Marco Guarnieri

Panagiotis Georgios Pennas, Konstantinos Papaioannou, Marco Guarnieri et al.

Large Language Models (LLMs) rely on optimizations like Automatic Prefix Caching (APC) to accelerate inference. APC works by reusing previously computed states for the beginning part of a request (prefix), when another request starts with the same text. While APC improves throughput, it introduces timing side channels: cache hits are faster than misses, creating observable latency differences. In multi-tenant systems, attackers can exploit these differences to infer sensitive information, e.g., by incrementally reconstructing another user's request by observing hit/miss patterns. Current defenses take a sledgehammer approach: they disable APC and cache sharing, isolating users, and sacrificing efficiency for regular users. This paper presents CacheSolidarity, a system that secures multi-tenant LLM serving systems against APC side channels without sacrificing performance and efficiency. CacheSolidarity monitors cache reuse across users, flags suspicious sharing, and selectively isolates prefixes, restricting their reuse only when necessary. Evaluation shows that CacheSolidarity enables up to 70% higher cache reuse and 30% lower inference latency compared to existing defenses that isolate users. CacheSolidarity's lightweight design demonstrates how security in LLM serving does not have to come at the cost of unnecessarily reduced performance or unbearable overheads.

Isaac David, Marco Guarnieri, Arthur Gervais

Agentic security systems increasingly combine LLM planners with tools that can discover, validate, and report vulnerabilities. This creates an asymmetric control problem: the system should retain strong offensive capability inside an authorized engagement, while the same capabilities must be denied outside scope. Existing guardrails provide useful policy controls, but they do not make this boundary a first-class formal contract over observable effects. We introduce alignment contracts, a framework for specifying and enforcing behavioral constraints over observable effect traces. A contract defines scope, allowed and forbidden effects, resource budgets, and disclosure policies. We give the language finite-trace semantics, characterize satisfaction as a safety property with finite violation witnesses, develop refinement and one-way composition rules for modular contract engineering, and show that admissibility checking is decidable. We instantiate the framework for web-focused agentic security workflows and show how the same structure extends to other effect profiles. Under an explicit Effect Observability Assumption, where all $\SigmaEff$-effects are mediated, the soundness theorem quantifies over the agent model and gives guarantees for mediated $\SigmaEff$-effects, including enforcement soundness for monitor-realized traces. We also state an assumption-lifted adaptation result and formalize limits through undecidability transfer and observability-boundary theorems. A Lean 4 artifact checks the formal core theorems used by the paper.

Xaver Fabian, Marco Guarnieri, Boris Köpf et al.

Speculative execution enhances processor performance by predicting intermediate results and executing instructions based on these predictions. However, incorrect predictions can lead to security vulnerabilities, as speculative instructions leave traces in microarchitectural components that attackers can exploit. This is demonstrated by the family of Spectre attacks. Unfortunately, existing countermeasures to these attacks lack a formal security characterization, making it difficult to verify their effectiveness. In this paper, we propose a novel framework for detecting information flows introduced by speculative execution and reasoning about software defenses. The theoretical foundation of our approach is speculative non-interference (SNI), a novel semantic notion of security against speculative execution attacks. SNI relates information leakage observed under a standard non-speculative semantics to leakage arising under semantics that explicitly model speculative execution. To capture their combined effects, we extend our framework with a mechanism to safely compose multiple speculative semantics, each focussing on a single aspect of speculation. This allows us to analyze the complex interactions and resulting leaks that can arise when multiple speculative mechanisms operate together. On the practical side, we develop Spectector, a symbolic analysis tool that uses our compositional framework and leverages SMT solvers to detect vulnerabilities and verify program security with respect to multiple speculation mechanisms. We demonstrate the effectiveness of Spectector through evaluations on standard security benchmarks and new vulnerability scenarios.

