Commit 338c7dbadd26 ("KVM: Improve create VCPU parameter (CVE-2013-4587)")
introduced a check to prevent potential kernel memory corruption in case
the vcpu id is too great.
Unfortunately this check assumes vcpu ids grow in sequence with a common
difference of 1, which is wrong: archs are free to use vcpu id as they fit.
For example, QEMU originated vcpu ids for PowerPC cpus running in boot3s_hv
mode, can grow with a common difference of 2, 4 or 8: if KVM_MAX_VCPUS is
1024, guests may be limited down to 128 vcpus on POWER8.
This means the check does not belong here and should be moved to some arch
specific function: kvm_arch_vcpu_create() looks like a good candidate.
ARM and s390 already have such a check.
I could not spot any path in the PowerPC or common KVM code where a vcpu
id is used as described in the above commit: I believe PowerPC can live
without this check.
In the end, this patch simply moves the check to MIPS and x86.
Signed-off-by: Greg Kurz <[email protected]>
---
v2: move kzalloc() after the check in MIPS to avoid memory leak
arch/mips/kvm/mips.c | 5 ++++-
arch/x86/kvm/x86.c | 3 +++
virt/kvm/kvm_main.c | 3 ---
3 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c
index 70ef1a43c114..c1ce9d44d685 100644
--- a/arch/mips/kvm/mips.c
+++ b/arch/mips/kvm/mips.c
@@ -248,9 +248,12 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
int err, size, offset;
void *gebase;
int i;
+ struct kvm_vcpu *vcpu;
- struct kvm_vcpu *vcpu = kzalloc(sizeof(struct kvm_vcpu), GFP_KERNEL);
+ if (id >= KVM_MAX_VCPUS)
+ return -EINVAL;
+ vcpu = kzalloc(sizeof(struct kvm_vcpu), GFP_KERNEL);
if (!vcpu) {
err = -ENOMEM;
goto out;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 9b7798c7b210..f705d57b12ed 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7358,6 +7358,9 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
{
struct kvm_vcpu *vcpu;
+ if (id >= KVM_MAX_VCPUS)
+ return -EINVAL;
+
if (check_tsc_unstable() && atomic_read(&kvm->online_vcpus) != 0)
printk_once(KERN_WARNING
"kvm: SMP vm created on host with unstable TSC; "
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 4fd482fb9260..6b6cca3cb488 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -2272,9 +2272,6 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
int r;
struct kvm_vcpu *vcpu;
- if (id >= KVM_MAX_VCPUS)
- return -EINVAL;
-
vcpu = kvm_arch_vcpu_create(kvm, id);
if (IS_ERR(vcpu))
return PTR_ERR(vcpu);
Forget this post, I'll send a v3. Thanks to the kbuild test robot ! :)
On Wed, 20 Apr 2016 17:26:48 +0200
Greg Kurz <[email protected]> wrote:
> Commit 338c7dbadd26 ("KVM: Improve create VCPU parameter (CVE-2013-4587)")
> introduced a check to prevent potential kernel memory corruption in case
> the vcpu id is too great.
>
> Unfortunately this check assumes vcpu ids grow in sequence with a common
> difference of 1, which is wrong: archs are free to use vcpu id as they fit.
> For example, QEMU originated vcpu ids for PowerPC cpus running in boot3s_hv
> mode, can grow with a common difference of 2, 4 or 8: if KVM_MAX_VCPUS is
> 1024, guests may be limited down to 128 vcpus on POWER8.
>
> This means the check does not belong here and should be moved to some arch
> specific function: kvm_arch_vcpu_create() looks like a good candidate.
>
> ARM and s390 already have such a check.
>
> I could not spot any path in the PowerPC or common KVM code where a vcpu
> id is used as described in the above commit: I believe PowerPC can live
> without this check.
>
> In the end, this patch simply moves the check to MIPS and x86.
