Adrian Conlon
|
59d1e2789e
|
Whoops: missed reference to RapidJson
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-11 19:39:37 +01:00 |
|
Adrian Conlon
|
7a7b0046cd
|
Add an implementation that uses JsonCpp (pretty slow!)
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-11 19:13:05 +01:00 |
|
Adrian Conlon
|
f3f6452119
|
Add nlohmann json parser as an option to build.
1/2 speed, compared to boost.json
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-11 14:59:23 +01:00 |
|
Adrian Conlon
|
db106b1719
|
Performance: speed up message handling in TestRunner.
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-11 11:41:50 +01:00 |
|
Adrian Conlon
|
5686906583
|
Add nodiscard attributes, where needed
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-11 10:43:33 +01:00 |
|
Adrian Conlon
|
4be61a9d54
|
Separate concerns a little between low level classes and high level classes in terms of report generation.
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-11 10:20:18 +01:00 |
|
Adrian Conlon
|
760f5d5aec
|
Correct filler memoryRead accesses to use PC() rather than the last location read.
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-10 21:35:26 +01:00 |
|
Adrian Conlon
|
6a59bfbcd8
|
First stab at using the Harte randomised processor tests. Some failures detected in the M6502 run.
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-10 21:26:30 +01:00 |
|
Adrian Conlon
|
6c3ef821bf
|
Update for latest boost v1.77
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-10-10 21:25:21 +01:00 |
|
Adrian Conlon
|
22506ea56c
|
Extensive change warning: lots of "noexcept" and "constexpr" changes. Not sure if I'll keep all of them, but interesting...
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-07-18 14:28:40 +01:00 |
|
Adrian Conlon
|
2f76e901f9
|
More tidying of include files for VS2019 compatibility (plus more correct!)
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-05-29 12:18:13 +01:00 |
|
Adrian Conlon
|
cab29e3ce4
|
Tidy some code formatting. No functional changes.
|
2021-04-07 21:37:30 +01:00 |
|
Adrian Conlon
|
2992a0e78a
|
Event fire: use default arguments, where possible.
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-04-07 21:36:53 +01:00 |
|
Adrian Conlon
|
2fa9ffd1e3
|
Tidy up some C++ a little
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-04-07 21:36:09 +01:00 |
|
Adrian Conlon
|
b1ca06447f
|
Ensure 6502 uses memory rather than bus read/write mechanism
Signed-off-by: Adrian Conlon <adrian.conlon@gmail.com>
|
2021-03-08 16:44:09 +00:00 |
|
Adrian Conlon
|
6285a397ab
|
Tidy 6502 whitespace
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2020-02-22 09:02:37 +00:00 |
|
Adrian Conlon
|
c8bdabf34f
|
Reflect that the I/O for Intel style processors isn't part of the CPU, but attached to the Bus and access controlled by the CPU.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2020-02-09 11:51:58 +00:00 |
|
Adrian Conlon
|
d9466082ec
|
M-Cycle accurate Z80 modifications.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-12-29 01:18:54 +00:00 |
|
Adrian Conlon
|
d0467421ff
|
Improve compatibility with .net emulator code.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-11-09 18:58:23 +00:00 |
|
Adrian Conlon
|
ee3ecc682d
|
Add R/W pins to the MC6809 and MOS6502 processors. Hoping it'll ease peripheral development.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-09-15 12:49:32 +01:00 |
|
Adrian Conlon
|
254cfbe342
|
Tidied up pin management to be synchronised with the .Net code.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-09-06 23:55:57 +01:00 |
|
Adrian Conlon
|
6940a54355
|
Update all EightBit projects to VS2019 (Latest SDK, C++17)
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-08-17 11:04:29 +01:00 |
|
Adrian Conlon
|
5e9014997a
|
Upgraded to VS2019, default SDK.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-08-06 13:16:38 +01:00 |
|
Adrian Conlon
|
def1c58e9d
|
Tidy project settings across the EightBit library to be more consistent.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-07-06 22:45:16 +01:00 |
|
Adrian Conlon
|
f5582df402
|
Add some more M6502 documentation.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-05-05 10:30:33 +01:00 |
|
Adrian Conlon
|
1a0d3ad77a
|
Go through all projects for configuration consistency.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-05-05 10:25:44 +01:00 |
|
Adrian Conlon
|
f0376fa81e
|
Use macros to define our device pins.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-14 23:17:54 +00:00 |
|
Adrian Conlon
|
7f853ec73f
|
Add missing 6502 pin events
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-14 08:26:27 +00:00 |
|
Adrian Conlon
|
92d23d82d6
|
Start big refactor of device/CPU pin usage (to allow pin events throughout).
