KVM: avoid NULL pointer dereference in kvm_dirty_ring_push

kvm_vcpu_release() will call kvm_dirty_ring_free(), freeing
ring->dirty_gfns and setting it to NULL.  Afterwards, it calls
kvm_arch_vcpu_destroy().

However, if closing the file descriptor races with KVM_RUN in such away
that vcpu->arch.st.preempted == 0, the following call stack leads to a
NULL pointer dereference in kvm_dirty_run_push():

 mark_page_dirty_in_slot+0x192/0x270 arch/x86/kvm/../../../virt/kvm/kvm_main.c:3171
 kvm_steal_time_set_preempted arch/x86/kvm/x86.c:4600 [inline]
 kvm_arch_vcpu_put+0x34e/0x5b0 arch/x86/kvm/x86.c:4618
 vcpu_put+0x1b/0x70 arch/x86/kvm/../../../virt/kvm/kvm_main.c:211
 vmx_free_vcpu+0xcb/0x130 arch/x86/kvm/vmx/vmx.c:6985
 kvm_arch_vcpu_destroy+0x76/0x290 arch/x86/kvm/x86.c:11219
 kvm_vcpu_destroy arch/x86/kvm/../../../virt/kvm/kvm_main.c:441 [inline]

The fix is to release the dirty page ring after kvm_arch_vcpu_destroy
has run.

Reported-by: Qiuhao Li <[email protected]>
Reported-by: Gaoning Pan <[email protected]>
Reported-by: Yongkang Jia <[email protected]>
Cc: [email protected]
Signed-off-by: Paolo Bonzini <[email protected]>

virt/kvm/kvm_main.c

@@ -434,8 +434,8 @@ static void kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)
 
 static void kvm_vcpu_destroy(struct kvm_vcpu *vcpu)
 {
-	kvm_dirty_ring_free(&vcpu->dirty_ring);
 	kvm_arch_vcpu_destroy(vcpu);
+	kvm_dirty_ring_free(&vcpu->dirty_ring);
 
 	/*
 	 * No need for rcu_read_lock as VCPU_RUN is the only place that changes