Files
TeslaRel410/dpl3-revive/patha/xref.py
T
CydandClaude Fable 5 afc3fd839e Vendor dpl3-revive: the Division/DPL3 renderer, now ours
Bring the graphics-dev collaborator's dpl3-revive into the repo as first-class
project code (they've handed it off; it's ours now). This is the proven
Division renderer that our in-process rt_draw has been trying to be.

What's here:
- parser/  B2Z/V2Z/SVT/SCN/SPL/BGF/BMF/BSL decoders (pure Python).
- spec/    reverse-engineered format + the definitive VelociRender wire
           protocol (from the original DIVISION source, matches our live
           VPX node/action tables exactly).
- source-ref/  read-only copies of the original DIVISION C (BIZREAD.C,
           DPLTYPES.H, DPL.H) that define the formats.
- patha/   the "virtual VelociRender board": vrboard.py (24-action protocol
           server), vrview.py (numpy software rasterizer, the reference),
           vrview_gl.py (moderngl GPU backend, 832x512@60Hz), plus the
           run/replay/regress tooling and evidence renders. Drives FLYK/BLADE/
           Star Trek demos AND our btl4opt/rpl4opt game binaries.
- viewer/  WebGL archive generators (.py); prebuilt HTML/data regeneratable.
- samples/ small test models/textures.
- bt*.raw.bin  real BTL4OPT arena wire captures (kept for offline renderer
           testing/regression against OUR game).

.gitignore keeps the multi-hundred-MB demo capture dumps + debug logs +
regeneratable viewer bundles out of history (they stay on disk).

Phase 0 of the integration is validated: their board decodes our bt8 capture
with zero errors (3748 nodes, 507 instances, 4 mechs) and renders our arena
(terrain/dome/sky, correct Division DAC gamma). Plan + status in memory;
integration continues in emulator/RENDERER-COLLAB.md.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-05 22:06:25 -05:00

63 lines
2.2 KiB
Python

#!/usr/bin/env python3
"""
xref.py -- locate a string in FLYK.EXE's data object, find code that references its
address (absolute imm32 operands), and disassemble a window around each reference.
Used to reverse the boot-handshake path: anchor on the error/format strings we saw at
runtime, then read the surrounding logic to learn exactly what the board must send.
python xref.py FLYK.EXE "velocirender_input failed"
python xref.py FLYK.EXE 0x00041234 # disasm at a VA
"""
import sys, struct
from leimage import LE
from capstone import Cs, CS_ARCH_X86, CS_MODE_32
def le32(v): return struct.pack('<I', v & 0xffffffff)
def find_all(buf, needle, start=0):
hits=[]; i=buf.find(needle, start)
while i!=-1: hits.append(i); i=buf.find(needle, i+1)
return hits
def main():
path = sys.argv[1]; arg = sys.argv[2]
le = LE(path)
cbase, code = le.image(1) # object 1: code
dbase, data = le.image(2) # object 2: data
md = Cs(CS_ARCH_X86, CS_MODE_32); md.detail = False
def disasm(va, back=32, fwd=64):
off = va - cbase - back
if off < 0: off = 0
for ins in md.disasm(code[off:va-cbase+fwd], cbase+off):
mark = " <=" if ins.address >= va-3 and ins.address <= va+1 else ""
print(f" {ins.address:#010x} {ins.mnemonic:7} {ins.op_str}{mark}")
if arg.startswith('0x'):
va = int(arg, 16)
print(f"--- disasm @ {va:#x} ---"); disasm(va, 0, 160); return
needle = arg.encode('latin1')
soffs = find_all(data, needle)
if not soffs:
print("string not found in data object"); return
for so in soffs[:4]:
sva = dbase + so
# show the full string
end = data.find(b'\x00', so)
s = data[so:end].decode('latin1', 'replace')
print(f'\n=== string @ {sva:#x}: "{s}" ===')
# Watcom LE stores the OBJECT-RELATIVE offset in code; the fixup adds the base.
refs = find_all(code, le32(so)) or find_all(code, le32(sva))
if not refs:
print(" (no code references found)")
for r in refs[:6]:
rva = cbase + r
print(f" referenced by imm32 at code VA {rva:#x}:")
disasm(rva, 40, 40)
print()
if __name__ == '__main__':
main()