From patchwork Thu Sep 22 08:55:52 2011 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Liu, Jinsong" X-Patchwork-Id: 115908 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Received: from lists.gnu.org (lists.gnu.org [140.186.70.17]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTPS id ABCB9B6F82 for ; Thu, 22 Sep 2011 18:56:39 +1000 (EST) Received: from localhost ([::1]:39325 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1R6f5A-0004Yw-Ix for incoming@patchwork.ozlabs.org; Thu, 22 Sep 2011 04:56:36 -0400 Received: from eggs.gnu.org ([140.186.70.92]:50786) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1R6f53-0004Yr-Vs for qemu-devel@nongnu.org; Thu, 22 Sep 2011 04:56:31 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1R6f4x-0008BA-EP for qemu-devel@nongnu.org; Thu, 22 Sep 2011 04:56:29 -0400 Received: from mga02.intel.com ([134.134.136.20]:8796) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1R6f4w-0008B3-Px for qemu-devel@nongnu.org; Thu, 22 Sep 2011 04:56:23 -0400 Received: from orsmga002.jf.intel.com ([10.7.209.21]) by orsmga101.jf.intel.com with ESMTP; 22 Sep 2011 01:56:21 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos; i="4.67,351,1309762800"; d="scan'208,223"; a="53176448" Received: from pgsmsx601.gar.corp.intel.com ([10.221.43.69]) by orsmga002.jf.intel.com with ESMTP; 22 Sep 2011 01:56:20 -0700 Received: from pgsmsx102.gar.corp.intel.com (10.221.44.80) by pgsmsx601.gar.corp.intel.com (10.221.43.69) with Microsoft SMTP Server (TLS) id 8.2.255.0; Thu, 22 Sep 2011 16:55:55 +0800 Received: from shsmsx601.ccr.corp.intel.com (10.239.4.112) by PGSMSX102.gar.corp.intel.com (10.221.44.80) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 22 Sep 2011 16:55:54 +0800 Received: from shsmsx502.ccr.corp.intel.com ([10.239.4.96]) by shsmsx601.ccr.corp.intel.com ([10.239.4.112]) with mapi; Thu, 22 Sep 2011 16:55:54 +0800 From: "Liu, Jinsong" To: Avi Kivity , Marcelo Tosatti , "kvm@vger.kernel.org" , "qemu-devel@nongnu.org" Date: Thu, 22 Sep 2011 16:55:52 +0800 Thread-Topic: [PATCH] KVM: emulate lapic tsc deadline timer for guest Thread-Index: Acx5BW7d1CjNK6Y/TVe+0aqj2Q4c3Q== Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-detected-operating-system: by eggs.gnu.org: Genre and OS details not recognized. X-Received-From: 134.134.136.20 Cc: "Liu, Jinsong" , "Tian, Kevin" Subject: [Qemu-devel] [PATCH] KVM: emulate lapic tsc deadline timer for guest X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Sender: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org From 4d5b83aba40ce0d421add9a41a6c591a8590a32e Mon Sep 17 00:00:00 2001 From: Liu, Jinsong Date: Thu, 22 Sep 2011 14:00:08 +0800 Subject: [PATCH 2/2] KVM: emulate lapic tsc deadline timer for guest This patch emulate lapic tsc deadline timer for guest: Enumerate tsc deadline timer capability by CPUID; Enable tsc deadline timer mode by lapic MMIO; Start tsc deadline timer by WRMSR; Signed-off-by: Liu, Jinsong --- arch/x86/include/asm/kvm_host.h | 2 + arch/x86/kvm/kvm_timer.h | 2 + arch/x86/kvm/lapic.c | 123 ++++++++++++++++++++++++++++++++------- arch/x86/kvm/lapic.h | 3 + arch/x86/kvm/x86.c | 16 +++++- 5 files changed, 123 insertions(+), 23 deletions(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 6ab4241..b9d4291 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -673,6 +673,8 @@ u8 kvm_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn); extern bool tdp_enabled; +u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu); + /* control of guest tsc rate supported? */ extern bool kvm_has_tsc_control; /* minimum supported tsc_khz for guests */ diff --git a/arch/x86/kvm/kvm_timer.h b/arch/x86/kvm/kvm_timer.h index 64bc6ea..497dbaa 100644 --- a/arch/x86/kvm/kvm_timer.h +++ b/arch/x86/kvm/kvm_timer.h @@ -2,6 +2,8 @@ struct kvm_timer { struct hrtimer timer; s64 period; /* unit: ns */ + u32 timer_mode_mask; + u64 tscdeadline; atomic_t pending; /* accumulated triggered timers */ bool reinject; struct kvm_timer_ops *t_ops; diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 