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
|
/* $NetBSD: bpf.c,v 1.251 2023/02/08 01:37:53 gutteridge Exp $ */
/*
* Copyright (c) 1990, 1991, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from the Stanford/CMU enet packet filter,
* (net/enet.c) distributed as part of 4.3BSD, and code contributed
* to Berkeley by Steven McCanne and Van Jacobson both of Lawrence
* Berkeley Laboratory.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)bpf.c 8.4 (Berkeley) 1/9/95
* static char rcsid[] =
* "Header: bpf.c,v 1.67 96/09/26 22:00:52 leres Exp ";
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: bpf.c,v 1.251 2023/02/08 01:37:53 gutteridge Exp $");
#if defined(_KERNEL_OPT)
#include "opt_bpf.h"
#include "sl.h"
#include "opt_net_mpsafe.h"
#endif
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/mbuf.h>
#include <sys/buf.h>
#include <sys/time.h>
#include <sys/proc.h>
#include <sys/ioctl.h>
#include <sys/conf.h>
#include <sys/vnode.h>
#include <sys/queue.h>
#include <sys/stat.h>
#include <sys/module.h>
#include <sys/atomic.h>
#include <sys/cpu.h>
#include <sys/file.h>
#include <sys/filedesc.h>
#include <sys/tty.h>
#include <sys/uio.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/errno.h>
#include <sys/kernel.h>
#include <sys/poll.h>
#include <sys/sysctl.h>
#include <sys/kauth.h>
#include <sys/syslog.h>
#include <sys/percpu.h>
#include <sys/pserialize.h>
#include <sys/lwp.h>
#include <sys/xcall.h>
#include <net/if.h>
#include <net/slip.h>
#include <net/bpf.h>
#include <net/bpfdesc.h>
#include <net/bpfjit.h>
#include <net/if_arc.h>
#include <net/if_ether.h>
#include <net/if_types.h>
#include <netinet/in.h>
#include <netinet/if_inarp.h>
#include <compat/sys/sockio.h>
#ifndef BPF_BUFSIZE
/*
* 4096 is too small for FDDI frames. 8192 is too small for gigabit Ethernet
* jumbos (circa 9k), ATM, or Intel gig/10gig ethernet jumbos (16k).
*/
# define BPF_BUFSIZE 32768
#endif
#define PRINET 26 /* interruptible */
/*
* The default read buffer size, and limit for BIOCSBLEN, is sysctl'able.
* XXX the default values should be computed dynamically based
* on available memory size and available mbuf clusters.
*/
static int bpf_bufsize = BPF_BUFSIZE;
static int bpf_maxbufsize = BPF_DFLTBUFSIZE; /* XXX set dynamically, see above */
static bool bpf_jit = false;
struct bpfjit_ops bpfjit_module_ops = {
.bj_generate_code = NULL,
.bj_free_code = NULL
};
/*
* Global BPF statistics returned by net.bpf.stats sysctl.
*/
static struct percpu *bpf_gstats_percpu; /* struct bpf_stat */
#define BPF_STATINC(id) \
{ \
struct bpf_stat *__stats = \
percpu_getref(bpf_gstats_percpu); \
__stats->bs_##id++; \
percpu_putref(bpf_gstats_percpu); \
}
/*
* Locking notes:
* - bpf_mtx (adaptive mutex) protects:
* - Gobal lists: bpf_iflist and bpf_dlist
* - struct bpf_if
* - bpf_close
* - bpf_psz (pserialize)
* - struct bpf_d has two mutexes:
* - bd_buf_mtx (spin mutex) protects the buffers that can be accessed
* on packet tapping
* - bd_mtx (adaptive mutex) protects member variables other than the buffers
* - Locking order: bpf_mtx => bpf_d#bd_mtx => bpf_d#bd_buf_mtx
* - struct bpf_d obtained via fp->f_bpf in bpf_read and bpf_write is
* never freed because struct bpf_d is only freed in bpf_close and
* bpf_close never be called while executing bpf_read and bpf_write
* - A filter that is assigned to bpf_d can be replaced with another filter
* while tapping packets, so it needs to be done atomically
* - struct bpf_d is iterated on bpf_dlist with psz
* - struct bpf_if is iterated on bpf_iflist with psz or psref
*/
/*
* Use a mutex to avoid a race condition between gathering the stats/peers
* and opening/closing the device.
*/
static kmutex_t bpf_mtx;
static struct psref_class *bpf_psref_class __read_mostly;
static pserialize_t bpf_psz;
static inline void
bpf_if_acquire(struct bpf_if *bp, struct psref *psref)
{
psref_acquire(psref, &bp->bif_psref, bpf_psref_class);
}
static inline void
bpf_if_release(struct bpf_if *bp, struct psref *psref)
{
psref_release(psref, &bp->bif_psref, bpf_psref_class);
}
/*
* bpf_iflist is the list of interfaces; each corresponds to an ifnet
* bpf_dtab holds the descriptors, indexed by minor device #
*/
static struct pslist_head bpf_iflist;
static struct pslist_head bpf_dlist;
/* Macros for bpf_d on bpf_dlist */
#define BPF_DLIST_WRITER_INSERT_HEAD(__d) \
PSLIST_WRITER_INSERT_HEAD(&bpf_dlist, (__d), bd_bpf_dlist_entry)
#define BPF_DLIST_READER_FOREACH(__d) \
PSLIST_READER_FOREACH((__d), &bpf_dlist, struct bpf_d, \
bd_bpf_dlist_entry)
#define BPF_DLIST_WRITER_FOREACH(__d) \
PSLIST_WRITER_FOREACH((__d), &bpf_dlist, struct bpf_d, \
bd_bpf_dlist_entry)
#define BPF_DLIST_ENTRY_INIT(__d) \
PSLIST_ENTRY_INIT((__d), bd_bpf_dlist_entry)
#define BPF_DLIST_WRITER_REMOVE(__d) \
PSLIST_WRITER_REMOVE((__d), bd_bpf_dlist_entry)
#define BPF_DLIST_ENTRY_DESTROY(__d) \
PSLIST_ENTRY_DESTROY((__d), bd_bpf_dlist_entry)
/* Macros for bpf_if on bpf_iflist */
#define BPF_IFLIST_WRITER_INSERT_HEAD(__bp) \
PSLIST_WRITER_INSERT_HEAD(&bpf_iflist, (__bp), bif_iflist_entry)
#define BPF_IFLIST_READER_FOREACH(__bp) \
PSLIST_READER_FOREACH((__bp), &bpf_iflist, struct bpf_if, \
bif_iflist_entry)
#define BPF_IFLIST_WRITER_FOREACH(__bp) \
PSLIST_WRITER_FOREACH((__bp), &bpf_iflist, struct bpf_if, \
bif_iflist_entry)
#define BPF_IFLIST_WRITER_REMOVE(__bp) \
PSLIST_WRITER_REMOVE((__bp), bif_iflist_entry)
#define BPF_IFLIST_ENTRY_INIT(__bp) \
PSLIST_ENTRY_INIT((__bp), bif_iflist_entry)
#define BPF_IFLIST_ENTRY_DESTROY(__bp) \
PSLIST_ENTRY_DESTROY((__bp), bif_iflist_entry)
/* Macros for bpf_d on bpf_if#bif_dlist_pslist */
#define BPFIF_DLIST_READER_FOREACH(__d, __bp) \
PSLIST_READER_FOREACH((__d), &(__bp)->bif_dlist_head, struct bpf_d, \
bd_bif_dlist_entry)
#define BPFIF_DLIST_WRITER_INSERT_HEAD(__bp, __d) \
PSLIST_WRITER_INSERT_HEAD(&(__bp)->bif_dlist_head, (__d), \
bd_bif_dlist_entry)
#define BPFIF_DLIST_WRITER_REMOVE(__d) \
PSLIST_WRITER_REMOVE((__d), bd_bif_dlist_entry)
#define BPFIF_DLIST_ENTRY_INIT(__d) \
PSLIST_ENTRY_INIT((__d), bd_bif_dlist_entry)
#define BPFIF_DLIST_READER_EMPTY(__bp) \
(PSLIST_READER_FIRST(&(__bp)->bif_dlist_head, struct bpf_d, \
bd_bif_dlist_entry) == NULL)
#define BPFIF_DLIST_WRITER_EMPTY(__bp) \
(PSLIST_WRITER_FIRST(&(__bp)->bif_dlist_head, struct bpf_d, \
bd_bif_dlist_entry) == NULL)
#define BPFIF_DLIST_ENTRY_DESTROY(__d) \
PSLIST_ENTRY_DESTROY((__d), bd_bif_dlist_entry)
static int bpf_allocbufs(struct bpf_d *);
static u_int bpf_xfilter(struct bpf_filter **, void *, u_int, u_int);
static void bpf_deliver(struct bpf_if *,
void *(*cpfn)(void *, const void *, size_t),
void *, u_int, u_int, const u_int);
static void bpf_freed(struct bpf_d *);
static void bpf_free_filter(struct bpf_filter *);
static void bpf_ifname(struct ifnet *, struct ifreq *);
static void *bpf_mcpy(void *, const void *, size_t);
static int bpf_movein(struct ifnet *, struct uio *, int, uint64_t,
struct mbuf **, struct sockaddr *,
struct bpf_filter **);
static void bpf_attachd(struct bpf_d *, struct bpf_if *);
static void bpf_detachd(struct bpf_d *);
static int bpf_setif(struct bpf_d *, struct ifreq *);
static int bpf_setf(struct bpf_d *, struct bpf_program *, u_long);
static void bpf_timed_out(void *);
static inline void
bpf_wakeup(struct bpf_d *);
static int bpf_hdrlen(struct bpf_d *);
static void catchpacket(struct bpf_d *, u_char *, u_int, u_int,
void *(*)(void *, const void *, size_t), struct timespec *);
static void reset_d(struct bpf_d *);
static int bpf_getdltlist(struct bpf_d *, struct bpf_dltlist *);
static int bpf_setdlt(struct bpf_d *, u_int);
static int bpf_read(struct file *, off_t *, struct uio *, kauth_cred_t,
int);
static int bpf_write(struct file *, off_t *, struct uio *, kauth_cred_t,
int);
static int bpf_ioctl(struct file *, u_long, void *);
static int bpf_poll(struct file *, int);
static int bpf_stat(struct file *, struct stat *);
static int bpf_close(struct file *);
static int bpf_kqfilter(struct file *, struct knote *);
static const struct fileops bpf_fileops = {
.fo_name = "bpf",
.fo_read = bpf_read,
.fo_write = bpf_write,
.fo_ioctl = bpf_ioctl,
.fo_fcntl = fnullop_fcntl,
.fo_poll = bpf_poll,
.fo_stat = bpf_stat,
.fo_close = bpf_close,
.fo_kqfilter = bpf_kqfilter,
.fo_restart = fnullop_restart,
};
dev_type_open(bpfopen);
const struct cdevsw bpf_cdevsw = {
.d_open = bpfopen,
.d_close = noclose,
.d_read = noread,
.d_write = nowrite,
.d_ioctl = noioctl,
.d_stop = nostop,
.d_tty = notty,
.d_poll = nopoll,
.d_mmap = nommap,
.d_kqfilter = nokqfilter,
.d_discard = nodiscard,
.d_flag = D_OTHER | D_MPSAFE
};
bpfjit_func_t
bpf_jit_generate(bpf_ctx_t *bc, void *code, size_t size)
{
struct bpfjit_ops *ops = &bpfjit_module_ops;
bpfjit_func_t (*generate_code)(const bpf_ctx_t *,
const struct bpf_insn *, size_t);
generate_code = atomic_load_acquire(&ops->bj_generate_code);
if (generate_code != NULL) {
return generate_code(bc, code, size);
}
return NULL;
}
void
bpf_jit_freecode(bpfjit_func_t jcode)
{
KASSERT(bpfjit_module_ops.bj_free_code != NULL);
bpfjit_module_ops.bj_free_code(jcode);
}
static int
bpf_movein(struct ifnet *ifp, struct uio *uio, int linktype, uint64_t mtu, struct mbuf **mp,
struct sockaddr *sockp, struct bpf_filter **wfilter)
{
struct mbuf *m, *m0, *n;
int error;
size_t len;
size_t hlen;
size_t align;
u_int slen;
/*
* Build a sockaddr based on the data link layer type.