Marco Guarnieri, Marco Patrignani

Microarchitectural attacks exploit the abstraction gap between the Instruction Set Architecture (ISA) and how instructions are actually executed by processors to compromise the confidentiality and integrity of a system. To secure systems against microarchitectural attacks, programmers need to reason about and program against these microarchitectural side-effects. However, we cannot -- and should not -- expect programmers to manually tailor programs for specific processors and their security guarantees. Instead, we could rely on compilers (and the secure compilation community), as they can play a prominent role in bridging this gap: compilers should target specific processors microarchitectural security guarantees and they should leverage these guarantees to produce secure code. To achieve this, we outline the idea of Contract-Aware Secure COmpilation (CASCO) where compilers are parametric with respect to a hardware/software security-contract, an abstraction capturing a processor's security guarantees. That is, compilers will automatically leverage the guarantees formalized in the contract to ensure that program-level security properties are preserved at microarchitectural level.

Marco Guarnieri, Boris Köpf, Jan Reineke et al.

Since the discovery of Spectre, a large number of hardware mechanisms for secure speculation has been proposed. Intuitively, more defensive mechanisms are less efficient but can securely execute a larger class of programs, while more permissive mechanisms may offer more performance but require more defensive programming. Unfortunately, there are no hardware-software contracts that would turn this intuition into a basis for principled co-design. In this paper, we put forward a framework for specifying such contracts, and we demonstrate its expressiveness and flexibility. On the hardware side, we use the framework to provide the first formalization and comparison of the security guarantees provided by a representative class of mechanisms for secure speculation. On the software side, we use the framework to characterize program properties that guarantee secure co-design in two scenarios traditionally investigated in isolation: (1) ensuring that a benign program does not leak information while computing on confidential data, and (2) ensuring that a potentially malicious program cannot read outside of its designated sandbox. Finally, we show how the properties corresponding to both scenarios can be checked based on existing tools for software verification, and we use them to validate our findings on executable code.

Pepe Vila, Andreas Abel, Marco Guarnieri et al.

Flushing the cache, using instructions like clflush and wbinvd, is commonly proposed as a countermeasure against access-based cache attacks. In this report, we show that several Intel caches, specifically the L1 caches in some pre-Skylake processors and the L2 caches in some post-Broadwell processors, leak information even after being flushed through clflush and wbinvd instructions. That is, security-critical assumptions about the behavior of clflush and wbinvd instructions are incorrect, and countermeasures that rely on them should be revised.

Marco Guarnieri, Boris Köpf, José F. Morales et al.

Since the advent of SPECTRE, a number of countermeasures have been proposed and deployed. Rigorously reasoning about their effectiveness, however, requires a well-defined notion of security against speculative execution attacks, which has been missing until now. In this paper (1) we put forward speculative non-interference, the first semantic notion of security against speculative execution attacks, and (2) we develop SPECTECTOR, an algorithm based on symbolic execution to automatically prove speculative non-interference, or to detect violations. We implement SPECTECTOR in a tool, which we use to detect subtle leaks and optimizations opportunities in the way major compilers place SPECTRE countermeasures. A scalability analysis indicates that checking speculative non-interference does not exhibit fundamental bottlenecks beyond those inherited by symbolic execution.

Marco Guarnieri, Srdjan Marinovic, David Basin

Databases can leak confidential information when users combine query results with probabilistic data dependencies and prior knowledge. Current research offers mechanisms that either handle a limited class of dependencies or lack tractable enforcement algorithms. We propose a foundation for Database Inference Control based on ProbLog, a probabilistic logic programming language. We leverage this foundation to develop Angerona, a provably secure enforcement mechanism that prevents information leakage in the presence of probabilistic dependencies. We then provide a tractable inference algorithm for a practically relevant fragment of ProbLog. We empirically evaluate Angerona's performance showing that it scales to relevant security-critical problems.

Marco Guarnieri, Srdjan Marinovic, David Basin

Existing SQL access control mechanisms are extremely limited. Attackers can leak information and escalate their privileges using advanced database features such as views, triggers, and integrity constraints. This is not merely a problem of vendors lagging behind the state-of-the-art. The theoretical foundations for database security lack adequate security definitions and a realistic attacker model, both of which are needed to evaluate the security of modern databases. We address these issues and present a provably secure access control mechanism that prevents attacks that defeat popular SQL database systems.