disruptor-mp 0.1.0-alpha.2

Ultra-low-latency and high-throughput multiprocess ring-buffer substrate over SHM and mmap (LMAX Disruptor design, extended cross-process), with cross-process producer/consumer coordination, required-consumer liveness, and observability counters.
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
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
//! Builder surface for creating and attaching multiprocess transports.
//!
//! [`SharedDisruptorBuilder`] is the main configuration surface for:
//!
//! - shared-memory or mmap setup
//! - startup coordination
//! - consumer discovery
//! - automatic event-handler threads
//! - wait-policy selection
//!
//! The builder supports two broad usage styles:
//!
//! - automatic event handling via `handle_events_with(...)`
//! - manual polling via `build_consumer()`

use super::consumer::SharedConsumer;
use super::consumer_barrier::{auto_consumer_id, consumer_registration_cursor_name, DiscoveryMode};
use super::producer::{CoordinationMode, SharedProducer};
use crate::env::{read, runtime as runtime_env};
use crate::{MultiProcessResult, SharedCursor, SharedMemoryConfig, SharedRingBuffer};
use disruptor_core::Sequence;
use std::env;
use std::sync::atomic::{AtomicUsize, Ordering};
use std::thread::{self, JoinHandle};
use std::time::Duration;

/// Global counter for generating unique consumer IDs within a process
static CONSUMER_COUNTER: AtomicUsize = AtomicUsize::new(0);

fn uses_registered_auto_ids(discovery_mode: &DiscoveryMode) -> bool {
    matches!(
        discovery_mode,
        DiscoveryMode::Enabled {
            consumer_prefix: None,
            ..
        }
    )
}

fn create_consumer_registration_cursor(
    base_name: &str,
    discovery_mode: &DiscoveryMode,
) -> MultiProcessResult<Option<SharedCursor>> {
    if !uses_registered_auto_ids(discovery_mode) {
        return Ok(None);
    }

    let registration_name = consumer_registration_cursor_name(base_name);
    let cursor = SharedCursor::new_or_attach(&registration_name, 0)?;
    Ok(Some(cursor))
}

fn allocate_registered_consumer_id(base_name: &str) -> Option<String> {
    let registration_name = consumer_registration_cursor_name(base_name);
    let registration = SharedCursor::attach(&registration_name).ok()?;
    let slot = registration.fetch_add(1, Ordering::AcqRel);
    if slot < 0 {
        return None;
    }

    Some(auto_consumer_id(slot as usize))
}

fn default_consumer_id(base_name: &str) -> String {
    if let Some(consumer_id) = allocate_registered_consumer_id(base_name) {
        return consumer_id;
    }

    let process_id = std::process::id();
    let consumer_counter = CONSUMER_COUNTER.fetch_add(1, Ordering::Relaxed);
    format!("c{}_{}", process_id % 10000, consumer_counter)
}

/// Wait strategy for automatic event handlers
#[derive(Debug, Clone, Default)]
pub enum AutoWaitStrategy {
    /// Maximum performance busy spinning (100% CPU usage) - true busy spin, no hints
    BusySpin,
    /// High performance busy spinning with spin loop hints (slightly lower CPU usage)
    BusySpinWithSpinLoopHint,
    /// Hybrid: spin N iterations, then yield the thread
    SpinThenYield {
        /// Number of spin-loop iterations before yielding once.
        spins: usize,
    },
    /// CPU efficient with configurable sleep duration
    Sleep(Duration),
    /// Blocking with efficient waiting (balanced performance/CPU)
    #[default]
    Block,
}

impl AutoWaitStrategy {
    /// Create a high performance wait strategy (true busy spin)
    pub fn high_performance() -> Self {
        AutoWaitStrategy::BusySpin
    }

    /// Create a high performance wait strategy with spin loop hints
    pub fn high_performance_with_hints() -> Self {
        AutoWaitStrategy::BusySpinWithSpinLoopHint
    }

    /// Create a hybrid strategy that spins N times then yields
    pub fn spin_then_yield(spins: usize) -> Self {
        AutoWaitStrategy::SpinThenYield { spins }
    }

    /// Create a CPU efficient wait strategy
    pub fn cpu_efficient() -> Self {
        AutoWaitStrategy::Sleep(Duration::from_micros(1))
    }

    /// Create a custom sleep-based wait strategy
    pub fn sleep(duration: Duration) -> Self {
        AutoWaitStrategy::Sleep(duration)
    }

    /// Create a sleep-based wait strategy with nanosecond precision
    ///
    /// # Special Values
    /// - `0` = use `spin_loop()` instead of sleep (high performance)
    /// - `1..` = Sleep for the specified nanoseconds (lower performance, CPU efficient)
    ///
    /// Note: Actual sleep precision depends on the operating system.
    /// Very small durations (< 1000ns) may not sleep at all on some systems.
    pub fn sleep_nanos(nanos: u64) -> Self {
        if nanos == 0 {
            AutoWaitStrategy::BusySpinWithSpinLoopHint
        } else {
            AutoWaitStrategy::Sleep(Duration::from_nanos(nanos))
        }
    }

    /// Create a sleep-based wait strategy with microsecond precision
    ///
    /// # Special Values
    /// - `0` = use `spin_loop()` instead of sleep (high performance)
    /// - `1..` = Sleep for the specified microseconds (lower performance, CPU efficient)
    pub fn sleep_micros(micros: u64) -> Self {
        if micros == 0 {
            AutoWaitStrategy::BusySpinWithSpinLoopHint
        } else {
            AutoWaitStrategy::Sleep(Duration::from_micros(micros))
        }
    }

    /// Create wait strategy from environment variables with fallback
    ///
    /// Checks these environment variables in order:
    /// 1. `MYELON_AUTO_WAIT_DELAY_NS` - nanosecond precision (`0` = `spin_loop`)
    /// 2. `MYELON_AUTO_WAIT_DELAY_US` - microsecond precision (`0` = `spin_loop`)
    /// 3. Falls back to the provided default
    ///
    /// # Examples
    /// ```bash
    /// # Use spin_loop (maximum performance)
    /// export MYELON_AUTO_WAIT_DELAY_NS=0
    ///
    /// # Sleep for 100 nanoseconds
    /// export MYELON_AUTO_WAIT_DELAY_NS=100
    ///
    /// # Sleep for 10 microseconds
    /// export MYELON_AUTO_WAIT_DELAY_US=10
    /// ```
    pub fn from_env_or(default: AutoWaitStrategy) -> Self {
        // Check for nanosecond precision first
        if let Some(nanos) = read::parse::<u64>(runtime_env::AUTO_WAIT_DELAY_NS) {
            return Self::sleep_nanos(nanos);
        }

