RFR(XS): 8025519: [TESTBUG] runtime/7107135/Test7107135.sh need to use COMPILEJAVA for javac task

Mikhailo Seledtsov mikhailo.seledtsov at oracle.com
Fri Feb 21 22:09:01 UTC 2014


Looks good.

Misha

On 2/21/2014 2:55 PM, Calvin Cheung wrote:
> I've modified test/TEST.groups so that all 3 tests are now under the 
> "compact1" group.
> I re-ran jtreg on those tests and they're working fine. (Note: I 
> needed to temporary remove the @ignore tag in the Test8017498.sh so 
> that I can verify the fix.)
>
> updated webrev: http://cr.openjdk.java.net/~ccheung/8025519/webrev/
>
> Calvin
>
> On 2/21/2014 11:16 AM, Mikhailo Seledtsov wrote:
>> David,
>>
>>  Thank you for this bit of knowledge. I was not sure about this, so I 
>> thought I raise a question.
>>
>> Misha
>>
>> On 2/21/2014 2:31 AM, David Holmes wrote:
>>> Hi Misha,
>>>
>>> On 21/02/2014 12:17 AM, Mikhailo Seledtsov wrote:
>>>> Hi Calvin,
>>>>
>>>>   The changes look good.
>>>> I also checked test/TEST.groups to make sure that the tests are under
>>>> "needs_jdk" category; two of them are.
>>>> However, the Test6929067.sh is under needs_full_vm_compact1. I 
>>>> wonder if
>>>> it should be moved to the "needs_jdk" category? *
>>>
>>> If the tests use COMPILEJAVA instead of TESTJAVA they no longer need 
>>> to be in the "needs_jdk" group. (assuming javac is all they were using)
>>>
>>> David
>>>
>>>> *Misha*
>>>>
>>>>
>>>> *
>>>> On 2/19/2014 7:29 PM, Calvin Cheung wrote:
>>>>> Some of the test shell scripts are using the TESTJAVA env variable
>>>>> which may point to a JRE, not to a JDK which contains the required
>>>>> tools such as javac and header files. This fix is to change TESTJAVA
>>>>> to COMPILEJAVA in 3 test shell scripts which have the problem.
>>>>>
>>>>> bug: https://bugs.openjdk.java.net/browse/JDK-8025519
>>>>>
>>>>> webrev: http://cr.openjdk.java.net/~ccheung/8025519/webrev/
>>>>>
>>>>> tests: (on linux-x64)
>>>>>     manually run the affected test cases
>>>>>     using jtreg to run the affected test cases
>>>>>
>>>>> thanks,
>>>>> Calvin
>>>>
>>
>



More information about the hotspot-runtime-dev mailing list