moving jtreg forward

[Seán Coffey]

No problems here either. I would say that the Quality group is impacted 
most by such a change. Best to check that they're in agreement with such 
changes.

regards,
Sean.

On 09/12/2014 00:38, Jonathan Gibbons wrote:
> It hasn't happened yet.
>
> But there is interesting in doing a bunch of improvements to jtreg 
> innards, especially timeout handling, which would benefit from being 
> able to use JDK 8 API.
>
> -- Jon
>
> On 12/08/2014 04:34 PM, Rob McKenna wrote:
>> We may sometimes test on the platforms that we use to build. In some 
>> cases it may not be possible to run 8 on those hosts, I don't have an 
>> objection to simply using an older version of jtreg in those 
>> instances however.
>>
>> Out of curiosity, what prompted the move to 8?
>>
>>     -Rob
>>
>> On 09/12/14 00:17, Jonathan Gibbons wrote:
>>> Sean, Rob,
>>>
>>> We're looking at advancing jtreg so that it will require JDK 8 (or 
>>> better) to run jtreg itself, although it will still be able to test 
>>> as far back as we want to go -- meaning JDK 5 or so. For refer4ence, 
>>> jtreg currently only requires JDK 5 or better.
>>>
>>> One consequence is that we would not be able to use -samevm mode for 
>>> anything less than JDK 8, but -agentvm should provide a satisfactory 
>>> replacement.
>>>
>>> Do you see any problems from the Sustaining side of the organization 
>>> if we require JDK 8 or better to be used to run jtreg?
>>>
>>> -- Jon
>>
>