On Sun, 15 Nov 2026 08:06:49 GMT, Alan Bateman <[email protected]> wrote:

>> The "blocked in write" tests in BlockingSocketOps and BlockingChannelOps are 
>> problematic as the underlying non-blocking socket may be polled as writable 
>> several times before the virtual thread writer blocks indefinitely. The 
>> tests were updated recently to additionally test the blocked thread's 
>> parkBlocker, which makes it harder to test when it wakes up several times.
>> 
>> The tests are simplified to avoid special casing the writing case. The 
>> runAfterParkedAsync method in both tests now polls the thread until it parks 
>> and the parkBlocker is set. BlockingChannelOps tests the expected 
>> parkBlocker whereas BlockingSocketOps just tests for a non-null parkBlocker.
>> 
>> ---------
>> - [x] I confirm that I make this contribution in accordance with the 
>> [OpenJDK Interim AI Policy](https://openjdk.org/legal/ai).
>
> Alan Bateman has updated the pull request incrementally with one additional 
> commit since the last revision:
> 
>   Fixed typos in test comments

LGTM. Dropped some minor remarks/questions, but it can be integrated in its 
current form.

I'd make sure, in addition to Tiers 1-2, @dfuch's reproducer (i.e., 
`test/jdk:jdk_net` with 50 repetitions) passes on all supported platforms.

test/jdk/java/net/vthread/BlockingSocketOps.java line 702:

> 700:      * @return the thread that executes the task
> 701:      */
> 702:     private static Thread runAfterParkedAsync(ThrowingRunnable task) {

Currently, `awaitParked(target); task.run()` failures are only communicated in 
the logs; they don't affect the test execution. You may consider returning a 
`CompletableFuture<Void>`, passing `e` at line 764 to it, and always finishing 
a `@Test` block with a `future.get()` to ensure no `task.run()` failures are 
discarded.

test/jdk/java/net/vthread/BlockingSocketOps.java line 721:

> 719:      */
> 720:     private static void awaitParked(Thread target) throws 
> InterruptedException {
> 721:         while (!isParked(target) || LockSupport.getBlocker(target) == 
> null) {

Curious: Why doesn't `isParked()` have a `LockSupport.getBlocker(target) != 
null` check, but `awaitParked()` does?

test/jdk/java/net/vthread/BlockingSocketOps.java line 757:

> 755:                         }
> 756:                     }
> 757:                 }

It is good to see this nudge is gone.

-------------

Marked as reviewed by vyazici (Reviewer).

PR Review: https://git.openjdk.org/jdk/pull/33907#pullrequestreview-5917077087
PR Review Comment: https://git.openjdk.org/jdk/pull/33907#discussion_r4583547492
PR Review Comment: https://git.openjdk.org/jdk/pull/33907#discussion_r4583507011
PR Review Comment: https://git.openjdk.org/jdk/pull/33907#discussion_r4583565210

Reply via email to