37 Commits

Author SHA1 Message Date
John Cuthbertson
3a2b2b4fd8 6814437: G1: remove the _new_refs array
The per-worker _new_refs array is used to hold references that point into the collection set. It is populated during RSet updating and subsequently processed. In the event of an evacuation failure it processed again to recreate the RSets of regions in the collection set. Remove the per-worker _new_refs array by processing the references directly. Use a DirtyCardQueue to hold the cards containing the references so that the RSets of regions in the collection set can be recreated when handling an evacuation failure.

Reviewed-by: iveresov, jmasa, tonyp
2010-08-02 12:51:43 -07:00
Antonios Printezis
dfc84e8c89 6944166: G1: explicit GCs are not always handled correctly
G1 was not handling explicit GCs correctly in many ways. It does now. See the CR for the list of improvements contained in this changeset.

Reviewed-by: iveresov, ysr, johnc
2010-06-28 14:13:17 -04:00
Antonios Printezis
3e2cf1a294 6946048: G1: improvements to +PrintGCDetails output
Small improvements to G1's PrintGCDetails output. It also includes minor formatting details.

Reviewed-by: ysr, johnc
2010-04-22 15:20:16 -04:00
Erik Trimble
ba7c173659 6941466: Oracle rebranding changes for Hotspot repositories
Change all the Sun copyrights to Oracle copyright

Reviewed-by: ohair
2010-05-27 19:08:38 -07:00
John Cuthbertson
b1867e0dd5 6819061: G1: eliminate serial Other times that are proportional to the collection set length
6871109: G1: remove the concept of the scan only prefix

Removed scan only regions and associated code. The young portion of the collection set is now constructed incrementally - when a young region is retired as the current allocation region it is added to the collection set.

Reviewed-by: apetrusenko, iveresov, tonyp
2010-04-22 10:02:38 -07:00
Andrey Petrusenko
99cdb36eca 6942253: G1: replace G1ParallelGCAllocBufferSize with YoungPLABSize and OldPLABSize
Replaces the G1-specific parameter with the existing ones that are used by the other GCs (YoungPLABSize and OldPLABSize)

Reviewed-by: tonyp, johnc
2010-04-16 08:48:16 -07:00
John Cuthbertson
e9f77f7ec1 6755988: G1: assert(new_obj != 0 || ... "should be forwarded")
A TLAB became large enough to be considered a humongous object allowing multiple objects to be allocated in a humongous region, which violates a basic assumption about humongous regions. The changes ensure that TLABs cannot be regarded as humongous.

Reviewed-by: iveresov, tonyp
2010-03-11 11:44:43 -08:00
John Coomes
4f82340476 4396719: Mark Sweep stack overflow on deeply nested Object arrays
Use an explicit stack for object arrays and process them in chunks.

Reviewed-by: iveresov, apetrusenko
2010-03-03 14:48:26 -08:00
Igor Veresov
52b92d5593 6923991: G1: improve scalability of RSet scanning
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
2010-02-11 15:52:19 -08:00
Y. Srinivas Ramakrishna
342a620a7f 6920090: G1: Disable ReduceInitialCardMarks at least until 6920109 is fixed
G1 now answers "no" to the query can_elide_initializing_store_barrier() in the product build. A debug flag allows alternate behaviour in debug builds.

Reviewed-by: iveresov, tonyp
2010-01-26 16:52:29 -08:00
Y. Srinivas Ramakrishna
331512606a 6896647: card marks can be deferred too long
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
2010-01-13 15:26:39 -08:00
Antonios Printezis
8a125d9913 6880903: G1: G1 reports incorrect Runtime.maxMemory()
G1 reports committed memory instead of reserved memory from the Runtime.maxMemory() method

Reviewed-by: ysr, jmasa
2009-12-04 07:44:35 -05:00
Y. Srinivas Ramakrishna
928ac69fcd 6888898: CMS: ReduceInitialCardMarks unsafe in the presence of cms precleaning
6889757: G1: enable card mark elision for initializing writes from compiled code (ReduceInitialCardMarks)

Defer the (compiler-elided) card-mark upon a slow-path allocation until after the store  and before the next subsequent safepoint; G1 now answers yes to can_elide_tlab_write_barriers().

Reviewed-by: jcoomes, kvn, never
2009-10-16 02:05:46 -07:00
Antonios Printezis
243367ec16 6819085: G1: use larger and/or user settable region size
Instead of the region size being hard-coded, allow the user to set it.