        // Check for microsecond precision
        if let Some(micros) = read::parse::<u64>(runtime_env::AUTO_WAIT_DELAY_US) {
            return Self::sleep_micros(micros);
        }

        // Fall back to default
        default
    }
}

/// Automatic consumer that runs in a background thread.
///
/// This provides a handle to a consumer running automatic event processing
/// in a separate thread. It supports graceful shutdown and cleanup operations.
pub struct AutoConsumer {
    join_handle: Option<JoinHandle<()>>,
    shutdown_signal: std::sync::Arc<std::sync::atomic::AtomicBool>,
}

impl AutoConsumer {
    fn new(
        join_handle: JoinHandle<()>,
        shutdown_signal: std::sync::Arc<std::sync::atomic::AtomicBool>,
    ) -> Self {
        Self {
            join_handle: Some(join_handle),
            shutdown_signal,
        }
    }

    /// Signal the consumer to shutdown
    pub fn shutdown(&self) {
        self.shutdown_signal
            .store(true, std::sync::atomic::Ordering::Release);
    }

    /// Wait for the consumer thread to finish
    pub fn join(&mut self) {
        if let Some(handle) = self.join_handle.take() {
            let _ = handle.join();
        }
    }

    /// Shutdown and wait for the consumer thread to finish
    pub fn shutdown_and_join(&mut self) {
        self.shutdown();
        self.join();
    }

    /// Check if the consumer thread is still running
    pub fn is_running(&self) -> bool {
        self.join_handle.is_some()
            && !self
                .shutdown_signal
                .load(std::sync::atomic::Ordering::Acquire)
    }
}

/// Automatic resource management for [`AutoConsumer`].
///
/// Ensures proper cleanup when the consumer goes out of scope, which
/// matters for embedded host runtimes (e.g. an FFI consumer wrapped by
/// a garbage-collected language) that expect automatic resource
/// management.
impl Drop for AutoConsumer {
    fn drop(&mut self) {
        // Signal shutdown and wait for thread to finish
        self.shutdown_signal
            .store(true, std::sync::atomic::Ordering::Release);

        if let Some(handle) = self.join_handle.take() {
            // Give the thread a moment to see the shutdown signal
            std::thread::sleep(super::wait::SLEEP_CONFIG.shutdown_grace_duration());

            // Wait for clean shutdown (with timeout for safety)
            match handle.join() {
                Ok(_) => {
                    // Clean shutdown successful
                }
                Err(_) => {
                    // Thread panicked - this is unexpected but not fatal
                    eprintln!(
                        "Warning: AutoConsumer thread terminated unexpectedly during cleanup"
                    );
                }
            }
        }
    }
}

/// Builder for creating multi-process disruptors.
///
/// This builder provides a fluent API for configuring and creating shared memory
/// disruptors with various coordination modes, discovery options, and event handling
/// strategies. It supports both producer and consumer creation with automatic
/// shared memory management.
///
/// ## Key features
///
/// - **Automatic event delivery**: `handle_events_with()` creates background processing
/// - **Automatic resource management**: proper cleanup via `Drop` implementations
/// - **Configurable timeouts**: customizable coordination and discovery timeouts
/// - **Error handling**: clear error messages and robust failure handling
///
/// ## Usage Patterns
///
/// ```rust,no_run
/// use disruptor_mp::*;
/// use std::time::Duration;
///
/// #[derive(Copy, Clone, Default)]
/// struct Event { data: i64 }
///
/// # fn main() -> Result<(), Box<dyn std::error::Error>> {
/// // Producer with automatic coordination
/// let mut producer = build_shared_single_producer::<Event>("test", 1024)
///     .enable_discovery(2)
///     .with_coordination_timeout(Duration::from_secs(30))
///     .build_producer(Event::default)?;
///
/// // Consumer with automatic event delivery  
/// let consumer = attach_shared_consumer::<Event>("test", 1024)
///     .handle_events_with(|event, seq, eob| {
///         // Events delivered automatically
///     })?;
/// # Ok(())
/// # }
/// ```
pub struct SharedDisruptorBuilder<E> {
    config: SharedMemoryConfig,
    coordination_mode: Option<CoordinationMode>,
    discovery_mode: Option<DiscoveryMode>,
    consumer_id: Option<String>,
    process_core: Option<usize>,
    consumer_core: Option<usize>,
    coordination_timeout: Option<Duration>,
    _phantom: std::marker::PhantomData<E>,
}

#[derive(Copy, Clone)]
enum ProcessRole {
    Producer,
    Consumer,
}

impl ProcessRole {
    fn env_var(self) -> &'static str {
        match self {
            ProcessRole::Producer => runtime_env::PRODUCER_CORE,
            ProcessRole::Consumer => runtime_env::CONSUMER_CORE,
        }
    }

    fn label(self) -> &'static str {
        match self {
            ProcessRole::Producer => "producer process",
            ProcessRole::Consumer => "consumer process",
        }
    }
}

impl<E> SharedDisruptorBuilder<E>
where
    E: Copy + Default + 'static,
{
    /// Create a new builder with the given configuration
    pub fn new(config: SharedMemoryConfig) -> Self {
        Self {
            config,
            coordination_mode: None,
            discovery_mode: None,
            consumer_id: None,
            process_core: None,
            consumer_core: None,
            coordination_timeout: None,
            _phantom: std::marker::PhantomData,
        }
    }

    /// Set a custom coordination timeout
    ///
    /// This overrides the default adaptive timeout behavior for consumer coordination.
    /// Useful for environments with specific timing requirements.
    ///
    /// # Examples
    /// ```rust,ignore
    /// use std::time::Duration;
    /// let builder = builder.with_coordination_timeout(Duration::from_secs(60));
    /// ```
    pub fn with_coordination_timeout(mut self, timeout: Duration) -> Self {
        self.coordination_timeout = Some(timeout);
        self
    }

