mirror of https://github.com/zeldaret/mm.git
23d63852de
* Restore padding in ObjVisiblock struct * Copy overlays out of compiled `code.elf` to put into the ROM This is just a "tempory fix" It seems like the built code/assets in `code.elf` should not be copied into `build/baserom/...` (over the original ROM's files) but instead into a `build/decomp/...` tree or similar. `dmadata_table.txt` would also need to be updated to read from the correct location. * Use dmadata_table.txt to generate build rules `makerom_files.txt` & `makerom_uncompressed_files.txt` contained a lot of the same data in `dmadata_table.txt`, so I added a small python script to generate this information into `build/` Segments are no longer dumped out of `code.elf` into `build/baserom/``, instead they are put in `build/binary/`. `linker_scripts/dmadata_script.txt` was checked in, but generated by `dmadata.py`. I deleted it / moved it to `build/dmadata_script.txt.pre`. I also introduced some sentinel files (`dep`). I ended up needing these to make incremental builds work smoothly? (Without them, there were a lot of steps that were getting re-triggered on every build.) If this style isn't welcome, I can try to fiddle with the Makefile more to try to avoid having them? * Restore padding in BgLbfshot struct * Touch sentinel file before command; rm on failure * Restore padding in ObjKepnKoya struct * Ensure asm/ directories exist before disasm steps * Clean up Makefile rules * Set default goal; silent objcopy; fix code_script path * Fix ovl_En_Ginko_Man, ovl_Obj_Lightswitch merge ovl_En_Encount2 still needs work to bring back to matching * Fix ovl_En_Encount2 merge |
||
|---|---|---|
| .. | ||
| dmadata_table.txt | ||
| files.txt | ||
| functions.txt | ||
| objects.txt | ||
| variables.txt | ||
| vrom_variables.txt | ||