- dev - passt

[PATCH v2 0/4] Add vhost-user support to passt. (part 3)
by Laurent Vivier 15 Aug '24

15 Aug '24

This series of patches adds vhost-user support to passt and then allows passt to connect to QEMU network backend using virtqueue rather than a socket. With QEMU, rather than using to connect: -netdev stream,id=s,server=off,addr.type=unix,addr.path=/tmp/passt_1.socket we will use: -chardev socket,id=chr0,path=/tmp/passt_1.socket -netdev vhost-user,id=netdev0,chardev=chr0 -device virtio-net,netdev=netdev0 -object memory-backend-memfd,id=memfd0,share=on,size=$RAMSIZE -numa node,memdev=memfd0 The memory backend is needed to share data between passt and QEMU. Performance comparison between "-netdev stream" and "-netdev vhost-user": $ iperf3 -c localhost -p 10001 -t 60 -6 -u -b 50G socket: [ 5] 0.00-60.05 sec 95.6 GBytes 13.7 Gbits/sec 0.017 ms 6998988/10132413 (69%) receiver vhost-user: [ 5] 0.00-60.04 sec 237 GBytes 33.9 Gbits/sec 0.006 ms 53673/7813770 (0.69%) receiver $ iperf3 -c localhost -p 10001 -t 60 -4 -u -b 50G socket: [ 5] 0.00-60.05 sec 98.9 GBytes 14.1 Gbits/sec 0.018 ms 6260735/9501832 (66%) receiver vhost-user: [ 5] 0.00-60.05 sec 235 GBytes 33.7 Gbits/sec 0.008 ms 37581/7752699 (0.48%) receiver $ iperf3 -c localhost -p 10001 -t 60 -6 socket: [ 5] 0.00-60.00 sec 17.3 GBytes 2.48 Gbits/sec 0 sender [ 5] 0.00-60.06 sec 17.3 GBytes 2.48 Gbits/sec receiver vhost-user: [ 5] 0.00-60.00 sec 191 GBytes 27.4 Gbits/sec 0 sender [ 5] 0.00-60.05 sec 191 GBytes 27.3 Gbits/sec receiver $ iperf3 -c localhost -p 10001 -t 60 -4 socket: [ 5] 0.00-60.00 sec 15.6 GBytes 2.24 Gbits/sec 0 sender [ 5] 0.00-60.06 sec 15.6 GBytes 2.24 Gbits/sec receiver vhost-user: [ 5] 0.00-60.00 sec 189 GBytes 27.1 Gbits/sec 0 sender [ 5] 0.00-60.04 sec 189 GBytes 27.0 Gbits/sec receiver v2: - remove PATCH 4 - rewrite PATCH 2 and 3 to follow passt coding style - move some code from PATCH 3 to PATCH 4 (previously PATCH 5) - partially addressed David's comment on PATCH 5 Laurent Vivier (4): packet: replace struct desc by struct iovec vhost-user: introduce virtio API vhost-user: introduce vhost-user API vhost-user: add vhost-user Makefile | 4 +- checksum.c | 1 - conf.c | 24 +- iov.c | 1 - isolation.c | 15 +- packet.c | 97 ++-- packet.h | 16 +- passt.c | 16 +- passt.h | 10 + pcap.c | 1 - tap.c | 114 ++++- tap.h | 5 +- tcp.c | 17 +- tcp_vu.c | 560 +++++++++++++++++++++ tcp_vu.h | 12 + udp.c | 54 +- udp_internal.h | 39 ++ udp_vu.c | 240 +++++++++ udp_vu.h | 11 + util.h | 11 + vhost_user.c | 1273 ++++++++++++++++++++++++++++++++++++++++++++++++ vhost_user.h | 197 ++++++++ virtio.c | 605 +++++++++++++++++++++++ virtio.h | 190 ++++++++ 24 files changed, 3392 insertions(+), 121 deletions(-) create mode 100644 tcp_vu.c create mode 100644 tcp_vu.h create mode 100644 udp_internal.h create mode 100644 udp_vu.c create mode 100644 udp_vu.h create mode 100644 vhost_user.c create mode 100644 vhost_user.h create mode 100644 virtio.c create mode 100644 virtio.h -- 2.45.2

3 12

[PATCH] test: Speed up by cutting on eye candy and performance test duration
by Stefano Brivio 15 Aug '24

