Trusted execution environments (TEEs) protect applications running inside of them from untrusted host systems. The host can not access or modify the memory of applications protected by a TEE. Intel TDX is a recently introduced TEE that allows for the execution of arbitrary code, including entire operating systems, inside a protected environment. Prior work has attacked AMD SEV-SNP, AMD's counterpart to Intel TDX, using performance counters, by leaking sensitive information through them, e.g., which branches are taken. Intel TDX is thought not to be affected by such attacks, as it disables performance counters when entering the protected guests (TDs) to mitigate these attacks.

In recent years, the use of GPUs for general-purpose computations has steadily increased. As security-critical computations like AES are becoming more common on GPUs, the scrutiny must also increase. At the same time, new technologies like WebGPU put easy access to compute shaders in every web browser. Prior work has shown that GPU caches are vulnerable to the same eviction-based attacks as CPUs, e.g., Prime+Probe, from native code.

Interrupts are fundamental for inter-process and cross-core communication in modern systems. Controlling these communication mechanisms historically requires switches into the kernel or hypervisor, incurring high-performance costs. To alleviate these costs, Intel introduced new hardware mechanisms to send inter-processor interrupts (IPIs) from user space without switching into the kernel and from virtual machines without switching into the hypervisor. However, it is unclear whether this direct, unsupervised interaction between unprivileged (or virtualized) workloads and the underlying hardware introduces a significant change in the attack surface.

The number of Linux kernel vulnerabilities discovered has increased drastically over the past years. In the kernel, even simple memory safety vulnerabilities can have devastating consequences, e.g., compromising the entire system. Efforts to mitigate these vulnerabilities have so far focused mainly on control-flow hijacking attacks in the kernel. Yet, data-oriented attacks remain largely unmitigated in practice as existing mitigations are limited in providing robust security guarantees at reasonable performance overhead for multiple sensitive data objects.