June 2026 - dev

[PATCH v2 2/6] flow: Indent flow details messages
by David Gibson 17 Jun '26

17 Jun '26

flow_log__() has the option to append an extra log message with address details of the flow, to give more context to errors or warnings. Add an indent to this part of the message, to emphasise that it's additional information related to a nearby message, rather than a message in its own right. Suggested-by: Stefano Brivio <sbrivio(a)redhat.com> Signed-off-by: David Gibson <david(a)gibson.dropbear.id.au> --- flow.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/flow.c b/flow.c index e71f3de7..2f1530b5 100644 --- a/flow.c +++ b/flow.c @@ -310,7 +310,7 @@ void flow_log__(const struct flow_common *f, int pri, bool perror, bool details, if (state >= FLOW_STATE_TGT) { flow_log__(f, pri, false, false, state, -"%s [%s]:%hu -> [%s]:%hu => %s [%s]:%hu -> [%s]:%hu", +" %s [%s]:%hu -> [%s]:%hu => %s [%s]:%hu -> [%s]:%hu", pif_name(f->pif[INISIDE]), inany_ntop(&ini->eaddr, estr0, sizeof(estr0)), ini->eport, @@ -323,7 +323,7 @@ void flow_log__(const struct flow_common *f, int pri, bool perror, bool details, tgt->eport); } else if (state >= FLOW_STATE_INI) { flow_log__(f, pri, false, false, state, - "%s [%s]:%hu -> [%s]:%hu => ?", + " %s [%s]:%hu -> [%s]:%hu => ?", pif_name(f->pif[INISIDE]), inany_ntop(&ini->eaddr, estr0, sizeof(estr0)), ini->eport, -- 2.54.0

1 0

[PATCH 1/4] flow: Regularise flow specific logging helpers
by David Gibson 17 Jun '26

