ringfire 0.5.2

Zero-copy lock-free inter-process communication (IPC) ring buffer and shared memory bus in Rust, with ring mirroring across hosts
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
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356

use ringfire::{CleanupMode, RingConsumer, RingProducer, RingProducerBuilder};
use std::thread;

#[derive(Debug, Clone, Copy, PartialEq, Eq)]
#[repr(C)]
struct TradeMsg {
    id: u64,
    price: u64,
    qty: u32,
    side: u8,
}

#[test]
fn test_spmc_broadcast_to_multiple_consumers() {
    let tmp_path = std::env::temp_dir().join("test_spmc_broadcast.shm");
    let _ = std::fs::remove_file(&tmp_path);

    let capacity = 4096;
    let mut producer = RingProducer::<TradeMsg>::create(&tmp_path, capacity).unwrap();

    let num_consumers = 4;
    let msg_count = 2000;

    let consumers: Vec<_> = (0..num_consumers)
        .map(|_| RingConsumer::<TradeMsg>::attach(&tmp_path).unwrap())
        .collect();

    // Spawn consumer threads
    let mut handles = Vec::new();
    for (cid, mut consumer) in consumers.into_iter().enumerate() {
        let handle = thread::spawn(move || {
            let mut received = 0;
            let mut last_id = 0;
            while received < msg_count {
                if let Some(msg) = consumer.try_recv() {
                    assert_eq!(msg.id, last_id + 1, "Consumer {} sequence discontinuity", cid);
                    assert_eq!(msg.price, (last_id + 1) * 100);
                    last_id = msg.id;
                    received += 1;
                } else {
                    core::hint::spin_loop();
                }
            }
            received
        });
        handles.push(handle);
    }

    // Publish messages
    for i in 1..=msg_count {
        let msg = TradeMsg {
            id: i,
            price: i * 100,
            qty: 10,
            side: if i % 2 == 0 { b'B' } else { b'S' },
        };
        producer.push(&msg);
    }

    for handle in handles {
        let count = handle.join().unwrap();
        assert_eq!(count, msg_count);
    }
}

#[test]
fn test_exclusive_producer_file_locking() {
    let tmp_path = std::env::temp_dir().join("test_spmc_flock.shm");
    let _ = std::fs::remove_file(&tmp_path);

    let p1 = RingProducer::<u64>::create(&tmp_path, 1024).unwrap();

    // Attempting to create another producer on the same path must fail with ProducerAlreadyExists
    let p2_res = RingProducer::<u64>::create(&tmp_path, 1024);
    assert!(p2_res.is_err());

    drop(p1);

    // After p1 drops and releases flock, we can create p3
    let p3 = RingProducer::<u64>::create(&tmp_path, 1024);
    assert!(p3.is_ok());
}

#[test]
fn test_cleanup_mode_unlink_vs_persistent() {
    let tmp_unlink = std::env::temp_dir().join("test_cleanup_unlink.shm");
    let tmp_persist = std::env::temp_dir().join("test_cleanup_persist.shm");
    let _ = std::fs::remove_file(&tmp_unlink);
    let _ = std::fs::remove_file(&tmp_persist);

    {
        let _p = RingProducerBuilder::new(1024)
            .cleanup_mode(CleanupMode::UnlinkOnDrop)
            .build::<u64, _>(&tmp_unlink)
            .unwrap();
        assert!(tmp_unlink.exists());
    }
    // Should be unlinked after drop
    assert!(!tmp_unlink.exists());

    {
        let _p = RingProducerBuilder::new(1024)
            .cleanup_mode(CleanupMode::Persistent)
            .build::<u64, _>(&tmp_persist)
            .unwrap();
        assert!(tmp_persist.exists());
    }
    // Should persist after drop
    assert!(tmp_persist.exists());
    let _ = std::fs::remove_file(&tmp_persist);
}

#[test]
fn test_consumer_start_mode_latest_and_head() {
    let tmp_path = std::env::temp_dir().join(format!("test_spmc_modes_{}.shm", std::process::id()));
    let _ = std::fs::remove_file(&tmp_path);

    let mut producer = RingProducer::<u64>::create(&tmp_path, 1024).unwrap();
    // Publish 10 items
    for i in 1..=10 {
        producer.push(&i);
    }

