netring 0.29.0

High-performance zero-copy packet I/O for Linux (AF_PACKET TPACKET_V3 + AF_XDP)
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257

//! End-to-end Monitor dispatch path under fully synthetic events.
//!
//! Doesn't open AF_PACKET, doesn't need CAP_NET_RAW — drives the
//! [`Dispatcher`] directly with crafted `FlowStarted<P>` / `Http`
//! payloads. Exercises the sync + async + layered-sink chain
//! together.

#![cfg(all(feature = "tokio", feature = "flow"))]

use std::net::{IpAddr, Ipv4Addr, SocketAddr};
use std::sync::Arc;
use std::sync::atomic::{AtomicU32, Ordering};

use flowscope::Timestamp;
use netring::anomaly::sink::NoopSink;
use netring::ctx::{CounterRegistry, Ctx, SourceIdx, StateMap};
use netring::monitor::{Dispatcher, HandlerRegistry};
use netring::protocol::builtin::Tcp;
use netring::protocol::event_typed::FlowStarted;

fn dummy_flow_started(port_a: u16, port_b: u16) -> FlowStarted<Tcp> {
    let key = flowscope::extract::FiveTupleKey::new(
        flowscope::L4Proto::Tcp,
        SocketAddr::new(IpAddr::V4(Ipv4Addr::new(10, 0, 0, 1)), port_a),
        SocketAddr::new(IpAddr::V4(Ipv4Addr::new(10, 0, 0, 2)), port_b),
    );
    FlowStarted::<Tcp>::new(key, Some(flowscope::L4Proto::Tcp), Timestamp::new(0, 0))
}

fn fresh_ctx<'a>(
    state: &'a mut StateMap,
    sink: &'a mut NoopSink,
    counters: &'a mut CounterRegistry,
    flow_states: &'a mut netring::ctx::FlowStateRegistry,
) -> Ctx<'a> {
    Ctx::new(
        None,
        Timestamp::new(0, 0),
        SourceIdx(0),
        state,
        sink,
        counters,
        flow_states,
    )
}

#[test]
fn sync_dispatch_fires_handler_once_per_event() {
    let count = Arc::new(AtomicU32::new(0));
    let c = Arc::clone(&count);

    let mut reg = HandlerRegistry::default();
    reg.register::<FlowStarted<Tcp>, _, _>(move |_evt: &FlowStarted<Tcp>| {
        c.fetch_add(1, Ordering::Relaxed);
        Ok(())
    });
    let mut disp: Dispatcher = reg.into_dispatcher().unwrap();

    let mut state = StateMap::default();
    let mut sink = NoopSink;
    let mut counters = CounterRegistry::default();
    let mut flow_states = netring::ctx::FlowStateRegistry::default();
    let mut ctx = fresh_ctx(&mut state, &mut sink, &mut counters, &mut flow_states);

    for _ in 0..7 {
        disp.dispatch::<FlowStarted<Tcp>>(&dummy_flow_started(12345, 80), &mut ctx)
            .unwrap();
    }
    assert_eq!(count.load(Ordering::Relaxed), 7);
}

#[test]
fn sync_two_handlers_same_event_fire_in_registration_order() {
    let order = Arc::new(std::sync::Mutex::new(Vec::new()));
    let o1 = Arc::clone(&order);
    let o2 = Arc::clone(&order);

    let mut reg = HandlerRegistry::default();
    reg.register::<FlowStarted<Tcp>, _, _>(move |_evt: &FlowStarted<Tcp>| {
        o1.lock().unwrap().push("a");
        Ok(())
    });
    reg.register::<FlowStarted<Tcp>, _, _>(move |_evt: &FlowStarted<Tcp>| {
        o2.lock().unwrap().push("b");
        Ok(())
    });
    let mut disp = reg.into_dispatcher().unwrap();

    let mut state = StateMap::default();
    let mut sink = NoopSink;
    let mut counters = CounterRegistry::default();
    let mut flow_states = netring::ctx::FlowStateRegistry::default();
    let mut ctx = fresh_ctx(&mut state, &mut sink, &mut counters, &mut flow_states);

    disp.dispatch::<FlowStarted<Tcp>>(&dummy_flow_started(12345, 80), &mut ctx)
        .unwrap();
    assert_eq!(*order.lock().unwrap(), vec!["a", "b"]);
}

#[test]
fn sync_handler_error_short_circuits_remaining() {
    let fired_after = Arc::new(AtomicU32::new(0));
    let f = Arc::clone(&fired_after);

    let mut reg = HandlerRegistry::default();
    reg.register::<FlowStarted<Tcp>, _, _>(|_evt: &FlowStarted<Tcp>| {
        Err(netring::Error::Config("synthetic".into()))
    });
    reg.register::<FlowStarted<Tcp>, _, _>(move |_evt: &FlowStarted<Tcp>| {
        f.fetch_add(1, Ordering::Relaxed);
        Ok(())
    });
    let mut disp = reg.into_dispatcher().unwrap();

    let mut state = StateMap::default();
    let mut sink = NoopSink;
    let mut counters = CounterRegistry::default();
    let mut flow_states = netring::ctx::FlowStateRegistry::default();
    let mut ctx = fresh_ctx(&mut state, &mut sink, &mut counters, &mut flow_states);

    let res = disp.dispatch::<FlowStarted<Tcp>>(&dummy_flow_started(12345, 80), &mut ctx);
    assert!(res.is_err());
    assert_eq!(
        fired_after.load(Ordering::Relaxed),
        0,
        "second handler must not fire after first errored"
    );
}

