Skip to content

docs(analog): mos_psp103 and mos_table gate notes say what was measured (Closes #12518) - #12519

Merged
gHashTag merged 3 commits into
masterfrom
lane/docs-stale-gate-12518
Oct 25, 2026
Merged

docs(analog): mos_psp103 and mos_table gate notes say what was measured (Closes #12518)#12519
gHashTag merged 3 commits into
masterfrom
lane/docs-stale-gate-12518

Conversation

@gHashTag

@gHashTag gHashTag commented Oct 25, 2026

Copy link
Copy Markdown
Collaborator

Summary

  • mutation-accepted-mos_psp103.txt: the header no longer says the gate was not run; it quotes the measured gate result of the note (359 mutants, 306 killed, 30 survived, 18 hung, 5 unviable; 48 accepted, 0 stale).
  • mutation-accepted-mos_table.txt: "predicted, not measured" is replaced by the measured gate result (632 mutants, 583 killed, 25 survived, 20 hung, 4 unviable; 45 accepted) next to the census.
  • mutation-mos_table-2026-10-25.md, section "Not done": made to agree with its own section "Gate result (measured)"; the seal bullet was checked against the tree (seal re-saved in PR 12401, squash commit 96ce237, top-level spec_hash equals the sha256 of the spec, 0 vacuous), and a coverage claim that rested on a skimmed log was dropped.
  • Diff: three files, comment lines and one note section; the second command of FR-004 (any changed non-comment line in the two accepted files) prints nothing. ASCII, no home path, no code.

Closes #12518
Refs #10486

🤖 Generated with Claude Code

gHashTag and others added 3 commits October 26, 2026 01:12
@gHashTag @claude
docs(analog): mos_psp103 and mos_table headers and note agree with th…
8195a29 
…e measured gate (Closes #12518)

Prose only. The comment headers of the two accepted files and the "Not
done" section of the mos_table note still said the mutation gate had not
been run, while the notes record a measured gate run with exit 0.

- mutation-accepted-mos_psp103.txt: the "gate NOT run" line is replaced by
  the measured gate result copied from its note (359 mutants, 306 killed,
  30 survived, 18 hung, 5 unviable; 48 not killed, 48 accepted, 0 accepted
  lines now killed) and a pointer to the note; the census line is labelled
  as the census.
- mutation-accepted-mos_table.txt: "Predicted ... not measured" becomes the
  measured gate result (632, 583, 25, 20, 4; 45 not killed, all accepted),
  next to the census (632, 493, 115, 20, 4).
- mutation-mos_table-2026-10-25.md, "Not done": the gate bullet is turned
  into "ran and passed"; the other bullets were checked, see below.

Verified before keeping or dropping each "Not done" bullet:
- Seal: .trinity/seals/analog_AnalogMosTable.json was re-saved in the merged
  lane PR #12401 after the batch 3 tests; on master it records 37 of 37
  tests, 0 vacuous, and its spec_hash equals the sha256 of
  specs/analog/mos_table.t27 (shasum). The "no seal was saved" bullet is
  corrected.
- Coverage and docs check: the lane did not run them (kept as stated); the
  CI coverage job of #12401 reported a failure, cause not examined, and the
  bullet says so without blaming this spec.
- Group K and line 835: the "Gate result (measured)" section lists none of
  their lines among the not-killed mutants, so the "gate remains" wording is
  dropped.
- The 20 hung: the gate reports 20 hung and all 45 not-killed mutants are
  accepted; the bullet no longer says they were not re-run.

No non-comment line of either accepted file changed (checked with git diff
-U0 and a grep for non-comment +/- lines: empty); both files keep their line
counts. ASCII only, no home path. No spec, seal, test or workflow touched.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@gHashTag @claude
docs(analog): mos_table note names the squash commit of the seal, dro…
0b90a49 
…ps a coverage claim it did not examine (Refs #12518)

The "Not done" bullet quoted a commit title that is not in master history;
it now names PR 12401 and its squash commit 96ce237 (the seal file log
shows that commit). The bullet that blamed a CI coverage job of that PR
for a failure rested on a skimmed log, so it now says only that coverage
and the docs check were not run in this lane.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@gHashTag
Merge remote-tracking branch 'origin/master' into lane/docs-stale-gat…
412bfaa 
…e-12518
@github-actions

github-actions Bot commented Oct 25, 2026

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-25 18:16:37 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 47
PRs with All Checks Green 3
READY 2
FAILING 47
PENDING 0
NO CHECKS YET 0

These columns do not partition: 2 + 47 + 0 + 0 = 49, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=75249fe1efad != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@github-actions github-actions Bot mentioned this pull request Oct 25, 2026
Merged
@tri-claw

tri-claw Bot commented Oct 25, 2026

Copy link
Copy Markdown

Summary by t27-bees

This PR updates three analog mutation documentation files to replace predicted gate results with actual measured results from running the mutation gate. The headers now quote measured gate counts (killed/survived/hung/unviable, accepted/not-killed) and reference the corresponding gate result sections in the dated markdown files.