    // 1. Latest mode: jumps directly to the latest published item (10), then reads subsequent items
    let mut cons_latest = RingConsumer::<u64>::attach_latest(&tmp_path).unwrap();
    assert_eq!(cons_latest.try_recv(), Some(10));
    assert_eq!(cons_latest.try_recv(), None);

    // 2. Head mode: strictly waits for future items (arriving after attach)
    let mut cons_head = RingConsumer::<u64>::builder()
        .start_from_head()
        .attach(&tmp_path)
        .unwrap();
    assert_eq!(cons_head.try_recv(), None);

    // Push new items 11 and 12
    producer.push(&11);
    producer.push(&12);

    assert_eq!(cons_latest.try_recv(), Some(11));
    assert_eq!(cons_latest.try_recv(), Some(12));

    assert_eq!(cons_head.try_recv(), Some(11));
    assert_eq!(cons_head.try_recv(), Some(12));

    let _ = std::fs::remove_file(&tmp_path);
}

#[test]
fn test_consumer_start_mode_oldest() {
    let tmp_path = std::env::temp_dir().join(format!("test_spmc_oldest_{}.shm", std::process::id()));
    let _ = std::fs::remove_file(&tmp_path);

    let mut producer = RingProducer::<u64>::create(&tmp_path, 1024).unwrap();
    for i in 1..=10 {
        producer.push(&i);
    }

    // Oldest mode (default for attach) replays from item 1
    let mut cons = RingConsumer::<u64>::attach(&tmp_path).unwrap();
    for i in 1..=10 {
        assert_eq!(cons.try_recv(), Some(i));
    }
    assert_eq!(cons.try_recv(), None);

    let _ = std::fs::remove_file(&tmp_path);
}

#[test]
fn test_shm_offset_checkpoint_crash_and_resume() {
    let dir = std::env::temp_dir();
    let ring_path = dir.join(format!("test_spmc_resume_ring_{}.shm", std::process::id()));
    let offset_path = dir.join(format!("test_spmc_resume_offset_{}.offset", std::process::id()));
    let _ = std::fs::remove_file(&ring_path);
    let _ = std::fs::remove_file(&offset_path);

    let mut producer = RingProducer::<u64>::create(&ring_path, 1024).unwrap();
    for i in 1..=50 {
        producer.push(&i);
    }

    // Consumer 1: processes items 1..=25 and commits offset in shared memory
    {
        let mut cons1 = RingConsumer::<u64>::builder()
            .offset_shm(&offset_path)
            .consumer_name("worker_a")
            .start_from_oldest()
            .attach(&ring_path)
            .unwrap();

        for i in 1..=25 {
            assert_eq!(cons1.try_recv(), Some(i));
        }
        assert_eq!(cons1.last_processed_sequence(), 25);
        cons1.commit_offset().unwrap();
        // cons1 dropped (simulating process exit/crash)
    }

    // Consumer 2: starts with same offset file. Must resume strictly from 26!
    {
        let mut cons2 = RingConsumer::<u64>::builder()
            .offset_shm(&offset_path)
            .consumer_name("worker_a")
            .attach(&ring_path)
            .unwrap();

        for i in 26..=50 {
            assert_eq!(cons2.try_recv(), Some(i), "Discontinuity during resume at {}", i);
        }
        assert_eq!(cons2.try_recv(), None);
        assert_eq!(cons2.last_processed_sequence(), 50);
        cons2.commit_offset().unwrap();
    }

    let _ = std::fs::remove_file(&ring_path);
    let _ = std::fs::remove_file(&offset_path);
}

#[test]
fn test_shm_offset_lapping_recovery() {
    let dir = std::env::temp_dir();
    let ring_path = dir.join(format!("test_spmc_lap_rec_ring_{}.shm", std::process::id()));
    let offset_path = dir.join(format!("test_spmc_lap_rec_offset_{}.offset", std::process::id()));
    let _ = std::fs::remove_file(&ring_path);
    let _ = std::fs::remove_file(&offset_path);

    let capacity = 32;
    let mut producer = RingProducer::<u64>::create(&ring_path, capacity).unwrap();
    for i in 1..=10 {
        producer.push(&i);
    }

