- put memory initialization in src/emu/emualloc.c behind NO_MEMORY_INITIALIZATION define
- disable memory initialization when msan is used
- use origin tracking for msan when clang is used
Tested on: Linux x64, Mac OS X 10.8, Windows 8 x64 w/latest toolchain for both OSD=windows and OSD=sdl. MIDI should auto-disable on other OSes (*BSD) but there may be problems. Works fine on Linux when set to use the system's copy of the library as well, which many distros prefer.
fixed uninitialized field in ioport_field.m_chars (fixes stack
corruption with three PORT_CHAR definitions) / added some sanity
checks for PORT_CHAR adding and lookup by Firewave
Intel 28F320J3D flash support by Lukasz Markowski
snapquik.c fixed uninitialized variable that causes random assertion
failure by Sandro Ronco
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Instead, it has been replaced by a device line callback on the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")
- enforced short names for slot card devices
- updated validation, romverify and listxml output accordingly
- slotoptions now also contain shortnames so it's possible to link
slot option and device
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.
and paths consistently for devices, I/O ports, memory
regions, memory banks, and memory shares. [Aaron Giles]
NOTE: there are likely regressions lurking here, mostly
due to devices not being properly found. I have temporarily
added more logging to -verbose to help understand what's
going on. Please let me know ASAP if anything that is being
actively worked on got broken.
As before, the driver device is the root device and all
other devices are owned by it. Previously all devices
were kept in a single master list, and the hierarchy was
purely logical. With this change, each device owns its
own list of subdevices, and the hierarchy is explicitly
manifest. This means when a device is removed, all of its
subdevices are automatically removed as well.
A side effect of this is that walking the device list is
no longer simple. To address this, a new set of iterator
classes is provided, which walks the device tree in a depth
first manner. There is a general device_iterator class for
walking all devices, plus templates for a device_type_iterator
and a device_interface_iterator which are used to build
iterators for identifying only devices of a given type or
with a given interface. Typedefs for commonly-used cases
(e.g., screen_device_iterator, memory_interface_iterator)
are provided. Iterators can also provide counts, and can
perform indexed lookups.
All device name lookups are now done relative to another
device. The maching_config and running_machine classes now
have a root_device() method to get the root of the hierarchy.
The existing machine->device("name") is now equivalent to
machine->root_device().subdevice("name").
A proper and normalized device path structure is now
supported. Device names that start with a colon are
treated as absolute paths from the root device. Device
names can also use a caret (^) to refer to the owning
device. Querying the device's tag() returns the device's
full path from the root. A new method basetag() returns
just the final tag.
The new pathing system is built on top of the
device_t::subtag() method, so anyone using that will
automatically support the new pathing rules. Each device
has its own internal map to cache successful lookups so
that subsequent lookups should be very fast.
Updated every place I could find that referenced devices,
memory regions, I/O ports, memory banks and memory shares
to leverage subtag/subdevice (or siblingtag/siblingdevice
which are built on top).
Removed the device_list class, as it doesn't apply any
more. Moved some of its methods into running_machine
instead.
Simplified the device callback system since the new
pathing can describe all of the special-case devices that
were previously handled manually.
Changed the core output function callbacks to be delegates.
Completely rewrote the validity checking mechanism. The
validity checker is now a proper C++ class, and temporarily
takes over the error and warning outputs. All errors and
warnings are collected during a session, and then output in
a consistent manner, with an explicit driver and source file
listed for each one, as well as additional device and/or
I/O port contexts where appropriate. Validity checkers
should no longer explicitly output this information, just
the error, assuming that the context is provided.
Rewrote the software_list_device as a modern device, getting
rid of the software_list_config abstraction and simplifying
things.
Changed the way FLAC compiles so that it works like other
external libraries, and also compiles successfully for MSVC
builds.
out of log:
This way it is possible to link two or more separated executables with different
copyright/xml out/name/... in one compilation, just one step closer...
