From 9eefa64a501f78a93b0002b959168c31716e8023 Mon Sep 17 00:00:00 2001
From: kaikisegfault <133794518+kaikisegfault@users.noreply.github.com>
Date: Sun, 16 Aug 2026 16:45:15 +0200
Subject: [PATCH] docs(project): record the M3.10b merge facts and the matrix
 margin

The handoff named the run and the margin as "recorded below" before either
existed, because the slice's own commits could not carry the result of the run
that verified them. Both are now exact: the candidate at `66fcab8` and the
post-merge run on `cc2e8fc`, with per-job durations on each.

The margin is about ten and a half minutes and the slice moved it very little,
which is the expected result for a change that adds no translation unit. The
figure is recorded so M3.10c has a baseline, because that slice is C++ and build
time is the larger half of each job.
---
 docs/project/current-state.md | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/docs/project/current-state.md b/docs/project/current-state.md
index d1d90ce..5b9d59d 100644
--- a/docs/project/current-state.md
+++ b/docs/project/current-state.md
@@ -1838,11 +1838,21 @@ behavior.
 
 - Repository: `kaikisegfault/protocol-stack`.
 - Issue #153 and PR #173 are the M3.10b delivery, merged by rebase across
-  commits `bcad56e` through `089e9b4` on `main`. It adds the version-six
+  commits `19107df` through `cc2e8fc` on `main`. It adds the version-six
   execution model, 512 vectors, a verifier, ADR 0045, and two test modules, and
-  edits no accepted artifact. The complete hosted matrix passed on the exact
-  candidate and again post-merge on `main`; per-job durations and the resulting
-  margin are recorded below.
+  edits no accepted artifact. PR run 31952597793 on the final head `66fcab8`
+  passed the complete hosted matrix — scope classification `full`, GCC and Clang
+  debug, both sanitizers, and the aggregate required check — and post-merge run
+  31953123699 on `cc2e8fc` passed the same matrix.
+- **The margin is about ten and a half minutes, and the slice moved it very
+  little.** Per-job durations on the candidate against the 20-minute per-job
+  timeout: `gcc-debug` 5m48s, `clang-sanitizers` 7m21s, `clang-debug` 8m45s,
+  `gcc-sanitizers` 9m21s. Post-merge on `main`: `clang-debug` 8m15s, `gcc-debug`
+  8m17s, `clang-sanitizers` 8m50s, `gcc-sanitizers` 9m20s. The slice adds no
+  translation unit and two fast Python entries that `ctest --parallel` absorbs,
+  so the spread sits inside the roughly 12% run-to-run variance M3.7a measured
+  on identical code. **M3.10c is the one to watch**: it is C++, and build time is
+  the larger half of each job.
 - Issue #153's scope has been rebound twice. It was opened as M3.9b against
   version four, renumbered M3.9e and rebound to version five when version four's
   kind-11 defect pushed three slices in front of it, and finally rebound to