Re: [PATCH bpf-next v5 1/2] bpf: enable generic kfuncs for BPF_CGROUP_* programs

[Andrii Nakryiko]

On Tue, Aug 13, 2024 at 6:28 AM Matteo Croce <technoboy85@xxxxxxxxx> wrote:
>
> From: Matteo Croce <teknoraver@xxxxxxxx>
>
> These kfuncs are enabled even in BPF_PROG_TYPE_TRACING, so they
> should be safe also in BPF_CGROUP_* programs.
>
> In enum btf_kfunc_hook, rename BTF_KFUNC_HOOK_CGROUP_SKB to a more
> generic BTF_KFUNC_HOOK_CGROUP, since it's used for all the cgroup
> related program types.
>
> Signed-off-by: Matteo Croce <teknoraver@xxxxxxxx>
> ---
>  kernel/bpf/btf.c     | 8 ++++++--
>  kernel/bpf/helpers.c | 6 ++++++
>  2 files changed, 12 insertions(+), 2 deletions(-)
>
> diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
> index 95426d5b634e..08d094875f00 100644
> --- a/kernel/bpf/btf.c
> +++ b/kernel/bpf/btf.c
> @@ -212,7 +212,7 @@ enum btf_kfunc_hook {
>         BTF_KFUNC_HOOK_TRACING,
>         BTF_KFUNC_HOOK_SYSCALL,
>         BTF_KFUNC_HOOK_FMODRET,
> -       BTF_KFUNC_HOOK_CGROUP_SKB,
> +       BTF_KFUNC_HOOK_CGROUP,
>         BTF_KFUNC_HOOK_SCHED_ACT,
>         BTF_KFUNC_HOOK_SK_SKB,
>         BTF_KFUNC_HOOK_SOCKET_FILTER,
> @@ -8312,8 +8312,12 @@ static int bpf_prog_type_to_kfunc_hook(enum bpf_prog_type prog_type)
>         case BPF_PROG_TYPE_SYSCALL:
>                 return BTF_KFUNC_HOOK_SYSCALL;
>         case BPF_PROG_TYPE_CGROUP_SKB:
> +       case BPF_PROG_TYPE_CGROUP_SOCK:
> +       case BPF_PROG_TYPE_CGROUP_DEVICE:
>         case BPF_PROG_TYPE_CGROUP_SOCK_ADDR:
> -               return BTF_KFUNC_HOOK_CGROUP_SKB;
> +       case BPF_PROG_TYPE_CGROUP_SOCKOPT:
> +       case BPF_PROG_TYPE_CGROUP_SYSCTL:
> +               return BTF_KFUNC_HOOK_CGROUP;
>         case BPF_PROG_TYPE_SCHED_ACT:
>                 return BTF_KFUNC_HOOK_SCHED_ACT;
>         case BPF_PROG_TYPE_SK_SKB:
> diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
> index d02ae323996b..0d1d97d968b0 100644
> --- a/kernel/bpf/helpers.c
> +++ b/kernel/bpf/helpers.c
> @@ -3052,6 +3052,12 @@ static int __init kfunc_init(void)
>         ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_XDP, &generic_kfunc_set);
>         ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &generic_kfunc_set);
>         ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SYSCALL, &generic_kfunc_set);
> +       ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_SKB, &generic_kfunc_set);
> +       ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_SOCK, &generic_kfunc_set);
> +       ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_DEVICE, &generic_kfunc_set);
> +       ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_SOCK_ADDR, &generic_kfunc_set);
> +       ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_SYSCTL, &generic_kfunc_set);
> +       ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_CGROUP_SOCKOPT, &generic_kfunc_set);

So given all those CGROUP_xxx program types map to the same
cgroup-generic BTF_KFUNC_HOOK_CGROUP hook, why do we need 6
repetitions of the same thing?

I'd say let's keep just one (pick any, CGROUP_SKB, for example)
registration. And then we should follow up with cleaning up
register_btf_kfunc_id_set() to accept hook type, not program type. And
then it will be clean and will make most sense.

But other than that looks good to me.

pw-bot: cr


>         ret = ret ?: register_btf_id_dtor_kfuncs(generic_dtors,
>                                                   ARRAY_SIZE(generic_dtors),
>                                                   THIS_MODULE);
> --
> 2.46.0
>