summaryrefslogtreecommitdiffstats
path: root/src/vnet/tcp/tcp.c
blob: 7dce97ea0c2a8f619a42064b6035cebad7452422 (plain)
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

@media only all and (prefers-color-scheme: dark) {
.highlight .hll { background-color: #49483e }
.highlight .c { color: #75715e } /* Comment */
.highlight .err { color: #960050; background-color: #1e0010 } /* Error */
.highlight .k { color: #66d9ef } /* Keyword */
.highlight .l { color: #ae81ff } /* Literal */
.highlight .n { color: #f8f8f2 } /* Name */
.highlight .o { color: #f92672 } /* Operator */
.highlight .p { color: #f8f8f2 } /* Punctuation */
.highlight .ch { color: #75715e } /* Comment.Hashbang */
.highlight .cm { color: #75715e } /* Comment.Multiline */
.highlight .cp { color: #75715e } /* Comment.Preproc */
.highlight .cpf { color: #75715e } /* Comment.PreprocFile */
.highlight .c1 { color: #75715e } /* Comment.Single */
.highlight .cs { color: #75715e } /* Comment.Special */
.highlight .gd { color: #f92672 } /* Generic.Deleted */
.highlight .ge { font-style: italic } /* Generic.Emph */
.highlight .gi { color: #a6e22e } /* Generic.Inserted */
.highlight .gs { font-weight: bold } /* Generic.Strong */
.highlight .gu { color: #75715e } /* Generic.Subheading */
.highlight .kc { color: #66d9ef } /* Keyword.Constant */
.highlight .kd { color: #66d9ef } /* Keyword.Declaration */
.highlight .kn { color: #f92672 } /* Keyword.Namespace */
.highlight .kp { color: #66d9ef } /* Keyword.Pseudo */
.highlight .kr { color: #66d9ef } /* Keyword.Reserved */
.highlight .kt { color: #66d9ef } /* Keyword.Type */
.highlight .ld { color: #e6db74 } /* Literal.Date */
.highlight .m { color: #ae81ff } /* Literal.Number */
.highlight .s { color: #e6db74 } /* Literal.String */
.highlight .na { color: #a6e22e } /* Name.Attribute */
.highlight .nb { color: #f8f8f2 } /* Name.Builtin */
.highlight .nc { color: #a6e22e } /* Name.Class */
.highlight .no { color: #66d9ef } /* Name.Constant */
.highlight .nd { color: #a6e22e } /* Name.Decorator */
.highlight .ni { color: #f8f8f2 } /* Name.Entity */
.highlight .ne { color: #a6e22e } /* Name.Exception */
.highlight .nf { color: #a6e22e } /* Name.Function */
.highlight .nl { color: #f8f8f2 } /* Name.Label */
.highlight .nn { color: #f8f8f2 } /* Name.Namespace */
.highlight .nx { color: #a6e22e } /* Name.Other */
.highlight .py { color: #f8f8f2 } /* Name.Property */
.highlight .nt { color: #f92672 } /* Name.Tag */
.highlight .nv { color: #f8f8f2 } /* Name.Variable */
.highlight .ow { color: #f92672 } /* Operator.Word */
.highlight .w { color: #f8f8f2 } /* Text.Whitespace */
.highlight .mb { color: #ae81ff } /* Literal.Number.Bin */
.highlight .mf { color: #ae81ff } /* Literal.Number.Float */
.highlight .mh { color: #ae81ff } /* Literal.Number.Hex */
.highlight .mi { color: #ae81ff } /* Literal.Number.Integer */
.highlight .mo { color: #ae81ff } /* Literal.Number.Oct */
.highlight .sa { color: #e6db74 } /* Literal.String.Affix */
.highlight .sb { color: #e6db74 } /* Literal.String.Backtick */
.highlight .sc { color: #e6db74 } /* Literal.String.Char */
.highlight .dl { color: #e6db74 } /* Literal.String.Delimiter */
.highlight .sd { color: #e6db74 } /* Literal.String.Doc */
.highlight .s2 { color: #e6db74 } /* Literal.String.Double */
.highlight .se { color: #ae81ff } /* Literal.String.Escape */
.highlight .sh { color: #e6db74 } /* Literal.String.Heredoc */
.highlight .si { color: #e6db74 } /* Literal.String.Interpol */
.highlight .sx { color: #e6db74 } /* Literal.String.Other */
.highlight .sr { color: #e6db74 } /* Literal.String.Regex */
.highlight .s1 { color: #e6db74 } /* Literal.String.Single */
.highlight .ss { color: #e6db74 } /* Literal.String.Symbol */
.highlight .bp { color: #f8f8f2 } /* Name.Builtin.Pseudo */
.highlight .fm { color: #a6e22e } /* Name.Function.Magic */
.highlight .vc { color: #f8f8f2 } /* Name.Variable.Class */
.highlight .vg { color: #f8f8f2 } /* Name.Variable.Global */
.highlight .vi { color: #f8f8f2 } /* Name.Variable.Instance */
.highlight .vm { color: #f8f8f2 } /* Name.Variable.Magic */
.highlight .il { color: #ae81ff } /* Literal.Number.Integer.Long */
}
@media (prefers-color-scheme: light) {
.highlight .hll { background-color: #ffffcc }
.highlight .c { color: #888888 } /* Comment */
.highlight .err { color: #a61717; background-color: #e3d2d2 } /* Error */
.highlight .k { color: #008800; font-weight: bold } /* Keyword */
.highlight .ch { color: #888888 } /* Comment.Hashbang */
.highlight .cm { color: #888888 } /* Comment.Multiline */
.highlight .cp { color: #cc0000; font-weight: bold } /* Comment.Preproc */
.highlight .cpf { color: #888888 } /* Comment.PreprocFile */
.highlight .c1 { color: #888888 } /* Comment.Single */
.highlight .cs { color: #cc0000; font-weight: bold; background-color: #fff0f0 } /* Comment.Special */
.highlight .gd { color: #000000; background-color: #ffdddd } /* Generic.Deleted */
.highlight .ge { font-style: italic } /* Generic.Emph */
.highlight .gr { color: #aa0000 } /* Generic.Error */
.highlight .gh { color: #333333 } /* Generic.Heading */
.highlight .gi { color: #000000; background-color: #ddffdd } /* Generic.Inserted */
.highlight .go { color: #888888 } /* Generic.Output */
.highlight .gp { color: #555555 } /* Generic.Prompt */
.highlight .gs { font-weight: bold } /* Generic.Strong */
.highlight .gu { color: #666666 } /* Generic.Subheading */
.highlight .gt { color: #aa0000 } /* Generic.Traceback */
.highlight .kc { color: #008800; font-weight: bold } /* Keyword.Constant */
.highlight .kd { color: #008800; font-weight: bold } /* Keyword.Declaration */
.highlight .kn { color: #008800; font-weight: bold } /* Keyword.Namespace */
.highlight .kp { color: #008800 } /* Keyword.Pseudo */
.highlight .kr { color: #008800; font-weight: bold } /* Keyword.Reserved */
.highlight .kt { color: #888888; font-weight: bold } /* Keyword.Type */
.highlight .m { color: #0000DD; font-weight: bold } /* Literal.Number */
.highlight .s { color: #dd2200; background-color: #fff0f0 } /* Literal.String */
.highlight .na { color: #336699 } /* Name.Attribute */
.highlight .nb { color: #003388 } /* Name.Builtin */
.highlight .nc { color: #bb0066; font-weight: bold } /* Name.Class */
.highlight .no { color: #003366; font-weight: bold } /* Name.Constant */
.highlight .nd { color: #555555 } /* Name.Decorator */
.highlight .ne { color: #bb0066; font-weight: bold } /* Name.Exception */
.highlight .nf { color: #0066bb; font-weight: bold } /* Name.Function */
.highlight .nl { color: #336699; font-style: italic } /* Name.Label */
.highlight .nn { color: #bb0066; font-weight: bold } /* Name.Namespace */
.highlight .py { color: #336699; font-weight: bold } /* Name.Property */
.highlight .nt { color: #bb0066; font-weight: bold } /* Name.Tag */
.highlight .nv { color: #336699 } /* Name.Variable */
.highlight .ow { color: #008800 } /* Operator.Word */
.highlight .w { color: #bbbbbb } /* Text.Whitespace */
.highlight .mb { color: #0000DD; font-weight: bold } /* Literal.Number.Bin */
.highlight .mf { color: #0000DD; font-weight: bold } /* Literal.Number.Float */
.highlight .mh { color: #0000DD; font-weight: bold } /* Literal.Number.Hex */
.highlight .mi { color: #0000DD; font-weight: bold } /* Literal.Number.Integer */
.highlight .mo { color: #0000DD; font-weight: bold } /* Literal.Number.Oct */
.highlight .sa { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Affix */
.highlight .sb { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Backtick */
.highlight .sc { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Char */
.highlight .dl { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Delimiter */
.highlight .sd { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Doc */
.highlight .s2 { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Double */
.highlight .se { color: #0044dd; background-color: #fff0f0 } /* Literal.String.Escape */
.highlight .sh { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Heredoc */
.highlight .si { color: #3333bb; background-color: #fff0f0 } /* Literal.String.Interpol */
.highlight .sx { color: #22bb22; background-color: #f0fff0 } /* Literal.String.Other */
.highlight .sr { color: #008800; background-color: #fff0ff } /* Literal.String.Regex */
.highlight .s1 { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Single */
.highlight .ss { color: #aa6600; background-color: #fff0f0 } /* Literal.String.Symbol */
.highlight .bp { color: #003388 } /* Name.Builtin.Pseudo */
.highlight .fm { color: #0066bb; font-weight: bold } /* Name.Function.Magic */
.highlight .vc { color: #336699 } /* Name.Variable.Class */
.highlight .vg { color: #dd7700 } /* Name.Variable.Global */
.highlight .vi { color: #3333bb } /* Name.Variable.Instance */
.highlight .vm { color: #336699 } /* Name.Variable.Magic */
.highlight .il { color: #0000DD; font-weight: bold } /* Literal.Number.Integer.Long */
}
#!/bin/bash
# Copyright (c) 2017 Cisco and/or its affiliates.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at:
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

set -x

cat /etc/hostname
cat /etc/hosts

SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"

VIRL_SERVERS=("10.30.51.28" "10.30.51.29" "10.30.51.30")
VIRL_SERVER=""

VIRL_USERNAME=jenkins-in
VIRL_PKEY=priv_key
VIRL_SERVER_STATUS_FILE="status"
VIRL_SERVER_EXPECTED_STATUS="PRODUCTION"

JOB_ARCHIVE_ARTIFACTS=(log.html output.xml report.html honeycomb.log karaf.log)
LOG_ARCHIVE_ARTIFACTS=(log.html output.xml report.html honeycomb.log karaf.log)
JOB_ARCHIVE_DIR="archive"
LOG_ARCHIVE_DIR="$WORKSPACE/archives"
mkdir -p ${JOB_ARCHIVE_DIR}
mkdir -p ${LOG_ARCHIVE_DIR}

STREAM=$1
OS=$2
ODL=$3

if [ -f "/etc/redhat-release" ]; then
    DISTRO="CENTOS"
    sudo yum install -y python-devel python-virtualenv openssh-clients sshpass
    VIRL_TOPOLOGY=$(cat ${SCRIPT_DIR}/VIRL_TOPOLOGY_CENTOS)
    VIRL_RELEASE=$(cat ${SCRIPT_DIR}/VIRL_RELEASE_CENTOS)
else
    DISTRO="UBUNTU"
    export DEBIAN_FRONTEND=noninteractive
    sudo apt-get -y update
    sudo apt-get -y install libpython2.7-dev python-virtualenv
    VIRL_TOPOLOGY=$(cat ${SCRIPT_DIR}/VIRL_TOPOLOGY_UBUNTU)
    VIRL_RELEASE=$(cat ${SCRIPT_DIR}/VIRL_RELEASE_UBUNTU)
fi

SSH_OPTIONS="-i ${VIRL_PKEY} -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -o BatchMode=yes -o LogLevel=error"

function ssh_do() {
    echo
    echo "### "  ssh $@
    ssh ${SSH_OPTIONS} $@
}

rm -f ${VIRL_PKEY}
cat > ${VIRL_PKEY} <<EOF
-----BEGIN RSA PRIVATE KEY-----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-----END RSA PRIVATE KEY-----
EOF
chmod 600 ${VIRL_PKEY}

