CMS parallel initial mark; vm thread sneaking in Thread::try_lock()

[Thomas Schatzl]

Hi Hiroshi,

On Tue, 2013-06-25 at 19:05 -0700, Hiroshi Yamauchi wrote:
> Here's an updated
> webrev: http://cr.openjdk.java.net/~hiroshi/webrevs/edenchunks/webrev.01/
>
> Let me know if there are more comments.

Minor nitpick:

concurrentMarkSweepGeneration.cpp:

4516   if (_eden_chunk_array && !CMSEdenChunksRecordAlways) {

The != NULL has been forgotten here.

Thanks for considering to use the mutex in
CMSCollector::sample_eden_chunk(). It looks a lot cleaner now.

Looks good.

Thanks,
  Thomas