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
|
/*-
* Copyright (c) 2007-2009 Bruce Simpson.
* Copyright (c) 1988 Stephen Deering.
* Copyright (c) 1992, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* Stephen Deering of Stanford University.
*
* 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.
* 4. 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.
*
* @(#)igmp.c 8.1 (Berkeley) 7/19/93
*/
/*
* Internet Group Management Protocol (IGMP) routines.
* [RFC1112, RFC2236, RFC3376]
*
* Written by Steve Deering, Stanford, May 1988.
* Modified by Rosen Sharma, Stanford, Aug 1994.
* Modified by Bill Fenner, Xerox PARC, Feb 1995.
* Modified to fully comply to IGMPv2 by Bill Fenner, Oct 1995.
* Significantly rewritten for IGMPv3, VIMAGE, and SMP by Bruce Simpson.
*
* MULTICAST Revision: 3.5.1.4
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/module.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/protosw.h>
#include <sys/kernel.h>
#include <sys/sysctl.h>
#include <sys/ktr.h>
#include <sys/condvar.h>
#include <net/if.h>
#include <net/netisr.h>
#include <net/vnet.h>
#include <netinet/in.h>
#include <netinet/in_var.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/ip_var.h>
#include <netinet/ip_options.h>
#include <netinet/igmp.h>
#include <netinet/igmp_var.h>
#include <machine/in_cksum.h>
#include <security/mac/mac_framework.h>
#ifndef KTR_IGMPV3
#define KTR_IGMPV3 KTR_INET
#endif
static struct igmp_ifinfo *
igi_alloc_locked(struct ifnet *);
static void igi_delete_locked(const struct ifnet *);
static void igmp_dispatch_queue(struct ifqueue *, int, const int);
static void igmp_fasttimo_vnet(void);
static void igmp_final_leave(struct in_multi *, struct igmp_ifinfo *);
static int igmp_handle_state_change(struct in_multi *,
struct igmp_ifinfo *);
static int igmp_initial_join(struct in_multi *, struct igmp_ifinfo *);
static int igmp_input_v1_query(struct ifnet *, const struct ip *,
const struct igmp *);
static int igmp_input_v2_query(struct ifnet *, const struct ip *,
const struct igmp *);
static int igmp_input_v3_query(struct ifnet *, const struct ip *,
/*const*/ struct igmpv3 *);
static int igmp_input_v3_group_query(struct in_multi *,
struct igmp_ifinfo *, int, /*const*/ struct igmpv3 *);
static int igmp_input_v1_report(struct ifnet *, /*const*/ struct ip *,
/*const*/ struct igmp *);
static int igmp_input_v2_report(struct ifnet *, /*const*/ struct ip *,
/*const*/ struct igmp *);
static void igmp_intr(struct mbuf *);
static int igmp_isgroupreported(const struct in_addr);
static struct mbuf *
igmp_ra_alloc(void);
#ifdef KTR
static char * igmp_rec_type_to_str(const int);
#endif
static void igmp_set_version(struct igmp_ifinfo *, const int);
static void igmp_slowtimo_vnet(void);
static int igmp_v1v2_queue_report(struct in_multi *, const int);
static void igmp_v1v2_process_group_timer(struct in_multi *, const int);
static void igmp_v1v2_process_querier_timers(struct igmp_ifinfo *);
static void igmp_v2_update_group(struct in_multi *, const int);
static void igmp_v3_cancel_link_timers(struct igmp_ifinfo *);
static void igmp_v3_dispatch_general_query(struct igmp_ifinfo *);
static struct mbuf *
igmp_v3_encap_report(struct ifnet *, struct mbuf *);
static int igmp_v3_enqueue_group_record(struct ifqueue *,
struct in_multi *, const int, const int, const int);
static int igmp_v3_enqueue_filter_change(struct ifqueue *,
struct in_multi *);
static void igmp_v3_process_group_timers(struct igmp_ifinfo *,
struct ifqueue *, struct ifqueue *, struct in_multi *,
const int);
static int igmp_v3_merge_state_changes(struct in_multi *,
struct ifqueue *);
static void igmp_v3_suppress_group_record(struct in_multi *);
static int sysctl_igmp_default_version(SYSCTL_HANDLER_ARGS);
static int sysctl_igmp_gsr(SYSCTL_HANDLER_ARGS);
static int sysctl_igmp_ifinfo(SYSCTL_HANDLER_ARGS);
static const struct netisr_handler igmp_nh = {
.nh_name = "igmp",
.nh_handler = igmp_intr,
.nh_proto = NETISR_IGMP,
.nh_policy = NETISR_POLICY_SOURCE,
};
/*
* System-wide globals.
*
* Unlocked access to these is OK, except for the global IGMP output
* queue. The IGMP subsystem lock ends up being system-wide for the moment,
* because all VIMAGEs have to share a global output queue, as netisrs
* themselves are not virtualized.
*
* Locking:
* * The permitted lock order is: IN_MULTI_LOCK, IGMP_LOCK, IF_ADDR_LOCK.
* Any may be taken independently; if any are held at the same
* time, the above lock order must be followed.
* * All output is delegated to the netisr.
* Now that Giant has been eliminated, the netisr may be inlined.
* * IN_MULTI_LOCK covers in_multi.
* * IGMP_LOCK covers igmp_ifinfo and any global variables in this file,
* including the output queue.
* * IF_ADDR_LOCK covers if_multiaddrs, which is used for a variety of
* per-link state iterators.
* * igmp_ifinfo is valid as long as PF_INET is attached to the interface,
* therefore it is not refcounted.
* We allow unlocked reads of igmp_ifinfo when accessed via in_multi.
*
* Reference counting
* * IGMP acquires its own reference every time an in_multi is passed to
* it and the group is being joined for the first time.
* * IGMP releases its reference(s) on in_multi in a deferred way,
* because the operations which process the release run as part of
* a loop whose control variables are directly affected by the release
* (that, and not recursing on the IF_ADDR_LOCK).
*
* VIMAGE: Each in_multi corresponds to an ifp, and each ifp corresponds
* to a vnet in ifp->if_vnet.
*
* SMPng: XXX We may potentially race operations on ifma_protospec.
* The problem is that we currently lack a clean way of taking the
* IF_ADDR_LOCK() between the ifnet and in layers w/o recursing,
* as anything which modifies ifma needs to be covered by that lock.
* So check for ifma_protospec being NULL before proceeding.
*/
struct mtx igmp_mtx;
struct mbuf *m_raopt; /* Router Alert option */
static MALLOC_DEFINE(M_IGMP, "igmp", "igmp state");
/*
* VIMAGE-wide globals.
*
* The IGMPv3 timers themselves need to run per-image, however,
* protosw timers run globally (see tcp).
* An ifnet can only be in one vimage at a time, and the loopback
* ifnet, loif, is itself virtualized.
* It would otherwise be possible to seriously hose IGMP state,
* and create inconsistencies in upstream multicast routing, if you have
* multiple VIMAGEs running on the same link joining different multicast
* groups, UNLESS the "primary IP address" is different. This is because
* IGMP for IPv4 does not force link-local addresses to be used for each
* node, unlike MLD for IPv6.
* Obviously the IGMPv3 per-interface state has per-vimage granularity
* also as a result.
*
* FUTURE: Stop using IFP_TO_IA/INADDR_ANY, and use source address selection
* policy to control the address used by IGMP on the link.
*/
static VNET_DEFINE(int, interface_timers_running); /* IGMPv3 general
* query response */
static VNET_DEFINE(int, state_change_timers_running); /* IGMPv3 state-change
* retransmit */
static VNET_DEFINE(int, current_state_timers_running); /* IGMPv1/v2 host
* report; IGMPv3 g/sg
* query response */
#define V_interface_timers_running VNET(interface_timers_running)
#define V_state_change_timers_running VNET(state_change_timers_running)
#define V_current_state_timers_running VNET(current_state_timers_running)
static VNET_DEFINE(LIST_HEAD(, igmp_ifinfo), igi_head);
static VNET_DEFINE(struct igmpstat, igmpstat) = {
.igps_version = IGPS_VERSION_3,
.igps_len = sizeof(struct igmpstat),
};
static VNET_DEFINE(struct timeval, igmp_gsrdelay) = {10, 0};
#define V_igi_head VNET(igi_head)
#define V_igmpstat VNET(igmpstat)
#define V_igmp_gsrdelay VNET(igmp_gsrdelay)
static VNET_DEFINE(int, igmp_recvifkludge) = 1;
static VNET_DEFINE(int, igmp_sendra) = 1;
static VNET_DEFINE(int, igmp_sendlocal) = 1;
static VNET_DEFINE(int, igmp_v1enable) = 1;
static VNET_DEFINE(int, igmp_v2enable) = 1;
static VNET_DEFINE(int, igmp_legacysupp);
static VNET_DEFINE(int, igmp_default_version) = IGMP_VERSION_3;
#define V_igmp_recvifkludge VNET(igmp_recvifkludge)
#define V_igmp_sendra VNET(igmp_sendra)
#define V_igmp_sendlocal VNET(igmp_sendlocal)
#define V_igmp_v1enable VNET(igmp_v1enable)
#define V_igmp_v2enable VNET(igmp_v2enable)
#define V_igmp_legacysupp VNET(igmp_legacysupp)
#define V_igmp_default_version VNET(igmp_default_version)
/*
* Virtualized sysctls.
*/
SYSCTL_VNET_STRUCT(_net_inet_igmp, IGMPCTL_STATS, stats, CTLFLAG_RW,
&VNET_NAME(igmpstat), igmpstat, "");
SYSCTL_VNET_INT(_net_inet_igmp, OID_AUTO, recvifkludge, CTLFLAG_RW,
&VNET_NAME(igmp_recvifkludge), 0,
"Rewrite IGMPv1/v2 reports from 0.0.0.0 to contain subnet address");
SYSCTL_VNET_INT(_net_inet_igmp, OID_AUTO, sendra, CTLFLAG_RW,
&VNET_NAME(igmp_sendra), 0,
"Send IP Router Alert option in IGMPv2/v3 messages");
SYSCTL_VNET_INT(_net_inet_igmp, OID_AUTO, sendlocal, CTLFLAG_RW,
&VNET_NAME(igmp_sendlocal), 0,
"Send IGMP membership reports for 224.0.0.0/24 groups");
SYSCTL_VNET_INT(_net_inet_igmp, OID_AUTO, v1enable, CTLFLAG_RW,
&VNET_NAME(igmp_v1enable), 0,
"Enable backwards compatibility with IGMPv1");
SYSCTL_VNET_INT(_net_inet_igmp, OID_AUTO, v2enable, CTLFLAG_RW,
&VNET_NAME(igmp_v2enable), 0,
"Enable backwards compatibility with IGMPv2");
SYSCTL_VNET_INT(_net_inet_igmp, OID_AUTO, legacysupp, CTLFLAG_RW,
&VNET_NAME(igmp_legacysupp), 0,
"Allow v1/v2 reports to suppress v3 group responses");
SYSCTL_VNET_PROC(_net_inet_igmp, OID_AUTO, default_version,
CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_MPSAFE,
&VNET_NAME(igmp_default_version), 0, sysctl_igmp_default_version, "I",
"Default version of IGMP to run on each interface");
SYSCTL_VNET_PROC(_net_inet_igmp, OID_AUTO, gsrdelay,
CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_MPSAFE,
&VNET_NAME(igmp_gsrdelay.tv_sec), 0, sysctl_igmp_gsr, "I",
"Rate limit for IGMPv3 Group-and-Source queries in seconds");
/*
* Non-virtualized sysctls.
*/
static SYSCTL_NODE(_net_inet_igmp, OID_AUTO, ifinfo,
CTLFLAG_RD | CTLFLAG_MPSAFE, sysctl_igmp_ifinfo,
"Per-interface IGMPv3 state");
static __inline void
igmp_save_context(struct mbuf *m, struct ifnet *ifp)
{
#ifdef VIMAGE
m->m_pkthdr.header = ifp->if_vnet;
#endif /* VIMAGE */
m->m_pkthdr.flowid = ifp->if_index;
}
static __inline void
igmp_scrub_context(struct mbuf *m)
{
m->m_pkthdr.header = NULL;
m->m_pkthdr.flowid = 0;
}
#ifdef KTR
static __inline char *
inet_ntoa_haddr(in_addr_t haddr)
{
struct in_addr ia;
ia.s_addr = htonl(haddr);
return (inet_ntoa(ia));
}
#endif
/*
* Restore context from a queued IGMP output chain.
* Return saved ifindex.
*
* VIMAGE: The assertion is there to make sure that we
* actually called CURVNET_SET() with what's in the mbuf chain.
*/
static __inline uint32_t
igmp_restore_context(struct mbuf *m)
{
#ifdef notyet
#if defined(VIMAGE) && defined(INVARIANTS)
KASSERT(curvnet == (m->m_pkthdr.header),
("%s: called when curvnet was not restored", __func__));
#endif
#endif
return (m->m_pkthdr.flowid);
}
/*
* Retrieve or set default IGMP version.
*
* VIMAGE: Assume curvnet set by caller.
* SMPng: NOTE: Serialized by IGMP lock.
*/
static int
sysctl_igmp_default_version(SYSCTL_HANDLER_ARGS)
{
int error;
int new;
error = sysctl_wire_old_buffer(req, sizeof(int));
if (error)
return (error);
IGMP_LOCK();
new = V_igmp_default_version;
error = sysctl_handle_int(oidp, &new, 0, req);
if (error || !req->newptr)
goto out_locked;
if (new < IGMP_VERSION_1 || new > IGMP_VERSION_3) {
error = EINVAL;
goto out_locked;
}
CTR2(KTR_IGMPV3, "change igmp_default_version from %d to %d",
V_igmp_default_version, new);
V_igmp_default_version = new;
out_locked:
IGMP_UNLOCK();
return (error);
}
/*
* Retrieve or set threshold between group-source queries in seconds.
*
* VIMAGE: Assume curvnet set by caller.
* SMPng: NOTE: Serialized by IGMP lock.
*/
static int
sysctl_igmp_gsr(SYSCTL_HANDLER_ARGS)
{
int error;
int i;
error = sysctl_wire_old_buffer(req, sizeof(int));
if (error)
return (error);
IGMP_LOCK();
i = V_igmp_gsrdelay.tv_sec;
error = sysctl_handle_int(oidp, &i, 0, req);
if (error || !req->newptr)
goto out_locked;
if (i < -1 || i >= 60) {
error = EINVAL;
goto out_locked;
}
CTR2(KTR_IGMPV3, "change igmp_gsrdelay from %d to %d",
V_igmp_gsrdelay.tv_sec, i);
V_igmp_gsrdelay.tv_sec = i;
out_locked:
IGMP_UNLOCK();
return (error);
}
/*
* Expose struct igmp_ifinfo to userland, keyed by ifindex.
