diff options
author | Avi Kivity <avi@redhat.com> | 2009-03-29 16:31:25 +0300 |
---|---|---|
committer | Avi Kivity <avi@redhat.com> | 2009-04-22 13:52:09 +0300 |
commit | 99894a799f09cf9e28296bb16e75bd5830fd2c4e (patch) | |
tree | 33fa687fa78b3c8b89d032ca11b89443296449a2 | |
parent | 091069740304c979f957ceacec39c461d0192158 (diff) | |
download | op-kernel-dev-99894a799f09cf9e28296bb16e75bd5830fd2c4e.zip op-kernel-dev-99894a799f09cf9e28296bb16e75bd5830fd2c4e.tar.gz |
KVM: MMU: Fix off-by-one calculating large page count
The large page initialization code concludes there are two large pages spanned
by a slot covering 1 (small) page starting at gfn 1. This is incorrect, and
also results in incorrect write_count initialization in some cases (base = 1,
npages = 513 for example).
Cc: stable@kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
-rw-r--r-- | virt/kvm/kvm_main.c | 8 |
1 files changed, 3 insertions, 5 deletions
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 605697e..28d693a 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -920,6 +920,7 @@ int __kvm_set_memory_region(struct kvm *kvm, int r; gfn_t base_gfn; unsigned long npages; + int largepages; unsigned long i; struct kvm_memory_slot *memslot; struct kvm_memory_slot old, new; @@ -995,11 +996,8 @@ int __kvm_set_memory_region(struct kvm *kvm, new.userspace_addr = 0; } if (npages && !new.lpage_info) { - int largepages = npages / KVM_PAGES_PER_HPAGE; - if (npages % KVM_PAGES_PER_HPAGE) - largepages++; - if (base_gfn % KVM_PAGES_PER_HPAGE) - largepages++; + largepages = 1 + (base_gfn + npages - 1) / KVM_PAGES_PER_HPAGE; + largepages -= base_gfn / KVM_PAGES_PER_HPAGE; new.lpage_info = vmalloc(largepages * sizeof(*new.lpage_info)); |