RFR 8225582: Shenandoah: Enable concurrent evacuation of JNIHandles

Roman Kennke rkennke at redhat.com
Tue Jun 18 21:00:37 UTC 2019


> Thanks for reviewing.
> 
> On 6/18/19 4:18 PM, Aleksey Shipilev wrote:
>> On 6/17/19 9:15 PM, Zhengyu Gu wrote:
>>> Narrowed CR to only evacuating JNIHandles.
>>>
>>> Updated Webrev: http://cr.openjdk.java.net/~zgu/JDK-8225582/webrev.02/
>>
>> I know you have updated patch in works, but here are the things that
>> caught my eye in current patch.
>>
>> *) Do we really need separate shenandoahConcurrentRoots.*? I.e. is
>> there an additional value over
>> keeping those deciders in ShenandoahHeap, given they are accessing
>> ShenandoahHeap anyway?
> 
> I would prefer the separation. ShenandoahConcurrentRoots is an AllStatic
> class, and the decision is not solely based on heap, but some other
> factors, especially when concurrent class unloading comes in.

Doesn't it also help back-portability of this code?

Roman

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://mail.openjdk.java.net/pipermail/hotspot-gc-dev/attachments/20190618/1593529b/signature.asc>


More information about the hotspot-gc-dev mailing list