- dev - passt

[PATCH v2] tap: Trim Ethernet padding from short IPv4 frames instead of dropping them
by David du Colombier 22 Jun '26

22 Jun '26

tap4_handler() requires the L2 payload after the IP header to match the IP datagram length exactly. Guests whose drivers pad transmitted frames to the 60 byte Ethernet minimum, as real hardware requires and as drivers modelled on hardware do (Plan 9's virtio-net, for one), send pure ACK and FIN segments as 60 byte frames: 14 byte Ethernet header, 40 byte IPv4 datagram, 6 padding octets. Those frames fail the exact length check and are dropped without trace. passt then never sees such a guest's acknowledgements: it retransmits from the lowest unacknowledged sequence with exponential backoff while the guest, which received and acknowledged everything, waits. Every fresh connection stalls for minutes (a 1 MiB HTTP fetch over --map-host-loopback measured 248 s before this change, 0.27 s after; bulk transfer over established connections, whose ACKs ride data segments above the padding threshold, is unaffected). FIN segments are padded too, so teardown hangs as well. Note that tap_send_single() pads passt's own outbound frames to ETH_ZLEN, so the receive path was already stricter than the send path. Trim the trailing padding to the IP datagram length instead, using a new iov_tail_trim() helper, and keep dropping frames genuinely shorter than the datagram they claim to carry. IPv6 is unaffected: its minimal TCP frame is 74 bytes, above the padding threshold. Signed-off-by: David du Colombier <0intro(a)gmail.com> --- v2: - store the iov_tail_size() result in a local variable instead of computing it twice - pass ARRAY_SIZE(trim_iov) instead of UIO_MAXIOV iov.c | 35 +++++++++++++++++++++++++++++++++++ iov.h | 2 ++ tap.c | 15 +++++++++++++-- 3 files changed, 50 insertions(+), 2 deletions(-) diff --git a/iov.c b/iov.c index 6fd684a..968a365 100644 --- a/iov.c +++ b/iov.c @@ -450,3 +450,38 @@ ssize_t iov_tail_clone(struct iovec *dst_iov, size_t dst_iov_cnt, return j; } + +/** + * iov_tail_trim() - Limit a tail to @len bytes via a scratch iovec array + * @tail: Pointer to the iov_tail to trim; rebuilt on success to + * reference @scratch + * @len: Number of bytes to keep + * @scratch: Scratch iovec array backing the trimmed tail; must stay + * valid as long as the trimmed tail is in use + * @scratch_cnt: Number of elements in @scratch + * + * Return: true on success, false if @tail is shorter than @len or does + * not fit in @scratch (@tail is unchanged on failure) + */ +bool iov_tail_trim(struct iov_tail *tail, size_t len, + struct iovec *scratch, size_t scratch_cnt) +{ + ssize_t cnt = iov_tail_clone(scratch, scratch_cnt, tail); + size_t left = len; + unsigned int i; + + if (cnt < 0) + return false; + + for (i = 0; i < (size_t)cnt && left; i++) { + if (scratch[i].iov_len > left) + scratch[i].iov_len = left; + left -= scratch[i].iov_len; + } + + if (left) + return false; + + *tail = IOV_TAIL(scratch, i, 0); + return true; +} diff --git a/iov.h b/iov.h index 4fdf14a..3af467e 100644 --- a/iov.h +++ b/iov.h @@ -97,6 +97,8 @@ size_t iov_push_header_(struct iov_tail *tail, const void *v, size_t len); void *iov_remove_header_(struct iov_tail *tail, void *v, size_t len, size_t align); ssize_t iov_tail_clone(struct iovec *dst_iov, size_t dst_iov_cnt, struct iov_tail *tail); +bool iov_tail_trim(struct iov_tail *tail, size_t len, + struct iovec *scratch, size_t scratch_cnt); /** * IOV_PEEK_HEADER() - Get typed pointer to a header from an IOV tail diff --git a/tap.c b/tap.c index 4cba4c7..6ed7f8d 100644 --- a/tap.c +++ b/tap.c @@ -716,7 +716,8 @@ static int tap4_handler(struct ctx *c, const struct pool *in, i = 0; resume: for (seq_count = 0, seq = NULL; i < in->count; i++) { - size_t l3len, hlen, l4len; + struct iovec trim_iov[UIO_MAXIOV]; + size_t l3len, hlen, l4len, check; struct ethhdr eh_storage; struct iphdr iph_storage; struct udphdr uh_storage; @@ -775,7 +776,17 @@ resume: if (!iov_drop_header(&data, hlen)) continue; - if (iov_tail_size(&data) != l4len) + + check = iov_tail_size(&data); + if (check < l4len) + continue; + + /* Drivers modelled on real hardware (Plan 9's virtio, for + * one) pad short frames to the 60 byte Ethernet minimum: + * trim trailing padding instead of dropping the packet. + */ + if (check > l4len && + !iov_tail_trim(&data, l4len, trim_iov, ARRAY_SIZE(trim_iov))) continue; if (iph->protocol == IPPROTO_ICMP) { -- 2.54.0

