Re: [PATCH v2 2/3] x86: relax LDT check in arch_set_info_guest()

[Roger Pau Monné <roger.pau@xxxxxxxxxx>]

On Tue, May 19, 2020 at 11:12:49AM +0200, Jan Beulich wrote:
> On 19.05.2020 11:02, Roger Pau Monné wrote:
> > On Fri, Dec 20, 2019 at 02:50:06PM +0100, Jan Beulich wrote:
> >> It is wrong for us to check the base address when there's no LDT in the
> >> first place. Once we don't do this check anymore we can also set the
> >> base address to a non-canonical value when the LDT is empty.
> >>
> >> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> > 
> > Reviewed-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
> 
> Thanks.
> 
> > I wonder if a ldt_ents check should also be added to
> > pv_map_ldt_shadow_page in order to avoid trying to get the mapping of
> > the LDT.
> 
> We already have
> 
>     if ( unlikely((offset >> 3) >= curr->arch.pv.ldt_ents) )
>     {
>         ASSERT_UNREACHABLE();
>         return false;
>     }
> 
> What else do you mean?

Oh, I've missed that. I was searching for a check before accessing
ldt_base, which is done at initialization time. That's indeed fine.

Roger.