1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
|
/* Functions to determine/estimate number of iterations of a loop.
Copyright (C) 2004-2018 Free Software Foundation, Inc.
This file is part of GCC.
GCC is free software; you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by the
Free Software Foundation; either version 3, or (at your option) any
later version.
GCC is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
for more details.
You should have received a copy of the GNU General Public License
along with GCC; see the file COPYING3. If not see
<http://www.gnu.org/licenses/>. */
#include "config.h"
#include "system.h"
#include "coretypes.h"
#include "backend.h"
#include "rtl.h"
#include "tree.h"
#include "gimple.h"
#include "tree-pass.h"
#include "ssa.h"
#include "gimple-pretty-print.h"
#include "diagnostic-core.h"
#include "stor-layout.h"
#include "fold-const.h"
#include "calls.h"
#include "intl.h"
#include "gimplify.h"
#include "gimple-iterator.h"
#include "tree-cfg.h"
#include "tree-ssa-loop-ivopts.h"
#include "tree-ssa-loop-niter.h"
#include "tree-ssa-loop.h"
#include "cfgloop.h"
#include "tree-chrec.h"
#include "tree-scalar-evolution.h"
#include "params.h"
#include "tree-dfa.h"
/* The maximum number of dominator BBs we search for conditions
of loop header copies we use for simplifying a conditional
expression. */
#define MAX_DOMINATORS_TO_WALK 8
/*
Analysis of number of iterations of an affine exit test.
*/
/* Bounds on some value, BELOW <= X <= UP. */
struct bounds
{
mpz_t below, up;
};
static bool number_of_iterations_popcount (loop_p loop, edge exit,
enum tree_code code,
struct tree_niter_desc *niter);
/* Splits expression EXPR to a variable part VAR and constant OFFSET. */
static void
split_to_var_and_offset (tree expr, tree *var, mpz_t offset)
{
tree type = TREE_TYPE (expr);
tree op0, op1;
bool negate = false;
*var = expr;
mpz_set_ui (offset, 0);
switch (TREE_CODE (expr))
{
case MINUS_EXPR:
negate = true;
/* Fallthru. */
case PLUS_EXPR:
case POINTER_PLUS_EXPR:
op0 = TREE_OPERAND (expr, 0);
op1 = TREE_OPERAND (expr, 1);
if (TREE_CODE (op1) != INTEGER_CST)
break;
*var = op0;
/* Always sign extend the offset. */
wi::to_mpz (wi::to_wide (op1), offset, SIGNED);
if (negate)
mpz_neg (offset, offset);
break;
case INTEGER_CST:
*var = build_int_cst_type (type, 0);
wi::to_mpz (wi::to_wide (expr), offset, TYPE_SIGN (type));
break;
default:
break;
}
}
/* From condition C0 CMP C1 derives information regarding the value range
of VAR, which is of TYPE. Results are stored in to BELOW and UP. */
static void
refine_value_range_using_guard (tree type, tree var,
tree c0, enum tree_code cmp, tree c1,
mpz_t below, mpz_t up)
{
tree varc0, varc1, ctype;
mpz_t offc0, offc1;
mpz_t mint, maxt, minc1, maxc1;
wide_int minv, maxv;
bool no_wrap = nowrap_type_p (type);
bool c0_ok, c1_ok;
signop sgn = TYPE_SIGN (type);
switch (cmp)
{
case LT_EXPR:
case LE_EXPR:
case GT_EXPR:
case GE_EXPR:
STRIP_SIGN_NOPS (c0);
STRIP_SIGN_NOPS (c1);
ctype = TREE_TYPE (c0);
if (!useless_type_conversion_p (ctype, type))
return;
break;
case EQ_EXPR:
/* We could derive quite precise information from EQ_EXPR, however,
such a guard is unlikely to appear, so we do not bother with
handling it. */
return;
case NE_EXPR:
/* NE_EXPR comparisons do not contain much of useful information,
except for cases of comparing with bounds. */
if (TREE_CODE (c1) != INTEGER_CST
|| !INTEGRAL_TYPE_P (type))
return;
/* Ensure that the condition speaks about an expression in the same
type as X and Y. */
ctype = TREE_TYPE (c0);
if (TYPE_PRECISION (ctype) != TYPE_PRECISION (type))
return;
c0 = fold_convert (type, c0);
c1 = fold_convert (type, c1);
if (operand_equal_p (var, c0, 0))
{
mpz_t valc1;
/* Case of comparing VAR with its below/up bounds. */
mpz_init (valc1);
wi::to_mpz (wi::to_wide (c1), valc1, TYPE_SIGN (type));
if (mpz_cmp (valc1, below) == 0)
cmp = GT_EXPR;
if (mpz_cmp (valc1, up) == 0)
cmp = LT_EXPR;
mpz_clear (valc1);
}
else
{
/* Case of comparing with the bounds of the type. */
wide_int min = wi::min_value (type);
wide_int max = wi::max_value (type);
if (wi::to_wide (c1) == min)
cmp = GT_EXPR;
if (wi::to_wide (c1) == max)
cmp = LT_EXPR;
}
/* Quick return if no useful information. */
if (cmp == NE_EXPR)
return;
break;
default:
return;
}
mpz_init (offc0);
mpz_init (offc1);
split_to_var_and_offset (expand_simple_operations (c0), &varc0, offc0);
split_to_var_and_offset (expand_simple_operations (c1), &varc1, offc1);
/* We are only interested in comparisons of expressions based on VAR. */
if (operand_equal_p (var, varc1, 0))
{
std::swap (varc0, varc1);
mpz_swap (offc0, offc1);
cmp = swap_tree_comparison (cmp);
}
else if (!operand_equal_p (var, varc0, 0))
{
mpz_clear (offc0);
mpz_clear (offc1);
return;
}
mpz_init (mint);
mpz_init (maxt);
get_type_static_bounds (type, mint, maxt);
mpz_init (minc1);
mpz_init (maxc1);
/* Setup range information for varc1. */
if (integer_zerop (varc1))
{
wi::to_mpz (0, minc1, TYPE_SIGN (type));
wi::to_mpz (0, maxc1, TYPE_SIGN (type));
}
else if (TREE_CODE (varc1) == SSA_NAME
&& INTEGRAL_TYPE_P (type)
&& get_range_info (varc1, &minv, &maxv) == VR_RANGE)
{
gcc_assert (wi::le_p (minv, maxv, sgn));
wi::to_mpz (minv, minc1, sgn);
wi::to_mpz (maxv, maxc1, sgn);
}
else
{
mpz_set (minc1, mint);
mpz_set (maxc1, maxt);
}
/* Compute valid range information for varc1 + offc1. Note nothing
useful can be derived if it overflows or underflows. Overflow or
underflow could happen when:
offc1 > 0 && varc1 + offc1 > MAX_VAL (type)
offc1 < 0 && varc1 + offc1 < MIN_VAL (type). */
mpz_add (minc1, minc1, offc1);
mpz_add (maxc1, maxc1, offc1);
c1_ok = (no_wrap
|| mpz_sgn (offc1) == 0
|| (mpz_sgn (offc1) < 0 && mpz_cmp (minc1, mint) >= 0)
|| (mpz_sgn (offc1) > 0 && mpz_cmp (maxc1, maxt) <= 0));
if (!c1_ok)
goto end;
if (mpz_cmp (minc1, mint) < 0)
mpz_set (minc1, mint);
if (mpz_cmp (maxc1, maxt) > 0)
mpz_set (maxc1, maxt);
if (cmp == LT_EXPR)
{
cmp = LE_EXPR;
mpz_sub_ui (maxc1, maxc1, 1);
}
if (cmp == GT_EXPR)
{
cmp = GE_EXPR;
mpz_add_ui (minc1, minc1, 1);
}
/* Compute range information for varc0. If there is no overflow,
the condition implied that
(varc0) cmp (varc1 + offc1 - offc0)
We can possibly improve the upper bound of varc0 if cmp is LE_EXPR,
or the below bound if cmp is GE_EXPR.
To prove there is no overflow/underflow, we need to check below
four cases:
1) cmp == LE_EXPR && offc0 > 0
(varc0 + offc0) doesn't overflow
&& (varc1 + offc1 - offc0) doesn't underflow
2) cmp == LE_EXPR && offc0 < 0
(varc0 + offc0) doesn't underflow
&& (varc1 + offc1 - offc0) doesn't overfloe
In this case, (varc0 + offc0) will never underflow if we can
prove (varc1 + offc1 - offc0) doesn't overflow.
3) cmp == GE_EXPR && offc0 < 0
(varc0 + offc0) doesn't underflow
&& (varc1 + offc1 - offc0) doesn't overflow
4) cmp == GE_EXPR && offc0 > 0
(varc0 + offc0) doesn't overflow
&& (varc1 + offc1 - offc0) doesn't underflow
In this case, (varc0 + offc0) will never overflow if we can
prove (varc1 + offc1 - offc0) doesn't underflow.
Note we only handle case 2 and 4 in below code. */
mpz_sub (minc1, minc1, offc0);
mpz_sub (maxc1, maxc1, offc0);
c0_ok = (no_wrap
|| mpz_sgn (offc0) == 0
|| (cmp == LE_EXPR
&& mpz_sgn (offc0) < 0 && mpz_cmp (maxc1, maxt) <= 0)
|| (cmp == GE_EXPR
&& mpz_sgn (offc0) > 0 && mpz_cmp (minc1, mint) >= 0));
if (!c0_ok)
goto end;
if (cmp == LE_EXPR)
{
if (mpz_cmp (up, maxc1) > 0)
mpz_set (up, maxc1);
}
else
{
if (mpz_cmp (below, minc1) < 0)
mpz_set (below, minc1);
}
end:
mpz_clear (mint);
mpz_clear (maxt);
mpz_clear (minc1);
mpz_clear (maxc1);
mpz_clear (offc0);
mpz_clear (offc1);
}
/* Stores estimate on the minimum/maximum value of the expression VAR + OFF
in TYPE to MIN and MAX. */
static void
determine_value_range (struct loop *loop, tree type, tree var, mpz_t off,
mpz_t min, mpz_t max)
{
int cnt = 0;
mpz_t minm, maxm;
basic_block bb;
wide_int minv, maxv;
enum value_range_kind rtype = VR_VARYING;
/* If the expression is a constant, we know its value exactly. */
if (integer_zerop (var))
{
mpz_set (min, off);
mpz_set (max, off);
return;
}
get_type_static_bounds (type, min, max);
/* See if we have some range info from VRP. */
if (TREE_CODE (var) == SSA_NAME && INTEGRAL_TYPE_P (type))
{
edge e = loop_preheader_edge (loop);
signop sgn = TYPE_SIGN (type);
gphi_iterator gsi;
/* Either for VAR itself... */
rtype = get_range_info (var, &minv, &maxv);
/* Or for PHI results in loop->header where VAR is used as
PHI argument from the loop preheader edge. */
for (gsi = gsi_start_phis (loop->header); !gsi_end_p (gsi); gsi_next (&gsi))
{
gphi *phi = gsi.phi ();
wide_int minc, maxc;
if (PHI_ARG_DEF_FROM_EDGE (phi, e) == var
&& (get_range_info (gimple_phi_result (phi), &minc, &maxc)
== VR_RANGE))
{
if (rtype != VR_RANGE)
{
rtype = VR_RANGE;
minv = minc;
maxv = maxc;
}
else
{
minv = wi::max (minv, minc, sgn);
maxv = wi::min (maxv, maxc, sgn);
/* If the PHI result range are inconsistent with
the VAR range, give up on looking at the PHI
results. This can happen if VR_UNDEFINED is
involved. */
if (wi::gt_p (minv, maxv, sgn))
{
rtype = get_range_info (var, &minv, &maxv);
break;
}
}
}
}
mpz_init (minm);
mpz_init (maxm);
if (rtype != VR_RANGE)
{
mpz_set (minm, min);
mpz_set (maxm, max);
}
else
{
gcc_assert (wi::le_p (minv, maxv, sgn));
wi::to_mpz (minv, minm, sgn);
wi::to_mpz (maxv, maxm, sgn);
}
/* Now walk the dominators of the loop header and use the entry
guards to refine the estimates. */
for (bb = loop->header;
bb != ENTRY_BLOCK_PTR_FOR_FN (cfun) && cnt < MAX_DOMINATORS_TO_WALK;
bb = get_immediate_dominator (CDI_DOMINATORS, bb))
{
edge e;
tree c0, c1;
gimple *cond;
enum tree_code cmp;
if (!single_pred_p (bb))
continue;
e = single_pred_edge (bb);
if (!(e->flags & (EDGE_TRUE_VALUE | EDGE_FALSE_VALUE)))
continue;
cond = last_stmt (e->src);
c0 = gimple_cond_lhs (cond);
cmp = gimple_cond_code (cond);
c1 = gimple_cond_rhs (cond);
if (e->flags & EDGE_FALSE_VALUE)
cmp = invert_tree_comparison (cmp, false);
refine_value_range_using_guard (type, var, c0, cmp, c1, minm, maxm);
++cnt;
}
mpz_add (minm, minm, off);
mpz_add (maxm, maxm, off);
/* If the computation may not wrap or off is zero, then this
is always fine. If off is negative and minv + off isn't
smaller than type's minimum, or off is positive and
maxv + off isn't bigger than type's maximum, use the more
precise range too. */
if (nowrap_type_p (type)
|| mpz_sgn (off) == 0
|| (mpz_sgn (off) < 0 && mpz_cmp (minm, min) >= 0)
|| (mpz_sgn (off) > 0 && mpz_cmp (maxm, max) <= 0))
{
mpz_set (min, minm);
mpz_set (max, maxm);
mpz_clear (minm);
mpz_clear (maxm);
return;
}
mpz_clear (minm);
mpz_clear (maxm);
}
/* If the computation may wrap, we know nothing about the value, except for
the range of the type. */
if (!nowrap_type_p (type))
return;
/* Since the addition of OFF does not wrap, if OFF is positive, then we may
add it to MIN, otherwise to MAX. */
if (mpz_sgn (off) < 0)
mpz_add (max, max, off);
else
mpz_add (min, min, off);
}
/* Stores the bounds on the difference of the values of the expressions
(var + X) and (var + Y), computed in TYPE, to BNDS. */
static void
bound_difference_of_offsetted_base (tree type, mpz_t x, mpz_t y,
bounds *bnds)
{
int rel = mpz_cmp (x, y);
bool may_wrap = !nowrap_type_p (type);
mpz_t m;
/* If X == Y, then the expressions are always equal.
If X > Y, there are the following possibilities:
a) neither of var + X and var + Y overflow or underflow, or both of
them do. Then their difference is X - Y.
b) var + X overflows, and var + Y does not. Then the values of the
expressions are var + X - M and var + Y, where M is the range of
the type, and their difference is X - Y - M.
c) var + Y underflows and var + X does not. Their difference again
is M - X + Y.
Therefore, if the arithmetics in type does not overflow, then the
bounds are (X - Y, X - Y), otherwise they are (X - Y - M, X - Y)
Similarly, if X < Y, the bounds are either (X - Y, X - Y) or
(X - Y, X - Y + M). */
if (rel == 0)
{
mpz_set_ui (bnds->below, 0);
mpz_set_ui (bnds->up, 0);
return;
}
mpz_init (m);
wi::to_mpz (wi::minus_one (TYPE_PRECISION (type)), m, UNSIGNED);
mpz_add_ui (m, m, 1);
mpz_sub (bnds->up, x, y);
mpz_set (bnds->below, bnds->up);
if (may_wrap)
{
if (rel > 0)
mpz_sub (bnds->below, bnds->below, m);
else
mpz_add (bnds->up, bnds->up, m);
}
mpz_clear (m);
}
/* From condition C0 CMP C1 derives information regarding the
difference of values of VARX + OFFX and VARY + OFFY, computed in TYPE,
and stores it to BNDS. */
static void
refine_bounds_using_guard (tree type, tree varx, mpz_t offx,
tree vary, mpz_t offy,
tree c0, enum tree_code cmp, tree c1,
bounds *bnds)
{
tree varc0, varc1, ctype;
mpz_t offc0, offc1, loffx, loffy, bnd;
bool lbound = false;
bool no_wrap = nowrap_type_p (type);
bool x_ok, y_ok;
switch (cmp)
{
case LT_EXPR:
case LE_EXPR:
case GT_EXPR:
case GE_EXPR:
STRIP_SIGN_NOPS (c0);
STRIP_SIGN_NOPS (c1);
ctype = TREE_TYPE (c0);
if (!useless_type_conversion_p (ctype, type))
return;
break;
case EQ_EXPR:
/* We could derive quite precise information from EQ_EXPR, however, such
a guard is unlikely to appear, so we do not bother with handling
it. */
return;
case NE_EXPR:
/* NE_EXPR comparisons do not contain much of useful information, except for
special case of comparing with the bounds of the type. */
if (TREE_CODE (c1) != INTEGER_CST
|| !INTEGRAL_TYPE_P (type))
return;
/* Ensure that the condition speaks about an expression in the same type
as X and Y. */
ctype = TREE_TYPE (c0);
if (TYPE_PRECISION (ctype) != TYPE_PRECISION (type))
return;
c0 = fold_convert (type, c0);
c1 = fold_convert (type, c1);
if (TYPE_MIN_VALUE (type)
&& operand_equal_p (c1, TYPE_MIN_VALUE (type), 0))
{
cmp = GT_EXPR;
break;
}
if (TYPE_MAX_VALUE (type)
&& operand_equal_p (c1, TYPE_MAX_VALUE (type), 0))
{
cmp = LT_EXPR;
break;
}
return;
default:
return;
}
mpz_init (offc0);
mpz_init (offc1);
split_to_var_and_offset (expand_simple_operations (c0), &varc0, offc0);
split_to_var_and_offset (expand_simple_operations (c1), &varc1, offc1);
/* We are only interested in comparisons of expressions based on VARX and
VARY. TODO -- we might also be able to derive some bounds from
expressions containing just one of the variables. */
if (operand_equal_p (varx, varc1, 0))
{
std::swap (varc0, varc1);
mpz_swap (offc0, offc1);
cmp = swap_tree_comparison (cmp);
}
if (!operand_equal_p (varx, varc0, 0)
|| !operand_equal_p (vary, varc1, 0))
goto end;
mpz_init_set (loffx, offx);
mpz_init_set (loffy, offy);
if (cmp == GT_EXPR || cmp == GE_EXPR)
{
std::swap (varx, vary);
mpz_swap (offc0, offc1);
mpz_swap (loffx, loffy);
cmp = swap_tree_comparison (cmp);
lbound = true;
}
/* If there is no overflow, the condition implies that
(VARX + OFFX) cmp (VARY + OFFY) + (OFFX - OFFY + OFFC1 - OFFC0).