    // Commit offset at sequence 5
    {
        let mut cons1 = RingConsumer::<u64>::builder()
            .offset_shm(&offset_path)
            .consumer_name("slow_worker")
            .start_from_oldest()
            .attach(&ring_path)
            .unwrap();

        for i in 1..=5 {
            assert_eq!(cons1.try_recv(), Some(i));
        }
        cons1.commit_offset().unwrap();
    }

    // Producer laps consumer by writing up to 100 in capacity 32 buffer
    for i in 11..=100 {
        producer.push(&i);
    }

    // Surviving range in buffer is: 100 - 32 + 1 = 69..=100.
    // Consumer committed at 5, wants 6. But 6 was overwritten!
    // Consumer must detect lapping and catch up to oldest available (69)
    {
        let mut cons2 = RingConsumer::<u64>::builder()
            .offset_shm(&offset_path)
            .consumer_name("slow_worker")
            .attach(&ring_path)
            .unwrap();

        assert_eq!(cons2.lapped_count(), 69 - 6); // 63 messages skipped
        assert_eq!(cons2.try_recv(), Some(69));
        for i in 70..=100 {
            assert_eq!(cons2.try_recv(), Some(i));
        }
        assert_eq!(cons2.try_recv(), None);
    }

    let _ = std::fs::remove_file(&ring_path);
    let _ = std::fs::remove_file(&offset_path);
}

#[test]
fn test_flow_control_lossless_backpressure() {
    use ringfire::{FlowControl, RingfireError};

    let dir = std::env::temp_dir();
    let ring_path = dir.join(format!("test_spmc_backpressure_{}.shm", std::process::id()));
    let _ = std::fs::remove_file(&ring_path);

    let capacity = 16;
    let mut producer = RingProducer::<u64>::create_with_options(
        &ring_path,
        RingProducerBuilder::new(capacity)
            .flow_control(FlowControl::LosslessBackpressure)
            .max_readers(8),
    )
    .unwrap();

    assert_eq!(producer.flow_control(), FlowControl::LosslessBackpressure);
    assert_eq!(producer.headroom(), capacity);

    // Attach consumer
    let mut consumer = RingConsumer::<u64>::builder()
        .consumer_name("careful_reader")
        .start_from_oldest()
        .attach(&ring_path)
        .unwrap();

    assert!(consumer.registration().is_some());
    assert_eq!(producer.registry().unwrap().active_readers(0).len(), 1);

    // Push 16 items to fill the buffer exactly
    for i in 1..=16 {
        producer.push(&i);
    }

    assert_eq!(producer.headroom(), 0);
    assert_eq!(producer.reader_lag(), 16);

    // 17th item should fail with BackpressureBufferFull via try_push
    let res = producer.try_push(&17);
    match res {
        Err(RingfireError::BackpressureBufferFull) => {}
        other => panic!("Expected BackpressureBufferFull, got {:?}", other),
    }

    // Spawn thread to drain first 8 messages from consumer
    let drain_handle = thread::spawn(move || {
        thread::sleep(std::time::Duration::from_millis(20));
        let mut drained = Vec::new();
        for _ in 0..8 {
            if let Some(item) = consumer.try_recv() {
                drained.push(item);
            }
        }
        (consumer, drained)
    });

    // Blocking push(17) unblocks as soon as the reader drains its first item
    producer.push(&17);

    let (mut consumer, drained) = drain_handle.join().unwrap();
    // 8 drained, 17 published: 9 unread
    assert_eq!(producer.headroom(), 7);
    assert_eq!(drained.len(), 8);
    assert_eq!(drained, vec![1, 2, 3, 4, 5, 6, 7, 8]);

    // Drain remainder from consumer: 9..=17
    let mut rest = Vec::new();
    for _ in 9..=17 {
        if let Some(item) = consumer.try_recv() {
            rest.push(item);
        }
    }
    assert_eq!(rest, vec![9, 10, 11, 12, 13, 14, 15, 16, 17]);
    assert_eq!(consumer.lapped_count(), 0); // Zero dropped messages under backpressure

    // Drop consumer; dead reader slot is reclaimed and headroom returns to full capacity
    drop(consumer);
    assert_eq!(producer.headroom(), capacity);

    let _ = std::fs::remove_file(&ring_path);
}