = AR->getStepRecurrence(*SE);
// The memory is consecutive because the last index is consecutive
// and all other indices are loop invariant.
if (Step->isOne())
return 1;
if (Step->isAllOnesValue())
return -1;
}
return 0;
}
bool LoopVectorizationLegality::isUniform(Value *V) {
return (SE->isLoopInvariant(SE->getSCEV(V), TheLoop));
}
InnerLoopVectorizer::VectorParts&
InnerLoopVectorizer::getVectorValue(Value *V) {
assert(V != Induction && "The new induction variable should not be used.");
assert(!V->getType()->isVectorTy() && "Can't widen a vector");
// If we have this scalar in the map, return it.
if (WidenMap.has(V))
return WidenMap.get(V);
// If this scalar is unknown, assume that it is a constant or that it is
// loop invariant. Broadcast V and save the value for future uses.
Value *B = getBroadcastInstrs(V);
WidenMap.splat(V, B);
return WidenMap.get(V);
}
Constant*
InnerLoopVectorizer::getUniformVector(unsigned Val, Type* ScalarTy) {
return ConstantVector::getSplat(VF, ConstantInt::get(ScalarTy, Val, true));
}
Value *InnerLoopVectorizer::reverseVector(Value *Vec) {
assert(Vec->getType()->isVectorTy() && "Invalid type");
SmallVector<Constant*, 8> ShuffleMask;
for (unsigned i = 0; i < VF; ++i)
ShuffleMask.push_back(Builder.getInt32(VF - i - 1));
return Builder.CreateShuffleVector(Vec, UndefValue::get(Vec->getType()),
ConstantVector::get(ShuffleMask),
"reverse");
}
void InnerLoopVectorizer::scalarizeInstruction(Instruction *Instr) {
assert(!Instr->getType()->isAggregateType() && "Can't handle vectors");
// Holds vector parameters or scalars, in case of uniform vals.
SmallVector<VectorParts, 4> Params;
// Find all of the vectorized parameters.
for (unsigned op = 0, e = Instr->getNumOperands(); op != e; ++op) {
Value *SrcOp = Instr->getOperand(op);
// If we are accessing the old induction variable, use the new one.
if (SrcOp == OldInduction) {
Params.push_back(getVectorValue(SrcOp));
continue;
}
// Try using previously calculated values.
Instruction *SrcInst = dyn_cast<Instruction>(SrcOp);
// If the src is an instruction that appeared earlier in the basic block
// then it should already be vectorized.
if (SrcInst && OrigLoop->contains(SrcInst)) {
assert(WidenMap.has(SrcInst) && "Source operand is unavailable");
// The parameter is a vector value from earlier.
Params.push_back(WidenMap.get(SrcInst));
} else {
// The parameter is a scalar from outside the loop. Maybe even a constant.
VectorParts Scalars;
Scalars.append(UF, SrcOp);
Params.push_back(Scalars);
}
}
assert(Params.size() == Instr->getNumOperands() &&
"Invalid number of operands");
// Does this instruction return a value ?
bool IsVoidRetTy = Instr->getType()->isVoidTy();
Value *UndefVec = IsVoidRetTy ? 0 :
UndefValue::get(VectorType::get(Instr->getType(), VF));
// Create a new entry in the WidenMap and initialize it to Undef or Null.
VectorParts &VecResults = WidenMap.splat(Instr, UndefVec);
// For each scalar that we create:
for (unsigned Width = 0; Width < VF; ++Width) {
// For each vector unroll 'part':
for (unsigned Part = 0; Part < UF; ++Part) {
Instruction *Cloned = Instr->clone();
if (!IsVoidRetTy)
Cloned->setName(Instr->getName() + ".cloned");
// Replace the operands of the cloned instrucions with extracted scalars.
for (unsigned op = 0, e = Instr->getNumOperands(); op != e; ++op) {
Value *Op = Params[op][Part];
// Param is a vector. Need to extract the right lane.
if (Op->getType()->isVectorTy())
Op = Builder.CreateExtractElement(Op, Builder.getInt32(Width));
Cloned->setOperand(op, Op);
}
// Place the cloned scalar in the new loop.
Builder.Insert(Cloned);
// If the original scalar returns a value we need to place it in a vector
// so that future users will be able to use it.
if (!IsVoidRetTy)
VecResults[Part] = Builder.CreateInsertElement(VecResults[Part], Cloned,
Builder.getInt32(Width));
}
}
}
Value*
InnerLoopVectorizer::addRuntimeCheck(LoopVectorizationLegality *Legal,
Instruction *Loc) {
LoopVectorizationLegality::RuntimePointerCheck *PtrRtCheck =
Legal->getRuntimePointerCheck();
if (!PtrRtCheck->Need)
return NULL;
Value *MemoryRuntimeCheck = 0;
unsigned NumPointers = PtrRtCheck->Pointers.size();
SmallVector<Value* , 2> Starts;
SmallVector<Value* , 2> Ends;
SCEVExpander Exp(*SE, "induction");
// Use this type for pointer arithmetic.
Type* PtrArithTy = Type::getInt8PtrTy(Loc->getContext(), 0);
for (unsigned i = 0; i < NumPointers; ++i) {
Value *Ptr = PtrRtCheck->Pointers[i];
const SCEV *Sc = SE->getSCEV(Ptr);
if (SE->isLoopInvariant(Sc, OrigLoop)) {
DEBUG(dbgs() << "LV: Adding RT check for a loop invariant ptr:" <<
*Ptr <<"\n");
Starts.push_back(Ptr);
Ends.push_back(Ptr);
} else {
DEBUG(dbgs() << "LV: Adding RT check for range:" << *Ptr <<"\n");
Value *Start = Exp.expandCodeFor(PtrRtCheck->Starts[i], PtrArithTy, Loc);
Value *End = Exp.expandCodeFor(PtrRtCheck->Ends[i], PtrArithTy, Loc);
Starts.push_back(Start);
Ends.push_back(End);
}
}
for (unsigned i = 0; i < NumPointers; ++i) {
for (unsigned j = i+1; j < NumPointers; ++j) {
Instruction::CastOps Op = Instruction::BitCast;
Value *Start0 = CastInst::Create(Op, Starts[i], PtrArithTy, "bc", Loc);
Value *Start1 = CastInst::Create(Op, Starts[j], PtrArithTy, "bc", Loc);
Value *End0 = CastInst::Create(Op, Ends[i], PtrArithTy, "bc", Loc);
Value *End1 = CastInst::Create(Op, Ends[j], PtrArithTy, "bc", Loc);
Value *Cmp0 = CmpInst::Create(Instruction::ICmp, CmpInst::ICMP_ULE,
Start0, End1, "bound0", Loc);
Value *Cmp1 = CmpInst::Create(Instruction::ICmp, CmpInst::ICMP_ULE,
Start1, End0, "bound1", Loc);
Value *IsConflict = BinaryOperator::Create(Instruction::And, Cmp0, Cmp1,
"found.conflict", Loc);
if (MemoryRuntimeCheck)
MemoryRuntimeCheck = BinaryOperator::Create(Instruction::Or,
MemoryRuntimeCheck,
IsConflict,
"conflict.rdx", Loc);
else
MemoryRuntimeCheck = IsConflict;
}
}
return MemoryRuntimeCheck;
}
void
InnerLoopVectorizer::createEmptyLoop(LoopVectorizationLegality *Legal) {
/*
In this function we generate a new loop. The new loop will contain
the vectorized instructions while the old loop will continue to run the
scalar remainder.
[ ] <-- vector loop bypass.
/ |
/ v
| [ ] <-- vector pre header.
| |
| v
| [ ] \
| [ ]_| <-- vector loop.
| |
\ v
>[ ] <--- middle-block.
/ |
/ v
| [ ] <--- new preheader.
| |
| v
| [ ] \
| [ ]_| <-- old scalar loop to handle remainder.
\ |
\ v
>[ ] <-- exit block.
...
*/
BasicBlock *OldBasicBlock = OrigLoop->getHeader();
BasicBlock *BypassBlock = OrigLoop->getLoopPreheader();
BasicBlock *ExitBlock = OrigLoop->getExitBlock();
assert(ExitBlock && "Must have an exit block");
// Some loops have a single integer induction variable, while other loops
// don't. One example is c++ iterators that often have multiple pointer
// induction variables. In the code below we also support a case where we
// don't have a single induction variable.
OldInduction = Legal->getInduction();
Type *IdxTy = OldInduction ? OldInduction->getType() :
DL->getIntPtrType(SE->getContext());
// Find the loop boundaries.
const SCEV *ExitCount = SE->getExitCount(OrigLoop, OrigLoop->getLoopLatch());
assert(ExitCount != SE->getCouldNotCompute() && "Invalid loop count");
// Get the total trip count from the count by adding 1.
ExitCount = SE->getAddExpr(ExitCount,
SE->getConstant(ExitCount->getType(), 1));
// Expand the trip count and place the new instructions in the preheader.
// Notice that the pre-header does not change, only the loop body.
SCEVExpander Exp(*SE, "induction");
// Count holds the overall loop count (N).
Value *Count = Exp.expandCodeFor(ExitCount, ExitCount->getType(),
BypassBlock->getTerminator());
// The loop index does not have to start at Zero. Find the original start
// value from the induction PHI node. If we don't have an induction variable
// then we know that it starts at zero.
Value *StartIdx = OldInduction ?
OldInduction->getIncomingValueForBlock(BypassBlock):
ConstantInt::get(IdxTy, 0);
assert(BypassBlock && "Invalid loop structure");
// Generate the code that checks in runtime if arrays overlap.
Value *MemoryRuntimeCheck = addRuntimeCheck(Legal,
BypassBlock->getTerminator());
// Split the single block loop into the two loop structure described above.
BasicBlock *VectorPH =
BypassBlock->splitBasicBlock(BypassBlock->getTerminator(), "vector.ph");
BasicBlock *VecBody =
VectorPH->splitBasicBlock(VectorPH->getTerminator(), "vector.body");
BasicBlock *MiddleBlock =
VecBody->splitBasicBlock(VecBody->getTerminator(), "middle.block");
BasicBlock *ScalarPH =
MiddleBlock->splitBasicBlock(MiddleBlock->getTerminator(), "scalar.ph");
// This is the location in which we add all of the logic for bypassing
// the new vector loop.
Instruction *Loc = BypassBlock->getTerminator();
// Use this IR builder to create the loop instructions (Phi, Br, Cmp)
// inside the loop.
Builder.SetInsertPoint(VecBody->getFirstInsertionPt());
// Generate the induction variable.
Induction = Builder.CreatePHI(IdxTy, 2, "index");
// The loop step is equal to the vectorization factor (num of SIMD elements)
// times the unroll factor (num of SIMD instructions).
Constant *Step = ConstantInt::get(IdxTy, VF * UF);
// We may need to extend the index in case there is a type mismatch.
