July 2024 - dev

[PATCH v2 00/14] Clean up checksum and header generation for inbound packets
by David Gibson 10 Oct '25

10 Oct '25

The main packet "fast paths" for UDP and TCP mostly just forward packets rather than generating them from scratch. However the control paths for ICMP and DHCP sometimes generate packets more or less from scratch. Because these are relatively rare, it's not performance critical. The paths for sending these packets have some duplication of the header generation. There's also some layering violation in tap_ip_send() which both generates IP headers and updates the L4 (UDP or UCMP) checksum. Finally that checksum generation is a little awkward: it temporarily generates the IP pseudo header (or something close enough to serve) in the place of the actual header, generates the checksum, then replaces it with the real IP header. This approach seems to be causing miscompiles with some LTO optimization, because the stores to the pseudo header are being moved or elided across the code calculating the checksum. This series addresses all of these. We consolidate and clarify the packet sending helpers, and use them in some places there was previously duplicated code. In the process we use new checksum generation helpers which take a different approach which should avoid the LTO problems (this aspect I haven't tested yet though). Changes since v1: * Numerous minor style changes * Rename header generation helpers to make their behaviour clearer * Added several missing function doc comments * Corrected some erroneous statements and terms in comments David Gibson (14): Add csum_icmp6() helper for calculating ICMPv6 checksums Add csum_icmp4() helper for calculating ICMP checksums Add csum_udp6() helper for calculating UDP over IPv6 checksums Add csum_udp4() helper for calculating UDP over IPv4 checksums Add csum_ip4_header() helper to calculate IPv4 header checksums Add helpers for normal inbound packet destination addresses Remove support for TCP packets from tap_ip_send() tap: Remove unhelpeful vnet_pre optimization from tap_send() Split tap_ip_send() into IPv4 and IPv6 specific functions tap: Split tap_ip6_send() into UDP and ICMP variants ndp: Remove unneeded eh_source parameter ndp: Use tap_icmp6_send() helper tap: Split tap_ip4_send() into UDP and ICMP variants dhcp: Use tap_udp4_send() helper in dhcp() arp.c | 2 +- checksum.c | 120 ++++++++++++++++----- checksum.h | 15 ++- dhcp.c | 19 +--- dhcpv6.c | 21 +--- icmp.c | 12 +-- ndp.c | 28 +---- ndp.h | 3 +- tap.c | 312 ++++++++++++++++++++++++++++++++++------------------- tap.h | 19 +++- 10 files changed, 345 insertions(+), 206 deletions(-) -- 2.37.3

2 17

[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

[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

[PATCH 0/2] Avoid duplicated clock_gettime() in logging
by David Gibson 05 Aug '24

05 Aug '24

Remove a small nit in the logging code. David Gibson (2): log: Make logfile_write() private log: Avoid duplicate calls to clock_gettime() log.c | 344 +++++++++++++++++++++++++++++----------------------------- log.h | 1 - 2 files changed, 172 insertions(+), 173 deletions(-) -- 2.45.2

2 5

[PATCH v2] pasta.c: modify hostname when detaching new namespace
by Danish Prakash 30 Jul '24

30 Jul '24

When invoking pasta without any arguments, it's difficult to tell whether we are in the new namespace or not leaving users a bit confused. This change modifies the host namespace to add a prefix "pasta-" to make it a bit more obvious. Signed-off-by: Danish Prakash <contact(a)danishpraka.sh> --- pasta.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/pasta.c b/pasta.c index 31e1e000ee1b..7ce502d3f299 100644 --- a/pasta.c +++ b/pasta.c @@ -50,6 +50,8 @@ #include "netlink.h" #include "log.h" +#define HOSTNAME_PREFIX "pasta-" + /* PID of child, in case we created a namespace */ int pasta_child_pid; @@ -178,6 +180,7 @@ struct pasta_spawn_cmd_arg { */ static int pasta_spawn_cmd(void *arg) { + char hostname[HOST_NAME_MAX + 1] = HOSTNAME_PREFIX; const struct pasta_spawn_cmd_arg *a; sigset_t set; @@ -188,6 +191,13 @@ static int pasta_spawn_cmd(void *arg) if (write_file("/proc/sys/net/ipv4/ping_group_range", "0 0")) warn("Cannot set ping_group_range, ICMP requests might fail"); + if (!gethostname(hostname + sizeof(HOSTNAME_PREFIX) - 1, HOST_NAME_MAX + 1 - + sizeof(HOSTNAME_PREFIX)) || errno == ENAMETOOLONG) { + hostname[HOST_NAME_MAX] = '\0'; + if (sethostname(hostname, strlen(hostname))) + warn("Unable to set pasta-prefixed hostname"); + } + /* Wait for the parent to be ready: see main() */ sigemptyset(&set); sigaddset(&set, SIGUSR1); -- 2.45.2

