Mostly relocating rodata from data/* files to the appropriate
asm/nonmatching/* files, although there is some plain data
relocating to be found in here too.
Big change to note is the addition of the code_802AAA70.c file,
split from memory.c. Based on some padding in the ro/data for
memory it seems clear that a split is needed, although it is
not clear that exact spot chosen for the split is correct.
Still, it gets the file padding for the ro/data and text segments
correct so it can't be too wrong.
Also included is an update to the asm_processor tool to support
the .hword directive. I made a PR to the main repo of that
project with an identical change and it was accepted.
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Remove some bin2c calls from the Makefile
* Found a way to make n64graphics use shared tluts
* Added an option that makes n64graphics operate in a mode where it
takes an image and tlut (as pngs) and spits out the appropriate
CI8 indices inc.c form
* Turns out all the TLUTs for Lakitu are kept in one place, far away
from the textures that use them
* Names some textures and tluts
* Pretty helpful actually, helped identify where a few different
objects types are initialized. Should be a little useful to
whoever dives into the object stuff more deeply
* Minor cleanup in n64graphics
* Added some comments, gave more appropriate names to some
variables.
* Leaving the new mode argument as `-Z` for now
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Textures compile using makefile rules
* Patched n64 graphics memory leak and ci output
* Added memset16safe to initialize ci palettes to magic numbers (transparent pixels).
---------
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
Co-authored-by: MegaMech <MegaMech@users.noreply.github.com>
* Give names to some textures
* Match a couple functions
* Remove a use of bin2c from the makefile
* Its a little strange, but two of the portrait textures share
a TLUT. I believe this necessitates that the portraits be
part of the same PNG file.
* 2nd portrait referenced in undefined_syms.txt
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
Identified a couple more lists of object list indices, added some
comments indicating what they're used for. Those comments are
most likely non-exhaustive so there's still some investigation
to be done there.
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Complete matches in ceremony_and_credits
* struct_D_802876D0 is the real CinematicCamera
* Replaced a bunch of Camera pointer args with
CinematicCamera pointers instead. Only a couple functions use plain Camera pointers
* Moved some type definitions into the header file
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Matches, add permute, and add notice to quick runners
* 80281fa0 matches, render_courses, and create defines for audio
* Renamed 80281fa0.h, func naming, and deleted data_121DA0
* Added clang format
* Matched load_surface_map
* Match func in memory.c
* some matches in audio in port_eu, heap, and external.
* update port eu
* Add export lang=c to makefile
* near match for func_800CBB48
* Update src/audio/heap.c
A lot of these matches come from observing the external audio
stuff in SM64 and realizing that a bunch a couple of the structs
in MK64 are just expanded version of their SM64 counterpart.
Presumably all the expansions are for handling the doppler effect,
although I can't say that with certainty
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Lots of audio decomp
* effects and playback are 100% matched
* Took some notes from the SM64 Makefile to get the right compiler
options for a couple files
* Added the audio files to rename_symbol utility
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Updated the Makefile to make the VERSION_US file
get used when compiling stuff.
* Updated the abi.h file to match SM64, with some modifications to
support the currently targeted version (US).
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Matches and m2c refreshing for code_80057C60
* Convert some Vtx's into a Mtx
Based on the usage of D_0D008E98, it seems like it should be a
Mtx. So this commit changes some Vtx's into a Mtx
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
Couple matches functions, an update to a struct, a bunch of header
files adjustments, and some mips_to_c regenerations
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* More matches for code_800431B0
* Also added a header for the common_textures file since some other
files use the Gfx/Vtx variables defined there
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Required making an adjustment to the object struct, which was required to fix some other files.
* Decompiled some gfx
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Add a lot of header files
Did my best to move all relevant function prototypes and external
variable declarations into the header files too.
Matched a few functions in the process
* Updated progress.py
Some of the listed files/folders have been deleted/renamed,
so I reflected those changes in progress.py
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
* Match a lot of functions, add header for objects
* Refreshed a TON of functions and created/added to several
header files
* Add header for code_800431B0, address comments
Signed-off-by: Taggerung <tyler.taggerung@gmail.com>
Tyler McGavran