The overflows and underflows may complicate things a bit; each
overflow decreases the appropriate offset by M, and underflow
increases it by M. The above inequality would not necessarily be
true if
-- VARX + OFFX underflows and VARX + OFFC0 does not, or
VARX + OFFC0 overflows, but VARX + OFFX does not.
This may only happen if OFFX < OFFC0.
-- VARY + OFFY overflows and VARY + OFFC1 does not, or
VARY + OFFC1 underflows and VARY + OFFY does not.
This may only happen if OFFY > OFFC1. */
if (no_wrap)
{
x_ok = true;
y_ok = true;
}
else
{
x_ok = (integer_zerop (varx)
|| mpz_cmp (loffx, offc0) >= 0);
y_ok = (integer_zerop (vary)
|| mpz_cmp (loffy, offc1) <= 0);
}
if (x_ok && y_ok)
{
mpz_init (bnd);
mpz_sub (bnd, loffx, loffy);
mpz_add (bnd, bnd, offc1);
mpz_sub (bnd, bnd, offc0);
if (cmp == LT_EXPR)
mpz_sub_ui (bnd, bnd, 1);
if (lbound)
{
mpz_neg (bnd, bnd);
if (mpz_cmp (bnds->below, bnd) < 0)
mpz_set (bnds->below, bnd);
}
else
{
if (mpz_cmp (bnd, bnds->up) < 0)
mpz_set (bnds->up, bnd);
}
mpz_clear (bnd);
}
mpz_clear (loffx);
mpz_clear (loffy);
end:
mpz_clear (offc0);
mpz_clear (offc1);
}
/* Stores the bounds on the value of the expression X - Y in LOOP to BNDS.
The subtraction is considered to be performed in arbitrary precision,
without overflows.
We do not attempt to be too clever regarding the value ranges of X and
Y; most of the time, they are just integers or ssa names offsetted by
integer. However, we try to use the information contained in the
comparisons before the loop (usually created by loop header copying). */
static void
bound_difference (struct loop *loop, tree x, tree y, bounds *bnds)
{
tree type = TREE_TYPE (x);
tree varx, vary;
mpz_t offx, offy;
mpz_t minx, maxx, miny, maxy;
int cnt = 0;
edge e;
basic_block bb;
tree c0, c1;
gimple *cond;
enum tree_code cmp;
/* Get rid of unnecessary casts, but preserve the value of
the expressions. */
STRIP_SIGN_NOPS (x);
STRIP_SIGN_NOPS (y);
mpz_init (bnds->below);
mpz_init (bnds->up);
mpz_init (offx);
mpz_init (offy);
split_to_var_and_offset (x, &varx, offx);
split_to_var_and_offset (y, &vary, offy);
if (!integer_zerop (varx)
&& operand_equal_p (varx, vary, 0))
{
/* Special case VARX == VARY -- we just need to compare the
offsets. The matters are a bit more complicated in the
case addition of offsets may wrap. */
bound_difference_of_offsetted_base (type, offx, offy, bnds);
}
else
{
/* Otherwise, use the value ranges to determine the initial
estimates on below and up. */
mpz_init (minx);
mpz_init (maxx);
mpz_init (miny);
mpz_init (maxy);
determine_value_range (loop, type, varx, offx, minx, maxx);
determine_value_range (loop, type, vary, offy, miny, maxy);
mpz_sub (bnds->below, minx, maxy);
mpz_sub (bnds->up, maxx, miny);
mpz_clear (minx);
mpz_clear (maxx);
mpz_clear (miny);
mpz_clear (maxy);
}
/* If both X and Y are constants, we cannot get any more precise. */
if (integer_zerop (varx) && integer_zerop (vary))
goto end;
/* Now walk the dominators of the loop header and use the entry
guards to refine the estimates. */
for (bb = loop->header;
bb != ENTRY_BLOCK_PTR_FOR_FN (cfun) && cnt < MAX_DOMINATORS_TO_WALK;
bb = get_immediate_dominator (CDI_DOMINATORS, bb))
{
if (!single_pred_p (bb))
continue;
e = single_pred_edge (bb);
if (!(e->flags & (EDGE_TRUE_VALUE | EDGE_FALSE_VALUE)))
continue;
cond = last_stmt (e->src);
c0 = gimple_cond_lhs (cond);
cmp = gimple_cond_code (cond);
c1 = gimple_cond_rhs (cond);
if (e->flags & EDGE_FALSE_VALUE)
cmp = invert_tree_comparison (cmp, false);
refine_bounds_using_guard (type, varx, offx, vary, offy,
c0, cmp, c1, bnds);
++cnt;
}
end:
mpz_clear (offx);
mpz_clear (offy);
}
/* Update the bounds in BNDS that restrict the value of X to the bounds
that restrict the value of X + DELTA. X can be obtained as a
difference of two values in TYPE. */
static void
bounds_add (bounds *bnds, const widest_int &delta, tree type)
{
mpz_t mdelta, max;
mpz_init (mdelta);
wi::to_mpz (delta, mdelta, SIGNED);
mpz_init (max);
wi::to_mpz (wi::minus_one (TYPE_PRECISION (type)), max, UNSIGNED);
mpz_add (bnds->up, bnds->up, mdelta);
mpz_add (bnds->below, bnds->below, mdelta);
if (mpz_cmp (bnds->up, max) > 0)
mpz_set (bnds->up, max);
mpz_neg (max, max);
if (mpz_cmp (bnds->below, max) < 0)
mpz_set (bnds->below, max);
mpz_clear (mdelta);
mpz_clear (max);
}
/* Update the bounds in BNDS that restrict the value of X to the bounds
that restrict the value of -X. */
static void
bounds_negate (bounds *bnds)
{
mpz_t tmp;
mpz_init_set (tmp, bnds->up);
mpz_neg (bnds->up, bnds->below);
mpz_neg (bnds->below, tmp);
mpz_clear (tmp);
}
/* Returns inverse of X modulo 2^s, where MASK = 2^s-1. */
static tree
inverse (tree x, tree mask)
{
tree type = TREE_TYPE (x);
tree rslt;
unsigned ctr = tree_floor_log2 (mask);
if (TYPE_PRECISION (type) <= HOST_BITS_PER_WIDE_INT)
{
unsigned HOST_WIDE_INT ix;
unsigned HOST_WIDE_INT imask;
unsigned HOST_WIDE_INT irslt = 1;
gcc_assert (cst_and_fits_in_hwi (x));
gcc_assert (cst_and_fits_in_hwi (mask));
ix = int_cst_value (x);
imask = int_cst_value (mask);
for (; ctr; ctr--)
{
irslt *= ix;
ix *= ix;
}
irslt &= imask;
rslt = build_int_cst_type (type, irslt);
}
else
{
rslt = build_int_cst (type, 1);
for (; ctr; ctr--)
{
rslt = int_const_binop (MULT_EXPR, rslt, x);
x = int_const_binop (MULT_EXPR, x, x);
}
rslt = int_const_binop (BIT_AND_EXPR, rslt, mask);
}
return rslt;
}
/* Derives the upper bound BND on the number of executions of loop with exit
condition S * i <> C. If NO_OVERFLOW is true, then the control variable of
the loop does not overflow. EXIT_MUST_BE_TAKEN is true if we are guaranteed
that the loop ends through this exit, i.e., the induction variable ever
reaches the value of C.
The value C is equal to final - base, where final and base are the final and
initial value of the actual induction variable in the analysed loop. BNDS
bounds the value of this difference when computed in signed type with
unbounded range, while the computation of C is performed in an unsigned
type with the range matching the range of the type of the induction variable.
In particular, BNDS.up contains an upper bound on C in the following cases:
-- if the iv must reach its final value without overflow, i.e., if
NO_OVERFLOW && EXIT_MUST_BE_TAKEN is true, or
-- if final >= base, which we know to hold when BNDS.below >= 0. */
static void
number_of_iterations_ne_max (mpz_t bnd, bool no_overflow, tree c, tree s,
bounds *bnds, bool exit_must_be_taken)
{
widest_int max;
mpz_t d;
tree type = TREE_TYPE (c);
bool bnds_u_valid = ((no_overflow && exit_must_be_taken)
|| mpz_sgn (bnds->below) >= 0);
if (integer_onep (s)
|| (TREE_CODE (c) == INTEGER_CST
&& TREE_CODE (s) == INTEGER_CST
&& wi::mod_trunc (wi::to_wide (c), wi::to_wide (s),
TYPE_SIGN (type)) == 0)
|| (TYPE_OVERFLOW_UNDEFINED (type)
&& multiple_of_p (type, c, s)))
{
/* If C is an exact multiple of S, then its value will be reached before
the induction variable overflows (unless the loop is exited in some
other way before). Note that the actual induction variable in the
loop (which ranges from base to final instead of from 0 to C) may
overflow, in which case BNDS.up will not be giving a correct upper
bound on C; thus, BNDS_U_VALID had to be computed in advance. */
no_overflow = true;
exit_must_be_taken = true;
}
/* If the induction variable can overflow, the number of iterations is at
most the period of the control variable (or infinite, but in that case
the whole # of iterations analysis will fail). */
if (!no_overflow)
{
max = wi::mask <widest_int> (TYPE_PRECISION (type)
- wi::ctz (wi::to_wide (s)), false);
wi::to_mpz (max, bnd, UNSIGNED);
return;
}
/* Now we know that the induction variable does not overflow, so the loop
iterates at most (range of type / S) times. */
wi::to_mpz (wi::minus_one (TYPE_PRECISION (type)), bnd, UNSIGNED);
/* If the induction variable is guaranteed to reach the value of C before
overflow, ... */
if (exit_must_be_taken)
{
/* ... then we can strengthen this to C / S, and possibly we can use
the upper bound on C given by BNDS. */
if (TREE_CODE (c) == INTEGER_CST)
wi::to_mpz (wi::to_wide (c), bnd, UNSIGNED);
else if (bnds_u_valid)
mpz_set (bnd, bnds->up);
}
mpz_init (d);
wi::to_mpz (wi::to_wide (s), d, UNSIGNED);
mpz_fdiv_q (bnd, bnd, d);
mpz_clear (d);
}
/* Determines number of iterations of loop whose ending condition
is IV <> FINAL. TYPE is the type of the iv. The number of
iterations is stored to NITER. EXIT_MUST_BE_TAKEN is true if
we know that the exit must be taken eventually, i.e., that the IV
ever reaches the value FINAL (we derived this earlier, and possibly set
NITER->assumptions to make sure this is the case). BNDS contains the
bounds on the difference FINAL - IV->base. */
static bool
number_of_iterations_ne (struct loop *loop, tree type, affine_iv *iv,
tree final, struct tree_niter_desc *niter,
bool exit_must_be_taken, bounds *bnds)
{
tree niter_type = unsigned_type_for (type);
tree s, c, d, bits, assumption, tmp, bound;
mpz_t max;
niter->control = *iv;
niter->bound = final;
niter->cmp = NE_EXPR;
/* Rearrange the terms so that we get inequality S * i <> C, with S
positive. Also cast everything to the unsigned type. If IV does
not overflow, BNDS bounds the value of C. Also, this is the
case if the computation |FINAL - IV->base| does not overflow, i.e.,
if BNDS->below in the result is nonnegative. */
if (tree_int_cst_sign_bit (iv->step))
{
s = fold_convert (niter_type,
fold_build1 (NEGATE_EXPR, type, iv->step));
c = fold_build2 (MINUS_EXPR, niter_type,
fold_convert (niter_type, iv->base),
fold_convert (niter_type, final));
bounds_negate (bnds);
}
else
{
s = fold_convert (niter_type, iv->step);
c = fold_build2 (MINUS_EXPR, niter_type,
fold_convert (niter_type, final),
fold_convert (niter_type, iv->base));
}
mpz_init (max);
number_of_iterations_ne_max (max, iv->no_overflow, c, s, bnds,
exit_must_be_taken);
niter->max = widest_int::from (wi::from_mpz (niter_type, max, false),
TYPE_SIGN (niter_type));
mpz_clear (max);
/* Compute no-overflow information for the control iv. This can be
proven when below two conditions are satisfied:
1) IV evaluates toward FINAL at beginning, i.e:
base <= FINAL ; step > 0
base >= FINAL ; step < 0
2) |FINAL - base| is an exact multiple of step.
Unfortunately, it's hard to prove above conditions after pass loop-ch
because loop with exit condition (IV != FINAL) usually will be guarded
by initial-condition (IV.base - IV.step != FINAL). In this case, we
can alternatively try to prove below conditions:
1') IV evaluates toward FINAL at beginning, i.e:
new_base = base - step < FINAL ; step > 0
&& base - step doesn't underflow
new_base = base - step > FINAL ; step < 0
&& base - step doesn't overflow
2') |FINAL - new_base| is an exact multiple of step.
Please refer to PR34114 as an example of loop-ch's impact, also refer
to PR72817 as an example why condition 2') is necessary.
Note, for NE_EXPR, base equals to FINAL is a special case, in
which the loop exits immediately, and the iv does not overflow. */
if (!niter->control.no_overflow
&& (integer_onep (s) || multiple_of_p (type, c, s)))
{
tree t, cond, new_c, relaxed_cond = boolean_false_node;
if (tree_int_cst_sign_bit (iv->step))
{
cond = fold_build2 (GE_EXPR, boolean_type_node, iv->base, final);
if (TREE_CODE (type) == INTEGER_TYPE)
{
/* Only when base - step doesn't overflow. */
t = TYPE_MAX_VALUE (type);
t = fold_build2 (PLUS_EXPR, type, t, iv->step);
t = fold_build2 (GE_EXPR, boolean_type_node, t, iv->base);
if (integer_nonzerop (t))
{
t = fold_build2 (MINUS_EXPR, type, iv->base, iv->step);
new_c = fold_build2 (MINUS_EXPR, niter_type,
fold_convert (niter_type, t),
fold_convert (niter_type, final));
if (multiple_of_p (type, new_c, s))
relaxed_cond = fold_build2 (GT_EXPR, boolean_type_node,
t, final);
}
}
}
else
{
cond = fold_build2 (LE_EXPR, boolean_type_node, iv->base, final);
if (TREE_CODE (type) == INTEGER_TYPE)
{
/* Only when base - step doesn't underflow. */
t = TYPE_MIN_VALUE (type);
t = fold_build2 (PLUS_EXPR, type, t, iv->step);
t = fold_build2 (LE_EXPR, boolean_type_node, t, iv->base);
if (integer_nonzerop (t))
{
t = fold_build2 (MINUS_EXPR, type, iv->base, iv->step);
new_c = fold_build2 (MINUS_EXPR, niter_type,
fold_convert (niter_type, final),
fold_convert (niter_type, t));
if (multiple_of_p (type, new_c, s))
relaxed_cond = fold_build2 (LT_EXPR, boolean_type_node,
t, final);
}
}
}
t = simplify_using_initial_conditions (loop, cond);
if (!t || !integer_onep (t))
t = simplify_using_initial_conditions (loop, relaxed_cond);
if (t && integer_onep (t))
niter->control.no_overflow = true;
}
/* First the trivial cases -- when the step is 1. */
if (integer_onep (s))
{
niter->niter = c;
return true;
}
if (niter->control.no_overflow && multiple_of_p (type, c, s))
{
niter->niter = fold_build2 (FLOOR_DIV_EXPR, niter_type, c, s);
return true;
}
/* Let nsd (step, size of mode) = d. If d does not divide c, the loop
is infinite. Otherwise, the number of iterations is
(inverse(s/d) * (c/d)) mod (size of mode/d). */
bits = num_ending_zeros (s);
bound = build_low_bits_mask (niter_type,
(TYPE_PRECISION (niter_type)
- tree_to_uhwi (bits)));
d = fold_binary_to_constant (LSHIFT_EXPR, niter_type,
build_int_cst (niter_type, 1), bits);
s = fold_binary_to_constant (RSHIFT_EXPR, niter_type, s, bits);
if (!exit_must_be_taken)
{
/* If we cannot assume that the exit is taken eventually, record the
assumptions for divisibility of c. */
assumption = fold_build2 (FLOOR_MOD_EXPR, niter_type, c, d);
assumption = fold_build2 (EQ_EXPR, boolean_type_node,
assumption, build_int_cst (niter_type, 0));
if (!integer_nonzerop (assumption))
niter->assumptions = fold_build2 (TRUTH_AND_EXPR, boolean_type_node,
niter->assumptions, assumption);
}
c = fold_build2 (EXACT_DIV_EXPR, niter_type, c, d);
tmp = fold_build2 (MULT_EXPR, niter_type, c, inverse (s, bound));
niter->niter = fold_build2 (BIT_AND_EXPR, niter_type, tmp, bound);
return true;
}
/* Checks whether we can determine the final value of the control variable
of the loop with ending condition IV0 < IV1 (computed in TYPE).
DELTA is the difference IV1->base - IV0->base, STEP is the absolute value
of the step. The assumptions necessary to ensure that the computation
of the final value does not overflow are recorded in NITER. If we
find the final value, we adjust DELTA and return TRUE. Otherwise
we return false. BNDS bounds the value of IV1->base - IV0->base,
and will be updated by the same amount as DELTA. EXIT_MUST_BE_TAKEN is
true if we know that the exit must be taken eventually. */
static bool
number_of_iterations_lt_to_ne (tree type, affine_iv *iv0, affine_iv *iv1,
struct tree_niter_desc *niter,
tree *delta, tree step,
bool exit_must_be_taken, bounds *bnds)
{
tree niter_type = TREE_TYPE (step);
tree mod = fold_build2 (FLOOR_MOD_EXPR, niter_type, *delta, step);
tree tmod;
mpz_t mmod;
tree assumption = boolean_true_node, bound, noloop;
bool ret = false, fv_comp_no_overflow;
tree type1 = type;
if (POINTER_TYPE_P (type))
type1 = sizetype;
if (TREE_CODE (mod) != INTEGER_CST)
return false;
if (integer_nonzerop (mod))
mod = fold_build2 (MINUS_EXPR, niter_type, step, mod);
tmod = fold_convert (type1, mod);
mpz_init (mmod);
wi::to_mpz (wi::to_wide (mod), mmod, UNSIGNED);
mpz_neg (mmod, mmod);
/* If the induction variable does not overflow and the exit is taken,
then the computation of the final value does not overflow. This is
also obviously the case if the new final value is equal to the
current one. Finally, we postulate this for pointer type variables,
as the code cannot rely on the object to that the pointer points being
placed at the end of the address space (and more pragmatically,
TYPE_{MIN,MAX}_VALUE is not defined for pointers). */
if (integer_zerop (mod) || POINTER_TYPE_P (type))
fv_comp_no_overflow = true;
else if (!exit_must_be_taken)
fv_comp_no_overflow = false;
else
fv_comp_no_overflow =
(iv0->no_overflow && integer_nonzerop (iv0->step))
|| (iv1->no_overflow && integer_nonzerop (iv1->step));
if (integer_nonzerop (iv0->step))
{
/* The final value of the iv is iv1->base + MOD, assuming that this
computation does not overflow, and that
iv0->base <= iv1->base + MOD. */
if (!fv_comp_no_overflow)
{
bound = fold_build2 (MINUS_EXPR, type1,
TYPE_MAX_VALUE (type1), tmod);
assumption = fold_build2 (LE_EXPR, boolean_type_node,
iv1->base, bound);
if (integer_zerop (assumption))
goto end;
}
if (mpz_cmp (mmod, bnds->below) < 0)
noloop = boolean_false_node;
else if (POINTER_TYPE_P (type))
noloop = fold_build2 (GT_EXPR, boolean_type_node,
iv0->base,
fold_build_pointer_plus (iv1->base, tmod));
else
noloop = fold_build2 (GT_EXPR, boolean_type_node,
iv0->base,
fold_build2 (PLUS_EXPR, type1,
iv1->base, tmod));
}
else
{
/* The final value of the iv is iv0->base - MOD, assuming that this
computation does not overflow, and that
iv0->base - MOD <= iv1->base. */
if (!fv_comp_no_overflow)
{
bound = fold_build2 (PLUS_EXPR, type1,
TYPE_MIN_VALUE (type1), tmod);
assumption = fold_build2 (GE_EXPR, boolean_type_node,
iv0->base, bound);
if (integer_zerop (assumption))
goto end;
}
if (mpz_cmp (mmod, bnds->below) < 0)
noloop = boolean_false_node;
else if (POINTER_TYPE_P (type))
noloop = fold_build2 (GT_EXPR, boolean_type_node,
fold_build_pointer_plus (iv0->base,
fold_build1 (NEGATE_EXPR,
type1, tmod)),
iv1->base);
else
noloop = fold_build2 (GT_EXPR, boolean_type_node,
fold_build2 (MINUS_EXPR, type1,
iv0->base, tmod),
iv1->base);
}
if (!integer_nonzerop (assumption))
niter->assumptions = fold_build2 (TRUTH_AND_EXPR, boolean_type_node,
niter->assumptions,
assumption);
if (!integer_zerop (noloop))
niter->may_be_zero = fold_build2 (TRUTH_OR_EXPR, boolean_type_node,
niter->may_be_zero,
noloop);
bounds_add (bnds, wi::to_widest (mod), type);
*delta = fold_build2 (PLUS_EXPR, niter_type, *delta, mod);
ret = true;
end:
mpz_clear (mmod);
return ret;
}
/* Add assertions to NITER that ensure that the control variable of the loop
with ending condition IV0 < IV1 does not overflow. Types of IV0 and IV1
are TYPE. Returns false if we can prove that there is an overflow, true
otherwise. STEP is the absolute value of the step. */
static bool
assert_no_overflow_lt (tree type, affine_iv *iv0, affine_iv *iv1,
struct tree_niter_desc *niter, tree step)
{
tree bound, d, assumption, diff;
tree niter_type = TREE_TYPE (step);
if (integer_nonzerop (iv0->step))
{
/* for (i = iv0->base; i < iv1->base; i += iv0->step) */
if (iv0->no_overflow)
return true;
/* If iv0->base is a constant, we can determine the last value before
overflow precisely; otherwise we conservatively assume
MAX - STEP + 1. */
if (TREE_CODE (iv0->base) == INTEGER_CST)
{
d = fold_build2 (MINUS_EXPR, niter_type,
fold_convert (niter_type, TYPE_MAX_VALUE (type)),
fold_convert (niter_type, iv0->base));
diff = fold_build2 (FLOOR_MOD_EXPR, niter_type, d, step);
}
else
diff = fold_build2 (MINUS_EXPR, niter_type, step,
build_int_cst (niter_type, 1));
bound = fold_build2 (MINUS_EXPR, type,
TYPE_MAX_VALUE (type), fold_convert (type, diff));
assumption = fold_build2 (LE_EXPR, boolean_type_node,
iv1->base, bound);
}
else
{
/* for (i = iv1->base; i > iv0->base; i += iv1->step) */
if (iv1->no_overflow)
return true;
if (TREE_CODE (iv1->base) == INTEGER_CST)
{
d = fold_build2 (MINUS_EXPR, niter_type,
fold_convert (niter_type, iv1->base),
fold_convert (niter_type, TYPE_MIN_VALUE (type)));
diff = fold_build2 (FLOOR_MOD_EXPR, niter_type, d, step);
}
else
diff = fold_build2 (MINUS_EXPR, niter_type, step,
build_int_cst (niter_type, 1));
bound = fold_build2 (PLUS_EXPR, type,
TYPE_MIN_VALUE (type), fold_convert (type, diff));
assumption = fold_build2 (GE_EXPR, boolean_type_node,
iv0->base, bound);
}
if (integer_zerop (assumption))
return false;
if (!integer_nonzerop (assumption))
niter->assumptions = fold_build2 (TRUTH_AND_EXPR, boolean_type_node,
niter->assumptions, assumption);
iv0->no_overflow = true;
iv1->no_overflow = true;
return true;
}
/* Add an assumption to NITER that a loop whose ending condition
is IV0 < IV1 rolls. TYPE is the type of the control iv. BNDS
bounds the value of IV1->base - IV0->base. */
static void
assert_loop_rolls_lt (tree type, affine_iv *iv0, affine_iv *iv1,
struct tree_niter_desc *niter, bounds *bnds)
{
tree assumption = boolean_true_node, bound, diff;
tree mbz, mbzl, mbzr, type1;
bool rolls_p, no_overflow_p;
widest_int dstep;
mpz_t mstep, max;
/* We are going to compute the number of iterations as
(iv1->base - iv0->base + step - 1) / step, computed in the unsigned
variant of TYPE. This formula only works if
-step + 1 <= (iv1->base - iv0->base) <= MAX - step + 1
(where MAX is the maximum value of the unsigned variant of TYPE, and
the computations in this formula are performed in full precision,
i.e., without overflows).
Usually, for loops with exit condition iv0->base + step * i < iv1->base,
we have a condition of the form iv0->base - step < iv1->base before the loop,
and for loops iv0->base < iv1->base - step * i the condition
iv0->base < iv1->base + step, due to loop header copying, which enable us
to prove the lower bound.
The upper bound is more complicated. Unless the expressions for initial
and final value themselves contain enough information, we usually cannot
derive it from the context. */
/* First check whether the answer does not follow from the bounds we gathered
before. */
if (integer_nonzerop (iv0->step))
dstep = wi::to_widest (iv0->step);
else
{
dstep = wi::sext (wi::to_widest (iv1->step), TYPE_PRECISION (type));
dstep = -dstep;
}
mpz_init (mstep);
wi::to_mpz (dstep, mstep, UNSIGNED);
mpz_neg (mstep, mstep);
mpz_add_ui (mstep, mstep, 1);
rolls_p = mpz_cmp (mstep, bnds->below) <= 0;
mpz_init (max);
wi::to_mpz (wi::minus_one (TYPE_PRECISION (type)), max, UNSIGNED);
mpz_add (max, max, mstep);
no_overflow_p = (mpz_cmp (bnds->up, max) <= 0
/* For pointers, only values lying inside a single object
can be compared or manipulated by pointer arithmetics.
Gcc in general does not allow or handle objects larger
than half of the address space, hence the upper bound
is satisfied for pointers. */
|| POINTER_TYPE_P (type));
mpz_clear (mstep);
mpz_clear (max);
if (rolls_p && no_overflow_p)
return;
type1 = type;
if (POINTER_TYPE_P (type))
type1 = sizetype;
/* Now the hard part; we must formulate the assumption(s) as expressions, and
we must be careful not to introduce overflow. */
if (integer_nonzerop (iv0->step))
{
diff = fold_build2 (MINUS_EXPR, type1,
iv0->step, build_int_cst (type1, 1));
/* We need to know that iv0->base >= MIN + iv0->step - 1. Since
0 address never belongs to any object, we can assume this for
pointers. */
if (!POINTER_TYPE_P (type))
{
bound = fold_build2 (PLUS_EXPR, type1,
TYPE_MIN_VALUE (type), diff);
assumption = fold_build2 (GE_EXPR, boolean_type_node,
iv0->base, bound);
}
/* And then we can compute iv0->base - diff, and compare it with
iv1->base. */
mbzl = fold_build2 (MINUS_EXPR, type1,
fold_convert (type1, iv0->base), diff);
mbzr = fold_convert (type1, iv1->base);
}
else
{
diff = fold_build2 (PLUS_EXPR, type1,
iv1->step, build_int_cst (type1, 1));
if (!POINTER_TYPE_P (type))
{
bound = fold_build2 (PLUS_EXPR, type1,
TYPE_MAX_VALUE (type), diff);
assumption = fold_build2 (LE_EXPR, boolean_type_node,
iv1->base, bound);
}
mbzl = fold_convert (type1, iv0->base);
mbzr = fold_build2 (MINUS_EXPR, type1,
fold_convert (type1, iv1->base), diff);
}
if (!integer_nonzerop (assumption))
niter->assumptions = fold_build2 (TRUTH_AND_EXPR, boolean_type_node,
niter->assumptions, assumption);
if (!rolls_p)
{
mbz = fold_build2 (GT_EXPR, boolean_type_node, mbzl, mbzr);
niter->may_be_zero = fold_build2 (TRUTH_OR_EXPR, boolean_type_node,
niter->may_be_zero, mbz);
}
}
/* Determines number of iterations of loop whose ending condition
is IV0 < IV1. TYPE is the type of the iv. The number of
iterations is stored to NITER. BNDS bounds the difference
IV1->base - IV0->base. EXIT_MUST_BE_TAKEN is true if we know
that the exit must be taken eventually. */
static bool
number_of_iterations_lt (struct loop *loop, tree type, affine_iv *iv0,
affine_iv *iv1, struct tree_niter_desc *niter,
bool exit_must_be_taken, bounds *bnds)
{
tree niter_type = unsigned_type_for (type);
tree delta, step, s;
mpz_t mstep, tmp;
if (integer_nonzerop (iv0->step))
{
niter->control = *iv0;
niter->cmp = LT_EXPR;
niter->bound = iv1->base;
}
else
{
niter->control = *iv1;
niter->cmp = GT_EXPR;
niter->bound = iv0->base;
}
delta = fold_build2 (MINUS_EXPR, niter_type,
fold_convert (niter_type, iv1->base),
fold_convert (niter_type, iv0->base));
/* First handle the special case that the step is +-1. */
if ((integer_onep (iv0->step) && integer_zerop (iv1->step))
|| (integer_all_onesp (iv1->step) && integer_zerop (iv0->step)))
{
/* for (i = iv0->base; i < iv1->base; i++)
or
for (i = iv1->base; i > iv0->base; i--).
In both cases # of iterations is iv1->base - iv0->base, assuming that
iv1->base >= iv0->base.
First try to derive a lower bound on the value of
iv1->base - iv0->base, computed in full precision. If the difference
is nonnegative, we are done, otherwise we must record the
condition. */
if (mpz_sgn (bnds->below) < 0)
niter->may_be_zero = fold_build2 (LT_EXPR, boolean_type_node,
iv1->base, iv0->base);
niter->niter = delta;
niter->max = widest_int::from (wi::from_mpz (niter_type, bnds->up, false),
TYPE_SIGN (niter_type));
niter->control.no_overflow = true;
return true;
}
if (integer_nonzerop (iv0->step))
step = fold_convert (niter_type, iv0->step);
else
step = fold_convert (niter_type,
fold_build1 (NEGATE_EXPR, type, iv1->step));
/* If we can determine the final value of the control iv exactly, we can
transform the condition to != comparison. In particular, this will be
the case if DELTA is constant. */
if (number_of_iterations_lt_to_ne (type, iv0, iv1, niter, &delta, step,
exit_must_be_taken, bnds))
{
affine_iv zps;
zps.base = build_int_cst (niter_type, 0);
zps.step = step;
/* number_of_iterations_lt_to_ne will add assumptions that ensure that
zps does not overflow. */
zps.no_overflow = true;
return number_of_iterations_ne (loop, type, &zps,
delta, niter, true, bnds);
}
/* Make sure that the control iv does not overflow. */
if (!assert_no_overflow_lt (type, iv0, iv1, niter, step))
return false;
/* We determine the number of iterations as (delta + step - 1) / step. For
this to work, we must know that iv1->base >= iv0->base - step + 1,
otherwise the loop does not roll. */
assert_loop_rolls_lt (type, iv0, iv1, niter, bnds);
s = fold_build2 (MINUS_EXPR, niter_type,
step, build_int_cst (niter_type, 1));
delta = fold_build2 (PLUS_EXPR, niter_type, delta, s);
niter->niter = fold_build2 (FLOOR_DIV_EXPR, niter_type, delta, step);
mpz_init (mstep);
mpz_init (tmp);
wi::to_mpz (wi::to_wide (step), mstep, UNSIGNED);
mpz_add (tmp, bnds->up, mstep);
mpz_sub_ui (tmp, tmp, 1);
mpz_fdiv_q (tmp, tmp, mstep);
niter->max = widest_int::from (wi::from_mpz (niter_type, tmp, false),
TYPE_SIGN (niter_type));
mpz_clear (mstep);
mpz_clear (tmp);
return true;
}
/* Determines number of iterations of loop whose ending condition
is IV0 <= IV1. TYPE is the type of the iv. The number of
iterations is stored to NITER. EXIT_MUST_BE_TAKEN is true if
we know that this condition must eventually become false (we derived this
earlier, and possibly set NITER->assumptions to make sure this
is the case). BNDS bounds the difference IV1->base - IV0->base. */
static bool
number_of_iterations_le (struct loop *loop, tree type, affine_iv *iv0,
affine_iv *iv1, struct tree_niter_desc *niter,
bool exit_must_be_taken, bounds *bnds)
{
tree assumption;
tree type1 = type;
if (POINTER_TYPE_P (type))
type1 = sizetype;
/* Say that IV0 is the control variable. Then IV0 <= IV1 iff
IV0 < IV1 + 1, assuming that IV1 is not equal to the greatest
value of the type. This we must know anyway, since if it is
equal to this value, the loop rolls forever. We do not check
this condition for pointer type ivs, as the code cannot rely on
the object to that the pointer points being placed at the end of
the address space (and more pragmatically, TYPE_{MIN,MAX}_VALUE is
not defined for pointers). */
if (!exit_must_be_taken && !POINTER_TYPE_P (type))
{
if (integer_nonzerop (iv0->step))
assumption = fold_build2 (NE_EXPR, boolean_type_node,
iv1->base, TYPE_MAX_VALUE (type));
else
assumption = fold_build2 (NE_EXPR, boolean_type_node,
iv0->base, TYPE_MIN_VALUE (type));
if (integer_zerop (assumption))
return false;
if (!integer_nonzerop (assumption))
niter->assumptions = fold_build2 (TRUTH_AND_EXPR, boolean_type_node,
niter->assumptions, assumption);
}
if (integer_nonzerop (iv0->step))
{
if (POINTER_TYPE_P (type))
iv1->base = fold_build_pointer_plus_hwi (iv1->base, 1);
else
iv1->base = fold_build2 (PLUS_EXPR, type1, iv1->base,
build_int_cst (type1, 1));
}
else if (POINTER_TYPE_P (type))
iv0->base = fold_build_pointer_plus_hwi (iv0->base, -1);
else
iv0->base = fold_build2 (MINUS_EXPR, type1,
iv0->base, build_int_cst (type1, 1));
bounds_add (bnds, 1, type1);
return number_of_iterations_lt (loop, type, iv0, iv1, niter, exit_must_be_taken,
bnds);
}
/* Dumps description of affine induction variable IV to FILE. */
static void
dump_affine_iv (FILE *file, affine_iv *iv)
{
if (!integer_zerop (iv->step))
fprintf (file, "[");
print_generic_expr (dump_file, iv->base, TDF_SLIM);
if (!integer_zerop (iv->step))
{
fprintf (file, ", + , ");
print_generic_expr (dump_file, iv->step, TDF_SLIM);
fprintf (file, "]%s", iv->no_overflow ? "(no_overflow)" : "");
}
}
/* Determine the number of iterations according to condition (for staying
inside loop) which compares two induction variables using comparison
operator CODE. The induction variable on left side of the comparison
is IV0, the right-hand side is IV1. Both induction variables must have
type TYPE, which must be an integer or pointer type. The steps of the
ivs must be constants (or NULL_TREE, which is interpreted as constant zero).
LOOP is the loop whose number of iterations we are determining.
ONLY_EXIT is true if we are sure this is the only way the loop could be
exited (including possibly non-returning function calls, exceptions, etc.)
-- in this case we can use the information whether the control induction
variables can overflow or not in a more efficient way.
if EVERY_ITERATION is true, we know the test is executed on every iteration.
The results (number of iterations and assumptions as described in
comments at struct tree_niter_desc in tree-ssa-loop.h) are stored to NITER.
Returns false if it fails to determine number of iterations, true if it
was determined (possibly with some assumptions). */
static bool
number_of_iterations_cond (struct loop *loop,
tree type, affine_iv *iv0, enum tree_code code,
affine_iv *iv1, struct tree_niter_desc *niter,
bool only_exit, bool every_iteration)
{
bool exit_must_be_taken = false, ret;
bounds bnds;
/* If the test is not executed every iteration, wrapping may make the test
to pass again.
TODO: the overflow case can be still used as unreliable estimate of upper
bound. But we have no API to pass it down to number of iterations code
and, at present, it will not use it anyway. */
if (!every_iteration
&& (!iv0->no_overflow || !iv1->no_overflow
|| code == NE_EXPR || code == EQ_EXPR))
return false;
/* The meaning of these assumptions is this:
if !assumptions
then the rest of information does not have to be valid
if may_be_zero then the loop does not roll, even if
niter != 0. */
niter->assumptions = boolean_true_node;
niter->may_be_zero = boolean_false_node;
niter->niter = NULL_TREE;
niter->max = 0;
niter->bound = NULL_TREE;
niter->cmp = ERROR_MARK;
/* Make < comparison from > ones, and for NE_EXPR comparisons, ensure that
the control variable is on lhs. */
if (code == GE_EXPR || code == GT_EXPR
|| (code == NE_EXPR && integer_zerop (iv0->step)))
{
std::swap (iv0, iv1);
code = swap_tree_comparison (code);
}
if (POINTER_TYPE_P (type))
{
/* Comparison of pointers is undefined unless both iv0 and iv1 point
to the same object. If they do, the control variable cannot wrap
(as wrap around the bounds of memory will never return a pointer
that would be guaranteed to point to the same object, even if we
avoid undefined behavior by casting to size_t and back). */
iv0->no_overflow = true;
iv1->no_overflow = true;
}
/* If the control induction variable does not overflow and the only exit
from the loop is the one that we analyze, we know it must be taken
eventually. */
if (only_exit)
{
if (!integer_zerop (iv0->step) && iv0->no_overflow)
exit_must_be_taken = true;
else if (!integer_zerop (iv1->step) && iv1->no_overflow)
exit_must_be_taken = true;
}
/* We can handle cases which neither of the sides of the comparison is
invariant:
{iv0.base, iv0.step} cmp_code {iv1.base, iv1.step}
as if:
{iv0.base, iv0.step - iv1.step} cmp_code {iv1.base, 0}
provided that either below condition is satisfied:
a) the test is NE_EXPR;
b) iv0.step - iv1.step is integer and iv0/iv1 don't overflow.