* We do this at this level because the ethernet header
* is copied directly into the data field of the sockaddr.
* In the case of SLIP, there is no header and the packet
* is forwarded as is.
* Also, we are careful to leave room at the front of the mbuf
* for the link level header.
*/
switch (linktype) {
case DLT_SLIP:
sockp->sa_family = AF_INET;
hlen = 0;
align = 0;
break;
case DLT_PPP:
sockp->sa_family = AF_UNSPEC;
hlen = 0;
align = 0;
break;
case DLT_EN10MB:
sockp->sa_family = AF_UNSPEC;
/* XXX Would MAXLINKHDR be better? */
/* 6(dst)+6(src)+2(type) */
hlen = sizeof(struct ether_header);
align = 2;
break;
case DLT_ARCNET:
sockp->sa_family = AF_UNSPEC;
hlen = ARC_HDRLEN;
align = 5;
break;
case DLT_FDDI:
sockp->sa_family = AF_LINK;
/* XXX 4(FORMAC)+6(dst)+6(src) */
hlen = 16;
align = 0;
break;
case DLT_ECONET:
sockp->sa_family = AF_UNSPEC;
hlen = 6;
align = 2;
break;
case DLT_NULL:
sockp->sa_family = AF_UNSPEC;
if (ifp->if_type == IFT_LOOP) {
/* Set here to apply the following validations */
hlen = sizeof(uint32_t);
} else
hlen = 0;
align = 0;
break;
default:
return (EIO);
}
len = uio->uio_resid;
/*
* If there aren't enough bytes for a link level header or the
* packet length exceeds the interface mtu, return an error.
*/
if (len - hlen > mtu)
return (EMSGSIZE);
m0 = m = m_gethdr(M_WAIT, MT_DATA);
m_reset_rcvif(m);
m->m_pkthdr.len = (int)(len - hlen);
if (len + align > MHLEN) {
m_clget(m, M_WAIT);
if ((m->m_flags & M_EXT) == 0) {
error = ENOBUFS;
goto bad;
}
}
/* Ensure the data is properly aligned */
if (align > 0)
m->m_data += align;
for (;;) {
len = M_TRAILINGSPACE(m);
if (len > uio->uio_resid)
len = uio->uio_resid;
error = uiomove(mtod(m, void *), len, uio);
if (error)
goto bad;
m->m_len = len;
if (uio->uio_resid == 0)
break;
n = m_get(M_WAIT, MT_DATA);
m_clget(n, M_WAIT); /* if fails, there is no problem */
m->m_next = n;
m = n;
}
slen = bpf_xfilter(wfilter, mtod(m, u_char *), len, len);
if (slen == 0) {
error = EPERM;
goto bad;
}
if (hlen != 0) {
if (linktype == DLT_NULL && ifp->if_type == IFT_LOOP) {
uint32_t af;
/* the link header indicates the address family */
memcpy(&af, mtod(m0, void *), sizeof(af));
sockp->sa_family = af;
} else {
/* move link level header in the top of mbuf to sa_data */
memcpy(sockp->sa_data, mtod(m0, void *), hlen);
}
m0->m_data += hlen;
m0->m_len -= hlen;
}
*mp = m0;
return (0);
bad:
m_freem(m0);
return (error);
}
/*
* Attach file to the bpf interface, i.e. make d listen on bp.
*/
static void
bpf_attachd(struct bpf_d *d, struct bpf_if *bp)
{
struct bpf_event_tracker *t;
KASSERT(mutex_owned(&bpf_mtx));
KASSERT(mutex_owned(d->bd_mtx));
/*
* Point d at bp, and add d to the interface's list of listeners.
* Finally, point the driver's bpf cookie at the interface so
* it will divert packets to bpf.
*/
d->bd_bif = bp;
BPFIF_DLIST_WRITER_INSERT_HEAD(bp, d);
*bp->bif_driverp = bp;
SLIST_FOREACH(t, &bp->bif_trackers, bet_entries) {
t->bet_notify(bp, bp->bif_ifp, bp->bif_dlt,
BPF_TRACK_EVENT_ATTACH);
}
}
/*
* Detach a file from its interface.
*/
static void
bpf_detachd(struct bpf_d *d)
{
struct bpf_if *bp;
struct bpf_event_tracker *t;
KASSERT(mutex_owned(&bpf_mtx));
KASSERT(mutex_owned(d->bd_mtx));
bp = d->bd_bif;
/*
* Check if this descriptor had requested promiscuous mode.
* If so, turn it off.
*/
if (d->bd_promisc) {
int error __diagused;
d->bd_promisc = 0;
/*
* Take device out of promiscuous mode. Since we were
* able to enter promiscuous mode, we should be able
* to turn it off. But we can get an error if
* the interface was configured down, so only panic
* if we don't get an unexpected error.
*/
KERNEL_LOCK_UNLESS_NET_MPSAFE();
error = ifpromisc(bp->bif_ifp, 0);
KERNEL_UNLOCK_UNLESS_NET_MPSAFE();
#ifdef DIAGNOSTIC
if (error)
printf("%s: ifpromisc failed: %d", __func__, error);
#endif
}
/* Remove d from the interface's descriptor list. */
BPFIF_DLIST_WRITER_REMOVE(d);
pserialize_perform(bpf_psz);
if (BPFIF_DLIST_WRITER_EMPTY(bp)) {
/*
* Let the driver know that there are no more listeners.
*/
*d->bd_bif->bif_driverp = NULL;
}
d->bd_bif = NULL;
SLIST_FOREACH(t, &bp->bif_trackers, bet_entries) {
t->bet_notify(bp, bp->bif_ifp, bp->bif_dlt,
BPF_TRACK_EVENT_DETACH);
}
}
static void
bpf_init(void)
{
mutex_init(&bpf_mtx, MUTEX_DEFAULT, IPL_NONE);
bpf_psz = pserialize_create();
bpf_psref_class = psref_class_create("bpf", IPL_SOFTNET);
PSLIST_INIT(&bpf_iflist);
PSLIST_INIT(&bpf_dlist);
bpf_gstats_percpu = percpu_alloc(sizeof(struct bpf_stat));
return;
}
/*
* bpfilterattach() is called at boot time. We don't need to do anything
* here, since any initialization will happen as part of module init code.
*/
/* ARGSUSED */
void
bpfilterattach(int n)
{
}
/*
* Open ethernet device. Clones.
*/
/* ARGSUSED */
int
bpfopen(dev_t dev, int flag, int mode, struct lwp *l)
{
struct bpf_d *d;
struct file *fp;
int error, fd;
/* falloc() will fill in the descriptor for us. */
if ((error = fd_allocfile(&fp, &fd)) != 0)
return error;
d = kmem_zalloc(sizeof(*d), KM_SLEEP);
d->bd_bufsize = bpf_bufsize;
d->bd_direction = BPF_D_INOUT;
d->bd_feedback = 0;
d->bd_pid = l->l_proc->p_pid;
#ifdef _LP64
if (curproc->p_flag & PK_32)
d->bd_compat32 = 1;
#endif
getnanotime(&d->bd_btime);
d->bd_atime = d->bd_mtime = d->bd_btime;
callout_init(&d->bd_callout, CALLOUT_MPSAFE);
selinit(&d->bd_sel);
d->bd_jitcode = NULL;
d->bd_rfilter = NULL;
d->bd_wfilter = NULL;
d->bd_locked = 0;
BPF_DLIST_ENTRY_INIT(d);
BPFIF_DLIST_ENTRY_INIT(d);
d->bd_mtx = mutex_obj_alloc(MUTEX_DEFAULT, IPL_SOFTNET);
d->bd_buf_mtx = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NET);
cv_init(&d->bd_cv, "bpf");
mutex_enter(&bpf_mtx);
BPF_DLIST_WRITER_INSERT_HEAD(d);
mutex_exit(&bpf_mtx);
return fd_clone(fp, fd, flag, &bpf_fileops, d);
}
/*
* Close the descriptor by detaching it from its interface,
* deallocating its buffers, and marking it free.