* For use by ifmcstat(8).
*
* SMPng: NOTE: Does an unlocked ifindex space read.
* VIMAGE: Assume curvnet set by caller. The node handler itself
* is not directly virtualized.
*/
static int
sysctl_igmp_ifinfo(SYSCTL_HANDLER_ARGS)
{
int *name;
int error;
u_int namelen;
struct ifnet *ifp;
struct igmp_ifinfo *igi;
name = (int *)arg1;
namelen = arg2;
if (req->newptr != NULL)
return (EPERM);
if (namelen != 1)
return (EINVAL);
error = sysctl_wire_old_buffer(req, sizeof(struct igmp_ifinfo));
if (error)
return (error);
IN_MULTI_LOCK();
IGMP_LOCK();
if (name[0] <= 0 || name[0] > V_if_index) {
error = ENOENT;
goto out_locked;
}
error = ENOENT;
ifp = ifnet_byindex(name[0]);
if (ifp == NULL)
goto out_locked;
LIST_FOREACH(igi, &V_igi_head, igi_link) {
if (ifp == igi->igi_ifp) {
error = SYSCTL_OUT(req, igi,
sizeof(struct igmp_ifinfo));
break;
}
}
out_locked:
IGMP_UNLOCK();
IN_MULTI_UNLOCK();
return (error);
}
/*
* Dispatch an entire queue of pending packet chains
* using the netisr.
* VIMAGE: Assumes the vnet pointer has been set.
*/
static void
igmp_dispatch_queue(struct ifqueue *ifq, int limit, const int loop)
{
struct mbuf *m;
for (;;) {
_IF_DEQUEUE(ifq, m);
if (m == NULL)
break;
CTR3(KTR_IGMPV3, "%s: dispatch %p from %p", __func__, ifq, m);
if (loop)
m->m_flags |= M_IGMP_LOOP;
netisr_dispatch(NETISR_IGMP, m);
if (--limit == 0)
break;
}
}
/*
* Filter outgoing IGMP report state by group.
*
* Reports are ALWAYS suppressed for ALL-HOSTS (224.0.0.1).
* If the net.inet.igmp.sendlocal sysctl is 0, then IGMP reports are
* disabled for all groups in the 224.0.0.0/24 link-local scope. However,
* this may break certain IGMP snooping switches which rely on the old
* report behaviour.
*
* Return zero if the given group is one for which IGMP reports
* should be suppressed, or non-zero if reports should be issued.
*/
static __inline int
igmp_isgroupreported(const struct in_addr addr)
{
if (in_allhosts(addr) ||
((!V_igmp_sendlocal && IN_LOCAL_GROUP(ntohl(addr.s_addr)))))
return (0);
return (1);
}
/*
* Construct a Router Alert option to use in outgoing packets.
*/
static struct mbuf *
igmp_ra_alloc(void)
{
struct mbuf *m;
struct ipoption *p;
m = m_get(M_WAITOK, MT_DATA);
p = mtod(m, struct ipoption *);
p->ipopt_dst.s_addr = INADDR_ANY;
p->ipopt_list[0] = IPOPT_RA; /* Router Alert Option */
p->ipopt_list[1] = 0x04; /* 4 bytes long */
p->ipopt_list[2] = IPOPT_EOL; /* End of IP option list */
p->ipopt_list[3] = 0x00; /* pad byte */
m->m_len = sizeof(p->ipopt_dst) + p->ipopt_list[1];
return (m);
}
/*
* Attach IGMP when PF_INET is attached to an interface.
*/
struct igmp_ifinfo *
igmp_domifattach(struct ifnet *ifp)
{
struct igmp_ifinfo *igi;
CTR3(KTR_IGMPV3, "%s: called for ifp %p(%s)",
__func__, ifp, ifp->if_xname);
IGMP_LOCK();
igi = igi_alloc_locked(ifp);
if (!(ifp->if_flags & IFF_MULTICAST))
igi->igi_flags |= IGIF_SILENT;
IGMP_UNLOCK();
return (igi);
}
/*
* VIMAGE: assume curvnet set by caller.
*/
static struct igmp_ifinfo *
igi_alloc_locked(/*const*/ struct ifnet *ifp)
{
struct igmp_ifinfo *igi;
IGMP_LOCK_ASSERT();
igi = malloc(sizeof(struct igmp_ifinfo), M_IGMP, M_NOWAIT|M_ZERO);
if (igi == NULL)
goto out;
igi->igi_ifp = ifp;
igi->igi_version = V_igmp_default_version;
igi->igi_flags = 0;
igi->igi_rv = IGMP_RV_INIT;
igi->igi_qi = IGMP_QI_INIT;
igi->igi_qri = IGMP_QRI_INIT;
igi->igi_uri = IGMP_URI_INIT;
SLIST_INIT(&igi->igi_relinmhead);
/*
* Responses to general queries are subject to bounds.
*/
IFQ_SET_MAXLEN(&igi->igi_gq, IGMP_MAX_RESPONSE_PACKETS);
LIST_INSERT_HEAD(&V_igi_head, igi, igi_link);
CTR2(KTR_IGMPV3, "allocate igmp_ifinfo for ifp %p(%s)",
ifp, ifp->if_xname);
out:
return (igi);
}
/*
* Hook for ifdetach.
*
* NOTE: Some finalization tasks need to run before the protocol domain
* is detached, but also before the link layer does its cleanup.
*
* SMPNG: igmp_ifdetach() needs to take IF_ADDR_LOCK().
* XXX This is also bitten by unlocked ifma_protospec access.
*/
void
igmp_ifdetach(struct ifnet *ifp)
{
struct igmp_ifinfo *igi;
struct ifmultiaddr *ifma;
struct in_multi *inm, *tinm;
CTR3(KTR_IGMPV3, "%s: called for ifp %p(%s)", __func__, ifp,
ifp->if_xname);
IGMP_LOCK();
igi = ((struct in_ifinfo *)ifp->if_afdata[AF_INET])->ii_igmp;
if (igi->igi_version == IGMP_VERSION_3) {
IF_ADDR_RLOCK(ifp);
TAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
if (ifma->ifma_addr->sa_family != AF_INET ||
ifma->ifma_protospec == NULL)
continue;
#if 0
KASSERT(ifma->ifma_protospec != NULL,
("%s: ifma_protospec is NULL", __func__));
#endif
inm = (struct in_multi *)ifma->ifma_protospec;
if (inm->inm_state == IGMP_LEAVING_MEMBER) {
SLIST_INSERT_HEAD(&igi->igi_relinmhead,
inm, inm_nrele);
}
inm_clear_recorded(inm);
}
IF_ADDR_RUNLOCK(ifp);
/*
* Free the in_multi reference(s) for this IGMP lifecycle.
*/
SLIST_FOREACH_SAFE(inm, &igi->igi_relinmhead, inm_nrele,
tinm) {
SLIST_REMOVE_HEAD(&igi->igi_relinmhead, inm_nrele);
inm_release_locked(inm);
}
}
IGMP_UNLOCK();
}
/*
* Hook for domifdetach.
*/
void
igmp_domifdetach(struct ifnet *ifp)
{
struct igmp_ifinfo *igi;
CTR3(KTR_IGMPV3, "%s: called for ifp %p(%s)",
__func__, ifp, ifp->if_xname);
IGMP_LOCK();
igi = ((struct in_ifinfo *)ifp->if_afdata[AF_INET])->ii_igmp;
igi_delete_locked(ifp);
IGMP_UNLOCK();
}
static void
igi_delete_locked(const struct ifnet *ifp)
{
struct igmp_ifinfo *igi, *tigi;
CTR3(KTR_IGMPV3, "%s: freeing igmp_ifinfo for ifp %p(%s)",
__func__, ifp, ifp->if_xname);
IGMP_LOCK_ASSERT();
LIST_FOREACH_SAFE(igi, &V_igi_head, igi_link, tigi) {
if (igi->igi_ifp == ifp) {
/*
* Free deferred General Query responses.
*/
_IF_DRAIN(&igi->igi_gq);
LIST_REMOVE(igi, igi_link);
KASSERT(SLIST_EMPTY(&igi->igi_relinmhead),
("%s: there are dangling in_multi references",
__func__));
free(igi, M_IGMP);
return;
}
}
#ifdef INVARIANTS
panic("%s: igmp_ifinfo not found for ifp %p\n", __func__, ifp);
#endif
}
/*
* Process a received IGMPv1 query.
* Return non-zero if the message should be dropped.
*
* VIMAGE: The curvnet pointer is derived from the input ifp.
*/
static int
igmp_input_v1_query(struct ifnet *ifp, const struct ip *ip,
const struct igmp *igmp)
{
struct ifmultiaddr *ifma;
struct igmp_ifinfo *igi;
struct in_multi *inm;
/*
* IGMPv1 Host Mmembership Queries SHOULD always be addressed to
* 224.0.0.1. They are always treated as General Queries.
* igmp_group is always ignored. Do not drop it as a userland
* daemon may wish to see it.
* XXX SMPng: unlocked increments in igmpstat assumed atomic.
*/
if (!in_allhosts(ip->ip_dst) || !in_nullhost(igmp->igmp_group)) {
IGMPSTAT_INC(igps_rcv_badqueries);
return (0);
}
IGMPSTAT_INC(igps_rcv_gen_queries);
IN_MULTI_LOCK();
IGMP_LOCK();
igi = ((struct in_ifinfo *)ifp->if_afdata[AF_INET])->ii_igmp;
KASSERT(igi != NULL, ("%s: no igmp_ifinfo for ifp %p", __func__, ifp));
if (igi->igi_flags & IGIF_LOOPBACK) {
CTR2(KTR_IGMPV3, "ignore v1 query on IGIF_LOOPBACK ifp %p(%s)",
ifp, ifp->if_xname);
goto out_locked;
}
/*
* Switch to IGMPv1 host compatibility mode.
*/
igmp_set_version(igi, IGMP_VERSION_1);
CTR2(KTR_IGMPV3, "process v1 query on ifp %p(%s)", ifp, ifp->if_xname);
/*
* Start the timers in all of our group records
* for the interface on which the query arrived,
* except those which are already running.
*/
IF_ADDR_RLOCK(ifp);
TAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
if (ifma->ifma_addr->sa_family != AF_INET ||
ifma->ifma_protospec == NULL)
continue;
inm = (struct in_multi *)ifma->ifma_protospec;
if (inm->inm_timer != 0)
continue;
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
break;
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
case IGMP_REPORTING_MEMBER:
case IGMP_IDLE_MEMBER:
case IGMP_LAZY_MEMBER:
case IGMP_SLEEPING_MEMBER:
case IGMP_AWAKENING_MEMBER:
inm->inm_state = IGMP_REPORTING_MEMBER;
inm->inm_timer = IGMP_RANDOM_DELAY(
IGMP_V1V2_MAX_RI * PR_FASTHZ);
V_current_state_timers_running = 1;
break;
case IGMP_LEAVING_MEMBER:
break;
}
}
IF_ADDR_RUNLOCK(ifp);
out_locked:
IGMP_UNLOCK();
IN_MULTI_UNLOCK();
return (0);
}
/*
* Process a received IGMPv2 general or group-specific query.
*/
static int
igmp_input_v2_query(struct ifnet *ifp, const struct ip *ip,
const struct igmp *igmp)
{
struct ifmultiaddr *ifma;
struct igmp_ifinfo *igi;
struct in_multi *inm;
int is_general_query;
uint16_t timer;
is_general_query = 0;
/*
* Validate address fields upfront.
* XXX SMPng: unlocked increments in igmpstat assumed atomic.
*/
if (in_nullhost(igmp->igmp_group)) {
/*
* IGMPv2 General Query.
* If this was not sent to the all-hosts group, ignore it.
*/
if (!in_allhosts(ip->ip_dst))
return (0);
IGMPSTAT_INC(igps_rcv_gen_queries);
is_general_query = 1;
} else {
/* IGMPv2 Group-Specific Query. */
IGMPSTAT_INC(igps_rcv_group_queries);
}
IN_MULTI_LOCK();
IGMP_LOCK();
igi = ((struct in_ifinfo *)ifp->if_afdata[AF_INET])->ii_igmp;
KASSERT(igi != NULL, ("%s: no igmp_ifinfo for ifp %p", __func__, ifp));
if (igi->igi_flags & IGIF_LOOPBACK) {
CTR2(KTR_IGMPV3, "ignore v2 query on IGIF_LOOPBACK ifp %p(%s)",
ifp, ifp->if_xname);
goto out_locked;
}
/*
* Ignore v2 query if in v1 Compatibility Mode.
*/
if (igi->igi_version == IGMP_VERSION_1)
goto out_locked;
igmp_set_version(igi, IGMP_VERSION_2);
timer = igmp->igmp_code * PR_FASTHZ / IGMP_TIMER_SCALE;
if (timer == 0)
timer = 1;
if (is_general_query) {
/*
* For each reporting group joined on this
* interface, kick the report timer.
*/
CTR2(KTR_IGMPV3, "process v2 general query on ifp %p(%s)",
ifp, ifp->if_xname);
IF_ADDR_RLOCK(ifp);
TAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
if (ifma->ifma_addr->sa_family != AF_INET ||
ifma->ifma_protospec == NULL)
continue;
inm = (struct in_multi *)ifma->ifma_protospec;
igmp_v2_update_group(inm, timer);
}
IF_ADDR_RUNLOCK(ifp);
} else {
/*
* Group-specific IGMPv2 query, we need only
* look up the single group to process it.
*/
inm = inm_lookup(ifp, igmp->igmp_group);
if (inm != NULL) {
CTR3(KTR_IGMPV3, "process v2 query %s on ifp %p(%s)",
inet_ntoa(igmp->igmp_group), ifp, ifp->if_xname);
igmp_v2_update_group(inm, timer);
}
}
out_locked:
IGMP_UNLOCK();
IN_MULTI_UNLOCK();
return (0);
}
/*
* Update the report timer on a group in response to an IGMPv2 query.
*
* If we are becoming the reporting member for this group, start the timer.
* If we already are the reporting member for this group, and timer is
* below the threshold, reset it.
*
* We may be updating the group for the first time since we switched
* to IGMPv3. If we are, then we must clear any recorded source lists,
* and transition to REPORTING state; the group timer is overloaded
* for group and group-source query responses.
*
* Unlike IGMPv3, the delay per group should be jittered
* to avoid bursts of IGMPv2 reports.
