Repository navigation
adopt meld 0.65.1; linmem-capture derives its offsets (AFD-143) - #304
Merged
Merged
Conversation
…dcode (AFD-143) meld 0.62.0 -> 0.65.1, attributed in artifacts.pins: 0.64's 7-byte pack-rebase guard (meld#479, SR-102) adds 16 B per packed provider and moves the ekf return area 0xb024 -> 0xb064; 0.64 -> 0.65.1 is a no-op on the lowered object apart from attestation. On-target behaviour unchanged: QEMU ARM fold 6ec49e30 == wasmtime 6ec49e30, the 14-field return area identical. The prerequisite was linmem-capture.sh, which had broken twice without going red: a hardcoded export name (KeyError after AFD-141) and a hardcoded return-area offset (the fixed-offset precondition kept passing after the layout moved: |q| = 0.51 at the old 0xac18). It now derives the module, the export and the return area, requires |q| = 1, and has a --check-inputs mode (no QEMU) that CI runs in on-target-oracles. The 10 DIVERGENT words at 0xe000 are the harness's staged input vehicle-state, written only by the ARM boot code; byte-identical on the 0.62 and 0.65.1 builds, so pre-existing. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017EKh2onUHpWSUiAEqWxfy2
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Change: meld 0.62.0 → 0.65.1 (
artifacts.pins+ci.yml MELD_VERSION). Every byte is attributed in the pins comment:On-target behaviour is unchanged. QEMU ARM fold 6ec49e30 == wasmtime 6ec49e30, and the 14-field return area is identical.
Prerequisite fixed (AFD-143).
linmem-capture.shbroke twice without any check going red:It now derives the module, the export and the return area, and it requires |q| = 1. A new
--check-inputsmode needs no QEMU and runs in CI (on-target-oracles).Controls:
Classified, not introduced: the whole-memory diff shows 10 DIVERGENT words at 0xe000–0xe044. They are the harness's staged input vehicle-state (matches
cascade_ref.pyVEHICLE_STATE), written only by the ARM boot code. They are byte-identical on the 0.62 and 0.65.1 builds, so they predate this bump.rivet validate(0.35.0):Result: PASS.🤖 Generated with Claude Code
https://claude.ai/code/session_017EKh2onUHpWSUiAEqWxfy2