1) convert framebuffer's ram from 8 bit to 16 bit
2) convert dma command from 8 bit to 16 bit
3) convert gfx command from 8 bit to 16 bit
4) implemented all gfx mode: 1,2,4,8,16 bpp
for now the rendering is only 8 bpp
now we can use this gfx chip for adp driver that use 4bpp
streams and a delta-RLE pre-encoding. Added optimized
case for the Y/Cb/Y/Cr video encoding case. Cleaned up
the code.
Updated avcomp.c to use the new huffman.c functions.
Reworked configuration options to allow for both input
and output of naturally aligned data streams. Updated
chdman and laserdsc to use the new interfaces. New
compression gives an additional 3-7% over previous
attempt and compresses the dummy CHDs down significantly.
Marked madalien and madalina as Imperfect Graphics. They are missing an effect. When you shoot an enemy in the dark, the whole screen lights up. Basically the headlights fill the whole screen.
credit Guru, Derrick Renaud
I did not do much to the following files because I did not know the best way to name them.
aicadsp.h
sid.h
sidenvel.h
sidvoice.h
ymdeltat.h
I did not update structures only used in the src\emu\sound\*.c files. They are only used locally in the file so they were not worth the effort.
Credit Guru.
(This completes adding all of Guru's readmes to the MAME source. There remain ~20 readmes that I still need to compare to what's in MAME to see if they have any additional information, but beyond that, this little project is done!)
---------------------------------------------------
Killer Instinct (SNES bootleg) [Tomasz Slanina]
Final Fight 2 (SNES bootleg) [David Haywood, Tomasz Slanina, Kowal]
New games marked as GAME_NOT_WORKING
------------------------------------
Sonic Blast Man TURBO 2 (SNES bootleg) [Corrado Tomaselli]
- General clean up (merged memory maps, inputs, readme's)
- Marked vbowl program rom as bad, patched to not crash (fixes bug 00492)
- Protection emulation / patches, promoting to working:
dbc, xymg, lhba (fixes bug 02027), drgwrld3 (fixes bug 00491), chindrac, chindrah
New games added or promoted from NOT_WORKING status
---------------------------------------------------
Da Ban Cheng [Pierpaolo Prazzoli, Luca Elia]
Xing Yen Man Guan [Luca Elia, Wei Mingzhi]
New clones added
----------------
Zhong Guo Long (China, V010C) [Luca Elia]
Zhong Guo Long? (Hong Kong, V011H) [Luca Elia]
- Pass chip type around and restore Jarek's original behavior for YM2203/YM2610
- Always recalculate attack rate in set_ar_ksr, fixes Megadrive/Genesis "Batman & Robin"
What a PITA. :) I'll have to go back and do A-J to lower case the handlers to match the new structure names I already did.
Sorry Mess and other port developers, but the code should match what is stated on the official WIKI. But don't worry, I don't plan on updating anything but the sound cores.
- Reverse engineering of custom IC's 06B49P, 06B53P & 09R81P.
- Mapped the missing 3x SN76489.
- Measured and traced all clocks on the board.
- Measured and fixed the interrupt system.
- Implemented timings/clocks from custom 06B49P.
- Added sound support. All regular game sounds/musics are working.
- Implemented the ADPCM samples system through 09R81P + M5205 emulation.
- Added pinouts and technical notes about custom IC's 06B49P, 06B53P & 09R81P.
- Added flip screen mode.
- Inverted the order of double-up difficult DIP switches.
(Seems to be the opposite of the indicated in the instruction sheet).
- Changed 'Key In' to be active LOW instead of HIGH (checked in the PCB).
- Complete memory map and ports scheme.
- Created handlers for USART port and co-processor communication.
- Renamed the sets accordingly.
- Updated all notes.
- Cleaned-up the driver.
Please co-credit Grull Osgo.
You may have to do a clean compile, or delete the affected driver object files.
struct C140interface to c140_interface
struct cem3394_interface to cem3394_interface
struct ES5503interface to es5503_interface
struct ES5505interface to es5505_interface
struct ES5506interface to es5506_interface
struct ssg_callbacks to ssg_callbacks
struct gaelcosnd_interface to gaelcosnd_interface
struct ics2115_interface to ics2115_interface
I'll stop until my posted question about chip names and upper case is answered.
