Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755283AbaGCDf1 (ORCPT ); Wed, 2 Jul 2014 23:35:27 -0400 Received: from mga02.intel.com ([134.134.136.20]:18362 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755158AbaGCDfZ (ORCPT ); Wed, 2 Jul 2014 23:35:25 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.01,592,1400050800"; d="scan'208";a="567622749" From: Chen LinX To: a.p.zijlstra@chello.nl, paulus@samba.org, mingo@redhat.com, acme@ghostprotocols.net, linux-kernel@vger.kernel.org Cc: yanmin_zhang@linux.intel.com, "Chen LinX" , Yanmin Zhang Subject: [PATCH] perf: Don't enable the perf_event without in PERF_ATTACH_CONTEXT status Date: Thu, 3 Jul 2014 11:36:38 +0800 Message-Id: <1404358598-19186-1-git-send-email-linx.z.chen@intel.com> X-Mailer: git-send-email 1.9.1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: "Chen LinX" when do cpu hotplug test and run below perf test together, pmu may access freed perf_event while true; do perf record -a -g -f sleep 10 rm perf.* done the scenario is that when cpu offline firstly, the 'perf_cpu_notify' will disable event on the pmu and remove it from the context list. after cpu online, the perf app may enable the event that without linked in context list again. when cpu offine the second time, the 'perf_cpu_notify' can't disable event on the pmu as the event doesn't link to context list. the perf app may free this event later(the free procedure try to disable event on the pmu but as the cpu is offline, the 'cpu_function_call(event->cpu, __perf_remove_from_context, event)' is failed). then after cpu online again, pmu will access freed perf_event and hit panic. so adding PERF_ATTACH_CONTEXT flag check before enable event to avoid this scenario. [ 157.666035 ] BUG: unable to handle kernel paging request at ffffffe89af56004 [ 157.666166 ] IP: [] do_raw_spin_lock+0xc/0x130 [ 157.668086 ] Call Trace: [ 157.668122 ] [ 157.668156 ] [] _raw_spin_lock_irqsave+0x2a/0x40 [ 157.668268 ] [] __intel_shared_reg_get_constraints.isra.9+0x70/0x150 [ 157.668350 ] [] intel_get_event_constraints+0x8e/0x150 [ 157.668424 ] [] x86_schedule_events+0x81/0x200 [ 157.668495 ] [] ? intel_get_event_constraints+0xda/0x150 [ 157.668568 ] [] ? x86_schedule_events+0x81/0x200 [ 157.668640 ] [] ? flat_send_IPI_mask+0x88/0xa0 [ 157.668710 ] [] ? __enqueue_entity+0x78/0x80 [ 157.668777 ] [] ? enqueue_task_fair+0x90a/0xdd0 [ 157.668848 ] [] ? tracer_tracing_is_on+0x15/0x30 [ 157.668918 ] [] ? cpuacct_charge+0x61/0x70 [ 157.668984 ] [] ? flat_send_IPI_mask+0x88/0xa0 [ 157.669052 ] [] ? native_smp_send_reschedule+0x45/0x60 [ 157.669126 ] [] ? resched_task+0x69/0x70 [ 157.669192 ] [] ? tracer_tracing_is_on+0x15/0x30 [ 157.669262 ] [] ? perf_pmu_enable+0x13/0x30 [ 157.669328 ] [] ? x86_pmu_add+0xaf/0x150 [ 157.669393 ] [] x86_pmu_commit_txn+0x50/0xa0 [ 157.669462 ] [] ? native_sched_clock+0x24/0x80 [ 157.669531 ] [] ? native_sched_clock+0x24/0x80 [ 157.669598 ] [] ? sched_clock_cpu+0xbd/0x110 [ 157.669664 ] [] ? local_clock+0x3f/0x50 [ 157.669729 ] [] ? perf_event_update_userpage+0xe4/0x150 [ 157.669802 ] [] ? event_sched_in.isra.72+0x81/0x190 [ 157.669871 ] [] group_sched_in+0x19a/0x1e0 [ 157.669937 ] [] ? native_sched_clock+0x24/0x80 [ 157.670006 ] [] ctx_sched_in+0x1b8/0x1e0 [ 157.670071 ] [] perf_event_sched_in+0x22/0x80 [ 157.670138 ] [] __perf_install_in_context+0xff/0x170 [ 157.670212 ] [] remote_function+0x4b/0x60 [ 157.670282 ] [] generic_smp_call_function_single_interrupt+0x9d/0x120 [ 157.670363 ] [] ? __const_udelay+0x29/0x30 [ 157.670429 ] [] smp_call_function_single_interrupt+0x27/0x40 [ 157.670504 ] [] call_function_single_interrupt+0x6f/0x80 Change-Id: I7265d83159b9180e9be3a370ba50e067385547bd Signed-off-by: Yanmin Zhang Signed-off-by: Chen LinX --- kernel/events/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index e76e495..30f0095 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1783,7 +1783,7 @@ static int __perf_event_enable(void *info) * where the task could be killed and 'ctx' deactivated * by perf_event_exit_task. */ - if (!ctx->is_active) + if (!ctx->is_active || !(event->attach_state & PERF_ATTACH_CONTEXT)) return -EINVAL; raw_spin_lock(&ctx->lock); -- 1.7.9.5 -- 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/