*/
static void
igmp_v2_update_group(struct in_multi *inm, const int timer)
{
CTR4(KTR_IGMPV3, "%s: %s/%s timer=%d", __func__,
inet_ntoa(inm->inm_addr), inm->inm_ifp->if_xname, timer);
IN_MULTI_LOCK_ASSERT();
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
break;
case IGMP_REPORTING_MEMBER:
if (inm->inm_timer != 0 &&
inm->inm_timer <= timer) {
CTR1(KTR_IGMPV3, "%s: REPORTING and timer running, "
"skipping.", __func__);
break;
}
/* FALLTHROUGH */
case IGMP_SG_QUERY_PENDING_MEMBER:
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_IDLE_MEMBER:
case IGMP_LAZY_MEMBER:
case IGMP_AWAKENING_MEMBER:
CTR1(KTR_IGMPV3, "%s: ->REPORTING", __func__);
inm->inm_state = IGMP_REPORTING_MEMBER;
inm->inm_timer = IGMP_RANDOM_DELAY(timer);
V_current_state_timers_running = 1;
break;
case IGMP_SLEEPING_MEMBER:
CTR1(KTR_IGMPV3, "%s: ->AWAKENING", __func__);
inm->inm_state = IGMP_AWAKENING_MEMBER;
break;
case IGMP_LEAVING_MEMBER:
break;
}
}
/*
* Process a received IGMPv3 general, group-specific or
* group-and-source-specific query.
* Assumes m has already been pulled up to the full IGMP message length.
* Return 0 if successful, otherwise an appropriate error code is returned.
*/
static int
igmp_input_v3_query(struct ifnet *ifp, const struct ip *ip,
/*const*/ struct igmpv3 *igmpv3)
{
struct igmp_ifinfo *igi;
struct in_multi *inm;
int is_general_query;
uint32_t maxresp, nsrc, qqi;
uint16_t timer;
uint8_t qrv;
is_general_query = 0;
CTR2(KTR_IGMPV3, "process v3 query on ifp %p(%s)", ifp, ifp->if_xname);
maxresp = igmpv3->igmp_code; /* in 1/10ths of a second */
if (maxresp >= 128) {
maxresp = IGMP_MANT(igmpv3->igmp_code) <<
(IGMP_EXP(igmpv3->igmp_code) + 3);
}
/*
* Robustness must never be less than 2 for on-wire IGMPv3.
* FUTURE: Check if ifp has IGIF_LOOPBACK set, as we will make
* an exception for interfaces whose IGMPv3 state changes
* are redirected to loopback (e.g. MANET).
*/
qrv = IGMP_QRV(igmpv3->igmp_misc);
if (qrv < 2) {
CTR3(KTR_IGMPV3, "%s: clamping qrv %d to %d", __func__,
qrv, IGMP_RV_INIT);
qrv = IGMP_RV_INIT;
}
qqi = igmpv3->igmp_qqi;
if (qqi >= 128) {
qqi = IGMP_MANT(igmpv3->igmp_qqi) <<
(IGMP_EXP(igmpv3->igmp_qqi) + 3);
}
timer = maxresp * PR_FASTHZ / IGMP_TIMER_SCALE;
if (timer == 0)
timer = 1;
nsrc = ntohs(igmpv3->igmp_numsrc);
/*
* Validate address fields and versions upfront before
* accepting v3 query.
* XXX SMPng: Unlocked access to igmpstat counters here.
*/
if (in_nullhost(igmpv3->igmp_group)) {
/*
* IGMPv3 General Query.
*
* General Queries SHOULD be directed to 224.0.0.1.
* A general query with a source list has undefined
* behaviour; discard it.
*/
IGMPSTAT_INC(igps_rcv_gen_queries);
if (!in_allhosts(ip->ip_dst) || nsrc > 0) {
IGMPSTAT_INC(igps_rcv_badqueries);
return (0);
}
is_general_query = 1;
} else {
/* Group or group-source specific query. */
if (nsrc == 0)
IGMPSTAT_INC(igps_rcv_group_queries);
else
IGMPSTAT_INC(igps_rcv_gsr_queries);
}
IN_MULTI_LOCK();
IGMP_LOCK();
igi = ((struct in_ifinfo *)ifp->if_afdata[AF_INET])->ii_igmp;
KASSERT(igi != NULL, ("%s: no igmp_ifinfo for ifp %p", __func__, ifp));
if (igi->igi_flags & IGIF_LOOPBACK) {
CTR2(KTR_IGMPV3, "ignore v3 query on IGIF_LOOPBACK ifp %p(%s)",
ifp, ifp->if_xname);
goto out_locked;
}
/*
* Discard the v3 query if we're in Compatibility Mode.
* The RFC is not obviously worded that hosts need to stay in
* compatibility mode until the Old Version Querier Present
* timer expires.
*/
if (igi->igi_version != IGMP_VERSION_3) {
CTR3(KTR_IGMPV3, "ignore v3 query in v%d mode on ifp %p(%s)",
igi->igi_version, ifp, ifp->if_xname);
goto out_locked;
}
igmp_set_version(igi, IGMP_VERSION_3);
igi->igi_rv = qrv;
igi->igi_qi = qqi;
igi->igi_qri = maxresp;
CTR4(KTR_IGMPV3, "%s: qrv %d qi %d qri %d", __func__, qrv, qqi,
maxresp);
if (is_general_query) {
/*
* Schedule a current-state report on this ifp for
* all groups, possibly containing source lists.
* If there is a pending General Query response
* scheduled earlier than the selected delay, do
* not schedule any other reports.
* Otherwise, reset the interface timer.
*/
CTR2(KTR_IGMPV3, "process v3 general query on ifp %p(%s)",
ifp, ifp->if_xname);
if (igi->igi_v3_timer == 0 || igi->igi_v3_timer >= timer) {
igi->igi_v3_timer = IGMP_RANDOM_DELAY(timer);
V_interface_timers_running = 1;
}
} else {
/*
* Group-source-specific queries are throttled on
* a per-group basis to defeat denial-of-service attempts.
* Queries for groups we are not a member of on this
* link are simply ignored.
*/
inm = inm_lookup(ifp, igmpv3->igmp_group);
if (inm == NULL)
goto out_locked;
if (nsrc > 0) {
if (!ratecheck(&inm->inm_lastgsrtv,
&V_igmp_gsrdelay)) {
CTR1(KTR_IGMPV3, "%s: GS query throttled.",
__func__);
IGMPSTAT_INC(igps_drop_gsr_queries);
goto out_locked;
}
}
CTR3(KTR_IGMPV3, "process v3 %s query on ifp %p(%s)",
inet_ntoa(igmpv3->igmp_group), ifp, ifp->if_xname);
/*
* If there is a pending General Query response
* scheduled sooner than the selected delay, no
* further report need be scheduled.
* Otherwise, prepare to respond to the
* group-specific or group-and-source query.
*/
if (igi->igi_v3_timer == 0 || igi->igi_v3_timer >= timer)
igmp_input_v3_group_query(inm, igi, timer, igmpv3);
}
out_locked:
IGMP_UNLOCK();
IN_MULTI_UNLOCK();
return (0);
}
/*
* Process a recieved IGMPv3 group-specific or group-and-source-specific
* query.
* Return <0 if any error occured. Currently this is ignored.
*/
static int
igmp_input_v3_group_query(struct in_multi *inm, struct igmp_ifinfo *igi,
int timer, /*const*/ struct igmpv3 *igmpv3)
{
int retval;
uint16_t nsrc;
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
retval = 0;
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
case IGMP_SLEEPING_MEMBER:
case IGMP_LAZY_MEMBER:
case IGMP_AWAKENING_MEMBER:
case IGMP_IDLE_MEMBER:
case IGMP_LEAVING_MEMBER:
return (retval);
break;
case IGMP_REPORTING_MEMBER:
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
break;
}
nsrc = ntohs(igmpv3->igmp_numsrc);
/*
* Deal with group-specific queries upfront.
* If any group query is already pending, purge any recorded
* source-list state if it exists, and schedule a query response
* for this group-specific query.
*/
if (nsrc == 0) {
if (inm->inm_state == IGMP_G_QUERY_PENDING_MEMBER ||
inm->inm_state == IGMP_SG_QUERY_PENDING_MEMBER) {
inm_clear_recorded(inm);
timer = min(inm->inm_timer, timer);
}
inm->inm_state = IGMP_G_QUERY_PENDING_MEMBER;
inm->inm_timer = IGMP_RANDOM_DELAY(timer);
V_current_state_timers_running = 1;
return (retval);
}
/*
* Deal with the case where a group-and-source-specific query has
* been received but a group-specific query is already pending.
*/
if (inm->inm_state == IGMP_G_QUERY_PENDING_MEMBER) {
timer = min(inm->inm_timer, timer);
inm->inm_timer = IGMP_RANDOM_DELAY(timer);
V_current_state_timers_running = 1;
return (retval);
}
/*
* Finally, deal with the case where a group-and-source-specific
* query has been received, where a response to a previous g-s-r
* query exists, or none exists.
* In this case, we need to parse the source-list which the Querier
* has provided us with and check if we have any source list filter
* entries at T1 for these sources. If we do not, there is no need
* schedule a report and the query may be dropped.
* If we do, we must record them and schedule a current-state
* report for those sources.
* FIXME: Handling source lists larger than 1 mbuf requires that
* we pass the mbuf chain pointer down to this function, and use
* m_getptr() to walk the chain.
*/
if (inm->inm_nsrc > 0) {
const struct in_addr *ap;
int i, nrecorded;
ap = (const struct in_addr *)(igmpv3 + 1);
nrecorded = 0;
for (i = 0; i < nsrc; i++, ap++) {
retval = inm_record_source(inm, ap->s_addr);
if (retval < 0)
break;
nrecorded += retval;
}
if (nrecorded > 0) {
CTR1(KTR_IGMPV3,
"%s: schedule response to SG query", __func__);
inm->inm_state = IGMP_SG_QUERY_PENDING_MEMBER;
inm->inm_timer = IGMP_RANDOM_DELAY(timer);
V_current_state_timers_running = 1;
}
}
return (retval);
}
/*
* Process a received IGMPv1 host membership report.
*
* NOTE: 0.0.0.0 workaround breaks const correctness.
*/
static int
igmp_input_v1_report(struct ifnet *ifp, /*const*/ struct ip *ip,
/*const*/ struct igmp *igmp)
{
struct in_ifaddr *ia;
struct in_multi *inm;
IGMPSTAT_INC(igps_rcv_reports);
if (ifp->if_flags & IFF_LOOPBACK)
return (0);
if (!IN_MULTICAST(ntohl(igmp->igmp_group.s_addr)) ||
!in_hosteq(igmp->igmp_group, ip->ip_dst)) {
IGMPSTAT_INC(igps_rcv_badreports);
return (EINVAL);
}
/*
* RFC 3376, Section 4.2.13, 9.2, 9.3:
* Booting clients may use the source address 0.0.0.0. Some
* IGMP daemons may not know how to use IP_RECVIF to determine
* the interface upon which this message was received.
* Replace 0.0.0.0 with the subnet address if told to do so.
*/
if (V_igmp_recvifkludge && in_nullhost(ip->ip_src)) {
IFP_TO_IA(ifp, ia);
if (ia != NULL) {
ip->ip_src.s_addr = htonl(ia->ia_subnet);
ifa_free(&ia->ia_ifa);
}
}
CTR3(KTR_IGMPV3, "process v1 report %s on ifp %p(%s)",
inet_ntoa(igmp->igmp_group), ifp, ifp->if_xname);
/*
* IGMPv1 report suppression.
* If we are a member of this group, and our membership should be
* reported, stop our group timer and transition to the 'lazy' state.
*/
IN_MULTI_LOCK();
inm = inm_lookup(ifp, igmp->igmp_group);
if (inm != NULL) {
struct igmp_ifinfo *igi;
igi = inm->inm_igi;
if (igi == NULL) {
KASSERT(igi != NULL,
("%s: no igi for ifp %p", __func__, ifp));
goto out_locked;
}
IGMPSTAT_INC(igps_rcv_ourreports);
/*
* If we are in IGMPv3 host mode, do not allow the
* other host's IGMPv1 report to suppress our reports
* unless explicitly configured to do so.
*/
if (igi->igi_version == IGMP_VERSION_3) {
if (V_igmp_legacysupp)
igmp_v3_suppress_group_record(inm);
goto out_locked;
}
inm->inm_timer = 0;
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
break;
case IGMP_IDLE_MEMBER:
case IGMP_LAZY_MEMBER:
case IGMP_AWAKENING_MEMBER:
CTR3(KTR_IGMPV3,
"report suppressed for %s on ifp %p(%s)",
inet_ntoa(igmp->igmp_group), ifp, ifp->if_xname);
case IGMP_SLEEPING_MEMBER:
inm->inm_state = IGMP_SLEEPING_MEMBER;
break;
case IGMP_REPORTING_MEMBER:
CTR3(KTR_IGMPV3,
"report suppressed for %s on ifp %p(%s)",
inet_ntoa(igmp->igmp_group), ifp, ifp->if_xname);
if (igi->igi_version == IGMP_VERSION_1)
inm->inm_state = IGMP_LAZY_MEMBER;
else if (igi->igi_version == IGMP_VERSION_2)
inm->inm_state = IGMP_SLEEPING_MEMBER;
break;
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
case IGMP_LEAVING_MEMBER:
break;
}
}
out_locked:
IN_MULTI_UNLOCK();
return (0);
}
/*
* Process a received IGMPv2 host membership report.
*
* NOTE: 0.0.0.0 workaround breaks const correctness.
*/
static int
igmp_input_v2_report(struct ifnet *ifp, /*const*/ struct ip *ip,
/*const*/ struct igmp *igmp)
{
struct in_ifaddr *ia;
struct in_multi *inm;
/*
* Make sure we don't hear our own membership report. Fast
* leave requires knowing that we are the only member of a
* group.
*/
IFP_TO_IA(ifp, ia);
if (ia != NULL && in_hosteq(ip->ip_src, IA_SIN(ia)->sin_addr)) {
ifa_free(&ia->ia_ifa);
return (0);
}
IGMPSTAT_INC(igps_rcv_reports);
if (ifp->if_flags & IFF_LOOPBACK) {
if (ia != NULL)
ifa_free(&ia->ia_ifa);
return (0);
}
if (!IN_MULTICAST(ntohl(igmp->igmp_group.s_addr)) ||
!in_hosteq(igmp->igmp_group, ip->ip_dst)) {
if (ia != NULL)
ifa_free(&ia->ia_ifa);
IGMPSTAT_INC(igps_rcv_badreports);
return (EINVAL);
}
/*
* RFC 3376, Section 4.2.13, 9.2, 9.3:
* Booting clients may use the source address 0.0.0.0. Some
* IGMP daemons may not know how to use IP_RECVIF to determine
* the interface upon which this message was received.
