05051: timecrs2: MAME debugger MIPS3 cpu trace fails.

ID	Category [?]	Severity [?]	Reproducibility	Date Submitted	Last Update
05051	Debugger	Major	Always	Oct 29, 2012, 17:02	Oct 31, 2012, 17:34

Tester	hap	View Status	Public	Platform
Assigned To		Resolution	Open	OS	Windows Vista/7 (64-bit)
Status [?]	Acknowledged			Driver	namco/namcos23.cpp
Version	0.147u1	Fixed in Version		Build
		Fixed in Git Commit		Github Pull Request #

Summary	05051: timecrs2: MAME debugger MIPS3 cpu trace fails.
Description	MAME debugger MIPS3 cpu trace fails. This happens in both the DRC core, and the C core. namcos23.c timecrs2 is the test case here, but it's only logical that a lot more drivers/games are affected. BTW you won't need a DEBUG build for this. Anyway: - mame64 timecrs2 -debug - trace tim2.txt - a couple seconds after: trace off The disassembly shows up fine in the debugger window, so I'm expecting a tracelog in tim2.txt But instead, it's completely filled with a bogus opcode, while the programcounter looks ok. 80025CB8: sd $ra,-$1($ra) 80025CBC: sd $ra,-$1($ra) 80025CC0: sd $ra,-$1($ra) 80115C24: sd $ra,-$1($ra) 80115C28: sd $ra,-$1($ra) etc etc I also can't get the 'dump' command work with MIPS3, while the memory window does show the right values.
Steps To Reproduce
Additional Information
Github Commit

Flags
Regression Version
Affected Sets / Systems	timecrs2

Attached Files

No.09064 hap Developer Oct 31, 2012, 17:34	Until it is fixed, it looks like I can get a workable cputrace for namcos23 if I comment out emu/debug/debugcpu.c lines 922 and 923: /* translate to physical first */ // if (!debug_cpu_translate(space, TRANSLATE_FETCH_DEBUG, &address)) // return result;