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
|
/*-
* Copyright (c) 1982, 1986, 1988, 1990, 1993, 1995
* The Regents of the University of California. All rights reserved.
*
* 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.
*
* @(#)tcp_subr.c 8.2 (Berkeley) 5/24/95
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_compat.h"
#include "opt_inet.h"
#include "opt_inet6.h"
#include "opt_ipsec.h"
#include "opt_tcpdebug.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/callout.h>
#include <sys/eventhandler.h>
#include <sys/hhook.h>
#include <sys/kernel.h>
#include <sys/khelp.h>
#include <sys/sysctl.h>
#include <sys/jail.h>
#include <sys/malloc.h>
#include <sys/refcount.h>
#include <sys/mbuf.h>
#ifdef INET6
#include <sys/domain.h>
#endif
#include <sys/priv.h>
#include <sys/proc.h>
#include <sys/sdt.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/protosw.h>
#include <sys/random.h>
#include <vm/uma.h>
#include <net/route.h>
#include <net/if.h>
#include <net/if_var.h>
#include <net/vnet.h>
#include <netinet/in.h>
#include <netinet/in_fib.h>
#include <netinet/in_kdtrace.h>
#include <netinet/in_pcb.h>
#include <netinet/in_systm.h>
#include <netinet/in_var.h>
#include <netinet/ip.h>
#include <netinet/ip_icmp.h>
#include <netinet/ip_var.h>
#ifdef INET6
#include <netinet/ip6.h>
#include <netinet6/in6_fib.h>
#include <netinet6/in6_pcb.h>
#include <netinet6/ip6_var.h>
#include <netinet6/scope6_var.h>
#include <netinet6/nd6.h>
#endif
#ifdef TCP_RFC7413
#include <netinet/tcp_fastopen.h>
#endif
#include <netinet/tcp.h>
#include <netinet/tcp_fsm.h>
#include <netinet/tcp_seq.h>
#include <netinet/tcp_timer.h>
#include <netinet/tcp_var.h>
#include <netinet/tcp_syncache.h>
#include <netinet/cc/cc.h>
#ifdef INET6
#include <netinet6/tcp6_var.h>
#endif
#include <netinet/tcpip.h>
#ifdef TCPPCAP
#include <netinet/tcp_pcap.h>
#endif
#ifdef TCPDEBUG
#include <netinet/tcp_debug.h>
#endif
#ifdef INET6
#include <netinet6/ip6protosw.h>
#endif
#ifdef TCP_OFFLOAD
#include <netinet/tcp_offload.h>
#endif
#ifdef IPSEC
#include <netipsec/ipsec.h>
#include <netipsec/xform.h>
#ifdef INET6
#include <netipsec/ipsec6.h>
#endif
#include <netipsec/key.h>
#include <sys/syslog.h>
#endif /*IPSEC*/
#include <machine/in_cksum.h>
#include <sys/md5.h>
#include <security/mac/mac_framework.h>
VNET_DEFINE(int, tcp_mssdflt) = TCP_MSS;
#ifdef INET6
VNET_DEFINE(int, tcp_v6mssdflt) = TCP6_MSS;
#endif
struct rwlock tcp_function_lock;
static int
sysctl_net_inet_tcp_mss_check(SYSCTL_HANDLER_ARGS)
{
int error, new;
new = V_tcp_mssdflt;
error = sysctl_handle_int(oidp, &new, 0, req);
if (error == 0 && req->newptr) {
if (new < TCP_MINMSS)
error = EINVAL;
else
V_tcp_mssdflt = new;
}
return (error);
}
SYSCTL_PROC(_net_inet_tcp, TCPCTL_MSSDFLT, mssdflt,
CTLFLAG_VNET | CTLTYPE_INT | CTLFLAG_RW, &VNET_NAME(tcp_mssdflt), 0,
&sysctl_net_inet_tcp_mss_check, "I",
"Default TCP Maximum Segment Size");
#ifdef INET6
static int
sysctl_net_inet_tcp_mss_v6_check(SYSCTL_HANDLER_ARGS)
{
int error, new;
new = V_tcp_v6mssdflt;
error = sysctl_handle_int(oidp, &new, 0, req);
if (error == 0 && req->newptr) {
if (new < TCP_MINMSS)
error = EINVAL;
else
V_tcp_v6mssdflt = new;
}
return (error);
}
SYSCTL_PROC(_net_inet_tcp, TCPCTL_V6MSSDFLT, v6mssdflt,
CTLFLAG_VNET | CTLTYPE_INT | CTLFLAG_RW, &VNET_NAME(tcp_v6mssdflt), 0,
&sysctl_net_inet_tcp_mss_v6_check, "I",
"Default TCP Maximum Segment Size for IPv6");
#endif /* INET6 */
/*
* Minimum MSS we accept and use. This prevents DoS attacks where
* we are forced to a ridiculous low MSS like 20 and send hundreds
* of packets instead of one. The effect scales with the available
* bandwidth and quickly saturates the CPU and network interface
* with packet generation and sending. Set to zero to disable MINMSS
* checking. This setting prevents us from sending too small packets.
*/
VNET_DEFINE(int, tcp_minmss) = TCP_MINMSS;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, minmss, CTLFLAG_VNET | CTLFLAG_RW,
&VNET_NAME(tcp_minmss), 0,
"Minimum TCP Maximum Segment Size");
VNET_DEFINE(int, tcp_do_rfc1323) = 1;
SYSCTL_INT(_net_inet_tcp, TCPCTL_DO_RFC1323, rfc1323, CTLFLAG_VNET | CTLFLAG_RW,
&VNET_NAME(tcp_do_rfc1323), 0,
"Enable rfc1323 (high performance TCP) extensions");
static int tcp_log_debug = 0;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, log_debug, CTLFLAG_RW,
&tcp_log_debug, 0, "Log errors caused by incoming TCP segments");
static int tcp_tcbhashsize;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, tcbhashsize, CTLFLAG_RDTUN | CTLFLAG_NOFETCH,
&tcp_tcbhashsize, 0, "Size of TCP control-block hashtable");
static int do_tcpdrain = 1;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, do_tcpdrain, CTLFLAG_RW, &do_tcpdrain, 0,
"Enable tcp_drain routine for extra help when low on mbufs");
SYSCTL_UINT(_net_inet_tcp, OID_AUTO, pcbcount, CTLFLAG_VNET | CTLFLAG_RD,
&VNET_NAME(tcbinfo.ipi_count), 0, "Number of active PCBs");
static VNET_DEFINE(int, icmp_may_rst) = 1;
#define V_icmp_may_rst VNET(icmp_may_rst)
SYSCTL_INT(_net_inet_tcp, OID_AUTO, icmp_may_rst, CTLFLAG_VNET | CTLFLAG_RW,
&VNET_NAME(icmp_may_rst), 0,
"Certain ICMP unreachable messages may abort connections in SYN_SENT");
static VNET_DEFINE(int, tcp_isn_reseed_interval) = 0;
#define V_tcp_isn_reseed_interval VNET(tcp_isn_reseed_interval)
SYSCTL_INT(_net_inet_tcp, OID_AUTO, isn_reseed_interval, CTLFLAG_VNET | CTLFLAG_RW,
&VNET_NAME(tcp_isn_reseed_interval), 0,
"Seconds between reseeding of ISN secret");
static int tcp_soreceive_stream;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, soreceive_stream, CTLFLAG_RDTUN,
&tcp_soreceive_stream, 0, "Using soreceive_stream for TCP sockets");
#ifdef TCP_SIGNATURE
static int tcp_sig_checksigs = 1;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, signature_verify_input, CTLFLAG_RW,
&tcp_sig_checksigs, 0, "Verify RFC2385 digests on inbound traffic");
#endif
VNET_DEFINE(uma_zone_t, sack_hole_zone);
#define V_sack_hole_zone VNET(sack_hole_zone)
VNET_DEFINE(struct hhook_head *, tcp_hhh[HHOOK_TCP_LAST+1]);
static struct inpcb *tcp_notify(struct inpcb *, int);
static struct inpcb *tcp_mtudisc_notify(struct inpcb *, int);
static void tcp_mtudisc(struct inpcb *, int);
static char * tcp_log_addr(struct in_conninfo *inc, struct tcphdr *th,
void *ip4hdr, const void *ip6hdr);
static struct tcp_function_block tcp_def_funcblk = {
"default",
tcp_output,
tcp_do_segment,
tcp_default_ctloutput,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
0,
0
};
int t_functions_inited = 0;
struct tcp_funchead t_functions;
static struct tcp_function_block *tcp_func_set_ptr = &tcp_def_funcblk;
static void
init_tcp_functions(void)
{
if (t_functions_inited == 0) {
TAILQ_INIT(&t_functions);
rw_init_flags(&tcp_function_lock, "tcp_func_lock" , 0);
t_functions_inited = 1;
}
}
static struct tcp_function_block *
find_tcp_functions_locked(struct tcp_function_set *fs)
{
struct tcp_function *f;
struct tcp_function_block *blk=NULL;
TAILQ_FOREACH(f, &t_functions, tf_next) {
if (strcmp(f->tf_fb->tfb_tcp_block_name, fs->function_set_name) == 0) {
blk = f->tf_fb;
break;
}
}
return(blk);
}
static struct tcp_function_block *
find_tcp_fb_locked(struct tcp_function_block *blk, struct tcp_function **s)
{
struct tcp_function_block *rblk=NULL;
struct tcp_function *f;
TAILQ_FOREACH(f, &t_functions, tf_next) {
if (f->tf_fb == blk) {
rblk = blk;
if (s) {
*s = f;
}
break;
}
}
return (rblk);
}
struct tcp_function_block *
find_and_ref_tcp_functions(struct tcp_function_set *fs)
{
struct tcp_function_block *blk;
rw_rlock(&tcp_function_lock);
blk = find_tcp_functions_locked(fs);
if (blk)
refcount_acquire(&blk->tfb_refcnt);
rw_runlock(&tcp_function_lock);
return(blk);
}
struct tcp_function_block *
find_and_ref_tcp_fb(struct tcp_function_block *blk)
{
struct tcp_function_block *rblk;
rw_rlock(&tcp_function_lock);
rblk = find_tcp_fb_locked(blk, NULL);
if (rblk)
refcount_acquire(&rblk->tfb_refcnt);
rw_runlock(&tcp_function_lock);
return(rblk);
}
static int
sysctl_net_inet_default_tcp_functions(SYSCTL_HANDLER_ARGS)
{
int error=ENOENT;
struct tcp_function_set fs;
struct tcp_function_block *blk;
memset(&fs, 0, sizeof(fs));
rw_rlock(&tcp_function_lock);
blk = find_tcp_fb_locked(tcp_func_set_ptr, NULL);
if (blk) {
/* Found him */
strcpy(fs.function_set_name, blk->tfb_tcp_block_name);
fs.pcbcnt = blk->tfb_refcnt;
}
rw_runlock(&tcp_function_lock);
error = sysctl_handle_string(oidp, fs.function_set_name,
sizeof(fs.function_set_name), req);
/* Check for error or no change */
if (error != 0 || req->newptr == NULL)
return(error);
rw_wlock(&tcp_function_lock);
blk = find_tcp_functions_locked(&fs);
if ((blk == NULL) ||
(blk->tfb_flags & TCP_FUNC_BEING_REMOVED)) {
error = ENOENT;
goto done;
}
tcp_func_set_ptr = blk;
done:
rw_wunlock(&tcp_function_lock);
return (error);
}
SYSCTL_PROC(_net_inet_tcp, OID_AUTO, functions_default,
CTLTYPE_STRING | CTLFLAG_RW,
NULL, 0, sysctl_net_inet_default_tcp_functions, "A",
"Set/get the default TCP functions");
static int
sysctl_net_inet_list_available(SYSCTL_HANDLER_ARGS)
{
int error, cnt, linesz;
struct tcp_function *f;
char *buffer, *cp;
size_t bufsz, outsz;
cnt = 0;
rw_rlock(&tcp_function_lock);
TAILQ_FOREACH(f, &t_functions, tf_next) {
cnt++;
}
rw_runlock(&tcp_function_lock);
bufsz = (cnt+2) * (TCP_FUNCTION_NAME_LEN_MAX + 12) + 1;
buffer = malloc(bufsz, M_TEMP, M_WAITOK);
error = 0;
cp = buffer;
linesz = snprintf(cp, bufsz, "\n%-32s%c %s\n", "Stack", 'D', "PCB count");
cp += linesz;
bufsz -= linesz;
outsz = linesz;
rw_rlock(&tcp_function_lock);
TAILQ_FOREACH(f, &t_functions, tf_next) {
linesz = snprintf(cp, bufsz, "%-32s%c %u\n",
f->tf_fb->tfb_tcp_block_name,
(f->tf_fb == tcp_func_set_ptr) ? '*' : ' ',
f->tf_fb->tfb_refcnt);
if (linesz >= bufsz) {
error = EOVERFLOW;
break;
}
cp += linesz;
bufsz -= linesz;
outsz += linesz;
}
rw_runlock(&tcp_function_lock);
if (error == 0)
error = sysctl_handle_string(oidp, buffer, outsz + 1, req);
free(buffer, M_TEMP);
return (error);
}
SYSCTL_PROC(_net_inet_tcp, OID_AUTO, functions_available,
CTLTYPE_STRING|CTLFLAG_RD,
NULL, 0, sysctl_net_inet_list_available, "A",
"list available TCP Function sets");
/*
* Target size of TCP PCB hash tables. Must be a power of two.
