Add jampy target to workflows

[dakk]

Summary by CodeRabbit

• New Features

  • Added automated performance testing for jampy with scheduled and manual run options.

• Documentation

  • Updated README to include jampy in the performance table and added a performance badge/link.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 0ed6d60e-4afe-405b-8ac9-0a08ac2183e7

📥 Commits

Reviewing files that changed from the base of the PR and between b1ca7e9 and 5edc23c.

📒 Files selected for processing (1)

• .github/workflows/jampy-performance.yml

---

📝 Walkthrough

Walkthrough

Adds a new GitHub Actions workflow "Performance: jampy" (scheduled and manual) that reuses an existing reusable workflow; updates the Deploy Dashboard workflow to trigger after this workflow; README gains a jampy entry in the Minifuzz + Performance table.

Changes

Cohort / File(s)	Summary
Performance workflow ​.github/workflows/jampy-performance.yml	New workflow "Performance: jampy" added; scheduled (cron: '0 6 * * *') and workflow_dispatch triggers; job reuses .github/workflows/reusable-picofuzz.yml with parameters: target_name: jampy, docker_image: ghcr.io/dakk/jampy-target:0.7.2, docker_cmd: --socket-file {TARGET_SOCK}.
Deploy dashboard trigger ​.github/workflows/deploy-dashboard.yml	Dashboard workflow updated to include "Performance: jampy" in the workflow_run -> workflows list so it runs after the jampy performance workflow completes.
Documentation README.md	Added a new jampy entry/row in the Minifuzz + Performance table (performance badge/link).

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Poem

🐰 I hopped to the CI with a squeak and a grin,
I added a jampy run to fuzz all the things within.
Scheduled at dawn, and hand-run by choice,
Reusing a workflow to keep the same voice.
Hop, hop — performance badges now gleam!

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the main change: adding jampy target to workflows across multiple files including the new jampy-performance.yml workflow, deploy-dashboard.yml trigger, and README.md documentation.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

📝 Coding Plan

• Generate coding plan for human review comments

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[tomusdrw]

@dakk the image seems to be private / non-existent, can you check?

[dakk]

@dakk the image seems to be private / non-existent, can you check?

Fixed now

[tomusdrw]

@dakk

[jampy] [PYI-1:ERROR] Failed to load Python shared library '/opt/jampy-target-0.7.2_x86-64/_internal/libpython3.11.so.1.0': /lib/x86_64-linux-gnu/libm.so.6: version `GLIBC_2.38' not found (required by /opt/jampy-target-0.7.2_x86-64/_internal/libpython3.11.so.1.0)

https://github.com/FluffyLabs/jam-testing/actions/runs/23113614274/job/67135247558#step:5:33

[dakk]

I updated the docker image with a newer version of debian, and it seems to work now

[dakk]

@tomusdrw I don't get why at some point the ci hangs, and then timeout: https://github.com/FluffyLabs/jam-testing/actions/runs/23136052605/job/67200886358

for instance, in picofuzz - fallback it processed all the blocks and then hanged

[tomusdrw]

After the tests are finished we attempt to terminate the target first by sending SIGTERM and when that doesn't work we send SIGKILL. It seems to me that neither of this works for jampy after the 10x picofuzz runs.

Since it works fine for minifuzz, my guess is that the issue is casued by repeated runs. Each run closes and re-opens the socket. I'm running a test now to confirm my suspicion, see #10