Skip to content

scale-check fit fixes; big-run summary reads each check's output - #33

Merged
zmaril merged 1 commit into
mainfrom
mirth/run-fixes
Oct 10, 2026
Merged

zmaril merged 1 commit into
mainfrom
mirth/run-fixes

Conversation

@zmaril

@zmaril zmaril commented Oct 10, 2026

Copy link
Copy Markdown
Contributor

Fixes found by the big run (~/mirth-work/big-2026-10-10).

  • scale-check:
    • A shape capped below the size series now gets the series scaled to end at its cap. nested-generic (cap 120) used to get only 2 points and no fit; it now runs to N=120.
    • There is no time exponent when the largest compile takes under 0.5 s. nested-expr at O0 was flagging k_time = 1.67 on 0.1 s compiles.
  • big-run.sh: the summary reads each check's own output.
    • gate-check, abi-diff and scale-check don't write findings/, so they always showed "0 findings".
    • On today's run it now shows gate-check 8 problems, abi-diff seed 1 11,799 unlabelled differences (non-main targets; assembly-level triage is in progress separately) and scale-check 3 shapes over budget.
    • No Python.

🤖 Generated with Claude Code

https://claude.ai/code/session_01QXiEXbESemwqMLYKaWLDbT

…ested-generic had 2 points, no fit); no time exponent when the largest compile is under 0.5 s (nested-expr O0 flagged 1.67 on 0.1 s). big-run.sh: the summary reads each check's own output (gate-check, abi-diff, scale-check do not write findings/), without Python

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QXiEXbESemwqMLYKaWLDbT
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant