RFR (S) 8178336: Unnecessary SystemDictionary walk for Protection domain liveness

[coleen.phillimore at oracle.com]

On 4/11/17 2:05 AM, David Holmes wrote:
> Hi Coleen,
>
> On 11/04/2017 6:18 AM, coleen.phillimore at oracle.com wrote:
>> Summary: remove system dictionary walk and pass strong closure for
>> !ClassUnloading
>
> Can't really comment on functional change but noticed this comment:
>
> 241   // Then iterate over the protection domain cache to apply the 
> closure on the
>  242   // previously marked ones.
>  243   _pd_cache_table->oops_do(blk);

Thanks - I tried to follow and update all the comments, but apparently 
missed some:

   // Follow oops that are in the protection domain cache table

This is what I changed it to.
Coleen
>
> is no longer accurate: there is nothing "before" for the "then" to 
> follow, and there is no previous marking done.
>
> Thanks,
> David
>
>> See bug for more details:
>>
>> open webrev at http://cr.openjdk.java.net/~coleenp/8178336.01/webrev
>> bug link https://bugs.openjdk.java.net/browse/JDK-8178336
>>
>> Tested with nightly tier2-5 tests and jprt (runs all GCs) and runThese
>> with -XX:-ClassUnloading.
>>
>> Thanks,
>> Coleen