Drew T db4fddc44e feat(phase-15): T6g medium band — harvest+propagate 65 shared fns; fleet 42.14% -> 44.64%
- medium harvest (callee+data-sig-aware) on the 293 still-stub call-heavy fns (nins 41-150):
  30-agent Workflow + a 3-target gap-fill (mandatory produced-vs-expected reconcile). Gate:
  73 verified (25% — these are the harder GTE/matrix/regalloc functions; 220 quarantined to
  .run/drafts-T6c-fail for a later recovery pass).
- dedup_propagate.py: broaden the struct/union guard to skip ANY struct/union-touching body
  (not just definitions) — agents used colliding generic struct names in params/casts/externs
  too. 37 skipped; 65 propagated fleet-wide, 134 overlays byte-identical; dedup 1076 -> 1141
  groups, dedup_integrate --check 1141 validated / 0 failed.
- fleet REAL substantive 143,322 -> 151,908; byte-identical 42.14% -> 44.64%.
- R22 clean rebuild: 136/136 byte-identical, 0 extract failures.
- open tail (paused here): 285+220 quarantined drafts (regalloc/struct/§10 residuals), the
  struct-using shared fns (shared-types-header follow-up), 077-unique remainder, T7 giants.
2026-06-17 13:21:36 -06:00
2026-06-10 22:02:07 -06:00
2026-06-10 22:02:07 -06:00

BFM-decomp

A matching decompilation of Brave Fencer Musashi (PlayStation, SLUS-00726, USA 1998) — the first public decompilation effort for this game.

What "matching" means

The goal is C source code that, compiled with the original-era toolchain (PsyQ 4.x / GCC 2.7.2-family + ASPSX via maspsx), produces a byte-for-byte identical SLUS_007.26 and, eventually, byte-identical overlay binaries. SHA1 checksums are the ground truth; "functionally equivalent" does not count.

No ROM content

This repository contains no game assets, no disassembly output, and no ROM-derived data — only source code, build configuration, symbol names/addresses, hashes, and documentation. To build or contribute you must provide your own dump of the game disc (4-track BIN/CUE, redump layout). See .gitignore for the firewall.

Project status

Gen1 (foundation) complete — the matching pipeline is proven end-to-end. make extract && make build && make check rebuilds SLUS_007.26 byte-for-byte identical (SHA1 143dbb89…) from C + assembly, reproducibly across many sessions.

  • Compiler pinned by evidence: gcc-2.7.2-psx -O2 -G0 -mips1 -mcpu=3000 + maspsx --aspsx-version=2.56 --expand-div.
  • 52 functions hand-matched to byte-identical machine code — including the LZSS streaming decompressor — with a decomp-permuter + matching-cookbook "flywheel" to accelerate the next.
  • 959 PsyQ SDK functions linked byte-identical (libcd, libgs, libgte, libspu/libsnd, libgpu, libc2, libmcrd, libapi/libcard, libetc) straight from the real PsyQ 4.0 libraries instead of re-decompiling them — bringing byte-identical-from-source coverage of the EXE to ~50%.
  • File-loader / overlay system reverse-engineered, with the resident engine blob + location overlays' load addresses proven byte-identical against a live PCSX-Redux RAM dump.

About half the EXE is still INCLUDE_ASM stubs (correct bytes, not yet C), and the bulk of the game lives in compressed overlays inside the .CD archives — Gen2 (overlays & engine at scale) is underway:

  • The build toolchain is binary-agnostic (one parameterized pipeline builds any binary), and the always-resident engine blob rebuilds byte-for-byte from source (SHA1 8e17e02f…) — the second binary reconstructed exactly, after the EXE — and is now 86% hand-matched C (123 / 146 functions, up from 0): its scripting turned out to be compiled-MIPS state/mode dispatch, not a bytecode VM, and the save-file + sound (SQV) formats are documented. The harvest used a reusable swarm-of-agents + bit-for-bit byte-gate method (a wrong match can't be accepted) — tools/harvest_verify.py + tools/match_one.py, which carry straight into the overlay phase.
  • A cross-binary deduplication pipeline is live: a Ghidra-free signer fingerprints all 134 location overlays, and the report finds ~9,000 byte-identical function groups shared across binaries (~28 MB of collapsible code) — a single engine function is byte-identical in all 134 overlays. This is "one match unlocks many": each engine match will be auto-credited across the overlay fleet.

Current phase and detailed progress live in phase-ends/ (newest PhaseEnd_*.md = current state); methodology, rules, and the full roadmap are in PROJECT_CONTEXT.md; environment setup in docs/SETUP.md.

This project is developed primarily by Claude Code driving Ghidra through an MCP server; see CLAUDE.md.

License

Private repository for now. AGPL-3.0 is planned at public release, modeled on sotn-decomp. tools/brave-CUE/ is CUE's BRAVE extractor (GPL, source included) and retains its own license.

S
Description
No description provided
Readme AGPL-3.0 612 MiB
Languages
C 96.2%
Python 3.4%
Makefile 0.2%
Shell 0.1%