// We know that the count starts at zero and does not overflow.
if (Count->getType() != IdxTy) {
// The exit count can be of pointer type. Convert it to the correct
// integer type.
if (ExitCount->getType()->isPointerTy())
Count = CastInst::CreatePointerCast(Count, IdxTy, "ptrcnt.to.int", Loc);
else
Count = CastInst::CreateZExtOrBitCast(Count, IdxTy, "zext.cnt", Loc);
}
// Add the start index to the loop count to get the new end index.
Value *IdxEnd = BinaryOperator::CreateAdd(Count, StartIdx, "end.idx", Loc);
// Now we need to generate the expression for N - (N % VF), which is
// the part that the vectorized body will execute.
Value *R = BinaryOperator::CreateURem(Count, Step, "n.mod.vf", Loc);
Value *CountRoundDown = BinaryOperator::CreateSub(Count, R, "n.vec", Loc);
Value *IdxEndRoundDown = BinaryOperator::CreateAdd(CountRoundDown, StartIdx,
"end.idx.rnd.down", Loc);
// Now, compare the new count to zero. If it is zero skip the vector loop and
// jump to the scalar loop.
Value *Cmp = CmpInst::Create(Instruction::ICmp, CmpInst::ICMP_EQ,
IdxEndRoundDown,
StartIdx,
"cmp.zero", Loc);
// If we are using memory runtime checks, include them in.
if (MemoryRuntimeCheck)
Cmp = BinaryOperator::Create(Instruction::Or, Cmp, MemoryRuntimeCheck,
"CntOrMem", Loc);
BranchInst::Create(MiddleBlock, VectorPH, Cmp, Loc);
// Remove the old terminator.
Loc->eraseFromParent();
// We are going to resume the execution of the scalar loop.
// Go over all of the induction variables that we found and fix the
// PHIs that are left in the scalar version of the loop.
// The starting values of PHI nodes depend on the counter of the last
// iteration in the vectorized loop.
// If we come from a bypass edge then we need to start from the original
// start value.
// This variable saves the new starting index for the scalar loop.
PHINode *ResumeIndex = 0;
LoopVectorizationLegality::InductionList::iterator I, E;
LoopVectorizationLegality::InductionList *List = Legal->getInductionVars();
for (I = List->begin(), E = List->end(); I != E; ++I) {
PHINode *OrigPhi = I->first;
LoopVectorizationLegality::InductionInfo II = I->second;
PHINode *ResumeVal = PHINode::Create(OrigPhi->getType(), 2, "resume.val",
MiddleBlock->getTerminator());
Value *EndValue = 0;
switch (II.IK) {
case LoopVectorizationLegality::NoInduction:
llvm_unreachable("Unknown induction");
case LoopVectorizationLegality::IntInduction: {
// Handle the integer induction counter:
assert(OrigPhi->getType()->isIntegerTy() && "Invalid type");
assert(OrigPhi == OldInduction && "Unknown integer PHI");
// We know what the end value is.
EndValue = IdxEndRoundDown;
// We also know which PHI node holds it.
ResumeIndex = ResumeVal;
break;
}
case LoopVectorizationLegality::ReverseIntInduction: {
// Convert the CountRoundDown variable to the PHI size.
unsigned CRDSize = CountRoundDown->getType()->getScalarSizeInBits();
unsigned IISize = II.StartValue->getType()->getScalarSizeInBits();
Value *CRD = CountRoundDown;
if (CRDSize > IISize)
CRD = CastInst::Create(Instruction::Trunc, CountRoundDown,
II.StartValue->getType(),
"tr.crd", BypassBlock->getTerminator());
else if (CRDSize < IISize)
CRD = CastInst::Create(Instruction::SExt, CountRoundDown,
II.StartValue->getType(),
"sext.crd", BypassBlock->getTerminator());
// Handle reverse integer induction counter:
EndValue = BinaryOperator::CreateSub(II.StartValue, CRD, "rev.ind.end",
BypassBlock->getTerminator());
break;
}
case LoopVectorizationLegality::PtrInduction: {
// For pointer induction variables, calculate the offset using
// the end index.
EndValue = GetElementPtrInst::Create(II.StartValue, CountRoundDown,
"ptr.ind.end",
BypassBlock->getTerminator());
break;
}
}// end of case
// The new PHI merges the original incoming value, in case of a bypass,
// or the value at the end of the vectorized loop.
ResumeVal->addIncoming(II.StartValue, BypassBlock);
ResumeVal->addIncoming(EndValue, VecBody);
// Fix the scalar body counter (PHI node).
unsigned BlockIdx = OrigPhi->getBasicBlockIndex(ScalarPH);
OrigPhi->setIncomingValue(BlockIdx, ResumeVal);
}
// If we are generating a new induction variable then we also need to
// generate the code that calculates the exit value. This value is not
// simply the end of the counter because we may skip the vectorized body
// in case of a runtime check.
if (!OldInduction){
assert(!ResumeIndex && "Unexpected resume value found");
ResumeIndex = PHINode::Create(IdxTy, 2, "new.indc.resume.val",
MiddleBlock->getTerminator());
ResumeIndex->addIncoming(StartIdx, BypassBlock);
ResumeIndex->addIncoming(IdxEndRoundDown, VecBody);
}
// Make sure that we found the index where scalar loop needs to continue.
assert(ResumeIndex && ResumeIndex->getType()->isIntegerTy() &&
"Invalid resume Index");
// Add a check in the middle block to see if we have completed
// all of the iterations in the first vector loop.
// If (N - N%VF) == N, then we *don't* need to run the remainder.
Value *CmpN = CmpInst::Create(Instruction::ICmp, CmpInst::ICMP_EQ, IdxEnd,
ResumeIndex, "cmp.n",
MiddleBlock->getTerminator());
BranchInst::Create(ExitBlock, ScalarPH, CmpN, MiddleBlock->getTerminator());
// Remove the old terminator.
MiddleBlock->getTerminator()->eraseFromParent();
// Create i+1 and fill the PHINode.
Value *NextIdx = Builder.CreateAdd(Induction, Step, "index.next");
Induction->addIncoming(StartIdx, VectorPH);
Induction->addIncoming(NextIdx, VecBody);
// Create the compare.
Value *ICmp = Builder.CreateICmpEQ(NextIdx, IdxEndRoundDown);
Builder.CreateCondBr(ICmp, MiddleBlock, VecBody);
// Now we have two terminators. Remove the old one from the block.
VecBody->getTerminator()->eraseFromParent();
// Get ready to start creating new instructions into the vectorized body.
Builder.SetInsertPoint(VecBody->getFirstInsertionPt());
// Create and register the new vector loop.
Loop* Lp = new Loop();
Loop *ParentLoop = OrigLoop->getParentLoop();
// Insert the new loop into the loop nest and register the new basic blocks.
if (ParentLoop) {
ParentLoop->addChildLoop(Lp);
ParentLoop->addBasicBlockToLoop(ScalarPH, LI->getBase());
ParentLoop->addBasicBlockToLoop(VectorPH, LI->getBase());
ParentLoop->addBasicBlockToLoop(MiddleBlock, LI->getBase());
} else {
LI->addTopLevelLoop(Lp);
}
Lp->addBasicBlockToLoop(VecBody, LI->getBase());
// Save the state.
LoopVectorPreHeader = VectorPH;
LoopScalarPreHeader = ScalarPH;
LoopMiddleBlock = MiddleBlock;
LoopExitBlock = ExitBlock;
LoopVectorBody = VecBody;
LoopScalarBody = OldBasicBlock;
LoopBypassBlock = BypassBlock;
}
/// This function returns the identity element (or neutral element) for
/// the operation K.
static unsigned
getReductionIdentity(LoopVectorizationLegality::ReductionKind K) {
switch (K) {
case LoopVectorizationLegality::IntegerXor:
case LoopVectorizationLegality::IntegerAdd:
case LoopVectorizationLegality::IntegerOr:
// Adding, Xoring, Oring zero to a number does not change it.
return 0;
case LoopVectorizationLegality::IntegerMult:
// Multiplying a number by 1 does not change it.
return 1;
case LoopVectorizationLegality::IntegerAnd:
// AND-ing a number with an all-1 value does not change it.
return -1;
default:
llvm_unreachable("Unknown reduction kind");
}
}
static bool
isTriviallyVectorizableIntrinsic(Instruction *Inst) {
IntrinsicInst *II = dyn_cast<IntrinsicInst>(Inst);
if (!II)
return false;
switch (II->getIntrinsicID()) {
case Intrinsic::sqrt:
case Intrinsic::sin:
case Intrinsic::cos:
case Intrinsic::exp:
case Intrinsic::exp2:
case Intrinsic::log:
case Intrinsic::log10:
case Intrinsic::log2:
case Intrinsic::fabs:
case Intrinsic::floor:
case Intrinsic::ceil:
case Intrinsic::trunc:
case Intrinsic::rint:
case Intrinsic::nearbyint:
case Intrinsic::pow:
case Intrinsic::fma:
case Intrinsic::fmuladd:
return true;
default:
return false;
}
return false;
}
void
InnerLoopVectorizer::vectorizeLoop(LoopVectorizationLegality *Legal) {
//===------------------------------------------------===//
//
// Notice: any optimization or new instruction that go
// into the code below should be also be implemented in
// the cost-model.
//
//===------------------------------------------------===//
BasicBlock &BB = *OrigLoop->getHeader();
Constant *Zero =
ConstantInt::get(IntegerType::getInt32Ty(BB.getContext()), 0);
// In order to support reduction variables we need to be able to vectorize
// Phi nodes. Phi nodes have cycles, so we need to vectorize them in two
// stages. First, we create a new vector PHI node with no incoming edges.
// We use this value when we vectorize all of the instructions that use the
// PHI. Next, after all of the instructions in the block are complete we
// add the new incoming edges to the PHI. At this point all of the
// instructions in the basic block are vectorized, so we can use them to
// construct the PHI.
PhiVector RdxPHIsToFix;
// Scan the loop in a topological order to ensure that defs are vectorized
// before users.
LoopBlocksDFS DFS(OrigLoop);
DFS.perform(LI);
// Vectorize all of the blocks in the original loop.
for (LoopBlocksDFS::RPOIterator bb = DFS.beginRPO(),
be = DFS.endRPO(); bb != be; ++bb)
vectorizeBlockInLoop(Legal, *bb, &RdxPHIsToFix);
// At this point every instruction in the original loop is widened to
// a vector form. We are almost done. Now, we need to fix the PHI nodes
// that we vectorized. The PHI nodes are currently empty because we did
// not want to introduce cycles. Notice that the remaining PHI nodes
// that we need to fix are reduction variables.
// Create the 'reduced' values for each of the induction vars.
// The reduced values are the vector values that we scalarize and combine
// after the loop is finished.
for (PhiVector::iterator it = RdxPHIsToFix.begin(), e = RdxPHIsToFix.end();
it != e; ++it) {
PHINode *RdxPhi = *it;
assert(RdxPhi && "Unable to recover vectorized PHI");
// Find the reduction variable descriptor.
assert(Legal->getReductionVars()->count(RdxPhi) &&
"Unable to find the reduction variable");
LoopVectorizationLegality::ReductionDescriptor RdxDesc =
(*Legal->getReductionVars())[RdxPhi];
// We need to generate a reduction vector from the incoming scalar.