#
# Pick a random host from the array of VIRL servers, and attempt
# to reach it and verify it's status.
#
# The server must be reachable, and have a "status" file with
# the content "PRODUCTION", to be selected.
#
# If the server is not reachable, or does not have the correct
# status, remove it from the array and start again.
#
# Abort if there are no more servers left in the array.
#
while [[ ! "$VIRL_SERVER" ]]
do
    num_hosts=${#VIRL_SERVERS[@]}
    if [ $num_hosts == 0 ]
    then
        echo "No more VIRL candidate hosts available, failing."
        exit 127
    fi
    element=$[ $RANDOM % $num_hosts ]
    virl_server_candidate=${VIRL_SERVERS[$element]}
    virl_server_status=$(ssh ${SSH_OPTIONS} ${VIRL_USERNAME}@${virl_server_candidate} cat $VIRL_SERVER_STATUS_FILE 2>&1)
    echo VIRL HOST $virl_server_candidate status is \"$virl_server_status\"
    if [ "$virl_server_status" == "$VIRL_SERVER_EXPECTED_STATUS" ]
    then
        # Candidate is in good status. Select this server.
        VIRL_SERVER="$virl_server_candidate"
    else
        # Candidate is in bad status. Remove from array.
        VIRL_SERVERS=("${VIRL_SERVERS[@]:0:$element}" "${VIRL_SERVERS[@]:$[$element+1]}")
    fi
done

SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"

# Download VPP and HC packages from the current branch
echo Downloading packages...
bash ${SCRIPT_DIR}/resources/tools/scripts/download_hc_pkgs.sh ${STREAM} ${OS}

if [ "${OS}" == "centos7" ]; then
    VPP_PKGS=(*.rpm)
else
    VPP_PKGS=(*.deb)
fi
echo ${VPP_PKGS[@]}
VIRL_DIR_LOC="/tmp"
VPP_PKGS_FULL=(${VPP_PKGS[@]})

# Prepend directory location at remote host to package file list
for index in "${!VPP_PKGS_FULL[@]}"; do
    VPP_PKGS_FULL[${index}]=${VIRL_DIR_LOC}/${VPP_PKGS_FULL[${index}]}
done

echo "Updated file names: " ${VPP_PKGS_FULL[@]}

cat ${VIRL_PKEY}

# Copy the files to VIRL hosts
DONE=""
for index in "${!VIRL_SERVER[@]}"; do
    # Do not copy files in case they have already been copied to the VIRL host
    [[ "${DONE[@]}" =~ "${VIRL_SERVER[${index}]}" ]] && copy=0 || copy=1

    if [ "${copy}" -eq "0" ]; then
        echo "files have already been copied to the VIRL host ${VIRL_SERVER[${index}]}"
    else
        if [ "${OS}" == "centos7" ]; then
            scp ${SSH_OPTIONS} *.rpm \
                ${VIRL_USERNAME}@${VIRL_SERVER[${index}]}:${VIRL_DIR_LOC}/
        else
            scp ${SSH_OPTIONS} *.deb \
                ${VIRL_USERNAME}@${VIRL_SERVER[${index}]}:${VIRL_DIR_LOC}/
        fi
         result=$?
        if [ "${result}" -ne "0" ]; then
            echo "Failed to copy files to VIRL host ${VIRL_SERVER[${index}]}"
            echo ${result}
            exit ${result}
        else
            echo "files successfully copied to the VIRL host ${VIRL_SERVER[${index}]}"
        fi
        DONE+=(${VIRL_SERVER[${index}]})
    fi
done

# Start a simulation on VIRL server
echo "Starting simulation on VIRL server"

function stop_virl_simulation {
    ssh ${SSH_OPTIONS} ${VIRL_USERNAME}@${VIRL_SERVER}\
        "stop-testcase ${VIRL_SID}"
}

VIRL_SID=$(ssh ${SSH_OPTIONS} \
    ${VIRL_USERNAME}@${VIRL_SERVER} \
    "start-testcase -c ${VIRL_TOPOLOGY} -r ${VIRL_RELEASE} ${VPP_PKGS_FULL[@]}")
retval=$?
if [ "${retval}" -ne "0" ]; then
    echo "VIRL simulation start failed"
    exit ${retval}
fi

if [[ ! "${VIRL_SID}" =~ session-[a-zA-Z0-9_]{6} ]]; then
    echo "No VIRL session ID reported."
    exit 127
fi

# Upon script exit, cleanup the simulation execution
trap stop_virl_simulation EXIT
echo ${VIRL_SID}

ssh_do ${VIRL_USERNAME}@${VIRL_SERVER} cat /scratch/${VIRL_SID}/topology.yaml

# Download the topology file from virl session
scp ${SSH_OPTIONS} \
    ${VIRL_USERNAME}@${VIRL_SERVER}:/scratch/${VIRL_SID}/topology.yaml \
    topologies/enabled/topology.yaml

retval=$?
if [ "${retval}" -ne "0" ]; then
    echo "Failed to copy topology file from VIRL simulation"
    exit ${retval}
fi

virtualenv --system-site-packages env
. env/bin/activate

echo pip install
pip install -r ${SCRIPT_DIR}/requirements.txt

pykwalify -s ${SCRIPT_DIR}/resources/topology_schemas/3_node_topology.sch.yaml \
          -s ${SCRIPT_DIR}/resources/topology_schemas/topology.sch.yaml \
          -d ${SCRIPT_DIR}/topologies/enabled/topology.yaml \
          -vvv

if [ "$?" -ne "0" ]; then
    echo "Topology schema validation failed."
    echo "However, the tests will start."
fi

# TODO(CSIT-994): reenable NSH
PYTHONPATH=`pwd` pybot -L TRACE -W 136\
    -v TOPOLOGY_PATH:${SCRIPT_DIR}/topologies/enabled/topology.yaml \
    --suite "tests.honeycomb.func" \
    --include HC_FUNC \
    --exclude hc_nshORhc_rest_only \
    --noncritical EXPECTED_FAILING \
    --variable HC_ODL:${ODL} \
    tests/

RETURN_STATUS=$?

# Get Honeycomb log file from virl host
scp ${SSH_OPTIONS} \
    ${VIRL_USERNAME}@${VIRL_SERVER}:/scratch/${VIRL_SID}/honeycomb.log . || true

# Get ODL karaf log file from virl host
scp ${SSH_OPTIONS} \
    ${VIRL_USERNAME}@${VIRL_SERVER}:/scratch/${VIRL_SID}/karaf.log . || true

# Archive JOB artifacts in jenkins
for i in ${JOB_ARCHIVE_ARTIFACTS[@]}; do
    cp $( readlink -f ${i} | tr '\n' ' ' ) ${JOB_ARCHIVE_DIR}/
done
# Archive JOB artifacts to logs.fd.io
for i in ${LOG_ARCHIVE_ARTIFACTS[@]}; do
    cp $( readlink -f ${i} | tr '\n' ' ' ) ${LOG_ARCHIVE_DIR}/
done

exit ${RETURN_STATUS}
0'>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
/*
 * Copyright (c) 2016-2019 Cisco and/or its affiliates.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at:
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

/**
 * @file
 * @brief TCP host stack utilities
 */

#include <vnet/tcp/tcp.h>
#include <vnet/session/session.h>
#include <vnet/fib/fib.h>
#include <vnet/dpo/load_balance.h>
#include <vnet/dpo/receive_dpo.h>
#include <vnet/ip/ip6_neighbor.h>
#include <math.h>
#include <vnet/ethernet/arp.h>

tcp_main_t tcp_main;

typedef struct
{
  fib_protocol_t nh_proto;
  vnet_link_t link_type;
  ip46_address_t ip;
  u32 sw_if_index;
  u8 is_add;
} tcp_add_del_adj_args_t;

static void
tcp_add_del_adj_cb (tcp_add_del_adj_args_t * args)
{
  u32 ai;
  if (args->is_add)
    {
      adj_nbr_add_or_lock (args->nh_proto, args->link_type, &args->ip,
			   args->sw_if_index);
    }
  else
    {
      ai = adj_nbr_find (FIB_PROTOCOL_IP6, VNET_LINK_IP6, &args->ip,
			 args->sw_if_index);
      if (ai != ADJ_INDEX_INVALID)
	adj_unlock (ai);
    }
}

static void
tcp_add_del_adjacency (tcp_connection_t * tc, u8 is_add)
{
  tcp_add_del_adj_args_t args = {
    .nh_proto = FIB_PROTOCOL_IP6,
    .link_type = VNET_LINK_IP6,
    .ip = tc->c_rmt_ip,
    .sw_if_index = tc->sw_if_index,
    .is_add = is_add
  };
  vlib_rpc_call_main_thread (tcp_add_del_adj_cb, (u8 *) & args,
			     sizeof (args));
}

static void
tcp_cc_init (tcp_connection_t * tc)
{
  tc->cc_algo->init (tc);
}

static void
tcp_cc_cleanup (tcp_connection_t * tc)
{
  if (tc->cc_algo->cleanup)
    tc->cc_algo->cleanup (tc);
}

void
tcp_cc_algo_register (tcp_cc_algorithm_type_e type,
		      const tcp_cc_algorithm_t * vft)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  vec_validate (tm->cc_algos, type);

  tm->cc_algos[type] = *vft;
  hash_set_mem (tm->cc_algo_by_name, vft->name, type);
}

tcp_cc_algorithm_t *
tcp_cc_algo_get (tcp_cc_algorithm_type_e type)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  return &tm->cc_algos[type];
}

tcp_cc_algorithm_type_e
tcp_cc_algo_new_type (const tcp_cc_algorithm_t * vft)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  tcp_cc_algo_register (++tm->cc_last_type, vft);
  return tm->cc_last_type;
}

static u32
tcp_connection_bind (u32 session_index, transport_endpoint_t * lcl)
{
  tcp_main_t *tm = &tcp_main;
  tcp_connection_t *listener;
  void *iface_ip;

  pool_get (tm->listener_pool, listener);
  clib_memset (listener, 0, sizeof (*listener));

  listener->c_c_index = listener - tm->listener_pool;
  listener->c_lcl_port = lcl->port;

  /* If we are provided a sw_if_index, bind using one of its ips */
  if (ip_is_zero (&lcl->ip, 1) && lcl->sw_if_index != ENDPOINT_INVALID_INDEX)
    {
      if ((iface_ip = ip_interface_get_first_ip (lcl->sw_if_index,
						 lcl->is_ip4)))
	ip_set (&lcl->ip, iface_ip, lcl->is_ip4);
    }
  ip_copy (&listener->c_lcl_ip, &lcl->ip, lcl->is_ip4);
  listener->c_is_ip4 = lcl->is_ip4;
  listener->c_proto = TRANSPORT_PROTO_TCP;
  listener->c_s_index = session_index;
  listener->c_fib_index = lcl->fib_index;
  listener->state = TCP_STATE_LISTEN;
  listener->cc_algo = tcp_cc_algo_get (tcp_cfg.cc_algo);

  tcp_connection_timers_init (listener);

  TCP_EVT (TCP_EVT_BIND, listener);

  return listener->c_c_index;
}

static u32
tcp_session_bind (u32 session_index, transport_endpoint_t * tep)
{
  return tcp_connection_bind (session_index, tep);
}

static void
tcp_connection_unbind (u32 listener_index)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  tcp_connection_t *tc;

  tc = pool_elt_at_index (tm->listener_pool, listener_index);

  TCP_EVT (TCP_EVT_UNBIND, tc);

  /* Poison the entry */
  if (CLIB_DEBUG > 0)
    clib_memset (tc, 0xFA, sizeof (*tc));

  pool_put_index (tm->listener_pool, listener_index);
}

static u32
tcp_session_unbind (u32 listener_index)
{
  tcp_connection_unbind (listener_index);
  return 0;
}

static transport_connection_t *
tcp_session_get_listener (u32 listener_index)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  tcp_connection_t *tc;
  tc = pool_elt_at_index (tm->listener_pool, listener_index);
  return &tc->connection;
}

/**
 * Cleanup half-open connection
 *
 */
static void
tcp_half_open_connection_del (tcp_connection_t * tc)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  clib_spinlock_lock_if_init (&tm->half_open_lock);
  if (CLIB_DEBUG)
    clib_memset (tc, 0xFA, sizeof (*tc));
  pool_put (tm->half_open_connections, tc);
  clib_spinlock_unlock_if_init (&tm->half_open_lock);
}

