On Wed, 25 Mar 2026 21:49:11 GMT, Doug Lea <[email protected]> wrote:
>> Changes signal filtering to avoid possible starvation
>
> Doug Lea has updated the pull request with a new target base due to a merge
> or a rebase. The pull request now contains 80 commits:
>
> - Merge branch 'openjdk:master' into JDK-8373118
> - Merge branch 'openjdk:master' into JDK-8373118
> - Fix DelayScheduler quiescent shutdown; misc tweaks
> - Merge remote-tracking branch 'refs/remotes/origin/JDK-8373118' into
> JDK-8373118
> - Merge branch 'openjdk:master' into JDK-8373118
> - Merge branch 'openjdk:master' into JDK-8373118
> - Initial version of delegated lock redesign
> - reenable submissionQueue bypass
> - Improve performance under oversubscription, step 1
> - Reduce reliance on noUserHelp
> - ... and 70 more: https://git.openjdk.org/jdk/compare/88bdbb78...f2842dcd
src/java.base/share/classes/java/util/concurrent/DelayScheduler.java line 193:
> 191: outer: for (long triggerTime = Long.MIN_VALUE;;) {
> 192: Thread.interrupted(); // clear status
> 193: long d = (triggerTime == Long.MAX_VALUE) ? 0L : triggerTime
> - now();
This will always go to the else-branch on the first run of the loop, might be
"cheaper" to initialize `d` to Long.MIN_VALUE - now() in the
for-loop-initializer and then recalculate it after `process()`?
src/java.base/share/classes/java/util/concurrent/DelayScheduler.java line 194:
> 192: Thread.interrupted(); // clear status
> 193: long d = (triggerTime == Long.MAX_VALUE) ? 0L : triggerTime
> - now();
> 194: triggerTime = Long.MIN_VALUE;
Is the intent setting this to MIN_VALUE that if process() throws then the
finally-block breaks? If so, it might be better to move the write to just
before the call to `process()` and add a comment to indicate this.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/28797#discussion_r2994107022
PR Review Comment: https://git.openjdk.org/jdk/pull/28797#discussion_r2994095264
