public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: "Arjan van de Ven" <arjan@linux.intel.com>,
	"Linux Kernel Mailing List" <linux-kernel@vger.kernel.org>,
	"Ingo Molnar" <mingo@elte.hu>,
	"Kok, Auke-jan H" <auke-jan.h.kok@intel.com>,
	"Frédéric Weisbecker" <fweisbec@gmail.com>
Subject: Re: [PATCH] sched: Provide iowait counters
Date: Mon, 20 Jul 2009 22:38:36 +0200	[thread overview]
Message-ID: <1248122316.23509.48.camel@laptop> (raw)
In-Reply-To: <alpine.DEB.2.00.0907201622140.26208@gandalf.stny.rr.com>

On Mon, 2009-07-20 at 16:26 -0400, Steven Rostedt wrote:
> > +#define _TRACE_PROFILE(call, proto, args, assign)                    \
> >  static void ftrace_profile_##call(proto)                             \
> >  {                                                                    \
> > -     extern void perf_tpcounter_event(int);                          \
> > -     perf_tpcounter_event(event_##call.id);                          \
> > +     extern void perf_tpcounter_event(int, u64, u64);                \
> > +     u64 addr = 0, count = 1;                                        \
> > +     { assign; }                                                     \
> > +     perf_tpcounter_event(event_##call.id, addr, count);             \
> 
> The problem here is that the assign also will do the :
> 
>        TP_fast_assign(
>                __entry->pid    = p->pid;
>                __entry->time   = time;
> 
> part, thus you will probably get errors in processing the __entry part.

Oh, right, no macros wrapping that..

> What about doing instead:
> 
>  TRACE_EVENT(sched_iowait,
>  
>         TP_PROTO(struct task_struct *p, u64 time),
>  
>         TP_ARGS(p, time),
>  
>         TP_STRUCT__entry(
>                 __field(pid_t, pid )
>                 __field(u64,   time)
>         ),
>  
>         TP_fast_assign(
>                 __entry->pid    = p->pid;
>                 __entry->time   = time;
>  
>         ) __perf_count(time),
> 
>         TP_printk("task %d waited for IO for %Lu ns",
>                 __entry->pid, __entry->time)
>  );
>  
> Then we could simply do:
> 
> #undef __perf_count
> #define __perf_count(a)
> 
> [ do all the ftrace event work ]
> 
> #undef TP_fast_assign
> #define TP_fast_assign(a...)
> 
> #undef __perf_count
> #define __perf_count(c) count = (c);

Hmm, should we maybe do something like:

	TP_fast_assign(
		...
	)
	TP_perf_assign(
		addr = 0;
		count = time;
	),

So we can extend it over time?

But otherwise, yes, looks nice. Lets make it happen ;-)


  reply	other threads:[~2009-07-20 20:38 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-07-20 18:31 [PATCH] sched: Provide iowait counters Arjan van de Ven
2009-07-20 19:16 ` Peter Zijlstra
2009-07-20 19:31   ` Arjan van de Ven
2009-07-20 19:42   ` Steven Rostedt
2009-07-20 20:11     ` Peter Zijlstra
2009-07-20 20:26       ` Steven Rostedt
2009-07-20 20:38         ` Peter Zijlstra [this message]
2009-07-20 21:03           ` Steven Rostedt
2009-07-25  4:22 ` Andrew Morton
2009-07-25  4:33   ` Arjan van de Ven
2009-07-25  4:40     ` Andrew Morton
2009-07-25  4:48       ` Arjan van de Ven
2009-07-25  5:04         ` Andrew Morton
2009-07-25  6:05           ` Peter Zijlstra
2009-07-25  7:21             ` Andrew Morton
2009-07-25 16:42               ` Arjan van de Ven
2009-07-25 17:41                 ` Peter Zijlstra
2009-07-25 17:56                   ` Arjan van de Ven
2009-07-25 18:25                   ` Arjan van de Ven
2009-08-03 13:21 ` [tip:sched/core] " tip-bot for Arjan van de Ven
2009-09-02  7:00 ` tip-bot for Arjan van de Ven

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=1248122316.23509.48.camel@laptop \
    --to=peterz@infradead.org \
    --cc=arjan@linux.intel.com \
    --cc=auke-jan.h.kok@intel.com \
    --cc=fweisbec@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --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