2 1

[PATCH v2] Fix typo in README file
by AbdAlRahman Gad 29 Jul '24

29 Jul '24

- remove duplicated 'the' in the 'Services' section Signed-off-by: AbdAlRahman Gad <abdobngad(a)gmail.com> --- v2: Add Signed-off-by: tag README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index aa2afac..bb114ea 100644 --- a/README.md +++ b/README.md @@ -398,7 +398,7 @@ services: and nameserver using SLAAC * [DHCPv6 server](/passt/tree/dhcpv6.c): a simple implementation handing out one single IPv6 address to the guest or namespace, - namely, the the same address as the first one configured for the upstream host + namely, the same address as the first one configured for the upstream host interface, and passing the nameservers configured on the host ## Addresses -- 2.43.0

3 2

[PATCH] pasta.c: modify hostname when detaching new namespace
by Danish Prakash 29 Jul '24

29 Jul '24

When invoking pasta without any arguments, it's difficult to tell whether we are in the new namespace or not leaving users a bit confused. This change modifies the host namespace to add a prefix "pasta-" to make it a bit more obvious. Signed-off-by: Danish Prakash <contact(a)danishpraka.sh> --- pasta.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/pasta.c b/pasta.c index 31e1e00..840a2b1 100644 --- a/pasta.c +++ b/pasta.c @@ -180,6 +180,8 @@ static int pasta_spawn_cmd(void *arg) { const struct pasta_spawn_cmd_arg *a; sigset_t set; + char hostname[HOST_NAME_MAX+1], pasta_hostname[HOST_NAME_MAX+1]; + char *hostname_prefix = "pasta-"; /* We run in a detached PID and mount namespace: mount /proc over */ if (mount("", "/proc", "proc", 0, NULL)) @@ -188,6 +190,17 @@ static int pasta_spawn_cmd(void *arg) if (write_file("/proc/sys/net/ipv4/ping_group_range", "0 0")) warn("Cannot set ping_group_range, ICMP requests might fail"); + if (gethostname(hostname, HOST_NAME_MAX+1) == 0) { + if ((strlen(hostname) + strlen(hostname_prefix)) > HOST_NAME_MAX) { + hostname[strlen(hostname)-strlen(hostname_prefix)] = '\0'; + } + sprintf(pasta_hostname, "%s%s", hostname_prefix, hostname); + + if (sethostname(pasta_hostname, strlen(pasta_hostname)) != 0) { + warn("Unable to set pasta-prefixed hostname"); + } + } + /* Wait for the parent to be ready: see main() */ sigemptyset(&set); sigaddset(&set, SIGUSR1); -- 2.45.1

2 10

[PATCH 0/5] Fix assorted errors in the Qemu socket tap receive handler
by David Gibson 29 Jul '24

29 Jul '24

A long time ago Matej Hrica pointed out a possible buffer overrun when receiving data from the qemu socket. Stefano recently proposed a fix for this, but I don't think it's quite right. This series is a different approach to fixing that problem and a few adjacent ones. David Gibson (5): tap: Better report errors receiving from QEMU socket tap: Don't attempt to carry on if we get a bad frame length from qemu tap: Don't use EPOLLET on Qemu sockets tap: Correctly handle frames of odd length tap: Improve handling of partially received frames on qemu socket passt.h | 1 - tap.c | 72 ++++++++++++++++++++++++++++++--------------------------- util.h | 16 +++++++++++++ 3 files changed, 54 insertions(+), 35 deletions(-) -- 2.45.2

2 17

[PATCH 0/3] Fixes for log timestamps
by Stefano Brivio 29 Jul '24

29 Jul '24

Taken out of the "Minor assorted fixes, mostly logging and tests" series I recently posted. Stefano Brivio (3): log, util: Fix sub-second part in relative log time calculation log: Initialise timestamp for relative log time also if we use a log file log: Fetch log times with CLOCK_MONOTONIC, not CLOCK_REALTIME log.c | 46 +++++++++++++++++++++++----------------------- log.h | 1 + passt.c | 2 ++ util.c | 25 ++++++++++++++++++------- util.h | 1 + 5 files changed, 45 insertions(+), 30 deletions(-) -- 2.43.0

2 7

[PATCH] Fix typo in README file
by AbdAlRahman Gad 27 Jul '24

27 Jul '24

- remove duplicated 'the' in the 'Services' section --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index aa2afac..bb114ea 100644 --- a/README.md +++ b/README.md @@ -398,7 +398,7 @@ services: and nameserver using SLAAC * [DHCPv6 server](/passt/tree/dhcpv6.c): a simple implementation handing out one single IPv6 address to the guest or namespace, - namely, the the same address as the first one configured for the upstream host + namely, the same address as the first one configured for the upstream host interface, and passing the nameservers configured on the host ## Addresses -- 2.43.0

1 0