/**
 * Try to cleanup half-open connection
 *
 * If called from a thread that doesn't own tc, the call won't have any
 * effect.
 *
 * @param tc - connection to be cleaned up
 * @return non-zero if cleanup failed.
 */
int
tcp_half_open_connection_cleanup (tcp_connection_t * tc)
{
  /* Make sure this is the owning thread */
  if (tc->c_thread_index != vlib_get_thread_index ())
    return 1;
  tcp_timer_reset (tc, TCP_TIMER_RETRANSMIT_SYN);
  tcp_half_open_connection_del (tc);
  return 0;
}

static tcp_connection_t *
tcp_half_open_connection_new (void)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  tcp_connection_t *tc = 0;
  ASSERT (vlib_get_thread_index () == 0);
  pool_get (tm->half_open_connections, tc);
  clib_memset (tc, 0, sizeof (*tc));
  tc->c_c_index = tc - tm->half_open_connections;
  return tc;
}

/**
 * Cleans up connection state.
 *
 * No notifications.
 */
void
tcp_connection_cleanup (tcp_connection_t * tc)
{
  tcp_main_t *tm = &tcp_main;

  TCP_EVT (TCP_EVT_DELETE, tc);

  /* Cleanup local endpoint if this was an active connect */
  transport_endpoint_cleanup (TRANSPORT_PROTO_TCP, &tc->c_lcl_ip,
			      tc->c_lcl_port);

  /* Check if connection is not yet fully established */
  if (tc->state == TCP_STATE_SYN_SENT)
    {
      /* Try to remove the half-open connection. If this is not the owning
       * thread, tc won't be removed. Retransmit or establish timers will
       * eventually expire and call again cleanup on the right thread. */
      if (tcp_half_open_connection_cleanup (tc))
	tc->flags |= TCP_CONN_HALF_OPEN_DONE;
    }
  else
    {
      int thread_index = tc->c_thread_index;

      /* Make sure all timers are cleared */
      tcp_connection_timers_reset (tc);

      if (!tc->c_is_ip4 && ip6_address_is_link_local_unicast (&tc->c_rmt_ip6))
	tcp_add_del_adjacency (tc, 0);

      tcp_cc_cleanup (tc);
      vec_free (tc->snd_sacks);
      vec_free (tc->snd_sacks_fl);
      vec_free (tc->rcv_opts.sacks);
      pool_free (tc->sack_sb.holes);

      if (tc->cfg_flags & TCP_CFG_F_RATE_SAMPLE)
	tcp_bt_cleanup (tc);

      /* Poison the entry */
      if (CLIB_DEBUG > 0)
	clib_memset (tc, 0xFA, sizeof (*tc));
      pool_put (tm->connections[thread_index], tc);
    }
}

/**
 * Connection removal.
 *
 * This should be called only once connection enters CLOSED state. Note
 * that it notifies the session of the removal event, so if the goal is to
 * just remove the connection, call tcp_connection_cleanup instead.
 */
void
tcp_connection_del (tcp_connection_t * tc)
{
  session_transport_delete_notify (&tc->connection);
  tcp_connection_cleanup (tc);
}

tcp_connection_t *
tcp_connection_alloc (u8 thread_index)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  tcp_connection_t *tc;

  pool_get (tm->connections[thread_index], tc);
  clib_memset (tc, 0, sizeof (*tc));
  tc->c_c_index = tc - tm->connections[thread_index];
  tc->c_thread_index = thread_index;
  return tc;
}

tcp_connection_t *
tcp_connection_alloc_w_base (u8 thread_index, tcp_connection_t * base)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  tcp_connection_t *tc;

  pool_get (tm->connections[thread_index], tc);
  clib_memcpy_fast (tc, base, sizeof (*tc));
  tc->c_c_index = tc - tm->connections[thread_index];
  tc->c_thread_index = thread_index;
  return tc;
}

void
tcp_connection_free (tcp_connection_t * tc)
{
  tcp_main_t *tm = &tcp_main;
  if (CLIB_DEBUG)
    {
      u8 thread_index = tc->c_thread_index;
      clib_memset (tc, 0xFA, sizeof (*tc));
      pool_put (tm->connections[thread_index], tc);
      return;
    }
  pool_put (tm->connections[tc->c_thread_index], tc);
}

/** Notify session that connection has been reset.
 *
 * Switch state to closed and wait for session to call cleanup.
 */
void
tcp_connection_reset (tcp_connection_t * tc)
{
  TCP_EVT (TCP_EVT_RST_RCVD, tc);
  switch (tc->state)
    {
    case TCP_STATE_SYN_RCVD:
      /* Cleanup everything. App wasn't notified yet */
      session_transport_delete_notify (&tc->connection);
      tcp_connection_cleanup (tc);
      break;
    case TCP_STATE_SYN_SENT:
      session_stream_connect_notify (&tc->connection, 1 /* fail */ );
      tcp_connection_cleanup (tc);
      break;
    case TCP_STATE_ESTABLISHED:
      tcp_connection_timers_reset (tc);
      /* Set the cleanup timer, in case the session layer/app don't
       * cleanly close the connection */
      tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.closewait_time);
      session_transport_reset_notify (&tc->connection);
      tcp_cong_recovery_off (tc);
      tcp_connection_set_state (tc, TCP_STATE_CLOSED);
      session_transport_closed_notify (&tc->connection);
      break;
    case TCP_STATE_CLOSE_WAIT:
    case TCP_STATE_FIN_WAIT_1:
    case TCP_STATE_FIN_WAIT_2:
    case TCP_STATE_CLOSING:
    case TCP_STATE_LAST_ACK:
      tcp_connection_timers_reset (tc);
      tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.closewait_time);
      tcp_cong_recovery_off (tc);
      /* Make sure we mark the session as closed. In some states we may
       * be still trying to send data */
      tcp_connection_set_state (tc, TCP_STATE_CLOSED);
      session_transport_closed_notify (&tc->connection);
      break;
    case TCP_STATE_CLOSED:
    case TCP_STATE_TIME_WAIT:
      break;
    default:
      TCP_DBG ("reset state: %u", tc->state);
    }
}

/**
 * Begin connection closing procedure.
 *
 * If at the end the connection is not in CLOSED state, it is not removed.
 * Instead, we rely on on TCP to advance through state machine to either
 * 1) LAST_ACK (passive close) whereby when the last ACK is received
 * tcp_connection_del is called. This notifies session of the delete and
 * calls cleanup.
 * 2) TIME_WAIT (active close) whereby after 2MSL the 2MSL timer triggers
 * and cleanup is called.
 *
 * N.B. Half-close connections are not supported
 */
void
tcp_connection_close (tcp_connection_t * tc)
{
  TCP_EVT (TCP_EVT_CLOSE, tc);

  /* Send/Program FIN if needed and switch state */
  switch (tc->state)
    {
    case TCP_STATE_SYN_SENT:
      /* Try to cleanup. If not on the right thread, mark as half-open done.
       * Connection will be cleaned up when establish timer pops */
      tcp_connection_cleanup (tc);
      break;
    case TCP_STATE_SYN_RCVD:
      tcp_connection_timers_reset (tc);
      tcp_send_fin (tc);
      tcp_connection_set_state (tc, TCP_STATE_FIN_WAIT_1);
      tcp_timer_update (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.finwait1_time);
      break;
    case TCP_STATE_ESTABLISHED:
      /* If closing with unread data, reset the connection */
      if (transport_max_rx_dequeue (&tc->connection))
	{
	  tcp_send_reset (tc);
	  tcp_connection_timers_reset (tc);
	  tcp_connection_set_state (tc, TCP_STATE_CLOSED);
	  tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.closewait_time);
	  session_transport_closed_notify (&tc->connection);
	  break;
	}
      if (!transport_max_tx_dequeue (&tc->connection))
	tcp_send_fin (tc);
      else
	tc->flags |= TCP_CONN_FINPNDG;
      tcp_connection_set_state (tc, TCP_STATE_FIN_WAIT_1);
      /* Set a timer in case the peer stops responding. Otherwise the
       * connection will be stuck here forever. */
      ASSERT (tc->timers[TCP_TIMER_WAITCLOSE] == TCP_TIMER_HANDLE_INVALID);
      tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.finwait1_time);
      break;
    case TCP_STATE_CLOSE_WAIT:
      if (!transport_max_tx_dequeue (&tc->connection))
	{
	  tcp_send_fin (tc);
	  tcp_connection_timers_reset (tc);
	  tcp_connection_set_state (tc, TCP_STATE_LAST_ACK);
	  tcp_timer_update (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.lastack_time);
	}
      else
	tc->flags |= TCP_CONN_FINPNDG;
      break;
    case TCP_STATE_FIN_WAIT_1:
      tcp_timer_update (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.finwait1_time);
      break;
    case TCP_STATE_CLOSED:
      tcp_connection_timers_reset (tc);
      /* Delete connection but instead of doing it now wait until next
       * dispatch cycle to give the session layer a chance to clear
       * unhandled events */
      tcp_timer_update (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.cleanup_time);
      break;
    default:
      TCP_DBG ("state: %u", tc->state);
    }
}

static void
tcp_session_close (u32 conn_index, u32 thread_index)
{
  tcp_connection_t *tc;
  tc = tcp_connection_get (conn_index, thread_index);
  tcp_connection_close (tc);
}

static void
tcp_session_cleanup (u32 conn_index, u32 thread_index)
{
  tcp_connection_t *tc;
  tc = tcp_connection_get (conn_index, thread_index);
  if (!tc)
    return;
  tcp_connection_set_state (tc, TCP_STATE_CLOSED);
  tcp_connection_cleanup (tc);
}

static void
tcp_session_reset (u32 conn_index, u32 thread_index)
{
  tcp_connection_t *tc;
  tc = tcp_connection_get (conn_index, thread_index);
  session_transport_closed_notify (&tc->connection);
  tcp_send_reset (tc);
  tcp_connection_timers_reset (tc);
  tcp_cong_recovery_off (tc);
  tcp_connection_set_state (tc, TCP_STATE_CLOSED);
  tcp_timer_update (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.cleanup_time);
}

/**
 * Initialize all connection timers as invalid
 */
void
tcp_connection_timers_init (tcp_connection_t * tc)
{
  int i;

  /* Set all to invalid */
  for (i = 0; i < TCP_N_TIMERS; i++)
    {
      tc->timers[i] = TCP_TIMER_HANDLE_INVALID;
    }

  tc->rto = TCP_RTO_INIT;
}

/**
 * Stop all connection timers
 */
void
tcp_connection_timers_reset (tcp_connection_t * tc)
{
  int i;
  for (i = 0; i < TCP_N_TIMERS; i++)
    {
      tcp_timer_reset (tc, i);
    }
}

