Skip to content

Use XRPLF heavy runners#57

Open
Patel-Raj11 wants to merge 5 commits into
XRPLF:mainfrom
Patel-Raj11:build-shared-lib-improvements
Open

Use XRPLF heavy runners#57
Patel-Raj11 wants to merge 5 commits into
XRPLF:mainfrom
Patel-Raj11:build-shared-lib-improvements

Conversation

@Patel-Raj11

@Patel-Raj11 Patel-Raj11 commented Apr 27, 2026

Copy link
Copy Markdown
Contributor

Changes

  • Use heavy / heavy-arm64 for Linux builds (inside the clio-ci container, matching build-and-test)
  • Use macos15 for macOS arm64 builds
  • Use windows runner for Windows builds; remove explicit MSVC environment setup (ilammy/msvc-dev-cmd)
  • Replace setup-python + pip install with XRPLF/actions/prepare-runner
  • Add Cleanup workspace step for the persistent macos15 runner
  • Skip native-binaries on forks (matches build-and-test's pattern)

Still on GHA-hosted runners

  • macOS Intel (macos-15-intel) — no XRPLF Intel runner available
  • Linux s390x (ubuntu-latest) — QEMU setup needs privileged binfmt registration, which heavy doesn't allow.

container: '{ "image": "ghcr.io/xrplf/clio-ci:14342e087ceb8b593027198bf9ef06a43833c696" }'

# ── Linux s390x (IBM Z) — built via QEMU user-mode emulation ──
- os: ubuntu-latest

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We can and should still use heavy here

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@mathbunnyru I tried using heavy runners for s390x, but it keeps failing with the error seen in below mentioned run. After some back and forth with Claude, it seems that heavy runners are running in some sandbox/VM/namespace such that docker/setup-qemu-action is not working as intended.

https://github.com/XRPLF/mpt-crypto/actions/runs/24976270406/job/73128730834

Can you see if you can suggest something based on the error message? Or s390x is quite unique (it needs docker emulation) so can we keep using GHA runners?

Comment thread .github/workflows/build-shared-libs.yml
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.

2 participants