* [RFC PATCH] KVM: x86: Set PF_VCPU when processing IRQs to fix tick-based accounting
@ 2021-02-06 0:42 Sean Christopherson
2021-02-08 10:28 ` Vitaly Kuznetsov
0 siblings, 1 reply; 2+ messages in thread
From: Sean Christopherson @ 2021-02-06 0:42 UTC (permalink / raw)
To: Paolo Bonzini
Cc: Sean Christopherson, Vitaly Kuznetsov, Wanpeng Li, Jim Mattson,
Joerg Roedel, kvm, linux-kernel, Peter Zijlstra, Thomas Gleixner
Temporarily set PF_VCPU while processing IRQ VM-Exits so that a tick IRQ
accounts the time slice to the guest. Tick-based accounting of guest
time is currently broken as PF_VCPU is only set during the relatively
short VM-Enter sequence, which runs entirely with IRQs disabled, and IRQs
that occur in the guest are processed well after PF_VCPU is cleared.
Keep PF_VCPU set across both VMX's processing of already-acked IRQs
(handle_exit_irqoff()) and the explicit IRQ window (SVM's processing,
plus ticks that occur immediately after VM-Exit on both VMX and SVM).
Fixes: 87fa7f3e98a1 ("x86/kvm: Move context tracking where it belongs")
Cc: stable@vger.kernel.org
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
This is quite obnoxious, hence the RFC, but I can't think of a clever,
less ugly way to fix the accounting.
arch/x86/kvm/x86.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index d9f931c63293..6ddf341cd755 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -9118,6 +9118,13 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
vcpu->mode = OUTSIDE_GUEST_MODE;
smp_wmb();
+ /*
+ * Temporarily pretend this task is running a vCPU when potentially
+ * processing an IRQ exit, including the below opening of an IRQ
+ * window. Tick-based accounting of guest time relies on PF_VCPU
+ * being set when the tick IRQ handler runs.
+ */
+ current->flags |= PF_VCPU;
static_call(kvm_x86_handle_exit_irqoff)(vcpu);
/*
@@ -9132,6 +9139,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
++vcpu->stat.exits;
local_irq_disable();
kvm_after_interrupt(vcpu);
+ current->flags &= ~PF_VCPU;
if (lapic_in_kernel(vcpu)) {
s64 delta = vcpu->arch.apic->lapic_timer.advance_expire_delta;
--
2.30.0.478.g8a0d178c01-goog
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [RFC PATCH] KVM: x86: Set PF_VCPU when processing IRQs to fix tick-based accounting
2021-02-06 0:42 [RFC PATCH] KVM: x86: Set PF_VCPU when processing IRQs to fix tick-based accounting Sean Christopherson
@ 2021-02-08 10:28 ` Vitaly Kuznetsov
0 siblings, 0 replies; 2+ messages in thread
From: Vitaly Kuznetsov @ 2021-02-08 10:28 UTC (permalink / raw)
To: Sean Christopherson
Cc: Wanpeng Li, Jim Mattson, Joerg Roedel, kvm, linux-kernel,
Peter Zijlstra, Thomas Gleixner, Paolo Bonzini
Sean Christopherson <seanjc@google.com> writes:
> Temporarily set PF_VCPU while processing IRQ VM-Exits so that a tick IRQ
> accounts the time slice to the guest. Tick-based accounting of guest
> time is currently broken as PF_VCPU is only set during the relatively
> short VM-Enter sequence, which runs entirely with IRQs disabled, and IRQs
> that occur in the guest are processed well after PF_VCPU is cleared.
>
> Keep PF_VCPU set across both VMX's processing of already-acked IRQs
> (handle_exit_irqoff()) and the explicit IRQ window (SVM's processing,
> plus ticks that occur immediately after VM-Exit on both VMX and SVM).
>
> Fixes: 87fa7f3e98a1 ("x86/kvm: Move context tracking where it belongs")
> Cc: stable@vger.kernel.org
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Signed-off-by: Sean Christopherson <seanjc@google.com>
> ---
>
> This is quite obnoxious, hence the RFC, but I can't think of a clever,
> less ugly way to fix the accounting.
>
> arch/x86/kvm/x86.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index d9f931c63293..6ddf341cd755 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -9118,6 +9118,13 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
> vcpu->mode = OUTSIDE_GUEST_MODE;
> smp_wmb();
>
> + /*
> + * Temporarily pretend this task is running a vCPU when potentially
> + * processing an IRQ exit, including the below opening of an IRQ
> + * window. Tick-based accounting of guest time relies on PF_VCPU
> + * being set when the tick IRQ handler runs.
> + */
> + current->flags |= PF_VCPU;
Should we do it only when !vtime_accounting_enabled_cpu() maybe?
> static_call(kvm_x86_handle_exit_irqoff)(vcpu);
>
> /*
> @@ -9132,6 +9139,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
> ++vcpu->stat.exits;
> local_irq_disable();
> kvm_after_interrupt(vcpu);
> + current->flags &= ~PF_VCPU;
>
> if (lapic_in_kernel(vcpu)) {
> s64 delta = vcpu->arch.apic->lapic_timer.advance_expire_delta;
--
Vitaly
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-02-08 10:41 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-02-06 0:42 [RFC PATCH] KVM: x86: Set PF_VCPU when processing IRQs to fix tick-based accounting Sean Christopherson
2021-02-08 10:28 ` Vitaly Kuznetsov
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.