RFR (S): Reset bitmaps in MarkCompact should use parallel threads

[Roman Kennke]

Am Mittwoch, den 02.11.2016, 20:18 +0100 schrieb Aleksey Shipilev:
> On 11/02/2016 08:09 PM, Aleksey Shipilev wrote:
> > 
> > http://cr.openjdk.java.net/~shade/shenandoah/markcompact-par-reset-
> > bitmaps/webrev.01/
> 
> I have a question! When clearing the prev bitmap, should we go up
> to...
> 
> a) region->top_prev_mark_bitmap(): clearing only the set bits?
> 
> http://cr.openjdk.java.net/~shade/shenandoah/markcompact-par-reset-bi
> tmaps/webrev.01/
> 
> b) region->top(): clearing the entire region map defensively?
> 
> http://cr.openjdk.java.net/~shade/shenandoah/markcompact-par-reset-
> bitmaps/webrev.02/

a) please. Results in much less work. Requires some care to be taken
that the top-marked is correct.

Roman