paoda
/
zba
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
587 Commits 4 Branches 0 Tags 7.7 MiB
Commit Graph

121 Commits

Author SHA1 Message Date
Rekai Nyangadzayi Musuka e60b556f72 chore(ppu): remove BGR555 -> RGBA888 LUT 
LUT probably couldn't fit in CPU cache anyways.

TODO: Consider whether LUTs for separate channels (size 32 * 3 * 3
instead of std.math.maxInt(u15))
 2022-10-17 20:31:42 -03:00
Rekai Nyangadzayi Musuka 26dba16789 chore(gpio): add missing errdefer 2022-10-17 20:01:50 -03:00
Rekai Nyangadzayi Musuka 7b146ad7ca fix(bios): set addr_latch even if bios is skipped 2022-10-13 00:35:22 -03:00
Rekai Nyangadzayi Musuka 822eed1f3a fix(bus): make open bus impl aware of CPU pipeline 2022-10-13 00:35:22 -03:00
Rekai Nyangadzayi Musuka b37a14900c style(bus): cpu ptr doesn't need to be optional 2022-10-13 00:35:22 -03:00
Rekai Nyangadzayi Musuka f5bd20bc2a style: code cleanup 2022-10-13 00:35:22 -03:00
Rekai Nyangadzayi Musuka d3514b14f3 fix: resolve timing regressions 
make sure to use fetch timings when fetching instructions
 2022-10-13 00:35:20 -03:00
Rekai Nyangadzayi Musuka 06c60dad74 fix: rename Pipline to Pipeline 2022-10-13 00:34:18 -03:00
Rekai Nyangadzayi Musuka 870e991862 feat: working pipeline implementation 2022-10-13 00:34:18 -03:00
Rekai Nyangadzayi Musuka 5bb5bdf389 chore: refactor ARM/THUMB data processing instructions 2022-10-13 00:34:18 -03:00
Rekai Nyangadzayi Musuka a3996cbc58 fix: don't flush pipeline when reloading CPSR in ARM Data Processing 2022-10-13 00:34:18 -03:00
Rekai Nyangadzayi Musuka a948c6f900 chore: don't write to CPSR + swap with SPSR at the same time 2022-10-13 00:34:18 -03:00
Rekai Nyangadzayi Musuka e4451738b5 fix: advance r15, even when the pipeline is reloaded from the scheduler 
The PC would fall behind whenever an IRQ was called because the pipeline
was reloaded (+8 to PC), however that was never actually done by any code

Now, the PC is always incremented when the pipeline is reloaded
 2022-10-13 00:33:13 -03:00
Rekai Nyangadzayi Musuka 48b81c8e7a chore: dump pipeline state on cpu panic 2022-10-13 00:33:13 -03:00
Rekai Nyangadzayi Musuka 3cf1bf54e9 fix: reimpl THUMB.5 instructions 
pipeline branch now passes arm.gba and thumb.gba again

(TODO: Stop rewriting my commits away)
 2022-10-13 00:33:13 -03:00
Rekai Nyangadzayi Musuka 1f9eeedfe8 fix: impl workaround for stage2 miscompilation 2022-10-13 00:33:13 -03:00
Rekai Nyangadzayi Musuka 72a63eeb98 chore: instantly refill the pipeline on flush 
I believe this to be necessary in order to get hardware interrupts
working.

