mridulm commented on code in PR #50594:
URL: https://github.com/apache/spark/pull/50594#discussion_r2065293524
##########
core/src/main/scala/org/apache/spark/util/UninterruptibleThread.scala:
##########
@@ -69,10 +75,22 @@ private[spark] class UninterruptibleThread(
}
uninterruptibleLock.synchronized {
+ uninterruptible = true
+ }
+
+ while (uninterruptibleLock.synchronized {
// Clear the interrupted status if it's set.
shouldInterruptThread = Thread.interrupted() || shouldInterruptThread
- uninterruptible = true
+ // wait for super.interrupt() to be called
+ !shouldInterruptThread && awaitInterruptThread }) {
+ try {
+ Thread.sleep(100)
Review Comment:
`sleep` has much higher overhead than parking a thread - not to mention it
will suspend for atleast the specified duration until interrupted (which has
additional costs again), while unpark can come really quickly.
> won't be even called and if it is called, it will exit on
InterruptedException almost immediately
`InterruptedException` will not be thrown, as we have already set
`uninterruptible = true`. So it will sleep for the specified duration before
detection.
--
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: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