* Replace 0.0.0.0 with the subnet address if told to do so.
*/
if (V_igmp_recvifkludge && in_nullhost(ip->ip_src)) {
if (ia != NULL)
ip->ip_src.s_addr = htonl(ia->ia_subnet);
}
if (ia != NULL)
ifa_free(&ia->ia_ifa);
CTR3(KTR_IGMPV3, "process v2 report %s on ifp %p(%s)",
inet_ntoa(igmp->igmp_group), ifp, ifp->if_xname);
/*
* IGMPv2 report suppression.
* If we are a member of this group, and our membership should be
* reported, and our group timer is pending or about to be reset,
* stop our group timer by transitioning to the 'lazy' state.
*/
IN_MULTI_LOCK();
inm = inm_lookup(ifp, igmp->igmp_group);
if (inm != NULL) {
struct igmp_ifinfo *igi;
igi = inm->inm_igi;
KASSERT(igi != NULL, ("%s: no igi for ifp %p", __func__, ifp));
IGMPSTAT_INC(igps_rcv_ourreports);
/*
* If we are in IGMPv3 host mode, do not allow the
* other host's IGMPv1 report to suppress our reports
* unless explicitly configured to do so.
*/
if (igi->igi_version == IGMP_VERSION_3) {
if (V_igmp_legacysupp)
igmp_v3_suppress_group_record(inm);
goto out_locked;
}
inm->inm_timer = 0;
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
case IGMP_SLEEPING_MEMBER:
break;
case IGMP_REPORTING_MEMBER:
case IGMP_IDLE_MEMBER:
case IGMP_AWAKENING_MEMBER:
CTR3(KTR_IGMPV3,
"report suppressed for %s on ifp %p(%s)",
inet_ntoa(igmp->igmp_group), ifp, ifp->if_xname);
case IGMP_LAZY_MEMBER:
inm->inm_state = IGMP_LAZY_MEMBER;
break;
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
case IGMP_LEAVING_MEMBER:
break;
}
}
out_locked:
IN_MULTI_UNLOCK();
return (0);
}
void
igmp_input(struct mbuf *m, int off)
{
int iphlen;
struct ifnet *ifp;
struct igmp *igmp;
struct ip *ip;
int igmplen;
int minlen;
int queryver;
CTR3(KTR_IGMPV3, "%s: called w/mbuf (%p,%d)", __func__, m, off);
ifp = m->m_pkthdr.rcvif;
IGMPSTAT_INC(igps_rcv_total);
ip = mtod(m, struct ip *);
iphlen = off;
igmplen = ntohs(ip->ip_len) - off;
/*
* Validate lengths.
*/
if (igmplen < IGMP_MINLEN) {
IGMPSTAT_INC(igps_rcv_tooshort);
m_freem(m);
return;
}
/*
* Always pullup to the minimum size for v1/v2 or v3
* to amortize calls to m_pullup().
*/
minlen = iphlen;
if (igmplen >= IGMP_V3_QUERY_MINLEN)
minlen += IGMP_V3_QUERY_MINLEN;
else
minlen += IGMP_MINLEN;
if ((m->m_flags & M_EXT || m->m_len < minlen) &&
(m = m_pullup(m, minlen)) == 0) {
IGMPSTAT_INC(igps_rcv_tooshort);
return;
}
ip = mtod(m, struct ip *);
/*
* Validate checksum.
*/
m->m_data += iphlen;
m->m_len -= iphlen;
igmp = mtod(m, struct igmp *);
if (in_cksum(m, igmplen)) {
IGMPSTAT_INC(igps_rcv_badsum);
m_freem(m);
return;
}
m->m_data -= iphlen;
m->m_len += iphlen;
/*
* IGMP control traffic is link-scope, and must have a TTL of 1.
* DVMRP traffic (e.g. mrinfo, mtrace) is an exception;
* probe packets may come from beyond the LAN.
*/
if (igmp->igmp_type != IGMP_DVMRP && ip->ip_ttl != 1) {
IGMPSTAT_INC(igps_rcv_badttl);
m_freem(m);
return;
}
switch (igmp->igmp_type) {
case IGMP_HOST_MEMBERSHIP_QUERY:
if (igmplen == IGMP_MINLEN) {
if (igmp->igmp_code == 0)
queryver = IGMP_VERSION_1;
else
queryver = IGMP_VERSION_2;
} else if (igmplen >= IGMP_V3_QUERY_MINLEN) {
queryver = IGMP_VERSION_3;
} else {
IGMPSTAT_INC(igps_rcv_tooshort);
m_freem(m);
return;
}
switch (queryver) {
case IGMP_VERSION_1:
IGMPSTAT_INC(igps_rcv_v1v2_queries);
if (!V_igmp_v1enable)
break;
if (igmp_input_v1_query(ifp, ip, igmp) != 0) {
m_freem(m);
return;
}
break;
case IGMP_VERSION_2:
IGMPSTAT_INC(igps_rcv_v1v2_queries);
if (!V_igmp_v2enable)
break;
if (igmp_input_v2_query(ifp, ip, igmp) != 0) {
m_freem(m);
return;
}
break;
case IGMP_VERSION_3: {
struct igmpv3 *igmpv3;
uint16_t igmpv3len;
uint16_t srclen;
int nsrc;
IGMPSTAT_INC(igps_rcv_v3_queries);
igmpv3 = (struct igmpv3 *)igmp;
/*
* Validate length based on source count.
*/
nsrc = ntohs(igmpv3->igmp_numsrc);
srclen = sizeof(struct in_addr) * nsrc;
if (nsrc * sizeof(in_addr_t) > srclen) {
IGMPSTAT_INC(igps_rcv_tooshort);
return;
}
/*
* m_pullup() may modify m, so pullup in
* this scope.
*/
igmpv3len = iphlen + IGMP_V3_QUERY_MINLEN +
srclen;
if ((m->m_flags & M_EXT ||
m->m_len < igmpv3len) &&
(m = m_pullup(m, igmpv3len)) == NULL) {
IGMPSTAT_INC(igps_rcv_tooshort);
return;
}
igmpv3 = (struct igmpv3 *)(mtod(m, uint8_t *)
+ iphlen);
if (igmp_input_v3_query(ifp, ip, igmpv3) != 0) {
m_freem(m);
return;
}
}
break;
}
break;
case IGMP_v1_HOST_MEMBERSHIP_REPORT:
if (!V_igmp_v1enable)
break;
if (igmp_input_v1_report(ifp, ip, igmp) != 0) {
m_freem(m);
return;
}
break;
case IGMP_v2_HOST_MEMBERSHIP_REPORT:
if (!V_igmp_v2enable)
break;
if (!ip_checkrouteralert(m))
IGMPSTAT_INC(igps_rcv_nora);
if (igmp_input_v2_report(ifp, ip, igmp) != 0) {
m_freem(m);
return;
}
break;
case IGMP_v3_HOST_MEMBERSHIP_REPORT:
/*
* Hosts do not need to process IGMPv3 membership reports,
* as report suppression is no longer required.
*/
if (!ip_checkrouteralert(m))
IGMPSTAT_INC(igps_rcv_nora);
break;
default:
break;
}
/*
* Pass all valid IGMP packets up to any process(es) listening on a
* raw IGMP socket.
*/
rip_input(m, off);
}
/*
* Fast timeout handler (global).
* VIMAGE: Timeout handlers are expected to service all vimages.
*/
void
igmp_fasttimo(void)
{
VNET_ITERATOR_DECL(vnet_iter);
VNET_LIST_RLOCK_NOSLEEP();
VNET_FOREACH(vnet_iter) {
CURVNET_SET(vnet_iter);
igmp_fasttimo_vnet();
CURVNET_RESTORE();
}
VNET_LIST_RUNLOCK_NOSLEEP();
}
/*
* Fast timeout handler (per-vnet).
* Sends are shuffled off to a netisr to deal with Giant.
*
* VIMAGE: Assume caller has set up our curvnet.
*/
static void
igmp_fasttimo_vnet(void)
{
struct ifqueue scq; /* State-change packets */
struct ifqueue qrq; /* Query response packets */
struct ifnet *ifp;
struct igmp_ifinfo *igi;
struct ifmultiaddr *ifma;
struct in_multi *inm;
int loop, uri_fasthz;
loop = 0;
uri_fasthz = 0;
/*
* Quick check to see if any work needs to be done, in order to
* minimize the overhead of fasttimo processing.
* SMPng: XXX Unlocked reads.
*/
if (!V_current_state_timers_running &&
!V_interface_timers_running &&
!V_state_change_timers_running)
return;
IN_MULTI_LOCK();
IGMP_LOCK();
/*
* IGMPv3 General Query response timer processing.
*/
if (V_interface_timers_running) {
CTR1(KTR_IGMPV3, "%s: interface timers running", __func__);
V_interface_timers_running = 0;
LIST_FOREACH(igi, &V_igi_head, igi_link) {
if (igi->igi_v3_timer == 0) {
/* Do nothing. */
} else if (--igi->igi_v3_timer == 0) {
igmp_v3_dispatch_general_query(igi);
} else {
V_interface_timers_running = 1;
}
}
}
if (!V_current_state_timers_running &&
!V_state_change_timers_running)
goto out_locked;
V_current_state_timers_running = 0;
V_state_change_timers_running = 0;
CTR1(KTR_IGMPV3, "%s: state change timers running", __func__);
/*
* IGMPv1/v2/v3 host report and state-change timer processing.
* Note: Processing a v3 group timer may remove a node.
*/
LIST_FOREACH(igi, &V_igi_head, igi_link) {
ifp = igi->igi_ifp;
if (igi->igi_version == IGMP_VERSION_3) {
loop = (igi->igi_flags & IGIF_LOOPBACK) ? 1 : 0;
uri_fasthz = IGMP_RANDOM_DELAY(igi->igi_uri *
PR_FASTHZ);
memset(&qrq, 0, sizeof(struct ifqueue));
IFQ_SET_MAXLEN(&qrq, IGMP_MAX_G_GS_PACKETS);
memset(&scq, 0, sizeof(struct ifqueue));
IFQ_SET_MAXLEN(&scq, IGMP_MAX_STATE_CHANGE_PACKETS);
}
IF_ADDR_RLOCK(ifp);
TAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
if (ifma->ifma_addr->sa_family != AF_INET ||
ifma->ifma_protospec == NULL)
continue;
inm = (struct in_multi *)ifma->ifma_protospec;
switch (igi->igi_version) {
case IGMP_VERSION_1:
case IGMP_VERSION_2:
igmp_v1v2_process_group_timer(inm,
igi->igi_version);
break;
case IGMP_VERSION_3:
igmp_v3_process_group_timers(igi, &qrq,
&scq, inm, uri_fasthz);
break;
}
}
IF_ADDR_RUNLOCK(ifp);
if (igi->igi_version == IGMP_VERSION_3) {
struct in_multi *tinm;
igmp_dispatch_queue(&qrq, 0, loop);
igmp_dispatch_queue(&scq, 0, loop);
/*
* Free the in_multi reference(s) for this
* IGMP lifecycle.
*/
SLIST_FOREACH_SAFE(inm, &igi->igi_relinmhead,
inm_nrele, tinm) {
SLIST_REMOVE_HEAD(&igi->igi_relinmhead,
inm_nrele);
inm_release_locked(inm);
}
}
}
out_locked:
IGMP_UNLOCK();
IN_MULTI_UNLOCK();
}
/*
* Update host report group timer for IGMPv1/v2.
* Will update the global pending timer flags.
*/
static void
igmp_v1v2_process_group_timer(struct in_multi *inm, const int version)
{
int report_timer_expired;
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
if (inm->inm_timer == 0) {
report_timer_expired = 0;
} else if (--inm->inm_timer == 0) {
report_timer_expired = 1;
} else {
V_current_state_timers_running = 1;
return;
}
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
case IGMP_IDLE_MEMBER:
case IGMP_LAZY_MEMBER:
case IGMP_SLEEPING_MEMBER:
case IGMP_AWAKENING_MEMBER:
break;
case IGMP_REPORTING_MEMBER:
if (report_timer_expired) {
inm->inm_state = IGMP_IDLE_MEMBER;
(void)igmp_v1v2_queue_report(inm,
(version == IGMP_VERSION_2) ?
IGMP_v2_HOST_MEMBERSHIP_REPORT :
IGMP_v1_HOST_MEMBERSHIP_REPORT);
}
break;
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
case IGMP_LEAVING_MEMBER:
break;
}
}
/*
* Update a group's timers for IGMPv3.
* Will update the global pending timer flags.
* Note: Unlocked read from igi.
*/
static void
igmp_v3_process_group_timers(struct igmp_ifinfo *igi,
struct ifqueue *qrq, struct ifqueue *scq,
struct in_multi *inm, const int uri_fasthz)
{
int query_response_timer_expired;
int state_change_retransmit_timer_expired;
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
query_response_timer_expired = 0;
state_change_retransmit_timer_expired = 0;
/*
* During a transition from v1/v2 compatibility mode back to v3,
* a group record in REPORTING state may still have its group
* timer active. This is a no-op in this function; it is easier
* to deal with it here than to complicate the slow-timeout path.
*/
if (inm->inm_timer == 0) {
query_response_timer_expired = 0;
} else if (--inm->inm_timer == 0) {
query_response_timer_expired = 1;
} else {
V_current_state_timers_running = 1;
}
if (inm->inm_sctimer == 0) {
state_change_retransmit_timer_expired = 0;
} else if (--inm->inm_sctimer == 0) {
state_change_retransmit_timer_expired = 1;
} else {
V_state_change_timers_running = 1;
}
/* We are in fasttimo, so be quick about it. */
if (!state_change_retransmit_timer_expired &&
!query_response_timer_expired)
return;
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
case IGMP_SLEEPING_MEMBER:
case IGMP_LAZY_MEMBER:
case IGMP_AWAKENING_MEMBER:
case IGMP_IDLE_MEMBER:
break;
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
/*
* Respond to a previously pending Group-Specific
* or Group-and-Source-Specific query by enqueueing
* the appropriate Current-State report for
* immediate transmission.
*/
if (query_response_timer_expired) {
int retval;
retval = igmp_v3_enqueue_group_record(qrq, inm, 0, 1,
(inm->inm_state == IGMP_SG_QUERY_PENDING_MEMBER));
CTR2(KTR_IGMPV3, "%s: enqueue record = %d",
__func__, retval);
inm->inm_state = IGMP_REPORTING_MEMBER;
/* XXX Clear recorded sources for next time. */
inm_clear_recorded(inm);
}
/* FALLTHROUGH */
case IGMP_REPORTING_MEMBER:
case IGMP_LEAVING_MEMBER:
if (state_change_retransmit_timer_expired) {
/*
* State-change retransmission timer fired.