*
* Note that this can be overridden by the kernel environment
* variable net.inet.tcp.tcbhashsize
*/
#ifndef TCBHASHSIZE
#define TCBHASHSIZE 0
#endif
/*
* XXX
* Callouts should be moved into struct tcp directly. They are currently
* separate because the tcpcb structure is exported to userland for sysctl
* parsing purposes, which do not know about callouts.
*/
struct tcpcb_mem {
struct tcpcb tcb;
struct tcp_timer tt;
struct cc_var ccv;
struct osd osd;
};
static VNET_DEFINE(uma_zone_t, tcpcb_zone);
#define V_tcpcb_zone VNET(tcpcb_zone)
MALLOC_DEFINE(M_TCPLOG, "tcplog", "TCP address and flags print buffers");
MALLOC_DEFINE(M_TCPFUNCTIONS, "tcpfunc", "TCP function set memory");
static struct mtx isn_mtx;
#define ISN_LOCK_INIT() mtx_init(&isn_mtx, "isn_mtx", NULL, MTX_DEF)
#define ISN_LOCK() mtx_lock(&isn_mtx)
#define ISN_UNLOCK() mtx_unlock(&isn_mtx)
/*
* TCP initialization.
*/
static void
tcp_zone_change(void *tag)
{
uma_zone_set_max(V_tcbinfo.ipi_zone, maxsockets);
uma_zone_set_max(V_tcpcb_zone, maxsockets);
tcp_tw_zone_change();
}
static int
tcp_inpcb_init(void *mem, int size, int flags)
{
struct inpcb *inp = mem;
INP_LOCK_INIT(inp, "inp", "tcpinp");
return (0);
}
/*
* Take a value and get the next power of 2 that doesn't overflow.
* Used to size the tcp_inpcb hash buckets.
*/
static int
maketcp_hashsize(int size)
{
int hashsize;
/*
* auto tune.
* get the next power of 2 higher than maxsockets.
*/
hashsize = 1 << fls(size);
/* catch overflow, and just go one power of 2 smaller */
if (hashsize < size) {
hashsize = 1 << (fls(size) - 1);
}
return (hashsize);
}
int
register_tcp_functions(struct tcp_function_block *blk, int wait)
{
struct tcp_function_block *lblk;
struct tcp_function *n;
struct tcp_function_set fs;
if (t_functions_inited == 0) {
init_tcp_functions();
}
if ((blk->tfb_tcp_output == NULL) ||
(blk->tfb_tcp_do_segment == NULL) ||
(blk->tfb_tcp_ctloutput == NULL) ||
(strlen(blk->tfb_tcp_block_name) == 0)) {
/*
* These functions are required and you
* need a name.
*/
return (EINVAL);
}
if (blk->tfb_tcp_timer_stop_all ||
blk->tfb_tcp_timer_activate ||
blk->tfb_tcp_timer_active ||
blk->tfb_tcp_timer_stop) {
/*
* If you define one timer function you
* must have them all.
*/
if ((blk->tfb_tcp_timer_stop_all == NULL) ||
(blk->tfb_tcp_timer_activate == NULL) ||
(blk->tfb_tcp_timer_active == NULL) ||
(blk->tfb_tcp_timer_stop == NULL)) {
return (EINVAL);
}
}
n = malloc(sizeof(struct tcp_function), M_TCPFUNCTIONS, wait);
if (n == NULL) {
return (ENOMEM);
}
n->tf_fb = blk;
strcpy(fs.function_set_name, blk->tfb_tcp_block_name);
rw_wlock(&tcp_function_lock);
lblk = find_tcp_functions_locked(&fs);
if (lblk) {
/* Duplicate name space not allowed */
rw_wunlock(&tcp_function_lock);
free(n, M_TCPFUNCTIONS);
return (EALREADY);
}
refcount_init(&blk->tfb_refcnt, 0);
blk->tfb_flags = 0;
TAILQ_INSERT_TAIL(&t_functions, n, tf_next);
rw_wunlock(&tcp_function_lock);
return(0);
}
int
deregister_tcp_functions(struct tcp_function_block *blk)
{
struct tcp_function_block *lblk;
struct tcp_function *f;
int error=ENOENT;
if (strcmp(blk->tfb_tcp_block_name, "default") == 0) {
/* You can't un-register the default */
return (EPERM);
}
rw_wlock(&tcp_function_lock);
if (blk == tcp_func_set_ptr) {
/* You can't free the current default */
rw_wunlock(&tcp_function_lock);
return (EBUSY);
}
if (blk->tfb_refcnt) {
/* Still tcb attached, mark it. */
blk->tfb_flags |= TCP_FUNC_BEING_REMOVED;
rw_wunlock(&tcp_function_lock);
return (EBUSY);
}
lblk = find_tcp_fb_locked(blk, &f);
if (lblk) {
/* Found */
TAILQ_REMOVE(&t_functions, f, tf_next);
f->tf_fb = NULL;
free(f, M_TCPFUNCTIONS);
error = 0;
}
rw_wunlock(&tcp_function_lock);
return (error);
}
void
tcp_init(void)
{
const char *tcbhash_tuneable;
int hashsize;
tcbhash_tuneable = "net.inet.tcp.tcbhashsize";
if (hhook_head_register(HHOOK_TYPE_TCP, HHOOK_TCP_EST_IN,
&V_tcp_hhh[HHOOK_TCP_EST_IN], HHOOK_NOWAIT|HHOOK_HEADISINVNET) != 0)
printf("%s: WARNING: unable to register helper hook\n", __func__);
if (hhook_head_register(HHOOK_TYPE_TCP, HHOOK_TCP_EST_OUT,
&V_tcp_hhh[HHOOK_TCP_EST_OUT], HHOOK_NOWAIT|HHOOK_HEADISINVNET) != 0)
printf("%s: WARNING: unable to register helper hook\n", __func__);
hashsize = TCBHASHSIZE;
TUNABLE_INT_FETCH(tcbhash_tuneable, &hashsize);
if (hashsize == 0) {
/*
* Auto tune the hash size based on maxsockets.
* A perfect hash would have a 1:1 mapping
* (hashsize = maxsockets) however it's been
* suggested that O(2) average is better.
*/
hashsize = maketcp_hashsize(maxsockets / 4);
/*
* Our historical default is 512,
* do not autotune lower than this.
*/
if (hashsize < 512)
hashsize = 512;
if (bootverbose && IS_DEFAULT_VNET(curvnet))
printf("%s: %s auto tuned to %d\n", __func__,
tcbhash_tuneable, hashsize);
}
/*
* We require a hashsize to be a power of two.
* Previously if it was not a power of two we would just reset it
* back to 512, which could be a nasty surprise if you did not notice
* the error message.
* Instead what we do is clip it to the closest power of two lower
* than the specified hash value.
*/
if (!powerof2(hashsize)) {
int oldhashsize = hashsize;
hashsize = maketcp_hashsize(hashsize);
/* prevent absurdly low value */
if (hashsize < 16)
hashsize = 16;
printf("%s: WARNING: TCB hash size not a power of 2, "
"clipped from %d to %d.\n", __func__, oldhashsize,
hashsize);
}
in_pcbinfo_init(&V_tcbinfo, "tcp", &V_tcb, hashsize, hashsize,
"tcp_inpcb", tcp_inpcb_init, NULL, 0, IPI_HASHFIELDS_4TUPLE);
/*
* These have to be type stable for the benefit of the timers.
*/
V_tcpcb_zone = uma_zcreate("tcpcb", sizeof(struct tcpcb_mem),
NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, 0);
uma_zone_set_max(V_tcpcb_zone, maxsockets);
uma_zone_set_warning(V_tcpcb_zone, "kern.ipc.maxsockets limit reached");
tcp_tw_init();
syncache_init();
tcp_hc_init();
TUNABLE_INT_FETCH("net.inet.tcp.sack.enable", &V_tcp_do_sack);
V_sack_hole_zone = uma_zcreate("sackhole", sizeof(struct sackhole),
NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, 0);
/* Skip initialization of globals for non-default instances. */
if (!IS_DEFAULT_VNET(curvnet))
return;
tcp_reass_global_init();
/* XXX virtualize those bellow? */
tcp_delacktime = TCPTV_DELACK;
tcp_keepinit = TCPTV_KEEP_INIT;
tcp_keepidle = TCPTV_KEEP_IDLE;
tcp_keepintvl = TCPTV_KEEPINTVL;
tcp_maxpersistidle = TCPTV_KEEP_IDLE;
tcp_msl = TCPTV_MSL;
tcp_rexmit_min = TCPTV_MIN;
if (tcp_rexmit_min < 1)
tcp_rexmit_min = 1;
tcp_persmin = TCPTV_PERSMIN;
tcp_persmax = TCPTV_PERSMAX;
tcp_rexmit_slop = TCPTV_CPU_VAR;
tcp_finwait2_timeout = TCPTV_FINWAIT2_TIMEOUT;
tcp_tcbhashsize = hashsize;
/* Setup the tcp function block list */
init_tcp_functions();
register_tcp_functions(&tcp_def_funcblk, M_WAITOK);
if (tcp_soreceive_stream) {
#ifdef INET
tcp_usrreqs.pru_soreceive = soreceive_stream;
#endif
#ifdef INET6
tcp6_usrreqs.pru_soreceive = soreceive_stream;
#endif /* INET6 */
}
#ifdef INET6
#define TCP_MINPROTOHDR (sizeof(struct ip6_hdr) + sizeof(struct tcphdr))
#else /* INET6 */
#define TCP_MINPROTOHDR (sizeof(struct tcpiphdr))
#endif /* INET6 */
if (max_protohdr < TCP_MINPROTOHDR)
max_protohdr = TCP_MINPROTOHDR;
if (max_linkhdr + TCP_MINPROTOHDR > MHLEN)
panic("tcp_init");
#undef TCP_MINPROTOHDR
ISN_LOCK_INIT();
EVENTHANDLER_REGISTER(shutdown_pre_sync, tcp_fini, NULL,
SHUTDOWN_PRI_DEFAULT);
EVENTHANDLER_REGISTER(maxsockets_change, tcp_zone_change, NULL,
EVENTHANDLER_PRI_ANY);
#ifdef TCPPCAP
tcp_pcap_init();
#endif
#ifdef TCP_RFC7413
tcp_fastopen_init();
#endif
}
#ifdef VIMAGE
static void
tcp_destroy(void *unused __unused)
{
int error;
/*
* All our processes are gone, all our sockets should be cleaned
* up, which means, we should be past the tcp_discardcb() calls.
* Sleep to let all tcpcb timers really disappear and then cleanup.
* Timewait will cleanup its queue and will be ready to go.
* XXX-BZ In theory a few ticks should be good enough to make sure
* the timers are all really gone. We should see if we could use a
* better metric here and, e.g., check a tcbcb count as an optimization?
*/
DELAY(1000000 / hz);
tcp_hc_destroy();
syncache_destroy();
tcp_tw_destroy();
in_pcbinfo_destroy(&V_tcbinfo);
/* tcp_discardcb() clears the sack_holes up. */
uma_zdestroy(V_sack_hole_zone);
uma_zdestroy(V_tcpcb_zone);
#ifdef TCP_RFC7413
/*
* Cannot free the zone until all tcpcbs are released as we attach
* the allocations to them.
*/
tcp_fastopen_destroy();
#endif
error = hhook_head_deregister(V_tcp_hhh[HHOOK_TCP_EST_IN]);
if (error != 0) {
printf("%s: WARNING: unable to deregister helper hook "
"type=%d, id=%d: error %d returned\n", __func__,
HHOOK_TYPE_TCP, HHOOK_TCP_EST_IN, error);
}
error = hhook_head_deregister(V_tcp_hhh[HHOOK_TCP_EST_OUT]);
if (error != 0) {
printf("%s: WARNING: unable to deregister helper hook "
"type=%d, id=%d: error %d returned\n", __func__,
HHOOK_TYPE_TCP, HHOOK_TCP_EST_OUT, error);
}
}
VNET_SYSUNINIT(tcp, SI_SUB_PROTO_DOMAIN, SI_ORDER_FOURTH, tcp_destroy, NULL);
#endif
void
tcp_fini(void *xtp)
{
}
/*
* Fill in the IP and TCP headers for an outgoing packet, given the tcpcb.
* tcp_template used to store this data in mbufs, but we now recopy it out
* of the tcpcb each time to conserve mbufs.