*/
/* ARGSUSED */
static int
bpf_close(struct file *fp)
{
struct bpf_d *d;
mutex_enter(&bpf_mtx);
if ((d = fp->f_bpf) == NULL) {
mutex_exit(&bpf_mtx);
return 0;
}
/*
* Refresh the PID associated with this bpf file.
*/
d->bd_pid = curproc->p_pid;
mutex_enter(d->bd_mtx);
if (d->bd_state == BPF_WAITING)
callout_halt(&d->bd_callout, d->bd_mtx);
d->bd_state = BPF_IDLE;
if (d->bd_bif)
bpf_detachd(d);
mutex_exit(d->bd_mtx);
BPF_DLIST_WRITER_REMOVE(d);
pserialize_perform(bpf_psz);
mutex_exit(&bpf_mtx);
BPFIF_DLIST_ENTRY_DESTROY(d);
BPF_DLIST_ENTRY_DESTROY(d);
fp->f_bpf = NULL;
bpf_freed(d);
callout_destroy(&d->bd_callout);
seldestroy(&d->bd_sel);
mutex_obj_free(d->bd_mtx);
mutex_obj_free(d->bd_buf_mtx);
cv_destroy(&d->bd_cv);
kmem_free(d, sizeof(*d));
return (0);
}
/*
* Rotate the packet buffers in descriptor d. Move the store buffer
* into the hold slot, and the free buffer into the store slot.
* Zero the length of the new store buffer.
*/
#define ROTATE_BUFFERS(d) \
(d)->bd_hbuf = (d)->bd_sbuf; \
(d)->bd_hlen = (d)->bd_slen; \
(d)->bd_sbuf = (d)->bd_fbuf; \
(d)->bd_slen = 0; \
(d)->bd_fbuf = NULL;
/*
* bpfread - read next chunk of packets from buffers
*/
static int
bpf_read(struct file *fp, off_t *offp, struct uio *uio,
kauth_cred_t cred, int flags)
{
struct bpf_d *d = fp->f_bpf;
int timed_out;
int error;
/*
* Refresh the PID associated with this bpf file.
*/
d->bd_pid = curproc->p_pid;
getnanotime(&d->bd_atime);
/*
* Restrict application to use a buffer the same size as
* the kernel buffers.
*/
if (uio->uio_resid != d->bd_bufsize)
return (EINVAL);
mutex_enter(d->bd_mtx);
if (d->bd_state == BPF_WAITING)
callout_halt(&d->bd_callout, d->bd_mtx);
timed_out = (d->bd_state == BPF_TIMED_OUT);
d->bd_state = BPF_IDLE;
mutex_exit(d->bd_mtx);
/*
* If the hold buffer is empty, then do a timed sleep, which
* ends when the timeout expires or when enough packets
* have arrived to fill the store buffer.
*/
mutex_enter(d->bd_buf_mtx);
while (d->bd_hbuf == NULL) {
if (fp->f_flag & FNONBLOCK) {
if (d->bd_slen == 0) {
error = EWOULDBLOCK;
goto out;
}
ROTATE_BUFFERS(d);
break;
}
if ((d->bd_immediate || timed_out) && d->bd_slen != 0) {
/*
* A packet(s) either arrived since the previous
* read or arrived while we were asleep.
* Rotate the buffers and return what's here.
*/
ROTATE_BUFFERS(d);
break;
}
error = cv_timedwait_sig(&d->bd_cv, d->bd_buf_mtx, d->bd_rtout);
if (error == EINTR || error == ERESTART)
goto out;
if (error == EWOULDBLOCK) {
/*
* On a timeout, return what's in the buffer,
* which may be nothing. If there is something
* in the store buffer, we can rotate the buffers.
*/
if (d->bd_hbuf)
/*
* We filled up the buffer in between
* getting the timeout and arriving
* here, so we don't need to rotate.
*/
break;
if (d->bd_slen == 0) {
error = 0;
goto out;
}
ROTATE_BUFFERS(d);
break;
}
if (error != 0)
goto out;
}
/*
* At this point, we know we have something in the hold slot.
*/
mutex_exit(d->bd_buf_mtx);
/*
* Move data from hold buffer into user space.
* We know the entire buffer is transferred since
* we checked above that the read buffer is bpf_bufsize bytes.
*/
error = uiomove(d->bd_hbuf, d->bd_hlen, uio);
mutex_enter(d->bd_buf_mtx);
d->bd_fbuf = d->bd_hbuf;
d->bd_hbuf = NULL;
d->bd_hlen = 0;
out:
mutex_exit(d->bd_buf_mtx);
return (error);
}
/*
* If there are processes sleeping on this descriptor, wake them up.
*/
static inline void
bpf_wakeup(struct bpf_d *d)
{
mutex_enter(d->bd_buf_mtx);
cv_broadcast(&d->bd_cv);
mutex_exit(d->bd_buf_mtx);
if (d->bd_async)
fownsignal(d->bd_pgid, SIGIO, 0, 0, NULL);
selnotify(&d->bd_sel, 0, 0);
}
static void
bpf_timed_out(void *arg)
{
struct bpf_d *d = arg;
mutex_enter(d->bd_mtx);
if (d->bd_state == BPF_WAITING) {
d->bd_state = BPF_TIMED_OUT;
if (d->bd_slen != 0)
bpf_wakeup(d);
}
mutex_exit(d->bd_mtx);
}
static int
bpf_write(struct file *fp, off_t *offp, struct uio *uio,
kauth_cred_t cred, int flags)
{
struct bpf_d *d = fp->f_bpf;
struct bpf_if *bp;
struct ifnet *ifp;
struct mbuf *m, *mc;
int error;
static struct sockaddr_storage dst;
struct psref psref;
int bound;
/*
* Refresh the PID associated with this bpf file.
*/
d->bd_pid = curproc->p_pid;
m = NULL; /* XXX gcc */
bound = curlwp_bind();
mutex_enter(d->bd_mtx);
bp = d->bd_bif;
if (bp == NULL) {
mutex_exit(d->bd_mtx);
error = ENXIO;
goto out_bindx;
}
bpf_if_acquire(bp, &psref);
mutex_exit(d->bd_mtx);
getnanotime(&d->bd_mtime);
ifp = bp->bif_ifp;
if (if_is_deactivated(ifp)) {
error = ENXIO;
goto out;
}
if (uio->uio_resid == 0) {
error = 0;
goto out;
}
error = bpf_movein(ifp, uio, (int)bp->bif_dlt, ifp->if_mtu, &m,
(struct sockaddr *) &dst, &d->bd_wfilter);
if (error)
goto out;
if (m->m_pkthdr.len > ifp->if_mtu) {
m_freem(m);
error = EMSGSIZE;
goto out;
}
/*
* If writing to a loopback interface, the address family has
* already been specially computed in bpf_movein(), so don't
* clobber it, or the loopback will reject it in looutput().
*/
if (d->bd_hdrcmplt && ifp->if_type != IFT_LOOP)
dst.ss_family = pseudo_AF_HDRCMPLT;
if (d->bd_feedback) {
mc = m_dup(m, 0, M_COPYALL, M_NOWAIT);
if (mc != NULL)
m_set_rcvif(mc, ifp);
/* Set M_PROMISC for outgoing packets to be discarded. */
if (1 /*d->bd_direction == BPF_D_INOUT*/)
m->m_flags |= M_PROMISC;
} else
mc = NULL;
error = if_output_lock(ifp, ifp, m, (struct sockaddr *) &dst, NULL);
if (mc != NULL) {
if (error == 0) {
int s = splsoftnet();
KERNEL_LOCK_UNLESS_IFP_MPSAFE(ifp);
ifp->_if_input(ifp, mc);
KERNEL_UNLOCK_UNLESS_IFP_MPSAFE(ifp);
splx(s);
} else
m_freem(mc);
}
/*
* The driver frees the mbuf.
*/
out:
bpf_if_release(bp, &psref);
out_bindx:
curlwp_bindx(bound);
return error;
}
/*
* Reset a descriptor by flushing its packet buffer and clearing the
* receive and drop counts.
*/
static void
reset_d(struct bpf_d *d)
{
KASSERT(mutex_owned(d->bd_mtx));
mutex_enter(d->bd_buf_mtx);
if (d->bd_hbuf) {
/* Free the hold buffer. */
d->bd_fbuf = d->bd_hbuf;
d->bd_hbuf = NULL;
}
d->bd_slen = 0;
d->bd_hlen = 0;
d->bd_rcount = 0;
d->bd_dcount = 0;
d->bd_ccount = 0;
mutex_exit(d->bd_buf_mtx);
}
/*
* FIONREAD Check for read packet available.
* BIOCGBLEN Get buffer len [for read()].
* BIOCSETF Set ethernet read filter.
* BIOCFLUSH Flush read packet buffer.
* BIOCPROMISC Put interface into promiscuous mode.
* BIOCGDLT Get link layer type.
* BIOCGETIF Get interface name.
* BIOCSETIF Set interface.
* BIOCSRTIMEOUT Set read timeout.
* BIOCGRTIMEOUT Get read timeout.
* BIOCGSTATS Get packet stats.
* BIOCIMMEDIATE Set immediate mode.
* BIOCVERSION Get filter language version.
* BIOCGHDRCMPLT Get "header already complete" flag.
* BIOCSHDRCMPLT Set "header already complete" flag.
* BIOCSFEEDBACK Set packet feedback mode.
* BIOCGFEEDBACK Get packet feedback mode.
* BIOCGDIRECTION Get packet direction flag
* BIOCSDIRECTION Set packet direction flag
*/
/* ARGSUSED */
static int
bpf_ioctl(struct file *fp, u_long cmd, void *addr)
{
struct bpf_d *d = fp->f_bpf;
int error = 0;
/*
* Refresh the PID associated with this bpf file.
