Load Value Injection (LVI) speculative side channel

ISSUE DESCRIPTION

This is very closely related to the Microarchitectural Data Sampling vulnerabilities from May 2019.
Please see <a href=“https://xenbits.xen.org/xsa/advisory-297.html”>https://xenbits.xen.org/xsa/advisory-297.html</a> for details about MDS.
A new way of using the micro-architectural details behind MDS has been identified. Instead of simply trying to sample data from a different privilege context, an attacker can arrange for poisoned data to be consumed (speculatively) in a victim context.
This expands the range of tools by which an attacker can manipulate speculation in the victim context to leak data via a side channel.
For more details, see: <a href=“https://software.intel.com/security-software-guidance/insights/deep-dive-load-value-injection”>https://software.intel.com/security-software-guidance/insights/deep-dive-load-value-injection</a>

IMPACT

An attacker, which could include a malicious untrusted user process on a trusted guest, or an untrusted guest, can potentially cause a victim context (process, or guest, or guest kernel, or hypervisor) to leak secrets available to it.

VULNERABLE SYSTEMS

Systems running all versions of Xen are affected.
Only x86 processors are vulnerable. ARM processors are not believed to be vulnerable.
Only Intel based processors are potentially affected. Processors from other manufacturers (e.g. AMD) are not believed to be vulnerable.
Please consult the Intel Security Advisory for details on the affected processors.