[Soc-2009-dev] Weird bug

Jingyuan Huang jingyuan.huang at gmail.com
Mon Jun 22 19:00:12 CEST 2009

Hi all,
Sorry to mail to both mailing lists (2.5 and gsoc). I have a bug that
I cannot determine if it roots from 2.5 or from my light paint

The bug can easily be reproduced in my branch. It is more difficult to
reproduce in 2.5. Steps are listed on wiki:
http://wiki.blender.org/index.php/User:Yukishiro/Bug_List. Please try
the alternative way for 2.5 branch since it doesn't have light paint
(the alternative approach can reproduce the crash quite predictably in
my branch). If you cannot make 2.5 crash by loading one file, keep
repeating step 3 and 4, and it should eventually crash (at least
according to my experience so far).

You can get my branch at:
The build depends on lapack and blas. Please install those packages if
you want to build (or wait for another day or two till the libraries
are added to lib directory).

The stack trace is shown below.
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f8634536740 (LWP 9050)]
0x00007f8633df2985 in gzread () from /usr/lib/libz.so.1
(gdb) bt
#0  0x00007f8633df2985 in gzread () from /usr/lib/libz.so.1
#1  0x00000000009d4417 in BKE_read_exotic (scene=0x2503990,
"/home/jing/repository/blender/test248/render/glass.blend") at
#2  0x00000000005db0bf in WM_read_file (C=0x1da28b0,
reports=0x7f862400fd70) at
#3  0x00000000005e50dd in wm_open_mainfile_exec (C=0x1da28b0,
op=0x7f8624011a00) at
#4  0x00000000005de7f4 in wm_handler_fileselect_call (C=0x1da28b0,
handlers=0x1ee7398, handler=0x7f8624011090, event=0x7f8624013790) at
#5  0x00000000005dea5f in wm_handlers_do (C=0x1da28b0,
event=0x7f8624013790, handlers=0x1ee7398) at
#6  0x00000000005deee3 in wm_event_do_handlers (C=0x1da28b0) at
#7  0x00000000005dcc24 in WM_main (C=0x1da28b0) at
#8  0x00000000005dac7d in main (argc=1, argv=0x7fff3c668f38) at
(gdb) info threads
* 1 Thread 0x7f8634536740 (LWP 9050)  0x00007f8633df2985 in gzread ()
from /usr/lib/libz.so.1

I have absolutely no idea why it would fail there. Any insight is appreciated.

Best Wishes
Jingyuan Huang
Computer Graphics Lab
University of Waterloo

More information about the Soc-2009-dev mailing list