Files
fips/src/node/handlers/dispatch.rs
T
Martti MalmiandJohnathan Corgan 0a5c367edc data-plane perf overhaul: off-task encrypt + decrypt, GSO, connected UDP
Moves both AEAD layers (ChaCha20-Poly1305, one round per layer per
packet) plus the sendmsg syscall off the rx_loop task onto a per-shard
worker pool, adds per-peer connect(2)-ed UDP with SO_REUSEPORT, and
uses Linux UDP GSO (sendmsg+UDP_SEGMENT — kernel splits one super-skb
into N on-the-wire datagrams in a single TX-stack walk) when packets
in a batch are uniform-size. Same kernel primitive WireGuard's
in-kernel module and BoringTun use to hit 2.5–3.2 Gbps single-stream.

Single TCP stream on a 5-node docker-bridge mesh, 5 x 15 s x P=1:

  A→D:  1379 → 2708 Mbps  (1.96x, RTT +0.12 ms)
  A→E:  1394 → 2663 Mbps  (1.91x, RTT +0.11 ms)
  E→A:  1406 → 2624 Mbps  (1.87x, RTT +0.19 ms)

Static-peer pairs only — every CoV under 3%, 0 outliers, 0% ICMP
loss. The ~+100 µs RTT is the worker queue handoff cost; AEAD +
sendmmsg now run on a separate core in exchange.

What lands:

- src/node/encrypt_worker.rs: std::thread + crossbeam_channel
  workers; hash-by-destination dispatch pins a TCP flow to one
  worker so wire ordering is preserved; per-worker sendmmsg(2)
  batching up to 32; Linux uses sendmsg(2)+UDP_SEGMENT when
  packets in a group are uniform-size.

- src/node/decrypt_worker.rs: receive-side mirror. Each shard owns
  its session's recv cipher + replay window in a thread-local
  HashMap (no shared RwLock/Mutex). Sessions are handed off at
  promote_connection and re-registered on K-bit flip / rekey
  cutover.

- src/node/handlers/session.rs try_send_session_data_pipelined:
  FSP+FMP both seal in-place in the worker on one wire-buffer
  alloc; no intermediate inner_plaintext / fsp_payload Vecs.

- src/transport/udp/connected_peer.rs + peer_drain.rs: per-peer
  connect(2)-ed UDP socket with SO_REUSEPORT (set on the listen
  socket too — without that, EADDRINUSE on activation and every
  packet falls back to the wildcard path); the worker sends with
  msg_name=NULL and the kernel uses its cached 5-tuple. Tick-
  driven activation in handlers/connected_udp.rs, idempotent.

- src/transport/udp/mod.rs: mem::replace the recvmmsg backing buffer
  instead of buf.to_vec() per packet — single pointer swap, no
  MTU-sized memcpy.

- src/protocol/link.rs SessionDatagramRef: zero-copy borrowed view
  used by handle_session_datagram for the bulk local-delivery
  path; handle_session_payload takes the borrowed payload
  directly (no payload[35..].to_vec()).

- src/transport/mod.rs TransportAddr::from_socket_addr: collapses
  the two-alloc from_string(addr.to_string()) pattern to one.

- src/node/handlers/rx_loop.rs: decrypt-fallback drain promoted
  ahead of packet_rx in the select! (TCP ACK starvation fix);
  interleaved fallback drain every 32 packets inside the rx burst
  loop.

- noise::Session: send_cipher_clone / recv_cipher_clone /
  recv_replay_snapshot_owned / take_send_counter / accept_replay
  so off-task workers can hold a cloned cipher + reserved counter
  while the dispatcher keeps replay/counter sequencing serial.
  CipherState::cipher_clone returns a refcount-bumped LessSafeKey.
  AsyncUdpSocket: AsRawFd so workers issue raw sendmmsg / sendmsg
  without going through the tokio reactor.

- Worker pool sizing: both default to num_cpus, overridable via
  FIPS_ENCRYPT_WORKERS=N / FIPS_DECRYPT_WORKERS=N. Per-peer
  connected UDP can be disabled via FIPS_CONNECTED_UDP=0.

- src/perf_profile.rs: optional per-stage timing reporter under
  FIPS_PERF=1 (or FIPS_PIPELINE_TRACE=1). Off by default; zero
  overhead when disabled.

- All cfg(unix)-gated. Windows continues on the existing tokio-
  based send/recv.

Decrypt worker session lifecycle:

- Node::unregister_decrypt_worker_session mirrors the existing
  register helper. Wired at the two natural sites that already
  iterate peers_by_index: the rekey drain-completion block in
  handlers/rekey.rs (drops the worker entry for the old our_index
  once the drain window has expired and the cache_key is
  unreachable to any in-flight OLD-K packet), and remove_active_peer
  in handlers/dispatch.rs (drops the worker entry for each of the
  four index slots: current, rekey, pending, previous). Only
  our_index is normally registered; unregister_session is fire-
  and-forget for missing entries, so calling unconditionally on
  all four slots is correct and bounds the cleanup without per-
  slot accounting. Without these callers the per-worker sessions
  HashMap and the Node's decrypt_registered_sessions set would
  grow monotonically per rekey on long-lived peers.

