just tried to run ut2003_demo with radeon_dri.so from current DRI-cvs/MESA-cvs HEAD and got a SIGSEGV, just before the logo usually is displayed.
However ut2003_demo worked "well" with an older radeon_dri.so from xfree-cvs (20031226) I had lying around and it worked with radeon_dri.so from DRI-CVS before Mesa-newtree merge (20031130). (I tried to build DRI with Mesa-cvs mesa_6_0_branch: doesnt build; and DRI CVS HEAD without a mesa-tree doesnt build, too..; but thats another story...) I got this backtrace when just running ut2003_demo: Backtrace: [ 1] ./Core.so(appGetProcReturnCode__FPvPi+0x516) [0x40a0778a] [ 2] /lib/libpthread.so.0(pthread_kill+0x174) [0x40dacbc4] [ 3] /lib/libc.so.6(__libc_sigaction+0x138) [0x40bc07f8] [ 4] /usr/X11R6/lib/modules/dri/radeon_dri.so(__driUtilCreateScreen+0x58f4) [0x43db00a4] [ 5] /usr/X11R6/lib/modules/dri/radeon_dri.so(__driUtilCreateScreen+0xfb0a1) [0x43ea5851] [ 6] /usr/X11R6/lib/modules/dri/radeon_dri.so(__driUtilCreateScreen+0xfb5f2) [0x43ea5da2] [ 7] /home/as/ut2003_demo/System/OpenGLDrv.so(DrawPrimitive__22FOpenGLRenderInterface14EPrimitiveTypeiiii+0x373) [0x42f8faeb] [ 8] ./Engine.so(Render__20FStaticMeshBatchListP10FSceneNodeP16FRenderInterface+0x2b2) [0x40375d66] [ 9] ./Engine.so(RenderLevel__FP15FLevelSceneNodeP16FRenderInterface+0x22df) [0x40368f23] [10] ./Engine.so(Render__15FLevelSceneNodeP16FRenderInterface+0x7a2) [0x4034b38a] [11] ./Engine.so(Render__16FPlayerSceneNodeP16FRenderInterface+0x330) [0x403504ec] [12] ./Engine.so(Draw__11UGameEngineP9UViewportiPUcPi+0x3fe) [0x4028808a] [13] /home/as/ut2003_demo/System/SDLDrv.so(Repaint__12USDLViewporti+0x33) [0x42f5093b] [14] /home/as/ut2003_demo/System/SDLDrv.so(Tick__10USDLClient+0x85) [0x42f4f365] [15] ./Engine.so(Tick__11UGameEnginef+0x31bd) [0x4028f2e1] [16] ./ut2003-bin(_start+0x86d) [0x8051b1d] [17] ./ut2003-bin(main+0x328c) [0x8058b2c] [18] /lib/libc.so.6(__libc_start_main+0xbe) [0x40baf7ee] [19] ./ut2003-bin(_start+0x21) [0x80512d1] Signal: SIGSEGV [segmentation fault] Aborting. after modifying the ut2003_demo script to start gdb instead of ut2003-bin, I got this backtrace from gdb: (gdb) exec-file ../System/ut2003-bin (gdb) r Starting program: /home/as/ut2003_demo/System/../System/ut2003-bin (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...[New Thread 1024 (LWP 31522)] (no debugging symbols found)...[New Thread 2049 (LWP 31523)] Delayed SIGSTOP caught for LWP 31523. [New Thread 1026 (LWP 31524)] Delayed SIGSTOP caught for LWP 31524. (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...[New Thread 2051 (LWP 31525)] Delayed SIGSTOP caught for LWP 31525. Program received signal SIGFPE, Arithmetic exception. [Switching to Thread 1024 (LWP 31522)] 0x43ec44f4 in _mesa_test_os_sse_exception_support () from /usr/X11R6/lib/modules/dri/radeon_dri.so (gdb) thread [Current thread is 1 (Thread 1024 (LWP 31522))] (gdb) c Continuing. Xlib: extension "XiG-SUNDRY-NONSTANDARD" missing on display ":0.0". Program received signal SIGSEGV, Segmentation fault. 0x00000000 in ?? () (gdb) back #0 0x00000000 in ?? () #1 0x43ea593a in _tnl_draw_range_elements (ctx=0x92ac188, mode=4, max_index=32, index_count=48, indices=0x91ef0c0) at t_array_api.c:118 #2 0x43ea5d94 in _tnl_DrawRangeElements (mode=4, start=0, end=31, count=48, type=5123, indices=0x8f94140) at t_array_api.c:341 #3 0x42f8faeb in FOpenGLRenderInterface::DrawPrimitive () from /home/as/ut2003_demo/System/OpenGLDrv.so #4 0x40355532 in FBspDrawList::Render () from ./Engine.so #5 0x40368edf in RenderLevel () from ./Engine.so #6 0x4034b38a in FLevelSceneNode::Render () from ./Engine.so #7 0x403504ec in FPlayerSceneNode::Render () from ./Engine.so #8 0x4028abf2 in UGameEngine::Draw () from ./Engine.so #9 0x42f5093b in USDLViewport::Repaint () from /home/as/ut2003_demo/System/SDLDrv.so #10 0x42f4f365 in USDLClient::Tick () from /home/as/ut2003_demo/System/SDLDrv.so #11 0x4028f2e1 in UGameEngine::Tick () from ./Engine.so #12 0x08051b1d in ?? () #13 0x08058b2c in ?? () #14 0x40baf7ee in __libc_start_main () from /lib/libc.so.6 (gdb) thread [Current thread is 1 (Thread 1024 (LWP 31522))] (gdb) up #1 0x43ea593a in _tnl_draw_range_elements (ctx=0x92ac188, mode=4, max_index=32, index_count=48, indices=0x91ef0c0) at t_array_api.c:118 118 tnl->Driver.RunPipeline( ctx ); (gdb) list 113 */ 114 GLuint enabledArrays = ctx->Array._Enabled | (ctx->Array._Enabled >> 16); 115 /* Note that arrays may have changed before/after execution. 116 */ 117 tnl->pipeline.run_input_changes |= enabledArrays & 0xffff; 118 tnl->Driver.RunPipeline( ctx ); 119 tnl->pipeline.run_input_changes |= enabledArrays & 0xffff; 120 } 121 } 122 (gdb) p tnl->Driver.RunPipeline $1 = (void (*)()) 0 (gdb) the same thread that initialised mesa (at least that tested for sse) seems to be used for rendering, but isnt initialised properly in the end. any hints? greetings, Andreas btw: q3a runs fine ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn -- _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
