1#!/bin/bash 2# SPDX-License-Identifier: GPL-2.0 3# 4# Check that route PMTU values match expectations, and that initial device MTU 5# values are assigned correctly 6# 7# Tests currently implemented: 8# 9# - pmtu_ipv4 10# Set up two namespaces, A and B, with two paths between them over routers 11# R1 and R2 (also implemented with namespaces), with different MTUs: 12# 13# segment a_r1 segment b_r1 a_r1: 2000 14# .--------------R1--------------. b_r1: 1400 15# A B a_r2: 2000 16# '--------------R2--------------' b_r2: 1500 17# segment a_r2 segment b_r2 18# 19# Check that PMTU exceptions with the correct PMTU are created. Then 20# decrease and increase the MTU of the local link for one of the paths, 21# A to R1, checking that route exception PMTU changes accordingly over 22# this path. Also check that locked exceptions are created when an ICMP 23# message advertising a PMTU smaller than net.ipv4.route.min_pmtu is 24# received 25# 26# - pmtu_ipv6 27# Same as pmtu_ipv4, except for locked PMTU tests, using IPv6 28# 29# - pmtu_ipv4_dscp_icmp_exception 30# Set up the same network topology as pmtu_ipv4, but use non-default 31# routing table in A. A fib-rule is used to jump to this routing table 32# based on DSCP. Send ICMPv4 packets with the expected DSCP value and 33# verify that ECN doesn't interfere with the creation of PMTU exceptions. 34# 35# - pmtu_ipv4_dscp_udp_exception 36# Same as pmtu_ipv4_dscp_icmp_exception, but use UDP instead of ICMP. 37# 38# - pmtu_ipv4_vxlan4_exception 39# Set up the same network topology as pmtu_ipv4, create a VXLAN tunnel 40# over IPv4 between A and B, routed via R1. On the link between R1 and B, 41# set a MTU lower than the VXLAN MTU and the MTU on the link between A and 42# R1. Send IPv4 packets, exceeding the MTU between R1 and B, over VXLAN 43# from A to B and check that the PMTU exception is created with the right 44# value on A 45# 46# - pmtu_ipv6_vxlan4_exception 47# Same as pmtu_ipv4_vxlan4_exception, but send IPv6 packets from A to B 48# 49# - pmtu_ipv4_vxlan6_exception 50# Same as pmtu_ipv4_vxlan4_exception, but use IPv6 transport from A to B 51# 52# - pmtu_ipv6_vxlan6_exception 53# Same as pmtu_ipv4_vxlan6_exception, but send IPv6 packets from A to B 54# 55# - pmtu_ipv4_geneve4_exception 56# Same as pmtu_ipv4_vxlan4_exception, but using a GENEVE tunnel instead of 57# VXLAN 58# 59# - pmtu_ipv6_geneve4_exception 60# Same as pmtu_ipv6_vxlan4_exception, but using a GENEVE tunnel instead of 61# VXLAN 62# 63# - pmtu_ipv4_geneve6_exception 64# Same as pmtu_ipv4_vxlan6_exception, but using a GENEVE tunnel instead of 65# VXLAN 66# 67# - pmtu_ipv6_geneve6_exception 68# Same as pmtu_ipv6_vxlan6_exception, but using a GENEVE tunnel instead of 69# VXLAN 70# 71# - pmtu_ipv{4,6}_br_vxlan{4,6}_exception 72# Set up three namespaces, A, B, and C, with routing between A and B over 73# R1. R2 is unused in these tests. A has a veth connection to C, and is 74# connected to B via a VXLAN endpoint, which is directly bridged to C. 75# MTU on the B-R1 link is lower than other MTUs. 76# 77# Check that both C and A are able to communicate with B over the VXLAN 78# tunnel, and that PMTU exceptions with the correct values are created. 79# 80# segment a_r1 segment b_r1 b_r1: 4000 81# .--------------R1--------------. everything 82# C---veth A B else: 5000 83# ' bridge | 84# '---- - - - - - VXLAN - - - - - - - ' 85# 86# - pmtu_ipv{4,6}_br_geneve{4,6}_exception 87# Same as pmtu_ipv{4,6}_br_vxlan{4,6}_exception, with a GENEVE tunnel 88# instead. 89# 90# - pmtu_ipv{4,6}_ovs_vxlan{4,6}_exception 91# Set up two namespaces, B, and C, with routing between the init namespace 92# and B over R1. A and R2 are unused in these tests. The init namespace 93# has a veth connection to C, and is connected to B via a VXLAN endpoint, 94# which is handled by Open vSwitch and bridged to C. MTU on the B-R1 link 95# is lower than other MTUs. 96# 97# Check that C is able to communicate with B over the VXLAN tunnel, and 98# that PMTU exceptions with the correct values are created. 99# 100# segment a_r1 segment b_r1 b_r1: 4000 101# .--------------R1--------------. everything 102# C---veth init B else: 5000 103# '- ovs | 104# '---- - - - - - VXLAN - - - - - - - ' 105# 106# - pmtu_ipv{4,6}_ovs_geneve{4,6}_exception 107# Same as pmtu_ipv{4,6}_ovs_vxlan{4,6}_exception, with a GENEVE tunnel 108# instead. 109# 110# - pmtu_ipv{4,6}_fou{4,6}_exception 111# Same as pmtu_ipv4_vxlan4, but using a direct IPv4/IPv6 encapsulation 112# (FoU) over IPv4/IPv6, instead of VXLAN 113# 114# - pmtu_ipv{4,6}_fou{4,6}_exception 115# Same as pmtu_ipv4_vxlan4, but using a generic UDP IPv4/IPv6 116# encapsulation (GUE) over IPv4/IPv6, instead of VXLAN 117# 118# - pmtu_ipv{4,6}_ipv{4,6}_exception 119# Same as pmtu_ipv4_vxlan4, but using a IPv4/IPv6 tunnel over IPv4/IPv6, 120# instead of VXLAN 121# 122# - pmtu_vti4_exception 123# Set up vti tunnel on top of veth, with xfrm states and policies, in two 124# namespaces with matching endpoints. Check that route exception is not 125# created if link layer MTU is not exceeded, then exceed it and check that 126# exception is created with the expected PMTU. The approach described 127# below for IPv6 doesn't apply here, because, on IPv4, administrative MTU 128# changes alone won't affect PMTU 129# 130# - pmtu_vti4_udp_exception 131# Same as pmtu_vti4_exception, but using ESP-in-UDP 132# 133# - pmtu_vti4_udp_routed_exception 134# Set up vti tunnel on top of veth connected through routing namespace and 135# add xfrm states and policies with ESP-in-UDP encapsulation. Check that 136# route exception is not created if link layer MTU is not exceeded, then 137# lower MTU on second part of routed environment and check that exception 138# is created with the expected PMTU. 139# 140# - pmtu_vti6_exception 141# Set up vti6 tunnel on top of veth, with xfrm states and policies, in two 142# namespaces with matching endpoints. Check that route exception is 143# created by exceeding link layer MTU with ping to other endpoint. Then 144# decrease and increase MTU of tunnel, checking that route exception PMTU 145# changes accordingly 146# 147# - pmtu_vti6_udp_exception 148# Same as pmtu_vti6_exception, but using ESP-in-UDP 149# 150# - pmtu_vti6_udp_routed_exception 151# Same as pmtu_vti6_udp_routed_exception but with routing between vti 152# endpoints 153# 154# - pmtu_vti4_default_mtu 155# Set up vti4 tunnel on top of veth, in two namespaces with matching 156# endpoints. Check that MTU assigned to vti interface is the MTU of the 157# lower layer (veth) minus additional lower layer headers (zero, for veth) 158# minus IPv4 header length 159# 160# - pmtu_vti6_default_mtu 161# Same as above, for IPv6 162# 163# - pmtu_vti4_link_add_mtu 164# Set up vti4 interface passing MTU value at link creation, check MTU is 165# configured, and that link is not created with invalid MTU values 166# 167# - pmtu_vti6_link_add_mtu 168# Same as above, for IPv6 169# 170# - pmtu_vti6_link_change_mtu 171# Set up two dummy interfaces with different MTUs, create a vti6 tunnel 172# and check that configured MTU is used on link creation and changes, and 173# that MTU is properly calculated instead when MTU is not configured from 174# userspace 175# 176# - cleanup_ipv4_exception 177# Similar to pmtu_ipv4_vxlan4_exception, but explicitly generate PMTU 178# exceptions on multiple CPUs and check that the veth device tear-down 179# happens in a timely manner 180# 181# - cleanup_ipv6_exception 182# Same as above, but use IPv6 transport from A to B 183# 184# - list_flush_ipv4_exception 185# Using the same topology as in pmtu_ipv4, create exceptions, and check 186# they are shown when listing exception caches, gone after flushing them 187# 188# - list_flush_ipv6_exception 189# Using the same topology as in pmtu_ipv6, create exceptions, and check 190# they are shown when listing exception caches, gone after flushing them 191# 192# - pmtu_ipv4_route_change 193# Use the same topology as in pmtu_ipv4, but issue a route replacement 194# command and delete the corresponding device afterward. This tests for 195# proper cleanup of the PMTU exceptions by the route replacement path. 196# Device unregistration should complete successfully 197# 198# - pmtu_ipv6_route_change 199# Same as above but with IPv6 200 201source lib.sh 202source net_helper.sh 203 204PAUSE_ON_FAIL=no 205VERBOSE=0 206TRACING=0 207 208# Some systems don't have a ping6 binary anymore 209which ping6 > /dev/null 2>&1 && ping6=$(which ping6) || ping6=$(which ping) 210 211# Name Description re-run with nh 212tests=" 213 pmtu_ipv4_exception ipv4: PMTU exceptions 1 214 pmtu_ipv6_exception ipv6: PMTU exceptions 1 215 pmtu_ipv4_dscp_icmp_exception ICMPv4 with DSCP and ECN: PMTU exceptions 1 216 pmtu_ipv4_dscp_udp_exception UDPv4 with DSCP and ECN: PMTU exceptions 1 217 pmtu_ipv4_vxlan4_exception IPv4 over vxlan4: PMTU exceptions 1 218 pmtu_ipv6_vxlan4_exception IPv6 over vxlan4: PMTU exceptions 1 219 pmtu_ipv4_vxlan6_exception IPv4 over vxlan6: PMTU exceptions 1 220 pmtu_ipv6_vxlan6_exception IPv6 over vxlan6: PMTU exceptions 1 221 pmtu_ipv4_geneve4_exception IPv4 over geneve4: PMTU exceptions 1 222 pmtu_ipv6_geneve4_exception IPv6 over geneve4: PMTU exceptions 1 223 pmtu_ipv4_geneve6_exception IPv4 over geneve6: PMTU exceptions 1 224 pmtu_ipv6_geneve6_exception IPv6 over geneve6: PMTU exceptions 1 225 pmtu_ipv4_br_vxlan4_exception IPv4, bridged vxlan4: PMTU exceptions 1 226 pmtu_ipv6_br_vxlan4_exception IPv6, bridged vxlan4: PMTU exceptions 1 227 pmtu_ipv4_br_vxlan6_exception IPv4, bridged vxlan6: PMTU exceptions 1 228 pmtu_ipv6_br_vxlan6_exception IPv6, bridged vxlan6: PMTU exceptions 1 229 pmtu_ipv4_br_geneve4_exception IPv4, bridged geneve4: PMTU exceptions 1 230 pmtu_ipv6_br_geneve4_exception IPv6, bridged geneve4: PMTU exceptions 1 231 pmtu_ipv4_br_geneve6_exception IPv4, bridged geneve6: PMTU exceptions 1 232 pmtu_ipv6_br_geneve6_exception IPv6, bridged geneve6: PMTU exceptions 1 233 pmtu_ipv4_ovs_vxlan4_exception IPv4, OVS vxlan4: PMTU exceptions 1 234 pmtu_ipv6_ovs_vxlan4_exception IPv6, OVS vxlan4: PMTU exceptions 1 235 pmtu_ipv4_ovs_vxlan6_exception IPv4, OVS vxlan6: PMTU exceptions 1 236 pmtu_ipv6_ovs_vxlan6_exception IPv6, OVS vxlan6: PMTU exceptions 1 237 pmtu_ipv4_ovs_geneve4_exception IPv4, OVS geneve4: PMTU exceptions 1 238 pmtu_ipv6_ovs_geneve4_exception IPv6, OVS geneve4: PMTU exceptions 1 239 pmtu_ipv4_ovs_geneve6_exception