KVM: x86: Disallow hypercalls for guest callers in rings > 0 [CVE-2009-3290]
authorJan Kiszka <jan.kiszka@siemens.com>
Wed, 7 Oct 2009 21:40:32 +0000 (17:40 -0400)
committerGreg Kroah-Hartman <gregkh@suse.de>
Mon, 12 Oct 2009 18:33:20 +0000 (11:33 -0700)
[ backport to 2.6.27 by Chuck Ebbert <cebbert@redhat.com> ]

commit 07708c4af1346ab1521b26a202f438366b7bcffd upstream.

So far unprivileged guest callers running in ring 3 can issue, e.g., MMU
hypercalls. Normally, such callers cannot provide any hand-crafted MMU
command structure as it has to be passed by its physical address, but
they can still crash the guest kernel by passing random addresses.

To close the hole, this patch considers hypercalls valid only if issued
from guest ring 0. This may still be relaxed on a per-hypercall base in
the future once required.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Cc: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
arch/x86/kvm/x86.c
include/linux/kvm_para.h

index f7c7142d89ae2aa214c8d32faf936174249396d8..60ebfd7eb0cf84818d1eec81b7553e82a60e07d9 100644 (file)
@@ -2571,6 +2571,11 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
                a3 &= 0xFFFFFFFF;
        }
 
+       if (kvm_x86_ops->get_cpl(vcpu) != 0) {
+               ret = -KVM_EPERM;
+               goto out;
+       }
+
        switch (nr) {
        case KVM_HC_VAPIC_POLL_IRQ:
                ret = 0;
@@ -2582,6 +2587,7 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
                ret = -KVM_ENOSYS;
                break;
        }
+out:
        vcpu->arch.regs[VCPU_REGS_RAX] = ret;
        kvm_x86_ops->decache_regs(vcpu);
        ++vcpu->stat.hypercalls;
index 3ddce03766caf5796f84f71bd7462f117c9a064a..d73109243fda075cdd6ef6be7860b917c5612787 100644 (file)
@@ -13,6 +13,7 @@
 #define KVM_ENOSYS             1000
 #define KVM_EFAULT             EFAULT
 #define KVM_E2BIG              E2BIG
+#define KVM_EPERM              EPERM
 
 #define KVM_HC_VAPIC_POLL_IRQ          1
 #define KVM_HC_MMU_OP                  2