On Fri, Apr 29, 2022 at 2:15 PM Stanislav Fomichev <sdf@xxxxxxxxxx> wrote:
>
> We have two options:
> 1. Treat all BPF_LSM_CGROUP as the same, regardless of attach_btf_id
> 2. Treat BPF_LSM_CGROUP+attach_btf_id as a separate hook point
>
> I'm doing (2) here and adding attach_btf_id as a new BPF_PROG_QUERY
> argument. The downside is that it requires iterating over all possible
> bpf_lsm_ hook points in the userspace which might take some time.
>
> Signed-off-by: Stanislav Fomichev <sdf@xxxxxxxxxx>
> ---
> include/uapi/linux/bpf.h | 1 +
> kernel/bpf/cgroup.c | 43 ++++++++++++++++++++++++----------
> kernel/bpf/syscall.c | 3 ++-
> tools/include/uapi/linux/bpf.h | 1 +
> tools/lib/bpf/bpf.c | 42 ++++++++++++++++++++++++++-------
> tools/lib/bpf/bpf.h | 15 ++++++++++++
> tools/lib/bpf/libbpf.map | 1 +
please split kernel and libbpf changes into separate patches and mark
libbpf's with "libbpf: " prefix
> 7 files changed, 85 insertions(+), 21 deletions(-)
>
[...]
> diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
> index a9d292c106c2..f62823451b99 100644
> --- a/tools/lib/bpf/bpf.c
> +++ b/tools/lib/bpf/bpf.c
> @@ -946,28 +946,54 @@ int bpf_iter_create(int link_fd)
> return libbpf_err_errno(fd);
> }
>
> -int bpf_prog_query(int target_fd, enum bpf_attach_type type, __u32 query_flags,
> - __u32 *attach_flags, __u32 *prog_ids, __u32 *prog_cnt)
> +int bpf_prog_query2(int target_fd,
this would have to be named bpf_prog_query_opts()
> + enum bpf_attach_type type,
> + struct bpf_prog_query_opts *opts)
> {
> union bpf_attr attr;
> int ret;
>
> memset(&attr, 0, sizeof(attr));
> +
[...]
> diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
> index b5bc84039407..5e5bb3e437cc 100644
> --- a/tools/lib/bpf/libbpf.map
> +++ b/tools/lib/bpf/libbpf.map
> @@ -450,4 +450,5 @@ LIBBPF_0.8.0 {
> bpf_program__attach_usdt;
> libbpf_register_prog_handler;
> libbpf_unregister_prog_handler;
> + bpf_prog_query2;
this list is alphabetically ordered
> } LIBBPF_0.7.0;
> --
> 2.36.0.464.gb9c8b46e94-goog
>