// To do so, we need to generate the 'identity' vector and overide
// one of the elements with the incoming scalar reduction. We need
// to do it in the vector-loop preheader.
Builder.SetInsertPoint(LoopBypassBlock->getTerminator());
// This is the vector-clone of the value that leaves the loop.
VectorParts &VectorExit = getVectorValue(RdxDesc.LoopExitInstr);
Type *VecTy = VectorExit[0]->getType();
// Find the reduction identity variable. Zero for addition, or, xor,
// one for multiplication, -1 for And.
Constant *Identity = getUniformVector(getReductionIdentity(RdxDesc.Kind),
VecTy->getScalarType());
// This vector is the Identity vector where the first element is the
// incoming scalar reduction.
Value *VectorStart = Builder.CreateInsertElement(Identity,
RdxDesc.StartValue, Zero);
// Fix the vector-loop phi.
// We created the induction variable so we know that the
// preheader is the first entry.
BasicBlock *VecPreheader = Induction->getIncomingBlock(0);
// Reductions do not have to start at zero. They can start with
// any loop invariant values.
VectorParts &VecRdxPhi = WidenMap.get(RdxPhi);
BasicBlock *Latch = OrigLoop->getLoopLatch();
Value *LoopVal = RdxPhi->getIncomingValueForBlock(Latch);
VectorParts &Val = getVectorValue(LoopVal);
for (unsigned part = 0; part < UF; ++part) {
// Make sure to add the reduction stat value only to the
// first unroll part.
Value *StartVal = (part == 0) ? VectorStart : Identity;
cast<PHINode>(VecRdxPhi[part])->addIncoming(StartVal, VecPreheader);
cast<PHINode>(VecRdxPhi[part])->addIncoming(Val[part], LoopVectorBody);
}
// Before each round, move the insertion point right between
// the PHIs and the values we are going to write.
// This allows us to write both PHINodes and the extractelement
// instructions.
Builder.SetInsertPoint(LoopMiddleBlock->getFirstInsertionPt());
VectorParts RdxParts;
for (unsigned part = 0; part < UF; ++part) {
// This PHINode contains the vectorized reduction variable, or
// the initial value vector, if we bypass the vector loop.
VectorParts &RdxExitVal = getVectorValue(RdxDesc.LoopExitInstr);
PHINode *NewPhi = Builder.CreatePHI(VecTy, 2, "rdx.vec.exit.phi");
Value *StartVal = (part == 0) ? VectorStart : Identity;
NewPhi->addIncoming(StartVal, LoopBypassBlock);
NewPhi->addIncoming(RdxExitVal[part], LoopVectorBody);
RdxParts.push_back(NewPhi);
}
// Reduce all of the unrolled parts into a single vector.
Value *ReducedPartRdx = RdxParts[0];
for (unsigned part = 1; part < UF; ++part) {
switch (RdxDesc.Kind) {
case LoopVectorizationLegality::IntegerAdd:
ReducedPartRdx =
Builder.CreateAdd(RdxParts[part], ReducedPartRdx, "add.rdx");
break;
case LoopVectorizationLegality::IntegerMult:
ReducedPartRdx =
Builder.CreateMul(RdxParts[part], ReducedPartRdx, "mul.rdx");
break;
case LoopVectorizationLegality::IntegerOr:
ReducedPartRdx =
Builder.CreateOr(RdxParts[part], ReducedPartRdx, "or.rdx");
break;
case LoopVectorizationLegality::IntegerAnd:
ReducedPartRdx =
Builder.CreateAnd(RdxParts[part], ReducedPartRdx, "and.rdx");
break;
case LoopVectorizationLegality::IntegerXor:
ReducedPartRdx =
Builder.CreateXor(RdxParts[part], ReducedPartRdx, "xor.rdx");
break;
default:
llvm_unreachable("Unknown reduction operation");
}
}
// VF is a power of 2 so we can emit the reduction using log2(VF) shuffles
// and vector ops, reducing the set of values being computed by half each
// round.
assert(isPowerOf2_32(VF) &&
"Reduction emission only supported for pow2 vectors!");
Value *TmpVec = ReducedPartRdx;
SmallVector<Constant*, 32> ShuffleMask(VF, 0);
for (unsigned i = VF; i != 1; i >>= 1) {
// Move the upper half of the vector to the lower half.
for (unsigned j = 0; j != i/2; ++j)
ShuffleMask[j] = Builder.getInt32(i/2 + j);
// Fill the rest of the mask with undef.
std::fill(&ShuffleMask[i/2], ShuffleMask.end(),
UndefValue::get(Builder.getInt32Ty()));
Value *Shuf =
Builder.CreateShuffleVector(TmpVec,
UndefValue::get(TmpVec->getType()),
ConstantVector::get(ShuffleMask),
"rdx.shuf");
// Emit the operation on the shuffled value.
switch (RdxDesc.Kind) {
case LoopVectorizationLegality::IntegerAdd:
TmpVec = Builder.CreateAdd(TmpVec, Shuf, "add.rdx");
break;
case LoopVectorizationLegality::IntegerMult:
TmpVec = Builder.CreateMul(TmpVec, Shuf, "mul.rdx");
break;
case LoopVectorizationLegality::IntegerOr:
TmpVec = Builder.CreateOr(TmpVec, Shuf, "or.rdx");
break;
case LoopVectorizationLegality::IntegerAnd:
TmpVec = Builder.CreateAnd(TmpVec, Shuf, "and.rdx");
break;
case LoopVectorizationLegality::IntegerXor:
TmpVec = Builder.CreateXor(TmpVec, Shuf, "xor.rdx");
break;
default:
llvm_unreachable("Unknown reduction operation");
}
}
// The result is in the first element of the vector.
Value *Scalar0 = Builder.CreateExtractElement(TmpVec, Builder.getInt32(0));
// Now, we need to fix the users of the reduction variable
// inside and outside of the scalar remainder loop.
// We know that the loop is in LCSSA form. We need to update the
// PHI nodes in the exit blocks.
for (BasicBlock::iterator LEI = LoopExitBlock->begin(),
LEE = LoopExitBlock->end(); LEI != LEE; ++LEI) {
PHINode *LCSSAPhi = dyn_cast<PHINode>(LEI);
if (!LCSSAPhi) continue;
// All PHINodes need to have a single entry edge, or two if
// we already fixed them.
assert(LCSSAPhi->getNumIncomingValues() < 3 && "Invalid LCSSA PHI");
// We found our reduction value exit-PHI. Update it with the
// incoming bypass edge.
if (LCSSAPhi->getIncomingValue(0) == RdxDesc.LoopExitInstr) {
// Add an edge coming from the bypass.
LCSSAPhi->addIncoming(Scalar0, LoopMiddleBlock);
break;
}
}// end of the LCSSA phi scan.
// Fix the scalar loop reduction variable with the incoming reduction sum
// from the vector body and from the backedge value.
int IncomingEdgeBlockIdx =
(RdxPhi)->getBasicBlockIndex(OrigLoop->getLoopLatch());
assert(IncomingEdgeBlockIdx >= 0 && "Invalid block index");
// Pick the other block.
int SelfEdgeBlockIdx = (IncomingEdgeBlockIdx ? 0 : 1);
(RdxPhi)->setIncomingValue(SelfEdgeBlockIdx, Scalar0);
(RdxPhi)->setIncomingValue(IncomingEdgeBlockIdx, RdxDesc.LoopExitInstr);
}// end of for each redux variable.
// The Loop exit block may have single value PHI nodes where the incoming
// value is 'undef'. While vectorizing we only handled real values that
// were defined inside the loop. Here we handle the 'undef case'.
// See PR14725.
for (BasicBlock::iterator LEI = LoopExitBlock->begin(),
LEE = LoopExitBlock->end(); LEI != LEE; ++LEI) {
PHINode *LCSSAPhi = dyn_cast<PHINode>(LEI);
if (!LCSSAPhi) continue;
if (LCSSAPhi->getNumIncomingValues() == 1)
LCSSAPhi->addIncoming(UndefValue::get(LCSSAPhi->getType()),
LoopMiddleBlock);
}
}
InnerLoopVectorizer::VectorParts
InnerLoopVectorizer::createEdgeMask(BasicBlock *Src, BasicBlock *Dst) {
assert(std::find(pred_begin(Dst), pred_end(Dst), Src) != pred_end(Dst) &&
"Invalid edge");
VectorParts SrcMask = createBlockInMask(Src);
// The terminator has to be a branch inst!
BranchInst *BI = dyn_cast<BranchInst>(Src->getTerminator());
assert(BI && "Unexpected terminator found");
if (BI->isConditional()) {
VectorParts EdgeMask = getVectorValue(BI->getCondition());
if (BI->getSuccessor(0) != Dst)
for (unsigned part = 0; part < UF; ++part)
EdgeMask[part] = Builder.CreateNot(EdgeMask[part]);
for (unsigned part = 0; part < UF; ++part)
EdgeMask[part] = Builder.CreateAnd(EdgeMask[part], SrcMask[part]);
return EdgeMask;
}
return SrcMask;
}
InnerLoopVectorizer::VectorParts
InnerLoopVectorizer::createBlockInMask(BasicBlock *BB) {
assert(OrigLoop->contains(BB) && "Block is not a part of a loop");
// Loop incoming mask is all-one.
if (OrigLoop->getHeader() == BB) {
Value *C = ConstantInt::get(IntegerType::getInt1Ty(BB->getContext()), 1);
return getVectorValue(C);
}
// This is the block mask. We OR all incoming edges, and with zero.
Value *Zero = ConstantInt::get(IntegerType::getInt1Ty(BB->getContext()), 0);
VectorParts BlockMask = getVectorValue(Zero);
// For each pred:
for (pred_iterator it = pred_begin(BB), e = pred_end(BB); it != e; ++it) {
VectorParts EM = createEdgeMask(*it, BB);
for (unsigned part = 0; part < UF; ++part)
BlockMask[part] = Builder.CreateOr(BlockMask[part], EM[part]);
}
return BlockMask;
}
void
InnerLoopVectorizer::vectorizeBlockInLoop(LoopVectorizationLegality *Legal,
BasicBlock *BB, PhiVector *PV) {
Constant *Zero = Builder.getInt32(0);
// For each instruction in the old loop.
for (BasicBlock::iterator it = BB->begin(), e = BB->end(); it != e; ++it) {
VectorParts &Entry = WidenMap.get(it);
switch (it->getOpcode()) {
case Instruction::Br:
// Nothing to do for PHIs and BR, since we already took care of the
// loop control flow instructions.
continue;
case Instruction::PHI:{
PHINode* P = cast<PHINode>(it);
// Handle reduction variables:
if (Legal->getReductionVars()->count(P)) {
for (unsigned part = 0; part < UF; ++part) {
// This is phase one of vectorizing PHIs.
