Mon, 28 Nov 2011 09:49:05 -0800 7114303: G1: assert(_g1->mark_in_progress()) failed: shouldn't be here otherwise
johnc [Mon, 28 Nov 2011 09:49:05 -0800] rev 11175
7114303: G1: assert(_g1->mark_in_progress()) failed: shouldn't be here otherwise Summary: Race between the VM thread reading G1CollectedHeap::_mark_in_progress and it being set by the concurrent mark thread when concurrent marking is aborted by a full GC. Have the concurrent mark thread join the SuspendibleThreadSet before changing the marking state. Reviewed-by: tonyp, brutisso
Tue, 09 Aug 2011 10:16:01 -0700 6593758: RFE: Enhance GC ergonomics to dynamically choose ParallelGCThreads
jmasa [Tue, 09 Aug 2011 10:16:01 -0700] rev 11174
6593758: RFE: Enhance GC ergonomics to dynamically choose ParallelGCThreads Summary: Select number of GC threads dynamically based on heap usage and number of Java threads Reviewed-by: johnc, ysr, jcoomes
Tue, 22 Nov 2011 04:47:10 -0500 Merge
tonyp [Tue, 22 Nov 2011 04:47:10 -0500] rev 11173
Merge
Fri, 18 Nov 2011 12:27:10 -0800 7111795: G1: Various cleanups identified during walk through of changes for 6484965
johnc [Fri, 18 Nov 2011 12:27:10 -0800] rev 11172
7111795: G1: Various cleanups identified during walk through of changes for 6484965 Summary: Various cleanups and formatting changes identified during a code walk through of the changes for 6484965 ("G1: piggy-back liveness accounting phase on marking"). Reviewed-by: brutisso, tonyp
Mon, 21 Nov 2011 09:24:56 -0800 7110173: GCNotifier::pushNotification publishes stale data.
johnc [Mon, 21 Nov 2011 09:24:56 -0800] rev 11171
7110173: GCNotifier::pushNotification publishes stale data. Summary: GCNotifier::pushNotification() references GCMemoryManager::_last_gc_stat but is called from GCMemoryManager::gc_end() before GCMemoryManager::_last_gc_stat is set up using the values in GCMemoryManager::_current_gc_stat. As a result the GC notification code accesses unitialized or stale data. Move the notification call after GCMemoryManager::_las_gc_stat is set, but inside the same if-block. Reviewed-by: poonam, dholmes, fparain, mchung
Mon, 21 Nov 2011 07:47:34 +0100 7110718: -XX:MarkSweepAlwaysCompactCount=0 crashes the JVM
brutisso [Mon, 21 Nov 2011 07:47:34 +0100] rev 11170
7110718: -XX:MarkSweepAlwaysCompactCount=0 crashes the JVM Summary: Interpret MarkSweepAlwaysCompactCount < 1 as never do full compaction Reviewed-by: ysr, tonyp, jmasa, johnc
Fri, 18 Nov 2011 12:52:27 -0500 7097002: G1: remove a lot of unused / redundant code from the G1CollectorPolicy class
tonyp [Fri, 18 Nov 2011 12:52:27 -0500] rev 11169
7097002: G1: remove a lot of unused / redundant code from the G1CollectorPolicy class Summary: Major cleanup of the G1CollectorPolicy class. It removes a lot of unused fields and methods and also consolidates replicated information (mainly various ways of counting the number of CSet regions) into one copy. Reviewed-by: johnc, brutisso
Thu, 01 Dec 2011 13:42:41 -0500 Merge
coleenp [Thu, 01 Dec 2011 13:42:41 -0500] rev 11168
Merge
Wed, 30 Nov 2011 12:48:52 -0500 7116730: Revert 7116481: Commercial features in Hotspot must be gated by a switch
phh [Wed, 30 Nov 2011 12:48:52 -0500] rev 11167
7116730: Revert 7116481: Commercial features in Hotspot must be gated by a switch Summary: Revert 7116481 to current hsx/hotspot-main Reviewed-by: kamg
Tue, 29 Nov 2011 17:00:46 -0500 7116481: Commercial features in Hotspot must be gated by a switch
phh [Tue, 29 Nov 2011 17:00:46 -0500] rev 11166
7116481: Commercial features in Hotspot must be gated by a switch Summary: Add -XX:+UnlockCommercialVMOptions to gate use of commercial feature switches in the same way as -XX:UnlockDiagnosticVMOptions gates use of diagnostic feature switches. Reviewed-by: jwilhelm, kamg
(0) -10000 -3000 -1000 -300 -100 -10 +10 +100 +300 +1000 +3000 +10000 +30000 tip