RFA 8056248: Improve ForkJoin thread throttling

[Seán Coffey]

On 05/09/14 11:15, Paul Sandoz wrote:
> On Sep 5, 2014, at 12:09 PM, Seán Coffey <sean.coffey at oracle.com> wrote:
>
>> I don't see CCC approval for an 8u backport either. Please ensure that's been completed also.
>>
> Does it need a CCC if the property is not specified?
Are you saying that you're going to pull the 
java.util.concurrent.ForkJoinPool.common.maximumSpare property out or 
make it implementation specific ? If you're pulling it out, then CCC 
shouldn't be required. If you're making it impl specific, my 
understanding is that all new system properties should be documented via 
CCC (even if implementation specific). Joe or Alan might be able to 
correct me here.

regards,
Sean.

>
> Paul.
>
>> regards,
>> Sean.
>>
>> On 05/09/14 10:54, Alan Bateman wrote:
>>> On 05/09/2014 10:35, Paul Sandoz wrote:
>>>> Hi,
>>>>
>>>> I would like to backport the improvements to the fork join framework from 9 to 8u40:
>>>>
>>>> http://mail.openjdk.java.net/pipermail/core-libs-dev/2014-August/028419.html
>>>>
>>>>    https://bugs.openjdk.java.net/browse/JDK-8056248
>>>>
>>>> http://cr.openjdk.java.net/~psandoz/jdk9/JDK-8056248-fj-thread-throttling/webrev/
>>>>
>>>>    http://hg.openjdk.java.net/jdk9/dev/jdk/rev/7af64e3e095d
>>>>
>>>> The changeset (when shuffled) imports cleanly into jdk8u-dev/jdk.
>>>>
>>>> However, it also introduced a new system property ("java.util.concurrent.ForkJoinPool.common.maximumSpares"). I am not sure if we can add such a new feature to an 8u release, any advice?
>>> It updates the spec so I assume would require a MR if you want this to update the Java SE 8 spec.
>>>
>>> Maybe the simplest for 8u is to put this under an @implNote so that it's clear to anyone generating the javadoc for 8u that this property is JDK-implementation specific (at least in in 8u).
>>>
>>> -Alan.