Type *VecTy = VectorType::get(it->getType(), VF);
Entry[part] = PHINode::Create(VecTy, 2, "vec.phi",
LoopVectorBody-> getFirstInsertionPt());
}
PV->push_back(P);
continue;
}
// Check for PHI nodes that are lowered to vector selects.
if (P->getParent() != OrigLoop->getHeader()) {
// We know that all PHIs in non header blocks are converted into
// selects, so we don't have to worry about the insertion order and we
// can just use the builder.
// At this point we generate the predication tree. There may be
// duplications since this is a simple recursive scan, but future
// optimizations will clean it up.
VectorParts Cond = createEdgeMask(P->getIncomingBlock(0),
P->getParent());
for (unsigned part = 0; part < UF; ++part) {
VectorParts &In0 = getVectorValue(P->getIncomingValue(0));
VectorParts &In1 = getVectorValue(P->getIncomingValue(1));
Entry[part] = Builder.CreateSelect(Cond[part], In0[part], In1[part],
"predphi");
}
continue;
}
// This PHINode must be an induction variable.
// Make sure that we know about it.
assert(Legal->getInductionVars()->count(P) &&
"Not an induction variable");
LoopVectorizationLegality::InductionInfo II =
Legal->getInductionVars()->lookup(P);
switch (II.IK) {
case LoopVectorizationLegality::NoInduction:
llvm_unreachable("Unknown induction");
case LoopVectorizationLegality::IntInduction: {
assert(P == OldInduction && "Unexpected PHI");
Value *Broadcasted = getBroadcastInstrs(Induction);
// After broadcasting the induction variable we need to make the
// vector consecutive by adding 0, 1, 2 ...
for (unsigned part = 0; part < UF; ++part)
Entry[part] = getConsecutiveVector(Broadcasted, VF * part, false);
continue;
}
case LoopVectorizationLegality::ReverseIntInduction:
case LoopVectorizationLegality::PtrInduction:
// Handle reverse integer and pointer inductions.
Value *StartIdx = 0;
// If we have a single integer induction variable then use it.
// Otherwise, start counting at zero.
if (OldInduction) {
LoopVectorizationLegality::InductionInfo OldII =
Legal->getInductionVars()->lookup(OldInduction);
StartIdx = OldII.StartValue;
} else {
StartIdx = ConstantInt::get(Induction->getType(), 0);
}
// This is the normalized GEP that starts counting at zero.
Value *NormalizedIdx = Builder.CreateSub(Induction, StartIdx,
"normalized.idx");
// Handle the reverse integer induction variable case.
if (LoopVectorizationLegality::ReverseIntInduction == II.IK) {
IntegerType *DstTy = cast<IntegerType>(II.StartValue->getType());
Value *CNI = Builder.CreateSExtOrTrunc(NormalizedIdx, DstTy,
"resize.norm.idx");
Value *ReverseInd = Builder.CreateSub(II.StartValue, CNI,
"reverse.idx");
// This is a new value so do not hoist it out.
Value *Broadcasted = getBroadcastInstrs(ReverseInd);
// After broadcasting the induction variable we need to make the
// vector consecutive by adding ... -3, -2, -1, 0.
for (unsigned part = 0; part < UF; ++part)
Entry[part] = getConsecutiveVector(Broadcasted, -VF * part, true);
continue;
}
// Handle the pointer induction variable case.
assert(P->getType()->isPointerTy() && "Unexpected type.");
// This is the vector of results. Notice that we don't generate
// vector geps because scalar geps result in better code.
for (unsigned part = 0; part < UF; ++part) {
Value *VecVal = UndefValue::get(VectorType::get(P->getType(), VF));
for (unsigned int i = 0; i < VF; ++i) {
Constant *Idx = ConstantInt::get(Induction->getType(),
i + part * VF);
Value *GlobalIdx = Builder.CreateAdd(NormalizedIdx, Idx,
"gep.idx");
Value *SclrGep = Builder.CreateGEP(II.StartValue, GlobalIdx,
"next.gep");
VecVal = Builder.CreateInsertElement(VecVal, SclrGep,
Builder.getInt32(i),
"insert.gep");
}
Entry[part] = VecVal;
}
continue;
}
}// End of PHI.
case Instruction::Add:
case Instruction::FAdd:
case Instruction::Sub:
case Instruction::FSub:
case Instruction::Mul:
case Instruction::FMul:
case Instruction::UDiv:
case Instruction::SDiv:
case Instruction::FDiv:
case Instruction::URem:
case Instruction::SRem:
case Instruction::FRem:
case Instruction::Shl:
case Instruction::LShr:
case Instruction::AShr:
case Instruction::And:
case Instruction::Or:
case Instruction::Xor: {
// Just widen binops.
BinaryOperator *BinOp = dyn_cast<BinaryOperator>(it);
VectorParts &A = getVectorValue(it->getOperand(0));
VectorParts &B = getVectorValue(it->getOperand(1));
// Use this vector value for all users of the original instruction.
for (unsigned Part = 0; Part < UF; ++Part) {
Value *V = Builder.CreateBinOp(BinOp->getOpcode(), A[Part], B[Part]);
// Update the NSW, NUW and Exact flags.
BinaryOperator *VecOp = cast<BinaryOperator>(V);
if (isa<OverflowingBinaryOperator>(BinOp)) {
VecOp->setHasNoSignedWrap(BinOp->hasNoSignedWrap());
VecOp->setHasNoUnsignedWrap(BinOp->hasNoUnsignedWrap());
}
if (isa<PossiblyExactOperator>(VecOp))
VecOp->setIsExact(BinOp->isExact());
Entry[Part] = V;
}
break;
}
case Instruction::Select: {
// Widen selects.
// If the selector is loop invariant we can create a select
// instruction with a scalar condition. Otherwise, use vector-select.
bool InvariantCond = SE->isLoopInvariant(SE->getSCEV(it->getOperand(0)),
OrigLoop);
// The condition can be loop invariant but still defined inside the
// loop. This means that we can't just use the original 'cond' value.
// We have to take the 'vectorized' value and pick the first lane.
// Instcombine will make this a no-op.
VectorParts &Cond = getVectorValue(it->getOperand(0));
VectorParts &Op0 = getVectorValue(it->getOperand(1));
VectorParts &Op1 = getVectorValue(it->getOperand(2));
Value *ScalarCond = Builder.CreateExtractElement(Cond[0],
Builder.getInt32(0));
for (unsigned Part = 0; Part < UF; ++Part) {
Entry[Part] = Builder.CreateSelect(
InvariantCond ? ScalarCond : Cond[Part],
Op0[Part],
Op1[Part]);
}
break;
}
case Instruction::ICmp:
case Instruction::FCmp: {
// Widen compares. Generate vector compares.
bool FCmp = (it->getOpcode() == Instruction::FCmp);
CmpInst *Cmp = dyn_cast<CmpInst>(it);
VectorParts &A = getVectorValue(it->getOperand(0));
VectorParts &B = getVectorValue(it->getOperand(1));
for (unsigned Part = 0; Part < UF; ++Part) {
Value *C = 0;
if (FCmp)
C = Builder.CreateFCmp(Cmp->getPredicate(), A[Part], B[Part]);
else
C = Builder.CreateICmp(Cmp->getPredicate(), A[Part], B[Part]);
Entry[Part] = C;
}
break;
}
case Instruction::Store: {
// Attempt to issue a wide store.
StoreInst *SI = dyn_cast<StoreInst>(it);
Type *StTy = VectorType::get(SI->getValueOperand()->getType(), VF);
Value *Ptr = SI->getPointerOperand();
unsigned Alignment = SI->getAlignment();
assert(!Legal->isUniform(Ptr) &&
"We do not allow storing to uniform addresses");
int Stride = Legal->isConsecutivePtr(Ptr);
bool Reverse = Stride < 0;
if (Stride == 0) {
scalarizeInstruction(it);
break;
}
// Handle consecutive stores.
GetElementPtrInst *Gep = dyn_cast<GetElementPtrInst>(Ptr);
if (Gep) {
// The last index does not have to be the induction. It can be
// consecutive and be a function of the index. For example A[I+1];
unsigned NumOperands = Gep->getNumOperands();
Value *LastGepOperand = Gep->getOperand(NumOperands - 1);
VectorParts &GEPParts = getVectorValue(LastGepOperand);
Value *LastIndex = GEPParts[0];
LastIndex = Builder.CreateExtractElement(LastIndex, Zero);
// Create the new GEP with the new induction variable.
GetElementPtrInst *Gep2 = cast<GetElementPtrInst>(Gep->clone());
Gep2->setOperand(NumOperands - 1, LastIndex);
Ptr = Builder.Insert(Gep2);
} else {
// Use the induction element ptr.
assert(isa<PHINode>(Ptr) && "Invalid induction ptr");
VectorParts &PtrVal = getVectorValue(Ptr);
Ptr = Builder.CreateExtractElement(PtrVal[0], Zero);
}
VectorParts &StoredVal = getVectorValue(SI->getValueOperand());
for (unsigned Part = 0; Part < UF; ++Part) {
// Calculate the pointer for the specific unroll-part.
Value *PartPtr = Builder.CreateGEP(Ptr, Builder.getInt32(Part * VF));
if (Reverse) {
// If we store to reverse consecutive memory locations then we need
// to reverse the order of elements in the stored value.
StoredVal[Part] = reverseVector(StoredVal[Part]);
// If the address is consecutive but reversed, then the
// wide store needs to start at the last vector element.
PartPtr = Builder.CreateGEP(Ptr, Builder.getInt32(-Part * VF));
PartPtr = Builder.CreateGEP(PartPtr, Builder.getInt32(1 - VF));
}
Value *VecPtr = Builder.CreateBitCast(PartPtr, StTy->getPointerTo());
Builder.CreateStore(StoredVal[Part], VecPtr)->setAlignment(Alignment);
}
break;
}
case Instruction::Load: {
// Attempt to issue a wide load.
LoadInst *LI = dyn_cast<LoadInst>(it);
Type *RetTy = VectorType::get(LI->getType(), VF);
Value *Ptr = LI->getPointerOperand();
unsigned Alignment = LI->getAlignment();
// If the pointer is loop invariant or if it is non consecutive,
// scalarize the load.
int Stride = Legal->isConsecutivePtr(Ptr);
bool Reverse = Stride < 0;
if (Legal->isUniform(Ptr) || Stride == 0) {
scalarizeInstruction(it);
break;
}
GetElementPtrInst *Gep = dyn_cast<GetElementPtrInst>(Ptr);
if (Gep) {
// The last index does not have to be the induction. It can be
// consecutive and be a function of the index. For example A[I+1];
unsigned NumOperands = Gep->getNumOperands();
Value *LastGepOperand = Gep->getOperand(NumOperands - 1);
VectorParts &GEPParts = getVectorValue(LastGepOperand);
Value *LastIndex = GEPParts[0];
LastIndex = Builder.CreateExtractElement(LastIndex, Zero);
// Create the new GEP with the new induction variable.
