Bug 1392938 (CVE-2016-9381, xsa197)

ISSUE DESCRIPTION
=================

The compiler can emit optimizations in qemu which can lead to double
fetch vulnerabilities.  Specifically data on the rings shared between
qemu and the hypervisor (which the guest under control can obtain
mappings of) can be fetched twice (during which time the guest can
alter the contents) possibly leading to arbitrary code execution in
qemu.

IMPACT
======

Malicious administrators can exploit this vulnerability to take over
the qemu process, elevating its privilege to that of the qemu process.

In a system not using a device model stub domain (or other techniques
for deprivileging qemu), malicious guest administrators can thus
elevate their privilege to that of the host.

VULNERABLE SYSTEMS
==================

All Xen versions with all flavors of qemu are affected.

Only x86 HVM guests expose the vulnerability.  x86 PV guests do not
expose the vulnerability.

ARM systems are not vulnerable.

MITIGATION
==========

Running only PV guests will avoid the vulnerability.

Enabling stubdomains will mitigate this issue, by reducing the
escalation to only those privileges accorded to the service domain.
In a usual configuration, a service domain has only the privilege of
the guest, so this eliminates the vulnerability.

The vulnerability can be avoided if the guest kernel is controlled by
the host rather than guest administrator, provided that further steps
are taken to prevent the guest administrator from loading code into
the kernel (e.g. by disabling loadable modules etc) or from using
other mechanisms which allow them to run code at kernel privilege.

External References:

http://xenbits.xen.org/xsa/advisory-197.html

Acknowledgements:

Name: the Xen project
Upstream: yanghongke (Huawei Security Test Team)