From patchwork Fri Aug 20 08:07:30 2010 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Zachary Amsden X-Patchwork-Id: 120499 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by demeter.kernel.org (8.14.4/8.14.3) with ESMTP id o7K89iO9002787 for ; Fri, 20 Aug 2010 08:09:44 GMT Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752065Ab0HTIIn (ORCPT ); Fri, 20 Aug 2010 04:08:43 -0400 Received: from mx1.redhat.com ([209.132.183.28]:33131 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752055Ab0HTIIj (ORCPT ); Fri, 20 Aug 2010 04:08:39 -0400 Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) by mx1.redhat.com (8.13.8/8.13.8) with ESMTP id o7K88Z36004833 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Fri, 20 Aug 2010 04:08:35 -0400 Received: from mysore (vpn-9-158.rdu.redhat.com [10.11.9.158]) by int-mx01.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with ESMTP id o7K87qfA027969; Fri, 20 Aug 2010 04:08:32 -0400 From: Zachary Amsden To: kvm@vger.kernel.org Cc: Zachary Amsden , Avi Kivity , Marcelo Tosatti , Glauber Costa , Thomas Gleixner , John Stultz , linux-kernel@vger.kernel.org Subject: [KVM timekeeping 16/35] Fix a possible backwards warp of kvmclock Date: Thu, 19 Aug 2010 22:07:30 -1000 Message-Id: <1282291669-25709-17-git-send-email-zamsden@redhat.com> In-Reply-To: <1282291669-25709-1-git-send-email-zamsden@redhat.com> References: <1282291669-25709-1-git-send-email-zamsden@redhat.com> X-Scanned-By: MIMEDefang 2.67 on 10.5.11.11 Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org X-Greylist: IP, sender and recipient auto-whitelisted, not delayed by milter-greylist-4.2.3 (demeter.kernel.org [140.211.167.41]); Fri, 20 Aug 2010 08:09:45 +0000 (UTC) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 324e892..871800d 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -339,6 +339,8 @@ struct kvm_vcpu_arch { unsigned int time_offset; struct page *time_page; u64 last_host_tsc; + u64 last_guest_tsc; + u64 last_kernel_ns; bool nmi_pending; bool nmi_injected; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 1948c36..fe74b42 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -976,14 +976,15 @@ static int kvm_write_guest_time(struct kvm_vcpu *v) struct kvm_vcpu_arch *vcpu = &v->arch; void *shared_kaddr; unsigned long this_tsc_khz; - s64 kernel_ns; + s64 kernel_ns, max_kernel_ns; + u64 tsc_timestamp; if ((!vcpu->time_page)) return 0; /* Keep irq disabled to prevent changes to the clock */ local_irq_save(flags); - kvm_get_msr(v, MSR_IA32_TSC, &vcpu->hv_clock.tsc_timestamp); + kvm_get_msr(v, MSR_IA32_TSC, &tsc_timestamp); kernel_ns = get_kernel_ns(); this_tsc_khz = __get_cpu_var(cpu_tsc_khz); local_irq_restore(flags); @@ -993,13 +994,49 @@ static int kvm_write_guest_time(struct kvm_vcpu *v) return 1; } + /* + * Time as measured by the TSC may go backwards when resetting the base + * tsc_timestamp. The reason for this is that the TSC resolution is + * higher than the resolution of the other clock scales. Thus, many + * possible measurments of the TSC correspond to one measurement of any + * other clock, and so a spread of values is possible. This is not a + * problem for the computation of the nanosecond clock; with TSC rates + * around 1GHZ, there can only be a few cycles which correspond to one + * nanosecond value, and any path through this code will inevitably + * take longer than that. However, with the kernel_ns value itself, + * the precision may be much lower, down to HZ granularity. If the + * first sampling of TSC against kernel_ns ends in the low part of the + * range, and the second in the high end of the range, we can get: + * + * (TSC - offset_low) * S + kns_old > (TSC - offset_high) * S + kns_new + * + * As the sampling errors potentially range in the thousands of cycles, + * it is possible such a time value has already been observed by the + * guest. To protect against this, we must compute the system time as + * observed by the guest and ensure the new system time is greater. + */ + max_kernel_ns = 0; + if (vcpu->hv_clock.tsc_timestamp && vcpu->last_guest_tsc) { + max_kernel_ns = vcpu->last_guest_tsc - + vcpu->hv_clock.tsc_timestamp; + max_kernel_ns = pvclock_scale_delta(max_kernel_ns, + vcpu->hv_clock.tsc_to_system_mul, + vcpu->hv_clock.tsc_shift); + max_kernel_ns += vcpu->last_kernel_ns; + } + if (unlikely(vcpu->hw_tsc_khz != this_tsc_khz)) { kvm_set_time_scale(this_tsc_khz, &vcpu->hv_clock); vcpu->hw_tsc_khz = this_tsc_khz; } + if (max_kernel_ns > kernel_ns) + kernel_ns = max_kernel_ns; + /* With all the info we got, fill in the values */ + vcpu->hv_clock.tsc_timestamp = tsc_timestamp; vcpu->hv_clock.system_time = kernel_ns + v->kvm->arch.kvmclock_offset; + vcpu->last_kernel_ns = kernel_ns; vcpu->hv_clock.flags = 0; /* @@ -4931,6 +4968,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) if (hw_breakpoint_active()) hw_breakpoint_restore(); + kvm_get_msr(vcpu, MSR_IA32_TSC, &vcpu->arch.last_guest_tsc); + atomic_set(&vcpu->guest_mode, 0); smp_wmb(); local_irq_enable();