1#!/bin/bash
2# SPDX-License-Identifier: GPL-2.0
3#
4# Check xfrm policy resolution.  Topology:
5#
6# 1.2   1.1   3.1  3.10    2.1   2.2
7# eth1  eth1 veth0 veth0 eth1   eth1
8# ns1 ---- ns3 ----- ns4 ---- ns2
9#
10# ns3 and ns4 are connected via ipsec tunnel.
11# pings from ns1 to ns2 (and vice versa) are supposed to work like this:
12# ns1: ping 10.0.2.2: passes via ipsec tunnel.
13# ns2: ping 10.0.1.2: passes via ipsec tunnel.
14
15# ns1: ping 10.0.1.253: passes via ipsec tunnel (direct policy)
16# ns2: ping 10.0.2.253: passes via ipsec tunnel (direct policy)
17#
18# ns1: ping 10.0.2.254: does NOT pass via ipsec tunnel (exception)
19# ns2: ping 10.0.1.254: does NOT pass via ipsec tunnel (exception)
20
21# Kselftest framework requirement - SKIP code is 4.
22ksft_skip=4
23ret=0
24policy_checks_ok=1
25
26KEY_SHA=0xdeadbeef1234567890abcdefabcdefabcdefabcd
27KEY_AES=0x0123456789abcdef0123456789012345
28SPI1=0x1
29SPI2=0x2
30
31do_esp_policy() {
32    local ns=$1
33    local me=$2
34    local remote=$3
35    local lnet=$4
36    local rnet=$5
37
38    # to encrypt packets as they go out (includes forwarded packets that need encapsulation)
39    ip -net $ns xfrm policy add src $lnet dst $rnet dir out tmpl src $me dst $remote proto esp mode tunnel priority 100 action allow
40    # to fwd decrypted packets after esp processing:
41    ip -net $ns xfrm policy add src $rnet dst $lnet dir fwd tmpl src $remote dst $me proto esp mode tunnel priority 100 action allow
42}
43
44do_esp() {
45    local ns=$1
46    local me=$2
47    local remote=$3
48    local lnet=$4
49    local rnet=$5
50    local spi_out=$6
51    local spi_in=$7
52
53    ip -net $ns xfrm state add src $remote dst $me proto esp spi $spi_in  enc aes $KEY_AES  auth sha1 $KEY_SHA  mode tunnel sel src $rnet dst $lnet
54    ip -net $ns xfrm state add src $me  dst $remote proto esp spi $spi_out enc aes $KEY_AES auth sha1 $KEY_SHA mode tunnel sel src $lnet dst $rnet
55
56    do_esp_policy $ns $me $remote $lnet $rnet
57}
58
59# add policies with different netmasks, to make sure kernel carries
60# the policies contained within new netmask over when search tree is
61# re-built.
62# peer netns that are supposed to be encapsulated via esp have addresses
63# in the 10.0.1.0/24 and 10.0.2.0/24 subnets, respectively.
64#
65# Adding a policy for '10.0.1.0/23' will make it necessary to
66# alter the prefix of 10.0.1.0 subnet.
67# In case new prefix overlaps with existing node, the node and all
68# policies it carries need to be merged with the existing one(s).
69#
70# Do that here.
71do_overlap()
72{
73    local ns=$1
74
75    # adds new nodes to tree (neither network exists yet in policy database).
76    ip -net $ns xfrm policy add src 10.1.0.0/24 dst 10.0.0.0/24 dir fwd priority 200 action block
77
78    # adds a new node in the 10.0.0.0/24 tree (dst node exists).
79    ip -net $ns xfrm policy add src 10.2.0.0/24 dst 10.0.0.0/24 dir fwd priority 200 action block
80
81    # adds a 10.2.0.0/23 node, but for different dst.
82    ip -net $ns xfrm policy add src 10.2.0.0/23 dst 10.0.1.0/24 dir fwd priority 200 action block
83
84    # dst now overlaps with the 10.0.1.0/24 ESP policy in fwd.
85    # kernel must 'promote' existing one (10.0.0.0/24) to 10.0.0.0/23.
86    # But 10.0.0.0/23 also includes existing 10.0.1.0/24, so that node
87    # also has to be merged too, including source-sorted subtrees.
88    # old:
89    # 10.0.0.0/24 (node 1 in dst tree of the bin)
90    #    10.1.0.0/24 (node in src tree of dst node 1)
91    #    10.2.0.0/24 (node in src tree of dst node 1)
92    # 10.0.1.0/24 (node 2 in dst tree of the bin)
93    #    10.0.2.0/24 (node in src tree of dst node 2)
94    #    10.2.0.0/24 (node in src tree of dst node 2)
95    #
96    # The next 'policy add' adds dst '10.0.0.0/23', which means
97    # that dst node 1 and dst node 2 have to be merged including
98    # the sub-tree.  As no duplicates are allowed, policies in
99    # the two '10.0.2.0/24' are also merged.
100    #
101    # after the 'add', internal search tree should look like this:
102    # 10.0.0.0/23 (node in dst tree of bin)
103    #     10.0.2.0/24 (node in src tree of dst node)
104    #     10.1.0.0/24 (node in src tree of dst node)
105    #     10.2.0.0/24 (node in src tree of dst node)
106    #
107    # 10.0.0.0/24 and 10.0.1.0/24 nodes have been merged as 10.0.0.0/23.
108    ip -net $ns xfrm policy add src 10.1.0.0/24 dst 10.0.0.0/23 dir fwd priority 200 action block
109
110    # similar to above: add policies (with partially random address), with shrinking prefixes.
111    for p in 29 28 27;do
112      for k in $(seq 1 32); do
113       ip -net $ns xfrm policy add src 10.253.1.$((RANDOM%255))/$p dst 10.254.1.$((RANDOM%255))/$p dir fwd priority $((200+k)) action block 2>/dev/null
114      done
115    done
116}
117
118do_esp_policy_get_check() {
119    local ns=$1
120    local lnet=$2
121    local rnet=$3
122
123    ip -net $ns xfrm policy get src $lnet dst $rnet dir out > /dev/null
124    if [ $? -ne 0 ] && [ $policy_checks_ok -eq 1 ] ;then
125        policy_checks_ok=0
126        echo "FAIL: ip -net $ns xfrm policy get src $lnet dst $rnet dir out"
127        ret=1
128    fi
129
130    ip -net $ns xfrm policy get src $rnet dst $lnet dir fwd > /dev/null
131    if [ $? -ne 0 ] && [ $policy_checks_ok -eq 1 ] ;then
132        policy_checks_ok=0
133        echo "FAIL: ip -net $ns xfrm policy get src $rnet dst $lnet dir fwd"
134        ret=1
135    fi
136}
137
138do_exception() {
139    local ns=$1
140    local me=$2
141    local remote=$3
142    local encryptip=$4
143    local plain=$5
144
145    # network $plain passes without tunnel
146    ip -net $ns xfrm policy add dst $plain dir out priority 10 action allow
147
148    # direct policy for $encryptip, use tunnel, higher prio takes precedence
149    ip -net $ns xfrm policy add dst $encryptip dir out tmpl src $me dst $remote proto esp mode tunnel priority 1 action allow
150}
151
152# policies that are not supposed to match any packets generated in this test.
153do_dummies4() {
154    local ns=$1
155
156    for i in $(seq 10 16);do
157      # dummy policy with wildcard src/dst.
158      echo netns exec $ns ip xfrm policy add src 0.0.0.0/0 dst 10.$i.99.0/30 dir out action block
159      echo netns exec $ns ip xfrm policy add src 10.$i.99.0/30 dst 0.0.0.0/0 dir out action block
160      for j in $(seq 32 64);do
161        echo netns exec $ns ip xfrm policy add src 10.$i.1.0/30 dst 10.$i.$j.0/30 dir out action block
162        # silly, as it encompasses the one above too, but its allowed:
163        echo netns exec $ns ip xfrm policy add src 10.$i.1.0/29 dst 10.$i.$j.0/29 dir out action block
164        # and yet again, even more broad one.