Reviewed-by: jmasa, johnc, apetrusenko
2009-07-30 16:22:58 -04:00
Y. Srinivas Ramakrishna
c6763b5bad 4957990: Perm heap bloat in JVM
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
2009-09-02 00:04:29 -07:00
Antonios Printezis
26c66d0015 6859911: G1: assert(Heap_lock->owner() = NULL, "Should be owned on this thread's behalf")
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
2009-07-15 12:22:59 -04:00
Y. Srinivas Ramakrishna
075c1335cb 6700789: G1: Enable use of compressed oops with G1 heaps
Modifications to G1 so as to allow the use of compressed oops.

Reviewed-by: apetrusenko, coleenp, jmasa, kvn, never, phh, tonyp
2009-07-14 15:40:39 -07:00
Antonios Printezis
1215bc6717 6855834: G1: minimize the output when -XX:+PrintHeapAtGC is set
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
2009-07-07 14:23:00 -04:00
Antonios Printezis
f741e2fceb 6850846: G1: extend G1 marking verification
Extend G1 marking verification to use either the "prev" or "next" marking information, as appropriate.

Reviewed-by: johnc, ysr
2009-06-12 16:20:16 -04:00
John Cuthbertson
70fe3ca4f5 6843694: G1: assert(index < _vs.committed_size(),"bad index"), g1BlockOffsetTable.inline.hpp:55
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
2009-06-11 17:19:33 -07:00
Andrey Petrusenko
b7eda61fb6 6819065: G1: eliminate high serial card table clearing time
Reviewed-by: iveresov, tonyp
2009-05-19 04:05:31 -07:00
Igor Veresov
74e0691df5 6822263: G1: JVMTI heap iteration fails
Make object_iterate() traverse the perm gen

Reviewed-by: apetrusenko, tonyp
2009-03-26 08:51:32 -07:00
Andrey Petrusenko
cf2ae8d98d 6543938: G1: remove the concept of popularity
Reviewed-by: iveresov, tonyp
2009-03-25 13:10:54 -07:00
Antonios Printezis
f8c3115e4e Merge 2009-03-18 11:37:48 -04:00
Antonios Printezis
ecdb99412d 6604422: G1: re-use half-promoted regions
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
2009-03-15 22:03:38 -04:00
Erik Trimble
8d6035660e Merge 2009-03-12 18:16:36 -07:00
Xiomara Jayasena
2be2e027b0 6814575: Update copyright year
Update copyright for files that have been modified in 2009, up to 03/09

Reviewed-by: katleman, tbell, ohair
2009-03-09 13:28:46 -07:00
Igor Veresov
416d70bf91 6720309: G1: don't synchronously update RSet during evacuation pauses
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
2009-03-06 13:50:14 -08:00
Andrey Petrusenko
61181babf1 6484959: G1: introduce survivor spaces
6797754: G1: combined bugfix

Implemented a policy to control G1 survivor space parameters.

Reviewed-by: tonyp, iveresov
2009-02-06 01:38:50 +03:00
Antonios Printezis
5d486c3a66 6484956: G1: improve evacuation pause efficiency
A bunch of performance optimizations to decrease GC pause times in G1.

Reviewed-by: apetrusenko, jmasa, iveresov
2009-01-16 13:02:20 -05:00
John Coomes
2ba0d5b055 Merge 2009-01-14 14:12:55 -08:00
Jon Masamitsu
3474623106 6689653: JMapPerm fails with UseConcMarkSweepIncGC and compressed oops off
Added safe_object_iterate() for use by JMapPerm.

Reviewed-by: tonyp
2009-01-06 07:05:05 -08:00
Xiomara Jayasena
99f0eac1a2 6785258: Update copyright year
Update copyright for files that have been modified starting July 2008 to Dec 2008

Reviewed-by: katleman, ohair, tbell
2008-12-15 16:55:11 -08:00
Antonios Printezis
6f5e28aca1 6752248: G1: introduce parallel heap verification
Introduce parallel heap verification in G1.

Reviewed-by: jcoomes, apetrusenko
2008-10-06 13:16:35 -04:00
Antonios Printezis
e2b3310cb6 6722565: G1: assert !r->is_on_unclean_list() fires
Under certain circumstances, two cleanup threads can claim and process the same region.

Reviewed-by: apetrusenko, ysr
2008-08-06 11:57:31 -04:00
Igor Veresov
2703662478 6723570: G1: assertion failure: p == current_top or oop(p)->is_oop(),"p is not a block start" (revisited!)
Fixed the incorrect assigment to G1OffsetTableContigSpace::_gc_time_stamp. Also added a little more paranoia to operations on a global timestamp.

Reviewed-by: tonyp
2008-07-10 09:29:54 -07:00
Y. Srinivas Ramakrishna
18f3386a98 6711316: Open source the Garbage-First garbage collector
First mercurial integration of the code for the Garbage-First garbage collector.

Reviewed-by: apetrusenko, iveresov, jmasa, sgoldman, tonyp, ysr
2008-06-05 15:57:56 -07:00