linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jiri Olsa <jolsa@redhat.com>
To: Andrii Nakryiko <andrii.nakryiko@gmail.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
	Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Andrii Nakryiko <andrii@kernel.org>,
	Networking <netdev@vger.kernel.org>, bpf <bpf@vger.kernel.org>,
	lkml <linux-kernel@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>,
	Steven Rostedt <rostedt@goodmis.org>,
	"Naveen N . Rao" <naveen.n.rao@linux.ibm.com>,
	Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>,
	"David S . Miller" <davem@davemloft.net>
Subject: Re: [PATCH v7 00/10] fprobe: Introduce fprobe function entry/exit probe
Date: Wed, 2 Feb 2022 10:03:07 +0100	[thread overview]
Message-ID: <YfpIy+/Pqlw5EcZk@krava> (raw)
In-Reply-To: <CAEf4BzYRJuTLJc=Z9P-p3BtuKu_74MtA2MyrY_ceBxofuKuzHQ@mail.gmail.com>

On Tue, Feb 01, 2022 at 04:09:05PM -0800, Andrii Nakryiko wrote:
> On Tue, Feb 1, 2022 at 4:02 PM Jiri Olsa <jolsa@redhat.com> wrote:
> >
> > On Mon, Jan 31, 2022 at 02:00:24PM +0900, Masami Hiramatsu wrote:
> > > Hi,
> > >
> > > Here is the 7th version of fprobe. This version fixes unregister_fprobe()
> > > ensures that exit_handler is not called after returning from the
> > > unregister_fprobe(), and fixes some comments and documents.
> > >
> > > The previous version is here[1];
> > >
> > > [1] https://lore.kernel.org/all/164338031590.2429999.6203979005944292576.stgit@devnote2/T/#u
> > >
> > > This series introduces the fprobe, the function entry/exit probe
> > > with multiple probe point support. This also introduces the rethook
> > > for hooking function return as same as the kretprobe does. This
> > > abstraction will help us to generalize the fgraph tracer,
> > > because we can just switch to it from the rethook in fprobe,
> > > depending on the kernel configuration.
> > >
> > > The patch [1/10] is from Jiri's series[2].
> > >
> > > [2] https://lore.kernel.org/all/20220104080943.113249-1-jolsa@kernel.org/T/#u
> > >
> > > And the patch [9/10] adds the FPROBE_FL_KPROBE_SHARED flag for the case
> > > if user wants to share the same code (or share a same resource) on the
> > > fprobe and the kprobes.
> >
> > hi,
> > it works fine for bpf selftests, but when I use it through bpftrace
> > to attach more probes with:
> >
> >   # ./src/bpftrace -e 'kprobe:ksys_* { }'
> >   Attaching 27 probes
> >
> > I'm getting stalls like:
> >
> > krava33 login: [  988.574069] INFO: task bpftrace:4137 blocked for more than 122 seconds.
> > [  988.577577]       Not tainted 5.16.0+ #89
> > [  988.580173] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> > [  988.585538] task:bpftrace        state:D stack:    0 pid: 4137 ppid:  4123 flags:0x00004004
> > [  988.589869] Call Trace:
> > [  988.591312]  <TASK>
> > [  988.592577]  __schedule+0x3a8/0xd30
> > [  988.594469]  ? wait_for_completion+0x84/0x110
> > [  988.596753]  schedule+0x4e/0xc0
> > [  988.598480]  schedule_timeout+0xed/0x130
> > [  988.600524]  ? rcu_read_lock_sched_held+0x12/0x70
> > [  988.602901]  ? lock_release+0x253/0x4a0
> > [  988.604935]  ? lock_acquired+0x1b7/0x410
> > [  988.607041]  ? trace_hardirqs_on+0x1b/0xe0
> > [  988.609202]  wait_for_completion+0xae/0x110
> > [  988.613762]  __wait_rcu_gp+0x127/0x130
> > [  988.615787]  synchronize_rcu_tasks_generic+0x46/0xa0
> > [  988.618329]  ? call_rcu_tasks+0x20/0x20
> > [  988.620600]  ? rcu_tasks_pregp_step+0x10/0x10
> > [  988.623232]  ftrace_shutdown.part.0+0x174/0x210
> > [  988.625820]  unregister_ftrace_function+0x37/0x60
> > [  988.628480]  unregister_fprobe+0x2d/0x50
> > [  988.630928]  bpf_link_free+0x4e/0x70
> > [  988.633126]  bpf_link_release+0x11/0x20
> > [  988.635249]  __fput+0xae/0x270
> > [  988.637022]  task_work_run+0x5c/0xa0
> > [  988.639016]  exit_to_user_mode_prepare+0x251/0x260
> > [  988.641294]  syscall_exit_to_user_mode+0x16/0x50
> > [  988.646249]  do_syscall_64+0x48/0x90
> > [  988.648218]  entry_SYSCALL_64_after_hwframe+0x44/0xae
> > [  988.650787] RIP: 0033:0x7f9079e95fbb
> > [  988.652761] RSP: 002b:00007ffd474fa3b0 EFLAGS: 00000293 ORIG_RAX: 0000000000000003
> > [  988.656718] RAX: 0000000000000000 RBX: 00000000011bf8d0 RCX: 00007f9079e95fbb
> > [  988.660110] RDX: 0000000000000000 RSI: 00007ffd474fa3b0 RDI: 0000000000000019
> > [  988.663512] RBP: 00007ffd474faaf0 R08: 0000000000000000 R09: 000000000000001a
> > [  988.666673] R10: 0000000000000064 R11: 0000000000000293 R12: 0000000000000001
> > [  988.669770] R13: 00000000004a19a1 R14: 00007f9083428c00 R15: 00000000008c02d8
> > [  988.672601]  </TASK>
> > [  988.675763] INFO: lockdep is turned off.
> >
> > I have't investigated yet, any idea?
> >
> 
> Do you happen to have a CPU count that's not a power of 2? Check if
> you have [0] in your tree, it might be that.
> 
>   [0] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a773abf72eb0cac008743891068ca6edecc44683

