aarch64: CONFIG_BPF_SYSCALL=y is missing

[kakkoyun]

What Happened?

I'm trying to run an eBPF application on Minikube on a Mac using QEMU (--drive=qemu2).

The first thing I realized when I tried to mount /sys/fs/bpf, it wasn't there.

Then when I tried to load an eBPF program, I got the following error.

libbpf: Error in bpf_object__probe_loading():Function not implemented(38). 
Couldn't load trivial BPF program. 
Make sure your kernel supports BPF (CONFIG_BPF_SYSCALL=y) and/or that RLIMIT_MEMLOCK is set to big enough value.
libbpf: failed to load object 'XXX'

This is what the config looks like:

$ cat /proc/config.gz | gunzip | grep CONFIG_BPF
CONFIG_BPF=y
# CONFIG_BPF_SYSCALL is not set
CONFIG_BPF_JIT_DEFAULT_ON=y
# CONFIG_BPFILTER is not set
CONFIG_BPF_JIT=y

FWIW, the same program flawlessly runs on minikube on a linux/amd64 platform.
I had assumed with #14316, this should work regardless of the driver.

I might be missing something trivial. I'd be glad if you could nudge me in the right direction.

cc @eiffel-fl

Attach the log file

log.txt

Operating System

macOS (Default)

Driver

N/A

[afbjorklund]

FWIW, the same program flawlessly runs on minikube on a linux/amd64 platform.

But that kernel config has CONFIG_BPF_SYSCALL=y set, where the linux/arm64 does not

[kakkoyun]

FWIW, the same program flawlessly runs on minikube on a linux/amd64 platform.

But that kernel config has CONFIG_BPF_SYSCALL=y set, where the linux/arm64 does not

Now that I checked. Yes, it is defined for x86_64

minikube/deploy/iso/minikube-iso/board/minikube/x86_64/linux_x86_64_defconfig

Line 44 in 632839e

CONFIG_BPF_SYSCALL=y

But it is not defined for aarch64
https://github.com/kubernetes/minikube/blob/master/deploy/iso/minikube-iso/board/minikube/aarch64/linux_aarch64_defconfig

I think it's overlooked. If you can confirm, I'm happy to contribute.

[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

[eiffel-fl]

/remove-lifecycle stale

I will take a look it but I just need to find a bit of time!

[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

[eiffel-fl]

No, this is not stale, I still want to take a look at it but I cannot really know how to debug this and I cannot create time.

[k8s-triage-robot]

The Kubernetes project currently lacks enough active 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 rotten
• Close this issue with /close
• Offer to help out with Issue Triage

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

/lifecycle rotten

[eiffel-fl]

/remove-lifecycle rotten

[carlosonunez]

Not stale. This is preventing Calico from working on "qemu2" minikube nodes (which you'll need to use if you want VM-based nodes on Apple Silicon).