Porting tools (b): liveness, don't-care declarations, poisoning (fixes #38) #52
Loading…
Reference in a new issue
No description provided.
Delete branch "liveness"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Liveness per routine (
tools/liveness.py, US 4 031 / JP 4 092 routines, conservative), drafts drop dead computations and bridges store back live values only,tools/dontcare_check.pyas testrom.dontcare, shadow skips declared don't-care values,--port poison=on. Reuses the user's CT liveness tool.Evidence: 9 field drafts 1 744 -> 1 709 lines, flag assignments 230 -> 170; draft_check title_to_mines_save + field_menu_tour US+JP 291 836 calls, 0 mismatch, 0 fallback, also with poison on (end hashes match); 10 registry don't-care declarations, 0 refuted; false claim on GetObjMapPtr caught by poison. debug and clang
-LE slow27/27, 0 warnings. Known gap: routine addresses stored as numbers in data tables are not seen (documented).Stacked on #49's branch. Fixes #38
🤖 Generated with Claude Code