01913: All sets in model3.c: hang

ID	Category [?]	Severity [?]	Reproducibility	Date Submitted	Last Update
01913	Crash/Freeze	Critical (emulation)	Always	Jun 19, 2008, 22:42	Sep 21, 2012, 10:08

Tester	Firewave	View Status	Public	Platform	MAME (Self-compiled)
Assigned To		Resolution	Fixed	OS
Status [?]	Resolved			Driver	sega/model3.cpp
Version	0.125u6	Fixed in Version		Build	Debug
		Fixed in Git Commit		Github Pull Request #

Summary	01913: All sets in model3.c: hang
Description	They now hang and you have to kill the MAME process. In 125u3 they were reporting different errors: dayto2pe, daytona2 pci_device_set_reg: Device 16, unknown reg 01, 00000002 spikeofe ----------------------------------------------------- Exception at EIP=007E7D08 (model3_palette_r+0x02c4): ACCESS VIOLATION While attempting to read memory at 29051F80 ----------------------------------------------------- EAX=29051F80 EBX=00000000 ECX=0000003E EDX=1A841FFC ESI=1A845EF8 EDI=0000001F EBP=0022F56C ESP=0022F514 spikeout was working
Steps To Reproduce
Additional Information	First time was >0.125u3 regression and fixed in 0.129u2.
Github Commit

Flags
Regression Version	0.130
Affected Sets / Systems	All sets in model3.c

Attached Files

No.03738 Firewave Senior Tester Feb 8, 2009, 20:01	Some time after 0.129u1 they started working again.
No.03778 couriersud Developer Feb 22, 2009, 00:21	These are not hanging as of 0129u4 ==> Resolved
No.04341 Tafoid Administrator May 7, 2009, 22:38 edited on: May 7, 2009, 22:56	This regressed again starting in 0.130 to hang in all sets in model3.c. Reopening.
No.04595 Firewave Senior Tester Jul 2, 2009, 18:50	No more hangs in 0.132u2. Guess it was fixed by the PPC DRC changes.
No.04607 Tafoid Administrator Jul 3, 2009, 01:39 edited on: Jul 3, 2009, 01:42	As of 132u3: spikeout, spikeofe, daytona2, dayto2pe still hang. Once I've been able to test all sets, I'll update bug accordingly.
No.04610 Firewave Senior Tester Jul 3, 2009, 06:42	Those hangs are also noted in the source. It is possible those are hanging again in 0132u3. But I am getting the model3_palette_r crashes noted above in vf3, although it is supposed to be working according to the source.
No.04611 Haze Senior Tester Jul 3, 2009, 08:38	vf3 hardlocks mame here, I actually had to reboot the machine to get back to the desktop..
No.04614 Tafoid Administrator Jul 3, 2009, 13:52	Let's give model3 a bit of a rest. There is a lot of work being done currently and it looks like some attention is there at last - so we'll see if things do not improve soon. I'll leave this bug as a catch all for anything model3 related, rather than creating new ones all the time. I can confirm the hang, on vf3 - just hits a few seconds in.
No.06365 Tafoid Administrator Jul 12, 2010, 14:00 edited on: Jul 12, 2010, 14:04	fvipers2 - crash since 0.132u3: Program received signal SIGSEGV, Segmentation fault. 0x00970d71 in get_texture (machine=0x2583f0, page=0, texx=29, texy=12, texwidth=0, texheight=6, format=6) at src/mame/video/model3.c:645 645 UINT8 grayvalue = pal4bit(texsrc[x] >> 12); ----------------------------------------------------- Exception at EIP=00970D71 (get_texture(running_machine, int, int, int, int, int, int)+0x058e): ACCESS VIOLATION While attempting to read memory at 23681760 ----------------------------------------------------- EAX=23681760 EBX=0010000C ECX=00000000 EDX=245F0028 ESI=00000006 EDI=00000006 EBP=0022F0C0 ESP=0022F078 ----------------------------------------------------- Stack crawl: 0022F0C0: 00970D71 (get_texture(running_machine, int, int, int, int, int, int)+0x058e) 0022F110: 00973A65 (render_one(running_machine, TRIANGLE)+0x017f) 0022F510: 009748F2 (draw_model(running_machine, unsigned int)+0x0d50) 0022F530: 00974C17 (process_link(running_machine, unsigned int, unsigned int)+0x0083) 0022F5B0: 00974D92 (draw_block(running_machine, unsigned int)+0x0144) 0022F5E0: 00974B77 (traverse_list(running_machine, unsigned int)+0x00a8) 0022F600: 00974C30 (process_link(running_machine, unsigned int, unsigned int)+0x009c) 0022F650: 00975073 (draw_viewport(running_machine, int, unsigned int)+0x02ae) 0022F680: 009750A9 (real3d_traverse_display_list(running_machine)+0x002e) 0022F6B0: 00971499 (real3d_display_list_end(running_machine)+0x00ce) 0022F6F0: 008F757F (real3d_dma_callback(running_machine, unsigned int, unsigned int, int, int)+0x0061) 0022F770: 008F72FA (real3d_dma_w(_address_space const, unsigned int, unsigned long long, unsigned long long)+0x01e9) 0022F7E0: 0141C4D0 (write_qword_generic(_address_space const, unsigned int, unsigned long long, unsigned long long)+0x0173) 0022F820: 0142866D (memory_write_dword_64be(_address_space const, unsigned int, unsigned int)+0x0070) C200008F: 1BCC8D4E (__tcf_0+0x1a33100e)
No.08930 Firewave Senior Tester Sep 21, 2012, 07:10	No more hard freezes since quite a while. I also cannot get fvipers2 to crash. It gets stuck at the character selection screen though. So this can be closed.
No.08934 Tafoid Administrator Sep 21, 2012, 10:08	I guess we can clean slate model3.c for now and if anyone comes across hangs or crashes, we can then get them documented.