3 2

[PATCH v4 0/4] Add --dhcp-boot and --dhcp-opt options
by Anshu Kumari 22 Jun '26

22 Jun '26

This series adds support for custom DHCP options in passt, enabling network boot (PXE/UEFI HTTP Boot) and arbitrary DHCP option injection. Two new command-line flags are introduced: --dhcp-boot URL Sets the boot file URL (DHCP option 67 and the legacy boot file field) --dhcp-opt CODE,VALUE Sets any DHCP option by numeric code, with type-aware parsing per RFC 2132 The DHCP reply path is extended with option overload support (RFC 2132 option 52), allowing options to overflow into the file and sname fields when the standard options area is full. v4: Restructured series from 6 patches to 4: - 1/4 refactor fill_one() - 2/4 option overload - 3/4 --dhcp-opt with parser and man page - 4/4 --dhcp-boot with man page Anshu Kumari (4): dhcp: Refactor fill_one() to operate on a generic buffer dhcp: Add option overload dhcp: Add --dhcp-opt with option table and value parser conf: Add --dhcp-boot command-line option conf.c | 50 ++++++++- dhcp.c | 321 ++++++++++++++++++++++++++++++++++++++++++++++++++++---- dhcp.h | 2 + passt.1 | 49 +++++++++ passt.h | 6 ++ 5 files changed, 407 insertions(+), 21 deletions(-) -- 2.54.0

2 12

[PATCH] udp, icmp: Remove unused timer_run fields from protocol contexts
by Laurent Vivier 22 Jun '26

22 Jun '26

Since 260075bde769 ("tcp, udp, fwd: Run all port scanning from a single timer"), CALL_PROTO_HANDLER() is only used for TCP. UDP and ICMP timers now run from flow_defer_handler(), making the timer_run fields in struct udp_ctx and struct icmp_ctx dead stores. Remove the unused fields, drop struct icmp_ctx entirely (it has no remaining members), and stop initialising them in timer_init(). Signed-off-by: Laurent Vivier <lvivier(a)redhat.com> --- icmp.h | 7 ------- passt.c | 2 +- passt.h | 2 -- udp.h | 2 -- 4 files changed, 1 insertion(+), 12 deletions(-) diff --git a/icmp.h b/icmp.h index e8b928b61370..556260461995 100644 --- a/icmp.h +++ b/icmp.h @@ -19,12 +19,5 @@ int icmp_tap_handler(const struct ctx *c, uint8_t pif, sa_family_t af, struct iov_tail *data, const struct timespec *now); void icmp_init(void); -/** - * struct icmp_ctx - Execution context for ICMP routines - * @timer_run: Timestamp of most recent timer run - */ -struct icmp_ctx { - struct timespec timer_run; -}; #endif /* ICMP_H */ diff --git a/passt.c b/passt.c index b1d35d5bf988..677c8471a105 100644 --- a/passt.c +++ b/passt.c @@ -141,7 +141,7 @@ static void random_init(struct ctx *c) */ static void timer_init(struct ctx *c, const struct timespec *now) { - c->tcp.timer_run = c->udp.timer_run = c->icmp.timer_run = *now; + c->tcp.timer_run = *now; } /** diff --git a/passt.h b/passt.h index 3a072945c382..205d7eecede7 100644 --- a/passt.h +++ b/passt.h @@ -194,7 +194,6 @@ struct ip6_ctx { * @no_udp: Disable UDP operation * @udp: Context for UDP protocol handler * @no_icmp: Disable ICMP operation - * @icmp: Context for ICMP protocol handler * @mtu: MTU passed via DHCP/NDP * @no_dns: Do not source/use DNS servers for any purpose * @no_dns_search: Do not source/use domain search lists for any purpose @@ -279,7 +278,6 @@ struct ctx { int no_udp; struct udp_ctx udp; int no_icmp; - struct icmp_ctx icmp; int no_dns; int no_dns_search; diff --git a/udp.h b/udp.h index 9ea7de682caf..b50283eab198 100644 --- a/udp.h +++ b/udp.h @@ -26,14 +26,12 @@ void udp_update_l2_buf(const unsigned char *eth_d); * struct udp_ctx - Execution context for UDP * @scan_in: Port scanning state for inbound packets * @scan_out: Port scanning state for outbound packets - * @timer_run: Timestamp of most recent timer run * @timeout: Timeout for unidirectional flows (in s) * @stream_timeout: Timeout for stream-like flows (in s) */ struct udp_ctx { struct fwd_scan scan_in; struct fwd_scan scan_out; - struct timespec timer_run; int timeout; int stream_timeout; }; -- 2.54.0