This rarely occurs in practice, but it is simple enough to manage. */
if (!integer_zerop (iv0->step) && !integer_zerop (iv1->step))
{
tree step_type = POINTER_TYPE_P (type) ? sizetype : type;
tree step = fold_binary_to_constant (MINUS_EXPR, step_type,
iv0->step, iv1->step);
/* No need to check sign of the new step since below code takes care
of this well. */
if (code != NE_EXPR
&& (TREE_CODE (step) != INTEGER_CST
|| !iv0->no_overflow || !iv1->no_overflow))
return false;
iv0->step = step;
if (!POINTER_TYPE_P (type))
iv0->no_overflow = false;
iv1->step = build_int_cst (step_type, 0);
iv1->no_overflow = true;
}
/* If the result of the comparison is a constant, the loop is weird. More
precise handling would be possible, but the situation is not common enough
to waste time on it. */
if (integer_zerop (iv0->step) && integer_zerop (iv1->step))
return false;
/* Ignore loops of while (i-- < 10) type. */
if (code != NE_EXPR)
{
if (iv0->step && tree_int_cst_sign_bit (iv0->step))
return false;
if (!integer_zerop (iv1->step) && !tree_int_cst_sign_bit (iv1->step))
return false;
}
/* If the loop exits immediately, there is nothing to do. */
tree tem = fold_binary (code, boolean_type_node, iv0->base, iv1->base);
if (tem && integer_zerop (tem))
{
niter->niter = build_int_cst (unsigned_type_for (type), 0);
niter->max = 0;
return true;
}
/* OK, now we know we have a senseful loop. Handle several cases, depending
on what comparison operator is used. */
bound_difference (loop, iv1->base, iv0->base, &bnds);
if (dump_file && (dump_flags & TDF_DETAILS))
{
fprintf (dump_file,
"Analyzing # of iterations of loop %d\n", loop->num);
fprintf (dump_file, " exit condition ");
dump_affine_iv (dump_file, iv0);
fprintf (dump_file, " %s ",
code == NE_EXPR ? "!="
: code == LT_EXPR ? "<"
: "<=");
dump_affine_iv (dump_file, iv1);
fprintf (dump_file, "\n");
fprintf (dump_file, " bounds on difference of bases: ");
mpz_out_str (dump_file, 10, bnds.below);
fprintf (dump_file, " ... ");
mpz_out_str (dump_file, 10, bnds.up);
fprintf (dump_file, "\n");
}
switch (code)
{
case NE_EXPR:
gcc_assert (integer_zerop (iv1->step));
ret = number_of_iterations_ne (loop, type, iv0, iv1->base, niter,
exit_must_be_taken, &bnds);
break;
case LT_EXPR:
ret = number_of_iterations_lt (loop, type, iv0, iv1, niter,
exit_must_be_taken, &bnds);
break;
case LE_EXPR:
ret = number_of_iterations_le (loop, type, iv0, iv1, niter,
exit_must_be_taken, &bnds);
break;
default:
gcc_unreachable ();
}
mpz_clear (bnds.up);
mpz_clear (bnds.below);
if (dump_file && (dump_flags & TDF_DETAILS))
{
if (ret)
{
fprintf (dump_file, " result:\n");
if (!integer_nonzerop (niter->assumptions))
{
fprintf (dump_file, " under assumptions ");
print_generic_expr (dump_file, niter->assumptions, TDF_SLIM);
fprintf (dump_file, "\n");
}
if (!integer_zerop (niter->may_be_zero))
{
fprintf (dump_file, " zero if ");
print_generic_expr (dump_file, niter->may_be_zero, TDF_SLIM);
fprintf (dump_file, "\n");
}
fprintf (dump_file, " # of iterations ");
print_generic_expr (dump_file, niter->niter, TDF_SLIM);
fprintf (dump_file, ", bounded by ");
print_decu (niter->max, dump_file);
fprintf (dump_file, "\n");
}
else
fprintf (dump_file, " failed\n\n");
}
return ret;
}
/* Substitute NEW for OLD in EXPR and fold the result. */
static tree
simplify_replace_tree (tree expr, tree old, tree new_tree)
{
unsigned i, n;
tree ret = NULL_TREE, e, se;
if (!expr)
return NULL_TREE;
/* Do not bother to replace constants. */
if (CONSTANT_CLASS_P (old))
return expr;
if (expr == old
|| operand_equal_p (expr, old, 0))
return unshare_expr (new_tree);
if (!EXPR_P (expr))
return expr;
n = TREE_OPERAND_LENGTH (expr);
for (i = 0; i < n; i++)
{
e = TREE_OPERAND (expr, i);
se = simplify_replace_tree (e, old, new_tree);
if (e == se)
continue;
if (!ret)
ret = copy_node (expr);
TREE_OPERAND (ret, i) = se;
}
return (ret ? fold (ret) : expr);
}
/* Expand definitions of ssa names in EXPR as long as they are simple
enough, and return the new expression. If STOP is specified, stop
expanding if EXPR equals to it. */
tree
expand_simple_operations (tree expr, tree stop)
{
unsigned i, n;
tree ret = NULL_TREE, e, ee, e1;
enum tree_code code;
gimple *stmt;
if (expr == NULL_TREE)
return expr;
if (is_gimple_min_invariant (expr))
return expr;
code = TREE_CODE (expr);
if (IS_EXPR_CODE_CLASS (TREE_CODE_CLASS (code)))
{
n = TREE_OPERAND_LENGTH (expr);
for (i = 0; i < n; i++)
{
e = TREE_OPERAND (expr, i);
ee = expand_simple_operations (e, stop);
if (e == ee)
continue;
if (!ret)
ret = copy_node (expr);
TREE_OPERAND (ret, i) = ee;
}
if (!ret)
return expr;
fold_defer_overflow_warnings ();
ret = fold (ret);
fold_undefer_and_ignore_overflow_warnings ();
return ret;
}
/* Stop if it's not ssa name or the one we don't want to expand. */
if (TREE_CODE (expr) != SSA_NAME || expr == stop)
return expr;
stmt = SSA_NAME_DEF_STMT (expr);
if (gimple_code (stmt) == GIMPLE_PHI)
{
basic_block src, dest;
if (gimple_phi_num_args (stmt) != 1)
return expr;
e = PHI_ARG_DEF (stmt, 0);
/* Avoid propagating through loop exit phi nodes, which
could break loop-closed SSA form restrictions. */
dest = gimple_bb (stmt);
src = single_pred (dest);
if (TREE_CODE (e) == SSA_NAME
&& src->loop_father != dest->loop_father)
return expr;
return expand_simple_operations (e, stop);
}
if (gimple_code (stmt) != GIMPLE_ASSIGN)
return expr;
/* Avoid expanding to expressions that contain SSA names that need
to take part in abnormal coalescing. */
ssa_op_iter iter;
FOR_EACH_SSA_TREE_OPERAND (e, stmt, iter, SSA_OP_USE)
if (SSA_NAME_OCCURS_IN_ABNORMAL_PHI (e))
return expr;
e = gimple_assign_rhs1 (stmt);
code = gimple_assign_rhs_code (stmt);
if (get_gimple_rhs_class (code) == GIMPLE_SINGLE_RHS)
{
if (is_gimple_min_invariant (e))
return e;
if (code == SSA_NAME)
return expand_simple_operations (e, stop);
else if (code == ADDR_EXPR)
{
poly_int64 offset;
tree base = get_addr_base_and_unit_offset (TREE_OPERAND (e, 0),
&offset);
if (base
&& TREE_CODE (base) == MEM_REF)
{
ee = expand_simple_operations (TREE_OPERAND (base, 0), stop);
return fold_build2 (POINTER_PLUS_EXPR, TREE_TYPE (expr), ee,
wide_int_to_tree (sizetype,
mem_ref_offset (base)
+ offset));
}
}
return expr;
}
switch (code)
{
CASE_CONVERT:
/* Casts are simple. */
ee = expand_simple_operations (e, stop);
return fold_build1 (code, TREE_TYPE (expr), ee);
case PLUS_EXPR:
case MINUS_EXPR:
if (ANY_INTEGRAL_TYPE_P (TREE_TYPE (expr))
&& TYPE_OVERFLOW_TRAPS (TREE_TYPE (expr)))
return expr;
/* Fallthru. */
case POINTER_PLUS_EXPR:
/* And increments and decrements by a constant are simple. */
e1 = gimple_assign_rhs2 (stmt);
if (!is_gimple_min_invariant (e1))
return expr;
ee = expand_simple_operations (e, stop);
return fold_build2 (code, TREE_TYPE (expr), ee, e1);
default:
return expr;
}
}
/* Tries to simplify EXPR using the condition COND. Returns the simplified
expression (or EXPR unchanged, if no simplification was possible). */
static tree
tree_simplify_using_condition_1 (tree cond, tree expr)
{
bool changed;
tree e, e0, e1, e2, notcond;
enum tree_code code = TREE_CODE (expr);
if (code == INTEGER_CST)
return expr;
if (code == TRUTH_OR_EXPR
|| code == TRUTH_AND_EXPR
|| code == COND_EXPR)
{
changed = false;
e0 = tree_simplify_using_condition_1 (cond, TREE_OPERAND (expr, 0));
if (TREE_OPERAND (expr, 0) != e0)
changed = true;
e1 = tree_simplify_using_condition_1 (cond, TREE_OPERAND (expr, 1));
if (TREE_OPERAND (expr, 1) != e1)
changed = true;
if (code == COND_EXPR)
{
e2 = tree_simplify_using_condition_1 (cond, TREE_OPERAND (expr, 2));
if (TREE_OPERAND (expr, 2) != e2)
changed = true;
}
else
e2 = NULL_TREE;
if (changed)
{
if (code == COND_EXPR)
expr = fold_build3 (code, boolean_type_node, e0, e1, e2);
else
expr = fold_build2 (code, boolean_type_node, e0, e1);
}
return expr;
}
/* In case COND is equality, we may be able to simplify EXPR by copy/constant
propagation, and vice versa. Fold does not handle this, since it is
considered too expensive. */
if (TREE_CODE (cond) == EQ_EXPR)
{
e0 = TREE_OPERAND (cond, 0);
e1 = TREE_OPERAND (cond, 1);
/* We know that e0 == e1. Check whether we cannot simplify expr
using this fact. */
e = simplify_replace_tree (expr, e0, e1);
if (integer_zerop (e) || integer_nonzerop (e))
return e;
e = simplify_replace_tree (expr, e1, e0);
if (integer_zerop (e) || integer_nonzerop (e))
return e;
}
if (TREE_CODE (expr) == EQ_EXPR)
{
e0 = TREE_OPERAND (expr, 0);
e1 = TREE_OPERAND (expr, 1);
/* If e0 == e1 (EXPR) implies !COND, then EXPR cannot be true. */
e = simplify_replace_tree (cond, e0, e1);
if (integer_zerop (e))
return e;
e = simplify_replace_tree (cond, e1, e0);
if (integer_zerop (e))
return e;
}
if (TREE_CODE (expr) == NE_EXPR)
{
e0 = TREE_OPERAND (expr, 0);
e1 = TREE_OPERAND (expr, 1);
/* If e0 == e1 (!EXPR) implies !COND, then EXPR must be true. */
e = simplify_replace_tree (cond, e0, e1);
if (integer_zerop (e))
return boolean_true_node;
e = simplify_replace_tree (cond, e1, e0);
if (integer_zerop (e))
return boolean_true_node;
}
/* Check whether COND ==> EXPR. */
notcond = invert_truthvalue (cond);
e = fold_binary (TRUTH_OR_EXPR, boolean_type_node, notcond, expr);
if (e && integer_nonzerop (e))
return e;
/* Check whether COND ==> not EXPR. */
e = fold_binary (TRUTH_AND_EXPR, boolean_type_node, cond, expr);
if (e && integer_zerop (e))
return e;
return expr;
}
/* Tries to simplify EXPR using the condition COND. Returns the simplified
expression (or EXPR unchanged, if no simplification was possible).
Wrapper around tree_simplify_using_condition_1 that ensures that chains
of simple operations in definitions of ssa names in COND are expanded,
so that things like casts or incrementing the value of the bound before
the loop do not cause us to fail. */
static tree
tree_simplify_using_condition (tree cond, tree expr)
{
cond = expand_simple_operations (cond);
return tree_simplify_using_condition_1 (cond, expr);
}
/* Tries to simplify EXPR using the conditions on entry to LOOP.
Returns the simplified expression (or EXPR unchanged, if no
simplification was possible). */
tree
simplify_using_initial_conditions (struct loop *loop, tree expr)
{
edge e;
basic_block bb;
gimple *stmt;
tree cond, expanded, backup;
int cnt = 0;
if (TREE_CODE (expr) == INTEGER_CST)
return expr;
backup = expanded = expand_simple_operations (expr);
/* Limit walking the dominators to avoid quadraticness in
the number of BBs times the number of loops in degenerate
cases. */
for (bb = loop->header;
bb != ENTRY_BLOCK_PTR_FOR_FN (cfun) && cnt < MAX_DOMINATORS_TO_WALK;
bb = get_immediate_dominator (CDI_DOMINATORS, bb))
{
if (!single_pred_p (bb))
continue;
e = single_pred_edge (bb);
if (!(e->flags & (EDGE_TRUE_VALUE | EDGE_FALSE_VALUE)))
continue;
stmt = last_stmt (e->src);
cond = fold_build2 (gimple_cond_code (stmt),
boolean_type_node,
gimple_cond_lhs (stmt),
gimple_cond_rhs (stmt));
if (e->flags & EDGE_FALSE_VALUE)
cond = invert_truthvalue (cond);
expanded = tree_simplify_using_condition (cond, expanded);
/* Break if EXPR is simplified to const values. */
if (expanded
&& (integer_zerop (expanded) || integer_nonzerop (expanded)))
return expanded;
++cnt;
}
/* Return the original expression if no simplification is done. */
return operand_equal_p (backup, expanded, 0) ? expr : expanded;
}
/* Tries to simplify EXPR using the evolutions of the loop invariants
in the superloops of LOOP. Returns the simplified expression
(or EXPR unchanged, if no simplification was possible). */
static tree
simplify_using_outer_evolutions (struct loop *loop, tree expr)
{
enum tree_code code = TREE_CODE (expr);
bool changed;
tree e, e0, e1, e2;
if (is_gimple_min_invariant (expr))
return expr;
if (code == TRUTH_OR_EXPR
|| code == TRUTH_AND_EXPR
|| code == COND_EXPR)
{
changed = false;
e0 = simplify_using_outer_evolutions (loop, TREE_OPERAND (expr, 0));
if (TREE_OPERAND (expr, 0) != e0)
changed = true;
e1 = simplify_using_outer_evolutions (loop, TREE_OPERAND (expr, 1));
if (TREE_OPERAND (expr, 1) != e1)
changed = true;
if (code == COND_EXPR)
{
e2 = simplify_using_outer_evolutions (loop, TREE_OPERAND (expr, 2));
if (TREE_OPERAND (expr, 2) != e2)
changed = true;
}
else
e2 = NULL_TREE;
if (changed)
{
if (code == COND_EXPR)
expr = fold_build3 (code, boolean_type_node, e0, e1, e2);
else
expr = fold_build2 (code, boolean_type_node, e0, e1);
}
return expr;
}
e = instantiate_parameters (loop, expr);
if (is_gimple_min_invariant (e))
return e;
return expr;
}
/* Returns true if EXIT is the only possible exit from LOOP. */
bool
loop_only_exit_p (const struct loop *loop, const_edge exit)
{
basic_block *body;
gimple_stmt_iterator bsi;
unsigned i;
if (exit != single_exit (loop))
return false;
body = get_loop_body (loop);
for (i = 0; i < loop->num_nodes; i++)
{
for (bsi = gsi_start_bb (body[i]); !gsi_end_p (bsi); gsi_next (&bsi))
if (stmt_can_terminate_bb_p (gsi_stmt (bsi)))
{
free (body);
return true;
}
}
free (body);
return true;
}
/* Stores description of number of iterations of LOOP derived from
EXIT (an exit edge of the LOOP) in NITER. Returns true if some useful
information could be derived (and fields of NITER have meaning described
in comments at struct tree_niter_desc declaration), false otherwise.
When EVERY_ITERATION is true, only tests that are known to be executed
every iteration are considered (i.e. only test that alone bounds the loop).
