https://bugs.freedesktop.org/show_bug.cgi?id=36919
--- Comment #12 from José Fonseca <[email protected]> 2011-05-10 05:46:02 PDT --- I reproduced the bug here: (gdb) bt #0 0x00007ffff681aaa7 in _debug_assert_fail (expr=0x7ffff70285d7 "reg.File != TGSI_FILE_NULL", file=0x7ffff70285b0 "src/gallium/auxiliary/tgsi/tgsi_ureg.h", line=863, function=0x7ffff7028c51 "ureg_writemask") at src/gallium/auxiliary/util/u_debug.c:282 #1 0x00007ffff6765607 in ureg_writemask (reg=..., writemask=7) at src/gallium/auxiliary/tgsi/tgsi_ureg.h:863 #2 0x00007ffff6766667 in translate_dst (t=0x7fffffffc1d0, DstReg=0xd217f34, saturate=0 '\000') at src/mesa/state_tracker/st_mesa_to_tgsi.c:310 #3 0x00007ffff6767699 in compile_instruction (t=0x7fffffffc1d0, inst=0xd217f18) at src/mesa/state_tracker/st_mesa_to_tgsi.c:673 #4 0x00007ffff67691bb in st_translate_mesa_program (ctx=0xf444b0, procType=0, ureg=0xf3ef2d0, program=0x111b7010, numInputs=2, inputMapping=0x7fffffffd3a0, inputSemanticName=0x7fffffffd4b0 "\001\003", <incomplete sequence \372>, inputSemanticIndex=0x7fffffffd490 "", interpMode=0x7fffffffd320, numOutputs=0, outputMapping=0x7fffffffd420, outputSemanticName=0x7fffffffd470 "\002", outputSemanticIndex=0x7fffffffd450 "\240\324\377\377\377\177", passthrough_edgeflags=0 '\000') at src/mesa/state_tracker/st_mesa_to_tgsi.c:1230 #5 0x00007ffff66a48c0 in st_translate_fragment_program (st=0xf94810, stfp=0x111b7010, key=0x7fffffffd5a0) at src/mesa/state_tracker/st_program.c:609 #6 0x00007ffff66a4a23 in st_get_fp_variant (st=0xf94810, stfp=0x111b7010, key=0x7fffffffd5a0) at src/mesa/state_tracker/st_program.c:667 #7 0x00007ffff674dc6d in update_fp (st=0xf94810) at src/mesa/state_tracker/st_atom_shader.c:86 #8 0x00007ffff6749e05 in st_validate_state (st=0xf94810) at src/mesa/state_tracker/st_atom.c:196 #9 0x00007ffff67559ed in st_Clear (ctx=0xf444b0, mask=16) at src/mesa/state_tracker/st_cb_clear.c:497 #10 0x00007ffff66cb7df in _mesa_Clear (mask=256) at src/mesa/main/clear.c:250 #11 0x00000000006fbd36 in KX_KetsjiEngine::RenderShadowBuffers(KX_Scene*) () #12 0x00000000006fc1f3 in KX_KetsjiEngine::Render() () #13 0x000000000055ef12 in GPG_Application::processEvent(GHOST_IEvent*) () #14 0x000000000056be14 in GHOST_EventManager::dispatchEvent(GHOST_IEvent*) () #15 0x000000000056bd1d in GHOST_EventManager::dispatchEvent() () #16 0x000000000056bd6c in GHOST_EventManager::dispatchEvents() () #17 0x000000000056b548 in GHOST_System::dispatchEvents() () #18 0x000000000055d906 in main () (gdb) call _mesa_print_instruction(&program->Instructions[i]) LRP OUTPUT[2].xyz, TEMP[1].xxxx, TEMP[0], STATE[1]; (gdb) call _mesa_print_program(program) # Fragment Program/Shader 0 0: MAD_SAT TEMP[1].x, INPUT[3].xxxx, STATE[0].xxxx, STATE[0].yyyy; 1: LRP OUTPUT[2].xyz, TEMP[1].xxxx, TEMP[0], STATE[1]; 2: MOV OUTPUT[2].w, TEMP[0]; 3: END (gdb) The problem is that stfp->Base.Base.OutputsWritten is 0, which causes st_translate_mesa_program() is receiving a bad numOutputs=0, and eventually the assertion. (gdb) p stfp->Base.Base.OutputsWritten $18 = 0 (gdb) p stfp->Base.Base $19 = { Id = 0, String = 0x0, RefCount = 3, Target = 34820, Format = 34933, Resident = 1 '\001', Instructions = 0xd217ed0, InputsRead = 10, OutputsWritten = 0, SystemValuesRead = 0, InputFlags = {0 <repeats 32 times>}, OutputFlags = {0 <repeats 64 times>}, TexturesUsed = {0 <repeats 16 times>}, SamplersUsed = 0, ShadowSamplers = 0, Parameters = 0xc8e2a90, LocalParams = {{0, 0, 0, 0} <repeats 1024 times>}, Varying = 0x0, Attributes = 0x0, SamplerUnits = "\000\001\002\003\004\005\006\a\b\t\n\v\f\r\016\017", SamplerTargets = {TEXTURE_BUFFER_INDEX <repeats 16 times>}, IndirectRegisterFiles = 0, NumInstructions = 4, NumTemporaries = 2, NumParameters = 0, NumAttributes = 0, NumAddressRegs = 0, NumAluInstructions = 1, NumTexInstructions = 0, NumTexIndirections = 1, NumNativeInstructions = 0, NumNativeTemporaries = 0, NumNativeParameters = 0, NumNativeAttributes = 0, NumNativeAddressRegs = 0, NumNativeAluInstructions = 0, NumNativeTexInstructions = 0, NumNativeTexIndirections = 0 } -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. _______________________________________________ mesa-dev mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/mesa-dev