IPv4, OVS geneve6: PMTU exceptions 1 240 pmtu_ipv6_ovs_geneve6_exception IPv6, OVS geneve6: PMTU exceptions 1 241 pmtu_ipv4_fou4_exception IPv4 over fou4: PMTU exceptions 1 242 pmtu_ipv6_fou4_exception IPv6 over fou4: PMTU exceptions 1 243 pmtu_ipv4_fou6_exception IPv4 over fou6: PMTU exceptions 1 244 pmtu_ipv6_fou6_exception IPv6 over fou6: PMTU exceptions 1 245 pmtu_ipv4_gue4_exception IPv4 over gue4: PMTU exceptions 1 246 pmtu_ipv6_gue4_exception IPv6 over gue4: PMTU exceptions 1 247 pmtu_ipv4_gue6_exception IPv4 over gue6: PMTU exceptions 1 248 pmtu_ipv6_gue6_exception IPv6 over gue6: PMTU exceptions 1 249 pmtu_ipv4_ipv4_exception IPv4 over IPv4: PMTU exceptions 1 250 pmtu_ipv6_ipv4_exception IPv6 over IPv4: PMTU exceptions 1 251 pmtu_ipv4_ipv6_exception IPv4 over IPv6: PMTU exceptions 1 252 pmtu_ipv6_ipv6_exception IPv6 over IPv6: PMTU exceptions 1 253 pmtu_vti6_exception vti6: PMTU exceptions 0 254 pmtu_vti4_exception vti4: PMTU exceptions 0 255 pmtu_vti6_udp_exception vti6: PMTU exceptions (ESP-in-UDP) 0 256 pmtu_vti4_udp_exception vti4: PMTU exceptions (ESP-in-UDP) 0 257 pmtu_vti6_udp_routed_exception vti6: PMTU exceptions, routed (ESP-in-UDP) 0 258 pmtu_vti4_udp_routed_exception vti4: PMTU exceptions, routed (ESP-in-UDP) 0 259 pmtu_vti4_default_mtu vti4: default MTU assignment 0 260 pmtu_vti6_default_mtu vti6: default MTU assignment 0 261 pmtu_vti4_link_add_mtu vti4: MTU setting on link creation 0 262 pmtu_vti6_link_add_mtu vti6: MTU setting on link creation 0 263 pmtu_vti6_link_change_mtu vti6: MTU changes on link changes 0 264 cleanup_ipv4_exception ipv4: cleanup of cached exceptions 1 265 cleanup_ipv6_exception ipv6: cleanup of cached exceptions 1 266 list_flush_ipv4_exception ipv4: list and flush cached exceptions 1 267 list_flush_ipv6_exception ipv6: list and flush cached exceptions 1 268 pmtu_ipv4_route_change ipv4: PMTU exception w/route replace 1 269 pmtu_ipv6_route_change ipv6: PMTU exception w/route replace 1" 270 271# Addressing and routing for tests with routers: four network segments, with 272# index SEGMENT between 1 and 4, a common prefix (PREFIX4 or PREFIX6) and an 273# identifier ID, which is 1 for hosts (A and B), 2 for routers (R1 and R2). 274# Addresses are: 275# - IPv4: PREFIX4.SEGMENT.ID (/24) 276# - IPv6: PREFIX6:SEGMENT::ID (/64) 277prefix4="10.0" 278prefix6="fc00" 279a_r1=1 280a_r2=2 281b_r1=3 282b_r2=4 283# ns peer segment 284routing_addrs=" 285 A R1 ${a_r1} 286 A R2 ${a_r2} 287 B R1 ${b_r1} 288 B R2 ${b_r2} 289" 290# Traffic from A to B goes through R1 by default, and through R2, if destined to 291# B's address on the b_r2 segment. 292# Traffic from B to A goes through R1. 293# ns destination gateway 294routes=" 295 A default ${prefix4}.${a_r1}.2 296 A ${prefix4}.${b_r2}.1 ${prefix4}.${a_r2}.2 297 B default ${prefix4}.${b_r1}.2 298 299 A default ${prefix6}:${a_r1}::2 300 A ${prefix6}:${b_r2}::1 ${prefix6}:${a_r2}::2 301 B default ${prefix6}:${b_r1}::2 302" 303USE_NH="no" 304# ns family nh id destination gateway 305nexthops=" 306 A 4 41 ${prefix4}.${a_r1}.2 veth_A-R1 307 A 4 42 ${prefix4}.${a_r2}.2 veth_A-R2 308 B 4 41 ${prefix4}.${b_r1}.2 veth_B-R1 309 310 A 6 61 ${prefix6}:${a_r1}::2 veth_A-R1 311 A 6 62 ${prefix6}:${a_r2}::2 veth_A-R2 312 B 6 61 ${prefix6}:${b_r1}::2 veth_B-R1 313" 314 315# nexthop id correlates to id in nexthops config above 316# ns family prefix nh id 317routes_nh=" 318 A 4 default 41 319 A 4 ${prefix4}.${b_r2}.1 42 320 B 4 default 41 321 322 A 6 default 61 323 A 6 ${prefix6}:${b_r2}::1 62 324 B 6 default 61 325" 326 327policy_mark=0x04 328rt_table=main 329 330veth4_a_addr="192.168.1.1" 331veth4_b_addr="192.168.1.2" 332veth4_c_addr="192.168.2.10" 333veth4_mask="24" 334veth6_a_addr="fd00:1::a" 335veth6_b_addr="fd00:1::b" 336veth6_c_addr="fd00:2::c" 337veth6_mask="64" 338 339tunnel4_a_addr="192.168.2.1" 340tunnel4_b_addr="192.168.2.2" 341tunnel4_mask="24" 342tunnel6_a_addr="fd00:2::a" 343tunnel6_b_addr="fd00:2::b" 344tunnel6_mask="64" 345 346dummy6_0_prefix="fc00:1000::" 347dummy6_1_prefix="fc00:1001::" 348dummy6_mask="64" 349 350err_buf= 351tcpdump_pids= 352nettest_pids= 353socat_pids= 354tmpoutfile= 355 356err() { 357 err_buf="${err_buf}${1} 358" 359} 360 361err_flush() { 362 echo -n "${err_buf}" 363 err_buf= 364} 365 366run_cmd() { 367 cmd="$*" 368 369 if [ "$VERBOSE" = "1" ]; then 370 printf " COMMAND: $cmd\n" 371 fi 372 373 out="$($cmd 2>&1)" 374 rc=$? 375 if [ "$VERBOSE" = "1" -a -n "$out" ]; then 376 echo " $out" 377 echo 378 fi 379 380 return $rc 381} 382 383run_cmd_bg() { 384 cmd="$*" 385 386 if [ "$VERBOSE" = "1" ]; then 387 printf " COMMAND: %s &\n" "${cmd}" 388 fi 389 390 $cmd 2>&1 & 391} 392 393# Find the auto-generated name for this namespace 394nsname() { 395 eval echo \$NS_$1 396} 397 398setup_fou_or_gue() { 399 outer="${1}" 400 inner="${2}" 401 encap="${3}" 402 403 if [ "${outer}" = "4" ]; then 404 modprobe fou || return $ksft_skip 405 a_addr="${prefix4}.${a_r1}.1" 406 b_addr="${prefix4}.${b_r1}.1" 407 if [ "${inner}" = "4" ]; then 408 type="ipip" 409 ipproto="4" 410 else 411 type="sit" 412 ipproto="41" 413 fi 414 else 415 modprobe fou6 || return $ksft_skip 416 a_addr="${prefix6}:${a_r1}::1" 417 b_addr="${prefix6}:${b_r1}::1" 418 if [ "${inner}" = "4" ]; then 419 type="ip6tnl" 420 mode="mode ipip6" 421 ipproto="4 -6" 422 else 423 type="ip6tnl" 424 mode="mode ip6ip6" 425 ipproto="41 -6" 426 fi 427 fi 428 429 run_cmd ${ns_a} ip fou add port 5555 ipproto ${ipproto} || return $ksft_skip 430 run_cmd ${ns_a} ip link add ${encap}_a type ${type} ${mode} local ${a_addr} remote ${b_addr} encap ${encap} encap-sport auto encap-dport 5556 || return $ksft_skip 431 432 run_cmd ${ns_b} ip fou add port 5556 ipproto ${ipproto} 433 run_cmd ${ns_b} ip link add ${encap}_b type ${type} ${mode} local ${b_addr} remote ${a_addr} encap ${encap} encap-sport auto encap-dport 5555 434 435 if [ "${inner}" = "4" ]; then 436 run_cmd ${ns_a} ip addr add ${tunnel4_a_addr}/${tunnel4_mask} dev ${encap}_a 437 run_cmd ${ns_b} ip addr add ${tunnel4_b_addr}/${tunnel4_mask} dev ${encap}_b 438 else 439 run_cmd ${ns_a} ip addr add ${tunnel6_a_addr}/${tunnel6_mask} dev ${encap}_a 440 run_cmd ${ns_b} ip addr add ${tunnel6_b_addr}/${tunnel6_mask} dev ${encap}_b 441 fi 442 443 run_cmd ${ns_a} ip link set ${encap}_a up 444 run_cmd ${ns_b} ip link set ${encap}_b up 445} 446 447setup_fou44() { 448 setup_fou_or_gue 4 4 fou 449} 450 451setup_fou46() { 452 setup_fou_or_gue 4 6 fou 453} 454 455setup_fou64() { 456 setup_fou_or_gue 6 4 fou 457} 458 459setup_fou66() { 460 setup_fou_or_gue 6 6 fou 461} 462 463setup_gue44() { 464 setup_fou_or_gue 4 4 gue 465} 466 467setup_gue46() { 468 setup_fou_or_gue 4 6 gue 469} 470 471setup_gue64() { 472 setup_fou_or_gue 6 4 gue 473} 474 475setup_gue66() { 476 setup_fou_or_gue 6 6 gue 477} 478 479setup_ipvX_over_ipvY() { 480 inner=${1} 481 outer=${2} 482 483 if [ "${outer}" -eq 4 ]; then 484 a_addr="${prefix4}.${a_r1}.1" 485 b_addr="${prefix4}.${b_r1}.1" 486 if [ "${inner}" -eq 4 ]; then 487 type="ipip" 488 mode="ipip" 489 else 490 type="sit" 491 mode="ip6ip" 492 fi 493 else 494 a_addr="${prefix6}:${a_r1}::1" 495 b_addr="${prefix6}:${b_r1}::1" 496 type="ip6tnl" 497 if [ "${inner}" -eq 4 ]; then 498 mode="ipip6" 499 else 500 mode="ip6ip6" 501 fi 502 fi 503 504 run_cmd ${ns_a} ip link add ip_a type ${type} local ${a_addr} remote ${b_addr} mode ${mode} || return $ksft_skip 505 run_cmd ${ns_b} ip link add ip_b type ${type} local ${b_addr} remote ${a_addr} mode ${mode} 506 507 run_cmd ${ns_a} ip link set ip_a up 508 run_cmd ${ns_b} ip link set ip_b up 509 510 if [ "${inner}" = "4" ]; then 511 run_cmd ${ns_a} ip addr add ${tunnel4_a_addr}/${tunnel4_mask} dev ip_a 512 run_cmd ${ns_b} ip addr add ${tunnel4_b_addr}/${tunnel4_mask} dev ip_b 513 else 514 run_cmd ${ns_a} ip addr add ${tunnel6_a_addr}/${tunnel6_mask} dev ip_a 515 run_cmd ${ns_b} ip addr add ${tunnel6_b_addr}/${tunnel6_mask} dev ip_b 516 fi 517} 518 519setup_ip4ip4() { 520 setup_ipvX_over_ipvY 4 4 521} 522 523setup_ip6ip4() { 524 setup_ipvX_over_ipvY 6 4 525} 526 527setup_ip4ip6() { 528 setup_ipvX_over_ipvY 4 6 529} 530 531setup_ip6ip6() { 532 setup_ipvX_over_ipvY 6 6 533} 534 535setup_namespaces() { 536 setup_ns NS_A NS_B NS_C NS_R1 NS_R2 537 for n in ${NS_A} ${NS_B} ${NS_C} ${NS_R1} ${NS_R2}; do 538 # Disable DAD, so that we don't have to wait to use the 539 # configured IPv6 addresses 540 ip netns exec ${n} sysctl -q net/ipv6/conf/default/accept_dad=0 541 done 542 ns_a="ip netns exec ${NS_A}" 543 ns_b="ip netns exec ${NS_B}" 544 ns_c="ip netns exec ${NS_C}" 545 ns_r1="ip netns exec ${NS_R1}" 546 ns_r2="ip netns exec ${NS_R2}" 547} 548 549setup_veth() { 550 run_cmd ${ns_a} ip link add veth_a type veth peer name veth_b || return 1 551 run_cmd ${ns_a} ip link set veth_b netns ${NS_B} 552 553 run_cmd ${ns_a} ip addr add ${veth4_a_addr}/${veth4_mask} dev veth_a 554 run_cmd ${ns_b} ip addr add ${veth4_b_addr}/${veth4_mask} dev veth_b 555 556 run_cmd ${ns_a} ip addr add ${veth6_a_addr}/${veth6_mask} dev veth_a 557 run_cmd ${ns_b} ip addr add ${veth6_b_addr}/${veth6_mask} dev veth_b 558 559 run_cmd ${ns_a} ip link set veth_a up 560 run_cmd ${ns_b} ip link set veth_b up 561} 562 563setup_vti() { 564 proto=${1} 565 veth_a_addr="${2}" 566 veth_b_addr="${3}" 567 vti_a_addr="${4}" 568 vti_b_addr="${5}" 569 vti_mask=${6} 570 571 [ ${proto} -eq 6 ] && vti_type="vti6" || vti_type="vti" 572 573 run_cmd ${ns_a} ip link add vti${proto}_a type ${vti_type} local ${veth_a_addr} remote ${veth_b_addr} key 10 || return 1 574 run_cmd ${ns_b} ip link add vti${proto}_b type ${vti_type} local ${veth_b_addr} remote ${veth_a_addr} key 10 575 576 run_cmd ${ns_a} ip addr add ${vti_a_addr}/${vti_mask} dev vti${proto}_a 577 run_cmd ${ns_b} ip addr add ${vti_b_addr}/${vti_mask} dev vti${proto}_b 578 579 run_cmd ${ns_a} ip link set vti${proto}_a up 580 run_cmd ${ns_b} ip link set vti${proto}_b up 581} 582 583setup_vti4() { 584 setup_vti 4 ${veth4_a_addr} ${veth4_b_addr} ${tunnel4_a_addr} ${tunnel4_b_addr} ${tunnel4_mask} 585} 586 587setup_vti6() { 588 setup_vti 6 ${veth6_a_addr} ${veth6_b_addr} ${tunnel6_a_addr} ${tunnel6_b_addr} ${tunnel6_mask} 589} 590 591setup_vti4routed() { 592 setup_vti 4 ${prefix4}.${a_r1}.1 ${prefix4}.