Re: [PATCH 2/2] sched/debug: add sched_update_nr_running tracepoint

[Alexei Starovoitov]

On Wed, Sep 4, 2019 at 8:33 AM Joel Fernandes <joel@xxxxxxxxxxxxxxxxx> wrote:
>
> On Wed, Sep 04, 2019 at 08:26:52AM -0700, Alexei Starovoitov wrote:
> > On Wed, Sep 4, 2019 at 6:14 AM Joel Fernandes <joel@xxxxxxxxxxxxxxxxx> wrote:
> > >
> > > True. However, for kprobes-based BPF program - it does check for kernel
> > > version to ensure that the BPF program is built against the right kernel
> > > version (in order to ensure the program is built against the right set of
> > > kernel headers). If it is not, then BPF refuses to load the program.
> >
> > This is not true anymore. Users found few ways to workaround that check
> > in practice. It became useless and it was deleted some time ago.
>
> Wow, Ok! Interesting!

the other part of your email says about kernel header requirement.
This is not true any more as well :)
BTF relocations are already supported by the kernel, llvm, libbpf,
bpftool, pahole.
We'll be posting sample code soon.