GetElementPtrInst *Gep2 = cast<GetElementPtrInst>(Gep->clone());
Gep2->setOperand(NumOperands - 1, LastIndex);
Ptr = Builder.Insert(Gep2);
} else {
// Use the induction element ptr.
assert(isa<PHINode>(Ptr) && "Invalid induction ptr");
VectorParts &PtrVal = getVectorValue(Ptr);
Ptr = Builder.CreateExtractElement(PtrVal[0], Zero);
}
for (unsigned Part = 0; Part < UF; ++Part) {
// Calculate the pointer for the specific unroll-part.
Value *PartPtr = Builder.CreateGEP(Ptr, Builder.getInt32(Part * VF));
if (Reverse) {
// If the address is consecutive but reversed, then the
// wide store needs to start at the last vector element.
PartPtr = Builder.CreateGEP(Ptr, Builder.getInt32(-Part * VF));
PartPtr = Builder.CreateGEP(PartPtr, Builder.getInt32(1 - VF));
}
Value *VecPtr = Builder.CreateBitCast(PartPtr, RetTy->getPointerTo());
Value *LI = Builder.CreateLoad(VecPtr, "wide.load");
cast<LoadInst>(LI)->setAlignment(Alignment);
Entry[Part] = Reverse ? reverseVector(LI) : LI;
}
break;
}
case Instruction::ZExt:
case Instruction::SExt:
case Instruction::FPToUI:
case Instruction::FPToSI:
case Instruction::FPExt:
case Instruction::PtrToInt:
case Instruction::IntToPtr:
case Instruction::SIToFP:
case Instruction::UIToFP:
case Instruction::Trunc:
case Instruction::FPTrunc:
case Instruction::BitCast: {
CastInst *CI = dyn_cast<CastInst>(it);
/// Optimize the special case where the source is the induction
/// variable. Notice that we can only optimize the 'trunc' case
/// because: a. FP conversions lose precision, b. sext/zext may wrap,
/// c. other casts depend on pointer size.
if (CI->getOperand(0) == OldInduction &&
it->getOpcode() == Instruction::Trunc) {
Value *ScalarCast = Builder.CreateCast(CI->getOpcode(), Induction,
CI->getType());
Value *Broadcasted = getBroadcastInstrs(ScalarCast);
for (unsigned Part = 0; Part < UF; ++Part)
Entry[Part] = getConsecutiveVector(Broadcasted, VF * Part, false);
break;
}
/// Vectorize casts.
Type *DestTy = VectorType::get(CI->getType()->getScalarType(), VF);
VectorParts &A = getVectorValue(it->getOperand(0));
for (unsigned Part = 0; Part < UF; ++Part)
Entry[Part] = Builder.CreateCast(CI->getOpcode(), A[Part], DestTy);
break;
}
case Instruction::Call: {
assert(isTriviallyVectorizableIntrinsic(it));
Module *M = BB->getParent()->getParent();
IntrinsicInst *II = cast<IntrinsicInst>(it);
Intrinsic::ID ID = II->getIntrinsicID();
for (unsigned Part = 0; Part < UF; ++Part) {
SmallVector<Value*, 4> Args;
for (unsigned i = 0, ie = II->getNumArgOperands(); i != ie; ++i) {
VectorParts &Arg = getVectorValue(II->getArgOperand(i));
Args.push_back(Arg[Part]);
}
Type *Tys[] = { VectorType::get(II->getType()->getScalarType(), VF) };
Function *F = Intrinsic::getDeclaration(M, ID, Tys);
Entry[Part] = Builder.CreateCall(F, Args);
}
break;
}
default:
// All other instructions are unsupported. Scalarize them.
scalarizeInstruction(it);
break;
}// end of switch.
}// end of for_each instr.
}
void InnerLoopVectorizer::updateAnalysis() {
// Forget the original basic block.
SE->forgetLoop(OrigLoop);
// Update the dominator tree information.
assert(DT->properlyDominates(LoopBypassBlock, LoopExitBlock) &&
"Entry does not dominate exit.");
DT->addNewBlock(LoopVectorPreHeader, LoopBypassBlock);
DT->addNewBlock(LoopVectorBody, LoopVectorPreHeader);
DT->addNewBlock(LoopMiddleBlock, LoopBypassBlock);
DT->addNewBlock(LoopScalarPreHeader, LoopMiddleBlock);
DT->changeImmediateDominator(LoopScalarBody, LoopScalarPreHeader);
DT->changeImmediateDominator(LoopExitBlock, LoopMiddleBlock);
DEBUG(DT->verifyAnalysis());
}
bool LoopVectorizationLegality::canVectorizeWithIfConvert() {
if (!EnableIfConversion)
return false;
assert(TheLoop->getNumBlocks() > 1 && "Single block loops are vectorizable");
std::vector<BasicBlock*> &LoopBlocks = TheLoop->getBlocksVector();
// Collect the blocks that need predication.
for (unsigned i = 0, e = LoopBlocks.size(); i < e; ++i) {
BasicBlock *BB = LoopBlocks[i];
// We don't support switch statements inside loops.
if (!isa<BranchInst>(BB->getTerminator()))
return false;
// We must have at most two predecessors because we need to convert
// all PHIs to selects.
unsigned Preds = std::distance(pred_begin(BB), pred_end(BB));
if (Preds > 2)
return false;
// We must be able to predicate all blocks that need to be predicated.
if (blockNeedsPredication(BB) && !blockCanBePredicated(BB))
return false;
}
// We can if-convert this loop.
return true;
}
bool LoopVectorizationLegality::canVectorize() {
assert(TheLoop->getLoopPreheader() && "No preheader!!");
// We can only vectorize innermost loops.
if (TheLoop->getSubLoopsVector().size())
return false;
// We must have a single backedge.
if (TheLoop->getNumBackEdges() != 1)
return false;
// We must have a single exiting block.
if (!TheLoop->getExitingBlock())
return false;
unsigned NumBlocks = TheLoop->getNumBlocks();
// Check if we can if-convert non single-bb loops.
if (NumBlocks != 1 && !canVectorizeWithIfConvert()) {
DEBUG(dbgs() << "LV: Can't if-convert the loop.\n");
return false;
}
// We need to have a loop header.
BasicBlock *Latch = TheLoop->getLoopLatch();
DEBUG(dbgs() << "LV: Found a loop: " <<
TheLoop->getHeader()->getName() << "\n");
// ScalarEvolution needs to be able to find the exit count.
const SCEV *ExitCount = SE->getExitCount(TheLoop, Latch);
if (ExitCount == SE->getCouldNotCompute()) {
DEBUG(dbgs() << "LV: SCEV could not compute the loop exit count.\n");
return false;
}
// Do not loop-vectorize loops with a tiny trip count.
unsigned TC = SE->getSmallConstantTripCount(TheLoop, Latch);
if (TC > 0u && TC < TinyTripCountThreshold) {
DEBUG(dbgs() << "LV: Found a loop with a very small trip count. " <<
"This loop is not worth vectorizing.\n");
return false;
}
// Check if we can vectorize the instructions and CFG in this loop.
if (!canVectorizeInstrs()) {
DEBUG(dbgs() << "LV: Can't vectorize the instructions or CFG\n");
return false;
}
// Go over each instruction and look at memory deps.
if (!canVectorizeMemory()) {
DEBUG(dbgs() << "LV: Can't vectorize due to memory conflicts\n");
return false;
}
// Collect all of the variables that remain uniform after vectorization.
collectLoopUniforms();
DEBUG(dbgs() << "LV: We can vectorize this loop" <<
(PtrRtCheck.Need ? " (with a runtime bound check)" : "")
<<"!\n");
// Okay! We can vectorize. At this point we don't have any other mem analysis
// which may limit our maximum vectorization factor, so just return true with
// no restrictions.
return true;
}
bool LoopVectorizationLegality::canVectorizeInstrs() {
BasicBlock *PreHeader = TheLoop->getLoopPreheader();
BasicBlock *Header = TheLoop->getHeader();
// For each block in the loop.
for (Loop::block_iterator bb = TheLoop->block_begin(),
be = TheLoop->block_end(); bb != be; ++bb) {
// Scan the instructions in the block and look for hazards.
for (BasicBlock::iterator it = (*bb)->begin(), e = (*bb)->end(); it != e;
++it) {
if (PHINode *Phi = dyn_cast<PHINode>(it)) {
// This should not happen because the loop should be normalized.
if (Phi->getNumIncomingValues() != 2) {
DEBUG(dbgs() << "LV: Found an invalid PHI.\n");
return false;
}
// Check that this PHI type is allowed.
if (!Phi->getType()->isIntegerTy() &&
!Phi->getType()->isPointerTy()) {
DEBUG(dbgs() << "LV: Found an non-int non-pointer PHI.\n");
return false;
}
// If this PHINode is not in the header block, then we know that we
// can convert it to select during if-conversion. No need to check if
// the PHIs in this block are induction or reduction variables.
if (*bb != Header)
continue;
// This is the value coming from the preheader.
Value *StartValue = Phi->getIncomingValueForBlock(PreHeader);
// Check if this is an induction variable.
InductionKind IK = isInductionVariable(Phi);
if (NoInduction != IK) {
// Int inductions are special because we only allow one IV.
if (IK == IntInduction) {
if (Induction) {
DEBUG(dbgs() << "LV: Found too many inductions."<< *Phi <<"\n");
return false;
}
Induction = Phi;
}
DEBUG(dbgs() << "LV: Found an induction variable.\n");
Inductions[Phi] = InductionInfo(StartValue, IK);
continue;
}
if (AddReductionVar(Phi, IntegerAdd)) {
DEBUG(dbgs() << "LV: Found an ADD reduction PHI."<< *Phi <<"\n");
continue;
}
if (AddReductionVar(Phi, IntegerMult)) {
DEBUG(dbgs() << "LV: Found a MUL reduction PHI."<< *Phi <<"\n");
continue;
}
if (AddReductionVar(Phi, IntegerOr)) {
DEBUG(dbgs() << "LV: Found an OR reduction PHI."<< *Phi <<"\n");
continue;
}
if (AddReductionVar(Phi, IntegerAnd)) {
DEBUG(dbgs() << "LV: Found an AND reduction PHI."<< *Phi <<"\n");
continue;
}
if (AddReductionVar(Phi, IntegerXor)) {
DEBUG(dbgs() << "LV: Found a XOR reduction PHI."<< *Phi <<"\n");
continue;
}
DEBUG(dbgs() << "LV: Found an unidentified PHI."<< *Phi <<"\n");
return false;
}// end of PHI handling
// We still don't handle functions.
