Researchers have extracted the secret key that encrypts the update to classify Intel CPUs, a feat that could have widespread consequences for the way chips are used and, possibly, the way they are protected.
The key makes it possible to decrypt the microcode update that Intel provides to fix security vulnerabilities and other types of bugs. Having a decrypted copy of an update allows hackers to reverse engineer it and learn exactly what patches it exploits. The key may also allow parties other than Intel – say a malicious hacker or a hobbyist – to update the chips with their own microcode, although that customized version will not survive a reboot.
“At the moment, the security impact is quite difficult to assess,” independent researchers Maxim Gorachi Said in a direct message. “But in any case, this is the first time in the history of Intel processors that you can execute your microcode inside and analyze updates.” Goryachi and two other researchers-Dmitry Skilarov And Mark ErmolovBoth, jointly with the security firm Positive Technologies – worked on the project.
Keys can be extracted for any chip – be it a Celeron, Pentium, or Atom – based on Intel’s Goldmont architecture.
Rabbit hole fall down
The genesis for the discovery came three years ago when Goryachy and Ermolov found a critical vulnerability, indexed as Intel SA-00086, which allowed them to execute the code of their choice within an independent core of chips, Which included a subsystem as an Intel management engine. Intel fixed the bug and released a patch, but because the chips can always be rolled back to an earlier firmware version and then exploited, there is no way to effectively eliminate the vulnerability.
Five months ago, the trio were able to use the vulnerability to access “Red Unlock”, a service mode (see page 6 here) embedded in Intel chips. The company’s engineers use this mode to debug microcodes before the chip is released publicly. In one stroke Mold The film, the researchers named their device for accessing this previously unspecified debugger chip Red Pill, because it allows researchers to experience the inner workings of the chip that are usually off-limits. The technology works using a USB cable or special Intel adapters that pipe data into a weak CPU.
By accessing a Goldmont-based CPU in red unlock mode, researchers created a special ROM field known as MSROM, which is considered small for the microcode sequencer ROM. From there, they embarked on the painstaking process of reverse engineering microcodes. After months of analysis, it uses the update process and the RC4 key. However, the analysis did not use a signed key, which Intel uses to authenticate the authenticity of an update.
In a statement, Intel executives wrote:
The described problem does not represent a security risk for customers, and we do not rely on the information behind Red Unlock as a security measure. In addition to Intel-SA-00086 mitigation, OEMs have reduced the OEM-specific unlock capabilities required for this research following Intel’s manufacturing guidance.
The private key used to authenticate the microcode does not reside in silicon, and an attacker cannot load an unauthenticated patch on a remote system.
So far impossible
This means that attackers cannot use Chip Red Pill and the decryption key exposes this hackable CPU remotely, at least not without pursuing it for other vulnerabilities that are currently unknown. Likewise, attackers cannot use these technologies to infect the supply chain of Goldmont-based devices. But the technology holds open possibilities for hackers who have physical access to a computer running one of these CPUs.
KN White, Head of Product Security at MongoDB stated, “There is a common misconception that modern CPUs are mostly fixed in-place from the factory, and sometimes they narrowly scoped microcode updates, especially for egoistic bugs Will receive. ” “But to the extent that this is true (and to a large extent not true), there are very few practical limitations about an engineer Might Do that with the state key for silicone. ”
One possibility could be hobbyists who want to root their CPU the way people jailbreak or root iPhones and Android devices or hack Sony’s PlayStation 3 console.
In theory, it may also be possible to use the Chip Red Pill in a rogue maid attack, in which a person with fleeting access to a device hacks it. But in any of these cases, the hack will be tethered, meaning it will only last as long as the device is powered on. After restarting, the chip will return to its normal state. In some cases, the ability to execute arbitrary microcodes inside the CPU may also be useful for attacks on cryptography keys, such as those used in trusted platform modules.
“So far there is only one but very important result: independent analysis of microcode patches that until now was impossible,” said Mark Ermolov, researcher at Positive Technologies. “Now, researchers can see how Intel fixes one or the other bug / vulnerability. And this is great. Encryption of microcode patches is a type of protection through ambiguity. ”