Published on July 30th, 2017 📆 | 3661 Views ⚑1
REMOTELY COMPROMISING ANDROID AND IOS VIA A BUG IN BROADCOM’S WI-FI CHIPSETS
At the Black Hat security conference, Artenstein demonstrated proof-of-concept attack code that exploited a vulnerability in Wi-Fi chips manufactured by Broadcom. It fills the airwaves with probes that request connections to nearby computing devices. When the specially devised requests reach a device using the BCM43xx family of Wi-Fi chipsets, the attack rewrites the firmware that controls the chip. The compromised chip then sends the same malicious packets to other vulnerable devices, setting off a potential chain reaction. Until early July and last week—when Google and Apple issued patches respectively—an estimated 1 billion devices were vulnerable to the attack. Artenstein has dubbed the worm "Broadpwn."
Although the flaw is now closed, the hack has important lessons as engineers continue their quest to secure mobile phones and other computing devices. Security protections such as address space layout randomization and data execution prevention have now become standard parts of the operating systems and apps. As a result, attackers have to work hard to exploit buffer overflowsand other types of software vulnerabilities. That extra work largely makes self-replicating worms impossible. Artenstein's exploit, however, suggests that such worms are by no means impossible.
"This research is an attempt to demonstrate what such an attack, and such a bug, will look like," the researcher wrote in a detailed blog post. "Broadpwn is a fully remote attack against Broadcom’s BCM43xx family of Wi-Fi chipsets, which allows for code execution on the main application processor in both Android and iOS. It is based on an unusually powerful 0-day that allowed us to leverage it into a reliable, fully remote exploit."
Making a comeback
In sharp contrast to the kernels in iOS and Android, the Broadcom chips Artenstein targeted aren't protected by ASLR or DEP. That meant he could reliably know where his malicious code would be loaded in chip memory so he could ensure it got executed. Additionally, he found a flaw across various chipset firmware versions that allowed his code to work universally rather than having to be customized for each firmware build. Making the attack even more potent, targets didn't have to connect to the attacker's Wi-Fi network. Simply having Wi-Fi turned on was sufficient to being hacked.
Artenstein said his attack worked on a wide range of phones, including all iPhones since the iPhone 5, Google's Nexus 5, 6, 6X and 6P models, Samsung Notes 3 devices, and Samsung Galaxy devices from S3 to S8. After he privately reported the flaw, Google and Apple released patches that closed the underlying vulnerability that made the attack possible. Because Wi-Fi chipsets in laptop and desktop computers have more limited access to the computer's networking functions, the researcher doesn't believe they are vulnerable to the same attack. While Artenstein's proof of concept didn't spread from the Wi-Fi chip to infect the phone's kernel, he said that additional step is well within the means of determined hackers.
The remote code-execution vulnerability is the second one to be fixed by Broadcom this year. In April, both Apple and Google patched a separate critical flaw in the manufacturer's Wi-Fi chipsets. Gal Beniamini, the Google Project Zero researcher who discovered the vulnerability, said the absence of security mitigations made his proof-of-concept exploit relatively easy to develop. Together, the flaws suggest a potentially more promising avenue for attackers targeting smart phones.
"Old school hackers often miss the 'good old days' of the early 2000s, when remotely exploitable bugs were abundant, no mitigations were in place to stop them, and worms and malware ran rampant," Artenstein wrote. "But with new research opening previously unknown attack surface such as the BCM Wi-Fi chip, those times may just be making a comeback."