System-Level Optimization Beyond Cryptographic Kernels: An ML-KEM Case Study on Arm Cortex-M7 arxiv.org
ML-KEM on Cortex-M7, where moving bytes around buys 2.5% unless you also recycle public data like it's 1999.
trending4
01 02 Ported AIM2-style SIMD tricks to AIMer v3 for AVX2/AVX-512 and Cortex-M55, because apparently one ISA wasn't enough.03 High-Performance Implementations of HAETAE and SMAUG-T on Arm Cortex-M55 Using MVE and Slothy eprint.iacr.orgPQC on Cortex-M55 gets the usual treatment: MVE, hand-tuned dataflow, Slothy, and some actual speedups on STM32N657.04 Benchmarking Post-Quantum Cryptography in Lightweight Virtualization Environments on Embedded Hardware arxiv.orgHow much do PQC handshakes cost on embedded ARM, and does Docker or a unikernel actually move the needle?