15 Aug '24

We have a number of delays when we switch to new layouts that were added to make the tests visually easier to follow, together with blinking status bars. Shorten the delays and avoid blinking the status bar if $FAST is set to 1 (no demo mode). Shorten delays in busy loops to 10ms, instead of 100ms, and skip the one-second fixed delay when we wait for the status of a command. Cut the duration of throughput and latency tests to one second, down from ten. Somewhat surprisingly, the results we get are rather consistent, and not significantly different from what we'd get with 10 seconds. This, together with Podman's commit 20f3e8909e3a ("test/system: pasta_test_do add explicit port check"), cuts the time needed on my setup for full test run from approximately 37 minutes to...: $ time ./run [exited] PASS: 165, FAIL: 0 Log at /home/sbrivio/passt/test/test_logs/test.log real 15m34.253s user 0m0.011s sys 0m0.011s Signed-off-by: Stefano Brivio <sbrivio(a)redhat.com> --- test/lib/layout | 14 +++++++------- test/lib/term | 31 +++++++++++++++---------------- test/lib/test | 2 +- test/pasta_options/log_to_file | 4 ++-- test/perf/passt_tcp | 18 +++++++++--------- test/perf/passt_udp | 2 +- test/perf/pasta_tcp | 34 +++++++++++++++++----------------- test/perf/pasta_udp | 2 +- 8 files changed, 53 insertions(+), 54 deletions(-) diff --git a/test/lib/layout b/test/lib/layout index f9a1cf1..4d03572 100644 --- a/test/lib/layout +++ b/test/lib/layout @@ -15,7 +15,7 @@ # layout_pasta() - Panes for host, pasta, and separate one for namespace layout_pasta() { - sleep 3 + sleep 1 tmux kill-pane -a -t 0 cmd_write 0 clear @@ -46,7 +46,7 @@ layout_pasta() { # layout_passt() - Panes for host, passt, and guest layout_passt() { - sleep 3 + sleep 1 tmux kill-pane -a -t 0 cmd_write 0 clear @@ -77,7 +77,7 @@ layout_passt() { # layout_passt_in_pasta() - Host, passt within pasta, namespace and guest layout_passt_in_pasta() { - sleep 3 + sleep 1 tmux kill-pane -a -t 0 cmd_write 0 clear @@ -113,7 +113,7 @@ layout_passt_in_pasta() { # layout_two_guests() - Two guest panes, two passt panes, plus host and log layout_two_guests() { - sleep 3 + sleep 1 tmux kill-pane -a -t 0 cmd_write 0 clear @@ -152,7 +152,7 @@ layout_two_guests() { # layout_demo_pasta() - Four panes for pasta demo layout_demo_pasta() { - sleep 3 + sleep 1 cmd_write 0 cd ${BASEPATH} cmd_write 0 clear @@ -188,7 +188,7 @@ layout_demo_pasta() { # layout_demo_passt() - Four panes for passt demo layout_demo_passt() { - sleep 3 + sleep 1 cmd_write 0 cd ${BASEPATH} cmd_write 0 clear @@ -224,7 +224,7 @@ layout_demo_passt() { # layout_demo_podman() - Four panes for pasta demo with Podman layout_demo_podman() { - sleep 3 + sleep 1 cmd_write 0 cd ${BASEPATH} cmd_write 0 clear diff --git a/test/lib/term b/test/lib/term index 262937e..3834092 100755 --- a/test/lib/term +++ b/test/lib/term @@ -97,7 +97,6 @@ display_delay() { switch_pane() { tmux select-pane -t ${1} PR_DELAY=${PR_DELAY_INIT} - display_delay "0.2" } # cmd_write() - Write a command to a pane, letter by letter, and execute it @@ -199,7 +198,7 @@ pane_run() { # $1: Pane name pane_wait() { __lc="$(echo "${1}" | tr [A-Z] [a-z])" - sleep 0.1 || sleep 1 + sleep 0.01 || sleep 1 __done=0 while @@ -207,7 +206,7 @@ pane_wait() { case ${__l} in *"$ " | *"# ") return ;; esac - do sleep 0.1 || sleep 1; done + do sleep 0.01 || sleep 1; done } # pane_parse() - Print last line, @EMPTY@ if command had no output @@ -231,7 +230,7 @@ pane_status() { __status="$(pane_parse "${1}")" while ! [ "${__status}" -eq "${__status}" ] 2>/dev/null; do - sleep 1 + sleep 0.01 || sleep 1 pane_run "${1}" 'echo $?' pane_wait "${1}" __status="$(pane_parse "${1}")" @@ -383,6 +382,16 @@ info_check_failed() { printf " < failed.\n" >> "${LOGFILE}" } +# status_bar_blink() - Make status bar blink +status_bar_blink() { + for i in `seq 1 3`; do + tmux set status-right-style 'bg=colour1 fg=colour196 bold' + sleep 0.1 || sleep 1 + tmux set status-right-style 'bg=colour1 fg=colour233 bold' + sleep 0.1 || sleep 1 + done +} + # info_passed() - Display, log, and make status bar blink when a test passes info_passed() { switch_pane ${PANE_INFO} @@ -391,12 +400,7 @@ info_passed() { log "...passed." log - for i in `seq 1 3`; do - tmux set status-right-style 'bg=colour1 fg=colour2 bold' - sleep "0.1" - tmux set status-right-style 'bg=colour1 fg=colour233 bold' - sleep "0.1" - done + [ ${FAST} -eq 1 ] || status_bar_blink } # info_failed() - Display, log, and make status bar blink when a test passes @@ -407,12 +411,7 @@ info_failed() { log "...failed." log - for i in `seq 1 3`; do - tmux set status-right-style 'bg=colour1 fg=colour196 bold' - sleep "0.1" - tmux set status-right-style 'bg=colour1 fg=colour233 bold' - sleep "0.1" - done + [ ${FAST} -eq 1 ] || status_bar_blink pause_continue \ "Press any key to pause test session" \ diff --git a/test/lib/test b/test/lib/test index c525f8e..e6726be 100755 --- a/test/lib/test +++ b/test/lib/test @@ -33,7 +33,7 @@ test_iperf3k() { pane_or_context_run "${__sctx}" 'kill -INT $(cat s.pid); rm s.pid' - sleep 3 # Wait for kernel to free up ports + sleep 1 # Wait for kernel to free up ports } # test_iperf3() - Ugly helper for iperf3 directive diff --git a/test/pasta_options/log_to_file b/test/pasta_options/log_to_file index fe50e50..3ead06c 100644 --- a/test/pasta_options/log_to_file +++ b/test/pasta_options/log_to_file @@ -19,7 +19,7 @@ sleep 1 endef def flood_log_client -host tcp_crr --nolog -P 10001 -C 10002 -6 -c -H ::1 +host tcp_crr --nolog -l1 -P 10001 -C 10002 -6 -c -H ::1 endef def check_log_size_mountns @@ -42,7 +42,7 @@ pout PID2 echo $! check head -1 __LOG_FILE__ | grep '^pasta .* [(]__PID2__[)]$' test Maximum log size -passtb ./pasta --config-net -d -f -l __LOG_FILE__ --log-size $((100 * 1024)) -- sh -c 'while true; do tcp_crr --nolog -P 10001 -C 10002 -6; done' +passtb ./pasta --config-net -d -f -l __LOG_FILE__ --log-size $((100 * 1024)) -- sh -c 'while true; do tcp_crr --nolog -l1 -P 10001 -C 10002 -6; done' sleep 1 flood_log_client diff --git a/test/perf/passt_tcp b/test/perf/passt_tcp index 14343cb..695479f 100644 --- a/test/perf/passt_tcp +++ b/test/perf/passt_tcp @@ -38,7 +38,7 @@ hout FREQ_CPUFREQ (echo "scale=1"; printf '( %i + 10^5 / 2 ) / 10^6\n' $(cat /sy hout FREQ [ -n "__FREQ_CPUFREQ__" ] && echo __FREQ_CPUFREQ__ || echo __FREQ_PROCFS__ set THREADS 4 -set TIME 10 +set TIME 1 set OMIT 0.1 set OPTS -Z -P __THREADS__ -l 1M -O__OMIT__ @@ -75,7 +75,7 @@ lat - lat - lat - nsb tcp_rr --nolog -6 -gout LAT tcp_rr --nolog -6 -c -H __GW6__%__IFNAME__ | sed -n 's/^throughput=$.*$/\1/p' +gout LAT tcp_rr --nolog -l1 -6 -c -H __GW6__%__IFNAME__ | sed -n 's/^throughput=$.*$/\1/p' lat __LAT__ 200 150 tl TCP CRR latency over IPv6: guest to host @@ -85,7 +85,7 @@ lat - lat - lat - nsb tcp_crr --nolog -6 -gout LAT tcp_crr --nolog -6 -c -H __GW6__%__IFNAME__ | sed -n 's/^throughput=$.*$/\1/p' +gout LAT tcp_crr --nolog -l1 -6 -c -H __GW6__%__IFNAME__ | sed -n 's/^throughput=$.*$/\1/p' lat __LAT__ 500 400 tr TCP throughput over IPv4: guest to host @@ -119,7 +119,7 @@ lat - lat - lat - nsb tcp_rr --nolog -4 -gout LAT tcp_rr --nolog -4 -c -H __GW__ | sed -n 's/^throughput=$.*$/\1/p' +gout LAT tcp_rr --nolog -l1 -4 -c -H __GW__ | sed -n 's/^throughput=$.*$/\1/p' lat __LAT__ 200 150 tl TCP CRR latency over IPv4: guest to host @@ -129,7 +129,7 @@ lat - lat - lat - nsb tcp_crr --nolog -4 -gout LAT tcp_crr --nolog -4 -c -H __GW__ | sed -n 's/^throughput=$.*$/\1/p' +gout LAT tcp_crr --nolog -l1 -4 -c -H __GW__ | sed -n 's/^throughput=$.