RFR: 8138920: Refactor the sampling thread from ConcurrentG1RefineThread

Bengt Rutisson bengt.rutisson at oracle.com
Mon Oct 26 08:12:08 UTC 2015


Hi Derek,

Modulo Per's comments I'm fine with this change as well.

Just like Per, I don't need to see an updated webrev if you want to go 
ahead and push.

Thanks,
Bengt

On 2015-10-26 08:09, Per Liden wrote:
> Hi Derek,
>
> On 2015-10-23 19:43, Derek White wrote:
>> RFR 5.
>>
>> This responds to Per's last comments and Thomas's suggestions for better
>> methods names and comments.
>>
>> Webrev: http://cr.openjdk.java.net/~drwhite/8138920/webrev.05/
>> Webrev 05 vs 04:
>> http://cr.openjdk.java.net/~drwhite/8138920/webrev.05.vs.04/
>>
>> RFE: JDK-8138920 <https://bugs.openjdk.java.net/browse/JDK-8138920>
>> Refactor the sampling thread from ConcurrentG1RefineThread
>
> The patch looks good, but it looks like you forgot to address the 
> following two comments.
>
> [...]
>>>>> g1YoungListSampleThread.cpp
>>>>> ---------------------------
>>>>>
>>>>>   59   _monitor = new Monitor(Mutex::nonleaf,
>>>>>   60                          "Sample thread monitor",
>>>>>   61                          true,
>>>>>   62 Monitor::_safepoint_check_never);
>>>>>
>>>>> It looks like _monitor could be a value member instead of a pointer,
>>>>> to avoid the extra allocation here.
>>>>>
>>>>>   66   set_name("G1 Sample");
>>>>>
>>>>> This should probably be updated to reflect the new name.
>
> I don't need to see a new webrev, but please address these before 
> pushing.
>
> cheers,
> /Per




More information about the hotspot-gc-dev mailing list