thumb.gba test 108 fails but I'm committing anyways (despite the
regression) because this is kind of rebase/merge hell and I have
something that at least sort of works rn
 2022-10-13 00:33:13 -03:00
Rekai Nyangadzayi Musuka 2799c3f202 fix: reimpl handleInterrupt code 2022-10-13 00:33:13 -03:00
Rekai Nyangadzayi Musuka b3ada64e64 feat: implement basic pipeline 
passes arm.gba, thumb.gb and armwrestler, fails in actual games
TODO: run FuzzARM debug specific titles
 2022-10-13 00:33:11 -03:00
Rekai Nyangadzayi Musuka 97f48c730e chore(emu): refactor code 2022-10-13 00:29:51 -03:00
Rekai Nyangadzayi Musuka 293fbd9f55 feat(config): add support for (and read from) TOML config file 2022-10-13 00:29:48 -03:00
Rekai Nyangadzayi Musuka 21eddac31e style: improve code quality 2022-10-13 00:23:58 -03:00
Rekai Nyangadzayi Musuka 785135a074 feat: rewrite device ticks 2022-10-13 00:23:58 -03:00
Rekai Nyangadzayi Musuka fd38fd6506 style(scheduler): rename scheduler event handlers 2022-10-13 00:23:58 -03:00
Rekai Nyangadzayi Musuka bcacac64df style: code refactoring 2022-10-13 00:23:58 -03:00
Rekai Nyangadzayi Musuka dc7cad9691 style(apu): split apu.zig into multiple files + refactor 2022-10-13 00:23:58 -03:00
Rekai Nyangadzayi Musuka b5d8a65e69 style(backup): refactor code 2022-10-10 12:01:49 -03:00
Rekai Nyangadzayi Musuka 8028394105 style(flash): move flash code into it's own file 2022-10-10 12:01:49 -03:00
Rekai Nyangadzayi Musuka cb0eb67e4b style(eeprom): move eeprom code to it's own file 2022-10-10 12:00:45 -03:00
Rekai Nyangadzayi Musuka 13f6ee8ec4 style(bus): refactor several hardware abstractions 2022-10-10 11:57:57 -03:00
Rekai Nyangadzayi Musuka 92cfc763c0 chore: move util.zig 2022-09-19 16:07:19 -03:00
Rekai Nyangadzayi Musuka e192c6712f chore: disable audio sync by default 
forgot SDL2 AudioStream doesn't work well for my use-case
 2022-09-18 09:20:01 -03:00
Rekai Nyangadzayi Musuka 3466bf6c0a chore: change default settings 2022-09-18 06:30:39 -03:00
Rekai Nyangadzayi Musuka fbe3de0eb3 chore: reimpl util.escape 
should make use of stdlib when I can
 2022-09-18 06:23:30 -03:00
Rekai Nyangadzayi Musuka 4af144fca2 fix: Detect FRAM ROMs 2022-09-18 06:19:05 -03:00
Rekai Nyangadzayi Musuka 9a8aaba1ab chore: improve util and Gui API 2022-09-18 05:55:15 -03:00
Rekai Nyangadzayi Musuka fa3b9c21b9 chore: move Gpio and Clock structs to separate file 2022-09-18 00:37:45 -03:00
Rekai Nyangadzayi Musuka 50adb5fbac feat: add option to force-enable RTC 2022-09-17 20:27:17 -03:00
Rekai Nyangadzayi Musuka 19d78b9292 feat: auto-detect RTC in commercial ROMS 2022-09-17 20:23:49 -03:00
Rekai Nyangadzayi Musuka a2e702c366 fix: account for lateness in RTC scheduler event 2022-09-17 09:07:31 -03:00
Rekai Nyangadzayi Musuka 12c138364d fix: RTC day is 6 bits wide, not 3 2022-09-16 10:59:41 -03:00
Rekai Nyangadzayi Musuka 7783c11fac feat: put RTC Sync on Scheduler 
TODO: Database to see what games have what GPIO devices
 2022-09-16 10:39:02 -03:00
Rekai Nyangadzayi Musuka 3fc3366c8a chore: import datetime library + default time for RTC 2022-09-16 10:39:02 -03:00
Rekai Nyangadzayi Musuka d6b182f245 fix: ignore RTC Time/DateTime writes 
this falls in-line with better emulators
 2022-09-16 10:39:02 -03:00
Rekai Nyangadzayi Musuka 3857c44e68 chore: use Clock.Writer for Command parsing, delete Clock.Command 2022-09-16 10:39:02 -03:00
Rekai Nyangadzayi Musuka 089c5fa025 feat: implement RTC Read/Writes 2022-09-16 10:39:02 -03:00
Rekai Nyangadzayi Musuka c977f3f965 feat: implement force irqs for GPIO/RTC 2022-09-16 10:38:51 -03:00
Rekai Nyangadzayi Musuka 92417025e9 fix: properly resovle stack UAF 2022-09-16 02:10:41 -03:00
Rekai Nyangadzayi Musuka 1c52c0bf91 chore: shorten `orelse @panic` to `.?` 2022-09-16 02:10:41 -03:00
Rekai Nyangadzayi Musuka 617f7f4690 fix: update GpioData extern union 
u4's are no longer supported in extern unions :\
 2022-09-16 02:10:41 -03:00
Rekai Nyangadzayi Musuka 434a0dfac9 tmp: incomplete impl of GPIO + RTC 2022-09-16 02:10:41 -03:00
Rekai Nyangadzayi Musuka 4ec8dab460 chore: Guilty Gear X expects these I/O Registers 2022-09-14 11:38:26 -03:00
Rekai Nyangadzayi Musuka 59c9ff910e feat: implement open bus for unmapped i/o 2022-09-12 23:18:29 -03:00
Rekai Nyangadzayi Musuka 0027d3f8a3 chore: comment open bus impl 2022-09-11 07:38:55 -03:00
Rekai Nyangadzayi Musuka 65cfc97f28 feat: reimplement audio sync 
APU will now drop samples if the Audio Queue is already full, therefore
creating a "sped-up" effect when the emulator runs faster than 100%
 2022-09-08 20:38:42 -03:00
Rekai Nyangadzayi Musuka fa862f095a chore: move arm/thumb lut idx functions 2022-09-06 23:58:24 -03:00
Rekai Nyangadzayi Musuka 3fb7f2f814 chore: better conform to zig idioms 2022-09-03 18:30:48 -03:00
Rekai Nyangadzayi Musuka 59669ba3a5 chore: rename arm7tdmi variables to just cpu 
Less verbose, specifying arm7tdmi doesn't really do much when there's
no other CPU in the system
 2022-09-03 17:56:37 -03:00
Rekai Nyangadzayi Musuka 6a798d2c9d chore: allocate sprite array on heap 
Each Sprite optional is 10 bytes meaning I'm allocating 1.28Kb on the
stack which isn't necessary.
 2022-08-29 01:07:25 -05:00
Rekai Nyangadzayi Musuka 5f8c6833f4 chore: improve init/deinit methods 2022-08-29 01:07:25 -05:00
Rekai Nyangadzayi Musuka aa52bb5917 chore: reorganize some code 2022-08-26 14:13:49 -05:00
Rekai Nyangadzayi Musuka e57f918856 chore: pass the allocator as an argument more often 
As of right now, I think the only cases where I shouldn't explicitly pass an allocator
are in read/write functions and deinits
 2022-08-26 13:54:38 -05:00
Rekai Nyangadzayi Musuka e5b7441740 fix: resolve use-afer-free in backup.zig 
This worked fine on stage1, and works fine in debug in stage3.
However, stage3 ReleaseSafe would panic due to what I assume must
have been an undefined behaviour optimization.