    /// Set the coordination mode for the producer
    ///
    /// This enables internal coordination to eliminate the need for external coordination structures.
    ///
    /// # Examples
    /// ```rust,ignore
    /// use std::time::Duration;
    /// use disruptor_mp::producer::CoordinationMode;
    ///
    /// // Wait for single consumer with 30 second timeout
    /// let builder = builder.with_coordination(
    ///     CoordinationMode::wait_for_single_consumer(Duration::from_secs(30))
    /// );
    ///
    /// // Wait for multiple consumers with 45 second timeout
    /// let builder = builder.with_coordination(
    ///     CoordinationMode::wait_for_consumers(3, Duration::from_secs(45))
    /// );
    /// ```
    pub fn with_coordination(mut self, coordination_mode: CoordinationMode) -> Self {
        self.coordination_mode = Some(coordination_mode);
        self
    }

    /// Configure consumer discovery mode
    ///
    /// # Discovery Modes (Default: Disabled)
    /// - `DiscoveryMode::Disabled` - No discovery (default), maximum performance for externally coordinated scenarios
    /// - `DiscoveryMode::Enabled { consumer_prefix: None }` - Deterministic slot-based discovery
    ///   when coordination is present, with PID scanning as legacy fallback
    /// - `DiscoveryMode::Enabled { consumer_prefix: Some("DIS_SM") }` - Optimized prefix-based discovery
    ///
    /// # Examples
    /// ```rust,ignore
    /// use disruptor_mp::lock_free::DiscoveryMode;
    ///
    /// // Enable basic discovery for 2 consumers (fixed topology)
    /// let builder = builder.enable_discovery(2);
    ///
    /// // Use consumer naming convention for efficient discovery of 5 consumers
    /// let builder = builder.discover_consumer_with_prefix(5, "DIS_SM");
    /// ```
    pub fn with_discovery(mut self, discovery_mode: DiscoveryMode) -> Self {
        self.discovery_mode = Some(discovery_mode);
        self
    }

    /// Disable consumer discovery (default behavior)
    ///
    /// This is ideal for externally coordinated scenarios like benchmarks
    /// where consumer coordination is handled outside the disruptor.
    /// Discovery is disabled by default for maximum performance.
    pub fn disable_discovery(self) -> Self {
        self.with_discovery(DiscoveryMode::Disabled)
    }

    /// Enable basic consumer discovery for fixed topology (convenience method)
    ///
    /// When startup coordination is present, consumers are assigned deterministic
    /// slot-based IDs (`ad_0`, `ad_1`, ...) so discovery does not rely on PID
    /// proximity. Without coordination, discovery falls back to PID scanning with
    /// the legacy naming pattern `c{pid}_{counter}`.
    /// Stops scanning once the expected number of consumers are discovered.
    /// Default scan interval: 100ms.
    pub fn enable_discovery(self, max_consumers: usize) -> Self {
        self.with_discovery(DiscoveryMode::enabled(max_consumers))
    }

    /// Enable optimized discovery with consumer naming convention for fixed topology (convenience method)
    ///
    /// This uses consumer name prefixes for much faster consumer discovery
    /// compared to PID scanning. Consumers will be discovered using names
    /// like `DIS_SM_1`, `DIS_SM_2`, etc.
    /// The prefix identifies the consumer group/application, not the OS process name.
    /// Stops scanning once the expected number of consumers are discovered, saving CPU cycles.
    /// Default scan interval: 100ms.
    pub fn discover_consumer_with_prefix(self, max_consumers: usize, prefix: &str) -> Self {
        self.with_discovery(DiscoveryMode::with_consumer_prefix(
            max_consumers,
            prefix.to_string(),
        ))
    }

    /// Enable discovery with custom scan interval for fixed topology (convenience method)
    ///
    /// This configures how often the producer scans for new consumers.
    /// Stops scanning once the expected number of consumers are discovered, saving CPU cycles.
    /// Shorter intervals provide faster consumer detection but use more CPU.
    /// Longer intervals reduce CPU usage but may delay consumer detection.
    pub fn with_discovery_interval(self, max_consumers: usize, scan_interval: Duration) -> Self {
        self.with_discovery(DiscoveryMode::with_scan_interval(
            max_consumers,
            scan_interval,
        ))
    }

    /// Enable discovery with consumer prefix and custom scan interval for fixed topology (convenience method)
    ///
    /// Combines optimized consumer prefix discovery with configurable scan timing.
    /// Stops scanning once the expected number of consumers are discovered, saving CPU cycles.
    pub fn discover_consumer_with_prefix_and_interval(
        self,
        max_consumers: usize,
        prefix: &str,
        scan_interval: Duration,
    ) -> Self {
        self.with_discovery(DiscoveryMode::with_consumer_prefix_and_interval(
            max_consumers,
            prefix.to_string(),
            scan_interval,
        ))
    }

    /// Enable coordination for single consumer scenarios (convenience method)
    ///
    /// This is equivalent to:
    /// ```rust,ignore
    /// use std::time::Duration;
    /// use disruptor_mp::producer::CoordinationMode;
    /// builder.with_coordination(CoordinationMode::wait_for_single_consumer(timeout))
    /// ```
    pub fn wait_for_single_consumer(self, timeout: Duration) -> Self {
        self.with_coordination(CoordinationMode::wait_for_single_consumer(timeout))
    }

    /// Enable coordination for multiple consumer scenarios (convenience method)
    pub fn wait_for_consumers(self, min_consumers: i64, timeout: Duration) -> Self {
        self.with_coordination(CoordinationMode::wait_for_consumers(min_consumers, timeout))
    }

    /// Set a custom consumer ID for this consumer instance
    ///
    /// This is useful for prefix-based consumer discovery where consumers need
    /// specific naming patterns. If not set, a default ID will be generated
    /// using the pattern "c{pid}_{counter}".
    ///
    /// # Example
    /// ```rust,ignore
    /// let consumer = builder
    ///     .with_consumer_id("TEST_CONSUMER_1")
    ///     .build_consumer()?;
    /// ```
    pub fn with_consumer_id(mut self, consumer_id: &str) -> Self {
        self.consumer_id = Some(consumer_id.to_string());
        self
    }

    /// Bind automatic consumer thread to a specific CPU core.
    ///
    /// Uses Linux core affinity on Linux builds. On macOS and other non-Linux
    /// platforms this configuration is accepted, ignored, and logged as a
    /// best-effort no-op so consumer bring-up still succeeds.
    ///
    /// The builder resolves the effective affinity order:
    /// 1. Explicit builder value.
    /// 2. `MYELON_AUTO_CONSUMER_CORE` environment variable (if set).
    pub fn with_consumer_core(mut self, core_id: usize) -> Self {
        self.consumer_core = Some(core_id);
        self
    }