#[test]
fn unknown_event_type_is_a_silent_noop() {
    let mut reg = HandlerRegistry::default();
    reg.register::<FlowStarted<Tcp>, _, _>(|_evt: &FlowStarted<Tcp>| Ok(()));
    let mut disp = reg.into_dispatcher().unwrap();

    let mut state = StateMap::default();
    let mut sink = NoopSink;
    let mut counters = CounterRegistry::default();
    let mut flow_states = netring::ctx::FlowStateRegistry::default();
    let mut ctx = fresh_ctx(&mut state, &mut sink, &mut counters, &mut flow_states);

    // Dispatch a payload type that has no handlers — quiet success.
    let unrelated: u64 = 42;
    assert!(disp.dispatch::<u64>(&unrelated, &mut ctx).is_ok());
}

#[tokio::test(flavor = "current_thread")]
async fn async_dispatch_runs_arc_capturing_handler() {
    let pool_count = Arc::new(AtomicU32::new(0));
    let p = Arc::clone(&pool_count);

    let mut reg = HandlerRegistry::default();
    reg.register_async::<FlowStarted<Tcp>, _>(move |_evt: &FlowStarted<Tcp>| {
        let p = Arc::clone(&p);
        async move {
            tokio::task::yield_now().await;
            p.fetch_add(1, Ordering::Relaxed);
            Ok(())
        }
    });
    let mut disp = reg.into_dispatcher().unwrap();

    for _ in 0..3 {
        disp.dispatch_async::<FlowStarted<Tcp>>(&dummy_flow_started(12345, 80))
            .await
            .unwrap();
    }
    assert_eq!(pool_count.load(Ordering::Relaxed), 3);
}

#[tokio::test(flavor = "current_thread")]
async fn sync_and_async_handlers_for_same_event_both_fire() {
    let sync_count = Arc::new(AtomicU32::new(0));
    let async_count = Arc::new(AtomicU32::new(0));
    let s = Arc::clone(&sync_count);
    let a = Arc::clone(&async_count);

    let mut reg = HandlerRegistry::default();
    reg.register::<FlowStarted<Tcp>, _, _>(move |_evt: &FlowStarted<Tcp>| {
        s.fetch_add(1, Ordering::Relaxed);
        Ok(())
    });
    reg.register_async::<FlowStarted<Tcp>, _>(move |_evt: &FlowStarted<Tcp>| {
        let a = Arc::clone(&a);
        async move {
            a.fetch_add(1, Ordering::Relaxed);
            Ok(())
        }
    });
    let mut disp = reg.into_dispatcher().unwrap();

    let mut state = StateMap::default();
    let mut sink = NoopSink;
    let mut counters = CounterRegistry::default();
    let mut flow_states = netring::ctx::FlowStateRegistry::default();
    let mut ctx = fresh_ctx(&mut state, &mut sink, &mut counters, &mut flow_states);

    let evt = dummy_flow_started(12345, 80);
    for _ in 0..5 {
        disp.dispatch::<FlowStarted<Tcp>>(&evt, &mut ctx).unwrap();
        disp.dispatch_async::<FlowStarted<Tcp>>(&evt).await.unwrap();
    }
    assert_eq!(sync_count.load(Ordering::Relaxed), 5);
    assert_eq!(async_count.load(Ordering::Relaxed), 5);
}

#[test]
fn dispatcher_routes_tick_payload_to_on_tick_handler() {
    // Phase F.2: users may register tick handlers via either
    // `.tick(period, handler)` (boxed registration; run loop
    // drives) OR `.on::<Tick>(handler)` (dispatcher slot).
    // The latter is what gets exercised here — verifies the run
    // loop's `dispatcher.dispatch::<Tick>` call hits the right
    // slot for a synthetic Tick payload.
    use netring::protocol::event_typed::Tick;
    let count = Arc::new(AtomicU32::new(0));
    let c = Arc::clone(&count);
    let mut reg = HandlerRegistry::default();
    reg.register::<Tick, _, _>(move |_t: &Tick| {
        c.fetch_add(1, Ordering::Relaxed);
        Ok(())
    });
    let mut disp = reg.into_dispatcher().unwrap();

    let mut state = StateMap::default();
    let mut sink = NoopSink;
    let mut counters = CounterRegistry::default();
    let mut flow_states = netring::ctx::FlowStateRegistry::default();
    let mut ctx = fresh_ctx(&mut state, &mut sink, &mut counters, &mut flow_states);

    // `Tick` is `#[non_exhaustive]` so external code can't use
    // field-init syntax; the `#[doc(hidden)] pub fn new` lets
    // integration tests synthesise one (same pattern as
    // `FlowStarted::<P>::new` used by the dispatcher tests
    // above).
    let tick = Tick::new(
        Timestamp::new(123, 456),
        std::time::Duration::from_millis(100),
    );
    for _ in 0..4 {
        disp.dispatch::<Tick>(&tick, &mut ctx).unwrap();
    }
    assert_eq!(count.load(Ordering::Relaxed), 4);
}

#[test]
fn dispatcher_carries_correct_type_and_handler_counts() {
    let mut reg = HandlerRegistry::default();
    reg.register::<FlowStarted<Tcp>, _, _>(|_evt: &FlowStarted<Tcp>| Ok(()));
    reg.register::<FlowStarted<Tcp>, _, _>(|_evt: &FlowStarted<Tcp>| Ok(()));
    reg.register_async::<FlowStarted<Tcp>, _>(|_evt: &FlowStarted<Tcp>| async { Ok(()) });
    let disp = reg.into_dispatcher().unwrap();

    assert_eq!(disp.type_count(), 1);
    assert_eq!(disp.handler_count(), 2);
    assert_eq!(disp.async_handler_count(), 1);
}