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-14 02:10:17 +00:00 |
|
Adrian Conlon
|
68030610d8
|
Move to an event driven clock tick event for all CPUs
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-09 23:24:33 +00:00 |
|
Adrian Conlon
|
8b187e7614
|
The clock still has to tick, even while held on RDY low, otherwise cycle timing won't work.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-09 09:05:12 +00:00 |
|
Adrian Conlon
|
a90ca6ba38
|
Move RDY initialisation to the CPU power-on sequence. Not strictly correct, but OK.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-09 00:33:56 +00:00 |
|
Adrian Conlon
|
87d86bcd84
|
Whoops: missed unneeded method declaration.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-09 00:08:03 +00:00 |
|
Adrian Conlon
|
01175cf9eb
|
Add support for emulated SYNC and RDY lines. Untested, but feel close.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-08 23:55:27 +00:00 |
|
Adrian Conlon
|
047babbe7c
|
Rearrange the RESET handler for cycle accuracy. Use more of the general interrupt handler, but with "dummy" stack write access.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-08 23:09:52 +00:00 |
|
Adrian Conlon
|
25321e78e7
|
Now that HALT/RESET/NMI/IRQ and BRK have a unified architecture, I think this wraps up the instruction handler of the 6502.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-08 01:32:43 +00:00 |
|
Adrian Conlon
|
3faec680b0
|
I think this fixes one of my NES issues. The fix-up required for the PC is handled by the fetchByte associated with the BRK instruction.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-07 01:28:23 +00:00 |
|
Adrian Conlon
|
d7763d8215
|
Profile mode no longer exists on the 6502 test program.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-07 01:08:23 +00:00 |
|
Adrian Conlon
|
06e2a5c947
|
Start unifying 6502 interrupt handling. NMI/IRQ/BRK are all doing mostly the same work.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-07 01:06:07 +00:00 |
|
Adrian Conlon
|
b7b7c93a77
|
This tidies the last of 6502 cycle accurate instruction implementations.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-06 22:34:53 +00:00 |
|
Adrian Conlon
|
ad644f7013
|
Tidy the 6502 instruction switch statement a little to lessen the width a little.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-06 20:51:13 +00:00 |
|
Adrian Conlon
|
c3d2ef51d9
|
Cycle accuracy for a couple of the indexed write (store) instructions.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-06 20:39:37 +00:00 |
|
Adrian Conlon
|
4d3be9e756
|
Use correct boundary crossing conditional for Read/Modify/Write Absolute,X addressing mode.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-06 12:58:13 +00:00 |
|
Adrian Conlon
|
baf32cef89
|
Get the memory access more correct for more of boundary crossing indexed addressing modes.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-06 12:17:43 +00:00 |
|
Adrian Conlon
|
a13ad5042a
|
Correct constructions of register16_t: the structure is "#ifdef"ed for different endian arrangements.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-06 11:27:43 +00:00 |
|
Adrian Conlon
|
3749585398
|
Cycle accuracy changes for BRK/RTI instructions.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-06 11:13:49 +00:00 |
|
Adrian Conlon
|
3337f57747
|
More cycle accuracy changes, concentrating on stack operations.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-05 23:21:43 +00:00 |
|
Adrian Conlon
|
f5125b2a35
|
Add some documentation regarding instruction cycle timings.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-05 23:21:12 +00:00 |
|
Adrian Conlon
|
143e9a9e68
|
More cycle accuracy changes:
1) implied instruction, pointless fetch
2) branch pointless fetch when condition is met
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-05 17:23:50 +00:00 |
|
Adrian Conlon
|
3b7cec9c69
|
Spot a couple more addCycle patterns.
Signed-off-by: Adrian Conlon <Adrian.conlon@gmail.com>
|
2019-01-04 09:17:59 +00:00 |
|