From: Wanpeng Li wanpengli@tencent.com

The bugzilla https://bugzilla.kernel.org/show_bug.cgi?id=209831 reported that the guest time remains 0 when running a while true loop in the guest.

The commit 87fa7f3e98a131 ("x86/kvm: Move context tracking where it belongs") moves guest_exit_irqoff() close to vmexit breaks the tick-based time accouting when the ticks that happen after IRQs are disabled are incorrectly accounted to the host/system time. This is because we exit the guest state too early.

Keep context tracking around the actual vmentry/exit code, the time accounting logic is separated out by preposed patch from guest_enter/exit_irqoff(). Keep vtime-based time accounting around the actual vmentry/exit code when vtime_accounting_enabled_this_cpu() is true, leave PF_VCPU clearing after handle_exit_irqoff() and explicit IRQ window for tick-based time accouting.

Fixes: 87fa7f3e98a131 ("x86/kvm: Move context tracking where it belongs") Cc: Thomas Gleixner tglx@linutronix.de Cc: Sean Christopherson seanjc@google.com Cc: Michael Tokarev mjt@tls.msk.ru Cc: stable@vger.kernel.org#v5.9-rc1+ Suggested-by: Thomas Gleixner tglx@linutronix.de Signed-off-by: Wanpeng Li wanpengli@tencent.com --- arch/x86/kvm/svm/svm.c | 6 ++++-- arch/x86/kvm/vmx/vmx.c | 6 ++++-- arch/x86/kvm/x86.c | 1 + 3 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index 48b396f3..2a4e284 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -3726,11 +3726,12 @@ static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu) * accordingly. */ instrumentation_begin(); + vtime_account_guest_enter(); trace_hardirqs_on_prepare(); lockdep_hardirqs_on_prepare(CALLER_ADDR0); instrumentation_end();

- guest_enter_irqoff(); + context_guest_enter_irqoff(); lockdep_hardirqs_on(CALLER_ADDR0);

if (sev_es_guest(vcpu->kvm)) { @@ -3758,10 +3759,11 @@ static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu) * into world and some more. */ lockdep_hardirqs_off(CALLER_ADDR0); - guest_exit_irqoff(); + context_guest_exit_irqoff();

instrumentation_begin(); trace_hardirqs_off_finish(); + __vtime_account_guest_exit(); instrumentation_end(); }

diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index c05e6e2..23be956 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -6613,11 +6613,12 @@ static noinstr void vmx_vcpu_enter_exit(struct kvm_vcpu *vcpu, * accordingly. */ instrumentation_begin(); + vtime_account_guest_enter(); trace_hardirqs_on_prepare(); lockdep_hardirqs_on_prepare(CALLER_ADDR0); instrumentation_end();

- guest_enter_irqoff(); + context_guest_enter_irqoff(); lockdep_hardirqs_on(CALLER_ADDR0);

/* L1D Flush includes CPU buffer clear to mitigate MDS */ @@ -6647,10 +6648,11 @@ static noinstr void vmx_vcpu_enter_exit(struct kvm_vcpu *vcpu, * into world and some more. */ lockdep_hardirqs_off(CALLER_ADDR0); - guest_exit_irqoff(); + context_guest_exit_irqoff();

instrumentation_begin(); trace_hardirqs_off_finish(); + __vtime_account_guest_exit(); instrumentation_end(); }

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ce9a1d2..0d2dd3f 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -9245,6 +9245,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) ++vcpu->stat.exits; local_irq_disable(); kvm_after_interrupt(vcpu); + vcpu_account_guest_exit();

if (lapic_in_kernel(vcpu)) { s64 delta = vcpu->arch.apic->lapic_timer.advance_expire_delta;