FWIW, I first check for variables named name_interface and rename them as name_cofig, so they don't conflict with the new interface names.
struct YMF262interface to ymf262_interface
struct YM2151interface to ym2151_interface
struct YM2203interface to ym2203_interface
struct YM2608interface to ym2608_interface
struct YM2610interface to ym2610_interface
struct YM2612interface to ym2612_interface
struct YM3812interface to ym3812_interface
struct Y8950interface to y8950_interface
struct YM3526interface to ym3526_interface
struct TMS5110interface to tms5110_interface
struct TMS5220interface to tms5220_interface
struct AICAinterface to aica_interface
Added a new Japanese version of Ghouls'n Ghosts.
This could be a hack since, running on a 91634B board, it must have been made at least three years after the initial release of the game. However, if if it's a hack, it's exceptionally well made since all ROM stickers look original and the B-board DAM63B PAL label is printed on the chip.
Also, hack or not, the code alignment of none of the already supported versions matches this one so it appears to at least be based on a new version.
The dip switches need to be checked. The Lives dips, and the two unused dips after those, also change the starting level. This could be further evidence that the game is a hack.
New clones added
----------------
Dai Makai-Mura (Japan new version)
struct CustomSound_interface to custom_sound_interface
struct Samplesinterface to samples_interface
struct SN76477interface to SN76477_interface
struct AY8910interface to AY8910_interface
The algorithm is more complex than it should be, but I haven't been able to simplify it more.
The keys are still embedded in the source. It would now be reasonable to move them to external files; however, for several games we only have partial keys so the files would need to be BAD_DUMP.
memory regions. Extended error codes to report incorrect memory
spaces, memory names, or memory sizes. Added verification callback
to the debugger to validate CPU and memory region names, as well
as verifying that a requested address space exists for a given
CPU.
Added support for oneshot cheats (those with only an "on" script).
They are activated via UI_SELECT in the cheat menu, and pop up a
message when activated. Also added a "Reset All" item in the cheat
menu to reset all cheats back to their default state, and added
support for UI_SELECT on a non-oneshot cheat so that it resets that
cheat to its default value.
Restored previous behavior that allowed popmessage() messages to
overlay menus and other UI.
and reports if no white flags were seen.
Fixed bug in aviio.c which could cause an infinite loop if attempting
to read sound beyond the end of the file.
Rewrote the white flag detection to use a histogram for outlier detection
and peak identification.
+ plus
- minus
* times or mul
/ div
% mod
! not
~ bnot
&& and
& band
|| or
| bor
^ bxor
lt <
le <=
gt >
ge >=
eq ==
ne !=
Changed cheat escaping to automatically escape && & < and <=
to and band lt and le.
Subject: New russian game
Dear Aaron,
thanks to the very kind and good guys at the "Soviet Arcade Games Museum"
(www.15kop.ru), namely Alex and Max; and thanks to Antro (a good friend of
mine), we have a new dump for a previously unknown Russian game: S.O.S. by Terminal.
It works on the same hardware as Konek (already in MAME), it's a simple rom
swap. I still have to chek about PROMs.
A diff file is included, together with roms.
All the best
Antonio (f205v)
Subject: some diplocations
Hi, enclosed please find a patch to add diplocations to the some
drivers which still missed them. The patch assumes my changes to
drivers starting with C,D,E have been applied (port01.diff -
>port09.diff I sent on Aug 1st... there is at least crshrace.c w
hich could fail to correctly patch, due to possibly different tags,
and probably 1 or 2 more... not many, but maybe annoying anyway).
It adds diplocations to the following drivers: combatsc.c, copsnrob.c,
cosmic.c, crgolf.c, crimfght.c, crospang.c (here I also added service
dips to heuksun and bestri, which allowed me to verify locations),
crshrace.c, dambustr.c, darkseal.c, dassault.c, dblewing.c, ddrible.c,
deadang.c
It also adds locations to the following games: ddragon & ddragon2 on
ddragon.c and docastle, dorunrun & dowild in docastle.c
Regards,
Fabio Priuli
From: Fabio Priuli [mailto:doge.fabio@gmail.com]
Subject: to simplify cischeat inputs
Hi,
the enclosed diff has a twofold effect on cischeat.c:
on one side: it adds diplocations to bigrun, scudhamm,
cischeat & f1gpstar
on the other side: it removes the current hacky handling of
f1gpstar coinage dips (check video/cischeat.c to see what I
mean with hacky!) to use conditional ports instead. BUT in
order to do this, I had to add more possible values for the
PORTCOND in iptport.c. The problem is that f1gpstar has a set
of coinage settings when you set the region to JPN or USA and
another one when you set the region to EUR or FRA. To implement
this I added the following self-explanatory new PORTCOND:
ISLARGER (>), ISNOTLARGER (<=), ISSMALLER (<), ISNOTSMALLER (<=).
