Re: [Xen-devel] xsm_pci_config_permission() placement

[Daniel De Graaf <dgdegra@xxxxxxxxxxxxx>]

On 05/10/15 10:04, Jan Beulich wrote:
> Daniel,
>
> now that we have MMCFG write intercepts in place, wouldn't it make
> sense to move that hook invocation into pci_conf_write_intercept()
> for the write case, so that it also covers MMCFG-based writes?
>
> Thanks, Jan
Yes, I think this would be a good idea.  The read check wouldn't be
covered, but I think the protections of that XSM hook are only really
important on writes.  The read check could also be duplicated as a
more conservative alternative.

--
Daniel De Graaf
National Security Agency

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel