RFR: 8254015: copy_to_survivor_space should use in-hand klass for scanning

Kim Barrett kim.barrett at oracle.com
Tue Oct 6 13:47:01 UTC 2020


> On Oct 5, 2020, at 9:33 AM, Albert Mingkun Yang <ayang at openjdk.java.net> wrote:
> 
> Does it make sense to drop `oop_iterate_backwards(OopClosureType* cl)` and only have the method with two args so that
> klass will never be re-fetched?

I don't think so. Not all uses both already have the klass in hand and are
performance critical. Though there are very few uses of this function.

I don't think so. Not all uses both already have the klass in hand and are
performance critical to care. Though there are very few uses of this
function, the other oop_iterate variants probably ought to have similar
treatment. I was planning to file an RFE to add klass-taking overloads for
them. When you include those I think the convenience for not requiring the
klass argument is more obvious.





More information about the hotspot-gc-dev mailing list