* [PATCH] KVM: x86: Return "unsupported" instead of "invalid" on access to unsupported PV MSR
@ 2025-12-30 20:59 Sean Christopherson
2026-01-05 18:25 ` Jim Mattson
2026-01-12 17:38 ` Sean Christopherson
0 siblings, 2 replies; 3+ messages in thread
From: Sean Christopherson @ 2025-12-30 20:59 UTC (permalink / raw)
To: Sean Christopherson, Paolo Bonzini; +Cc: kvm, linux-kernel
Return KVM_MSR_RET_UNSUPPORTED instead of '1' (which for all intents and
purposes means "invalid") when rejecting accesses to KVM PV MSRs to adhere
to KVM's ABI of allowing host reads and writes of '0' to MSRs that are
advertised to userspace via KVM_GET_MSR_INDEX_LIST, even if the vCPU model
doesn't support the MSR.
E.g. running a QEMU VM with
-cpu host,-kvmclock,kvm-pv-enforce-cpuid
yields:
qemu: error: failed to set MSR 0x12 to 0x0
qemu: target/i386/kvm/kvm.c:3301: kvm_buf_set_msrs:
Assertion `ret == cpu->kvm_msr_buf->nmsrs' failed.
Fixes: 66570e966dd9 ("kvm: x86: only provide PV features if enabled in guest's CPUID")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
arch/x86/kvm/x86.c | 40 ++++++++++++++++++++--------------------
1 file changed, 20 insertions(+), 20 deletions(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index ff8812f3a129..dd0b5be1514d 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -4096,47 +4096,47 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
break;
case MSR_KVM_WALL_CLOCK_NEW:
if (!guest_pv_has(vcpu, KVM_FEATURE_CLOCKSOURCE2))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
vcpu->kvm->arch.wall_clock = data;
kvm_write_wall_clock(vcpu->kvm, data, 0);
break;
case MSR_KVM_WALL_CLOCK:
if (!guest_pv_has(vcpu, KVM_FEATURE_CLOCKSOURCE))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
vcpu->kvm->arch.wall_clock = data;
kvm_write_wall_clock(vcpu->kvm, data, 0);
break;
case MSR_KVM_SYSTEM_TIME_NEW:
if (!guest_pv_has(vcpu, KVM_FEATURE_CLOCKSOURCE2))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
kvm_write_system_time(vcpu, data, false, msr_info->host_initiated);
break;
case MSR_KVM_SYSTEM_TIME:
if (!guest_pv_has(vcpu, KVM_FEATURE_CLOCKSOURCE))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
kvm_write_system_time(vcpu, data, true, msr_info->host_initiated);
break;
case MSR_KVM_ASYNC_PF_EN:
if (!guest_pv_has(vcpu, KVM_FEATURE_ASYNC_PF))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
if (kvm_pv_enable_async_pf(vcpu, data))
return 1;
break;
case MSR_KVM_ASYNC_PF_INT:
if (!guest_pv_has(vcpu, KVM_FEATURE_ASYNC_PF_INT))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
if (kvm_pv_enable_async_pf_int(vcpu, data))
return 1;
break;
case MSR_KVM_ASYNC_PF_ACK:
if (!guest_pv_has(vcpu, KVM_FEATURE_ASYNC_PF_INT))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
if (data & 0x1) {
/*
* Pairs with the smp_mb__after_atomic() in
@@ -4149,7 +4149,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
break;
case MSR_KVM_STEAL_TIME:
if (!guest_pv_has(vcpu, KVM_FEATURE_STEAL_TIME))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
if (unlikely(!sched_info_on()))
return 1;
@@ -4167,7 +4167,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
break;
case MSR_KVM_PV_EOI_EN:
if (!guest_pv_has(vcpu, KVM_FEATURE_PV_EOI))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
if (kvm_lapic_set_pv_eoi(vcpu, data, sizeof(u8)))
return 1;
@@ -4175,7 +4175,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
case MSR_KVM_POLL_CONTROL:
if (!guest_pv_has(vcpu, KVM_FEATURE_POLL_CONTROL))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
/* only enable bit supported */
if (data & (-1ULL << 1))
@@ -4476,61 +4476,61 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
break;
case MSR_KVM_WALL_CLOCK:
if (!guest_pv_has(vcpu, KVM_FEATURE_CLOCKSOURCE))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->kvm->arch.wall_clock;
break;
case MSR_KVM_WALL_CLOCK_NEW:
if (!guest_pv_has(vcpu, KVM_FEATURE_CLOCKSOURCE2))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->kvm->arch.wall_clock;
break;
case MSR_KVM_SYSTEM_TIME:
if (!guest_pv_has(vcpu, KVM_FEATURE_CLOCKSOURCE))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->arch.time;
