Repository navigation
match(pal): add twenty-four strictly verified overlay routines - #133
Conversation
Codex Review Summary: ✅ CompletedThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
ByteMatch progress reportSCUS_974.65
No function changes. SCES_532.85
✅ Newly matched (24)
Commit e6f98c0 · Ratchet: Deadlocked on ByteMatch |
Ten PAL level functions verified byte-identical (strict MATCH) against SCES_532.85: six multiplayer-menu helpers in level_00, a cheats panel routine in level_04, two chain routines in level_08 and one in level_11. The sources are m2c output cleaned of solver macros.
-Wuninitialized does not affect code generation; the rebuilt objects are byte-identical and still strict MATCH.
Shaar-games
changed the title
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e6f98c0a58
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| func_L11_003FF000(i); | ||
| result = *(s32 *)(i + 172); | ||
| } | ||
| return result; |
There was a problem hiding this comment.
When the list is empty (D_00220F50 == 0, also its initializer here), the loop never assigns result, but this line returns it. This violates the matching protocol in docs/QUEUE.md, which explicitly forbids “read of a local that was never assigned” and requires candidates needing such reads to remain in nonmatching/, even if their generated bytes compare exactly. Recover the retail-supported return contract and reverify the candidate, or stop publishing this implementation as a match.
AGENTS.md reference: AGENTS.md:L4-L5
Useful? React with 👍 / 👎.
Strictly verified (pal-solver verify, byte-identical linked code and verified data/BSS) against PAL SCES_532.85: main/multiplayer: 00484F88 net: 01EEE13C L00: 005300C8 00530410 00567138 00605A70 00605B58 006A09B8 006A0F88 L02: 004CF080 L05: 003E47B8 004EF810 004F0FD8 004F4E68 L06: 004C5DF0 L08: 004B9CC0 L11: 00408630 L15: 004EDFA0 L41: 00410098 00410298 00444C58 00444D18 004469F8 PAL catalogue 8,044 -> 8,067 of 18,134 functions (incl. OpenRAC#133 baseline).
Decodes 24 RAC4 PAL (SCES_532.85) functions, each strictly verified (byte-identical compiled code) and registered in the maps.
Original 14:
New in this PR:
All are exact matches in the regenerated maps; both progress reports and the README table are updated.