- Removed LOADBYNAME, since it is deprecated by using per device rom load_software_part_region
- Created makedev tool to generate array of devices, and created lst file according to current devices usage.
- Changed listxml command to output device roms too
using one cothread for each executable device. This functionality is
strictly internal and should result in no changes in scheduling
behavior. However, it now becomes possible to exit out of a device's
execution in the midst of an instruction by calling
machine().scheduler().make_active(). Note that this is somewhat dangerous
for cores that have logic at the start of their execute loop to check for
interrupts or other conditions, as those checks will be skipped upon
resume, so consider this sort of early exiting to be highly experimental
for now. [Aaron Giles, Miodrag Milanovic, byuu]
loader rewrite, which is still in progress....)
Replaced mamedriv.c with a new driver list mechanism that is generated
by the build tools. The emulator core now expects the presence of a
file called src/$(TARGET)/$(SUBTARGET).lst which is just a raw list of
driver names, one per line. C and C++ comments are still permitted.
This file is parsed by a new build tool makelist which extracts the
driver names, sorts them, and generates a file called drivlist.c, which
is consumed by the core. [Aaron Giles]
Added new osdcore function osd_malloc_array() which is identical to
osd_malloc() but obviously hints that the underlying allocation is for
an array. Updated all callers to use the appropriate form. Modified the
Windows allocator to only use guard pages for array-style allocations,
allowing us to enable them once again in debug builds. [Aaron Giles]
Created new static class driver_list to wrap accesses to the list of
available drivers. Improved speed of driver lookups by relying on the
presorting done by makelist. [Aaron Giles]
Created helper class driver_enumerator as a helper for iterating through
the list of drivers. This class supports basic filtering and iteration,
and also serves as a temporary cache of machine_configs. [Aaron Giles]
Created cli_frontend object to wrap all the CLI handling code in
clifront.c. Updated/simplified all the code to take advantage of the
driver_enumerator. [Aaron Giles]
Created media_auditor object to wrap all the auditing functions in
audit.c. Updated all users to the new interface. Note that the new
auditing mechanism is slightly out of sync with the romload code in
terms of finding ROMs owned by devices, so it may mis-report some
issues until the new ROM loading code is in. [Aaron Giles]
Added concept of a per-device searchpath. For most devices, their
searchpath is just the short name of the device. For driver_devices, the
searchpath is driver[;parent[;bios]]. This searchpath will eventually be
used by the rom loader to find ROMs. For now it is used by the media
auditor only. [Aaron Giles]
Created info_xml_creator object to wrap all the info generation functions
in info.c. Converted the file to C++ and cleaned up the input processing
code. [Aaron Giles]
(not for whatsnew ... Known issues: auditing of CHDs appears busted, and
debug builds report unfreed memory if you use the built-in game picker)
Not for whatsnew: I added -Wno-conversion unconditionally to disable the
warnings Thomas reported. That setting is the default for GCC out-of-the-box
but apparently not on NetBSD. As far as I know it shouldn't cause a problem
with any GCC version back to at least 4.0.0 so we're safe even on PPC OSX,
but do let me know if hilarity ensues.
Does basic dependency analysis and supports excluding paths.
Makefile now supports a depend target which will run makedep
on the src/$(TARGET) tree and create dependencies. It deliberately
excludes the root of src/emu as well as the osd directories in
order to keep the dependency file down to a reasonable size, so
if a core header file changes, you still need to clean and build
all.
DECLARE_LEGACY_CPU_DEVICE and DEFINE_LEGACY_CPU_DEVICE. Changed CPUs
to be their own device types, rather than all of type CPU with a
special internal subtype. Note that as part of this process I removed
the CPU_ prefix from the ALL-CAPS device name, so CPU_Z80 is just
plain old Z80 now. This required changing a couple of names like
8080 to I8080 so that there was an alphabetic first character.
Added memory interfaces to the list of fast-access interfaces. To do
this properly I had to add a separate method to devices which is
called immediately after construction, when it is possible to perform
dynamic_casts on fully-constructed objects. (This is just internal,
no changes necessary to the devices themselves.)