17 Jun '26

flow.h has a collection of logging helpers that automatically include information about a specific flow. Which variants are present are a bit ad-hoc, based on what we happened to want use (e.g. there are no LOG_WARNING level versions, at present). There's also a rather awkward and only occasionally used flow_log_details_() helper to print additional log messages with more details of the flow (basically its addresses). It's particularly awkward to try to combine that with ratelimiting. Re-organise this to be based around a flow_log__() internal helper, which has bool parameters to include strerror() / perror information and/or the extra details. Add wrapper macros for all combinations of perror, ratelimiting and DEBUG/WARNING/ERR priorities. Be a little more consistent about parameter order between the various functions / macros / wrappers while we're at it. Signed-off-by: David Gibson <david(a)gibson.dropbear.id.au> --- flow.c | 89 ++++++++++++++++++++++++++++------------------------------ flow.h | 70 +++++++++++++++++++++++---------------------- tcp.c | 4 +-- udp.c | 6 ++-- 4 files changed, 85 insertions(+), 84 deletions(-) diff --git a/flow.c b/flow.c index 565ed2b2..6cf3905a 100644 --- a/flow.c +++ b/flow.c @@ -270,17 +270,20 @@ int flowside_connect(const struct ctx *c, int s, return connect(s, &sa.sa, socklen_inany(&sa)); } -/** flow_log_ - Log flow-related message +/** flow_log__ - Log flow-related message, internal helper * @f: flow the message is related to - * @newline: Append newline at the end of the message, if missing * @pri: Log priority + * @perror: Append strerror(errno) output + * @details: Add lines with flow addresses + * @state: State to display details for (used during state changes) * @fmt: Format string * @...: printf-arguments */ -void flow_log_(const struct flow_common *f, bool newline, int pri, - const char *fmt, ...) +void flow_log__(const struct flow_common *f, int pri, bool perror, bool details, + enum flow_state state, const char *fmt, ...) { const char *type_or_state; + int errno_ = errno; char msg[BUFSIZ]; va_list args; @@ -294,45 +297,41 @@ void flow_log_(const struct flow_common *f, bool newline, int pri, else type_or_state = FLOW_TYPE(f); - logmsg(newline, false, pri, + logmsg(!perror, false, pri, "Flow %u (%s): %s", flow_idx(f), type_or_state, msg); -} -/** flow_log_details_() - Log the details of a flow - * @f: flow to log - * @pri: Log priority - * @state: State to log details according to - * - * Logs the details of the flow: endpoints, interfaces, type etc. - */ -void flow_log_details_(const struct flow_common *f, int pri, - enum flow_state state) -{ - char estr0[INANY_ADDRSTRLEN], fstr0[INANY_ADDRSTRLEN]; - char estr1[INANY_ADDRSTRLEN], fstr1[INANY_ADDRSTRLEN]; - const struct flowside *ini = &f->side[INISIDE]; - const struct flowside *tgt = &f->side[TGTSIDE]; - - if (state >= FLOW_STATE_TGT) - flow_log_(f, true, pri, - "%s [%s]:%hu -> [%s]:%hu => %s [%s]:%hu -> [%s]:%hu", - pif_name(f->pif[INISIDE]), - inany_ntop(&ini->eaddr, estr0, sizeof(estr0)), - ini->eport, - inany_ntop(&ini->oaddr, fstr0, sizeof(fstr0)), - ini->oport, - pif_name(f->pif[TGTSIDE]), - inany_ntop(&tgt->oaddr, fstr1, sizeof(fstr1)), - tgt->oport, - inany_ntop(&tgt->eaddr, estr1, sizeof(estr1)), - tgt->eport); - else if (state >= FLOW_STATE_INI) - flow_log_(f, true, pri, "%s [%s]:%hu -> [%s]:%hu => ?", - pif_name(f->pif[INISIDE]), - inany_ntop(&ini->eaddr, estr0, sizeof(estr0)), - ini->eport, - inany_ntop(&ini->oaddr, fstr0, sizeof(fstr0)), - ini->oport); + if (perror) + logmsg(true, true, pri, ": %s", strerror_(errno_)); + + if (details) { + char estr0[INANY_ADDRSTRLEN], fstr0[INANY_ADDRSTRLEN]; + char estr1[INANY_ADDRSTRLEN], fstr1[INANY_ADDRSTRLEN]; + const struct flowside *ini = &f->side[INISIDE]; + const struct flowside *tgt = &f->side[TGTSIDE]; + + if (state >= FLOW_STATE_TGT) { + flow_log__(f, pri, false, false, state, +"%s [%s]:%hu -> [%s]:%hu => %s [%s]:%hu -> [%s]:%hu", + pif_name(f->pif[INISIDE]), + inany_ntop(&ini->eaddr, estr0, sizeof(estr0)), + ini->eport, + inany_ntop(&ini->oaddr, fstr0, sizeof(fstr0)), + ini->oport, + pif_name(f->pif[TGTSIDE]), + inany_ntop(&tgt->oaddr, fstr1, sizeof(fstr1)), + tgt->oport, + inany_ntop(&tgt->eaddr, estr1, sizeof(estr1)), + tgt->eport); + } else if (state >= FLOW_STATE_INI) { + flow_log__(f, pri, false, false, state, + "%s [%s]:%hu -> [%s]:%hu => ?", + pif_name(f->pif[INISIDE]), + inany_ntop(&ini->eaddr, estr0, sizeof(estr0)), + ini->eport, + inany_ntop(&ini->oaddr, fstr0, sizeof(fstr0)), + ini->oport); + } + } } /** @@ -348,10 +347,9 @@ static void flow_set_state(struct flow_common *f, enum flow_state state) assert(oldstate < FLOW_NUM_STATES); f->state = state; - flow_log_(f, true, LOG_DEBUG, "%s -> %s", flow_state_str[oldstate], + flow_log__(f, LOG_DEBUG, false, true, MAX(state, oldstate), + "%s -> %s", flow_state_str[oldstate], FLOW_STATE(f)); - - flow_log_details_(f, LOG_DEBUG, MAX(state, oldstate)); } /** @@ -552,8 +550,7 @@ norule: /* This shouldn't happen, because