RFR (XS): Run SPECjbb in headless mode in JPRT
Vladimir Kozlov
vladimir.kozlov at oracle.com
Tue May 29 15:24:07 PDT 2012
We usually don't add Bug ID into code change. I think the comment could be
general: "Prevent popups during testing".
Vladimir
Mikael Vidstedt wrote:
>
> New webrev below. The list of targets now matches the linux x64 list
> (apart from the platform name that is).
>
> http://cr.openjdk.java.net/~mikael/7155453/webrev.02/
>
> Can I get a couple of Reviews and a sponsor?
>
> Thanks,
> Mikael
>
> On 2012-05-23 17:27, Vladimir Kozlov wrote:
>> You mean missing "comma and backslash"?:
>>
>> ${jprt.my.macosx.x64}-{product|fastdebug}-c2-GCOld_ParOldGC, \
>>
>> Vladimir
>>
>> Mikael Vidstedt wrote:
>>>
>>> After addressing the missing backslash it now passes through JPRT (on
>>> all platforms).
>>>
>>> Cheers,
>>> Mikael
>>>
>>> On 2012-05-22 17:20, Vladimir Kozlov wrote:
>>>> Did you run it through JPRT testing?
>>>>
>>>> Vladimir
>>>>
>>>> Mikael Vidstedt wrote:
>>>>>
>>>>> Thanks for the review. I updated based on your feedback.
>>>>>
>>>>> This new webrev also re-enables the JBB testing on OSX:
>>>>>
>>>>> http://cr.openjdk.java.net/~mikael/7155453/webrev.01
>>>>>
>>>>> Cheers,
>>>>> Mikael
>>>>>
>>>>> On 2012-05-22 13:24, Daniel D. Daugherty wrote:
>>>>>> On 5/22/12 2:16 PM, Mikael Vidstedt wrote:
>>>>>>>
>>>>>>> All,
>>>>>>>
>>>>>>> I'd like to get a couple of reviews of this change.
>>>>>>>
>>>>>>> We've seen problems with intermittent and hard to debug popups
>>>>>>> when running SPECjbb on lights out OSX machines. To prevent the
>>>>>>> popups from blocking the completion of the tests this change
>>>>>>> makes SPECjbb run in headless mode in JPRT.
>>>>>>>
>>>>>>> http://cr.openjdk.java.net/~mikael/7155453/webrev.00
>>>>>>
>>>>>> Thumbs up.
>>>>>>
>>>>>> make/jprt.properties
>>>>>> typo: "popup:s" -> "popups"
>>>>>>
>>>>>> The comment kind of implies that this is a MacOS X only change,
>>>>>> but it isn't. Maybe add one more line:
>>>>>>
>>>>>> # but the work-around is added to all platforms to be consistent
>>>>>>
>>>>>> Dan
>>>>>>
>>>>>>
>>>>>
>>>
>
More information about the hotspot-runtime-dev
mailing list