*$/\1/p' lat __LAT__ 500 400 tr TCP throughput over IPv6: host to guest @@ -153,7 +153,7 @@ lat - lat - guestb tcp_rr --nolog -P 10001 -C 10011 -6 sleep 1 -nsout LAT tcp_rr --nolog -P 10001 -C 10011 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_rr --nolog -l1 -P 10001 -C 10011 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' lat __LAT__ 200 150 tl TCP CRR latency over IPv6: host to guest @@ -164,7 +164,7 @@ lat - lat - guestb tcp_crr --nolog -P 10001 -C 10011 -6 sleep 1 -nsout LAT tcp_crr --nolog -P 10001 -C 10011 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_crr --nolog -l1 -P 10001 -C 10011 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' lat __LAT__ 500 350 @@ -189,7 +189,7 @@ lat - lat - guestb tcp_rr --nolog -P 10001 -C 10011 -4 sleep 1 -nsout LAT tcp_rr --nolog -P 10001 -C 10011 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_rr --nolog -l1 -P 10001 -C 10011 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' lat __LAT__ 200 150 tl TCP CRR latency over IPv6: host to guest @@ -200,7 +200,7 @@ lat - lat - guestb tcp_crr --nolog -P 10001 -C 10011 -4 sleep 1 -nsout LAT tcp_crr --nolog -P 10001 -C 10011 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_crr --nolog -l1 -P 10001 -C 10011 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' lat __LAT__ 500 300 te diff --git a/test/perf/passt_udp b/test/perf/passt_udp index 8919280..f25c903 100644 --- a/test/perf/passt_udp +++ b/test/perf/passt_udp @@ -31,7 +31,7 @@ hout FREQ_CPUFREQ (echo "scale=1"; printf '( %i + 10^5 / 2 ) / 10^6\n' $(cat /sy hout FREQ [ -n "__FREQ_CPUFREQ__" ] && echo __FREQ_CPUFREQ__ || echo __FREQ_PROCFS__ set THREADS 2 -set TIME 10 +set TIME 1 set OPTS -u -P __THREADS__ --pacing-timer 1000 info Throughput in Gbps, latency in µs, __THREADS__ threads at __FREQ__ GHz diff --git a/test/perf/pasta_tcp b/test/perf/pasta_tcp index 8d2f911..a443f5a 100644 --- a/test/perf/pasta_tcp +++ b/test/perf/pasta_tcp @@ -22,7 +22,7 @@ ns /sbin/sysctl -w net.ipv4.tcp_timestamps=0 set THREADS 4 -set TIME 10 +set TIME 1 set OMIT 0.1 set OPTS -Z -w 4M -l 1M -P __THREADS__ -O__OMIT__ @@ -46,13 +46,13 @@ iperf3k host tl TCP RR latency over IPv6: ns to host hostb tcp_rr --nolog -P 10003 -C 10013 -6 -nsout LAT tcp_rr --nolog -P 10003 -C 10013 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_rr --nolog -l1 -P 10003 -C 10013 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' hostw lat __LAT__ 150 100 tl TCP CRR latency over IPv6: ns to host hostb tcp_crr --nolog -P 10003 -C 10013 -6 -nsout LAT tcp_crr --nolog -P 10003 -C 10013 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_crr --nolog -l1 -P 10003 -C 10013 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' hostw lat __LAT__ 500 350 @@ -67,13 +67,13 @@ iperf3k host tl TCP RR latency over IPv4: ns to host hostb tcp_rr --nolog -P 10003 -C 10013 -4 -nsout LAT tcp_rr --nolog -P 10003 -C 10013 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_rr --nolog -l1 -P 10003 -C 10013 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' hostw lat __LAT__ 150 100 tl TCP CRR latency over IPv4: ns to host hostb tcp_crr --nolog -P 10003 -C 10013 -4 -nsout LAT tcp_crr --nolog -P 10003 -C 10013 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_crr --nolog -l1 -P 10003 -C 10013 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' hostw lat __LAT__ 500 350 @@ -87,13 +87,13 @@ iperf3k ns tl TCP RR latency over IPv6: host to ns nsb tcp_rr --nolog -P 10002 -C 10012 -6 -hout LAT tcp_rr --nolog -P 10002 -C 10012 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' +hout LAT tcp_rr --nolog -l1 -P 10002 -C 10012 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' nsw lat __LAT__ 150 100 tl TCP CRR latency over IPv6: host to ns nsb tcp_crr --nolog -P 10002 -C 10012 -6 -hout LAT tcp_crr --nolog -P 10002 -C 10012 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' +hout LAT tcp_crr --nolog -l1 -P 10002 -C 10012 -6 -c -H ::1 | sed -n 's/^throughput=$.*$/\1/p' nsw lat __LAT__ 1000 700 @@ -108,13 +108,13 @@ iperf3k ns tl TCP RR latency over IPv4: host to ns nsb tcp_rr --nolog -P 10002 -C 10012 -4 -hout LAT tcp_rr --nolog -P 10002 -C 10012 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' +hout LAT tcp_rr --nolog -l1 -P 10002 -C 10012 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' nsw lat __LAT__ 150 100 tl TCP CRR latency over IPv4: host to ns nsb tcp_crr --nolog -P 10002 -C 10012 -4 -hout LAT tcp_crr --nolog -P 10002 -C 10012 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' +hout LAT tcp_crr --nolog -l1 -P 10002 -C 10012 -4 -c -H 127.0.0.1 | sed -n 's/^throughput=$.*$/\1/p' nsw lat __LAT__ 1000 700 @@ -156,7 +156,7 @@ lat - lat - lat - hostb tcp_rr --nolog -P 10003 -C 10013 -6 -nsout LAT tcp_rr --nolog -P 10003 -C 10013 -6 -c -H __GW6__%__IFNAME__ | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_rr --nolog -l1 -P 10003 -C 10013 -6 -c -H __GW6__%__IFNAME__ | sed -n 's/^throughput=$.*$/\1/p' hostw lat __LAT__ 150 100 @@ -165,7 +165,7 @@ lat - lat - lat - hostb tcp_crr --nolog -P 10003 -C 10013 -6 -nsout LAT tcp_crr --nolog -P 10003 -C 10013 -6 -c -H __GW6__%__IFNAME__ | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_crr --nolog -l1 -P 10003 -C 10013 -6 -c -H __GW6__%__IFNAME__ | sed -n 's/^throughput=$.*$/\1/p' hostw lat __LAT__ 1500 500 @@ -193,7 +193,7 @@ lat - lat - lat - hostb tcp_rr --nolog -P 10003 -C 10013 -4 -nsout LAT tcp_rr --nolog -P 10003 -C 10013 -4 -c -H __GW__ | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_rr --nolog -l1 -P 10003 -C 10013 -4 -c -H __GW__ | sed -n 's/^throughput=$.*$/\1/p' hostw lat __LAT__ 150 100 @@ -202,7 +202,7 @@ lat - lat - lat - hostb tcp_crr --nolog -P 10003 -C 10013 -4 -nsout LAT tcp_crr --nolog -P 10003 -C 10013 -4 -c -H __GW__ | sed -n 's/^throughput=$.*$/\1/p' +nsout LAT tcp_crr --nolog -l1 -P 10003 -C 10013 -4 -c -H __GW__ | sed -n 's/^throughput=$.*$/\1/p' hostw lat __LAT__ 1500 500 @@ -224,7 +224,7 @@ lat - lat - lat - nsb tcp_rr --nolog -P 10002 -C 10012 -6 -hout LAT tcp_rr --nolog -P 10002 -C 10012 -6 -c -H __ADDR6__ | sed -n 's/^throughput=$.*$/\1/p' +hout LAT tcp_rr --nolog -l1 -P 10002 -C 10012 -6 -c -H __ADDR6__ | sed -n 's/^throughput=$.*$/\1/p' nsw lat __LAT__ 150 100 @@ -234,7 +234,7 @@ lat - lat - sleep 1 nsb tcp_crr --nolog -P 10002 -C 10012 -6 -hout LAT tcp_crr --nolog -P 10002 -C 10012 -6 -c -H __ADDR6__ | sed -n 's/^throughput=$.*$/\1/p' +hout LAT tcp_crr --nolog -l1 -P 10002 -C 10012 -6 -c -H __ADDR6__ | sed -n 's/^throughput=$.*$/\1/p' nsw lat __LAT__ 5000 10000 @@ -256,7 +256,7 @@ lat - lat - lat - nsb tcp_rr --nolog -P 10002 -C 10012 -4 -hout LAT tcp_rr --nolog -P 10002 -C 10012 -4 -c -H __ADDR__ | sed -n 's/^throughput=$.*$/\1/p' +hout LAT tcp_rr --nolog -l1 -P 10002 -C 10012 -4 -c -H __ADDR__ | sed -n 's/^throughput=$.*$/\1/p' nsw lat __LAT__ 150 100 @@ -266,7 +266,7 @@ lat - lat - sleep 1 nsb tcp_crr --nolog -P 10002 -C 10012 -4 -hout LAT tcp_crr --nolog -P 10002 -C 10012 -4 -c -H __ADDR__ | sed -n 's/^throughput=$.*$/\1/p' +hout LAT tcp_crr --nolog -l1 -P 10002 -C 10012 -4 -c -H __ADDR__ | sed -n 's/^throughput=$.*$/\1/p' nsw lat __LAT__ 5000 10000 diff --git a/test/perf/pasta_udp b/test/perf/pasta_udp index 6acbfd3..9fed62e 100644 --- a/test/perf/pasta_udp +++ b/test/perf/pasta_udp @@ -21,7 +21,7 @@ hout FREQ_CPUFREQ (echo "scale=1"; printf '( %i + 10^5 / 2 ) / 10^6\n' $(cat /sy hout FREQ [ -n "__FREQ_CPUFREQ__" ] && echo __FREQ_CPUFREQ__ || echo __FREQ_PROCFS__ set THREADS 1 -set TIME 10 +set TIME 1 set OPTS -u -P __THREADS__ info Throughput in Gbps, latency in µs, one thread at __FREQ__ GHz -- 2.43.0

