openjdk crashes while installing jdeveloper

Wed Feb 9 04:37:09 PST 2011

Nikolas and I are working on this off list. The bsd_zero apple
atomic operation implementations such as atomic_swap and
atomic_cas are not thread safe. SMP makes that problem
easier to see. We tried using the gcc 4.2 builtin atomic operations
however the are now deadlocks to debug.

On 2/3/11 6:09 AM, Java House wrote:
> Kurt I can't tell you how much I appreciate the effort you put on this.
>  I made several tries and I have attached the files for each try.
> I didn't zip them because they are relative small.
>
>
> this is the first try with jdeveloper 3
>
>
> quad:tests nikoniko$ export 
> PATH=/Volumes/Data/Users/nikoniko/Applications/Java/openjdk7-macppc-2011-02-02/bin/:$PATH
> quad:tests nikoniko$ export 
> JAVA_HOME=/Volumes/Data/Users/nikoniko/Applications/Java/openjdk7-macppc-2011-02-02/
> quad:tests nikoniko$
> quad:tests nikoniko$
> quad:tests nikoniko$ which java
> /Volumes/Data/Users/nikoniko/Applications/Java/openjdk7-macppc-2011-02-02/bin//java
> quad:tests nikoniko$ java -version
> openjdk version "1.7.0-internal-fastdebug"
> OpenJDK Runtime Environment (build 
> 1.7.0-internal-fastdebug-truk_2011_02_02_14_07-b00)
> OpenJDK Zero VM (build 20.0-b06-fastdebug, interpreted mode)
> quad:tests nikoniko$ java -version
> openjdk version "1.7.0-internal-fastdebug"
> OpenJDK Runtime Environment (build 
> 1.7.0-internal-fastdebug-truk_2011_02_02_14_07-b00)
> OpenJDK Zero VM (build 20.0-b06-fastdebug, interpreted mode)
> quad:tests nikoniko$ pwd
> /Volumes/Data/Users/nikoniko/temp/tests
> quad:tests nikoniko$ ls
> hotspot.log
> quad:tests nikoniko$ rm hotspot.log
> quad:tests nikoniko$ java -jar ~/Downloads/jdevstudio11113install.jar
> Unable to instantiate GUI, defaulting to console mode.
> Extracting 
> 0%....................................................................................................100%
> # To suppress the following error report, specify this argument
> # after -XX: or in .hotspotrc:  SuppressErrorAt=/objectMonitor.cpp:297
> #
> # A fatal error has been detected by the Java Runtime Environment:
> #
> #  Internal Error 
> (/Users/truk/jdk/bsd-port/hotspot/src/share/vm/runtime/objectMonitor.cpp:297), 
> pid=865, tid=4031803392
> #  assert(_owner == Self) failed: invariant
> #
> # JRE version: 7.0
> # Java VM: OpenJDK Zero VM (20.0-b06-fastdebug interpreted mode bsd-ppc )
> # An error report file with more information is saved as:
> # /Volumes/Data/Users/nikoniko/temp/tests/hs_err_pid865.log
> #
> # If you would like to submit a bug report, please visit:
> # http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Current thread is 4031803392
> Dumping core ...
>
>
> the two log files
>
>
>
>
>
> --------------------------------
>
> Next try with the newest jdeveloper 4
> first try I had to stop with CTRL-C as it stop responding, two files 
> attached hotspot.897.log and hs_err_pid897.log
>
>
>
>
>
>
>
> -------------------------------
>
> Second try generated core
>
>
> quad:tests nikoniko$ java -jar ~/Downloads/jdevstudio11114install.jar
> Unable to instantiate GUI, defaulting to console mode.
> Extracting 
> 0%....................................................................................................100%
> # To suppress the following error report, specify this argument
> # after -XX: or in .hotspotrc:  SuppressErrorAt=/objectMonitor.cpp:297
> #
> # A fatal error has been detected by the Java Runtime Environment:
> #
> #  Internal Error 
> (/Users/truk/jdk/bsd-port/hotspot/src/share/vm/runtime/objectMonitor.cpp:297), 
> pid=928, tid=4030222336
> #  assert(_owner == Self) failed: invariant
> #
> # JRE version: 7.0
> # Java VM: OpenJDK Zero VM (20.0-b06-fastdebug interpreted mode bsd-ppc )
> # An error report file with more information is saved as:
> # /Volumes/Data/Users/nikoniko/temp/tests/hs_err_pid928.log
> #
> # If you would like to submit a bug report, please visit:
> # http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Current thread is 4030222336
> Dumping core ...