Only two were really needed to implement f1gpstar dips, but
the other two seemed costless to me while offering even more
flexibility to the PORT_CONDITION macro.
Also notice that the handling of conditional ports in 'TAB>Dip
Switches' menu doesn't work well (in the current source, not
affected by my patch) with contracted expressions like
PORT_DIPUNKNOWN_DIPLOC: the wrong items remain listed in the
UI menu even when you change the condition! If you leave the
DEF_STR( Unknown ) with ON/OFF cases, everything is displayed
correctly in the menu. However, I was not able to track down
what part of code is responsible for this bug.
Regards,
Fabio Priuli
Subject: another small fix
Hi,
with this last patch I hope to have fixed all regressions I could
had introduced with my past submissions. This time it was a small
problem due to my fix to bug 1565 (alleymas duplicated inputs):
the game was unplayable in u3 (even if not reported to MT).
Regards,
Fabio Priuli
Subject: fix for MT bug 2056
I failed to see this wrong tag for almost 2 days!! One would guess that
looking for the cause of coins problems, an handler with the comment "coin
inputs" should have been noticed immediately. I saw it only this morning :(
However, fix in attachment.
Regards,
Fabio Priuli
PORT_START_TAG to PORT_START.
From: Fabio Priuli [mailto:doge.fabio@gmail.com]
Subject: let's say goodbye to untagged input ports ;)
Hi,
enclosed please find a few patches against 126u3 source
* port01.diff -> port09.diff : these patches continues the
cleanup of inputs in drivers starting with C (the remaining ones),
D and E. No ambiguous situations this time, so I guess there should
be no problem with the changes. Improvements include conditional dips
added to dynduke and diplocations for chinagat, chqflag, circus,
citycon, cloak, contra and dynduke
* tag01.diff -> tag09.diff : these patches add tag to EVERY input port
in drivers starting with F->Z. Notice that only tags are added, no
AM_READ_PORT, nor PORT_INCLUDE (and I saw a lot of drivers would need
it ;) ), nor other cleanups contained in the patches above. I'll keep
cleaning input next week, probably, but at least we can definitely
remove the untagged version of PORT_START :) Notice I was able to
compile the source commenting out the definition of PORT_START, so
I guess every occurrence is covered...
- new tag <comment> (within <cheat>) is read and preserved
when saved
- removed variable attribute from <parameter>; it is now
assumed to be 'param'
- added default attribute for <parameter>
Added new variable 'frame' accessible from cheat scripts. This
can be used in the conditional to prevent execution on every
frame, or for other effects (like displaying temporary messages).
Added new variable 'argindex' which is the index when processing
an <argument> with a count attribute greater than 1. Can be used
in expressions like:
<argument count="3">main.pb@(1000+argindex)</argument>
Reinstated the cheat menu. It now displays all loaded cheats and
allows for them to be activated. All known cheat behaviors should
be working now.
* moved a couple of exp(x) from _step into _reset where appropriate
* use tabulated function values for the transfer function in DSS_INVERTER_OSC
==> 40% speed increase in dkong - from 520% to
$ ./mame64 -noautosave -nothrottle -str 30 -video none dkong -nomt -nosleep
Average speed: 722.48% (29 seconds)
tool which walks through either an AVI or CHD capture from a
laserdisc and ensures that there are no anomalies lurking in
the VBI data.
Added new option to chdman, where if you use -createav with
a special filename (either 2:2 or 3:2) it will create a fake
laserdisc files with the equivalent cadence. This can be used
for testing until full captures are available. Most games used
2:2 apart from Dragon's Lair and Space Ace, which were 3:2.
Note that even though these files are essentially blank, they
are Huffman-compressed, meaning that the maximum compression
ratio you will get is 8:1, so they still end up ~5GB.
Moved error strings from chdman.c to chd.c, where they can be
fetched from any caller via the new chd_error_string()
function.
Updated vbiparse to improve Manchester code decoding. It now
assigns a confidence level per bit based on how solid the
data is. Also added a new function vbi_parse_all which parses
the white flag and all three lines of metadata from a
laserdisc frame. It then compares line 17 and 18 against each
other and selects the most likely candidate based on per-bit
confidence and other factors.
Added frame number display to laserdsc.c. It is off by
default, though most players can be configured to turn it on
(not sure if any actually do). It can be manually toggled via
the backslash key during playback.
