From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753117Ab3BRStE (ORCPT ); Mon, 18 Feb 2013 13:49:04 -0500 Received: from mga14.intel.com ([143.182.124.37]:8652 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755072Ab3BRSs1 (ORCPT ); Mon, 18 Feb 2013 13:48:27 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.84,689,1355126400"; d="scan'208";a="258471108" From: Andi Kleen To: mingo@kernel.org Cc: linux-kernel@vger.kernel.org, Andi Kleen Subject: [PATCH 5/5] perf, x86: Move NMI clearing to end of PMI handler after the counter registers are reset Date: Mon, 18 Feb 2013 10:48:07 -0800 Message-Id: <1361213287-7636-6-git-send-email-andi@firstfloor.org> X-Mailer: git-send-email 1.7.7.6 In-Reply-To: <1361213287-7636-1-git-send-email-andi@firstfloor.org> References: <1361213287-7636-1-git-send-email-andi@firstfloor.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Andi Kleen This avoids some problems with spurious PMIs on Haswell. Haswell seems to behave more like P4 in this regard. Do the same thing as the P4 perf handler by unmasking the NMI only at the end. Shouldn't make any difference for earlier family 6 cores. Tested on Haswell, IvyBridge, Westmere, Saltwell (Atom) Signed-off-by: Andi Kleen --- arch/x86/kernel/cpu/perf_event_intel.c | 16 ++++++---------- 1 files changed, 6 insertions(+), 10 deletions(-) diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index f4d3377..23e1dcb 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -1122,16 +1122,6 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) cpuc = &__get_cpu_var(cpu_hw_events); - /* - * Some chipsets need to unmask the LVTPC in a particular spot - * inside the nmi handler. As a result, the unmasking was pushed - * into all the nmi handlers. - * - * This handler doesn't seem to have any issues with the unmasking - * so it was left at the top. - */ - apic_write(APIC_LVTPC, APIC_DM_NMI); - intel_pmu_disable_all(); handled = intel_pmu_drain_bts_buffer(); status = intel_pmu_get_status(); @@ -1191,6 +1181,12 @@ again: done: intel_pmu_enable_all(0); + /* + * Only unmask the NMI after the overflow counters + * have been reset. This avoids spurious NMIs on + * Haswell CPUs. + */ + apic_write(APIC_LVTPC, APIC_DM_NMI); return handled; } -- 1.7.7.6