If AT_STMT is not NULL, this function stores LOOP's condition statement in
it when returning true. */
bool
number_of_iterations_exit_assumptions (struct loop *loop, edge exit,
struct tree_niter_desc *niter,
gcond **at_stmt, bool every_iteration)
{
gimple *last;
gcond *stmt;
tree type;
tree op0, op1;
enum tree_code code;
affine_iv iv0, iv1;
bool safe;
/* Nothing to analyze if the loop is known to be infinite. */
if (loop_constraint_set_p (loop, LOOP_C_INFINITE))
return false;
safe = dominated_by_p (CDI_DOMINATORS, loop->latch, exit->src);
if (every_iteration && !safe)
return false;
niter->assumptions = boolean_false_node;
niter->control.base = NULL_TREE;
niter->control.step = NULL_TREE;
niter->control.no_overflow = false;
last = last_stmt (exit->src);
if (!last)
return false;
stmt = dyn_cast <gcond *> (last);
if (!stmt)
return false;
/* We want the condition for staying inside loop. */
code = gimple_cond_code (stmt);
if (exit->flags & EDGE_TRUE_VALUE)
code = invert_tree_comparison (code, false);
switch (code)
{
case GT_EXPR:
case GE_EXPR:
case LT_EXPR:
case LE_EXPR:
case NE_EXPR:
break;
default:
return false;
}
op0 = gimple_cond_lhs (stmt);
op1 = gimple_cond_rhs (stmt);
type = TREE_TYPE (op0);
if (TREE_CODE (type) != INTEGER_TYPE
&& !POINTER_TYPE_P (type))
return false;
tree iv0_niters = NULL_TREE;
if (!simple_iv_with_niters (loop, loop_containing_stmt (stmt),
op0, &iv0, safe ? &iv0_niters : NULL, false))
return number_of_iterations_popcount (loop, exit, code, niter);
tree iv1_niters = NULL_TREE;
if (!simple_iv_with_niters (loop, loop_containing_stmt (stmt),
op1, &iv1, safe ? &iv1_niters : NULL, false))
return false;
/* Give up on complicated case. */
if (iv0_niters && iv1_niters)
return false;
/* We don't want to see undefined signed overflow warnings while
computing the number of iterations. */
fold_defer_overflow_warnings ();
iv0.base = expand_simple_operations (iv0.base);
iv1.base = expand_simple_operations (iv1.base);
if (!number_of_iterations_cond (loop, type, &iv0, code, &iv1, niter,
loop_only_exit_p (loop, exit), safe))
{
fold_undefer_and_ignore_overflow_warnings ();
return false;
}
/* Incorporate additional assumption implied by control iv. */
tree iv_niters = iv0_niters ? iv0_niters : iv1_niters;
if (iv_niters)
{
tree assumption = fold_build2 (LE_EXPR, boolean_type_node, niter->niter,
fold_convert (TREE_TYPE (niter->niter),
iv_niters));
if (!integer_nonzerop (assumption))
niter->assumptions = fold_build2 (TRUTH_AND_EXPR, boolean_type_node,
niter->assumptions, assumption);
/* Refine upper bound if possible. */
if (TREE_CODE (iv_niters) == INTEGER_CST
&& niter->max > wi::to_widest (iv_niters))
niter->max = wi::to_widest (iv_niters);
}
/* There is no assumptions if the loop is known to be finite. */
if (!integer_zerop (niter->assumptions)
&& loop_constraint_set_p (loop, LOOP_C_FINITE))
niter->assumptions = boolean_true_node;
if (optimize >= 3)
{
niter->assumptions = simplify_using_outer_evolutions (loop,
niter->assumptions);
niter->may_be_zero = simplify_using_outer_evolutions (loop,
niter->may_be_zero);
niter->niter = simplify_using_outer_evolutions (loop, niter->niter);
}
niter->assumptions
= simplify_using_initial_conditions (loop,
niter->assumptions);
niter->may_be_zero
= simplify_using_initial_conditions (loop,
niter->may_be_zero);
fold_undefer_and_ignore_overflow_warnings ();
/* If NITER has simplified into a constant, update MAX. */
if (TREE_CODE (niter->niter) == INTEGER_CST)
niter->max = wi::to_widest (niter->niter);
if (at_stmt)
*at_stmt = stmt;
return (!integer_zerop (niter->assumptions));
}
/* Utility function to check if OP is defined by a stmt
that is a val - 1. */
static bool
ssa_defined_by_minus_one_stmt_p (tree op, tree val)
{
gimple *stmt;
return (TREE_CODE (op) == SSA_NAME
&& (stmt = SSA_NAME_DEF_STMT (op))
&& is_gimple_assign (stmt)
&& (gimple_assign_rhs_code (stmt) == PLUS_EXPR)
&& val == gimple_assign_rhs1 (stmt)
&& integer_minus_onep (gimple_assign_rhs2 (stmt)));
}
/* See if LOOP is a popcout implementation, determine NITER for the loop
We match:
<bb 2>
goto <bb 4>
<bb 3>
_1 = b_11 + -1
b_6 = _1 & b_11
<bb 4>
b_11 = PHI <b_5(D)(2), b_6(3)>
exit block
if (b_11 != 0)
goto <bb 3>
else
goto <bb 5>
OR we match copy-header version:
if (b_5 != 0)
goto <bb 3>
else
goto <bb 4>
<bb 3>
b_11 = PHI <b_5(2), b_6(3)>
_1 = b_11 + -1
b_6 = _1 & b_11
exit block
if (b_6 != 0)
goto <bb 3>
else
goto <bb 4>
If popcount pattern, update NITER accordingly.
i.e., set NITER to __builtin_popcount (b)
return true if we did, false otherwise.
*/
static bool
number_of_iterations_popcount (loop_p loop, edge exit,
enum tree_code code,
struct tree_niter_desc *niter)
{
bool adjust = true;
tree iter;
HOST_WIDE_INT max;
adjust = true;
tree fn = NULL_TREE;
/* Check loop terminating branch is like
if (b != 0). */
gimple *stmt = last_stmt (exit->src);
if (!stmt
|| gimple_code (stmt) != GIMPLE_COND
|| code != NE_EXPR
|| !integer_zerop (gimple_cond_rhs (stmt))
|| TREE_CODE (gimple_cond_lhs (stmt)) != SSA_NAME)
return false;
gimple *and_stmt = SSA_NAME_DEF_STMT (gimple_cond_lhs (stmt));
/* Depending on copy-header is performed, feeding PHI stmts might be in
the loop header or loop latch, handle this. */
if (gimple_code (and_stmt) == GIMPLE_PHI
&& gimple_bb (and_stmt) == loop->header
&& gimple_phi_num_args (and_stmt) == 2
&& (TREE_CODE (gimple_phi_arg_def (and_stmt,
loop_latch_edge (loop)->dest_idx))
== SSA_NAME))
{
/* SSA used in exit condition is defined by PHI stmt
b_11 = PHI <b_5(D)(2), b_6(3)>
from the PHI stmt, get the and_stmt
b_6 = _1 & b_11. */
tree t = gimple_phi_arg_def (and_stmt, loop_latch_edge (loop)->dest_idx);
and_stmt = SSA_NAME_DEF_STMT (t);
adjust = false;
}
/* Make sure it is indeed an and stmt (b_6 = _1 & b_11). */
if (!is_gimple_assign (and_stmt)
|| gimple_assign_rhs_code (and_stmt) != BIT_AND_EXPR)
return false;
tree b_11 = gimple_assign_rhs1 (and_stmt);
tree _1 = gimple_assign_rhs2 (and_stmt);
/* Check that _1 is defined by _b11 + -1 (_1 = b_11 + -1).
Also make sure that b_11 is the same in and_stmt and _1 defining stmt.
Also canonicalize if _1 and _b11 are revrsed. */
if (ssa_defined_by_minus_one_stmt_p (b_11, _1))
std::swap (b_11, _1);
else if (ssa_defined_by_minus_one_stmt_p (_1, b_11))
;
else
return false;
/* Check the recurrence:
... = PHI <b_5(2), b_6(3)>. */
gimple *phi = SSA_NAME_DEF_STMT (b_11);
if (gimple_code (phi) != GIMPLE_PHI
|| (gimple_bb (phi) != loop_latch_edge (loop)->dest)
|| (gimple_assign_lhs (and_stmt)
!= gimple_phi_arg_def (phi, loop_latch_edge (loop)->dest_idx)))
return false;
/* We found a match. Get the corresponding popcount builtin. */
tree src = gimple_phi_arg_def (phi, loop_preheader_edge (loop)->dest_idx);
if (TYPE_PRECISION (TREE_TYPE (src)) == TYPE_PRECISION (integer_type_node))
fn = builtin_decl_implicit (BUILT_IN_POPCOUNT);
else if (TYPE_PRECISION (TREE_TYPE (src)) == TYPE_PRECISION
(long_integer_type_node))
fn = builtin_decl_implicit (BUILT_IN_POPCOUNTL);
else if (TYPE_PRECISION (TREE_TYPE (src)) == TYPE_PRECISION
(long_long_integer_type_node))
fn = builtin_decl_implicit (BUILT_IN_POPCOUNTLL);
/* ??? Support promoting char/short to int. */
if (!fn)
return false;
/* Update NITER params accordingly */
tree utype = unsigned_type_for (TREE_TYPE (src));
src = fold_convert (utype, src);
tree call = fold_convert (utype, build_call_expr (fn, 1, src));
if (adjust)
iter = fold_build2 (MINUS_EXPR, utype,
call,
build_int_cst (utype, 1));
else
iter = call;
if (TREE_CODE (call) == INTEGER_CST)
max = tree_to_uhwi (call);
else
{
max = TYPE_PRECISION (TREE_TYPE (src));
if (adjust)
max = max - 1;
}
niter->niter = iter;
niter->assumptions = boolean_true_node;
if (adjust)
{
tree may_be_zero = fold_build2 (EQ_EXPR, boolean_type_node, src,
build_zero_cst
(TREE_TYPE (src)));
niter->may_be_zero =
simplify_using_initial_conditions (loop, may_be_zero);
}
else
niter->may_be_zero = boolean_false_node;
niter->max = max;
niter->bound = NULL_TREE;
niter->cmp = ERROR_MARK;
return true;
}
/* Like number_of_iterations_exit_assumptions, but return TRUE only if
the niter information holds unconditionally. */
bool
number_of_iterations_exit (struct loop *loop, edge exit,
struct tree_niter_desc *niter,
bool warn, bool every_iteration)
{
gcond *stmt;
if (!number_of_iterations_exit_assumptions (loop, exit, niter,
&stmt, every_iteration))
return false;
if (integer_nonzerop (niter->assumptions))
return true;
if (warn)
dump_printf_loc (MSG_MISSED_OPTIMIZATION, stmt,
"missed loop optimization: niters analysis ends up "
"with assumptions.\n");
return false;
}
/* Try to determine the number of iterations of LOOP. If we succeed,
expression giving number of iterations is returned and *EXIT is
set to the edge from that the information is obtained. Otherwise
chrec_dont_know is returned. */
tree
find_loop_niter (struct loop *loop, edge *exit)
{
unsigned i;
vec<edge> exits = get_loop_exit_edges (loop);
edge ex;
tree niter = NULL_TREE, aniter;
struct tree_niter_desc desc;
*exit = NULL;
FOR_EACH_VEC_ELT (exits, i, ex)
{
if (!number_of_iterations_exit (loop, ex, &desc, false))
continue;
if (integer_nonzerop (desc.may_be_zero))
{
/* We exit in the first iteration through this exit.
We won't find anything better. */
niter = build_int_cst (unsigned_type_node, 0);
*exit = ex;
break;
}
if (!integer_zerop (desc.may_be_zero))
continue;
aniter = desc.niter;
if (!niter)
{
/* Nothing recorded yet. */
niter = aniter;
*exit = ex;
continue;
}
/* Prefer constants, the lower the better. */
if (TREE_CODE (aniter) != INTEGER_CST)
continue;
if (TREE_CODE (niter) != INTEGER_CST)
{
niter = aniter;
*exit = ex;
continue;
}
if (tree_int_cst_lt (aniter, niter))
{
niter = aniter;
*exit = ex;
continue;
}
}
exits.release ();
return niter ? niter : chrec_dont_know;
}
/* Return true if loop is known to have bounded number of iterations. */
bool
finite_loop_p (struct loop *loop)
{
widest_int nit;
int flags;
flags = flags_from_decl_or_type (current_function_decl);
if ((flags & (ECF_CONST|ECF_PURE)) && !(flags & ECF_LOOPING_CONST_OR_PURE))
{
if (dump_file && (dump_flags & TDF_DETAILS))
fprintf (dump_file, "Found loop %i to be finite: it is within pure or const function.\n",
loop->num);
return true;
}
if (loop->any_upper_bound
|| max_loop_iterations (loop, &nit))
{
if (dump_file && (dump_flags & TDF_DETAILS))
fprintf (dump_file, "Found loop %i to be finite: upper bound found.\n",
loop->num);
return true;
}
return false;
}
/*
Analysis of a number of iterations of a loop by a brute-force evaluation.
*/
/* Bound on the number of iterations we try to evaluate. */
#define MAX_ITERATIONS_TO_TRACK \
((unsigned) PARAM_VALUE (PARAM_MAX_ITERATIONS_TO_TRACK))
/* Returns the loop phi node of LOOP such that ssa name X is derived from its
result by a chain of operations such that all but exactly one of their
operands are constants. */
static gphi *
chain_of_csts_start (struct loop *loop, tree x)
{
gimple *stmt = SSA_NAME_DEF_STMT (x);
tree use;
basic_block bb = gimple_bb (stmt);
enum tree_code code;
if (!bb
|| !flow_bb_inside_loop_p (loop, bb))
return NULL;
if (gimple_code (stmt) == GIMPLE_PHI)
{
if (bb == loop->header)
return as_a <gphi *> (stmt);
return NULL;
}
if (gimple_code (stmt) != GIMPLE_ASSIGN
|| gimple_assign_rhs_class (stmt) == GIMPLE_TERNARY_RHS)
return NULL;
code = gimple_assign_rhs_code (stmt);
if (gimple_references_memory_p (stmt)
|| TREE_CODE_CLASS (code) == tcc_reference
|| (code == ADDR_EXPR
&& !is_gimple_min_invariant (gimple_assign_rhs1 (stmt))))
return NULL;
use = SINGLE_SSA_TREE_OPERAND (stmt, SSA_OP_USE);
if (use == NULL_TREE)
return NULL;
return chain_of_csts_start (loop, use);
}
/* Determines whether the expression X is derived from a result of a phi node
in header of LOOP such that
* the derivation of X consists only from operations with constants
* the initial value of the phi node is constant
* the value of the phi node in the next iteration can be derived from the
value in the current iteration by a chain of operations with constants,
or is also a constant
If such phi node exists, it is returned, otherwise NULL is returned. */
static gphi *
get_base_for (struct loop *loop, tree x)
{
gphi *phi;
tree init, next;
if (is_gimple_min_invariant (x))
return NULL;
phi = chain_of_csts_start (loop, x);
if (!phi)
return NULL;
init = PHI_ARG_DEF_FROM_EDGE (phi, loop_preheader_edge (loop));
next = PHI_ARG_DEF_FROM_EDGE (phi, loop_latch_edge (loop));
if (!is_gimple_min_invariant (init))
return NULL;
if (TREE_CODE (next) == SSA_NAME
&& chain_of_csts_start (loop, next) != phi)
return NULL;
return phi;
}
/* Given an expression X, then
* if X is NULL_TREE, we return the constant BASE.
* if X is a constant, we return the constant X.
* otherwise X is a SSA name, whose value in the considered loop is derived
by a chain of operations with constant from a result of a phi node in
the header of the loop. Then we return value of X when the value of the
result of this phi node is given by the constant BASE. */
static tree
get_val_for (tree x, tree base)
{
gimple *stmt;
gcc_checking_assert (is_gimple_min_invariant (base));
if (!x)
return base;
else if (is_gimple_min_invariant (x))
return x;
stmt = SSA_NAME_DEF_STMT (x);
if (gimple_code (stmt) == GIMPLE_PHI)
return base;
gcc_checking_assert (is_gimple_assign (stmt));
/* STMT must be either an assignment of a single SSA name or an
expression involving an SSA name and a constant. Try to fold that
expression using the value for the SSA name. */
if (gimple_assign_ssa_name_copy_p (stmt))
return get_val_for (gimple_assign_rhs1 (stmt), base);
else if (gimple_assign_rhs_class (stmt) == GIMPLE_UNARY_RHS
&& TREE_CODE (gimple_assign_rhs1 (stmt)) == SSA_NAME)
return fold_build1 (gimple_assign_rhs_code (stmt),
gimple_expr_type (stmt),
get_val_for (gimple_assign_rhs1 (stmt), base));
else if (gimple_assign_rhs_class (stmt) == GIMPLE_BINARY_RHS)
{
tree rhs1 = gimple_assign_rhs1 (stmt);
tree rhs2 = gimple_assign_rhs2 (stmt);
if (TREE_CODE (rhs1) == SSA_NAME)
rhs1 = get_val_for (rhs1, base);
else if (TREE_CODE (rhs2) == SSA_NAME)
rhs2 = get_val_for (rhs2, base);
else
gcc_unreachable ();
return fold_build2 (gimple_assign_rhs_code (stmt),
gimple_expr_type (stmt), rhs1, rhs2);
}
else
gcc_unreachable ();
}
/* Tries to count the number of iterations of LOOP till it exits by EXIT
by brute force -- i.e. by determining the value of the operands of the
condition at EXIT in first few iterations of the loop (assuming that
these values are constant) and determining the first one in that the
condition is not satisfied. Returns the constant giving the number
of the iterations of LOOP if successful, chrec_dont_know otherwise. */
tree
loop_niter_by_eval (struct loop *loop, edge exit)
{
tree acnd;
tree op[2], val[2], next[2], aval[2];
gphi *phi;
gimple *cond;
unsigned i, j;
enum tree_code cmp;
cond = last_stmt (exit->src);
if (!cond || gimple_code (cond) != GIMPLE_COND)
return chrec_dont_know;
cmp = gimple_cond_code (cond);
if (exit->flags & EDGE_TRUE_VALUE)
cmp = invert_tree_comparison (cmp, false);
switch (cmp)
{
case EQ_EXPR:
case NE_EXPR:
case GT_EXPR:
case GE_EXPR:
case LT_EXPR:
case LE_EXPR:
op[0] = gimple_cond_lhs (cond);
op[1] = gimple_cond_rhs (cond);
break;
default:
return chrec_dont_know;
}
for (j = 0; j < 2; j++)
{
if (is_gimple_min_invariant (op[j]))
{
val[j] = op[j];
next[j] = NULL_TREE;
op[j] = NULL_TREE;
}
else
{
phi = get_base_for (loop, op[j]);
if (!phi)
return chrec_dont_know;
val[j] = PHI_ARG_DEF_FROM_EDGE (phi, loop_preheader_edge (loop));
next[j] = PHI_ARG_DEF_FROM_EDGE (phi, loop_latch_edge (loop));
}
}
/* Don't issue signed overflow warnings. */
fold_defer_overflow_warnings ();
for (i = 0; i < MAX_ITERATIONS_TO_TRACK; i++)
{
for (j = 0; j < 2; j++)
aval[j] = get_val_for (op[j], val[j]);
acnd = fold_binary (cmp, boolean_type_node, aval[0], aval[1]);
if (acnd && integer_zerop (acnd))
{
fold_undefer_and_ignore_overflow_warnings ();
if (dump_file && (dump_flags & TDF_DETAILS))
fprintf (dump_file,
"Proved that loop %d iterates %d times using brute force.\n",
loop->num, i);
return build_int_cst (unsigned_type_node, i);
}
for (j = 0; j < 2; j++)
{
aval[j] = val[j];
val[j] = get_val_for (next[j], val[j]);
if (!is_gimple_min_invariant (val[j]))
{
fold_undefer_and_ignore_overflow_warnings ();
return chrec_dont_know;
}
}
/* If the next iteration would use the same base values
as the current one, there is no point looping further,
all following iterations will be the same as this one. */
if (val[0] == aval[0] && val[1] == aval[1])
break;
}
fold_undefer_and_ignore_overflow_warnings ();
return chrec_dont_know;
}
/* Finds the exit of the LOOP by that the loop exits after a constant
number of iterations and stores the exit edge to *EXIT. The constant
giving the number of iterations of LOOP is returned. The number of
iterations is determined using loop_niter_by_eval (i.e. by brute force
evaluation). If we are unable to find the exit for that loop_niter_by_eval
determines the number of iterations, chrec_dont_know is returned. */
tree
find_loop_niter_by_eval (struct loop *loop, edge *exit)
{
unsigned i;
vec<edge> exits = get_loop_exit_edges (loop);
edge ex;
tree niter = NULL_TREE, aniter;
*exit = NULL;
/* Loops with multiple exits are expensive to handle and less important. */
if (!flag_expensive_optimizations
&& exits.length () > 1)
{
exits.release ();
return chrec_dont_know;
}
FOR_EACH_VEC_ELT (exits, i, ex)
{
if (!just_once_each_iteration_p (loop, ex->src))
continue;
aniter = loop_niter_by_eval (loop, ex);
if (chrec_contains_undetermined (aniter))
continue;
if (niter
&& !tree_int_cst_lt (aniter, niter))
continue;
niter = aniter;
*exit = ex;
}
exits.release ();
return niter ? niter : chrec_dont_know;
}
/*
Analysis of upper bounds on number of iterations of a loop.
