From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756476Ab1KPQlh (ORCPT ); Wed, 16 Nov 2011 11:41:37 -0500 Received: from hrndva-omtalb.mail.rr.com ([71.74.56.124]:62539 "EHLO hrndva-omtalb.mail.rr.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753545Ab1KPQlg (ORCPT ); Wed, 16 Nov 2011 11:41:36 -0500 X-Authority-Analysis: v=2.0 cv=JOm5Qr2b c=1 sm=0 a=ZycB6UtQUfgMyuk2+PxD7w==:17 a=l4wf8OQxcBQA:10 a=5SG0PmZfjMsA:10 a=IkcTkHD0fZMA:10 a=QyXUC8HyAAAA:8 a=oK_g3dvKfHLur5cet_4A:9 a=QEXdDO2ut3YA:10 a=dGJ0OcVc7YAA:10 a=ZycB6UtQUfgMyuk2+PxD7w==:117 X-Cloudmark-Score: 0 X-Originating-IP: 74.67.80.29 Subject: Re: [PATCH] tracing: add trace console From: Steven Rostedt To: Johannes Berg Cc: Christoph Hellwig , LKML , Frederic Weisbecker , Ingo Molnar In-Reply-To: <1321456652.4502.12.camel@jlt3.sipsolutions.net> References: <1321438728.4773.16.camel@jlt3.sipsolutions.net> <20111116151037.GA26166@infradead.org> <1321456652.4502.12.camel@jlt3.sipsolutions.net> Content-Type: text/plain; charset="UTF-8" Date: Wed, 16 Nov 2011 11:41:33 -0500 Message-ID: <1321461693.4181.26.camel@frodo> Mime-Version: 1.0 X-Mailer: Evolution 2.32.3 (2.32.3-1.fc14) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2011-11-16 at 16:17 +0100, Johannes Berg wrote: > On Wed, 2011-11-16 at 10:10 -0500, Christoph Hellwig wrote: > > On Wed, Nov 16, 2011 at 11:18:48AM +0100, Johannes Berg wrote: > > > From: Johannes Berg > > > > > > As described in the Kconfig entry, logging printk > > > output is useful to correlate (existing) printk > > > debugging with (existing) tracing. The easiest way > > > to achieve this is to register a console that just > > > calls trace_printk(), which this module does. > > > > I'd much prefer if we could have it built in all the time, but had a > > way to enable it at runtime, just like all the "real" tracepoints. > > That's possible, basically adding a tracepoints to printk -- that would > even catch everything and not require setting the console level (which > has its advantages too since other consoles might be slow). > > It's a completely different thing though. I did it this way because it > was trivial to do out of tree for a quick test I was doing yesterday. > Doing it the other way is obviously more intrusive in core code. I'm not sure it would be too invasive. Think you could get something working? -- Steve