RFR: 8166276: Refactor gen_process_roots to allow simpler fix for 8165949

[Stefan Johansson]

On 2016-09-20 14:28, Stefan Johansson wrote:
> Thanks for reviewing Mikael,
>
> On 2016-09-20 14:11, Mikael Gerdin wrote:
>> Hi Stefan,
>>
>>
>> On 2016-09-20 11:08, Stefan Johansson wrote:
>>> Hi all,
>>>
>>> While working on JDK-8165949 [0] I realized that gen_process_roots
>>> needed to be cleaned/split up to allow it to be fixed. This work had in
>>> big parts already been done as part of an enhancement [1] that never 
>>> got
>>> pushed. That enhancement included another cleanup as well but I have
>>> split it out to only include the necessary part in this fix. It splits
>>> gen_process_roots into two different versions, one used for young
>>> collections and one for old. This make the two methods much easier to
>>> follow and removes some special cases. I will post a second review for
>>> the actual bug-fix.
>>>
>>> Webrev:
>>> http://cr.openjdk.java.net/~sjohanss/8166276/hotspot.00/
>>
>> Overall I think this looks good but it might be a good idea to have a 
>> look at the comment just above the definition of enum ScanningOption, 
>> it describes concepts which you are removing (such as not_older_gens)
>>
> Good catch, removed parts of it and move the other part down to 
> old_process_roots:
> Full: http://wikifiles.se.oracle.com/dev/sjohanss/8166276/hotspot.01
> Inc: http://wikifiles.se.oracle.com/dev/sjohanss/8166276/hotspot.00-01/
Correct links:
http://cr.openjdk.java.net/~sjohanss/8166276/hotspot.01/
http://cr.openjdk.java.net/~sjohanss/8166276/hotspot.00-01/

Stefan
>
> Thanks,
> Stefan
>
>> /Mikael
>>
>>>
>>> Subtask:
>>> https://bugs.openjdk.java.net/browse/JDK-8166276
>>>
>>> Testing:
>>> * JTREG tests locally
>>> * RBT on HS-gc tier 2-4
>>>
>>> Thanks,
>>> Stefan
>>>
>>> [0] https://bugs.openjdk.java.net/browse/JDK-8165949
>>> [1] https://bugs.openjdk.java.net/browse/JDK-8072809
>