if there's no rule for it we should * have no listening socket that would let us get here */ - flow_dbg(flow, "Missing forward rule"); - flow_log_details_(f, LOG_DEBUG, f->state); + flow_log(flow, LOG_DEBUG, false, true, "Missing forward rule"); nofwd: flow_err(flow, "No rules to forward %s %s [%s]:%hu -> [%s]:%hu", diff --git a/flow.h b/flow.h index 6c6a9260..d168a35a 100644 --- a/flow.h +++ b/flow.h @@ -280,42 +280,46 @@ int flow_migrate_source(struct ctx *c, const struct migrate_stage *stage, int flow_migrate_target(struct ctx *c, const struct migrate_stage *stage, int fd); -void flow_log_(const struct flow_common *f, bool newline, int pri, - const char *fmt, ...) - __attribute__((format(printf, 4, 5))); - -#define flow_log(f_, pri, ...) flow_log_(&(f_)->f, true, (pri), __VA_ARGS__) -#define flow_dbg(f, ...) flow_log((f), LOG_DEBUG, __VA_ARGS__) -#define flow_err(f, ...) flow_log((f), LOG_ERR, __VA_ARGS__) - -#define flow_trace(f, ...) \ +void flow_log__(const struct flow_common *f, int pri, bool perror, bool details, + enum flow_state state, const char *fmt, ...); + +#define flow_log_(f_, pri_, perror_, details_, ...) \ + flow_log__((f_), (pri_), (perror_), (details_), (f_)->state, \ + __VA_ARGS__) + +#define flow_log(flow_, pri_, perror_, details_, ...) \ + flow_log_(&(flow_)->f, (pri_), (perror_), (details_), __VA_ARGS__) + +#define flow_dbg(flow_, ...) \ + flow_log((flow_), LOG_DEBUG, false, false, __VA_ARGS__) +#define flow_warn(flow_, ...) \ + flow_log((flow_), LOG_WARNING, false, false, __VA_ARGS__) +#define flow_err(flow_, ...) \ + flow_log((flow_), LOG_ERR, false, false, __VA_ARGS__) +#define flow_trace(flow_, ...) \ do { \ if (log_trace) \ - flow_dbg((f), __VA_ARGS__); \ - } while (0) - -#define flow_log_perror_(f, pri, ...) \ - do { \ - int errno_ = errno; \ - flow_log_((f), false, (pri), __VA_ARGS__); \ - logmsg(true, true, (pri), ": %s", strerror_(errno_)); \ + flow_dbg((flow_), __VA_ARGS__); \ } while (0) -#define flow_dbg_perror(f_, ...) flow_log_perror_(&(f_)->f, LOG_DEBUG, __VA_ARGS__) -#define flow_perror(f_, ...) flow_log_perror_(&(f_)->f, LOG_ERR, __VA_ARGS__) - -void flow_log_details_(const struct flow_common *f, int pri, - enum flow_state state); -#define flow_log_details(f_, pri) \ - flow_log_details_(&((f_)->f), (pri), (f_)->f.state) -#define flow_dbg_details(f_) flow_log_details((f_), LOG_DEBUG) -#define flow_err_details(f_) flow_log_details((f_), LOG_ERR) - -#define flow_dbg_ratelimit(f, now, ...) \ - logmsg_ratelimit(flow_dbg, debug, now, f, __VA_ARGS__) -#define flow_err_ratelimit(f, now, ...) \ - logmsg_ratelimit(flow_err, err, now, f, __VA_ARGS__) -#define flow_perror_ratelimit(f, now, ...) \ - logmsg_ratelimit(flow_perror, err, now, f, __VA_ARGS__) +#define flow_dbg_perror(flow_, ...) \ + flow_log((flow_), LOG_DEBUG, true, false, __VA_ARGS__) +#define flow_warn_perror(flow_, ...) \ + flow_log((flow_), LOG_WARNING, true, false, __VA_ARGS__) +#define flow_perror(flow_, ...) \ + flow_log((flow_), LOG_ERR, true, false, __VA_ARGS__) + +#define flow_dbg_ratelimit(flow_, now_, ...) \ + logmsg_ratelimit(flow_dbg, debug, (now_), (flow_), __VA_ARGS__) +#define flow_dbg_perror_ratelimit(flow_, now_, ...) \ + logmsg_ratelimit(flow_dbg_perror, debug, (now_), (flow_), __VA_ARGS__) +#define flow_warn_ratelimit(flow_, now_, ...) \ + logmsg_ratelimit(flow_warn, warn, (now_), (flow_), __VA_ARGS__) +#define flow_warn_perror_ratelimit(flow_, now_, ...) \ + logmsg_ratelimit(flow_warn_perror, warn, (now_), (flow_), __VA_ARGS__) +#define flow_err_ratelimit(flow_, now_, ...) \ + logmsg_ratelimit(flow_err, err, (now_), (flow_), __VA_ARGS__) +#define flow_perror_ratelimit(flow_, now_, ...) \ + logmsg_ratelimit(flow_perror, err, (now_), (flow_), __VA_ARGS__) #endif /* FLOW_H */ diff --git a/tcp.c b/tcp.c index c4000754..6fba865f 100644 --- a/tcp.c +++ b/tcp.c @@ -1404,8 +1404,8 @@ void tcp_linger0_(const struct flow_common *f, int s) */ if (setsockopt(s, SOL_SOCKET, SO_LINGER, &linger0, sizeof(linger0)) < 0) { - flow_log_perror_(f, LOG_DEBUG, - "SO_LINGER failed, may not send RST to peer"); + flow_log_(f, LOG_DEBUG, true, false, + "SO_LINGER failed, may not send RST to peer"); } } diff --git a/udp.c b/udp.c index c28d6ee2..f29ca3da 100644 --- a/udp.c +++ b/udp.c @@ -943,7 +943,8 @@ void udp_sock_handler(const struct ctx *c, union epoll_ref ref, if (events & EPOLLERR) { if (udp_sock_errs(c, ref.fd, ref.flowside, PIF_NONE, 0) < 0) { - flow_err(uflow, "Unrecoverable error on flow socket"); + flow_log(uflow, LOG_ERR, false, true, + "Unrecoverable error on flow socket"); goto fail; } } @@ -974,7 +975,7 @@ void udp_sock_handler(const struct ctx *c, union epoll_ref ref, udp_buf_sock_to_tap(c, s, n, tosidx); } } else { - flow_err(uflow, + flow_log(uflow, LOG_ERR, false, true, "No support for forwarding UDP from %s to %s", pif_name(pif_at_sidx(ref.flowside)), pif_name(topif)); @@ -984,7 +985,6 @@ void udp_sock_handler(const struct ctx *c, union epoll_ref ref, return; fail: - flow_err_details(uflow); udp_flow_close(c, uflow); } -- 2.54.0

2 2

[PATCH v2 0/4] Assorted minor socket creation cleanups
by David Gibson 17 Jun '26

17 Jun '26

As discussed on our recent call, I was looking again at bug 167. I discovered it's still fairly fiddly to address this, but while investigating spotted a number of cleanups to make in the vicinity. I think they make sense even without fixing bug 167 (yet), so here they are. Tested on top of my recent cppcheck series, but I don't think they conflict. v2: * Fixed some cppcheck warnings I missed initially David Gibson (4): flow: Correct misleading signature of flowside_sock_l4() Makefile: Remove unused DUAL_STACK_SOCKETS define fwd, pif: Remove duplicated logic between tcp_listen() and udp_listen() pif, util: Move listen(2) call from sock_l4_() to pif_listen() Makefile | 5 ----- flow.c | 11 +++++------ flow.h | 2 +- fwd.c | 8 +------- pif.c | 41 +++++++++++++++++++++++++++++++++++------ pif.h | 2 +- tcp.c | 38 -------------------------------------- tcp.h | 2 -- udp.c | 38 -------------------------------------- udp.h | 2 -- util.c | 7 ------- 11 files changed, 43 insertions(+), 113 deletions(-) -- 2.54.0

2 5

[PATCH 0/5] Fix cppcheck-2.21.0 warnings
by David Gibson 17 Jun '26

17 Jun '26

Another cppcheck version, another batch of false positives :(. These appears to be cppcheck regressions, I've filed two bugs on their tracker. Still, here are workarounds in the passt code. At least I managed to make a few genuine if minor improvements to the code along the way. David Gibson (5): cppcheck: Remove unused CPPCHECK_6936 cppcheck: Add workaround for cppcheck bug 14847 tcp: Merge common sequence logic from tcp_{buf,vu}_data_from_sock() tcp: Avoid SEQ_*() comparisons against 0 tcp: MAX_WINDOW should be unsigned Makefile | 3 +-- tcp.c | 50 +++++++++++++++++++++++++++++++++++--------------- tcp_buf.c | 28 ++++------------------------ tcp_buf.h | 3 ++- tcp_internal.h | 4 ++-- tcp_vu.c | 27 ++++----------------------- tcp_vu.h | 3 ++- util.c | 6 ++++-- 8 files changed, 54 insertions(+), 70 deletions(-) -- 2.54.0