Testing:

- testing/static/scripts/bench-multirun.sh: multi-run iperf3 +
  ping bench. N reruns (default 5), median / min / max / CoV % /
  per-run outlier flag, avg ping RTT, ICMP loss %, TCP retransmit
  total. Plain client→dest labels + topology header. Pre-bench
  peer-convergence check (FIPS_BENCH_CONVERGE_SECS, default 15);
  per-path route verification via stats.bytes_sent deltas — fails
  fast if traffic exits via a non-static-peer link.

- testing/static/docker-compose.yml: passes FIPS_ENCRYPT_WORKERS /
  FIPS_DECRYPT_WORKERS / FIPS_PERF through to containers for A/B
  benchmarking without rebuilds.

- testing/static/scripts/iperf-test.sh: same plain client→dest
  labels + topology header (was multihop/direct/N hop, which
  conflated topology distance with on-wire path).

- .config/nextest.toml: synthetic UDP node tests serialized
  through a max-threads=1 test group. Localhost handshakes drop
  on shared CI runners under parallel load; one-at-a-time keeps
  assertions reliable.

- src/node/tests/spanning_tree.rs: repair_missing_edge_handshakes
  — retries up to 5 times for synthetic edges whose msg1 was
  dropped, with a drain after each edge retry instead of after
  each attempt's full burst.

- src/node/decrypt_worker.rs::tests: two unit tests asserting
  WorkerMsg::UnregisterSession removes the worker-thread session
  HashMap entry (handle_msg_unregister_session_removes_entry) and
  is a no-op for never-seen cache_keys
  (handle_msg_unregister_session_idempotent_on_unknown_key), which
  is the safety invariant the unconditional unregister calls at
  the four index slots in remove_active_peer rely on.

- src/node/encrypt_worker.rs::unix_tests
  pipelined_send_wire_layout_roundtrips_canonical_decoders: mirrors
  the encoder geometry of try_send_session_data_pipelined (no
  coords, the common established-session path), runs the worker's
  real seal + send via flush_direct_batch_sync, and decodes the
  resulting wire packet using only canonical receive-side decoders
  (EncryptedHeader::parse, SessionDatagramRef::decode, FSP header
  parse, noise::open). Any divergence between the hand-rolled
  encoder offsets (fsp_aad_offset, fsp_plaintext_offset) and the
  decoders fails at one of the parse / open / decode steps before
  the inner-plaintext assertion fires. Complements the existing
  fsp_preseal_runs_before_outer_fmp_seal test which covers the
  seal-ordering invariant with synthetic headers but does not
  exercise the wire-layout invariant.

CHANGELOG.md [Unreleased] # Changed entry added describing the
worker-pool threading model, hash-by-destination dispatch,
sendmmsg/UDP_GSO, per-peer connected UDP, the operator-facing env
vars, and the bench numbers above.

Cherry-picks from mmalmi/master (paths translated from
crates/fips-core/src/ to src/): 9b7c723, 0deb5cb, 13f7339, e036c0e,
3740a68, 3792f83, 8510193, 4910b07, e53f545, e4e2896, 5fe4af5,
1d01ada, 8c37008, e12469e, 6eb2860.

Co-authored-by: Johnathan Corgan <johnathan@corganlabs.com>
2026-05-19 20:53:31 +00:00