*/
d->bd_pid = curproc->p_pid;
#ifdef _LP64
if (curproc->p_flag & PK_32)
d->bd_compat32 = 1;
else
d->bd_compat32 = 0;
#endif
mutex_enter(d->bd_mtx);
if (d->bd_state == BPF_WAITING)
callout_halt(&d->bd_callout, d->bd_mtx);
d->bd_state = BPF_IDLE;
mutex_exit(d->bd_mtx);
if (d->bd_locked) {
switch (cmd) {
case BIOCGBLEN: /* FALLTHROUGH */
case BIOCFLUSH: /* FALLTHROUGH */
case BIOCGDLT: /* FALLTHROUGH */
case BIOCGDLTLIST: /* FALLTHROUGH */
case BIOCGETIF: /* FALLTHROUGH */
case BIOCGRTIMEOUT: /* FALLTHROUGH */
case BIOCGSTATS: /* FALLTHROUGH */
case BIOCVERSION: /* FALLTHROUGH */
case BIOCGHDRCMPLT: /* FALLTHROUGH */
case FIONREAD: /* FALLTHROUGH */
case BIOCLOCK: /* FALLTHROUGH */
case BIOCSRTIMEOUT: /* FALLTHROUGH */
case BIOCIMMEDIATE: /* FALLTHROUGH */
case TIOCGPGRP:
break;
default:
return EPERM;
}
}
switch (cmd) {
default:
error = EINVAL;
break;
/*
* Check for read packet available.
*/
case FIONREAD:
{
int n;
mutex_enter(d->bd_buf_mtx);
n = d->bd_slen;
if (d->bd_hbuf)
n += d->bd_hlen;
mutex_exit(d->bd_buf_mtx);
*(int *)addr = n;
break;
}
/*
* Get buffer len [for read()].
*/
case BIOCGBLEN:
*(u_int *)addr = d->bd_bufsize;
break;
/*
* Set buffer length.
*/
case BIOCSBLEN:
/*
* Forbid to change the buffer length if buffers are already
* allocated.
*/
mutex_enter(d->bd_mtx);
mutex_enter(d->bd_buf_mtx);
if (d->bd_bif != NULL || d->bd_sbuf != NULL)
error = EINVAL;
else {
u_int size = *(u_int *)addr;
if (size > bpf_maxbufsize)
*(u_int *)addr = size = bpf_maxbufsize;
else if (size < BPF_MINBUFSIZE)
*(u_int *)addr = size = BPF_MINBUFSIZE;
d->bd_bufsize = size;
}
mutex_exit(d->bd_buf_mtx);
mutex_exit(d->bd_mtx);
break;
/*
* Set link layer read filter.
*/
case BIOCSETF: /* FALLTHROUGH */
case BIOCSETWF:
error = bpf_setf(d, addr, cmd);
break;
case BIOCLOCK:
d->bd_locked = 1;
break;
/*
* Flush read packet buffer.
*/
case BIOCFLUSH:
mutex_enter(d->bd_mtx);
reset_d(d);
mutex_exit(d->bd_mtx);
break;
/*
* Put interface into promiscuous mode.
*/
case BIOCPROMISC:
mutex_enter(d->bd_mtx);
if (d->bd_bif == NULL) {
mutex_exit(d->bd_mtx);
/*
* No interface attached yet.
*/
error = EINVAL;
break;
}
if (d->bd_promisc == 0) {
KERNEL_LOCK_UNLESS_NET_MPSAFE();
error = ifpromisc(d->bd_bif->bif_ifp, 1);
KERNEL_UNLOCK_UNLESS_NET_MPSAFE();
if (error == 0)
d->bd_promisc = 1;
}
mutex_exit(d->bd_mtx);
break;
/*
* Get device parameters.
*/
case BIOCGDLT:
mutex_enter(d->bd_mtx);
if (d->bd_bif == NULL)
error = EINVAL;
else
*(u_int *)addr = d->bd_bif->bif_dlt;
mutex_exit(d->bd_mtx);
break;
/*
* Get a list of supported device parameters.
*/
case BIOCGDLTLIST:
mutex_enter(d->bd_mtx);
if (d->bd_bif == NULL)
error = EINVAL;
else
error = bpf_getdltlist(d, addr);
mutex_exit(d->bd_mtx);
break;
/*
* Set device parameters.
*/
case BIOCSDLT:
mutex_enter(&bpf_mtx);
mutex_enter(d->bd_mtx);
if (d->bd_bif == NULL)
error = EINVAL;
else
error = bpf_setdlt(d, *(u_int *)addr);
mutex_exit(d->bd_mtx);
mutex_exit(&bpf_mtx);
break;
/*
* Set interface name.
*/
#ifdef OBIOCGETIF
case OBIOCGETIF:
#endif
case BIOCGETIF:
mutex_enter(d->bd_mtx);
if (d->bd_bif == NULL)
error = EINVAL;
else
bpf_ifname(d->bd_bif->bif_ifp, addr);
mutex_exit(d->bd_mtx);
break;
/*
* Set interface.
*/
#ifdef OBIOCSETIF
case OBIOCSETIF:
#endif
case BIOCSETIF:
mutex_enter(&bpf_mtx);
error = bpf_setif(d, addr);
mutex_exit(&bpf_mtx);
break;
/*
* Set read timeout.
*/
case BIOCSRTIMEOUT:
{
struct timeval *tv = addr;
/* Compute number of ticks. */
if (tv->tv_sec < 0 ||
tv->tv_usec < 0 || tv->tv_usec >= 1000000) {
error = EINVAL;
break;
} else if (tv->tv_sec > INT_MAX/hz - 1) {
d->bd_rtout = INT_MAX;
} else {
d->bd_rtout = tv->tv_sec * hz
+ tv->tv_usec / tick;
}
if ((d->bd_rtout == 0) && (tv->tv_usec != 0))
d->bd_rtout = 1;
break;
}
#ifdef BIOCGORTIMEOUT
/*
* Get read timeout.
*/
case BIOCGORTIMEOUT:
{
struct timeval50 *tv = addr;
tv->tv_sec = d->bd_rtout / hz;
tv->tv_usec = (d->bd_rtout % hz) * tick;
break;
}
#endif
#ifdef BIOCSORTIMEOUT
/*
* Set read timeout.
*/
case BIOCSORTIMEOUT:
{
struct timeval50 *tv = addr;
/* Compute number of ticks. */
if (tv->tv_sec < 0 ||
tv->tv_usec < 0 || tv->tv_usec >= 1000000) {
error = EINVAL;
break;
} else if (tv->tv_sec > INT_MAX/hz - 1) {
d->bd_rtout = INT_MAX;
} else {
d->bd_rtout = tv->tv_sec * hz
+ tv->tv_usec / tick;
}
if ((d->bd_rtout == 0) && (tv->tv_usec != 0))
d->bd_rtout = 1;
break;
}
#endif
/*
* Get read timeout.
*/
case BIOCGRTIMEOUT:
{
struct timeval *tv = addr;
tv->tv_sec = d->bd_rtout / hz;
tv->tv_usec = (d->bd_rtout % hz) * tick;
break;
}
/*
* Get packet stats.
*/
case BIOCGSTATS:
{
struct bpf_stat *bs = addr;
bs->bs_recv = d->bd_rcount;
bs->bs_drop = d->bd_dcount;
bs->bs_capt = d->bd_ccount;
break;
}
case BIOCGSTATSOLD:
{
struct bpf_stat_old *bs = addr;
bs->bs_recv = d->bd_rcount;
bs->bs_drop = d->bd_dcount;
break;
}
/*
* Set immediate mode.
*/
case BIOCIMMEDIATE:
d->bd_immediate = *(u_int *)addr;
break;
case BIOCVERSION:
{
struct bpf_version *bv = addr;
bv->bv_major = BPF_MAJOR_VERSION;
bv->bv_minor = BPF_MINOR_VERSION;
break;
}
case BIOCGHDRCMPLT: /* get "header already complete" flag */
*(u_int *)addr = d->bd_hdrcmplt;
break;
case BIOCSHDRCMPLT: /* set "header already complete" flag */
d->bd_hdrcmplt = *(u_int *)addr ? 1 : 0;
break;
/*
* Get packet direction flag
*/
case BIOCGDIRECTION:
*(u_int *)addr = d->bd_direction;
break;
/*
* Set packet direction flag
*/
case BIOCSDIRECTION:
{
u_int direction;
direction = *(u_int *)addr;
switch (direction) {
case BPF_D_IN:
case BPF_D_INOUT:
case BPF_D_OUT:
d->bd_direction = direction;
break;
default:
error = EINVAL;
}
}
break;
/*
* Set "feed packets from bpf back to input" mode
*/
case BIOCSFEEDBACK:
d->bd_feedback = *(u_int *)addr;
break;
/*
* Get "feed packets from bpf back to input" mode
*/
case BIOCGFEEDBACK:
*(u_int *)addr = d->bd_feedback;
break;
case FIONBIO: /* Non-blocking I/O */
/*
* No need to do anything special as we use IO_NDELAY in
* bpfread() as an indication of whether or not to block
* the read.
*/
break;
case FIOASYNC: /* Send signal on receive packets */
mutex_enter(d->bd_mtx);
d->bd_async = *(int *)addr;
mutex_exit(d->bd_mtx);
break;
case TIOCSPGRP: /* Process or group to send signals to */
case FIOSETOWN:
error = fsetown(&d->bd_pgid, cmd, addr);
break;
case TIOCGPGRP:
case FIOGETOWN:
error = fgetown(d->bd_pgid, cmd, addr);
break;
}
return (error);
}
/*
* Set d's packet filter program to fp. If this file already has a filter,
* free it and replace it. Returns EINVAL for bogus requests.
