From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751648AbdBBQU4 (ORCPT ); Thu, 2 Feb 2017 11:20:56 -0500 Received: from mx1.redhat.com ([209.132.183.28]:32826 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750777AbdBBQUz (ORCPT ); Thu, 2 Feb 2017 11:20:55 -0500 Date: Thu, 2 Feb 2017 14:20:51 -0200 From: Arnaldo Carvalho de Melo To: Ingo Molnar Cc: Ravi Bangoria , alexis.berlemont@gmail.com, linux-kernel@vger.kernel.org, peterz@infradead.org, mingo@redhat.com, alexander.shishkin@linux.intel.com, mpe@ellerman.id.au, naveen.n.rao@linux.vnet.ibm.com, mhiramat@kernel.org, maddy@linux.vnet.ibm.com Subject: Re: [PATCH 1/5] perf/sdt: Show proper hint Message-ID: <20170202162051.GA2266@redhat.com> References: <20161214000732.1710-1-alexis.berlemont@gmail.com> <20170202111143.14319-1-ravi.bangoria@linux.vnet.ibm.com> <20170202111143.14319-2-ravi.bangoria@linux.vnet.ibm.com> <20170202134048.GB3393@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170202134048.GB3393@gmail.com> X-Url: http://acmel.wordpress.com User-Agent: Mutt/1.5.20 (2009-12-10) X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.25]); Thu, 02 Feb 2017 16:20:55 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Em Thu, Feb 02, 2017 at 02:40:48PM +0100, Ingo Molnar escreveu: > * Ravi Bangoria wrote: > > After patch: > > $ perf record -e sdt_glib:main__after_check > > event syntax error: 'sdt_glib:idle__add' > > \___ unknown tracepoint > > Error: File /sys/kernel/debug/tracing/events/sdt_glib/idle__add ... > > Hint: SDT event has to be probed before recording it. > > So what is the command line option to 'probe' an SDT event? Could you give an > example that would make the above command work? I also think 'has to be probed' looks confusing, it should be, I suggest, "put in place using 'perf probe'", what will case it to really be _probed_ is when it is activated via a tool like 'perf record'. So yes, I think the message _and_ the example in this commit log must show what needs to be done, in detail, to overcome the "problem" mentioned in the message above. And after that its up to someone to do the logical next step: if it needs to be put in place, well, do it, automagically. - Arnaldo