#if 0
typedef struct ip4_tcp_hdr
{
  ip4_header_t ip;
  tcp_header_t tcp;
} ip4_tcp_hdr_t;

typedef struct ip6_tcp_hdr
{
  ip6_header_t ip;
  tcp_header_t tcp;
} ip6_tcp_hdr_t;

static void
tcp_connection_select_lb_bucket (tcp_connection_t * tc, const dpo_id_t * dpo,
				 dpo_id_t * result)
{
  const dpo_id_t *choice;
  load_balance_t *lb;
  int hash;

  lb = load_balance_get (dpo->dpoi_index);
  if (tc->c_is_ip4)
    {
      ip4_tcp_hdr_t hdr;
      clib_memset (&hdr, 0, sizeof (hdr));
      hdr.ip.protocol = IP_PROTOCOL_TCP;
      hdr.ip.address_pair.src.as_u32 = tc->c_lcl_ip.ip4.as_u32;
      hdr.ip.address_pair.dst.as_u32 = tc->c_rmt_ip.ip4.as_u32;
      hdr.tcp.src_port = tc->c_lcl_port;
      hdr.tcp.dst_port = tc->c_rmt_port;
      hash = ip4_compute_flow_hash (&hdr.ip, lb->lb_hash_config);
    }
  else
    {
      ip6_tcp_hdr_t hdr;
      clib_memset (&hdr, 0, sizeof (hdr));
      hdr.ip.protocol = IP_PROTOCOL_TCP;
      clib_memcpy_fast (&hdr.ip.src_address, &tc->c_lcl_ip.ip6,
			sizeof (ip6_address_t));
      clib_memcpy_fast (&hdr.ip.dst_address, &tc->c_rmt_ip.ip6,
			sizeof (ip6_address_t));
      hdr.tcp.src_port = tc->c_lcl_port;
      hdr.tcp.dst_port = tc->c_rmt_port;
      hash = ip6_compute_flow_hash (&hdr.ip, lb->lb_hash_config);
    }
  choice = load_balance_get_bucket_i (lb, hash & lb->lb_n_buckets_minus_1);
  dpo_copy (result, choice);
}

fib_node_index_t
tcp_lookup_rmt_in_fib (tcp_connection_t * tc)
{
  fib_prefix_t prefix;
  u32 fib_index;

  clib_memcpy_fast (&prefix.fp_addr, &tc->c_rmt_ip, sizeof (prefix.fp_addr));
  prefix.fp_proto = tc->c_is_ip4 ? FIB_PROTOCOL_IP4 : FIB_PROTOCOL_IP6;
  prefix.fp_len = tc->c_is_ip4 ? 32 : 128;
  fib_index = fib_table_find (prefix.fp_proto, tc->c_fib_index);
  return fib_table_lookup (fib_index, &prefix);
}

static int
tcp_connection_stack_on_fib_entry (tcp_connection_t * tc)
{
  dpo_id_t choice = DPO_INVALID;
  u32 output_node_index;
  fib_entry_t *fe;

  fe = fib_entry_get (tc->c_rmt_fei);
  if (fe->fe_lb.dpoi_type != DPO_LOAD_BALANCE)
    return -1;

  tcp_connection_select_lb_bucket (tc, &fe->fe_lb, &choice);

  output_node_index =
    tc->c_is_ip4 ? tcp4_output_node.index : tcp6_output_node.index;
  dpo_stack_from_node (output_node_index, &tc->c_rmt_dpo, &choice);
  return 0;
}

/** Stack tcp connection on peer's fib entry.
 *
 * This ultimately populates the dpo the connection will use to send packets.
 */
static void
tcp_connection_fib_attach (tcp_connection_t * tc)
{
  tc->c_rmt_fei = tcp_lookup_rmt_in_fib (tc);

  ASSERT (tc->c_rmt_fei != FIB_NODE_INDEX_INVALID);

  tcp_connection_stack_on_fib_entry (tc);
}
#endif /* 0 */

/**
 * Generate random iss as per rfc6528
 */
static u32
tcp_generate_random_iss (tcp_connection_t * tc)
{
  tcp_main_t *tm = &tcp_main;
  u64 tmp;

  if (tc->c_is_ip4)
    tmp = (u64) tc->c_lcl_ip.ip4.as_u32 << 32 | (u64) tc->c_rmt_ip.ip4.as_u32;
  else
    tmp = tc->c_lcl_ip.ip6.as_u64[0] ^ tc->c_lcl_ip.ip6.as_u64[1]
      ^ tc->c_rmt_ip.ip6.as_u64[0] ^ tc->c_rmt_ip.ip6.as_u64[1];

  tmp ^= tm->iss_seed.first | ((u64) tc->c_lcl_port << 16 | tc->c_rmt_port);
  tmp ^= tm->iss_seed.second;
  tmp = clib_xxhash (tmp) + clib_cpu_time_now ();
  return ((tmp >> 32) ^ (tmp & 0xffffffff));
}

/**
 * Initialize max segment size we're able to process.
 *
 * The value is constrained by the output interface's MTU and by the size
 * of the IP and TCP headers (see RFC6691). It is also what we advertise
 * to our peer.
 */
static void
tcp_init_rcv_mss (tcp_connection_t * tc)
{
  u8 ip_hdr_len;

  ip_hdr_len = tc->c_is_ip4 ? sizeof (ip4_header_t) : sizeof (ip6_header_t);
  tc->mss = tcp_cfg.default_mtu - sizeof (tcp_header_t) - ip_hdr_len;
}

static void
tcp_init_mss (tcp_connection_t * tc)
{
  u16 default_min_mss = 536;

  tcp_init_rcv_mss (tc);

  /* TODO consider PMTU discovery */
  tc->snd_mss = clib_min (tc->rcv_opts.mss, tc->mss);

  if (tc->snd_mss < 45)
    {
      /* Assume that at least the min default mss works */
      tc->snd_mss = default_min_mss;
      tc->rcv_opts.mss = default_min_mss;
    }

  /* We should have enough space for 40 bytes of options */
  ASSERT (tc->snd_mss > 45);

  /* If we use timestamp option, account for it */
  if (tcp_opts_tstamp (&tc->rcv_opts))
    tc->snd_mss -= TCP_OPTION_LEN_TIMESTAMP;
}

/**
 * Initialize connection send variables.
 */
void
tcp_init_snd_vars (tcp_connection_t * tc)
{
  /*
   * We use the time to randomize iss and for setting up the initial
   * timestamp. Make sure it's updated otherwise syn and ack in the
   * handshake may make it look as if time has flown in the opposite
   * direction for us.
   */
  tcp_set_time_now (tcp_get_worker (vlib_get_thread_index ()));

  tcp_init_rcv_mss (tc);
  tc->iss = tcp_generate_random_iss (tc);
  tc->snd_una = tc->iss;
  tc->snd_nxt = tc->iss + 1;
  tc->snd_una_max = tc->snd_nxt;
  tc->srtt = 100;		/* 100 ms */

  if (!tcp_cfg.csum_offload)
    tc->cfg_flags |= TCP_CFG_F_NO_CSUM_OFFLOAD;
}

void
tcp_enable_pacing (tcp_connection_t * tc)
{
  u32 byte_rate;
  byte_rate = tc->cwnd / (tc->srtt * TCP_TICK);
  transport_connection_tx_pacer_init (&tc->connection, byte_rate, tc->cwnd);
  tc->mrtt_us = (u32) ~ 0;
}

/** Initialize tcp connection variables
 *
 * Should be called after having received a msg from the peer, i.e., a SYN or
 * a SYNACK, such that connection options have already been exchanged. */
void
tcp_connection_init_vars (tcp_connection_t * tc)
{
  tcp_connection_timers_init (tc);
  tcp_init_mss (tc);
  scoreboard_init (&tc->sack_sb);
  if (tc->state == TCP_STATE_SYN_RCVD)
    tcp_init_snd_vars (tc);

  tcp_cc_init (tc);

  if (!tc->c_is_ip4 && ip6_address_is_link_local_unicast (&tc->c_rmt_ip6))
    tcp_add_del_adjacency (tc, 1);

  /*  tcp_connection_fib_attach (tc); */

  if (transport_connection_is_tx_paced (&tc->connection)
      || tcp_cfg.enable_tx_pacing)
    tcp_enable_pacing (tc);

  if (tc->cfg_flags & TCP_CFG_F_RATE_SAMPLE)
    tcp_bt_init (tc);

  if (!tcp_cfg.allow_tso)
    tc->cfg_flags |= TCP_CFG_F_NO_TSO;

  tc->start_ts = tcp_time_now_us (tc->c_thread_index);
}

static int
tcp_alloc_custom_local_endpoint (tcp_main_t * tm, ip46_address_t * lcl_addr,
				 u16 * lcl_port, u8 is_ip4)
{
  int index, port;
  if (is_ip4)
    {
      index = tm->last_v4_addr_rotor++;
      if (tm->last_v4_addr_rotor >= vec_len (tcp_cfg.ip4_src_addrs))
	tm->last_v4_addr_rotor = 0;
      lcl_addr->ip4.as_u32 = tcp_cfg.ip4_src_addrs[index].as_u32;
    }
  else
    {
      index = tm->last_v6_addr_rotor++;
      if (tm->last_v6_addr_rotor >= vec_len (tcp_cfg.ip6_src_addrs))
	tm->last_v6_addr_rotor = 0;
      clib_memcpy_fast (&lcl_addr->ip6, &tcp_cfg.ip6_src_addrs[index],
			sizeof (ip6_address_t));
    }
  port = transport_alloc_local_port (TRANSPORT_PROTO_TCP, lcl_addr);
  if (port < 1)
    {
      clib_warning ("Failed to allocate src port");
      return -1;
    }
  *lcl_port = port;
  return 0;
}

static int
tcp_session_open (transport_endpoint_cfg_t * rmt)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  tcp_connection_t *tc;
  ip46_address_t lcl_addr;
  u16 lcl_port;
  int rv;

  /*
   * Allocate local endpoint
   */
  if ((rmt->is_ip4 && vec_len (tcp_cfg.ip4_src_addrs))
      || (!rmt->is_ip4 && vec_len (tcp_cfg.ip6_src_addrs)))
    rv = tcp_alloc_custom_local_endpoint (tm, &lcl_addr, &lcl_port,
					  rmt->is_ip4);
  else
    rv = transport_alloc_local_endpoint (TRANSPORT_PROTO_TCP,
					 rmt, &lcl_addr, &lcl_port);

  if (rv)
    return -1;

  /*
   * Create connection and send SYN
   */
  clib_spinlock_lock_if_init (&tm->half_open_lock);
  tc = tcp_half_open_connection_new ();
  ip_copy (&tc->c_rmt_ip, &rmt->ip, rmt->is_ip4);
  ip_copy (&tc->c_lcl_ip, &lcl_addr, rmt->is_ip4);
  tc->c_rmt_port = rmt->port;
  tc->c_lcl_port = clib_host_to_net_u16 (lcl_port);
  tc->c_is_ip4 = rmt->is_ip4;
  tc->c_proto = TRANSPORT_PROTO_TCP;
  tc->c_fib_index = rmt->fib_index;
  tc->cc_algo = tcp_cc_algo_get (tcp_cfg.cc_algo);
  /* The other connection vars will be initialized after SYN ACK */
  tcp_connection_timers_init (tc);

  TCP_EVT (TCP_EVT_OPEN, tc);
  tc->state = TCP_STATE_SYN_SENT;
  tcp_init_snd_vars (tc);
  tcp_send_syn (tc);
  clib_spinlock_unlock_if_init (&tm->half_open_lock);

  return tc->c_c_index;
}

const char *tcp_fsm_states[] = {
#define _(sym, str) str,
  foreach_tcp_fsm_state
#undef _
};

u8 *
format_tcp_state (u8 * s, va_list * args)
{
  u32 state = va_arg (*args, u32);

  if (state < TCP_N_STATES)
    s = format (s, "%s", tcp_fsm_states[state]);
  else
    s = format (s, "UNKNOWN (%d (0x%x))", state, state);
  return s;
}

const char *tcp_cfg_flags_str[] = {
#define _(sym, str) str,
  foreach_tcp_cfg_flag
#undef _
};

static u8 *
format_tcp_cfg_flags (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  int i, last = -1;

  for (i = 0; i < TCP_CFG_N_FLAG_BITS; i++)
    if (tc->cfg_flags & (1 << i))
      last = i;
  for (i = 0; i < last; i++)
    {
      if (tc->cfg_flags & (1 << i))
	s = format (s, "%s, ", tcp_cfg_flags_str[i]);
    }
  if (last >= 0)
    s = format (s, "%s", tcp_cfg_flags_str[last]);
  return s;
}

const char *tcp_connection_flags_str[] = {
#define _(sym, str) str,
  foreach_tcp_connection_flag
#undef _
};

static u8 *
format_tcp_connection_flags (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  int i, last = -1;

  for (i = 0; i < TCP_CONN_N_FLAG_BITS; i++)
    if (tc->flags & (1 << i))
      last = i;
  for (i = 0; i < last; i++)
    {
      if (tc->flags & (1 << i))
	s = format (s, "%s, ", tcp_connection_flags_str[i]);
    }
  if (last >= 0)
    s = format (s, "%s", tcp_connection_flags_str[last]);
  return s;
}

const char *tcp_conn_timers[] = {
#define _(sym, str) str,
  foreach_tcp_timer
#undef _
};

static u8 *
format_tcp_timers (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  int i, last = -1;

  for (i = 0; i < TCP_N_TIMERS; i++)
    if (tc->timers[i] != TCP_TIMER_HANDLE_INVALID)
      last = i;