*/
static int
bpf_setf(struct bpf_d *d, struct bpf_program *fp, u_long cmd)
{
struct bpf_insn *fcode;
bpfjit_func_t jcode;
size_t flen, size = 0;
struct bpf_filter *oldf, *newf, **storef;
jcode = NULL;
flen = fp->bf_len;
if ((fp->bf_insns == NULL && flen) || flen > BPF_MAXINSNS) {
return EINVAL;
}
if (flen) {
/*
* Allocate the buffer, copy the byte-code from
* userspace and validate it.
*/
size = flen * sizeof(*fp->bf_insns);
fcode = kmem_alloc(size, KM_SLEEP);
if (copyin(fp->bf_insns, fcode, size) != 0 ||
!bpf_validate(fcode, (int)flen)) {
kmem_free(fcode, size);
return EINVAL;
}
if (bpf_jit)
jcode = bpf_jit_generate(NULL, fcode, flen);
} else {
fcode = NULL;
}
newf = kmem_alloc(sizeof(*newf), KM_SLEEP);
newf->bf_insn = fcode;
newf->bf_size = size;
newf->bf_jitcode = jcode;
if (cmd == BIOCSETF)
d->bd_jitcode = jcode; /* XXX just for kvm(3) users */
/* Need to hold bpf_mtx for pserialize_perform */
mutex_enter(&bpf_mtx);
mutex_enter(d->bd_mtx);
if (cmd == BIOCSETWF) {
oldf = d->bd_wfilter;
storef = &d->bd_wfilter;
} else {
oldf = d->bd_rfilter;
storef = &d->bd_rfilter;
}
atomic_store_release(storef, newf);
reset_d(d);
pserialize_perform(bpf_psz);
mutex_exit(d->bd_mtx);
mutex_exit(&bpf_mtx);
if (oldf != NULL)
bpf_free_filter(oldf);
return 0;
}
/*
* Detach a file from its current interface (if attached at all) and attach
* to the interface indicated by the name stored in ifr.
* Return an errno or 0.
*/
static int
bpf_setif(struct bpf_d *d, struct ifreq *ifr)
{
struct bpf_if *bp;
char *cp;
int unit_seen, i, error;
KASSERT(mutex_owned(&bpf_mtx));
/*
* Make sure the provided name has a unit number, and default
* it to '0' if not specified.
* XXX This is ugly ... do this differently?
*/
unit_seen = 0;
cp = ifr->ifr_name;
cp[sizeof(ifr->ifr_name) - 1] = '\0'; /* sanity */
while (*cp++)
if (*cp >= '0' && *cp <= '9')
unit_seen = 1;
if (!unit_seen) {
/* Make sure to leave room for the '\0'. */
for (i = 0; i < (IFNAMSIZ - 1); ++i) {
if ((ifr->ifr_name[i] >= 'a' &&
ifr->ifr_name[i] <= 'z') ||
(ifr->ifr_name[i] >= 'A' &&
ifr->ifr_name[i] <= 'Z'))
continue;
ifr->ifr_name[i] = '0';
}
}
/*
* Look through attached interfaces for the named one.
*/
BPF_IFLIST_WRITER_FOREACH(bp) {
struct ifnet *ifp = bp->bif_ifp;
if (ifp == NULL ||
strcmp(ifp->if_xname, ifr->ifr_name) != 0)
continue;
/* skip additional entry */
if (bp->bif_driverp != &ifp->if_bpf)
continue;
/*
* We found the requested interface.
* Allocate the packet buffers if we need to.
* If we're already attached to requested interface,
* just flush the buffer.
*/
/*
* bpf_allocbufs is called only here. bpf_mtx ensures that
* no race condition happen on d->bd_sbuf.
*/
if (d->bd_sbuf == NULL) {
error = bpf_allocbufs(d);
if (error != 0)
return (error);
}
mutex_enter(d->bd_mtx);
if (bp != d->bd_bif) {
if (d->bd_bif) {
/*
* Detach if attached to something else.
*/
bpf_detachd(d);
BPFIF_DLIST_ENTRY_INIT(d);
}
bpf_attachd(d, bp);
}
reset_d(d);
mutex_exit(d->bd_mtx);
return (0);
}
/* Not found. */
return (ENXIO);
}
/*
* Copy the interface name to the ifreq.
*/
static void
bpf_ifname(struct ifnet *ifp, struct ifreq *ifr)
{
memcpy(ifr->ifr_name, ifp->if_xname, IFNAMSIZ);
}
static int
bpf_stat(struct file *fp, struct stat *st)
{
struct bpf_d *d = fp->f_bpf;
(void)memset(st, 0, sizeof(*st));
mutex_enter(d->bd_mtx);
st->st_dev = makedev(cdevsw_lookup_major(&bpf_cdevsw), d->bd_pid);
st->st_atimespec = d->bd_atime;
st->st_mtimespec = d->bd_mtime;
st->st_ctimespec = st->st_birthtimespec = d->bd_btime;
st->st_uid = kauth_cred_geteuid(fp->f_cred);
st->st_gid = kauth_cred_getegid(fp->f_cred);
st->st_mode = S_IFCHR;
mutex_exit(d->bd_mtx);
return 0;
}
/*
* Support for poll() system call
*
* Return true iff the specific operation will not block indefinitely - with
* the assumption that it is safe to positively acknowledge a request for the
* ability to write to the BPF device.
* Otherwise, return false but make a note that a selnotify() must be done.
*/
static int
bpf_poll(struct file *fp, int events)
{
struct bpf_d *d = fp->f_bpf;
int revents;
/*
* Refresh the PID associated with this bpf file.
*/
mutex_enter(&bpf_mtx);
d->bd_pid = curproc->p_pid;
revents = events & (POLLOUT | POLLWRNORM);
if (events & (POLLIN | POLLRDNORM)) {
/*
* An imitation of the FIONREAD ioctl code.
*/
mutex_enter(d->bd_mtx);
if (d->bd_hlen != 0 ||
((d->bd_immediate || d->bd_state == BPF_TIMED_OUT) &&
d->bd_slen != 0)) {
revents |= events & (POLLIN | POLLRDNORM);
} else {
selrecord(curlwp, &d->bd_sel);
/* Start the read timeout if necessary */
if (d->bd_rtout > 0 && d->bd_state == BPF_IDLE) {
callout_reset(&d->bd_callout, d->bd_rtout,
bpf_timed_out, d);
d->bd_state = BPF_WAITING;
}
}
mutex_exit(d->bd_mtx);
}
mutex_exit(&bpf_mtx);
return (revents);
}
static void
filt_bpfrdetach(struct knote *kn)
{
struct bpf_d *d = kn->kn_hook;
mutex_enter(d->bd_buf_mtx);
selremove_knote(&d->bd_sel, kn);
mutex_exit(d->bd_buf_mtx);
}
static int
filt_bpfread(struct knote *kn, long hint)
{
struct bpf_d *d = kn->kn_hook;
int rv;
/*
* Refresh the PID associated with this bpf file.
*/
d->bd_pid = curproc->p_pid;
mutex_enter(d->bd_buf_mtx);
kn->kn_data = d->bd_hlen;
if (d->bd_immediate)
kn->kn_data += d->bd_slen;
rv = (kn->kn_data > 0);
mutex_exit(d->bd_buf_mtx);
return rv;
}
static const struct filterops bpfread_filtops = {
.f_flags = FILTEROP_ISFD,
.f_attach = NULL,
.f_detach = filt_bpfrdetach,
.f_event = filt_bpfread,
};
static int
bpf_kqfilter(struct file *fp, struct knote *kn)
{
struct bpf_d *d = fp->f_bpf;
switch (kn->kn_filter) {
case EVFILT_READ:
kn->kn_fop = &bpfread_filtops;
break;
default:
return (EINVAL);
}
kn->kn_hook = d;
mutex_enter(d->bd_buf_mtx);
selrecord_knote(&d->bd_sel, kn);
mutex_exit(d->bd_buf_mtx);
return (0);
}
/*
* Copy data from an mbuf chain into a buffer. This code is derived
* from m_copydata in sys/uipc_mbuf.c.
*/
static void *
bpf_mcpy(void *dst_arg, const void *src_arg, size_t len)
{
const struct mbuf *m;
u_int count;
u_char *dst;
m = src_arg;
dst = dst_arg;
while (len > 0) {
if (m == NULL)
panic("bpf_mcpy");
count = uimin(m->m_len, len);
memcpy(dst, mtod(m, const void *), count);
m = m->m_next;
dst += count;
len -= count;
}
return dst_arg;
}
static inline u_int
bpf_xfilter(struct bpf_filter **filter, void *pkt, u_int pktlen, u_int buflen)
{
struct bpf_filter *filt;
uint32_t mem[BPF_MEMWORDS];
bpf_args_t args = {
.pkt = (const uint8_t *)pkt,
.wirelen = pktlen,
.buflen = buflen,
.mem = mem,
.arg = NULL
};
u_int slen;
filt = atomic_load_consume(filter);
if (filt == NULL) /* No filter means accept all. */
return (u_int)-1;
if (filt->bf_jitcode != NULL)
slen = filt->bf_jitcode(NULL, &args);
else
slen = bpf_filter_ext(NULL, filt->bf_insn, &args);
return slen;
}
/*
* Dispatch a packet to all the listeners on interface bp.
*
* pkt pointer to the packet, either a data buffer or an mbuf chain
* buflen buffer length, if pkt is a data buffer
* cpfn a function that can copy pkt into the listener's buffer
* pktlen length of the packet
* direction BPF_D_IN or BPF_D_OUT
*/
static inline void
bpf_deliver(struct bpf_if *bp, void *(*cpfn)(void *, const void *, size_t),
void *pkt, u_int pktlen, u_int buflen, const u_int direction)
{
bool gottime = false;
struct timespec ts;
struct bpf_d *d;
int s;
u_int slen;
KASSERT(!cpu_intr_p());
/*
* Note that the IPL does not have to be raised at this point.
* The only problem that could arise here is that if two different
* interfaces shared any data. This is not the case.
