Hardware Security and Side Channels

Stub. Trusted computing and hardware roots of trust (TPMs, secure enclaves) as a way to anchor security below the OS, and side-channel attacks (timing, power, cache, and related channels) that extract secrets by observing a system's physical behavior rather than attacking its stated interface.

· 1 min read · 170 words

Software-level security assumes the hardware underneath is trustworthy and side-channel-free — this note is about both halves of what happens when that assumption is examined. Trusted computing (TPMs, secure/trusted execution environments) tries to establish a hardware root of trust that software can build on, via measured boot and hardware-isolated execution. Side-channel attacks go the other way, extracting secrets (cryptographic keys, passwords) not from a system’s logical interface but from physical signals it leaks incidentally — timing (how long an operation takes depending on secret data), power draw, electromagnetic emission, and cache-occupancy patterns (shared with other tenants on the same hardware) — which is exactly why constant-time implementations matter for cryptographic code covered elsewhere in this topic.

Outline (TODO — flesh out each)

  • Trusted computing — hardware roots of trust, measured boot
  • Trusted execution environments / secure enclaves
  • Timing side channels
  • Power analysis side channels
  • Cache-based side channels
  • Defenses: constant-time implementations, isolation

← All notes