Some additional notes:
* SH2 and SH4 had typedefs that conflicted with their CPU_-less names
so I bulk renamed to structures to sh2_state and sh4_state; RB, feel
free to choose alternate names if you don't like 'em
* SCSP was caught doing something to the 3rd indexed CPU. Since several
systems that use SCSP don't even have 3 CPUs, I had no idea what
this was supposed to do, so I changed to it reference "audiocpu"
assuming that stv was the assumed target. This is really gross and
should be a configuration parameter, not a hard-coded assumption.
* created dynamic_bind<> template class to handle dynamically binding to
optionally-supported functions
* wrapped stack walking code in a class
* wrapped symbol lookup code in a class
* added support for parsing objdump-produced symbol dumps which include
non-global functions for much better stack dumps and profiling in gcc builds
Also: modified makefile for win32 targets to automatically run objdump and
produce a .sym file if SYMBOLS is enabled.
Added a check for the OPTION_READCONFIG option before executing
the code which would attempt to incorporate configuration file
settings into the current configuration, because if OPTION_READCONFIG
is set to false, then there is no reason to even try to do this as
every single configuration file will be ignored (because config files
have been turned off by OPTION_READCONFIG). [Bryan Ischo]
Fixed small memory leak in mame.c. [Bryan Ischo]
Fixed double-free error in render.c. [Bryan Ischo]
Made core_strdup use osd_malloc instead of malloc. [Bryan Ischo]
As previously discussed, this will be used for the MC680x0 FPU in
order to handle high-precision floats portably. The license is
included in README.txt and is MAME compatible.
1. If OSD is not defined, we look for the built-in Windows
environment variable 'OS' and if it is set to 'Windows_NT',
we assume OSD=windows; otherwise we assume OSD=sdl.
2. If TARGETOS is not defined, we look for the same variable
and set TARGETOS=win32 if it is set to 'Windows_NT'.
3. Windows SDL builds have an 'sdl' prefix on the executable
so they can live side-by-side.
The upshot of this is that on Windows systems:
make -> standard Windows build producing mame.exe
make OSD=sdl -> SDL Windows builds producing sdlmame.exe
On non-Windows systems:
make -> SDL build by default (don't need to say OSD=sdl)
- Changed INLINE to map to "static inline"
- (Windows only) Made -static-libgcc standard for both 32-bit and 64-bit
- (Windows only) Some fixes for MSVC builds
1. Removed CPP_COMPILE option. All files (except expat and zlib)
are now compiled as C++ by default. For now, imagine nothing has
changed. The goal is not to go hog-wild with C++isms, but to
leverage it where it makes the most sense.
2. Mapped INLINE to plain old C++ inline now, fixing several
cases where this was problematic.
3. Marked global const structures explicitly extern since consts
are locally-scoped by default in C++.
4. Added new 'default' make target which just builds the emulator.
Use 'make all' to build everything including the tools.
5. 64-bit builds now get a '64' suffix on them. We might want to
just make this true for Windows builds, but it's on for everyone
at the moment.
6. (Windows) Removed UNICODE option. UNICODE is enabled by default
on all Windows builds now. The 32-bit version links against
libunicows.a for continued Win9x compatibility.
7. (Windows) Removed hacks surrounding unicode handling of main().
They are no longer necessary with the new tools.
---------- Forwarded message ----------
From: Oliver Stöneberg <oliverst@online.de>
Date: Sat, Dec 12, 2009 at 3:27 PM
Subject: includes cleanup
To: submit@mamedev.org
This patch removes the global "includes" include and explcitly
specifies the "includes" path in all #include occurances.
- added unidasm to the tools build
- split the disassemblers out of libcpu and into new libdasm
- ensured the disassembly entry points for all disassemblers are
in the source file for the disassembler (sometimes new generic
versions were created)
Still needs command line options and file loading, but the
fundamentals are present, and it links.
