TenuringThreshold being adaptive when UseAdaptiveSizePolicy is off

Jungwoo Ha jwha at google.com
Thu Mar 12 22:30:39 UTC 2015


Nice finding. Thank you Wessam.

On Thu, Mar 12, 2015 at 10:49 AM, Wessam Hassanein <wessam at google.com>
wrote:

> Jungwoo, I think the answer to your question of fixing the tenuring
> threshold in CMS for jdk8 is to set the TargetSurvivorRatio to 100. This
> will result in always using the MaxTenuringThreshold as the tenuring
> threshold and the desired_survivor_size as the full survivor_capacity
> which would match ParallelScavange with UseAdaptiveSizePolicy disabled.
>
> Regards,
>
> Wessam
>
> On Tue, Mar 10, 2015 at 11:37 AM, Jon Masamitsu <jon.masamitsu at oracle.com>
> wrote:
>
>>
>> On 3/10/2015 4:59 AM, Jungwoo Ha wrote:
>>
>> Tao,
>>
>>  CMS no longer use InitialTenuringThreshold.
>> IIRC, even before when CMS used InitialTenuringThreshold, setting initial
>> and max the same value didn't turn off the adaptiveness.
>>
>>
>> I think we're removing that  code from CMS in jdk9
>>
>> compute_survivor_space_size_and_threshold()
>>
>> The implementation of adaptive size policy for CMS never
>> worked correctly.  Too many times it does the exact wrong
>> thing.
>>
>> Jon
>>
>>
>>
>>
>> On Tue, Mar 10, 2015 at 12:23 AM, Tao Mao <yiyeguhu at gmail.com> wrote:
>>
>>> Hi Jungwoo,
>>>
>>>  Would fixing InitialTenuringThreshold and MaxTenuringThreshold to the
>>> same value work for you?
>>>
>>>  Thanks.
>>>  Tao Mao
>>>
>>>
>>> On Mon, Mar 9, 2015 at 10:39 PM, Jungwoo Ha <jwha at google.com> wrote:
>>>
>>>> We were investigating some features and figured out that
>>>> ParallelScavange and ParNew has different TenuringTreshold heuristics.
>>>> We thought it would be nice to have a flag that can control the
>>>> adaptiveness, so that we can fix it on purpose to isolate other performance
>>>> factors.
>>>>  --Jungwoo
>>>>
>>>> On Mon, Mar 9, 2015 at 11:27 AM, Jon Masamitsu <
>>>> jon.masamitsu at oracle.com> wrote:
>>>>
>>>>>  Jungwoo,
>>>>>
>>>>> On 03/06/2015 06:03 PM, Jungwoo Ha wrote:
>>>>>
>>>>> I think it is the same for all JDK{7,8,9}.
>>>>>
>>>>>  arguments.cpp disabled UseAdaptiveSizePolicy for CMS.
>>>>> parallel scavange initializes _tenuring_threshold with
>>>>>   _tenuring_threshold = (UseAdaptiveSizePolicy) ?
>>>>> InitialTenuringThreshold : MaxTenuringThreshold;
>>>>> whereas CMS uses
>>>>>   _tenuring_threshold = MaxTenuringThreshold;
>>>>>
>>>>>  adjust_desired_tenuring_threshold() is called inside
>>>>> ParNewGeneration::collect(), which calls compute_tenuring_threshold.
>>>>> It only uses MaxTenuringThreshold and TargetSurvivorRatio.
>>>>> I don't see any other flags here.
>>>>>
>>>>>
>>>>>  Yes, this is the intended behavior.
>>>>>
>>>>> You're saying we're sometimes wasting survivor space?
>>>>>
>>>>> Jon
>>>>>
>>>>>
>>>>> On Fri, Mar 6, 2015 at 4:28 PM, Jon Masamitsu <
>>>>> jon.masamitsu at oracle.com> wrote:
>>>>>
>>>>>> This is in jdk9?
>>>>>>
>>>>>> Jon
>>>>>>
>>>>>>
>>>>>> On 3/6/2015 4:21 PM, Jungwoo Ha wrote:
>>>>>>
>>>>>>> On CMS when UseAdaptiveSizePolicy is off, still tenuring_threshold
>>>>>>> is being changed.
>>>>>>> Is this an intended behavior?
>>>>>>> (yes. I know it is not "size" related heuristics)
>>>>>>>
>>>>>>> It would be nice to have a flag that can disable adjusting tenuring
>>>>>>> threshold,
>>>>>>> or at least set minimum threshold. When the default heuristics
>>>>>>> doesn't work,
>>>>>>> there are nothing much we can do other than changing the Max.
>>>>>>>
>>>>>>> --Jungwoo
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>
>>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.java.net/pipermail/hotspot-gc-dev/attachments/20150312/0f5a062a/attachment.htm>


More information about the hotspot-gc-dev mailing list