Uploaded image for project: 'JDK'
  1. JDK
  2. JDK-6211040

vmark client crashed with mustang b16 CMS GC

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Not an Issue
    • Icon: P2 P2
    • 6
    • 6
    • hotspot
    • None
    • gc
    • x86
    • solaris_10


      VolanoMark client crashed intermittenly on x86-bigwolf with Mustang b16 CMS GC.
      JVM flags used: -server -XX:+UseConcMarkSweepGC -XX:+ShowMessageBoxOnError
      -XX:CompileThreshold=100

      x86-bigwolf is a x86 running sol 10 b70.

      stack trace shows that it might be a GC bug.
      =>[1] ___nanosleep(0x64, 0xd2e26698), at 0xd2f6c9ec
        [2] _sleep(0x64), at 0xd2f6126c
        [3] os::message_box(0xd2d96f5c, 0xd2e26698), at 0xd2cb25e9
        [4] VMError::show_message_box(0xd1f391c8, 0xd2e26698, 0x7d0), at 0xd2cfe584
        [5] VMError::report_and_die(), at 0xd2cfe165
        [6] JVM_handle_solaris_signal(0xb, 0xd1f39614, 0xd1f39414, 0x1), at 0xd297a37
      e
        [7] signalHandler(0xb, 0xd1f39614, 0xd1f39414), at 0xd2979c60
        [8] __sighndlr(0xb, 0xd1f39614, 0xd1f39414, 0xd2979c40), at 0xd2f6c8cf
        [9] call_user_handler(0xb, 0xd1f39614, 0xd1f39414), at 0xd2f62f20
        [10] sigacthandler(), at 0xd2f630a0
        ---- called from signal handler with signal 11 (SIGSEGV) ------
        [11] ParScanClosure::do_oop_work(), at 0xd2cb8dfd
        [12] ParRootScanWithoutBarrierClosure::do_oop(0xd1f39b5c, 0x857e560), at 0xd2
      cb8856
        [13] chunk_oops_do(0xd1f39b5c, 0x857e558, 0x857e574), at 0xd292a1fe
        [14] HandleArea::oops_do(0x857b830, 0xd1f39b5c), at 0xd292a166
        [15] Thread::oops_do(0x857dc68, 0xd1f39b5c), at 0xd292a251
        [16] JavaThread::oops_do(0x857dc68, 0xd1f39b5c), at 0xd292ba5b
        [17] Threads::possibly_parallel_oops_do(0xd1f39b5c), at 0xd2cda32a
        [18] GenCollectedHeap::process_strong_roots(0x80eacd0, 0x0, 0x1, 0x0, 0x1, 0x
      d1f39b80, 0xd1f39b5c), at 0xd2b33898
        [19] ParNewGenTask::work(0xcf7fd8c8, 0x0), at 0xd2cb84bd
        [20] GangWorker::run(0x80eb8b0), at 0xd2d0056e
        [21] _start(0x80eb8b0), at 0xd2cb10e2
        [22] _thr_setup(0xd2eb8800), at 0xd2f6c492
        [23] _lwp_start(), at 0xd2f6c780

      The problem occured on x86-bigwolf three times. Not sure if sol 10 b70 OS
      could be the culprit. Recently I have seen some bugs in sol 10, but stack traces of those bugs clearly indicate the bugs are in sol 10 OS.
      Running the test with the same jvm options on three other x86 machines now.

      ###@###.### 2004-12-21 06:47:25 GMT
      ###@###.### 2004-12-21 18:35:24 GMT

            ysr Y. Ramakrishna
            jzhongsunw June Zhong (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

              Created:
              Updated:
              Resolved:
              Imported:
              Indexed: