Controlled-Channel Attacks: Deterministic Side Channels for Untrusted Operating Systems

By Yuanzhong Xu, Weidong Cui, Marcus Peinado. 2015

876 citations recorded by the source corpus.

The presence of large numbers of security vulnerabilities in popular feature-rich commodity operating systems has inspired a long line of work on excluding these operating systems from the trusted computing base of applications, while retaining many of their benefits. Legacy applications continue to run on the untrusted operating system, while a small hyper visor or trusted hardware prevents the operating system from accessing the applications' memory. In this paper, we introduce controlled-channel attacks, a new type of side-channel attack that allows an untrusted operating system to extract large amounts of sensitive information from protected applications on systems like Overshadow, Ink Tag or Haven. We implement the attacks on Haven and Ink Tag and demonstrate their power by extracting complete text documents and outlines of JPEG images from widely deployed application libraries. Given these attacks, it is unclear if Over shadow's vision of protecting unmodified legacy applications from legacy operating systems running on off-the-shelf hardware is still tenable.

Subfields: Artificial Intelligence, Hardware and Architecture, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Hardware and Architecture

In Signal Processing

Citations in the corpus

Ekiden: A Platform for Confidentiality-Preserving, Trustworthy, and Performant Smart Contracts (2018, neutral): “Also, the current SGX implementation is vulnerable to sidechannel attacks [77], [60].”
Prochlo: Strong Privacy for Analytics in the Crowd (2017, methodology): “We also took care to minimize side channels [12, 43, 76]; for example, we ensure that control paths involving real and dummy items are the same, and write the same private memory pages.”
AsyncShock: Exploiting Synchronisation Bugs in Intel SGX Enclaves (2016, background): “Fine-grained page tracking can be used for powerful side channel attacks [28].”
Komodo: Using verification to disentangle secure-enclave hardware from software (2017, background): “Komodo is not vulnerable to controlled-channel attacks [78, 88] merely by virtue of the fact that it does not yet support demand-paging of enclave memory.”
DAWG: A Defense Against Cache Timing Attacks in Speculative Execution Processors (2018, background): “Dirty pages can be prepared for CoW sharing eagerly, or lazily (but cautiously [57]) by installing non-present pages in the consumer domain mapping.”
T-SGX: Eradicating Controlled-Channel Attacks Against Enclave Programs (2017, methodology): “CONTROLLED-CHANNEL ATTACK REVISITED In this section, we briefly explain the controlled-channel attack [65] (called pigeonhole attack in [54]) that allows a malicious OS to infer sensitive computation and data inside a TEE such as Haven [4] and InkTag [21].”
Plundervolt: Software-based Fault Injection Attacks against Intel SGX (2020, background): “Note, however, that SGX clears the least signiﬁcant 12 bits in the reported page fault address to limit page fault side-channel exposure [78].”
Hacking in Darkness: Return-oriented Programming against Secure Enclaves (2017, background): “Several authors [26, 33, 38] have identified side channels that can leak large amounts of sensitive information out of the application’s isolated execution”
Foreshadow: Extracting the Keys to the Intel SGX Kingdom with Transient Out-of-Order Execution (2018, positive): “Our exploit targets the more challenging tmp buffer to demonstrate Foreshadow’s strength in combination with state-of-the-art enclave execution control frameworks [57, 60].”
Malware Guard Extension: Using SGX to Conceal Cache Attacks (2017, background): “[32] show that SGX is vulnerable to controlled channel attacks in which a malicious operating system triggers and monitors enclave page faults [33].”
ZombieLoad: Cross-Privilege-Boundary Data Sampling (2019, methodology): “rs that co-locate with a victim enclave on the same physical CPU. As a system attacker, we can increase ZombieLoad’s temporal resolution by leveraging previous research results exploiting page faults [71, 78] or interrupts [54, 70] to regulate the victim enclave’s execution. We use the SGX-Step [69]”
BranchScope: A New Side-Channel Attack on Directional Branch Predictor (2018, methodology): “Attacks on libjpeg were previously demonstrated using the page fault side channel [27, 54] by counting the number of times the optimization can be applied, resulting in recovery of an original image.”
Translation Leak-aside Buffer: Defeating Cache Side-channel Protections with TLB Attacks (2018, methodology): “In particular, Wang et al. [56] showed that it is possible to bypass existing defenses [51] against controlled channel attacks [58] by ﬂushing the TLB to force page table walks without trapping SGX enclaves.”
Sanctum: Minimal Hardware Extensions for Strong Software Isolation (2016, background): “3), for a runtime to handle page faults and implement demand paging securely, and robust against the attacks described in [60].”
Reverse Engineering Convolutional Neural Networks Through Side-channel Information Leaks (2018, background): “A compromised OS can also observe memory accesses through side channels such as page faults and cache conicts [17] or by repeated reading memory to detect changes [10].”
Secure Multiparty Computation from SGX (2017, background): “In the former, software co-located in the machine observes timing channels based on memory access patterns, control flow, branch prediction, cache-based based attacks [15], page-fault side channels [38], etc.”