165        echo netns exec $ns ip xfrm policy add src 10.$i.1.0/24 dst 10.$i.$j.0/24 dir out action block
166        echo netns exec $ns ip xfrm policy add src 10.$i.$j.0/24 dst 10.$i.1.0/24 dir fwd action block
167      done
168    done | ip -batch /dev/stdin
169}
170
171do_dummies6() {
172    local ns=$1
173
174    for i in $(seq 10 16);do
175      for j in $(seq 32 64);do
176       echo netns exec $ns ip xfrm policy add src dead:$i::/64 dst dead:$i:$j::/64 dir out action block
177       echo netns exec $ns ip xfrm policy add src dead:$i:$j::/64 dst dead:$i::/24 dir fwd action block
178      done
179    done | ip -batch /dev/stdin
180}
181
182check_ipt_policy_count()
183{
184	ns=$1
185
186	ip netns exec $ns iptables-save -c |grep policy | ( read c rest
187		ip netns exec $ns iptables -Z
188		if [ x"$c" = x'[0:0]' ]; then
189			exit 0
190		elif [ x"$c" = x ]; then
191			echo "ERROR: No counters"
192			ret=1
193			exit 111
194		else
195			exit 1
196		fi
197	)
198}
199
200check_xfrm() {
201	# 0: iptables -m policy rule count == 0
202	# 1: iptables -m policy rule count != 0
203	rval=$1
204	ip=$2
205	local lret=0
206
207	ip netns exec ns1 ping -q -c 1 10.0.2.$ip > /dev/null
208
209	check_ipt_policy_count ns3
210	if [ $? -ne $rval ] ; then
211		lret=1
212	fi
213	check_ipt_policy_count ns4
214	if [ $? -ne $rval ] ; then
215		lret=1
216	fi
217
218	ip netns exec ns2 ping -q -c 1 10.0.1.$ip > /dev/null
219
220	check_ipt_policy_count ns3
221	if [ $? -ne $rval ] ; then
222		lret=1
223	fi
224	check_ipt_policy_count ns4
225	if [ $? -ne $rval ] ; then
226		lret=1
227	fi
228
229	return $lret
230}
231
232check_exceptions()
233{
234	logpostfix="$1"
235	local lret=0
236
237	# ping to .254 should be excluded from the tunnel (exception is in place).
238	check_xfrm 0 254
239	if [ $? -ne 0 ]; then
240		echo "FAIL: expected ping to .254 to fail ($logpostfix)"
241		lret=1
242	else
243		echo "PASS: ping to .254 bypassed ipsec tunnel ($logpostfix)"
244	fi
245
246	# ping to .253 should use use ipsec due to direct policy exception.
247	check_xfrm 1 253
248	if [ $? -ne 0 ]; then
249		echo "FAIL: expected ping to .253 to use ipsec tunnel ($logpostfix)"
250		lret=1
251	else
252		echo "PASS: direct policy matches ($logpostfix)"
253	fi
254
255	# ping to .2 should use ipsec.
256	check_xfrm 1 2
257	if [ $? -ne 0 ]; then
258		echo "FAIL: expected ping to .2 to use ipsec tunnel ($logpostfix)"
259		lret=1
260	else
261		echo "PASS: policy matches ($logpostfix)"
262	fi
263
264	return $lret
265}
266
267check_hthresh_repeat()
268{
269	local log=$1
270	i=0
271
272	for i in $(seq 1 10);do
273		ip -net ns1 xfrm policy update src e000:0001::0000 dst ff01::0014:0000:0001 dir in tmpl src :: dst :: proto esp mode tunnel priority 100 action allow || break
274		ip -net ns1 xfrm policy set hthresh6 0 28 || break
275
276		ip -net ns1 xfrm policy update src e000:0001::0000 dst ff01::01 dir in tmpl src :: dst :: proto esp mode tunnel priority 100 action allow || break
277		ip -net ns1 xfrm policy set hthresh6 0 28 || break
278	done
279
280	if [ $i -ne 10 ] ;then
281		echo "FAIL: $log" 1>&2
282		ret=1
283		return 1
284	fi
285
286	echo "PASS: $log"
287	return 0
288}
289
290# insert non-overlapping policies in a random order and check that
291# all of them can be fetched using the traffic selectors.
