Re: SIGSEGV during adding directory tree

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: SIGSEGV during adding directory tree

Karol Szkudlarek
Peter N. Lundblad wrote:

> On Fri, 22 Apr 2005, Karol Szkudlarek wrote:
>
>
>>I have svn compiled from trunk (linked with apr 0.9.5) and when I want
>>add directory tree (recursive) via command: svn add * I got randomly
>>SIGSEGV. See my example of gdb session:
>>
>>(gdb) karol@karol:~/pgr/thirdparty/boost_1_32_0/boost/numeric> gdb svn
>
> ...
>
>>(gdb) r add *
>>Starting program: /home/karol/usr/subversion/bin/svn add *
>>[Thread debugging using libthread_db enabled]
>>[New Thread 182924926816 (LWP 32100)]
>>A         conversion
>
> ...
>
>>Program received signal SIGSEGV, Segmentation fault.
>>[Switching to Thread 182924926816 (LWP 32100)]
>>0x0000002a97119df0 in strlen () from /lib64/tls/libc.so.6
>>(gdb) where
>>#0  0x0000002a97119df0 in strlen () from /lib64/tls/libc.so.6
>>#1  0x0000002a968ca720 in apr_vformatter () from /usr/lib64/libapr-0.so.0
>>#2  0x0000002a968da16b in apr_pvsprintf () from /usr/lib64/libapr-0.so.0
>>#3  0x0000002a9626431b in svn_error_wrap_apr (status=Variable "status"
>>is not available.
>>) at ../repossvn/trunk/subversion/libsvn_subr/error.c:171
>
>
> Thanks for the report. I've looked at the code and there is nothing
> obvious. Could you debug a little more? For example, is the argument to
> strlen NULL, or something else? And what are the arguments to
> apr_pvsprintf?
>
> A complete reproduction recipe would be good ofcourse, but I'm not sure if
> this is system dependent.
>
> Regards,
> //Peter
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]

Hi!

Currently I compiled svn-trunk with the latest version (0.9.x) of apr
library and from now I can't reproduce that SIGFAULT.....
So I've been watching that problem yet.

Greets,
Karol


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Loading...