From: Randy Dunlap <randy.dunlap@oracle.com>
To: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Ingo Molnar <mingo@elte.hu>, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/3] tracing: Reduce overhead of module tracepoints
Date: Wed, 24 Mar 2010 16:41:31 -0700 [thread overview]
Message-ID: <4BAAA32B.3050405@oracle.com> (raw)
In-Reply-To: <4BA97FA7.6040406@cn.fujitsu.com>
On 03/23/10 19:57, Li Zefan wrote:
> Remove the @refcnt argument, because it has side-effects, and
> arguments with side-effects are not skipped by the jump over
> disabled instrumentation, so it adds overhead even when the
> tracepoints are disabled.
>
> Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Tested-by: Randy Dunlap <randy.dunlap@oracle.com>
Thanks.
> ---
> include/linux/module.h | 6 ++----
> include/trace/events/module.h | 14 +++++++-------
> kernel/module.c | 3 +--
> 3 files changed, 10 insertions(+), 13 deletions(-)
>
> diff --git a/include/linux/module.h b/include/linux/module.h
> index 5e869ff..393ec39 100644
> --- a/include/linux/module.h
> +++ b/include/linux/module.h
> @@ -460,8 +460,7 @@ static inline void __module_get(struct module *module)
> if (module) {
> preempt_disable();
> __this_cpu_inc(module->refptr->count);
> - trace_module_get(module, _THIS_IP_,
> - __this_cpu_read(module->refptr->count));
> + trace_module_get(module, _THIS_IP_);
> preempt_enable();
> }
> }
> @@ -475,8 +474,7 @@ static inline int try_module_get(struct module *module)
>
> if (likely(module_is_live(module))) {
> __this_cpu_inc(module->refptr->count);
> - trace_module_get(module, _THIS_IP_,
> - __this_cpu_read(module->refptr->count));
> + trace_module_get(module, _THIS_IP_);
> }
> else
> ret = 0;
> diff --git a/include/trace/events/module.h b/include/trace/events/module.h
> index 4b0f48b..a585f81 100644
> --- a/include/trace/events/module.h
> +++ b/include/trace/events/module.h
> @@ -53,9 +53,9 @@ TRACE_EVENT(module_free,
>
> DECLARE_EVENT_CLASS(module_refcnt,
>
> - TP_PROTO(struct module *mod, unsigned long ip, int refcnt),
> + TP_PROTO(struct module *mod, unsigned long ip),
>
> - TP_ARGS(mod, ip, refcnt),
> + TP_ARGS(mod, ip),
>
> TP_STRUCT__entry(
> __field( unsigned long, ip )
> @@ -65,7 +65,7 @@ DECLARE_EVENT_CLASS(module_refcnt,
>
> TP_fast_assign(
> __entry->ip = ip;
> - __entry->refcnt = refcnt;
> + __entry->refcnt = __this_cpu_read(mod->refptr->count);
> __assign_str(name, mod->name);
> ),
>
> @@ -75,16 +75,16 @@ DECLARE_EVENT_CLASS(module_refcnt,
>
> DEFINE_EVENT(module_refcnt, module_get,
>
> - TP_PROTO(struct module *mod, unsigned long ip, int refcnt),
> + TP_PROTO(struct module *mod, unsigned long ip),
>
> - TP_ARGS(mod, ip, refcnt)
> + TP_ARGS(mod, ip)
> );
>
> DEFINE_EVENT(module_refcnt, module_put,
>
> - TP_PROTO(struct module *mod, unsigned long ip, int refcnt),
> + TP_PROTO(struct module *mod, unsigned long ip),
>
> - TP_ARGS(mod, ip, refcnt)
> + TP_ARGS(mod, ip)
> );
>
> TRACE_EVENT(module_request,
> diff --git a/kernel/module.c b/kernel/module.c
> index c968d36..21591ad 100644
> --- a/kernel/module.c
> +++ b/kernel/module.c
> @@ -800,8 +800,7 @@ void module_put(struct module *module)
> preempt_disable();
> __this_cpu_dec(module->refptr->count);
>
> - trace_module_put(module, _RET_IP_,
> - __this_cpu_read(module->refptr->count));
> + trace_module_put(module, _RET_IP_);
> /* Maybe they're waiting for us to drop reference? */
> if (unlikely(!module_is_live(module)))
> wake_up_process(module->waiter);
--
~Randy
next prev parent reply other threads:[~2010-03-24 23:42 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-18 23:26 2.6.33 GP fault only when built with tracing Randy Dunlap
2010-03-18 23:55 ` Steven Rostedt
2010-03-19 0:08 ` Randy Dunlap
2010-03-19 0:59 ` Mathieu Desnoyers
2010-03-19 18:22 ` Randy Dunlap
2010-03-19 18:46 ` Mathieu Desnoyers
2010-03-23 15:26 ` Randy Dunlap
2010-03-24 1:20 ` Mathieu Desnoyers
2010-03-24 1:42 ` Li Zefan
2010-03-24 20:21 ` Randy Dunlap
2010-03-24 20:31 ` Steven Rostedt
2010-03-20 0:12 ` Randy Dunlap
2010-03-19 0:01 ` Frederic Weisbecker
2010-03-19 2:15 ` Steven Rostedt
2010-03-19 16:10 ` Randy Dunlap
2010-03-24 2:57 ` [PATCH 1/3] tracing: Reduce overhead of module tracepoints Li Zefan
2010-03-24 2:58 ` [PATCH 2/3] tracing: Convert some signal events to DEFINE_TRACE Li Zefan
2010-03-24 3:07 ` Steven Rostedt
2010-03-24 3:17 ` Masami Hiramatsu
2010-04-02 19:03 ` [tip:tracing/core] " tip-bot for Li Zefan
2010-03-24 2:58 ` [PATCH 3/3] tracing: Update comments Li Zefan
2010-03-24 3:07 ` Steven Rostedt
2010-04-02 19:04 ` [tip:tracing/core] " tip-bot for Li Zefan
2010-03-24 3:05 ` [PATCH 1/3] tracing: Reduce overhead of module tracepoints Steven Rostedt
2010-03-24 10:24 ` Mathieu Desnoyers
2010-03-24 23:41 ` Randy Dunlap [this message]
2010-03-27 2:03 ` [tip:tracing/urgent] tracing: Remove side effect from module tracepoints that caused a GPF tip-bot for Li Zefan
2010-03-27 4:10 ` Mathieu Desnoyers
2010-03-27 4:23 ` Steven Rostedt
2010-04-02 19:04 ` [tip:tracing/core] " tip-bot for Li Zefan
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=4BAAA32B.3050405@oracle.com \
--to=randy.dunlap@oracle.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lizf@cn.fujitsu.com \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@elte.hu \
--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