292check_random_order()
293{
294	local ns=$1
295	local log=$2
296
297	for i in $(seq 100); do
298		ip -net $ns xfrm policy flush
299		for j in $(seq 0 16 255 | sort -R); do
300			ip -net $ns xfrm policy add dst $j.0.0.0/24 dir out priority 10 action allow
301		done
302		for j in $(seq 0 16 255); do
303			if ! ip -net $ns xfrm policy get dst $j.0.0.0/24 dir out > /dev/null; then
304				echo "FAIL: $log" 1>&2
305				return 1
306			fi
307		done
308	done
309
310	for i in $(seq 100); do
311		ip -net $ns xfrm policy flush
312		for j in $(seq 0 16 255 | sort -R); do
313			local addr=$(printf "e000:0000:%02x00::/56" $j)
314			ip -net $ns xfrm policy add dst $addr dir out priority 10 action allow
315		done
316		for j in $(seq 0 16 255); do
317			local addr=$(printf "e000:0000:%02x00::/56" $j)
318			if ! ip -net $ns xfrm policy get dst $addr dir out > /dev/null; then
319				echo "FAIL: $log" 1>&2
320				return 1
321			fi
322		done
323	done
324
325	ip -net $ns xfrm policy flush
326
327	echo "PASS: $log"
328	return 0
329}
330
331#check for needed privileges
332if [ "$(id -u)" -ne 0 ];then
333	echo "SKIP: Need root privileges"
334	exit $ksft_skip
335fi
336
337ip -Version 2>/dev/null >/dev/null
338if [ $? -ne 0 ];then
339	echo "SKIP: Could not run test without the ip tool"
340	exit $ksft_skip
341fi
342
343# needed to check if policy lookup got valid ipsec result
344iptables --version 2>/dev/null >/dev/null
345if [ $? -ne 0 ];then
346	echo "SKIP: Could not run test without iptables tool"
347	exit $ksft_skip
348fi
349
350for i in 1 2 3 4; do
351    ip netns add ns$i
352    ip -net ns$i link set lo up
353done
354
355DEV=veth0
356ip link add $DEV netns ns1 type veth peer name eth1 netns ns3
357ip link add $DEV netns ns2 type veth peer name eth1 netns ns4
358
359ip link add $DEV netns ns3 type veth peer name veth0 netns ns4
360
361DEV=veth0
362for i in 1 2; do
363    ip -net ns$i link set $DEV up
364    ip -net ns$i addr add 10.0.$i.2/24 dev $DEV
365    ip -net ns$i addr add dead:$i::2/64 dev $DEV
366
367    ip -net ns$i addr add 10.0.$i.253 dev $DEV
368    ip -net ns$i addr add 10.0.$i.254 dev $DEV
369    ip -net ns$i addr add dead:$i::fd dev $DEV
370    ip -net ns$i addr add dead:$i::fe dev $DEV
371done
372
373for i in 3 4; do
374ip -net ns$i link set eth1 up
375ip -net ns$i link set veth0 up
376done
377
378ip -net ns1 route add default via 10.0.1.1
379ip -net ns2 route add default via 10.0.2.1
380
381ip -net ns3 addr add 10.0.1.1/24 dev eth1
382ip -net ns3 addr add 10.0.3.1/24 dev veth0
383ip -net ns3 addr add 2001:1::1/64 dev eth1
384ip -net ns3 addr add 2001:3::1/64 dev veth0
385
386ip -net ns3 route add default via 10.0.3.10
387
388ip -net ns4 addr add 10.0.2.1/24 dev eth1
389ip -net ns4 addr add 10.0.3.10/24 dev veth0
390ip -net ns4 addr add 2001:2::1/64 dev eth1
391ip -net ns4 addr add 2001:3::10/64 dev veth0
392ip -net ns4 route add default via 10.0.3.1
393
394for j in 4 6; do
395	for i in 3 4;do
396		ip netns exec ns$i sysctl net.ipv$j.conf.eth1.forwarding=1 > /dev/null