* If there are any further pending retransmissions,
* set the global pending state-change flag, and
* reset the timer.
*/
if (--inm->inm_scrv > 0) {
inm->inm_sctimer = uri_fasthz;
V_state_change_timers_running = 1;
}
/*
* Retransmit the previously computed state-change
* report. If there are no further pending
* retransmissions, the mbuf queue will be consumed.
* Update T0 state to T1 as we have now sent
* a state-change.
*/
(void)igmp_v3_merge_state_changes(inm, scq);
inm_commit(inm);
CTR3(KTR_IGMPV3, "%s: T1 -> T0 for %s/%s", __func__,
inet_ntoa(inm->inm_addr), inm->inm_ifp->if_xname);
/*
* If we are leaving the group for good, make sure
* we release IGMP's reference to it.
* This release must be deferred using a SLIST,
* as we are called from a loop which traverses
* the in_ifmultiaddr TAILQ.
*/
if (inm->inm_state == IGMP_LEAVING_MEMBER &&
inm->inm_scrv == 0) {
inm->inm_state = IGMP_NOT_MEMBER;
SLIST_INSERT_HEAD(&igi->igi_relinmhead,
inm, inm_nrele);
}
}
break;
}
}
/*
* Suppress a group's pending response to a group or source/group query.
*
* Do NOT suppress state changes. This leads to IGMPv3 inconsistency.
* Do NOT update ST1/ST0 as this operation merely suppresses
* the currently pending group record.
* Do NOT suppress the response to a general query. It is possible but
* it would require adding another state or flag.
*/
static void
igmp_v3_suppress_group_record(struct in_multi *inm)
{
IN_MULTI_LOCK_ASSERT();
KASSERT(inm->inm_igi->igi_version == IGMP_VERSION_3,
("%s: not IGMPv3 mode on link", __func__));
if (inm->inm_state != IGMP_G_QUERY_PENDING_MEMBER ||
inm->inm_state != IGMP_SG_QUERY_PENDING_MEMBER)
return;
if (inm->inm_state == IGMP_SG_QUERY_PENDING_MEMBER)
inm_clear_recorded(inm);
inm->inm_timer = 0;
inm->inm_state = IGMP_REPORTING_MEMBER;
}
/*
* Switch to a different IGMP version on the given interface,
* as per Section 7.2.1.
*/
static void
igmp_set_version(struct igmp_ifinfo *igi, const int version)
{
int old_version_timer;
IGMP_LOCK_ASSERT();
CTR4(KTR_IGMPV3, "%s: switching to v%d on ifp %p(%s)", __func__,
version, igi->igi_ifp, igi->igi_ifp->if_xname);
if (version == IGMP_VERSION_1 || version == IGMP_VERSION_2) {
/*
* Compute the "Older Version Querier Present" timer as per
* Section 8.12.
*/
old_version_timer = igi->igi_rv * igi->igi_qi + igi->igi_qri;
old_version_timer *= PR_SLOWHZ;
if (version == IGMP_VERSION_1) {
igi->igi_v1_timer = old_version_timer;
igi->igi_v2_timer = 0;
} else if (version == IGMP_VERSION_2) {
igi->igi_v1_timer = 0;
igi->igi_v2_timer = old_version_timer;
}
}
if (igi->igi_v1_timer == 0 && igi->igi_v2_timer > 0) {
if (igi->igi_version != IGMP_VERSION_2) {
igi->igi_version = IGMP_VERSION_2;
igmp_v3_cancel_link_timers(igi);
}
} else if (igi->igi_v1_timer > 0) {
if (igi->igi_version != IGMP_VERSION_1) {
igi->igi_version = IGMP_VERSION_1;
igmp_v3_cancel_link_timers(igi);
}
}
}
/*
* Cancel pending IGMPv3 timers for the given link and all groups
* joined on it; state-change, general-query, and group-query timers.
*
* Only ever called on a transition from v3 to Compatibility mode. Kill
* the timers stone dead (this may be expensive for large N groups), they
* will be restarted if Compatibility Mode deems that they must be due to
* query processing.
*/
static void
igmp_v3_cancel_link_timers(struct igmp_ifinfo *igi)
{
struct ifmultiaddr *ifma;
struct ifnet *ifp;
struct in_multi *inm, *tinm;
CTR3(KTR_IGMPV3, "%s: cancel v3 timers on ifp %p(%s)", __func__,
igi->igi_ifp, igi->igi_ifp->if_xname);
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
/*
* Stop the v3 General Query Response on this link stone dead.
* If fasttimo is woken up due to V_interface_timers_running,
* the flag will be cleared if there are no pending link timers.
*/
igi->igi_v3_timer = 0;
/*
* Now clear the current-state and state-change report timers
* for all memberships scoped to this link.
*/
ifp = igi->igi_ifp;
IF_ADDR_RLOCK(ifp);
TAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
if (ifma->ifma_addr->sa_family != AF_INET ||
ifma->ifma_protospec == NULL)
continue;
inm = (struct in_multi *)ifma->ifma_protospec;
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
case IGMP_IDLE_MEMBER:
case IGMP_LAZY_MEMBER:
case IGMP_SLEEPING_MEMBER:
case IGMP_AWAKENING_MEMBER:
/*
* These states are either not relevant in v3 mode,
* or are unreported. Do nothing.
*/
break;
case IGMP_LEAVING_MEMBER:
/*
* If we are leaving the group and switching to
* compatibility mode, we need to release the final
* reference held for issuing the INCLUDE {}, and
* transition to REPORTING to ensure the host leave
* message is sent upstream to the old querier --
* transition to NOT would lose the leave and race.
*/
SLIST_INSERT_HEAD(&igi->igi_relinmhead, inm, inm_nrele);
/* FALLTHROUGH */
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
inm_clear_recorded(inm);
/* FALLTHROUGH */
case IGMP_REPORTING_MEMBER:
inm->inm_state = IGMP_REPORTING_MEMBER;
break;
}
/*
* Always clear state-change and group report timers.
* Free any pending IGMPv3 state-change records.
*/
inm->inm_sctimer = 0;
inm->inm_timer = 0;
_IF_DRAIN(&inm->inm_scq);
}
IF_ADDR_RUNLOCK(ifp);
SLIST_FOREACH_SAFE(inm, &igi->igi_relinmhead, inm_nrele, tinm) {
SLIST_REMOVE_HEAD(&igi->igi_relinmhead, inm_nrele);
inm_release_locked(inm);
}
}
/*
* Update the Older Version Querier Present timers for a link.
* See Section 7.2.1 of RFC 3376.
*/
static void
igmp_v1v2_process_querier_timers(struct igmp_ifinfo *igi)
{
IGMP_LOCK_ASSERT();
if (igi->igi_v1_timer == 0 && igi->igi_v2_timer == 0) {
/*
* IGMPv1 and IGMPv2 Querier Present timers expired.
*
* Revert to IGMPv3.
*/
if (igi->igi_version != IGMP_VERSION_3) {
CTR5(KTR_IGMPV3,
"%s: transition from v%d -> v%d on %p(%s)",
__func__, igi->igi_version, IGMP_VERSION_3,
igi->igi_ifp, igi->igi_ifp->if_xname);
igi->igi_version = IGMP_VERSION_3;
}
} else if (igi->igi_v1_timer == 0 && igi->igi_v2_timer > 0) {
/*
* IGMPv1 Querier Present timer expired,
* IGMPv2 Querier Present timer running.
* If IGMPv2 was disabled since last timeout,
* revert to IGMPv3.
* If IGMPv2 is enabled, revert to IGMPv2.
*/
if (!V_igmp_v2enable) {
CTR5(KTR_IGMPV3,
"%s: transition from v%d -> v%d on %p(%s)",
__func__, igi->igi_version, IGMP_VERSION_3,
igi->igi_ifp, igi->igi_ifp->if_xname);
igi->igi_v2_timer = 0;
igi->igi_version = IGMP_VERSION_3;
} else {
--igi->igi_v2_timer;
if (igi->igi_version != IGMP_VERSION_2) {
CTR5(KTR_IGMPV3,
"%s: transition from v%d -> v%d on %p(%s)",
__func__, igi->igi_version, IGMP_VERSION_2,
igi->igi_ifp, igi->igi_ifp->if_xname);
igi->igi_version = IGMP_VERSION_2;
igmp_v3_cancel_link_timers(igi);
}
}
} else if (igi->igi_v1_timer > 0) {
/*
* IGMPv1 Querier Present timer running.
* Stop IGMPv2 timer if running.
*
* If IGMPv1 was disabled since last timeout,
* revert to IGMPv3.
* If IGMPv1 is enabled, reset IGMPv2 timer if running.
*/
if (!V_igmp_v1enable) {
CTR5(KTR_IGMPV3,
"%s: transition from v%d -> v%d on %p(%s)",
__func__, igi->igi_version, IGMP_VERSION_3,
igi->igi_ifp, igi->igi_ifp->if_xname);
igi->igi_v1_timer = 0;
igi->igi_version = IGMP_VERSION_3;
} else {
--igi->igi_v1_timer;
}
if (igi->igi_v2_timer > 0) {
CTR3(KTR_IGMPV3,
"%s: cancel v2 timer on %p(%s)",
__func__, igi->igi_ifp, igi->igi_ifp->if_xname);
igi->igi_v2_timer = 0;
}
}
}
/*
* Global slowtimo handler.
* VIMAGE: Timeout handlers are expected to service all vimages.
*/
void
igmp_slowtimo(void)
{
VNET_ITERATOR_DECL(vnet_iter);
VNET_LIST_RLOCK_NOSLEEP();
VNET_FOREACH(vnet_iter) {
CURVNET_SET(vnet_iter);
igmp_slowtimo_vnet();
CURVNET_RESTORE();
}
VNET_LIST_RUNLOCK_NOSLEEP();
}
/*
* Per-vnet slowtimo handler.
*/
static void
igmp_slowtimo_vnet(void)
{
struct igmp_ifinfo *igi;
IGMP_LOCK();
LIST_FOREACH(igi, &V_igi_head, igi_link) {
igmp_v1v2_process_querier_timers(igi);
}
IGMP_UNLOCK();
}
/*
* Dispatch an IGMPv1/v2 host report or leave message.
* These are always small enough to fit inside a single mbuf.
*/
static int
igmp_v1v2_queue_report(struct in_multi *inm, const int type)
{
struct ifnet *ifp;
struct igmp *igmp;
struct ip *ip;
struct mbuf *m;
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
ifp = inm->inm_ifp;
m = m_gethdr(M_NOWAIT, MT_DATA);
if (m == NULL)
return (ENOMEM);
MH_ALIGN(m, sizeof(struct ip) + sizeof(struct igmp));
m->m_pkthdr.len = sizeof(struct ip) + sizeof(struct igmp);
m->m_data += sizeof(struct ip);
m->m_len = sizeof(struct igmp);
igmp = mtod(m, struct igmp *);
igmp->igmp_type = type;
igmp->igmp_code = 0;
igmp->igmp_group = inm->inm_addr;
igmp->igmp_cksum = 0;
igmp->igmp_cksum = in_cksum(m, sizeof(struct igmp));
m->m_data -= sizeof(struct ip);
m->m_len += sizeof(struct ip);
ip = mtod(m, struct ip *);
ip->ip_tos = 0;
ip->ip_len = htons(sizeof(struct ip) + sizeof(struct igmp));
ip->ip_off = 0;
ip->ip_p = IPPROTO_IGMP;
ip->ip_src.s_addr = INADDR_ANY;
if (type == IGMP_HOST_LEAVE_MESSAGE)
ip->ip_dst.s_addr = htonl(INADDR_ALLRTRS_GROUP);
else
ip->ip_dst = inm->inm_addr;
igmp_save_context(m, ifp);
m->m_flags |= M_IGMPV2;
if (inm->inm_igi->igi_flags & IGIF_LOOPBACK)
m->m_flags |= M_IGMP_LOOP;
CTR2(KTR_IGMPV3, "%s: netisr_dispatch(NETISR_IGMP, %p)", __func__, m);
netisr_dispatch(NETISR_IGMP, m);
return (0);
}
/*
* Process a state change from the upper layer for the given IPv4 group.
*
* Each socket holds a reference on the in_multi in its own ip_moptions.
* The socket layer will have made the necessary updates to.the group
* state, it is now up to IGMP to issue a state change report if there
* has been any change between T0 (when the last state-change was issued)
* and T1 (now).
*
* We use the IGMPv3 state machine at group level. The IGMP module
* however makes the decision as to which IGMP protocol version to speak.
* A state change *from* INCLUDE {} always means an initial join.
* A state change *to* INCLUDE {} always means a final leave.
*
* FUTURE: If IGIF_V3LITE is enabled for this interface, then we can
* save ourselves a bunch of work; any exclusive mode groups need not
* compute source filter lists.
*
* VIMAGE: curvnet should have been set by caller, as this routine
* is called from the socket option handlers.
*/
int
igmp_change_state(struct in_multi *inm)
{
struct igmp_ifinfo *igi;
struct ifnet *ifp;
int error;
IN_MULTI_LOCK_ASSERT();
error = 0;
/*
* Try to detect if the upper layer just asked us to change state
* for an interface which has now gone away.
*/
KASSERT(inm->inm_ifma != NULL, ("%s: no ifma", __func__));
ifp = inm->inm_ifma->ifma_ifp;
/*
* Sanity check that netinet's notion of ifp is the
* same as net's.
*/
KASSERT(inm->inm_ifp == ifp, ("%s: bad ifp", __func__));
IGMP_LOCK();
igi = ((struct in_ifinfo *)ifp->if_afdata[AF_INET])->ii_igmp;
KASSERT(igi != NULL, ("%s: no igmp_ifinfo for ifp %p", __func__, ifp));
/*
* If we detect a state transition to or from MCAST_UNDEFINED
* for this group, then we are starting or finishing an IGMP
* life cycle for this group.
*/
if (inm->inm_st[1].iss_fmode != inm->inm_st[0].iss_fmode) {
CTR3(KTR_IGMPV3, "%s: inm transition %d -> %d", __func__,
inm->inm_st[0].iss_fmode, inm->inm_st[1].iss_fmode);
if (inm->inm_st[0].iss_fmode == MCAST_UNDEFINED) {
CTR1(KTR_IGMPV3, "%s: initial join", __func__);
error = igmp_initial_join(inm, igi);
goto out_locked;
} else if (inm->inm_st[1].iss_fmode == MCAST_UNDEFINED) {
CTR1(KTR_IGMPV3, "%s: final leave", __func__);
igmp_final_leave(inm, igi);
goto out_locked;
}
} else {
CTR1(KTR_IGMPV3, "%s: filter set change", __func__);
}
error = igmp_handle_state_change(inm, igi);
out_locked:
IGMP_UNLOCK();
return (error);
}
/*
* Perform the initial join for an IGMP group.