  for (i = 0; i < last; i++)
    {
      if (tc->timers[i] != TCP_TIMER_HANDLE_INVALID)
	s = format (s, "%s,", tcp_conn_timers[i]);
    }

  if (last >= 0)
    s = format (s, "%s", tcp_conn_timers[i]);

  return s;
}

static u8 *
format_tcp_congestion_status (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  if (tcp_in_recovery (tc))
    s = format (s, "recovery");
  else if (tcp_in_fastrecovery (tc))
    s = format (s, "fastrecovery");
  else
    s = format (s, "none");
  return s;
}

static i32
tcp_rcv_wnd_available (tcp_connection_t * tc)
{
  return (i32) tc->rcv_wnd - (tc->rcv_nxt - tc->rcv_las);
}

static u8 *
format_tcp_congestion (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  u32 indent = format_get_indent (s), prr_space = 0;

  s = format (s, "%U ", format_tcp_congestion_status, tc);
  s = format (s, "algo %s cwnd %u ssthresh %u bytes_acked %u\n",
	      tc->cc_algo->name, tc->cwnd, tc->ssthresh, tc->bytes_acked);
  s = format (s, "%Ucc space %u prev_cwnd %u prev_ssthresh %u\n",
	      format_white_space, indent, tcp_available_cc_snd_space (tc),
	      tc->prev_cwnd, tc->prev_ssthresh);
  s = format (s, "%Usnd_cong %u dupack %u limited_tx %u\n",
	      format_white_space, indent, tc->snd_congestion - tc->iss,
	      tc->rcv_dupacks, tc->limited_transmit - tc->iss);
  s = format (s, "%Urxt_bytes %u rxt_delivered %u rxt_head %u rxt_ts %u\n",
	      format_white_space, indent, tc->snd_rxt_bytes,
	      tc->rxt_delivered, tc->rxt_head - tc->iss,
	      tcp_time_now_w_thread (tc->c_thread_index) - tc->snd_rxt_ts);
  if (tcp_in_fastrecovery (tc))
    prr_space = tcp_fastrecovery_prr_snd_space (tc);
  s = format (s, "%Uprr_start %u prr_delivered %u prr space %u\n",
	      format_white_space, indent, tc->prr_start - tc->iss,
	      tc->prr_delivered, prr_space);
  return s;
}

static u8 *
format_tcp_stats (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  u32 indent = format_get_indent (s);
  s = format (s, "in segs %lu dsegs %lu bytes %lu dupacks %u\n",
	      tc->segs_in, tc->data_segs_in, tc->bytes_in, tc->dupacks_in);
  s = format (s, "%Uout segs %lu dsegs %lu bytes %lu dupacks %u\n",
	      format_white_space, indent, tc->segs_out,
	      tc->data_segs_out, tc->bytes_out, tc->dupacks_out);
  s = format (s, "%Ufr %u tr %u rxt segs %lu bytes %lu duration %.3f\n",
	      format_white_space, indent, tc->fr_occurences,
	      tc->tr_occurences, tc->segs_retrans, tc->bytes_retrans,
	      tcp_time_now_us (tc->c_thread_index) - tc->start_ts);
  s = format (s, "%Uerr wnd data below %u above %u ack below %u above %u",
	      format_white_space, indent, tc->errors.below_data_wnd,
	      tc->errors.above_data_wnd, tc->errors.below_ack_wnd,
	      tc->errors.above_ack_wnd);
  return s;
}

static u8 *
format_tcp_vars (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  s = format (s, " index: %u cfg: %U flags: %U timers: %U\n", tc->c_c_index,
	      format_tcp_cfg_flags, tc, format_tcp_connection_flags, tc,
	      format_tcp_timers, tc);
  s = format (s, " snd_una %u snd_nxt %u snd_una_max %u",
	      tc->snd_una - tc->iss, tc->snd_nxt - tc->iss,
	      tc->snd_una_max - tc->iss);
  s = format (s, " rcv_nxt %u rcv_las %u\n",
	      tc->rcv_nxt - tc->irs, tc->rcv_las - tc->irs);
  s = format (s, " snd_wnd %u rcv_wnd %u rcv_wscale %u ",
	      tc->snd_wnd, tc->rcv_wnd, tc->rcv_wscale);
  s = format (s, "snd_wl1 %u snd_wl2 %u\n", tc->snd_wl1 - tc->irs,
	      tc->snd_wl2 - tc->iss);
  s = format (s, " flight size %u out space %u rcv_wnd_av %u",
	      tcp_flight_size (tc), tcp_available_output_snd_space (tc),
	      tcp_rcv_wnd_available (tc));
  s = format (s, " tsval_recent %u\n", tc->tsval_recent);
  s = format (s, " tsecr %u tsecr_last_ack %u tsval_recent_age %u",
	      tc->rcv_opts.tsecr, tc->tsecr_last_ack,
	      tcp_time_now () - tc->tsval_recent_age);
  s = format (s, " snd_mss %u\n", tc->snd_mss);
  s = format (s, " rto %u rto_boff %u srtt %u us %.3f rttvar %u rtt_ts %.4f",
	      tc->rto, tc->rto_boff, tc->srtt, tc->mrtt_us * 1000, tc->rttvar,
	      tc->rtt_ts);
  s = format (s, " rtt_seq %u\n", tc->rtt_seq - tc->iss);
  s = format (s, " next_node %u opaque 0x%x\n", tc->next_node_index,
	      tc->next_node_opaque);
  s = format (s, " cong:   %U", format_tcp_congestion, tc);

  if (tc->state >= TCP_STATE_ESTABLISHED)
    {
      s = format (s, " sboard: %U\n", format_tcp_scoreboard, &tc->sack_sb,
		  tc);
      s = format (s, " stats: %U\n", format_tcp_stats, tc);
    }
  if (vec_len (tc->snd_sacks))
    s = format (s, " sacks tx: %U\n", format_tcp_sacks, tc);

  return s;
}

static u8 *
format_tcp_connection_id (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  if (!tc)
    return s;
  if (tc->c_is_ip4)
    {
      s = format (s, "[%d:%d][%s] %U:%d->%U:%d", tc->c_thread_index,
		  tc->c_s_index, "T", format_ip4_address, &tc->c_lcl_ip4,
		  clib_net_to_host_u16 (tc->c_lcl_port), format_ip4_address,
		  &tc->c_rmt_ip4, clib_net_to_host_u16 (tc->c_rmt_port));
    }
  else
    {
      s = format (s, "[%d:%d][%s] %U:%d->%U:%d", tc->c_thread_index,
		  tc->c_s_index, "T", format_ip6_address, &tc->c_lcl_ip6,
		  clib_net_to_host_u16 (tc->c_lcl_port), format_ip6_address,
		  &tc->c_rmt_ip6, clib_net_to_host_u16 (tc->c_rmt_port));
    }

  return s;
}

u8 *
format_tcp_connection (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  u32 verbose = va_arg (*args, u32);

  if (!tc)
    return s;
  s = format (s, "%-50U", format_tcp_connection_id, tc);
  if (verbose)
    {
      s = format (s, "%-15U", format_tcp_state, tc->state);
      if (verbose > 1)
	s = format (s, "\n%U", format_tcp_vars, tc);
    }

  return s;
}

static u8 *
format_tcp_session (u8 * s, va_list * args)
{
  u32 tci = va_arg (*args, u32);
  u32 thread_index = va_arg (*args, u32);
  u32 verbose = va_arg (*args, u32);
  tcp_connection_t *tc;

  tc = tcp_connection_get (tci, thread_index);
  if (tc)
    s = format (s, "%U", format_tcp_connection, tc, verbose);
  else
    s = format (s, "empty\n");
  return s;
}

static u8 *
format_tcp_listener_session (u8 * s, va_list * args)
{
  u32 tci = va_arg (*args, u32);
  u32 __clib_unused thread_index = va_arg (*args, u32);
  u32 verbose = va_arg (*args, u32);
  tcp_connection_t *tc = tcp_listener_get (tci);
  s = format (s, "%-50U", format_tcp_connection_id, tc);
  if (verbose)
    s = format (s, "%-15U", format_tcp_state, tc->state);
  return s;
}

static u8 *
format_tcp_half_open_session (u8 * s, va_list * args)
{
  u32 tci = va_arg (*args, u32);
  u32 __clib_unused thread_index = va_arg (*args, u32);
  tcp_connection_t *tc = tcp_half_open_connection_get (tci);
  return format (s, "%U", format_tcp_connection_id, tc);
}

u8 *
format_tcp_sacks (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  sack_block_t *sacks = tc->snd_sacks;
  sack_block_t *block;
  int i, len = 0;

  len = vec_len (sacks);
  for (i = 0; i < len - 1; i++)
    {
      block = &sacks[i];
      s = format (s, " start %u end %u\n", block->start - tc->irs,
		  block->end - tc->irs);
    }
  if (len)
    {
      block = &sacks[len - 1];
      s = format (s, " start %u end %u", block->start - tc->irs,
		  block->end - tc->irs);
    }
  return s;
}

u8 *
format_tcp_rcv_sacks (u8 * s, va_list * args)
{
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  sack_block_t *sacks = tc->rcv_opts.sacks;
  sack_block_t *block;
  int i, len = 0;

  len = vec_len (sacks);
  for (i = 0; i < len - 1; i++)
    {
      block = &sacks[i];
      s = format (s, " start %u end %u\n", block->start - tc->iss,
		  block->end - tc->iss);
    }
  if (len)
    {
      block = &sacks[len - 1];
      s = format (s, " start %u end %u", block->start - tc->iss,
		  block->end - tc->iss);
    }
  return s;
}

static u8 *
format_tcp_sack_hole (u8 * s, va_list * args)
{
  sack_scoreboard_hole_t *hole = va_arg (*args, sack_scoreboard_hole_t *);
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  if (tc)
    s = format (s, "  [%u, %u]", hole->start - tc->iss, hole->end - tc->iss);
  else
    s = format (s, "  [%u, %u]", hole->start, hole->end);
  return s;
}

u8 *
format_tcp_scoreboard (u8 * s, va_list * args)
{
  sack_scoreboard_t *sb = va_arg (*args, sack_scoreboard_t *);
  tcp_connection_t *tc = va_arg (*args, tcp_connection_t *);
  sack_scoreboard_hole_t *hole;
  u32 indent = format_get_indent (s);

  s = format (s, "sacked %u last_sacked %u lost %u last_lost %u"
	      " rxt_sacked %u\n",
	      sb->sacked_bytes, sb->last_sacked_bytes, sb->lost_bytes,
	      sb->last_lost_bytes, sb->rxt_sacked);
  s = format (s, "%Ulast_delivered %u high_sacked %u is_reneging %u\n",
	      format_white_space, indent, sb->last_bytes_delivered,
	      sb->high_sacked - tc->iss, sb->is_reneging);
  s = format (s, "%Ucur_rxt_hole %u high_rxt %u rescue_rxt %u",
	      format_white_space, indent, sb->cur_rxt_hole,
	      sb->high_rxt - tc->iss, sb->rescue_rxt - tc->iss);

  hole = scoreboard_first_hole (sb);
  if (hole)
    s = format (s, "\n%Uhead %u tail %u %u holes:\n%U", format_white_space,
		indent, sb->head, sb->tail, pool_elts (sb->holes),
		format_white_space, indent);

  while (hole)
    {
      s = format (s, "%U", format_tcp_sack_hole, hole, tc);
      hole = scoreboard_next_hole (sb, hole);
    }

  return s;
}

static transport_connection_t *
tcp_session_get_transport (u32 conn_index, u32 thread_index)
{
  tcp_connection_t *tc = tcp_connection_get (conn_index, thread_index);
  if (PREDICT_FALSE (!tc))
    return 0;
  return &tc->connection;
}

static transport_connection_t *
tcp_half_open_session_get_transport (u32 conn_index)
{
  tcp_connection_t *tc = tcp_half_open_connection_get (conn_index);
  return &tc->connection;
}

static u16
tcp_session_cal_goal_size (tcp_connection_t * tc)
{
  u16 goal_size = tc->snd_mss;

  goal_size = TCP_MAX_GSO_SZ - tc->snd_mss % TCP_MAX_GSO_SZ;
  goal_size = clib_min (goal_size, tc->snd_wnd / 2);

  return goal_size > tc->snd_mss ? goal_size : tc->snd_mss;
}

/**
 * Compute maximum segment size for session layer.
 *
 * Since the result needs to be the actual data length, it first computes
 * the tcp options to be used in the next burst and subtracts their
 * length from the connection's snd_mss.
 */
static u16
tcp_session_send_mss (transport_connection_t * trans_conn)
{
  tcp_connection_t *tc = (tcp_connection_t *) trans_conn;