*/
s = pserialize_read_enter();
BPFIF_DLIST_READER_FOREACH(d, bp) {
if (direction == BPF_D_IN) {
if (d->bd_direction == BPF_D_OUT)
continue;
} else { /* BPF_D_OUT */
if (d->bd_direction == BPF_D_IN)
continue;
}
atomic_inc_ulong(&d->bd_rcount);
BPF_STATINC(recv);
slen = bpf_xfilter(&d->bd_rfilter, pkt, pktlen, buflen);
if (slen == 0)
continue;
if (!gottime) {
gottime = true;
nanotime(&ts);
}
/* Assume catchpacket doesn't sleep */
catchpacket(d, pkt, pktlen, slen, cpfn, &ts);
}
pserialize_read_exit(s);
}
/*
* Incoming linkage from device drivers, when the head of the packet is in
* a buffer, and the tail is in an mbuf chain.
*/
static void
_bpf_mtap2(struct bpf_if *bp, void *data, u_int dlen, struct mbuf *m,
u_int direction)
{
u_int pktlen;
struct mbuf mb;
/* Skip outgoing duplicate packets. */
if ((m->m_flags & M_PROMISC) != 0 && m->m_pkthdr.rcvif_index == 0) {
m->m_flags &= ~M_PROMISC;
return;
}
pktlen = m_length(m) + dlen;
/*
* Craft on-stack mbuf suitable for passing to bpf_filter.
* Note that we cut corners here; we only set up what's
* absolutely needed--this mbuf should never go anywhere else.
*/
(void)memset(&mb, 0, sizeof(mb));
mb.m_type = MT_DATA;
mb.m_next = m;
mb.m_data = data;
mb.m_len = dlen;
bpf_deliver(bp, bpf_mcpy, &mb, pktlen, 0, direction);
}
/*
* Incoming linkage from device drivers, when packet is in an mbuf chain.
*/
static void
_bpf_mtap(struct bpf_if *bp, struct mbuf *m, u_int direction)
{
void *(*cpfn)(void *, const void *, size_t);
u_int pktlen, buflen;
void *marg;
/* Skip outgoing duplicate packets. */
if ((m->m_flags & M_PROMISC) != 0 && m->m_pkthdr.rcvif_index == 0) {
m->m_flags &= ~M_PROMISC;
return;
}
pktlen = m_length(m);
/* Skip zero-sized packets. */
if (__predict_false(pktlen == 0)) {
return;
}
if (pktlen == m->m_len) {
cpfn = (void *)memcpy;
marg = mtod(m, void *);
buflen = pktlen;
KASSERT(buflen != 0);
} else {
cpfn = bpf_mcpy;
marg = m;
buflen = 0;
}
bpf_deliver(bp, cpfn, marg, pktlen, buflen, direction);
}
/*
* We need to prepend the address family as
* a four byte field. Cons up a dummy header
* to pacify bpf. This is safe because bpf
* will only read from the mbuf (i.e., it won't
* try to free it or keep a pointer a to it).
*/
static void
_bpf_mtap_af(struct bpf_if *bp, uint32_t af, struct mbuf *m, u_int direction)
{
struct mbuf m0;
m0.m_type = MT_DATA;
m0.m_flags = 0;
m0.m_next = m;
m0.m_nextpkt = NULL;
m0.m_owner = NULL;
m0.m_len = 4;
m0.m_data = (char *)⁡
_bpf_mtap(bp, &m0, direction);
}
/*
* Put the SLIP pseudo-"link header" in place.
* Note this M_PREPEND() should never fail,
* since we know we always have enough space
* in the input buffer.
*/
static void
_bpf_mtap_sl_in(struct bpf_if *bp, u_char *chdr, struct mbuf **m)
{
u_char *hp;
M_PREPEND(*m, SLIP_HDRLEN, M_DONTWAIT);
if (*m == NULL)
return;
hp = mtod(*m, u_char *);
hp[SLX_DIR] = SLIPDIR_IN;
(void)memcpy(&hp[SLX_CHDR], chdr, CHDR_LEN);
_bpf_mtap(bp, *m, BPF_D_IN);
m_adj(*m, SLIP_HDRLEN);
}
/*
* Put the SLIP pseudo-"link header" in
* place. The compressed header is now
* at the beginning of the mbuf.
*/
static void
_bpf_mtap_sl_out(struct bpf_if *bp, u_char *chdr, struct mbuf *m)
{
struct mbuf m0;
u_char *hp;
m0.m_type = MT_DATA;
m0.m_flags = 0;
m0.m_next = m;
m0.m_nextpkt = NULL;
m0.m_owner = NULL;
m0.m_data = m0.m_dat;
m0.m_len = SLIP_HDRLEN;
hp = mtod(&m0, u_char *);
hp[SLX_DIR] = SLIPDIR_OUT;
(void)memcpy(&hp[SLX_CHDR], chdr, CHDR_LEN);
_bpf_mtap(bp, &m0, BPF_D_OUT);
m_freem(m);
}
static struct mbuf *
bpf_mbuf_enqueue(struct bpf_if *bp, struct mbuf *m)
{
struct mbuf *dup;
dup = m_dup(m, 0, M_COPYALL, M_NOWAIT);
if (dup == NULL)
return NULL;
if (bp->bif_mbuf_tail != NULL) {
bp->bif_mbuf_tail->m_nextpkt = dup;
} else {
bp->bif_mbuf_head = dup;
}
bp->bif_mbuf_tail = dup;
#ifdef BPF_MTAP_SOFTINT_DEBUG
log(LOG_DEBUG, "%s: enqueued mbuf=%p to %s\n",
__func__, dup, bp->bif_ifp->if_xname);
#endif
return dup;
}
static struct mbuf *
bpf_mbuf_dequeue(struct bpf_if *bp)
{
struct mbuf *m;
int s;
/* XXX NOMPSAFE: assumed running on one CPU */
s = splnet();
m = bp->bif_mbuf_head;
if (m != NULL) {
bp->bif_mbuf_head = m->m_nextpkt;
m->m_nextpkt = NULL;
if (bp->bif_mbuf_head == NULL)
bp->bif_mbuf_tail = NULL;
#ifdef BPF_MTAP_SOFTINT_DEBUG
log(LOG_DEBUG, "%s: dequeued mbuf=%p from %s\n",
__func__, m, bp->bif_ifp->if_xname);
#endif
}
splx(s);
return m;
}
static void
bpf_mtap_si(void *arg)
{
struct bpf_if *bp = arg;
struct mbuf *m;
while ((m = bpf_mbuf_dequeue(bp)) != NULL) {
#ifdef BPF_MTAP_SOFTINT_DEBUG
log(LOG_DEBUG, "%s: tapping mbuf=%p on %s\n",
__func__, m, bp->bif_ifp->if_xname);
#endif
bpf_ops->bpf_mtap(bp, m, BPF_D_IN);
m_freem(m);
}
}
static void
_bpf_mtap_softint(struct ifnet *ifp, struct mbuf *m)
{
struct bpf_if *bp = ifp->if_bpf;
struct mbuf *dup;
KASSERT(cpu_intr_p());
/* To avoid extra invocations of the softint */
if (BPFIF_DLIST_READER_EMPTY(bp))
return;
KASSERT(bp->bif_si != NULL);
dup = bpf_mbuf_enqueue(bp, m);
if (dup != NULL)
softint_schedule(bp->bif_si);
}
static int
bpf_hdrlen(struct bpf_d *d)
{
int hdrlen = d->bd_bif->bif_hdrlen;
/*
* Compute the length of the bpf header. This is not necessarily
* equal to SIZEOF_BPF_HDR because we want to insert spacing such
* that the network layer header begins on a longword boundary (for
* performance reasons and to alleviate alignment restrictions).
*/
#ifdef _LP64
if (d->bd_compat32)
return (BPF_WORDALIGN32(hdrlen + SIZEOF_BPF_HDR32) - hdrlen);
else
#endif
return (BPF_WORDALIGN(hdrlen + SIZEOF_BPF_HDR) - hdrlen);
}
/*
* Move the packet data from interface memory (pkt) into the
* store buffer. Call the wakeup functions if it's time to wake up
* a listener (buffer full), "cpfn" is the routine called to do the
* actual data transfer. memcpy is passed in to copy contiguous chunks,
* while bpf_mcpy is passed in to copy mbuf chains. In the latter case,
* pkt is really an mbuf.
*/
static void
catchpacket(struct bpf_d *d, u_char *pkt, u_int pktlen, u_int snaplen,
void *(*cpfn)(void *, const void *, size_t), struct timespec *ts)
{
char *h;
int totlen, curlen, caplen;
int hdrlen = bpf_hdrlen(d);
int do_wakeup = 0;
atomic_inc_ulong(&d->bd_ccount);
BPF_STATINC(capt);
/*
* Figure out how many bytes to move. If the packet is
* greater or equal to the snapshot length, transfer that
* much. Otherwise, transfer the whole packet (unless
* we hit the buffer size limit).
*/
totlen = hdrlen + uimin(snaplen, pktlen);
if (totlen > d->bd_bufsize)
totlen = d->bd_bufsize;
/*
* If we adjusted totlen to fit the bufsize, it could be that
* totlen is smaller than hdrlen because of the link layer header.
*/
caplen = totlen - hdrlen;
if (caplen < 0)
caplen = 0;
mutex_enter(d->bd_buf_mtx);
/*
* Round up the end of the previous packet to the next longword.
*/
#ifdef _LP64
if (d->bd_compat32)
curlen = BPF_WORDALIGN32(d->bd_slen);
else
#endif
curlen = BPF_WORDALIGN(d->bd_slen);
if (curlen + totlen > d->bd_bufsize) {
/*
* This packet will overflow the storage buffer.
* Rotate the buffers if we can, then wakeup any
* pending reads.
*/
if (d->bd_fbuf == NULL) {
mutex_exit(d->bd_buf_mtx);
/*
* We haven't completed the previous read yet,
* so drop the packet.
