You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@harmony.apache.org by "Igor V Chebykin (JIRA)" <ji...@apache.org> on 2007/03/15 14:08:09 UTC
[jira] Commented: (HARMONY-3179) [drlvm][jit][opt] specjvm98
_213_javac benchmark fails, segmentation fault
[ https://issues.apache.org/jira/browse/HARMONY-3179?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12481143 ]
Igor V Chebykin commented on HARMONY-3179:
------------------------------------------
After applying HARMONY-3389 the bug is not reproducible.
Analyzing shows that the issue is due to wrong value of vt_offset field in struct ManagedObject.
But wrong value is not actually wrong - it is equals to right value + 2.
It seems that gc is something doing with this while vm_monitor_exit_synchronized_method()
So, it seems that HARMONY-3389 is right patch for HARMONY-3179 too.
Vera, I think we can close the issue now.
What do you think?
> [drlvm][jit][opt] specjvm98 _213_javac benchmark fails, segmentation fault
> ---------------------------------------------------------------------------
>
> Key: HARMONY-3179
> URL: https://issues.apache.org/jira/browse/HARMONY-3179
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Environment: Linux x86_64
> Reporter: Vera Volynets
>
> Our machine fails on:
> java -Xem:opt -cp "." SpecApplication -a -m10 -M10 _213_javac
> Output:
> ======= _213_javac 10% execution - timing ignored
> Run 1 start. Total memory=73662464 free memory=10211996
> Javac benchmark starting...
> Javac benchmark starting...
> Javac benchmark starting...
> Segmentation fault (core dumped)
> Stack trace:
> #0 0x0000002a969dd59c in Class::get_vtable (this=0x2aaaa8) at Class.h:1131
> #1 0x0000002a96aa9260 in managed_object_object_is_valid_unsafe (p_obj=0x2a9bfba0f0)
> at /nfs/ims/proj/drl/mrt1/users/verav/trunk/working_vm/vm/vmcore/src/object/object_handles.cpp:183
> #2 0x0000002a96aa92de in managed_object_is_valid (p_obj=0x2a9bfba0f0)
> at /nfs/ims/proj/drl/mrt1/users/verav/trunk/working_vm/vm/vmcore/src/object/object_handles.cpp:190
> #3 0x0000002a96aa57df in vm_monitor_exit_default (p_obj=0x2a9bfba0f0)
> at /nfs/ims/proj/drl/mrt1/users/verav/trunk/working_vm/vm/vmcore/src/thread/mon_enter_exit.cpp:129
> #4 0x0000002a96aa56ec in vm_monitor_exit_synchronized_method (si=0x2aab4fe460)
> at /nfs/ims/proj/drl/mrt1/users/verav/trunk/working_vm/vm/vmcore/src/thread/mon_enter_exit.cpp:102
> #5 0x0000002a96a301e9 in exn_propagate_exception (si=0x2aab4fe460, exn_obj=0x7fbfffc1b0, exn_class=0x2aaa3be968, exn_constr=0x0,
> jit_exn_constr_args=0x0, vm_exn_constr_args=0x0)
> at /nfs/ims/proj/drl/mrt1/users/verav/trunk/working_vm/vm/vmcore/src/exception/exceptions_jit.cpp:343
> #6 0x0000002a96a304f0 in exn_throw_for_JIT (exn_obj=0x2a99bf3b00, exn_class=0x0, exn_constr=0x0, jit_exn_constr_args=0x0,
> vm_exn_constr_args=0x0) at /nfs/ims/proj/drl/mrt1/users/verav/trunk/working_vm/vm/vmcore/src/exception/exceptions_jit.cpp:439
> #7 0x0000002a96a30582 in exn_athrow (exn_obj=0x2a99bf3b00, exn_class=0x0, exn_constr=0x0, exn_constr_args=0x0)
> at /nfs/ims/proj/drl/mrt1/users/verav/trunk/working_vm/vm/vmcore/src/exception/exceptions_jit.cpp:462
> I suppose that jit returns wrong p_this in funtion vm_monitor_exit_synchronized_method.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.