${b_r1}.1 ${tunnel4_a_addr} ${tunnel4_b_addr} ${tunnel4_mask} 593} 594 595setup_vti6routed() { 596 setup_vti 6 ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1 ${tunnel6_a_addr} ${tunnel6_b_addr} ${tunnel6_mask} 597} 598 599setup_vxlan_or_geneve() { 600 type="${1}" 601 a_addr="${2}" 602 b_addr="${3}" 603 opts="${4}" 604 br_if_a="${5}" 605 606 if [ "${type}" = "vxlan" ]; then 607 opts="${opts} ttl 64 dstport 4789" 608 opts_a="local ${a_addr}" 609 opts_b="local ${b_addr}" 610 else 611 opts_a="" 612 opts_b="" 613 fi 614 615 run_cmd ${ns_a} ip link add ${type}_a type ${type} id 1 ${opts_a} remote ${b_addr} ${opts} || return 1 616 run_cmd ${ns_b} ip link add ${type}_b type ${type} id 1 ${opts_b} remote ${a_addr} ${opts} 617 618 if [ -n "${br_if_a}" ]; then 619 run_cmd ${ns_a} ip addr add ${tunnel4_a_addr}/${tunnel4_mask} dev ${br_if_a} 620 run_cmd ${ns_a} ip addr add ${tunnel6_a_addr}/${tunnel6_mask} dev ${br_if_a} 621 run_cmd ${ns_a} ip link set ${type}_a master ${br_if_a} 622 else 623 run_cmd ${ns_a} ip addr add ${tunnel4_a_addr}/${tunnel4_mask} dev ${type}_a 624 run_cmd ${ns_a} ip addr add ${tunnel6_a_addr}/${tunnel6_mask} dev ${type}_a 625 fi 626 627 run_cmd ${ns_b} ip addr add ${tunnel4_b_addr}/${tunnel4_mask} dev ${type}_b 628 run_cmd ${ns_b} ip addr add ${tunnel6_b_addr}/${tunnel6_mask} dev ${type}_b 629 630 run_cmd ${ns_a} ip link set ${type}_a up 631 run_cmd ${ns_b} ip link set ${type}_b up 632} 633 634setup_geneve4() { 635 setup_vxlan_or_geneve geneve ${prefix4}.${a_r1}.1 ${prefix4}.${b_r1}.1 "df set" 636} 637 638setup_vxlan4() { 639 setup_vxlan_or_geneve vxlan ${prefix4}.${a_r1}.1 ${prefix4}.${b_r1}.1 "df set" 640} 641 642setup_geneve6() { 643 setup_vxlan_or_geneve geneve ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1 "" 644} 645 646setup_vxlan6() { 647 setup_vxlan_or_geneve vxlan ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1 "" 648} 649 650setup_bridged_geneve4() { 651 setup_vxlan_or_geneve geneve ${prefix4}.${a_r1}.1 ${prefix4}.${b_r1}.1 "df set" "br0" 652} 653 654setup_bridged_vxlan4() { 655 setup_vxlan_or_geneve vxlan ${prefix4}.${a_r1}.1 ${prefix4}.${b_r1}.1 "df set" "br0" 656} 657 658setup_bridged_geneve6() { 659 setup_vxlan_or_geneve geneve ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1 "" "br0" 660} 661 662setup_bridged_vxlan6() { 663 setup_vxlan_or_geneve vxlan ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1 "" "br0" 664} 665 666setup_xfrm() { 667 proto=${1} 668 veth_a_addr="${2}" 669 veth_b_addr="${3}" 670 encap=${4} 671 672 run_cmd ${ns_a} ip -${proto} xfrm state add src ${veth_a_addr} dst ${veth_b_addr} spi 0x1000 proto esp aead 'rfc4106(gcm(aes))' 0x0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f 128 mode tunnel ${encap} || return 1 673 run_cmd ${ns_a} ip -${proto} xfrm state add src ${veth_b_addr} dst ${veth_a_addr} spi 0x1001 proto esp aead 'rfc4106(gcm(aes))' 0x0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f 128 mode tunnel ${encap} 674 run_cmd ${ns_a} ip -${proto} xfrm policy add dir out mark 10 tmpl src ${veth_a_addr} dst ${veth_b_addr} proto esp mode tunnel 675 run_cmd ${ns_a} ip -${proto} xfrm policy add dir in mark 10 tmpl src ${veth_b_addr} dst ${veth_a_addr} proto esp mode tunnel 676 677 run_cmd ${ns_b} ip -${proto} xfrm state add src ${veth_a_addr} dst ${veth_b_addr} spi 0x1000 proto esp aead 'rfc4106(gcm(aes))' 0x0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f 128 mode tunnel ${encap} 678 run_cmd ${ns_b} ip -${proto} xfrm state add src ${veth_b_addr} dst ${veth_a_addr} spi 0x1001 proto esp aead 'rfc4106(gcm(aes))' 0x0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f0f 128 mode tunnel ${encap} 679 run_cmd ${ns_b} ip -${proto} xfrm policy add dir out mark 10 tmpl src ${veth_b_addr} dst ${veth_a_addr} proto esp mode tunnel 680 run_cmd ${ns_b} ip -${proto} xfrm policy add dir in mark 10 tmpl src ${veth_a_addr} dst ${veth_b_addr} proto esp mode tunnel 681} 682 683setup_nettest_xfrm() { 684 check_gen_prog "nettest" 685 686 [ ${1} -eq 6 ] && proto="-6" || proto="" 687 port=${2} 688 689 run_cmd_bg "${ns_a}" nettest "${proto}" -q -D -s -x -p "${port}" -t 5 690 nettest_pids="${nettest_pids} $!" 691 692 run_cmd_bg "${ns_b}" nettest "${proto}" -q -D -s -x -p "${port}" -t 5 693 nettest_pids="${nettest_pids} $!" 694} 695 696setup_xfrm4() { 697 setup_xfrm 4 ${veth4_a_addr} ${veth4_b_addr} 698} 699 700setup_xfrm6() { 701 setup_xfrm 6 ${veth6_a_addr} ${veth6_b_addr} 702} 703 704setup_xfrm4udp() { 705 setup_xfrm 4 ${veth4_a_addr} ${veth4_b_addr} "encap espinudp 4500 4500 0.0.0.0" && \ 706 setup_nettest_xfrm 4 4500 707} 708 709setup_xfrm6udp() { 710 setup_xfrm 6 ${veth6_a_addr} ${veth6_b_addr} "encap espinudp 4500 4500 0.0.0.0" && \ 711 setup_nettest_xfrm 6 4500 712} 713 714setup_xfrm4udprouted() { 715 setup_xfrm 4 ${prefix4}.${a_r1}.1 ${prefix4}.${b_r1}.1 "encap espinudp 4500 4500 0.0.0.0" && \ 716 setup_nettest_xfrm 4 4500 717} 718 719setup_xfrm6udprouted() { 720 setup_xfrm 6 ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1 "encap espinudp 4500 4500 0.0.0.0" && \ 721 setup_nettest_xfrm 6 4500 722} 723 724setup_routing_old() { 725 for i in ${routes}; do 726 [ "${ns}" = "" ] && ns="${i}" && continue 727 [ "${addr}" = "" ] && addr="${i}" && continue 728 [ "${gw}" = "" ] && gw="${i}" 729 730 ns_name="$(nsname ${ns})" 731 732 ip -n "${ns_name}" route add "${addr}" table "${rt_table}" via "${gw}" 733 734 ns=""; addr=""; gw="" 735 done 736} 737 738setup_routing_new() { 739 for i in ${nexthops}; do 740 [ "${ns}" = "" ] && ns="${i}" && continue 741 [ "${fam}" = "" ] && fam="${i}" && continue 742 [ "${nhid}" = "" ] && nhid="${i}" && continue 743 [ "${gw}" = "" ] && gw="${i}" && continue 744 [ "${dev}" = "" ] && dev="${i}" 745 746 ns_name="$(nsname ${ns})" 747 748 ip -n ${ns_name} -${fam} nexthop add id ${nhid} via ${gw} dev ${dev} 749 750 ns=""; fam=""; nhid=""; gw=""; dev="" 751 752 done 753 754 for i in ${routes_nh}; do 755 [ "${ns}" = "" ] && ns="${i}" && continue 756 [ "${fam}" = "" ] && fam="${i}" && continue 757 [ "${addr}" = "" ] && addr="${i}" && continue 758 [ "${nhid}" = "" ] && nhid="${i}" 759 760 ns_name="$(nsname ${ns})" 761 762 ip -n "${ns_name}" -"${fam}" route add "${addr}" table "${rt_table}" nhid "${nhid}" 763 764 ns=""; fam=""; addr=""; nhid="" 765 done 766} 767 768setup_routing() { 769 for i in ${NS_R1} ${NS_R2}; do 770 ip netns exec ${i} sysctl -q net/ipv4/ip_forward=1 771 ip netns exec ${i} sysctl -q net/ipv6/conf/all/forwarding=1 772 done 773 774 for i in ${routing_addrs}; do 775 [ "${ns}" = "" ] && ns="${i}" && continue 776 [ "${peer}" = "" ] && peer="${i}" && continue 777 [ "${segment}" = "" ] && segment="${i}" 778 779 ns_name="$(nsname ${ns})" 780 peer_name="$(nsname ${peer})" 781 if="veth_${ns}-${peer}" 782 ifpeer="veth_${peer}-${ns}" 783 784 # Create veth links 785 ip link add ${if} up netns ${ns_name} type veth peer name ${ifpeer} netns ${peer_name} || return 1 786 ip -n ${peer_name} link set dev ${ifpeer} up 787 788 # Add addresses 789 ip -n ${ns_name} addr add ${prefix4}.${segment}.1/24 dev ${if} 790 ip -n ${ns_name} addr add ${prefix6}:${segment}::1/64 dev ${if} 791 792 ip -n ${peer_name} addr add ${prefix4}.${segment}.2/24 dev ${ifpeer} 793 ip -n ${peer_name} addr add ${prefix6}:${segment}::2/64 dev ${ifpeer} 794 795 ns=""; peer=""; segment="" 796 done 797 798 if [ "$USE_NH" = "yes" ]; then 799 setup_routing_new 800 else 801 setup_routing_old 802 fi 803 804 return 0 805} 806 807setup_policy_routing() { 808 setup_routing 809 810 ip -netns "${NS_A}" -4 rule add dsfield "${policy_mark}" \ 811 table "${rt_table}" 812 813 # Set the IPv4 Don't Fragment bit with tc, since socat doesn't seem to 814 # have an option do to it. 815 tc -netns "${NS_A}" qdisc replace dev veth_A-R1 root prio 816 tc -netns "${NS_A}" qdisc replace dev veth_A-R2 root prio 817 tc -netns "${NS_A}" filter add dev veth_A-R1 \ 818 protocol ipv4 flower ip_proto udp \ 819 action pedit ex munge ip df set 0x40 pipe csum ip and udp 820 tc -netns "${NS_A}" filter add dev veth_A-R2 \ 821 protocol ipv4 flower ip_proto udp \ 822 action pedit ex munge ip df set 0x40 pipe csum ip and udp 823} 824 825setup_bridge() { 826 run_cmd ${ns_a} ip link add br0 type bridge || return $ksft_skip 827 run_cmd ${ns_a} ip link set br0 up 828 829 run_cmd ${ns_c} ip link add veth_C-A type veth peer name veth_A-C 830 run_cmd ${ns_c} ip link set veth_A-C netns ${NS_A} 831 832 run_cmd ${ns_a} ip link set veth_A-C up 833 run_cmd ${ns_c} ip link set veth_C-A up 834 run_cmd ${ns_c} ip addr add ${veth4_c_addr}/${veth4_mask} dev veth_C-A 835 run_cmd ${ns_c} ip addr add ${veth6_c_addr}/${veth6_mask} dev veth_C-A 836 run_cmd ${ns_a} ip link set veth_A-C master br0 837} 838 839setup_ovs_via_internal_utility() { 840 type="${1}" 841 a_addr="${2}" 842 b_addr="${3}" 843 dport="${4}" 844 845 run_cmd python3 ./openvswitch/ovs-dpctl.py add-if ovs_br0 ${type}_a -t ${type} || return 1 846 847 ports=$(python3 ./openvswitch/ovs-dpctl.py show) 848 br0_port=$(echo "$ports" | grep -E "\sovs_br0" | sed -e 's@port @@' | cut -d: -f1 | xargs) 849 type_a_port=$(echo "$ports" | grep ${type}_a | sed -e 's@port @@' | cut -d: -f1 | xargs) 850 veth_a_port=$(echo "$ports" | grep veth_A | sed -e 's@port @@' | cut -d: -f1 | xargs) 851 852 v4_a_tun="${prefix4}.${a_r1}.1" 853 v4_b_tun="${prefix4}.