*/
void
tcpip_fillheaders(struct inpcb *inp, void *ip_ptr, void *tcp_ptr)
{
struct tcphdr *th = (struct tcphdr *)tcp_ptr;
INP_WLOCK_ASSERT(inp);
#ifdef INET6
if ((inp->inp_vflag & INP_IPV6) != 0) {
struct ip6_hdr *ip6;
ip6 = (struct ip6_hdr *)ip_ptr;
ip6->ip6_flow = (ip6->ip6_flow & ~IPV6_FLOWINFO_MASK) |
(inp->inp_flow & IPV6_FLOWINFO_MASK);
ip6->ip6_vfc = (ip6->ip6_vfc & ~IPV6_VERSION_MASK) |
(IPV6_VERSION & IPV6_VERSION_MASK);
ip6->ip6_nxt = IPPROTO_TCP;
ip6->ip6_plen = htons(sizeof(struct tcphdr));
ip6->ip6_src = inp->in6p_laddr;
ip6->ip6_dst = inp->in6p_faddr;
}
#endif /* INET6 */
#if defined(INET6) && defined(INET)
else
#endif
#ifdef INET
{
struct ip *ip;
ip = (struct ip *)ip_ptr;
ip->ip_v = IPVERSION;
ip->ip_hl = 5;
ip->ip_tos = inp->inp_ip_tos;
ip->ip_len = 0;
ip->ip_id = 0;
ip->ip_off = 0;
ip->ip_ttl = inp->inp_ip_ttl;
ip->ip_sum = 0;
ip->ip_p = IPPROTO_TCP;
ip->ip_src = inp->inp_laddr;
ip->ip_dst = inp->inp_faddr;
}
#endif /* INET */
th->th_sport = inp->inp_lport;
th->th_dport = inp->inp_fport;
th->th_seq = 0;
th->th_ack = 0;
th->th_x2 = 0;
th->th_off = 5;
th->th_flags = 0;
th->th_win = 0;
th->th_urp = 0;
th->th_sum = 0; /* in_pseudo() is called later for ipv4 */
}
/*
* Create template to be used to send tcp packets on a connection.
* Allocates an mbuf and fills in a skeletal tcp/ip header. The only
* use for this function is in keepalives, which use tcp_respond.
*/
struct tcptemp *
tcpip_maketemplate(struct inpcb *inp)
{
struct tcptemp *t;
t = malloc(sizeof(*t), M_TEMP, M_NOWAIT);
if (t == NULL)
return (NULL);
tcpip_fillheaders(inp, (void *)&t->tt_ipgen, (void *)&t->tt_t);
return (t);
}
/*
* Send a single message to the TCP at address specified by
* the given TCP/IP header. If m == NULL, then we make a copy
* of the tcpiphdr at th and send directly to the addressed host.
* This is used to force keep alive messages out using the TCP
* template for a connection. If flags are given then we send
* a message back to the TCP which originated the segment th,
* and discard the mbuf containing it and any other attached mbufs.
*
* In any case the ack and sequence number of the transmitted
* segment are as specified by the parameters.
*
* NOTE: If m != NULL, then th must point to *inside* the mbuf.
*/
void
tcp_respond(struct tcpcb *tp, void *ipgen, struct tcphdr *th, struct mbuf *m,
tcp_seq ack, tcp_seq seq, int flags)
{
struct tcpopt to;
struct inpcb *inp;
struct ip *ip;
struct mbuf *optm;
struct tcphdr *nth;
u_char *optp;
#ifdef INET6
struct ip6_hdr *ip6;
int isipv6;
#endif /* INET6 */
int optlen, tlen, win;
bool incl_opts;
KASSERT(tp != NULL || m != NULL, ("tcp_respond: tp and m both NULL"));
#ifdef INET6
isipv6 = ((struct ip *)ipgen)->ip_v == (IPV6_VERSION >> 4);
ip6 = ipgen;
#endif /* INET6 */
ip = ipgen;
if (tp != NULL) {
inp = tp->t_inpcb;
KASSERT(inp != NULL, ("tcp control block w/o inpcb"));
INP_WLOCK_ASSERT(inp);
} else
inp = NULL;
incl_opts = false;
win = 0;
if (tp != NULL) {
if (!(flags & TH_RST)) {
win = sbspace(&inp->inp_socket->so_rcv);
if (win > (long)TCP_MAXWIN << tp->rcv_scale)
win = (long)TCP_MAXWIN << tp->rcv_scale;
}
if ((tp->t_flags & TF_NOOPT) == 0)
incl_opts = true;
}
if (m == NULL) {
m = m_gethdr(M_NOWAIT, MT_DATA);
if (m == NULL)
return;
m->m_data += max_linkhdr;
#ifdef INET6
if (isipv6) {
bcopy((caddr_t)ip6, mtod(m, caddr_t),
sizeof(struct ip6_hdr));
ip6 = mtod(m, struct ip6_hdr *);
nth = (struct tcphdr *)(ip6 + 1);
} else
#endif /* INET6 */
{
bcopy((caddr_t)ip, mtod(m, caddr_t), sizeof(struct ip));
ip = mtod(m, struct ip *);
nth = (struct tcphdr *)(ip + 1);
}
bcopy((caddr_t)th, (caddr_t)nth, sizeof(struct tcphdr));
flags = TH_ACK;
} else if (!M_WRITABLE(m)) {
struct mbuf *n;
/* Can't reuse 'm', allocate a new mbuf. */
n = m_gethdr(M_NOWAIT, MT_DATA);
if (n == NULL) {
m_freem(m);
return;
}
if (!m_dup_pkthdr(n, m, M_NOWAIT)) {
m_freem(m);
m_freem(n);
return;
}
n->m_data += max_linkhdr;
/* m_len is set later */
#define xchg(a,b,type) { type t; t=a; a=b; b=t; }
#ifdef INET6
if (isipv6) {
bcopy((caddr_t)ip6, mtod(n, caddr_t),
sizeof(struct ip6_hdr));
ip6 = mtod(n, struct ip6_hdr *);
xchg(ip6->ip6_dst, ip6->ip6_src, struct in6_addr);
nth = (struct tcphdr *)(ip6 + 1);
} else
#endif /* INET6 */
{
bcopy((caddr_t)ip, mtod(n, caddr_t), sizeof(struct ip));
ip = mtod(n, struct ip *);
xchg(ip->ip_dst.s_addr, ip->ip_src.s_addr, uint32_t);
nth = (struct tcphdr *)(ip + 1);
}
bcopy((caddr_t)th, (caddr_t)nth, sizeof(struct tcphdr));
xchg(nth->th_dport, nth->th_sport, uint16_t);
th = nth;
m_freem(m);
m = n;
} else {
/*
* reuse the mbuf.
* XXX MRT We inherit the FIB, which is lucky.
*/
m_freem(m->m_next);
m->m_next = NULL;
m->m_data = (caddr_t)ipgen;
/* m_len is set later */
#ifdef INET6
if (isipv6) {
xchg(ip6->ip6_dst, ip6->ip6_src, struct in6_addr);
nth = (struct tcphdr *)(ip6 + 1);
} else
#endif /* INET6 */
{
xchg(ip->ip_dst.s_addr, ip->ip_src.s_addr, uint32_t);
nth = (struct tcphdr *)(ip + 1);
}
if (th != nth) {
/*
* this is usually a case when an extension header
* exists between the IPv6 header and the
* TCP header.
*/
nth->th_sport = th->th_sport;
nth->th_dport = th->th_dport;
}
xchg(nth->th_dport, nth->th_sport, uint16_t);
#undef xchg
}
tlen = 0;
#ifdef INET6
if (isipv6)
tlen = sizeof (struct ip6_hdr) + sizeof (struct tcphdr);
#endif
#if defined(INET) && defined(INET6)
else
#endif
#ifdef INET
tlen = sizeof (struct tcpiphdr);
#endif
#ifdef INVARIANTS
m->m_len = 0;
KASSERT(M_TRAILINGSPACE(m) >= tlen,
("Not enough trailing space for message (m=%p, need=%d, have=%ld)",
m, tlen, (long)M_TRAILINGSPACE(m)));
#endif
m->m_len = tlen;
to.to_flags = 0;
if (incl_opts) {
/* Make sure we have room. */
if (M_TRAILINGSPACE(m) < TCP_MAXOLEN) {
m->m_next = m_get(M_NOWAIT, MT_DATA);
if (m->m_next) {
optp = mtod(m->m_next, u_char *);
optm = m->m_next;
} else
incl_opts = false;
} else {
optp = (u_char *) (nth + 1);
optm = m;
}
}
if (incl_opts) {
/* Timestamps. */
if (tp->t_flags & TF_RCVD_TSTMP) {
to.to_tsval = tcp_ts_getticks() + tp->ts_offset;
to.to_tsecr = tp->ts_recent;
to.to_flags |= TOF_TS;
}
#ifdef TCP_SIGNATURE
/* TCP-MD5 (RFC2385). */
if (tp->t_flags & TF_SIGNATURE)
to.to_flags |= TOF_SIGNATURE;
#endif
/* Add the options. */
tlen += optlen = tcp_addoptions(&to, optp);
/* Update m_len in the correct mbuf. */
optm->m_len += optlen;
} else
optlen = 0;
#ifdef INET6
if (isipv6) {
ip6->ip6_flow = 0;
ip6->ip6_vfc = IPV6_VERSION;
ip6->ip6_nxt = IPPROTO_TCP;
ip6->ip6_plen = htons(tlen - sizeof(*ip6));
}
#endif
#if defined(INET) && defined(INET6)
else
#endif
#ifdef INET
{
ip->ip_len = htons(tlen);
ip->ip_ttl = V_ip_defttl;
if (V_path_mtu_discovery)
ip->ip_off |= htons(IP_DF);
}
#endif
m->m_pkthdr.len = tlen;
m->m_pkthdr.rcvif = NULL;
#ifdef MAC
if (inp != NULL) {
/*
* Packet is associated with a socket, so allow the
* label of the response to reflect the socket label.
*/
INP_WLOCK_ASSERT(inp);
mac_inpcb_create_mbuf(inp, m);
} else {
/*
* Packet is not associated with a socket, so possibly
* update the label in place.
*/
mac_netinet_tcp_reply(m);
}
#endif
nth->th_seq = htonl(seq);
nth->th_ack = htonl(ack);
nth->th_x2 = 0;
nth->th_off = (sizeof (struct tcphdr) + optlen) >> 2;
nth->th_flags = flags;
if (tp != NULL)
nth->th_win = htons((u_short) (win >> tp->rcv_scale));
else
nth->th_win = htons((u_short)win);
nth->th_urp = 0;
#ifdef TCP_SIGNATURE
if (to.to_flags & TOF_SIGNATURE) {
tcp_signature_compute(m, 0, 0, optlen, to.to_signature,
IPSEC_DIR_OUTBOUND);
}
#endif
m->m_pkthdr.csum_data = offsetof(struct tcphdr, th_sum);
#ifdef INET6
if (isipv6) {
m->m_pkthdr.csum_flags = CSUM_TCP_IPV6;
nth->th_sum = in6_cksum_pseudo(ip6,
tlen - sizeof(struct ip6_hdr), IPPROTO_TCP, 0);
ip6->ip6_hlim = in6_selecthlim(tp != NULL ? tp->t_inpcb :
NULL, NULL);
}
#endif /* INET6 */
#if defined(INET6) && defined(INET)
else
#endif
#ifdef INET
{
m->m_pkthdr.csum_flags = CSUM_TCP;
nth->th_sum = in_pseudo(ip->ip_src.s_addr, ip->ip_dst.s_addr,
htons((u_short)(tlen - sizeof(struct ip) + ip->ip_p)));
}
#endif /* INET */
#ifdef TCPDEBUG
if (tp == NULL || (inp->inp_socket->so_options & SO_DEBUG))
tcp_trace(TA_OUTPUT, 0, tp, mtod(m, void *), th, 0);
#endif
TCP_PROBE3(debug__output, tp, th, mtod(m, const char *));
if (flags & TH_RST)
TCP_PROBE5(accept__refused, NULL, NULL, mtod(m, const char *),
tp, nth);
TCP_PROBE5(send, NULL, tp, mtod(m, const char *), tp, nth);
#ifdef INET6
if (isipv6)
(void) ip6_output(m, NULL, NULL, 0, NULL, NULL, inp);
#endif /* INET6 */
#if defined(INET) && defined(INET6)
else
#endif
#ifdef INET
(void) ip_output(m, NULL, NULL, 0, NULL, inp);
#endif
}
/*
* Create a new TCP control block, making an
* empty reassembly queue and hooking it to the argument
* protocol control block. The `inp' parameter must have
* come from the zone allocator set up in tcp_init().
