From patchwork Mon Jul 6 02:17:47 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Shannon Zhao X-Patchwork-Id: 6719311 Return-Path: X-Original-To: patchwork-kvm@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork1.web.kernel.org (Postfix) with ESMTP id 65A789F38C for ; Mon, 6 Jul 2015 02:19:26 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 6578620642 for ; Mon, 6 Jul 2015 02:19:25 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 49BA82049C for ; Mon, 6 Jul 2015 02:19:24 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753672AbbGFCTW (ORCPT ); Sun, 5 Jul 2015 22:19:22 -0400 Received: from mail-pa0-f44.google.com ([209.85.220.44]:35759 "EHLO mail-pa0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753665AbbGFCTT (ORCPT ); Sun, 5 Jul 2015 22:19:19 -0400 Received: by pactm7 with SMTP id tm7so86963850pac.2 for ; Sun, 05 Jul 2015 19:19:19 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=K+9VXeFH80ALR+SXG8oIBzzxq5vf726b+YassUSm5Yk=; b=U/UMkRtSLcpF85xbZJuAsbPCzQLFqMl4gTgtOqBk72LVTQrniaX64/tSZs7En22PUX j1RTpF1ZXTrde6Q3DH7o4wowXhj0a9FCg7bheUCNsBpmGaCa4rSu47BGTKzyPOROO249 QavJT4Y64HqlaRPDAT27IHt4GL0DXffKCGhamsaLM4wIqEiyzT4wqRJjnj55E4V8PTS7 ZHLF3ad8DRf0OyJys08Me6wfaUmfyEe/c7YwxUnJm7T8L7CW00Q2tClayzOemn8PjLj6 AxAGn0G8hGfSkgv4+U/AR3RiBfWTRBCq2uLvHiuf8wxT/KwdZwAli9GjwsauX347S52u 0uJw== X-Gm-Message-State: ALoCoQllxV1nQ/Jer57cFdwrjUT6W9cQ8T7hQtl7EaJOCLdG7fXm3Tu3GzYPcg/ftsLlBUUyy2TY X-Received: by 10.68.192.74 with SMTP id he10mr42217222pbc.64.1436149159009; Sun, 05 Jul 2015 19:19:19 -0700 (PDT) Received: from localhost ([120.136.34.248]) by mx.google.com with ESMTPSA id xs13sm16248652pac.3.2015.07.05.19.19.16 (version=TLSv1 cipher=RC4-SHA bits=128/128); Sun, 05 Jul 2015 19:19:17 -0700 (PDT) From: shannon.zhao@linaro.org To: kvmarm@lists.cs.columbia.edu Cc: linux-arm-kernel@lists.infradead.org, kvm@vger.kernel.org, christoffer.dall@linaro.org, will.deacon@arm.com, marc.zyngier@arm.com, alex.bennee@linaro.org, shannon.zhao@linaro.org, zhaoshenglong@huawei.com Subject: [PATCH 17/18] KVM: ARM64: Add PMU overflow interrupt routing Date: Mon, 6 Jul 2015 10:17:47 +0800 Message-Id: <1436149068-3784-18-git-send-email-shannon.zhao@linaro.org> X-Mailer: git-send-email 1.9.5.msysgit.1 In-Reply-To: <1436149068-3784-1-git-send-email-shannon.zhao@linaro.org> References: <1436149068-3784-1-git-send-email-shannon.zhao@linaro.org> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org X-Spam-Status: No, score=-7.5 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=ham version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Shannon Zhao When calling perf_event_create_kernel_counter to create perf_event, assign a overflow handler. Then when perf event overflows, if vcpu doesn't run, call irq_work_queue to wake up vcpu. Otherwise call kvm_vgic_inject_irq to inject the interrupt. Signed-off-by: Shannon Zhao --- arch/arm/kvm/arm.c | 2 ++ include/kvm/arm_pmu.h | 2 ++ virt/kvm/arm/pmu.