You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@harmony.apache.org by "Elena Semukhina (JIRA)" <ji...@apache.org> on 2007/01/22 12:06:30 UTC

[jira] Created: (HARMONY-3033) [drlvm][ipf] smoke test gc.Mark crashes with seg fault

[drlvm][ipf] smoke test gc.Mark crashes with seg fault
------------------------------------------------------

                 Key: HARMONY-3033
                 URL: https://issues.apache.org/jira/browse/HARMONY-3033
             Project: Harmony
          Issue Type: Bug
          Components: DRLVM
            Reporter: Elena Semukhina


The test fails in the interpreter mode with the following message:

allocating 400 Mb on 2 workers with 10 sleepers, live size 100 Mb, pause threshold 120 ms
Segmentation fault



-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

[jira] Commented: (HARMONY-3033) [drlvm][ipf] smoke test gc.Mark crashes with seg fault

Posted by "Ivan Volosyuk (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HARMONY-3033?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12471270 ] 

Ivan Volosyuk commented on HARMONY-3033:
----------------------------------------

Here the stack trace at crash point:

java/net/URL hashCode ()I
java/util/HashMap keysEqual (Ljava/lang/Object;Ljava/util/HashMap$Entry;)Z
java/util/HashMap findEntry (Ljava/lang/Object;I)Ljava/util/HashMap$Entry;
java/util/HashMap getEntry (Ljava/lang/Object;)Ljava/util/HashMap$Entry;
java/util/HashMap get (Ljava/lang/Object;)Ljava/lang/Object;
java/net/URLClassLoader explore (Ljava/net/URL;I)[Ljava/net/URL;
java/net/URLClassLoader findClassImpl ([Ljava/net/URL;Ljava/lang/String;)Ljava/lang/Class;
java/net/URLClassLoader$4 run ()Ljava/lang/Class;
java/net/URLClassLoader$4 run ()Ljava/lang/Object;
java/security/AccessController doPrivilegedImpl (Ljava/security/PrivilegedAction;Ljava/security/AccessControlContext;)Ljava/lang/Object;
java/security/AccessController doPrivileged (Ljava/security/PrivilegedAction;Ljava/security/AccessControlContext;)Ljava/lang/Object;
java/net/URLClassLoader findClass (Ljava/lang/String;)Ljava/lang/Class;
java/lang/ClassLoader loadClass (Ljava/lang/String;Z)Ljava/lang/Class;
java/lang/ClassLoader$SystemClassLoader loadClass (Ljava/lang/String;Z)Ljava/lang/Class;
java/lang/ClassLoader loadClass (Ljava/lang/String;)Ljava/lang/Class;
<VM code>
UserDefinedClassLoader::DoLoadClass
Class::load_ancestors
Java_java_lang_VMClassRegistry_defineClass


> [drlvm][ipf] smoke test gc.Mark crashes with seg fault
> ------------------------------------------------------
>
>                 Key: HARMONY-3033
>                 URL: https://issues.apache.org/jira/browse/HARMONY-3033
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>            Reporter: Elena Semukhina
>
> The test fails in the interpreter mode with the following message:
> allocating 400 Mb on 2 workers with 10 sleepers, live size 100 Mb, pause threshold 120 ms
> Segmentation fault

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


[jira] Updated: (HARMONY-3033) [drlvm][exception][smoke] smoke test gc.Mark crashes with seg fault

Posted by "Gregory Shimansky (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HARMONY-3033?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Gregory Shimansky updated HARMONY-3033:
---------------------------------------

    Summary: [drlvm][exception][smoke] smoke test gc.Mark crashes with seg fault  (was: [drlvm][exception] smoke test gc.Mark crashes with seg fault)

> [drlvm][exception][smoke] smoke test gc.Mark crashes with seg fault
> -------------------------------------------------------------------
>
>                 Key: HARMONY-3033
>                 URL: https://issues.apache.org/jira/browse/HARMONY-3033
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>         Environment: Linux/IPF
>            Reporter: Elena Semukhina
>
> The test fails in the interpreter mode with the following message:
> allocating 400 Mb on 2 workers with 10 sleepers, live size 100 Mb, pause threshold 120 ms
> Segmentation fault

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


[jira] Updated: (HARMONY-3033) [drlvm][exception] smoke test gc.Mark crashes with seg fault

Posted by "Gregory Shimansky (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HARMONY-3033?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Gregory Shimansky updated HARMONY-3033:
---------------------------------------

    Environment: Linux/IPF
        Summary: [drlvm][exception] smoke test gc.Mark crashes with seg fault  (was: [drlvm][ipf] smoke test gc.Mark crashes with seg fault)

Set exception category according to evaluation that it looks like SOE. Similar problem is described in HARMONY-3019.

