bpf.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jiri Olsa <jolsa@kernel.org>
To: Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Andrii Nakryiko <andrii@kernel.org>
Cc: bpf@vger.kernel.org, Martin KaFai Lau <kafai@fb.com>,
	Song Liu <songliubraving@fb.com>, Yonghong Song <yhs@fb.com>,
	John Fastabend <john.fastabend@gmail.com>,
	KP Singh <kpsingh@chromium.org>,
	Stanislav Fomichev <sdf@google.com>, Hao Luo <haoluo@google.com>,
	Hou Tao <houtao1@huawei.com>, Daniel Xu <dxu@dxuuu.xyz>
Subject: [PATCH bpf-next 00/12] bpf: Add missed stats for kprobes
Date: Mon, 28 Aug 2023 09:55:25 +0200	[thread overview]
Message-ID: <20230828075537.194192-1-jolsa@kernel.org> (raw)

hi,
at the moment we can't retrieve the number of missed kprobe
executions and subsequent execution of BPF programs.

This patchset adds:
  - counting of missed execution on attach layer for:
    . kprobes attached through perf link (kprobe/ftrace)
    . kprobes attached through kprobe.multi link (fprobe)
  - counting of recursion_misses for BPF kprobe programs
  - counting runtime stats (kernel.bpf_stats_enabled=1) for BPF programs
    executed through bpf_prog_run_array - kprobes, perf events, trace
    syscall probes


It's still technically possible to create kprobe without perf link (using
SET_BPF perf ioctl) in which case we don't have a way to retrieve the kprobe's
'missed' count. However both libbpf and cilium/ebpf libraries use perf link
if it's available, and for old kernels without perf link support we can use
BPF program to retrieve the kprobe missed count.

Also available at:
  https://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf.git
  bpf/missed_stats

thanks,
jirka


---
Jiri Olsa (12):
      bpf: Move update_prog_stats to syscall object
      bpf: Move bpf_prog_start_time to linux/filter.h
      bpf: Count stats for kprobe_multi programs
      bpf: Add missed value to kprobe_multi link info
      bpf: Add missed value to kprobe perf link info
      bpf: Count missed stats in trace_call_bpf
      bpf: Move bpf_prog_run_array down in the header file
      bpf: Count run stats in bpf_prog_run_array
      bpftool: Display missed count for kprobe_multi link
      bpftool: Display missed count for kprobe perf link
      selftests/bpf: Add test missed counts of perf event link kprobe
      elftests/bpf: Add test recursion stats of perf event link kprobe

 include/linux/bpf.h                                         | 106 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++--------------------------
 include/linux/trace_events.h                                |   6 ++++--
 include/uapi/linux/bpf.h                                    |   2 ++
 kernel/bpf/syscall.c                                        |  36 +++++++++++++++++++++++++------
 kernel/bpf/trampoline.c                                     |  45 +++++----------------------------------
 kernel/trace/bpf_trace.c                                    |  17 ++++++++++++---
 kernel/trace/trace_kprobe.c                                 |   5 ++++-
 tools/bpf/bpftool/link.c                                    |   8 ++++++-
 tools/include/uapi/linux/bpf.h                              |   2 ++
 tools/testing/selftests/bpf/DENYLIST.aarch64                |   1 +
 tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c       |   5 +++++
 tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h |   2 ++
 tools/testing/selftests/bpf/prog_tests/missed.c             |  97 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/missed_kprobe.c           |  30 ++++++++++++++++++++++++++
 tools/testing/selftests/bpf/progs/missed_kprobe_recursion.c |  48 +++++++++++++++++++++++++++++++++++++++++
 15 files changed, 327 insertions(+), 83 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/missed.c
 create mode 100644 tools/testing/selftests/bpf/progs/missed_kprobe.c
 create mode 100644 tools/testing/selftests/bpf/progs/missed_kprobe_recursion.c

             reply	other threads:[~2023-08-28  7:55 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-28  7:55 Jiri Olsa [this message]
2023-08-28  7:55 ` [PATCH bpf-next 01/12] bpf: Move update_prog_stats to syscall object Jiri Olsa
2023-08-28 17:25   ` Alexei Starovoitov
2023-08-29  8:00     ` Jiri Olsa
2023-08-28  7:55 ` [PATCH bpf-next 02/12] bpf: Move bpf_prog_start_time to linux/filter.h Jiri Olsa
2023-09-04 13:12   ` Hou Tao
2023-08-28  7:55 ` [PATCH bpf-next 03/12] bpf: Count stats for kprobe_multi programs Jiri Olsa
2023-09-04 13:30   ` Hou Tao
2023-09-05  6:15   ` Hou Tao
2023-09-05  7:19     ` Jiri Olsa
2023-08-28  7:55 ` [PATCH bpf-next 04/12] bpf: Add missed value to kprobe_multi link info Jiri Olsa
2023-09-04 13:33   ` Hou Tao
2023-08-28  7:55 ` [PATCH bpf-next 05/12] bpf: Add missed value to kprobe perf " Jiri Olsa
2023-09-05  2:23   ` Hou Tao
2023-09-05  7:19     ` Jiri Olsa
2023-08-28  7:55 ` [PATCH bpf-next 06/12] bpf: Count missed stats in trace_call_bpf Jiri Olsa
2023-08-28 17:32   ` Alexei Starovoitov
2023-08-29  8:04     ` Jiri Olsa
2023-08-28  7:55 ` [PATCH bpf-next 07/12] bpf: Move bpf_prog_run_array down in the header file Jiri Olsa
2023-08-28  7:55 ` [PATCH bpf-next 08/12] bpf: Count run stats in bpf_prog_run_array Jiri Olsa
2023-09-05  2:40   ` Hou Tao
2023-09-05  7:19     ` Jiri Olsa
2023-08-28  7:55 ` [PATCH bpf-next 09/12] bpftool: Display missed count for kprobe_multi link Jiri Olsa
2023-08-29 16:40   ` Quentin Monnet
2023-08-30 10:46     ` Jiri Olsa
2023-08-28  7:55 ` [PATCH bpf-next 10/12] bpftool: Display missed count for kprobe perf link Jiri Olsa
2023-08-29 16:42   ` Quentin Monnet
2023-08-30 16:01     ` Jiri Olsa
2023-08-28  7:55 ` [PATCH bpf-next 11/12] selftests/bpf: Add test missed counts of perf event link kprobe Jiri Olsa
2023-09-05  3:14   ` Hou Tao
2023-08-28  7:55 ` [PATCH bpf-next 12/12] selftests/bpf: Add test recursion stats " Jiri Olsa
2023-09-05  3:23   ` Hou Tao
2023-09-05  3:30 ` [PATCH bpf-next 00/12] bpf: Add missed stats for kprobes Hou Tao
2023-09-05  7:20   ` Jiri Olsa

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230828075537.194192-1-jolsa@kernel.org \
    --to=jolsa@kernel.org \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=dxu@dxuuu.xyz \
    --cc=haoluo@google.com \
    --cc=houtao1@huawei.com \
    --cc=john.fastabend@gmail.com \
    --cc=kafai@fb.com \
    --cc=kpsingh@chromium.org \
    --cc=sdf@google.com \
    --cc=songliubraving@fb.com \
    --cc=yhs@fb.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).