Skip to content

docs: lock accepted P0 MMS runtime checkpoint after #185 and #187 #188

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
masarray merged 1 commit into from
Oct 16, 2026
Merged

docs: lock accepted P0 MMS runtime checkpoint after #185 and #187 #188

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 9 additions & 0 deletions docs/WORKBENCH_CURRENT_HANDOFF.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,15 @@
> **Primary strategy and ownership contract:** [WORKBENCH_ARSAS_GUI_SSOT_STRATEGY.md](WORKBENCH_ARSAS_GUI_SSOT_STRATEGY.md).
> **Scope:** installed native ARStack Workbench in apps/ied_simulator. Not a web app, WPF clone, or the separate ESP32-P4/SMV Studio track.

## Latest accepted P0 checkpoint — 2026-10-16 (WIB)

- **Main HEAD after integration:** `aae37327d8e9c40314f4a86b6127a12d38248420` (merge of #187). This supersedes the older baseline in the historical section below; always re-query GitHub before further writes.
- **[PR #185](https://github.com/masarray/arstack61850/pull/185) MERGED** at `8fba05d19e02d94af333fae02313f7430a6696e1`, exact PR head `3903d88623a4dfe66320967d0da8bd8c512dc7e7`, **12/12 exact-head workflow SUCCESS** and no unresolved reviews. Native Workbench exposes immutable measured GetNameList/GVAA/DataSetDirectory/RCB Read confirmed-request counts on completed live discovery. No additional wire traffic, GUI polling, or protocol concurrency. Post-merge CI should be verified separately at the merge SHA; exact-PR evidence is not an installed-package or field IED test.
- **[PR #187](https://github.com/masarray/arstack61850/pull/187) MERGED** at `aae37327d8e9c40314f4a86b6127a12d38248420`, exact PR head `882dd307a2452ca3c0d62f9d3779e8588a5dc7db`, **13/13 exact-head workflow SUCCESS** (GCC, Clang, MSVC, native Qt, interoperability, security, naming guard), zero unresolved review threads. Fixes a confirmed-response integrity hazard: an out-of-order/unexpected invoke ID or an idle-orphan cached response must not acknowledge a future, unsent request. The serial runtime faults and closes the association instead of replaying stale evidence. Added two targeted MMS runtime negative tests; all 48 CTest cases passed on Linux GCC and Clang. No Write/Control/GI/RCB enable, extra Read, new polling, or naive concurrency.
- **Transport ownership audit:** `exchange_confirmed`, `poll_once`, and `try_poll_once_for` call a synchronous association-owned receive primitive. The thread-safe invoke queue is **not** a thread-safe transport dispatcher; no safe simultaneous confirmed exchange or receive route exists. The negotiated `maximum_outstanding_calling` is decoded and retained but not yet enforced as an in-flight window. InformationReport routing remains separately bounded; cancellation, timeout, reconnect, and per-exchange send receipt must retain their established semantics. Detailed source audit is recorded in [PR #187 discussion](https://github.com/masarray/arstack61850/pull/187).
- **Next P0 engineering slice:** Only after an explicit association-owned send/receive dispatcher, invoke lifecycle registration, bounded negotiated concurrency, frame-write serialization, cancellation/reconnect fences and deterministic model parity tests exist, consider opt-in concurrent GetNameList chains. Prefer a measured non-concurrent optimization or additional negative regression first. Compare exact LD/LN/DO/DA, ordered FCDA, RCB, type coverage, wire-budget, failure behavior, CPU/latency and time-to-first-model on the same device.
- **Acceptance still open:** post-merge exact main check completion, installed Windows Workbench verification, real relay/IED trials, and hardware ESP32-P4/SV/PTP qualification. Native Workbench `apps/ied_simulator` and separate Studio `apps/arstack_studio` remain distinct products. No physical interop or speedup claim has been established.

## P0 MMS Discovery interoperability and failure containment — 2026-10-16

- **Integrated baseline**: main `6cc64b384b643fe85b71b9e2082daa1e805766c8`. Source remains the portable IEC 61850 C++ engine under `src/mms/` and `include/ariec61850/mms/`, consumed by the native Qt Workbench; the ESP32-P4 SV Studio is a separate product.
Expand Down
Loading