3 6

[PATCH 0/8] multithreading: Prepare data structures for concurrent queue pair workers
by Laurent Vivier 16 Jun '26

16 Jun '26

This series makes the passt data structures safe for concurrent access by multiple worker threads, each handling a different queue pair. It builds on the multiqueue series which threads the queue pair parameter through the network stack. The changes fall into three categories: 1. Per-queue-pair data isolation: - Convert packet pools (pool_tap4, pool_tap6) to per-queue-pair arrays - Convert L4 sequence batching arrays (tap4_l4, tap6_l4) similarly - Move flow and TCP timer state out of global/context structures into per-caller parameters 2. Stack-local buffers: - Move static iovec and virtqueue buffers onto the stack in tcp.c, tcp_vu.c, and udp_vu.c so each thread operates on its own copy 3. Locking for remaining shared state: - pthread_mutex on pre-opened socket pools (init_sock_pool4/6) - pthread_rwlock on the flow table, hash table, and free list, with read locks for lookups and write locks for mutations - _Thread_local for flow_new_entry so each thread independently tracks its in-progress allocation - Per-qpair filtering in flow_defer_handler() so each worker only processes its own flows Global operations that don't need per-queue-pair handling (tcp_payload_flush, socket pool refills) are guarded to run only on queue pair 0. Based-on: 20260616125130.1324274-1-lvivier(a)redhat.com Laurent Vivier (8): tap: Convert packet pools to per-queue-pair arrays for multiqueue tap: Make L4 sequence pools per-qpair for thread safety tcp: Make static buffers stack-local for thread safety udp_vu: Make virtqueue buffers stack-local for thread safety flow: Make flow timer per-caller for thread safety tcp: Make TCP timer state per-caller and guard global tasks tcp: Protect init socket pools with mutex for thread safety flow: Add mutex and per-qpair filtering to flow table operations flow.c | 68 +++++++++++++++++++++++++------- flow.h | 2 +- flow_table.h | 2 +- passt.c | 37 +++++++++-------- tap.c | 109 +++++++++++++++++++++++++++++---------------------- tap.h | 2 +- tcp.c | 64 +++++++++++++++++++----------- tcp.h | 9 +---- tcp_vu.c | 33 ++++++++++------ udp_vu.c | 4 +- vu_common.c | 2 +- 11 files changed, 203 insertions(+), 129 deletions(-) -- 2.54.0

1 8

[PATCH] tap: Trim Ethernet padding from short IPv4 frames instead of dropping them
by David du Colombier 15 Jun '26

15 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> --- iov.c | 35 +++++++++++++++++++++++++++++++++++ iov.h | 2 ++ tap.c | 11 ++++++++++- 3 files changed, 47 insertions(+), 1 deletion(-) 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..b929b21 100644 --- a/tap.c +++ b/tap.c @@ -716,6 +716,7 @@ static int tap4_handler(struct ctx *c, const struct pool *in, i = 0; resume: for (seq_count = 0, seq = NULL; i < in->count; i++) { + struct iovec trim_iov[UIO_MAXIOV]; size_t l3len, hlen, l4len; struct ethhdr eh_storage; struct iphdr iph_storage; @@ -775,7 +776,15 @@ resume: if (!iov_drop_header(&data, hlen)) continue; - if (iov_tail_size(&data) != l4len) + if (iov_tail_size(&data) < 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 (iov_tail_size(&data) > l4len && + !iov_tail_trim(&data, l4len, trim_iov, UIO_MAXIOV)) continue; if (iph->protocol == IPPROTO_ICMP) { -- 2.54.0

3 2

[PATCH 2/5] cppcheck: Add workaround for cppcheck bug 14847
by David Gibson 15 Jun '26

