Gatchan/FF6Portage
Watch
1
0
Fork
You've already forked FF6Portage
 0
Code Issues 16 Pull requests Projects Releases Packages Wiki Activity Actions

Frame model: per-port coroutine for routines that wait for frames #37

New issue
Closed
opened 2026-11-02 17:38:40 +00:00 by Gatchan · 1 comment
Gatchan commented 2026-11-02 17:38:40 +00:00
Owner

Every heavy field/battle/menu routine waits for NMI or spans frames (docs/coverage.md, timing-census.md). Brief §5.1 and workdocs/execution-model.md describe the model (libco coroutines, wait_frame() as a scheduling point) but core/ has no co_create yet.

Scope: design and implement the frame model for ported code (a ported routine that waits for a frame yields to the host; the interpreter and the port interleave correctly), with one real port that waits for a frame as the proof, validated in shadow US+JP. Needed before FieldMain/FieldNMI and any multi-frame loader.

From the 2026-11-02 plan review.

Every heavy field/battle/menu routine waits for NMI or spans frames (docs/coverage.md, timing-census.md). Brief §5.1 and workdocs/execution-model.md describe the model (libco coroutines, `wait_frame()` as a scheduling point) but `core/` has no `co_create` yet. Scope: design and implement the frame model for ported code (a ported routine that waits for a frame yields to the host; the interpreter and the port interleave correctly), with one real port that waits for a frame as the proof, validated in shadow US+JP. Needed before `FieldMain`/`FieldNMI` and any multi-frame loader. From the 2026-11-02 plan review.
Gatchan commented 2026-11-06 04:48:05 +00:00
Author
Owner

Merged in PR #65 (355a2d9). Frame model: tm_wait_frame + per-port libco coroutines; shadow compares a waiting call at its return (clone follows the original cycle by cycle). Ports: field/WaitVblank, btlgfx/WaitVblank. Shadow, chapter 1 routes US+JP once each: btlgfx 39108/39108 US, 37025/37025 JP; field 46586/46589 US, 42888/42891 JP (last call per route pending at movie end), 0 mismatch, 0 fallback. portcheck all PASS (136). debug ctest -LE slow 27/27. Side bug #64 (LakeSnes FastROM table after state load) fixed in core/hw.

Merged in PR #65 (355a2d9). Frame model: tm_wait_frame + per-port libco coroutines; shadow compares a waiting call at its return (clone follows the original cycle by cycle). Ports: field/WaitVblank, btlgfx/WaitVblank. Shadow, chapter 1 routes US+JP once each: btlgfx 39108/39108 US, 37025/37025 JP; field 46586/46589 US, 42888/42891 JP (last call per route pending at movie end), 0 mismatch, 0 fallback. portcheck all PASS (136). debug ctest -LE slow 27/27. Side bug #64 (LakeSnes FastROM table after state load) fixed in core/hw.
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
No results found.
Labels
Clear labels   
bug
Something produces a wrong result

  
jp
Japanese version specific

  
perf
Speed or resource use

  
port
Hand port of original routines

  
resources
Resource pipeline and pack

  
task
Planned work item

  
tools
Python tools, extract/pack, generators

  
upstream
Defect in a reference or third-party component (disassembly, LakeSnes)

  
validation
Oracles, shadow, tests, routes

No labels
bug
jp
perf
port
resources
task
tools
upstream
validation
Milestone
Clear milestone
No items
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
Gatchan/FF6Portage#37
No description provided.