JDK-4950157 : CMS: vtest died with tiger b26
  • Type: Bug
  • Component: hotspot
  • Sub-Component: gc
  • Affected Version: 1.4.2_02,1.4.2_04,5.0
  • Priority: P3
  • Status: Resolved
  • Resolution: Fixed
  • OS: generic,solaris_8
  • CPU: generic,sparc
  • Submitted: 2003-11-06
  • Updated: 2004-08-17
  • Resolved: 2004-01-21
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
Other
1.4.2_06 06Fixed
Related Reports
Duplicate :  
JDK-5013589 - CMS: HotSpot 1.4.2_02 dies with a seg fault.
Relates :  
JDK-5093457 - CMS: crash in nightly RT_QUICK tests 8/26/04 "Adding a bad referent"
Description
On j2se-bigapps.west, (SunOS j2se-bigapps 5.9 Generic_112233-09 sun4u sparc SUNW,Sun-Fire   20 cpu * 1050 MHZ, 98 GB memory )
vtest died with CMS gc on j2se-bigapps.west after 4 days.
Flags used: -client -Xcomp -XX:+UseConcMarkSweepGC
log files are under /bt/VolanoTestrun.25402.-client ( I may move the log
to /export/archive/tiger_b26/ later)

  ---- called from signal handler with signal 11 (SIGSEGV) ------
  [10] MarkSweep::IsAliveClosure::do_object_b(0xff12fbbc, 0x0, 0x6000, 0xff02c1ec,
0x35e900, 0x6314), at 0xfedbdfd8
  [11] ReferenceProcessor::process_discovered_reflist(0x28e760, 0xfedbdfd8, 0x0,
0x0, 0xff12fbbc, 0xff13226c), at 0xff02c1fc
  [12] ReferenceProcessor::process_discovered_references(0x28e760, 0xf8, 0x7640,
0xff1289a0, 0x0, 0x7400), at 0xff02bbcc
  [13] GenMarkSweep::invoke_at_safepoint(0x1, 0x7000, 0x7168, 0x28e760, 0x6c00,
0xc1e38), at 0xfedbb50c
  [14] CMSCollector::do_compaction_work(0x1d0190, 0x0, 0xeb800000, 0x0, 0x1d032c,
0x3600000), at 0xfee53900
  [15] CMSCollector::acquire_control_and_collect(0x1d0190, 0x2ac90, 0x0, 0x9,
0xff130510, 0x1d0268), at 0xfee53690
  [16] ConcurrentMarkSweepGeneration::collect(0x145d28, 0x0, 0x0, 0x2c2c54,
0xfee54f64, 0x1d0190), at 0xfee53400
  [17] GenCollectedHeap::do_collection(0x28ea98, 0x0, 0xc1e80, 0x10, 0xc1e84, 0x0),
at 0xfee7dd00
  [18] TwoGenerationCollectorPolicy::satisfy_failed_allocation(0xc1ca0, 0x10, 0x0,
0x0, 0xdcf7f3b4, 0xff116000), at 0xfee42f58
  [19] VM_GenCollectForAllocation::doit(0xdcf7f398, 0x62, 0xfed745d8, 0xff1338dc,
0x6c00, 0xff13aa74), at 0xfedb025c
  [20] VM_Operation::evaluate(0xdcf7f398, 0xc80e8, 0x28ef34, 0x28eb40, 0x28eb48,
0x7f9a91), at 0xfed768d4
  [21] VMThread::run(0xff130eac, 0x28eb48, 0x28ea98, 0x28ef34, 0xff13a924,
0xff13a920), at 0xfecdaae0
  [22] _start(0x28ea98, 0x4c80, 0x6400, 0x6710, 0x4c00, 0xff116000), at 0xfecda098

I also ran vtest with -server -XX:CompileThreshold=100 and -d64. So far
the failure only happens with -client -Xcomp.

###@###.### 2003-11-06
Comments
CONVERTED DATA BugTraq+ Release Management Values COMMIT TO FIX: 1.4.2_06 generic tiger-beta2 FIXED IN: 1.4.2_06 tiger-beta2 INTEGRATED IN: 1.4.2_06 tiger-b35 tiger-beta2 VERIFIED IN: tiger-beta2
18-08-2004
SUGGESTED FIX Before the mark-sweep-compact code picks up the collection, clear references with NULL referents out of the discovered lists. The diffs for this bug-fix are available at: http://analemma.sfbay.sun.com/net/prt-archiver/export2/archived_workspaces/main/gc_baseline/2004/20040112104357.jmasa.gc_baseline_4950157/workspace/webrevs/webrev-2004.01.12/ Please contact ###@###.### or ###@###.### if there are any issues if/when backporting this fix from 1.5 to 1.4.2_XX.
18-08-2004
PUBLIC COMMENTS no comment
18-08-2004
EVALUATION Root cause not yet known; trying to reproduce problem. ###@###.### 2003-12-16 I have not been able to reproduce this failure. In the comments is a description of a scenario that could lead to the stack trace in the description.
16-12-2003