RFR(M): 8212243: More gc interface tweaks for arraycopy and some other smaller changes in preparation for Shenandoah

Erik Österlund erik.osterlund at oracle.com
Thu Oct 18 15:41:43 UTC 2018

Hi Roland,

On 2018-10-18 17:37, Roland Westrelin wrote:
>> I prefer the first one.
> Then you have to test the type of the C2Access being dealt with. Would
> you do that with virtual methods in C2Access (is_parse_access(),
> is_opt_access()) or throught the _decorators maybe?

Yes. As for how to generate the RAW access inside of BarrierSetC2, I 
would prefer to check if it is_parse_access etc as you say.
As for the semantics, using decorators would make sense if there is some 
reasonable semantic property we can catch.
To determine that, I wonder exactly what barriers you need here at the 
write. All I know is what barriers you do not know (no SATB, no 


> Roland.

More information about the hotspot-compiler-dev mailing list