*/
static widest_int derive_constant_upper_bound_ops (tree, tree,
enum tree_code, tree);
/* Returns a constant upper bound on the value of the right-hand side of
an assignment statement STMT. */
static widest_int
derive_constant_upper_bound_assign (gimple *stmt)
{
enum tree_code code = gimple_assign_rhs_code (stmt);
tree op0 = gimple_assign_rhs1 (stmt);
tree op1 = gimple_assign_rhs2 (stmt);
return derive_constant_upper_bound_ops (TREE_TYPE (gimple_assign_lhs (stmt)),
op0, code, op1);
}
/* Returns a constant upper bound on the value of expression VAL. VAL
is considered to be unsigned. If its type is signed, its value must
be nonnegative. */
static widest_int
derive_constant_upper_bound (tree val)
{
enum tree_code code;
tree op0, op1, op2;
extract_ops_from_tree (val, &code, &op0, &op1, &op2);
return derive_constant_upper_bound_ops (TREE_TYPE (val), op0, code, op1);
}
/* Returns a constant upper bound on the value of expression OP0 CODE OP1,
whose type is TYPE. The expression is considered to be unsigned. If
its type is signed, its value must be nonnegative. */
static widest_int
derive_constant_upper_bound_ops (tree type, tree op0,
enum tree_code code, tree op1)
{
tree subtype, maxt;
widest_int bnd, max, cst;
gimple *stmt;
if (INTEGRAL_TYPE_P (type))
maxt = TYPE_MAX_VALUE (type);
else
maxt = upper_bound_in_type (type, type);
max = wi::to_widest (maxt);
switch (code)
{
case INTEGER_CST:
return wi::to_widest (op0);
CASE_CONVERT:
subtype = TREE_TYPE (op0);
if (!TYPE_UNSIGNED (subtype)
/* If TYPE is also signed, the fact that VAL is nonnegative implies
that OP0 is nonnegative. */
&& TYPE_UNSIGNED (type)
&& !tree_expr_nonnegative_p (op0))
{
/* If we cannot prove that the casted expression is nonnegative,
we cannot establish more useful upper bound than the precision
of the type gives us. */
return max;
}
/* We now know that op0 is an nonnegative value. Try deriving an upper
bound for it. */
bnd = derive_constant_upper_bound (op0);
/* If the bound does not fit in TYPE, max. value of TYPE could be
attained. */
if (wi::ltu_p (max, bnd))
return max;
return bnd;
case PLUS_EXPR:
case POINTER_PLUS_EXPR:
case MINUS_EXPR:
if (TREE_CODE (op1) != INTEGER_CST
|| !tree_expr_nonnegative_p (op0))
return max;
/* Canonicalize to OP0 - CST. Consider CST to be signed, in order to
choose the most logical way how to treat this constant regardless
of the signedness of the type. */
cst = wi::sext (wi::to_widest (op1), TYPE_PRECISION (type));
if (code != MINUS_EXPR)
cst = -cst;
bnd = derive_constant_upper_bound (op0);
if (wi::neg_p (cst))
{
cst = -cst;
/* Avoid CST == 0x80000... */
if (wi::neg_p (cst))
return max;
/* OP0 + CST. We need to check that
BND <= MAX (type) - CST. */
widest_int mmax = max - cst;
if (wi::leu_p (bnd, mmax))
return max;
return bnd + cst;
}
else
{
/* OP0 - CST, where CST >= 0.
If TYPE is signed, we have already verified that OP0 >= 0, and we
know that the result is nonnegative. This implies that
VAL <= BND - CST.
If TYPE is unsigned, we must additionally know that OP0 >= CST,
otherwise the operation underflows.
*/
/* This should only happen if the type is unsigned; however, for
buggy programs that use overflowing signed arithmetics even with
-fno-wrapv, this condition may also be true for signed values. */
if (wi::ltu_p (bnd, cst))
return max;
if (TYPE_UNSIGNED (type))
{
tree tem = fold_binary (GE_EXPR, boolean_type_node, op0,
wide_int_to_tree (type, cst));
if (!tem || integer_nonzerop (tem))
return max;
}
bnd -= cst;
}
return bnd;
case FLOOR_DIV_EXPR:
case EXACT_DIV_EXPR:
if (TREE_CODE (op1) != INTEGER_CST
|| tree_int_cst_sign_bit (op1))
return max;
bnd = derive_constant_upper_bound (op0);
return wi::udiv_floor (bnd, wi::to_widest (op1));
case BIT_AND_EXPR:
if (TREE_CODE (op1) != INTEGER_CST
|| tree_int_cst_sign_bit (op1))
return max;
return wi::to_widest (op1);
case SSA_NAME:
stmt = SSA_NAME_DEF_STMT (op0);
if (gimple_code (stmt) != GIMPLE_ASSIGN
|| gimple_assign_lhs (stmt) != op0)
return max;
return derive_constant_upper_bound_assign (stmt);
default:
return max;
}
}
/* Emit a -Waggressive-loop-optimizations warning if needed. */
static void
do_warn_aggressive_loop_optimizations (struct loop *loop,
widest_int i_bound, gimple *stmt)
{
/* Don't warn if the loop doesn't have known constant bound. */
if (!loop->nb_iterations
|| TREE_CODE (loop->nb_iterations) != INTEGER_CST
|| !warn_aggressive_loop_optimizations
/* To avoid warning multiple times for the same loop,
only start warning when we preserve loops. */
|| (cfun->curr_properties & PROP_loops) == 0
/* Only warn once per loop. */
|| loop->warned_aggressive_loop_optimizations
/* Only warn if undefined behavior gives us lower estimate than the
known constant bound. */
|| wi::cmpu (i_bound, wi::to_widest (loop->nb_iterations)) >= 0
/* And undefined behavior happens unconditionally. */
|| !dominated_by_p (CDI_DOMINATORS, loop->latch, gimple_bb (stmt)))
return;
edge e = single_exit (loop);
if (e == NULL)
return;
gimple *estmt = last_stmt (e->src);
char buf[WIDE_INT_PRINT_BUFFER_SIZE];
print_dec (i_bound, buf, TYPE_UNSIGNED (TREE_TYPE (loop->nb_iterations))
? UNSIGNED : SIGNED);
auto_diagnostic_group d;
if (warning_at (gimple_location (stmt), OPT_Waggressive_loop_optimizations,
"iteration %s invokes undefined behavior", buf))
inform (gimple_location (estmt), "within this loop");
loop->warned_aggressive_loop_optimizations = true;
}
/* Records that AT_STMT is executed at most BOUND + 1 times in LOOP. IS_EXIT
is true if the loop is exited immediately after STMT, and this exit
is taken at last when the STMT is executed BOUND + 1 times.
REALISTIC is true if BOUND is expected to be close to the real number
of iterations. UPPER is true if we are sure the loop iterates at most
BOUND times. I_BOUND is a widest_int upper estimate on BOUND. */
static void
record_estimate (struct loop *loop, tree bound, const widest_int &i_bound,
gimple *at_stmt, bool is_exit, bool realistic, bool upper)
{
widest_int delta;
if (dump_file && (dump_flags & TDF_DETAILS))
{
fprintf (dump_file, "Statement %s", is_exit ? "(exit)" : "");
print_gimple_stmt (dump_file, at_stmt, 0, TDF_SLIM);
fprintf (dump_file, " is %sexecuted at most ",
upper ? "" : "probably ");
print_generic_expr (dump_file, bound, TDF_SLIM);
fprintf (dump_file, " (bounded by ");
print_decu (i_bound, dump_file);
fprintf (dump_file, ") + 1 times in loop %d.\n", loop->num);
}
/* If the I_BOUND is just an estimate of BOUND, it rarely is close to the
real number of iterations. */
if (TREE_CODE (bound) != INTEGER_CST)
realistic = false;
else
gcc_checking_assert (i_bound == wi::to_widest (bound));
/* If we have a guaranteed upper bound, record it in the appropriate
list, unless this is an !is_exit bound (i.e. undefined behavior in
at_stmt) in a loop with known constant number of iterations. */
if (upper
&& (is_exit
|| loop->nb_iterations == NULL_TREE
|| TREE_CODE (loop->nb_iterations) != INTEGER_CST))
{
struct nb_iter_bound *elt = ggc_alloc<nb_iter_bound> ();
elt->bound = i_bound;
elt->stmt = at_stmt;
elt->is_exit = is_exit;
elt->next = loop->bounds;
loop->bounds = elt;
}
/* If statement is executed on every path to the loop latch, we can directly
infer the upper bound on the # of iterations of the loop. */
if (!dominated_by_p (CDI_DOMINATORS, loop->latch, gimple_bb (at_stmt)))
upper = false;
/* Update the number of iteration estimates according to the bound.
If at_stmt is an exit then the loop latch is executed at most BOUND times,
otherwise it can be executed BOUND + 1 times. We will lower the estimate
later if such statement must be executed on last iteration */
if (is_exit)
delta = 0;
else
delta = 1;
widest_int new_i_bound = i_bound + delta;
/* If an overflow occurred, ignore the result. */
if (wi::ltu_p (new_i_bound, delta))
return;
if (upper && !is_exit)
do_warn_aggressive_loop_optimizations (loop, new_i_bound, at_stmt);
record_niter_bound (loop, new_i_bound, realistic, upper);
}
/* Records the control iv analyzed in NITER for LOOP if the iv is valid
and doesn't overflow. */
static void
record_control_iv (struct loop *loop, struct tree_niter_desc *niter)
{
struct control_iv *iv;
if (!niter->control.base || !niter->control.step)
return;
if (!integer_onep (niter->assumptions) || !niter->control.no_overflow)
return;
iv = ggc_alloc<control_iv> ();
iv->base = niter->control.base;
iv->step = niter->control.step;
iv->next = loop->control_ivs;
loop->control_ivs = iv;
return;
}
/* This function returns TRUE if below conditions are satisfied:
1) VAR is SSA variable.
2) VAR is an IV:{base, step} in its defining loop.
3) IV doesn't overflow.
4) Both base and step are integer constants.
5) Base is the MIN/MAX value depends on IS_MIN.
Store value of base to INIT correspondingly. */
static bool
get_cst_init_from_scev (tree var, wide_int *init, bool is_min)
{
if (TREE_CODE (var) != SSA_NAME)
return false;
gimple *def_stmt = SSA_NAME_DEF_STMT (var);
struct loop *loop = loop_containing_stmt (def_stmt);
if (loop == NULL)
return false;
affine_iv iv;
if (!simple_iv (loop, loop, var, &iv, false))
return false;
if (!iv.no_overflow)
return false;
if (TREE_CODE (iv.base) != INTEGER_CST || TREE_CODE (iv.step) != INTEGER_CST)
return false;
if (is_min == tree_int_cst_sign_bit (iv.step))
return false;
*init = wi::to_wide (iv.base);
return true;
}
/* Record the estimate on number of iterations of LOOP based on the fact that
the induction variable BASE + STEP * i evaluated in STMT does not wrap and
its values belong to the range <LOW, HIGH>. REALISTIC is true if the
estimated number of iterations is expected to be close to the real one.
UPPER is true if we are sure the induction variable does not wrap. */
static void
record_nonwrapping_iv (struct loop *loop, tree base, tree step, gimple *stmt,
tree low, tree high, bool realistic, bool upper)
{
tree niter_bound, extreme, delta;
tree type = TREE_TYPE (base), unsigned_type;
tree orig_base = base;
if (TREE_CODE (step) != INTEGER_CST || integer_zerop (step))
return;
if (dump_file && (dump_flags & TDF_DETAILS))
{
fprintf (dump_file, "Induction variable (");
print_generic_expr (dump_file, TREE_TYPE (base), TDF_SLIM);
fprintf (dump_file, ") ");
print_generic_expr (dump_file, base, TDF_SLIM);
fprintf (dump_file, " + ");
print_generic_expr (dump_file, step, TDF_SLIM);
fprintf (dump_file, " * iteration does not wrap in statement ");
print_gimple_stmt (dump_file, stmt, 0, TDF_SLIM);
fprintf (dump_file, " in loop %d.\n", loop->num);
}
unsigned_type = unsigned_type_for (type);
base = fold_convert (unsigned_type, base);
step = fold_convert (unsigned_type, step);
if (tree_int_cst_sign_bit (step))
{
wide_int min, max;
extreme = fold_convert (unsigned_type, low);
if (TREE_CODE (orig_base) == SSA_NAME
&& TREE_CODE (high) == INTEGER_CST
&& INTEGRAL_TYPE_P (TREE_TYPE (orig_base))
&& (get_range_info (orig_base, &min, &max) == VR_RANGE
|| get_cst_init_from_scev (orig_base, &max, false))
&& wi::gts_p (wi::to_wide (high), max))
base = wide_int_to_tree (unsigned_type, max);
else if (TREE_CODE (base) != INTEGER_CST
&& dominated_by_p (CDI_DOMINATORS,
loop->latch, gimple_bb (stmt)))
base = fold_convert (unsigned_type, high);
delta = fold_build2 (MINUS_EXPR, unsigned_type, base, extreme);
step = fold_build1 (NEGATE_EXPR, unsigned_type, step);
}
else
{
wide_int min, max;
extreme = fold_convert (unsigned_type, high);
if (TREE_CODE (orig_base) == SSA_NAME
&& TREE_CODE (low) == INTEGER_CST
&& INTEGRAL_TYPE_P (TREE_TYPE (orig_base))
&& (get_range_info (orig_base, &min, &max) == VR_RANGE
|| get_cst_init_from_scev (orig_base, &min, true))
&& wi::gts_p (min, wi::to_wide (low)))
base = wide_int_to_tree (unsigned_type, min);
else if (TREE_CODE (base) != INTEGER_CST
&& dominated_by_p (CDI_DOMINATORS,
loop->latch, gimple_bb (stmt)))
base = fold_convert (unsigned_type, low);
delta = fold_build2 (MINUS_EXPR, unsigned_type, extreme, base);
}
/* STMT is executed at most NITER_BOUND + 1 times, since otherwise the value
would get out of the range. */
niter_bound = fold_build2 (FLOOR_DIV_EXPR, unsigned_type, delta, step);
widest_int max = derive_constant_upper_bound (niter_bound);
record_estimate (loop, niter_bound, max, stmt, false, realistic, upper);
}
/* Determine information about number of iterations a LOOP from the index
IDX of a data reference accessed in STMT. RELIABLE is true if STMT is
guaranteed to be executed in every iteration of LOOP. Callback for
for_each_index. */
struct ilb_data
{
struct loop *loop;
gimple *stmt;
};
static bool
idx_infer_loop_bounds (tree base, tree *idx, void *dta)
{
struct ilb_data *data = (struct ilb_data *) dta;
tree ev, init, step;
tree low, high, type, next;
bool sign, upper = true, at_end = false;
struct loop *loop = data->loop;
if (TREE_CODE (base) != ARRAY_REF)
return true;
/* For arrays at the end of the structure, we are not guaranteed that they
do not really extend over their declared size. However, for arrays of
size greater than one, this is unlikely to be intended. */
if (array_at_struct_end_p (base))
{
at_end = true;
upper = false;
}
struct loop *dloop = loop_containing_stmt (data->stmt);
if (!dloop)
return true;
ev = analyze_scalar_evolution (dloop, *idx);
ev = instantiate_parameters (loop, ev);
init = initial_condition (ev);
step = evolution_part_in_loop_num (ev, loop->num);
if (!init
|| !step
|| TREE_CODE (step) != INTEGER_CST
|| integer_zerop (step)
|| tree_contains_chrecs (init, NULL)
|| chrec_contains_symbols_defined_in_loop (init, loop->num))
return true;
low = array_ref_low_bound (base);
high = array_ref_up_bound (base);
/* The case of nonconstant bounds could be handled, but it would be
complicated. */
if (TREE_CODE (low) != INTEGER_CST
|| !high
|| TREE_CODE (high) != INTEGER_CST)
return true;
sign = tree_int_cst_sign_bit (step);
type = TREE_TYPE (step);
/* The array of length 1 at the end of a structure most likely extends
beyond its bounds. */
if (at_end
&& operand_equal_p (low, high, 0))
return true;
/* In case the relevant bound of the array does not fit in type, or
it does, but bound + step (in type) still belongs into the range of the
array, the index may wrap and still stay within the range of the array
(consider e.g. if the array is indexed by the full range of
unsigned char).