CallInst *CI = dyn_cast<CallInst>(it);
if (CI && !isTriviallyVectorizableIntrinsic(it)) {
DEBUG(dbgs() << "LV: Found a call site.\n");
return false;
}
// Check that the instruction return type is vectorizable.
if (!VectorType::isValidElementType(it->getType()) &&
!it->getType()->isVoidTy()) {
DEBUG(dbgs() << "LV: Found unvectorizable type." << "\n");
return false;
}
// Check that the stored type is vectorizable.
if (StoreInst *ST = dyn_cast<StoreInst>(it)) {
Type *T = ST->getValueOperand()->getType();
if (!VectorType::isValidElementType(T))
return false;
}
// Reduction instructions are allowed to have exit users.
// All other instructions must not have external users.
if (!AllowedExit.count(it))
//Check that all of the users of the loop are inside the BB.
for (Value::use_iterator I = it->use_begin(), E = it->use_end();
I != E; ++I) {
Instruction *U = cast<Instruction>(*I);
// This user may be a reduction exit value.
if (!TheLoop->contains(U)) {
DEBUG(dbgs() << "LV: Found an outside user for : "<< *U << "\n");
return false;
}
}
} // next instr.
}
if (!Induction) {
DEBUG(dbgs() << "LV: Did not find one integer induction var.\n");
assert(getInductionVars()->size() && "No induction variables");
}
return true;
}
void LoopVectorizationLegality::collectLoopUniforms() {
// We now know that the loop is vectorizable!
// Collect variables that will remain uniform after vectorization.
std::vector<Value*> Worklist;
BasicBlock *Latch = TheLoop->getLoopLatch();
// Start with the conditional branch and walk up the block.
Worklist.push_back(Latch->getTerminator()->getOperand(0));
while (Worklist.size()) {
Instruction *I = dyn_cast<Instruction>(Worklist.back());
Worklist.pop_back();
// Look at instructions inside this loop.
// Stop when reaching PHI nodes.
// TODO: we need to follow values all over the loop, not only in this block.
if (!I || !TheLoop->contains(I) || isa<PHINode>(I))
continue;
// This is a known uniform.
Uniforms.insert(I);
// Insert all operands.
for (int i = 0, Op = I->getNumOperands(); i < Op; ++i) {
Worklist.push_back(I->getOperand(i));
}
}
}
bool LoopVectorizationLegality::canVectorizeMemory() {
typedef SmallVector<Value*, 16> ValueVector;
typedef SmallPtrSet<Value*, 16> ValueSet;
// Holds the Load and Store *instructions*.
ValueVector Loads;
ValueVector Stores;
PtrRtCheck.Pointers.clear();
PtrRtCheck.Need = false;
// For each block.
for (Loop::block_iterator bb = TheLoop->block_begin(),
be = TheLoop->block_end(); bb != be; ++bb) {
// Scan the BB and collect legal loads and stores.
for (BasicBlock::iterator it = (*bb)->begin(), e = (*bb)->end(); it != e;
++it) {
// If this is a load, save it. If this instruction can read from memory
// but is not a load, then we quit. Notice that we don't handle function
// calls that read or write.
if (it->mayReadFromMemory()) {
LoadInst *Ld = dyn_cast<LoadInst>(it);
if (!Ld) return false;
if (!Ld->isSimple()) {
DEBUG(dbgs() << "LV: Found a non-simple load.\n");
return false;
}
Loads.push_back(Ld);
continue;
}
// Save 'store' instructions. Abort if other instructions write to memory.
if (it->mayWriteToMemory()) {
StoreInst *St = dyn_cast<StoreInst>(it);
if (!St) return false;
if (!St->isSimple()) {
DEBUG(dbgs() << "LV: Found a non-simple store.\n");
return false;
}
Stores.push_back(St);
}
} // next instr.
} // next block.
// Now we have two lists that hold the loads and the stores.
// Next, we find the pointers that they use.
// Check if we see any stores. If there are no stores, then we don't
// care if the pointers are *restrict*.
if (!Stores.size()) {
DEBUG(dbgs() << "LV: Found a read-only loop!\n");
return true;
}
// Holds the read and read-write *pointers* that we find.
ValueVector Reads;
ValueVector ReadWrites;
// Holds the analyzed pointers. We don't want to call GetUnderlyingObjects
// multiple times on the same object. If the ptr is accessed twice, once
// for read and once for write, it will only appear once (on the write
// list). This is okay, since we are going to check for conflicts between
// writes and between reads and writes, but not between reads and reads.
ValueSet Seen;
ValueVector::iterator I, IE;
for (I = Stores.begin(), IE = Stores.end(); I != IE; ++I) {
StoreInst *ST = cast<StoreInst>(*I);
Value* Ptr = ST->getPointerOperand();
if (isUniform(Ptr)) {
DEBUG(dbgs() << "LV: We don't allow storing to uniform addresses\n");
return false;
}
// If we did *not* see this pointer before, insert it to
// the read-write list. At this phase it is only a 'write' list.
if (Seen.insert(Ptr))
ReadWrites.push_back(Ptr);
}
for (I = Loads.begin(), IE = Loads.end(); I != IE; ++I) {
LoadInst *LD = cast<LoadInst>(*I);
Value* Ptr = LD->getPointerOperand();
// If we did *not* see this pointer before, insert it to the
// read list. If we *did* see it before, then it is already in
// the read-write list. This allows us to vectorize expressions
// such as A[i] += x; Because the address of A[i] is a read-write
// pointer. This only works if the index of A[i] is consecutive.
// If the address of i is unknown (for example A[B[i]]) then we may
// read a few words, modify, and write a few words, and some of the
// words may be written to the same address.
if (Seen.insert(Ptr) || 0 == isConsecutivePtr(Ptr))
Reads.push_back(Ptr);
}
// If we write (or read-write) to a single destination and there are no
// other reads in this loop then is it safe to vectorize.
if (ReadWrites.size() == 1 && Reads.size() == 0) {
DEBUG(dbgs() << "LV: Found a write-only loop!\n");
return true;
}
// Find pointers with computable bounds. We are going to use this information
// to place a runtime bound check.
bool CanDoRT = true;
for (I = ReadWrites.begin(), IE = ReadWrites.end(); I != IE; ++I)
if (hasComputableBounds(*I)) {
PtrRtCheck.insert(SE, TheLoop, *I);
DEBUG(dbgs() << "LV: Found a runtime check ptr:" << **I <<"\n");
} else {
CanDoRT = false;
break;
}
for (I = Reads.begin(), IE = Reads.end(); I != IE; ++I)
if (hasComputableBounds(*I)) {
PtrRtCheck.insert(SE, TheLoop, *I);
DEBUG(dbgs() << "LV: Found a runtime check ptr:" << **I <<"\n");
} else {
CanDoRT = false;
break;
}
// Check that we did not collect too many pointers or found a
// unsizeable pointer.
if (!CanDoRT || PtrRtCheck.Pointers.size() > RuntimeMemoryCheckThreshold) {
PtrRtCheck.reset();
CanDoRT = false;
}
if (CanDoRT) {
DEBUG(dbgs() << "LV: We can perform a memory runtime check if needed.\n");
}
bool NeedRTCheck = false;
// Now that the pointers are in two lists (Reads and ReadWrites), we
// can check that there are no conflicts between each of the writes and
// between the writes to the reads.
ValueSet WriteObjects;
ValueVector TempObjects;
// Check that the read-writes do not conflict with other read-write
// pointers.
bool AllWritesIdentified = true;
for (I = ReadWrites.begin(), IE = ReadWrites.end(); I != IE; ++I) {
GetUnderlyingObjects(*I, TempObjects, DL);
for (ValueVector::iterator it=TempObjects.begin(), e=TempObjects.end();
it != e; ++it) {
if (!isIdentifiedObject(*it)) {
DEBUG(dbgs() << "LV: Found an unidentified write ptr:"<< **it <<"\n");
NeedRTCheck = true;
AllWritesIdentified = false;
}
if (!WriteObjects.insert(*it)) {
DEBUG(dbgs() << "LV: Found a possible write-write reorder:"
<< **it <<"\n");
return false;
}
}
TempObjects.clear();
}
/// Check that the reads don't conflict with the read-writes.
for (I = Reads.begin(), IE = Reads.end(); I != IE; ++I) {
GetUnderlyingObjects(*I, TempObjects, DL);
for (ValueVector::iterator it=TempObjects.begin(), e=TempObjects.end();
it != e; ++it) {
// If all of the writes are identified then we don't care if the read
// pointer is identified or not.
if (!AllWritesIdentified && !isIdentifiedObject(*it)) {
DEBUG(dbgs() << "LV: Found an unidentified read ptr:"<< **it <<"\n");
NeedRTCheck = true;
}
if (WriteObjects.count(*it)) {
DEBUG(dbgs() << "LV: Found a possible read/write reorder:"
<< **it <<"\n");
return false;
}
}
TempObjects.clear();
}
PtrRtCheck.Need = NeedRTCheck;
if (NeedRTCheck && !CanDoRT) {
DEBUG(dbgs() << "LV: We can't vectorize because we can't find " <<
"the array bounds.\n");
PtrRtCheck.reset();
return false;
}
DEBUG(dbgs() << "LV: We "<< (NeedRTCheck ? "" : "don't") <<
" need a runtime memory check.\n");
return true;
}
bool LoopVectorizationLegality::AddReductionVar(PHINode *Phi,
ReductionKind Kind) {
if (Phi->getNumIncomingValues() != 2)
return false;
// Reduction variables are only found in the loop header block.
if (Phi->getParent() != TheLoop->getHeader())
return false;
// Obtain the reduction start value from the value that comes from the loop
// preheader.
Value *RdxStart = Phi->getIncomingValueForBlock(TheLoop->getLoopPreheader());
// ExitInstruction is the single value which is used outside the loop.
// We only allow for a single reduction value to be used outside the loop.
// This includes users of the reduction, variables (which form a cycle
// which ends in the phi node).
Instruction *ExitInstruction = 0;
// Iter is our iterator. We start with the PHI node and scan for all of the
// users of this instruction. All users must be instructions that can be
// used as reduction variables (such as ADD). We may have a single
// out-of-block user. The cycle must end with the original PHI.
Instruction *Iter = Phi;
while (true) {
// If the instruction has no users then this is a broken
// chain and can't be a reduction variable.
if (Iter->use_empty())
return false;
// Did we find a user inside this loop already ?
bool FoundInBlockUser = false;
// Did we reach the initial PHI node already ?
bool FoundStartPHI = false;
// For each of the *users* of iter.
for (Value::use_iterator it = Iter->use_begin(), e = Iter->use_end();
it != e; ++it) {
Instruction *U = cast<Instruction>(*it);
// We already know that the PHI is a user.
if (U == Phi) {
FoundStartPHI = true;
continue;
}
// Check if we found the exit user.