>
> Signed-off-by: Greg Kurz <[email protected]>
> ---
> v2: move kzalloc() after the check in MIPS to avoid memory leak
>
> arch/mips/kvm/mips.c | 5 ++++-
> arch/x86/kvm/x86.c | 3 +++
> virt/kvm/kvm_main.c | 3 ---
> 3 files changed, 7 insertions(+), 4 deletions(-)
>
> diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c
> index 70ef1a43c114..c1ce9d44d685 100644
> --- a/arch/mips/kvm/mips.c
> +++ b/arch/mips/kvm/mips.c
> @@ -248,9 +248,12 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
> int err, size, offset;
> void *gebase;
> int i;
> + struct kvm_vcpu *vcpu;
>
> - struct kvm_vcpu *vcpu = kzalloc(sizeof(struct kvm_vcpu), GFP_KERNEL);
> + if (id >= KVM_MAX_VCPUS)
> + return -EINVAL;
>
> + vcpu = kzalloc(sizeof(struct kvm_vcpu), GFP_KERNEL);
> if (!vcpu) {
> err = -ENOMEM;
> goto out;
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 9b7798c7b210..f705d57b12ed 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -7358,6 +7358,9 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
> {
> struct kvm_vcpu *vcpu;
>
> + if (id >= KVM_MAX_VCPUS)
> + return -EINVAL;
> +
> if (check_tsc_unstable() && atomic_read(&kvm->online_vcpus) != 0)
> printk_once(KERN_WARNING
> "kvm: SMP vm created on host with unstable TSC; "
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 4fd482fb9260..6b6cca3cb488 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -2272,9 +2272,6 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
> int r;
> struct kvm_vcpu *vcpu;
>
> - if (id >= KVM_MAX_VCPUS)
> - return -EINVAL;
> -
> vcpu = kvm_arch_vcpu_create(kvm, id);
> if (IS_ERR(vcpu))
> return PTR_ERR(vcpu);
>
>
On Wed, 20 Apr 2016 17:26:48 +0200
Greg Kurz <[email protected]> wrote:
> Commit 338c7dbadd26 ("KVM: Improve create VCPU parameter (CVE-2013-4587)")
> introduced a check to prevent potential kernel memory corruption in case
> the vcpu id is too great.
>
> Unfortunately this check assumes vcpu ids grow in sequence with a common
> difference of 1, which is wrong: archs are free to use vcpu id as they fit.
> For example, QEMU originated vcpu ids for PowerPC cpus running in boot3s_hv
> mode, can grow with a common difference of 2, 4 or 8: if KVM_MAX_VCPUS is
> 1024, guests may be limited down to 128 vcpus on POWER8.
>
> This means the check does not belong here and should be moved to some arch
> specific function: kvm_arch_vcpu_create() looks like a good candidate.
>
> ARM and s390 already have such a check.
>
> I could not spot any path in the PowerPC or common KVM code where a vcpu
> id is used as described in the above commit: I believe PowerPC can live
> without this check.
>
> In the end, this patch simply moves the check to MIPS and x86.
The concept makes sense to me...
>
> Signed-off-by: Greg Kurz <[email protected]>
> ---
> v2: move kzalloc() after the check in MIPS to avoid memory leak
>
> arch/mips/kvm/mips.c | 5 ++++-
> arch/x86/kvm/x86.c | 3 +++
> virt/kvm/kvm_main.c | 3 ---
> 3 files changed, 7 insertions(+), 4 deletions(-)
>
> diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c
> index 70ef1a43c114..c1ce9d44d685 100644
> --- a/arch/mips/kvm/mips.c
> +++ b/arch/mips/kvm/mips.c
> @@ -248,9 +248,12 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
> int err, size, offset;
> void *gebase;
> int i;
> + struct kvm_vcpu *vcpu;
>
> - struct kvm_vcpu *vcpu = kzalloc(sizeof(struct kvm_vcpu), GFP_KERNEL);
> + if (id >= KVM_MAX_VCPUS)
> + return -EINVAL;
...but you need to use ERR_PTR().
>
> + vcpu = kzalloc(sizeof(struct kvm_vcpu), GFP_KERNEL);
> if (!vcpu) {
> err = -ENOMEM;
> goto out;
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 9b7798c7b210..f705d57b12ed 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -7358,6 +7358,9 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
> {
> struct kvm_vcpu *vcpu;
>
> + if (id >= KVM_MAX_VCPUS)
> + return -EINVAL;
Here as well.
> +
> if (check_tsc_unstable() && atomic_read(&kvm->online_vcpus) != 0)
> printk_once(KERN_WARNING
> "kvm: SMP vm created on host with unstable TSC; "