*/
struct tcpcb *
tcp_newtcpcb(struct inpcb *inp)
{
struct tcpcb_mem *tm;
struct tcpcb *tp;
#ifdef INET6
int isipv6 = (inp->inp_vflag & INP_IPV6) != 0;
#endif /* INET6 */
tm = uma_zalloc(V_tcpcb_zone, M_NOWAIT | M_ZERO);
if (tm == NULL)
return (NULL);
tp = &tm->tcb;
/* Initialise cc_var struct for this tcpcb. */
tp->ccv = &tm->ccv;
tp->ccv->type = IPPROTO_TCP;
tp->ccv->ccvc.tcp = tp;
rw_rlock(&tcp_function_lock);
tp->t_fb = tcp_func_set_ptr;
refcount_acquire(&tp->t_fb->tfb_refcnt);
rw_runlock(&tcp_function_lock);
if (tp->t_fb->tfb_tcp_fb_init) {
(*tp->t_fb->tfb_tcp_fb_init)(tp);
}
/*
* Use the current system default CC algorithm.
*/
CC_LIST_RLOCK();
KASSERT(!STAILQ_EMPTY(&cc_list), ("cc_list is empty!"));
CC_ALGO(tp) = CC_DEFAULT();
CC_LIST_RUNLOCK();
if (CC_ALGO(tp)->cb_init != NULL)
if (CC_ALGO(tp)->cb_init(tp->ccv) > 0) {
if (tp->t_fb->tfb_tcp_fb_fini)
(*tp->t_fb->tfb_tcp_fb_fini)(tp);
refcount_release(&tp->t_fb->tfb_refcnt);
uma_zfree(V_tcpcb_zone, tm);
return (NULL);
}
tp->osd = &tm->osd;
if (khelp_init_osd(HELPER_CLASS_TCP, tp->osd)) {
if (tp->t_fb->tfb_tcp_fb_fini)
(*tp->t_fb->tfb_tcp_fb_fini)(tp);
refcount_release(&tp->t_fb->tfb_refcnt);
uma_zfree(V_tcpcb_zone, tm);
return (NULL);
}
#ifdef VIMAGE
tp->t_vnet = inp->inp_vnet;
#endif
tp->t_timers = &tm->tt;
/* LIST_INIT(&tp->t_segq); */ /* XXX covered by M_ZERO */
tp->t_maxseg =
#ifdef INET6
isipv6 ? V_tcp_v6mssdflt :
#endif /* INET6 */
V_tcp_mssdflt;
/* Set up our timeouts. */
callout_init(&tp->t_timers->tt_rexmt, 1);
callout_init(&tp->t_timers->tt_persist, 1);
callout_init(&tp->t_timers->tt_keep, 1);
callout_init(&tp->t_timers->tt_2msl, 1);
callout_init(&tp->t_timers->tt_delack, 1);
if (V_tcp_do_rfc1323)
tp->t_flags = (TF_REQ_SCALE|TF_REQ_TSTMP);
if (V_tcp_do_sack)
tp->t_flags |= TF_SACK_PERMIT;
TAILQ_INIT(&tp->snd_holes);
/*
* The tcpcb will hold a reference on its inpcb until tcp_discardcb()
* is called.
*/
in_pcbref(inp); /* Reference for tcpcb */
tp->t_inpcb = inp;
/*
* Init srtt to TCPTV_SRTTBASE (0), so we can tell that we have no
* rtt estimate. Set rttvar so that srtt + 4 * rttvar gives
* reasonable initial retransmit time.
*/
tp->t_srtt = TCPTV_SRTTBASE;
tp->t_rttvar = ((TCPTV_RTOBASE - TCPTV_SRTTBASE) << TCP_RTTVAR_SHIFT) / 4;
tp->t_rttmin = tcp_rexmit_min;
tp->t_rxtcur = TCPTV_RTOBASE;
tp->snd_cwnd = TCP_MAXWIN << TCP_MAX_WINSHIFT;
tp->snd_ssthresh = TCP_MAXWIN << TCP_MAX_WINSHIFT;
tp->t_rcvtime = ticks;
/*
* IPv4 TTL initialization is necessary for an IPv6 socket as well,
* because the socket may be bound to an IPv6 wildcard address,
* which may match an IPv4-mapped IPv6 address.
*/
inp->inp_ip_ttl = V_ip_defttl;
inp->inp_ppcb = tp;
#ifdef TCPPCAP
/*
* Init the TCP PCAP queues.
*/
tcp_pcap_tcpcb_init(tp);
#endif
return (tp); /* XXX */
}
/*
* Switch the congestion control algorithm back to NewReno for any active
* control blocks using an algorithm which is about to go away.
* This ensures the CC framework can allow the unload to proceed without leaving
* any dangling pointers which would trigger a panic.
* Returning non-zero would inform the CC framework that something went wrong
* and it would be unsafe to allow the unload to proceed. However, there is no
* way for this to occur with this implementation so we always return zero.
*/
int
tcp_ccalgounload(struct cc_algo *unload_algo)
{
struct cc_algo *tmpalgo;
struct inpcb *inp;
struct tcpcb *tp;
VNET_ITERATOR_DECL(vnet_iter);
/*
* Check all active control blocks across all network stacks and change
* any that are using "unload_algo" back to NewReno. If "unload_algo"
* requires cleanup code to be run, call it.
*/
VNET_LIST_RLOCK();
VNET_FOREACH(vnet_iter) {
CURVNET_SET(vnet_iter);
INP_INFO_WLOCK(&V_tcbinfo);
/*
* New connections already part way through being initialised
* with the CC algo we're removing will not race with this code
* because the INP_INFO_WLOCK is held during initialisation. We
* therefore don't enter the loop below until the connection
* list has stabilised.
*/
LIST_FOREACH(inp, &V_tcb, inp_list) {
INP_WLOCK(inp);
/* Important to skip tcptw structs. */
if (!(inp->inp_flags & INP_TIMEWAIT) &&
(tp = intotcpcb(inp)) != NULL) {
/*
* By holding INP_WLOCK here, we are assured
* that the connection is not currently
* executing inside the CC module's functions
* i.e. it is safe to make the switch back to
* NewReno.
*/
if (CC_ALGO(tp) == unload_algo) {
tmpalgo = CC_ALGO(tp);
/* NewReno does not require any init. */
CC_ALGO(tp) = &newreno_cc_algo;
if (tmpalgo->cb_destroy != NULL)
tmpalgo->cb_destroy(tp->ccv);
}
}
INP_WUNLOCK(inp);
}
INP_INFO_WUNLOCK(&V_tcbinfo);
CURVNET_RESTORE();
}
VNET_LIST_RUNLOCK();
return (0);
}
/*
* Drop a TCP connection, reporting
* the specified error. If connection is synchronized,
* then send a RST to peer.
*/
struct tcpcb *
tcp_drop(struct tcpcb *tp, int errno)
{
struct socket *so = tp->t_inpcb->inp_socket;
INP_INFO_LOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(tp->t_inpcb);
if (TCPS_HAVERCVDSYN(tp->t_state)) {
tcp_state_change(tp, TCPS_CLOSED);
(void) tp->t_fb->tfb_tcp_output(tp);
TCPSTAT_INC(tcps_drops);
} else
TCPSTAT_INC(tcps_conndrops);
if (errno == ETIMEDOUT && tp->t_softerror)
errno = tp->t_softerror;
so->so_error = errno;
return (tcp_close(tp));
}
void
tcp_discardcb(struct tcpcb *tp)
{
struct inpcb *inp = tp->t_inpcb;
struct socket *so = inp->inp_socket;
#ifdef INET6
int isipv6 = (inp->inp_vflag & INP_IPV6) != 0;
#endif /* INET6 */
int released;
INP_WLOCK_ASSERT(inp);
/*
* Make sure that all of our timers are stopped before we delete the
* PCB.
*
* If stopping a timer fails, we schedule a discard function in same
* callout, and the last discard function called will take care of
* deleting the tcpcb.
*/
tp->t_timers->tt_draincnt = 0;
tcp_timer_stop(tp, TT_REXMT);
tcp_timer_stop(tp, TT_PERSIST);
tcp_timer_stop(tp, TT_KEEP);
tcp_timer_stop(tp, TT_2MSL);
tcp_timer_stop(tp, TT_DELACK);
if (tp->t_fb->tfb_tcp_timer_stop_all) {
/*
* Call the stop-all function of the methods,
* this function should call the tcp_timer_stop()
* method with each of the function specific timeouts.
* That stop will be called via the tfb_tcp_timer_stop()
* which should use the async drain function of the
* callout system (see tcp_var.h).
*/
tp->t_fb->tfb_tcp_timer_stop_all(tp);
}
/*
* If we got enough samples through the srtt filter,
* save the rtt and rttvar in the routing entry.
* 'Enough' is arbitrarily defined as 4 rtt samples.
* 4 samples is enough for the srtt filter to converge
* to within enough % of the correct value; fewer samples
* and we could save a bogus rtt. The danger is not high
* as tcp quickly recovers from everything.
* XXX: Works very well but needs some more statistics!
*/
if (tp->t_rttupdated >= 4) {
struct hc_metrics_lite metrics;
u_long ssthresh;
bzero(&metrics, sizeof(metrics));
/*
* Update the ssthresh always when the conditions below
* are satisfied. This gives us better new start value
* for the congestion avoidance for new connections.
* ssthresh is only set if packet loss occurred on a session.
*
* XXXRW: 'so' may be NULL here, and/or socket buffer may be
* being torn down. Ideally this code would not use 'so'.
*/
ssthresh = tp->snd_ssthresh;
if (ssthresh != 0 && ssthresh < so->so_snd.sb_hiwat / 2) {
/*
* convert the limit from user data bytes to
* packets then to packet data bytes.
*/
ssthresh = (ssthresh + tp->t_maxseg / 2) / tp->t_maxseg;
if (ssthresh < 2)
ssthresh = 2;
ssthresh *= (u_long)(tp->t_maxseg +
#ifdef INET6
(isipv6 ? sizeof (struct ip6_hdr) +
sizeof (struct tcphdr) :
#endif
sizeof (struct tcpiphdr)
#ifdef INET6
)
#endif
);
} else
ssthresh = 0;
metrics.rmx_ssthresh = ssthresh;
metrics.rmx_rtt = tp->t_srtt;
metrics.rmx_rttvar = tp->t_rttvar;
metrics.rmx_cwnd = tp->snd_cwnd;
metrics.rmx_sendpipe = 0;
metrics.rmx_recvpipe = 0;
tcp_hc_update(&inp->inp_inc, &metrics);
}
/* free the reassembly queue, if any */
tcp_reass_flush(tp);
#ifdef TCP_OFFLOAD
/* Disconnect offload device, if any. */
if (tp->t_flags & TF_TOE)
tcp_offload_detach(tp);
#endif
tcp_free_sackholes(tp);
#ifdef TCPPCAP
/* Free the TCP PCAP queues. */
tcp_pcap_drain(&(tp->t_inpkts));
tcp_pcap_drain(&(tp->t_outpkts));
#endif
/* Allow the CC algorithm to clean up after itself. */
if (CC_ALGO(tp)->cb_destroy != NULL)
CC_ALGO(tp)->cb_destroy(tp->ccv);
khelp_destroy_osd(tp->osd);
CC_ALGO(tp) = NULL;
inp->inp_ppcb = NULL;
if (tp->t_timers->tt_draincnt == 0) {
/* We own the last reference on tcpcb, let's free it. */
if (tp->t_fb->tfb_tcp_fb_fini)
(*tp->t_fb->tfb_tcp_fb_fini)(tp);
refcount_release(&tp->t_fb->tfb_refcnt);
tp->t_inpcb = NULL;
uma_zfree(V_tcpcb_zone, tp);
released = in_pcbrele_wlocked(inp);
KASSERT(!released, ("%s: inp %p should not have been released "
"here", __func__, inp));
}
}
void
tcp_timer_discard(void *ptp)
{
struct inpcb *inp;
struct tcpcb *tp;
tp = (struct tcpcb *)ptp;
CURVNET_SET(tp->t_vnet);
INP_INFO_RLOCK(&V_tcbinfo);
inp = tp->t_inpcb;
KASSERT(inp != NULL, ("%s: tp %p tp->t_inpcb == NULL",
__func__, tp));
INP_WLOCK(inp);
KASSERT((tp->t_timers->tt_flags & TT_STOPPED) != 0,
("%s: tcpcb has to be stopped here", __func__));
tp->t_timers->tt_draincnt--;
if (tp->t_timers->tt_draincnt == 0) {
/* We own the last reference on this tcpcb, let's free it. */
if (tp->t_fb->tfb_tcp_fb_fini)
(*tp->t_fb->tfb_tcp_fb_fini)(tp);
refcount_release(&tp->t_fb->tfb_refcnt);
tp->t_inpcb = NULL;
uma_zfree(V_tcpcb_zone, tp);
if (in_pcbrele_wlocked(inp)) {
INP_INFO_RUNLOCK(&V_tcbinfo);
CURVNET_RESTORE();
return;
}
}
INP_WUNLOCK(inp);
INP_INFO_RUNLOCK(&V_tcbinfo);
CURVNET_RESTORE();
}
/*
* Attempt to close a TCP control block, marking it as dropped, and freeing
* the socket if we hold the only reference.
