Copilot commented on code in PR #1240:
URL: 
https://github.com/apache/rocketmq-client-go/pull/1240#discussion_r3923562014


##########
consumer/lock_test.go:
##########
@@ -0,0 +1,90 @@
+/*
+Licensed to the Apache Software Foundation (ASF) under one or more
+contributor license agreements.  See the NOTICE file distributed with
+this work for additional information regarding copyright ownership.
+The ASF licenses this file to You under the Apache License, Version 2.0
+(the "License"); you may not use this file except in compliance with
+the License.  You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package consumer
+
+import (
+       "sync"
+       "testing"
+       "time"
+
+       "github.com/apache/rocketmq-client-go/v2/primitive"
+)
+
+// TestFetchLockReturnsSameLockForSameQueue is the deterministic proof of the 
bug: fetchLock must
+// return the SAME lock instance for the same queue so that per-queue 
processing is serialized.
+// With a value receiver on QueueLock (which embeds sync.Map), every call 
operates on a copy of the
+// map, so LoadOrStore never persists into the shared map and a brand-new 
*sync.Mutex is returned
+// each time.
+//
+//     buggy (value receiver)   -> l1 != l2 -> FAIL
+//     fixed (pointer receiver) -> l1 == l2 -> PASS
+func TestFetchLockReturnsSameLockForSameQueue(t *testing.T) {
+       ql := newQueueLock()
+       mq := primitive.MessageQueue{Topic: "t", BrokerName: "b", QueueId: 0}
+
+       l1 := ql.fetchLock(mq)
+       l2 := ql.fetchLock(mq)
+
+       if l1 != l2 {
+               t.Fatalf("fetchLock must return the same lock for the same 
queue, got two different instances (%p vs %p)", l1, l2)
+       }
+}
+
+// TestFetchLockProvidesMutualExclusion demonstrates the real-world impact: 
consumeMessageOrderly
+// relies on fetchLock(mq).Lock() to guarantee that only one goroutine 
processes a given queue at a
+// time (FIFO). If every call returns a fresh mutex, there is no mutual 
exclusion and multiple
+// goroutines enter the critical section concurrently.
+//
+//     buggy   -> maxConcurrent > 1 -> FAIL
+//     fixed   -> maxConcurrent == 1 -> PASS
+func TestFetchLockProvidesMutualExclusion(t *testing.T) {
+       ql := newQueueLock()
+       mq := primitive.MessageQueue{Topic: "t", BrokerName: "b", QueueId: 0}
+
+       var stateMu sync.Mutex
+       concurrent, maxConcurrent := 0, 0
+
+       var wg sync.WaitGroup
+       for i := 0; i < 20; i++ {
+               wg.Add(1)
+               go func() {
+                       defer wg.Done()
+                       lock := ql.fetchLock(mq)
+                       lock.Lock()
+                       defer lock.Unlock()
+
+                       stateMu.Lock()
+                       concurrent++
+                       if concurrent > maxConcurrent {
+                               maxConcurrent = concurrent
+                       }
+                       stateMu.Unlock()
+
+                       time.Sleep(2 * time.Millisecond)
+
+                       stateMu.Lock()
+                       concurrent--
+                       stateMu.Unlock()
+               }()
+       }

Review Comment:
   The mutual exclusion test currently relies on goroutine scheduling and a 
small `time.Sleep` to create overlap. That can be timing-sensitive and can 
(rarely) create false negatives if goroutines don't overlap enough. To make 
this more robust, add a start barrier (e.g., a `start := make(chan struct{})` 
that each goroutine waits on before attempting to lock, then `close(start)` 
after spawning) and consider avoiding fixed sleeps by using a short loop with 
`runtime.Gosched()` (or similar) inside the critical section to encourage 
interleaving without depending on wall-clock timing.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to