Security researchers from the Vrije Universiteit Amsterdam and ETH Zurich discovered a new security vulnerability in Intel's Xeon processors. The vulnerability concerns DDIO (Data-Direct I/O), a performance feature that was introduced in 2011 to let network cards and other peripherlas communicate directly with the CPU's last-level cache. By skipping the main system memory, this optimization results in increased performance, lower latency, and lower power consumption.
But unfortunately, DDIO can also enable the stealing of data. The researchers called the vulnerability NetCAT, which is short for Network Cache ATtack. NetCAT is primarily a concern for webhosting and cloud providers.
Now, researchers are warning that, in certain scenarios, attackers can abuse DDIO to obtain keystrokes and possibly other types of sensitive data that flow through the memory of vulnerable servers. The most serious form of attack can take place in data centers and cloud environments that have both DDIO and remote direct memory access enabled to allow servers to exchange data. A server leased by a malicious hacker could abuse the vulnerability to attack other customers. To prove their point, the researchers devised an attack that allows a server to steal keystrokes typed into the protected SSH (or secure shell session) established between another server and an application server.
More details can be read at ARS Technica. The researchers admit NetCAT is a rather complex attack but they recommend that in server settings with untrusted clients, where security matters more than performance, DDIO is disabled.