Revisit the Go Std Library Install Step in kubecross Dockerfile

[jeremyrickard]

Currently in the kubecross Dockerfile, we precompile the standard library for a number of platforms:

ENV KUBE_CROSSPLATFORMS \
  linux/386 \
  linux/arm linux/arm64 \
  linux/ppc64le \
  linux/s390x \
  darwin/amd64 darwin/386 \
  windows/amd64 windows/386

We only do this when the targetArch is amd64:

RUN targetArch=$(echo $TARGETPLATFORM | cut -f2 -d '/') \
    && if [ ${targetArch} = "amd64" ]; then \
    for platform in ${KUBE_CROSSPLATFORMS}; do GOOS=${platform%/*} GOARCH=${platform##*/} go install std; done \
    && go clean -cache; \
fi

This was likely done as a way to improve efficiency with amd64 build hosts, but there are a number of arm64 platforms now that we should consider. We should also validate if this is actually useful, or if we end up recompiling the std lib anyway due to the different build flags and modes (cgo vs non-cgo) in use. If it's useful, we should consider the list of platforms and if the current conditional to only do this on amd64 is sufficient.

Removing this current step could reduce the overall size of the kubecross images and would be positive, if we don't need to do it.

/kind cleanup
/sig release
/area release-eng

[k8s-ci-robot]

@jeremyrickard: The label(s) area/release-end cannot be applied, because the repository doesn't have them.

In response to this:

Currently in the kubecross Dockerfile, we precompile the standard library for a number of platforms:

ENV KUBE_CROSSPLATFORMS \
 linux/386 \
 linux/arm linux/arm64 \
 linux/ppc64le \
 linux/s390x \
 darwin/amd64 darwin/386 \
 windows/amd64 windows/386

We only do this when the targetArch is amd64:

RUN targetArch=$(echo $TARGETPLATFORM | cut -f2 -d '/') \
   && if [ ${targetArch} = "amd64" ]; then \
   for platform in ${KUBE_CROSSPLATFORMS}; do GOOS=${platform%/*} GOARCH=${platform##*/} go install std; done \
   && go clean -cache; \
fi

This was likely done as a way to improve efficiency with amd64 build hosts, but there are a number of arm64 platforms now that we should consider. We should also validate if this is actually useful, or if we end up recompiling the std lib anyway due to the different build flags and modes (cgo vs non-cgo) in use.

Removing this current step could reduce the overall size of the kubecross images and would be positive, if we don't need to do it.

/kind cleanup
/sig release
/area release-end

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[jeremyrickard]

/cc @BenTheElder

[jeremyrickard]

/area release-eng

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[BenTheElder]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[BenTheElder]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[vaibhav2107]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[xmudrii]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[BenTheElder]

/lifecycle frozen

not high prio but I don't want to just ignore this, we're probably complicating and slowing down the builds of these images without improving the performance of actually building releases, but we need to confirm this and then either close the issue because it's wrong or drop this and win on simpler smaller images

I can probably say I can look by like EOY but not tighter than that, workload fluctuates a lot but I'd realllly like to see us get some wins simplifying the build now that we only have one to maintain