| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
| |
callback types. timer_callback is a class containing a timer_expired_delegate plus a user-supplied pointer, and accounting information for registration. timer_instance is a single active timer in the scheduler. persistent_timer is what you get back from timer_alloc() now, and has methods for dynamic changes. transient_timer_factory is what you use to create transient timers. Devices now own a transient_timer_factory for transient device timers. persistent_timers can be emedded and initialized directly.
|
| |
|
|
| |
parameters to u64. Change first parameter of device timers to be emu_timer const. Updated all device timers to follow suit.
|
| |
|
| |
-mahine/upd765.cpp: Implemented sector-based termination conditions for read commands.
|
| | |
|
| |
|
|
| |
* hx5102: Add addressable latches
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
| |
- dp8473, pc_fdc_superio: Unmap registers not actually provided
- upd72065, i82072, smc37c78: Increase number of steps for recalibrate command
- wd37c65c: Duplicate write protect flag in SR3
|
| | |
|
| |
|
|
| |
Discovered pseudo-dma behaviour on Sony NEWS 68k system; read/write to the data port during DMA transfer may behave differently if DACK is not asserted.
|
| | |
|
| | |
|
| |
|
|
|
|
| |
- pc_fdc_interface has been eliminated. The XT and AT hookups have been moved into the ISA devices.
- pc_fdc_xt_device remains for now only because two drivers depend on it.
- europc still prints "disk error" on boot. It should still be possible to fix this.
|
| | |
|
| |
|
|
|
|
| |
* upd765, next, sun3x: fix FDC mode being treated as clock frequency (MCFG removal gone wrong)
* h01x: translate some comments from Chinese to English
* snes/sfx: trampoline constructors should be private
|
| | |
|
| | |
|
| |
|
|
|
| |
- Deselect drives upon reset, or select drive 0 for older FDCs with multiplexed select lines
- Deselect other floppies when using set_floppy
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
This effectively reverts b380514764cf857469bae61c11143a19f79a74c5 and
c24473ddff715ecec2e258a6eb38960cf8c8e98e, restoring the state at
598cd5227223c3b04ca31f0dbc1981256d9ea3ff.
Before pushing, please check that what you're about to push is sane.
Check your local commit log and ensure there isn't anything out-of-place
before pushing to mainline. When things like this happen, it wastes
everyone's time. I really don't need this in a week when real work™ is
busting my balls and I'm behind where I want to be with preparing for
MAME release.
|
| |
|
|
|
| |
This reverts commit c24473ddff715ecec2e258a6eb38960cf8c8e98e, reversing
changes made to 009cba4fb8102102168ef32870892438327f3705.
|
| |
|
|
|
|
|
|
|
| |
Result from Read ID must be the sector ID information during execution phase. So result must be:
- Zeroes if not ready (execution phase is not reached).
- Command codes if ready but scan failed (execution phase is reached but ID is not found). This patch fixes this.
- If scan is correct then result is the ID information from sector found.
Fix MT 06611
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
upd765: Kludge in an output callback for the index pulse, which doesn't deserve to be attached to this device but is done this way because of MAME's FDC model. (nw)
|
| |
|
|
| |
This is intended for future use by machines in which the US/DS lines are neither directly connected to the drives nor wholly disconnected from them.
|
| |
|
|
| |
These clocks are currently unused, and their sources and dividers are often unclear. In some cases they are clearly software-configurable, which has not been emulated at all.
|
| |
|
|
| |
through popular FDC headers (nw)
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
unbreak x68000 floppy loading (nw)
|
| |
|
|
|
| |
Mostly tiying up and simplifying the i82072 behaviour; the only change that could potentially impact other upd765 users is marking the st0_filled after every interrupt-generating command ends, rather than only the non-data ones. This appears to be correct behaviour, and will only affect systems which actually do a "sense interrupt status" after a command-completion interrupt (like the InterPro does).
* report valid interrupt sense data after all interrupt-generating commands
* improve i82072 drive busy handling and motor control
|
| |
|
|
| |
stick, also const (nw)
|
| | |
|
| | |
|
| |
|
|
|
|
|
| |
suddenly silenced
* streamline templates in addrmap.h
* get rid of overloads on read/write member names - this will become even more important in the near future
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
Also, a lot more freedom happened, that's going to be more visible
soon.
|
| | |
|
| |
|
|
| |
Fixes http://mametesters.org/view.php?id=6689.
|
| |
|
|
| |
UTF-8 will be nuked, and cleaning it up by hand is annoying
|
| |
|
| |
Yet another attempt at implementing the i82072 sense interrupt status logic. This time, I've separated the logic into the derived class, so it won't affect the other devices. Also included motor on/off logic (unique to i82072) - motor on delay is incomplete.
|
| | |
|
| |
|
|
|
|
|
|
|
| |
This currently only works if you set the drive select lines correctly,
which the upd765 does now. Hook it up to the psi98 driver and use the
TEAC FD-55F drive instead of the generic one.
This is some basic first work to better model the real hardware and
floppy drive setups in the future.
|