Re: [PATCH v3 bpf-next 4/7] libbpf: prog load: allow to use fd_array_cnt

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, Nov 29, 2024 at 5:29 AM Anton Protopopov <aspsk@xxxxxxxxxxxxx> wrote:
>
> Add new fd_array_cnt field to bpf_prog_load_opts
> and pass it in bpf_attr, if set.
>
> Signed-off-by: Anton Protopopov <aspsk@xxxxxxxxxxxxx>
> ---
>  tools/lib/bpf/bpf.c      | 5 +++--
>  tools/lib/bpf/bpf.h      | 5 ++++-
>  tools/lib/bpf/features.c | 2 +-
>  3 files changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
> index becdfa701c75..0e7f59224936 100644
> --- a/tools/lib/bpf/bpf.c
> +++ b/tools/lib/bpf/bpf.c
> @@ -105,7 +105,7 @@ int sys_bpf_prog_load(union bpf_attr *attr, unsigned int size, int attempts)
>   */
>  int probe_memcg_account(int token_fd)
>  {
> -       const size_t attr_sz = offsetofend(union bpf_attr, prog_token_fd);
> +       const size_t attr_sz = offsetofend(union bpf_attr, fd_array_cnt);

Thankfully this function doesn't set fd_array_cnt below.
Otherwise the detection of memcg would fail on older kernels.
Let's avoid people mindlessly adding init of fd_array_cnt here by accident.
Simply keep this line as-is.
offsetofend(.., prog_token_fd) is fine.

>         struct bpf_insn insns[] = {
>                 BPF_EMIT_CALL(BPF_FUNC_ktime_get_coarse_ns),
>                 BPF_EXIT_INSN(),
> @@ -238,7 +238,7 @@ int bpf_prog_load(enum bpf_prog_type prog_type,
>                   const struct bpf_insn *insns, size_t insn_cnt,
>                   struct bpf_prog_load_opts *opts)
>  {
> -       const size_t attr_sz = offsetofend(union bpf_attr, prog_token_fd);
> +       const size_t attr_sz = offsetofend(union bpf_attr, fd_array_cnt);
>         void *finfo = NULL, *linfo = NULL;
>         const char *func_info, *line_info;
>         __u32 log_size, log_level, attach_prog_fd, attach_btf_obj_fd;
> @@ -311,6 +311,7 @@ int bpf_prog_load(enum bpf_prog_type prog_type,
>         attr.line_info_cnt = OPTS_GET(opts, line_info_cnt, 0);
>
>         attr.fd_array = ptr_to_u64(OPTS_GET(opts, fd_array, NULL));
> +       attr.fd_array_cnt = OPTS_GET(opts, fd_array_cnt, 0);
>
>         if (log_level) {
>                 attr.log_buf = ptr_to_u64(log_buf);
> diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
> index a4a7b1ad1b63..435da95d2058 100644
> --- a/tools/lib/bpf/bpf.h
> +++ b/tools/lib/bpf/bpf.h
> @@ -107,9 +107,12 @@ struct bpf_prog_load_opts {
>          */
>         __u32 log_true_size;
>         __u32 token_fd;
> +
> +       /* if set, provides the length of fd_array */
> +       __u32 fd_array_cnt;
>         size_t :0;
>  };
> -#define bpf_prog_load_opts__last_field token_fd
> +#define bpf_prog_load_opts__last_field fd_array_cnt
>
>  LIBBPF_API int bpf_prog_load(enum bpf_prog_type prog_type,
>                              const char *prog_name, const char *license,
> diff --git a/tools/lib/bpf/features.c b/tools/lib/bpf/features.c
> index 760657f5224c..5afc9555d9ac 100644
> --- a/tools/lib/bpf/features.c
> +++ b/tools/lib/bpf/features.c
> @@ -22,7 +22,7 @@ int probe_fd(int fd)
>
>  static int probe_kern_prog_name(int token_fd)
>  {
> -       const size_t attr_sz = offsetofend(union bpf_attr, prog_token_fd);
> +       const size_t attr_sz = offsetofend(union bpf_attr, fd_array_cnt);

Same here. Don't change it.





[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux