RFR: 8234131: Miscellaneous changes imported from jsr166 CVS 2020-12 [v3]

[Martin Buchholz]

Thanks, Pavel!

I committed this to CVS, but am not planning to include it in this
integration:
(not sure if re-generating will invalidate my "Ready" status)
(Having to cast to Future makes this more problematic, but in typical
usages the user will know that the queue contains Futures)

--- src/main/java/util/concurrent/ThreadPoolExecutor.java 8 Dec 2020
20:33:21 -0000 1.195
+++ src/main/java/util/concurrent/ThreadPoolExecutor.java 9 Dec 2020
22:05:21 -0000 1.197
@@ -2066,10 +2066,12 @@
      * <pre> {@code
      * new RejectedExecutionHandler() {
      *   public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
-     *     Future<?> dropped = e.getQueue().poll();
-     *     if (dropped != null)
-     *        dropped.cancel(false); // also consider logging the failure
-     *     e.execute(r);             // retry
+     *     Runnable dropped = e.getQueue().poll();
+     *     if (dropped instanceof Future<?>) {
+     *       ((Future<?>)dropped).cancel(false);
+     *       // also consider logging the failure
+     *     }
+     *     e.execute(r);  // retry
      * }}}</pre>
      */
     public static class DiscardOldestPolicy implements
RejectedExecutionHandler {

On Wed, Dec 9, 2020 at 10:46 AM Pavel Rappo <prappo at openjdk.java.net> wrote:

> On Tue, 8 Dec 2020 21:15:48 GMT, Martin Buchholz <martin at openjdk.org>
> wrote:
>
> >> 8234131: Miscellaneous changes imported from jsr166 CVS 2020-12
> >
> > Martin Buchholz has updated the pull request with a new target base due
> to a merge or a rebase. The pull request now contains one commit:
> >
> >   JDK-8234131
>
> The changes to doc comments look good.
>
> Thanks for incorporating my earlier code snippet suggestions published on
> [concurrency-interest](
> http://cs.oswego.edu/pipermail/concurrency-interest/2020-November/017264.html
> )!
>
> src/java.base/share/classes/java/util/concurrent/ThreadPoolExecutor.java
> line 2102:
>
> > 2100:      *        dropped.cancel(false); // also consider logging the
> failure
> > 2101:      *     e.execute(r);             // retry
> > 2102:      * }}}</pre>
>
> I tried to reify that code snippet:
>
> $ cat MySnippet.java
> import java.util.concurrent.Future;
> import java.util.concurrent.RejectedExecutionHandler;
> import java.util.concurrent.ThreadPoolExecutor;
>
> public class MySnippet {
>     public static void main(String[] args) {
>         new RejectedExecutionHandler() {
>             public void rejectedExecution(Runnable r, ThreadPoolExecutor
> e) {
>                 Future<?> dropped = e.getQueue().poll();
>                 if (dropped != null)
>                     dropped.cancel(false); // also consider logging the
> failure
>                 e.execute(r);              // retry
>             }
>         };
>     }
> }
> $ javac MySnippet.java
> MySnippet.java:9: error: incompatible types: Runnable cannot be converted
> to Future<?>
>                 Future<?> dropped = e.getQueue().poll();
>                                                      ^
> 1 error
> $
> Separately, it seems that the `if` statement uses a 3-space indent which
> is surprising to see in the JSR 166 code base.
>
> -------------
>
> Marked as reviewed by prappo (Reviewer).
>
> PR: https://git.openjdk.java.net/jdk/pull/1647
>