> [drlvm][exception] smoke test gc.Mark crashes with seg fault
> ------------------------------------------------------------
>
>                 Key: HARMONY-3033
>                 URL: https://issues.apache.org/jira/browse/HARMONY-3033
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>         Environment: Linux/IPF
>            Reporter: Elena Semukhina
>
> The test fails in the interpreter mode with the following message:
> allocating 400 Mb on 2 workers with 10 sleepers, live size 100 Mb, pause threshold 120 ms
> Segmentation fault

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


[jira] Commented: (HARMONY-3033) [drlvm][ipf] smoke test gc.Mark crashes with seg fault

Posted by "Ivan Volosyuk (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HARMONY-3033?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12471335 ] 

Ivan Volosyuk commented on HARMONY-3033:
----------------------------------------

May be r12 contains the correct value. Very strange situation.

Stack trace:
#0  Opcode_GETFIELD (frame=@0x0)
    at /a/nfs/ims/proj/drl/mrt1/users/ivanrt/harmony/drlvm/trunk/vm/interpreter/src/interpreter.cpp:1949
#1  0x200000000452a440 in interpreter (frame=@0x2000000015d403b0)
    at /a/nfs/ims/proj/drl/mrt1/users/ivanrt/harmony/drlvm/trunk/vm/interpreter/src/interpreter.cpp:2883
#2  0x200000000452fd70 in interpreterInvoke (prevFrame=@0x2000000015d40900, method=0x60000000006392c0, args=1,
    obj=0x2000000004aef140, intf=false)
    at /a/nfs/ims/proj/drl/mrt1/users/ivanrt/harmony/drlvm/trunk/vm/interpreter/src/interpreter.cpp:3382

Frame is zero at upper frame. But looking at the prologue of the method:
(gdb) i r ip
ip             0x2000000004558c40
(gdb) disas
Dump of assembler code for function Opcode_GETFIELD:
0x2000000004558c20 <Opcode_GETFIELD+0>: [MII]       alloc r36=ar.pfs,13,7,0
0x2000000004558c21 <Opcode_GETFIELD+1>:             mov r37=r12
0x2000000004558c22 <Opcode_GETFIELD+2>:             adds r12=-304,r12
0x2000000004558c30 <Opcode_GETFIELD+16>:        [MII]       mov r38=r1
0x2000000004558c31 <Opcode_GETFIELD+17>:                    mov r35=b0;;
0x2000000004558c32 <Opcode_GETFIELD+18>:                    adds r14=-288,r37;;
0x2000000004558c40 <Opcode_GETFIELD+32>:        [MII]       st8 [r14]=r32
0x2000000004558c41 <Opcode_GETFIELD+33>:                    adds r33=-280,r37
0x2000000004558c42 <Opcode_GETFIELD+34>:                    adds r15=-288,r37;;

Looks like the prologue is not finished so the values represented by debug may be invalid.
r14 initialized from r12 (stack frame register)
(gdb) i r r12 r14
r12            0x2000000015d3ffe0       2305843009579909088
r14            0x2000000015d3fff0       2305843009579909104

cat /proc/PID/maps:
2000000015c44000-2000000015d3c000 rwxp 2000000015c44000 00:00 0
2000000015d3c000-2000000015d40000 ---p 2000000015d3c000 00:00 0
2000000015d40000-2000000015dc0000 rwxp 2000000015d40000 00:00 0

It seems that both r12 and r14 point to inaccessible memory region.

Last and first StackFrame:
(StackFrame *) 0x2000000015d403b0
(StackFrame *) 0x2000000015dbcb20
354 frames total. 509808 bytes.

Looks like stack overflow.


> [drlvm][ipf] smoke test gc.Mark crashes with seg fault
> ------------------------------------------------------
>
>                 Key: HARMONY-3033
>                 URL: https://issues.apache.org/jira/browse/HARMONY-3033
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>            Reporter: Elena Semukhina
>
> The test fails in the interpreter mode with the following message:
> allocating 400 Mb on 2 workers with 10 sleepers, live size 100 Mb, pause threshold 120 ms
> Segmentation fault

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


[jira] Commented: (HARMONY-3033) [drlvm][ipf] smoke test gc.Mark crashes with seg fault

Posted by "Ivan Volosyuk (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HARMONY-3033?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12471299 ] 

Ivan Volosyuk commented on HARMONY-3033:
----------------------------------------

Looks like something broken after this call:
   java/lang/System arraycopy(Ljava/lang/Object;ILjava/lang/Object;II)

r12 register which points to stack contains wrong value.

> [drlvm][ipf] smoke test gc.Mark crashes with seg fault
> ------------------------------------------------------
>
>                 Key: HARMONY-3033
>                 URL: https://issues.apache.org/jira/browse/HARMONY-3033
>             Project: Harmony
>          Issue Type: Bug
>          Components: DRLVM
>            Reporter: Elena Semukhina
>
> The test fails in the interpreter mode with the following message:
> allocating 400 Mb on 2 workers with 10 sleepers, live size 100 Mb, pause threshold 120 ms
> Segmentation fault

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