    /// Bind this process (producer or consumer) to a specific CPU core.
    ///
    /// On Linux this pins the current thread before ring setup. For the normal
    /// single-threaded producer/consumer entrypoints this acts as process-role
    /// affinity, and newly spawned auto-consumer threads inherit the mask unless
    /// they are pinned separately with `with_consumer_core()`. On macOS and
    /// other non-Linux platforms this request is accepted, ignored, and logged
    /// as a best-effort no-op so producer/consumer bring-up still succeeds.
    ///
    /// Resolution order:
    /// 1. Explicit builder value via `with_process_core()`
    /// 2. Role specific env var (`MYELON_PRODUCER_CORE`
    ///    or `MYELON_CONSUMER_CORE`)
    /// 3. Generic env var `MYELON_PROCESS_CORE`
    pub fn with_process_core(mut self, core_id: usize) -> Self {
        self.process_core = Some(core_id);
        self
    }

    fn maybe_pin_process_to_core(&self, role: ProcessRole) {
        if let Some(core_id) = resolve_process_core(self.process_core, role) {
            pin_current_thread_to_core(core_id, role.label());
        }
    }

    /// Build a consumer with automatic batch event handling
    ///
    /// This provides automatic event delivery with configurable wait strategies.
    /// Events are processed in batches for optimal performance.
    ///
    /// # Wait Strategies
    /// - `AutoWaitStrategy::BusySpin` - Maximum performance (100% CPU, true busy spin)
    /// - `AutoWaitStrategy::BusySpinWithSpinLoopHint` - High performance with CPU hints
    /// - `AutoWaitStrategy::Block` - Balanced performance/CPU (default)
    /// - `AutoWaitStrategy::Sleep(duration)` - CPU efficient with custom sleep
    ///
    /// # Examples
    /// ```rust,ignore
    /// use std::time::Duration;
    /// use disruptor_mp::builder::AutoWaitStrategy;
    ///
    /// // Maximum performance for AI inference (true busy spin)
    /// consumer.handle_events_batch(handler, AutoWaitStrategy::high_performance())
    ///
    /// // High performance with CPU hints (slightly more efficient)
    /// consumer.handle_events_batch(handler, AutoWaitStrategy::high_performance_with_hints())
    ///
    /// // CPU efficient
    /// consumer.handle_events_batch(handler, AutoWaitStrategy::cpu_efficient())
    ///
    /// // Custom sleep duration
    /// consumer.handle_events_batch(handler, AutoWaitStrategy::sleep(Duration::from_micros(10)))
    ///
    /// // Balanced (default)
    /// consumer.handle_events_batch(handler, AutoWaitStrategy::default())
    /// ```
    pub fn handle_events_batch<EH>(
        self,
        mut event_handler: EH,
        wait_strategy: AutoWaitStrategy,
    ) -> MultiProcessResult<AutoConsumer>
    where
        EH: 'static + Send + FnMut(&E, Sequence, bool),
    {
        self.maybe_pin_process_to_core(ProcessRole::Consumer);

        let ring_buffer: SharedRingBuffer<E> = SharedRingBuffer::attach(self.config.clone())?;

        // Attach to existing shared atomics
        let producer_sequence_name = format!("{}_producer_seq", self.config.name);
        let producer_sequence = SharedCursor::attach(&producer_sequence_name)?;

        // Use custom consumer ID if provided, otherwise generate a deterministic
        // slot-based ID when coordinated discovery is available.
        let consumer_id = if let Some(custom_id) = self.consumer_id {
            custom_id
        } else {
            default_consumer_id(&self.config.name)
        };

        // Create this consumer's own sequence tracker
        let consumer_sequence_name = format!("{}_{}_seq", self.config.name, consumer_id);
        let mut consumer_sequence = SharedCursor::new_or_attach(&consumer_sequence_name, -1)?;
        if consumer_sequence.is_owner() {
            consumer_sequence.set_owner(false);
        }

        // Create the consumer with coordination support
        let mut consumer = SharedConsumer::new_with_coordination(
            ring_buffer,
            producer_sequence,
            consumer_sequence,
            consumer_id,
            Some(self.config.name.clone()),
        );

        // Create shutdown signal
        let shutdown_signal = std::sync::Arc::new(std::sync::atomic::AtomicBool::new(false));
        let shutdown_signal_clone = std::sync::Arc::clone(&shutdown_signal);
        let consumer_core = resolve_auto_consumer_core(self.consumer_core);
        let thread_name = format!("multiprocess-consumer-{}", std::process::id());

        // Spawn background thread for automatic event processing
        let join_handle = thread::Builder::new()
            .name(thread_name.clone())
            .spawn(move || {
                if let Some(core_id) = consumer_core {
                    pin_current_thread_to_core(core_id, &thread_name);
                }

                loop {
                    let processed = match wait_strategy {
                        AutoWaitStrategy::BusySpin => {
                            // Maximum performance: True busy spin
                            // TRUE BUSY SPIN: Do nothing when no events (like single-process BusySpin)
                            // This achieves maximum throughput at 100% CPU usage
                            consumer.process_available(|event, seq| {
                                // Approximate end_of_batch as false for maximum performance
                                event_handler(event, seq, false);
                            })
                        }
                        AutoWaitStrategy::BusySpinWithSpinLoopHint => {
                            // High performance: Busy spin with hints
                            let processed = consumer.process_available(|event, seq| {
                                // Approximate end_of_batch as false for maximum performance
                                event_handler(event, seq, false);
                            });

                            if processed == 0 {
                                std::hint::spin_loop();
                            }
                            processed
                        }
                        AutoWaitStrategy::SpinThenYield { spins } => {
                            // Hybrid: spin N times then yield to reduce tail latencies
                            let processed = consumer.process_available(|event, seq| {
                                event_handler(event, seq, false);
                            });

                            if processed == 0 {
                                // Spin N iterations using spin_loop hint
                                for _ in 0..spins {
                                    std::hint::spin_loop();
                                }
                                // Then yield the thread to reduce scheduler starvation
                                std::thread::yield_now();
                            }
                            processed
                        }
                        AutoWaitStrategy::Block => {
                            // ⚖️ BALANCED: Batch processing with non-blocking check first
                            // Check if events are immediately available
                            let processed = consumer.process_available(|event, seq| {
                                // Approximate end_of_batch as false for simplicity
                                event_handler(event, seq, false);
                            });

                            if processed == 0 {
                                // No events available, wait a bit before trying again
                                // This prevents the infinite spin in consume_next()
                                super::wait::perform_default_block_wait();
                            }