2 1

[PATCH] test/pasta/ndp: Fix NDP test for --config-net pasta setup
by Jon Maloy 21 Jun '26

21 Jun '26

--config-net sets IPv6 addresses directly via netlink rather than relying on kernel SLAAC, so addresses won't have the "kernel_ra" protocol tag. Match global-scope addresses instead, which works for both --config-net and SLAAC configurations. Signed-off-by: Jon Maloy <jmaloy(a)redhat.com> --- test/pasta/ndp | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/test/pasta/ndp b/test/pasta/ndp index 1d385c73..7169f57a 100644 --- a/test/pasta/ndp +++ b/test/pasta/ndp @@ -18,11 +18,11 @@ test Interface name nsout IFNAME ip -j link show | jq -rM '.[] | select(.link_type == "ether").ifname' check [ -n "__IFNAME__" ] ns ip link set dev __IFNAME__ up -# Wait for SLAAC & DAD to complete -ns while ! ip -j -6 addr show dev __IFNAME__ | jq -e '.[].addr_info.[] | select(.protocol == "kernel_ra")'; do sleep 0.1; done +# Wait for global IPv6 address to appear (via --config-net or SLAAC) +ns while ! ip -j -6 addr show dev __IFNAME__ | jq -e '.[].addr_info[] | select(.scope == "global")'; do sleep 0.1; done test SLAAC: prefix -nsout ADDR6 ip -j -6 addr show|jq -rM '[.[] | select(.ifname == "__IFNAME__").addr_info[] | select(.protocol == "kernel_ra") | .local + "/" + (.prefixlen | tostring)] | .[0]' +nsout ADDR6 ip -j -6 addr show|jq -rM '[.[] | select(.ifname == "__IFNAME__").addr_info[] | select(.scope == "global") | .local + "/" + (.prefixlen | tostring)] | .[0]' nsout PREFIX6 sipcalc __ADDR6__ | grep prefix | cut -d' ' -f4 hout HOST_ADDR6 ip -j -6 addr show|jq -rM ['.[] | select(.ifname == "__IFNAME__").addr_info[] | select(.scope != "host" and .scope != "link" and .deprecated != true).local]| join("/64 ")' hout HOST_PREFIX6 sipcalc __HOST_ADDR6__/64 | grep prefix | cut -d' ' -f4 | tr '\n' ' ' -- 2.52.0

2 1

[PATCH] test/pasta/dhcp: Fix DHCP tests for --config-net pasta setup
by Jon Maloy 21 Jun '26

