RFR(M): Backports of G1 bugs 8009940, 8008301, 8010463, 8010780, 8012335, and 8012715 to hs24

[Thomas Schatzl]

Hi,

On Thu, 2013-05-02 at 12:04 -0700, John Cuthbertson wrote:
> Hi Everyone,
> 
> Some more backports that did not apply cleanly to review:
> 
> 8009940: G1: assert(_finger == _heap_end) failed,
> concurrentMark.cpp:809
> Backport:
> http://cr.openjdk.java.net/~johnc/hsx24-backports/webrev.2.8009940.hsx24.patch/
> Original: http://cr.openjdk.java.net/~johnc/8009940/webrev.1/
> 
> The changes did not apply cleanly because of CMTask::_worker_id being
> renamed to CMTask::_task_id.
> 
> Note these changes are generated w.r.t the backports for 8005032 and
> 8009536 being applied.

Good.
> 
> 8010463: G1: Crashes with -UseTLAB and heap verification
> Backport:
> http://cr.openjdk.java.net/~johnc/hsx24-backports/webrev.4.8010463.hsx24.patch/
> Original: http://cr.openjdk.java.net/~johnc/8010463/webrev.0/
> 
> I'm not sure why this did not apply cleanly but the rejected chunk was
> in g1CollectedHeap.cpp.

Looks okay.

> 
> 8010780: G1: Eden occupancy/capacity output wrong after a full GC
> Backport:
> http://cr.openjdk.java.net/~johnc/hsx24-backports/webrev.5.8010780.hsx24.patch/
> Original: http://cr.openjdk.java.net/~johnc/8010780/webrev.0/
> 
> The changes (especially in g1CollectedHeap.cpp) conflicted with the
> tracing changes.

Seem all good.

Hth,
Thomas