Source library
Memory region allocator — red team v5 (panel re-review of v6)
docs/ps1/memory-region-allocator-red-team-v5.md
This is the website shelf page for docs/ps1/memory-region-allocator-red-team-v5.md.
It exists because the project has hundreds of Markdown files and the website
should not pretend the interesting work only happened in the polished essays.
Why this file matters
Active PS1 reference. This is active engineering documentation for the PS1 port. It is close enough to the code that stale claims here become real bugs.
Same panel reconvenes. v6 closed 18 of 18 v4 findings. This pass looks for new issues introduced by the v6 fixes. Loop 3 of the goal-driven iteration. Convergence approaching.
File facts
| Field | Value |
|---|---|
| Source path | docs/ps1/memory-region-allocator-red-team-v5.md |
| Lines | 340 |
| Approx. words | 1819 |
| Code fences | 4 |
Outline
- Memory region allocator — red team v5 (panel re-review of v6)
- Reviewer 1 — Pat, Embedded Systems Veteran
- Closed in v6
- New concerns
- 🟠 P17. ps1IsMainContext() itself doesn’t exist yet
- 🟠 P18. PsyQ toolchain may not support -Wglobal-constructors
- 🟡 P19. builtinunreachable() after while(1) is redundant
- Reviewer 2 — Sarah, SRE / Operations
- Closed in v6
- New concerns
- 🟠 S14. MEMREGIONRATIONALE grep gate has a false-positive failure mode
- 🟠 S15. Pre-Phase-1 gates can’t all be closed before Phase 1
- 🟡 S16. Phase 1 checklist drifts from plan
- Reviewer 3 — Mateo, Future Maintainer