mirrors/arti.git/crates/tor-proto/src/circuit.rs, branch arti-v1.4.0

tor-proto: Turn some CtrlMsgs into CtrlCmds.

2025-02-06T13:33:05Z

All `CtrlMsg`s to which the reactor did not respond by sending a cell down the channel are now `CtrlCmd`s. Most of this is code motion, best reviewed with `--color-moved`.

tor-proto: Make shutdown a CtrlCmd.

2025-02-06T13:33:05Z

proto: Remove ConversationInHandler

2025-02-04T16:08:43Z

It is unused but most importantly it allows any RELAY cell to be sent from anywhere in the code which is really not desirable because it is skipping congestion control. It also allows us to remove the `control_tx` from the reactor which is one less channel to track/understand/think about. This opens up the door to all sorts of problems especially side channel that can be exploited if we are not careful. We can always bring this back if we need it but for now, it is unused and allows us to remove the `CtrlMsg::SendRelayCell` control message. No code behavior change. Signed-off-by: David Goulet

tor-proto: Ensure rx end of chan reactor output sink is not dropped in test.

2025-02-03T11:30:41Z

This hopefully fixes #1832. I haven't been able to repro the failure from #1832 locally, but judging from the logs, I suspect it's triggered by interleavings like: - `extend_fut` executes up until `circ.extend_ntor(..).await`, where it's blocked `await`ing a completion notification from the circuit reactor - `reply_fut` starts running: * `CircuitExtender` installs a meta-handler in the circuit reactor, and waits for an `EXTENDED2` cell * `test_extend()` reads a cell from the receiving end of the fake channel's output sink (`rx`) * `test_extend()` sends the `EXTENDED2` cell to the circuit reactor over `CircuitRxSender` * `rx` is dropped - `reply_fut` completes - in the channel reactor's `run_once()`, `select_biased!` resolves to an error, because the receiving end of the `output` sink (`rx`) was dropped. The reactor logs `got sink error: SendError { kind: Disconnected }` (see [`SendError`]), and exits - the circuit reactor exits too, because the receiving end of its `chan_sender` (which was in the channel reactor) was dropped Returning the `rx` of the fake channel's output sink from `reply_fut` *should* stop it from being dropped, and fix #1832. [`SendError`]: https://docs.rs/futures/latest/futures/channel/mpsc/struct.SendError.html#method.is_disconnected

tor-proto: Rewrite circuit reactor run_once() loop to use select!.

2025-01-29T14:34:58Z

This rewrites the circuit reactor main loop to use `select_biased!` to poll multiple futures simultaneously. The new `run_once()`, like the old, first waits for an initial `CtrlMsg::Create`. Then, it uses a `select_biased!` to poll the `chan_sender` sink and shutdown channel for readiness. When the channel sink is ready, we poll the `control` and `input` channels like before, as well as the new `ready_streams` `Stream` (`ready_streams` is a `futures::Stream` that replaces the previous `send_outbound()` function). Most of the implementation remains unchanged, except the `handle_input`, `handle_cell` and `handle_control` functions no longer send anything on the `chan_sender` channel. Instead, they may do some (synchronous) processing, and send instructions for the remaining work that needs to be done (for example, for writing the cell to the `chan_sender` channel). These instructions are handled at the end of `run_once()`, and are encoded in the `RunOnceCmdInner` enum. What this change does **not** do: * the control channel *still* bypasses congestion control. We could fix this by making the various reactor functions send the `RunOnceCmdInner` commands to `run_once()` via a channel (instead of returning them). This would enable the reactor to stop reading the commands (except for handle `Sendme`, which would be handled separately) if it's blocked on congestion control.

tor-proto: Use SendRelayCell within CtrlMsg::SendRelayCell.

2025-01-29T14:26:27Z

tor-proto: Derive Default for MutableState.

2025-01-28T14:07:53Z

This enables us to get rid of two lines from the circuit reactor constructor.

tor-proto: Replace sleep() in test with advance_until_stalled().

