KVM: MMU: Fix free memory accounting race in mmu_alloc_roots()
We drop the mmu lock between freeing memory and allocating the roots; this
allows some other vcpu to sneak in and allocate memory.
While the race is benign (resulting only in temporary overallocation, not oom)
it is simple and easy to fix by moving the freeing close to the allocation.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 4a02dee..d7aebaf 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2094,6 +2094,7 @@
root_gfn = 0;
}
spin_lock(&vcpu->kvm->mmu_lock);
+ kvm_mmu_free_some_pages(vcpu->kvm);
sp = kvm_mmu_get_page(vcpu, root_gfn, 0,
PT64_ROOT_LEVEL, direct,
ACC_ALL, NULL);
@@ -2124,6 +2125,7 @@
root_gfn = i << 30;
}
spin_lock(&vcpu->kvm->mmu_lock);
+ kvm_mmu_free_some_pages(vcpu->kvm);
sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30,
PT32_ROOT_LEVEL, direct,
ACC_ALL, NULL);
@@ -2496,9 +2498,6 @@
r = mmu_topup_memory_caches(vcpu);
if (r)
goto out;
- spin_lock(&vcpu->kvm->mmu_lock);
- kvm_mmu_free_some_pages(vcpu);
- spin_unlock(&vcpu->kvm->mmu_lock);
r = mmu_alloc_roots(vcpu);
spin_lock(&vcpu->kvm->mmu_lock);
mmu_sync_roots(vcpu);