Re: [PATCH bpf-next v1 3/5] bpf: Introduce cgroup iter

[Hao Luo]

On Mon, May 23, 2022 at 4:58 PM Andrii Nakryiko
<andrii.nakryiko@xxxxxxxxx> wrote:
>
> On Fri, May 20, 2022 at 7:35 PM Hao Luo <haoluo@xxxxxxxxxx> wrote:
> >
> > On Fri, May 20, 2022 at 5:59 PM Yonghong Song <yhs@xxxxxx> wrote:
> > > On 5/20/22 3:57 PM, Tejun Heo wrote:
> > > > Hello,
> > > >
> > > > On Fri, May 20, 2022 at 03:19:19PM -0700, Alexei Starovoitov wrote:
> > > >> We have bpf_map iterator that walks all bpf maps.
> > > >> When map iterator is parametrized with map_fd the iterator walks
> > > >> all elements of that map.
> > > >> cgroup iterator should have similar semantics.
> > > >> When non-parameterized it will walk all cgroups and their descendent
> > > >> depth first way. I believe that's what Yonghong is proposing.
> > > >> When parametrized it will start from that particular cgroup and
> > > >> walk all descendant of that cgroup only.
> > > >> The bpf prog can stop the iteration right away with ret 1.
> > > >> Maybe we can add two parameters. One -> cgroup_fd to use and another ->
> > > >> the order of iteration css_for_each_descendant_pre vs _post.
> > > >> wdyt?
> > > >
> > > > Sounds perfectly reasonable to me.
> > >
> > > This works for me too. Thanks!
> > >
> >
> > This sounds good to me. Thanks. Let's try to do it in the next iteration.
>
> Can we, in addition to descendant_pre and descendant_post walk
> algorithms also add the one that does ascendants walk (i.e., start
> from specified cgroup and walk up to the root cgroup)? I don't have
> specific example, but it seems natural to include it for "cgroup
> iterator" in general. Hopefully it won't add much code to the
> implementation.

Yep. Sounds reasonable and doable. It's just adding a flag to specify
traversal order, like:

{
  WALK_DESCENDANT_PRE,
  WALK_DESCENDANT_POST,
  WALK_PARENT_UP,
};

In bpf_iter's seq_next(), change the algorithm to yield the parent of
the current cgroup.