397		ip netns exec ns$i sysctl net.ipv$j.conf.veth0.forwarding=1 > /dev/null
398	done
399done
400
401# abuse iptables rule counter to check if ping matches a policy
402ip netns exec ns3 iptables -p icmp -A FORWARD -m policy --dir out --pol ipsec
403ip netns exec ns4 iptables -p icmp -A FORWARD -m policy --dir out --pol ipsec
404if [ $? -ne 0 ];then
405	echo "SKIP: Could not insert iptables rule"
406	for i in 1 2 3 4;do ip netns del ns$i;done
407	exit $ksft_skip
408fi
409
410#          localip  remoteip  localnet    remotenet
411do_esp ns3 10.0.3.1 10.0.3.10 10.0.1.0/24 10.0.2.0/24 $SPI1 $SPI2
412do_esp ns3 dead:3::1 dead:3::10 dead:1::/64 dead:2::/64 $SPI1 $SPI2
413do_esp ns4 10.0.3.10 10.0.3.1 10.0.2.0/24 10.0.1.0/24 $SPI2 $SPI1
414do_esp ns4 dead:3::10 dead:3::1 dead:2::/64 dead:1::/64 $SPI2 $SPI1
415
416do_dummies4 ns3
417do_dummies6 ns4
418
419do_esp_policy_get_check ns3 10.0.1.0/24 10.0.2.0/24
420do_esp_policy_get_check ns4 10.0.2.0/24 10.0.1.0/24
421do_esp_policy_get_check ns3 dead:1::/64 dead:2::/64
422do_esp_policy_get_check ns4 dead:2::/64 dead:1::/64
423
424# ping to .254 should use ipsec, exception is not installed.
425check_xfrm 1 254
426if [ $? -ne 0 ]; then
427	echo "FAIL: expected ping to .254 to use ipsec tunnel"
428	ret=1
429else
430	echo "PASS: policy before exception matches"
431fi
432
433# installs exceptions
434#                localip  remoteip   encryptdst  plaindst
435do_exception ns3 10.0.3.1 10.0.3.10 10.0.2.253 10.0.2.240/28
436do_exception ns4 10.0.3.10 10.0.3.1 10.0.1.253 10.0.1.240/28
437
438do_exception ns3 dead:3::1 dead:3::10 dead:2::fd  dead:2:f0::/96
439do_exception ns4 dead:3::10 dead:3::1 dead:1::fd  dead:1:f0::/96
440
441check_exceptions "exceptions"
442if [ $? -ne 0 ]; then
443	ret=1
444fi
445
446# insert block policies with adjacent/overlapping netmasks
447do_overlap ns3
448
449check_exceptions "exceptions and block policies"
450if [ $? -ne 0 ]; then
451	ret=1
452fi
453
454for n in ns3 ns4;do
455	ip -net $n xfrm policy set hthresh4 28 24 hthresh6 126 125
456	sleep $((RANDOM%5))
457done
458
459check_exceptions "exceptions and block policies after hresh changes"
460
461# full flush of policy db, check everything gets freed incl. internal meta data
462ip -net ns3 xfrm policy flush
463
464do_esp_policy ns3 10.0.3.1 10.0.3.10 10.0.1.0/24 10.0.2.0/24
465do_exception ns3 10.0.3.1 10.0.3.10 10.0.2.253 10.0.2.240/28
466
467# move inexact policies to hash table
468ip -net ns3 xfrm policy set hthresh4 16 16
469
470sleep $((RANDOM%5))
471check_exceptions "exceptions and block policies after hthresh change in ns3"
472
473# restore original hthresh settings -- move policies back to tables
474for n in ns3 ns4;do
475	ip -net $n xfrm policy set hthresh4 32 32 hthresh6 128 128
476	sleep $((RANDOM%5))
477done
478check_exceptions "exceptions and block policies after htresh change to normal"
479
480check_hthresh_repeat "policies with repeated htresh change"
481
482check_random_order ns3 "policies inserted in random order"
483
484for i in 1 2 3 4;do ip netns del ns$i;done
485
486exit $ret
487