*/
struct tcpcb *
tcp_close(struct tcpcb *tp)
{
struct inpcb *inp = tp->t_inpcb;
struct socket *so;
INP_INFO_LOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(inp);
#ifdef TCP_OFFLOAD
if (tp->t_state == TCPS_LISTEN)
tcp_offload_listen_stop(tp);
#endif
#ifdef TCP_RFC7413
/*
* This releases the TFO pending counter resource for TFO listen
* sockets as well as passively-created TFO sockets that transition
* from SYN_RECEIVED to CLOSED.
*/
if (tp->t_tfo_pending) {
tcp_fastopen_decrement_counter(tp->t_tfo_pending);
tp->t_tfo_pending = NULL;
}
#endif
in_pcbdrop(inp);
TCPSTAT_INC(tcps_closed);
TCPSTATES_DEC(tp->t_state);
KASSERT(inp->inp_socket != NULL, ("tcp_close: inp_socket NULL"));
so = inp->inp_socket;
soisdisconnected(so);
if (inp->inp_flags & INP_SOCKREF) {
KASSERT(so->so_state & SS_PROTOREF,
("tcp_close: !SS_PROTOREF"));
inp->inp_flags &= ~INP_SOCKREF;
INP_WUNLOCK(inp);
ACCEPT_LOCK();
SOCK_LOCK(so);
so->so_state &= ~SS_PROTOREF;
sofree(so);
return (NULL);
}
return (tp);
}
void
tcp_drain(void)
{
VNET_ITERATOR_DECL(vnet_iter);
if (!do_tcpdrain)
return;
VNET_LIST_RLOCK_NOSLEEP();
VNET_FOREACH(vnet_iter) {
CURVNET_SET(vnet_iter);
struct inpcb *inpb;
struct tcpcb *tcpb;
/*
* Walk the tcpbs, if existing, and flush the reassembly queue,
* if there is one...
* XXX: The "Net/3" implementation doesn't imply that the TCP
* reassembly queue should be flushed, but in a situation
* where we're really low on mbufs, this is potentially
* useful.
*/
INP_INFO_WLOCK(&V_tcbinfo);
LIST_FOREACH(inpb, V_tcbinfo.ipi_listhead, inp_list) {
if (inpb->inp_flags & INP_TIMEWAIT)
continue;
INP_WLOCK(inpb);
if ((tcpb = intotcpcb(inpb)) != NULL) {
tcp_reass_flush(tcpb);
tcp_clean_sackreport(tcpb);
}
INP_WUNLOCK(inpb);
}
INP_INFO_WUNLOCK(&V_tcbinfo);
CURVNET_RESTORE();
}
VNET_LIST_RUNLOCK_NOSLEEP();
}
/*
* Notify a tcp user of an asynchronous error;
* store error as soft error, but wake up user
* (for now, won't do anything until can select for soft error).
*
* Do not wake up user since there currently is no mechanism for
* reporting soft errors (yet - a kqueue filter may be added).
*/
static struct inpcb *
tcp_notify(struct inpcb *inp, int error)
{
struct tcpcb *tp;
INP_INFO_LOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(inp);
if ((inp->inp_flags & INP_TIMEWAIT) ||
(inp->inp_flags & INP_DROPPED))
return (inp);
tp = intotcpcb(inp);
KASSERT(tp != NULL, ("tcp_notify: tp == NULL"));
/*
* Ignore some errors if we are hooked up.
* If connection hasn't completed, has retransmitted several times,
* and receives a second error, give up now. This is better
* than waiting a long time to establish a connection that
* can never complete.
*/
if (tp->t_state == TCPS_ESTABLISHED &&
(error == EHOSTUNREACH || error == ENETUNREACH ||
error == EHOSTDOWN)) {
if (inp->inp_route.ro_rt) {
RTFREE(inp->inp_route.ro_rt);
inp->inp_route.ro_rt = (struct rtentry *)NULL;
}
return (inp);
} else if (tp->t_state < TCPS_ESTABLISHED && tp->t_rxtshift > 3 &&
tp->t_softerror) {
tp = tcp_drop(tp, error);
if (tp != NULL)
return (inp);
else
return (NULL);
} else {
tp->t_softerror = error;
return (inp);
}
#if 0
wakeup( &so->so_timeo);
sorwakeup(so);
sowwakeup(so);
#endif
}
static int
tcp_pcblist(SYSCTL_HANDLER_ARGS)
{
int error, i, m, n, pcb_count;
struct inpcb *inp, **inp_list;
inp_gen_t gencnt;
struct xinpgen xig;
/*
* The process of preparing the TCB list is too time-consuming and
* resource-intensive to repeat twice on every request.
*/
if (req->oldptr == NULL) {
n = V_tcbinfo.ipi_count +
counter_u64_fetch(V_tcps_states[TCPS_SYN_RECEIVED]);
n += imax(n / 8, 10);
req->oldidx = 2 * (sizeof xig) + n * sizeof(struct xtcpcb);
return (0);
}
if (req->newptr != NULL)
return (EPERM);
/*
* OK, now we're committed to doing something.
*/
INP_LIST_RLOCK(&V_tcbinfo);
gencnt = V_tcbinfo.ipi_gencnt;
n = V_tcbinfo.ipi_count;
INP_LIST_RUNLOCK(&V_tcbinfo);
m = counter_u64_fetch(V_tcps_states[TCPS_SYN_RECEIVED]);
error = sysctl_wire_old_buffer(req, 2 * (sizeof xig)
+ (n + m) * sizeof(struct xtcpcb));
if (error != 0)
return (error);
xig.xig_len = sizeof xig;
xig.xig_count = n + m;
xig.xig_gen = gencnt;
xig.xig_sogen = so_gencnt;
error = SYSCTL_OUT(req, &xig, sizeof xig);
if (error)
return (error);
error = syncache_pcblist(req, m, &pcb_count);
if (error)
return (error);
inp_list = malloc(n * sizeof *inp_list, M_TEMP, M_WAITOK);
INP_INFO_WLOCK(&V_tcbinfo);
for (inp = LIST_FIRST(V_tcbinfo.ipi_listhead), i = 0;
inp != NULL && i < n; inp = LIST_NEXT(inp, inp_list)) {
INP_WLOCK(inp);
if (inp->inp_gencnt <= gencnt) {
/*
* XXX: This use of cr_cansee(), introduced with
* TCP state changes, is not quite right, but for
* now, better than nothing.
*/
if (inp->inp_flags & INP_TIMEWAIT) {
if (intotw(inp) != NULL)
error = cr_cansee(req->td->td_ucred,
intotw(inp)->tw_cred);
else
error = EINVAL; /* Skip this inp. */
} else
error = cr_canseeinpcb(req->td->td_ucred, inp);
if (error == 0) {
in_pcbref(inp);
inp_list[i++] = inp;
}
}
INP_WUNLOCK(inp);
}
INP_INFO_WUNLOCK(&V_tcbinfo);
n = i;
error = 0;
for (i = 0; i < n; i++) {
inp = inp_list[i];
INP_RLOCK(inp);
if (inp->inp_gencnt <= gencnt) {
struct xtcpcb xt;
void *inp_ppcb;
bzero(&xt, sizeof(xt));
xt.xt_len = sizeof xt;
/* XXX should avoid extra copy */
bcopy(inp, &xt.xt_inp, sizeof *inp);
inp_ppcb = inp->inp_ppcb;
if (inp_ppcb == NULL)
bzero((char *) &xt.xt_tp, sizeof xt.xt_tp);
else if (inp->inp_flags & INP_TIMEWAIT) {
bzero((char *) &xt.xt_tp, sizeof xt.xt_tp);
xt.xt_tp.t_state = TCPS_TIME_WAIT;
} else {
bcopy(inp_ppcb, &xt.xt_tp, sizeof xt.xt_tp);
if (xt.xt_tp.t_timers)
tcp_timer_to_xtimer(&xt.xt_tp, xt.xt_tp.t_timers, &xt.xt_timer);
}
if (inp->inp_socket != NULL)
sotoxsocket(inp->inp_socket, &xt.xt_socket);
else {
bzero(&xt.xt_socket, sizeof xt.xt_socket);
xt.xt_socket.xso_protocol = IPPROTO_TCP;
}
xt.xt_inp.inp_gencnt = inp->inp_gencnt;
INP_RUNLOCK(inp);
error = SYSCTL_OUT(req, &xt, sizeof xt);
} else
INP_RUNLOCK(inp);
}
INP_INFO_RLOCK(&V_tcbinfo);
for (i = 0; i < n; i++) {
inp = inp_list[i];
INP_RLOCK(inp);
if (!in_pcbrele_rlocked(inp))
INP_RUNLOCK(inp);
}
INP_INFO_RUNLOCK(&V_tcbinfo);
if (!error) {
/*
* Give the user an updated idea of our state.
* If the generation differs from what we told
* her before, she knows that something happened
* while we were processing this request, and it
* might be necessary to retry.
*/
INP_LIST_RLOCK(&V_tcbinfo);
xig.xig_gen = V_tcbinfo.ipi_gencnt;
xig.xig_sogen = so_gencnt;
xig.xig_count = V_tcbinfo.ipi_count + pcb_count;
INP_LIST_RUNLOCK(&V_tcbinfo);
error = SYSCTL_OUT(req, &xig, sizeof xig);
}
free(inp_list, M_TEMP);
return (error);
}
SYSCTL_PROC(_net_inet_tcp, TCPCTL_PCBLIST, pcblist,
CTLTYPE_OPAQUE | CTLFLAG_RD, NULL, 0,
tcp_pcblist, "S,xtcpcb", "List of active TCP connections");
#ifdef INET
static int
tcp_getcred(SYSCTL_HANDLER_ARGS)
{
struct xucred xuc;
struct sockaddr_in addrs[2];
struct inpcb *inp;
int error;
error = priv_check(req->td, PRIV_NETINET_GETCRED);
if (error)
return (error);
error = SYSCTL_IN(req, addrs, sizeof(addrs));
if (error)
return (error);
inp = in_pcblookup(&V_tcbinfo, addrs[1].sin_addr, addrs[1].sin_port,
addrs[0].sin_addr, addrs[0].sin_port, INPLOOKUP_RLOCKPCB, NULL);
if (inp != NULL) {
if (inp->inp_socket == NULL)
error = ENOENT;
if (error == 0)
error = cr_canseeinpcb(req->td->td_ucred, inp);
if (error == 0)
cru2x(inp->inp_cred, &xuc);
INP_RUNLOCK(inp);
} else
error = ENOENT;
if (error == 0)
error = SYSCTL_OUT(req, &xuc, sizeof(struct xucred));
return (error);
}
SYSCTL_PROC(_net_inet_tcp, OID_AUTO, getcred,
CTLTYPE_OPAQUE|CTLFLAG_RW|CTLFLAG_PRISON, 0, 0,
tcp_getcred, "S,xucred", "Get the xucred of a TCP connection");
#endif /* INET */
#ifdef INET6
static int
tcp6_getcred(SYSCTL_HANDLER_ARGS)
{
struct xucred xuc;
struct sockaddr_in6 addrs[2];
struct inpcb *inp;
int error;
#ifdef INET
int mapped = 0;
#endif
error = priv_check(req->td, PRIV_NETINET_GETCRED);
if (error)
return (error);
error = SYSCTL_IN(req, addrs, sizeof(addrs));
if (error)
return (error);
if ((error = sa6_embedscope(&addrs[0], V_ip6_use_defzone)) != 0 ||
(error = sa6_embedscope(&addrs[1], V_ip6_use_defzone)) != 0) {
return (error);
}
if (IN6_IS_ADDR_V4MAPPED(&addrs[0].sin6_addr)) {
#ifdef INET
if (IN6_IS_ADDR_V4MAPPED(&addrs[1].sin6_addr))
mapped = 1;
else
#endif
return (EINVAL);
}
#ifdef INET
if (mapped == 1)
inp = in_pcblookup(&V_tcbinfo,
*(struct in_addr *)&addrs[1].sin6_addr.s6_addr[12],
addrs[1].sin6_port,
*(struct in_addr *)&addrs[0].sin6_addr.s6_addr[12],
addrs[0].sin6_port, INPLOOKUP_RLOCKPCB, NULL);
else
#endif
inp = in6_pcblookup(&V_tcbinfo,
&addrs[1].sin6_addr, addrs[1].sin6_port,
&addrs[0].sin6_addr, addrs[0].sin6_port,
INPLOOKUP_RLOCKPCB, NULL);
if (inp != NULL) {
if (inp->inp_socket == NULL)
error = ENOENT;
if (error == 0)
error = cr_canseeinpcb(req->td->td_ucred, inp);
if (error == 0)
cru2x(inp->inp_cred, &xuc);
INP_RUNLOCK(inp);
} else
error = ENOENT;
if (error == 0)
error = SYSCTL_OUT(req, &xuc, sizeof(struct xucred));
return (error);
}
SYSCTL_PROC(_net_inet6_tcp6, OID_AUTO, getcred,
CTLTYPE_OPAQUE|CTLFLAG_RW|CTLFLAG_PRISON, 0, 0,
tcp6_getcred, "S,xucred", "Get the xucred of a TCP6 connection");
#endif /* INET6 */
#ifdef INET
void
tcp_ctlinput(int cmd, struct sockaddr *sa, void *vip)
{
struct ip *ip = vip;
struct tcphdr *th;
struct in_addr faddr;
struct inpcb *inp;
struct tcpcb *tp;
struct inpcb *(*notify)(struct inpcb *, int) = tcp_notify;
struct icmp *icp;
struct in_conninfo inc;
tcp_seq icmp_tcp_seq;
int mtu;
faddr = ((struct sockaddr_in *)sa)->sin_addr;
if (sa->sa_family != AF_INET || faddr.s_addr == INADDR_ANY)
return;
if (cmd == PRC_MSGSIZE)
notify = tcp_mtudisc_notify;
else if (V_icmp_may_rst && (cmd == PRC_UNREACH_ADMIN_PROHIB ||
cmd == PRC_UNREACH_PORT || cmd == PRC_TIMXCEED_INTRANS) && ip)
notify = tcp_drop_syn_sent;
else if (PRC_IS_REDIRECT(cmd)) {
/* signal EHOSTDOWN, as it flushes the cached route */
in_pcbnotifyall(&V_tcbinfo, faddr, EHOSTDOWN, notify);
return;
}
/*
* Hostdead is ugly because it goes linearly through all PCBs.
