From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756705AbaJXNVt (ORCPT ); Fri, 24 Oct 2014 09:21:49 -0400 Received: from mail.kernel.org ([198.145.19.201]:33460 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756655AbaJXNVs (ORCPT ); Fri, 24 Oct 2014 09:21:48 -0400 Date: Fri, 24 Oct 2014 10:21:41 -0300 From: Arnaldo Carvalho de Melo To: Namhyung Kim Cc: Adrian Hunter , Peter Zijlstra , linux-kernel@vger.kernel.org, David Ahern , Frederic Weisbecker , Jiri Olsa , Paul Mackerras , Stephane Eranian Subject: Re: [PATCH 02/16] perf pmu: Let pmu's with no events show up on perf list Message-ID: <20141024132141.GB14687@kernel.org> References: <1414061124-26830-1-git-send-email-adrian.hunter@intel.com> <1414061124-26830-3-git-send-email-adrian.hunter@intel.com> <87y4s6cak7.fsf@sejong.aot.lge.com> <20141024125702.GZ14687@kernel.org> <20141024130320.GA14687@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20141024130320.GA14687@kernel.org> X-Url: http://acmel.wordpress.com User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Em Fri, Oct 24, 2014 at 10:03:20AM -0300, Arnaldo Carvalho de Melo escreveu: > Em Fri, Oct 24, 2014 at 09:57:02AM -0300, Arnaldo Carvalho de Melo escreveu: > > Em Fri, Oct 24, 2014 at 02:15:52PM +0900, Namhyung Kim escreveu: > > > On Thu, 23 Oct 2014 13:45:10 +0300, Adrian Hunter wrote: > > > > + if (pmu->selectable) { > > > > + scnprintf(buf, sizeof(buf), "%s//", pmu->name); > > > > + aliases[j] = strdup(buf); > > > > > > You need to check the return value here (and above too). > > > > Well spotted, fixing this up. > > Oh well, this print_pmu_events() function needs some care, it starts by > trying to alloc the array, if it fails, it silently returns, does that > mean that there are no pmu events? Or that memory allocation failed? > > Ok, will do the fixes in a separate patch... > > - Arnaldo The patch below should check everything and warn the user, even maintaining that void return... Acked-by tags welcome as always :-) - Arnaldo diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 91dca60..881b754 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -753,9 +753,9 @@ void print_pmu_events(const char *event_glob, bool name_only) if (pmu->selectable) len++; } - aliases = malloc(sizeof(char *) * len); + aliases = zalloc(sizeof(char *) * len); if (!aliases) - return; + goto out_enomem; pmu = NULL; j = 0; while ((pmu = perf_pmu__scan(pmu)) != NULL) { @@ -768,16 +768,20 @@ void print_pmu_events(const char *event_glob, bool name_only) (!is_cpu && strglobmatch(alias->name, event_glob)))) continue; - aliases[j] = name; + if (is_cpu && !name_only) - aliases[j] = format_alias_or(buf, sizeof(buf), - pmu, alias); - aliases[j] = strdup(aliases[j]); + name = format_alias_or(buf, sizeof(buf), pmu, alias); + + aliases[j] = strdup(name); + if (aliases[j] == NULL) + goto out_enomem; j++; } if (pmu->selectable) { - scnprintf(buf, sizeof(buf), "%s//", pmu->name); - aliases[j] = strdup(buf); + char *s; + if (asprintf(&s, "%s//", pmu->name) < 0) + goto out_enomem; + aliases[j] = s; j++; } } @@ -789,12 +793,20 @@ void print_pmu_events(const char *event_glob, bool name_only) continue; } printf(" %-50s [Kernel PMU event]\n", aliases[j]); - zfree(&aliases[j]); printed++; } if (printed) printf("\n"); - free(aliases); +out_free: + for (j = 0; j < len; j++) + zfree(&aliases[j]); + zfree(&aliases); + return; + +out_enomem: + printf("FATAL: not enough memory to print PMU events\n"); + if (aliases) + goto out_free; } bool pmu_have_event(const char *pname, const char *name)