RFR 8150521: Test test/java/net/InetAddress/getOriginalHostName.java fails to load JavaNetInetAddressAccess from SharedSecrets

[Chris Hegarty]

> On 2 Mar 2016, at 07:44, vyom <vyom.tewari at oracle.com> wrote:
> 
> Hi Chris,
> 
> Thanks for review please find the updated webrev(http://cr.openjdk.java.net/~nkumar/vyom/8150521/webrev0.1/index.html <http://cr.openjdk.java.net/%7Enkumar/vyom/8150521/webrev0.1/index.html>).

Looks good to me Vyom.

-Chris.

> Thanks,
> Vyom
> 
>> On Tuesday 01 March 2016 03:35 PM, Chris Hegarty wrote:
>>> On 1 Mar 2016, at 09:18, vyom <vyom.tewari at oracle.com> wrote:
>>> 
>>> Hi All,
>>> 
>>> Please  review my code changes for testcase fix.
>>> 
>>> Bug: JDK: Test test/java/net/InetAddress/getOriginalHostName.java fails to load JavaNetInetAddressAccess from         SharedSecrets
>>> 
>>> Webrev: http://cr.openjdk.java.net/~nkumar/vyom/8150521/webrev0.0/index.html <http://cr.openjdk.java.net/%7Enkumar/vyom/8150521/webrev0.0/index.html>
>>> 
>>> This code changes ensures that  InetAdress class is loaded before we call "SharedSecrets.getJavaNetInetAddressAccess()”.
>> I wonder if the test failure is exposing an actual product bug?
>> Should SharedSecrets.getJavaNetInetAddressAccess ensure that the
>> InetAddress class is initialized before returning? This is a common
>> pattern in most of the SS.getXXX methods, and I think it appropriate
>> here, given its usage.
>> 
>> -Chris.
>