[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH] xen/x86: Remap text/data/bss with appropriate permissions

>>> On 17.03.16 at 17:15, <andrew.cooper3@xxxxxxxxxx> wrote:
> On 17/03/16 15:32, Jan Beulich wrote:
>>>>> On 17.03.16 at 15:44, <andrew.cooper3@xxxxxxxxxx> wrote:
>>> On 17/03/16 14:31, Jan Beulich wrote:
>>>> Also - do we really want to make this code dependent on
>>>> map_pages_to_xen() not intermediately zapping the mappings
>>>> being changed?
>>> Do you mean "immediately"?
>> No.
>>> As far as I can tell, it is guaranteed to be safe, even when remapping
>>> the code section.  Updates to the live pagetables are using atomic
>>> writes, and I didn't spot a point which would end up with a transient
>>> non-present mapping.
>> But we may, at some point and for whatever reason, come to make
>> the function zap the mapping (i.e. clear the present bit), flush, and
>> only the re-establish the new mapping.
> This change is temporary until I can fix the legacy boot issue and
> reintroduce the proper 2M functionality.
> If someone in the future wants to change the behaviour of
> map_pages_to_xen() then we can reconsider.  However, I think it is
> unlikely that this will actually happen at all, and if it ever does, I
> hope to have already fixed the 2M alignment and deleted this change.
> This change is a big security improvement, and absolutely should be
> taken, especially as the current implementation of map_pages_to_xen() is
> safe.

I by no means intend to reject this change just because of this
aspect - I merely wanted to make the slight concern explicit.


Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.