2 1

[net] tcp: add SO_PEEK_OFF socket option tor TCPv6
by jmaloy＠redhat.com 14 Aug '24

14 Aug '24

From: Jon Maloy <jmaloy(a)redhat.com> When we added the SO_PEEK_OFF socket option to TCP we forgot to add it even for TCP on IPv6. We do that here. Fixes: 05ea491641d3 ("tcp: add support for SO_PEEK_OFF socket option") Signed-off-by: Jon Maloy <jmaloy(a)redhat.com> --- net/ipv6/af_inet6.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 90d2c7e3f5e9..ba69b86f1c7d 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -708,6 +708,7 @@ const struct proto_ops inet6_stream_ops = { .splice_eof = inet_splice_eof, .sendmsg_locked = tcp_sendmsg_locked, .splice_read = tcp_splice_read, + .set_peek_off = sk_set_peek_off, .read_sock = tcp_read_sock, .read_skb = tcp_read_skb, .peek_len = tcp_peek_len, -- 2.45.2

5 4

passt: new version 2024_08_14.61c0b0d available
by Stefano Brivio 14 Aug '24

14 Aug '24

The new version with tag 2024_08_14.61c0b0d includes the following changes: 61c0b0d flow: Don't crash if guest attempts to connect to port 0 baba284 conf: Don't ignore -t and -u options after -D c16141e ndp.c: Turn NDP responder into more declarative implementation f6d5a52 conf: Delay handling -D option until after addresses are configured 86bdd96 Correct inaccurate comments on ip[46]_ctx::addr fecb1b6 log: Don't prefix message with timestamp on --debug if it's a continuation baccfb9 conf: Stop parsing options at first non-option argument 09603ca passt, util: Close any open file that the parent might have leaked 755f9fd nstool: Propagate SIGTERM to processes executed in the namespace 5ca61c2 nstool: Fix some trivial typos a628cb9 log: Avoid duplicate calls to logtime() 2c7558d log: Handle errors from clock_gettime() b91bae1 log: Correct formatting of timestamps 95569e4 util: Some corrections for timespec_diff_us fbb0c95 conf, pasta: Make -g and -a skip route/addresses copy for matching IP version only https://passt.top/passt/log/?qt=range&q=2024_08_06.ee36266..2024_08_14.61c0… Packages: - Alpine Linux: https://pkgs.alpinelinux.org/packages?name=passt - Arch Linux: https://www.archlinux.org/packages/extra/x86_64/passt/ https://archlinuxarm.org/packages/aarch64/passt https://archlinuxarm.org/packages/armv7h/passt - Chimera: https://pkgs.chimera-linux.org/packages?name=passt - Clear Linux: https://github.com/clearlinux-pkgs/passt/ - Debian tracker: https://tracker.debian.org/pkg/passt - Copr (CentOS Stream, EPEL, Fedora, Mageia): https://copr.fedorainfracloud.org/coprs/sbrivio/passt/build/7907497/ permanent mirror: https://passt.top/builds/copr/0^20240814.g61c0b0d/ - Fedora updates: https://bodhi.fedoraproject.org/updates/?packages=passt - Gentoo versions: https://packages.gentoo.org/packages/net-misc/passt - GNU Guix: https://packages.guix.gnu.org/packages/passt/ - Homebrew: https://formulae.brew.sh/formula/passt - NixOS: https://github.com/NixOS/nixpkgs/tree/nixos-unstable/pkgs/by-name/pa/passt - openSUSE: https://software.opensuse.org/package/passt - PLD Linux: https://git.pld-linux.org/cgi-bin/gitweb.cgi?p=packages/passt.git - Solus: https://github.com/getsolus/packages/tree/main/packages/p/passt - Ubuntu tracker: https://launchpad.net/ubuntu/+source/passt - Void Linux: https://voidlinux.org/packages/?q=passt - Static builds: - Package for other RPM-based distributions, x86_64 only: https://passt.top/builds/latest/x86_64/passt-g61c0b0d-1.x86_64.rpm - x86_64 static binaries: https://passt.top/builds/latest/x86_64/ - Debian package, from x86_64 static build: https://passt.top/builds/latest/x86_64/passt_61c0b0d-1_all.deb -- Stefano

1 0

[PATCH 00/16] Cleanups and fixes related to "our" addresses
by David Gibson 14 Aug '24

14 Aug '24

There are some places where we have addresses that are "ours" in the sense that they're local to passt on at least one interface. But in some cases it wasn't clear which addresses those were or how to use them. Make a number of renames, cleanups and small fixes related to that. ..and also an assortment of slightly related things that I encountered along the way. Note that 1/16 is an important fix for a bug I introduced in the last series I sent. For the rest, apply as many as you're happy with and I'll respin what's left as necessary. David Gibson (16): conf: Don't ignore -t and -u options after -D treewide: Use "our address" instead of "forwarding address" util: Helper for formatting MAC addresses treewide: Rename MAC address fields for clarity treewide: Use struct assignment instead of memcpy() for IP addresses conf: Use array indices rather than pointers for DNS array slots conf: More accurately count entries added in get_dns() conf: Move DNS array bounds checks into add_dns[46] conf: Move adding of a nameserver from resolv.conf into subfunction conf: Correct setting of dns_match address in add_dns6() conf: Treat --dns addresses as guest visible addresses conf: Remove incorrect initialisation of addr_ll_seen util: Correct sock_l4() binding for link local addresses treewide: Change misleading 'addr_ll' name Clarify which addresses in ip[46]_ctx are meaningful where Initialise our_tap_ll to ip6.gw when suitable arp.c | 4 +- conf.c | 181 ++++++++++++++++++++++++++++--------------------- dhcp.c | 5 +- dhcpv6.c | 21 +++--- flow.c | 74 ++++++++++---------- flow.h | 18 ++--- fwd.c | 70 +++++++++---------- icmp.c | 4 +- ndp.c | 9 +-- passt.1 | 14 ++-- passt.c | 2 +- passt.h | 22 +++--- pasta.c | 8 +-- tap.c | 12 ++-- tcp.c | 33 ++++----- tcp_internal.h | 2 +- udp.c | 12 ++-- util.c | 22 +++++- util.h | 3 + 19 files changed, 285 insertions(+), 231 deletions(-) -- 2.46.0

2 19

[PATCH] flow: Don't crash if guest attempts to connect to port 0
by David Gibson 14 Aug '24

14 Aug '24

Using a zero port on TCP or UDP is dubious, and we can't really deal with forwarding such a flow within the constraints of the socket API. Hence we ASSERT()ed that we had non-zero ports in flow_hash(). The intention was to make sure that the protocol code sanitizes such ports before completing a flow entry. Unfortunately, flow_hash() is also called on new packets to see if they have an existing flow, so the unsanitized guest packet can crash passt with the assert. Correct this by moving the assert from flow_hash() to flow_sidx_hash() which is only used on entries already in the table, not on unsanitized data. Signed-off-by: David Gibson <david(a)gibson.dropbear.id.au> --- flow.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/flow.c b/flow.c index 687e9fd0..93b687dc 100644 --- a/flow.c +++ b/flow.c @@ -561,12 +561,6 @@ static uint64_t flow_hash(const struct ctx *c, uint8_t proto, uint8_t pif, { struct siphash_state state = SIPHASH_INIT(c->hash_secret); - /* For the hash table to work, we need complete endpoint information, - * and at least a forwarding port. - */ - ASSERT(pif != PIF_NONE && !inany_is_unspecified(&side->eaddr) && - side->eport != 0 && side->fport != 0); - inany_siphash_feed(&state, &side->faddr); inany_siphash_feed(&state, &side->eaddr); @@ -586,8 +580,16 @@ static uint64_t flow_hash(const struct ctx *c, uint8_t proto, uint8_t pif, static uint64_t flow_sidx_hash(const struct ctx *c, flow_sidx_t sidx) { const struct flow_common *f = &flow_at_sidx(sidx)->f; - return flow_hash(c, FLOW_PROTO(f), - f->pif[sidx.sidei], &f->side[sidx.sidei]); + const struct flowside *side = &f->side[sidx.sidei]; + uint8_t pif = f->pif[sidx.sidei]; + + /* For the hash table to work, entries must have complete endpoint + * information, and at least a forwarding port. + */ + ASSERT(pif != PIF_NONE && !inany_is_unspecified(&side->eaddr) && + side->eport != 0 && side->fport != 0); + + return flow_hash(c, FLOW_PROTO(f), pif, side); } /** -- 2.46.0

2 1

[PATCH v2] ndp.c: Turn NDP responder into more declarative implementation
by AbdAlRahman Gad 13 Aug '24