57dcbd4..66b64b8 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -135,9 +135,23 @@ static inline int apic_lvt_vector(struct kvm_lapic *apic, int lvt_type) return apic_get_reg(apic, lvt_type) & APIC_VECTOR_MASK; } +static inline int apic_lvtt_oneshot(struct kvm_lapic *apic) +{ + return ((apic_get_reg(apic, APIC_LVTT) & + apic->lapic_timer.timer_mode_mask) == APIC_LVT_TIMER_ONESHOT); +} + static inline int apic_lvtt_period(struct kvm_lapic *apic) { - return apic_get_reg(apic, APIC_LVTT) & APIC_LVT_TIMER_PERIODIC; + return ((apic_get_reg(apic, APIC_LVTT) & + apic->lapic_timer.timer_mode_mask) == APIC_LVT_TIMER_PERIODIC); +} + +static inline int apic_lvtt_tscdeadline(struct kvm_lapic *apic) +{ + return ((apic_get_reg(apic, APIC_LVTT) & + apic->lapic_timer.timer_mode_mask) == + APIC_LVT_TIMER_TSCDEADLINE); } static inline int apic_lvt_nmi_mode(u32 lvt_val) @@ -166,7 +180,7 @@ static inline int apic_x2apic_mode(struct kvm_lapic *apic) } static unsigned int apic_lvt_mask[APIC_LVT_NUM] = { - LVT_MASK | APIC_LVT_TIMER_PERIODIC, /* LVTT */ + LVT_MASK , /* part LVTT mask, timer mode mask added at runtime */ LVT_MASK | APIC_MODE_MASK, /* LVTTHMR */ LVT_MASK | APIC_MODE_MASK, /* LVTPC */ LINT_MASK, LINT_MASK, /* LVT0-1 */ @@ -570,6 +584,9 @@ static u32 __apic_read(struct kvm_lapic *apic, unsigned int offset) break; case APIC_TMCCT: /* Timer CCR */ + if (apic_lvtt_tscdeadline(apic)) + return 0; + val = apic_get_tmcct(apic); break; @@ -664,29 +681,32 @@ static void update_divide_count(struct kvm_lapic *apic) static void start_apic_timer(struct kvm_lapic *apic) { - ktime_t now = apic->lapic_timer.timer.base->get_time(); - - apic->lapic_timer.period = (u64)apic_get_reg(apic, APIC_TMICT) * - APIC_BUS_CYCLE_NS * apic->divide_count; + ktime_t now; atomic_set(&apic->lapic_timer.pending, 0); - if (!apic->lapic_timer.period) - return; - /* - * Do not allow the guest to program periodic timers with small - * interval, since the hrtimers are not throttled by the host - * scheduler. - */ - if (apic_lvtt_period(apic)) { - if (apic->lapic_timer.period < NSEC_PER_MSEC/2) - apic->lapic_timer.period = NSEC_PER_MSEC/2; - } + if (apic_lvtt_period(apic) || apic_lvtt_oneshot(apic)) { + /* lapic timer in oneshot or peroidic mode */ + now = apic->lapic_timer.timer.base->get_time(); + apic->lapic_timer.period = (u64)apic_get_reg(apic, APIC_TMICT) + * APIC_BUS_CYCLE_NS * apic->divide_count; + + if (!apic->lapic_timer.period) + return; + /* + * Do not allow the guest to program periodic timers with small + * interval, since the hrtimers are not throttled by the host + * scheduler. + */ + if (apic_lvtt_period(apic)) { + if (apic->lapic_timer.period < NSEC_PER_MSEC/2) + apic->lapic_timer.period = NSEC_PER_MSEC/2; + } - hrtimer_start(&apic->lapic_timer.timer, - ktime_add_ns(now, apic->lapic_timer.period), - HRTIMER_MODE_ABS); + hrtimer_start(&apic->lapic_timer.timer, + ktime_add_ns(now, apic->lapic_timer.period), + HRTIMER_MODE_ABS); - apic_debug("%s: bus cycle is %" PRId64 "ns, now 0x%016" + apic_debug("%s: bus cycle is %" PRId64 "ns, now 0x%016" PRIx64 ", " "timer initial count 0x%x, period %lldns, " "expire @ 0x%016" PRIx64 ".\n", __func__, @@ -695,6 +715,29 @@ static void start_apic_timer(struct kvm_lapic *apic) apic->lapic_timer.period, ktime_to_ns(ktime_add_ns(now, apic->lapic_timer.period))); + } else if (apic_lvtt_tscdeadline(apic)) { + /* lapic timer in tsc deadline mode */ + u64 guest_tsc, tscdeadline = apic->lapic_timer.tscdeadline; + u64 ns = 0; + struct kvm_vcpu *vcpu = apic->vcpu; + unsigned long this_tsc_khz = vcpu_tsc_khz(vcpu); + unsigned long flags; + + if (unlikely(!tscdeadline || !this_tsc_khz)) + return; + + local_irq_save(flags); + + now = apic->lapic_timer.timer.base->get_time(); + guest_tsc = kvm_x86_ops->read_l1_tsc(vcpu); + if (likely(tscdeadline > guest_tsc)) + ns = (tscdeadline - guest_tsc) + * 1000000L / this_tsc_khz; + hrtimer_start(&apic->lapic_timer.timer, + ktime_add_ns(now, ns), HRTIMER_MODE_ABS); + + local_irq_restore(flags); + } } static void