* XXX: We never get this from ICMP, otherwise it makes an
* excellent DoS attack on machines with many connections.
*/
else if (cmd == PRC_HOSTDEAD)
ip = NULL;
else if ((unsigned)cmd >= PRC_NCMDS || inetctlerrmap[cmd] == 0)
return;
if (ip == NULL) {
in_pcbnotifyall(&V_tcbinfo, faddr, inetctlerrmap[cmd], notify);
return;
}
icp = (struct icmp *)((caddr_t)ip - offsetof(struct icmp, icmp_ip));
th = (struct tcphdr *)((caddr_t)ip + (ip->ip_hl << 2));
INP_INFO_RLOCK(&V_tcbinfo);
inp = in_pcblookup(&V_tcbinfo, faddr, th->th_dport, ip->ip_src,
th->th_sport, INPLOOKUP_WLOCKPCB, NULL);
if (inp != NULL) {
if (!(inp->inp_flags & INP_TIMEWAIT) &&
!(inp->inp_flags & INP_DROPPED) &&
!(inp->inp_socket == NULL)) {
icmp_tcp_seq = ntohl(th->th_seq);
tp = intotcpcb(inp);
if (SEQ_GEQ(icmp_tcp_seq, tp->snd_una) &&
SEQ_LT(icmp_tcp_seq, tp->snd_max)) {
if (cmd == PRC_MSGSIZE) {
/*
* MTU discovery:
* If we got a needfrag set the MTU
* in the route to the suggested new
* value (if given) and then notify.
*/
mtu = ntohs(icp->icmp_nextmtu);
/*
* If no alternative MTU was
* proposed, try the next smaller
* one.
*/
if (!mtu)
mtu = ip_next_mtu(
ntohs(ip->ip_len), 1);
if (mtu < V_tcp_minmss +
sizeof(struct tcpiphdr))
mtu = V_tcp_minmss +
sizeof(struct tcpiphdr);
/*
* Only process the offered MTU if it
* is smaller than the current one.
*/
if (mtu < tp->t_maxseg +
sizeof(struct tcpiphdr)) {
bzero(&inc, sizeof(inc));
inc.inc_faddr = faddr;
inc.inc_fibnum =
inp->inp_inc.inc_fibnum;
tcp_hc_updatemtu(&inc, mtu);
tcp_mtudisc(inp, mtu);
}
} else
inp = (*notify)(inp,
inetctlerrmap[cmd]);
}
}
if (inp != NULL)
INP_WUNLOCK(inp);
} else {
bzero(&inc, sizeof(inc));
inc.inc_fport = th->th_dport;
inc.inc_lport = th->th_sport;
inc.inc_faddr = faddr;
inc.inc_laddr = ip->ip_src;
syncache_unreach(&inc, th);
}
INP_INFO_RUNLOCK(&V_tcbinfo);
}
#endif /* INET */
#ifdef INET6
void
tcp6_ctlinput(int cmd, struct sockaddr *sa, void *d)
{
struct tcphdr th;
struct inpcb *(*notify)(struct inpcb *, int) = tcp_notify;
struct ip6_hdr *ip6;
struct mbuf *m;
struct ip6ctlparam *ip6cp = NULL;
const struct sockaddr_in6 *sa6_src = NULL;
int off;
struct tcp_portonly {
u_int16_t th_sport;
u_int16_t th_dport;
} *thp;
if (sa->sa_family != AF_INET6 ||
sa->sa_len != sizeof(struct sockaddr_in6))
return;
if (cmd == PRC_MSGSIZE)
notify = tcp_mtudisc_notify;
else if (!PRC_IS_REDIRECT(cmd) &&
((unsigned)cmd >= PRC_NCMDS || inet6ctlerrmap[cmd] == 0))
return;
/* if the parameter is from icmp6, decode it. */
if (d != NULL) {
ip6cp = (struct ip6ctlparam *)d;
m = ip6cp->ip6c_m;
ip6 = ip6cp->ip6c_ip6;
off = ip6cp->ip6c_off;
sa6_src = ip6cp->ip6c_src;
} else {
m = NULL;
ip6 = NULL;
off = 0; /* fool gcc */
sa6_src = &sa6_any;
}
if (ip6 != NULL) {
struct in_conninfo inc;
/*
* XXX: We assume that when IPV6 is non NULL,
* M and OFF are valid.
*/
/* check if we can safely examine src and dst ports */
if (m->m_pkthdr.len < off + sizeof(*thp))
return;
bzero(&th, sizeof(th));
m_copydata(m, off, sizeof(*thp), (caddr_t)&th);
in6_pcbnotify(&V_tcbinfo, sa, th.th_dport,
(struct sockaddr *)ip6cp->ip6c_src,
th.th_sport, cmd, NULL, notify);
bzero(&inc, sizeof(inc));
inc.inc_fport = th.th_dport;
inc.inc_lport = th.th_sport;
inc.inc6_faddr = ((struct sockaddr_in6 *)sa)->sin6_addr;
inc.inc6_laddr = ip6cp->ip6c_src->sin6_addr;
inc.inc_flags |= INC_ISIPV6;
INP_INFO_RLOCK(&V_tcbinfo);
syncache_unreach(&inc, &th);
INP_INFO_RUNLOCK(&V_tcbinfo);
} else
in6_pcbnotify(&V_tcbinfo, sa, 0, (const struct sockaddr *)sa6_src,
0, cmd, NULL, notify);
}
#endif /* INET6 */
/*
* Following is where TCP initial sequence number generation occurs.
*
* There are two places where we must use initial sequence numbers:
* 1. In SYN-ACK packets.
* 2. In SYN packets.
*
* All ISNs for SYN-ACK packets are generated by the syncache. See
* tcp_syncache.c for details.
*
* The ISNs in SYN packets must be monotonic; TIME_WAIT recycling
* depends on this property. In addition, these ISNs should be
* unguessable so as to prevent connection hijacking. To satisfy
* the requirements of this situation, the algorithm outlined in
* RFC 1948 is used, with only small modifications.
*
* Implementation details:
*
* Time is based off the system timer, and is corrected so that it
* increases by one megabyte per second. This allows for proper
* recycling on high speed LANs while still leaving over an hour
* before rollover.
*
* As reading the *exact* system time is too expensive to be done
* whenever setting up a TCP connection, we increment the time
* offset in two ways. First, a small random positive increment
* is added to isn_offset for each connection that is set up.
* Second, the function tcp_isn_tick fires once per clock tick
* and increments isn_offset as necessary so that sequence numbers
* are incremented at approximately ISN_BYTES_PER_SECOND. The
* random positive increments serve only to ensure that the same
* exact sequence number is never sent out twice (as could otherwise
* happen when a port is recycled in less than the system tick
* interval.)
*
* net.inet.tcp.isn_reseed_interval controls the number of seconds
* between seeding of isn_secret. This is normally set to zero,
* as reseeding should not be necessary.
*
* Locking of the global variables isn_secret, isn_last_reseed, isn_offset,
* isn_offset_old, and isn_ctx is performed using the TCP pcbinfo lock. In
* general, this means holding an exclusive (write) lock.
*/
#define ISN_BYTES_PER_SECOND 1048576
#define ISN_STATIC_INCREMENT 4096
#define ISN_RANDOM_INCREMENT (4096 - 1)
static VNET_DEFINE(u_char, isn_secret[32]);
static VNET_DEFINE(int, isn_last);
static VNET_DEFINE(int, isn_last_reseed);
static VNET_DEFINE(u_int32_t, isn_offset);
static VNET_DEFINE(u_int32_t, isn_offset_old);
#define V_isn_secret VNET(isn_secret)
#define V_isn_last VNET(isn_last)
#define V_isn_last_reseed VNET(isn_last_reseed)
#define V_isn_offset VNET(isn_offset)
#define V_isn_offset_old VNET(isn_offset_old)
tcp_seq
tcp_new_isn(struct tcpcb *tp)
{
MD5_CTX isn_ctx;
u_int32_t md5_buffer[4];
tcp_seq new_isn;
u_int32_t projected_offset;
INP_WLOCK_ASSERT(tp->t_inpcb);
ISN_LOCK();
/* Seed if this is the first use, reseed if requested. */
if ((V_isn_last_reseed == 0) || ((V_tcp_isn_reseed_interval > 0) &&
(((u_int)V_isn_last_reseed + (u_int)V_tcp_isn_reseed_interval*hz)
< (u_int)ticks))) {
read_random(&V_isn_secret, sizeof(V_isn_secret));
V_isn_last_reseed = ticks;
}
/* Compute the md5 hash and return the ISN. */
MD5Init(&isn_ctx);
MD5Update(&isn_ctx, (u_char *) &tp->t_inpcb->inp_fport, sizeof(u_short));
MD5Update(&isn_ctx, (u_char *) &tp->t_inpcb->inp_lport, sizeof(u_short));
#ifdef INET6
if ((tp->t_inpcb->inp_vflag & INP_IPV6) != 0) {
MD5Update(&isn_ctx, (u_char *) &tp->t_inpcb->in6p_faddr,
sizeof(struct in6_addr));
MD5Update(&isn_ctx, (u_char *) &tp->t_inpcb->in6p_laddr,
sizeof(struct in6_addr));
} else
#endif
{
MD5Update(&isn_ctx, (u_char *) &tp->t_inpcb->inp_faddr,
sizeof(struct in_addr));
MD5Update(&isn_ctx, (u_char *) &tp->t_inpcb->inp_laddr,
sizeof(struct in_addr));
}
MD5Update(&isn_ctx, (u_char *) &V_isn_secret, sizeof(V_isn_secret));
MD5Final((u_char *) &md5_buffer, &isn_ctx);
new_isn = (tcp_seq) md5_buffer[0];
V_isn_offset += ISN_STATIC_INCREMENT +
(arc4random() & ISN_RANDOM_INCREMENT);
if (ticks != V_isn_last) {
projected_offset = V_isn_offset_old +
ISN_BYTES_PER_SECOND / hz * (ticks - V_isn_last);
if (SEQ_GT(projected_offset, V_isn_offset))
V_isn_offset = projected_offset;
V_isn_offset_old = V_isn_offset;
V_isn_last = ticks;
}
new_isn += V_isn_offset;
ISN_UNLOCK();
return (new_isn);
}
/*
* When a specific ICMP unreachable message is received and the
* connection state is SYN-SENT, drop the connection. This behavior
* is controlled by the icmp_may_rst sysctl.
*/
struct inpcb *
tcp_drop_syn_sent(struct inpcb *inp, int errno)
{
struct tcpcb *tp;
INP_INFO_RLOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(inp);
if ((inp->inp_flags & INP_TIMEWAIT) ||
(inp->inp_flags & INP_DROPPED))
return (inp);
tp = intotcpcb(inp);
if (tp->t_state != TCPS_SYN_SENT)
return (inp);
tp = tcp_drop(tp, errno);
if (tp != NULL)
return (inp);
else
return (NULL);
}
/*
* When `need fragmentation' ICMP is received, update our idea of the MSS
* based on the new value. Also nudge TCP to send something, since we
* know the packet we just sent was dropped.
* This duplicates some code in the tcp_mss() function in tcp_input.c.