13 Aug '24

- Add structs for NA, RA, NS, MTU, prefix info, option header, link-layer address, RDNSS, DNSSL and link-layer for RA message. - Turn NA message from purely imperative, going byte by byte, to declarative by filling it's struct. - Turn part of RA message into declarative. - Move packet_add() to be before the call of ndp() in tap6_handler() if the protocol of the packet is ICMPv6. - Add a pool of packets as an additional parameter to ndp(). - Check the size of NS packet with packet_get() before sending an NA packet. - Add documentation for the structs. - Add an enum for NDP option types. Link: https://bugs.passt.top/show_bug.cgi?id=21 Signed-off-by: AbdAlRahman Gad <abdobngad(a)gmail.com> --- ndp.c | 317 ++++++++++++++++++++++++++++++++++++++++++++-------------- ndp.h | 3 +- tap.c | 5 +- 3 files changed, 246 insertions(+), 79 deletions(-) diff --git a/ndp.c b/ndp.c index cea3df5..9a80397 100644 --- a/ndp.c +++ b/ndp.c @@ -38,22 +38,193 @@ #define NS 135 #define NA 136 +enum ndp_option_types { + OPT_SRC_L2_ADDR = 1, + OPT_TARGET_L2_ADDR = 2, + OPT_PREFIX_INFO = 3, + OPT_MTU = 5, + OPT_RDNSS_TYPE = 25, + OPT_DNSSL_TYPE = 31, +}; + +/** + * struct opt_header - Option header + * @type: Option type + * @len: Option length, in units of 8 bytes +*/ +struct opt_header { + uint8_t type; + uint8_t len; +} __attribute__((packed)); + +/** + * struct opt_l2_addr - Link-layer address + * @header: Option header + * @mac: MAC address + */ +struct opt_l2_addr { + struct opt_header header; + unsigned char mac[ETH_ALEN]; +} __attribute__((packed)); + +/** + * struct ndp_na - NDP Neighbor Advertisement (NA) message + * @ih: ICMPv6 header + * @target_addr: Target IPv6 address + * @target_l2_addr: Target link-layer address + */ +struct ndp_na { + struct icmp6hdr ih; + struct in6_addr target_addr; + struct opt_l2_addr target_l2_addr; +} __attribute__((packed)); + +/** + * struct opt_prefix_info - Prefix Information option + * @header: Option header + * @prefix_len: The number of leading bits in the Prefix that are valid + * @prefix_flags: Flags associated with the prefix + * @valid_lifetime: Valid lifetime (ms) + * @pref_lifetime: Preferred lifetime (ms) + * @reserved: Unused + */ +struct opt_prefix_info { + struct opt_header header; + uint8_t prefix_len; + uint8_t prefix_flags; + uint32_t valid_lifetime; + uint32_t pref_lifetime; + uint32_t reserved; +} __attribute__((packed)); + +/** + * struct mtu_opt - Maximum transmission unit (MTU) option + * @header: Option header + * @reserved: Unused + * @value: MTU value, network order + */ +struct mtu_opt { + struct opt_header header; + uint16_t reserved; + uint32_t value; +} __attribute__((packed)); + +/** + * struct rdnss - Recursive DNS Server (RDNSS) option + * @header: Option header + * @reserved: Unused + * @lifetime: Validity time (s) + * @dns: List of DNS server addresses + */ +struct opt_rdnss { + struct opt_header header; + uint16_t reserved; + uint32_t lifetime; + struct in6_addr dns[MAXNS + 1]; +} __attribute__((packed)); + +/** + * struct dnssl - DNS Search List (DNSSL) option + * @header: Option header + * @reserved: Unused + * @lifetime: Validity time (s) + * @domains: List of NULL-seperated search domains + */ +struct opt_dnssl { + struct opt_header header; + uint16_t reserved; + uint32_t lifetime; + unsigned char domains[MAXDNSRCH * NS_MAXDNAME]; +} __attribute__((packed)); + +/** + * struct ndp_ra - NDP Router Advertisement (RA) message + * @ih: ICMPv6 header + * @reachable: Reachability time, after confirmation (ms) + * @retrans: Time between retransmitted NS messages (ms) + * @prefix_info: Prefix Information option + * @prefix: IPv6 prefix + * @mtu: MTU option + * @source_ll: Target link-layer address + * @var: Variable fields + */ +struct ndp_ra { + struct icmp6hdr ih; + uint32_t reachable; + uint32_t retrans; + struct opt_prefix_info prefix_info; + struct in6_addr prefix; + struct opt_l2_addr source_ll; + + unsigned char var[sizeof(struct mtu_opt) +sizeof(struct opt_rdnss) + + sizeof(struct opt_dnssl)]; +} __attribute__((packed)); + +/** + * struct ndp_ns - NDP Neighbor Solicitation (NS) message + * @ih: ICMPv6 header + * @target_addr: Target IPv6 address + */ +struct ndp_ns { + struct icmp6hdr ih; + struct in6_addr target_addr; +} __attribute__((packed)); + /** * ndp() - Check for NDP solicitations, reply as needed * @c: Execution context * @ih: ICMPv6 header - * @saddr Source IPv6 address + * @saddr: Source IPv6 address + * @p: Packet pool * * Return: 0 if not handled here, 1 if handled, -1 on failure */ -int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr) +int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr, + const struct pool *p) { + struct ndp_na na = { + .ih = { + .icmp6_type = NA, + .icmp6_code = 0, + .icmp6_router = 1, + .icmp6_solicited = 1, + .icmp6_override = 1, + }, + .target_l2_addr = { + .header = { + .type = OPT_TARGET_L2_ADDR, + .len = 1, + }, + } + }; + struct ndp_ra ra = { + .ih = { + .icmp6_type = RA, + .icmp6_code = 0, + .icmp6_hop_limit = 255, + /* RFC 8319 */ + .icmp6_rt_lifetime = htons_constant(65535), + .icmp6_addrconf_managed = 1, + }, + .prefix_info = { + .header = { + .type = OPT_PREFIX_INFO, + .len = 4, + }, + .prefix_len = 64, + .prefix_flags = 0xc0, /* prefix flags: L, A */ + .valid_lifetime = ~0U, + .pref_lifetime = ~0U, + }, + .source_ll = { + .header = { + .type = OPT_SRC_L2_ADDR, + .len = 1, + }, + }, + }; const struct in6_addr *rsaddr; /* src addr for reply */ - char buf[BUFSIZ] = { 0 }; - struct ipv6hdr *ip6hr; - struct icmp6hdr *ihr; - struct ethhdr *ehr; - unsigned char *p; + unsigned char *ptr = NULL; size_t dlen; if (ih->icmp6_type < RS || ih->icmp6_type > NA) @@ -62,28 +233,22 @@ int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr) if (c->no_ndp) return 1; - ehr = (struct ethhdr *)buf; - ip6hr = (struct ipv6hdr *)(ehr + 1); - ihr = (struct icmp6hdr *)(ip6hr + 1); - if (ih->icmp6_type == NS) { + struct ndp_ns *ns = packet_get(p, 0, 0, sizeof(struct ndp_ns), + NULL); + + if (!ns) + return -1; + if (IN6_IS_ADDR_UNSPECIFIED(saddr)) return 1; info("NDP: received NS, sending NA"); - ihr->icmp6_type = NA; - ihr->icmp6_code = 0; - ihr->icmp6_router = 1; - ihr->icmp6_solicited = 1; - ihr->icmp6_override = 1; - - p = (unsigned char *)(ihr + 1); - memcpy(p, ih + 1, sizeof(struct in6_addr)); /* target address */ - p += 16; - *p++ = 2; /* target ll */ - *p++ = 1; /* length */ - memcpy(p, c->mac, ETH_ALEN); - p += 6; + + memcpy(&na.target_addr, &ns->target_addr, + sizeof(na.target_addr)); + memcpy(na.target_l2_addr.mac, c->mac, ETH_ALEN); + } else if (ih->icmp6_type == RS) { size_t dns_s_len = 0; int i, n; @@ -92,31 +257,20 @@ int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr) return 1; info("NDP: received RS, sending RA"); - ihr->icmp6_type = RA; - ihr->icmp6_code = 0; - ihr->icmp6_hop_limit = 255; - ihr->icmp6_rt_lifetime = htons(65535); /* RFC 8319 */ - ihr->icmp6_addrconf_managed = 1; - - p = (unsigned char *)(ihr + 1); - p += 8; /* reachable, retrans time */ - *p++ = 3; /* prefix */ - *p++ = 4; /* length */ - *p++ = 64; /* prefix length */ - *p++ = 0xc0; /* prefix flags: L, A */ - *(uint32_t *)p = (uint32_t)~0U; /* lifetime */ - p += 4; - *(uint32_t *)p = (uint32_t)~0U; /* preferred lifetime */ - p += 8; - memcpy(p, &c->ip6.addr, 8); /* prefix */ - p += 16; + memcpy(&ra.prefix, &c->ip6.addr, sizeof(ra.prefix)); + + ptr = &ra.var[0]; if (c->mtu != -1) { - *p++ = 5; /* type */ - *p++ = 1; /* length */ - p += 2; /* reserved */ - *(uint32_t *)p = htonl(c->mtu); /* MTU */ - p += 4; + struct mtu_opt *mtu = (struct mtu_opt *)ptr; + *mtu = (struct mtu_opt) { + .header = { + .type = OPT_MTU, + .len = 1, + }, + .value = htonl(c->mtu), + }; + ptr += sizeof(struct mtu_opt); } if (c->no_dhcp_dns) @@ -124,59 +278,64 @@ int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr) for (n = 0; !IN6_IS_ADDR_UNSPECIFIED(&c->ip6.dns[n]); n++); if (n) { - *p++ = 25; /* RDNSS */ - *p++ = 1 + 2 * n; /* length */ - p += 2; /* reserved */ - *(uint32_t *)p = (uint32_t)~0U; /* lifetime */ - p += 4; - + struct opt_rdnss *rdnss = (struct opt_rdnss *)ptr; + *rdnss = (struct opt_rdnss) { + .header = { + .type = OPT_RDNSS_TYPE, + .len = 1 + 2 * n, + }, + .lifetime = ~0U, + }; for (i = 0; i < n; i++) { - memcpy(p, &c->ip6.dns[i], 16); /* address */ - p += 16; + memcpy(&rdnss->dns[i], &c->ip6.dns[i], + sizeof(rdnss->dns[i])); } + ptr += offsetof(struct opt_rdnss, dns) + + i * sizeof(rdnss->dns[0]); for (n = 0; *c->dns_search[n].n; n++) dns_s_len += strlen(c->dns_search[n].n) + 2; } if (!c->no_dhcp_dns_search && dns_s_len) { - *p++ = 31; /* DNSSL */ - *p++ = (dns_s_len + 8 - 1) / 8 + 1; /* length */ - p += 2; /* reserved */ - *(uint32_t *)p = (uint32_t)~0U; /* lifetime */ - p += 4; + struct opt_dnssl *dnssl = (struct opt_dnssl *)ptr; + *dnssl = (struct opt_dnssl) { + .header = { + .type = OPT_DNSSL_TYPE, + .len = DIV_ROUND_UP(dns_s_len, 8) + 1, + }, + .lifetime = ~0U, + }; + ptr = dnssl->domains; for (i = 0; i < n; i++) { + size_t len; char *dot; - *(p++) = '.'; + *(ptr++) = '.'; - strncpy((char *)p, c->dns_search[i].n, - sizeof(buf) - - ((intptr_t)p - (intptr_t)buf)); - for (dot = (char *)p - 1; *dot; dot++) { + len = sizeof(dnssl->domains) - + (ptr - dnssl->domains); + + strncpy((char *)ptr, c->dns_search[i].n, len); + for (dot = (char *)ptr - 1; *dot; dot++) { if (*dot == '.') *dot = strcspn(dot + 1, "."); } - p += strlen(c->dns_search[i].n); - *(p++) = 0; + ptr += strlen(c->dns_search[i].n); + *(ptr++) = 0; } - memset(p, 0, 8 - dns_s_len % 8); /* padding */ - p += 8 - dns_s_len % 8; + memset(ptr, 0, 8 - dns_s_len % 8); /* padding */ + ptr += 8 - dns_s_len % 8; } dns_done: - *p++ = 1; /* source ll */ - *p++ = 1; /* length */ - memcpy(p, c->mac, ETH_ALEN); - p += 6; + memcpy(&ra.source_ll.mac, c->mac, ETH_ALEN); } else { return 1; } - dlen = (uintptr_t)p - (uintptr_t)ihr - sizeof(*ihr); - if (IN6_IS_ADDR_LINKLOCAL(saddr)) c->ip6.addr_ll_seen = *saddr; else @@ -187,7 +346,13 @@ dns_done: else rsaddr = &c->ip6.addr_ll; - tap_icmp6_send(c, rsaddr, saddr, ihr, dlen + sizeof(*ihr)); + if (ih->icmp6_type == NS) { + dlen = sizeof(struct ndp_na); + tap_icmp6_send(c, rsaddr, saddr, &na, dlen); + } else if (ih->icmp6_type == RS) { + dlen = ptr - (unsigned char *)&ra; + tap_icmp6_send(c, rsaddr, saddr, &ra, dlen); + } return 1; } diff --git a/ndp.h b/ndp.h index b33cd69..a786441 100644 --- a/ndp.h +++ b/ndp.h @@ -6,6 +6,7 @@ #ifndef NDP_H #define NDP_H -int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr); +int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr, + const struct pool *p); #endif /* NDP_H */ diff --git a/tap.c b/tap.c index 5852705..87be3a6 100644 --- a/tap.c +++ b/tap.c @@ -808,12 +808,13 @@ resume: if (l4len < sizeof(struct icmp6hdr)) continue; - if (ndp(c, (struct icmp6hdr *)l4h, saddr)) + packet_add(pkt, l4len, l4h); + + if (ndp(c, (struct icmp6hdr *)l4h, saddr, pkt)) continue; tap_packet_debug(NULL, ip6h, NULL, proto, NULL, 1); - packet_add(pkt, l4len, l4h); icmp_tap_handler(c, PIF_TAP, AF_INET6, saddr, daddr, pkt, now); continue; -- 2.46.0

