fix: sync reproducibility artifacts for v3.2.1 registry#117
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: d38f82a538
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| "parameter_registry": { | ||
| "path": "src/factorforge/registry/current_parameter_registry.yaml", | ||
| "sha256": "da95907eced327f90eb463e359f87f316bf2bffbb29b022ee3656b79905e40e2" | ||
| "sha256": "849cbdc8f60b71543dd95b9018ecfc02f16265797549ceaa0800ddf8c029f37f" |
There was a problem hiding this comment.
Pin the v3.2.0 benchmark manifest to its registry
When users verify or reproduce this benchmark_v0.5.1 evidence pack, the same manifest still declares factorforge_version/archive version 3.2.0 and pins git_commit to 6409108921d538396b67c7edd3434b43e547f666; the registry file at that commit hashes to the previous da959... value, not this new current-3.2.1 hash. Updating only this SHA makes the manifest describe parameters that were not present in the pinned software release, so verification against the recorded commit fails or misattributes the benchmark run; keep this entry tied to the recorded release/commit, or archive a separate registry file if the manifest should hash current bytes.
Useful? React with 👍 / 👎.
Summary
Verification
Local result: 589 passed, 1 skipped.