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

[Jan Beulich <jbeulich@xxxxxxxx>]

On 27.12.2019 16:26, Andrew Cooper wrote:
> On 20/12/2019 13:50, 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>
> 
> I've only just spotted, but this is a semantic change to the guest. 
> Previously, base with ents=0 would be preserved via arch_get_info_guest().

I've done (extended from v1) this upon your request; I did notice
this side effect of the change. This is (partly) why I've made an
adjustment to arch_get_info_guest() in the first place.

> Finally, a similar concern about changes which a guest can't actually
> make use of, even if this one seems rather more minor.

Like for the GDT case, the goal isn't so much to allow guests more
relaxed behavior (albeit for ones not caring about being compatible
with older Xen this is still a secondary goal), but to get behavior
in Xen into an overall more consistent shape.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel