Skip to content

Avoid copying packed scientific evidence before encoding - #97

Open
tpn wants to merge 1 commit into
mainfrom
codex/scientific-evidence-buffer-20261003
Open

tpn wants to merge 1 commit into
mainfrom
codex/scientific-evidence-buffer-20261003

Conversation

@tpn

@tpn tpn commented Oct 4, 2026 •

Copy link
Copy Markdown
Collaborator

Summary

Publishing xScan scientific evidence copies the complete packed float64 array solely to feed SHA-256 and Base64. Pass a view of the canonical NumPy buffer to both consumers, avoiding that temporary allocation and copy while preserving the encoded payload, checksum, counts and receipt format.

CPU regression tests compare the complete receipt for contiguous and strided read-only inputs, including signed zeros, infinities and distinct NaN bit patterns.

Validation

  • Focused artifact/device-pipeline tests pass; GPU cases skip without CUDA.
  • uv lock --check and all repository pre-commit checks pass.
  • make lint passes. make test-cpu: 2,789 passed, 210 skipped (GPU, optional backends and observational data); 40 expected Astropy FITS-keyword warnings.
  • Source distribution builds with the native extension disabled; expected manifest notices exclude absent local-only guidance files.

Repository hygiene

  • No credentials, private paths, local datasets, notebook outputs, or generated run artifacts were added.
  • No large binary artifacts were added.
  • Security-sensitive information is not included.
  • Receipt format and user-visible behavior are unchanged.
  • No dependencies were added.

Pass the canonical host buffer directly to SHA-256 and Base64.
Retain the exact receipt format while avoiding a full-size byte copy.

Signed-off-by: Trent Nelson <trentn@nvidia.com>
@tpn
tpn requested a review from melo-gonzo as a code owner October 4, 2026 06:58
@coderabbitai

coderabbitai Bot commented Oct 4, 2026

Copy link
Copy Markdown

Important

Review skipped

Auto reviews are limited based on label configuration.

🏷️ Required labels (at least one) (1)
  • ai-review

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration
  • Configuration used: Repository: NVIDIA/cuPhoton/.coderabbit.yaml
  • Review profile: CHILL
  • Plan: Enterprise
  • Run ID: 386c9561-ddb0-47c5-9da4-1cf071d1b40b

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • 🔍 Trigger review
  • Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Comment @coderabbitai help to get the list of available commands.

This branch has not been deployed

No deployments
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