Hi
I'm seeing a crash in ubuntu 8.10 now after updating to R2.9 with rkward 0.5c (the deb package I built locally) after a svn-source install (/usr/local/bin) it works fine again.

I have done the tests suggested by Thomas on the good and the bad version and can send the output it they are still needed.

Regards
Finn

On 2009-04-20 12:57, Pierre-Yves wrote:
On Fri, 2009-04-17 at 23:15 +0200, Thomas Friedrichsmeier wrote:
Hi Pierre,

On Wednesday 15 April 2009, Pierre-Yves wrote:
Do you have by any change an idea to solve it ?
I have no good idea on why this is happening. Perhaps one of these pointers will provide a useful cue to work with:

1. The symbol "_ZNK7QString11lastIndexOfER7QRegExpi" should be in Qt. More precisely in libQtCore.so . You can run something like
        strings /usr/lib/libQtCore.so | grep "lastIndexOf"
to verify it really exists (with that precise name).

Interestingly, here (on Debian), the symbol is called "_ZNK7QString11lastIndexOfERK7QRegExpi". That's real close, but note the missing "K".

@pingouLab ~]$ strings /usr/lib64/libQtCore.so | grep "lastIndexOf"
_ZNK10QByteArray11lastIndexOfEci
_ZNK10QByteArray11lastIndexOfERKS_i
_ZNK7QString11lastIndexOfE5QChariN2Qt15CaseSensitivityE
_ZNK7QString11lastIndexOfERK7QRegExpi
_ZNK7QString11lastIndexOfERKS_iN2Qt15CaseSensitivityE
_ZN9QtPrivate23QStringList_lastIndexOfEPK11QStringListRK7QRegExpi

On the computer where rkward crashes (after installing qt-devel)
@pingouRed ~]$ strings /usr/lib/libQtCore.so | grep "lastIndexOf"
_ZNK10QByteArray11lastIndexOfEci
_ZNK10QByteArray11lastIndexOfERKS_i
_ZNK7QString11lastIndexOfE5QChariN2Qt15CaseSensitivityE
_ZNK7QString11lastIndexOfERK7QRegExpi
_ZNK7QString11lastIndexOfERKS_iN2Qt15CaseSensitivityE
_ZN9QtPrivate23QStringList_lastIndexOfEPK11QStringListRK7QRegExpi


1b. Running
        strings /path/to/rkward.bin | grep "lastIndexOf"
should contain that precise string as well.


I have some weird results here:
@pingouRed ~]$ strings /usr/bin/rkward.bin | grep "lastIndexOf"
_ZNK7QString11lastIndexOfE5QChariN2Qt15CaseSensitivityE
_ZNK7QString11lastIndexOfERKS_iN2Qt15CaseSensitivityE
_ZNK7QString11lastIndexOfER7QRegExpi
----- From the computer on which rkward crashes
@pingouLab ~]$ strings /usr/bin/rkward.bin | grep "lastIndexOf"
_ZNK7QString11lastIndexOfE5QChariN2Qt15CaseSensitivityE
_ZNK7QString11lastIndexOfERKS_iN2Qt15CaseSensitivityE
_ZNK7QString11lastIndexOfERK7QRegExpi
----- From the computer on which rkward does *not* crash

2. If you manage to have one version of rkward with that crash and one without, running
        ldd /path/to/rkward.bin
on both binaries could possibly show a relevant difference.

Where it crashes:
http://fpaste.org/paste/9599
Where is does not crash:
http://fpaste.org/paste/9600

3. Assuming something is going wrong inside rkward code, perhaps trashing some memory locations in a strange way, it would be helpful to have the precise context. Try running
        rkward --debug-level 5
produce the crash and post the last ~20 lines just before the "undefined symbol" message. Please do this several times to make sure the context stays the same. If it does not, post several versions.
First trial:
http://fpaste.org/paste/9596
Second:
http://fpaste.org/paste/9597
Third:
http://fpaste.org/paste/9598

I am wondering if I don't miss the requirement on qt-devel when I build
the package, I will do some test in this direction.

Thanks for your help,
Best regards,
Pierre



------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
RKWard-devel mailing list
RKWard-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rkward-devel

------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
RKWard-devel mailing list
RKWard-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rkward-devel

Reply via email to