Commander (MACH 3 hardware). Old laserdisc hacks are now removed.
The code now talks to the standard Pioneer PR-8210 interface. Also
removed hacky "target list" from the MACH 3 ROMs; this information
is now decoded on the fly from the right channel laserdisc audio.
Other Gottlieb cleanups:
- moved sound inputs to audio/gottlieb and included them in
all relevant drivers
- ordered input ports and ROM definitions consistently
Other laserdisc changes:
- changed PR-8210 interface to work consistently for both Gottlieb
and Stern games
- added audio callback mechanism to allow drivers to peek at the
raw audio streams
- extended the VBI parser to be even more lenient
Instead, they are identified by the region they were loaded in. This
generally means one disk per region. get_disk_handle() now takes a
region name in place of an index. Updated all callers to do this.
The SCSI modules in particular posed a challenge to make this work,
so watch out for potential bugs there.
Changed the IDE interfaces to default to choosing the region with the
same tag as the device for its master disk (assuming no slave). Added
support for specifying explicit master/slave disk regions as part of
the configuration, though slave disks are still not supported yet.
Change the laserdisc interface to no longer require a disk number or
a custom sound device. Both are now assumed to be tagged the same as
the laserdisc device. Updated all drivers accordingly.
maps and behaviors according to the schematics. Hooked up coin
counters.
Added new functions cputag_set_input_line() to assert input lines
via CPU tag instead of index.
Formerly known as pokerigs in iqblock.c (wrong driver).
I've renamed it to cpokert and moved it to csk.c.
It works now.
New clones added
----------------
Champion Poker (v200G) [Luca Elia]
now simply requires an AVI file input.
Relaxed VBI Manchester code clocking to allow for a little bit
of wiggle when finding the clock.
Changed laserdisc support to parse white flags and other data
from the VBI data directly, rather than relying on the textual
metadata. Expanded video frame cache to 3. Changed the way
frames are assembled to decrease the likelihood of getting an
interlaced weave. Fixed sound creation so that it is done at
reset time instead of device start, when the sound devices aren't
yet live.
Fixed bug in winwork that caused the creation of single work
items to return NULL, and thus lead to massive memory leaks when
using A/V CHDs.
Removed the Enable line from the DISCRETE_ADJUSTMENT modules. They should never be disabled.
Some before/after speed tests:
dkong - from 203% to 204%
firetrk - from 418% to 429%
phoenix - from 253% to 257%
polaris - from 448% to 476%
(roms in current/j/jingbell.zip)
The background (slot reels) is not emulated properly,
hence it's not playable and marked as not working.
New games marked as GAME_NOT_WORKING
------------------------------------
Jingle Bell [Luca Elia]
Like some other Space games (mosaic.c), it is protected by an MCU
that supplies the addresses to jump to.
I've done some basic simulation that at least shows the service mode.
This is a candidate for decapping / trojaning, or for Nicola to examine
like he did with Mosaic ;-)
Note to Guru: u118 is probably bad, it fails the rom test.
New games marked as GAME_NOT_WORKING
------------------------------------
Quiz Punch 2 [Luca Elia]
to whole functions. The other one I'm not so sure about. Commented
code is usually hilighted differently, making it very easy to spot.)
Hi mamedev,
This set of patches has one aim only, to identify chunks of code that
have been disabled by the use of C/C++ comments, and to disable them
instead by using the preprocessor. The C comment approach to
disabling code isn't safe (embedded comments will trip it up), and the
C++ comment approach isn't elegant (you shouldn't need to touch every
line to disable a chunk of code). Using #if...#endif is preferable
always, excepting perhaps if (0) { ... }.
The patch has three parts. The first only handles cases where full
functions were disabled. The second handles cases where parts of
functions were disabled. The third then tries to restore the
whitespace that the use of comments converted from tabs to spaces via
srcclean.exe. It also cleans up the whitespace in a handful of the
files in areas outside of the original two patches.
~aa
Subject: a few cleanups
This patch contains:
- removal of unprintable chars (newlines) in SH-2 disassembler (I
submitted this in the past nd it wasn't included)
- a few unnecessary checks after malloc_ort_die() calls
- changes two romload.c warnings to use GAMENOUN instead
- adds "deprecat.h" in a few src/mame/drivers files (would be
necessary, if the debugger.h one would be removed)
- cleans up the mame.mak by adding all missing defines and grouping
them based on cpu.mak
- renamed video_exit() to winvideo_exit() for consistency in function
names
files. Removed old option 'cheat_file'.