                            processed
                        }
                        AutoWaitStrategy::Sleep(duration) => {
                            // CPU efficient: Batch processing with configurable sleep
                            let processed = consumer.process_available(|event, seq| {
                                // Approximate end_of_batch as false for simplicity
                                event_handler(event, seq, false);
                            });

                            if processed == 0 {
                                super::wait::sleep_or_yield(duration);
                            }
                            processed
                        }
                    };

                    // Continue processing - the wait strategy handles timing

                    // Performance optimization: Only check shutdown when idle (no events processed)
                    // This avoids atomic loads in the hot path when processing events
                    if processed == 0
                        && shutdown_signal_clone.load(std::sync::atomic::Ordering::Acquire)
                    {
                        break;
                    }
                }
            })
            .expect("Should spawn consumer thread");

        Ok(AutoConsumer::new(join_handle, shutdown_signal))
    }

    /// Build a consumer with automatic event handling (convenience method)
    ///
    /// Uses the default balanced wait strategy (Block).
    /// For performance tuning, use `handle_events_batch()` with explicit wait strategy.
    pub fn handle_events_with<EH>(self, event_handler: EH) -> MultiProcessResult<AutoConsumer>
    where
        EH: 'static + Send + FnMut(&E, Sequence, bool),
    {
        self.handle_events_batch(event_handler, AutoWaitStrategy::default())
    }

    /// Build a producer (creates shared memory segments)
    ///
    /// Note: Only one producer can exist per shared memory segment.
    /// For multi-process scenarios, use Single Producer Multiple Consumer (SPMC) pattern.
    pub fn build_producer<F>(self, event_factory: F) -> MultiProcessResult<SharedProducer<E>>
    where
        F: FnMut() -> E,
    {
        #[cfg(dst)]
        if crate::dst::buggify(file!(), line!()) {
            std::thread::sleep(Duration::from_millis(100));
        }

        self.maybe_pin_process_to_core(ProcessRole::Producer);

        let discovery_mode = self.discovery_mode.unwrap_or_default();
        let consumer_registration =
            create_consumer_registration_cursor(&self.config.name, &discovery_mode)?;

        let ring_buffer: SharedRingBuffer<E> =
            SharedRingBuffer::new(self.config.clone(), event_factory)?;

        // Create shared atomic for producer sequence
        let producer_sequence_name = format!("{}_producer_seq", self.config.name);
        let producer_sequence = SharedCursor::new(&producer_sequence_name, -1)?;

        // Automatic coordination when discovery is enabled
        // This eliminates the need for users to manually configure coordination
        let coordination_mode = self.coordination_mode.unwrap_or_else(|| {
            // Auto-enable coordination when discovery is enabled
            match &discovery_mode {
                DiscoveryMode::Enabled { max_consumers, .. } => {
                    // Use custom timeout if provided, otherwise use adaptive timeout
                    let timeout = self.coordination_timeout.unwrap_or_else(|| {
                        // FRAMEWORK INTELLIGENCE: Adaptive timeout based on consumer count
                        match *max_consumers {
                            1 => Duration::from_secs(15),     // SPSC: Quick startup
                            2 => Duration::from_secs(20),     // SPMC-2: Proven sweet spot
                            3..=4 => Duration::from_secs(25), // SPMC-3/4: Moderate
                            5..=8 => Duration::from_secs(30), // SPMC-5+: More time for coordination
                            _ => Duration::from_secs(45),     // SPMC-9+: Maximum timeout
                        }
                    });

                    CoordinationMode::wait_for_consumers(*max_consumers as i64, timeout)
                }
                _ => CoordinationMode::default(), // No discovery = immediate start
            }
        });

        let mut producer = SharedProducer::new_with_coordination_and_discovery(
            ring_buffer,
            producer_sequence,
            self.config.name.clone(), // Pass base name for consumer discovery
            coordination_mode.clone(),
            discovery_mode,
            consumer_registration,
        );

        // Handle coordination during producer creation, not first publish
        match coordination_mode {
            CoordinationMode::Immediate => {
                // No coordination - for external coordination benchmarks
                producer.coordination_completed = true;
            }
            CoordinationMode::WaitForConsumers {
                min_consumers,
                timeout,
            } => {
                println!(
                    "Framework coordinating startup: waiting for {} consumers (timeout: {:?})...",
                    min_consumers, timeout
                );
                let coordination_ready = producer
                    .consumer_barrier
                    .wait_for_consumers_ready(min_consumers, timeout);
                if !coordination_ready {
                    eprintln!("Warning: Timed out waiting for {} consumers after {:?}. Producer created anyway.",
                        min_consumers, timeout);
                    println!(
                        "Framework coordination incomplete - producer continuing without {} ready consumers",
                        min_consumers
                    );
                } else {
                    println!(
                        "Framework coordination completed - {} consumers ready",
                        min_consumers
                    );
                }
                // Mark coordination as completed
                producer.coordination_completed = true;
            }
            CoordinationMode::BufferUntilConsumers { .. } => {
                // Future enhancement - for now, treat as completed
                producer.coordination_completed = true;
            }
        }

        Ok(producer)
    }

    /// Build a consumer (attaches to existing shared memory segments)
    pub fn build_consumer(self) -> MultiProcessResult<SharedConsumer<E>> {
        #[cfg(dst)]
        if crate::dst::buggify(file!(), line!()) {
            std::thread::sleep(Duration::from_millis(100));
        }

        self.maybe_pin_process_to_core(ProcessRole::Consumer);

        let ring_buffer: SharedRingBuffer<E> = SharedRingBuffer::attach(self.config.clone())?;

        // Attach to existing shared atomics
        let producer_sequence_name = format!("{}_producer_seq", self.config.name);
        let producer_sequence = SharedCursor::attach(&producer_sequence_name)?;

        // Use custom consumer ID if provided, otherwise generate a deterministic
        // slot-based ID when coordinated discovery is available.
        let consumer_id = if let Some(custom_id) = self.consumer_id {
            custom_id
        } else {
            default_consumer_id(&self.config.name)
        };

        // Create this consumer's own sequence tracker
        // Note: Uses new_or_attach because multiple consumers might start simultaneously
        // and try to create the same sequence name (different from coordination structures)
        let consumer_sequence_name = format!("{}_{}_seq", self.config.name, consumer_id);
        let mut consumer_sequence = SharedCursor::new_or_attach(&consumer_sequence_name, -1)?;
        if consumer_sequence.is_owner() {
            consumer_sequence.set_owner(false);
        }