*
* When joining a group:
* If the group should have its IGMP traffic suppressed, do nothing.
* IGMPv1 starts sending IGMPv1 host membership reports.
* IGMPv2 starts sending IGMPv2 host membership reports.
* IGMPv3 will schedule an IGMPv3 state-change report containing the
* initial state of the membership.
*/
static int
igmp_initial_join(struct in_multi *inm, struct igmp_ifinfo *igi)
{
struct ifnet *ifp;
struct ifqueue *ifq;
int error, retval, syncstates;
CTR4(KTR_IGMPV3, "%s: initial join %s on ifp %p(%s)",
__func__, inet_ntoa(inm->inm_addr), inm->inm_ifp,
inm->inm_ifp->if_xname);
error = 0;
syncstates = 1;
ifp = inm->inm_ifp;
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
KASSERT(igi && igi->igi_ifp == ifp, ("%s: inconsistent ifp", __func__));
/*
* Groups joined on loopback or marked as 'not reported',
* e.g. 224.0.0.1, enter the IGMP_SILENT_MEMBER state and
* are never reported in any IGMP protocol exchanges.
* All other groups enter the appropriate IGMP state machine
* for the version in use on this link.
* A link marked as IGIF_SILENT causes IGMP to be completely
* disabled for the link.
*/
if ((ifp->if_flags & IFF_LOOPBACK) ||
(igi->igi_flags & IGIF_SILENT) ||
!igmp_isgroupreported(inm->inm_addr)) {
CTR1(KTR_IGMPV3,
"%s: not kicking state machine for silent group", __func__);
inm->inm_state = IGMP_SILENT_MEMBER;
inm->inm_timer = 0;
} else {
/*
* Deal with overlapping in_multi lifecycle.
* If this group was LEAVING, then make sure
* we drop the reference we picked up to keep the
* group around for the final INCLUDE {} enqueue.
*/
if (igi->igi_version == IGMP_VERSION_3 &&
inm->inm_state == IGMP_LEAVING_MEMBER)
inm_release_locked(inm);
inm->inm_state = IGMP_REPORTING_MEMBER;
switch (igi->igi_version) {
case IGMP_VERSION_1:
case IGMP_VERSION_2:
inm->inm_state = IGMP_IDLE_MEMBER;
error = igmp_v1v2_queue_report(inm,
(igi->igi_version == IGMP_VERSION_2) ?
IGMP_v2_HOST_MEMBERSHIP_REPORT :
IGMP_v1_HOST_MEMBERSHIP_REPORT);
if (error == 0) {
inm->inm_timer = IGMP_RANDOM_DELAY(
IGMP_V1V2_MAX_RI * PR_FASTHZ);
V_current_state_timers_running = 1;
}
break;
case IGMP_VERSION_3:
/*
* Defer update of T0 to T1, until the first copy
* of the state change has been transmitted.
*/
syncstates = 0;
/*
* Immediately enqueue a State-Change Report for
* this interface, freeing any previous reports.
* Don't kick the timers if there is nothing to do,
* or if an error occurred.
*/
ifq = &inm->inm_scq;
_IF_DRAIN(ifq);
retval = igmp_v3_enqueue_group_record(ifq, inm, 1,
0, 0);
CTR2(KTR_IGMPV3, "%s: enqueue record = %d",
__func__, retval);
if (retval <= 0) {
error = retval * -1;
break;
}
/*
* Schedule transmission of pending state-change
* report up to RV times for this link. The timer
* will fire at the next igmp_fasttimo (~200ms),
* giving us an opportunity to merge the reports.
*/
if (igi->igi_flags & IGIF_LOOPBACK) {
inm->inm_scrv = 1;
} else {
KASSERT(igi->igi_rv > 1,
("%s: invalid robustness %d", __func__,
igi->igi_rv));
inm->inm_scrv = igi->igi_rv;
}
inm->inm_sctimer = 1;
V_state_change_timers_running = 1;
error = 0;
break;
}
}
/*
* Only update the T0 state if state change is atomic,
* i.e. we don't need to wait for a timer to fire before we
* can consider the state change to have been communicated.
*/
if (syncstates) {
inm_commit(inm);
CTR3(KTR_IGMPV3, "%s: T1 -> T0 for %s/%s", __func__,
inet_ntoa(inm->inm_addr), inm->inm_ifp->if_xname);
}
return (error);
}
/*
* Issue an intermediate state change during the IGMP life-cycle.
*/
static int
igmp_handle_state_change(struct in_multi *inm, struct igmp_ifinfo *igi)
{
struct ifnet *ifp;
int retval;
CTR4(KTR_IGMPV3, "%s: state change for %s on ifp %p(%s)",
__func__, inet_ntoa(inm->inm_addr), inm->inm_ifp,
inm->inm_ifp->if_xname);
ifp = inm->inm_ifp;
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
KASSERT(igi && igi->igi_ifp == ifp, ("%s: inconsistent ifp", __func__));
if ((ifp->if_flags & IFF_LOOPBACK) ||
(igi->igi_flags & IGIF_SILENT) ||
!igmp_isgroupreported(inm->inm_addr) ||
(igi->igi_version != IGMP_VERSION_3)) {
if (!igmp_isgroupreported(inm->inm_addr)) {
CTR1(KTR_IGMPV3,
"%s: not kicking state machine for silent group", __func__);
}
CTR1(KTR_IGMPV3, "%s: nothing to do", __func__);
inm_commit(inm);
CTR3(KTR_IGMPV3, "%s: T1 -> T0 for %s/%s", __func__,
inet_ntoa(inm->inm_addr), inm->inm_ifp->if_xname);
return (0);
}
_IF_DRAIN(&inm->inm_scq);
retval = igmp_v3_enqueue_group_record(&inm->inm_scq, inm, 1, 0, 0);
CTR2(KTR_IGMPV3, "%s: enqueue record = %d", __func__, retval);
if (retval <= 0)
return (-retval);
/*
* If record(s) were enqueued, start the state-change
* report timer for this group.
*/
inm->inm_scrv = ((igi->igi_flags & IGIF_LOOPBACK) ? 1 : igi->igi_rv);
inm->inm_sctimer = 1;
V_state_change_timers_running = 1;
return (0);
}
/*
* Perform the final leave for an IGMP group.
*
* When leaving a group:
* IGMPv1 does nothing.
* IGMPv2 sends a host leave message, if and only if we are the reporter.
* IGMPv3 enqueues a state-change report containing a transition
* to INCLUDE {} for immediate transmission.
*/
static void
igmp_final_leave(struct in_multi *inm, struct igmp_ifinfo *igi)
{
int syncstates;
syncstates = 1;
CTR4(KTR_IGMPV3, "%s: final leave %s on ifp %p(%s)",
__func__, inet_ntoa(inm->inm_addr), inm->inm_ifp,
inm->inm_ifp->if_xname);
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
case IGMP_LEAVING_MEMBER:
/* Already leaving or left; do nothing. */
CTR1(KTR_IGMPV3,
"%s: not kicking state machine for silent group", __func__);
break;
case IGMP_REPORTING_MEMBER:
case IGMP_IDLE_MEMBER:
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
if (igi->igi_version == IGMP_VERSION_2) {
#ifdef INVARIANTS
if (inm->inm_state == IGMP_G_QUERY_PENDING_MEMBER ||
inm->inm_state == IGMP_SG_QUERY_PENDING_MEMBER)
panic("%s: IGMPv3 state reached, not IGMPv3 mode",
__func__);
#endif
igmp_v1v2_queue_report(inm, IGMP_HOST_LEAVE_MESSAGE);
inm->inm_state = IGMP_NOT_MEMBER;
} else if (igi->igi_version == IGMP_VERSION_3) {
/*
* Stop group timer and all pending reports.
* Immediately enqueue a state-change report
* TO_IN {} to be sent on the next fast timeout,
* giving us an opportunity to merge reports.
*/
_IF_DRAIN(&inm->inm_scq);
inm->inm_timer = 0;
if (igi->igi_flags & IGIF_LOOPBACK) {
inm->inm_scrv = 1;
} else {
inm->inm_scrv = igi->igi_rv;
}
CTR4(KTR_IGMPV3, "%s: Leaving %s/%s with %d "
"pending retransmissions.", __func__,
inet_ntoa(inm->inm_addr),
inm->inm_ifp->if_xname, inm->inm_scrv);
if (inm->inm_scrv == 0) {
inm->inm_state = IGMP_NOT_MEMBER;
inm->inm_sctimer = 0;
} else {
int retval;
inm_acquire_locked(inm);
retval = igmp_v3_enqueue_group_record(
&inm->inm_scq, inm, 1, 0, 0);
KASSERT(retval != 0,
("%s: enqueue record = %d", __func__,
retval));
inm->inm_state = IGMP_LEAVING_MEMBER;
inm->inm_sctimer = 1;
V_state_change_timers_running = 1;
syncstates = 0;
}
break;
}
break;
case IGMP_LAZY_MEMBER:
case IGMP_SLEEPING_MEMBER:
case IGMP_AWAKENING_MEMBER:
/* Our reports are suppressed; do nothing. */
break;
}
if (syncstates) {
inm_commit(inm);
CTR3(KTR_IGMPV3, "%s: T1 -> T0 for %s/%s", __func__,
inet_ntoa(inm->inm_addr), inm->inm_ifp->if_xname);
inm->inm_st[1].iss_fmode = MCAST_UNDEFINED;
CTR3(KTR_IGMPV3, "%s: T1 now MCAST_UNDEFINED for %s/%s",
__func__, inet_ntoa(inm->inm_addr), inm->inm_ifp->if_xname);
}
}
/*
* Enqueue an IGMPv3 group record to the given output queue.
*
* XXX This function could do with having the allocation code
* split out, and the multiple-tree-walks coalesced into a single
* routine as has been done in igmp_v3_enqueue_filter_change().
*
* If is_state_change is zero, a current-state record is appended.
* If is_state_change is non-zero, a state-change report is appended.
*
* If is_group_query is non-zero, an mbuf packet chain is allocated.
* If is_group_query is zero, and if there is a packet with free space
* at the tail of the queue, it will be appended to providing there
* is enough free space.
* Otherwise a new mbuf packet chain is allocated.
*
* If is_source_query is non-zero, each source is checked to see if
* it was recorded for a Group-Source query, and will be omitted if
* it is not both in-mode and recorded.
*
* The function will attempt to allocate leading space in the packet
* for the IP/IGMP header to be prepended without fragmenting the chain.
*
* If successful the size of all data appended to the queue is returned,
* otherwise an error code less than zero is returned, or zero if
* no record(s) were appended.
*/
static int
igmp_v3_enqueue_group_record(struct ifqueue *ifq, struct in_multi *inm,
const int is_state_change, const int is_group_query,
const int is_source_query)
{
struct igmp_grouprec ig;
struct igmp_grouprec *pig;
struct ifnet *ifp;
struct ip_msource *ims, *nims;
struct mbuf *m0, *m, *md;
int error, is_filter_list_change;
int minrec0len, m0srcs, msrcs, nbytes, off;
int record_has_sources;
int now;
int type;
in_addr_t naddr;
uint8_t mode;
IN_MULTI_LOCK_ASSERT();
error = 0;
ifp = inm->inm_ifp;
is_filter_list_change = 0;
m = NULL;
m0 = NULL;
m0srcs = 0;
msrcs = 0;
nbytes = 0;
nims = NULL;
record_has_sources = 1;
pig = NULL;
type = IGMP_DO_NOTHING;
mode = inm->inm_st[1].iss_fmode;
/*
* If we did not transition out of ASM mode during t0->t1,
* and there are no source nodes to process, we can skip
* the generation of source records.
*/
if (inm->inm_st[0].iss_asm > 0 && inm->inm_st[1].iss_asm > 0 &&
inm->inm_nsrc == 0)
record_has_sources = 0;
if (is_state_change) {
/*
* Queue a state change record.
* If the mode did not change, and there are non-ASM
* listeners or source filters present,
* we potentially need to issue two records for the group.
* If we are transitioning to MCAST_UNDEFINED, we need
* not send any sources.
* If there are ASM listeners, and there was no filter
* mode transition of any kind, do nothing.
*/
if (mode != inm->inm_st[0].iss_fmode) {
if (mode == MCAST_EXCLUDE) {
CTR1(KTR_IGMPV3, "%s: change to EXCLUDE",
__func__);
type = IGMP_CHANGE_TO_EXCLUDE_MODE;
} else {
CTR1(KTR_IGMPV3, "%s: change to INCLUDE",
__func__);
type = IGMP_CHANGE_TO_INCLUDE_MODE;
if (mode == MCAST_UNDEFINED)
record_has_sources = 0;
}
} else {
if (record_has_sources) {
is_filter_list_change = 1;
} else {
type = IGMP_DO_NOTHING;
}
}
} else {
/*
* Queue a current state record.
*/
if (mode == MCAST_EXCLUDE) {
type = IGMP_MODE_IS_EXCLUDE;
} else if (mode == MCAST_INCLUDE) {
type = IGMP_MODE_IS_INCLUDE;
KASSERT(inm->inm_st[1].iss_asm == 0,
("%s: inm %p is INCLUDE but ASM count is %d",
__func__, inm, inm->inm_st[1].iss_asm));
}
}
/*
* Generate the filter list changes using a separate function.
*/
if (is_filter_list_change)
return (igmp_v3_enqueue_filter_change(ifq, inm));
if (type == IGMP_DO_NOTHING) {
CTR3(KTR_IGMPV3, "%s: nothing to do for %s/%s",
__func__, inet_ntoa(inm->inm_addr),
inm->inm_ifp->if_xname);
return (0);
}
/*
* If any sources are present, we must be able to fit at least
* one in the trailing space of the tail packet's mbuf,
* ideally more.
*/
minrec0len = sizeof(struct igmp_grouprec);
if (record_has_sources)
minrec0len += sizeof(in_addr_t);
CTR4(KTR_IGMPV3, "%s: queueing %s for %s/%s", __func__,
igmp_rec_type_to_str(type), inet_ntoa(inm->inm_addr),
inm->inm_ifp->if_xname);
/*
* Check if we have a packet in the tail of the queue for this
* group into which the first group record for this group will fit.
* Otherwise allocate a new packet.
* Always allocate leading space for IP+RA_OPT+IGMP+REPORT.
* Note: Group records for G/GSR query responses MUST be sent
* in their own packet.
