From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8F08CC433F5 for ; Mon, 3 Jan 2022 15:15:59 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233440AbiACPP5 (ORCPT ); Mon, 3 Jan 2022 10:15:57 -0500 Received: from ams.source.kernel.org ([145.40.68.75]:38616 "EHLO ams.source.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229995AbiACPPw (ORCPT ); Mon, 3 Jan 2022 10:15:52 -0500 Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ams.source.kernel.org (Postfix) with ESMTPS id 55623B80EFC; Mon, 3 Jan 2022 15:15:51 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id A17F0C36AED; Mon, 3 Jan 2022 15:15:49 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1641222950; bh=JlsPW+XrEU2GAaGgYw+gRc7fXysdOHvhf4ucIuTXWwA=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=j+TKAu/VWbwINQiN+cy4dJvGI4Pp93ZQQ9Xw8Q0568LoeqlZQ9HMtxajwxbzg7Q7N rFWdSJqWRFOu04e8JZGN3suKf7UCYUtKJfAq0/mCZyp0Fj5bGT2xIfEv9m2u0zLfuC bojiet0Jg7yx6/5cngumnJHmKjsgMvQK1WFGnccMMcbvGWt/ejIpqbNz1xkff8fVyN KDF8tv55FQgwdSmvShKUAU59yPRuFWt/4cod+CPoiBlYAPowu3xNF3TUj+mBBNim7m ytP6oO8zJB62z1UZP9R7wWLP3dmhMZElhYqOqmgm112d4xWXhUOTl1JGf2uGJ46B7c ViYJuDJq/xxAw== Received: by quaco.ghostprotocols.net (Postfix, from userid 1000) id 0889840B92; Mon, 3 Jan 2022 12:15:47 -0300 (-03) Date: Mon, 3 Jan 2022 12:15:46 -0300 From: Arnaldo Carvalho de Melo To: Jiri Slaby Cc: Ian Rogers , Peter Zijlstra , Ingo Molnar , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, linux-trace-devel@vger.kernel.org, Tzvetomir Stoyanov , Steven Rostedt , "torvalds@linux-foundation.org" , eranian@google.com Subject: Re: [was: libtraceevent: Increase libtraceevent logging when verbose] Message-ID: References: <20210923001024.550263-1-irogers@google.com> <20210923001024.550263-4-irogers@google.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: X-Url: http://acmel.wordpress.com Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Em Mon, Jan 03, 2022 at 09:25:26AM +0100, Jiri Slaby escreveu: > Ping -- perf build is still broken in 5.16-rc8 due to this. > > On 16. 12. 21, 9:07, Jiri Slaby wrote: > > Hi, > > > > On 23. 09. 21, 2:10, Ian Rogers wrote: > > > libtraceevent has added more levels of debug printout and with changes > > > like: > > > https://lore.kernel.org/linux-trace-devel/20210507095022.1079364-3-tz.stoyanov@gmail.com > > > > > > previously generated output like "registering plugin" is no longer > > > displayed. This change makes it so that if perf's verbose debug output > > > is enabled then the debug and info libtraceevent messages can be > > > displayed. > > > The code is conditionally enabled based on the libtraceevent version as > > > discussed in the RFC: > > > https://lore.kernel.org/lkml/20210610060643.595673-1-irogers@google.com/ > > > > > > v2. Is a rebase and handles the case of building without > > >      LIBTRACEEVENT_DYNAMIC. > > > > It breaks build with LIBTRACEEVENT_DYNAMIC and version 1.3.0: > > > util/debug.c: In function ‘perf_debug_option’: > > > util/debug.c:243:17: error: implicit declaration of function > > ‘tep_set_loglevel’ [-Werror=implicit-function-declaration] > > >   243 |                 tep_set_loglevel(TEP_LOG_INFO); > > >       |                 ^~~~~~~~~~~~~~~~ > > > util/debug.c:243:34: error: ‘TEP_LOG_INFO’ undeclared (first use in > > this function); did you mean ‘TEP_PRINT_INFO’? > > >   243 |                 tep_set_loglevel(TEP_LOG_INFO); > > >       |                                  ^~~~~~~~~~~~ > > >       |                                  TEP_PRINT_INFO > > > util/debug.c:243:34: note: each undeclared identifier is reported > > only once for each function it appears in > > > util/debug.c:245:34: error: ‘TEP_LOG_DEBUG’ undeclared (first use in > > this function) > > >   245 |                 tep_set_loglevel(TEP_LOG_DEBUG); > > >       |                                  ^~~~~~~~~~~~~ > > > util/debug.c:247:34: error: ‘TEP_LOG_ALL’ undeclared (first use in > > this function) > > >   247 |                 tep_set_loglevel(TEP_LOG_ALL); > > >       |                                  ^~~~~~~~~~~ > > > > It is because the gcc's command line looks like: > > gcc > > ... > > -I/home/abuild/rpmbuild/BUILD/tools/lib/ > > ... > > -DLIBTRACEEVENT_VERSION=65790 > > ... > > > > So: > > > > > > > --- a/tools/perf/util/debug.c > > > +++ b/tools/perf/util/debug.c > > > @@ -24,6 +24,16 @@ > > >   #include "util/parse-sublevel-options.h" > > >   #include > > > +#include > > > > /home/abuild/rpmbuild/BUILD/tools/lib/traceevent/event-parse.h is taken > > here. > > > > > @@ -228,6 +238,15 @@ int perf_debug_option(const char *str) > > >       /* Allow only verbose value in range (0, 10), otherwise set 0. */ > > >       verbose = (verbose < 0) || (verbose > 10) ? 0 : verbose; > > > +#if MAKE_LIBTRACEEVENT_VERSION(1, 3, 0) <= LIBTRACEEVENT_VERSION > > > > But > > -DLIBTRACEEVENT_VERSION=65790 > > is taken here (which is 1.3.0). > > > > > +    if (verbose == 1) > > > +        tep_set_loglevel(TEP_LOG_INFO); > > > +    else if (verbose == 2) > > > +        tep_set_loglevel(TEP_LOG_DEBUG); > > > +    else if (verbose >= 3) > > > +        tep_set_loglevel(TEP_LOG_ALL); > > > +#endif > > > + > > >       return 0; > > >   } Tried reproducing here on fedora 34 but it is at 1.1.1: make: Leaving directory '/var/home/acme/git/perf/tools/perf' Performance counter stats for 'make -k LIBTRACEEVENT_DYNAMIC=1 BUILD_BPF_SKEL=1 CORESIGHT=1 PYTHON=python3 O=/tmp/build/perf -C tools/perf install-bin': 397,176,418,790 cycles:u 493,885,896,752 instructions:u # 1.24 insn per cycle 8.664957770 seconds time elapsed 97.706894000 seconds user 23.714293000 seconds sys On branch perf/urgent Your branch is ahead of 'quaco/perf/urgent' by 369 commits. (use "git push" to publish your local commits) nothing to commit, working tree clean 19: 'import perf' in python : Ok ⬢[acme@toolbox perf]$ git diff ⬢[acme@toolbox perf]$ rpm -q libtraceevent-devel libtraceevent-devel-1.1.1-2.fc34.x86_64 Lemme see in a container so that I can also test the fix. - Arnaldo