device_rtc_interface: Cleanups and refinements (nw)
- Give RTCs their own phase of machine initialization, right after NVRAM loading
- Make RTC feature flag overrides const, including one new one
- Make rtc_clock_updated a required override
* Make stream_format return characters printed
* Add iostreams with std::vector storage
* Move to type-safe templates for logerror and popmessage
* Remove now-unnecessary I64FMT from calls to logerror/popmessage
* Put some lib/util stuff in util:: namespace
* Some fixes to Japanese translation
Per-game parameters are (tag, value) pairs that follow the same tag
structure than everything else. They're added within a ROM section
with ROM_PARAMETER(tag, value). You look them up through the device
method parameter(tag) which returns the string, or "" if not present.
A *lot* of drivers can benefit from that stuff. It should kill most
game-name keyed tables, and the alternative approach of
bazillon-of-driver-inits.
Possible evolutions, if someone wants to do it:
- output them in -listxml
- push the softlist parameters present in the xml through that
- maybe allow to change/set them through the command line
-Updated LUA engine to run in machine_manager instead of being initialized per machine
-Added "-console" option so emulator can be started with LUA enabled console
-Update LUA to version 5.2.3
-Enabled SQLite3 to be compiled and added LUA module for it
Moved graphics decoding to a new device interface class: device_gfx_interface.
The gfxdecode device is now a device that simply inherits this interface and
does nothing else. Devices that draw tilemaps or sprites using gfx_elements
should in time be updated to use this interface rather than connect to a
machine-global gfxdecode device. Updated toaplan_scu.c as an example (also
fixed off-by-one sprite alignment in twincobr and rallybik while I was at it).
gfx_elements are normally created in interface_post_start(), making it
possible to dynamically create or modify the graphics decoding info during
device_start() if you need to. On the other hand, if you need the gfx_elements
during device_start(), you can directly call decode_gfx() to create them early.
This interface also provides a standard and init-order-safe way to connect to
a palette device (similarly to how device_video_interface helps devices
connect to a screen), so it's handy for any device that does palettized
drawing even if it doesn't use gfx_elements. Updated k053250.c as an example
of this usage.
gfxdecode info entries can now reference shared RAM regions by tag as well as
ROM regions, automatically handle endianness, and have some other new
capabilities. Updated nemesis.c and pgm.c to showcase the new features.
Removed validate_display() (it was just a commented out stub already) since
its only function, checking that drivers don't have an ind16 screen without
a palette, is now done by screen_device::device_validity_check().
Updated obsolete comments about GFXLAYOUT_RAW (cps1.c hasn't used raw gfx for
years, and "to save memory" is no longer a good reason to use it)
more generally useful than just in tilemaps. Code is now in memarray.*
Converted the Atari RLE motion objects device from a half-assed
device into a full-assed device, leveraging the memory_array class.
house a screen tag and to find the screen at startup, providing an m_screen
object that can be used. One nice feature is that if there is only one
screen and no screen has been specified, it will auto configure to that
screen. This removes the need to explicitly specify a screen in the
configuration for a large chunk of drivers (though doing so never hurts).
A new macro MCFG_VIDEO_SET_SCREEN is provided, though devices are
encouraged to define their own that maps there so it is obvious which
device is being targeted. The device_video_interface's validation
function will error if an invalid screen is specified or if no screen
is provided but there are multiple screens present.
Updated all devices that currently had an m_screen in them to use the
device_video_interface instead. This also has the nice benefit of flagging
video-related devices for categorization purposes. It also means all
these devices inherit the same screen-finding behaviors. For devices
that had interfaces that specified a screen tag, those have been removed
and all existing structs updated.
Added an optional_device<screen_device> m_screen to the base driver_device.
If you name your screen "screen" (as most drivers do), you will have free
access to your screen this way.
Future updates include:
* Updating all devices referencing machine.primary_screen to use the
device_video_interface instead
* Updating all drivers referencing machine.primary_screen to use the
m_screen instead
* Removing machine.primary_screen entirely
Tilemap system: numerous changes:
* Moved remaining legacy macros and typedefs to tilelgcy.h. This revealed
a few drivers mixing and matching modern & legcy, which have now been
fixed.
* Changed get info callback signature to no longer pass the user_data
pointer, but instead pass a reference to the tilemap object itself.
Updated those few drivers using user_data to pull it out of the
tilemap object with the new user_data() getter method.
* Changed get info and mapping callbacks to be device_delegates so that
they can be described at config time.
* Added tilemap_memory object that is used internally for reading/
writing to memory that backs a tilemap. This object is used to track
a memory pointer that backs tilemap memory, and also is designed to
transparently handle all bus width and endianness associated with
reading and writing data in a tilemap.
