jprt build failures on windows

[Chris Hegarty]

There was a cut'n'paste error in the original change for CR 7098719. 
super.create(stream) should be reinstated.

After the above, now the (closed) regression test amended for CR 7098719 
fails. The reason is that TwoStacksPlainSocketImpl overrides the close() 
method, from AbstractPlainSocketImpl, and does not decrements the 
ResourceManagers count. This problem existed before CR 7098719, it is 
just now being seen as the amended test catches this situation. The 
solution is to simply invoke ResourceManager.afterUdpClose() for 
non-stream sockets, just as is done in AbstractPlainSocketImpl.close().

Webrev for CR 7099488:
   http://cr.openjdk.java.net/~chegar/7099488/webrev.00/webrev/

Thanks,
-Chris.

On 10/11/11 10:36 AM, Chris Hegarty wrote:
> Thanks Max,
>
> I filed the following for this:
> CR 7099488: "TwoStacksPlainSocketImpl should invoke
> super.create(stream), typo in fix for 7098719"
>
> Too many changes in too many repos!
>
> -Chris.
>
> On 10/11/11 06:32 AM, Weijun Wang wrote:
>> http://prt-web.us.oracle.com//archive/2011/10/2011-10-11-051159.ww155710.jdk//JobStatus.txt
>>
>>
>>
>> where T is a type-variable:
>> T extends Object declared in class Constructor
>> ..\..\..\src\windows\classes\java\net\TwoStacksPlainSocketImpl.java:85:
>> error: no suitable method found for create()
>> super.create();
>> ^
>> method AbstractPlainSocketImpl.create(boolean) is not applicable
>> (actual and formal argument lists differ in length)
>>
>> Looks like something changed in a recent change
>>
>> http://hg.openjdk.java.net/jdk8/tl/jdk/rev/1e89a13d9d8f
>>
>> - super.create(stream);
>> + try {
>> + super.create();
>>
>> -Max