BasicBlock *Parent = U->getParent();
if (!TheLoop->contains(Parent)) {
// Exit if you find multiple outside users.
if (ExitInstruction != 0)
return false;
ExitInstruction = Iter;
}
// We allow in-loop PHINodes which are not the original reduction PHI
// node. If this PHI is the only user of Iter (happens in IF w/ no ELSE
// structure) then don't skip this PHI.
if (isa<PHINode>(Iter) && isa<PHINode>(U) &&
U->getParent() != TheLoop->getHeader() &&
TheLoop->contains(U) &&
Iter->getNumUses() > 1)
continue;
// We can't have multiple inside users.
if (FoundInBlockUser)
return false;
FoundInBlockUser = true;
// Any reduction instr must be of one of the allowed kinds.
if (!isReductionInstr(U, Kind))
return false;
// Reductions of instructions such as Div, and Sub is only
// possible if the LHS is the reduction variable.
if (!U->isCommutative() && U->getOperand(0) != Iter)
return false;
Iter = U;
}
// We found a reduction var if we have reached the original
// phi node and we only have a single instruction with out-of-loop
// users.
if (FoundStartPHI && ExitInstruction) {
// This instruction is allowed to have out-of-loop users.
AllowedExit.insert(ExitInstruction);
// Save the description of this reduction variable.
ReductionDescriptor RD(RdxStart, ExitInstruction, Kind);
Reductions[Phi] = RD;
return true;
}
// If we've reached the start PHI but did not find an outside user then
// this is dead code. Abort.
if (FoundStartPHI)
return false;
}
}
bool
LoopVectorizationLegality::isReductionInstr(Instruction *I,
ReductionKind Kind) {
switch (I->getOpcode()) {
default:
return false;
case Instruction::PHI:
// possibly.
return true;
case Instruction::Sub:
case Instruction::Add:
return Kind == IntegerAdd;
case Instruction::SDiv:
case Instruction::UDiv:
case Instruction::Mul:
return Kind == IntegerMult;
case Instruction::And:
return Kind == IntegerAnd;
case Instruction::Or:
return Kind == IntegerOr;
case Instruction::Xor:
return Kind == IntegerXor;
}
}
LoopVectorizationLegality::InductionKind
LoopVectorizationLegality::isInductionVariable(PHINode *Phi) {
Type *PhiTy = Phi->getType();
// We only handle integer and pointer inductions variables.
if (!PhiTy->isIntegerTy() && !PhiTy->isPointerTy())
return NoInduction;
// Check that the PHI is consecutive and starts at zero.
const SCEV *PhiScev = SE->getSCEV(Phi);
const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(PhiScev);
if (!AR) {
DEBUG(dbgs() << "LV: PHI is not a poly recurrence.\n");
return NoInduction;
}
const SCEV *Step = AR->getStepRecurrence(*SE);
// Integer inductions need to have a stride of one.
if (PhiTy->isIntegerTy()) {
if (Step->isOne())
return IntInduction;
if (Step->isAllOnesValue())
return ReverseIntInduction;
return NoInduction;
}
// Calculate the pointer stride and check if it is consecutive.
const SCEVConstant *C = dyn_cast<SCEVConstant>(Step);
if (!C)
return NoInduction;
assert(PhiTy->isPointerTy() && "The PHI must be a pointer");
uint64_t Size = DL->getTypeAllocSize(PhiTy->getPointerElementType());
if (C->getValue()->equalsInt(Size))
return PtrInduction;
return NoInduction;
}
bool LoopVectorizationLegality::isInductionVariable(const Value *V) {
Value *In0 = const_cast<Value*>(V);
PHINode *PN = dyn_cast_or_null<PHINode>(In0);
if (!PN)
return false;
return Inductions.count(PN);
}
bool LoopVectorizationLegality::blockNeedsPredication(BasicBlock *BB) {
assert(TheLoop->contains(BB) && "Unknown block used");
// Blocks that do not dominate the latch need predication.
BasicBlock* Latch = TheLoop->getLoopLatch();
return !DT->dominates(BB, Latch);
}
bool LoopVectorizationLegality::blockCanBePredicated(BasicBlock *BB) {
for (BasicBlock::iterator it = BB->begin(), e = BB->end(); it != e; ++it) {
// We don't predicate loads/stores at the moment.
if (it->mayReadFromMemory() || it->mayWriteToMemory() || it->mayThrow())
return false;
// The instructions below can trap.
switch (it->getOpcode()) {
default: continue;
case Instruction::UDiv:
case Instruction::SDiv:
case Instruction::URem:
case Instruction::SRem:
return false;
}
}
return true;
}
bool LoopVectorizationLegality::hasComputableBounds(Value *Ptr) {
const SCEV *PhiScev = SE->getSCEV(Ptr);
const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(PhiScev);
if (!AR)
return false;
return AR->isAffine();
}
unsigned
LoopVectorizationCostModel::selectVectorizationFactor(bool OptForSize,
unsigned UserVF) {
if (OptForSize && Legal->getRuntimePointerCheck()->Need) {
DEBUG(dbgs() << "LV: Aborting. Runtime ptr check is required in Os.\n");
return 1;
}
// Find the trip count.
unsigned TC = SE->getSmallConstantTripCount(TheLoop, TheLoop->getLoopLatch());
DEBUG(dbgs() << "LV: Found trip count:"<<TC<<"\n");
unsigned VF = MaxVectorSize;
// If we optimize the program for size, avoid creating the tail loop.
if (OptForSize) {
// If we are unable to calculate the trip count then don't try to vectorize.
if (TC < 2) {
DEBUG(dbgs() << "LV: Aborting. A tail loop is required in Os.\n");
return 1;
}
// Find the maximum SIMD width that can fit within the trip count.
VF = TC % MaxVectorSize;
if (VF == 0)
VF = MaxVectorSize;
// If the trip count that we found modulo the vectorization factor is not
// zero then we require a tail.
if (VF < 2) {
DEBUG(dbgs() << "LV: Aborting. A tail loop is required in Os.\n");
return 1;
}
}
if (UserVF != 0) {
assert(isPowerOf2_32(UserVF) && "VF needs to be a power of two");
DEBUG(dbgs() << "LV: Using user VF "<<UserVF<<".\n");
return UserVF;
}
if (!TTI) {
DEBUG(dbgs() << "LV: No vector target information. Not vectorizing. \n");
return 1;
}
float Cost = expectedCost(1);
unsigned Width = 1;
DEBUG(dbgs() << "LV: Scalar loop costs: "<< (int)Cost << ".\n");
for (unsigned i=2; i <= VF; i*=2) {
// Notice that the vector loop needs to be executed less times, so
// we need to divide the cost of the vector loops by the width of
// the vector elements.
float VectorCost = expectedCost(i) / (float)i;
DEBUG(dbgs() << "LV: Vector loop of width "<< i << " costs: " <<
(int)VectorCost << ".\n");
if (VectorCost < Cost) {
Cost = VectorCost;
Width = i;
}
}
DEBUG(dbgs() << "LV: Selecting VF = : "<< Width << ".\n");
return Width;
}
unsigned
LoopVectorizationCostModel::selectUnrollFactor(bool OptForSize,
unsigned UserUF) {
// Use the user preference, unless 'auto' is selected.
if (UserUF != 0)
return UserUF;
// When we optimize for size we don't unroll.
if (OptForSize)
return 1;
unsigned TargetVectorRegisters = TTI->getNumberOfRegisters(true);
DEBUG(dbgs() << "LV: The target has " << TargetVectorRegisters <<
" vector registers\n");
LoopVectorizationCostModel::RegisterUsage R = calculateRegisterUsage();
// We divide by these constants so assume that we have at least one
// instruction that uses at least one register.
R.MaxLocalUsers = std::max(R.MaxLocalUsers, 1U);
R.NumInstructions = std::max(R.NumInstructions, 1U);
// We calculate the unroll factor using the following formula.
// Subtract the number of loop invariants from the number of available
// registers. These registers are used by all of the unrolled instances.
// Next, divide the remaining registers by the number of registers that is
// required by the loop, in order to estimate how many parallel instances
// fit without causing spills.
unsigned UF = (TargetVectorRegisters - R.LoopInvariantRegs) / R.MaxLocalUsers;
// We don't want to unroll the loops to the point where they do not fit into
// the decoded cache. Assume that we only allow 32 IR instructions.
UF = std::min(UF, (32 / R.NumInstructions));
// Clamp the unroll factor ranges to reasonable factors.
if (UF > MaxUnrollSize)
UF = MaxUnrollSize;
else if (UF < 1)
UF = 1;
return UF;
}
LoopVectorizationCostModel::RegisterUsage
LoopVectorizationCostModel::calculateRegisterUsage() {
// This function calculates the register usage by measuring the highest number
// of values that are alive at a single location. Obviously, this is a very
// rough estimation. We scan the loop in a topological order in order and
// assign a number to each instruction. We use RPO to ensure that defs are
// met before their users. We assume that each instruction that has in-loop
// users starts an interval. We record every time that an in-loop value is
// used, so we have a list of the first and last occurrences of each
// instruction. Next, we transpose this data structure into a multi map that
// holds the list of intervals that *end* at a specific location. This multi
// map allows us to perform a linear search. We scan the instructions linearly
// and record each time that a new interval starts, by placing it in a set.
// If we find this value in the multi-map then we remove it from the set.
// The max register usage is the maximum size of the set.
// We also search for instructions that are defined outside the loop, but are
// used inside the loop. We need this number separately from the max-interval
// usage number because when we unroll, loop-invariant values do not take
// more register.
LoopBlocksDFS DFS(TheLoop);
DFS.perform(LI);
RegisterUsage R;
R.NumInstructions = 0;
// Each 'key' in the map opens a new interval. The values
// of the map are the index of the 'last seen' usage of the
// instruction that is the key.
typedef DenseMap<Instruction*, unsigned> IntervalMap;
// Maps instruction to its index.
DenseMap<unsigned, Instruction*> IdxToInstr;
// Marks the end of each interval.
IntervalMap EndPoint;
// Saves the list of instruction indices that are used in the loop.
SmallSet<Instruction*, 8> Ends;
// Saves the list of values that are used in the loop but are
// defined outside the loop, such as arguments and constants.
SmallPtrSet<Value*, 8> LoopInvariants;
unsigned Index = 0;
for (LoopBlocksDFS::RPOIterator bb = DFS.beginRPO(),
be = DFS.endRPO(); bb != be; ++bb) {
R.NumInstructions += (*bb)->size();
for (BasicBlock::iterator it = (*bb)->begin(), e = (*bb)->end(); it != e;
++it) {
Instruction *I = it;
IdxToInstr[Index++] = I;
// Save the end location of each USE.
for (unsigned i = 0; i < I->getNumOperands(); ++i) {
Value *U = I->getOperand(i);
Instruction *Instr = dyn_cast<Instruction>(U);
// Ignore non-instruction values such as arguments, constants, etc.
if (!Instr) continue;
// If this instruction is outside the loop then record it and continue.
if (!TheLoop->contains(Instr)) {
LoopInvariants.insert(Instr);
continue;
}
// Overwrite previous end points.
