| Commit message (Collapse) | Author | Age | Files | Lines |
| | |
|
| |
|
|
| |
* fix booting in most if not all hash/smc777.xml FDC failures
|
| |
|
|
|
|
| |
at compile time. (#15224)
* netlist/nl_setup.cpp: Use C++20 __VA_OPT__ to simplify device registration code.
* netlist/build/create_devinc.py: Generate check for correct number of arguments.
|
| |
|
|
|
|
|
|
|
|
| |
This reverts commit 7689ad57e746a33dfeae9299821992e15dd63ea9.
The nl_exception doesn't inherently log anything, and whatever catches
it may send its log output to a different place to the netlist devices'
loggers if it end up logging the message. All the error conditions are
supposed to be sent to the netlist devices' own loggers to ensure you
have visibility.
|
| |
|
|
|
| |
* netlist: Remove redundant usages of fatal logger
* nl_factory.h: Fix Wunused compiling error
|
| |
|
|
|
|
|
| |
This was supposed to allow calling i686 "thiscall" member functions via
a free function pointer. It had been disabled for years as it stopped
working in GCC. Simply having the code present breaks compiling with
clang 22.1 even when it isn't enabled.
|
| |
|
|
|
|
|
| |
Also updated compiling documentation. Mentioned the possibility to use
the x86-64 clang/libc++/ucrt environment with MSYS2 on Windows. Removed
link to tools at mamedev.org - it's trivially easy to get an up-to-date
MSYS2 environment by following the instructions.
|
| |
|
|
|
|
|
|
|
|
|
| |
(#15182)
* Hackery to allow UTF-8 strings as well as plain strings:
- emu/ioport.h: Allow char8_t for names in field configuration helper.
- emu/device.h, emu/gamedrv.h: Allow char8_t for descriptions and manufacturers.
- util/language.h: Allow char8_t message input.
* util/strformat.h: Allow char8_t format and string arguments with char output.
* ui/videoopt.cpp: Deal with UTF-8 strings as a distinct type.
* osd/windows: Assume Windows 8 or later.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
when getting PC MFM track data. (#15165)
Ensoniq VFX and EPS family floppies use 10 sectors per track,
starting at sector 0 rather than the more common sector 1.
The code in `esq16_dsk.cpp` handles this when loading floppy images,
but not when saving them. This attempts to fix this in a way that
is not only specific to Ensoniq's floppy images,
but to possibly other future ones as well.
In `flopimg.cpp`, allow specifying explicit start and end sectors
when getting PC MFM track data.
In `esq16_dsk.cpp`, use this to explicitly specifying sectors
0 to 9 when getting the track data to save.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
ui/info.cpp, imagedev/cassette.cpp: Deal with UTF-8 strings as a distint
type.
ui/inputmap.cpp, debugger/qt/dasmwindow.cpp, debugger/qt/mainwindow.cpp,
sound/coreaudio_sound.cpp, cpu/drcbec.cpp, mit/tx0_v.cpp,
konami/3dom2.cpp: machine/mc68328.cpp, cpu/mips/mips1.cpp,
cpu/mips/r4000.cpp, cpu/romp, machine/cammu.cpp, machine/ns32081.cpp:
Avoid arithmetic between different enum types.
dec/pdp1.cpp, konami/firebeat.cpp, mit/tx0.cpp, sound/lc7535.cpp,
sound/spkrdev.cpp: Avoid arithmetic between enum and floating point.
gaelco/gaelco3d_m.cpp: Fixed deprecated uses of volatile variables.
sound/discrete.h: Avoid comparing enum to floatint point.
ui/toolbar.ipp: Don't use UTF-8 qualifier on pure ASCII strings, just
assume char is ASCII-like.
cpu/unsp: Use default constructor for compiler_state.
sgi/pm2_mmu.cpp: Avoid conflict between file static access and
identically named function in unistd.h.
osd/interface/audio.h: Provide an explicit constructor.
util/server_http_impl.hpp: Fixed uninitialised class member warning.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
- Added FD-5 floppies (emulation works but only with ROM hack)
- Updated Sord m5 www links
- Added fd5 utility disk to software list - not original dump, made from program listings
- Added support of optional SRAM in Basic-F and Basic-G cartridges. Works only if shortname(softlist) is used
- Marked 32/64KB RAM expansions EM-5, 64KBI, 64KBF, 64KRX as supported in software list
- Reenabled and refactored memory banking
- Added support of E and F block types for cassettes
- m5p_brno: Switched to ROM including basic-i
- m5p_brno: Reenabled and fixed memory banking
- Repaired all not working ROMs except one and all cassettes
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
-cpu/powerpc: Modernised recompiler front-end code, improved instruction
analysis logging.
-cpu/sh: Cleaned up instruction analysis logging a little
-util/client_ws.hpp, util/server_ws_impl.hpp, capcom/cps2comm.cpp:
Updated some use of deprecation ASIO APIs.
-nakajima/nakajies.cpp: Use range-based loops.
|
| | |
|
| |
|
|
|
|
| |
The new function was just the old function copy/pasted with the comments
stripped out and some constants changed. That just causes duplicate
maintenance.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
(format + motor-gate) (#14976)
* Fix MZ-80B cassette loading: add 6000 Hz format for 1800 baud
MZ-80B uses 1800 baud Sharp PWM (Logic 0=332.75µs, Logic 1=667µs),
not 1200 baud. Adds mz80b_cassette_formats with 6000 Hz sample rate.
