byuu says:
(Windows: compile with -fpermissive to silence an annoying error. I'll
fix it in the next WIP.)
I completely replaced the time management system in higan and overhauled
the scheduler.
Before, processor threads would have "int64 clock"; and there would
be a 1:1 relationship between two threads. When thread A ran for X
cycles, it'd subtract X * B.Frequency from clock; and when thread B ran
for Y cycles, it'd add Y * A.Frequency from clock. This worked well
and allowed perfect precision; but it doesn't work when you have more
complicated relationships: eg the 68K can sync to the Z80 and PSG; the
Z80 to the 68K and PSG; so the PSG needs two counters.
The new system instead uses a "uint64 clock" variable that represents
time in attoseconds. Every time the scheduler exits, it subtracts
the smallest clock count from all threads, to prevent an overflow
scenario. The only real downside is that rounding errors mean that
roughly every 20 minutes, we have a rounding error of one clock cycle
(one 20,000,000th of a second.) However, this only applies to systems
with multiple oscillators, like the SNES. And when you're in that
situation ... there's no such thing as a perfect oscillator anyway. A
real SNES will be thousands of times less out of spec than 1hz per 20
minutes.
The advantages are pretty immense. First, we obviously can now support
more complex relationships between threads. Second, we can build a
much more abstracted scheduler. All of libco is now abstracted away
completely, which may permit a state-machine / coroutine version of
Thread in the future. We've basically gone from this:
auto SMP::step(uint clocks) -> void {
clock += clocks * (uint64)cpu.frequency;
dsp.clock -= clocks;
if(dsp.clock < 0 && !scheduler.synchronizing()) co_switch(dsp.thread);
if(clock >= 0 && !scheduler.synchronizing()) co_switch(cpu.thread);
}
To this:
auto SMP::step(uint clocks) -> void {
Thread::step(clocks);
synchronize(dsp);
synchronize(cpu);
}
As you can see, we don't have to do multiple clock adjustments anymore.
This is a huge win for the SNES CPU that had to update the SMP, DSP, all
peripherals and all coprocessors. Likewise, we don't have to synchronize
all coprocessors when one runs, now we can just synchronize the active
one to the CPU.
Third, when changing the frequencies of threads (think SGB speed setting
modes, GBC double-speed mode, etc), it no longer causes the "int64
clock" value to be erroneous.
Fourth, this results in a fairly decent speedup, mostly across the
board. Aside from the GBA being mostly a wash (for unknown reasons),
it's about an 8% - 12% speedup in every other emulation core.
Now, all of this said ... this was an unbelievably massive change, so
... you know what that means >_> If anyone can help test all types of
SNES coprocessors, and some other system games, it'd be appreciated.
----
Lastly, we have a bitchin' new about screen. It unfortunately adds
~200KiB onto the binary size, because the PNG->C++ header file
transformation doesn't compress very well, and I want to keep the
original resource files in with the higan archive. I might try some
things to work around this file size increase in the future, but for now
... yeah, slightly larger archive sizes, sorry.
The logo's a bit busted on Windows (the Label control's background
transparency and alignment settings aren't working), but works well on
GTK. I'll have to fix Windows before the next official release. For now,
look on my Twitter feed if you want to see what it's supposed to look
like.
----
EDIT: forgot about ICD2::Enter. It's doing some weird inverse
run-to-save thing that I need to implement support for somehow. So, save
states on the SGB core probably won't work with this WIP.
byuu says:
Changelog:
- GB core code cleanup completed
- GBA core code cleanup completed
- some more cleanup on missed processor/arm functions/variables
- fixed FC loading icarus bug
- "Load ROM File" icarus functionality restored
- minor code unification efforts all around (not perfect yet)
- MMIO->IO
- mmio.cpp->io.cpp
- read,write->readIO,writeIO
It's been a very long work in progress ... starting all the way back with
v094r09, but the major part of the higan code cleanup is now completed! Of
course, it's very important to note that this is only for the basic style:
- under_score functions and variables are now camelCase
- return-type function-name() are now auto function-name() -> return-type
- Natural<T>/Integer<T> replace (u)intT_n types where possible
- signed/unsigned are now int/uint
- most of the x==true,x==false tests changed to x,!x
A lot of spot improvements to consistency, simplicity and quality have
gone in along the way, of course. But we'll probably never fully finishing
beautifying every last line of code in the entire codebase. Still,
this is a really great start. Going forward, WIP diffs should start
being smaller and of higher quality once again.
