From: Steven Rostedt <rostedt@goodmis.org>
To: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>, LKML <linux-kernel@vger.kernel.org>,
Li Zefan <lizf@cn.fujitsu.com>,
Lai Jiangshan <laijs@cn.fujitsu.com>
Subject: Re: [RFC PATCH 02/10] ftrace: Ensure tracing has really stopped before leaving unregister_ftrace_graph
Date: Thu, 21 Jan 2010 20:51:48 -0500 [thread overview]
Message-ID: <1264125108.31321.304.camel@gandalf.stny.rr.com> (raw)
In-Reply-To: <1264122982-1553-3-git-send-regression-fweisbec@gmail.com>
On Fri, 2010-01-22 at 02:16 +0100, Frederic Weisbecker wrote:
> When we run under dynamic tracing, we know that after calling
> unregister_ftrace_graph(), tracing has really stopped because of
> the hot patching and use of stop_machine().
This is incorrect. Even after unregister_ftrace_graph() with
stop_machine(), we still have no guarantee that a call back is not being
called. This is the reason I use sub tracing instead of NULLs. The call
to the trace function could have been loaded in a register and then
preempted. Even after stop_machine() that trace function can be called.
This is also the reason that I never let modules add hooks to the
function tracer (although I can easily make a wrapper to do so).
>
> But in static tracing case, we only set stub callbacks. This is
> not sufficient on archs that have weak memory ordering to assume
> the older callbacks won't be called right after we leave
> unregister_ftrace_graph().
>
> Insert a read/write memory barrier in the end of
> unregister_ftrace_graph() so that the code that follow can safely
> assume tracing has really stopped. This can avoid its older tracing
> callbacks to perform checks about various states like ensuring
> needed buffers have been allocated, etc...
There's no guarantee, even with a smp_mb() that a trace function will
not be called after being unregistered.
-- Steve
next prev parent reply other threads:[~2010-01-22 1:51 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-22 1:16 [RFC PATCH 00/10] Ftrace functions hashlist Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 01/10] ftrace: Generalize the function hashlist from function profiler Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 02/10] ftrace: Ensure tracing has really stopped before leaving unregister_ftrace_graph Frederic Weisbecker
2010-01-22 1:51 ` Steven Rostedt [this message]
2010-01-22 2:04 ` Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 03/10] ftrace: Drop the ftrace_profile_enabled checks in tracing hot path Frederic Weisbecker
2010-01-22 2:05 ` Steven Rostedt
2010-01-22 2:28 ` Mathieu Desnoyers
2010-01-22 3:12 ` Steven Rostedt
2010-01-22 4:09 ` Mathieu Desnoyers
2010-01-22 4:52 ` Steven Rostedt
2010-01-22 12:34 ` Mathieu Desnoyers
2010-01-22 14:54 ` Masami Hiramatsu
2010-01-25 20:58 ` Frederic Weisbecker
2010-01-25 22:14 ` Masami Hiramatsu
2010-01-26 0:41 ` Frederic Weisbecker
2010-01-26 1:13 ` Mathieu Desnoyers
2010-01-26 1:37 ` Masami Hiramatsu
2010-01-27 21:55 ` [PATCH tracing/kprobes] kprobes: Disable booster when CONFIG_PREEMPT=y Masami Hiramatsu
2010-01-28 1:08 ` Mathieu Desnoyers
2010-01-28 4:21 ` Ananth N Mavinakayanahalli
2010-01-29 9:21 ` Ingo Molnar
2010-01-29 11:30 ` Peter Zijlstra
2010-01-29 14:52 ` Masami Hiramatsu
2010-01-29 17:08 ` Mathieu Desnoyers
2010-01-29 17:15 ` Peter Zijlstra
2010-01-29 17:27 ` Mathieu Desnoyers
2010-01-29 17:32 ` Masami Hiramatsu
2010-01-22 2:43 ` [RFC PATCH 03/10] ftrace: Drop the ftrace_profile_enabled checks in tracing hot path Frederic Weisbecker
2010-01-22 3:05 ` Steven Rostedt
2010-01-25 20:52 ` Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 04/10] ftrace: Ensure buffers are visibles to tracing callbacks right away Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 05/10] ftrace: Drop buffer check in function profiler callbacks Frederic Weisbecker
2010-01-25 6:17 ` Lai Jiangshan
2010-01-30 20:47 ` Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 06/10] ftrace: Release the function hlist if we don't need it anymore Frederic Weisbecker
2010-01-25 6:41 ` Lai Jiangshan
2010-01-30 21:14 ` Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 07/10] ftrace: Make the function hashlist concurrently usable Frederic Weisbecker
2010-01-22 1:16 ` [PATCH 08/10] tracing: Simplify test for function_graph tracing Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 09/10] tracing: Use the hashlist for graph function Frederic Weisbecker
2010-01-25 8:50 ` Lai Jiangshan
2010-01-30 21:19 ` Frederic Weisbecker
2010-01-22 1:16 ` [RFC PATCH 10/10] ftrace: Factorize search and insertion in the function hashlist Frederic Weisbecker
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=1264125108.31321.304.camel@gandalf.stny.rr.com \
--to=rostedt@goodmis.org \
--cc=fweisbec@gmail.com \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lizf@cn.fujitsu.com \
--cc=mingo@elte.hu \
/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
all inboxes | Powered by JetHome®