Player::CleanupResources compiles again, and it was a match all along

[andrewboudreau]

A match that has been sitting unreadable in the tree

src/_ZN6Player16CleanupResourcesEv.cpp has not compiled since it gained
#include "decl_common.h". That header declares, inside extern "C":

extern void func_ov002_020bdd2c(void*);
extern void func_02073244(void*, int, int, void (*)(void*));

and the file then declared the same names again taking char * and
(int, int, int, void (*)(void)). Same C linkage, different parameter types — so it
is not a redeclaration but an attempt to overload a C-linkage name, which mwccarm
rejects outright:

src/_ZN6Player16CleanupResourcesEv.cpp:17: illegal function overloading
... :18, :19, :20, :23
Errors caused tool to abort.

Nothing caught it, because the failure removes the file from every gate that could
have. A file that will not compile is not eligible, so it is not enrolled; its
delinks entry carries no complete, so dsd hands the range to the original ROM bytes
and the build stays 106/106 without the source ever being touched. It went unbuilt
through #919, #994, #1032 and #1382 — the last of which edited it.

This is #1367's NO-SYM category doing exactly what it was added for. It only
surfaced now because the fan-out of #1390's header change reached the file, and
because #1392 stops the validator crashing before it can say so.

The function is a byte match

That is the real result. Once it compiles, all 0x730 bytes at 0x020e32d4 in
ov002 reproduce under the pin with correct relocation destinations. Nothing about
the logic needed repair — only the declarations. So it is enrolled here rather than
left as rombytes, and the tree gains a function it has had all along.

The fix is to declare only what decl_common.h does not. Three names remain because
that header lacks them; _ZN7Vector3D1Ev is spelled with the this it actually
takes, which is also what makes the func_02073244 argument type-correct without a
cast.

Verified

gate	result
build_pin	(True, '2004/b56') — 0x020e32d4, size 0x730, ov002
pr_linkcheck	ok, 1 slot, correct destinations
eligible.py	10,814 / 11,162 — main is 10,813, and the diff is this symbol alone
enroll.py	1 delinks file changed, 1 line added
rombuild -j16	106/106 exact, 10,814 source-built, 0 mismatching
progress	1,943,084 → 1,944,924 code bytes, 87.88% → 87.96%
check_data_definitions	10,972 objects, none define a ROM data symbol
check_duplicate_sources	11,282 stems, none doubled
port_refcheck	393 references, all resolve
prepush_attribution	11,288 tracked, 0 changed, 0 lost

Worth a follow-up

331 delink entries are unbuilt and carry no NONMATCHING banner. Most will be
ineligible for ordinary reasons, but this file proves at least one of them was a
finished match held out of the ROM by a declaration mismatch. Sweeping them for the
same signature is a separate change; I have not measured how many are in that state
and am not claiming a number.

🤖 Generated with Claude Code

https://claude.ai/code/session_01XxDmkQ47fWa3GB6mj8xEQe

[tangos-validator]

✅ PR validation — Passed

Committed merge introduces no reconstruction or attribution regression.

Full merge validation

Check	Result
Committed test merge	yes
Matched functions	11,178 / 11,347 (98.5%, +0)
Matched code bytes	2,078,800 / 2,211,124 (94.0%, +0)
Tracked source enrollment	10,786 functions, 1,938,036 bytes (87.65%, +1,840)
Perfect source moves	0 R100
Contributor credit	0 added, 0 changed, 0 lost
Relocation check	1 checked; 1 VERIFIED
Port reference check	393 checked; 0 stale
Module fidelity	106/106 exact; 100.000000% compared bytes
Code linked from verified source	10,814 functions, 1,944,924 bytes (87.96%)

Per-file link-check detail

All 1 changed file(s) compile to the ROM byte-for-byte with correct relocation targets.

File	Symbol	Result	Slots checked
src/_ZN6Player16CleanupResourcesEv.cpp	_ZN6Player16CleanupResourcesEv	✅ verified	1

_{The private worker commits a test merge, builds the stock ROM profile, compares every executable module, measures matched and source-built code, checks contributor lineage, and verifies affected relocations. The mod profile is opt-in and is not part of this merge gate.}