netdev - Re: [PATCH bpf-next 2/3] samples: bpf: Replace attach_tracepoint() to attach() in xdp_redirect

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Sat, 10 Oct 2020 18:58:31 +0900
From:   "Daniel T. Lee" <danieltimlee@...il.com>
To:     Andrii Nakryiko <andrii.nakryiko@...il.com>
Cc:     Daniel Borkmann <daniel@...earbox.net>,
        Alexei Starovoitov <ast@...nel.org>,
        Jesper Dangaard Brouer <brouer@...hat.com>,
        Lorenzo Bianconi <lorenzo@...nel.org>,
        bpf <bpf@...r.kernel.org>, Networking <netdev@...r.kernel.org>,
        Xdp <xdp-newbies@...r.kernel.org>
Subject: Re: [PATCH bpf-next 2/3] samples: bpf: Replace attach_tracepoint() to
 attach() in xdp_redirect_cpu

On Sat, Oct 10, 2020 at 3:23 AM Andrii Nakryiko
<andrii.nakryiko@...il.com> wrote:
>
> On Fri, Oct 9, 2020 at 9:04 AM Daniel T. Lee <danieltimlee@...il.com> wrote:
> >
> > From commit d7a18ea7e8b6 ("libbpf: Add generic bpf_program__attach()"),
> > for some BPF programs, it is now possible to attach BPF programs
> > with __attach() instead of explicitly calling __attach_<type>().
> >
> > This commit refactors the __attach_tracepoint() with libbpf's generic
> > __attach() method. In addition, this refactors the logic of setting
> > the map FD to simplify the code. Also, the missing removal of
> > bpf_load.o in Makefile has been fixed.
> >
> > Signed-off-by: Daniel T. Lee <danieltimlee@...il.com>
> > ---
> >  samples/bpf/Makefile                |   2 +-
> >  samples/bpf/xdp_redirect_cpu_user.c | 138 +++++++++++++---------------
> >  2 files changed, 67 insertions(+), 73 deletions(-)
> >
>
> [...]
>
> >  #define NUM_TP 5
> > +#define NUM_MAP 9
> >  struct bpf_link *tp_links[NUM_TP] = { 0 };
>
> = {}
>
> > +static int map_fds[NUM_MAP];
> >  static int tp_cnt = 0;
> >
> >  /* Exit return codes */
>
> [...]
>
> > -static struct bpf_link * attach_tp(struct bpf_object *obj,
> > -                                  const char *tp_category,
> > -                                  const char* tp_name)
> > +static int init_tracepoints(struct bpf_object *obj)
> >  {
> > +       char *tp_section = "tracepoint/";
> >         struct bpf_program *prog;
> > -       struct bpf_link *link;
> > -       char sec_name[PATH_MAX];
> > -       int len;
> > +       const char *section;
> >
> > -       len = snprintf(sec_name, PATH_MAX, "tracepoint/%s/%s",
> > -                      tp_category, tp_name);
> > -       if (len < 0)
> > -               exit(EXIT_FAIL);
> > +       bpf_object__for_each_program(prog, obj) {
> > +               section = bpf_program__section_name(prog);
> > +               if (strncmp(section, tp_section, strlen(tp_section)) != 0)
> > +                       continue;
>
> that's a convoluted and error-prone way (you can also use "tp/bla/bla"
> for tracepoint programs, for example). Use
> bpf_program__is_tracepoint() check.
>

Thanks for the review!
I think that's a much better way. I will send the next patch with applying
that method.

> >
> > -       prog = bpf_object__find_program_by_title(obj, sec_name);
> > -       if (!prog) {
> > -               fprintf(stderr, "ERR: finding progsec: %s\n", sec_name);
> > -               exit(EXIT_FAIL_BPF);
> > +               tp_links[tp_cnt] = bpf_program__attach(prog);
> > +               if (libbpf_get_error(tp_links[tp_cnt])) {
> > +                       tp_links[tp_cnt] = NULL;
> > +                       return -EINVAL;
> > +               }
> > +               tp_cnt++;
> >         }
> >
>
> [...]