Posted on 03/28/2018 2:29:38 AM PDT by ShadowAce
Researchers have discovered a new side-channel attack method that can be launched against devices with Intel processors, and the patches released in response to the Spectre and Meltdown vulnerabilities might not prevent these types of attacks.
The new attack, dubbed BranchScope, has been identified and demonstrated by a team of researchers from the College of William & Mary, University of California Riverside, Carnegie Mellon University in Qatar, and Binghamton University.
Similar to Meltdown and Spectre, BranchScope can be exploited by an attacker to obtain potentially sensitive information they normally would not be able to access directly. The attacker needs to have access to the targeted system and they must be able to execute arbitrary code.
Researchers believe the requirements for such an attack are realistic, making it a serious threat to modern computers, on par with other side-channel attacks. The BranchScope attack has been demonstrated on devices with three types of Intel i5 and i7 CPUs based on Skylake, Haswell and Sandy Bridge microarchitectures.
Experts showed that the attack works even if the targeted application is running inside of an Intel SGX enclave. Intel SGX, or Software Guard Extensions, is a hardware-based isolated execution system designed to prevent code and data from getting leaked or modified.
BranchScope is similar to Spectre as they both target the directional branch predictors. Branch prediction units (BPUs) are used to improve the performance of pipelined processors by guessing the execution path of branch instructions. The problem is that when two processes are executed on the same physical CPU core, they share a BPU, potentially allowing a malicious process to manipulate the direction of a branch instruction executed by the targeted application.
The BPU has two main components a branch target buffer (BTB) and a directional predictor and manipulating either one of them can be used to obtain potentially sensitive data from the memory. Intel recently published a video providing a high level explanation of how these attacks work.
Researchers showed on several occasions in the past how BTB manipulation can be used for attacks, but BranchScope involves manipulation of branch predictors.
BranchScope is the first fine-grained attack on the directional branch predictor, expanding our understanding of the side channel vulnerability of the branch prediction unit, the researchers explained in their paper.
The researchers who identified the BranchScope attack method have proposed a series of countermeasures that include both software- and hardware-based solutions.
Dmitry Evtyushkin, one of the people involved in this research, told SecurityWeek that while they have not been tested, the microcode updates released by Intel in response to Meltdown and Spectre might only fix the BTB vector, which means BranchScope attacks could still be possible. However, Intel told the researchers that software guidance for mitigating Spectre Variant 1 could be effective against BranchScope attacks as well.
We have been working with these researchers and we have determined the method they describe is similar to previously known side channel exploits, Intel said in an emailed statement. We anticipate that existing software mitigations for previously known side channel exploits, such as the use of side channel resistant cryptography, will be similarly effective against the method described in this paper. We believe close partnership with the research community is one of the best ways to protect customers and their data, and we are appreciative of the work from these researchers.
BranchScope is not the only CPU side-channel attack method uncovered following the disclosure of Meltdown and Spectre. One of them, dubbed SgxPectre, shows how Spectre can be leveraged to defeat SGX.
Researchers have also demonstrated new variants of the Meltdown and Spectre attacks, which they have named MeltdownPrime and SpectrePrime.
Thanks to bitt for the linK!
My management team is going to start stroking out from all of these vulnerabilities.
The attacker needs to have access to the targeted system and they must be able to execute arbitrary code.
So if your systems are already locked down, you should be OK.
This is over one year old and mitigations are already out there.
https://www.usenix.org/system/files/conference/usenixsecurity17/sec17-gruss.pdf
https://arxiv.org/pdf/1709.09917.pdf
Tell that to PCI-DSS auditors.
This is a different attack, though.
rofl. Yes, I understand. I was talking about reality, though.
I have to deal with vuln reports all the time, and all the time, I have to put those auditors in their place. They don't actually know what they are doing.
So did Asian dominated grad students discover a new way for their friends back home to hack, or did more NSA secrets get discovered?
I have decades of IT experience and several industry certifications under my belt including a CISSP, and despite similar reassurances, they demand proof of remediation. I’ve learned to work with auditors, but that doesn’t mean I have to like them.
Thanks to ShadowAce and bitt for the pings!
No, it’s the same attack.
They sure are really good at marking off check boxes on forms though!
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.