Hi Antonio,

> Hi Carsten,
> 
> Carsten Neumann wrote:
> 
>>      Hi Antonio,
>>
>>On Wed, 2007-09-05 at 19:52 +0200, Antonio Bleile wrote:
>>
>>>Hi,
>>
>>[SNIP]
>>
>>>             LockCount       -17891601       long
>>>             RecursionCount  -17891602       long
>>>             OwningThread    0xfeeefeee      void *
>>>             LockSemaphore   0xfeeefeee      void *
>>>             SpinCount       4277075694      unsigned long
>>
>>>where OwningThread seems quite suspicious...
>>
>>indeed, but the interesting question is of course why it has that
>>strange value ;)
>>
>>
>>>If I remove the addRefCP and subRefCP calls, the osgExit works
>>>ok. osgInit, the rendering etc and osgExit run all in the same
>>>thread! The context and the PassiveWindow is deleted _before_
>>>calling osgExit(). Any ideas?
>>
>>No, sorry, none right now, only that I would not be too surprised if
>>this is related to the ThreadManager::shutdown crashes reported by
>>Gabriel yesterday.
>>The problem is that without being able to reproduce these cases (ok, I
>>haven't tried yet) it looks hard to figure out. You don't happen to have
>>a small reproducing testcase, do you ? ;)
> 
> 
> I'm trying very hard. But it doesn't crash.... I observed that the
> actual crash does not occur in the subRefCP but in the destructor
> of the instance which has some OSGPtr's as members. Using
> and deleting the instance withing the OpenSG tutorial's works fine
> though. I'm not able to see on which member destruction it crashes
> with the debugger...(if it does so...).
> 
> Regards,
> 
>    Toni

in the scons build system there exist a "invalid_pointer_check" option 
if you enable this the addRefCP, subRefCP, ... methods are replaced by 
some special macros and with these it is possible to detect a subRefCP 
call on a already destroyed pointer. Actually it will report the exact 
source code line with the invalid subRefCP call as a FATAL error 
message. As usual there er some drawbacks with this option, ok the check 
costs a bit processor time, and if you call addRefCP, subRefCP with the 
OSG namespace like OSG::subRefCP this won't work.
If someone knows a solution how to write a macro that works with a 
namespace ...

Andreas


-- 
VREC
Robert-Bosch-Straße 7
D-64293 Darmstadt
Tel. 06151-4921035

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users

Reply via email to