Review feedback changes 1

internals/overlord/overlord.go

@@ -328,19 +328,14 @@ func (o *Overlord) ensureTimerSetup() {
 	o.pruneTicker = time.NewTicker(pruneInterval)
 }
 
-func (o *Overlord) ensureTimerStopClearChannel() {
+func (o *Overlord) ensureTimerStop() {
 	o.ensureLock.Lock()
 	defer o.ensureLock.Unlock()
-	// The Go v1.x timer implementation does not guarantee that a Reset
-	// clears the channel. The documented proposal is to call stop and
-	// manually clear the channel if it already expired. This is required
-	// on loop entry because the timer is started much earlier, and
-	// managers are now allowed to call state.EnsureBefore() as soon as state
-	// is available. If the ensure timer expired before loop entry, this code
-	// ensures the channel is cleared and that only the ensure on entry is
-	// performed (and not a second ensure as a result of the channel).
-	// See: https://pkg.go.dev/time#Timer.Reset
 	o.ensureTimer.Stop()
+	// The Go version < 1.23 timer implementation does not guarantee that a
+	// Stop clears the channel. The documented proposal is to call stop and
+	// manually clear the channel if it already expired.
+	// See: https://pkg.go.dev/time#Timer.Stop
 	select {
 	case <-o.ensureTimer.C:
 	default:
@@ -384,10 +379,10 @@ func (o *Overlord) ensureBefore(d time.Duration) {
 // Loop runs a loop in a goroutine to ensure the current state regularly through StateEngine Ensure.
 func (o *Overlord) Loop() {
 	o.loopTomb.Go(func() error {
-		// We are about to perform the first ensure on loop entry. Stop and
-		// clear the timer channel, so we do not process a spurious second
+		// We are about to perform the first ensure on loop entry. Stop (and
+		// clear the timer channel), so we do not process a spurious second
 		// ensure due to a pending timer channel expiry.
-		o.ensureTimerStopClearChannel()
+		o.ensureTimerStop()
 		for {
 			// TODO: pass a proper context into Ensure
 			o.ensureTimerReset()