Page MenuHome

Yafray plugin crash.
Closed, ArchivedPublicKNOWN ISSUE

Description

1) Load the attached file.
2) Click RENDER (F12)

Result: Crash!!!!!

Tested on current CVS sources and current 2.37a release (Linux OS).

Event Timeline

Logged In: YES
user_id=103

Maybe Alfredo can check?

Logged In: YES
user_id=742

I checked, but no crash... and without any further indication of what and where it is going wrong, I can't really do much about it...

Logged In: YES
user_id=103

OK.

Renato: you then should provide us with more information. Does yafray even render at all? What yafray version do you use?

Collegato: no
user_id=3475

Ok, I did some more testing. I'm using a Fedora Core 4 Linux box. I compiled Blender with gcc 4.0.1 and I was using official Yafray 0.0.8 rpm distribution (so no debug symbols).
When I click "Render" with the attached file, I receive a SIGSEGV signal, with the following backtrace from GDB:

Program received signal SIGSEGV, Segmentation fault.
0x63617274 in ?? ()
(gdb) bt
#0 0x63617274 in ?? ()
#1 0xb242c931 in yafray::blenderShader_t::factory () from /usr/lib/yafray/libblendershader.so
#2 0xb2670dc2 in yafray::interfaceImpl_t::addShader () from /usr/lib/libyafrayplugin.so
#3 0x082e999f in yafrayPluginRender_t::writeShader (this=0x87e89a0, shader_name=@0x8b4ce20, matr=0x8aa82a4,
facetexname=@0xbfcadeac) at source/blender/yafray/intern/export_Plugin.cpp:941
#4 0x082e436c in yafrayPluginRender_t::writeMaterialsAndModulators (this=0x87e89a0)
at source/blender/yafray/intern/export_Plugin.cpp:1090
#5 0x082b7ab3 in yafrayRender_t::exportScene (this=0x87e89a0) at source/blender/yafray/intern/yafray_Render.cpp:52
#6 0x082b5b53 in YAF_exportScene () at source/blender/yafray/intern/api.cpp:13
#7 0x0828e228 in render () at source/blender/render/intern/source/initrender.c:865
#8 0x0828f5b9 in RE_initrender (ogl_render_view3d=0xbfcabb80) at source/blender/render/intern/source/initrender.c:1373
#9 0x081a5b48 in do_render (ogl_render_view3d=0x0, anim=0, force_dispwin=Variable "force_dispwin" is not available.
) at source/blender/src/renderwin.c:1072
#10 0x081a5c6a in BIF_do_render (anim=0) at source/blender/src/renderwin.c:1177
#11 0x081bd5e4 in scrarea_do_winhandle (area=0x8a97fcc, evt=0xbfcabb80) at source/blender/src/spacetypes.c:144
#12 0x08130f25 in screenmain () at source/blender/src/editscreen.c:563
#13 0x081009c3 in main (argc=1, argv=0xbfcae334) at source/creator/creator.c:584

So I decided to recompile Yafray 0.0.8 from sources with debug symbols, using gcc 4.0.1. The crash seems to disappear (!!!) but, if I follow the following procedure, with the attached file, I receive another crash:
1) Click Render. Wait until it is finished.
2) Close the render window.
3) Re-click Render
4) Re-close render window.
5) Re-click render again.
6) Crash.

The backtrace, this time, said that's a problem of mesa_dri.so:

Program received signal SIGSEGV, Segmentation fault.
0xb75afa28 in __driGarbageCollectDrawables () from /usr/X11R6/lib/modules/dri/radeon_dri.so
(gdb) bt
#0 0xb75afa28 in __driGarbageCollectDrawables () from /usr/X11R6/lib/modules/dri/radeon_dri.so
#1 0xb75b0588 in driCreateNewContext () from /usr/X11R6/lib/modules/dri/radeon_dri.so
#2 0xb7c01b76 in glXChooseFBConfigSGIX () from /usr/X11R6/lib/libGL.so.1
#3 0xb7c0206d in glXCreateContext () from /usr/X11R6/lib/libGL.so.1
#4 0x08446ae5 in GHOST_WindowX11::installDrawingContext (this=0x8a01118, type=GHOST_kDrawingContextTypeOpenGL)
at intern/ghost/intern/GHOST_WindowX11.cpp:563
#5 0x08447529 in GHOST_WindowX11 (this=0x8a01118, system=0x74696445, display=0x74696445, title=@0xbf84beb0, left=327,
top=231, width=400, height=318, state=GHOST_kWindowStateNormal, type=GHOST_kDrawingContextTypeOpenGL,
stereoVisual=false) at intern/ghost/intern/GHOST_WindowX11.cpp:198
#6 0x084457d9 in GHOST_SystemX11::createWindow (this=0x888d2c0, title=@0x74696445, left=1953064005, top=1953064005,
width=1953064005, height=1953064005, state=1953064005, type=1953064005, stereoVisual=false)
at intern/ghost/intern/GHOST_SystemX11.cpp:217
#7 0x08444469 in GHOST_CreateWindow (systemhandle=0x888d2c0, title=0x74696445 <Address 0x74696445 out of bounds>,
left=1953064005, top=1953064005, width=1953064005, height=1953064005, state=1953064005, type=1953064005,
stereoVisual=0) at intern/ghost/intern/GHOST_C-api.cpp:162
#8 0x081423a5 in window_open (title=0x74696445 <Address 0x74696445 out of bounds>, posx=327, posy=231, sizex=400,
sizey=318, start_maximized=0) at source/blender/src/ghostwinlay.c:342
#9 0x081a496f in open_renderwin (winpos=0xbf84bf70, winsize=0xbf84bf78) at source/blender/src/renderwin.c:577
#10 0x081a594e in renderwin_init_display_cb () at source/blender/src/renderwin.c:656
#11 0x081a59d1 in do_render (ogl_render_view3d=0x0, anim=0, force_dispwin=Variable "force_dispwin" is not available.
) at source/blender/src/renderwin.c:1041
#12 0x081a5c6a in BIF_do_render (anim=0) at source/blender/src/renderwin.c:1177
#13 0x081bd5e4 in scrarea_do_winhandle (area=0x8a9800c, evt=0x74696445) at source/blender/src/spacetypes.c:144
#14 0x08130f25 in screenmain () at source/blender/src/editscreen.c:563
#15 0x081009c3 in main (argc=1, argv=0xbf84c174) at source/creator/creator.c:584

I'm using an ATI Radeon Mobility IGP320M.

So the crash cannot be considered blender/yafray related?

Logged In: YES
user_id=103

No this is a crash related to the dri drivers. Please check on alternative drivers, especially from ati themselves. You can also do more testing with renders, I am sure also non-yafray renders will crash!

Logged In: YES
user_id=2923

please confirm that this is still a problem with current CVS builds any unconfirmed bugs by the end of the week will be closed (they can be reoppened though).

you can get a current cvs build from

http://blender.org/forum/viewforum.php?f=18

Collegato: no
user_id=3475

I think this report could be definitely closed. It's a problem of Mesa-DRI driver, not a Yafray's one.
Sorry for the bad report.

This is a generic request to test your bug report and see if it is still an issue in 2.5alpha2 if so please let me know by making a comment in this report ie 'also in 2.5alpha2' and I will add it to the 2.5 bug list.

Matt Ebb (broken) changed the task status from Unknown Status to Unknown Status.Mar 26 2010, 6:15 AM