21 Jun '26

setup_pasta() now passes --config-net, which pre-configures the namespace interface via NDP/SLAAC before dhclient runs. This caused several issues: - dhclient -6 using the host's dhclient-script failed inside user namespaces (NetworkManager hooks, resolv.conf writes), causing infinite SOLICIT/REPLY loops. Add a minimal dhclient-script that applies leases with 'ip addr replace' and exits cleanly. - The DHCPv6 address check filtered for prefixlen == 128, but --config-net already configures the address with /64 via SLAAC, and 'ip -6 addr replace' cannot change the prefix length of an existing address. Match by scope == "global" instead. - Copy dhclient to STATEDIR and use explicit lease files to avoid conflicts with the host's dhclient state. - Quote variable expansions in test comparisons to prevent failures on empty values. Signed-off-by: Jon Maloy <jmaloy(a)redhat.com> --- test/lib/dhclient-script | 17 +++++++++++++++++ test/pasta/dhcp | 14 ++++++++------ 2 files changed, 25 insertions(+), 6 deletions(-) create mode 100755 test/lib/dhclient-script diff --git a/test/lib/dhclient-script b/test/lib/dhclient-script new file mode 100755 index 00000000..c6f4c0fc --- /dev/null +++ b/test/lib/dhclient-script @@ -0,0 +1,17 @@ +#!/bin/sh +# +# SPDX-License-Identifier: GPL-2.0-or-later +# +# Minimal dhclient-script for pasta tests. The host's dhclient-script +# fails inside user namespaces (NetworkManager hooks, resolv.conf +# writes, etc.), so we use this stripped-down version instead. +# Based on the guest script in test/passt.mbuto. + +[ -n "${new_interface_mtu}" ] && ip link set dev ${interface} mtu ${new_interface_mtu} + +[ -n "${new_ip_address}" ] && ip addr replace ${new_ip_address}/${new_subnet_mask} dev ${interface} +[ -n "${new_routers}" ] && for r in ${new_routers}; do ip route replace default via ${r} dev ${interface}; done + +[ -n "${new_ip6_address}" ] && ip -6 addr replace ${new_ip6_address}/${new_ip6_prefixlen:-128} dev ${interface} nodad + +exit 0 diff --git a/test/pasta/dhcp b/test/pasta/dhcp index 366935ff..5d7e78b6 100644 --- a/test/pasta/dhcp +++ b/test/pasta/dhcp @@ -19,30 +19,32 @@ nsout IFNAME ip -j link show | jq -rM '.[] | select(.link_type == "ether").ifnam check [ -n "__IFNAME__" ] test DHCP: address -ns /sbin/dhclient -4 --no-pid __IFNAME__ +ns cp /sbin/dhclient __STATEDIR__/dhclient +ns __STATEDIR__/dhclient -4 --no-pid -lf __STATEDIR__/dhclient4.leases __IFNAME__ +sleep 1 nsout ADDR ip -j -4 addr show|jq -rM '.[] | select(.ifname == "__IFNAME__").addr_info[0].local' hout HOST_ADDR ip -j -4 addr show|jq -rM '.[] | select(.ifname == "__IFNAME__").addr_info[0].local' -check [ __ADDR__ = __HOST_ADDR__ ] +check [ "__ADDR__" = "__HOST_ADDR__" ] test DHCP: route nsout GW ip -j -4 route show|jq -rM '.[] | select(.dst == "default").gateway' hout HOST_GW ip -j -4 route show|jq -rM '[.[] | select(.dst == "default").gateway] | .[0]' -check [ __GW__ = __HOST_GW__ ] +check [ "__GW__" = "__HOST_GW__" ] test DHCP: MTU nsout MTU ip -j link show | jq -rM '.[] | select(.ifname == "__IFNAME__").mtu' check [ __MTU__ = 65520 ] test DHCPv6: address -ns /sbin/dhclient -6 --no-pid __IFNAME__ +ns __STATEDIR__/dhclient -6 -sf __BASEPATH__/lib/dhclient-script --no-pid -lf __STATEDIR__/dhclient6.leases __IFNAME__ # Wait for DAD to complete ns while ip -j -6 addr show tentative | jq -e '.[].addr_info'; do sleep 0.1; done hout HOST_IFNAME6 ip -j -6 route show|jq -rM '[.[] | select(.dst == "default").dev] | .[0]' -nsout ADDR6 ip -j -6 addr show|jq -rM '[.[] | select(.ifname == "__IFNAME__").addr_info[] | select(.prefixlen == 128).local] | .[0]' +nsout ADDR6 ip -j -6 addr show|jq -rM '[.[] | select(.ifname == "__IFNAME__").addr_info[] | select(.scope == "global").local] | .[0]' hout HOST_ADDR6 ip -j -6 addr show|jq -rM '[.[] | select(.ifname == "__HOST_IFNAME6__").addr_info[] | select(.scope != "host" and .scope != "link" and .deprecated != true).local] | join(" ")' check echo "__HOST_ADDR6__" | grep -wq "__ADDR6__" test DHCPv6: route nsout GW6 ip -j -6 route show|jq -rM '.[] | select(.dst == "default").gateway' hout HOST_GW6 ip -j -6 route show|jq -rM '[.[] | select(.dst == "default").gateway] | .[0]' -check [ __GW6__ = __HOST_GW6__ ] +check [ "__GW6__" = "__HOST_GW6__" ] -- 2.52.0

