irtt_server::ServerCore::handle_datagram is a public API that
always passes kernel_rx_timestamp = None; only the crate-private
entry point the Tokio runtime (the irtt-server applet) uses
supplies one. Say so explicitly instead of implying every use of
this crate's server gets the preference.
irtt-client can talk to any IRTT-compatible peer, not only this
repo's own irtt-server, and has no way to know what clock source
produced a peer-reported wall-clock endpoint. Restrict the
kernel-receive-timestamp-preference claim to when the peer is
this crate's own Linux irtt-server; state plainly that another
peer's reported value is used as-is otherwise.
adjusted_rtt_us/effective_rtt_us are derived in part from
server-reported processing time, which is computed from the same
server receive endpoint the client_to_server one-way-delay field
documents as kernel-preferred. Only raw RTT is guaranteed to stay
on userspace monotonic timing; say so precisely instead of
claiming RTT in general is unaffected.
On Linux, one-way delay measurement best-effort prefers kernel
send/receive timestamps over userspace timing, falling back
whenever a kernel reading is unavailable or implausible. RTT is
unaffected and remains userspace-timed.
Add a user-facing README note near the measurement/output
explanation, and bring OneWayDelaySample::client_to_server's
rustdoc in line with the send-endpoint behavior, matching the
style already used for server_to_client. Also fix a stale
socket_io module doc that still claimed the captured kernel
receive timestamp went unconsumed; ServerCore has used it via
preferred_receive_wall_ns since d3f93a9.
Moves the client userspace probe sent_at measurement to immediately
after a successful UDP send, improving raw RTT and the userspace
upstream-OWD fallback by excluding the socket send-call interval and
the small amount of local work preceding it.
All fallible probe-commit preflight (timeout deadline checked_add,
capacity/collision checks) still runs before the send, from a private
pre-send anchor (SessionMachine::finalize_probe_commit). The public
sent_at is now supplied afterward to an infallible
SessionMachine::commit_probe_sent, so DurationOverflow still guarantees
nothing was transmitted and a WouldBlock try_send still commits
nothing.
The operational probe timeout deadline stays anchored to the pre-send
timestamp on purpose, not to the new post-send sent_at, preserving
existing timeout/loss timing. Linux kernel TX_SOFTWARE plausibility
keeps its own pre-send wall-clock lower bound
(PendingProbe::tx_not_before_wall) instead of the post-send sent_at,
since a legitimate TX_SOFTWARE timestamp can be generated while the
send syscall is still in flight and would otherwise be wrongly
rejected.
Also flags (without touching production code) that irtt-stats'
ReplySample::from_reply_parts feeds the public sent_at directly into
IpdvSample.client_send_mono/client_send_wall_ns, so the upstream/send
IPDV component naturally shifts along with the sent_at semantics
change; per scope this is reported rather than compensated for.
An audit of the server-reported receive window found no defect in the
statistics pipeline: the bitmap is decoded raw, stored raw, never
interpreted, and never feeds the directional loss estimates, which use
the cumulative server receive count.
The public rustdoc was silent on all of that, leaving consumers of
PacketCounts free to read a window of 0x1 as a 63-packet loss burst.
Document the field's contract, name the count as the only server input
to LossStats, and cover the window-only case with a regression test.
Treat a decoded, authenticated, token-matched close flag as a session-level signal after classifying the packet. Classification events now deterministically precede SessionClosed for pending, duplicate, retained-timeout, evicted-late, and untracked sequences.
Wrong-token and bad-HMAC close packets remain non-terminal. Public event documentation records the ordering contract, and focused regressions cover every correlation class.
Allow one-way delay samples to represent negative values caused by wall-clock skew between client and server. Update CLI output and statistics collection to correctly handle and report these signed durations.
Hide internal fields to improve type safety and provide helper methods for unit conversions and negative value checks. Fix several feature-gated test compilation issues.
Simplify RttSample by merging signed and unsigned fields into a single SignedDuration for adjusted and effective RTT. This ensures accurate reporting and statistics when server processing time exceeds raw RTT.
It was an attempt at handling integer wraparound more cleanly keeping in mind that we have a continuous mode,
and so we can more easily reach seq > u32::MAX
However after running the numbers, if we run with a interval of 10ms, it's still gonna take ~497 days before we hit integer overflow, so for now it doesn't seem worth to have this extra logic to handle something that'll in 99.9% of cases be purely an edge case
Keeping in mind autorate usage, in sqm-autorate-rust each send loop is 500 ms by default, and the sleep is then 500 ms / 5 reflectors, so our interval is 100 ms. that means it'll take 4970 days or ~13.6 years to reach the limit. Not worth dealing with IMO
Apply the #[non-exhaustive] attribute to ClientError and WarningKind to prevent breaking changes when adding new variants. Handle the catch-all case in the CLI formatter.
Nils Andreas Svee