*/
atomic_inc_ulong(&d->bd_dcount);
BPF_STATINC(drop);
return;
}
ROTATE_BUFFERS(d);
do_wakeup = 1;
curlen = 0;
} else if (d->bd_immediate || d->bd_state == BPF_TIMED_OUT) {
/*
* Immediate mode is set, or the read timeout has
* already expired during a select call. A packet
* arrived, so the reader should be woken up.
*/
do_wakeup = 1;
}
/*
* Append the bpf header.
*/
h = (char *)d->bd_sbuf + curlen;
#ifdef _LP64
if (d->bd_compat32) {
struct bpf_hdr32 *hp32;
hp32 = (struct bpf_hdr32 *)h;
hp32->bh_tstamp.tv_sec = ts->tv_sec;
hp32->bh_tstamp.tv_usec = ts->tv_nsec / 1000;
hp32->bh_datalen = pktlen;
hp32->bh_hdrlen = hdrlen;
hp32->bh_caplen = caplen;
} else
#endif
{
struct bpf_hdr *hp;
hp = (struct bpf_hdr *)h;
hp->bh_tstamp.tv_sec = ts->tv_sec;
hp->bh_tstamp.tv_usec = ts->tv_nsec / 1000;
hp->bh_datalen = pktlen;
hp->bh_hdrlen = hdrlen;
hp->bh_caplen = caplen;
}
/*
* Copy the packet data into the store buffer and update its length.
*/
(*cpfn)(h + hdrlen, pkt, caplen);
d->bd_slen = curlen + totlen;
mutex_exit(d->bd_buf_mtx);
/*
* Call bpf_wakeup after bd_slen has been updated so that kevent(2)
* will cause filt_bpfread() to be called with it adjusted.
*/
if (do_wakeup)
bpf_wakeup(d);
}
/*
* Initialize all nonzero fields of a descriptor.
*/
static int
bpf_allocbufs(struct bpf_d *d)
{
d->bd_fbuf = kmem_zalloc(d->bd_bufsize, KM_NOSLEEP);
if (!d->bd_fbuf)
return (ENOBUFS);
d->bd_sbuf = kmem_zalloc(d->bd_bufsize, KM_NOSLEEP);
if (!d->bd_sbuf) {
kmem_free(d->bd_fbuf, d->bd_bufsize);
return (ENOBUFS);
}
d->bd_slen = 0;
d->bd_hlen = 0;
return (0);
}
static void
bpf_free_filter(struct bpf_filter *filter)
{
KASSERT(filter != NULL);
if (filter->bf_insn != NULL)
kmem_free(filter->bf_insn, filter->bf_size);
if (filter->bf_jitcode != NULL)
bpf_jit_freecode(filter->bf_jitcode);
kmem_free(filter, sizeof(*filter));
}
/*
* Free buffers currently in use by a descriptor.
* Called on close.
*/
static void
bpf_freed(struct bpf_d *d)
{
/*
* We don't need to lock out interrupts since this descriptor has
* been detached from its interface and it yet hasn't been marked
* free.
*/
if (d->bd_sbuf != NULL) {
kmem_free(d->bd_sbuf, d->bd_bufsize);
if (d->bd_hbuf != NULL)
kmem_free(d->bd_hbuf, d->bd_bufsize);
if (d->bd_fbuf != NULL)
kmem_free(d->bd_fbuf, d->bd_bufsize);
}
if (d->bd_rfilter != NULL) {
bpf_free_filter(d->bd_rfilter);
d->bd_rfilter = NULL;
}
if (d->bd_wfilter != NULL) {
bpf_free_filter(d->bd_wfilter);
d->bd_wfilter = NULL;
}
d->bd_jitcode = NULL;
}
/*
* Attach an interface to bpf. dlt is the link layer type;
* hdrlen is the fixed size of the link header for the specified dlt
* (variable length headers not yet supported).
*/
static void
_bpfattach(struct ifnet *ifp, u_int dlt, u_int hdrlen, struct bpf_if **driverp)
{
struct bpf_if *bp;
bp = kmem_alloc(sizeof(*bp), KM_SLEEP);
mutex_enter(&bpf_mtx);
bp->bif_driverp = driverp;
bp->bif_ifp = ifp;
bp->bif_dlt = dlt;
bp->bif_si = NULL;
BPF_IFLIST_ENTRY_INIT(bp);
PSLIST_INIT(&bp->bif_dlist_head);
psref_target_init(&bp->bif_psref, bpf_psref_class);
SLIST_INIT(&bp->bif_trackers);
BPF_IFLIST_WRITER_INSERT_HEAD(bp);
*bp->bif_driverp = NULL;
bp->bif_hdrlen = hdrlen;
mutex_exit(&bpf_mtx);
#if 0
printf("bpf: %s attached with dlt %x\n", ifp->if_xname, dlt);
#endif
}
static void
_bpf_mtap_softint_init(struct ifnet *ifp)
{
struct bpf_if *bp;
mutex_enter(&bpf_mtx);
BPF_IFLIST_WRITER_FOREACH(bp) {
if (bp->bif_ifp != ifp)
continue;
bp->bif_mbuf_head = NULL;
bp->bif_mbuf_tail = NULL;
bp->bif_si = softint_establish(SOFTINT_NET, bpf_mtap_si, bp);
if (bp->bif_si == NULL)
panic("%s: softint_establish() failed", __func__);
break;
}
mutex_exit(&bpf_mtx);
if (bp == NULL)
panic("%s: no bpf_if found for %s", __func__, ifp->if_xname);
}
/*
* Remove an interface from bpf.
*/
static void
_bpfdetach(struct ifnet *ifp)
{
struct bpf_if *bp;
struct bpf_d *d;
int s;
mutex_enter(&bpf_mtx);
/* Nuke the vnodes for any open instances */
again_d:
BPF_DLIST_WRITER_FOREACH(d) {
mutex_enter(d->bd_mtx);
if (d->bd_bif != NULL && d->bd_bif->bif_ifp == ifp) {
/*
* Detach the descriptor from an interface now.
* It will be free'ed later by close routine.
*/
bpf_detachd(d);
mutex_exit(d->bd_mtx);
goto again_d;
}
mutex_exit(d->bd_mtx);
}
again:
BPF_IFLIST_WRITER_FOREACH(bp) {
if (bp->bif_ifp == ifp) {
BPF_IFLIST_WRITER_REMOVE(bp);
pserialize_perform(bpf_psz);
psref_target_destroy(&bp->bif_psref, bpf_psref_class);
while (!SLIST_EMPTY(&bp->bif_trackers)) {
struct bpf_event_tracker *t =
SLIST_FIRST(&bp->bif_trackers);
SLIST_REMOVE_HEAD(&bp->bif_trackers,
bet_entries);
kmem_free(t, sizeof(*t));
}
BPF_IFLIST_ENTRY_DESTROY(bp);
if (bp->bif_si != NULL) {
/* XXX NOMPSAFE: assumed running on one CPU */
s = splnet();
while (bp->bif_mbuf_head != NULL) {
struct mbuf *m = bp->bif_mbuf_head;
bp->bif_mbuf_head = m->m_nextpkt;
m_freem(m);
}
splx(s);
softint_disestablish(bp->bif_si);
}
kmem_free(bp, sizeof(*bp));
goto again;
}
}
mutex_exit(&bpf_mtx);
}
/*
* Change the data link type of a interface.
*/
static void
_bpf_change_type(struct ifnet *ifp, u_int dlt, u_int hdrlen)
{
struct bpf_if *bp;
mutex_enter(&bpf_mtx);
BPF_IFLIST_WRITER_FOREACH(bp) {
if (bp->bif_driverp == &ifp->if_bpf)
break;
}
if (bp == NULL)
panic("bpf_change_type");
bp->bif_dlt = dlt;
bp->bif_hdrlen = hdrlen;
mutex_exit(&bpf_mtx);
}
/*
* Get a list of available data link type of the interface.
*/
static int
bpf_getdltlist(struct bpf_d *d, struct bpf_dltlist *bfl)
{
int n, error;
struct ifnet *ifp;
struct bpf_if *bp;
int s, bound;
KASSERT(mutex_owned(d->bd_mtx));
ifp = d->bd_bif->bif_ifp;
n = 0;
error = 0;
bound = curlwp_bind();
s = pserialize_read_enter();
BPF_IFLIST_READER_FOREACH(bp) {
if (bp->bif_ifp != ifp)
continue;
if (bfl->bfl_list != NULL) {
struct psref psref;
if (n >= bfl->bfl_len) {
pserialize_read_exit(s);
return ENOMEM;
}
bpf_if_acquire(bp, &psref);
pserialize_read_exit(s);
error = copyout(&bp->bif_dlt,
bfl->bfl_list + n, sizeof(u_int));
s = pserialize_read_enter();
bpf_if_release(bp, &psref);
}
n++;
}
pserialize_read_exit(s);
curlwp_bindx(bound);
bfl->bfl_len = n;
return error;
}
/*
* Set the data link type of a BPF instance.