Restores mz700 to 4400 Hz. MZ-2000 also uses 1800 baud.
* mz80b: Fix cassette loader hang on multi-block loads
* mz_cas.cpp: Add MZ-80B-specific format (fill_wave_mz80b) with 48 kHz,
LGAP 10000, SGAP 5000, 16/32 sample pulses
* mz2000.cpp: Gate PPI port B tape data bit (6) with motor state.
When motor stops, force Low to simulate signal decay. Fixes loader
'wait for silence' loop at \ hanging on frozen High waveform.
|
| |
|
|
| |
and inline for functions; util/palette.h: Don't use reserved _Names.
|
| |
|
|
| |
compare warnings for clang/arm64.
|
| |
|
|
|
|
|
|
|
|
| |
`entry:default_value()` was always returning string. A consequence of this problem is that the following LUA commands would error
```
manager.options.entries['beam_dot_size']:value(manager.options.entries['beam_dot_size']:default_value())
manager.options.entries['frameskip']:value(manager.options.entries['frameskip']:default_value())
```
With this change, `default_value()` will return objects of the correct type
|
| | |
|
| |
|
|
|
|
|
|
|
| |
* Moved float/integer conversions into the option entries themselves.
* Fixed another bug where float values would be converted to strings
with the global local, but converted from strings using the "classic"
locale.
* frontend/mame/luaengine.cpp: Got rid of duplicated numeric options
conversion code.
|
| |
|
|
|
|
| |
* Use "classic" locale for numeric conversions.
* Don't use capturing lambdas.
* Put integer overload of value() before float overload.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Lua considers anything other than null or false to be "truthy". The
tests against "null" were comparing to a non-existent global called
"null" that evaluated to nil.
* Got bus.lua back in alphabetical order and got at least most of the
trigger variables aligned with paths.
* At least don't put random stuff in the middle of the Acorn devices in
machine.lua.
-cpu/adsp2100: Corrected some tabulaton that had rotted.
-cpu/tms32032: Got rid of some superfluous parentheses.
-apple/lisa*.cpp: Added ATTR_COLD on stuff.
|
| | |
|
| |
|
|
|
|
|
|
| |
core: Better support for external mmus
Note: I didn't expect to push that one right now, it's missing
documentation among other things, but since some people seems to be
interested to work on that driver I don't want to hold them.
|
| | |
|
| | |
|
| |
|
|
| |
multibyte.h helpers instead of playing fast and loose with alignments
|
| | |
|
| |
|
|
| |
* fs_coco_os9.cpp: Add disk_id to volume metadata
|
| |
|
|
| |
every file or directory on the specified volume
|
| |
|
|
| |
and use it in one filesystem
|
| |
|
|
| |
in file_create; use std::min
|
| |
|
|
| |
than copies of the boot sector
|
| |
|
|
|
|
|
|
|
| |
* att6300p.cpp: Add support for 1.2MB drive (525hd)
* att6300p_fdc.cpp: Move FDC support into device-specific files.
* att6300p.cpp: Fix enabling/disabling of A20-A23 address lines.
* att6300p.cpp: Add comment explaining default drive densities.
|
| |
|
|
|
| |
- Raw pointer accessors (data, rodata) are no longer provided for fsblk_t::block_t. Other accessors used to be based on these methods, but now require their own internal overrides. This entails more copying of data in many cases, but will allow more flexible subclassing opportunities in the future.
- To compensate for the removal of raw pointer accessors, add methods to fsblk_t::block_t for checking whether a portion of a block is equal to a memory range or a string (eqmem, eqstr). The resulting changed behavior in formats/fs_oric_jasmin.cpp actually fixes a bug in the directory search routine.
|
| | |
|
| |
|
|
| |
improves directory navigation in the MAME UI file manager.
|
| |
|
|
|
|
|
|
|
|
| |
The previous code allocated 512 bytes to accomodate a FLUX chunk
if the resulting file has flux tracks, but the actual size of a
FLUX chunk is 168 bytes (8 bytes header + 160 bytes chunk).
When MAME allocated 512 bytes, this left 344 zero-bytes padding in
the file that other tools would interpret as invalid chunks.
See file format reference: https://applesaucefdc.com/moof-reference/
|
| | |
|
| | |
|
| |
|
|
|
| |
-------------------
Wren Executive [Nigel Barnes]
|
| |
|
|
|
|
|
|
|
|
| |
* bus/bk: Expose correct public interface for card devices.
* bus/qbus/terak_v.cpp: Fixed save state support, restrict drawing to
clipping rectangle.
* apple/apple2e.cpp: Try to actually show the value that would be read
in the debugger for C000 I/O.
* Got rid of pointless duplicate layout.
|
| | |
|
| |
|
|
|
|
| |
- Completely reimplement Adam .dsk floppy format, permuting sectors of 720KB and 1.44MB 3.5" formats into (and out of) the correct native interleave
- Split adam_fdc device into several sets, some now properly supporting 720KB and/or 1.44MB formats
- Remove the "Doug Slopsema" BIOS for the 320KB FDC variant upon determining it to be a copy of the 320ta BIOS with slight bitrot
|
| | |
|
| |
|
|
|
| |
- Add flophashes and hdhashes commands to compute CRC-32 and SHA-1 hashes for each file on a floppy or hard disk image.
- Add flopchmeta command to change the metadata for a file, directory or volume on a floppy image.
|