[jira] Commented: (HARMONY-5553) [drlvm][reliability] MappedByteBufferForceTest failed

["Pavel Rebriy (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HARMONY-5553?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12579369#action_12579369 ] 

Pavel Rebriy commented on HARMONY-5553:
---------------------------------------

I've investigated the failure and got the following crashes:

1. Crash in free_local_object_handles3() function:

#0  0x00002adce7441aa5 in raise () from /lib64/libc.so.6
#1  0x00002adce7442f90 in abort () from /lib64/libc.so.6
#2  0x00002adce747836b in __libc_message () from /lib64/libc.so.6
#3  0x00002adce747d37e in malloc_printerr () from /lib64/libc.so.6
#4  0x00002aaaaabc58bf in free_local_object_handles3 (head=<value optimized out>)
    at /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/vm/vmcore/src/object/object_handles.cpp:507
#5  0x00002aaaac7916db in ?? ()
#6  0x00002aaaaad12850 in FRAME_POP_NOW ()
   from /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/build/linux_x86_64_gcc_release/deploy/jdk/jre/
bin/default/libharmonyvm.so
#7  0x0000000000000100 in ?? ()
#8  0x00002aaaae6606e8 in ?? ()
#9  0x00000000011c4a68 in ?? ()
#10 0x00000000011fca08 in ?? ()
#11 0x00002aaaae6606b8 in ?? ()
#12 0x0000000000000000 in ?? ()

This crash happened due to invalid local handle stored in m2n frame.

2. Crash during enumeration of local handles in oh_enumerate_handles() function:

#0  oh_enumerate_handles (hs=<value optimized out>)
    at /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/vm/vmcore/src/object/object_handles.cpp:338
#1  0x00002aaaaabcf539 in vm_enumerate_root_set_single_thread_on_stack (si=0x2aaabe8a1800)
    at /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/vm/vmcore/src/gc/root_set_enum_common.cpp:347
#2  0x00002aaaaabd7720 in vm_enumerate_thread (thread=0x11ee250)
    at /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/vm/vmcore/src/gc/stop_the_world_root_set_enum.c
pp:221
#3  0x00002aaaaabd791a in stop_the_world_root_set_enumeration ()
    at /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/vm/vmcore/src/gc/stop_the_world_root_set_enum.c
pp:121
#4  0x00002aaaad516e4c in gc_prepare_rootset (gc=0x573530)
    at /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/vm/gc_gen/src/common/gc_common.cpp:408
#5  0x00002aaaad516fa2 in gc_reclaim_heap (gc=0x573530, gc_cause=<value optimized out>)
    at /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/vm/gc_gen/src/common/gc_common.cpp:440
#6  0x00002aaaad51aa1a in gc_force_gc ()
    at /nfs/ims/proj/drl/mrt2/users/psrebriy/git-tmp/vm/gc_gen/src/common/gc_for_vm.cpp:246
#7  0x00002aaabe49722e in ?? ()
#8  0x00002aaa00090009 in ?? ()
#9  0x0000000000000000 in ?? ()
Current language:  auto; currently c++

It looks like the same problem: incorrect value of local handle in local handles list.

There are 2 assumption of the failures: the 1st one is incorrect operations with local handles during  exception processing and the 2nd is a memory hit.
Have to note this is a intermittent failure I've reproduced on release build only.

> [drlvm][reliability] MappedByteBufferForceTest failed
> -----------------------------------------------------
>
>                 Key: HARMONY-5553
>                 URL: https://issues.apache.org/jira/browse/HARMONY-5553
>             Project: Harmony
>          Issue Type: Bug
>    Affects Versions: 5.0M5
>         Environment: Linux x86_64
> revision 629320
>            Reporter: Andrey Yakushev
>
> api.nio.channels.filechannel.MappedByteBufferForceTest from the relability test suite failed after 233 successful iterations with message:
> ...
> OK
> Thread 8: IOException while writing buffer
> Thread 9: IOException while writing buffer
> Thread 9: IOException while writing buffer
> thread #8 returned not PASS status
> *** iteration: 234
> I will check possible resource leakage later. Or it also could be race conditions.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.