[kaffe] Re: jthreads problem?

[Guilhem Lavaux]

Hi Ioannis,

I have downloaded the file. However I am not sure I will be able to 
reproduce your exact system state and this is a problem because core 
dumps depends on the exact libc, the exact gcc, ... So probably _I_ will 
not be able to exploit it. However _you_ can. ;-) I can guide you if you 
need. You can also come on irc.freenode.net (channel #kaffe). I'll 
probably be there on sunday.

I'll try to produce the same error too.

Regards,
Guilhem.

Ioannis Liverezas wrote:

>Hi Guilhem,
>
>I followed your instructions and produced a core dump. The problem appeared 
>immediately (~5minutes), possibly because I use jthreads. This time I used 
>the latest cvs.
>The file size of the core dump is 31MB, and you can download it from the 
>following link:
>http://skylab.inaccessnetworks.com/misc-downloads/kaffe-core.13590 
>
>After downloading the file inform me as soon as possible in order to remove it 
>from the server.
>
>I also attached to this mail the options that I pass to configure
>
>I noticed that I forgot to CC to kaffe at kaffe.org my previous mails to you and 
>as reasonable you replied only to me. So if it is ok with you, CC your reply 
>to this mail to the mailing list, so that anybody can follow this thread.
>
>Best regards,
>Ioannis Liverezas
>
>
>On Thursday 09 February 2006 22:54, you wrote:
>  
>
>>Hi Ioannis,
>>
>>Anyway I think the two problems may be related. The lists are badly
>>trashed sometimes. I fear the synchronization has some problems in
>>locks.c/getHeavyLock. The problem is that I need a proof of it and it's
>>hard to get it as it's about thread concurrency. :-/
>>
>>If you can have more information then I'll take them with pleasure.
>>Please try to get a core dump (ulimit -c unlimited or unlimit on tcsh)
>>and use a kaffe compiled with '-O0'. That way we'll be able to examine
>>the "full" status of kaffe.
>>
>>Regards,
>>Guilhem.
>>
>>Ioannis Liverezas wrote:
>>    
>>
>>>Hi Guilhem,
>>>
>>>I 'll check the test with the current cvs version of kaffe. However, what
>>>was the version that you used in order to reproduce the problem? It seems
>>>strange that you could not reproduce the bug. Also keep in mind that the
>>>time that the problem needs in order to appear is random, varying from
>>>10minutes to a couple of hours.
>>>I will run some tests in order to provide you with some feedback.
>>>
>>>On Thursday 09 February 2006 19:09, you wrote:
>>>      
>>>
>>>>Ioannis Liverezas wrote:
>>>>        
>>>>
>>>>>Hi all,
>>>>>
>>>>>Any progress made with the jthreads problem? Last post about this
>>>>>problem was on 13/01/2006.
>>>>>          
>>>>>
>>>>It's my first priority currently. I didn't manage to reproduce the
>>>>problem with jthreads atm however pthreads has a really painful bug. For
>>>>some reason the lists of the locking subsystem are sometimes completely
>>>>crashed. I think I'll manage to have some results soon though (~ 1 week).
>>>>
>>>>Guilhem.
>>>>        
>>>>