From: Masami Hiramatsu <mhiramat@kernel.org>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>,
Alexei Starovoitov <ast@kernel.org>,
Ingo Molnar <mingo@kernel.org>,
bpf@vger.kernel.org, linux-kernel@vger.kernel.org,
Daniel Borkmann <daniel@iogearbox.net>,
Arnaldo Carvalho de Melo <acme@kernel.org>,
"David S . Miller" <davem@davemloft.net>,
paulmck@kernel.org, joel@joelfernandes.org,
"Naveen N . Rao" <naveen.n.rao@linux.ibm.com>,
Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Subject: Re: [RFT PATCH 04/13] kprobes: Make optimizer delay to 1 second
Date: Wed, 22 Jan 2020 22:12:40 +0900 [thread overview]
Message-ID: <20200122221240.cef447446785f46862fee97a@kernel.org> (raw)
In-Reply-To: <20200122071115.28e3c763@gandalf.local.home>
On Wed, 22 Jan 2020 07:11:15 -0500
Steven Rostedt <rostedt@goodmis.org> wrote:
> On Wed, 22 Jan 2020 16:23:17 +0900
> Masami Hiramatsu <mhiramat@kernel.org> wrote:
>
> > On Tue, 21 Jan 2020 19:29:05 -0500
> > Steven Rostedt <rostedt@goodmis.org> wrote:
> >
> > > On Thu, 16 Jan 2020 23:44:52 +0900
> > > Masami Hiramatsu <mhiramat@kernel.org> wrote:
> > >
> > > > Since the 5 jiffies delay for the optimizer is too
> > > > short to wait for other probes, make it longer,
> > > > like 1 second.
> > >
> > > Hi Masami,
> > >
> > > Can you explain more *why* 5 jiffies is too short.
> >
> > Yes, I had introduced this 5 jiffies delay for multiple probe registration
> > and unregistration like systemtap, which will use array-based interface to
> > register/unregister. In that case, 5 jiffies will be enough for the delay
> > to wait for other kprobe registration/unregsitration.
> >
> > However, since perf and ftrace register/unregister probes one-by-one with
> > RCU synchronization interval, the optimizer will be started before
> > finishing to register/unregister all probes.
> > And the optimizer locks kprobe_mutex a while -- RCU-tasks synchronization.
> > Since the kprobe_mutex is also involved in disabling kprobes, this also
> > stops probe-event disabling.
> >
> > Maybe 5 jiffies is enough for adding/removing a few probe events, but
> > not enough for dozens of probe events.
> >
>
> Perhaps we should have a mechanism that can detect new probes being
> added, and just continue to delay the optimization, instead of having
> some arbitrary delay.
Yes, that is what [03/13] does :)
Anyway, it seems that the RCU-synchronization takes more than 5 jiffies.
And in that case, [03/13] still doesn't work. That's why I added this patch
after that.
Thank you,
--
Masami Hiramatsu <mhiramat@kernel.org>
next prev parent reply other threads:[~2020-01-22 13:12 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-16 14:44 [RFT PATCH 00/13] tracing: kprobes: Introduce async unregistration Masami Hiramatsu
2020-01-16 14:44 ` [RFT PATCH 01/13] kprobes: Fix to protect kick_kprobe_optimizer() by kprobe_mutex Masami Hiramatsu
2020-01-16 14:44 ` [RFT PATCH 02/13] kprobes: Remove redundant arch_disarm_kprobe() call Masami Hiramatsu
2020-01-16 14:44 ` [RFT PATCH 03/13] kprobes: Postpone optimizer until a bunch of probes (un)registered Masami Hiramatsu
2020-01-16 14:44 ` [RFT PATCH 04/13] kprobes: Make optimizer delay to 1 second Masami Hiramatsu
2020-01-22 0:29 ` Steven Rostedt
2020-01-22 7:23 ` Masami Hiramatsu
2020-01-22 12:11 ` Steven Rostedt
2020-01-22 13:12 ` Masami Hiramatsu [this message]
2020-01-22 16:54 ` Paul E. McKenney
2020-01-23 1:33 ` Masami Hiramatsu
2020-01-23 2:26 ` Paul E. McKenney
2020-01-23 6:13 ` Masami Hiramatsu
2020-01-23 10:03 ` Paul E. McKenney
2020-01-16 14:45 ` [RFT PATCH 05/13] tracing/kprobe: Use call_rcu to defer freeing event_file_link Masami Hiramatsu
2020-01-27 15:02 ` kbuild test robot
2020-01-27 15:02 ` [RFC PATCH] tracing/kprobe: trace_kprobe_disabled_finished can be static kbuild test robot
2020-01-28 15:02 ` Masami Hiramatsu
2020-01-16 14:45 ` [RFT PATCH 06/13] kprobes: Enable kprobe-booster with CONFIG_PREEMPT=y Masami Hiramatsu
2020-01-16 14:45 ` [RFT PATCH 07/13] kprobes: Use normal list traversal API if a mutex is held Masami Hiramatsu
2020-01-16 14:45 ` [RFT PATCH 08/13] kprobes: Use workqueue for reclaiming kprobe insn cache pages Masami Hiramatsu
2020-01-16 14:45 ` [RFT PATCH 09/13] kprobes: Free kprobe_insn_page asynchronously Masami Hiramatsu
2020-01-16 14:45 ` [RFT PATCH 10/13] kprobes: Make free_*insn_slot() mutex-less Masami Hiramatsu
2020-01-16 14:46 ` [RFT PATCH 11/13] kprobes: Add asynchronous unregistration APIs Masami Hiramatsu
2020-01-16 14:46 ` [RFT PATCH 12/13] tracing/kprobe: Free probe event asynchronously Masami Hiramatsu
2020-01-16 14:46 ` [RFT PATCH 13/13] tracing/kprobe: perf_event: Remove local kprobe " 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=20200122221240.cef447446785f46862fee97a@kernel.org \
--to=mhiramat@kernel.org \
--cc=acme@kernel.org \
--cc=anil.s.keshavamurthy@intel.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=brendan.d.gregg@gmail.com \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=joel@joelfernandes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=naveen.n.rao@linux.ibm.com \
--cc=paulmck@kernel.org \
--cc=rostedt@goodmis.org \
/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).