Skip to content

Commit

Permalink
KVM: Fix off by one in kvm_for_each_vcpu iteration
Browse files Browse the repository at this point in the history
This patch avoids gcc issuing the following warning when KVM_MAX_VCPUS=1:
warning: array subscript is above array bounds

kvm_for_each_vcpu currently checks to see if the index for the vcpu is
valid /after/ loading it. We don't run into problems because the address
is still inside the enclosing struct kvm and we never deference or write
to it, so this isn't a security issue.

The warning occurs when KVM_MAX_VCPUS=1 because the increment portion of
the loop will *always* cause the loop to load an invalid location since
++idx will always be > 0.

This patch moves the load so that the check occurs before the load and
we don't run into the compiler warning.

Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
  • Loading branch information
Jeff Mahoney authored and Avi Kivity committed May 11, 2011
1 parent 71f9833 commit b42fc3c
Showing 1 changed file with 4 additions and 3 deletions.
7 changes: 4 additions & 3 deletions include/linux/kvm_host.h
Original file line number Diff line number Diff line change
Expand Up @@ -296,9 +296,10 @@ static inline struct kvm_vcpu *kvm_get_vcpu(struct kvm *kvm, int i)
}

#define kvm_for_each_vcpu(idx, vcpup, kvm) \
for (idx = 0, vcpup = kvm_get_vcpu(kvm, idx); \
idx < atomic_read(&kvm->online_vcpus) && vcpup; \
vcpup = kvm_get_vcpu(kvm, ++idx))
for (idx = 0; \
idx < atomic_read(&kvm->online_vcpus) && \
(vcpup = kvm_get_vcpu(kvm, idx)) != NULL; \
idx++)

int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id);
void kvm_vcpu_uninit(struct kvm_vcpu *vcpu);
Expand Down

0 comments on commit b42fc3c

Please sign in to comment.