* Fix build of ldplayer on OS X. Since the CUSTOM sound module no longer exists, I
arbitrarily changed it to WAVE, as ar gets upset if it has no input files. I also
removed the -all_load flag for ldplayer from the main makefile as it upsets the linker
on OS X.
* Fix build for PPC64 Linux. (This slightly messes up static branch prediction hints
on OS X and AIX, but OS X for PPC64 is dead, and no- one builds MAME for AIX, and
it will still build, anyway.)
* Paramaterise the arguments to check for NULL in the ATTR_NONNULL macro rather than
just checking the first argument. This requires compiler support for C99 variadic
macros (MSVC2005 and GCC4 have this AFAIK).
Vas
compilation:
- new option CPP_COMPILE to trigger this (off by default)
- split CFLAGS into common, C-only, and C++-only flags
- when enabled, CPP_COMPILE causes 'pp' to be appended to
the target name
NOTE THAT THE SYSTEM CANNOT ACTUALLY BE COMPILED THIS WAY
YET. IT IS JUST AN EXPERIMENT.
Modified lib.mak to always build zlib/expat as C regardless
of CPP_COMPILE.
Modified windows.mak to fix warnings with MAXOPT=1, and to
leverage the new CFLAGs definitions.
Modified vconv.c to do appropriate conversions for new C++
options.
Updated sources so that libutil, libocore (Windows), and
libosd (Windows) can be cleanly compiled as C or C++. This
was mostly adding some casts against void *.
Fixed a few more general obvious problems at random
locations in the source:
- device->class is now device->devclass
- TYPES_COMPATIBLE uses typeid() when compiled for C++
- some functions with reserved names ('xor' in particular)
were renamed
- nested enums and structs were pulled out into separate
definitions (under C++ these would need to be scoped to
be referenced)
- TOKEN_VALUE cannot use .field=x initialization in C++ :(
macro from the source code. All MAME builds now include
the debugger, and it is enabled/disabled exclusively by
the runtime command-line/ini settings. This is a minor
speed hit for now, but will be further optimized going
forward.
Changed the 'd' suffix in the makefile to apply to DEBUG
builds (versus DEBUGGER builds as it did before).
Changed machine->debug_mode to machine->debug_flags.
These flags now indicate several things, such as whether
debugging is enabled, whether CPU cores should call the
debugger on each instruction, and whether there are live
watchpoints on each address space.
Redesigned a significant portion of debugcpu.c around
the concept of maintaining these flags globally and a
similar, more complete set of flags internally for each
CPU. All previous functionality should work as designed
but should be more robust and faster to work with.
Added new debugger hooks for starting/stopping CPU
execution. This allows the debugger to decide whether
or not a given CPU needs to call the debugger on each
instruction during the coming timeslice.
Added new debugger hook for reporting exceptions.
Proper exception breakpoints are not yet implemented.
Added new module debugger.c which is where global
debugger functions live.
