eiffel-fl

Thank you for the merge!

The [kernel bug fix](https://lkml.org/lkml/2022/7/1/889) should land in upstream 5.20.

The bug fix is being backported to all stable kernels (_e.g._ [5.15](https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-5.15.y&id=42eede3ae05bbf32cb0d87940b466ec5a76aca3f)).

> I'm back from the virus... Oh damn... I wish you a quick and good recovery!

> It looks like the libbpf profile should work the same way as profile.py. A middle ground would be to add this as an option, _.e.g._ `--sort-function-by-cpu-time`, in another PR...

> it appears all existing libbpf-tools tool are not name space aware, right. It is indeed the case but I am not sure if the goals of this tools are...

> I guess the changes you made can be directly copied/pasted/modified to existing libbpf-tools and it should work, right? I think you will need to add the `mntns` map handling...

@eiffel-fl (Just so I can remember of reviewing it.)

> That may be the reason, I guess > > https://discourse.ubuntu.com/t/unprivileged-ebpf-disabled-by-default-for-ubuntu-20-04-lts-18-04-lts-16-04-esm/27047 I am not 100% sure as if this problem occurs it would mean all the CO-RE gadgets would not...

> > I suspect more something about a BPF iterator we use in this gadget which is not available for AKS 5.4 kernel, a problem like the one describe in...