[7u-dev] Request for approval for CR 7132378: Race in FutureTask if used with explicit set and get ( not Runnable )

[Martin Buchholz]

As always, I would like to see more timely lower-overhead integration of
jsr166 CVS with jdk8 and jdk7u.
At Google, we have chosen to use the jsr166 version of FutureTask with
jdk7, as well as to use jsr166 versions of

./src/jdk7/java/util/concurrent/ScheduledThreadPoolExecutor.java
./src/jdk7/java/util/concurrent/DelayQueue.java
./src/jdk7/java/util/concurrent/ThreadPoolExecutor.java
./src/test/jtreg/util/concurrent/ScheduledThreadPoolExecutor/GCRetention.java

but our primary motivation is squashing unwanted memory retention.



On Thu, Jun 6, 2013 at 3:43 AM, Aleksey Shipilev <
aleksey.shipilev at oracle.com> wrote:

> Hi,
>
> Could you please approve the backport of CR 7132378 to 7u?
>
> External bug:
>   http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7132378
>
> The fix was in JDK 8 for more than a year already:
>   http://hg.openjdk.java.net/jdk8/tl/jdk/rev/f9fb8c4b4550
>
> Testing:
>  - "hg import" applies cleanly to jdk7u
>  - Linux x86_64 build is OK
>  - Linux x86_64 passes test/java/util/concurrent/FutureTask jtreg
>
> -Aleksey.
>