1 0

[PATCH] test/lib: Add --config-net to setup_pasta()
by Jon Maloy 21 Jun '26

21 Jun '26

Since commit b4dace8 ("fwd: Direct inbound spliced forwards to the guest's external address"), inbound splice connections target the guest's external address (addr_seen) rather than loopback. This requires the namespace tap interface to be configured with that address. setup_pasta() starts pasta without --config-net, so the tap interface in the namespace is never brought up and has no addresses or routes. Splice connections to addr_seen fail because the address is unreachable in the namespace, breaking all pasta TCP and UDP tests. Add --config-net to the pasta command line in setup_pasta(), matching what setup_passt_in_ns() and setup_two_guests() already do. Signed-off-by: Jon Maloy <jmaloy(a)redhat.com> --- test/lib/setup | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test/lib/setup b/test/lib/setup index 59945987..dbf8079d 100755 --- a/test/lib/setup +++ b/test/lib/setup @@ -107,7 +107,7 @@ setup_pasta() { [ ${DEBUG} -eq 1 ] && __opts="${__opts} -d" [ ${TRACE} -eq 1 ] && __opts="${__opts} --trace" - context_run_bg passt "./pasta ${__opts} -f -t 10002 -T 10003 -u 10002 -U 10003 -P ${STATESETUP}/passt.pid $(${NSTOOL} info -pw ${STATESETUP}/ns.hold)" + context_run_bg passt "./pasta ${__opts} -f -t 10002 -T 10003 -u 10002 -U 10003 --config-net -P ${STATESETUP}/passt.pid $(${NSTOOL} info -pw ${STATESETUP}/ns.hold)" # pidfile isn't created until pasta is ready wait_for [ -f "${STATESETUP}/passt.pid" ] -- 2.52.0

2 1

[PATCH] test/lib: Pre-create context log files before starting tail
by Jon Maloy 21 Jun '26

21 Jun '26

pane_watch_contexts() starts tail -f --retry on context log files, but those files may not exist yet when tail starts, resulting in error messages in the tmux pane. Pre-create them with touch so tail can open them immediately. Signed-off-by: Jon Maloy <jmaloy(a)redhat.com> --- test/lib/term | 1 + 1 file changed, 1 insertion(+) diff --git a/test/lib/term b/test/lib/term index 2d2d2182..23639e34 100755 --- a/test/lib/term +++ b/test/lib/term @@ -252,6 +252,7 @@ pane_watch_contexts() { tmux select-pane -t ${__pane_number} -T "${__desc}" __cmd="tail -f --retry" for c; do + touch "${LOGDIR}/context_${c}.log" __cmd="${__cmd} ${LOGDIR}/context_${c}.log" done cmd_write ${__pane_number} "${__cmd}" -- 2.52.0

2 1

[PATCH] inany: Fix doc comment to match u32 field, not u64
by Laurent Vivier 19 Jun '26

19 Jun '26

The original siphash incremental patch added a u64[2] member to union inany_addr for hashing. During review, this was changed to u32[4] to preserve 4-byte alignment: a u64 member would have forced 8-byte alignment on the union, mismatching struct in6_addr. The doc comment was not updated, fix @u64 references to @u32. Fixes: fc8f0f8c48ef ("siphash: Use incremental rather than all-at-once siphash functions") Signed-off-by: Laurent Vivier <lvivier(a)redhat.com> --- inany.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/inany.h b/inany.h index 73385b916240..6bf3ecaa0e7d 100644 --- a/inany.h +++ b/inany.h @@ -22,9 +22,9 @@ * @v4mapped.zero: All zero-bits for an IPv4 address * @v4mapped.one: All one-bits for an IPv4 address * @v4mapped.a4: If @a6 is an IPv4 mapped address, the IPv4 address - * @u64: As an array of u64s (solely for hashing) + * @u32: As an array of u32s (solely for hashing) * - * @v4mapped and @u64 shouldn't be accessed except via helpers. + * @v4mapped and @u32 shouldn't be accessed except via helpers. */ union inany_addr { struct in6_addr a6; -- 2.54.0

