KVM: X86: Update last_guest_tsc in vcpu_put
The last_guest_tsc is used in vcpu_load to adjust the
tsc_offset since tsc-scaling is merged. So the
last_guest_tsc needs to be updated in vcpu_put instead of
the the last_host_tsc. This is fixed with this patch.
Reported-by: Jan Kiszka <jan.kiszka@web.de>
Tested-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 3e03f37..e50bffc 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -393,7 +393,6 @@
unsigned int hw_tsc_khz;
unsigned int time_offset;
struct page *time_page;
- u64 last_host_tsc;
u64 last_guest_tsc;
u64 last_kernel_ns;
u64 last_tsc_nsec;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index b05e3fc..6aa1377 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2146,7 +2146,7 @@
{
kvm_x86_ops->vcpu_put(vcpu);
kvm_put_guest_fpu(vcpu);
- vcpu->arch.last_host_tsc = native_read_tsc();
+ kvm_get_msr(vcpu, MSR_IA32_TSC, &vcpu->arch.last_guest_tsc);
}
static int is_efer_nx(void)