2010-08-20 08:12:21

by Zachary Amsden

[permalink] [raw]
Subject: [KVM timekeeping 26/35] Catchup slower TSC to guest rate

Use the catchup code to continue adjusting the TSC when
running at lower than the guest rate

Signed-off-by: Zachary Amsden <[email protected]>
---
arch/x86/kvm/x86.c | 9 ++++++++-
1 files changed, 8 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index a4215d7..086d56a 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1013,8 +1013,11 @@ static int kvm_guest_time_update(struct kvm_vcpu *v)
kvm_x86_ops->adjust_tsc_offset(v, tsc-tsc_timestamp);
}
local_irq_restore(flags);
- if (catchup)
+ if (catchup) {
+ if (this_tsc_khz < v->kvm->arch.virtual_tsc_khz)
+ vcpu->tsc_rebase = 1;
return 0;
+ }

/*
* Time as measured by the TSC may go backwards when resetting the base
@@ -5022,6 +5025,10 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)

kvm_guest_exit();

+ /* Running on slower TSC without kvmclock, we must bump TSC */
+ if (vcpu->arch.tsc_rebase)
+ kvm_request_clock_update(vcpu);
+
preempt_enable();

vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
--
1.7.1