Re: [PATCH bpf-next 2/5] libbpf: support BPF_PROG_TYPE_USER programs

From: Andrii Nakryiko <andrii.nakryiko@gmail.com>
To: Song Liu <songliubraving@fb.com>
Cc: open list <linux-kernel@vger.kernel.org>,
	bpf <bpf@vger.kernel.org>, Networking <netdev@vger.kernel.org>,
	Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Kernel Team <kernel-team@fb.com>,
	john fastabend <john.fastabend@gmail.com>,
	KP Singh <kpsingh@chromium.org>,
	Jesper Dangaard Brouer <brouer@redhat.com>,
	Daniel Xu <dlxu@fb.com>
Subject: Re: [PATCH bpf-next 2/5] libbpf: support BPF_PROG_TYPE_USER programs
Date: Sun, 2 Aug 2020 18:40:27 -0700	[thread overview]
Message-ID: <CAEf4BzYp4gO1P+OrY7hGyQjdia3BuSu4DX2_z=UF6RfGNa+gkQ@mail.gmail.com> (raw)
In-Reply-To: <20200801084721.1812607-3-songliubraving@fb.com>

On Sat, Aug 1, 2020 at 1:50 AM Song Liu <songliubraving@fb.com> wrote:
>
> Add cpu_plus to bpf_prog_test_run_attr. Add BPF_PROG_SEC "user" for
> BPF_PROG_TYPE_USER programs.
>
> Signed-off-by: Song Liu <songliubraving@fb.com>
> ---
>  tools/lib/bpf/bpf.c           | 1 +
>  tools/lib/bpf/bpf.h           | 3 +++
>  tools/lib/bpf/libbpf.c        | 1 +
>  tools/lib/bpf/libbpf_probes.c | 1 +
>  4 files changed, 6 insertions(+)
>
> diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
> index e1bdf214f75fe..b28c3daa9c270 100644
> --- a/tools/lib/bpf/bpf.c
> +++ b/tools/lib/bpf/bpf.c
> @@ -693,6 +693,7 @@ int bpf_prog_test_run_xattr(struct bpf_prog_test_run_attr *test_attr)
>         attr.test.ctx_size_in = test_attr->ctx_size_in;
>         attr.test.ctx_size_out = test_attr->ctx_size_out;
>         attr.test.repeat = test_attr->repeat;
> +       attr.test.cpu_plus = test_attr->cpu_plus;
>
>         ret = sys_bpf(BPF_PROG_TEST_RUN, &attr, sizeof(attr));
>         test_attr->data_size_out = attr.test.data_size_out;
> diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
> index 6d367e01d05e9..0c799740df566 100644
> --- a/tools/lib/bpf/bpf.h
> +++ b/tools/lib/bpf/bpf.h
> @@ -205,6 +205,9 @@ struct bpf_prog_test_run_attr {
>         void *ctx_out;      /* optional */
>         __u32 ctx_size_out; /* in: max length of ctx_out
>                              * out: length of cxt_out */
> +       __u32 cpu_plus;     /* specify which cpu to run the test with
> +                            * cpu_plus = cpu_id + 1.
> +                            * If cpu_plus = 0, run on current cpu */

We can't do this due to ABI guarantees. We'll have to add a new API
using OPTS arguments.

>  };
>
>  LIBBPF_API int bpf_prog_test_run_xattr(struct bpf_prog_test_run_attr *test_attr);
> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> index b9f11f854985b..9ce175a486214 100644
> --- a/tools/lib/bpf/libbpf.c
> +++ b/tools/lib/bpf/libbpf.c
> @@ -6922,6 +6922,7 @@ static const struct bpf_sec_def section_defs[] = {
>         BPF_PROG_SEC("lwt_out",                 BPF_PROG_TYPE_LWT_OUT),
>         BPF_PROG_SEC("lwt_xmit",                BPF_PROG_TYPE_LWT_XMIT),
>         BPF_PROG_SEC("lwt_seg6local",           BPF_PROG_TYPE_LWT_SEG6LOCAL),
> +       BPF_PROG_SEC("user",                    BPF_PROG_TYPE_USER),

let's do "user/" for consistency with most other prog types (and nice
separation between prog type and custom user name)

>         BPF_APROG_SEC("cgroup_skb/ingress",     BPF_PROG_TYPE_CGROUP_SKB,
>                                                 BPF_CGROUP_INET_INGRESS),
>         BPF_APROG_SEC("cgroup_skb/egress",      BPF_PROG_TYPE_CGROUP_SKB,
> diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
> index 5a3d3f0784081..163013084000e 100644
> --- a/tools/lib/bpf/libbpf_probes.c
> +++ b/tools/lib/bpf/libbpf_probes.c
> @@ -112,6 +112,7 @@ probe_load(enum bpf_prog_type prog_type, const struct bpf_insn *insns,
>         case BPF_PROG_TYPE_STRUCT_OPS:
>         case BPF_PROG_TYPE_EXT:
>         case BPF_PROG_TYPE_LSM:
> +       case BPF_PROG_TYPE_USER:
>         default:
>                 break;
>         }
> --
> 2.24.1
>