To make things simpler, we require both bounds to fit into type, although
there are cases where this would not be strictly necessary. */
if (!int_fits_type_p (high, type)
|| !int_fits_type_p (low, type))
return true;
low = fold_convert (type, low);
high = fold_convert (type, high);
if (sign)
next = fold_binary (PLUS_EXPR, type, low, step);
else
next = fold_binary (PLUS_EXPR, type, high, step);
if (tree_int_cst_compare (low, next) <= 0
&& tree_int_cst_compare (next, high) <= 0)
return true;
/* If access is not executed on every iteration, we must ensure that overlow
may not make the access valid later. */
if (!dominated_by_p (CDI_DOMINATORS, loop->latch, gimple_bb (data->stmt))
&& scev_probably_wraps_p (NULL_TREE,
initial_condition_in_loop_num (ev, loop->num),
step, data->stmt, loop, true))
upper = false;
record_nonwrapping_iv (loop, init, step, data->stmt, low, high, false, upper);
return true;
}
/* Determine information about number of iterations a LOOP from the bounds
of arrays in the data reference REF accessed in STMT. RELIABLE is true if
STMT is guaranteed to be executed in every iteration of LOOP.*/
static void
infer_loop_bounds_from_ref (struct loop *loop, gimple *stmt, tree ref)
{
struct ilb_data data;
data.loop = loop;
data.stmt = stmt;
for_each_index (&ref, idx_infer_loop_bounds, &data);
}
/* Determine information about number of iterations of a LOOP from the way
arrays are used in STMT. RELIABLE is true if STMT is guaranteed to be
executed in every iteration of LOOP. */
static void
infer_loop_bounds_from_array (struct loop *loop, gimple *stmt)
{
if (is_gimple_assign (stmt))
{
tree op0 = gimple_assign_lhs (stmt);
tree op1 = gimple_assign_rhs1 (stmt);
/* For each memory access, analyze its access function
and record a bound on the loop iteration domain. */
if (REFERENCE_CLASS_P (op0))
infer_loop_bounds_from_ref (loop, stmt, op0);
if (REFERENCE_CLASS_P (op1))
infer_loop_bounds_from_ref (loop, stmt, op1);
}
else if (is_gimple_call (stmt))
{
tree arg, lhs;
unsigned i, n = gimple_call_num_args (stmt);
lhs = gimple_call_lhs (stmt);
if (lhs && REFERENCE_CLASS_P (lhs))
infer_loop_bounds_from_ref (loop, stmt, lhs);
for (i = 0; i < n; i++)
{
arg = gimple_call_arg (stmt, i);
if (REFERENCE_CLASS_P (arg))
infer_loop_bounds_from_ref (loop, stmt, arg);
}
}
}
/* Determine information about number of iterations of a LOOP from the fact
that pointer arithmetics in STMT does not overflow. */
static void
infer_loop_bounds_from_pointer_arith (struct loop *loop, gimple *stmt)
{
tree def, base, step, scev, type, low, high;
tree var, ptr;
if (!is_gimple_assign (stmt)
|| gimple_assign_rhs_code (stmt) != POINTER_PLUS_EXPR)
return;
def = gimple_assign_lhs (stmt);
if (TREE_CODE (def) != SSA_NAME)
return;
type = TREE_TYPE (def);
if (!nowrap_type_p (type))
return;
ptr = gimple_assign_rhs1 (stmt);
if (!expr_invariant_in_loop_p (loop, ptr))
return;
var = gimple_assign_rhs2 (stmt);
if (TYPE_PRECISION (type) != TYPE_PRECISION (TREE_TYPE (var)))
return;
struct loop *uloop = loop_containing_stmt (stmt);
scev = instantiate_parameters (loop, analyze_scalar_evolution (uloop, def));
if (chrec_contains_undetermined (scev))
return;
base = initial_condition_in_loop_num (scev, loop->num);
step = evolution_part_in_loop_num (scev, loop->num);
if (!base || !step
|| TREE_CODE (step) != INTEGER_CST
|| tree_contains_chrecs (base, NULL)
|| chrec_contains_symbols_defined_in_loop (base, loop->num))
return;
low = lower_bound_in_type (type, type);
high = upper_bound_in_type (type, type);
/* In C, pointer arithmetic p + 1 cannot use a NULL pointer, and p - 1 cannot
produce a NULL pointer. The contrary would mean NULL points to an object,
while NULL is supposed to compare unequal with the address of all objects.
Furthermore, p + 1 cannot produce a NULL pointer and p - 1 cannot use a
NULL pointer since that would mean wrapping, which we assume here not to
happen. So, we can exclude NULL from the valid range of pointer
arithmetic. */
if (flag_delete_null_pointer_checks && int_cst_value (low) == 0)
low = build_int_cstu (TREE_TYPE (low), TYPE_ALIGN_UNIT (TREE_TYPE (type)));
record_nonwrapping_iv (loop, base, step, stmt, low, high, false, true);
}
/* Determine information about number of iterations of a LOOP from the fact
that signed arithmetics in STMT does not overflow. */
static void
infer_loop_bounds_from_signedness (struct loop *loop, gimple *stmt)
{
tree def, base, step, scev, type, low, high;
if (gimple_code (stmt) != GIMPLE_ASSIGN)
return;
def = gimple_assign_lhs (stmt);
if (TREE_CODE (def) != SSA_NAME)
return;
type = TREE_TYPE (def);
if (!INTEGRAL_TYPE_P (type)
|| !TYPE_OVERFLOW_UNDEFINED (type))
return;
scev = instantiate_parameters (loop, analyze_scalar_evolution (loop, def));
if (chrec_contains_undetermined (scev))
return;
base = initial_condition_in_loop_num (scev, loop->num);
step = evolution_part_in_loop_num (scev, loop->num);
if (!base || !step
|| TREE_CODE (step) != INTEGER_CST
|| tree_contains_chrecs (base, NULL)
|| chrec_contains_symbols_defined_in_loop (base, loop->num))
return;
low = lower_bound_in_type (type, type);
high = upper_bound_in_type (type, type);
wide_int minv, maxv;
if (get_range_info (def, &minv, &maxv) == VR_RANGE)
{
low = wide_int_to_tree (type, minv);
high = wide_int_to_tree (type, maxv);
}
record_nonwrapping_iv (loop, base, step, stmt, low, high, false, true);
}
/* The following analyzers are extracting informations on the bounds
of LOOP from the following undefined behaviors:
- data references should not access elements over the statically
allocated size,
- signed variables should not overflow when flag_wrapv is not set.
*/
static void
infer_loop_bounds_from_undefined (struct loop *loop)
{
unsigned i;
basic_block *bbs;
gimple_stmt_iterator bsi;
basic_block bb;
bool reliable;
bbs = get_loop_body (loop);
for (i = 0; i < loop->num_nodes; i++)
{
bb = bbs[i];
/* If BB is not executed in each iteration of the loop, we cannot
use the operations in it to infer reliable upper bound on the
# of iterations of the loop. However, we can use it as a guess.
Reliable guesses come only from array bounds. */
reliable = dominated_by_p (CDI_DOMINATORS, loop->latch, bb);
for (bsi = gsi_start_bb (bb); !gsi_end_p (bsi); gsi_next (&bsi))
{
gimple *stmt = gsi_stmt (bsi);
infer_loop_bounds_from_array (loop, stmt);
if (reliable)
{
infer_loop_bounds_from_signedness (loop, stmt);
infer_loop_bounds_from_pointer_arith (loop, stmt);
}
}
}
free (bbs);
}
/* Compare wide ints, callback for qsort. */
static int
wide_int_cmp (const void *p1, const void *p2)
{
const widest_int *d1 = (const widest_int *) p1;
const widest_int *d2 = (const widest_int *) p2;
return wi::cmpu (*d1, *d2);
}
/* Return index of BOUND in BOUNDS array sorted in increasing order.
Lookup by binary search. */
static int
bound_index (vec<widest_int> bounds, const widest_int &bound)
{
unsigned int end = bounds.length ();
unsigned int begin = 0;
/* Find a matching index by means of a binary search. */
while (begin != end)
{
unsigned int middle = (begin + end) / 2;
widest_int index = bounds[middle];
if (index == bound)
return middle;
else if (wi::ltu_p (index, bound))
begin = middle + 1;
else
end = middle;
}
gcc_unreachable ();
}
/* We recorded loop bounds only for statements dominating loop latch (and thus
executed each loop iteration). If there are any bounds on statements not
dominating the loop latch we can improve the estimate by walking the loop
body and seeing if every path from loop header to loop latch contains
some bounded statement. */
static void
discover_iteration_bound_by_body_walk (struct loop *loop)
{
struct nb_iter_bound *elt;
auto_vec<widest_int> bounds;
vec<vec<basic_block> > queues = vNULL;
vec<basic_block> queue = vNULL;
ptrdiff_t queue_index;
ptrdiff_t latch_index = 0;
/* Discover what bounds may interest us. */
for (elt = loop->bounds; elt; elt = elt->next)
{
widest_int bound = elt->bound;
/* Exit terminates loop at given iteration, while non-exits produce undefined
effect on the next iteration. */
if (!elt->is_exit)
{
bound += 1;
/* If an overflow occurred, ignore the result. */
if (bound == 0)
continue;
}
if (!loop->any_upper_bound
|| wi::ltu_p (bound, loop->nb_iterations_upper_bound))
bounds.safe_push (bound);
}
/* Exit early if there is nothing to do. */
if (!bounds.exists ())
return;
if (dump_file && (dump_flags & TDF_DETAILS))
fprintf (dump_file, " Trying to walk loop body to reduce the bound.\n");
/* Sort the bounds in decreasing order. */
bounds.qsort (wide_int_cmp);
/* For every basic block record the lowest bound that is guaranteed to
terminate the loop. */
hash_map<basic_block, ptrdiff_t> bb_bounds;
for (elt = loop->bounds; elt; elt = elt->next)
{
widest_int bound = elt->bound;
if (!elt->is_exit)
{
bound += 1;
/* If an overflow occurred, ignore the result. */
if (bound == 0)
continue;
}
if (!loop->any_upper_bound
|| wi::ltu_p (bound, loop->nb_iterations_upper_bound))
{
ptrdiff_t index = bound_index (bounds, bound);
ptrdiff_t *entry = bb_bounds.get (gimple_bb (elt->stmt));
if (!entry)
bb_bounds.put (gimple_bb (elt->stmt), index);
else if ((ptrdiff_t)*entry > index)
*entry = index;
}
}
hash_map<basic_block, ptrdiff_t> block_priority;
/* Perform shortest path discovery loop->header ... loop->latch.
The "distance" is given by the smallest loop bound of basic block
present in the path and we look for path with largest smallest bound
on it.
To avoid the need for fibonacci heap on double ints we simply compress
double ints into indexes to BOUNDS array and then represent the queue
as arrays of queues for every index.
Index of BOUNDS.length() means that the execution of given BB has
no bounds determined.
VISITED is a pointer map translating basic block into smallest index
it was inserted into the priority queue with. */
latch_index = -1;
/* Start walk in loop header with index set to infinite bound. */
queue_index = bounds.length ();
queues.safe_grow_cleared (queue_index + 1);
queue.safe_push (loop->header);
queues[queue_index] = queue;
block_priority.put (loop->header, queue_index);
for (; queue_index >= 0; queue_index--)
{
if (latch_index < queue_index)
{
while (queues[queue_index].length ())
{
basic_block bb;
ptrdiff_t bound_index = queue_index;
edge e;
edge_iterator ei;
queue = queues[queue_index];
bb = queue.pop ();
/* OK, we later inserted the BB with lower priority, skip it. */
if (*block_priority.get (bb) > queue_index)
continue;
/* See if we can improve the bound. */
ptrdiff_t *entry = bb_bounds.get (bb);
if (entry && *entry < bound_index)
bound_index = *entry;
/* Insert succesors into the queue, watch for latch edge
and record greatest index we saw. */
FOR_EACH_EDGE (e, ei, bb->succs)
{
bool insert = false;
if (loop_exit_edge_p (loop, e))
continue;
if (e == loop_latch_edge (loop)
&& latch_index < bound_index)
latch_index = bound_index;
else if (!(entry = block_priority.get (e->dest)))
{
insert = true;
block_priority.put (e->dest, bound_index);
}
else if (*entry < bound_index)
{
insert = true;
*entry = bound_index;
}
if (insert)
queues[bound_index].safe_push (e->dest);
}
}
}
queues[queue_index].release ();
}
gcc_assert (latch_index >= 0);
if ((unsigned)latch_index < bounds.length ())
{
if (dump_file && (dump_flags & TDF_DETAILS))
{
fprintf (dump_file, "Found better loop bound ");
print_decu (bounds[latch_index], dump_file);
fprintf (dump_file, "\n");
}
record_niter_bound (loop, bounds[latch_index], false, true);
}
queues.release ();
}
/* See if every path cross the loop goes through a statement that is known
to not execute at the last iteration. In that case we can decrese iteration
count by 1. */
static void
maybe_lower_iteration_bound (struct loop *loop)
{
hash_set<gimple *> *not_executed_last_iteration = NULL;
struct nb_iter_bound *elt;
bool found_exit = false;
auto_vec<basic_block> queue;
bitmap visited;
/* Collect all statements with interesting (i.e. lower than
nb_iterations_upper_bound) bound on them.
TODO: Due to the way record_estimate choose estimates to store, the bounds
will be always nb_iterations_upper_bound-1. We can change this to record
also statements not dominating the loop latch and update the walk bellow
to the shortest path algorithm. */
for (elt = loop->bounds; elt; elt = elt->next)
{
if (!elt->is_exit
&& wi::ltu_p (elt->bound, loop->nb_iterations_upper_bound))
{
if (!not_executed_last_iteration)
not_executed_last_iteration = new hash_set<gimple *>;
not_executed_last_iteration->add (elt->stmt);
}
}
if (!not_executed_last_iteration)
return;
/* Start DFS walk in the loop header and see if we can reach the
loop latch or any of the exits (including statements with side
effects that may terminate the loop otherwise) without visiting
any of the statements known to have undefined effect on the last
iteration. */
queue.safe_push (loop->header);
visited = BITMAP_ALLOC (NULL);
bitmap_set_bit (visited, loop->header->index);
found_exit = false;
do
{
basic_block bb = queue.pop ();
gimple_stmt_iterator gsi;
bool stmt_found = false;
/* Loop for possible exits and statements bounding the execution. */
for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi))
{
gimple *stmt = gsi_stmt (gsi);
if (not_executed_last_iteration->contains (stmt))
{
stmt_found = true;
break;
}
if (gimple_has_side_effects (stmt))
{
found_exit = true;
break;
}
}
if (found_exit)
break;
/* If no bounding statement is found, continue the walk. */
if (!stmt_found)
{
edge e;
edge_iterator ei;
FOR_EACH_EDGE (e, ei, bb->succs)
{
if (loop_exit_edge_p (loop, e)
|| e == loop_latch_edge (loop))
{
found_exit = true;
break;
}
if (bitmap_set_bit (visited, e->dest->index))
queue.safe_push (e->dest);
}
}
}
while (queue.length () && !found_exit);
/* If every path through the loop reach bounding statement before exit,
then we know the last iteration of the loop will have undefined effect
and we can decrease number of iterations. */
if (!found_exit)
{
if (dump_file && (dump_flags & TDF_DETAILS))
fprintf (dump_file, "Reducing loop iteration estimate by 1; "
"undefined statement must be executed at the last iteration.\n");
record_niter_bound (loop, loop->nb_iterations_upper_bound - 1,
false, true);
}
BITMAP_FREE (visited);
delete not_executed_last_iteration;
}
/* Records estimates on numbers of iterations of LOOP. If USE_UNDEFINED_P
is true also use estimates derived from undefined behavior. */
void
estimate_numbers_of_iterations (struct loop *loop)
{
vec<edge> exits;
tree niter, type;
unsigned i;
struct tree_niter_desc niter_desc;
edge ex;
widest_int bound;
edge likely_exit;
/* Give up if we already have tried to compute an estimation. */
if (loop->estimate_state != EST_NOT_COMPUTED)
return;
loop->estimate_state = EST_AVAILABLE;
/* If we have a measured profile, use it to estimate the number of
iterations. Normally this is recorded by branch_prob right after
reading the profile. In case we however found a new loop, record the
information here.
Explicitly check for profile status so we do not report
wrong prediction hitrates for guessed loop iterations heuristics.
