Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755200Ab0H0SHf (ORCPT ); Fri, 27 Aug 2010 14:07:35 -0400 Received: from cam-admin0.cambridge.arm.com ([217.140.96.50]:45850 "EHLO cam-admin0.cambridge.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755191Ab0H0SHc (ORCPT ); Fri, 27 Aug 2010 14:07:32 -0400 Subject: Re: [PATCH 1/4] oprofile: Handle initialisation failure more gracefully From: Will Deacon To: Robert Richter Cc: Matt Fleming , linux-kernel@vger.kernel.org, Paul Mundt , Russell King , linux-arm-kernel@lists.infradead.org, linux-sh@vger.kernel.org, Peter Zijlstra , Ingo Molnar , Frederic Weisbecker , Arnaldo Carvalho de Melo , linux-arch@vger.kernel.org In-Reply-To: <20100827163839.GR22783@erda.amd.com> References: <442a16796990290ca3ebaaa3d0ab317e7b0a30a5.1282848651.git.matt@console-pimps.org> <20100827124344.GJ22783@erda.amd.com> <1282922125.22128.5.camel@e102144-lin.cambridge.arm.com> <20100827163839.GR22783@erda.amd.com> Content-Type: text/plain; charset="UTF-8" Date: Fri, 27 Aug 2010 19:06:45 +0100 Message-ID: <1282932405.2899.22.camel@e102144-lin.cambridge.arm.com> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 27 Aug 2010 18:06:48.0785 (UTC) FILETIME=[9E357410:01CB4612] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5563 Lines: 167 Hi Robert, Thanks for taking the time to look at this. On Fri, 2010-08-27 at 17:38 +0100, Robert Richter wrote: > On 27.08.10 11:15:25, Will Deacon wrote: > > How about something like this? This removes the exit call > > from the init code and sets pointers to NULL after they have been > > freed. We still have to do some checking so that we don't try to > > release a NULL perf event: > > > > diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c > > index 0691176..12253eb 100644 > > --- a/arch/arm/oprofile/common.c > > +++ b/arch/arm/oprofile/common.c > > @@ -275,10 +275,12 @@ out: > > return ret; > > } > > > > -static void exit_driverfs(void) > > +static void __exit exit_driverfs(void) > > { > > - platform_device_unregister(oprofile_pdev); > > - platform_driver_unregister(&oprofile_driver); > > + if (!IS_ERR_OR_NULL(oprofile_pdev)) { > > This check is obsolete here as we do not call oprofile_arch_exit() on > failure. > Sorry, I was being braindead and thought oprofile_arch_exit might be called via oprofile_exit. As it happens, this is only called because of module_exit so we know that initialisation will have completed. In light of this and your comments, I've simplified the code. Once people are happy with it, I'll post it as a couple of patches: diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c index 0691176..c2c4a2e 100644 --- a/arch/arm/oprofile/common.c +++ b/arch/arm/oprofile/common.c @@ -275,7 +275,7 @@ out: return ret; } -static void exit_driverfs(void) +static void __exit exit_driverfs(void) { platform_device_unregister(oprofile_pdev); platform_driver_unregister(&oprofile_driver); @@ -359,14 +359,13 @@ int __init oprofile_arch_init(struct oprofile_operations *ops) if (!counter_config) { pr_info("oprofile: failed to allocate %d " "counters\n", perf_num_counters); - return -ENOMEM; + ret = -ENOMEM; + goto out; } ret = init_driverfs(); - if (ret) { - kfree(counter_config); - return ret; - } + if (ret) + goto out; for_each_possible_cpu(cpu) { perf_events[cpu] = kcalloc(perf_num_counters, @@ -374,9 +373,8 @@ int __init oprofile_arch_init(struct oprofile_operations *ops) if (!perf_events[cpu]) { pr_info("oprofile: failed to allocate %d perf events " "for cpu %d\n", perf_num_counters, cpu); - while (--cpu >= 0) - kfree(perf_events[cpu]); - return -ENOMEM; + ret = -ENOMEM; + goto out; } } @@ -393,28 +391,33 @@ int __init oprofile_arch_init(struct oprofile_operations *ops) else pr_info("oprofile: using %s\n", ops->cpu_type); +out: + if (ret) { + kfree(counter_config); + for_each_possible_cpu(cpu) + kfree(perf_events[cpu]); + } + return ret; } -void oprofile_arch_exit(void) +void __exit oprofile_arch_exit(void) { int cpu, id; struct perf_event *event; - if (*perf_events) { - exit_driverfs(); - for_each_possible_cpu(cpu) { - for (id = 0; id < perf_num_counters; ++id) { - event = perf_events[cpu][id]; - if (event != NULL) - perf_event_release_kernel(event); - } - kfree(perf_events[cpu]); + for_each_possible_cpu(cpu) { + for (id = 0; id < perf_num_counters; ++id) { + event = perf_events[cpu][id]; + if (event) + perf_event_release_kernel(event); } + + kfree(perf_events[cpu]); } - if (counter_config) - kfree(counter_config); + kfree(counter_config); + exit_driverfs(); } #else int __init oprofile_arch_init(struct oprofile_operations *ops) @@ -422,5 +425,5 @@ int __init oprofile_arch_init(struct oprofile_operations *ops) pr_info("oprofile: hardware counters not available\n"); return -ENODEV; } -void oprofile_arch_exit(void) {} +void __exit oprofile_arch_exit(void) {} #endif /* CONFIG_HW_PERF_EVENTS */ diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c index b336cd9..b4a6857 100644 --- a/drivers/oprofile/oprof.c +++ b/drivers/oprofile/oprof.c @@ -257,16 +257,9 @@ static int __init oprofile_init(void) printk(KERN_INFO "oprofile: using timer interrupt.\n"); err = oprofile_timer_init(&oprofile_ops); if (err) - goto out_arch; + return err; } - err = oprofilefs_register(); - if (err) - goto out_arch; - return 0; - -out_arch: - oprofile_arch_exit(); - return err; + return oprofilefs_register(); } Once again, thanks for looking at this and have a good weekend, Will -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/