RFR (T): 8213997: Remove G1HRRSUseSparseTable flag

[Thomas Schatzl]

Hi,

On Tue, 2018-11-20 at 21:37 -0800, Man Cao wrote:
> Looks good to me, too. (I'm not a reviewer, though.)

  thanks for your review.

And btw, you are a reviewer (everyone who cares to look at changes is),
maybe not a "R"eviewer (yet). :)

Thanks,
  Thomas

> 
> -Man
> 
> On Fri, Nov 16, 2018 at 6:10 AM Thomas Schatzl
> <thomas.schatzl at oracle.com> wrote:
> > 
> > Hi all,
> > 
> >   can I have reviews to remove the G1HRRSUseSparseTable develop
> > flag?
> > 
> > It has never been used to my recollection, and if so only for doing
> > performance comparisons.
> > 
> > However in that case you need a modified build anyway as it is a
> > develop flag, i.e. you already need to modify the code.
> > 
> > Even if you disable Sparse PRTs you still pay some memory costs, so
> > it
> > does not allow easy memory comparisons anyway.
> > 
> > From performance/memory consumption POV sparse PRTs are always a
> > win,
> > in some cases we even recommend to let G1 store more remembered set
> > entries in sparse PRTs to save memory (without otherwise adverse
> > effects).
> > 
> > CR:
> > https://bugs.openjdk.java.net/browse/JDK-8213997
> > Webrev:
> > http://cr.openjdk.java.net/~tschatzl/8213997/webrev/
> > Testing:
> > local compilation
> > 
> > I consider this change trivial as it consists of code deletion and
> > some
> > reformatting only.
> > 
> > Thanks,
> >   Thomas
> > 
> >