${b_r1}.1" 854 855 v6_a_tun="${prefix6}:${a_r1}::1" 856 v6_b_tun="${prefix6}:${b_r1}::1" 857 858 if [ "${v4_a_tun}" = "${a_addr}" ]; then 859 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 860 "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x0800),ipv4()" \ 861 "set(tunnel(tun_id=1,dst=${v4_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}" 862 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 863 "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x86dd),ipv6()" \ 864 "set(tunnel(tun_id=1,dst=${v4_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}" 865 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 866 "recirc_id(0),tunnel(tun_id=1,src=${v4_b_tun},dst=${v4_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x0800),ipv4()" \ 867 "${veth_a_port}" 868 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 869 "recirc_id(0),tunnel(tun_id=1,src=${v4_b_tun},dst=${v4_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x86dd),ipv6()" \ 870 "${veth_a_port}" 871 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 872 "recirc_id(0),tunnel(tun_id=1,src=${v4_b_tun},dst=${v4_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x0806),arp()" \ 873 "${veth_a_port}" 874 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 875 "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x0806),arp(sip=${veth4_c_addr},tip=${tunnel4_b_addr})" \ 876 "set(tunnel(tun_id=1,dst=${v4_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}" 877 else 878 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 879 "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x0800),ipv4()" \ 880 "set(tunnel(tun_id=1,ipv6_dst=${v6_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}" 881 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 882 "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x86dd),ipv6()" \ 883 "set(tunnel(tun_id=1,ipv6_dst=${v6_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}" 884 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 885 "recirc_id(0),tunnel(tun_id=1,ipv6_src=${v6_b_tun},ipv6_dst=${v6_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x0800),ipv4()" \ 886 "${veth_a_port}" 887 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 888 "recirc_id(0),tunnel(tun_id=1,ipv6_src=${v6_b_tun},ipv6_dst=${v6_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x86dd),ipv6()" \ 889 "${veth_a_port}" 890 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 891 "recirc_id(0),tunnel(tun_id=1,ipv6_src=${v6_b_tun},ipv6_dst=${v6_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x0806),arp()" \ 892 "${veth_a_port}" 893 run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \ 894 "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x0806),arp(sip=${veth4_c_addr},tip=${tunnel4_b_addr})" \ 895 "set(tunnel(tun_id=1,ipv6_dst=${v6_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}" 896 fi 897} 898 899setup_ovs_via_vswitchd() { 900 type="${1}" 901 b_addr="${2}" 902 903 run_cmd ovs-vsctl add-port ovs_br0 ${type}_a -- \ 904 set interface ${type}_a type=${type} \ 905 options:remote_ip=${b_addr} options:key=1 options:csum=true || return 1 906} 907 908setup_ovs_vxlan_or_geneve() { 909 type="${1}" 910 a_addr="${2}" 911 b_addr="${3}" 912 dport="6081" 913 914 if [ "${type}" = "vxlan" ]; then 915 dport="4789" 916 opts="${opts} ttl 64 dstport 4789" 917 opts_b="local ${b_addr}" 918 fi 919 920 setup_ovs_via_internal_utility "${type}" "${a_addr}" "${b_addr}" \ 921 "${dport}" || \ 922 setup_ovs_via_vswitchd "${type}" "${b_addr}" || return 1 923 924 run_cmd ${ns_b} ip link add ${type}_b type ${type} id 1 ${opts_b} remote ${a_addr} ${opts} || return 1 925 926 run_cmd ${ns_b} ip addr add ${tunnel4_b_addr}/${tunnel4_mask} dev ${type}_b 927 run_cmd ${ns_b} ip addr add ${tunnel6_b_addr}/${tunnel6_mask} dev ${type}_b 928 929 run_cmd ip link set ${type}_a up 930 run_cmd ${ns_b} ip link set ${type}_b up 931} 932 933setup_ovs_geneve4() { 934 setup_ovs_vxlan_or_geneve geneve ${prefix4}.${a_r1}.1 ${prefix4}.${b_r1}.1 935} 936 937setup_ovs_vxlan4() { 938 setup_ovs_vxlan_or_geneve vxlan ${prefix4}.${a_r1}.1 ${prefix4}.${b_r1}.1 939} 940 941setup_ovs_geneve6() { 942 setup_ovs_vxlan_or_geneve geneve ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1 943} 944 945setup_ovs_vxlan6() { 946 setup_ovs_vxlan_or_geneve vxlan ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1 947} 948 949setup_ovs_br_internal() { 950 run_cmd python3 ./openvswitch/ovs-dpctl.py add-dp ovs_br0 || \ 951 return 1 952} 953 954setup_ovs_br_vswitchd() { 955 run_cmd ovs-vsctl add-br ovs_br0 || return 1 956} 957 958setup_ovs_add_if() { 959 ifname="${1}" 960 run_cmd python3 ./openvswitch/ovs-dpctl.py add-if ovs_br0 \ 961 "${ifname}" || \ 962 run_cmd ovs-vsctl add-port ovs_br0 "${ifname}" 963} 964 965setup_ovs_bridge() { 966 setup_ovs_br_internal || setup_ovs_br_vswitchd || return $ksft_skip 967 run_cmd ip link set ovs_br0 up 968 969 run_cmd ${ns_c} ip link add veth_C-A type veth peer name veth_A-C 970 run_cmd ${ns_c} ip link set veth_A-C netns 1 971 972 run_cmd ip link set veth_A-C up 973 run_cmd ${ns_c} ip link set veth_C-A up 974 run_cmd ${ns_c} ip addr add ${veth4_c_addr}/${veth4_mask} dev veth_C-A 975 run_cmd ${ns_c} ip addr add ${veth6_c_addr}/${veth6_mask} dev veth_C-A 976 setup_ovs_add_if veth_A-C 977 978 # Move veth_A-R1 to init 979 run_cmd ${ns_a} ip link set veth_A-R1 netns 1 980 run_cmd ip addr add ${prefix4}.${a_r1}.1/${veth4_mask} dev veth_A-R1 981 run_cmd ip addr add ${prefix6}:${a_r1}::1/${veth6_mask} dev veth_A-R1 982 run_cmd ip link set veth_A-R1 up 983 run_cmd ip route add ${prefix4}.${b_r1}.1 via ${prefix4}.${a_r1}.2 984 run_cmd ip route add ${prefix6}:${b_r1}::1 via ${prefix6}:${a_r1}::2 985} 986 987setup() { 988 [ "$(id -u)" -ne 0 ] && echo " need to run as root" && return $ksft_skip 989 990 for arg do 991 eval setup_${arg} || { echo " ${arg} not supported"; return 1; } 992 done 993} 994 995trace() { 996 [ $TRACING -eq 0 ] && return 997 998 for arg do 999 [ "${ns_cmd}" = "" ] && ns_cmd="${arg}" && continue 1000 ${ns_cmd} tcpdump --immediate-mode -s 0 -i "${arg}" -w "${name}_${arg}.pcap" 2> /dev/null & 1001 tcpdump_pids="${tcpdump_pids} $!" 1002 ns_cmd= 1003 done 1004 sleep 1 1005} 1006 1007cleanup_del_ovs_internal() { 1008 # squelch the output of the del-if commands since it can be wordy 1009 python3 ./openvswitch/ovs-dpctl.py del-if ovs_br0 -d true vxlan_a >/dev/null 2>&1 1010 python3 ./openvswitch/ovs-dpctl.py del-if ovs_br0 -d true geneve_a >/dev/null 2>&1 1011 python3 ./openvswitch/ovs-dpctl.py del-dp ovs_br0 >/dev/null 2>&1 1012} 1013 1014cleanup_del_ovs_vswitchd() { 1015 ovs-vsctl --if-exists del-port vxlan_a 2>/dev/null 1016 ovs-vsctl --if-exists del-br ovs_br0 2>/dev/null 1017} 1018 1019cleanup() { 1020 for pid in ${tcpdump_pids}; do 1021 kill ${pid} 1022 done 1023 tcpdump_pids= 1024 1025 for pid in ${nettest_pids}; do 1026 kill ${pid} 1027 done 1028 nettest_pids= 1029 1030 for pid in ${socat_pids}; do 1031 kill "${pid}" 1032 done 1033 socat_pids= 1034 1035 cleanup_all_ns 1036 1037 ip link del veth_A-C 2>/dev/null 1038 ip link del veth_A-R1 2>/dev/null 1039 cleanup_del_ovs_internal 1040 cleanup_del_ovs_vswitchd 1041 rm -f "$tmpoutfile" 1042} 1043 1044mtu() { 1045 ns_cmd="${1}" 1046 dev="${2}" 1047 mtu="${3}" 1048 1049 ${ns_cmd} ip link set dev ${dev} mtu ${mtu} 1050} 1051 1052mtu_parse() { 1053 input="${1}" 1054 1055 next=0 1056 for i in ${input}; do 1057 [ ${next} -eq 1 -a "${i}" = "lock" ] && next=2 && continue 1058 [ ${next} -eq 1 ] && echo "${i}" && return 1059 [ ${next} -eq 2 ] && echo "lock ${i}" && return 1060 [ "${i}" = "mtu" ] && next=1 1061 done 1062} 1063 1064link_get() { 1065 ns_cmd="${1}" 1066 name="${2}" 1067 1068 ${ns_cmd} ip link show dev "${name}" 1069} 1070 1071link_get_mtu() { 1072 ns_cmd="${1}" 1073 name="${2}" 1074 1075 mtu_parse "$(link_get "${ns_cmd}" ${name})" 1076} 1077 1078route_get_dst_exception() { 1079 ns_cmd="${1}" 1080 dst="${2}" 1081 dsfield="${3}" 1082 1083 if [ -z "${dsfield}" ]; then 1084 dsfield=0 1085 fi 1086 1087 ${ns_cmd} ip route get "${dst}" dsfield "${dsfield}" 1088} 1089 1090route_get_dst_pmtu_from_exception() { 1091 ns_cmd="${1}" 1092 dst="${2}" 1093 dsfield="${3}" 1094 1095 mtu_parse "$(route_get_dst_exception "${ns_cmd}" "${dst}" "${dsfield}")" 1096} 1097 1098check_pmtu_value() { 1099 expected="${1}" 1100 value="${2}" 1101 event="${3}" 1102 1103 [ "${expected}" = "any" ] && [ -n "${value}" ] && return 0 1104 [ "${value}" = "${expected}" ] && return 0 1105 [ -z "${value}" ] && err " PMTU exception wasn't created after ${event}" && return 1 1106 [ -z "${expected}" ] && err " PMTU exception shouldn't exist after ${event}" && return 1 1107 err " found PMTU exception with incorrect MTU ${value}, expected ${expected}, after ${event}" 1108 return 1 1109} 1110 1111test_pmtu_ipvX() { 1112 family=${1} 1113 1114 setup namespaces routing || return $ksft_skip 1115 trace "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 1116 "${ns_r1}" veth_R1-B "${ns_b}" veth_B-R1 \ 1117 "${ns_a}" veth_A-R2 "${ns_r2}" veth_R2-A \ 1118 "${ns_r2}" veth_R2-B "${ns_b}" veth_B-R2 1119 1120 if [ ${family} -eq 4 ]; then 1121 ping=ping 1122 dst1="${prefix4}.${b_r1}.1" 1123 dst2="${prefix4}.${b_r2}.1" 1124 else 1125 ping=${ping6} 1126 dst1="${prefix6}:${b_r1}::1" 1127 dst2="${prefix6}:${b_r2}::1" 1128 fi 1129 1130 # Set up initial MTU values 1131 mtu "${ns_a}" veth_A-R1 2000 1132 mtu "${ns_r1}" veth_R1-A 2000 1133 mtu "${ns_r1}" veth_R1-B 1400 1134 mtu "${ns_b}" veth_B-R1 1400 1135 1136 mtu "${ns_a}" veth_A-R2 2000 1137 mtu "${ns_r2}" veth_R2-A 2000 1138 mtu "${ns_r2}" veth_R2-B 1500 1139 mtu "${ns_b}" veth_B-R2 1500 1140 1141 # Create route exceptions 1142 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s 1800 ${dst1} 1143 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s 1800 ${dst2} 1144 1145 # Check that exceptions have been created with the correct PMTU 1146 pmtu_1="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst1})" 1147 check_pmtu_value "1400" "${pmtu_1}" "exceeding MTU" || return 1 1148 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 1149 check_pmtu_value "1500" "${pmtu_2}" "exceeding MTU" || return 1 1150 1151 # Decrease local MTU below PMTU, check for PMTU decrease in route exception 1152 mtu "${ns_a}" veth_A-R1 1300 1153 mtu "${ns_r1}" veth_R1-A 1300 1154 pmtu_1="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst1})" 1155 check_pmtu_value "1300" "${pmtu_1}" "decreasing local MTU" || return 1 1156 # Second exception shouldn't be modified 1157 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 1158 check_pmtu_value "1500" "${pmtu_2}" "changing local MTU on a link not on this path" || return 1 1159 1160 # Increase MTU, check for PMTU increase in route exception 1161 mtu "${ns_a}" veth_A-R1 1700 1162 mtu "${ns_r1}" veth_R1-A 1700 1163 pmtu_1="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst1})" 1164 check_pmtu_value "1700" "${pmtu_1}" "increasing local MTU" || return 1 1165 # Second exception shouldn't be modified 1166 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 1167 check_pmtu_value "1500" "${pmtu_2}" "changing local MTU on a link not on this path" || return 1 1168 1169 # Skip PMTU locking tests for IPv6 1170 [ $family -eq 6 ] && return 0 1171 1172 # Decrease remote MTU on path via R2, get new exception 1173 mtu "${ns_r2}" veth_R2-B 400 1174 mtu "${ns_b}" veth_B-R2 400 1175 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s 1400 ${dst2} 1176 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 1177 check_pmtu_value "lock 552" "${pmtu_2}" "exceeding MTU, with MTU < min_pmtu" || return 1 1178 1179 # Decrease local MTU below PMTU 1180 mtu "${ns_a}" veth_A-R2 500 1181 mtu "${ns_r2}" veth_R2-A 500 1182 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 1183 check_pmtu_value "500" "${pmtu_2}" "decreasing local MTU" || return 1 1184 1185 # Increase local MTU 1186 mtu "${ns_a}" veth_A-R2 1500 1187 mtu "${ns_r2}" veth_R2-A 1500 1188 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 1189 check_pmtu_value "1500" "${pmtu_2}" "increasing local MTU" || return 1 1190 1191 # Get new exception 1192 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s 1400 ${dst2} 1193 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 1194 check_pmtu_value "lock 552" "${pmtu_2}" "exceeding MTU, with MTU < min_pmtu" || return 1 1195} 1196 1197test_pmtu_ipv4_exception() { 1198 test_pmtu_ipvX 4 1199} 1200 1201test_pmtu_ipv6_exception() { 1202 test_pmtu_ipvX 6 1203} 1204 1205test_pmtu_ipv4_dscp_icmp_exception() { 1206 rt_table=100 1207 1208 setup namespaces policy_routing || return $ksft_skip 1209 trace "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 1210 "${ns_r1}" veth_R1-B "${ns_b}" veth_B-R1 \ 1211 "${ns_a}" veth_A-R2 "${ns_r2}" veth_R2-A \ 1212 "${ns_r2}" veth_R2-B "${ns_b}" veth_B-R2 1213 1214 # Set up initial MTU values 1215 mtu "${ns_a}" veth_A-R1 2000 1216 mtu "${ns_r1}" veth_R1-A 2000 1217 mtu "${ns_r1}" veth_R1-B 1400 1218 mtu "${ns_b}" veth_B-R1 1400 1219 1220 mtu "${ns_a}" veth_A-R2 2000 1221 mtu "${ns_r2}" veth_R2-A 2000 1222 mtu "${ns_r2}" veth_R2-B 1500 1223 mtu "${ns_b}" veth_B-R2 1500 1224 1225 len=$((2000 - 20 - 8)) # Fills MTU of veth_A-R1 1226 1227 dst1="${prefix4}.