EndPoint[Instr] = Index;
Ends.insert(Instr);
}
}
}
// Saves the list of intervals that end with the index in 'key'.
typedef SmallVector<Instruction*, 2> InstrList;
DenseMap<unsigned, InstrList> TransposeEnds;
// Transpose the EndPoints to a list of values that end at each index.
for (IntervalMap::iterator it = EndPoint.begin(), e = EndPoint.end();
it != e; ++it)
TransposeEnds[it->second].push_back(it->first);
SmallSet<Instruction*, 8> OpenIntervals;
unsigned MaxUsage = 0;
DEBUG(dbgs() << "LV(REG): Calculating max register usage:\n");
for (unsigned int i = 0; i < Index; ++i) {
Instruction *I = IdxToInstr[i];
// Ignore instructions that are never used within the loop.
if (!Ends.count(I)) continue;
// Remove all of the instructions that end at this location.
InstrList &List = TransposeEnds[i];
for (unsigned int j=0, e = List.size(); j < e; ++j)
OpenIntervals.erase(List[j]);
// Count the number of live interals.
MaxUsage = std::max(MaxUsage, OpenIntervals.size());
DEBUG(dbgs() << "LV(REG): At #" << i << " Interval # " <<
OpenIntervals.size() <<"\n");
// Add the current instruction to the list of open intervals.
OpenIntervals.insert(I);
}
unsigned Invariant = LoopInvariants.size();
DEBUG(dbgs() << "LV(REG): Found max usage: " << MaxUsage << " \n");
DEBUG(dbgs() << "LV(REG): Found invariant usage: " << Invariant << " \n");
DEBUG(dbgs() << "LV(REG): LoopSize: " << R.NumInstructions << " \n");
R.LoopInvariantRegs = Invariant;
R.MaxLocalUsers = MaxUsage;
return R;
}
unsigned LoopVectorizationCostModel::expectedCost(unsigned VF) {
unsigned Cost = 0;
// For each block.
for (Loop::block_iterator bb = TheLoop->block_begin(),
be = TheLoop->block_end(); bb != be; ++bb) {
unsigned BlockCost = 0;
BasicBlock *BB = *bb;
// For each instruction in the old loop.
for (BasicBlock::iterator it = BB->begin(), e = BB->end(); it != e; ++it) {
unsigned C = getInstructionCost(it, VF);
Cost += C;
DEBUG(dbgs() << "LV: Found an estimated cost of "<< C <<" for VF " <<
VF << " For instruction: "<< *it << "\n");
}
// We assume that if-converted blocks have a 50% chance of being executed.
// When the code is scalar then some of the blocks are avoided due to CF.
// When the code is vectorized we execute all code paths.
if (Legal->blockNeedsPredication(*bb) && VF == 1)
BlockCost /= 2;
Cost += BlockCost;
}
return Cost;
}
unsigned
LoopVectorizationCostModel::getInstructionCost(Instruction *I, unsigned VF) {
assert(TTI && "Invalid vector target transformation info");
// If we know that this instruction will remain uniform, check the cost of
// the scalar version.
if (Legal->isUniformAfterVectorization(I))
VF = 1;
Type *RetTy = I->getType();
Type *VectorTy = ToVectorTy(RetTy, VF);
// TODO: We need to estimate the cost of intrinsic calls.
switch (I->getOpcode()) {
case Instruction::GetElementPtr:
// We mark this instruction as zero-cost because scalar GEPs are usually
// lowered to the intruction addressing mode. At the moment we don't
// generate vector geps.
return 0;
case Instruction::Br: {
return TTI->getCFInstrCost(I->getOpcode());
}
case Instruction::PHI:
//TODO: IF-converted IFs become selects.
return 0;
case Instruction::Add:
case Instruction::FAdd:
case Instruction::Sub:
case Instruction::FSub:
case Instruction::Mul:
case Instruction::FMul:
case Instruction::UDiv:
case Instruction::SDiv:
case Instruction::FDiv:
case Instruction::URem:
case Instruction::SRem:
case Instruction::FRem:
case Instruction::Shl:
case Instruction::LShr:
case Instruction::AShr:
case Instruction::And:
case Instruction::Or:
case Instruction::Xor:
return TTI->getArithmeticInstrCost(I->getOpcode(), VectorTy);
case Instruction::Select: {
SelectInst *SI = cast<SelectInst>(I);
const SCEV *CondSCEV = SE->getSCEV(SI->getCondition());
bool ScalarCond = (SE->isLoopInvariant(CondSCEV, TheLoop));
Type *CondTy = SI->getCondition()->getType();
if (ScalarCond)
CondTy = VectorType::get(CondTy, VF);
return TTI->getCmpSelInstrCost(I->getOpcode(), VectorTy, CondTy);
}
case Instruction::ICmp:
case Instruction::FCmp: {
Type *ValTy = I->getOperand(0)->getType();
VectorTy = ToVectorTy(ValTy, VF);
return TTI->getCmpSelInstrCost(I->getOpcode(), VectorTy);
}
case Instruction::Store: {
StoreInst *SI = cast<StoreInst>(I);
Type *ValTy = SI->getValueOperand()->getType();
VectorTy = ToVectorTy(ValTy, VF);
if (VF == 1)
return TTI->getMemoryOpCost(I->getOpcode(), VectorTy,
SI->getAlignment(),
SI->getPointerAddressSpace());
// Scalarized stores.
int Stride = Legal->isConsecutivePtr(SI->getPointerOperand());
bool Reverse = Stride < 0;
if (0 == Stride) {
unsigned Cost = 0;
// The cost of extracting from the value vector and pointer vector.
Type *PtrTy = ToVectorTy(I->getOperand(0)->getType(), VF);
for (unsigned i = 0; i < VF; ++i) {
Cost += TTI->getVectorInstrCost(Instruction::ExtractElement,
VectorTy, i);
Cost += TTI->getVectorInstrCost(Instruction::ExtractElement,
PtrTy, i);
}
// The cost of the scalar stores.
Cost += VF * TTI->getMemoryOpCost(I->getOpcode(),
ValTy->getScalarType(),
SI->getAlignment(),
SI->getPointerAddressSpace());
return Cost;
}
// Wide stores.
unsigned Cost = TTI->getMemoryOpCost(I->getOpcode(), VectorTy,
SI->getAlignment(),
SI->getPointerAddressSpace());
if (Reverse)
Cost += TTI->getShuffleCost(TargetTransformInfo::SK_Reverse,
VectorTy, 0);
return Cost;
}
case Instruction::Load: {
LoadInst *LI = cast<LoadInst>(I);
if (VF == 1)
return TTI->getMemoryOpCost(I->getOpcode(), VectorTy,
LI->getAlignment(),
LI->getPointerAddressSpace());
// Scalarized loads.
int Stride = Legal->isConsecutivePtr(LI->getPointerOperand());
bool Reverse = Stride < 0;
if (0 == Stride) {
unsigned Cost = 0;
Type *PtrTy = ToVectorTy(I->getOperand(0)->getType(), VF);
// The cost of extracting from the pointer vector.
for (unsigned i = 0; i < VF; ++i)
Cost += TTI->getVectorInstrCost(Instruction::ExtractElement,
PtrTy, i);
// The cost of inserting data to the result vector.
for (unsigned i = 0; i < VF; ++i)
Cost += TTI->getVectorInstrCost(Instruction::InsertElement,
VectorTy, i);
// The cost of the scalar stores.
Cost += VF * TTI->getMemoryOpCost(I->getOpcode(),
RetTy->getScalarType(),
LI->getAlignment(),
LI->getPointerAddressSpace());
return Cost;
}
// Wide loads.
unsigned Cost = TTI->getMemoryOpCost(I->getOpcode(), VectorTy,
LI->getAlignment(),
LI->getPointerAddressSpace());
if (Reverse)
Cost += TTI->getShuffleCost(TargetTransformInfo::SK_Reverse,
VectorTy, 0);
return Cost;
}
case Instruction::ZExt:
case Instruction::SExt:
case Instruction::FPToUI:
case Instruction::FPToSI:
case Instruction::FPExt:
case Instruction::PtrToInt:
case Instruction::IntToPtr:
case Instruction::SIToFP:
case Instruction::UIToFP:
case Instruction::Trunc:
case Instruction::FPTrunc:
case Instruction::BitCast: {
// We optimize the truncation of induction variable.
// The cost of these is the same as the scalar operation.
if (I->getOpcode() == Instruction::Trunc &&
Legal->isInductionVariable(I->getOperand(0)))
return TTI->getCastInstrCost(I->getOpcode(), I->getType(),
I->getOperand(0)->getType());
Type *SrcVecTy = ToVectorTy(I->getOperand(0)->getType(), VF);
return TTI->getCastInstrCost(I->getOpcode(), VectorTy, SrcVecTy);
}
case Instruction::Call: {
assert(isTriviallyVectorizableIntrinsic(I));
IntrinsicInst *II = cast<IntrinsicInst>(I);
Type *RetTy = ToVectorTy(II->getType(), VF);
SmallVector<Type*, 4> Tys;
for (unsigned i = 0, ie = II->getNumArgOperands(); i != ie; ++i)
Tys.push_back(ToVectorTy(II->getArgOperand(i)->getType(), VF));
return TTI->getIntrinsicInstrCost(II->getIntrinsicID(), RetTy, Tys);
}
default: {
// We are scalarizing the instruction. Return the cost of the scalar
// instruction, plus the cost of insert and extract into vector
// elements, times the vector width.
unsigned Cost = 0;
if (!RetTy->isVoidTy() && VF != 1) {
unsigned InsCost = TTI->getVectorInstrCost(Instruction::InsertElement,
VectorTy);
unsigned ExtCost = TTI->getVectorInstrCost(Instruction::ExtractElement,
VectorTy);
// The cost of inserting the results plus extracting each one of the
// operands.
Cost += VF * (InsCost + ExtCost * I->getNumOperands());
}
// The cost of executing VF copies of the scalar instruction. This opcode
// is unknown. Assume that it is the same as 'mul'.
Cost += VF * TTI->getArithmeticInstrCost(Instruction::Mul, VectorTy);
return Cost;
}
}// end of switch.
}
Type* LoopVectorizationCostModel::ToVectorTy(Type *Scalar, unsigned VF) {
if (Scalar->isVoidTy() || VF == 1)
return Scalar;
return VectorType::get(Scalar, VF);
}
char LoopVectorize::ID = 0;
static const char lv_name[] = "Loop Vectorization";
INITIALIZE_PASS_BEGIN(LoopVectorize, LV_NAME, lv_name, false, false)
INITIALIZE_AG_DEPENDENCY(AliasAnalysis)
INITIALIZE_PASS_DEPENDENCY(ScalarEvolution)
INITIALIZE_PASS_DEPENDENCY(LoopSimplify)
INITIALIZE_PASS_END(LoopVectorize, LV_NAME, lv_name, false, false)
namespace llvm {
Pass *createLoopVectorizePass() {
return new LoopVectorize();
}
}
|