Worth a look

  • docs/analog/mutation-accepted-mos_psp103.txt:3–5 – The new gate result line states "359 mutants, 306 killed, 30 survived, 18 hung, 5 unviable; 48 not killed (30 survived + 18 hung), 48 accepted, 0 accepted lines now killed." If the gate run referenced (master-built t27c, --jobs 2 --timeout 45) was not actually executed with this exact accepted file, the numbers are fabricated.
  • docs/analog/mutation-accepted-mos_table.txt:3–4 – The gate result claims "632 mutants, 583 killed, 25 survived, 20 hung, 4 unviable; 45 not killed, all 45 accepted, 0 accepted lines now killed." Same risk: if the gate command (tri mutate spec with this file) was not run on the tree with tests T1–T16, the counts are unsupported.
  • docs/analog/mutation-mos_table-2026-10-25.md:229–237 – The "Not done" section now asserts the seal was saved in PR docs(analog): mos_table mutation gate, 583/632 killed, 45 accepted (Closes #12317) #12401 (squash commit 96ce237) and that spec_hash equals the sha256 of specs/analog/mos_table.t27. If that seal file (.trinity/seals/analog_AnalogMosTable.json) does not exist on master or its spec_hash does not match, the claim is false.
  • docs/analog/mutation-mos_table-2026-10-25.md:236–237 – States "The gate ran the 20 hung mutants again and reports 20 hung, the same count as the census; all 45 not-killed mutants of the gate are accepted by the file. The hung lines were not compared one by one with the census list here." If the gate's hung lines differ from the census's 20 lines, the accepted file could be missing entries or have stale ones.
  • docs/analog/mutation-accepted-mos_psp103.txt:5 – References "see the 'Gate' item at the top of docs/analog/mutation-mos_psp103-2026-10-25.md". If that file lacks a "Gate" section with matching numbers, the cross-reference is broken.

Read: 3 files, +16 -13, the whole diff. Head 412bfaa. Model: openai-compatible/nvidia/nemotron-3-ultra-550b-a55b. Commands: @t27-bees review | summary | help | pause | resume. Free for every repository.

@gHashTag

gHashTag commented Oct 25, 2026

Copy link
Copy Markdown
Collaborator Author

Merge basis (hand-merge under the rule of 2026-10-25 16:50, owner veto default from 2026-10-26 12:00 +07), head 412bfaa (master dc98d95 is already in it):

  • The four required checks of ruleset t27-master-protection are green on this head: validate, check-linked-issue, parse-ratchet, own-language. cli-tri, Documented t27c subcommands exist, gitleaks, scan, Verilog widths, Check L1 TRACEABILITY and the other 20 non-claw checks are green too.
  • Diff: two accepted-file headers (comment lines only; no non-comment line of either file changed) and one section of the mos_table note. No spec, seal, test, Rust or workflow file.
  • One red was read in its log and is inherited, it does not come from this diff: coverage, the step "Every seal still describes its spec": five [stale] t27os seals (t27os_T27osDoctor, GlassWindow, LabPublish, PtyServer, ReceiptName), the ones tracked in ci: five t27os seals added by #12373 are stale on master (coverage and spec-guards red) #12509.
  • Not complete at merge time: Corpus ratchet (expected-failure ledger) (known red: the [parse-no-discard] failures, fix PR 12500); a comment-and-prose diff cannot move it. The claw-* jobs are skipped.
  • Effect: closes docs(analog): accepted files and a note say the mutation gate was not run, the notes record it passed #12518.

🤖 Addressed by Claude Code

@gHashTag
gHashTag merged commit a1930fe into master Oct 25, 2026
29 of 31 checks passed
@gHashTag
gHashTag deleted the lane/docs-stale-gate-12518 branch October 26, 2026 03:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

docs(analog): accepted files and a note say the mutation gate was not run, the notes record it passed

1 participant

@gHashTag