Re: [PATCH bpf-next 0/3] Introduce pinnable bpf_link kernel abstraction

From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
To: Jakub Kicinski <kuba@kernel.org>
Cc: "Toke Høiland-Jørgensen" <toke@redhat.com>,
	"Alexei Starovoitov" <ast@fb.com>,
	"Daniel Borkmann" <daniel@iogearbox.net>,
	"Andrii Nakryiko" <andrii.nakryiko@gmail.com>,
	"Andrii Nakryiko" <andriin@fb.com>, bpf <bpf@vger.kernel.org>,
	Networking <netdev@vger.kernel.org>,
	"Kernel Team" <kernel-team@fb.com>
Subject: Re: [PATCH bpf-next 0/3] Introduce pinnable bpf_link kernel abstraction
Date: Thu, 5 Mar 2020 08:39:07 -0800	[thread overview]
Message-ID: <20200305163905.mumvxqmdotavd6t7@ast-mbp> (raw)
In-Reply-To: <20200305001620.204c292e@cakuba.hsd1.ca.comcast.net>

On Thu, Mar 05, 2020 at 12:16:20AM -0800, Jakub Kicinski wrote:
> > 
> > bpf program is an object. That object has an owner or multiple owners.
> > A user process that holds a pointer to that object is a shared owner.
> > FD is such pointer. FD == std::shared_ptr<bpf_prog>.
> > Holding that pointer guarantees that <bpf_prog> will not disappear,
> > but it says nothing that the program will keep running.
> > For [ku]probe,tp,fentry,fexit there was always <bpf_link> in the kernel.
> > It wasn't that formal in the past until most recent Andrii's patches,
> > but the concept existed for long time. FD == std::shared_ptr<bpf_link>
> > connects a kernel object with <bpf_prog>. When that kernel objects emits
> > an event the <bpf_link> guarantees that <bpf_prog> will be executed.
> 
> I see so the link is sort of [owner -> prog -> target].

No. Link has two pieces: kernel object and program. It connects the two.

> > For cgroups we don't have such concept. We thought that three attach modes we
> > introduced (default, allow-override, allow-multi) will cover all use cases. But
> > in practice turned out that it only works when there is a central daemon for
> > _all_ cgroup-bpf progs in the system otherwise different processes step on each
> > other. More so there has to be a central diff-review human authority otherwise
> > teams step on each other. That's sort-of works within one org, but doesn't
> > scale.
> > 
> > To avoid making systemd a central place to coordinate attaching xdp, tc, cgroup
> > progs the kernel has to provide a mechanism for an application to connect a
> > kernel object with a prog and hold the ownership of that link so that no other
> > process in the system can break that connection. 
> 
> To me for XDP the promise that nothing breaks the connection cannot be
> made without a daemon, because without the daemon the link has to be
> available somewhere/pinned to make changes to, and therefore is no
> longer safe.

This is not true. Nothing requires pinning.
Again. FD == shared_ptr. Two applications can share that link and
coordinate the changes to dispatcher prog without a daemon and without pinning.