> quad:tests nikoniko$
>
>
> two files attached hotspot.928.log hs_err_pid928.log
>
>
>
>
>
>
> ---------------------------------------------------------
>
> Now with the flag
>
> quad:tests nikoniko$ java -XX:+SuppressFatalErrorMessage -jar 
> ~/Downloads/jdevstudio11114install.jar
> VM option '+SuppressFatalErrorMessage'
> Unable to instantiate GUI, defaulting to console mode.
> Extracting 
> 0%....................................................................................................100%
> # To suppress the following error report, specify this argument
> # after -XX: or in .hotspotrc:  SuppressErrorAt=/objectMonitor.cpp:297
> #
> # A fatal error has been detected by the Java Runtime Environment:
> #
> #  Internal Error 
> (/Users/truk/jdk/bsd-port/hotspot/src/share/vm/runtime/objectMonitor.cpp:297), 
> pid=970, tid=4031803392
> #  assert(_owner == Self) failed: invariant
> #
> # JRE version: 7.0
> # Java VM: OpenJDK Zero VM (20.0-b06-fastdebug interpreted mode bsd-ppc )
> # An error report file with more information is saved as:
> # /Volumes/Data/Users/nikoniko/temp/tests/hs_err_pid970.log
> #
> # If you would like to submit a bug report, please visit:
> # http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Current thread is 4031803392
> Dumping core ...
> quad:tests nikoniko$ ls
>
> attached files hotspot.970.log hs_err_pid970.log
>
>
>
>
>
>
> now running the same under xterm
>
> bash-3.2$ java -jar ../../Downloads/jdevstudio11114install.jar
> Unable to instantiate GUI, defaulting to console mode.
> # To suppress the following error report, specify this argument
> # after -XX: or in .hotspotrc:  SuppressErrorAt=/mutex.cpp:901
> #
> # A fatal error has been detected by the Java Runtime Environment:
> #
> #  Internal Error 
> (/Users/truk/jdk/bsd-port/hotspot/src/share/vm/runtime/mutex.cpp:901), 
> pid=1196, tid=4038660096
> #  assert(owner() == NULL) failed: invariant
> #
> # JRE version: 7.0
> # Java VM: OpenJDK Zero VM (20.0-b06-fastdebug interpreted mode bsd-ppc )
> # An error report file with more information is saved as:
> # /Volumes/Data/Users/nikoniko/temp/tests/hs_err_pid1196.log
> #
> # If you would like to submit a bug report, please visit:
> # http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Current thread is 4038660096
> Dumping core ...
>
>
> the files are hotspot.1196.log  hs_err_pid1196.log
>
>
>
>
>
>
> under xterm with the flag
> bash-3.2$ java -XX:+SuppressFatalErrorMessage -jar 
> ../../Downloads/jdevstudio11114install.jar
> VM option '+SuppressFatalErrorMessage'
> Unable to instantiate GUI, defaulting to console mode.
> # To suppress the following error report, specify this argument
> # after -XX: or in .hotspotrc:  SuppressErrorAt=/mutex.cpp:1274
> #
> # A fatal error has been detected by the Java Runtime Environment:
> #
> #  Internal Error 
> (/Users/truk/jdk/bsd-port/hotspot/src/share/vm/runtime/mutex.cpp:1274), pid=1219, 
> tid=4030222336
> #  assert(_owner == NULL) failed: setting the owner thread of an 
> already owned mutex
> #
> # JRE version: 7.0
> # Java VM: OpenJDK Zero VM (20.0-b06-fastdebug interpreted mode bsd-ppc )
> # An error report file with more information is saved as:
> # /Volumes/Data/Users/nikoniko/temp/tests/hs_err_pid1219.log
> #
> # If you would like to submit a bug report, please visit:
> # http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Current thread is 4030222336
> Dumping core ...
>
>
>
>
>
>
>
>
>
>
> On 03 ??? 2011, at 3:18 ??, Kurt Miller wrote:
>
>> Hi Nikolas,
>>
>> Thank you for testing and the details. I have tried around twenty times
>> to get this to reproduce for me. It always works. I suspect your SMP
>> setup is tripping over a race condition that I can't see with my single
>> processor setup.
>>
>> I have built the fastdebug version of the jdk and uploaded it to
>>
>> http://www.intricatesoftware.com/OpenJDK/macppc/openjdk7-macppc-2011-02-02.tar.bz2
>>
>> I adjusted the "fatal error: caught unhandled signal 10" string to 
>> include
>> the faulting address. Could you make it crash again two more times?
>>
>> The first time as before where it generates the hs_err_pidXXXX.log file
>> and the second time with this argument: -xx:+SuppressFatalErrorMessage
>> That argument should avoid the crash report and instead leave a core 
>> file.
>>
>> The core file and the new hs_err_pidXXX.log file should help diagnose 
>> this
>> more.
>>
>> Thanks,
>> -Kurt
>>
>> On 2/1/11 6:17 PM, Java House wrote:
>>> Hello Kurt,
>>>
>>> I tried first with the jdevstudio11113install.jar.
>>> Most of the times the installation with this package just hangs with 
>>> no error as shown bellow so I just kill the process
>>>
>>> quad:Downloads xxx$ java -version
>>> openjdk version "1.7.0-internal"
>>> OpenJDK Runtime Environment (build 
>>> 1.7.0-internal-truk_2011_01_31_21_31-b00)
>>> OpenJDK Zero VM (build 20.0-b06, interpreted mode)
>>>
>>>
>>> quad:Downloads xxx$ java -jar jdevstudio11113install.jar
>>> Unable to instantiate GUI, defaulting to console mode.
>>> Extracting 
>>> 0%...................................................................................................#
>>> # A fatal error has been detected by the Java Runtime Environment:
>>> #
>>> #  Internal Error (os_bsd_zero.cpp:266), pid=1635, tid=4043403264
>>> #  fatal error: caught unhandled signal 10
>>> #
>>> # JRE version: 7.0
>>> # Java VM: OpenJDK Zero VM (20.0-b06 interpreted mode bsd-ppc )
>>> # An error report file with more information is saved as:
>>> # /Volumes/Data/Users/xxx/Downloads/hs_err_pid1635.log
>>> #
>>> # If you would like to submit a bug report, please visit:
>>> # http://java.sun.com/webapps/bugreport/crash.jsp
>>> #
>>> quad:Downloads nikoniko$ java -jar jdevstudio11113install.jar
>>> Unable to instantiate GUI, defaulting to console mode.
>>> Extracting 
>>> 0%....................................................................................................100%
>>>
>>>
>>>
>>>
>>>
>>> <----------- Oracle Installer - Oracle JDeveloper 11g (11.1.1.3.0) 
>>> ----------->
>>>
>>> Welcome:
>>> --------
>>>
>>> This installer will guide you through the installation of Oracle 
>>> JDeveloper 11g (11.1.1.3.0). Type "Next" or enter to proceed to the 
>>> next prompt.  If you want
>>> to change data entered previously, type "Previous".  You may quit 
>>> the installer at any time by typing "Exit".
>>>
>>>
>>>
>>>
>>> Enter [Exit][Next]>
>>> ^Cquad:Downloads xxx$
>>>
>>>
>>>
>>> But sometimes the JVM crashes like bellow. I attached the log files.
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> quad:Downloads nikoniko$ java -jar jdevstudio11113install.jar
>>> Unable to instantiate GUI, defaulting to console mode.