Updated xmlfile.c to specify the line number for each node, for more
accurate error reporting.
Removed old cheat.c/.h and replaced them with code that will read and
write the new XML-based cheat format using the new cheatpath option
(cheats still need to be enabled in order to load at all). Note that
the cheats are not implemented yet, but they are fully parsed. To test
saving, whenever a cheat file is successfully loaded, it is immediately
written back out as 'output.xml' to the cheat directory for validation.
Game is NOT WORKING because it uses an H8/3644 MCU plus an IDE HDD to emulate the home PS1 version's redbook audio tracks. Cross your fingers for decapping.
Roms are in current/g/galaxi.zip.
New games added or promoted from NOT_WORKING status
---------------------------------------------------
Galaxi [Luca Elia]
The ROL/ROR/SHL/SHR opcodes provide the carry flags typically generated by
almost all CPUs. The RORC/ROLC opcdes map directly to the rotate through
carry of most CPUs as well.
auto-filled by the device manager before starting the device. This
simplifies the auto-binding process for devices.
Added new selector for device-specific ROM regions. Not yet wired
up in the ROM loader.
1) refactors many global and static variables in video\dc.c into a
structure to hold the tile accelerator state
2) textures should be better decoded with more pixel formats
3) adds placeholders for some register probably used by the dimm board
4) corrects a little error in a rom board register (very important, now at
least the gdrom games try to start)
Sent: Sunday, July 27, 2008 2:46 PM
To: submit@mamedev.org
Subject: misc patches
Hi,
enclosed please find a few more patches based on 126u2. It's just a few
minor bugfixes because I'm quite busy and I still have to test the other
changes to tagged inputs (I will send them in the next few days)
In misc2.zip you'll find a few corrections to my previous patches and a
few new bugfixes:
* cachat.diff: It fixes bug 2046. This is a bug I introduced misreading
the default value of the flip dip when updating taito_l.c to use
taitoipt.h definitions. My bad, sorry.
* witch.diff: It fixes another small 'mistake' I introduced when tagging
the ports in witch.c. I searched and replaced a few too many 'port'
occurrences, making some comments unreadable. Again, sorry for the
inconvenience.
* btime.diff: It's an updated version of my previous patch, following
your suggestions. 'Off' comes always first now and manual settings which
do not work on the emulated game are commented out
* skyfox.diff: It fixes bug 2050. Being at it, I also added dip
locations to the driver and updated memory maps to use AM_READ_PORT
* combatsc.diff: It fixes bug 456, solving both the regression which
made disappear all sprites (they were covered by the non-transparent
text layer since 118u1) and the corruption of the graphics in title
screen (due to the wrong drawing order). Mind that bug 400 (scrolling
problems present since 0.79) is not fixed yet, unfortunately.
Regards,
Fabio
From: Atari Ace [mailto:atari_ace@verizon.net]
Subject: [patch] Stricter ADDRESS_MAP checks
Hi mamedev,
The current implementation of address maps is lacking some useful
checks of the initial tokens. In particular, it doesn't validate that
a particular entry doesn't try to define the same handler multiple
times.
The attached patch adds this and some other validations as fatalerrors
in address_map_detokenize, and fixes the affected maps. The errors
generally fall into the following categories.
1. AM_RAM AM_WRITE(...). Should have been AM_RAM_WRITE(...)
2. AM_RAM AM_READWRITE(...). The AM_RAM is a NOP.
3. AM_RAM AM_RAMBANK(...). The AM_RAM is a NOP.
4. AM_ROM AM_ROMBANK(...). The AM_ROM is a NOP.
5. AM_ROM AM_READ(...). The AM_ROM is a NOP.
One peculiar error was in equites.c, where there were two AM_BASE
entries.
~aa
second SH2 for ST-V games.
Fixed buffer overrun in latest SCSP code. Not sure how that ever
worked.
colmns97 still asserts in SCSPDSP, but this happened before the
region change.
- Pre-defined CPU and SND clocks.
- Switched the color system to RESNET calculations.
- Completed the remaining DIP switches.
- Added lamps support. Created a layout to show them.
- Changes on the interrupt system (need to be verified on the PCB).
- Renamed the graphics regions to more descriptive names.
- Corrected the manufacturer's name.
- Splitted the driver to driver + video.
- Updated technical notes.
integer value, regions are now referred to by a region class and
a region tag. The class specifies the type of region (one of CPU,
gfx, sound, user, disk, prom, pld) while the tag uniquely specifies
the region. This change required updating all the ROM region
definitions in the project to specify the class/tag instead of
region number.