*/
static struct inpcb *
tcp_mtudisc_notify(struct inpcb *inp, int error)
{
tcp_mtudisc(inp, -1);
return (inp);
}
static void
tcp_mtudisc(struct inpcb *inp, int mtuoffer)
{
struct tcpcb *tp;
struct socket *so;
INP_WLOCK_ASSERT(inp);
if ((inp->inp_flags & INP_TIMEWAIT) ||
(inp->inp_flags & INP_DROPPED))
return;
tp = intotcpcb(inp);
KASSERT(tp != NULL, ("tcp_mtudisc: tp == NULL"));
tcp_mss_update(tp, -1, mtuoffer, NULL, NULL);
so = inp->inp_socket;
SOCKBUF_LOCK(&so->so_snd);
/* If the mss is larger than the socket buffer, decrease the mss. */
if (so->so_snd.sb_hiwat < tp->t_maxseg)
tp->t_maxseg = so->so_snd.sb_hiwat;
SOCKBUF_UNLOCK(&so->so_snd);
TCPSTAT_INC(tcps_mturesent);
tp->t_rtttime = 0;
tp->snd_nxt = tp->snd_una;
tcp_free_sackholes(tp);
tp->snd_recover = tp->snd_max;
if (tp->t_flags & TF_SACK_PERMIT)
EXIT_FASTRECOVERY(tp->t_flags);
tp->t_fb->tfb_tcp_output(tp);
}
#ifdef INET
/*
* Look-up the routing entry to the peer of this inpcb. If no route
* is found and it cannot be allocated, then return 0. This routine
* is called by TCP routines that access the rmx structure and by
* tcp_mss_update to get the peer/interface MTU.
*/
u_long
tcp_maxmtu(struct in_conninfo *inc, struct tcp_ifcap *cap)
{
struct nhop4_extended nh4;
struct ifnet *ifp;
u_long maxmtu = 0;
KASSERT(inc != NULL, ("tcp_maxmtu with NULL in_conninfo pointer"));
if (inc->inc_faddr.s_addr != INADDR_ANY) {
if (fib4_lookup_nh_ext(inc->inc_fibnum, inc->inc_faddr,
NHR_REF, 0, &nh4) != 0)
return (0);
ifp = nh4.nh_ifp;
maxmtu = nh4.nh_mtu;
/* Report additional interface capabilities. */
if (cap != NULL) {
if (ifp->if_capenable & IFCAP_TSO4 &&
ifp->if_hwassist & CSUM_TSO) {
cap->ifcap |= CSUM_TSO;
cap->tsomax = ifp->if_hw_tsomax;
cap->tsomaxsegcount = ifp->if_hw_tsomaxsegcount;
cap->tsomaxsegsize = ifp->if_hw_tsomaxsegsize;
}
}
fib4_free_nh_ext(inc->inc_fibnum, &nh4);
}
return (maxmtu);
}
#endif /* INET */
#ifdef INET6
u_long
tcp_maxmtu6(struct in_conninfo *inc, struct tcp_ifcap *cap)
{
struct nhop6_extended nh6;
struct in6_addr dst6;
uint32_t scopeid;
struct ifnet *ifp;
u_long maxmtu = 0;
KASSERT(inc != NULL, ("tcp_maxmtu6 with NULL in_conninfo pointer"));
if (!IN6_IS_ADDR_UNSPECIFIED(&inc->inc6_faddr)) {
in6_splitscope(&inc->inc6_faddr, &dst6, &scopeid);
if (fib6_lookup_nh_ext(inc->inc_fibnum, &dst6, scopeid, 0,
0, &nh6) != 0)
return (0);
ifp = nh6.nh_ifp;
maxmtu = nh6.nh_mtu;
/* Report additional interface capabilities. */
if (cap != NULL) {
if (ifp->if_capenable & IFCAP_TSO6 &&
ifp->if_hwassist & CSUM_TSO) {
cap->ifcap |= CSUM_TSO;
cap->tsomax = ifp->if_hw_tsomax;
cap->tsomaxsegcount = ifp->if_hw_tsomaxsegcount;
cap->tsomaxsegsize = ifp->if_hw_tsomaxsegsize;
}
}
fib6_free_nh_ext(inc->inc_fibnum, &nh6);
}
return (maxmtu);
}
#endif /* INET6 */
/*
* Calculate effective SMSS per RFC5681 definition for a given TCP
* connection at its current state, taking into account SACK and etc.
*/
u_int
tcp_maxseg(const struct tcpcb *tp)
{
u_int optlen;
if (tp->t_flags & TF_NOOPT)
return (tp->t_maxseg);
/*
* Here we have a simplified code from tcp_addoptions(),
* without a proper loop, and having most of paddings hardcoded.
* We might make mistakes with padding here in some edge cases,
* but this is harmless, since result of tcp_maxseg() is used
* only in cwnd and ssthresh estimations.
*/
#define PAD(len) ((((len) / 4) + !!((len) % 4)) * 4)
if (TCPS_HAVEESTABLISHED(tp->t_state)) {
if (tp->t_flags & TF_RCVD_TSTMP)
optlen = TCPOLEN_TSTAMP_APPA;
else
optlen = 0;
#ifdef TCP_SIGNATURE
if (tp->t_flags & TF_SIGNATURE)
optlen += PAD(TCPOLEN_SIGNATURE);
#endif
if ((tp->t_flags & TF_SACK_PERMIT) && tp->rcv_numsacks > 0) {
optlen += TCPOLEN_SACKHDR;
optlen += tp->rcv_numsacks * TCPOLEN_SACK;
optlen = PAD(optlen);
}
} else {
if (tp->t_flags & TF_REQ_TSTMP)
optlen = TCPOLEN_TSTAMP_APPA;
else
optlen = PAD(TCPOLEN_MAXSEG);
if (tp->t_flags & TF_REQ_SCALE)
optlen += PAD(TCPOLEN_WINDOW);
#ifdef TCP_SIGNATURE
if (tp->t_flags & TF_SIGNATURE)
optlen += PAD(TCPOLEN_SIGNATURE);
#endif
if (tp->t_flags & TF_SACK_PERMIT)
optlen += PAD(TCPOLEN_SACK_PERMITTED);
}
#undef PAD
optlen = min(optlen, TCP_MAXOLEN);
return (tp->t_maxseg - optlen);
}
#ifdef IPSEC
/* compute ESP/AH header size for TCP, including outer IP header. */
size_t
ipsec_hdrsiz_tcp(struct tcpcb *tp)
{
struct inpcb *inp;
struct mbuf *m;
size_t hdrsiz;
struct ip *ip;
#ifdef INET6
struct ip6_hdr *ip6;
#endif
struct tcphdr *th;
if ((tp == NULL) || ((inp = tp->t_inpcb) == NULL) ||
(!key_havesp(IPSEC_DIR_OUTBOUND)))
return (0);
m = m_gethdr(M_NOWAIT, MT_DATA);
if (!m)
return (0);
#ifdef INET6
if ((inp->inp_vflag & INP_IPV6) != 0) {
ip6 = mtod(m, struct ip6_hdr *);
th = (struct tcphdr *)(ip6 + 1);
m->m_pkthdr.len = m->m_len =
sizeof(struct ip6_hdr) + sizeof(struct tcphdr);
tcpip_fillheaders(inp, ip6, th);
hdrsiz = ipsec_hdrsiz(m, IPSEC_DIR_OUTBOUND, inp);
} else
#endif /* INET6 */
{
ip = mtod(m, struct ip *);
th = (struct tcphdr *)(ip + 1);
m->m_pkthdr.len = m->m_len = sizeof(struct tcpiphdr);
tcpip_fillheaders(inp, ip, th);
hdrsiz = ipsec_hdrsiz(m, IPSEC_DIR_OUTBOUND, inp);
}
m_free(m);
return (hdrsiz);
}
#endif /* IPSEC */
#ifdef TCP_SIGNATURE
/*
* Callback function invoked by m_apply() to digest TCP segment data
* contained within an mbuf chain.
*/
static int
tcp_signature_apply(void *fstate, void *data, u_int len)
{
MD5Update(fstate, (u_char *)data, len);
return (0);
}
/*
* XXX The key is retrieved from the system's PF_KEY SADB, by keying a
* search with the destination IP address, and a 'magic SPI' to be
* determined by the application. This is hardcoded elsewhere to 1179
*/
struct secasvar *
tcp_get_sav(struct mbuf *m, u_int direction)
{
union sockaddr_union dst;
struct secasvar *sav;
struct ip *ip;
#ifdef INET6
struct ip6_hdr *ip6;
char ip6buf[INET6_ADDRSTRLEN];
#endif
/* Extract the destination from the IP header in the mbuf. */
bzero(&dst, sizeof(union sockaddr_union));
ip = mtod(m, struct ip *);
#ifdef INET6
ip6 = NULL; /* Make the compiler happy. */
#endif
switch (ip->ip_v) {
#ifdef INET
case IPVERSION:
dst.sa.sa_len = sizeof(struct sockaddr_in);
dst.sa.sa_family = AF_INET;
dst.sin.sin_addr = (direction == IPSEC_DIR_INBOUND) ?
ip->ip_src : ip->ip_dst;
break;
#endif
#ifdef INET6
case (IPV6_VERSION >> 4):
ip6 = mtod(m, struct ip6_hdr *);
dst.sa.sa_len = sizeof(struct sockaddr_in6);
dst.sa.sa_family = AF_INET6;
dst.sin6.sin6_addr = (direction == IPSEC_DIR_INBOUND) ?
ip6->ip6_src : ip6->ip6_dst;
break;
#endif
default:
return (NULL);
/* NOTREACHED */
break;
}
/* Look up an SADB entry which matches the address of the peer. */
sav = KEY_ALLOCSA(&dst, IPPROTO_TCP, htonl(TCP_SIG_SPI));
if (sav == NULL) {
ipseclog((LOG_ERR, "%s: SADB lookup failed for %s\n", __func__,
(ip->ip_v == IPVERSION) ? inet_ntoa(dst.sin.sin_addr) :
#ifdef INET6
(ip->ip_v == (IPV6_VERSION >> 4)) ?
ip6_sprintf(ip6buf, &dst.sin6.sin6_addr) :
#endif
"(unsupported)"));
}
return (sav);
}
/*
* Compute TCP-MD5 hash of a TCP segment. (RFC2385)
*
* Parameters:
* m pointer to head of mbuf chain
* len length of TCP segment data, excluding options
* optlen length of TCP segment options
* buf pointer to storage for computed MD5 digest
* sav pointer to security assosiation
*
* We do this over ip, tcphdr, segment data, and the key in the SADB.
* When called from tcp_input(), we can be sure that th_sum has been
* zeroed out and verified already.
*
* Releases reference to SADB key before return.
*
* Return 0 if successful, otherwise return -1.
*
*/
int
tcp_signature_do_compute(struct mbuf *m, int len, int optlen,
u_char *buf, struct secasvar *sav)
{
#ifdef INET
struct ippseudo ippseudo;
#endif
MD5_CTX ctx;
int doff;
struct ip *ip;
#ifdef INET
struct ipovly *ipovly;
#endif
struct tcphdr *th;
#ifdef INET6
struct ip6_hdr *ip6;
struct in6_addr in6;
uint32_t plen;
uint16_t nhdr;
#endif
u_short savecsum;
KASSERT(m != NULL, ("NULL mbuf chain"));
KASSERT(buf != NULL, ("NULL signature pointer"));
/* Extract the destination from the IP header in the mbuf. */
ip = mtod(m, struct ip *);
#ifdef INET6
ip6 = NULL; /* Make the compiler happy. */
#endif
MD5Init(&ctx);
/*
* Step 1: Update MD5 hash with IP(v6) pseudo-header.
*
* XXX The ippseudo header MUST be digested in network byte order,
* or else we'll fail the regression test. Assume all fields we've
* been doing arithmetic on have been in host byte order.
* XXX One cannot depend on ipovly->ih_len here. When called from
* tcp_output(), the underlying ip_len member has not yet been set.
*/
switch (ip->ip_v) {
#ifdef INET
case IPVERSION:
ipovly = (struct ipovly *)ip;
ippseudo.ippseudo_src = ipovly->ih_src;
ippseudo.ippseudo_dst = ipovly->ih_dst;
ippseudo.ippseudo_pad = 0;
ippseudo.ippseudo_p = IPPROTO_TCP;
ippseudo.ippseudo_len = htons(len + sizeof(struct tcphdr) +
optlen);
MD5Update(&ctx, (char *)&ippseudo, sizeof(struct ippseudo));
th = (struct tcphdr *)((u_char *)ip + sizeof(struct ip));
doff = sizeof(struct ip) + sizeof(struct tcphdr) + optlen;
break;
#endif
#ifdef INET6
/*
* RFC 2385, 2.0 Proposal
* For IPv6, the pseudo-header is as described in RFC 2460, namely the
* 128-bit source IPv6 address, 128-bit destination IPv6 address, zero-
* extended next header value (to form 32 bits), and 32-bit segment
* length.
* Note: Upper-Layer Packet Length comes before Next Header.