  /* Ensure snd_mss does accurately reflect the amount of data we can push
   * in a segment. This also makes sure that options are updated according to
   * the current state of the connection. */
  tcp_update_burst_snd_vars (tc);

  if (PREDICT_FALSE (tc->cfg_flags & TCP_CFG_F_TSO))
    return tcp_session_cal_goal_size (tc);

  return tc->snd_mss;
}

always_inline u32
tcp_round_snd_space (tcp_connection_t * tc, u32 snd_space)
{
  if (PREDICT_FALSE (tc->snd_wnd < tc->snd_mss))
    {
      return tc->snd_wnd <= snd_space ? tc->snd_wnd : 0;
    }

  /* If not snd_wnd constrained and we can't write at least a segment,
   * don't try at all */
  if (PREDICT_FALSE (snd_space < tc->snd_mss))
    return snd_space < tc->cwnd ? 0 : snd_space;

  /* round down to mss multiple */
  return snd_space - (snd_space % tc->snd_mss);
}

/**
 * Compute tx window session is allowed to fill.
 *
 * Takes into account available send space, snd_mss and the congestion
 * state of the connection. If possible, the value returned is a multiple
 * of snd_mss.
 *
 * @param tc tcp connection
 * @return number of bytes session is allowed to write
 */
static inline u32
tcp_snd_space_inline (tcp_connection_t * tc)
{
  int snd_space;

  if (PREDICT_FALSE (tcp_in_fastrecovery (tc)
		     || tc->state == TCP_STATE_CLOSED))
    return 0;

  snd_space = tcp_available_output_snd_space (tc);

  /* If we got dupacks or sacked bytes but we're not yet in recovery, try
   * to force the peer to send enough dupacks to start retransmitting as
   * per Limited Transmit (RFC3042)
   */
  if (PREDICT_FALSE (tc->rcv_dupacks != 0 || tc->sack_sb.sacked_bytes))
    {
      if (tc->limited_transmit != tc->snd_nxt
	  && (seq_lt (tc->limited_transmit, tc->snd_nxt - 2 * tc->snd_mss)
	      || seq_gt (tc->limited_transmit, tc->snd_nxt)))
	tc->limited_transmit = tc->snd_nxt;

      ASSERT (seq_leq (tc->limited_transmit, tc->snd_nxt));

      int snt_limited = tc->snd_nxt - tc->limited_transmit;
      snd_space = clib_max ((int) 2 * tc->snd_mss - snt_limited, 0);
    }
  return tcp_round_snd_space (tc, snd_space);
}

u32
tcp_snd_space (tcp_connection_t * tc)
{
  return tcp_snd_space_inline (tc);
}

static u32
tcp_session_send_space (transport_connection_t * trans_conn)
{
  tcp_connection_t *tc = (tcp_connection_t *) trans_conn;
  return clib_min (tcp_snd_space_inline (tc),
		   tc->snd_wnd - (tc->snd_nxt - tc->snd_una));
}

static u32
tcp_session_tx_fifo_offset (transport_connection_t * trans_conn)
{
  tcp_connection_t *tc = (tcp_connection_t *) trans_conn;

  ASSERT (seq_geq (tc->snd_nxt, tc->snd_una));

  /* This still works if fast retransmit is on */
  return (tc->snd_nxt - tc->snd_una);
}

static void
tcp_update_time (f64 now, u8 thread_index)
{
  tcp_worker_ctx_t *wrk = tcp_get_worker (thread_index);

  tcp_set_time_now (wrk);
  tw_timer_expire_timers_16t_2w_512sl (&wrk->timer_wheel, now);
  tcp_flush_frames_to_output (wrk);
}

static void
tcp_session_flush_data (transport_connection_t * tconn)
{
  tcp_connection_t *tc = (tcp_connection_t *) tconn;
  if (tc->flags & TCP_CONN_PSH_PENDING)
    return;
  tc->flags |= TCP_CONN_PSH_PENDING;
  tc->psh_seq = tc->snd_una + transport_max_tx_dequeue (tconn) - 1;
}

/* *INDENT-OFF* */
const static transport_proto_vft_t tcp_proto = {
  .enable = vnet_tcp_enable_disable,
  .start_listen = tcp_session_bind,
  .stop_listen = tcp_session_unbind,
  .push_header = tcp_session_push_header,
  .get_connection = tcp_session_get_transport,
  .get_listener = tcp_session_get_listener,
  .get_half_open = tcp_half_open_session_get_transport,
  .connect = tcp_session_open,
  .close = tcp_session_close,
  .cleanup = tcp_session_cleanup,
  .reset = tcp_session_reset,
  .send_mss = tcp_session_send_mss,
  .send_space = tcp_session_send_space,
  .update_time = tcp_update_time,
  .tx_fifo_offset = tcp_session_tx_fifo_offset,
  .flush_data = tcp_session_flush_data,
  .custom_tx = tcp_session_custom_tx,
  .format_connection = format_tcp_session,
  .format_listener = format_tcp_listener_session,
  .format_half_open = format_tcp_half_open_session,
  .transport_options = {
    .tx_type = TRANSPORT_TX_PEEK,
    .service_type = TRANSPORT_SERVICE_VC,
  },
};
/* *INDENT-ON* */

void
tcp_connection_tx_pacer_update (tcp_connection_t * tc)
{
  if (!transport_connection_is_tx_paced (&tc->connection))
    return;

  f64 srtt = clib_min ((f64) tc->srtt * TCP_TICK, tc->mrtt_us);

  transport_connection_tx_pacer_update (&tc->connection,
					tcp_cc_get_pacing_rate (tc),
					srtt * CLIB_US_TIME_FREQ);
}

void
tcp_connection_tx_pacer_reset (tcp_connection_t * tc, u32 window,
			       u32 start_bucket)
{
  f64 srtt = clib_min ((f64) tc->srtt * TCP_TICK, tc->mrtt_us);
  transport_connection_tx_pacer_reset (&tc->connection,
				       tcp_cc_get_pacing_rate (tc),
				       start_bucket,
				       srtt * CLIB_US_TIME_FREQ);
}

static void
tcp_timer_waitclose_handler (u32 conn_index, u32 thread_index)
{
  tcp_connection_t *tc;

  tc = tcp_connection_get (conn_index, thread_index);
  if (!tc)
    return;

  switch (tc->state)
    {
    case TCP_STATE_CLOSE_WAIT:
      tcp_connection_timers_reset (tc);
      session_transport_closed_notify (&tc->connection);

      if (!(tc->flags & TCP_CONN_FINPNDG))
	{
	  tcp_connection_set_state (tc, TCP_STATE_CLOSED);
	  tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.cleanup_time);
	  break;
	}

      /* Session didn't come back with a close. Send FIN either way
       * and switch to LAST_ACK. */
      tcp_cong_recovery_off (tc);
      /* Make sure we don't try to send unsent data */
      tc->snd_nxt = tc->snd_una;
      tcp_send_fin (tc);
      tcp_connection_set_state (tc, TCP_STATE_LAST_ACK);

      /* Make sure we don't wait in LAST ACK forever */
      tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.lastack_time);

      /* Don't delete the connection yet */
      break;
    case TCP_STATE_FIN_WAIT_1:
      tcp_connection_timers_reset (tc);
      session_transport_closed_notify (&tc->connection);
      if (tc->flags & TCP_CONN_FINPNDG)
	{
	  /* If FIN pending, we haven't sent everything, but we did try.
	   * Notify session layer that transport is closed. */
	  tcp_connection_set_state (tc, TCP_STATE_CLOSED);
	  tcp_send_reset (tc);
	  tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.cleanup_time);
	}
      else
	{
	  /* We've sent the fin but no progress. Close the connection and
	   * to make sure everything is flushed, setup a cleanup timer */
	  tcp_connection_set_state (tc, TCP_STATE_CLOSED);
	  tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.cleanup_time);
	}
      break;
    case TCP_STATE_LAST_ACK:
    case TCP_STATE_CLOSING:
      tcp_connection_timers_reset (tc);
      tcp_connection_set_state (tc, TCP_STATE_CLOSED);
      tcp_timer_set (tc, TCP_TIMER_WAITCLOSE, tcp_cfg.cleanup_time);
      session_transport_closed_notify (&tc->connection);
      break;
    default:
      tcp_connection_del (tc);
      break;
    }
}

/* *INDENT-OFF* */
static timer_expiration_handler *timer_expiration_handlers[TCP_N_TIMERS] =
{
    tcp_timer_retransmit_handler,
    tcp_timer_delack_handler,
    tcp_timer_persist_handler,
    tcp_timer_waitclose_handler,
    tcp_timer_retransmit_syn_handler,
};
/* *INDENT-ON* */

static void
tcp_expired_timers_dispatch (u32 * expired_timers)
{
  u32 thread_index = vlib_get_thread_index ();
  u32 connection_index, timer_id;
  tcp_connection_t *tc;
  int i;

  /*
   * Invalidate all timer handles before dispatching. This avoids dangling
   * index references to timer wheel pool entries that have been freed.
   */
  for (i = 0; i < vec_len (expired_timers); i++)
    {
      connection_index = expired_timers[i] & 0x0FFFFFFF;
      timer_id = expired_timers[i] >> 28;

      if (timer_id != TCP_TIMER_RETRANSMIT_SYN)
	tc = tcp_connection_get (connection_index, thread_index);
      else
	tc = tcp_half_open_connection_get (connection_index);

      TCP_EVT (TCP_EVT_TIMER_POP, connection_index, timer_id);

      tc->timers[timer_id] = TCP_TIMER_HANDLE_INVALID;
    }

  /*
   * Dispatch expired timers
   */
  for (i = 0; i < vec_len (expired_timers); i++)
    {
      connection_index = expired_timers[i] & 0x0FFFFFFF;
      timer_id = expired_timers[i] >> 28;
      (*timer_expiration_handlers[timer_id]) (connection_index, thread_index);
    }
}

static void
tcp_initialize_timer_wheels (tcp_main_t * tm)
{
  tw_timer_wheel_16t_2w_512sl_t *tw;
  /* *INDENT-OFF* */
  foreach_vlib_main (({
    tw = &tm->wrk_ctx[ii].timer_wheel;
    tw_timer_wheel_init_16t_2w_512sl (tw, tcp_expired_timers_dispatch,
                                      TCP_TIMER_TICK, ~0);
    tw->last_run_time = vlib_time_now (this_vlib_main);
  }));
  /* *INDENT-ON* */
}

static void
tcp_initialize_iss_seed (tcp_main_t * tm)
{
  u32 default_seed = random_default_seed ();
  u64 time_now = clib_cpu_time_now ();

  tm->iss_seed.first = (u64) random_u32 (&default_seed) << 32;
  tm->iss_seed.second = random_u64 (&time_now);
}

static clib_error_t *
tcp_main_enable (vlib_main_t * vm)
{
  vlib_thread_main_t *vtm = vlib_get_thread_main ();
  u32 num_threads, n_workers, prealloc_conn_per_wrk;
  tcp_connection_t *tc __attribute__ ((unused));
  tcp_main_t *tm = vnet_get_tcp_main ();
  clib_error_t *error = 0;
  int thread;

  if ((error = vlib_call_init_function (vm, ip_main_init)))
    return error;
  if ((error = vlib_call_init_function (vm, ip4_lookup_init)))
    return error;
  if ((error = vlib_call_init_function (vm, ip6_lookup_init)))
    return error;

  /*
   * Registrations
   */

  ip4_register_protocol (IP_PROTOCOL_TCP, tcp4_input_node.index);
  ip6_register_protocol (IP_PROTOCOL_TCP, tcp6_input_node.index);

