From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756971AbZHGHHW (ORCPT ); Fri, 7 Aug 2009 03:07:22 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755314AbZHGHHV (ORCPT ); Fri, 7 Aug 2009 03:07:21 -0400 Received: from bilbo.ozlabs.org ([203.10.76.25]:50993 "EHLO bilbo.ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754982AbZHGHHV (ORCPT ); Fri, 7 Aug 2009 03:07:21 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19067.53473.239250.278692@cargo.ozlabs.ibm.com> Date: Fri, 7 Aug 2009 16:59:45 +1000 From: Paul Mackerras To: Ingo Molnar CC: Peter Zijlstra , linux-kernel@vger.kernel.org, benh@kernel.crashing.org Subject: [PATCH] perf_counter/powerpc: Avoid oops on cpus without perf_counter hardware support X-Mailer: VM 8.0.12 under 22.2.1 (i486-pc-linux-gnu) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org If we have the powerpc perf_counter backend compiled in, but the cpu we are running on is one where we don't support the PMU, we currently oops in hw_perf_group_sched_in if we try to use any counters, because ppmu is NULL in that case, and we unconditionally dereference ppmu. This fixes the problem by adding a check if ppmu is NULL at the beginning of hw_perf_group_sched_in, and also at the beginning of the other functions that get called from the perf_counter core, i.e. hw_perf_disable, hw_perf_enable, and hw_perf_counter_setup. Signed-off-by: Paul Mackerras --- This also needs to go in 2.6.31. arch/powerpc/kernel/perf_counter.c | 8 ++++++++ 1 files changed, 8 insertions(+), 0 deletions(-) diff --git a/arch/powerpc/kernel/perf_counter.c b/arch/powerpc/kernel/perf_counter.c index 809fdf9..70e1f57 100644 --- a/arch/powerpc/kernel/perf_counter.c +++ b/arch/powerpc/kernel/perf_counter.c @@ -518,6 +518,8 @@ void hw_perf_disable(void) struct cpu_hw_counters *cpuhw; unsigned long flags; + if (!ppmu) + return; local_irq_save(flags); cpuhw = &__get_cpu_var(cpu_hw_counters); @@ -572,6 +574,8 @@ void hw_perf_enable(void) int n_lim; int idx; + if (!ppmu) + return; local_irq_save(flags); cpuhw = &__get_cpu_var(cpu_hw_counters); if (!cpuhw->disabled) { @@ -737,6 +741,8 @@ int hw_perf_group_sched_in(struct perf_counter *group_leader, long i, n, n0; struct perf_counter *sub; + if (!ppmu) + return 0; cpuhw = &__get_cpu_var(cpu_hw_counters); n0 = cpuhw->n_counters; n = collect_events(group_leader, ppmu->n_counter - n0, @@ -1281,6 +1287,8 @@ void hw_perf_counter_setup(int cpu) { struct cpu_hw_counters *cpuhw = &per_cpu(cpu_hw_counters, cpu); + if (!ppmu) + return; memset(cpuhw, 0, sizeof(*cpuhw)); cpuhw->mmcr[0] = MMCR0_FC; } -- 1.6.0.4