While I'm happy that I was quickly made aware of the issue thanks to
the safety checks in ReleaseSafe I do wish that this issue showed itself
in Debug, since I *am* using the GPA
 2022-08-26 13:04:09 -05:00
Rekai Nyangadzayi Musuka 2ab8769b7a feat: Get ZBA working on Zig's new stage2/stage3 compiler 2022-08-21 12:28:31 -05:00
Rekai Nyangadzayi Musuka 3c3c0d32dd chore: move window scale const to emu.zig 2022-08-08 11:03:23 +02:00
Rekai Nyangadzayi Musuka 739db99c83 fix: reimpl debug reads w/out throwing away *const Self 2022-08-07 05:11:29 -05:00
Rekai Nyangadzayi Musuka 2c8616f610 feat: reimplement cpu logging 2022-07-27 14:50:28 -03:00
Rekai Nyangadzayi Musuka 53eec5c3ff chore: don't init bus in Arm7tdmi init 2022-07-27 13:44:24 -03:00
Rekai Nyangadzayi Musuka c397b7069d feat: move arm instr decoding to module 2022-07-27 13:23:29 -03:00
Rekai Nyangadzayi Musuka 9d037fdc3e feat: move thumb instr decoding to module 2022-07-27 13:10:58 -03:00
Rekai Nyangadzayi Musuka 53191b0eeb chore: change directory structure 2022-07-22 21:11:19 -03:00