If vmx_vcpu_up() failed, __vmxon() would generally not have got (successfully) executed, and in that case __vmxoff() will #UD. Additionally, any panic() during early resume (namely the tboot related one) would cause vmx_cpu_down() to get executed without vmx_cpu_up() having run before. Signed-off-by: Jan Beulich --- a/xen/arch/x86/hvm/vmx/vmcs.c +++ b/xen/arch/x86/hvm/vmx/vmcs.c @@ -71,6 +71,7 @@ bool_t cpu_has_vmx_ins_outs_instr_info _ static DEFINE_PER_CPU_READ_MOSTLY(struct vmcs_struct *, vmxon_region); static DEFINE_PER_CPU(struct vmcs_struct *, current_vmcs); static DEFINE_PER_CPU(struct list_head, active_vmcs_list); +static DEFINE_PER_CPU(bool_t, vmxon); static u32 vmcs_revision_id __read_mostly; @@ -519,6 +520,7 @@ int vmx_cpu_up(void) printk("CPU%d: unexpected VMXON failure\n", cpu); return -EINVAL; case 0: /* success */ + this_cpu(vmxon) = 1; break; default: BUG(); @@ -540,6 +542,9 @@ void vmx_cpu_down(void) struct list_head *active_vmcs_list = &this_cpu(active_vmcs_list); unsigned long flags; + if ( !this_cpu(vmxon) ) + return; + local_irq_save(flags); while ( !list_empty(active_vmcs_list) ) @@ -547,6 +552,7 @@ void vmx_cpu_down(void) struct vcpu, arch.hvm_vmx.active_list)); BUG_ON(!(read_cr4() & X86_CR4_VMXE)); + this_cpu(vmxon) = 0; __vmxoff(); local_irq_restore(flags);