Re: [PATCH 03/10] x86/shadow: monitor table is HVM-only

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

On 17/04/2020 15:26, Jan Beulich wrote:
> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
>
> --- a/xen/arch/x86/mm/shadow/common.c
> +++ b/xen/arch/x86/mm/shadow/common.c
> @@ -2376,7 +2376,6 @@ void sh_reset_l3_up_pointers(struct vcpu
>  static void sh_update_paging_modes(struct vcpu *v)
>  {
>      struct domain *d = v->domain;
> -    const struct paging_mode *old_mode = v->arch.paging.mode;
>  
>      ASSERT(paging_locked_by_me(d));
>  
> @@ -2421,11 +2420,14 @@ static void sh_update_paging_modes(struc
>      if ( v->arch.paging.mode )
>          v->arch.paging.mode->shadow.detach_old_tables(v);
>  
> +#ifdef CONFIG_HVM
>      if ( !is_pv_domain(d) )
>      {
>          ///
>          /// HVM guest
>          ///

Can we drop this comment while we're here?  The ifdef and
!is_pv_domain() are crystal clear.

~Andrew