Use the more permissive set_klass_or_null() and klass_or_null() interfaces in ParNew's workqueue overflow code that manipulates the klass-word.
Reviewed-by: coleenp
For current soft-ref clearing policies, we can decide at marking time if a soft-reference will definitely not be cleared, postponing the decision of whether it will definitely be cleared to the final reference processing phase. This can be especially beneficial in the case of concurrent collectors where the marking is usually concurrent but reference processing is usually not.
Reviewed-by: jmasa
When we encounter marking stack overflow during precleaning of Reference lists, we were using the overflow list mechanism, which can cause problems on account of mutating the mark word of the header because of conflicts with mutator accesses and updates of that field. Instead we should use the usual mechanism for overflow handling in concurrent phases, namely dirtying of the card on which the overflowed object lies. Since precleaning effectively does a form of discovered list processing, albeit with discovery enabled, we needed to adjust some code to be correct in the face of interleaved processing and discovery.
Reviewed-by: apetrusenko, jcoomes
6660573: G1: BigApps Failure : guarantee(satb_mq_set.completed_buffers_num() == 0,"invariant")
When exiting a mutator thread is removed from the thread list before it has a chance to flush its SATB and barrier queues. If GC happens at this moment the objects that are refererred from these queues can be moved, which will case a crash. The fix is simply to flush the buffers before removing a thread from the list.
Reviewed-by: jcoomes, tonyp
The per-lgrp chuck size can be incorrectly computed (causing an assertion failure) because of the non-associativity of the floating point operations. The fix is to rearrange the operations.
Reviewed-by: ysr
An update buffer is not processed correctly, which causes roots into the collection set not to be scanned and, hence, for the heap to be corrupted. The cause is that an object is accessed after it has been explicitly deleted, which causes a race.
Reviewed-by: jcoomes, ysr
Fixed CMSConcMarkingTask::reset() to store the restart address upon a marking stack overflow and to use it as the base, suitably aligned, for restarting the scan in CMSConcMarkingTask::do_scan_and_mark().
Reviewed-by: jcoomes, tonyp
6622418: G1: assert(false,"Non-balanced monitor enter/exit!") fails
The mark-sweep compact (which we use for full gc) wrapper did not save the mark words for biased locked objects. The fix is to trivially call the appropriate methods.
Reviewed-by: tonyp, ysr
When an object array overflows during precleaning, we should have been marking the entire array dirty, not just its first card.
Reviewed-by: jmasa, poonam, tonyp
Allocate alignment-sized page(s) below java heap so that memory accesses at heap_base+1page give signal and cause an implicit null check
Reviewed-by: kvn, jmasa, phh, jcoomes
6723229: NUMA allocator: assert(lgrp_num > 0, "There should be at least one locality group")
The fix takes care of the assertion triggered during TLAB resizing after reconfiguration. Also it now handles a defect in the topology graph, in which a single leaf node doesn't have memory.
Reviewed-by: jmasa
Fixed the incorrect assigment to G1OffsetTableContigSpace::_gc_time_stamp. Also added a little more paranoia to operations on a global timestamp.
Reviewed-by: tonyp
Do not coalesce dead and moved objects when removing self-forwarding pointers during the evacuation failure. Also fixed a issue in a BOT refinement code for TLABs.
Reviewed-by: tonyp, jcoomes
Experimental() flags will protect features of an experimental nature that are not supported in the regular product build. Made UseG1GC an experimental flag.
Reviewed-by: jmasa, kamg, coleenp
