RFR (S) 8140650: Method::is_accessor should cover getters and setters for all types

Aleksey Shipilev aleksey.shipilev at oracle.com
Tue Nov 10 22:26:14 UTC 2015

Thanks for re-reviews, Coleen, Vladimir and Severin!

Pushing this through hs-comp now.


On 11/09/2015 11:28 PM, Coleen Phillimore wrote:
> On 11/9/15 1:51 PM, Aleksey Shipilev wrote:
>> Hi Severin,
>> On 11/09/2015 08:57 PM, Severin Gehwolf wrote:
>>> Here is a partial implementation of the Zero bits. In fact, the code
>>> was already there to handle other getters too. I don't know why it was
>>> guarded with those asserts and with is_simple_accessor().
>>> It does not yet implement the setter logic, but at least we don't have
>>> to keep is_simple_accessor() around. The patch builds on top of your
>>> latest 8140650/webrev.02/ webrev.
>>> The patch I'm talking about is:
>>> http://cr.openjdk.java.net/~sgehwolf/webrevs/JDK-8140650/JDK-8140650-is_accessor_zero_cleanup.hotspot.patch
>> Excellent, thanks! Incorporated:
>>    http://cr.openjdk.java.net/~shade/8140650/webrev.03/
>> Coleen, are you still good with these changes?
> Yes, this looks fine.
> I can't see how there's any value performance-wise to implement the
> is_setter entry point in Zero, that would be worth writing the code for
> the setter entry point.
> Coleen
>> The RBT tests (hotspot_all, vm.{gc|regression|runtime}.testlist, and
>> some others) came clean for a previous version, and this version touches
>> only Zero code, so I am inclined to think this change does not need
>> re-testing.
>> Thanks,
>> -Aleksey

More information about the hotspot-dev mailing list