  /*
   * Initialize data structures
   */

  num_threads = 1 /* main thread */  + vtm->n_threads;
  vec_validate (tm->connections, num_threads - 1);
  vec_validate (tm->wrk_ctx, num_threads - 1);
  n_workers = num_threads == 1 ? 1 : vtm->n_threads;
  prealloc_conn_per_wrk = tcp_cfg.preallocated_connections / n_workers;

  for (thread = 0; thread < num_threads; thread++)
    {
      vec_validate (tm->wrk_ctx[thread].pending_deq_acked, 255);
      vec_validate (tm->wrk_ctx[thread].pending_disconnects, 255);
      vec_reset_length (tm->wrk_ctx[thread].pending_deq_acked);
      vec_reset_length (tm->wrk_ctx[thread].pending_disconnects);
      tm->wrk_ctx[thread].vm = vlib_mains[thread];

      /*
       * Preallocate connections. Assume that thread 0 won't
       * use preallocated threads when running multi-core
       */
      if ((thread > 0 || num_threads == 1) && prealloc_conn_per_wrk)
	pool_init_fixed (tm->connections[thread], prealloc_conn_per_wrk);
    }

  /*
   * Use a preallocated half-open connection pool?
   */
  if (tcp_cfg.preallocated_half_open_connections)
    pool_init_fixed (tm->half_open_connections,
		     tcp_cfg.preallocated_half_open_connections);

  /* Initialize clocks per tick for TCP timestamp. Used to compute
   * monotonically increasing timestamps. */
  tm->tstamp_ticks_per_clock = vm->clib_time.seconds_per_clock
    / TCP_TSTAMP_RESOLUTION;

  if (num_threads > 1)
    {
      clib_spinlock_init (&tm->half_open_lock);
    }

  tcp_initialize_timer_wheels (tm);
  tcp_initialize_iss_seed (tm);

  tm->bytes_per_buffer = vlib_buffer_get_default_data_size (vm);
  tm->cc_last_type = TCP_CC_LAST;
  return error;
}

clib_error_t *
vnet_tcp_enable_disable (vlib_main_t * vm, u8 is_en)
{
  if (is_en)
    {
      if (tcp_main.is_enabled)
	return 0;

      return tcp_main_enable (vm);
    }
  else
    {
      tcp_main.is_enabled = 0;
    }

  return 0;
}

void
tcp_punt_unknown (vlib_main_t * vm, u8 is_ip4, u8 is_add)
{
  tcp_main_t *tm = &tcp_main;
  if (is_ip4)
    tm->punt_unknown4 = is_add;
  else
    tm->punt_unknown6 = is_add;
}

/**
 * Initialize default values for tcp parameters
 */
static void
tcp_configuration_init (void)
{
  /* Initial wnd for SYN. Fifos are not allocated at that point so use some
   * predefined value. For SYN-ACK we still want the scale to be computed in
   * the same way */
  tcp_cfg.max_rx_fifo = 32 << 20;
  tcp_cfg.min_rx_fifo = 4 << 10;

  tcp_cfg.default_mtu = 1500;
  tcp_cfg.initial_cwnd_multiplier = 0;
  tcp_cfg.enable_tx_pacing = 1;
  tcp_cfg.allow_tso = 0;
  tcp_cfg.csum_offload = 1;
  tcp_cfg.cc_algo = TCP_CC_NEWRENO;
  tcp_cfg.rwnd_min_update_ack = 1;

  /* Time constants defined as timer tick (100ms) multiples */
  tcp_cfg.delack_time = 1;	/* 0.1s */
  tcp_cfg.closewait_time = 20;	/* 2s */
  tcp_cfg.timewait_time = 100;	/* 10s */
  tcp_cfg.finwait1_time = 600;	/* 60s */
  tcp_cfg.lastack_time = 300;	/* 30s */
  tcp_cfg.finwait2_time = 300;	/* 30s */
  tcp_cfg.closing_time = 300;	/* 30s */
  tcp_cfg.cleanup_time = 1;	/* 0.1s */
}

static clib_error_t *
tcp_init (vlib_main_t * vm)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  ip_main_t *im = &ip_main;
  ip_protocol_info_t *pi;

  /* Session layer, and by implication tcp, are disabled by default */
  tm->is_enabled = 0;

  /* Register with IP for header parsing */
  pi = ip_get_protocol_info (im, IP_PROTOCOL_TCP);
  if (pi == 0)
    return clib_error_return (0, "TCP protocol info AWOL");
  pi->format_header = format_tcp_header;
  pi->unformat_pg_edit = unformat_pg_tcp_header;

  /* Register as transport with session layer */
  transport_register_protocol (TRANSPORT_PROTO_TCP, &tcp_proto,
			       FIB_PROTOCOL_IP4, tcp4_output_node.index);
  transport_register_protocol (TRANSPORT_PROTO_TCP, &tcp_proto,
			       FIB_PROTOCOL_IP6, tcp6_output_node.index);

  tcp_api_reference ();
  tcp_configuration_init ();

  tm->cc_algo_by_name = hash_create_string (0, sizeof (uword));

  return 0;
}

VLIB_INIT_FUNCTION (tcp_init);

uword
unformat_tcp_cc_algo (unformat_input_t * input, va_list * va)
{
  tcp_cc_algorithm_type_e *result = va_arg (*va, tcp_cc_algorithm_type_e *);
  tcp_main_t *tm = &tcp_main;
  char *cc_algo_name;
  u8 found = 0;
  uword *p;

  if (unformat (input, "%s", &cc_algo_name)
      && ((p = hash_get_mem (tm->cc_algo_by_name, cc_algo_name))))
    {
      *result = *p;
      found = 1;
    }

  vec_free (cc_algo_name);
  return found;
}

uword
unformat_tcp_cc_algo_cfg (unformat_input_t * input, va_list * va)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  tcp_cc_algorithm_t *cc_alg;
  unformat_input_t sub_input;
  int found = 0;

  vec_foreach (cc_alg, tm->cc_algos)
  {
    if (!unformat (input, cc_alg->name))
      continue;

    if (cc_alg->unformat_cfg
	&& unformat (input, "%U", unformat_vlib_cli_sub_input, &sub_input))
      {
	if (cc_alg->unformat_cfg (&sub_input))
	  found = 1;
      }
  }
  return found;
}

static clib_error_t *
tcp_config_fn (vlib_main_t * vm, unformat_input_t * input)
{
  u32 cwnd_multiplier, tmp_time;
  uword memory_size;

  while (unformat_check_input (input) != UNFORMAT_END_OF_INPUT)
    {
      if (unformat (input, "preallocated-connections %d",
		    &tcp_cfg.preallocated_connections))
	;
      else if (unformat (input, "preallocated-half-open-connections %d",
			 &tcp_cfg.preallocated_half_open_connections))
	;
      else if (unformat (input, "buffer-fail-fraction %f",
			 &tcp_cfg.buffer_fail_fraction))
	;
      else if (unformat (input, "max-rx-fifo %U", unformat_memory_size,
			 &memory_size))
	{
	  if (memory_size >= 0x100000000)
	    {
	      return clib_error_return
		(0, "max-rx-fifo %llu (0x%llx) too large", memory_size,
		 memory_size);
	    }
	  tcp_cfg.max_rx_fifo = memory_size;
	}
      else if (unformat (input, "min-rx-fifo %U", unformat_memory_size,
			 &memory_size))
	{
	  if (memory_size >= 0x100000000)
	    {
	      return clib_error_return
		(0, "min-rx-fifo %llu (0x%llx) too large", memory_size,
		 memory_size);
	    }
	  tcp_cfg.min_rx_fifo = memory_size;
	}
      else if (unformat (input, "mtu %u", &tcp_cfg.default_mtu))
	;
      else if (unformat (input, "rwnd-min-update-ack %d",
			 &tcp_cfg.rwnd_min_update_ack))
	;
      else if (unformat (input, "initial-cwnd-multiplier %u",
			 &cwnd_multiplier))
	tcp_cfg.initial_cwnd_multiplier = cwnd_multiplier;
      else if (unformat (input, "no-tx-pacing"))
	tcp_cfg.enable_tx_pacing = 0;
      else if (unformat (input, "tso"))
	tcp_cfg.allow_tso = 1;
      else if (unformat (input, "no-csum-offload"))
	tcp_cfg.csum_offload = 0;
      else if (unformat (input, "cc-algo %U", unformat_tcp_cc_algo,
			 &tcp_cfg.cc_algo))
	;
      else if (unformat (input, "%U", unformat_tcp_cc_algo_cfg))
	;
      else if (unformat (input, "closewait-time %u", &tmp_time))
	tcp_cfg.closewait_time = tmp_time / TCP_TIMER_TICK;
      else if (unformat (input, "timewait-time %u", &tmp_time))
	tcp_cfg.timewait_time = tmp_time / TCP_TIMER_TICK;
      else if (unformat (input, "finwait1-time %u", &tmp_time))
	tcp_cfg.finwait1_time = tmp_time / TCP_TIMER_TICK;
      else if (unformat (input, "finwait2-time %u", &tmp_time))
	tcp_cfg.finwait2_time = tmp_time / TCP_TIMER_TICK;
      else if (unformat (input, "lastack-time %u", &tmp_time))
	tcp_cfg.lastack_time = tmp_time / TCP_TIMER_TICK;
      else if (unformat (input, "closing-time %u", &tmp_time))
	tcp_cfg.closing_time = tmp_time / TCP_TIMER_TICK;
      else if (unformat (input, "cleanup-time %u", &tmp_time))
	tcp_cfg.cleanup_time = tmp_time / TCP_TIMER_TICK;
      else
	return clib_error_return (0, "unknown input `%U'",
				  format_unformat_error, input);
    }
  return 0;
}

VLIB_CONFIG_FUNCTION (tcp_config_fn, "tcp");


/**
 * \brief Configure an ipv4 source address range
 * @param vm vlib_main_t pointer
 * @param start first ipv4 address in the source address range
 * @param end last ipv4 address in the source address range
 * @param table_id VRF / table ID, 0 for the default FIB
 * @return 0 if all OK, else an error indication from api_errno.h
 */

int
tcp_configure_v4_source_address_range (vlib_main_t * vm,
				       ip4_address_t * start,
				       ip4_address_t * end, u32 table_id)
{
  vnet_main_t *vnm = vnet_get_main ();
  u32 start_host_byte_order, end_host_byte_order;
  fib_prefix_t prefix;
  fib_node_index_t fei;
  u32 fib_index = 0;
  u32 sw_if_index;
  int rv;

  clib_memset (&prefix, 0, sizeof (prefix));

  fib_index = fib_table_find (FIB_PROTOCOL_IP4, table_id);

  if (fib_index == ~0)
    return VNET_API_ERROR_NO_SUCH_FIB;

  start_host_byte_order = clib_net_to_host_u32 (start->as_u32);
  end_host_byte_order = clib_net_to_host_u32 (end->as_u32);

  /* sanity check for reversed args or some such */
  if ((end_host_byte_order - start_host_byte_order) > (10 << 10))
    return VNET_API_ERROR_INVALID_ARGUMENT;

  /* Lookup the last address, to identify the interface involved */
  prefix.fp_len = 32;
  prefix.fp_proto = FIB_PROTOCOL_IP4;
  memcpy (&prefix.fp_addr.ip4, end, sizeof (ip4_address_t));

  fei = fib_table_lookup (fib_index, &prefix);

  /* Couldn't find route to destination. Bail out. */
  if (fei == FIB_NODE_INDEX_INVALID)
    return VNET_API_ERROR_NEXT_HOP_NOT_IN_FIB;

  sw_if_index = fib_entry_get_resolving_interface (fei);

  /* Configure proxy arp across the range */
  rv = vnet_proxy_arp_add_del (start, end, fib_index, 0 /* is_del */ );

  if (rv)
    return rv;

  rv = vnet_proxy_arp_enable_disable (vnm, sw_if_index, 1);

  if (rv)
    return rv;

  do
    {
      dpo_id_t dpo = DPO_INVALID;

      vec_add1 (tcp_cfg.ip4_src_addrs, start[0]);