>>> Extracting 
>>> 0%...................................................................................................#
>>> # A fatal error has been detected by the Java Runtime Environment:
>>> #
>>> #  Internal Error (os_bsd_zero.cpp:266), pid=1635, tid=4043403264
>>> #  fatal error: caught unhandled signal 10
>>> #
>>> # JRE version: 7.0
>>> # Java VM: OpenJDK Zero VM (20.0-b06 interpreted mode bsd-ppc )
>>> # An error report file with more information is saved as:
>>> # /Volumes/Data/Users/nikoniko/Downloads/hs_err_pid1635.log
>>> #
>>> # If you would like to submit a bug report, please visit:
>>> # http://java.sun.com/webapps/bugreport/crash.jsp
>>> #
>>>
>>>
>>> And with the newer version of JDeveloper
>>>
>>> Unable to instantiate GUI, defaulting to console mode.
>>> Extracting 
>>> 0%....................................................................................................100%
>>> #
>>> # A fatal error has been detected by the Java Runtime Environment:
>>> #
>>> #  Internal Error (os_bsd_zero.cpp:266), pid=1768, tid=4041895936
>>> #  fatal error: caught unhandled signal 10
>>> #
>>> # JRE version: 7.0
>>> # Java VM: OpenJDK Zero VM (20.0-b06 interpreted mode bsd-ppc )
>>> # An error report file with more information is saved as:
>>> # /Volumes/Data/Users/nikoniko/Downloads/hs_err_pid1768.log
>>> #
>>> # If you would like to submit a bug report, please visit:
>>> # http://java.sun.com/webapps/bugreport/crash.jsp
>>> #
>>>
>>>
>>> brg
>>> Nikolas
>>>
>>> On 01 ??? 2011, at 2:44 ??, Kurt Miller wrote:
>>>
>>>> Hi Jav,
>>>>
>>>> I have make the current macppc build available here:
>>>>
>>>> http://www.intricatesoftware.com/OpenJDK/macppc/openjdk7-macppc-2011-01-31.tar.bz2
>>>>
>>>> It includes the SIGBUS patch that I sent previously. Could
>>>> you give it a try with jdeveloper and report back success or
>>>> failure here?
>>>>
>>>> Thanks,
>>>> -Kurt
>>>>
>>>> On 12/24/10 8:40 AM, Kurt Miller wrote:
>>>>> Hello Jav,
>>>>>
>>>>> OS X raises SIGBUS signal instead of SIGSEGV for
>>>>> certain memory access exceptions.
>>>>>
>>>>> Please apply this diff to your hotspot src tree,
>>>>> rebuild and test. If this corrects your crash, please
>>>>> let me know and I will commit it to bsd-port.
>>>>>
>>>>> Regards,
>>>>> -Kurt
>>>>>
>>>>> On 12/23/10 02:29, Jav House wrote:
>>>>>> Hello all
>>>>>> As most people with PPC I am stack on JavaSE 5 but I need to use
>>>>>> JDeveloper and it requires at least 1.6.
>>>>>> While trying to install jdeveloper with openjdk7-macppc-2009-12-16-b4 on
>>>>>> MacOSX 10.5.8 PPC Quad with 16GB RAM
>>>>>>
>>>>>> #
>>>>>> # A fatal error has been detected by the Java Runtime Environment:
>>>>>> #
>>>>>> #  Internal Error (os_bsd_zero.cpp:232), pid=595, tid=4034990080
>>>>>> #  Error: caught unhandled signal 10
>>>>>> #
>>>>>> # JRE version: 7.0
>>>>>> # Java VM: OpenJDK Zero VM (17.0-b05 interpreted mode bsd-ppc )
>>>>>> # An error report file with more information is saved as:
>>>>>> #
>>>>>> /Volumes/Data/Users/nikoniko/Downloads/openjdk7-macppc-2009-12-16-b4/bin/hs_err_pid595.log
>>>>>> #
>>>>>> # If you would like to submit a bug report, please visit:
>>>>>> #http://java.sun.com/webapps/bugreport/crash.jsp
>>>>>> #
>>>>>>
>>>>>>
>>>>>> Has anyone succeed on installing Oracle's JDeveloper on PPC
>>>>>> with penjdk7-macppc-2009-12-16-b4?
>>>>>> Is there any new release coming? This is a year old release.
>>>>>> regards
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>>
>>
>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/bsd-port-dev/attachments/20110209/50f50e49/attachment.html 