Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756735Ab0GMC0K (ORCPT ); Mon, 12 Jul 2010 22:26:10 -0400 Received: from mx1.redhat.com ([209.132.183.28]:20819 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754091Ab0GMC0F (ORCPT ); Mon, 12 Jul 2010 22:26:05 -0400 From: Zachary Amsden To: KVM , Avi Kivity , Marcelo Tosatti , Glauber Costa Cc: Linux-kernel , Zachary Amsden , Avi Kivity , Marcelo Tosatti , Glauber Costa Subject: [PATCH 10/18] Keep SMP VMs more in sync on unstable TSC Date: Mon, 12 Jul 2010 16:25:30 -1000 Message-Id: <1278987938-23873-11-git-send-email-zamsden@redhat.com> In-Reply-To: <1278987938-23873-1-git-send-email-zamsden@redhat.com> References: <1278987938-23873-1-git-send-email-zamsden@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2086 Lines: 58 SMP VMs on machines with unstable TSC have their TSC offset adjusted by the local offset delta from last measurement. This does not take into account how long it has been since the measurement, leading to drift. Minimize the drift by accounting for any time difference the kernel has observed. Signed-off-by: Zachary Amsden --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/x86.c | 8 +++++++- 2 files changed, 8 insertions(+), 1 deletions(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 324e892..d9e0aa0 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -339,6 +339,7 @@ struct kvm_vcpu_arch { unsigned int time_offset; struct page *time_page; u64 last_host_tsc; + u64 last_host_ns; bool nmi_pending; bool nmi_injected; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 51d3f3e..a4d7cb9 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1888,8 +1888,13 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) native_read_tsc() - vcpu->arch.last_host_tsc; if (tsc_delta < 0) mark_tsc_unstable("KVM discovered backwards TSC"); - if (check_tsc_unstable()) + if (check_tsc_unstable()) { + /* Subtract elapsed cycle time */ + u64 ns = !vcpu->arch.last_host_ns ? 0 : + get_kernel_ns() - vcpu->arch.last_host_ns; + tsc_delta -= nsec_to_cycles(ns); kvm_x86_ops->adjust_tsc_offset(vcpu, -tsc_delta); + } kvm_migrate_timers(vcpu); kvm_request_guest_time_update(vcpu); vcpu->cpu = cpu; @@ -1901,6 +1906,7 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) kvm_x86_ops->vcpu_put(vcpu); kvm_put_guest_fpu(vcpu); vcpu->arch.last_host_tsc = native_read_tsc(); + vcpu->arch.last_host_ns = get_kernel_ns(); } static int is_efer_nx(void) -- 1.7.1 -- 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/