${b_r1}.1" 1228 dst2="${prefix4}.${b_r2}.1" 1229 1230 # Create route exceptions 1231 dsfield=${policy_mark} # No ECN bit set (Not-ECT) 1232 run_cmd "${ns_a}" ping -q -M want -Q "${dsfield}" -c 1 -w 1 -s "${len}" "${dst1}" 1233 1234 dsfield=$(printf "%#x" $((policy_mark + 0x02))) # ECN=2 (ECT(0)) 1235 run_cmd "${ns_a}" ping -q -M want -Q "${dsfield}" -c 1 -w 1 -s "${len}" "${dst2}" 1236 1237 # Check that exceptions have been created with the correct PMTU 1238 pmtu_1="$(route_get_dst_pmtu_from_exception "${ns_a}" "${dst1}" "${policy_mark}")" 1239 check_pmtu_value "1400" "${pmtu_1}" "exceeding MTU" || return 1 1240 1241 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" "${dst2}" "${policy_mark}")" 1242 check_pmtu_value "1500" "${pmtu_2}" "exceeding MTU" || return 1 1243} 1244 1245test_pmtu_ipv4_dscp_udp_exception() { 1246 rt_table=100 1247 1248 if ! which socat > /dev/null 2>&1; then 1249 echo "'socat' command not found; skipping tests" 1250 return $ksft_skip 1251 fi 1252 1253 setup namespaces policy_routing || return $ksft_skip 1254 trace "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 1255 "${ns_r1}" veth_R1-B "${ns_b}" veth_B-R1 \ 1256 "${ns_a}" veth_A-R2 "${ns_r2}" veth_R2-A \ 1257 "${ns_r2}" veth_R2-B "${ns_b}" veth_B-R2 1258 1259 # Set up initial MTU values 1260 mtu "${ns_a}" veth_A-R1 2000 1261 mtu "${ns_r1}" veth_R1-A 2000 1262 mtu "${ns_r1}" veth_R1-B 1400 1263 mtu "${ns_b}" veth_B-R1 1400 1264 1265 mtu "${ns_a}" veth_A-R2 2000 1266 mtu "${ns_r2}" veth_R2-A 2000 1267 mtu "${ns_r2}" veth_R2-B 1500 1268 mtu "${ns_b}" veth_B-R2 1500 1269 1270 len=$((2000 - 20 - 8)) # Fills MTU of veth_A-R1 1271 1272 dst1="${prefix4}.${b_r1}.1" 1273 dst2="${prefix4}.${b_r2}.1" 1274 1275 # Create route exceptions 1276 run_cmd_bg "${ns_b}" socat UDP-LISTEN:50000 OPEN:/dev/null,wronly=1 1277 socat_pids="${socat_pids} $!" 1278 1279 dsfield=${policy_mark} # No ECN bit set (Not-ECT) 1280 run_cmd "${ns_a}" socat OPEN:/dev/zero,rdonly=1,readbytes="${len}" \ 1281 UDP:"${dst1}":50000,tos="${dsfield}" 1282 1283 dsfield=$(printf "%#x" $((policy_mark + 0x02))) # ECN=2 (ECT(0)) 1284 run_cmd "${ns_a}" socat OPEN:/dev/zero,rdonly=1,readbytes="${len}" \ 1285 UDP:"${dst2}":50000,tos="${dsfield}" 1286 1287 # Check that exceptions have been created with the correct PMTU 1288 pmtu_1="$(route_get_dst_pmtu_from_exception "${ns_a}" "${dst1}" "${policy_mark}")" 1289 check_pmtu_value "1400" "${pmtu_1}" "exceeding MTU" || return 1 1290 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" "${dst2}" "${policy_mark}")" 1291 check_pmtu_value "1500" "${pmtu_2}" "exceeding MTU" || return 1 1292} 1293 1294test_pmtu_ipvX_over_vxlanY_or_geneveY_exception() { 1295 type=${1} 1296 family=${2} 1297 outer_family=${3} 1298 ll_mtu=4000 1299 1300 if [ ${outer_family} -eq 4 ]; then 1301 setup namespaces routing ${type}4 || return $ksft_skip 1302 # IPv4 header UDP header VXLAN/GENEVE header Ethernet header 1303 exp_mtu=$((${ll_mtu} - 20 - 8 - 8 - 14)) 1304 else 1305 setup namespaces routing ${type}6 || return $ksft_skip 1306 # IPv6 header UDP header VXLAN/GENEVE header Ethernet header 1307 exp_mtu=$((${ll_mtu} - 40 - 8 - 8 - 14)) 1308 fi 1309 1310 trace "${ns_a}" ${type}_a "${ns_b}" ${type}_b \ 1311 "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 1312 "${ns_b}" veth_B-R1 "${ns_r1}" veth_R1-B 1313 1314 if [ ${family} -eq 4 ]; then 1315 ping=ping 1316 dst=${tunnel4_b_addr} 1317 else 1318 ping=${ping6} 1319 dst=${tunnel6_b_addr} 1320 fi 1321 1322 # Create route exception by exceeding link layer MTU 1323 mtu "${ns_a}" veth_A-R1 $((${ll_mtu} + 1000)) 1324 mtu "${ns_r1}" veth_R1-A $((${ll_mtu} + 1000)) 1325 mtu "${ns_b}" veth_B-R1 ${ll_mtu} 1326 mtu "${ns_r1}" veth_R1-B ${ll_mtu} 1327 1328 mtu "${ns_a}" ${type}_a $((${ll_mtu} + 1000)) 1329 mtu "${ns_b}" ${type}_b $((${ll_mtu} + 1000)) 1330 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s $((${ll_mtu} + 500)) ${dst} 1331 1332 # Check that exception was created 1333 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst})" 1334 check_pmtu_value ${exp_mtu} "${pmtu}" "exceeding link layer MTU on ${type} interface" 1335} 1336 1337test_pmtu_ipv4_vxlan4_exception() { 1338 test_pmtu_ipvX_over_vxlanY_or_geneveY_exception vxlan 4 4 1339} 1340 1341test_pmtu_ipv6_vxlan4_exception() { 1342 test_pmtu_ipvX_over_vxlanY_or_geneveY_exception vxlan 6 4 1343} 1344 1345test_pmtu_ipv4_geneve4_exception() { 1346 test_pmtu_ipvX_over_vxlanY_or_geneveY_exception geneve 4 4 1347} 1348 1349test_pmtu_ipv6_geneve4_exception() { 1350 test_pmtu_ipvX_over_vxlanY_or_geneveY_exception geneve 6 4 1351} 1352 1353test_pmtu_ipv4_vxlan6_exception() { 1354 test_pmtu_ipvX_over_vxlanY_or_geneveY_exception vxlan 4 6 1355} 1356 1357test_pmtu_ipv6_vxlan6_exception() { 1358 test_pmtu_ipvX_over_vxlanY_or_geneveY_exception vxlan 6 6 1359} 1360 1361test_pmtu_ipv4_geneve6_exception() { 1362 test_pmtu_ipvX_over_vxlanY_or_geneveY_exception geneve 4 6 1363} 1364 1365test_pmtu_ipv6_geneve6_exception() { 1366 test_pmtu_ipvX_over_vxlanY_or_geneveY_exception geneve 6 6 1367} 1368 1369test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception() { 1370 type=${1} 1371 family=${2} 1372 outer_family=${3} 1373 ll_mtu=4000 1374 1375 if [ ${outer_family} -eq 4 ]; then 1376 setup namespaces routing bridge bridged_${type}4 || return $ksft_skip 1377 # IPv4 header UDP header VXLAN/GENEVE header Ethernet header 1378 exp_mtu=$((${ll_mtu} - 20 - 8 - 8 - 14)) 1379 else 1380 setup namespaces routing bridge bridged_${type}6 || return $ksft_skip 1381 # IPv6 header UDP header VXLAN/GENEVE header Ethernet header 1382 exp_mtu=$((${ll_mtu} - 40 - 8 - 8 - 14)) 1383 fi 1384 1385 trace "${ns_a}" ${type}_a "${ns_b}" ${type}_b \ 1386 "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 1387 "${ns_b}" veth_B-R1 "${ns_r1}" veth_R1-B \ 1388 "${ns_a}" br0 "${ns_a}" veth-A-C \ 1389 "${ns_c}" veth_C-A 1390 1391 if [ ${family} -eq 4 ]; then 1392 ping=ping 1393 dst=${tunnel4_b_addr} 1394 else 1395 ping=${ping6} 1396 dst=${tunnel6_b_addr} 1397 fi 1398 1399 # Create route exception by exceeding link layer MTU 1400 mtu "${ns_a}" veth_A-R1 $((${ll_mtu} + 1000)) 1401 mtu "${ns_a}" br0 $((${ll_mtu} + 1000)) 1402 mtu "${ns_a}" veth_A-C $((${ll_mtu} + 1000)) 1403 mtu "${ns_c}" veth_C-A $((${ll_mtu} + 1000)) 1404 mtu "${ns_r1}" veth_R1-A $((${ll_mtu} + 1000)) 1405 mtu "${ns_b}" veth_B-R1 ${ll_mtu} 1406 mtu "${ns_r1}" veth_R1-B ${ll_mtu} 1407 1408 mtu "${ns_a}" ${type}_a $((${ll_mtu} + 1000)) 1409 mtu "${ns_b}" ${type}_b $((${ll_mtu} + 1000)) 1410 1411 run_cmd ${ns_c} ${ping} -q -M want -i 0.1 -c 10 -s $((${ll_mtu} + 500)) ${dst} || return 1 1412 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s $((${ll_mtu} + 500)) ${dst} || return 1 1413 1414 # Check that exceptions were created 1415 pmtu="$(route_get_dst_pmtu_from_exception "${ns_c}" ${dst})" 1416 check_pmtu_value ${exp_mtu} "${pmtu}" "exceeding link layer MTU on bridged ${type} interface" 1417 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst})" 1418 check_pmtu_value ${exp_mtu} "${pmtu}" "exceeding link layer MTU on locally bridged ${type} interface" 1419 1420 tmpoutfile=$(mktemp) 1421 1422 # Flush Exceptions, retry with TCP 1423 run_cmd ${ns_a} ip route flush cached ${dst} 1424 run_cmd ${ns_b} ip route flush cached ${dst} 1425 run_cmd ${ns_c} ip route flush cached ${dst} 1426 1427 for target in "${ns_a}" "${ns_c}" ; do 1428 if [ ${family} -eq 4 ]; then 1429 TCPDST=TCP:${dst}:50000 1430 else 1431 TCPDST="TCP:[${dst}]:50000" 1432 fi 1433 ${ns_b} socat -T 3 -u -6 TCP-LISTEN:50000,reuseaddr STDOUT > $tmpoutfile & 1434 local socat_pid=$! 