        Ok(SharedConsumer::new_with_coordination(
            ring_buffer,
            producer_sequence,
            consumer_sequence,
            consumer_id,
            Some(self.config.name.clone()),
        ))
    }
}

fn resolve_auto_consumer_core(consumer_core: Option<usize>) -> Option<usize> {
    if let Some(core_id) = consumer_core {
        return Some(core_id);
    }

    resolve_core_from_env_vars(&[runtime_env::AUTO_CONSUMER_CORE])
}

fn resolve_process_core(process_core: Option<usize>, role: ProcessRole) -> Option<usize> {
    if let Some(core_id) = process_core {
        return Some(core_id);
    }

    resolve_core_from_env_vars(&[role.env_var(), runtime_env::PROCESS_CORE])
}

fn resolve_core_from_env_vars(env_vars: &[&str]) -> Option<usize> {
    for env_var in env_vars {
        match env::var(env_var) {
            Ok(raw) => match raw.parse::<usize>() {
                Ok(core_id) => return Some(core_id),
                Err(_) => {
                    eprintln!(
                        "Invalid {}='{}'. Expected a non-negative integer.",
                        env_var, raw
                    );
                    return None;
                }
            },
            Err(env::VarError::NotPresent) => {}
            Err(env::VarError::NotUnicode(_)) => {
                eprintln!("Invalid {}: value is not valid Unicode.", env_var);
                return None;
            }
        }
    }

    None
}

#[cfg(target_os = "linux")]
fn pin_current_thread_to_core(core_id: usize, thread_name: &str) {
    let available_cores = core_affinity::get_core_ids().unwrap_or_default();
    let core = core_affinity::CoreId { id: core_id };

    if !available_cores
        .iter()
        .any(|candidate| candidate.id == core_id)
    {
        eprintln!(
            "Could not pin {} to core {}: core not available.",
            thread_name, core_id
        );
        return;
    }

    if !core_affinity::set_for_current(core) {
        eprintln!("Could not pin {} to core {}.", thread_name, core_id);
    }
}

#[cfg(not(target_os = "linux"))]
fn pin_current_thread_to_core(core_id: usize, thread_name: &str) {
    eprintln!(
        "CPU affinity support is Linux-only in disruptor-mp. Requested pinning {} -> core {} ignored; bring-up continues without pinning on this platform.",
        thread_name, core_id
    );
}

/// Create a shared single producer for multi-process communication
///
/// This creates a Single Producer Multiple Consumer (SPMC) setup where:
/// - One process creates and owns the producer
/// - Multiple processes can attach as consumers (each sees all events)
///
/// Note: `SharedProducer` cannot be cloned across processes. Each shared memory
/// segment supports exactly one producer process.
///
/// For automatic coordination, use the builder pattern:
/// ```rust,ignore
/// use std::time::Duration;
/// use disruptor_mp::build_shared_single_producer;
///
/// let producer = build_shared_single_producer::<Event>("test", 1024)
///     .wait_for_single_consumer(Duration::from_secs(30))
///     .build_producer(Event::default)?;
/// ```
/// Create a builder for a shared single producer with the given name and size.
///
/// This is a convenience function that creates a [`SharedDisruptorBuilder`] configured
/// for producer creation with the specified shared memory segment name and buffer size.
///
/// # Arguments
/// * `name` - Shared memory segment name (keep under 10 characters for cross-platform compatibility)
/// * `size` - Ring buffer size (must be power of 2)
///
/// # Examples
/// ```rust
/// use disruptor_mp::build_shared_single_producer;
///
/// #[derive(Copy, Clone, Default)]
/// struct Event { data: i64 }
///
/// let builder = build_shared_single_producer::<Event>("ring123", 1024);
/// let producer = builder.build_producer(|| Event::default())?;
/// # Ok::<(), Box<dyn std::error::Error>>(())
/// ```
pub fn build_shared_single_producer<E: Copy + Default + 'static>(
    name: &str,
    size: usize,
) -> SharedDisruptorBuilder<E> {
    let config = SharedMemoryConfig {
        name: name.to_string(),
        buffer_size: size,
        element_size: std::mem::size_of::<E>(),
        create: true,
    };

    SharedDisruptorBuilder::new(config)
}

/// Attach to an existing shared disruptor as a consumer
///
/// This allows multiple consumer processes to attach to a shared memory segment
/// created by a producer process. Each consumer will see all events (broadcast semantics).
/// Create a builder for attaching to an existing shared consumer.
///
/// This is a convenience function that creates a [`SharedDisruptorBuilder`] configured
/// for consumer attachment to an existing shared memory segment created by a producer.
///
/// # Arguments
/// * `name` - Shared memory segment name (must match the producer's name)
/// * `size` - Ring buffer size (must match the producer's size)
///
/// # Examples
/// ```rust,no_run
/// use disruptor_mp::attach_shared_consumer;
///
/// #[derive(Copy, Clone, Default)]
/// struct Event { data: i64 }
///
/// let builder = attach_shared_consumer::<Event>("ring123", 1024);
/// let consumer = builder.build_consumer()?;
/// # Ok::<(), Box<dyn std::error::Error>>(())
/// ```
pub fn attach_shared_consumer<E: Copy + Default + 'static>(
    name: &str,
    size: usize,
) -> SharedDisruptorBuilder<E> {
    let config = SharedMemoryConfig {
        name: name.to_string(),
        buffer_size: size,
        element_size: std::mem::size_of::<E>(),
        create: false,
    };

    SharedDisruptorBuilder::new(config)
}

#[cfg(test)]
mod tests {
    use super::*;
    use std::ffi::OsString;
    use std::sync::atomic::{AtomicUsize, Ordering};
    use std::sync::Arc;
    use std::sync::{Mutex, OnceLock};
    use std::time::{Duration, Instant};

