netdev - Re: [PATCH bpf-next 2/7] libbpf: add ability to attach/detach BPF to perf event

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

Message-ID: <CAEf4Bzaxh4tZB1eVUu-G5bGzwe7ehLVbb5M49c_YeeSuFibTTA@mail.gmail.com>
Date:   Thu, 20 Jun 2019 21:28:30 -0700
From:   Andrii Nakryiko <andrii.nakryiko@...il.com>
To:     Stanislav Fomichev <sdf@...ichev.me>
Cc:     Andrii Nakryiko <andriin@...com>, Alexei Starovoitov <ast@...com>,
        Daniel Borkmann <daniel@...earbox.net>,
        Networking <netdev@...r.kernel.org>, bpf <bpf@...r.kernel.org>,
        Kernel Team <kernel-team@...com>
Subject: Re: [PATCH bpf-next 2/7] libbpf: add ability to attach/detach BPF to
 perf event

On Thu, Jun 20, 2019 at 5:01 PM Stanislav Fomichev <sdf@...ichev.me> wrote:
>
> On 06/20, Andrii Nakryiko wrote:
> > bpf_program__attach_perf_event allows to attach BPF program to existing
> > perf event, providing most generic and most low-level way to attach BPF
> > programs.
> >
> > libbpf_perf_event_disable_and_close API is added to disable and close
> > existing perf event by its FD.
> >
> > Signed-off-by: Andrii Nakryiko <andriin@...com>
> > ---
> >  tools/lib/bpf/libbpf.c   | 41 ++++++++++++++++++++++++++++++++++++++++
> >  tools/lib/bpf/libbpf.h   |  4 ++++
> >  tools/lib/bpf/libbpf.map |  2 ++
> >  3 files changed, 47 insertions(+)
> >
> > diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> > index 8ce3beba8551..2bb1fa008be3 100644
> > --- a/tools/lib/bpf/libbpf.c
> > +++ b/tools/lib/bpf/libbpf.c
> > @@ -32,6 +32,7 @@
> >  #include <linux/limits.h>
> >  #include <linux/perf_event.h>
> >  #include <linux/ring_buffer.h>
> > +#include <sys/ioctl.h>
> >  #include <sys/stat.h>
> >  #include <sys/types.h>
> >  #include <sys/vfs.h>
> > @@ -3928,6 +3929,46 @@ int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr,
> >       return 0;
> >  }
> >
> [..]
> > +int libbpf_perf_event_disable_and_close(int pfd)
> nit: why not call it libbpf_perf_event_detach[_and_close]?
> It's usually attach/detach.

I think detach is actually confusing for perf event. Here's what you
do for tracing:

1. open perf event
2. enable perf event
3. attach BPF program to perf event
...
4. <is there a way to detach BPF program?>
5. disable perf event
6. close perf event

So open/close event, enable/disable event, attach / (auto-detach on
close right now) BPF program.

It seems like there should be explicit "detach this BPF program from
perf event without killing event itself", but I haven't found it.

But my point is that for event open/close and enable/disable seems
very complementary.


>
> > +{
> > +     int err;
> > +
> > +     if (pfd < 0)
> > +             return 0;
> > +
> > +     err = ioctl(pfd, PERF_EVENT_IOC_DISABLE, 0);
> > +     close(pfd);
> > +     return err;
> > +}
> > +
> > +int bpf_program__attach_perf_event(struct bpf_program *prog, int pfd)
> > +{
> > +     char errmsg[STRERR_BUFSIZE];
> > +     int bpf_fd, err;
> > +
> > +     bpf_fd = bpf_program__fd(prog);
> > +     if (bpf_fd < 0) {
> > +             pr_warning("program '%s': can't attach before loaded\n",
> > +                        bpf_program__title(prog, false));
> > +             return -EINVAL;
> > +     }
> > +     if (ioctl(pfd, PERF_EVENT_IOC_SET_BPF, bpf_fd) < 0) {
> > +             err = -errno;
> > +             pr_warning("program '%s': failed to attach to pfd %d: %s\n",
> > +                        bpf_program__title(prog, false), pfd,
> > +                        libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
> > +             return err;
> > +     }
> > +     if (ioctl(pfd, PERF_EVENT_IOC_ENABLE, 0) < 0) {
> > +             err = -errno;
> > +             pr_warning("program '%s': failed to enable pfd %d: %s\n",
> > +                        bpf_program__title(prog, false), pfd,
> > +                        libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
> > +             return err;
> > +     }
> > +     return 0;
> > +}
> > +
> >  enum bpf_perf_event_ret
> >  bpf_perf_event_read_simple(void *mmap_mem, size_t mmap_size, size_t page_size,
> >                          void **copy_mem, size_t *copy_size,
> > diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
> > index d639f47e3110..76db1bbc0dac 100644
> > --- a/tools/lib/bpf/libbpf.h
> > +++ b/tools/lib/bpf/libbpf.h
> > @@ -165,6 +165,10 @@ LIBBPF_API int bpf_program__pin(struct bpf_program *prog, const char *path);
> >  LIBBPF_API int bpf_program__unpin(struct bpf_program *prog, const char *path);
> >  LIBBPF_API void bpf_program__unload(struct bpf_program *prog);
> >
> > +LIBBPF_API int libbpf_perf_event_disable_and_close(int pfd);
> > +LIBBPF_API int bpf_program__attach_perf_event(struct bpf_program *prog,
> > +                                           int pfd);
> > +
> >  struct bpf_insn;
> >
> >  /*
> > diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
> > index 2c6d835620d2..d27406982b5a 100644
> > --- a/tools/lib/bpf/libbpf.map
> > +++ b/tools/lib/bpf/libbpf.map
> > @@ -172,5 +172,7 @@ LIBBPF_0.0.4 {
> >               btf_dump__new;
> >               btf__parse_elf;
> >               bpf_object__load_xattr;
> > +             bpf_program__attach_perf_event;
> >               libbpf_num_possible_cpus;
> > +             libbpf_perf_event_disable_and_close;
> >  } LIBBPF_0.0.3;
> > --
> > 2.17.1
> >