I know the joke is, "until my coding style changes again", but ... this
was way too stressful, way too time consuming, and way too risky. I'm
too old and tired now for extreme upheavel like this again. The only
major change I'm slowly mulling over would be renaming the using
Natural<T>/Integer<T> = (u)intT; shorthand to something that isn't as
easily confused with the (u)int_t types ... but we'll see. I'll definitely
continue to change small things all the time, but for the larger picture,
I need to just accept the style I have and live with it.
byuu says:
Changelog:
- fixed FC AxROM / VRC7 regression
- BitField split to BooleanBitField/NaturalBitField (in preparation
for IntegerBitField)
- BitFieldReference removed
- GB CPU cleaned up
- GB Cartridge + Mappers cleaned up
- SFC CGRAM is now emulated as uint15[256] instead of uint[512]
- sfc/ppu/memory.cpp no longer needed; removed
- purged SFC Debugger hooks for now (some of the operator[] calls were
bypassing them anyway)
Unfortunately, for reasons that defy all semblance of logic, the CGRAM
change caused a slight speed hit. As have the last few changes. We're
now down to around 129.5fps compared to 123.fps for v099 and 134.5fps
at our peak (v099r01-r02).
I really like the style I came up with for the Game Boy mappers to settle
the purpose(ROM,RAM) vs (rom,ram)Purpose naming convention. If I ever get
around to redoing the NES mappers, that's likely the approach I'll take.
byuu says:
Changelog:
- fixed Super Game Boy regression from v096r04 with bottom tile row
flickering
- fixed GB STAT IRQ regression from previous WIP
- Altered Space is now playable
- GBVideoPlayer isn't; but nobody seems to know exactly what weird
hardware quirk that one relies on to work
- ~3-4% speed improvement in SuperFX games by eliminating function<>
callback on register assignments
- most noticeable in Doom in-game; least noticeable on Yoshi's Island
title screen (darn)
- finished GSU core and SuperFX coprocessor code cleanups
- did some more work cleaning up the LR35902 core and GB CPU code
Just a fair warning: don't get your hopes up on these GB
fixes. Cliffhanger now hangs completely (har har), and none of the
other bugs are fixed. We pretty much did all this work just for Altered
Space. So, I hope you like playing Altered Space.
byuu says:
Changelog:
- GNUmakefile: reverted $(call unique,) to $(strip)
- processor/r6502: removed templates; reduces object size from 146.5kb
to 107.6kb
- processor/lr35902: removed templates; reduces object size from 386.2kb
to 197.4kb
- processor/spc700: merged op macros for switch table declarations
- sfc/coprocessor/sa1: partial cleanups; flattened directory structure
- sfc/coprocessor/superfx: partial cleanups; flattened directory structure
- sfc/coprocessor/icd2: flattened directory structure
- gb/ppu: changed behavior of STAT IRQs
Major caveat! The GB/GBC STAT IRQ changes has a major bug in it somewhere
that's seriously breaking most games. I'm pushing the WIP anyway, because
I believe the changes to be mostly correct. I'd like to get more people
looking at these changes, and also try more heavy-handed hacking and
diff comparison logging between the previous WIP and this one.
byuu says:
I refactored my schedulers. Added about ten lines to each scheduler, and
removed about 100 lines of calling into internal state in the scheduler
for the FC,SFC cores and about 30-40 lines for the other cores. All of
its state is now private.
Also reworked all of the entry points to static auto Enter() and auto
main(). Where Enter() handles all the synchronization stuff, and main()
doesn't need the while(true); loop forcing another layer of indentation
everywhere.
Took a few hours to do, but totally worth it. I'm surprised I didn't do
this sooner.
Also updated icarus gmake install rule to copy over the database.
byuu says:
This release features improvements to all emulation cores, but most
substantially for the Game Boy core. All of blargg's test ROMs that pass
in gambatte now either pass in higan, or are off by 1-2 clocks (the
actual behaviors are fully emulated.) I consider the Game Boy core to
now be fairly accurate, but there's still more improvements to be had.
Also, what's sure to be a major feature for some: higan now has full
support for loading and playing ordinary ROM files, whether they have
copier headers, weird extensions, or are inside compressed archives. You
can load these games from the command-line, from the main Library menu
(via Load ROM Image), or via drag-and-drop on the main higan window. Of
course, fans of game folders and the library need not worry: that's
still there as well.
Also new, you can drop the (uncompressed) Game Boy Advance BIOS onto the
higan main window to install it into the correct location with the
correct file name.
Lastly, this release technically restores Mac OS X support. However,
it's still not very stable, so I have decided against releasing binaries
at this time. I'd rather not rush this and leave a bad first impression
for OS X users.