* Incorporated two tilemap_memory objects (basemem and extmem) into the
tilemap object and added accessors to them, as well as read/write
handlers for reading/writing to entries stored in the memory. This
means that tilemap get info callbacks can now easily read data out of
the tilemap in a generic way.
* Rejiggered the initialization sequence for tilemap objects so that
the tilemap_manager is not required to be present at instantiation.
* Created a new tilemap_device, which can be used to declare a tilemap
in the machine config, and which also is a tilemap object itself.
The tilemap device will look for shared memory regions called
"<tag>" and "<tag>_ext" and automatically plug them into the tilemap.
The device also provides write handlers that can be used to write
to the tilemap memory and mark tiles dirty, saving the need for each
driver to write their own.
Device system: moved required/optional device finders to a new header
devfind.h.
Atari drivers: removed all playfield and alpha memory and tilemap
variables, apart from those needed by atarivc-using games (this will
become a device in a future update). Updated all Atari 16-bit drivers to
use the new tilemap_device instead, which provides all the needed
functionality in a more generic way.
types. Created a binding_type_exception which is thrown when
a bind attempt fails due to mismatched types.
Added helper templates to driver_device to wrap legacy
device read/write handlers into driver_device member functions.
This should help move some things forward until more common
code is converted into proper devices.
Introduce new module devcb2 which contains modernized
versions of devcb. Compared to previous implementation
this one is simpler overall, trampolining calls through
a single internal set of adapter functions. The new
versions are also designed to be specified in the
machine_config rather than in structures, so they are
no longer simple POD types. Additional new/changed
features:
* reads and writes can map to delegates for line or 8/16/32/64-bit
* reads and writes can map to an I/O port
* reads can be mapped to a constant value, with or without logging
* writes can be mapped to a device's input line
* all reads/writes can have a shift, mask, and/or xor applied
* devices can opt to make the functions safe-if-NULL when resolving
* only member function types are supported
Rewrote the YM2151 interface to be fully modernized, and
removed the ym2151_interface struct in favor of inline configs
using the new devcb2 mechanism. In many cases, removed
no longer needed trampolines, instead taking advantage of
direct support for input line writes.
non-templatized helper class so that the code can live
co-located, rather than invading device.h.
Changed the read/write delegates to derive from
device_delegate. Updated the address map macros to create
these properly.
Removed remnants of the old AM_BASE/SIZE macros from the
memory system.
intended differences from previous behavior. For drivers,
the main change is that input_port_read() no longer exists.
Instead, the port must be fetched from the appropriate device,
and then read() is called.
For member functions, this is actually simpler/cleaner:
value = ioport("tag")->read()
For legacy functions which have a driver_data state, it goes:
value = state->ioport("tag")->read()
For other legacy functions, they need to fetch the root device:
value = machine.root_device().ioport("tag")->read()
The other big change for drivers is that IPT_VBLANK is gone.
Instead, it has been replaced by a device line callback on the
screen device. There's a new macro PORT_VBLANK("tag") which
automatically points things to the right spot.
Here's a set of imperfect search & replace strings to convert
the input_port_read calls and fix up IPT_VBLANK:
input_port_read( *\( *)(machine\(\)) *, *([^)]+ *\))
ioport\1\3->read\(\)
input_port_read( *\( *)(.*machine[()]*) *, *([^)]+ *\))
\2\.root_device\(\)\.ioport\1\3->read\(\)
(state = .*driver_data[^}]+)space->machine\(\)\.root_device\(\)\.
\1state->
(state = .*driver_data[^}]+)device->machine\(\)\.root_device\(\)\.
\1state->
input_port_read_safe( *\( *)(machine\(\)) *, *([^,]+), *([^)]+\))
ioport\1\3->read_safe\(\4\)
IPT_VBLANK( *\))
IPT_CUSTOM\1 PORT_VBLANK("screen")
Renamed driver overrides to MCFG_MACHINE/SOUND/VIDEO_START_OVERRIDE to
explicitly indicate they are overriding the default behavior.
Put liberatr back the way it used to be.
Normalized soundlatch helper function names.
Created delegates for machine/sound/video_start/reset callbacks and
added necessary infrastructure to use them going forward.
reading is now done using the FLAC wrapper. There is now a
samples_iterator class to centralize the logic for handling
the sample list walking.
