Patchwork APPLIED: ACK: APPLIED: [PATCH 1/1] [CVE-2012-1601] [ONEIRIC] [NATTY] KVM: Ensure all vcpus are consistent with in-kernel irqchip settings

login
register
mail settings
Submitter Tim Gardner
Date April 26, 2012, 1:39 p.m.
Message ID <4F99500B.7060907@canonical.com>
Download mbox | patch
Permalink /patch/155268/
State New
Headers show

Comments

Tim Gardner - April 26, 2012, 1:39 p.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

On 04/26/2012 07:28 AM, Stefan Bader wrote:
> On 26.04.2012 15:24, Tim Gardner wrote:
>> On 04/26/2012 06:57 AM, Stefan Bader wrote:
>>> On 26.04.2012 14:52, Tim Gardner wrote:
>>>> So this patch causes a compile error. I've attached the
>>>> correction. I suggest that it just be squashed into the
>>>> original commit.
>>>> 
>>>> rtg
>>>> 
>>>> 
>>>> 
>>>> 
>>> Maybe we want to pick up the patch that introduced it:
>>> 
>>> 
>>> commit d780592b99d7d8a5ff905f6bacca519d4a342c76 Author: Jan
>>> Kiszka <jan.kiszka@siemens.com> Date:   Mon May 23 10:33:05
>>> 2011 +0200
>>> 
>>> KVM: Clean up error handling during VCPU creation
>>> 
>>> Not sure all the jumps to vcpu_destroy before would otherwise
>>> do the right thing...
>>> 
>>> 
>>> 
>>> 
>> 
>> OK, here is the real solution.
>> d780592b99d7d8a5ff905f6bacca519d4a342c76 and
>> 3e515705a1f46beb1c942bb8043c16f8ac7b1e9e are clean cherry-picks.
>> No backport required.
>> 
>> rtg
> 
> fwiw... and likely going back...
> 

Straight cherry-picks
- -- 
Tim Gardner tim.gardner@canonical.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
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=gS4s
-----END PGP SIGNATURE-----

Patch

From 08a5694c9cce61f44bc4909ecb947586105374a3 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@redhat.com>
Date: Mon, 5 Mar 2012 14:23:29 +0200
Subject: [PATCH 2/2] KVM: Ensure all vcpus are consistent with in-kernel
 irqchip settings

BugLink: http://bugs.launchpad.net/bugs/971685

CVE-2012-1601

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>
(cherry picked from commit 3e515705a1f46beb1c942bb8043c16f8ac7b1e9e)

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Acked-by: Stefan Bader <stefan.bader@canonical.com>
Acked-by: Herton Krzesinski <herton.krzesinski@canonical.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(+)

diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index 8213efe..a874213 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -1168,6 +1168,11 @@  out:
 
 #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 d829351..bf59dfc 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3439,6 +3439,9 @@  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) {
@@ -6218,6 +6221,11 @@  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 31ebb59..d70a250 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -730,6 +730,13 @@  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 inline 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 7d6b8e3..f9013d1 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1618,6 +1618,10 @@  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;
-- 
1.7.9.5