    #[derive(Copy, Clone, Default)]
    struct TestEvent {
        value: i64,
    }

    fn env_lock() -> &'static Mutex<()> {
        static ENV_LOCK: OnceLock<Mutex<()>> = OnceLock::new();
        ENV_LOCK.get_or_init(|| Mutex::new(()))
    }

    fn with_env_vars<const N: usize, F, R>(
        vars: [(&'static str, Option<&'static str>); N],
        f: F,
    ) -> R
    where
        F: FnOnce() -> R,
    {
        let _guard = env_lock().lock().expect("environment lock poisoned");
        let previous: [(&'static str, Option<OsString>); N] =
            vars.map(|(key, _)| (key, std::env::var_os(key)));

        for (key, value) in vars {
            match value {
                Some(value) => std::env::set_var(key, value),
                None => std::env::remove_var(key),
            }
        }

        let result = f();

        for (key, value) in previous {
            match value {
                Some(value) => std::env::set_var(key, value),
                None => std::env::remove_var(key),
            }
        }

        result
    }

    /// Test that Block wait strategy doesn't hang when no events are available
    #[test]
    fn test_block_wait_strategy_no_hang() {
        let segment_name = format!("test_blk_{}", std::process::id() % 10000);
        let buffer_size = 128;

        // Create producer
        let mut producer = build_shared_single_producer::<TestEvent>(&segment_name, buffer_size)
            .build_producer(TestEvent::default)
            .expect("Failed to create producer");

        // Create consumer with Block wait strategy
        let events_received = Arc::new(AtomicUsize::new(0));
        let events_clone = Arc::clone(&events_received);

        let consumer = attach_shared_consumer::<TestEvent>(&segment_name, buffer_size)
            .handle_events_batch(
                move |_event: &TestEvent, _seq, _eob| {
                    events_clone.fetch_add(1, Ordering::Relaxed);
                },
                AutoWaitStrategy::Block,
            )
            .expect("Failed to create consumer");

        // Wait a bit to ensure consumer doesn't hang
        std::thread::sleep(Duration::from_millis(100));

        // Publish an event
        producer.publish(|event| {
            event.value = 42;
        });

        // Wait for event to be processed
        std::thread::sleep(Duration::from_millis(50));

        // Verify event was received
        assert_eq!(events_received.load(Ordering::Relaxed), 1);

        // Shutdown consumer
        drop(consumer);
    }

    /// Test that `BusySpin` wait strategy works correctly
    #[test]
    fn test_busy_spin_wait_strategy() {
        let segment_name = format!("test_spn_{}", std::process::id() % 10000);
        let buffer_size = 128;

        // Create producer
        let mut producer = build_shared_single_producer::<TestEvent>(&segment_name, buffer_size)
            .build_producer(TestEvent::default)
            .expect("Failed to create producer");

        // Create consumer with BusySpin wait strategy
        let events_received = Arc::new(AtomicUsize::new(0));
        let events_clone = Arc::clone(&events_received);

        let consumer = attach_shared_consumer::<TestEvent>(&segment_name, buffer_size)
            .handle_events_batch(
                move |_event: &TestEvent, _seq, _eob| {
                    events_clone.fetch_add(1, Ordering::Relaxed);
                },
                AutoWaitStrategy::BusySpin,
            )
            .expect("Failed to create consumer");

        // Publish multiple events
        for i in 0..10 {
            producer.publish(|event| {
                event.value = i;
            });
        }

        // Wait for events to be processed
        std::thread::sleep(Duration::from_millis(50));

        // Verify all events were received
        assert_eq!(events_received.load(Ordering::Relaxed), 10);

        // Shutdown consumer
        drop(consumer);
    }

    /// Test that Sleep wait strategy works correctly
    #[test]
    fn test_sleep_wait_strategy() {
        let segment_name = format!("test_slp_{}", std::process::id() % 10000);
        let buffer_size = 128;

        // Create producer
        let mut producer = build_shared_single_producer::<TestEvent>(&segment_name, buffer_size)
            .build_producer(TestEvent::default)
            .expect("Failed to create producer");

        // Create consumer with Sleep wait strategy
        let events_received = Arc::new(AtomicUsize::new(0));
        let events_clone = Arc::clone(&events_received);

        let consumer = attach_shared_consumer::<TestEvent>(&segment_name, buffer_size)
            .handle_events_batch(
                move |_event: &TestEvent, _seq, _eob| {
                    events_clone.fetch_add(1, Ordering::Relaxed);
                },
                AutoWaitStrategy::Sleep(Duration::from_millis(5)),
            )
            .expect("Failed to create consumer");

        // Publish events
        for i in 0..5 {
            producer.publish(|event| {
                event.value = i;
            });
            std::thread::sleep(Duration::from_millis(10));
        }

        // Wait for events to be processed
        std::thread::sleep(Duration::from_millis(100));

        // Verify all events were received
        assert_eq!(events_received.load(Ordering::Relaxed), 5);

        // Shutdown consumer
        drop(consumer);
    }

    /// Test `AutoConsumer` shutdown mechanism
    #[test]
    fn test_auto_consumer_shutdown() {
        let segment_name = format!("test_sht_{}", std::process::id() % 10000);
        let buffer_size = 128;

        // Create producer
        let mut producer = build_shared_single_producer::<TestEvent>(&segment_name, buffer_size)
            .build_producer(TestEvent::default)
            .expect("Failed to create producer");

        // Create consumer
        let events_received = Arc::new(AtomicUsize::new(0));
        let events_clone = Arc::clone(&events_received);

        let mut consumer = attach_shared_consumer::<TestEvent>(&segment_name, buffer_size)
            .handle_events_batch(
                move |_event: &TestEvent, _seq, _eob| {
                    events_clone.fetch_add(1, Ordering::Relaxed);
                },
                AutoWaitStrategy::Block,
            )
            .expect("Failed to create consumer");

        // Publish some events
        for i in 0..5 {
            producer.publish(|event| {
                event.value = i;
            });
        }

        // Wait for events to be processed
        std::thread::sleep(Duration::from_millis(50));
        assert_eq!(events_received.load(Ordering::Relaxed), 5);

        // Shutdown consumer
        consumer.shutdown();

        // Publish more events
        for i in 5..10 {
            producer.publish(|event| {
                event.value = i;
            });
        }

        // Wait and verify no new events are processed
        std::thread::sleep(Duration::from_millis(50));
        assert_eq!(events_received.load(Ordering::Relaxed), 5);

        // Clean up
        consumer.join();
    }

    /// Test that `AutoConsumer` processes events correctly
    /// Note: The batch tracking with `end_of_batch` flag is not reliable in the current
    /// implementation as it's approximated for performance reasons
    #[test]
    fn test_auto_consumer_batch_processing() {
        let segment_name = format!("test_bat_{}", std::process::id() % 10000);
        let buffer_size = 1024;

        // Track events processed (simpler test)
        let events_processed = Arc::new(AtomicUsize::new(0));
        let events_clone = Arc::clone(&events_processed);