      /* Add local adjacencies for the range */

      receive_dpo_add_or_lock (DPO_PROTO_IP4, ~0 /* sw_if_index */ ,
			       NULL, &dpo);
      prefix.fp_len = 32;
      prefix.fp_proto = FIB_PROTOCOL_IP4;
      prefix.fp_addr.ip4.as_u32 = start->as_u32;

      fib_table_entry_special_dpo_update (fib_index,
					  &prefix,
					  FIB_SOURCE_API,
					  FIB_ENTRY_FLAG_EXCLUSIVE, &dpo);
      dpo_reset (&dpo);

      start_host_byte_order++;
      start->as_u32 = clib_host_to_net_u32 (start_host_byte_order);
    }
  while (start_host_byte_order <= end_host_byte_order);

  return 0;
}

/**
 * \brief Configure an ipv6 source address range
 * @param vm vlib_main_t pointer
 * @param start first ipv6 address in the source address range
 * @param end last ipv6 address in the source address range
 * @param table_id VRF / table ID, 0 for the default FIB
 * @return 0 if all OK, else an error indication from api_errno.h
 */

int
tcp_configure_v6_source_address_range (vlib_main_t * vm,
				       ip6_address_t * start,
				       ip6_address_t * end, u32 table_id)
{
  fib_prefix_t prefix;
  u32 fib_index = 0;
  fib_node_index_t fei;
  u32 sw_if_index;

  clib_memset (&prefix, 0, sizeof (prefix));

  fib_index = fib_table_find (FIB_PROTOCOL_IP6, table_id);

  if (fib_index == ~0)
    return VNET_API_ERROR_NO_SUCH_FIB;

  while (1)
    {
      int i;
      ip6_address_t tmp;
      dpo_id_t dpo = DPO_INVALID;

      /* Remember this address */
      vec_add1 (tcp_cfg.ip6_src_addrs, start[0]);

      /* Lookup the prefix, to identify the interface involved */
      prefix.fp_len = 128;
      prefix.fp_proto = FIB_PROTOCOL_IP6;
      memcpy (&prefix.fp_addr.ip6, start, sizeof (ip6_address_t));

      fei = fib_table_lookup (fib_index, &prefix);

      /* Couldn't find route to destination. Bail out. */
      if (fei == FIB_NODE_INDEX_INVALID)
	return VNET_API_ERROR_NEXT_HOP_NOT_IN_FIB;

      sw_if_index = fib_entry_get_resolving_interface (fei);

      if (sw_if_index == (u32) ~ 0)
	return VNET_API_ERROR_NO_MATCHING_INTERFACE;

      /* Add a proxy neighbor discovery entry for this address */
      ip6_neighbor_proxy_add_del (sw_if_index, start, 0 /* is_del */ );

      /* Add a receive adjacency for this address */
      receive_dpo_add_or_lock (DPO_PROTO_IP6, ~0 /* sw_if_index */ ,
			       NULL, &dpo);

      fib_table_entry_special_dpo_update (fib_index,
					  &prefix,
					  FIB_SOURCE_API,
					  FIB_ENTRY_FLAG_EXCLUSIVE, &dpo);
      dpo_reset (&dpo);

      /* Done with the entire range? */
      if (!memcmp (start, end, sizeof (start[0])))
	break;

      /* Increment the address. DGMS. */
      tmp = start[0];
      for (i = 15; i >= 0; i--)
	{
	  tmp.as_u8[i] += 1;
	  if (tmp.as_u8[i] != 0)
	    break;
	}
      start[0] = tmp;
    }
  return 0;
}

static clib_error_t *
tcp_src_address (vlib_main_t * vm,
		 unformat_input_t * input, vlib_cli_command_t * cmd_arg)
{
  ip4_address_t v4start, v4end;
  ip6_address_t v6start, v6end;
  u32 table_id = 0;
  int v4set = 0;
  int v6set = 0;
  int rv;

  while (unformat_check_input (input) != UNFORMAT_END_OF_INPUT)
    {
      if (unformat (input, "%U - %U", unformat_ip4_address, &v4start,
		    unformat_ip4_address, &v4end))
	v4set = 1;
      else if (unformat (input, "%U", unformat_ip4_address, &v4start))
	{
	  memcpy (&v4end, &v4start, sizeof (v4start));
	  v4set = 1;
	}
      else if (unformat (input, "%U - %U", unformat_ip6_address, &v6start,
			 unformat_ip6_address, &v6end))
	v6set = 1;
      else if (unformat (input, "%U", unformat_ip6_address, &v6start))
	{
	  memcpy (&v6end, &v6start, sizeof (v6start));
	  v6set = 1;
	}
      else if (unformat (input, "fib-table %d", &table_id))
	;
      else
	break;
    }

  if (!v4set && !v6set)
    return clib_error_return (0, "at least one v4 or v6 address required");

  if (v4set)
    {
      rv = tcp_configure_v4_source_address_range (vm, &v4start, &v4end,
						  table_id);
      switch (rv)
	{
	case 0:
	  break;

	case VNET_API_ERROR_NO_SUCH_FIB:
	  return clib_error_return (0, "Invalid table-id %d", table_id);

	case VNET_API_ERROR_INVALID_ARGUMENT:
	  return clib_error_return (0, "Invalid address range %U - %U",
				    format_ip4_address, &v4start,
				    format_ip4_address, &v4end);
	default:
	  return clib_error_return (0, "error %d", rv);
	  break;
	}
    }
  if (v6set)
    {
      rv = tcp_configure_v6_source_address_range (vm, &v6start, &v6end,
						  table_id);
      switch (rv)
	{
	case 0:
	  break;

	case VNET_API_ERROR_NO_SUCH_FIB:
	  return clib_error_return (0, "Invalid table-id %d", table_id);

	default:
	  return clib_error_return (0, "error %d", rv);
	  break;
	}
    }
  return 0;
}

/* *INDENT-OFF* */
VLIB_CLI_COMMAND (tcp_src_address_command, static) =
{
  .path = "tcp src-address",
  .short_help = "tcp src-address <ip-addr> [- <ip-addr>] add src address range",
  .function = tcp_src_address,
};
/* *INDENT-ON* */

static u8 *
tcp_scoreboard_dump_trace (u8 * s, sack_scoreboard_t * sb)
{
#if TCP_SCOREBOARD_TRACE

  scoreboard_trace_elt_t *block;
  int i = 0;

  if (!sb->trace)
    return s;

  s = format (s, "scoreboard trace:");
  vec_foreach (block, sb->trace)
  {
    s = format (s, "{%u, %u, %u, %u, %u}, ", block->start, block->end,
		block->ack, block->snd_una_max, block->group);
    if ((++i % 3) == 0)
      s = format (s, "\n");
  }
  return s;
#else
  return 0;
#endif
}

static clib_error_t *
tcp_show_scoreboard_trace_fn (vlib_main_t * vm, unformat_input_t * input,
			      vlib_cli_command_t * cmd_arg)
{
  transport_connection_t *tconn = 0;
  tcp_connection_t *tc;
  u8 *s = 0;
  while (unformat_check_input (input) != UNFORMAT_END_OF_INPUT)
    {
      if (unformat (input, "%U", unformat_transport_connection, &tconn,
		    TRANSPORT_PROTO_TCP))
	;
      else
	return clib_error_return (0, "unknown input `%U'",
				  format_unformat_error, input);
    }

  if (!TCP_SCOREBOARD_TRACE)
    {
      vlib_cli_output (vm, "scoreboard tracing not enabled");
      return 0;
    }

  tc = tcp_get_connection_from_transport (tconn);
  s = tcp_scoreboard_dump_trace (s, &tc->sack_sb);
  vlib_cli_output (vm, "%v", s);
  return 0;
}

/* *INDENT-OFF* */
VLIB_CLI_COMMAND (tcp_show_scoreboard_trace_command, static) =
{
  .path = "show tcp scoreboard trace",
  .short_help = "show tcp scoreboard trace <connection>",
  .function = tcp_show_scoreboard_trace_fn,
};
/* *INDENT-ON* */

u8 *
tcp_scoreboard_replay (u8 * s, tcp_connection_t * tc, u8 verbose)
{
  int i, trace_len;
  scoreboard_trace_elt_t *trace;
  u32 next_ack, left, group, has_new_ack = 0;
  tcp_connection_t _dummy_tc, *dummy_tc = &_dummy_tc;
  sack_block_t *block;

  if (!TCP_SCOREBOARD_TRACE)
    {
      s = format (s, "scoreboard tracing not enabled");
      return s;
    }

  if (!tc)
    return s;

  clib_memset (dummy_tc, 0, sizeof (*dummy_tc));
  tcp_connection_timers_init (dummy_tc);
  scoreboard_init (&dummy_tc->sack_sb);
  dummy_tc->rcv_opts.flags |= TCP_OPTS_FLAG_SACK;

#if TCP_SCOREBOARD_TRACE
  trace = tc->sack_sb.trace;
  trace_len = vec_len (tc->sack_sb.trace);
#endif

  for (i = 0; i < trace_len; i++)
    {
      if (trace[i].ack != 0)
	{
	  dummy_tc->snd_una = trace[i].ack - 1448;
	  dummy_tc->snd_una_max = trace[i].ack;
	}
    }

  left = 0;
  while (left < trace_len)
    {
      group = trace[left].group;
      vec_reset_length (dummy_tc->rcv_opts.sacks);
      has_new_ack = 0;
      while (trace[left].group == group)
	{
	  if (trace[left].ack != 0)
	    {
	      if (verbose)
		s = format (s, "Adding ack %u, snd_una_max %u, segs: ",
			    trace[left].ack, trace[left].snd_una_max);
	      dummy_tc->snd_una_max = trace[left].snd_una_max;
	      next_ack = trace[left].ack;
	      has_new_ack = 1;
	    }
	  else
	    {
	      if (verbose)
		s = format (s, "[%u, %u], ", trace[left].start,
			    trace[left].end);
	      vec_add2 (dummy_tc->rcv_opts.sacks, block, 1);
	      block->start = trace[left].start;
	      block->end = trace[left].end;
	    }
	  left++;
	}

      /* Push segments */
      tcp_rcv_sacks (dummy_tc, next_ack);
      if (has_new_ack)
	dummy_tc->snd_una = next_ack;

      if (verbose)
	s = format (s, "result: %U", format_tcp_scoreboard,
		    &dummy_tc->sack_sb);

    }
  s = format (s, "result: %U", format_tcp_scoreboard, &dummy_tc->sack_sb);

  return s;
}

static clib_error_t *
tcp_scoreboard_trace_fn (vlib_main_t * vm, unformat_input_t * input,
			 vlib_cli_command_t * cmd_arg)
{
  transport_connection_t *tconn = 0;
  tcp_connection_t *tc = 0;
  u8 *str = 0;
  while (unformat_check_input (input) != UNFORMAT_END_OF_INPUT)
    {
      if (unformat (input, "%U", unformat_transport_connection, &tconn,
		    TRANSPORT_PROTO_TCP))
	;
      else
	return clib_error_return (0, "unknown input `%U'",
				  format_unformat_error, input);
    }

  if (!TCP_SCOREBOARD_TRACE)
    {
      vlib_cli_output (vm, "scoreboard tracing not enabled");
      return 0;
    }

  tc = tcp_get_connection_from_transport (tconn);
  if (!tc)
    {
      vlib_cli_output (vm, "connection not found");
      return 0;
    }
  str = tcp_scoreboard_replay (str, tc, 1);
  vlib_cli_output (vm, "%v", str);
  return 0;
}

/* *INDENT-OFF* */
VLIB_CLI_COMMAND (tcp_replay_scoreboard_command, static) =
{
  .path = "tcp replay scoreboard",
  .short_help = "tcp replay scoreboard <connection>",
  .function = tcp_scoreboard_trace_fn,
};
/* *INDENT-ON* */

static clib_error_t *
show_tcp_punt_fn (vlib_main_t * vm, unformat_input_t * input,
		  vlib_cli_command_t * cmd_arg)
{
  tcp_main_t *tm = vnet_get_tcp_main ();
  if (unformat_check_input (input) != UNFORMAT_END_OF_INPUT)
    return clib_error_return (0, "unknown input `%U'", format_unformat_error,
			      input);
  vlib_cli_output (vm, "IPv4 TCP punt: %s",
		   tm->punt_unknown4 ? "enabled" : "disabled");
  vlib_cli_output (vm, "IPv6 TCP punt: %s",
		   tm->punt_unknown6 ? "enabled" : "disabled");
  return 0;
}
/* *INDENT-OFF* */
VLIB_CLI_COMMAND (show_tcp_punt_command, static) =
{
  .path = "show tcp punt",
  .short_help = "show tcp punt",
  .function = show_tcp_punt_fn,
};
/* *INDENT-ON* */

/*
 * fd.io coding-style-patch-verification: ON
 *
 * Local Variables:
 * eval: (c-set-style "gnu")
 * End:
 */