15 Jun '26

It seems the function pointer argument confuses cppcheck 2.21.0 (at least), causing it to think do_clone()'s definition has different argument names than it's declaration, even though that's not the case. I made a minimal reproducer and filed a cppcheck bug for this (see link). In the meanwhile work around it with an explicit suppression. Link: https://trac.cppcheck.net/ticket/14847 Signed-off-by: David Gibson <david(a)gibson.dropbear.id.au> --- util.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/util.c b/util.c index 43977d42..ed874f1c 100644 --- a/util.c +++ b/util.c @@ -713,8 +713,10 @@ int __clone2(int (*fn)(void *), void *stack_base, size_t stack_size, int flags, * * Return: thread ID of child, -1 on failure */ -int do_clone(int (*fn)(void *), char *stack_area, size_t stack_size, int flags, - void *arg) +int do_clone(int (*fn)(void *), +/* false positive, see https://trac.cppcheck.net/ticket/14847 */ +/* cppcheck-suppress [funcArgNamesDifferentUnnamed,unmatchedSuppression] */ + char *stack_area, size_t stack_size, int flags, void *arg) { #ifdef __ia64__ return __clone2(fn, stack_area + stack_size / 2, stack_size / 2, -- 2.54.0

1 0

[PATCH 1/5] cppcheck: Remove unused CPPCHECK_6936
by David Gibson 15 Jun '26

15 Jun '26

Our flags for cppcheck include -D CPPCHECK_6936. This used to enable a workaround for a cppcheck bug, but the code where we used it went away in commit 9153aca15bc1. Remove the now unused flag. Fixes: 9153aca15bc1 ("util: Add abort_with_msg() and ASSERT_WITH_MSG() helpers") Signed-off-by: David Gibson <david(a)gibson.dropbear.id.au> --- Makefile | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/Makefile b/Makefile index 0a0a60b0..c19383f3 100644 --- a/Makefile +++ b/Makefile @@ -208,8 +208,7 @@ CPPCHECK_FLAGS = --std=c11 --error-exitcode=1 --enable=all --force \ else \ echo ""; \ fi) \ - --suppress=missingIncludeSystem \ - -D CPPCHECK_6936 + --suppress=missingIncludeSystem cppcheck: passt.cppcheck passt-repair.cppcheck pesto.cppcheck -- 2.54.0

1 0

[PATCH v3 0/6] Add --dhcp-boot and --dhcp-opt options
by Anshu Kumari 15 Jun '26

15 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. Anshu Kumari (6): conf: Add --dhcp-opt command-line option conf: Add --dhcp-boot command-line option dhcp: Add option type table and value parser dhcp: Refactor fill_one() to operate on a generic buffer dhcp: Add option overload doc: Add --dhcp-boot and --dhcp-opt to man page conf.c | 29 ++++- dhcp.c | 322 ++++++++++++++++++++++++++++++++++++++++++++++++++++---- dhcp.h | 1 + passt.1 | 41 ++++++++ passt.h | 16 +++ 5 files changed, 389 insertions(+), 20 deletions(-) -- 2.54.0

3 22

[PATCH 0/4] Assorted minor socket creation cleanups
by David Gibson 15 Jun '26

15 Jun '26

As discussed on our recent call, I was looking again at bug 167. I discovered it's still fairly fiddly to address this, but while investigating spotted a number of cleanups to make in the vicinity. I think they make sense even without fixing bug 167 (yet), so here they are. David Gibson (4): flow: Correct misleading signature of flowside_sock_l4() Makefile: Remove unused DUAL_STACK_SOCKETS define fwd, pif: Remove duplicated logic between tcp_listen() and udp_listen() pif, util: Move listen(2) call from sock_l4_() to pif_listen() Makefile | 5 ----- flow.c | 11 +++++------ fwd.c | 8 +------- pif.c | 45 ++++++++++++++++++++++++++++++++++++--------- pif.h | 2 +- tcp.c | 38 -------------------------------------- tcp.h | 2 -- udp.c | 38 -------------------------------------- udp.h | 2 -- util.c | 7 ------- 10 files changed, 43 insertions(+), 115 deletions(-) -- 2.54.0

2 5