*/
static int
bpf_setdlt(struct bpf_d *d, u_int dlt)
{
int error, opromisc;
struct ifnet *ifp;
struct bpf_if *bp;
KASSERT(mutex_owned(&bpf_mtx));
KASSERT(mutex_owned(d->bd_mtx));
if (d->bd_bif->bif_dlt == dlt)
return 0;
ifp = d->bd_bif->bif_ifp;
BPF_IFLIST_WRITER_FOREACH(bp) {
if (bp->bif_ifp == ifp && bp->bif_dlt == dlt)
break;
}
if (bp == NULL)
return EINVAL;
opromisc = d->bd_promisc;
bpf_detachd(d);
BPFIF_DLIST_ENTRY_INIT(d);
bpf_attachd(d, bp);
reset_d(d);
if (opromisc) {
KERNEL_LOCK_UNLESS_NET_MPSAFE();
error = ifpromisc(bp->bif_ifp, 1);
KERNEL_UNLOCK_UNLESS_NET_MPSAFE();
if (error)
printf("%s: bpf_setdlt: ifpromisc failed (%d)\n",
bp->bif_ifp->if_xname, error);
else
d->bd_promisc = 1;
}
return 0;
}
static int
sysctl_net_bpf_maxbufsize(SYSCTLFN_ARGS)
{
int newsize, error;
struct sysctlnode node;
node = *rnode;
node.sysctl_data = &newsize;
newsize = bpf_maxbufsize;
error = sysctl_lookup(SYSCTLFN_CALL(&node));
if (error || newp == NULL)
return (error);
if (newsize < BPF_MINBUFSIZE || newsize > BPF_MAXBUFSIZE)
return (EINVAL);
bpf_maxbufsize = newsize;
return (0);
}
#if defined(MODULAR) || defined(BPFJIT)
static int
sysctl_net_bpf_jit(SYSCTLFN_ARGS)
{
bool newval;
int error;
struct sysctlnode node;
node = *rnode;
node.sysctl_data = &newval;
newval = bpf_jit;
error = sysctl_lookup(SYSCTLFN_CALL(&node));
if (error != 0 || newp == NULL)
return error;
bpf_jit = newval;
if (newval && bpfjit_module_ops.bj_generate_code == NULL) {
printf("JIT compilation is postponed "
"until after bpfjit module is loaded\n");
}
return 0;
}
#endif
static int
sysctl_net_bpf_peers(SYSCTLFN_ARGS)
{
int error, elem_count;
struct bpf_d *dp;
struct bpf_d_ext dpe;
size_t len, needed, elem_size, out_size;
char *sp;
if (namelen == 1 && name[0] == CTL_QUERY)
return (sysctl_query(SYSCTLFN_CALL(rnode)));
if (namelen != 2)
return (EINVAL);
/* BPF peers is privileged information. */
error = kauth_authorize_network(l->l_cred, KAUTH_NETWORK_INTERFACE,
KAUTH_REQ_NETWORK_INTERFACE_GETPRIV, NULL, NULL, NULL);
if (error)
return (EPERM);
len = (oldp != NULL) ? *oldlenp : 0;
sp = oldp;
elem_size = name[0];
elem_count = name[1];
out_size = MIN(sizeof(dpe), elem_size);
needed = 0;
if (elem_size < 1 || elem_count < 0)
return (EINVAL);
mutex_enter(&bpf_mtx);
BPF_DLIST_WRITER_FOREACH(dp) {
if (len >= elem_size && elem_count > 0) {
#define BPF_EXT(field) dpe.bde_ ## field = dp->bd_ ## field
BPF_EXT(bufsize);
BPF_EXT(promisc);
BPF_EXT(state);
BPF_EXT(immediate);
BPF_EXT(hdrcmplt);
BPF_EXT(direction);
BPF_EXT(pid);
BPF_EXT(rcount);
BPF_EXT(dcount);
BPF_EXT(ccount);
#undef BPF_EXT
mutex_enter(dp->bd_mtx);
if (dp->bd_bif)
(void)strlcpy(dpe.bde_ifname,
dp->bd_bif->bif_ifp->if_xname,
IFNAMSIZ - 1);
else
dpe.bde_ifname[0] = '\0';
dpe.bde_locked = dp->bd_locked;
mutex_exit(dp->bd_mtx);
error = copyout(&dpe, sp, out_size);
if (error)
break;
sp += elem_size;
len -= elem_size;
}
needed += elem_size;
if (elem_count > 0 && elem_count != INT_MAX)
elem_count--;
}
mutex_exit(&bpf_mtx);
*oldlenp = needed;
return (error);
}
static void
bpf_stats(void *p, void *arg, struct cpu_info *ci __unused)
{
struct bpf_stat *const stats = p;
struct bpf_stat *sum = arg;
int s = splnet();
sum->bs_recv += stats->bs_recv;
sum->bs_drop += stats->bs_drop;
sum->bs_capt += stats->bs_capt;
splx(s);
}
static int
bpf_sysctl_gstats_handler(SYSCTLFN_ARGS)
{
struct sysctlnode node;
int error;
struct bpf_stat sum;
memset(&sum, 0, sizeof(sum));
node = *rnode;
percpu_foreach_xcall(bpf_gstats_percpu, XC_HIGHPRI_IPL(IPL_SOFTNET),
bpf_stats, &sum);
node.sysctl_data = ∑
node.sysctl_size = sizeof(sum);
error = sysctl_lookup(SYSCTLFN_CALL(&node));
if (error != 0 || newp == NULL)
return error;
return 0;
}
SYSCTL_SETUP(sysctl_net_bpf_setup, "bpf sysctls")
{
const struct sysctlnode *node;
node = NULL;
sysctl_createv(clog, 0, NULL, &node,
CTLFLAG_PERMANENT,
CTLTYPE_NODE, "bpf",
SYSCTL_DESCR("BPF options"),
NULL, 0, NULL, 0,
CTL_NET, CTL_CREATE, CTL_EOL);
if (node != NULL) {
#if defined(MODULAR) || defined(BPFJIT)
sysctl_createv(clog, 0, NULL, NULL,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
CTLTYPE_BOOL, "jit",
SYSCTL_DESCR("Toggle Just-In-Time compilation"),
sysctl_net_bpf_jit, 0, &bpf_jit, 0,
CTL_NET, node->sysctl_num, CTL_CREATE, CTL_EOL);
#endif
sysctl_createv(clog, 0, NULL, NULL,
CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
CTLTYPE_INT, "maxbufsize",
SYSCTL_DESCR("Maximum size for data capture buffer"),
sysctl_net_bpf_maxbufsize, 0, &bpf_maxbufsize, 0,
CTL_NET, node->sysctl_num, CTL_CREATE, CTL_EOL);
sysctl_createv(clog, 0, NULL, NULL,
CTLFLAG_PERMANENT,
CTLTYPE_STRUCT, "stats",
SYSCTL_DESCR("BPF stats"),
bpf_sysctl_gstats_handler, 0, NULL, 0,
CTL_NET, node->sysctl_num, CTL_CREATE, CTL_EOL);
sysctl_createv(clog, 0, NULL, NULL,
CTLFLAG_PERMANENT,
CTLTYPE_STRUCT, "peers",
SYSCTL_DESCR("BPF peers"),
sysctl_net_bpf_peers, 0, NULL, 0,
CTL_NET, node->sysctl_num, CTL_CREATE, CTL_EOL);
}
}
static int
_bpf_register_track_event(struct bpf_if **driverp,
void (*_fun)(struct bpf_if *, struct ifnet *, int, int))
{
struct bpf_if *bp;
struct bpf_event_tracker *t;
int ret = ENOENT;
t = kmem_zalloc(sizeof(*t), KM_SLEEP);
if (!t)
return ENOMEM;
t->bet_notify = _fun;
mutex_enter(&bpf_mtx);
BPF_IFLIST_WRITER_FOREACH(bp) {
if (bp->bif_driverp != driverp)
continue;
SLIST_INSERT_HEAD(&bp->bif_trackers, t, bet_entries);
ret = 0;
break;
}
mutex_exit(&bpf_mtx);
return ret;
}
static int
_bpf_deregister_track_event(struct bpf_if **driverp,
void (*_fun)(struct bpf_if *, struct ifnet *, int, int))
{
struct bpf_if *bp;
struct bpf_event_tracker *t = NULL;
int ret = ENOENT;
mutex_enter(&bpf_mtx);
BPF_IFLIST_WRITER_FOREACH(bp) {
if (bp->bif_driverp != driverp)
continue;
SLIST_FOREACH(t, &bp->bif_trackers, bet_entries) {
if (t->bet_notify == _fun) {
ret = 0;
break;
}
}
if (ret == 0)
break;
}
if (ret == 0 && t && t->bet_notify == _fun) {
SLIST_REMOVE(&bp->bif_trackers, t, bpf_event_tracker,
bet_entries);
}
mutex_exit(&bpf_mtx);
if (ret == 0)
kmem_free(t, sizeof(*t));
return ret;
}
struct bpf_ops bpf_ops_kernel = {
.bpf_attach = _bpfattach,
.bpf_detach = _bpfdetach,
.bpf_change_type = _bpf_change_type,
.bpf_register_track_event = _bpf_register_track_event,
.bpf_deregister_track_event = _bpf_deregister_track_event,
.bpf_mtap = _bpf_mtap,
.bpf_mtap2 = _bpf_mtap2,
.bpf_mtap_af = _bpf_mtap_af,
.bpf_mtap_sl_in = _bpf_mtap_sl_in,
.bpf_mtap_sl_out = _bpf_mtap_sl_out,
.bpf_mtap_softint = _bpf_mtap_softint,
.bpf_mtap_softint_init = _bpf_mtap_softint_init,
};
MODULE(MODULE_CLASS_DRIVER, bpf, "bpf_filter");
static int
bpf_modcmd(modcmd_t cmd, void *arg)
{
#ifdef _MODULE
devmajor_t bmajor, cmajor;
#endif
int error = 0;
switch (cmd) {
case MODULE_CMD_INIT:
bpf_init();
#ifdef _MODULE
bmajor = cmajor = NODEVMAJOR;
error = devsw_attach("bpf", NULL, &bmajor,
&bpf_cdevsw, &cmajor);
if (error)
break;
#endif
bpf_ops_handover_enter(&bpf_ops_kernel);
atomic_swap_ptr(&bpf_ops, &bpf_ops_kernel);
bpf_ops_handover_exit();
break;
case MODULE_CMD_FINI:
/*
* While there is no reference counting for bpf callers,
* unload could at least in theory be done similarly to
* system call disestablishment. This should even be
* a little simpler:
*
* 1) replace op vector with stubs
* 2) post update to all cpus with xc
* 3) check that nobody is in bpf anymore
* (it's doubtful we'd want something like l_sysent,
* but we could do something like *signed* percpu
* counters. if the sum is 0, we're good).
* 4) if fail, unroll changes
*
* NOTE: change won't be atomic to the outside. some
* packets may be not captured even if unload is
* not successful. I think packet capture not working
* is a perfectly logical consequence of trying to
* disable packet capture.
*/
error = EOPNOTSUPP;
break;
default:
error = ENOTTY;
break;
}
return error;
}
|