Re: [PATCH bpf-next v5 1/2] bpf: support BPF_PROG_QUERY for progs attached to sockmap

From: zhudi (E)
Date: Fri Jan 14 2022 - 00:45:03 EST


> On Thu, Jan 13, 2022 at 10:00 AM CET, Di Zhu wrote:
> > Right now there is no way to query whether BPF programs are
> > attached to a sockmap or not.
> >
> > we can use the standard interface in libbpf to query, such as:
> > bpf_prog_query(mapFd, BPF_SK_SKB_STREAM_PARSER, 0, NULL, ...);
> > the mapFd is the fd of sockmap.
> >
> > Signed-off-by: Di Zhu <zhudi2@xxxxxxxxxx>
> > Acked-by: Yonghong Song <yhs@xxxxxx>
> > ---
> > include/linux/bpf.h | 9 +++++
> > kernel/bpf/syscall.c | 5 +++
> > net/core/sock_map.c | 78
> ++++++++++++++++++++++++++++++++++++++++----
> > 3 files changed, 85 insertions(+), 7 deletions(-)
> >
> > diff --git a/include/linux/bpf.h b/include/linux/bpf.h
> > index 6e947cd91152..c4ca14c9f838 100644
> > --- a/include/linux/bpf.h
> > +++ b/include/linux/bpf.h
> > @@ -2071,6 +2071,9 @@ int bpf_prog_test_run_syscall(struct bpf_prog
> *prog,
> > int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog
> *prog);
> > int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type
> ptype);
> > int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value,
> u64 flags);


.......


> > +int sock_map_bpf_prog_query(const union bpf_attr *attr,
> > + union bpf_attr __user *uattr)
> > +{
> > + __u32 __user *prog_ids = u64_to_user_ptr(attr->query.prog_ids);
> > + u32 prog_cnt = 0, flags = 0, ufd = attr->target_fd;
> > + struct bpf_prog **pprog;
> > + struct bpf_prog *prog;
> > + struct bpf_map *map;
> > + struct fd f;
> > + u32 id = 0;
> > + int ret;
> > +
> > + if (attr->query.query_flags)
> > + return -EINVAL;
> > +
> > + f = fdget(ufd);
> > + map = __bpf_map_get(f);
> > + if (IS_ERR(map))
> > + return PTR_ERR(map);
> > +
> > + rcu_read_lock();
> > +
> > + ret = sock_map_prog_lookup(map, &pprog, attr->query.attach_type);
> > + if (ret)
> > + goto end;
> > +
> > + prog = *pprog;
> > + prog_cnt = !prog ? 0 : 1;
> > +
> > + if (!attr->query.prog_cnt || !prog_ids || !prog_cnt)
> > + goto end;
> > +
> > + id = prog->aux->id;
>
> ^ This looks like a concurrent read/write.
>
> Would wrap with READ_ONCE() and corresponding WRITE_ONCE() in
> bpf_prog_free_id(). See [1] for rationale.
>
> [1]
> https://github.com/google/kernel-sanitizers/blob/master/other/READ_WRITE_O
> NCE.md


Thanks for your advice, I will modify this code.


> > +
> > + /* we do not hold the refcnt, the bpf prog may be released
> > + * asynchronously and the id would be set to 0.
> > + */
> > + if (id == 0)
> > + prog_cnt = 0;
> > +
> > +end:
> > + rcu_read_unlock();
> > +
> > + if (copy_to_user(&uattr->query.attach_flags, &flags, sizeof(flags)) ||
> > + (id != 0 && copy_to_user(prog_ids, &id, sizeof(u32))) ||
> > + copy_to_user(&uattr->query.prog_cnt, &prog_cnt, sizeof(prog_cnt)))
> > + ret = -EFAULT;
> > +
> > + fdput(f);
> > + return ret;
> > +}
> > +
> > static void sock_map_unlink(struct sock *sk, struct sk_psock_link *link)
> > {
> > switch (link->map->map_type) {