Re: [PATCH] tracing/markers: make markers select tracepoints

From: "Frank Ch. Eigler" <fche@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>,
	Randy Dunlap <randy.dunlap@oracle.com>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	linux-kernel@vger.kernel.org, zippel@linux-m68k.org,
	linux-kbuild@vger.kernel.org
Subject: Re: [PATCH] tracing/markers: make markers select tracepoints
Date: Sun, 22 Feb 2009 07:24:25 -0500	[thread overview]
Message-ID: <20090222122424.GA9951@redhat.com> (raw)
In-Reply-To: <1235304876.4632.10.camel@laptop>

Hi -

On Sun, Feb 22, 2009 at 01:14:36PM +0100, Peter Zijlstra wrote:
> > > I would like to never merge an ftrace_printk() user... just as I'd like
> > > to get rid of every marker.
> > 
> > But why?  They solve a problem well enough that Ingo had in effect
> > reinvented them on Friday.
> 
> Because after a printk() debug spree, I don't commit them, I toss them
> out and keep the fix.

Markers solve a problem closer to tracepoints than to debugging
printk's.

In this context, the main difference between tracepoints is that
markers need almost no hand-written glue code of the sort that make up
ftrace engines that just trace simple values.  Simpler & smaller code
for the same output seems like a win.

- FChE