break;
case MSR_KVM_SYSTEM_TIME_NEW:
if (!guest_pv_has(vcpu, KVM_FEATURE_CLOCKSOURCE2))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->arch.time;
break;
case MSR_KVM_ASYNC_PF_EN:
if (!guest_pv_has(vcpu, KVM_FEATURE_ASYNC_PF))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->arch.apf.msr_en_val;
break;
case MSR_KVM_ASYNC_PF_INT:
if (!guest_pv_has(vcpu, KVM_FEATURE_ASYNC_PF_INT))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->arch.apf.msr_int_val;
break;
case MSR_KVM_ASYNC_PF_ACK:
if (!guest_pv_has(vcpu, KVM_FEATURE_ASYNC_PF_INT))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = 0;
break;
case MSR_KVM_STEAL_TIME:
if (!guest_pv_has(vcpu, KVM_FEATURE_STEAL_TIME))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->arch.st.msr_val;
break;
case MSR_KVM_PV_EOI_EN:
if (!guest_pv_has(vcpu, KVM_FEATURE_PV_EOI))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->arch.pv_eoi.msr_val;
break;
case MSR_KVM_POLL_CONTROL:
if (!guest_pv_has(vcpu, KVM_FEATURE_POLL_CONTROL))
- return 1;
+ return KVM_MSR_RET_UNSUPPORTED;
msr_info->data = vcpu->arch.msr_kvm_poll_control;
break;
base-commit: 9448598b22c50c8a5bb77a9103e2d49f134c9578
--
2.52.0.351.gbe84eed79e-goog
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] KVM: x86: Return "unsupported" instead of "invalid" on access to unsupported PV MSR
2025-12-30 20:59 [PATCH] KVM: x86: Return "unsupported" instead of "invalid" on access to unsupported PV MSR Sean Christopherson
@ 2026-01-05 18:25 ` Jim Mattson
2026-01-12 17:38 ` Sean Christopherson
1 sibling, 0 replies; 3+ messages in thread
From: Jim Mattson @ 2026-01-05 18:25 UTC (permalink / raw)
To: Sean Christopherson; +Cc: Paolo Bonzini, kvm, linux-kernel
On Tue, Dec 30, 2025 at 12:59 PM Sean Christopherson <seanjc@google.com> wrote:
>
> Return KVM_MSR_RET_UNSUPPORTED instead of '1' (which for all intents and
> purposes means "invalid") when rejecting accesses to KVM PV MSRs to adhere
> to KVM's ABI of allowing host reads and writes of '0' to MSRs that are
> advertised to userspace via KVM_GET_MSR_INDEX_LIST, even if the vCPU model
> doesn't support the MSR.
>
> E.g. running a QEMU VM with
>
> -cpu host,-kvmclock,kvm-pv-enforce-cpuid
>
> yields:
>
> qemu: error: failed to set MSR 0x12 to 0x0
> qemu: target/i386/kvm/kvm.c:3301: kvm_buf_set_msrs:
> Assertion `ret == cpu->kvm_msr_buf->nmsrs' failed.
>
> Fixes: 66570e966dd9 ("kvm: x86: only provide PV features if enabled in guest's CPUID")
> Cc: stable@vger.kernel.org
> Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] KVM: x86: Return "unsupported" instead of "invalid" on access to unsupported PV MSR
2025-12-30 20:59 [PATCH] KVM: x86: Return "unsupported" instead of "invalid" on access to unsupported PV MSR Sean Christopherson
2026-01-05 18:25 ` Jim Mattson
@ 2026-01-12 17:38 ` Sean Christopherson
1 sibling, 0 replies; 3+ messages in thread
From: Sean Christopherson @ 2026-01-12 17:38 UTC (permalink / raw)
To: Sean Christopherson, Paolo Bonzini; +Cc: kvm, linux-kernel
On Tue, 30 Dec 2025 12:59:48 -0800, Sean Christopherson wrote:
> Return KVM_MSR_RET_UNSUPPORTED instead of '1' (which for all intents and
> purposes means "invalid") when rejecting accesses to KVM PV MSRs to adhere
> to KVM's ABI of allowing host reads and writes of '0' to MSRs that are
> advertised to userspace via KVM_GET_MSR_INDEX_LIST, even if the vCPU model
> doesn't support the MSR.
>
> E.g. running a QEMU VM with
>
> [...]
Applied to kvm-x86 misc, thanks!
[1/1] KVM: x86: Return "unsupported" instead of "invalid" on access to unsupported PV MSR
https://github.com/kvm-x86/linux/commit/5bb9ac186512
--
https://github.com/kvm-x86/linux/tree/next
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2026-01-12 17:40 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-12-30 20:59 [PATCH] KVM: x86: Return "unsupported" instead of "invalid" on access to unsupported PV MSR Sean Christopherson
2026-01-05 18:25 ` Jim Mattson
2026-01-12 17:38 ` Sean Christopherson
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®