Re: [Xen-devel] [PATCH RFC v1] x86/emulate: Send vm_event form emulate

[Roger Pau Monné <royger@xxxxxxxxxxx>]

On Thu, Jan 10, 2019 at 5:40 PM Alexandru Stefan ISAILA
<aisaila@xxxxxxxxxxxxxxx> wrote:
>
>
> >>>
> >>> Newline.
> >>>
> >>>> +    default:
> >>>> +        return false;
> >>>> +    }
> >>>
> >>> I'm not sure the switch is needed, you can't have a PFEC_write_access
> >>> for example if the p2m type is p2m_access_w or p2m_access_rw, hence it
> >>> seems like it could be simplified to only take the pfec into account?
> >>
> >> It is possible to have PFEC_write_access and p2m_access_w but then there
> >> is no violation and the event will not be sent.
> >
> > I'm not sure I follow. PFEC is a way to signal a page fault, hence it
> > would make no sense to for example get PFEC_write_access if the access
> > is p2m_access_w?
> >
> Yes, normally pfec (page fault error code) does what the name implies
> but in this case the pfec var is constructed in the caller function.
>
> Let's take for example hvmemul_write() pfec here has always
> PFEC_write_access so every emulated write will have this no matter if is
> a fault or not and that is what I am checking. If in fact there is a
> fault then the event is sent.

OK, this seems weird but you are right that hvmemul_write will always
pass PFEC_write_access to hvmemul_map_linear_addr.

Roger.

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