apic_manage_nmi_watchdog(struct kvm_lapic *apic, u32 lvt0_val) @@ -782,7 +825,6 @@ static int apic_reg_write(struct kvm_lapic *apic, u32 reg, u32 val) case APIC_LVT0: apic_manage_nmi_watchdog(apic, val); - case APIC_LVTT: case APIC_LVTTHMR: case APIC_LVTPC: case APIC_LVT1: @@ -796,7 +838,22 @@ static int apic_reg_write(struct kvm_lapic *apic, u32 reg, u32 val) break; + case APIC_LVTT: + if ((apic_get_reg(apic, APIC_LVTT) & + apic->lapic_timer.timer_mode_mask) != + (val & apic->lapic_timer.timer_mode_mask)) + hrtimer_cancel(&apic->lapic_timer.timer); + + if (!apic_sw_enabled(apic)) + val |= APIC_LVT_MASKED; + val &= (apic_lvt_mask[0] | apic->lapic_timer.timer_mode_mask); + apic_set_reg(apic, APIC_LVTT, val); + break; + case APIC_TMICT: + if (apic_lvtt_tscdeadline(apic)) + break; + hrtimer_cancel(&apic->lapic_timer.timer); apic_set_reg(apic, APIC_TMICT, val); start_apic_timer(apic); @@ -883,6 +940,28 @@ void kvm_free_lapic(struct kvm_vcpu *vcpu) *---------------------------------------------------------------------- */ +u64 kvm_get_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu) +{ + struct kvm_lapic *apic = vcpu->arch.apic; + + if (apic_lvtt_oneshot(apic) || apic_lvtt_period(apic)) + return 0; + + return apic->lapic_timer.tscdeadline; +} + +void kvm_set_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu, u64 data) +{ + struct kvm_lapic *apic = vcpu->arch.apic; + + if (apic_lvtt_oneshot(apic) || apic_lvtt_period(apic)) + return; + + hrtimer_cancel(&apic->lapic_timer.timer); + apic->lapic_timer.tscdeadline = data; + start_apic_timer(apic); +} + void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8) { struct kvm_lapic *apic = vcpu->arch.apic; diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h index 52c9e6b..10c6ee6 100644 --- a/arch/x86/kvm/lapic.h +++ b/arch/x86/kvm/lapic.h @@ -41,6 +41,9 @@ int kvm_lapic_enabled(struct kvm_vcpu *vcpu); bool kvm_apic_present(struct kvm_vcpu *vcpu); int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu); +u64 kvm_get_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu); +void kvm_set_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu, u64 data); + void kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr); void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu); void kvm_lapic_sync_to_vapic(struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 6b37f18..382c0b8 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -610,6 +610,13 @@ static void update_cpuid(struct kvm_vcpu *vcpu) if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE)) best->ecx |= bit(X86_FEATURE_OSXSAVE); } + + if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && + best->function == 0x1) { + best->ecx |= bit(X86_FEATURE_TSC_DEADLINE_TIMER); + vcpu->arch.apic->lapic_timer.timer_mode_mask = (3 << 17); + } else + vcpu->arch.apic->lapic_timer.timer_mode_mask = (1 << 17); } int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) @@ -825,6 +832,7 @@ static u32 msrs_to_save[] = { static unsigned num_msrs_to_save; static u32 emulated_msrs[] = { + MSR_IA32_TSCDEADLINE, MSR_IA32_MISC_ENABLE, MSR_IA32_MCG_STATUS, MSR_IA32_MCG_CTL, @@ -1000,7 +1008,7 @@ static inline int kvm_tsc_changes_freq(void) return ret; } -static u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu) +u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu) { if (vcpu->arch.virtual_tsc_khz) return vcpu->arch.virtual_tsc_khz; @@ -1564,6 +1572,9 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) break; case APIC_BASE_MSR ... APIC_BASE_MSR + 0x3ff: return kvm_x2apic_msr_write(vcpu, msr, data); + case MSR_IA32_TSCDEADLINE: + kvm_set_lapic_tscdeadline_msr(vcpu, data); + break; case MSR_IA32_MISC_ENABLE: vcpu->arch.ia32_misc_enable_msr = data; break; @@ -1893,6 +1904,9 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case APIC_BASE_MSR ... APIC_BASE_MSR + 0x3ff: return kvm_x2apic_msr_read(vcpu, msr, pdata); break; + case MSR_IA32_TSCDEADLINE: + data = kvm_get_lapic_tscdeadline_msr(vcpu); + break; case MSR_IA32_MISC_ENABLE: data = vcpu->arch.ia32_misc_enable_msr; break;