[PATCH v3] kvm: x86: Disable KVM_HC_CLOCK_PAIRING if tsc is in always catchup mode

[PATCH v3] kvm: x86: Disable KVM_HC_CLOCK_PAIRING if tsc is in always catchup mode

[Anton Romanov]

If vcpu has tsc_always_catchup set each request updates pvclock data.
KVM_HC_CLOCK_PAIRING consumers such as ptp_kvm_x86 rely on tsc read on
host's side and do hypercall inside pvclock_read_retry loop leading to
infinite loop in such situation.

v3:
    Removed warn
    Changed return code to KVM_EFAULT
v2:
    Added warn

Signed-off-by: Anton Romanov <romanton@google.com>
---
 arch/x86/kvm/x86.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 7131d735b1ef..d0b31b115922 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -8945,6 +8945,13 @@ static int kvm_pv_clock_pairing(struct kvm_vcpu *vcpu, gpa_t paddr,
 	if (!kvm_get_walltime_and_clockread(&ts, &cycle))
 		return -KVM_EOPNOTSUPP;
 
+	/*
+	 * When tsc is in permanent catchup mode guests won't be able to use
+	 * pvclock_read_retry loop to get consistent view of pvclock
+	 */
+	if (vcpu->arch.tsc_always_catchup)
+		return -KVM_EFAULT;
+
 	clock_pairing.sec = ts.tv_sec;
 	clock_pairing.nsec = ts.tv_nsec;
 	clock_pairing.tsc = kvm_read_l1_tsc(vcpu, cycle);
-- 
2.35.1.265.g69c8d7142f-goog

Re: [PATCH v3] kvm: x86: Disable KVM_HC_CLOCK_PAIRING if tsc is in always catchup mode

[Paolo Bonzini]

On 2/16/22 19:26, Anton Romanov wrote:
> If vcpu has tsc_always_catchup set each request updates pvclock data.
> KVM_HC_CLOCK_PAIRING consumers such as ptp_kvm_x86 rely on tsc read on
> host's side and do hypercall inside pvclock_read_retry loop leading to
> infinite loop in such situation.
> 
> v3:
>      Removed warn
>      Changed return code to KVM_EFAULT
> v2:
>      Added warn
> 
> Signed-off-by: Anton Romanov <romanton@google.com>
> ---
>   arch/x86/kvm/x86.c | 7 +++++++
>   1 file changed, 7 insertions(+)
> 
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 7131d735b1ef..d0b31b115922 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -8945,6 +8945,13 @@ static int kvm_pv_clock_pairing(struct kvm_vcpu *vcpu, gpa_t paddr,
>   	if (!kvm_get_walltime_and_clockread(&ts, &cycle))
>   		return -KVM_EOPNOTSUPP;
>   
> +	/*
> +	 * When tsc is in permanent catchup mode guests won't be able to use
> +	 * pvclock_read_retry loop to get consistent view of pvclock
> +	 */
> +	if (vcpu->arch.tsc_always_catchup)
> +		return -KVM_EFAULT;
> +

KVM_EFAULT is for invalid addresses.  I changed it back to EOPNOTSUPP 
and queued the patch.

Paolo