217 lines
8.2 KiB
Rust

//! Link message dispatch and peer removal.
use crate::NodeAddr;
use crate::node::Node;
use tracing::{debug, info, trace};
impl Node {
/// Dispatch a decrypted link message to the appropriate handler.
///
/// Link messages are protocol messages exchanged between authenticated peers.
pub(in crate::node) async fn dispatch_link_message(
&mut self,
from: &NodeAddr,
plaintext: &[u8],
ce_flag: bool,
) {
if plaintext.is_empty() {
return;
}
let msg_type = plaintext[0];
let payload = &plaintext[1..];
match msg_type {
0x00 => {
// SessionDatagram
self.handle_session_datagram(from, payload, ce_flag).await;
}
0x01 => {
// SenderReport
self.handle_sender_report(from, payload);
}
0x02 => {
// ReceiverReport
self.handle_receiver_report(from, payload).await;
}
0x10 => {
// TreeAnnounce
self.handle_tree_announce(from, payload).await;
}
0x20 => {
// FilterAnnounce
self.handle_filter_announce(from, payload).await;
}
0x30 => {
// LookupRequest
self.handle_lookup_request(from, payload).await;
}
0x31 => {
// LookupResponse
self.handle_lookup_response(from, payload).await;
}
0x50 => {
// Disconnect
self.handle_disconnect(from, payload);
}
0x51 => {
// Heartbeat — no-op, last_recv_time already updated by record_recv()
trace!(peer = %self.peer_display_name(from), "Received heartbeat");
}
_ => {
debug!(msg_type = msg_type, "Unknown link message type");
}
}
}
/// Handle a Disconnect notification from a peer.
///
/// The peer is signaling an orderly departure. We immediately remove
/// them from all state rather than waiting for timeout detection, and
/// schedule a reconnect if the peer is configured as auto-connect.
/// Without this, a graceful upstream shutdown orphans auto-connect
/// entries — other removal paths (link-dead, decrypt failure, peer
/// restart) all schedule reconnect.
pub(in crate::node) fn handle_disconnect(&mut self, from: &NodeAddr, payload: &[u8]) {
let disconnect = match crate::protocol::Disconnect::decode(payload) {
Ok(msg) => msg,
Err(e) => {
debug!(from = %self.peer_display_name(from), error = %e, "Malformed disconnect message");
return;
}
};
info!(
peer = %self.peer_display_name(from),
reason = %disconnect.reason,
"Peer sent disconnect notification"
);
let addr = *from;
self.remove_active_peer(from);
let now_ms = std::time::SystemTime::now()
.duration_since(std::time::UNIX_EPOCH)
.map(|d| d.as_millis() as u64)
.unwrap_or(0);
self.schedule_reconnect(addr, now_ms);
}
/// Remove an active peer and clean up all associated state.
///
/// Frees session index, removes link and address mappings. Used for
/// both graceful disconnect and timeout-based eviction.
///
/// Also handles tree state cleanup: if the removed peer was our parent,
/// selects an alternative or becomes root, and marks remaining peers
/// for pending tree announce (delivered on next tick).
pub(in crate::node) fn remove_active_peer(&mut self, node_addr: &NodeAddr) {
let peer = match self.peers.remove(node_addr) {
Some(p) => p,
None => {
debug!(peer = %self.peer_display_name(node_addr), "Peer already removed");
return;
}
};
// Log suppressed replay detection summary before teardown
let suppressed = peer.replay_suppressed_count();
if suppressed > 0 {
debug!(
peer = %self.peer_display_name(node_addr),
count = suppressed,
"Suppressed replay detections during link transition"
);
}
// MMP teardown log (before we drop the peer)
let peer_name = self
.peer_aliases
.get(node_addr)
.cloned()
.unwrap_or_else(|| peer.identity().short_npub());
if let Some(mmp) = peer.mmp() {
Self::log_mmp_teardown(&peer_name, mmp);
}
// Remove any end-to-end session associated with this peer.
//
// Sessions are tracked separately from peers (self.sessions vs self.peers).
// Leaving a stale session alive after removing the peer causes:
// 1. check_session_mmp_reports() keeps logging stale "MMP session metrics"
// with frozen counters until purge_idle_sessions() eventually fires.
// 2. initiate_session() finds is_established() == true on the stale entry
// and silently returns Ok(()), preventing a new session from being
// established even after the link layer reconnects successfully.
if let Some(session_entry) = self.sessions.remove(node_addr)
&& let Some(mmp) = session_entry.mmp()
{
Self::log_session_mmp_teardown(&peer_name, mmp);
}
self.pending_tun_packets.remove(node_addr);
let link_id = peer.link_id();
let transport_id = peer.transport_id();
// Free session indices (current, rekey, pending, previous)
if let Some(tid) = transport_id {
if let Some(idx) = peer.our_index() {
let cache_key = (tid, idx.as_u32());
self.peers_by_index.remove(&cache_key);
#[cfg(unix)]
self.unregister_decrypt_worker_session(cache_key);
let _ = self.index_allocator.free(idx);
}
if let Some(idx) = peer.rekey_our_index() {
let cache_key = (tid, idx.as_u32());
self.pending_outbound.remove(&cache_key);
self.peers_by_index.remove(&cache_key);
#[cfg(unix)]
self.unregister_decrypt_worker_session(cache_key);
let _ = self.index_allocator.free(idx);
}
if let Some(idx) = peer.pending_our_index() {
let cache_key = (tid, idx.as_u32());
self.peers_by_index.remove(&cache_key);
#[cfg(unix)]
self.unregister_decrypt_worker_session(cache_key);
let _ = self.index_allocator.free(idx);
}
if let Some(idx) = peer.previous_our_index() {
let cache_key = (tid, idx.as_u32());
self.peers_by_index.remove(&cache_key);
#[cfg(unix)]
self.unregister_decrypt_worker_session(cache_key);
let _ = self.index_allocator.free(idx);
}
}
// Remove link and address mapping
self.remove_link(&link_id);
if let Some(transport_id) = transport_id {
self.cleanup_bootstrap_transport_if_unused(transport_id);
}
// Tree state cleanup
let tree_changed = self.handle_peer_removal_tree_cleanup(node_addr);
if tree_changed {
// Mark all remaining peers for pending tree announce.
// These will be sent on the next tick via check_tree_state().
for peer in self.peers.values_mut() {
peer.mark_tree_announce_pending();
}
}
// Bloom filter cleanup: clear state for removed peer, mark all remaining peers
self.bloom_state.remove_peer_state(node_addr);
let remaining_peers: Vec<NodeAddr> = self.peers.keys().copied().collect();
self.bloom_state.mark_all_updates_needed(remaining_peers);
info!(
peer = %self.peer_display_name(node_addr),
link_id = %link_id,
tree_changed = tree_changed,
"Peer removed and state cleaned up"
);
}
}