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 56 insertions(+), 1 deletion(-) diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index 4e82625..41eb063 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -551,6 +551,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) preempt_enable(); kvm_timer_sync_hwstate(vcpu); kvm_vgic_sync_hwstate(vcpu); + kvm_pmu_sync_hwstate(vcpu); continue; } @@ -595,6 +596,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) kvm_timer_sync_hwstate(vcpu); kvm_vgic_sync_hwstate(vcpu); + kvm_pmu_sync_hwstate(vcpu); ret = handle_exit(vcpu, run, ret); } diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h index 6985809..5bcf27b 100644 --- a/include/kvm/arm_pmu.h +++ b/include/kvm/arm_pmu.h @@ -45,6 +45,7 @@ struct kvm_pmu { }; #ifdef CONFIG_KVM_ARM_PMU +void kvm_pmu_sync_hwstate(struct kvm_vcpu *vcpu); void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu); void kvm_pmu_set_counter_value(struct kvm_vcpu *vcpu, unsigned long select_idx, unsigned long val); @@ -59,6 +60,7 @@ void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, unsigned long data, unsigned long select_idx); void kvm_pmu_init(struct kvm_vcpu *vcpu); #else +void kvm_pmu_sync_hwstate(struct kvm_vcpu *vcpu) {} static inline void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu) {} void kvm_pmu_set_counter_value(struct kvm_vcpu *vcpu, unsigned long select_idx, unsigned long val) {} diff --git a/virt/kvm/arm/pmu.c b/virt/kvm/arm/pmu.c index e655426..f957b85 100644 --- a/virt/kvm/arm/pmu.c +++ b/virt/kvm/arm/pmu.c @@ -20,6 +20,7 @@ #include #include #include +#include /* PMU HW events mapping. */ static struct kvm_pmu_hw_event_map { @@ -81,6 +82,23 @@ static void kvm_pmu_stop_counter(struct kvm_vcpu *vcpu, } /** + * kvm_pmu_sync_hwstate - sync pmu state for cpu + * @vcpu: The vcpu pointer + * + * Inject virtual PMU IRQ if IRQ is pending for this cpu. + */ +void kvm_pmu_sync_hwstate(struct kvm_vcpu *vcpu) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + + if (pmu->irq_pending) { + kvm_vgic_inject_irq(vcpu->kvm, vcpu->vcpu_id, pmu->irq_num, 1); + pmu->irq_pending = 0; + return; + } +} + +/** * kvm_pmu_vcpu_reset - reset pmu state for cpu * @vcpu: The vcpu pointer * @@ -96,6 +114,37 @@ void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu) pmu->irq_pending = false; } +static void kvm_pum_trigger_pmi(struct irq_work *irq_work) +{ + struct kvm_pmu *pmu = container_of(irq_work, struct kvm_pmu, irq_work); + struct kvm_vcpu *vcpu = container_of(pmu, struct kvm_vcpu, arch.pmu); + + kvm_vgic_inject_irq(vcpu->kvm, vcpu->vcpu_id, pmu->irq_num, 1); +} + +/** + * When perf event overflows, if vcpu doesn't run, call irq_work_queue to wake + * up vcpu. Otherwise call kvm_vgic_inject_irq to inject the interrupt. + */ +static void kvm_pmu_perf_overflow(struct perf_event *perf_event, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + struct kvm_pmc *pmc = perf_event->overflow_handler_context; + struct kvm_vcpu *vcpu = pmc->vcpu; + struct kvm_pmu *pmu = &vcpu->arch.pmu; + + if (pmc->interrupt == true) { + __set_bit(pmc->idx, (unsigned long *)&pmu->overflow_status); + pmu->irq_pending = 1; + if (vcpu->mode != IN_GUEST_MODE) + irq_work_queue(&pmu->irq_work); + else + kvm_vgic_inject_irq(vcpu->kvm, vcpu->vcpu_id, + pmu->irq_num, 1); + } +} + /** * kvm_pmu_set_counter_value - set PMU counter value * @vcpu: The vcpu pointer @@ -322,7 +371,8 @@ void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, unsigned long data, attr.config = config; attr.sample_period = (-pmc->counter) & (((u64)1 << 32) - 1); - event = perf_event_create_kernel_counter(&attr, -1, current, NULL, pmc); + event = perf_event_create_kernel_counter(&attr, -1, current, + kvm_pmu_perf_overflow, pmc); if (IS_ERR(event)) { kvm_err("kvm: pmu event creation failed %ld\n", PTR_ERR(event)); @@ -351,4 +401,5 @@ void kvm_pmu_init(struct kvm_vcpu *vcpu) } pmu->irq_num = -1; + init_irq_work(&pmu->irq_work, kvm_pum_trigger_pmi); }