Merge

8134631: G1DummyRegionsPerGC fires assert of assert(words <= filler_array_max_size()) failed: too big for a single object Summary: Change filler max value temporarily for G1DummyRegionsPerGC flag. Reviewed-by: tbenson, tschatzl

8139424: SIGSEGV, Problematic frame: # V [libjvm.so+0xd0c0cc] void InstanceKlass::oop_oop_iterate_oop_maps_specialized<true,oopDesc*,MarkAndPushClosure> Summary: The crash was caused by a faulty eager humongous reclaim. The reason for reclaiming a live object was that the call to cleanupHRRS was done after dirtying cards and clearing the remembered sets for the humongous object. This could lead to one or many cards being missed. Reviewed-by: tbenson, kbarrett, tschatzl

8138684: G1 decision about taking regions into the collection set is too aggressive Summary: Factor in expected waste and uncertainty of our guess in the decision whether to take another region into the collection set. Reviewed-by: mgerdin, jmasa

8141356: Explicitly stop CMS threads during VM termination. Summary: GenCollectedHeap::stop() is not implemented which is supposed to stop CMS threads during VM termination. Reviewed-by: jmasa, kbarrett Contributed-by: jwha@google.com

8142403: Make G1CollectorPolicy::predictor const Reviewed-by: stefank, tschatzl

8139883: Add virtual destructor G1ParScanThreadState Reviewed-by: tschatzl, mgerdin, stefank

8142475: Remove TraceParallelOldGCTasks Reviewed-by: tschatzl, david

8142472: Remove debugging code guarded by CMSPrintPromoBlockInfo Reviewed-by: david, ehelin

Merge