Do not recompute already recorded bounds - we ought to be better on
updating iteration bounds than updating profile in general and thus
recomputing iteration bounds later in the compilation process will just
introduce random roundoff errors. */
if (!loop->any_estimate
&& loop->header->count.reliable_p ())
{
gcov_type nit = expected_loop_iterations_unbounded (loop);
bound = gcov_type_to_wide_int (nit);
record_niter_bound (loop, bound, true, false);
}
/* Ensure that loop->nb_iterations is computed if possible. If it turns out
to be constant, we avoid undefined behavior implied bounds and instead
diagnose those loops with -Waggressive-loop-optimizations. */
number_of_latch_executions (loop);
exits = get_loop_exit_edges (loop);
likely_exit = single_likely_exit (loop);
FOR_EACH_VEC_ELT (exits, i, ex)
{
if (!number_of_iterations_exit (loop, ex, &niter_desc, false, false))
continue;
niter = niter_desc.niter;
type = TREE_TYPE (niter);
if (TREE_CODE (niter_desc.may_be_zero) != INTEGER_CST)
niter = build3 (COND_EXPR, type, niter_desc.may_be_zero,
build_int_cst (type, 0),
niter);
record_estimate (loop, niter, niter_desc.max,
last_stmt (ex->src),
true, ex == likely_exit, true);
record_control_iv (loop, &niter_desc);
}
exits.release ();
if (flag_aggressive_loop_optimizations)
infer_loop_bounds_from_undefined (loop);
discover_iteration_bound_by_body_walk (loop);
maybe_lower_iteration_bound (loop);
/* If we know the exact number of iterations of this loop, try to
not break code with undefined behavior by not recording smaller
maximum number of iterations. */
if (loop->nb_iterations
&& TREE_CODE (loop->nb_iterations) == INTEGER_CST)
{
loop->any_upper_bound = true;
loop->nb_iterations_upper_bound = wi::to_widest (loop->nb_iterations);
}
}
/* Sets NIT to the estimated number of executions of the latch of the
LOOP. If CONSERVATIVE is true, we must be sure that NIT is at least as
large as the number of iterations. If we have no reliable estimate,
the function returns false, otherwise returns true. */
bool
estimated_loop_iterations (struct loop *loop, widest_int *nit)
{
/* When SCEV information is available, try to update loop iterations
estimate. Otherwise just return whatever we recorded earlier. */
if (scev_initialized_p ())
estimate_numbers_of_iterations (loop);
return (get_estimated_loop_iterations (loop, nit));
}
/* Similar to estimated_loop_iterations, but returns the estimate only
if it fits to HOST_WIDE_INT. If this is not the case, or the estimate
on the number of iterations of LOOP could not be derived, returns -1. */
HOST_WIDE_INT
estimated_loop_iterations_int (struct loop *loop)
{
widest_int nit;
HOST_WIDE_INT hwi_nit;
if (!estimated_loop_iterations (loop, &nit))
return -1;
if (!wi::fits_shwi_p (nit))
return -1;
hwi_nit = nit.to_shwi ();
return hwi_nit < 0 ? -1 : hwi_nit;
}
/* Sets NIT to an upper bound for the maximum number of executions of the
latch of the LOOP. If we have no reliable estimate, the function returns
false, otherwise returns true. */
bool
max_loop_iterations (struct loop *loop, widest_int *nit)
{
/* When SCEV information is available, try to update loop iterations
estimate. Otherwise just return whatever we recorded earlier. */
if (scev_initialized_p ())
estimate_numbers_of_iterations (loop);
return get_max_loop_iterations (loop, nit);
}
/* Similar to max_loop_iterations, but returns the estimate only
if it fits to HOST_WIDE_INT. If this is not the case, or the estimate
on the number of iterations of LOOP could not be derived, returns -1. */
HOST_WIDE_INT
max_loop_iterations_int (struct loop *loop)
{
widest_int nit;
HOST_WIDE_INT hwi_nit;
if (!max_loop_iterations (loop, &nit))
return -1;
if (!wi::fits_shwi_p (nit))
return -1;
hwi_nit = nit.to_shwi ();
return hwi_nit < 0 ? -1 : hwi_nit;
}
/* Sets NIT to an likely upper bound for the maximum number of executions of the
latch of the LOOP. If we have no reliable estimate, the function returns
false, otherwise returns true. */
bool
likely_max_loop_iterations (struct loop *loop, widest_int *nit)
{
/* When SCEV information is available, try to update loop iterations
estimate. Otherwise just return whatever we recorded earlier. */
if (scev_initialized_p ())
estimate_numbers_of_iterations (loop);
return get_likely_max_loop_iterations (loop, nit);
}
/* Similar to max_loop_iterations, but returns the estimate only
if it fits to HOST_WIDE_INT. If this is not the case, or the estimate
on the number of iterations of LOOP could not be derived, returns -1. */
HOST_WIDE_INT
likely_max_loop_iterations_int (struct loop *loop)
{
widest_int nit;
HOST_WIDE_INT hwi_nit;
if (!likely_max_loop_iterations (loop, &nit))
return -1;
if (!wi::fits_shwi_p (nit))
return -1;
hwi_nit = nit.to_shwi ();
return hwi_nit < 0 ? -1 : hwi_nit;
}
/* Returns an estimate for the number of executions of statements
in the LOOP. For statements before the loop exit, this exceeds
the number of execution of the latch by one. */
HOST_WIDE_INT
estimated_stmt_executions_int (struct loop *loop)
{
HOST_WIDE_INT nit = estimated_loop_iterations_int (loop);
HOST_WIDE_INT snit;
if (nit == -1)
return -1;
snit = (HOST_WIDE_INT) ((unsigned HOST_WIDE_INT) nit + 1);
/* If the computation overflows, return -1. */
return snit < 0 ? -1 : snit;
}
/* Sets NIT to the maximum number of executions of the latch of the
LOOP, plus one. If we have no reliable estimate, the function returns
false, otherwise returns true. */
bool
max_stmt_executions (struct loop *loop, widest_int *nit)
{
widest_int nit_minus_one;
if (!max_loop_iterations (loop, nit))
return false;
nit_minus_one = *nit;
*nit += 1;
return wi::gtu_p (*nit, nit_minus_one);
}
/* Sets NIT to the estimated maximum number of executions of the latch of the
LOOP, plus one. If we have no likely estimate, the function returns
false, otherwise returns true. */
bool
likely_max_stmt_executions (struct loop *loop, widest_int *nit)
{
widest_int nit_minus_one;
if (!likely_max_loop_iterations (loop, nit))
return false;
nit_minus_one = *nit;
*nit += 1;
return wi::gtu_p (*nit, nit_minus_one);
}
/* Sets NIT to the estimated number of executions of the latch of the
LOOP, plus one. If we have no reliable estimate, the function returns
false, otherwise returns true. */
bool
estimated_stmt_executions (struct loop *loop, widest_int *nit)
{
widest_int nit_minus_one;
if (!estimated_loop_iterations (loop, nit))
return false;
nit_minus_one = *nit;
*nit += 1;
return wi::gtu_p (*nit, nit_minus_one);
}
/* Records estimates on numbers of iterations of loops. */
void
estimate_numbers_of_iterations (function *fn)
{
struct loop *loop;
/* We don't want to issue signed overflow warnings while getting
loop iteration estimates. */
fold_defer_overflow_warnings ();
FOR_EACH_LOOP_FN (fn, loop, 0)
estimate_numbers_of_iterations (loop);
fold_undefer_and_ignore_overflow_warnings ();
}
/* Returns true if statement S1 dominates statement S2. */
bool
stmt_dominates_stmt_p (gimple *s1, gimple *s2)
{
basic_block bb1 = gimple_bb (s1), bb2 = gimple_bb (s2);
if (!bb1
|| s1 == s2)
return true;
if (bb1 == bb2)
{
gimple_stmt_iterator bsi;
if (gimple_code (s2) == GIMPLE_PHI)
return false;
if (gimple_code (s1) == GIMPLE_PHI)
return true;
for (bsi = gsi_start_bb (bb1); gsi_stmt (bsi) != s2; gsi_next (&bsi))
if (gsi_stmt (bsi) == s1)
return true;
return false;
}
return dominated_by_p (CDI_DOMINATORS, bb2, bb1);
}
/* Returns true when we can prove that the number of executions of
STMT in the loop is at most NITER, according to the bound on
the number of executions of the statement NITER_BOUND->stmt recorded in
NITER_BOUND and fact that NITER_BOUND->stmt dominate STMT.
??? This code can become quite a CPU hog - we can have many bounds,
and large basic block forcing stmt_dominates_stmt_p to be queried
many times on a large basic blocks, so the whole thing is O(n^2)
for scev_probably_wraps_p invocation (that can be done n times).
It would make more sense (and give better answers) to remember BB
bounds computed by discover_iteration_bound_by_body_walk. */
static bool
n_of_executions_at_most (gimple *stmt,
struct nb_iter_bound *niter_bound,
tree niter)
{
widest_int bound = niter_bound->bound;
tree nit_type = TREE_TYPE (niter), e;
enum tree_code cmp;
gcc_assert (TYPE_UNSIGNED (nit_type));
/* If the bound does not even fit into NIT_TYPE, it cannot tell us that
the number of iterations is small. */
if (!wi::fits_to_tree_p (bound, nit_type))
return false;
/* We know that NITER_BOUND->stmt is executed at most NITER_BOUND->bound + 1
times. This means that:
-- if NITER_BOUND->is_exit is true, then everything after
it at most NITER_BOUND->bound times.
-- If NITER_BOUND->is_exit is false, then if we can prove that when STMT
is executed, then NITER_BOUND->stmt is executed as well in the same
iteration then STMT is executed at most NITER_BOUND->bound + 1 times.
If we can determine that NITER_BOUND->stmt is always executed
after STMT, then STMT is executed at most NITER_BOUND->bound + 2 times.
We conclude that if both statements belong to the same
basic block and STMT is before NITER_BOUND->stmt and there are no
statements with side effects in between. */
if (niter_bound->is_exit)
{
if (stmt == niter_bound->stmt
|| !stmt_dominates_stmt_p (niter_bound->stmt, stmt))
return false;
cmp = GE_EXPR;
}
else
{
if (!stmt_dominates_stmt_p (niter_bound->stmt, stmt))
{
gimple_stmt_iterator bsi;
if (gimple_bb (stmt) != gimple_bb (niter_bound->stmt)
|| gimple_code (stmt) == GIMPLE_PHI
|| gimple_code (niter_bound->stmt) == GIMPLE_PHI)
return false;
/* By stmt_dominates_stmt_p we already know that STMT appears
before NITER_BOUND->STMT. Still need to test that the loop
can not be terinated by a side effect in between. */
for (bsi = gsi_for_stmt (stmt); gsi_stmt (bsi) != niter_bound->stmt;
gsi_next (&bsi))
if (gimple_has_side_effects (gsi_stmt (bsi)))
return false;
bound += 1;
if (bound == 0
|| !wi::fits_to_tree_p (bound, nit_type))
return false;
}
cmp = GT_EXPR;
}
e = fold_binary (cmp, boolean_type_node,
niter, wide_int_to_tree (nit_type, bound));
return e && integer_nonzerop (e);
}
/* Returns true if the arithmetics in TYPE can be assumed not to wrap. */
bool
nowrap_type_p (tree type)
{
if (ANY_INTEGRAL_TYPE_P (type)
&& TYPE_OVERFLOW_UNDEFINED (type))
return true;
if (POINTER_TYPE_P (type))
return true;
return false;
}
/* Return true if we can prove LOOP is exited before evolution of induction
variable {BASE, STEP} overflows with respect to its type bound. */
static bool
loop_exits_before_overflow (tree base, tree step,
gimple *at_stmt, struct loop *loop)
{
widest_int niter;
struct control_iv *civ;
struct nb_iter_bound *bound;
tree e, delta, step_abs, unsigned_base;
tree type = TREE_TYPE (step);
tree unsigned_type, valid_niter;
/* Don't issue signed overflow warnings. */
fold_defer_overflow_warnings ();
/* Compute the number of iterations before we reach the bound of the
type, and verify that the loop is exited before this occurs. */
unsigned_type = unsigned_type_for (type);
unsigned_base = fold_convert (unsigned_type, base);
if (tree_int_cst_sign_bit (step))
{
tree extreme = fold_convert (unsigned_type,
lower_bound_in_type (type, type));
delta = fold_build2 (MINUS_EXPR, unsigned_type, unsigned_base, extreme);
step_abs = fold_build1 (NEGATE_EXPR, unsigned_type,
fold_convert (unsigned_type, step));
}
else
{
tree extreme = fold_convert (unsigned_type,
upper_bound_in_type (type, type));
delta = fold_build2 (MINUS_EXPR, unsigned_type, extreme, unsigned_base);
step_abs = fold_convert (unsigned_type, step);
}
valid_niter = fold_build2 (FLOOR_DIV_EXPR, unsigned_type, delta, step_abs);
estimate_numbers_of_iterations (loop);
if (max_loop_iterations (loop, &niter)
&& wi::fits_to_tree_p (niter, TREE_TYPE (valid_niter))
&& (e = fold_binary (GT_EXPR, boolean_type_node, valid_niter,
wide_int_to_tree (TREE_TYPE (valid_niter),
niter))) != NULL
&& integer_nonzerop (e))
{
fold_undefer_and_ignore_overflow_warnings ();
return true;
}
if (at_stmt)
for (bound = loop->bounds; bound; bound = bound->next)
{
if (n_of_executions_at_most (at_stmt, bound, valid_niter))
{
fold_undefer_and_ignore_overflow_warnings ();
return true;
}
}
fold_undefer_and_ignore_overflow_warnings ();
/* Try to prove loop is exited before {base, step} overflows with the
help of analyzed loop control IV. This is done only for IVs with
constant step because otherwise we don't have the information. */
if (TREE_CODE (step) == INTEGER_CST)
{
for (civ = loop->control_ivs; civ; civ = civ->next)
{
enum tree_code code;
tree civ_type = TREE_TYPE (civ->step);
/* Have to consider type difference because operand_equal_p ignores
that for constants. */
if (TYPE_UNSIGNED (type) != TYPE_UNSIGNED (civ_type)
|| element_precision (type) != element_precision (civ_type))
continue;
/* Only consider control IV with same step. */
if (!operand_equal_p (step, civ->step, 0))
continue;
/* Done proving if this is a no-overflow control IV. */
if (operand_equal_p (base, civ->base, 0))
return true;
/* Control IV is recorded after expanding simple operations,
Here we expand base and compare it too. */
tree expanded_base = expand_simple_operations (base);
if (operand_equal_p (expanded_base, civ->base, 0))
return true;
/* If this is a before stepping control IV, in other words, we have
{civ_base, step} = {base + step, step}
Because civ {base + step, step} doesn't overflow during loop
iterations, {base, step} will not overflow if we can prove the
operation "base + step" does not overflow. Specifically, we try
to prove below conditions are satisfied:
base <= UPPER_BOUND (type) - step ;;step > 0
base >= LOWER_BOUND (type) - step ;;step < 0
by proving the reverse conditions are false using loop's initial
condition. */
if (POINTER_TYPE_P (TREE_TYPE (base)))
code = POINTER_PLUS_EXPR;
else
code = PLUS_EXPR;
tree stepped = fold_build2 (code, TREE_TYPE (base), base, step);
tree expanded_stepped = fold_build2 (code, TREE_TYPE (base),
expanded_base, step);
if (operand_equal_p (stepped, civ->base, 0)
|| operand_equal_p (expanded_stepped, civ->base, 0))
{
tree extreme;
if (tree_int_cst_sign_bit (step))
{
code = LT_EXPR;
extreme = lower_bound_in_type (type, type);
}
else
{
code = GT_EXPR;
extreme = upper_bound_in_type (type, type);
}
extreme = fold_build2 (MINUS_EXPR, type, extreme, step);
e = fold_build2 (code, boolean_type_node, base, extreme);
e = simplify_using_initial_conditions (loop, e);
if (integer_zerop (e))
return true;
}
}
}
return false;
}
/* VAR is scev variable whose evolution part is constant STEP, this function
proves that VAR can't overflow by using value range info. If VAR's value
range is [MIN, MAX], it can be proven by:
MAX + step doesn't overflow ; if step > 0
or
MIN + step doesn't underflow ; if step < 0.
We can only do this if var is computed in every loop iteration, i.e, var's
definition has to dominate loop latch. Consider below example:
{
unsigned int i;
<bb 3>:
<bb 4>:
# RANGE [0, 4294967294] NONZERO 65535
# i_21 = PHI <0(3), i_18(9)>
if (i_21 != 0)
goto <bb 6>;
else
goto <bb 8>;
<bb 6>:
# RANGE [0, 65533] NONZERO 65535
_6 = i_21 + 4294967295;
# RANGE [0, 65533] NONZERO 65535
_7 = (long unsigned int) _6;
# RANGE [0, 524264] NONZERO 524280
_8 = _7 * 8;
# PT = nonlocal escaped
_9 = a_14 + _8;
*_9 = 0;
<bb 8>:
# RANGE [1, 65535] NONZERO 65535
i_18 = i_21 + 1;
if (i_18 >= 65535)
goto <bb 10>;
else
goto <bb 9>;
<bb 9>:
goto <bb 4>;
<bb 10>:
return;
}
VAR _6 doesn't overflow only with pre-condition (i_21 != 0), here we
can't use _6 to prove no-overlfow for _7. In fact, var _7 takes value
sequence (4294967295, 0, 1, ..., 65533) in loop life time, rather than
(4294967295, 4294967296, ...). */
static bool
scev_var_range_cant_overflow (tree var, tree step, struct loop *loop)
{
tree type;
wide_int minv, maxv, diff, step_wi;
enum value_range_kind rtype;
if (TREE_CODE (step) != INTEGER_CST || !INTEGRAL_TYPE_P (TREE_TYPE (var)))
return false;
/* Check if VAR evaluates in every loop iteration. It's not the case
if VAR is default definition or does not dominate loop's latch. */
basic_block def_bb = gimple_bb (SSA_NAME_DEF_STMT (var));
if (!def_bb || !dominated_by_p (CDI_DOMINATORS, loop->latch, def_bb))
return false;
rtype = get_range_info (var, &minv, &maxv);
if (rtype != VR_RANGE)
return false;
/* VAR is a scev whose evolution part is STEP and value range info
is [MIN, MAX], we can prove its no-overflowness by conditions:
type_MAX - MAX >= step ; if step > 0
MIN - type_MIN >= |step| ; if step < 0.
Or VAR must take value outside of value range, which is not true. */
step_wi = wi::to_wide (step);
type = TREE_TYPE (var);
if (tree_int_cst_sign_bit (step))
{
diff = minv - wi::to_wide (lower_bound_in_type (type, type));
step_wi = - step_wi;
}
else
diff = wi::to_wide (upper_bound_in_type (type, type)) - maxv;
return (wi::geu_p (diff, step_wi));
}
/* Return false only when the induction variable BASE + STEP * I is
known to not overflow: i.e. when the number of iterations is small
enough with respect to the step and initial condition in order to
keep the evolution confined in TYPEs bounds. Return true when the
iv is known to overflow or when the property is not computable.
USE_OVERFLOW_SEMANTICS is true if this function should assume that
the rules for overflow of the given language apply (e.g., that signed
arithmetics in C does not overflow).
If VAR is a ssa variable, this function also returns false if VAR can
be proven not overflow with value range info. */
bool
scev_probably_wraps_p (tree var, tree base, tree step,
gimple *at_stmt, struct loop *loop,
bool use_overflow_semantics)
{
/* FIXME: We really need something like
http://gcc.gnu.org/ml/gcc-patches/2005-06/msg02025.html.
We used to test for the following situation that frequently appears
during address arithmetics:
D.1621_13 = (long unsigned intD.4) D.1620_12;
D.1622_14 = D.1621_13 * 8;
D.1623_15 = (doubleD.29 *) D.1622_14;
And derived that the sequence corresponding to D_14
can be proved to not wrap because it is used for computing a
memory access; however, this is not really the case -- for example,
if D_12 = (unsigned char) [254,+,1], then D_14 has values
2032, 2040, 0, 8, ..., but the code is still legal. */
if (chrec_contains_undetermined (base)
|| chrec_contains_undetermined (step))
return true;
if (integer_zerop (step))
return false;
/* If we can use the fact that signed and pointer arithmetics does not
wrap, we are done. */
if (use_overflow_semantics && nowrap_type_p (TREE_TYPE (base)))
return false;
/* To be able to use estimates on number of iterations of the loop,
we must have an upper bound on the absolute value of the step. */
if (TREE_CODE (step) != INTEGER_CST)
return true;
/* Check if var can be proven not overflow with value range info. */
if (var && TREE_CODE (var) == SSA_NAME
&& scev_var_range_cant_overflow (var, step, loop))
return false;
if (loop_exits_before_overflow (base, step, at_stmt, loop))
return false;
/* At this point we still don't have a proof that the iv does not
overflow: give up. */
return true;
}
/* Frees the information on upper bounds on numbers of iterations of LOOP. */
void
free_numbers_of_iterations_estimates (struct loop *loop)
{
struct control_iv *civ;
struct nb_iter_bound *bound;
loop->nb_iterations = NULL;
loop->estimate_state = EST_NOT_COMPUTED;
for (bound = loop->bounds; bound;)
{
struct nb_iter_bound *next = bound->next;
ggc_free (bound);
bound = next;
}
loop->bounds = NULL;
for (civ = loop->control_ivs; civ;)
{
struct control_iv *next = civ->next;
ggc_free (civ);
civ = next;
}
loop->control_ivs = NULL;
}
/* Frees the information on upper bounds on numbers of iterations of loops. */
void
free_numbers_of_iterations_estimates (function *fn)
{
struct loop *loop;
FOR_EACH_LOOP_FN (fn, loop, 0)
free_numbers_of_iterations_estimates (loop);
}
/* Substitute value VAL for ssa name NAME inside expressions held
at LOOP. */
void
substitute_in_loop_info (struct loop *loop, tree name, tree val)
{
loop->nb_iterations = simplify_replace_tree (loop->nb_iterations, name, val);
}
|