- rewrote PowerPC implementation as a dynamic recompiler on top
of the universal recompiler engine
- wrote a front-end to analyze PowerPC code paths and register usage
- wrote a common shared module with C implementations of tricky
CPU behaviors
- added separate CPU types for the variants supported, instead of
relying on a hidden model enum
- rewrote the serial port emulation for the 4xx series to be more
accurate and not rely on separate DMA handlers
- rewrote the MMU handling to implement a software TLB that faults
in pages and handles changed bits appropriately
- implemented emulation of the PowerPC 603's software TLB, which
allows the model 3 games to run without a hack to disable the MMU
Updated the PowerPC disassembler to share constants with the rest of
the core, and to more aggressively use simplified mnemonics, especially
for branches. [Aaron Giles]
Universal recompiler:
- fixed frontend to handle opcode widths different from bus width
- added several new opcodes:
* (D)GETFLGS - copies the UML flags to a destination operand
* FDRNDS - rounds a double precision value to single precision
- renamed several opcodes:
* SETC -> CARRY
* XTRACT -> ROLAND
* INSERT -> ROLINS
- consolidated the following opcodes:
* LOAD?U -> LOAD
* LOAD?S -> LOADS
* STORE? -> STORE
* READ?U -> READ
* READ?M -> READM
* WRITE? -> WRITE
* WRITM? -> WRITEM
* SEXT? -> SEXT
* FTOI?? -> FTOINT
* FFRI? -> FFRINT
* FFRF? -> FFRFLT
- removed some opcodes:
* FLAGS - can be done with GETFLGS/LOAD4/ROLINS
* ZEXT - can be achieved with AND
* READ?S - can be achieved with READ/SEXT
- updated C, x86, and x64 back-ends to support these opcode changes
- updated disassembler to support these opcode changes
MIPS3 dynamic recompiler:
- updated to use new/changed opcode forms
- changed context switch so that it only swaps a single pointer
Konami Hornet changes: [Aaron Giles]
- updated to new PowerPC configurations
- updated some memory handlers to be native 8-bit handlers
- cleaned up JVS implementation to work with new serial code
- added fast RAM for the work RAM to give a small speed boost
Konami GTI Club changes: [Aaron Giles]
- updated to new PowerPC configurations
- updated some memory handlers to be native 8-bit handlers
Konami Viper/ZR107 changes: [Aaron Giles]
- updated to new PowerPC configurations
Sega Model 3 changes: [Aaron Giles]
- updated to new PowerPC configurations
- reimplemented/centralized interrupt handling
- these games are broken for the moment
Fixed crasher due to some Konami games using 8 layers in
the K056832 implementation, even though it was only written
for 4. [Aaron Giles]
Added fisttp opcode to i386 disassembler. [Aaron Giles]
is drcuml.c, which defines a universal machine language
syntax that can be generated by a frontend recompiler and
then retargeted via a generic backend interface to any of
a number of different architectures. A disassembler for the
UML is also included to allow examination of the generated
UML code.
Currently supported backend architectures include 32-bit x86,
64-bit x86, and a platform-neutral interpreted C backend that
can be used as a fallback for platforms without native
support. The C backend also performs additional validation
to ensure assumptions are met.
Along with the new architecture is a new MIPS III/IV
recompiler frontend. This frontend has been rewritten from
the old x64-specific recompiler to generate UML opcodes
instead. This means that the single recompiler can be used
to target multiple backend architectures and should in
theory produce identical results across all of them.
The old 32-bit and 64-bit MIPS recompilers are now officially
retired. The new system provides similar performance (within
5% generally) to the old system and has similar compatibility.
The only currently known issues are some problems with the
two Gauntlet 3D games.
Subject: Minor patch submission - verbose build information for development in IDE
I am submitting this minor patch that adds verbose build information to the
build output when the "VERBOSE" flag is set to "1", rather than the simple
output normally seen, by adding the "-v" flag to the CFLAGS variable. This
is most useful for those that like to use an IDE (Eclipse in my case) for
source browsing and development. The IDE usually will parse the build
information, which is currently not available, to determine include file
locations, build definitions, etc. and will use that information to provide
more advanced navigation features.
a new compile-time define (ENABLE_DEBUGGER). This means that MAME_DEBUG no longer means
"enable debugger", it simply enables debugging features such as assertions and debug code
in drivers.
Also removed the various levels of opbase protection in memory.h and always just turned
on full bounds checking.
Fixed build break due to missing ampoker.lay -> ampoker2.lay renaming.
- removed years from copyright notices
- removed redundant (c) from copyright notices
- updated "the MAME Team" to be "Nicola Salmoria and the MAME Team"
beastrzb097u2red.patch
- added GAME_NOT_WORKING flag, so this bogus bug report can be
removed (not even worth 1 buck!)
20080104_1.patch
- fixes a memory in src/mame/machine/stvcd.c
memoryleak0118u1ora.patch
- fixes a memory leak, that only happened when no RawInput API was
available
memory_leak0111u6ora_1.patch
- only a partial patch - model3.c still missing