*/
case (IPV6_VERSION >> 4):
in6 = ip6->ip6_src;
in6_clearscope(&in6);
MD5Update(&ctx, (char *)&in6, sizeof(struct in6_addr));
in6 = ip6->ip6_dst;
in6_clearscope(&in6);
MD5Update(&ctx, (char *)&in6, sizeof(struct in6_addr));
plen = htonl(len + sizeof(struct tcphdr) + optlen);
MD5Update(&ctx, (char *)&plen, sizeof(uint32_t));
nhdr = 0;
MD5Update(&ctx, (char *)&nhdr, sizeof(uint8_t));
MD5Update(&ctx, (char *)&nhdr, sizeof(uint8_t));
MD5Update(&ctx, (char *)&nhdr, sizeof(uint8_t));
nhdr = IPPROTO_TCP;
MD5Update(&ctx, (char *)&nhdr, sizeof(uint8_t));
th = (struct tcphdr *)((u_char *)ip6 + sizeof(struct ip6_hdr));
doff = sizeof(struct ip6_hdr) + sizeof(struct tcphdr) + optlen;
break;
#endif
default:
KEY_FREESAV(&sav);
return (-1);
/* NOTREACHED */
break;
}
/*
* Step 2: Update MD5 hash with TCP header, excluding options.
* The TCP checksum must be set to zero.
*/
savecsum = th->th_sum;
th->th_sum = 0;
MD5Update(&ctx, (char *)th, sizeof(struct tcphdr));
th->th_sum = savecsum;
/*
* Step 3: Update MD5 hash with TCP segment data.
* Use m_apply() to avoid an early m_pullup().
*/
if (len > 0)
m_apply(m, doff, len, tcp_signature_apply, &ctx);
/*
* Step 4: Update MD5 hash with shared secret.
*/
MD5Update(&ctx, sav->key_auth->key_data, _KEYLEN(sav->key_auth));
MD5Final(buf, &ctx);
key_sa_recordxfer(sav, m);
KEY_FREESAV(&sav);
return (0);
}
/*
* Compute TCP-MD5 hash of a TCP segment. (RFC2385)
*
* Return 0 if successful, otherwise return -1.
*/
int
tcp_signature_compute(struct mbuf *m, int _unused, int len, int optlen,
u_char *buf, u_int direction)
{
struct secasvar *sav;
if ((sav = tcp_get_sav(m, direction)) == NULL)
return (-1);
return (tcp_signature_do_compute(m, len, optlen, buf, sav));
}
/*
* Verify the TCP-MD5 hash of a TCP segment. (RFC2385)
*
* Parameters:
* m pointer to head of mbuf chain
* len length of TCP segment data, excluding options
* optlen length of TCP segment options
* buf pointer to storage for computed MD5 digest
* direction direction of flow (IPSEC_DIR_INBOUND or OUTBOUND)
*
* Return 1 if successful, otherwise return 0.
*/
int
tcp_signature_verify(struct mbuf *m, int off0, int tlen, int optlen,
struct tcpopt *to, struct tcphdr *th, u_int tcpbflag)
{
char tmpdigest[TCP_SIGLEN];
if (tcp_sig_checksigs == 0)
return (1);
if ((tcpbflag & TF_SIGNATURE) == 0) {
if ((to->to_flags & TOF_SIGNATURE) != 0) {
/*
* If this socket is not expecting signature but
* the segment contains signature just fail.
*/
TCPSTAT_INC(tcps_sig_err_sigopt);
TCPSTAT_INC(tcps_sig_rcvbadsig);
return (0);
}
/* Signature is not expected, and not present in segment. */
return (1);
}
/*
* If this socket is expecting signature but the segment does not
* contain any just fail.
*/
if ((to->to_flags & TOF_SIGNATURE) == 0) {
TCPSTAT_INC(tcps_sig_err_nosigopt);
TCPSTAT_INC(tcps_sig_rcvbadsig);
return (0);
}
if (tcp_signature_compute(m, off0, tlen, optlen, &tmpdigest[0],
IPSEC_DIR_INBOUND) == -1) {
TCPSTAT_INC(tcps_sig_err_buildsig);
TCPSTAT_INC(tcps_sig_rcvbadsig);
return (0);
}
if (bcmp(to->to_signature, &tmpdigest[0], TCP_SIGLEN) != 0) {
TCPSTAT_INC(tcps_sig_rcvbadsig);
return (0);
}
TCPSTAT_INC(tcps_sig_rcvgoodsig);
return (1);
}
#endif /* TCP_SIGNATURE */
static int
sysctl_drop(SYSCTL_HANDLER_ARGS)
{
/* addrs[0] is a foreign socket, addrs[1] is a local one. */
struct sockaddr_storage addrs[2];
struct inpcb *inp;
struct tcpcb *tp;
struct tcptw *tw;
struct sockaddr_in *fin, *lin;
#ifdef INET6
struct sockaddr_in6 *fin6, *lin6;
#endif
int error;
inp = NULL;
fin = lin = NULL;
#ifdef INET6
fin6 = lin6 = NULL;
#endif
error = 0;
if (req->oldptr != NULL || req->oldlen != 0)
return (EINVAL);
if (req->newptr == NULL)
return (EPERM);
if (req->newlen < sizeof(addrs))
return (ENOMEM);
error = SYSCTL_IN(req, &addrs, sizeof(addrs));
if (error)
return (error);
switch (addrs[0].ss_family) {
#ifdef INET6
case AF_INET6:
fin6 = (struct sockaddr_in6 *)&addrs[0];
lin6 = (struct sockaddr_in6 *)&addrs[1];
if (fin6->sin6_len != sizeof(struct sockaddr_in6) ||
lin6->sin6_len != sizeof(struct sockaddr_in6))
return (EINVAL);
if (IN6_IS_ADDR_V4MAPPED(&fin6->sin6_addr)) {
if (!IN6_IS_ADDR_V4MAPPED(&lin6->sin6_addr))
return (EINVAL);
in6_sin6_2_sin_in_sock((struct sockaddr *)&addrs[0]);
in6_sin6_2_sin_in_sock((struct sockaddr *)&addrs[1]);
fin = (struct sockaddr_in *)&addrs[0];
lin = (struct sockaddr_in *)&addrs[1];
break;
}
error = sa6_embedscope(fin6, V_ip6_use_defzone);
if (error)
return (error);
error = sa6_embedscope(lin6, V_ip6_use_defzone);
if (error)
return (error);
break;
#endif
#ifdef INET
case AF_INET:
fin = (struct sockaddr_in *)&addrs[0];
lin = (struct sockaddr_in *)&addrs[1];
if (fin->sin_len != sizeof(struct sockaddr_in) ||
lin->sin_len != sizeof(struct sockaddr_in))
return (EINVAL);
break;
#endif
default:
return (EINVAL);
}
INP_INFO_RLOCK(&V_tcbinfo);
switch (addrs[0].ss_family) {
#ifdef INET6
case AF_INET6:
inp = in6_pcblookup(&V_tcbinfo, &fin6->sin6_addr,
fin6->sin6_port, &lin6->sin6_addr, lin6->sin6_port,
INPLOOKUP_WLOCKPCB, NULL);
break;
#endif
#ifdef INET
case AF_INET:
inp = in_pcblookup(&V_tcbinfo, fin->sin_addr, fin->sin_port,
lin->sin_addr, lin->sin_port, INPLOOKUP_WLOCKPCB, NULL);
break;
#endif
}
if (inp != NULL) {
if (inp->inp_flags & INP_TIMEWAIT) {
/*
* XXXRW: There currently exists a state where an
* inpcb is present, but its timewait state has been
* discarded. For now, don't allow dropping of this
* type of inpcb.
*/
tw = intotw(inp);
if (tw != NULL)
tcp_twclose(tw, 0);
else
INP_WUNLOCK(inp);
} else if (!(inp->inp_flags & INP_DROPPED) &&
!(inp->inp_socket->so_options & SO_ACCEPTCONN)) {
tp = intotcpcb(inp);
tp = tcp_drop(tp, ECONNABORTED);
if (tp != NULL)
INP_WUNLOCK(inp);
} else
INP_WUNLOCK(inp);
} else
error = ESRCH;
INP_INFO_RUNLOCK(&V_tcbinfo);
return (error);
}
SYSCTL_PROC(_net_inet_tcp, TCPCTL_DROP, drop,
CTLFLAG_VNET | CTLTYPE_STRUCT | CTLFLAG_WR | CTLFLAG_SKIP, NULL,
0, sysctl_drop, "", "Drop TCP connection");
/*
* Generate a standardized TCP log line for use throughout the
* tcp subsystem. Memory allocation is done with M_NOWAIT to
* allow use in the interrupt context.
*
* NB: The caller MUST free(s, M_TCPLOG) the returned string.
* NB: The function may return NULL if memory allocation failed.
*
* Due to header inclusion and ordering limitations the struct ip
* and ip6_hdr pointers have to be passed as void pointers.
*/
char *
tcp_log_vain(struct in_conninfo *inc, struct tcphdr *th, void *ip4hdr,
const void *ip6hdr)
{
/* Is logging enabled? */
if (tcp_log_in_vain == 0)
return (NULL);
return (tcp_log_addr(inc, th, ip4hdr, ip6hdr));
}
char *
tcp_log_addrs(struct in_conninfo *inc, struct tcphdr *th, void *ip4hdr,
const void *ip6hdr)
{
/* Is logging enabled? */
if (tcp_log_debug == 0)
return (NULL);
return (tcp_log_addr(inc, th, ip4hdr, ip6hdr));
}
static char *
tcp_log_addr(struct in_conninfo *inc, struct tcphdr *th, void *ip4hdr,
const void *ip6hdr)
{
char *s, *sp;
size_t size;
struct ip *ip;
#ifdef INET6
const struct ip6_hdr *ip6;
ip6 = (const struct ip6_hdr *)ip6hdr;
#endif /* INET6 */
ip = (struct ip *)ip4hdr;
/*
* The log line looks like this:
* "TCP: [1.2.3.4]:50332 to [1.2.3.4]:80 tcpflags 0x2<SYN>"
*/
size = sizeof("TCP: []:12345 to []:12345 tcpflags 0x2<>") +
sizeof(PRINT_TH_FLAGS) + 1 +
#ifdef INET6
2 * INET6_ADDRSTRLEN;
#else
2 * INET_ADDRSTRLEN;
#endif /* INET6 */
s = malloc(size, M_TCPLOG, M_ZERO|M_NOWAIT);
if (s == NULL)
return (NULL);
strcat(s, "TCP: [");
sp = s + strlen(s);
if (inc && ((inc->inc_flags & INC_ISIPV6) == 0)) {
inet_ntoa_r(inc->inc_faddr, sp);
sp = s + strlen(s);
sprintf(sp, "]:%i to [", ntohs(inc->inc_fport));
sp = s + strlen(s);
inet_ntoa_r(inc->inc_laddr, sp);
sp = s + strlen(s);
sprintf(sp, "]:%i", ntohs(inc->inc_lport));
#ifdef INET6
} else if (inc) {
ip6_sprintf(sp, &inc->inc6_faddr);
sp = s + strlen(s);
sprintf(sp, "]:%i to [", ntohs(inc->inc_fport));
sp = s + strlen(s);
ip6_sprintf(sp, &inc->inc6_laddr);
sp = s + strlen(s);
sprintf(sp, "]:%i", ntohs(inc->inc_lport));
} else if (ip6 && th) {
ip6_sprintf(sp, &ip6->ip6_src);
sp = s + strlen(s);
sprintf(sp, "]:%i to [", ntohs(th->th_sport));
sp = s + strlen(s);
ip6_sprintf(sp, &ip6->ip6_dst);
sp = s + strlen(s);
sprintf(sp, "]:%i", ntohs(th->th_dport));
#endif /* INET6 */
#ifdef INET
} else if (ip && th) {
inet_ntoa_r(ip->ip_src, sp);
sp = s + strlen(s);
sprintf(sp, "]:%i to [", ntohs(th->th_sport));
sp = s + strlen(s);
inet_ntoa_r(ip->ip_dst, sp);
sp = s + strlen(s);
sprintf(sp, "]:%i", ntohs(th->th_dport));
#endif /* INET */
} else {
free(s, M_TCPLOG);
return (NULL);
}
sp = s + strlen(s);
if (th)
sprintf(sp, " tcpflags 0x%b", th->th_flags, PRINT_TH_FLAGS);
if (*(s + size - 1) != '\0')
panic("%s: string too long", __func__);
return (s);
}
/*
* A subroutine which makes it easy to track TCP state changes with DTrace.
* This function shouldn't be called for t_state initializations that don't
* correspond to actual TCP state transitions.
*/
void
tcp_state_change(struct tcpcb *tp, int newstate)
{
#if defined(KDTRACE_HOOKS)
int pstate = tp->t_state;
#endif
TCPSTATES_DEC(tp->t_state);
TCPSTATES_INC(newstate);
tp->t_state = newstate;
TCP_PROBE6(state__change, NULL, tp, NULL, tp, NULL, pstate);
}
|