ci: build multi-arch image on native runners (no QEMU)#60
Merged
Conversation
The release image build was the slow part of the release pipeline: it built linux/arm64 under QEMU emulation, and emulating a Rust release compile is extremely slow (the native per-arch binary builds finish in ~1-2 min, while the emulated docker arm64 layer can take many times longer). Rework the single `docker` job into the canonical native-runner pattern: - `docker-build`: a matrix that builds each architecture on its own native runner (linux/amd64 on ubuntu-latest, linux/arm64 on ubuntu-24.04-arm), pushing each by digest. No QEMU, so both archs compile at native speed. - `docker-merge`: assembles the multi-arch manifest list (semver + latest tags) from the per-arch digests with `docker buildx imagetools create`. GHA build cache is scoped per platform to avoid cross-arch collisions. The now-unused QEMU setup step is dropped. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Speeds up the slowest part of the release pipeline. The release image previously built
linux/arm64under QEMU emulation, and emulating a Rust release compile is extremely slow. (For comparison, the native per-arch binary builds finish in ~1–2 min.)Change
Rework the single
dockerjob into the canonical Docker native-runner pattern:docker-build— matrix building each arch on its own native runner (linux/amd64→ubuntu-latest,linux/arm64→ubuntu-24.04-arm, free for public repos), pushing by digest. No QEMU → both archs compile at native speed.docker-merge— assembles the multi-arch manifest (semver +latest) from the per-arch digests viadocker buildx imagetools create.GHA build cache is scoped per platform; the unused QEMU setup step is removed. All actions remain SHA-pinned.
Notes
ci:change → it does not trigger a release.docker-buildcheck). The new part is the digest/merge orchestration (standard Docker docs pattern); it gets its first real exercise on the next release. A failure there would only mean "no image for that release" (recoverable) — the release itself is created by release-please independently.🤖 Generated with Claude Code