[v2,bpf-next,0/3] Introduce pinnable bpf_link kernel abstraction

This patch series adds bpf_link abstraction, analogous to libbpf's already
existing bpf_link abstraction. This formalizes and makes more uniform existing
bpf_link-like BPF program link (attachment) types (raw tracepoint and tracing
links), which are FD-based objects that are automatically detached when last
file reference is closed. These types of BPF program links are switched to
using bpf_link framework.

FD-based bpf_link approach provides great safety guarantees, by ensuring there
is not going to be an abandoned BPF program attached, if user process suddenly
exits or forgets to clean up after itself. This is especially important in
production environment and is what all the recent new BPF link types followed.

One of the previously existing  inconveniences of FD-based approach, though,
was the scenario in which user process wants to install BPF link and exit, but
let attached BPF program run. Now, with bpf_link abstraction in place, it's
easy to support pinning links in BPF FS, which is done as part of the same
patch #1. This allows FD-based BPF program links to survive exit of a user
process and original file descriptor being closed, by creating an file entry
in BPF FS. This provides great safety by default, with simple way to opt out
for cases where it's needed.

Corresponding libbpf APIs are added in the same patch set, as well as
selftests for this functionality.

Other types of BPF program attachments (XDP, cgroup, perf_event, etc) are
going to be converted in subsequent patches to follow similar approach.

v1->v2:
- use bpf_link_new_fd() uniformly (Alexei).

Andrii Nakryiko (3):
  bpf: introduce pinnable bpf_link abstraction
  libbpf: add bpf_link pinning/unpinning
  selftests/bpf: add link pinning selftests

 include/linux/bpf.h                           |  13 +
 kernel/bpf/inode.c                            |  42 +++-
 kernel/bpf/syscall.c                          | 223 ++++++++++++++----
 tools/lib/bpf/libbpf.c                        | 131 +++++++---
 tools/lib/bpf/libbpf.h                        |   5 +
 tools/lib/bpf/libbpf.map                      |   5 +
 .../selftests/bpf/prog_tests/link_pinning.c   | 105 +++++++++
 .../selftests/bpf/progs/test_link_pinning.c   |  25 ++
 8 files changed, 476 insertions(+), 73 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/link_pinning.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_link_pinning.c

On Mon, Mar 2, 2020 at 8:32 PM Andrii Nakryiko <andriin@fb.com> wrote:
>
> This patch series adds bpf_link abstraction, analogous to libbpf's already
> existing bpf_link abstraction. This formalizes and makes more uniform existing
> bpf_link-like BPF program link (attachment) types (raw tracepoint and tracing
> links), which are FD-based objects that are automatically detached when last
> file reference is closed. These types of BPF program links are switched to
> using bpf_link framework.
>
> FD-based bpf_link approach provides great safety guarantees, by ensuring there
> is not going to be an abandoned BPF program attached, if user process suddenly
> exits or forgets to clean up after itself. This is especially important in
> production environment and is what all the recent new BPF link types followed.
>
> One of the previously existing  inconveniences of FD-based approach, though,
> was the scenario in which user process wants to install BPF link and exit, but
> let attached BPF program run. Now, with bpf_link abstraction in place, it's
> easy to support pinning links in BPF FS, which is done as part of the same
> patch #1. This allows FD-based BPF program links to survive exit of a user
> process and original file descriptor being closed, by creating an file entry
> in BPF FS. This provides great safety by default, with simple way to opt out
> for cases where it's needed.
>
> Corresponding libbpf APIs are added in the same patch set, as well as
> selftests for this functionality.
>
> Other types of BPF program attachments (XDP, cgroup, perf_event, etc) are
> going to be converted in subsequent patches to follow similar approach.

Applied. Thanks.