3 2

[PATCH] tap: don't let overheard traffic move addr_seen when serving a fixed address
by EJ Campbell 19 Jun '26

19 Jun '26

The latest source address seen on the tap interface (ip4.addr_seen) is taken to be the guest's and is used as the target for inbound port forwarding. The update accepts any source address. On a point-to-point tap that is always correct. But when the tap sits on a shared L2 segment -- for example when it is bridged together with other hosts -- pasta also sees frames from those other hosts. With a static configuration (-a / address set, DHCP disabled), an overheard frame from another host silently moves addr_seen to that host, and every subsequent inbound forwarded connection is sent there instead of to the guest, so connections are reset or hang. I hit this running pasta with its tap enslaved to a Linux bridge that also carries other traffic: inbound port forwarding to the guest broke as soon as another address was seen on the segment. Restrict the update so that, when DHCP is disabled (no_dhcp), only frames sourced from the configured guest address may move addr_seen. With DHCP enabled -- the default -- behaviour is unchanged, since the guest's address is whatever pasta leases and tracking it is still correct. The same reasoning applies to ip6.addr_seen on a shared segment; I left IPv6 alone for now since I only exercise IPv4 in this setup, and would be glad to extend it the same way if you prefer. Signed-off-by: EJ Campbell <ej.campbell(a)gmail.com> --- tap.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/tap.c b/tap.c index 4cba4c7..514379e 100644 --- a/tap.c +++ b/tap.c @@ -770,7 +770,19 @@ resume: continue; } - if (iph->saddr && c->ip4.addr_seen.s_addr != iph->saddr) + /* The latest source address seen on the tap is taken to be the + * guest's, and becomes the target for inbound forwarding. On a + * point-to-point tap that's always correct. But if the tap sits + * on a shared L2 segment (for example bridged together with + * other hosts) and we serve a fixed address (DHCP disabled), a + * frame overheard from another host on that segment must not + * move addr_seen and silently retarget forwarded connections + * away from the guest. When DHCP is disabled, only the + * configured guest address may update addr_seen; with DHCP + * enabled (the default) behaviour is unchanged. + */ + if (iph->saddr && c->ip4.addr_seen.s_addr != iph->saddr && + (!c->no_dhcp || iph->saddr == c->ip4.addr.s_addr)) c->ip4.addr_seen.s_addr = iph->saddr; if (!iov_drop_header(&data, hlen)) base-commit: 4b2823784aab04a70dfc295b16fd6f0592955790 -- 2.53.0

3 6

[PATCH v2 0/6] Improvements to flow specific logging
by David Gibson 19 Jun '26

19 Jun '26

While working on podman bug 23739, I found some deficiences in how and when we report various socket errors related to specific flows. Here are some preliminary patches to improve that. I think patches 1..3 are pretty straightforward and safe. 4/4 I think is a good idea in principle - promoting the priority of a number of messages now that we have ratelimiting. However, plumbing 'now' all the places we need it resuls in a *lot* of churn, so we might want to postpone or split it up. Still, current drafted posted for consideration. v2: * Added patch 2/6 At Stefano's suggestion * Simplified the errno Handling fix at Stefano's suggestion * Added patch 6/6 which I noticed while revising 5/6. * Made a handful of minor corrections based on feedback. David Gibson (6): flow: Regularise flow specific logging helpers flow: Indent flow details messages flow: Include flow details with higher priority log messages flow, udp: Fix errno handling in udp_flow_sock() flow, treewide: Promote priority of selected flow-linked messages udp: Improve messages for errors getting errors flow.c | 103 ++++++------- flow.h | 72 ++++----- icmp.c | 27 ++-- icmp.h | 3 +- migrate.c | 27 ++-- migrate.h | 8 +- passt.c | 8 +- tcp.c | 387 ++++++++++++++++++++++++++++--------------------- tcp.h | 5 +- tcp_buf.c | 67 +++++---- tcp_buf.h | 7 +- tcp_conn.h | 8 +- tcp_internal.h | 27 ++-- tcp_splice.c | 59 +++++--- tcp_splice.h | 3 +- tcp_vu.c | 31 ++-- tcp_vu.h | 5 +- udp.c | 61 +++++--- udp_flow.c | 18 ++- 19 files changed, 528 insertions(+), 398 deletions(-) -- 2.54.0

2 7