Implemented block-based work stealing. Moved copying during the rset scanning phase to the main copying phase. Made the size of rset table depend on the region size.
Reviewed-by: apetrusenko, tonyp
Deferred card marks are now flushed during the gc prologue. Parallel[Scavege,OldGC] and SerialGC no longer defer card marks generated by COMPILER2 as a result of ReduceInitialCardMarks. For these cases, introduced a diagnostic option to defer the card marks, only for the purposes of testing and diagnostics. CMS and G1 continue to defer card marks. Potential performance concern related to single-threaded flushing of deferred card marks in the gc prologue will be addressed in the future.
Reviewed-by: never, johnc
ScavengeALot now causes an incremental (but possibly partially young, in the G1 sense) collection. Some such collections may be abandoned on account of MMU specs. Band-aided a native leak associated with abandoned pauses, as well as an MMU tracker overflow related to frequent scavenge events in the face of a large MMU denominator interval; the latter is protected by a product flag that defaults to false.
Reviewed-by: tonyp
The fix addresses two memory leaks in G1 code: (1) _evac_failure_scan_stack - a resource object allocated on the C heap was not freed; (2) RSHashTable were linked into deleted list which was only cleared at full GC.
Reviewed-by: tonyp, iveresov
It tidies up the G1 heap verification a bit. In particular, when the verification is done in parallel and there is a failure, this is propagated to the top level and the heap is dumped at the end, not by every thread that encounters a failure.
Reviewed-by: johnc, jmasa
The first worker thread is delayed when entering the GC because it clears the card count table that is used in identifying hot cards. Replace the card count table with a dynamically sized evicting hash table that includes an epoch based counter.
Reviewed-by: iveresov, tonyp
Have the GC worker threads clear the hot card cache in parallel by having each worker thread claim a chunk of the card cache and process the cards in that chunk. The size of the chunks that each thread will claim is determined at VM initialization from the size of the card cache and the number of worker threads.
Reviewed-by: jmasa, tonyp
Treat ProfileData in MDO's as a source of weak, not strong, roots. Fixes the bug for stop-world collection -- the case of concurrent collection will be fixed separately.
Reviewed-by: jcoomes, jmasa, kvn, never
Removing the concurrent overhead tracker from G1, along with the GC overhead reporter and the G1AccountConcurrentOverhead (both of which rely on the the concurrent overhead tracker).
Reviewed-by: iveresov, johnc
It adds parameter -XX:+G1SummarizeRSetStatsPeriod that causes update buffer processing information to be printed periodically. It also includes two small formatting changes.
Reviewed-by: jmasa, jcoomes, ysr
The used() method assumes that the heap lock is held when it is called. However, when used() is called from print_on(), this is not the case.
Reviewed-by: ysr, jmasa
Changing the behavior of -XX:+PrintHeapAtGC for G1 from printing lengthy, per-region information to instead printing a concise summary.
Reviewed-by: ysr, apetrusenko, jcoomes
For heaps larger than 32Gb, the number of heap regions overflows the data type used to hold the region index in the SparsePRT structure. Changed the region indexes, card indexes, and RSet hash table buckets to ints and added some size overflow guarantees.
Reviewed-by: ysr, tonyp
Enabled full verification for humongous regions. Also made sure that the VerifyAfterGC works with deferred updates and G1HRRSFlushLogBuffersOnVerify.
Reviewed-by: tonyp
6728271: G1: Cleanup G1CollectedHeap::get_gc_alloc_regions()
It allows the last half-full region to be allocated to during a GC to be reused during the next GC.
Reviewed-by: apetrusenko, jcoomes
Use zero based compressed oops if java heap is below 32gb and unscaled compressed oops if java heap is below 4gb.
Reviewed-by: never, twisti, jcoomes, coleenp
The length of the fast cset test vector is decided at the beginning of a GC, but more regions can be added during the GC. The simple fix is to set the length of the fast cset test vector to the max.
Reviewed-by: iveresov
6720334: G1: don't update RSets of collection set regions during an evacuation pause
Introduced a deferred update mechanism for delaying the rset updates during the collection pause
Reviewed-by: apetrusenko, tonyp
Under certain circumstances (evacuation failure) the pause time is not communicated to the policy and, as a result, the pause time field is not initialized properly.
Reviewed-by: jmasa