*/
m0 = ifq->ifq_tail;
if (!is_group_query &&
m0 != NULL &&
(m0->m_pkthdr.PH_vt.vt_nrecs + 1 <= IGMP_V3_REPORT_MAXRECS) &&
(m0->m_pkthdr.len + minrec0len) <
(ifp->if_mtu - IGMP_LEADINGSPACE)) {
m0srcs = (ifp->if_mtu - m0->m_pkthdr.len -
sizeof(struct igmp_grouprec)) / sizeof(in_addr_t);
m = m0;
CTR1(KTR_IGMPV3, "%s: use existing packet", __func__);
} else {
if (_IF_QFULL(ifq)) {
CTR1(KTR_IGMPV3, "%s: outbound queue full", __func__);
return (-ENOMEM);
}
m = NULL;
m0srcs = (ifp->if_mtu - IGMP_LEADINGSPACE -
sizeof(struct igmp_grouprec)) / sizeof(in_addr_t);
if (!is_state_change && !is_group_query) {
m = m_getcl(M_NOWAIT, MT_DATA, M_PKTHDR);
if (m)
m->m_data += IGMP_LEADINGSPACE;
}
if (m == NULL) {
m = m_gethdr(M_NOWAIT, MT_DATA);
if (m)
MH_ALIGN(m, IGMP_LEADINGSPACE);
}
if (m == NULL)
return (-ENOMEM);
igmp_save_context(m, ifp);
CTR1(KTR_IGMPV3, "%s: allocated first packet", __func__);
}
/*
* Append group record.
* If we have sources, we don't know how many yet.
*/
ig.ig_type = type;
ig.ig_datalen = 0;
ig.ig_numsrc = 0;
ig.ig_group = inm->inm_addr;
if (!m_append(m, sizeof(struct igmp_grouprec), (void *)&ig)) {
if (m != m0)
m_freem(m);
CTR1(KTR_IGMPV3, "%s: m_append() failed.", __func__);
return (-ENOMEM);
}
nbytes += sizeof(struct igmp_grouprec);
/*
* Append as many sources as will fit in the first packet.
* If we are appending to a new packet, the chain allocation
* may potentially use clusters; use m_getptr() in this case.
* If we are appending to an existing packet, we need to obtain
* a pointer to the group record after m_append(), in case a new
* mbuf was allocated.
* Only append sources which are in-mode at t1. If we are
* transitioning to MCAST_UNDEFINED state on the group, do not
* include source entries.
* Only report recorded sources in our filter set when responding
* to a group-source query.
*/
if (record_has_sources) {
if (m == m0) {
md = m_last(m);
pig = (struct igmp_grouprec *)(mtod(md, uint8_t *) +
md->m_len - nbytes);
} else {
md = m_getptr(m, 0, &off);
pig = (struct igmp_grouprec *)(mtod(md, uint8_t *) +
off);
}
msrcs = 0;
RB_FOREACH_SAFE(ims, ip_msource_tree, &inm->inm_srcs, nims) {
CTR2(KTR_IGMPV3, "%s: visit node %s", __func__,
inet_ntoa_haddr(ims->ims_haddr));
now = ims_get_mode(inm, ims, 1);
CTR2(KTR_IGMPV3, "%s: node is %d", __func__, now);
if ((now != mode) ||
(now == mode && mode == MCAST_UNDEFINED)) {
CTR1(KTR_IGMPV3, "%s: skip node", __func__);
continue;
}
if (is_source_query && ims->ims_stp == 0) {
CTR1(KTR_IGMPV3, "%s: skip unrecorded node",
__func__);
continue;
}
CTR1(KTR_IGMPV3, "%s: append node", __func__);
naddr = htonl(ims->ims_haddr);
if (!m_append(m, sizeof(in_addr_t), (void *)&naddr)) {
if (m != m0)
m_freem(m);
CTR1(KTR_IGMPV3, "%s: m_append() failed.",
__func__);
return (-ENOMEM);
}
nbytes += sizeof(in_addr_t);
++msrcs;
if (msrcs == m0srcs)
break;
}
CTR2(KTR_IGMPV3, "%s: msrcs is %d this packet", __func__,
msrcs);
pig->ig_numsrc = htons(msrcs);
nbytes += (msrcs * sizeof(in_addr_t));
}
if (is_source_query && msrcs == 0) {
CTR1(KTR_IGMPV3, "%s: no recorded sources to report", __func__);
if (m != m0)
m_freem(m);
return (0);
}
/*
* We are good to go with first packet.
*/
if (m != m0) {
CTR1(KTR_IGMPV3, "%s: enqueueing first packet", __func__);
m->m_pkthdr.PH_vt.vt_nrecs = 1;
_IF_ENQUEUE(ifq, m);
} else
m->m_pkthdr.PH_vt.vt_nrecs++;
/*
* No further work needed if no source list in packet(s).
*/
if (!record_has_sources)
return (nbytes);
/*
* Whilst sources remain to be announced, we need to allocate
* a new packet and fill out as many sources as will fit.
* Always try for a cluster first.
*/
while (nims != NULL) {
if (_IF_QFULL(ifq)) {
CTR1(KTR_IGMPV3, "%s: outbound queue full", __func__);
return (-ENOMEM);
}
m = m_getcl(M_NOWAIT, MT_DATA, M_PKTHDR);
if (m)
m->m_data += IGMP_LEADINGSPACE;
if (m == NULL) {
m = m_gethdr(M_NOWAIT, MT_DATA);
if (m)
MH_ALIGN(m, IGMP_LEADINGSPACE);
}
if (m == NULL)
return (-ENOMEM);
igmp_save_context(m, ifp);
md = m_getptr(m, 0, &off);
pig = (struct igmp_grouprec *)(mtod(md, uint8_t *) + off);
CTR1(KTR_IGMPV3, "%s: allocated next packet", __func__);
if (!m_append(m, sizeof(struct igmp_grouprec), (void *)&ig)) {
if (m != m0)
m_freem(m);
CTR1(KTR_IGMPV3, "%s: m_append() failed.", __func__);
return (-ENOMEM);
}
m->m_pkthdr.PH_vt.vt_nrecs = 1;
nbytes += sizeof(struct igmp_grouprec);
m0srcs = (ifp->if_mtu - IGMP_LEADINGSPACE -
sizeof(struct igmp_grouprec)) / sizeof(in_addr_t);
msrcs = 0;
RB_FOREACH_FROM(ims, ip_msource_tree, nims) {
CTR2(KTR_IGMPV3, "%s: visit node %s", __func__,
inet_ntoa_haddr(ims->ims_haddr));
now = ims_get_mode(inm, ims, 1);
if ((now != mode) ||
(now == mode && mode == MCAST_UNDEFINED)) {
CTR1(KTR_IGMPV3, "%s: skip node", __func__);
continue;
}
if (is_source_query && ims->ims_stp == 0) {
CTR1(KTR_IGMPV3, "%s: skip unrecorded node",
__func__);
continue;
}
CTR1(KTR_IGMPV3, "%s: append node", __func__);
naddr = htonl(ims->ims_haddr);
if (!m_append(m, sizeof(in_addr_t), (void *)&naddr)) {
if (m != m0)
m_freem(m);
CTR1(KTR_IGMPV3, "%s: m_append() failed.",
__func__);
return (-ENOMEM);
}
++msrcs;
if (msrcs == m0srcs)
break;
}
pig->ig_numsrc = htons(msrcs);
nbytes += (msrcs * sizeof(in_addr_t));
CTR1(KTR_IGMPV3, "%s: enqueueing next packet", __func__);
_IF_ENQUEUE(ifq, m);
}
return (nbytes);
}
/*
* Type used to mark record pass completion.
* We exploit the fact we can cast to this easily from the
* current filter modes on each ip_msource node.
*/
typedef enum {
REC_NONE = 0x00, /* MCAST_UNDEFINED */
REC_ALLOW = 0x01, /* MCAST_INCLUDE */
REC_BLOCK = 0x02, /* MCAST_EXCLUDE */
REC_FULL = REC_ALLOW | REC_BLOCK
} rectype_t;
/*
* Enqueue an IGMPv3 filter list change to the given output queue.
*
* Source list filter state is held in an RB-tree. When the filter list
* for a group is changed without changing its mode, we need to compute
* the deltas between T0 and T1 for each source in the filter set,
* and enqueue the appropriate ALLOW_NEW/BLOCK_OLD records.
*
* As we may potentially queue two record types, and the entire R-B tree
* needs to be walked at once, we break this out into its own function
* so we can generate a tightly packed queue of packets.
*
* XXX This could be written to only use one tree walk, although that makes
* serializing into the mbuf chains a bit harder. For now we do two walks
* which makes things easier on us, and it may or may not be harder on
* the L2 cache.
*
* If successful the size of all data appended to the queue is returned,
* otherwise an error code less than zero is returned, or zero if
* no record(s) were appended.
*/
static int
igmp_v3_enqueue_filter_change(struct ifqueue *ifq, struct in_multi *inm)
{
static const int MINRECLEN =
sizeof(struct igmp_grouprec) + sizeof(in_addr_t);
struct ifnet *ifp;
struct igmp_grouprec ig;
struct igmp_grouprec *pig;
struct ip_msource *ims, *nims;
struct mbuf *m, *m0, *md;
in_addr_t naddr;
int m0srcs, nbytes, npbytes, off, rsrcs, schanged;
int nallow, nblock;
uint8_t mode, now, then;
rectype_t crt, drt, nrt;
IN_MULTI_LOCK_ASSERT();
if (inm->inm_nsrc == 0 ||
(inm->inm_st[0].iss_asm > 0 && inm->inm_st[1].iss_asm > 0))
return (0);
ifp = inm->inm_ifp; /* interface */
mode = inm->inm_st[1].iss_fmode; /* filter mode at t1 */
crt = REC_NONE; /* current group record type */
drt = REC_NONE; /* mask of completed group record types */
nrt = REC_NONE; /* record type for current node */
m0srcs = 0; /* # source which will fit in current mbuf chain */
nbytes = 0; /* # of bytes appended to group's state-change queue */
npbytes = 0; /* # of bytes appended this packet */
rsrcs = 0; /* # sources encoded in current record */
schanged = 0; /* # nodes encoded in overall filter change */
nallow = 0; /* # of source entries in ALLOW_NEW */
nblock = 0; /* # of source entries in BLOCK_OLD */
nims = NULL; /* next tree node pointer */
/*
* For each possible filter record mode.
* The first kind of source we encounter tells us which
* is the first kind of record we start appending.
* If a node transitioned to UNDEFINED at t1, its mode is treated
* as the inverse of the group's filter mode.
*/
while (drt != REC_FULL) {
do {
m0 = ifq->ifq_tail;
if (m0 != NULL &&
(m0->m_pkthdr.PH_vt.vt_nrecs + 1 <=
IGMP_V3_REPORT_MAXRECS) &&
(m0->m_pkthdr.len + MINRECLEN) <
(ifp->if_mtu - IGMP_LEADINGSPACE)) {
m = m0;
m0srcs = (ifp->if_mtu - m0->m_pkthdr.len -
sizeof(struct igmp_grouprec)) /
sizeof(in_addr_t);
CTR1(KTR_IGMPV3,
"%s: use previous packet", __func__);
} else {
m = m_getcl(M_NOWAIT, MT_DATA, M_PKTHDR);
if (m)
m->m_data += IGMP_LEADINGSPACE;
if (m == NULL) {
m = m_gethdr(M_NOWAIT, MT_DATA);
if (m)
MH_ALIGN(m, IGMP_LEADINGSPACE);
}
if (m == NULL) {
CTR1(KTR_IGMPV3,
"%s: m_get*() failed", __func__);
return (-ENOMEM);
}
m->m_pkthdr.PH_vt.vt_nrecs = 0;
igmp_save_context(m, ifp);
m0srcs = (ifp->if_mtu - IGMP_LEADINGSPACE -
sizeof(struct igmp_grouprec)) /
sizeof(in_addr_t);
npbytes = 0;
CTR1(KTR_IGMPV3,
"%s: allocated new packet", __func__);
}
/*
* Append the IGMP group record header to the
* current packet's data area.
* Recalculate pointer to free space for next
* group record, in case m_append() allocated
* a new mbuf or cluster.
*/
memset(&ig, 0, sizeof(ig));
ig.ig_group = inm->inm_addr;
if (!m_append(m, sizeof(ig), (void *)&ig)) {
if (m != m0)
m_freem(m);
CTR1(KTR_IGMPV3,
"%s: m_append() failed", __func__);
return (-ENOMEM);
}
npbytes += sizeof(struct igmp_grouprec);
if (m != m0) {
/* new packet; offset in c hain */
md = m_getptr(m, npbytes -
sizeof(struct igmp_grouprec), &off);
pig = (struct igmp_grouprec *)(mtod(md,
uint8_t *) + off);
} else {
/* current packet; offset from last append */
md = m_last(m);
pig = (struct igmp_grouprec *)(mtod(md,
uint8_t *) + md->m_len -
sizeof(struct igmp_grouprec));
}
/*
* Begin walking the tree for this record type
* pass, or continue from where we left off
* previously if we had to allocate a new packet.
* Only report deltas in-mode at t1.
* We need not report included sources as allowed
* if we are in inclusive mode on the group,
* however the converse is not true.
*/
rsrcs = 0;
if (nims == NULL)
nims = RB_MIN(ip_msource_tree, &inm->inm_srcs);
RB_FOREACH_FROM(ims, ip_msource_tree, nims) {
CTR2(KTR_IGMPV3, "%s: visit node %s",
__func__, inet_ntoa_haddr(ims->ims_haddr));
now = ims_get_mode(inm, ims, 1);
then = ims_get_mode(inm, ims, 0);
CTR3(KTR_IGMPV3, "%s: mode: t0 %d, t1 %d",
__func__, then, now);
if (now == then) {
CTR1(KTR_IGMPV3,
"%s: skip unchanged", __func__);
continue;
}
if (mode == MCAST_EXCLUDE &&
now == MCAST_INCLUDE) {
CTR1(KTR_IGMPV3,
"%s: skip IN src on EX group",
__func__);
continue;
}
nrt = (rectype_t)now;
if (nrt == REC_NONE)
nrt = (rectype_t)(~mode & REC_FULL);
if (schanged++ == 0) {
crt = nrt;
} else if (crt != nrt)
continue;
naddr = htonl(ims->ims_haddr);
if (!m_append(m, sizeof(in_addr_t),
(void *)&naddr)) {
if (m != m0)
m_freem(m);
CTR1(KTR_IGMPV3,
"%s: m_append() failed", __func__);
return (-ENOMEM);
}
nallow += !!(crt == REC_ALLOW);
nblock += !!(crt == REC_BLOCK);
if (++rsrcs == m0srcs)
break;
}
/*
* If we did not append any tree nodes on this
* pass, back out of allocations.
