Summary: Evaluate and improve object copy time by micro-optimizations and splitting out slow and fast paths aggressively.
Reviewed-by: kbarrett, mgerdin, jmasa
Contributed-by: NTony Printezis &lt;tprintezis@twitter.com&gt;, Thomas Schatzl <thomas.schatzl@oracle.com>

Reviewed-by: brutisso, ehelin

Summary: Stop using save_marks in G1 related code and make setting the replacement field less racy.
Reviewed-by: brutisso, tschatzl

8062063: Usage of UseHugeTLBFS, UseLargePagesInMetaspace and huge SurvivorAlignmentInBytes cause crashes in CMBitMapClosure::do_bit
Summary: Making sure committed memory is cleared when re-committed, even if using large pages.
Reviewed-by: jwilhelm, tschatzl

8062063: Usage of UseHugeTLBFS, UseLargePagesInMetaspace and huge SurvivorAlignmentInBytes cause crashes in CMBitMapClosure::do_bit
Summary: Making sure committed memory is cleared when re-committed, even if using large pages.
Reviewed-by: jwilhelm, tschatzl

Summary: Moved allocation context update from remark to the cleanup phase.
Reviewed-by: mgerdin, jmasa

Summary: Added a G1CollectorPolicyExt where it is possible to extend the class.
Reviewed-by: sjohanss, tschatzl

Reviewed-by: brutisso, tschatzl

Summary: Read _top before _gc_time_stamp in saved_mark_word() with LoadLoad order to ensure we get a consistent view
Reviewed-by: brutisso, dcubed, dholmes, stefank

8062036: ConcurrentMarkThread::slt may be invoked before ConcurrentMarkThread::makeSurrogateLockerThread causing intermittent crashes
Summary: Suppress gc_alot during VM init, improve error for SLT uninitialized.
Reviewed-by: jmasa, brutisso, tschatzl

8064556: G1: ParallelGCThreads=0 may cause assert(!MetadataOnStackMark::has_buffer_for_thread(Thread::current())) failed: Should be empty
Reviewed-by: mgerdin, tschatzl

8058801: G1TraceReclaimDeadHumongousObjectsAtYoungGC only prints humongous object liveness output when there is at least one candidate humongous object
Summary: If G1TraceReclaimDeadHumongousObjectsAtYoungGC is enabled, always print humongous object liveness output.
Reviewed-by: tschatzl
Contributed-by: sangheon.kim@oracle.com

Reviewed-by: mgerdin, coleenp, bdelsart

8052172: Evacuation failure handling in G1 does not evacuate all objects if -XX:-G1DeferredRSUpdate is set
Summary: Remove -XX:-G1DeferredRSUpdate functionality as it is racy. During evacuation failure handling, threads where evacuation failure handling occurred may try to add remembered sets to regions which remembered sets are currently being scanned. The iterator to handle the remembered set scan does not support addition of entries during scan and so may skip valid references.
Reviewed-by: iveresov, brutisso, mgerdin

Summary: Ergonomics updates flags for number of GC threads
Reviewed-by: tschatzl, jwilhelm
Contributed-by: sangheon.kim@oracle.com

Reviewed-by: brutisso, tschatzl

Summary: Enable yc to be forced for stat update.
Reviewed-by: mgerdin, jcoomes

Reviewed-by: tschatzl, stefank

Reviewed-by: brutisso, jwilhelm

Summary: Replace G1CodeRootSet with a Hashtable based implementation, merge Code Root Migration phase into Code Root Scanning
Reviewed-by: jmasa, brutisso, tschatzl

Summary: Splitting out a g1 allocator class to simply specialized allocators which can associate each allocation with a given context.
Reviewed-by: mgerdin, brutisso

8055635: Missing include in g1RegionToSpaceMapper.hpp results in unresolved symbol of fastdebug build without precompiled headers
Reviewed-by: mgerdin, ehelin

Summary: If large pages are enabled and configured in Linux, the VM always pre-commits the entire space. The VM fails verification of the commit of the initial heap because some internal data structure marked all memory pages of the heap as committed during initialization. This makes the code think that we attempted a double-commit during first allocation of the heap. Remove the initial marking of memory pages of the heap to committed.
Reviewed-by: mgerdin

Summary: Allow G1 to decommit memory of arbitrary regions within the heap and their associated auxiliary data structures card table, BOT, hot card cache, and mark bitmaps.
Reviewed-by: mgerdin, brutisso, jwilhelm

Summary: Let HeapRegionSeq manage the heap region and auxiliary data to decrease the amount of responsibilities of G1CollectedHeap, and encapsulate this work from other code.
Reviewed-by: jwilhelm, jmasa, mgerdin, brutisso

Reviewed-by: tschatzl, jmasa

Summary: Fixup the remaining variable names.
Reviewed-by: tonyp, jwilhelm

Summary: The verification code that checked whether no bitmap mark had been found re-read HeapRegion::end() after the check on the bitmap. Concurrent humongous object allocation could have changed HeapRegion::end() in the meantime. Fix this by using the actual end of the region instead of HeapRegion::end() for comparison.
Reviewed-by: brutisso, jmasa

Summary: Enable shrinking within the heap by removing some code added for JDK-8054818. Enable the test case that checks that again too.
Reviewed-by: jwilhelm, jmasa

Reviewed-by: jmasa, jwilhelm

Reviewed-by: jwilhelm, jmasa

Reviewed-by: pliden, jmasa

Summary: Extend the WhiteBox API to provide information about the size and age of objects. Further add a mechanism to trigger a young GC.
Reviewed-by: tschatzl, sjohanss
Contributed-by: NLeonid Mesnik <leonid.mesnik@oracle.com>

Summary: Enhance scalability of native memory tracking
Reviewed-by: coleenp, ctornqvi, gtriantafill

Summary: Changing copy_allocation_context_stats to return if there are more stats available after the copy.
Reviewed-by: rriggs, jcoomes

Reviewed-by: brutisso, tschatzl

Reviewed-by: brutisso, sjohanss

Summary: Remove dead code.
Reviewed-by: stefank, brutisso

Summary: Remove dead code.
Reviewed-by: stefank, brutisso

Summary: Do not try to verify timing measures for the redirty logged cards phase when it is not executed.
Reviewed-by: brutisso, jmasa