Also redid the cheesy half-baked votrax device since it relied
on some old samples-based handling. Until we have a real
implementation, it would be good to route the various clients
through the current one to at least wire it up properly, even
if it just plays samples in the end. Will look into that
shortly.
from 4 to 5. This means any diff CHDs will no longer work. If you
absolutely need to keep the data for any existing ones you have,
find both the diff CHD and the original CHD for the game in question
and upgrade using these commands:
rename diff\game.dif diff\game-old.dif
chdman copy -i diff\game-old.dif -ip roms\game.chd -o diff\game.dif -op roms\game.chd -c none
Specifics regarding this change:
Defined a new CHD version 5. New features/behaviors of this version:
- support for up to 4 codecs; each block can use 1 of the 4
- new LZMA codec, which tends to do better than zlib overall
- new FLAC codec, primarily used for CDs (but can be applied anywhere)
- upgraded AVHuff codec now uses FLAC for encoding audio
- new Huffman codec, used to catch more nearly-uncompressable blocks
- compressed CHDs now use a compressed map for significant savings
- CHDs now are aware of a "unit" size; each hunk holds 1 or more units
(in general units map to sectors for hard disks/CDs)
- diff'ing against a parent now diffs at the unit level, greatly
improving compression
Rewrote and modernized chd.c. CHD versions prior to 3 are unsupported,
and version 3/4 CHDs are only supported for reading. Creating a new
CHD now leaves the file open. Added methods to read and write at the
unit and byte level, removing the need to handle this manually. Added
metadata access methods that pass astrings and dynamic_buffers to
simplify the interfaces. A companion class chd_compressor now
implements full multithreaded compression, analyzing and compressing
multiple hunks independently in parallel. Split the codec
implementations out into a separate file chdcodec.*
Updated harddisk.c and cdrom.c to rely on the caching/byte-level read/
write capabilities of the chd_file class. cdrom.c (and chdman) now also
pad CDs to 4-frame boundaries instead of hunk boundaries, ensuring that
the same SHA1 hashes are produced regardless of the hunk size.
Rewrote chdman.exe entirely, switching from positional parameters to
proper options. Use "chdman help" to get a list of commands, and
"chdman help <command>" to get help for any particular command. Many
redundant commands were removed now that additional flexibility is
available. Some basic mappings:
Old: chdman -createblankhd <out.chd> <cyls> <heads> <secs>
New: chdman createhd -o <out.chd> -chs <cyls>,<heads>,<secs>
Old: chdman -createuncomphd <in.raw> <out.chd> ....
New: chdman createhd -i <in.raw> -o <out.chd> -c none ....
Old: chdman -verifyfix <in.chd>
New: chdman verify -i <in.chd> -f
Old: chdman -merge <parent.chd> <diff.chd> <out.chd>
New: chdman copy -i <diff.chd> -ip <parent.chd> -o <out.chd>
Old: chdman -diff <parent.chd> <compare.chd> <diff.chd>
New: chdman copy -i <compare.chd> -o <diff.chd> -op <parent.chd>
Old: chdman -update <in.chd> <out.chd>
New: chdman copy -i <in.chd> -o <out.chd>
Added new core file coretmpl.h to hold core template classes. For now
just one class, dynamic_array<> is defined, which acts like an array
of a given object but which can be appended to and/or resized. Also
defines dynamic_buffer as dynamic_array<UINT8> for holding an
arbitrary buffer of bytes. Expect to see these used a lot.
Added new core helper hashing.c/.h which defines classes for each of
the common hashing methods and creator classes to wrap the
computation of these hashes. A future work item is to reimplement
the core emulator hashing code using these.
Split bit buffer helpers out into C++ classes and into their own
public header in bitstream.h.
Updated huffman.c/.h to C++, and changed the interface to make it
more flexible to use in nonstandard ways. Also added huffman compression
of the static tree for slightly better compression rates.
Created flac.c/.h as simplified C++ wrappers around the FLAC interface.
A future work item is to convert the samples sound device to a modern
device and leverage this for reading FLAC files.
Renamed avcomp.* to avhuff.*, updated to C++, and added support for
FLAC as the audio encoding mechanism. The old huffman audio is still
supported for decode only.
Added a variant of core_fload that loads to a dynamic_buffer.
Tweaked winwork.c a bit to not limit the maximum number of processors
unless the work queue was created with the WORK_QUEUE_FLAG_HIGH_FREQ
option. Further adjustments here are likely going to be necessary.
Fixed bug in aviio.c which caused errors when reading some AVI files.
macros with bitmap->pix* functions, and moved bitmap_fill() to bitmap->fill()
among other similar changes. Bitmap fields now only available via accessors.
Replaced sect_rect with &= and union_rect with |= operators for rectangle
classes. Some general cleanup as a result of these changes. [Aaron Giles]