        // Create producer WITHOUT discovery (since buffer wrapping is now fixed)
        let mut producer = build_shared_single_producer::<TestEvent>(&segment_name, buffer_size)
            .build_producer(TestEvent::default)
            .expect("Failed to create producer");

        // Create consumer after producer (now safe with buffer wrapping fix)
        let consumer = attach_shared_consumer::<TestEvent>(&segment_name, buffer_size)
            .handle_events_batch(
                move |_event: &TestEvent, _seq, _end_of_batch| {
                    events_clone.fetch_add(1, Ordering::Relaxed);
                },
                AutoWaitStrategy::Block,
            )
            .expect("Failed to create consumer");

        // Publish events in bursts
        for burst in 0..3 {
            for i in 0..10 {
                producer.publish(|event| {
                    event.value = burst * 10 + i;
                });
            }
            std::thread::sleep(Duration::from_millis(50));
        }

        // Wait for processing
        std::thread::sleep(Duration::from_millis(200)); // Slightly longer wait

        // Verify all events were processed
        let total_events = events_processed.load(Ordering::Relaxed);
        assert_eq!(total_events, 30, "Should have processed all 30 events");

        // Shutdown consumer
        drop(consumer);
    }

    /// Test performance characteristics of different wait strategies
    #[test]
    #[ignore] // Ignore by default as this is a performance test
    fn test_wait_strategy_performance() {
        let buffer_size = 8192;
        let num_events = 100_000;

        // Test each wait strategy
        let strategies = vec![
            ("BusySpin", AutoWaitStrategy::BusySpin),
            (
                "BusySpinWithHint",
                AutoWaitStrategy::BusySpinWithSpinLoopHint,
            ),
            ("Block", AutoWaitStrategy::Block),
            (
                "Sleep_1us",
                AutoWaitStrategy::Sleep(Duration::from_micros(1)),
            ),
            (
                "Sleep_100us",
                AutoWaitStrategy::Sleep(Duration::from_micros(100)),
            ),
        ];

        for (name, strategy) in strategies {
            let segment_name = format!("tst_p_{}_{}", name, std::process::id() % 10000);

            // Create producer
            let mut producer =
                build_shared_single_producer::<TestEvent>(&segment_name, buffer_size)
                    .build_producer(TestEvent::default)
                    .expect("Failed to create producer");

            // Create consumer
            let events_received = Arc::new(AtomicUsize::new(0));
            let events_clone = Arc::clone(&events_received);
            let start = Instant::now();

            let consumer = attach_shared_consumer::<TestEvent>(&segment_name, buffer_size)
                .handle_events_batch(
                    move |_event: &TestEvent, _seq, _eob| {
                        events_clone.fetch_add(1, Ordering::Relaxed);
                    },
                    strategy,
                )
                .expect("Failed to create consumer");

            // Publish events
            for i in 0..num_events {
                producer.publish(|event| {
                    event.value = i;
                });
            }

            // Wait for all events to be processed
            while events_received.load(Ordering::Relaxed) < num_events as usize {
                std::thread::sleep(Duration::from_millis(1));
            }

            let elapsed = start.elapsed();
            let events_per_sec = num_events as f64 / elapsed.as_secs_f64();

            println!("{} strategy: {:.0} events/sec", name, events_per_sec);

            // Shutdown consumer
            drop(consumer);
        }
    }

    #[test]
    fn test_consumer_core_override_is_preserved() {
        let segment_name = format!("test_cpu_{}", std::process::id() % 10000);
        let buffer_size = 128;
        let builder =
            attach_shared_consumer::<TestEvent>(&segment_name, buffer_size).with_consumer_core(3);

        assert_eq!(builder.consumer_core, Some(3));
    }

    #[test]
    fn test_consumer_core_resolve_from_builder_overrides_env() {
        let override_core = 7usize;
        let resolved = resolve_auto_consumer_core(Some(override_core));
        assert_eq!(resolved, Some(override_core));
    }

    #[test]
    fn test_consumer_core_resolve_from_env_var() {
        with_env_vars([(runtime_env::AUTO_CONSUMER_CORE, Some("9"))], || {
            let resolved = resolve_auto_consumer_core(None);
            assert_eq!(resolved, Some(9));
        });
    }

    #[test]
    fn test_consumer_core_resolve_from_invalid_env_var() {
        with_env_vars([(runtime_env::AUTO_CONSUMER_CORE, Some("invalid"))], || {
            let resolved = resolve_auto_consumer_core(None);
            assert_eq!(resolved, None);
        });
    }

    #[test]
    fn test_process_core_override_is_preserved() {
        let segment_name = format!("test_prc_{}", std::process::id() % 10000);
        let buffer_size = 128;
        let builder = build_shared_single_producer::<TestEvent>(&segment_name, buffer_size)
            .with_process_core(5);

        assert_eq!(builder.process_core, Some(5));
    }

    #[test]
    fn test_process_core_resolve_from_builder_overrides_env() {
        with_env_vars(
            [
                (runtime_env::PRODUCER_CORE, Some("7")),
                (runtime_env::PROCESS_CORE, Some("9")),
            ],
            || {
                let resolved = resolve_process_core(Some(11), ProcessRole::Producer);
                assert_eq!(resolved, Some(11));
            },
        );
    }

    #[test]
    fn test_process_core_resolve_from_role_specific_env_var() {
        with_env_vars(
            [
                (runtime_env::PRODUCER_CORE, Some("6")),
                (runtime_env::PROCESS_CORE, Some("8")),
            ],
            || {
                let resolved = resolve_process_core(None, ProcessRole::Producer);
                assert_eq!(resolved, Some(6));
            },
        );
    }

    #[test]
    fn test_process_core_resolve_from_generic_env_var() {
        with_env_vars(
            [
                (runtime_env::PRODUCER_CORE, None),
                (runtime_env::PROCESS_CORE, Some("10")),
            ],
            || {
                let resolved = resolve_process_core(None, ProcessRole::Producer);
                assert_eq!(resolved, Some(10));
            },
        );
    }

    #[test]
    fn test_process_core_resolve_from_invalid_role_specific_env_var() {
        with_env_vars(
            [
                (runtime_env::CONSUMER_CORE, Some("invalid")),
                (runtime_env::PROCESS_CORE, Some("12")),
            ],
            || {
                let resolved = resolve_process_core(None, ProcessRole::Consumer);
                assert_eq!(resolved, None);
            },
        );
    }
}