[PATCH] KVM: Ensure all vcpus are consistent with in-kernel irqchip settings

[PATCH] KVM: Ensure all vcpus are consistent with in-kernel irqchip settings

[Avi Kivity]

If some vcpus are created before KVM_CREATE_IRQCHIP, then
irqchip_in_kernel() and vcpu->arch.apic will be inconsistent, leading
to potential NULL pointer dereferences.

Fix by:
- ensuring that no vcpus are installed when KVM_CREATE_IRQCHIP is called
- ensuring that a vcpu has an apic if it is installed after KVM_CREATE_IRQCHIP

This is somewhat long winded because vcpu->arch.apic is created without
kvm->lock held.

Based on earlier patch by Michael Ellerman.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/ia64/kvm/kvm-ia64.c |    5 +++++
 arch/x86/kvm/x86.c       |    8 ++++++++
 include/linux/kvm_host.h |    7 +++++++
 virt/kvm/kvm_main.c      |    4 ++++
 4 files changed, 24 insertions(+), 0 deletions(-)

diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index d8ddbba..f5104b7 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -1172,6 +1172,11 @@ static enum hrtimer_restart hlt_timer_fn(struct hrtimer *data)
 
 #define PALE_RESET_ENTRY    0x80000000ffffffb0UL
 
+bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu)
+{
+	return irqchip_in_kernel(vcpu->kcm) == (vcpu->arch.apic != NULL);
+}
+
 int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 {
 	struct kvm_vcpu *v;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 3ee008f..be9594a 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3199,6 +3199,9 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = -EEXIST;
 		if (kvm->arch.vpic)
 			goto create_irqchip_unlock;
+		r = -EINVAL;
+		if (atomic_read(&kvm->online_vcpus))
+			goto create_irqchip_unlock;
 		r = -ENOMEM;
 		vpic = kvm_create_pic(kvm);
 		if (vpic) {
@@ -6107,6 +6110,11 @@ void kvm_arch_check_processor_compat(void *rtn)
 	kvm_x86_ops->check_processor_compatibility(rtn);
 }
 
+bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu)
+{
+	return irqchip_in_kernel(vcpu->kvm) == (vcpu->arch.apic != NULL);
+}
+
 int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 {
 	struct page *page;
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 355e445..759fa26 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -805,6 +805,13 @@ static inline bool kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu)
 {
 	return vcpu->kvm->bsp_vcpu_id == vcpu->vcpu_id;
 }
+
+bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu);
+
+#else
+
+static bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu) { return true; }
+
 #endif
 
 #ifdef __KVM_HAVE_DEVICE_ASSIGNMENT
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index e4431ad..94e148e 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1651,6 +1651,10 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
 		goto vcpu_destroy;
 
 	mutex_lock(&kvm->lock);
+	if (!kvm_vcpu_compatible(vcpu)) {
+		r = -EINVAL;
+		goto unlock_vcpu_destroy;
+	}
 	if (atomic_read(&kvm->online_vcpus) == KVM_MAX_VCPUS) {
 		r = -EINVAL;
 		goto unlock_vcpu_destroy;
-- 
1.7.9

Re: [PATCH] KVM: Ensure all vcpus are consistent with in-kernel irqchip settings

[Michael Ellerman]

[-- Attachment #1: Type: text/plain, Size: 1049 bytes --]

On Mon, 2012-03-05 at 14:29 +0200, Avi Kivity wrote:
> If some vcpus are created before KVM_CREATE_IRQCHIP, then
> irqchip_in_kernel() and vcpu->arch.apic will be inconsistent, leading
> to potential NULL pointer dereferences.
> 
> Fix by:
> - ensuring that no vcpus are installed when KVM_CREATE_IRQCHIP is called
> - ensuring that a vcpu has an apic if it is installed after KVM_CREATE_IRQCHIP
> 
> This is somewhat long winded because vcpu->arch.apic is created without
> kvm->lock held.

Hi Avi,

Thanks for following up on this. This looks OK to me.

I wonder if we will end up needing to add other sanity tests at the same
point, ie. when we install the vcpu, in which case we might need a
generic "sanity hook". But better to keep it specific until we need
something generalised.

When we do irqchip-in-kernel on powerpc we'll need to rework the #ifdef
in kvm_host.h, because we don't want CONFIG_KVM_APIC_ARCHITECTURE, but
we will need our own kvm_vcpu_compatible(). But again we'll do that at
the time.

cheers


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] KVM: Ensure all vcpus are consistent with in-kernel irqchip settings

[Avi Kivity]

On 03/06/2012 07:30 AM, Michael Ellerman wrote:
> On Mon, 2012-03-05 at 14:29 +0200, Avi Kivity wrote:
> > If some vcpus are created before KVM_CREATE_IRQCHIP, then
> > irqchip_in_kernel() and vcpu->arch.apic will be inconsistent, leading
> > to potential NULL pointer dereferences.
> > 
> > Fix by:
> > - ensuring that no vcpus are installed when KVM_CREATE_IRQCHIP is called
> > - ensuring that a vcpu has an apic if it is installed after KVM_CREATE_IRQCHIP
> > 
> > This is somewhat long winded because vcpu->arch.apic is created without
> > kvm->lock held.
>
> Hi Avi,
>
> Thanks for following up on this. This looks OK to me.
>
> I wonder if we will end up needing to add other sanity tests at the same
> point, ie. when we install the vcpu, in which case we might need a
> generic "sanity hook". But better to keep it specific until we need
> something generalised.

Right, ARM will also have the same problem (with/without GIC).

> When we do irqchip-in-kernel on powerpc we'll need to rework the #ifdef
> in kvm_host.h, because we don't want CONFIG_KVM_APIC_ARCHITECTURE, but
> we will need our own kvm_vcpu_compatible(). But again we'll do that at
> the time.
>

-- 
error compiling committee.c: too many arguments to function