2 2

[PATCH] ndp.c: Turn NDP responder into more declarative implementation
by AbdAlRahman Gad 12 Aug '24

12 Aug '24

- Add structs for NA, RA, NS, MTU, prefix info, option header, link-layer address, RDNSS, DNSSL and link-layer for RA message. - Turn NA message from purely imperative, going byte by byte, to declarative by filling it's struct. - Turn part of RA message into declarative. - Move packet_add() to be before the call of ndp() in tap6_handler() if the protocol of the packet is ICMPv6. - Add a pool of packets as an additional parameter to ndp(). - Check the size of NS packet with packet_get() before sending an NA packet. - Add documentation for the structs. - Add an enum for NDP option types. Link: https://bugs.passt.top/show_bug.cgi?id=21 Signed-off-by: AbdAlRahman Gad <abdobngad(a)gmail.com> --- ndp.c | 320 +++++++++++++++++++++++++++++++++++++++++++--------------- ndp.h | 3 +- tap.c | 5 +- 3 files changed, 246 insertions(+), 82 deletions(-) diff --git a/ndp.c b/ndp.c index cea3df5..5e9cb78 100644 --- a/ndp.c +++ b/ndp.c @@ -38,22 +38,199 @@ #define NS 135 #define NA 136 +enum ndp_option_types { + OPT_SRC_L2_ADDR = 1, + OPT_TARGET_L2_ADDR = 2, + OPT_PREFIX_INFO = 3, + OPT_MTU = 5, + OPT_RDNSS_TYPE = 25, + OPT_DNSSL_TYPE = 31, +}; + +/** + * struct opt_header - Option header + * @type: Option type + * @len: Option length, in units of 8 bytes +*/ +struct opt_header { + uint8_t type; + uint8_t len; +} __attribute__((packed)); + +/** + * struct opt_l2_addr - Link-layer address + * @header: Option header + * @mac: MAC address + */ +struct opt_l2_addr { + struct opt_header header; + unsigned char mac[ETH_ALEN]; +} __attribute__((packed)); + +/** + * struct ndp_na - NDP Neighbor Advertisement (NA) message + * @ih: ICMPv6 header + * @target_addr: Target IPv6 address + * @target_l2_addr: Target link-layer address + */ +struct ndp_na { + struct icmp6hdr ih; + struct in6_addr target_addr; + struct opt_l2_addr target_l2_addr; +} __attribute__((packed)); + +/** + * struct opt_prefix_info - Prefix Information option + * @header: Option header + * @prefix_len: The number of leading bits in the Prefix that are valid + * @prefix_flags: Flags associated with the prefix + * @valid_lifetime: Valid lifetime (ms) + * @pref_lifetime: Preferred lifetime (ms) + * @reserved: Unused + */ +struct opt_prefix_info { + struct opt_header header; + uint8_t prefix_len; + uint8_t prefix_flags; + uint32_t valid_lifetime; + uint32_t pref_lifetime; + uint32_t reserved; +} __attribute__((packed)); + +/** + * struct mtu_opt - Maximum transmission unit (MTU) option + * @header: Option header + * @reserved: Unused + * @value: MTU value, network order + */ +struct mtu_opt { + struct opt_header header; + uint16_t reserved; + uint32_t value; +} __attribute__((packed)); + +/** + * struct rdnss - Recursive DNS Server (RDNSS) option + * @header: Option header + * @reserved: Unused + * @lifetime: Validity time (s) + * @dns: List of DNS server addresses + */ +struct opt_rdnss { + struct opt_header header; + uint16_t reserved; + uint32_t lifetime; + struct in6_addr dns[MAXNS + 1]; +} __attribute__((packed)); + +/** + * struct dnssl - DNS Search List (DNSSL) option + * @header: Option header + * @reserved: Unused + * @lifetime: Validity time (s) + * @domains: List of NULL-seperated search domains + */ +struct opt_dnssl { + struct opt_header header; + uint16_t reserved; + uint32_t lifetime; + unsigned char domains[MAXDNSRCH * NS_MAXDNAME]; +} __attribute__((packed)); + +/** + * struct ndp_ra - NDP Router Advertisement (RA) message + * @ih: ICMPv6 header + * @reachable: Reachability time, after confirmation (ms) + * @retrans: Time between retransmitted NS messages (ms) + * @prefix_info: Prefix Information option + * @prefix: IPv6 prefix + * @mtu: MTU option + * @source_ll: Target link-layer address + * @dns: RDNSS and DNSSL options + */ +struct ndp_ra { + struct icmp6hdr ih; + uint32_t reachable; + uint32_t retrans; + struct opt_prefix_info prefix_info; + struct in6_addr prefix; + struct mtu_opt mtu; + struct opt_l2_addr source_ll; + + unsigned char dns[sizeof(struct opt_rdnss) + sizeof(struct opt_dnssl)]; +} __attribute__((packed)); + +/** + * struct ndp_ns - NDP Neighbor Solicitation (NS) message + * @ih: ICMPv6 header + * @target_addr: Target IPv6 address + */ +struct ndp_ns { + struct icmp6hdr ih; + struct in6_addr target_addr; +} __attribute__((packed)); + /** * ndp() - Check for NDP solicitations, reply as needed * @c: Execution context * @ih: ICMPv6 header - * @saddr Source IPv6 address + * @saddr: Source IPv6 address + * @p: Packet pool * * Return: 0 if not handled here, 1 if handled, -1 on failure */ -int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr) +int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr, + const struct pool *p) { + struct ndp_na na = { + .ih = { + .icmp6_type = NA, + .icmp6_code = 0, + .icmp6_router = 1, + .icmp6_solicited = 1, + .icmp6_override = 1, + }, + .target_l2_addr = { + .header = { + .type = OPT_TARGET_L2_ADDR, + .len = 1, + }, + } + }; + struct ndp_ra ra = { + .ih = { + .icmp6_type = RA, + .icmp6_code = 0, + .icmp6_hop_limit = 255, + /* RFC 8319 */ + .icmp6_rt_lifetime = htons_constant(65535), + .icmp6_addrconf_managed = 1 + }, + .prefix_info = { + .header = { + .type = OPT_PREFIX_INFO, + .len = 4, + }, + .prefix_len = 64, + .prefix_flags = 0xc0, /* prefix flags: L, A */ + .valid_lifetime = ~0U, + .pref_lifetime = ~0U, + }, + .mtu = { + .header = { + .type = OPT_MTU, + .len = 1, + }, + }, + .source_ll = { + .header = { + .type = OPT_SRC_L2_ADDR, + .len = 1, + }, + }, + }; const struct in6_addr *rsaddr; /* src addr for reply */ - char buf[BUFSIZ] = { 0 }; - struct ipv6hdr *ip6hr; - struct icmp6hdr *ihr; - struct ethhdr *ehr; - unsigned char *p; + unsigned char *ptr = NULL; size_t dlen; if (ih->icmp6_type < RS || ih->icmp6_type > NA) @@ -62,28 +239,22 @@ int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr) if (c->no_ndp) return 1; - ehr = (struct ethhdr *)buf; - ip6hr = (struct ipv6hdr *)(ehr + 1); - ihr = (struct icmp6hdr *)(ip6hr + 1); - if (ih->icmp6_type == NS) { + struct ndp_ns *ns = packet_get(p, 0, 0, sizeof(struct ndp_ns), + NULL); + + if (!ns) + return -1; + if (IN6_IS_ADDR_UNSPECIFIED(saddr)) return 1; info("NDP: received NS, sending NA"); - ihr->icmp6_type = NA; - ihr->icmp6_code = 0; - ihr->icmp6_router = 1; - ihr->icmp6_solicited = 1; - ihr->icmp6_override = 1; - - p = (unsigned char *)(ihr + 1); - memcpy(p, ih + 1, sizeof(struct in6_addr)); /* target address */ - p += 16; - *p++ = 2; /* target ll */ - *p++ = 1; /* length */ - memcpy(p, c->mac, ETH_ALEN); - p += 6; + + memcpy(&na.target_addr, &ns->target_addr, + sizeof(na.target_addr)); + memcpy(na.target_l2_addr.mac, c->mac, ETH_ALEN); + } else if (ih->icmp6_type == RS) { size_t dns_s_len = 0; int i, n; @@ -92,91 +263,76 @@ int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr) return 1; info("NDP: received RS, sending RA"); - ihr->icmp6_type = RA; - ihr->icmp6_code = 0; - ihr->icmp6_hop_limit = 255; - ihr->icmp6_rt_lifetime = htons(65535); /* RFC 8319 */ - ihr->icmp6_addrconf_managed = 1; - - p = (unsigned char *)(ihr + 1); - p += 8; /* reachable, retrans time */ - *p++ = 3; /* prefix */ - *p++ = 4; /* length */ - *p++ = 64; /* prefix length */ - *p++ = 0xc0; /* prefix flags: L, A */ - *(uint32_t *)p = (uint32_t)~0U; /* lifetime */ - p += 4; - *(uint32_t *)p = (uint32_t)~0U; /* preferred lifetime */ - p += 8; - memcpy(p, &c->ip6.addr, 8); /* prefix */ - p += 16; - - if (c->mtu != -1) { - *p++ = 5; /* type */ - *p++ = 1; /* length */ - p += 2; /* reserved */ - *(uint32_t *)p = htonl(c->mtu); /* MTU */ - p += 4; - } + memcpy(&ra.prefix, &c->ip6.addr, sizeof(ra.prefix)); + + if (c->mtu != -1) + ra.mtu.value = htonl(c->mtu); + + ptr = &ra.dns[0]; if (c->no_dhcp_dns) goto dns_done; for (n = 0; !IN6_IS_ADDR_UNSPECIFIED(&c->ip6.dns[n]); n++); if (n) { - *p++ = 25; /* RDNSS */ - *p++ = 1 + 2 * n; /* length */ - p += 2; /* reserved */ - *(uint32_t *)p = (uint32_t)~0U; /* lifetime */ - p += 4; - + struct opt_rdnss *rdnss = (struct opt_rdnss *)ptr; + *rdnss = (struct opt_rdnss) { + .header = { + .type = OPT_RDNSS_TYPE, + .len = 1 + 2 * n, + }, + .lifetime = ~0U, + }; for (i = 0; i < n; i++) { - memcpy(p, &c->ip6.dns[i], 16); /* address */ - p += 16; + memcpy(&rdnss->dns[i], &c->ip6.dns[i], + sizeof(rdnss->dns[i])); } + ptr += offsetof(struct opt_rdnss, dns) + + i * sizeof(rdnss->dns[0]); for (n = 0; *c->dns_search[n].n; n++) dns_s_len += strlen(c->dns_search[n].n) + 2; } if (!c->no_dhcp_dns_search && dns_s_len) { - *p++ = 31; /* DNSSL */ - *p++ = (dns_s_len + 8 - 1) / 8 + 1; /* length */ - p += 2; /* reserved */ - *(uint32_t *)p = (uint32_t)~0U; /* lifetime */ - p += 4; + struct opt_dnssl *dnssl = (struct opt_dnssl *)ptr; + *dnssl = (struct opt_dnssl) { + .header = { + .type = OPT_DNSSL_TYPE, + .len = DIV_ROUND_UP(dns_s_len, 8) + 1, + }, + .lifetime = ~0U, + }; + ptr = dnssl->domains; for (i = 0; i < n; i++) { + size_t len; char *dot; - *(p++) = '.'; + *(ptr++) = '.'; - strncpy((char *)p, c->dns_search[i].n, - sizeof(buf) - - ((intptr_t)p - (intptr_t)buf)); - for (dot = (char *)p - 1; *dot; dot++) { + len = sizeof(dnssl->domains) - + (ptr - dnssl->domains); + + strncpy((char *)ptr, c->dns_search[i].n, len); + for (dot = (char *)ptr - 1; *dot; dot++) { if (*dot == '.') *dot = strcspn(dot + 1, "."); } - p += strlen(c->dns_search[i].n); - *(p++) = 0; + ptr += strlen(c->dns_search[i].n); + *(ptr++) = 0; } - memset(p, 0, 8 - dns_s_len % 8); /* padding */ - p += 8 - dns_s_len % 8; + memset(ptr, 0, 8 - dns_s_len % 8); /* padding */ + ptr += 8 - dns_s_len % 8; } dns_done: - *p++ = 1; /* source ll */ - *p++ = 1; /* length */ - memcpy(p, c->mac, ETH_ALEN); - p += 6; + memcpy(&ra.source_ll.mac, c->mac, ETH_ALEN); } else { return 1; } - dlen = (uintptr_t)p - (uintptr_t)ihr - sizeof(*ihr); - if (IN6_IS_ADDR_LINKLOCAL(saddr)) c->ip6.addr_ll_seen = *saddr; else @@ -187,7 +343,13 @@ dns_done: else rsaddr = &c->ip6.addr_ll; - tap_icmp6_send(c, rsaddr, saddr, ihr, dlen + sizeof(*ihr)); + if (ih->icmp6_type == NS) { + dlen = sizeof(struct ndp_na); + tap_icmp6_send(c, rsaddr, saddr, &na, dlen); + } else if (ih->icmp6_type == RS) { + dlen = ptr - (unsigned char *)&ra; + tap_icmp6_send(c, rsaddr, saddr, &ra, dlen); + } return 1; } diff --git a/ndp.h b/ndp.h index b33cd69..a786441 100644 --- a/ndp.h +++ b/ndp.h @@ -6,6 +6,7 @@ #ifndef NDP_H #define NDP_H -int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr); +int ndp(struct ctx *c, const struct icmp6hdr *ih, const struct in6_addr *saddr, + const struct pool *p); #endif /* NDP_H */ diff --git a/tap.c b/tap.c index 5852705..87be3a6 100644 --- a/tap.c +++ b/tap.c @@ -808,12 +808,13 @@ resume: if (l4len < sizeof(struct icmp6hdr)) continue; - if (ndp(c, (struct icmp6hdr *)l4h, saddr)) + packet_add(pkt, l4len, l4h); + + if (ndp(c, (struct icmp6hdr *)l4h, saddr, pkt)) continue; tap_packet_debug(NULL, ip6h, NULL, proto, NULL, 1); - packet_add(pkt, l4len, l4h); icmp_tap_handler(c, PIF_TAP, AF_INET6, saddr, daddr, pkt, now); continue; -- 2.46.0

