From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1753471Ab1G1OIR (ORCPT <rfc822;w@1wt.eu>);
	Thu, 28 Jul 2011 10:08:17 -0400
Received: from mail-wy0-f174.google.com ([74.125.82.174]:34846 "EHLO
	mail-wy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752192Ab1G1OIP (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Thu, 28 Jul 2011 10:08:15 -0400
Date: Thu, 28 Jul 2011 16:08:10 +0200
From: Frederic Weisbecker <fweisbec@gmail.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Vaibhav Nagarnaik <vnagarnaik@google.com>,
        Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@redhat.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Michael Rubin <mrubin@google.com>, David Sharp <dhsharp@google.com>,
        linux-kernel@vger.kernel.org, x86@kernel.org
Subject: Re: [PATCH v2 0/6] Add tracepoints to trace all system IRQs
Message-ID: <20110728140808.GG11820@somewhere.redhat.com>
References: <1310679495-29104-1-git-send-email-vnagarnaik@google.com>
 <1311793038-15155-1-git-send-email-vnagarnaik@google.com>
 <1311855169.2617.168.camel@laptop>
 <20110728123131.GA11820@somewhere.redhat.com>
 <1311862204.2617.179.camel@laptop>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1311862204.2617.179.camel@laptop>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Thu, Jul 28, 2011 at 04:10:04PM +0200, Peter Zijlstra wrote:
> On Thu, 2011-07-28 at 14:31 +0200, Frederic Weisbecker wrote:
> > 
> > Because that requires to put the tracepoints in every interrupt sources
> > for every archs instead of putting one in the generic space. 
> 
> To sum up from IRC, yes that's exactly what you need to do, because this
> patch set assumes two things:
> 
>  1) that there is a 1:1 mapping between interrupt sources and generic
> handlers, and
> 
>  2) that there is a neglectable amount of arch specific code for each
> generic handler.
> 
> Both are false in generic. Therefore there is no option but to
> instrument each and every arch specific interrupt handler.
> 
> Also, please use a single tracepoint for all that instead of this horrid
> forest of random crap. Use something like arch_interrupt_enter(line_nr),
> arch_interrupt_exit(line_nr).

Right, so I guess the first iteration of the patchset was good in fact.