RFR (S/M): 8159440: Move marking of promoted objects during initial mark into the concurrent phas

[Thomas Schatzl]

Hi,

On Mon, 2018-12-03 at 15:04 +0100, Stefan Johansson wrote:
> Hi Thomas,
> 
> Nice clean up, looks good in general. Just one minor nit.
> 
> On 2018-11-29 11:23, Thomas Schatzl wrote:
> > Hi all,
> > 
> >    can I have reviews for this cleanup change that unifies how we
> > handle "roots" from survivor and old gen regions?
> > 
> > [...]
> >
> > This change makes handling of the promoted objects (from ntams to
> > top) equal to survivor regions.
> > 
> > For this reason we extend the existing G1RootRegions data structure
> > to remember the regions (both survivor and promoted-into old gen
> > regions). That saves quite a bit of code.
> > 
> > CR:
> > https://bugs.openjdk.java.net/browse/JDK-8159440
> > Webrev:
> > http://cr.openjdk.java.net/~tschatzl/8159440/webrev/
> 
> src/hotspot/share/gc/g1/g1ConcurrentMark.hpp
> --------------------------------------------
>   239   uint _max_regions;
>   240   volatile size_t _cur_regions;
>   ...
>   245   volatile size_t _claimed_root_regions;
> 
> I would like these names to feel more connected. Maybe rename 
> _cur_regions to _used_root_regions or _num_root_regions (as the
> getter).  Is there any reason _max_regions can't be size_t as well to
> be even more consistent?

I moved _claimed_root_regions up a bit, did the renaming to
_num_root_regions. Also made _max_regions size_t.

http://cr.openjdk.java.net/~tschatzl/8159440/webrev.0_to_1/ (diff)
http://cr.openjdk.java.net/~tschatzl/8159440/webrev.1/ (full)

Thanks,
  Thomas