RFR (S): 8176885: Clear Claimed Marks log message at wrong location in the log

Stefan Johansson stefan.johansson at oracle.com
Mon Mar 27 12:26:38 UTC 2017


Looks good,
Thanks for cleaning this up,
StefanJ

On 2017-03-24 13:30, Thomas Schatzl wrote:
> Hi,
>
>    could someone have a look at this "move some code around" change?
>
> Thanks,
>    Thomas
>
> On Thu, 2017-03-16 at 18:53 +0100, Thomas Schatzl wrote:
>> Hi all,
>>
>>    can I have quick reviews for this change to the order of the "Clear
>> Claimed Marks" log message?
>>
>> Two issues: it's at the wrong location, printed in the "Post Evacuate
>> Collection Set" section as it is executed during the "Evacuate
>> Collection Set" phase, and in the middle of the "(Parallel) Redirty"
>> log output.
>>
>> Second, this task should actually be part of the "Pre Evacuate
>> Collection Set" phase, as the "Evacuate Collection Set" phase should
>> only contain the bare (parallel) evacuation itself.
>>
>> I moved the log message and the actual code to the "pre evacuate
>> collection set" phase.
>>
>> CR:
>> https://bugs.openjdk.java.net/browse/JDK-8176885
>> Webrev:
>> http://cr.openjdk.java.net/~tschatzl/8176885/webrev/
>> Testing:
>> local testing, inspecting log
>>
>> Thanks,
>>    Thomas
>>




More information about the hotspot-gc-dev mailing list