Changelog (since v096):
- higan: project source code hierarchy restructured; icarus directly
integrated
- higan: added software emulation of color-bleed, LCD-refresh,
scanlines, interlacing
- icarus: you can now load and import ROM files/archives from the main
higan menu
- NES: fixed manifest parsing for board mirroring and VRC pinouts
- SNES: fixed manifest for Star Ocean
- SNES: fixed manifest for Rockman X2,X3
- GB: enabling LCD restarts frame
- GB: emulated extra OAM STAT IRQ quirk required for GBVideoPlayer
(Shonumi)
- GB: VBK, BGPI, OBPI are readable
- GB: OAM DMA happens inside PPU core instead of CPU core
- GB: fixed APU length and sweep operations
- GB: emulated wave RAM quirks when accessing while channel is enabled
- GB: improved timings of several CPU opcodes (gekkio)
- GB: improved timings of OAM DMA refresh (gekkio)
- GB: CPU uses open collector logic; return 0xFF for unmapped memory
(gekkio)
- GBA: fixed sequencer enable flags; fixes audio in Zelda - Minish Cap
(Jonas Quinn)
- GBA: fixed disassembler masking error (Lioncash)
- hiro: Cocoa support added; higan can now be compiled on Mac OS X 10.7+
- nall: improved program path detection on Windows
- higan/Windows: moved configuration data from %appdata% to
%localappdata%
- higan/Linux,BSD: moved configuration data from ~/.config/higan to
~/.local/higan
byuu says:
Changelog:
- fixed S-DD1 RAM writes (Star Ocean audio fixed)
- applied all of the DMG test ROM fixes discussed earlier; passes many
more test ROMs now
- at least until the GBVideoPlayer is working: for debugging purposes,
CPU/PPU single-step now instead of sync just-in-time (~30% slower)
- fixed OS X crash on NSTextView (hopefully, would be very odd if not)
Unfortunately passing these test ROMs caused my favorite GB/GBC game to
break all of its graphics =(
Shin Megami Tensei - Devichil - Kuro no Sho (Japan) is all garbled now.
I'm really quite bummed by this ... but I guess I'll go through and
revert r04's fixes one at a time until I find what's causing it.
On the plus side, Astro Rabby is playable now. Still acts weird when
pressing B/A on the first screen, but the start button will start the
game.
EDIT: got it. Shin Megami Tensei - Devichil requires FF4F (VBK) to be
readable. Before, it was always returning 0x00. With my return 0xFF
patch, that broke. But it should be returning the VBK value, which also
fixes it. Also need to handle FF68/FF6A reads. Was really hoping that'd
help GBVideoPlayer too, but nope. It doesn't read any of those three
registers.
byuu says:
Changelog:
- restructured the project and removed a whole bunch of old/dead
directives from higan/GNUmakefile
- huge amounts of work on hiro/cocoa (compiles but ~70% of the
functionality is commented out)
- fixed a masking error in my ARM CPU disassembler [Lioncash]
- SFC: decided to change board cic=(411,413) back to board
region=(ntsc,pal) ... the former was too obtuse
If you rename Boolean (it's a problem with an include from ruby, not
from hiro) and disable all the ruby drivers, you can compile an
OS X binary, but obviously it's not going to do anything.
It's a boring WIP, I just wanted to push out the project structure
change now at the start of this WIP cycle.
byuu says:
Changelog:
- added Cocoa target: higan can now be compiled for OS X Lion
[Cydrak, byuu]
- SNES/accuracy profile hires color blending improvements - fixes
Marvelous text [AWJ]
- fixed a slight bug in SNES/SA-1 VBR support caused by a typo
- added support for multi-pass shaders that can load external textures
(requires OpenGL 3.2+)
- added game library path (used by ananke->Import Game) to
Settings->Advanced
- system profiles, shaders and cheats database can be stored in "all
users" shared folders now (eg /usr/share on Linux)
- all configuration files are in BML format now, instead of XML (much
easier to read and edit this way)
- main window supports drag-and-drop of game folders (but not game files
/ ZIP archives)
- audio buffer clears when entering a modal loop on Windows (prevents
audio repetition with DirectSound driver)
- a substantial amount of code clean-up (probably the biggest
refactoring to date)
One highly desired target for this release was to default to the optimal
drivers instead of the safest drivers, but because AMD drivers don't
seem to like my OpenGL 3.2 driver, I've decided to postpone that. AMD
has too big a market share. Hopefully with v093 officially released, we
can get some public input on what AMD doesn't like.
byuu says:
This will be another massive diff from the previous version.
All of higan was updated to use the new foo& bar syntax, and I also
updated switch statements to be consistent as well (but not in the
disassemblers, was starting to get an RSI just from what I already did.)
phoenix/{windows, cocoa, qt} need to be updated to use "string foo"
instead of "const string& foo", and after that, the major diffs should
be finished.
This archive is the first time I'm posting my copy-on-write,
size+capacity nall::string class, so any feedback on that is welcome as
well.
byuu says:
Basically just a project rename, with s/bsnes/higan and the new icon
from lowkee added in.
It won't compile on Windows because I forgot to update the resource.rc
file, and a path transform command isn't working on Windows.
It was really just meant as a starting point, so that v091 WIPs can flow
starting from .00 with the new name (it overshadows bsnes v091, so
publicly speaking this "shouldn't exist" and will probably be deleted
from Google Code when v092 is ready.)
Tim Allen