1435 1436 wait_local_port_listen ${NS_B} 50000 tcp 1437 1438 dd if=/dev/zero status=none bs=1M count=1 | ${target} socat -T 3 -u STDIN $TCPDST,connect-timeout=3 1439 1440 wait ${socat_pid} 1441 size=$(du -sb $tmpoutfile) 1442 size=${size%%/tmp/*} 1443 1444 [ $size -ne 1048576 ] && err "File size $size mismatches expected value in locally bridged vxlan test" && return 1 1445 done 1446 1447 rm -f "$tmpoutfile" 1448 1449 # Check that exceptions were created 1450 pmtu="$(route_get_dst_pmtu_from_exception "${ns_c}" ${dst})" 1451 check_pmtu_value ${exp_mtu} "${pmtu}" "tcp: exceeding link layer MTU on bridged ${type} interface" 1452 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst})" 1453 check_pmtu_value ${exp_mtu} "${pmtu}" "tcp exceeding link layer MTU on locally bridged ${type} interface" 1454} 1455 1456test_pmtu_ipv4_br_vxlan4_exception() { 1457 test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception vxlan 4 4 1458} 1459 1460test_pmtu_ipv6_br_vxlan4_exception() { 1461 test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception vxlan 6 4 1462} 1463 1464test_pmtu_ipv4_br_geneve4_exception() { 1465 test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception geneve 4 4 1466} 1467 1468test_pmtu_ipv6_br_geneve4_exception() { 1469 test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception geneve 6 4 1470} 1471 1472test_pmtu_ipv4_br_vxlan6_exception() { 1473 test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception vxlan 4 6 1474} 1475 1476test_pmtu_ipv6_br_vxlan6_exception() { 1477 test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception vxlan 6 6 1478} 1479 1480test_pmtu_ipv4_br_geneve6_exception() { 1481 test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception geneve 4 6 1482} 1483 1484test_pmtu_ipv6_br_geneve6_exception() { 1485 test_pmtu_ipvX_over_bridged_vxlanY_or_geneveY_exception geneve 6 6 1486} 1487 1488test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception() { 1489 type=${1} 1490 family=${2} 1491 outer_family=${3} 1492 ll_mtu=4000 1493 1494 if [ "${type}" = "vxlan" ]; then 1495 tun_a="vxlan_sys_4789" 1496 elif [ "${type}" = "geneve" ]; then 1497 tun_a="genev_sys_6081" 1498 fi 1499 1500 if [ ${outer_family} -eq 4 ]; then 1501 setup namespaces routing ovs_bridge ovs_${type}4 || return $ksft_skip 1502 # IPv4 header UDP header VXLAN/GENEVE header Ethernet header 1503 exp_mtu=$((${ll_mtu} - 20 - 8 - 8 - 14)) 1504 else 1505 setup namespaces routing ovs_bridge ovs_${type}6 || return $ksft_skip 1506 # IPv6 header UDP header VXLAN/GENEVE header Ethernet header 1507 exp_mtu=$((${ll_mtu} - 40 - 8 - 8 - 14)) 1508 fi 1509 1510 trace "" ${type}_a "${ns_b}" ${type}_b \ 1511 "" veth_A-R1 "${ns_r1}" veth_R1-A \ 1512 "${ns_b}" veth_B-R1 "${ns_r1}" veth_R1-B \ 1513 "" ovs_br0 "" veth-A_C \ 1514 "${ns_c}" veth_C-A "" "${tun_a}" 1515 1516 if [ ${family} -eq 4 ]; then 1517 ping=ping 1518 dst=${tunnel4_b_addr} 1519 else 1520 ping=${ping6} 1521 dst=${tunnel6_b_addr} 1522 fi 1523 1524 # Create route exception by exceeding link layer MTU 1525 mtu "" veth_A-R1 $((${ll_mtu} + 1000)) 1526 mtu "" ovs_br0 $((${ll_mtu} + 1000)) 1527 mtu "" veth_A-C $((${ll_mtu} + 1000)) 1528 mtu "${ns_c}" veth_C-A $((${ll_mtu} + 1000)) 1529 mtu "${ns_r1}" veth_R1-A $((${ll_mtu} + 1000)) 1530 mtu "${ns_b}" veth_B-R1 ${ll_mtu} 1531 mtu "${ns_r1}" veth_R1-B ${ll_mtu} 1532 1533 mtu "" ${tun_a} $((${ll_mtu} + 1000)) 2>/dev/null || \ 1534 mtu "" ${type}_a $((${ll_mtu} + 1000)) 2>/dev/null 1535 mtu "${ns_b}" ${type}_b $((${ll_mtu} + 1000)) 1536 1537 run_cmd ${ns_c} ${ping} -q -M want -i 0.1 -c 20 -s $((${ll_mtu} + 500)) ${dst} || return 1 1538 1539 # Check that exceptions were created 1540 pmtu="$(route_get_dst_pmtu_from_exception "${ns_c}" ${dst})" 1541 check_pmtu_value ${exp_mtu} "${pmtu}" "exceeding link layer MTU on Open vSwitch ${type} interface" 1542} 1543 1544test_pmtu_ipv4_ovs_vxlan4_exception() { 1545 test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception vxlan 4 4 1546} 1547 1548test_pmtu_ipv6_ovs_vxlan4_exception() { 1549 test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception vxlan 6 4 1550} 1551 1552test_pmtu_ipv4_ovs_geneve4_exception() { 1553 test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception geneve 4 4 1554} 1555 1556test_pmtu_ipv6_ovs_geneve4_exception() { 1557 test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception geneve 6 4 1558} 1559 1560test_pmtu_ipv4_ovs_vxlan6_exception() { 1561 test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception vxlan 4 6 1562} 1563 1564test_pmtu_ipv6_ovs_vxlan6_exception() { 1565 test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception vxlan 6 6 1566} 1567 1568test_pmtu_ipv4_ovs_geneve6_exception() { 1569 test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception geneve 4 6 1570} 1571 1572test_pmtu_ipv6_ovs_geneve6_exception() { 1573 test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception geneve 6 6 1574} 1575 1576test_pmtu_ipvX_over_fouY_or_gueY() { 1577 inner_family=${1} 1578 outer_family=${2} 1579 encap=${3} 1580 ll_mtu=4000 1581 1582 setup namespaces routing ${encap}${outer_family}${inner_family} || return $ksft_skip 1583 trace "${ns_a}" ${encap}_a "${ns_b}" ${encap}_b \ 1584 "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 1585 "${ns_b}" veth_B-R1 "${ns_r1}" veth_R1-B 1586 1587 if [ ${inner_family} -eq 4 ]; then 1588 ping=ping 1589 dst=${tunnel4_b_addr} 1590 else 1591 ping=${ping6} 1592 dst=${tunnel6_b_addr} 1593 fi 1594 1595 if [ "${encap}" = "gue" ]; then 1596 encap_overhead=4 1597 else 1598 encap_overhead=0 1599 fi 1600 1601 if [ ${outer_family} -eq 4 ]; then 1602 # IPv4 header UDP header 1603 exp_mtu=$((${ll_mtu} - 20 - 8 - ${encap_overhead})) 1604 else 1605 # IPv6 header Option 4 UDP header 1606 exp_mtu=$((${ll_mtu} - 40 - 8 - 8 - ${encap_overhead})) 1607 fi 1608 1609 # Create route exception by exceeding link layer MTU 1610 mtu "${ns_a}" veth_A-R1 $((${ll_mtu} + 1000)) 1611 mtu "${ns_r1}" veth_R1-A $((${ll_mtu} + 1000)) 1612 mtu "${ns_b}" veth_B-R1 ${ll_mtu} 1613 mtu "${ns_r1}" veth_R1-B ${ll_mtu} 1614 1615 mtu "${ns_a}" ${encap}_a $((${ll_mtu} + 1000)) 1616 mtu "${ns_b}" ${encap}_b $((${ll_mtu} + 1000)) 1617 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s $((${ll_mtu} + 500)) ${dst} 1618 1619 # Check that exception was created 1620 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst})" 1621 check_pmtu_value ${exp_mtu} "${pmtu}" "exceeding link layer MTU on ${encap} interface" 1622} 1623 1624test_pmtu_ipv4_fou4_exception() { 1625 test_pmtu_ipvX_over_fouY_or_gueY 4 4 fou 1626} 1627 1628test_pmtu_ipv6_fou4_exception() { 1629 test_pmtu_ipvX_over_fouY_or_gueY 6 4 fou 1630} 1631 1632test_pmtu_ipv4_fou6_exception() { 1633 test_pmtu_ipvX_over_fouY_or_gueY 4 6 fou 1634} 1635 1636test_pmtu_ipv6_fou6_exception() { 1637 test_pmtu_ipvX_over_fouY_or_gueY 6 6 fou 1638} 1639 1640test_pmtu_ipv4_gue4_exception() { 1641 test_pmtu_ipvX_over_fouY_or_gueY 4 4 gue 1642} 1643 1644test_pmtu_ipv6_gue4_exception() { 1645 test_pmtu_ipvX_over_fouY_or_gueY 6 4 gue 1646} 1647 1648test_pmtu_ipv4_gue6_exception() { 1649 test_pmtu_ipvX_over_fouY_or_gueY 4 6 gue 1650} 1651 1652test_pmtu_ipv6_gue6_exception() { 1653 test_pmtu_ipvX_over_fouY_or_gueY 6 6 gue 1654} 1655 1656test_pmtu_ipvX_over_ipvY_exception() { 1657 inner=${1} 1658 outer=${2} 1659 ll_mtu=4000 1660 1661 setup namespaces routing ip${inner}ip${outer} || return $ksft_skip 1662 1663 trace "${ns_a}" ip_a "${ns_b}" ip_b \ 1664 "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 1665 "${ns_b}" veth_B-R1 "${ns_r1}" veth_R1-B 1666 1667 if [ ${inner} -eq 4 ]; then 1668 ping=ping 1669 dst=${tunnel4_b_addr} 1670 else 1671 ping=${ping6} 1672 dst=${tunnel6_b_addr} 1673 fi 1674 1675 if [ ${outer} -eq 4 ]; then 1676 # IPv4 header 1677 exp_mtu=$((${ll_mtu} - 20)) 1678 else 1679 # IPv6 header Option 4 1680 exp_mtu=$((${ll_mtu} - 40 - 8)) 1681 fi 1682 1683 # Create route exception by exceeding link layer MTU 1684 mtu "${ns_a}" veth_A-R1 $((${ll_mtu} + 1000)) 1685 mtu "${ns_r1}" veth_R1-A $((${ll_mtu} + 1000)) 1686 mtu "${ns_b}" veth_B-R1 ${ll_mtu} 1687 mtu "${ns_r1}" veth_R1-B ${ll_mtu} 1688 1689 mtu "${ns_a}" ip_a $((${ll_mtu} + 1000)) || return 1690 mtu "${ns_b}" ip_b $((${ll_mtu} + 1000)) || return 1691 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s $((${ll_mtu} + 500)) ${dst} 1692 1693 # Check that exception was created 1694 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst})" 1695 check_pmtu_value ${exp_mtu} "${pmtu}" "exceeding link layer MTU on ip${inner}ip${outer} interface" 1696} 1697 1698test_pmtu_ipv4_ipv4_exception() { 1699 test_pmtu_ipvX_over_ipvY_exception 4 4 1700} 1701 1702test_pmtu_ipv6_ipv4_exception() { 1703 test_pmtu_ipvX_over_ipvY_exception 6 4 1704} 1705 1706test_pmtu_ipv4_ipv6_exception() { 1707 test_pmtu_ipvX_over_ipvY_exception 4 6 1708} 1709 1710test_pmtu_ipv6_ipv6_exception() { 1711 test_pmtu_ipvX_over_ipvY_exception 6 6 1712} 1713 1714test_pmtu_vti4_exception() { 1715 setup namespaces veth vti4 xfrm4 || return $ksft_skip 1716 trace "${ns_a}" veth_a "${ns_b}" veth_b \ 1717 "${ns_a}" vti4_a "${ns_b}" vti4_b 1718 1719 veth_mtu=1500 1720 vti_mtu=$((veth_mtu - 20)) 1721 1722 # SPI SN IV ICV pad length next header 1723 esp_payload_rfc4106=$((vti_mtu - 4 - 4 - 8 - 16 - 1 - 1)) 1724 ping_payload=$((esp_payload_rfc4106 - 28)) 1725 1726 mtu "${ns_a}" veth_a ${veth_mtu} 1727 mtu "${ns_b}" veth_b ${veth_mtu} 1728 mtu "${ns_a}" vti4_a ${vti_mtu} 1729 mtu "${ns_b}" vti4_b ${vti_mtu} 1730 1731 # Send DF packet without exceeding link layer MTU, check that no 1732 # exception is created 1733 run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s ${ping_payload} ${tunnel4_b_addr} 1734 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel4_b_addr})" 1735 check_pmtu_value "" "${pmtu}" "sending packet smaller than PMTU (IP payload length ${esp_payload_rfc4106})" || return 1 1736 1737 # Now exceed link layer MTU by one byte, check that exception is created 1738 # with the right PMTU value 1739 run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s $((ping_payload + 1)) ${tunnel4_b_addr} 1740 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel4_b_addr})" 1741 check_pmtu_value "${esp_payload_rfc4106}" "${pmtu}" "exceeding PMTU (IP payload length $((esp_payload_rfc4106 + 1)))" 1742} 1743 1744test_pmtu_vti6_exception() { 1745 setup namespaces veth vti6 xfrm6 || return $ksft_skip 1746 trace "${ns_a}" veth_a "${ns_b}" veth_b \ 1747 "${ns_a}" vti6_a "${ns_b}" vti6_b 1748 fail=0 1749 1750 # Create route exception by exceeding link layer MTU 1751 mtu "${ns_a}" veth_a 4000 1752 mtu "${ns_b}" veth_b 4000 1753 mtu "${ns_a}" vti6_a 5000 1754 mtu "${ns_b}" vti6_b 5000 1755 run_cmd ${ns_a} ${ping6} -q -i 0.1 -w 1 -s 60000 ${tunnel6_b_addr} 1756 1757 # Check that exception was created 1758 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel6_b_addr})" 1759 check_pmtu_value any "${pmtu}" "creating tunnel exceeding link layer MTU" || return 1 1760 1761 # Decrease tunnel MTU, check for PMTU decrease in route exception 1762 mtu "${ns_a}" vti6_a 3000 1763 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel6_b_addr})" 1764 check_pmtu_value "3000" "${pmtu}" "decreasing tunnel MTU" || fail=1 1765 1766 # Increase tunnel MTU, check for PMTU increase in route exception 1767 mtu "${ns_a}" vti6_a 9000 1768 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel6_b_addr})" 1769 check_pmtu_value "9000" "${pmtu}" "increasing tunnel MTU" || fail=1 1770 1771 return ${fail} 1772} 1773 1774test_pmtu_vti4_udp_exception() { 1775 setup namespaces veth vti4 xfrm4udp || return $ksft_skip 1776 trace "${ns_a}" veth_a "${ns_b}" veth_b \ 1777 "${ns_a}" vti4_a "${ns_b}" vti4_b 1778 1779 veth_mtu=1500 1780 vti_mtu=$((veth_mtu - 20)) 1781 1782 # UDP SPI SN IV ICV pad length next header 1783 esp_payload_rfc4106=$((vti_mtu - 8 - 4 - 4 - 8 - 16 - 1 - 1)) 1784 ping_payload=$((esp_payload_rfc4106 - 28)) 1785 1786 mtu "${ns_a}" veth_a ${veth_mtu} 1787 mtu "${ns_b}" veth_b ${veth_mtu} 1788 mtu "${ns_a}" vti4_a ${vti_mtu} 1789 mtu "${ns_b}" vti4_b ${vti_mtu} 1790 1791 # Send DF packet without exceeding link layer MTU, check that no 1792 # exception is created 1793 run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s ${ping_payload} ${tunnel4_b_addr} 1794 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel4_b_addr})" 1795 check_pmtu_value "" "${pmtu}" "sending packet smaller than PMTU (IP payload length ${esp_payload_rfc4106})" || return 1 1796 1797 # Now exceed link layer MTU by one byte, check that exception is created 1798 # with the right PMTU value 1799 run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s $((ping_payload + 1)) ${tunnel4_b_addr} 1800 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel4_b_addr})" 1801 check_pmtu_value "${esp_payload_rfc4106}" "${pmtu}" "exceeding PMTU (IP payload length $((esp_payload_rfc4106 + 1)))" 1802} 1803 1804test_pmtu_vti6_udp_exception() { 1805 setup namespaces veth vti6 xfrm6udp || return $ksft_skip 1806 trace "${ns_a}" veth_a "${ns_b}" veth_b \ 1807 "${ns_a}" vti6_a "${ns_b}" vti6_b 1808 fail=0 1809 1810 # Create route exception by exceeding link layer MTU 1811 mtu "${ns_a}" veth_a 4000 1812 mtu "${ns_b}" veth_b 4000 1813 mtu "${ns_a}" vti6_a 5000 1814 mtu "${ns_b}" vti6_b 5000 1815 run_cmd ${ns_a} ${ping6} -q -i 0.1 -w 1 -s 60000 ${tunnel6_b_addr} 1816 1817 # Check that exception was created 1818 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel6_b_addr})" 1819 check_pmtu_value any "${pmtu}" "creating tunnel exceeding link layer MTU" || return 1 1820 1821 # Decrease tunnel MTU, check for PMTU decrease in route exception 1822 mtu "${ns_a}" vti6_a 3000 1823 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel6_b_addr})" 1824 check_pmtu_value "3000" "${pmtu}" "decreasing tunnel MTU" || fail=1 1825 1826 # Increase tunnel MTU, check for PMTU increase in route exception 1827 mtu "${ns_a}" vti6_a 9000 1828 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel6_b_addr})" 1829 check_pmtu_value "9000" "${pmtu}" "increasing tunnel MTU" || fail=1 1830 1831 return ${fail} 1832} 1833 1834test_pmtu_vti4_udp_routed_exception() { 1835 setup namespaces routing vti4routed xfrm4udprouted || return $ksft_skip 1836 trace "${ns_a}" veth_A-R1 "${ns_b}" veth_B-R1 \ 1837 "${ns_a}" vti4_a "${ns_b}" vti4_b 1838 1839 veth_mtu=1500 1840 vti_mtu=$((veth_mtu - 20)) 1841 1842 # UDP SPI SN IV ICV pad length next header 1843 esp_payload_rfc4106=$((vti_mtu - 8 - 4 - 4 - 8 - 16 - 1 - 1)) 1844 ping_payload=$((esp_payload_rfc4106 - 28)) 1845 1846 mtu "${ns_a}" veth_A-R1 ${veth_mtu} 1847 mtu "${ns_r1}" veth_R1-A ${veth_mtu} 1848 mtu "${ns_b}" veth_B-R1 ${veth_mtu} 1849 mtu "${ns_r1}" veth_R1-B ${veth_mtu} 1850 1851 mtu "${ns_a}" vti4_a ${vti_mtu} 1852 mtu "${ns_b}" vti4_b ${vti_mtu} 1853 1854 # Send DF packet without exceeding link layer MTU, check that no 1855 # exception is created 1856 run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s ${ping_payload} ${tunnel4_b_addr} 1857 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel4_b_addr})" 1858 check_pmtu_value "" "${pmtu}" "sending packet smaller than PMTU (IP payload length ${esp_payload_rfc4106})" || return 1 1859 1860 # Now decrease link layer MTU by 8 bytes on R1, check that exception is created 1861 # with the right PMTU value 1862 mtu "${ns_r1}" veth_R1-B $((veth_mtu - 8)) 1863 run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s $((ping_payload)) ${tunnel4_b_addr} 1864 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel4_b_addr})" 1865 check_pmtu_value "$((esp_payload_rfc4106 - 8))" "${pmtu}" "exceeding PMTU (IP payload length $((esp_payload_rfc4106)))" 1866} 1867 1868test_pmtu_vti6_udp_routed_exception() { 1869 setup namespaces routing vti6routed xfrm6udprouted || return $ksft_skip 1870 trace "${ns_a}" veth_A-R1 "${ns_b}" veth_B-R1 \ 1871 "${ns_a}" vti6_a "${ns_b}" vti6_b 1872 1873 veth_mtu=1500 1874 vti_mtu=$((veth_mtu - 40)) 1875 1876 # UDP SPI SN IV ICV pad length next header 1877 esp_payload_rfc4106=$((vti_mtu - 8 - 4 - 4 - 8 - 16 - 1 - 1)) 1878 ping_payload=$((esp_payload_rfc4106 - 48)) 1879 1880 mtu "${ns_a}" veth_A-R1 ${veth_mtu} 1881 mtu "${ns_r1}" veth_R1-A ${veth_mtu} 1882 mtu "${ns_b}" veth_B-R1 ${veth_mtu} 1883 mtu "${ns_r1}" veth_R1-B ${veth_mtu} 1884 1885 # mtu "${ns_a}" vti6_a ${vti_mtu} 1886 # mtu "${ns_b}" vti6_b ${vti_mtu} 1887 1888 run_cmd ${ns_a} ${ping6} -q -M want -i 0.1 -w 1 -s ${ping_payload} ${tunnel6_b_addr} 1889 1890 # Check that exception was not created 1891 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel6_b_addr})" 1892 check_pmtu_value "" "${pmtu}" "sending packet smaller than PMTU (IP payload length ${esp_payload_rfc4106})" || return 1 1893 1894 # Now decrease link layer MTU by 8 bytes on R1, check that exception is created 1895 # with the right PMTU value 1896 mtu "${ns_r1}" veth_R1-B $((veth_mtu - 8)) 1897 run_cmd ${ns_a} ${ping6} -q -M want -i 0.1 -w 1 -s $((ping_payload)) ${tunnel6_b_addr} 1898 pmtu="$(route_get_dst_pmtu_from_exception "${ns_a}" ${tunnel6_b_addr})" 1899 check_pmtu_value "$((esp_payload_rfc4106 - 8))" "${pmtu}" "exceeding PMTU (IP payload length $((esp_payload_rfc4106)))" 1900 1901} 1902 1903test_pmtu_vti4_default_mtu() { 1904 setup namespaces veth vti4 || return $ksft_skip 1905 1906 # Check that MTU of vti device is MTU of veth minus IPv4 header length 1907 veth_mtu="$(link_get_mtu "${ns_a}" veth_a)" 1908 vti4_mtu="$(link_get_mtu "${ns_a}" vti4_a)" 1909 if [ $((veth_mtu - vti4_mtu)) -ne 20 ]; then 1910 err " vti MTU ${vti4_mtu} is not veth MTU ${veth_mtu} minus IPv4 header length" 1911 return 1 1912 fi 1913} 1914 1915test_pmtu_vti6_default_mtu() { 1916 setup namespaces veth vti6 || return $ksft_skip 1917 1918 # Check that MTU of vti device is MTU of veth minus IPv6 header length 1919 veth_mtu="$(link_get_mtu "${ns_a}" veth_a)" 1920 vti6_mtu="$(link_get_mtu "${ns_a}" vti6_a)" 1921 if [ $((veth_mtu - vti6_mtu)) -ne 40 ]; then 1922 err " vti MTU ${vti6_mtu} is not veth MTU ${veth_mtu} minus IPv6 header length" 1923 return 1 1924 fi 1925} 1926 1927test_pmtu_vti4_link_add_mtu() { 1928 setup namespaces || return $ksft_skip 1929 1930 run_cmd ${ns_a} ip link add vti4_a type vti local ${veth4_a_addr} remote ${veth4_b_addr} key 10 1931 [ $? -ne 0 ] && err " vti not supported" && return $ksft_skip 1932 run_cmd ${ns_a} ip link del vti4_a 1933 1934 fail=0 1935 1936 min=68 1937 max=$((65535 - 20)) 1938 # Check invalid values first 1939 for v in $((min - 1)) $((max + 1)); do 1940 run_cmd ${ns_a} ip link add vti4_a mtu ${v} type vti local ${veth4_a_addr} remote ${veth4_b_addr} key 10 1941 # This can fail, or MTU can be adjusted to a proper value 1942 [ $? -ne 0 ] && continue 1943 mtu="$(link_get_mtu "${ns_a}" vti4_a)" 1944 if [ ${mtu} -lt ${min} -o ${mtu} -gt ${max} ]; then 1945 err " vti tunnel created with invalid MTU ${mtu}" 1946 fail=1 1947 fi 1948 run_cmd ${ns_a} ip link del vti4_a 1949 done 1950 1951 # Now check valid values 1952 for v in ${min} 1300 ${max}; do 1953 run_cmd ${ns_a} ip link add vti4_a mtu ${v} type vti local ${veth4_a_addr} remote ${veth4_b_addr} key 10 1954 mtu="$(link_get_mtu "${ns_a}" vti4_a)" 1955 run_cmd ${ns_a} ip link del vti4_a 1956 if [ "${mtu}" != "${v}" ]; then 1957 err " vti MTU ${mtu} doesn't match configured value ${v}" 1958 fail=1 1959 fi 1960 done 1961 1962 return ${fail} 1963} 1964 1965test_pmtu_vti6_link_add_mtu() { 1966 setup namespaces || return $ksft_skip 1967 1968 run_cmd ${ns_a} ip link add vti6_a type vti6 local ${veth6_a_addr} remote ${veth6_b_addr} key 10 1969 [ $? -ne 0 ] && err " vti6 not supported" && return $ksft_skip 1970 run_cmd ${ns_a} ip link del vti6_a 1971 1972 fail=0 1973 1974 min=68 # vti6 can carry IPv4 packets too 1975 max=$((65535 - 40)) 1976 # Check invalid values first 1977 for v in $((min - 1)) $((max + 1)); do 1978 run_cmd ${ns_a} ip link add vti6_a mtu ${v} type vti6 local ${veth6_a_addr} remote ${veth6_b_addr} key 10 1979 # This can fail, or MTU can be adjusted to a proper value 1980 [ $? -ne 0 ] && continue 1981 mtu="$(link_get_mtu "${ns_a}" vti6_a)" 1982 if [ ${mtu} -lt ${min} -o ${mtu} -gt ${max} ]; then 1983 err " vti6 tunnel created with invalid MTU ${v}" 1984 fail=1 1985 fi 1986 run_cmd ${ns_a} ip link del vti6_a 1987 done 1988 1989 # Now check valid values 1990 for v in 68 1280 1300 $((65535 - 40)); do 1991 run_cmd ${ns_a} ip link add vti6_a mtu ${v} type vti6 local ${veth6_a_addr} remote ${veth6_b_addr} key 10 1992 mtu="$(link_get_mtu "${ns_a}" vti6_a)" 1993 run_cmd ${ns_a} ip link del vti6_a 1994 if [ "${mtu}" != "${v}" ]; then 1995 err " vti6 MTU ${mtu} doesn't match configured value ${v}" 1996 fail=1 1997 fi 1998 done 1999 2000 return ${fail} 2001} 2002 2003test_pmtu_vti6_link_change_mtu() { 2004 setup namespaces || return $ksft_skip 2005 2006 run_cmd ${ns_a} ip link add dummy0 mtu 1500 type dummy 2007 [ $? -ne 0 ] && err " dummy not supported" && return $ksft_skip 2008 run_cmd ${ns_a} ip link add dummy1 mtu 3000 type dummy 2009 run_cmd ${ns_a} ip link set dummy0 up 2010 run_cmd ${ns_a} ip link set dummy1 up 2011 2012 run_cmd ${ns_a} ip addr add ${dummy6_0_prefix}1/${dummy6_mask} dev dummy0 2013 run_cmd ${ns_a} ip addr add ${dummy6_1_prefix}1/${dummy6_mask} dev dummy1 2014 2015 fail=0 2016 2017 # Create vti6 interface bound to device, passing MTU, check it 2018 run_cmd ${ns_a} ip link add vti6_a mtu 1300 type vti6 remote ${dummy6_0_prefix}2 local ${dummy6_0_prefix}1 2019 mtu="$(link_get_mtu "${ns_a}" vti6_a)" 2020 if [ ${mtu} -ne 1300 ]; then 2021 err " vti6 MTU ${mtu} doesn't match configured value 1300" 2022 fail=1 2023 fi 2024 2025 # Move to another device with different MTU, without passing MTU, check 2026 # MTU is adjusted 2027 run_cmd ${ns_a} ip link set vti6_a type vti6 remote ${dummy6_1_prefix}2 local ${dummy6_1_prefix}1 2028 mtu="$(link_get_mtu "${ns_a}" vti6_a)" 2029 if [ ${mtu} -ne $((3000 - 40)) ]; then 2030 err " vti MTU ${mtu} is not dummy MTU 3000 minus IPv6 header length" 2031 fail=1 2032 fi 2033 2034 # Move it back, passing MTU, check MTU is not overridden 2035 run_cmd ${ns_a} ip link set vti6_a mtu 1280 type vti6 remote ${dummy6_0_prefix}2 local ${dummy6_0_prefix}1 2036 mtu="$(link_get_mtu "${ns_a}" vti6_a)" 2037 if [ ${mtu} -ne 1280 ]; then 2038 err " vti6 MTU ${mtu} doesn't match configured value 1280" 2039 fail=1 2040 fi 2041 2042 return ${fail} 2043} 2044 2045check_command() { 2046 cmd=${1} 2047 2048 if ! which ${cmd} > /dev/null 2>&1; then 2049 err " missing required command: '${cmd}'" 2050 return 1 2051 fi 2052 return 0 2053} 2054 2055check_running() { 2056 pid=${1} 2057 cmd=${2} 2058 2059 [ "$(cat /proc/${pid}/cmdline 2>/dev/null | tr -d '\0')" = "{cmd}" ] 2060} 2061 2062test_cleanup_vxlanX_exception() { 2063 outer="${1}" 2064 encap="vxlan" 2065 ll_mtu=4000 2066 2067 check_command taskset || return $ksft_skip 2068 cpu_list=$(grep -m 2 processor /proc/cpuinfo | cut -d ' ' -f 2) 2069 2070 setup namespaces routing ${encap}${outer} || return $ksft_skip 2071 trace "${ns_a}" ${encap}_a "${ns_b}" ${encap}_b \ 2072 "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 2073 "${ns_b}" veth_B-R1 "${ns_r1}" veth_R1-B 2074 2075 # Create route exception by exceeding link layer MTU 2076 mtu "${ns_a}" veth_A-R1 $((${ll_mtu} + 1000)) 2077 mtu "${ns_r1}" veth_R1-A $((${ll_mtu} + 1000)) 2078 mtu "${ns_b}" veth_B-R1 ${ll_mtu} 2079 mtu "${ns_r1}" veth_R1-B ${ll_mtu} 2080 2081 mtu "${ns_a}" ${encap}_a $((${ll_mtu} + 1000)) 2082 mtu "${ns_b}" ${encap}_b $((${ll_mtu} + 1000)) 2083 2084 # Fill exception cache for multiple CPUs (2) 2085 # we can always use inner IPv4 for that 2086 for cpu in ${cpu_list}; do 2087 run_cmd taskset --cpu-list ${cpu} ${ns_a} ping -q -M want -i 0.1 -w 1 -s $((${ll_mtu} + 500)) ${tunnel4_b_addr} 2088 done 2089 2090 ${ns_a} ip link del dev veth_A-R1 & 2091 iplink_pid=$! 