2025-01-16T18:03:31Z

This test spuriously failed on my !2720 branch (https://gitlab.torproject.org/gabi-250/arti/-/jobs/811495): ``` failures: ---- circuit::test::invalid_circ_sendme stdout ---- E tor_proto::channel::reactor: UniqId(71): Running reactor E tor_proto::circuit::reactor: Circ 23.17: Running circuit reactor E tor_proto::circuit::reactor: Circ 23.17: reactor received AddFakeHop { relay_cell_format: V0, fwd_lasthop: false, rev_lasthop: false, params: CircParameters { extend_by_ed25519_id: true, ccontrol: CongestionControlParams { alg: FixedWindow(FixedWindowParams { circ_window_start: 1000, circ_window_min: 100, circ_window_max: 1000 }), cwnd_params: CongestionWindowParams { cwnd_init: 124, cwnd_inc_pct_ss: Percentage { value: 100 }, cwnd_inc: 1, cwnd_inc_rate: 31, cwnd_min: 124, cwnd_max: 4294967295, sendme_inc: 31 }, rtt_params: RoundTripEstimatorParams { ewma_cwnd_pct: Percentage { value: 50 }, ewma_max: 10, ewma_ss_max: 2, rtt_reset_pct: Percentage { value: 100 } } } }, done: Sender { complete: false } } E tor_proto::circuit::reactor: Circ 23.17: reactor received AddFakeHop { relay_cell_format: V0, fwd_lasthop: true, rev_lasthop: true, params: CircParameters { extend_by_ed25519_id: true, ccontrol: CongestionControlParams { alg: FixedWindow(FixedWindowParams { circ_window_start: 1000, circ_window_min: 100, circ_window_max: 1000 }), cwnd_params: CongestionWindowParams { cwnd_init: 124, cwnd_inc_pct_ss: Percentage { value: 100 }, cwnd_inc: 1, cwnd_inc_rate: 31, cwnd_min: 124, cwnd_max: 4294967295, sendme_inc: 31 }, rtt_params: RoundTripEstimatorParams { ewma_cwnd_pct: Percentage { value: 50 }, ewma_max: 10, ewma_ss_max: 2, rtt_reset_pct: Percentage { value: 100 } } } }, done: Sender { complete: false } } E tor_proto::circuit::reactor: Circ 23.17: reactor received BeginStream { hop_num: HopNum(2), message: Begin(Begin { addr: [119, 119, 119, 46, 101, 120, 97, 109, 112, 108, 101, 46, 99, 111, 109], port: 443, flags: BeginFlags(IPV6_OKAY) }), sender: Sender { tx: Sender { closed: false }, mq: TypedParticipation(Participation(Noop)) }, rx: Receiver { inner: ReceiverInner { state: Mutex { data: Ok(ReceiverState { rx: StreamUnobtrusivePeeker { buffered: None, poll_waker: None, inner: Some(Receiver { closed: false }) }, mq: TypedParticipation(Participation(Noop)), collapse_callbacks: 0 }), poisoned: false, .. } } }, done: Sender { complete: false }, cmd_checker: DataCmdChecker { expecting_connected: true } } Using RNG seed ARTI_TEST_PRNG=62d085c0fb1213c4f41d1a0a5c3f92dd10223a27c42bfda3caedfe60e011e95d E tor_proto::circuit::reactor: Circ 23.17: handling cell: Relay(Relay { body: .. }) Using RNG seed ARTI_TEST_PRNG=fd2f2f045a7340f1189972b13645346943efa2c42f43afd2161420c692bb4ff0 E tor_proto::circuit::reactor: Circ 23.17: handling cell: Relay(Relay { body: .. }) E tor_proto::circuit::reactor: Circ 23.17: Circuit reactor stopped: Err(CircProto("Mismatched tag on circuit SENDME")) E tor_proto::channel::reactor: UniqId(71): reactor received CloseCircuit(CircId(128)) E tor_proto::channel::reactor: UniqId(71): Circuit 128 is gone; sending DESTROY thread 'circuit::test::invalid_circ_sendme' panicked at crates/tor-proto/src/circuit.rs:2262:21: reactor continued running after invalid sendme failures: circuit::test::invalid_circ_sendme test result: FAILED. 142 passed; 1 failed; 1 ignored; 0 measured; 0 filtered out; finished in 4.93s ``` This is probably because of the `sleep()`-based check. This branch changes the test to use `MockRuntime`, replacing the `sleep()` with `advance_until_stalled()`.

congestion: Remove FixedWindow start window default value

2025-01-16T14:16:01Z

In unit tests, we set a 1000 by default but else, we don't set a default so our implementation doesn't forget to set the right value from the consensus. Signed-off-by: David Goulet

test: Fix circuit unit tests

2025-01-16T14:16:01Z

Signed-off-by: David Goulet