Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757516Ab2JQRFB (ORCPT ); Wed, 17 Oct 2012 13:05:01 -0400 Received: from mail-qa0-f46.google.com ([209.85.216.46]:61118 "EHLO mail-qa0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754150Ab2JQRE7 (ORCPT ); Wed, 17 Oct 2012 13:04:59 -0400 Date: Wed, 17 Oct 2012 13:05:45 -0400 (EDT) From: Vince Weaver To: linux-kernel@vger.kernel.org cc: Peter Zijlstra , Paul Mackerras , Ingo Molnar , Arnaldo Carvalho de Melo , eranian@gmail.com, "Meadows, Lawrence F" Subject: [PATCH 3/3] perf_event: Enable overflow on KNC with a custom knc_pmu_handle_irq() In-Reply-To: Message-ID: References: User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3041 Lines: 120 Although based on the P6 design, the interrupt mechnanism for KNC more closely resembles the intel architectural perfmon one. We can't just re-use that code though, because KNC has different MSR numbers for the status and ack registers. In this case we just cut-and paste from perf_event_intel.c with some minor changes, as it looks like it would not be worth the trouble to change that code to be MSR-configurable. Signed-off-by: Vince Weaver diff -ur linux-3.7-rc1.orig/arch/x86/kernel/cpu/perf_event_knc.c linux-3.7-rc1/arch/x86/kernel/cpu/perf_event_knc.c --- linux-3.7-rc1.orig/arch/x86/kernel/cpu/perf_event_knc.c 2012-10-14 17:41:04.000000000 -0400 +++ linux-3.7-rc1/arch/x86/kernel/cpu/perf_event_knc.c 2012-10-17 12:11:48.604393067 -0400 @@ -3,6 +3,8 @@ #include #include +#include + #include "perf_event.h" static const u64 knc_perfmon_event_map[] = @@ -197,6 +199,80 @@ (void)wrmsrl_safe(hwc->config_base + hwc->idx, val); } +static inline u64 knc_pmu_get_status(void) +{ + u64 status; + + rdmsrl(MSR_KNC_IA32_PERF_GLOBAL_STATUS, status); + + return status; +} + +static inline void knc_pmu_ack_status(u64 ack) +{ + wrmsrl(MSR_KNC_IA32_PERF_GLOBAL_OVF_CONTROL, ack); +} + +static int knc_pmu_handle_irq(struct pt_regs *regs) +{ + struct perf_sample_data data; + struct cpu_hw_events *cpuc; + int bit, loops; + u64 status; + int handled=0; + + cpuc = &__get_cpu_var(cpu_hw_events); + + knc_pmu_disable_all(); + + status = knc_pmu_get_status(); + if (!status) { + knc_pmu_enable_all(0); + return handled; + } + + loops = 0; +again: + knc_pmu_ack_status(status); + if (++loops > 100) { + WARN_ONCE(1, "perfevents: irq loop stuck!\n"); + perf_event_print_debug(); + goto done; + } + + inc_irq_stat(apic_perf_irqs); + + for_each_set_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) { + struct perf_event *event = cpuc->events[bit]; + + handled++; + + if (!test_bit(bit, cpuc->active_mask)) + continue; + + if (!intel_pmu_save_and_restart(event)) + continue; + + perf_sample_data_init(&data, 0, event->hw.last_period); + + if (perf_event_overflow(event, &data, regs)) + x86_pmu_stop(event, 0); + } + + /* + * Repeat if there is more work to be done: + */ + status = knc_pmu_get_status(); + if (status) + goto again; + +done: + knc_pmu_enable_all(0); + + return handled; +} + + PMU_FORMAT_ATTR(event, "config:0-7" ); PMU_FORMAT_ATTR(umask, "config:8-15" ); PMU_FORMAT_ATTR(edge, "config:18" ); @@ -214,7 +290,7 @@ static __initconst struct x86_pmu knc_pmu = { .name = "knc", - .handle_irq = x86_pmu_handle_irq, + .handle_irq = knc_pmu_handle_irq, .disable_all = knc_pmu_disable_all, .enable_all = knc_pmu_enable_all, .enable = knc_pmu_enable_event, -- 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/