2092 for i in $(seq 1 20); do 2093 check_running ${iplink_pid} "iplinkdeldevveth_A-R1" || return 0 2094 sleep 0.1 2095 done 2096 err " can't delete veth device in a timely manner, PMTU dst likely leaked" 2097 return 1 2098} 2099 2100test_cleanup_ipv6_exception() { 2101 test_cleanup_vxlanX_exception 6 2102} 2103 2104test_cleanup_ipv4_exception() { 2105 test_cleanup_vxlanX_exception 4 2106} 2107 2108run_test() { 2109 ( 2110 tname="$1" 2111 tdesc="$2" 2112 2113 unset IFS 2114 2115 # Since cleanup() relies on variables modified by this subshell, it 2116 # has to run in this context. 2117 trap cleanup EXIT 2118 2119 if [ "$VERBOSE" = "1" ]; then 2120 printf "\n##########################################################################\n\n" 2121 fi 2122 2123 eval test_${tname} 2124 ret=$? 2125 2126 if [ $ret -eq 0 ]; then 2127 printf "TEST: %-60s [ OK ]\n" "${tdesc}" 2128 elif [ $ret -eq 1 ]; then 2129 printf "TEST: %-60s [FAIL]\n" "${tdesc}" 2130 if [ "${PAUSE_ON_FAIL}" = "yes" ]; then 2131 echo 2132 echo "Pausing. Hit enter to continue" 2133 read a 2134 fi 2135 err_flush 2136 exit 1 2137 elif [ $ret -eq $ksft_skip ]; then 2138 printf "TEST: %-60s [SKIP]\n" "${tdesc}" 2139 err_flush 2140 fi 2141 2142 return $ret 2143 ) 2144 ret=$? 2145 case $ret in 2146 0) 2147 all_skipped=false 2148 [ $exitcode -eq $ksft_skip ] && exitcode=0 2149 ;; 2150 $ksft_skip) 2151 [ $all_skipped = true ] && exitcode=$ksft_skip 2152 ;; 2153 *) 2154 all_skipped=false 2155 exitcode=1 2156 ;; 2157 esac 2158 2159 return $ret 2160} 2161 2162run_test_nh() { 2163 tname="$1" 2164 tdesc="$2" 2165 2166 USE_NH=yes 2167 run_test "${tname}" "${tdesc} - nexthop objects" 2168 USE_NH=no 2169} 2170 2171test_list_flush_ipv4_exception() { 2172 setup namespaces routing || return $ksft_skip 2173 trace "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 2174 "${ns_r1}" veth_R1-B "${ns_b}" veth_B-R1 \ 2175 "${ns_a}" veth_A-R2 "${ns_r2}" veth_R2-A \ 2176 "${ns_r2}" veth_R2-B "${ns_b}" veth_B-R2 2177 2178 dst_prefix1="${prefix4}.${b_r1}." 2179 dst2="${prefix4}.${b_r2}.1" 2180 2181 # Set up initial MTU values 2182 mtu "${ns_a}" veth_A-R1 2000 2183 mtu "${ns_r1}" veth_R1-A 2000 2184 mtu "${ns_r1}" veth_R1-B 1500 2185 mtu "${ns_b}" veth_B-R1 1500 2186 2187 mtu "${ns_a}" veth_A-R2 2000 2188 mtu "${ns_r2}" veth_R2-A 2000 2189 mtu "${ns_r2}" veth_R2-B 1500 2190 mtu "${ns_b}" veth_B-R2 1500 2191 2192 fail=0 2193 2194 # Add 100 addresses for veth endpoint on B reached by default A route 2195 for i in $(seq 100 199); do 2196 run_cmd ${ns_b} ip addr add "${dst_prefix1}${i}" dev veth_B-R1 2197 done 2198 2199 # Create 100 cached route exceptions for path via R1, one via R2. Note 2200 # that with IPv4 we need to actually cause a route lookup that matches 2201 # the exception caused by ICMP, in order to actually have a cached 2202 # route, so we need to ping each destination twice 2203 for i in $(seq 100 199); do 2204 run_cmd ${ns_a} ping -q -M want -i 0.1 -c 2 -s 1800 "${dst_prefix1}${i}" 2205 done 2206 run_cmd ${ns_a} ping -q -M want -i 0.1 -c 2 -s 1800 "${dst2}" 2207 2208 if [ "$(${ns_a} ip -oneline route list cache | wc -l)" -ne 101 ]; then 2209 err " can't list cached exceptions" 2210 fail=1 2211 fi 2212 2213 run_cmd ${ns_a} ip route flush cache 2214 pmtu1="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst_prefix}1)" 2215 pmtu2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst_prefix}2)" 2216 if [ -n "${pmtu1}" ] || [ -n "${pmtu2}" ] || \ 2217 [ -n "$(${ns_a} ip route list cache)" ]; then 2218 err " can't flush cached exceptions" 2219 fail=1 2220 fi 2221 2222 return ${fail} 2223} 2224 2225test_list_flush_ipv6_exception() { 2226 setup namespaces routing || return $ksft_skip 2227 trace "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 2228 "${ns_r1}" veth_R1-B "${ns_b}" veth_B-R1 \ 2229 "${ns_a}" veth_A-R2 "${ns_r2}" veth_R2-A \ 2230 "${ns_r2}" veth_R2-B "${ns_b}" veth_B-R2 2231 2232 dst_prefix1="${prefix6}:${b_r1}::" 2233 dst2="${prefix6}:${b_r2}::1" 2234 2235 # Set up initial MTU values 2236 mtu "${ns_a}" veth_A-R1 2000 2237 mtu "${ns_r1}" veth_R1-A 2000 2238 mtu "${ns_r1}" veth_R1-B 1500 2239 mtu "${ns_b}" veth_B-R1 1500 2240 2241 mtu "${ns_a}" veth_A-R2 2000 2242 mtu "${ns_r2}" veth_R2-A 2000 2243 mtu "${ns_r2}" veth_R2-B 1500 2244 mtu "${ns_b}" veth_B-R2 1500 2245 2246 fail=0 2247 2248 # Add 100 addresses for veth endpoint on B reached by default A route 2249 for i in $(seq 100 199); do 2250 run_cmd ${ns_b} ip addr add "${dst_prefix1}${i}" dev veth_B-R1 2251 done 2252 2253 # Create 100 cached route exceptions for path via R1, one via R2 2254 for i in $(seq 100 199); do 2255 run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s 1800 "${dst_prefix1}${i}" 2256 done 2257 run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s 1800 "${dst2}" 2258 if [ "$(${ns_a} ip -oneline -6 route list cache | wc -l)" -ne 101 ]; then 2259 err " can't list cached exceptions" 2260 fail=1 2261 fi 2262 2263 run_cmd ${ns_a} ip -6 route flush cache 2264 pmtu1="$(route_get_dst_pmtu_from_exception "${ns_a}" "${dst_prefix1}100")" 2265 pmtu2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 2266 if [ -n "${pmtu1}" ] || [ -n "${pmtu2}" ] || \ 2267 [ -n "$(${ns_a} ip -6 route list cache)" ]; then 2268 err " can't flush cached exceptions" 2269 fail=1 2270 fi 2271 2272 return ${fail} 2273} 2274 2275test_pmtu_ipvX_route_change() { 2276 family=${1} 2277 2278 setup namespaces routing || return 2 2279 trace "${ns_a}" veth_A-R1 "${ns_r1}" veth_R1-A \ 2280 "${ns_r1}" veth_R1-B "${ns_b}" veth_B-R1 \ 2281 "${ns_a}" veth_A-R2 "${ns_r2}" veth_R2-A \ 2282 "${ns_r2}" veth_R2-B "${ns_b}" veth_B-R2 2283 2284 if [ ${family} -eq 4 ]; then 2285 ping=ping 2286 dst1="${prefix4}.${b_r1}.1" 2287 dst2="${prefix4}.${b_r2}.1" 2288 gw="${prefix4}.${a_r1}.2" 2289 else 2290 ping=${ping6} 2291 dst1="${prefix6}:${b_r1}::1" 2292 dst2="${prefix6}:${b_r2}::1" 2293 gw="${prefix6}:${a_r1}::2" 2294 fi 2295 2296 # Set up initial MTU values 2297 mtu "${ns_a}" veth_A-R1 2000 2298 mtu "${ns_r1}" veth_R1-A 2000 2299 mtu "${ns_r1}" veth_R1-B 1400 2300 mtu "${ns_b}" veth_B-R1 1400 2301 2302 mtu "${ns_a}" veth_A-R2 2000 2303 mtu "${ns_r2}" veth_R2-A 2000 2304 mtu "${ns_r2}" veth_R2-B 1500 2305 mtu "${ns_b}" veth_B-R2 1500 2306 2307 # Create route exceptions 2308 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s 1800 ${dst1} 2309 run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s 1800 ${dst2} 2310 2311 # Check that exceptions have been created with the correct PMTU 2312 pmtu_1="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst1})" 2313 check_pmtu_value "1400" "${pmtu_1}" "exceeding MTU" || return 1 2314 pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})" 2315 check_pmtu_value "1500" "${pmtu_2}" "exceeding MTU" || return 1 2316 2317 # Replace the route from A to R1 2318 run_cmd ${ns_a} ip route change default via ${gw} 2319 2320 # Delete the device in A 2321 run_cmd ${ns_a} ip link del "veth_A-R1" 2322} 2323 2324test_pmtu_ipv4_route_change() { 2325 test_pmtu_ipvX_route_change 4 2326} 2327 2328test_pmtu_ipv6_route_change() { 2329 test_pmtu_ipvX_route_change 6 2330} 2331 2332usage() { 2333 echo 2334 echo "$0 [OPTIONS] [TEST]..." 2335 echo "If no TEST argument is given, all tests will be run." 2336 echo 2337 echo "Options" 2338 echo " --trace: capture traffic to TEST_INTERFACE.pcap" 2339 echo 2340 echo "Available tests${tests}" 2341 exit 1 2342} 2343 2344################################################################################ 2345# 2346exitcode=0 2347desc=0 2348all_skipped=true 2349 2350while getopts :ptv o 2351do 2352 case $o in 2353 p) PAUSE_ON_FAIL=yes;; 2354 v) VERBOSE=1;; 2355 t) if which tcpdump > /dev/null 2>&1; then 2356 TRACING=1 2357 else 2358 echo "=== tcpdump not available, tracing disabled" 2359 fi 2360 ;; 2361 *) usage;; 2362 esac 2363done 2364shift $(($OPTIND-1)) 2365 2366IFS=" 2367" 2368 2369for arg do 2370 # Check first that all requested tests are available before running any 2371 command -v > /dev/null "test_${arg}" || { echo "=== Test ${arg} not found"; usage; } 2372done 2373 2374trap cleanup EXIT 2375 2376# start clean 2377cleanup 2378 2379HAVE_NH=no 2380ip nexthop ls >/dev/null 2>&1 2381[ $? -eq 0 ] && HAVE_NH=yes 2382 2383name="" 2384desc="" 2385rerun_nh=0 2386for t in ${tests}; do 2387 [ "${name}" = "" ] && name="${t}" && continue 2388 [ "${desc}" = "" ] && desc="${t}" && continue 2389 2390 if [ "${HAVE_NH}" = "yes" ]; then 2391 rerun_nh="${t}" 2392 fi 2393 2394 run_this=1 2395 for arg do 2396 [ "${arg}" != "${arg#--*}" ] && continue 2397 [ "${arg}" = "${name}" ] && run_this=1 && break 2398 run_this=0 2399 done 2400 if [ $run_this -eq 1 ]; then 2401 run_test "${name}" "${desc}" 2402 # if test was skipped no need to retry with nexthop objects 2403 [ $? -eq $ksft_skip ] && rerun_nh=0 2404 2405 if [ "${rerun_nh}" = "1" ]; then 2406 run_test_nh "${name}" "${desc}" 2407 fi 2408 fi 2409 name="" 2410 desc="" 2411 rerun_nh=0 2412done 2413 2414exit ${exitcode} 2415