yes, that helped, thanks

jirka


  reply	other threads:[~2022-02-02  9:03 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-31  5:00 [PATCH v7 00/10] fprobe: Introduce fprobe function entry/exit probe Masami Hiramatsu
2022-01-31  5:00 ` [PATCH v7 01/10] ftrace: Add ftrace_set_filter_ips function Masami Hiramatsu
2022-01-31  5:00 ` [PATCH v7 02/10] fprobe: Add ftrace based probe APIs Masami Hiramatsu
2022-01-31  5:01 ` [PATCH v7 03/10] rethook: Add a generic return hook Masami Hiramatsu
2022-01-31  5:01 ` [PATCH v7 04/10] rethook: x86: Add rethook x86 implementation Masami Hiramatsu
2022-01-31  5:01 ` [PATCH v7 05/10] ARM: rethook: Add rethook arm implementation Masami Hiramatsu
2022-01-31  5:01 ` [PATCH v7 06/10] arm64: rethook: Add arm64 rethook implementation Masami Hiramatsu
2022-01-31  5:01 ` [PATCH v7 07/10] fprobe: Add exit_handler support Masami Hiramatsu
2022-01-31  5:01 ` [PATCH v7 08/10] fprobe: Add sample program for fprobe Masami Hiramatsu
2022-01-31  5:02 ` [PATCH v7 09/10] fprobe: Introduce FPROBE_FL_KPROBE_SHARED flag " Masami Hiramatsu
2022-01-31  5:02 ` [PATCH v7 10/10] docs: fprobe: Add fprobe description to ftrace-use.rst Masami Hiramatsu
2022-01-31  9:36 ` [PATCH v7 00/10] fprobe: Introduce fprobe function entry/exit probe Masami Hiramatsu
2022-01-31 11:45   ` Jiri Olsa
2022-02-02  0:02 ` Jiri Olsa
2022-02-02  0:09   ` Andrii Nakryiko
2022-02-02  9:03     ` Jiri Olsa [this message]
2022-02-02  7:29   ` Masami Hiramatsu
  -- strict thread matches above, loose matches on Subject: below --
2022-01-31  4:59 Masami Hiramatsu

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=YfpIy+/Pqlw5EcZk@krava \
    --to=jolsa@redhat.com \
    --cc=andrii.nakryiko@gmail.com \
    --cc=andrii@kernel.org \
    --cc=anil.s.keshavamurthy@intel.com \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=john.fastabend@gmail.com \
    --cc=kafai@fb.com \
    --cc=kpsingh@chromium.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mhiramat@kernel.org \
    --cc=naveen.n.rao@linux.ibm.com \
    --cc=netdev@vger.kernel.org \
    --cc=rostedt@goodmis.org \
    --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).