2 1

[PATCH] log: Don't prefix message with timestamp on --debug if it's a continuation
by Stefano Brivio 12 Aug '24

12 Aug '24

If we prefix the second part of messages printed through logmsg_perror() by the timestamp, on debug, we'll have two timestamps and a weird separator in the result, such as this beauty: 0.0013: Failed to clone process with detached namespaces0.0013: : Operation not permitted Add a parameter to logmsg() and vlogmsg() which indicates a message continuation. If that's set, don't print the timestamp in vlogmsg(). Link: https://github.com/moby/moby/issues/48257#issuecomment-2282875092 Signed-off-by: Stefano Brivio <sbrivio(a)redhat.com> --- flow.c | 3 ++- log.c | 14 ++++++++------ log.h | 22 +++++++++++----------- 3 files changed, 21 insertions(+), 18 deletions(-) diff --git a/flow.c b/flow.c index bd5fa2b..687e9fd 100644 --- a/flow.c +++ b/flow.c @@ -279,7 +279,8 @@ void flow_log_(const struct flow_common *f, int pri, const char *fmt, ...) else type_or_state = FLOW_TYPE(f); - logmsg(true, pri, "Flow %u (%s): %s", flow_idx(f), type_or_state, msg); + logmsg(true, false, pri, + "Flow %u (%s): %s", flow_idx(f), type_or_state, msg); } /** diff --git a/log.c b/log.c index 0c377cc..e7202d0 100644 --- a/log.c +++ b/log.c @@ -253,11 +253,12 @@ static void logfile_write(bool newline, int pri, const struct timespec *now, /** * vlogmsg() - Print or send messages to log or output files as configured * @newline: Append newline at the end of the message, if missing + * @cont: Continuation of a previous message, on the same line * @pri: Facility and level map, same as priority for vsyslog() * @format: Message * @ap: Variable argument list */ -void vlogmsg(bool newline, int pri, const char *format, va_list ap) +void vlogmsg(bool newline, bool cont, int pri, const char *format, va_list ap) { bool debug_print = (log_mask & LOG_MASK(LOG_DEBUG)) && log_file == -1; const struct timespec *now; @@ -265,7 +266,7 @@ void vlogmsg(bool newline, int pri, const char *format, va_list ap) now = logtime(&ts); - if (debug_print) { + if (debug_print && !cont) { char timestr[LOGTIME_STRLEN]; logtime_fmt(timestr, sizeof(timestr), now); @@ -295,15 +296,16 @@ void vlogmsg(bool newline, int pri, const char *format, va_list ap) /** * logmsg() - vlogmsg() wrapper for variable argument lists * @newline: Append newline at the end of the message, if missing + * @cont: Continuation of a previous message, on the same line * @pri: Facility and level map, same as priority for vsyslog() * @format: Message */ -void logmsg(bool newline, int pri, const char *format, ...) +void logmsg(bool newline, bool cont, int pri, const char *format, ...) { va_list ap; va_start(ap, format); - vlogmsg(newline, pri, format, ap); + vlogmsg(newline, cont, pri, format, ap); va_end(ap); } @@ -318,10 +320,10 @@ void logmsg_perror(int pri, const char *format, ...) va_list ap; va_start(ap, format); - vlogmsg(false, pri, format, ap); + vlogmsg(false, false, pri, format, ap); va_end(ap); - logmsg(true, pri, ": %s", strerror(errno_copy)); + logmsg(true, true, pri, ": %s", strerror(errno_copy)); } /** diff --git a/log.h b/log.h index ba17f79..a30b091 100644 --- a/log.h +++ b/log.h @@ -13,21 +13,21 @@ #define LOGFILE_CUT_RATIO 30 /* When full, cut ~30% size */ #define LOGFILE_SIZE_MIN (5UL * MAX(BUFSIZ, PAGE_SIZE)) -void vlogmsg(bool newline, int pri, const char *format, va_list ap); -void logmsg(bool newline, int pri, const char *format, ...) - __attribute__((format(printf, 3, 4))); +void vlogmsg(bool newline, bool cont, int pri, const char *format, va_list ap); +void logmsg(bool newline, bool cont, int pri, const char *format, ...) + __attribute__((format(printf, 4, 5))); void logmsg_perror(int pri, const char *format, ...) __attribute__((format(printf, 2, 3))); -#define err(...) logmsg(true, LOG_ERR, __VA_ARGS__) -#define warn(...) logmsg(true, LOG_WARNING, __VA_ARGS__) -#define info(...) logmsg(true, LOG_INFO, __VA_ARGS__) -#define debug(...) logmsg(true, LOG_DEBUG, __VA_ARGS__) +#define err(...) logmsg(true, false, LOG_ERR, __VA_ARGS__) +#define warn(...) logmsg(true, false, LOG_WARNING, __VA_ARGS__) +#define info(...) logmsg(true, false, LOG_INFO, __VA_ARGS__) +#define debug(...) logmsg(true, false, LOG_DEBUG, __VA_ARGS__) -#define err_perror(...) logmsg_perror( LOG_ERR, __VA_ARGS__) -#define warn_perror(...) logmsg_perror( LOG_WARNING, __VA_ARGS__) -#define info_perror(...) logmsg_perror( LOG_INFO, __VA_ARGS__) -#define debug_perror(...) logmsg_perror( LOG_DEBUG, __VA_ARGS__) +#define err_perror(...) logmsg_perror( LOG_ERR, __VA_ARGS__) +#define warn_perror(...) logmsg_perror( LOG_WARNING, __VA_ARGS__) +#define info_perror(...) logmsg_perror( LOG_INFO, __VA_ARGS__) +#define debug_perror(...) logmsg_perror( LOG_DEBUG, __VA_ARGS__) #define die(...) \ do { \ -- 2.43.0

2 1

[PATCH v3] conf: Stop parsing options at first non-option argument
by Stefano Brivio 09 Aug '24

09 Aug '24

Given that pasta supports specifying a command to be executed on the command line, even without the usual -- separator as long as there's no ambiguity, we shouldn't eat up options that are not meant for us. Paul reports, for instance, that with: pasta --config-net ip -6 route -6 is taken by pasta to mean --ipv6-only, and we execute 'ip route'. That's because getopt_long(), by default, shuffles the argument list to shift non-option arguments at the end. Avoid that by adding '+' at the beginning of 'optstring'. Reported-by: Paul Holzinger <pholzing(a)redhat.com> Signed-off-by: Stefano Brivio <sbrivio(a)redhat.com> --- v3: Use '+' in optstring and drop first non-option tracking v2: Instead of overriding 'name' in the getopt_long() loop, to force exiting the loop, adjust the exit condition conf.c | 4 ++-- util.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/conf.c b/conf.c index 2f5d649..a79e7a6 100644 --- a/conf.c +++ b/conf.c @@ -1253,9 +1253,9 @@ void conf(struct ctx *c, int argc, char **argv) if (c->mode == MODE_PASTA) { c->no_dhcp_dns = c->no_dhcp_dns_search = 1; fwd_default = FWD_AUTO; - optstring = "dqfel:hF:I:p:P:m:a:n:M:g:i:o:D:S:46t:u:T:U:"; + optstring = "+dqfel:hF:I:p:P:m:a:n:M:g:i:o:D:S:46t:u:T:U:"; } else { - optstring = "dqfel:hs:F:p:P:m:a:n:M:g:i:o:D:S:461t:u:"; + optstring = "+dqfel:hs:F:p:P:m:a:n:M:g:i:o:D:S:461t:u:"; } c->tcp.fwd_in.mode = c->tcp.fwd_out.mode = FWD_UNSET; diff --git a/util.c b/util.c index 7761bd3..0b41404 100644 --- a/util.c +++ b/util.c @@ -710,7 +710,7 @@ void close_open_files(int argc, char **argv) int name, rc; do { - name = getopt_long(argc, argv, ":F", optfd, NULL); + name = getopt_long(argc, argv, "+:F", optfd, NULL); if (name == 'F') { errno = 0; -- 2.43.0

3 7