[7u] review and approval request for 8007515: update RMI tests on ProblemList.txt

[Stuart Marks]

Change pushed. We now have 134 RMI tests passing in jdk7u, with only 4 excluded 
-- same as jdk8!

Regarding the mailing list, we should probably replace jdk-regtest at sun.com with 
an OpenJDK mailing list. The problem is, which one? Core-libs-dev, jdk8-dev, 
jdk7u-dev, discuss, and quality-discuss have all been kicked around, but none 
of these is an obvious match.

s'marks

On 2/12/13 11:13 AM, Darryl Mocek wrote:
> Seems to me we need an Oracle equivalent.  Is there already one for jtreg or
> JDK tests in general?  Not a huge priority, but we should get rid of the
> sun.com email addresses when an easy opportunity arises.
>
> Darryl
>
> On 02/11/2013 06:06 PM, Stuart Marks wrote:
>> Arrgh, ya caught me! :-) I was aware of this issue but I wanted to leave it
>> out of this discussion since it's a potential distraction.
>>
>> No, jdk-regtest at sun.com (mentioned in the TEST.ROOT file) is not a valid
>> email address. At least, it bounced when I tried to send something to it.
>>
>> I did have a sidebar discussion on this with Jon Gibbons and Alan B recently
>> but we didn't come to a conclusion as to what it should be replaced with.
>>
>> This one actually seems like a separate bug, since it needs to be changed in
>> 8 (and backported to 7u) instead of just an adjustment to 7u to match 8.
>>
>> s'marks
>>
>> On 2/11/13 3:15 PM, Darryl Mocek wrote:
>>> Stuart,
>>>
>>>     is jdk-regtest at sun.com still a valid email address?
>>>
>>> Darryl
>>>
>>> On 02/11/2013 03:02 PM, Stuart Marks wrote:
>>>> On 2/9/13 7:49 AM, Alan Bateman wrote:
>>>>> On 09/02/2013 00:51, Stuart Marks wrote:
>>>>>> A simple request for code review *and* integration approval for the
>>>>>> following
>>>>>> bugfix:
>>>>>>
>>>>>> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=8007515
>>>>>>
>>>>>> http://cr.openjdk.java.net/~smarks/reviews/8007515/webrev.0/
>>>>>>
>>>>>> Now that all the RMI test fixes have been backported, this reconciles the
>>>>>> ProblemList.txt file between jdk8 and jdk7u so that the same RMI tests are
>>>>>> excluded on each release. This ends up removing a few dozen RMI tests from
>>>>>> the jdk7u problem list, leaving just four.
>>>>>>
>>>>> This looks fine to me. While you are there, would you mind adding the RMI
>>>>> directories to the othervm.dirs and exclusiveAccess.dirs list in TEST.ROOT?
>>>>> That way folks can run jtreg -agentvm without needing to special-case the RMI
>>>>> tests. I actually think we should just copy TEST.ROOT from jdk8 to jdk7u-dev
>>>>> but that could be another bug as it is not specific to RMI.
>>>>
>>>> I thought about dealing with TEST.ROOT in a separate bug but that just seems
>>>> like busy-work. Updating the problem list is at least as much about making
>>>> sure that the testing configurations between jdk8 and jdk7u are brought
>>>> closer together as much as it is about RMI, so updating TEST.ROOT is part of
>>>> that.
>>>>
>>>> I've refreshed jdk7u's TEST.ROOT from jdk8's and adjusted the bug accordingly.
>>>>
>>>> Here's the webrev:
>>>>
>>>> http://cr.openjdk.java.net/~smarks/reviews/8007515/webrev.1/
>>>>
>>>> s'marks
>>>
>