Updated the core memory_region_* functions to accept a class/tag
pair. Added new memory_region_next() function to allow for iteration
over all memory regions of a given class. Added new function
memory_region_class_name() to return the name for a given CPU
memory region class.
Changed the auto-binding behavior of CPU regions. Previously, the
first CPU would auto-bind to REGION_CPU1 (that is, any ROM references
would automatically assume that they lived in the corresponding
region). Now, each CPU automatically binds to the RGNCLASS_CPU region
with the same tag as the CPU itself. This behavior required ensuring
that all previous REGION_CPU* regions were changed to RGNCLASS_CPU
with the same tag as the CPU.
Introduced a new auto-binding mechanism for sound cores. This works
similarly to the CPU binding. Each sound core that requires a memory
region now auto-binds to the RGNCLASS_SOUND with the same tag as the
sound core. In almost all cases, this allowed for the removal of the
explicit region item in the sound configuration, which in turn
allowed for many sound configurations to removed altogether.
Updated the expression engine's memory reference behavior. A recent
update expanded the scope of memory references to allow for referencing
data in non-active CPU spaces, in memory regions, and in EEPROMs.
However, this previous update required an index, which is no longer
appropriate for regions and will become increasingly less appropriate
for CPUs over time. Instead, a new syntax is supported, of the form:
"[tag.][space]size@addr", where 'tag' is an optional tag for the CPU
or memory region you wish to access, followed by a period as a
separator; 'space' is the memory address space or region class you
wish to access (p/d/i for program/data/I/O spaces; o for opcode space;
r for direct RAM; c/u/g/s for CPU/user/gfx/sound regions; e for
EEPROMs); and 'size' is the usual b/w/d/q for byte/word/dword/qword.
Cleaned up ROM definition flags and removed some ugly hacks that had
existed previously. Expanded to support up to 256 BIOSes. Updated
ROM_COPY to support specifying class/tag for the source region.
Updated the address map AM_REGION macro to support specifying a
class/tag for the region.
Updated debugger windows to display the CPU and region tags where
appropriate.
Updated -listxml to output region class and tag for each ROM entry.
- Fixed ADPCM decoder to remove bad optimization and match YMZ280B. DC offset is fixed and output quality is higher
- Improved handling of slot monitor and timers B and C
- Simplified interpolation code
- Rewrote core logic, communications, and interfaces.
- Added three parallel memory moves to the disassembler.
- Initial interrupt logic in place.
Plygonet.c updates.
- All communication hacks have been removed.
- Memory maps have been temporarily reverted while new DSP56k cpu core catches up.
Subject: [patch] Conditional code cleanup resubmit
Hi mamedev,
This is a resubmit of a previous patch. The earlier version would not
compile with 32-bit MSVC, due to the fact that its linker required
external dependencies in dead code to be met before dead code
elimination was done, causing linker errors. The proper fix for this
would be to add the necessary dependencies, so I instead simply left
the conditional code in place in winalloc.c and chd.c.
~aa
Original submission email below:
----
Conditionally compiled code tends to bitrot, so MAME should try to
avoid it as much as possible. I sent a patch six months ago to
eliminate conditional code associated with logging, here's another
patch that does more of this. Some notes:
1. drc_ops.c: I couldn't find a LOG_CODE anywhere, so I used if (0).
2. romload.c: I converted all the users of debugload to use
LOG((...)) instead, following the traditional conditional logging
pattern.
3. windows/sound.c: I eliminated the separate sound log and directed
the few outputs to the error log.
~aa
Subject: [patch] .data removals to fix reset/multisession bugs
Hi mamedev,
One nice artifact of properly constifying data structures in MAME is
that it makes it relatively easy to spot a class of reset/multisession
bugs, namely that almost any object in .data is probably in error.
Unless the value is properly initialized in a reset routine the
initial non-zero value can't be relied upon, so there's no need to
have a non-zero value to begin with.
With that in mind, here's a patch to move more items out of .data by
either applying const, removing the non-zero initializer (if its
overwritten by init/reset) or by adding appropriate initialization
code. In most cases I tried to add initialization code to a reset
routine, but in some cases I chose an init routine, possibly leaving a
reset bug intact.
Some interesting bits:
1. tms9900 core. The use of .data to initialize the irq_level wasn't
correct in some cases as the layout of the structure was core
dependent.
2. bfcobra.c. By introducing a VIDEO_START routine a hack in
VIDEO_UPDATE could be removed.
~aa