*/
if (rsrcs == 0) {
npbytes -= sizeof(struct igmp_grouprec);
if (m != m0) {
CTR1(KTR_IGMPV3,
"%s: m_free(m)", __func__);
m_freem(m);
} else {
CTR1(KTR_IGMPV3,
"%s: m_adj(m, -ig)", __func__);
m_adj(m, -((int)sizeof(
struct igmp_grouprec)));
}
continue;
}
npbytes += (rsrcs * sizeof(in_addr_t));
if (crt == REC_ALLOW)
pig->ig_type = IGMP_ALLOW_NEW_SOURCES;
else if (crt == REC_BLOCK)
pig->ig_type = IGMP_BLOCK_OLD_SOURCES;
pig->ig_numsrc = htons(rsrcs);
/*
* Count the new group record, and enqueue this
* packet if it wasn't already queued.
*/
m->m_pkthdr.PH_vt.vt_nrecs++;
if (m != m0)
_IF_ENQUEUE(ifq, m);
nbytes += npbytes;
} while (nims != NULL);
drt |= crt;
crt = (~crt & REC_FULL);
}
CTR3(KTR_IGMPV3, "%s: queued %d ALLOW_NEW, %d BLOCK_OLD", __func__,
nallow, nblock);
return (nbytes);
}
static int
igmp_v3_merge_state_changes(struct in_multi *inm, struct ifqueue *ifscq)
{
struct ifqueue *gq;
struct mbuf *m; /* pending state-change */
struct mbuf *m0; /* copy of pending state-change */
struct mbuf *mt; /* last state-change in packet */
int docopy, domerge;
u_int recslen;
docopy = 0;
domerge = 0;
recslen = 0;
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
/*
* If there are further pending retransmissions, make a writable
* copy of each queued state-change message before merging.
*/
if (inm->inm_scrv > 0)
docopy = 1;
gq = &inm->inm_scq;
#ifdef KTR
if (gq->ifq_head == NULL) {
CTR2(KTR_IGMPV3, "%s: WARNING: queue for inm %p is empty",
__func__, inm);
}
#endif
m = gq->ifq_head;
while (m != NULL) {
/*
* Only merge the report into the current packet if
* there is sufficient space to do so; an IGMPv3 report
* packet may only contain 65,535 group records.
* Always use a simple mbuf chain concatentation to do this,
* as large state changes for single groups may have
* allocated clusters.
*/
domerge = 0;
mt = ifscq->ifq_tail;
if (mt != NULL) {
recslen = m_length(m, NULL);
if ((mt->m_pkthdr.PH_vt.vt_nrecs +
m->m_pkthdr.PH_vt.vt_nrecs <=
IGMP_V3_REPORT_MAXRECS) &&
(mt->m_pkthdr.len + recslen <=
(inm->inm_ifp->if_mtu - IGMP_LEADINGSPACE)))
domerge = 1;
}
if (!domerge && _IF_QFULL(gq)) {
CTR2(KTR_IGMPV3,
"%s: outbound queue full, skipping whole packet %p",
__func__, m);
mt = m->m_nextpkt;
if (!docopy)
m_freem(m);
m = mt;
continue;
}
if (!docopy) {
CTR2(KTR_IGMPV3, "%s: dequeueing %p", __func__, m);
_IF_DEQUEUE(gq, m0);
m = m0->m_nextpkt;
} else {
CTR2(KTR_IGMPV3, "%s: copying %p", __func__, m);
m0 = m_dup(m, M_NOWAIT);
if (m0 == NULL)
return (ENOMEM);
m0->m_nextpkt = NULL;
m = m->m_nextpkt;
}
if (!domerge) {
CTR3(KTR_IGMPV3, "%s: queueing %p to ifscq %p)",
__func__, m0, ifscq);
_IF_ENQUEUE(ifscq, m0);
} else {
struct mbuf *mtl; /* last mbuf of packet mt */
CTR3(KTR_IGMPV3, "%s: merging %p with ifscq tail %p)",
__func__, m0, mt);
mtl = m_last(mt);
m0->m_flags &= ~M_PKTHDR;
mt->m_pkthdr.len += recslen;
mt->m_pkthdr.PH_vt.vt_nrecs +=
m0->m_pkthdr.PH_vt.vt_nrecs;
mtl->m_next = m0;
}
}
return (0);
}
/*
* Respond to a pending IGMPv3 General Query.
*/
static void
igmp_v3_dispatch_general_query(struct igmp_ifinfo *igi)
{
struct ifmultiaddr *ifma;
struct ifnet *ifp;
struct in_multi *inm;
int retval, loop;
IN_MULTI_LOCK_ASSERT();
IGMP_LOCK_ASSERT();
KASSERT(igi->igi_version == IGMP_VERSION_3,
("%s: called when version %d", __func__, igi->igi_version));
ifp = igi->igi_ifp;
IF_ADDR_RLOCK(ifp);
TAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
if (ifma->ifma_addr->sa_family != AF_INET ||
ifma->ifma_protospec == NULL)
continue;
inm = (struct in_multi *)ifma->ifma_protospec;
KASSERT(ifp == inm->inm_ifp,
("%s: inconsistent ifp", __func__));
switch (inm->inm_state) {
case IGMP_NOT_MEMBER:
case IGMP_SILENT_MEMBER:
break;
case IGMP_REPORTING_MEMBER:
case IGMP_IDLE_MEMBER:
case IGMP_LAZY_MEMBER:
case IGMP_SLEEPING_MEMBER:
case IGMP_AWAKENING_MEMBER:
inm->inm_state = IGMP_REPORTING_MEMBER;
retval = igmp_v3_enqueue_group_record(&igi->igi_gq,
inm, 0, 0, 0);
CTR2(KTR_IGMPV3, "%s: enqueue record = %d",
__func__, retval);
break;
case IGMP_G_QUERY_PENDING_MEMBER:
case IGMP_SG_QUERY_PENDING_MEMBER:
case IGMP_LEAVING_MEMBER:
break;
}
}
IF_ADDR_RUNLOCK(ifp);
loop = (igi->igi_flags & IGIF_LOOPBACK) ? 1 : 0;
igmp_dispatch_queue(&igi->igi_gq, IGMP_MAX_RESPONSE_BURST, loop);
/*
* Slew transmission of bursts over 500ms intervals.
*/
if (igi->igi_gq.ifq_head != NULL) {
igi->igi_v3_timer = 1 + IGMP_RANDOM_DELAY(
IGMP_RESPONSE_BURST_INTERVAL);
V_interface_timers_running = 1;
}
}
/*
* Transmit the next pending IGMP message in the output queue.
*
* We get called from netisr_processqueue(). A mutex private to igmpoq
* will be acquired and released around this routine.
*
* VIMAGE: Needs to store/restore vnet pointer on a per-mbuf-chain basis.
* MRT: Nothing needs to be done, as IGMP traffic is always local to
* a link and uses a link-scope multicast address.
*/
static void
igmp_intr(struct mbuf *m)
{
struct ip_moptions imo;
struct ifnet *ifp;
struct mbuf *ipopts, *m0;
int error;
uint32_t ifindex;
CTR2(KTR_IGMPV3, "%s: transmit %p", __func__, m);
/*
* Set VNET image pointer from enqueued mbuf chain
* before doing anything else. Whilst we use interface
* indexes to guard against interface detach, they are
* unique to each VIMAGE and must be retrieved.
*/
CURVNET_SET((struct vnet *)(m->m_pkthdr.header));
ifindex = igmp_restore_context(m);
/*
* Check if the ifnet still exists. This limits the scope of
* any race in the absence of a global ifp lock for low cost
* (an array lookup).
*/
ifp = ifnet_byindex(ifindex);
if (ifp == NULL) {
CTR3(KTR_IGMPV3, "%s: dropped %p as ifindex %u went away.",
__func__, m, ifindex);
m_freem(m);
IPSTAT_INC(ips_noroute);
goto out;
}
ipopts = V_igmp_sendra ? m_raopt : NULL;
imo.imo_multicast_ttl = 1;
imo.imo_multicast_vif = -1;
imo.imo_multicast_loop = (V_ip_mrouter != NULL);
/*
* If the user requested that IGMP traffic be explicitly
* redirected to the loopback interface (e.g. they are running a
* MANET interface and the routing protocol needs to see the
* updates), handle this now.
*/
if (m->m_flags & M_IGMP_LOOP)
imo.imo_multicast_ifp = V_loif;
else
imo.imo_multicast_ifp = ifp;
if (m->m_flags & M_IGMPV2) {
m0 = m;
} else {
m0 = igmp_v3_encap_report(ifp, m);
if (m0 == NULL) {
CTR2(KTR_IGMPV3, "%s: dropped %p", __func__, m);
m_freem(m);
IPSTAT_INC(ips_odropped);
goto out;
}
}
igmp_scrub_context(m0);
m->m_flags &= ~(M_PROTOFLAGS);
m0->m_pkthdr.rcvif = V_loif;
#ifdef MAC
mac_netinet_igmp_send(ifp, m0);
#endif
error = ip_output(m0, ipopts, NULL, 0, &imo, NULL);
if (error) {
CTR3(KTR_IGMPV3, "%s: ip_output(%p) = %d", __func__, m0, error);
goto out;
}
IGMPSTAT_INC(igps_snd_reports);
out:
/*
* We must restore the existing vnet pointer before
* continuing as we are run from netisr context.
*/
CURVNET_RESTORE();
}
/*
* Encapsulate an IGMPv3 report.
*
* The internal mbuf flag M_IGMPV3_HDR is used to indicate that the mbuf
* chain has already had its IP/IGMPv3 header prepended. In this case
* the function will not attempt to prepend; the lengths and checksums
* will however be re-computed.
*
* Returns a pointer to the new mbuf chain head, or NULL if the
* allocation failed.
*/
static struct mbuf *
igmp_v3_encap_report(struct ifnet *ifp, struct mbuf *m)
{
struct igmp_report *igmp;
struct ip *ip;
int hdrlen, igmpreclen;
KASSERT((m->m_flags & M_PKTHDR),
("%s: mbuf chain %p is !M_PKTHDR", __func__, m));
igmpreclen = m_length(m, NULL);
hdrlen = sizeof(struct ip) + sizeof(struct igmp_report);
if (m->m_flags & M_IGMPV3_HDR) {
igmpreclen -= hdrlen;
} else {
M_PREPEND(m, hdrlen, M_NOWAIT);
if (m == NULL)
return (NULL);
m->m_flags |= M_IGMPV3_HDR;
}
CTR2(KTR_IGMPV3, "%s: igmpreclen is %d", __func__, igmpreclen);
m->m_data += sizeof(struct ip);
m->m_len -= sizeof(struct ip);
igmp = mtod(m, struct igmp_report *);
igmp->ir_type = IGMP_v3_HOST_MEMBERSHIP_REPORT;
igmp->ir_rsv1 = 0;
igmp->ir_rsv2 = 0;
igmp->ir_numgrps = htons(m->m_pkthdr.PH_vt.vt_nrecs);
igmp->ir_cksum = 0;
igmp->ir_cksum = in_cksum(m, sizeof(struct igmp_report) + igmpreclen);
m->m_pkthdr.PH_vt.vt_nrecs = 0;
m->m_data -= sizeof(struct ip);
m->m_len += sizeof(struct ip);
ip = mtod(m, struct ip *);
ip->ip_tos = IPTOS_PREC_INTERNETCONTROL;
ip->ip_len = htons(hdrlen + igmpreclen);
ip->ip_off = htons(IP_DF);
ip->ip_p = IPPROTO_IGMP;
ip->ip_sum = 0;
ip->ip_src.s_addr = INADDR_ANY;
if (m->m_flags & M_IGMP_LOOP) {
struct in_ifaddr *ia;
IFP_TO_IA(ifp, ia);
if (ia != NULL) {
ip->ip_src = ia->ia_addr.sin_addr;
ifa_free(&ia->ia_ifa);
}
}
ip->ip_dst.s_addr = htonl(INADDR_ALLRPTS_GROUP);
return (m);
}
#ifdef KTR
static char *
igmp_rec_type_to_str(const int type)
{
switch (type) {
case IGMP_CHANGE_TO_EXCLUDE_MODE:
return "TO_EX";
break;
case IGMP_CHANGE_TO_INCLUDE_MODE:
return "TO_IN";
break;
case IGMP_MODE_IS_EXCLUDE:
return "MODE_EX";
break;
case IGMP_MODE_IS_INCLUDE:
return "MODE_IN";
break;
case IGMP_ALLOW_NEW_SOURCES:
return "ALLOW_NEW";
break;
case IGMP_BLOCK_OLD_SOURCES:
return "BLOCK_OLD";
break;
default:
break;
}
return "unknown";
}
#endif
static void
igmp_init(void *unused __unused)
{
CTR1(KTR_IGMPV3, "%s: initializing", __func__);
IGMP_LOCK_INIT();
m_raopt = igmp_ra_alloc();
netisr_register(&igmp_nh);
}
SYSINIT(igmp_init, SI_SUB_PSEUDO, SI_ORDER_MIDDLE, igmp_init, NULL);
static void
igmp_uninit(void *unused __unused)
{
CTR1(KTR_IGMPV3, "%s: tearing down", __func__);
netisr_unregister(&igmp_nh);
m_free(m_raopt);
m_raopt = NULL;
IGMP_LOCK_DESTROY();
}
SYSUNINIT(igmp_uninit, SI_SUB_PSEUDO, SI_ORDER_MIDDLE, igmp_uninit, NULL);
static void
vnet_igmp_init(const void *unused __unused)
{
CTR1(KTR_IGMPV3, "%s: initializing", __func__);
LIST_INIT(&V_igi_head);
}
VNET_SYSINIT(vnet_igmp_init, SI_SUB_PSEUDO, SI_ORDER_ANY, vnet_igmp_init,
NULL);
static void
vnet_igmp_uninit(const void *unused __unused)
{
CTR1(KTR_IGMPV3, "%s: tearing down", __func__);
KASSERT(LIST_EMPTY(&V_igi_head),
("%s: igi list not empty; ifnets not detached?", __func__));
}
VNET_SYSUNINIT(vnet_igmp_uninit, SI_SUB_PSEUDO, SI_ORDER_ANY,
vnet_igmp_uninit, NULL);
static int
igmp_modevent(module_t mod, int type, void *unused __unused)
{
switch (type) {
case MOD_LOAD:
case MOD_UNLOAD:
break;
default:
return (EOPNOTSUPP);
}
return (0);
}
static moduledata_t igmp_mod = {
"igmp",
igmp_modevent,
0
};
DECLARE_MODULE(igmp, igmp_mod, SI_SUB_PSEUDO, SI_ORDER_ANY);
|