moa/todo.txt

68 lines
3.3 KiB
Plaintext
Raw Normal View History

* there is a problem where something writes to the rom area which causes a crash
At 0x16cde, a move writes an invalid value to 0 via an indirect %a2 reg. The value of the reg might have changed during an interrupt, but it definitely breaks when the next interrupt occurs
Before the loop is 0x16a0e which then calculates the count and such
0x16584 is where the memory address 0xffd11a is updated, which is then used for the bad 0x0000 address which causes the improper write. 0x16570 is a better start
On broken cycle: %a1 = 1df40, moves that location + 1 to %d0
* 0x1650e is where 0xffac08 is changed to 0xd100
* 0x16572 is where 0xffd100 is changed to 0
* what if it's a problem when turning the 0 into a full address (which should be ff0000 or ffff00 but instead ends up being 000000)
* the overflow bit is only correct for addition but not subtraction... in subtraction, two positives can result in a negative and vice versa
2021-10-31 18:00:14 +00:00
* fix ym7101 to better handle V/H interrupts (right now it sets and then the next step will clear, but it'd be nice if it could 'edge trigger')
2021-11-01 23:51:45 +00:00
* could have a remapper device, which takes a big swath of addresses in and maps them to another set of addresses (for Mac VIA generic to bus-hookup-in-mac adapter)
2021-10-31 18:00:14 +00:00
* how can you do devices that change their address map during operation, like mac which puts rom at 0 and ram at 600000 temporarily
2021-10-29 22:05:41 +00:00
* i need a better way of handling disperate reads/writes to I/O spaces, rather than having multiple devices or having a massive chunk of address space allocated, continuously
* should you modify Addressable to also take the absolute address as input? I'm thinking of how the same device could be mapped to multiple addresses in memory instead
of taking up a whole range of addresses
* could you use a generic sharable signal thing for sharing data, such as the VIA in mac128 where a single output bit determines the video mode (which would be a separate device)
So both could share the same Signal, one setting it and the other reading it, but how would you actually configure/build that?
* make it possible to set the frame sizes when creating the frame swapper
* should you rename devices.rs traits.rs?
2021-10-31 18:00:14 +00:00
* implement a Z80
* maybe see about a Mac 128k or something
2021-10-29 22:05:41 +00:00
* add instruction timing to M68k
* YM7101 timing is causing it to be very slow... speeding this up increasing rendering speed a lot, even though the frame shouldn't be drawn that often... not sure what's wrong with the timing
* make the ym7101 set/reset the v_int occurred flag based on the interrupt controller
2021-10-29 22:05:41 +00:00
* you could modify read()/write() in Addressable to return the number of bytes read or written for dynamic bus sizing used by the MC68020+
* should you simulate bus arbitration?
2021-10-26 19:17:59 +00:00
2021-10-29 22:05:41 +00:00
* make tests for each instruction
* check all instructions in the docs
* unimplemented: ABCD, ADDX, BFFFO, BFINS, BKPT, CHK, EXG, ILLEGAL, MOVEfromCCR, MOVEP, RTR, RTD, SBCD, SUBX
* >=MC68020 undecoded & unimplemented: CALLM, CAS, CAS2, CHK2, CMP2, RTM, PACK, TRAPcc, UNPK
* add support for MMU
* add support for FPU
* Coprocessor instructions: cpBcc, cpDBcc, cpGEN, cpScc, cpTRAPcc
2021-10-10 03:35:52 +00:00
* how can you have multple CPUs
* each device that can make a bus request should have a BusPort which is used to access the bus
* can you eventually make the system connections all configurable via a config file?