Contents

Secure boot and the chain of trust

How an ECU makes sure that only genuine, unmodified software runs — from an immutable root of trust in silicon through every boot stage.

Why it exists

If an attacker can replace ECU software, every other protection can be switched off. Secure boot ensures only software signed by the manufacturer runs.

What it is

  • A root of trust: immutable code (boot ROM) and a key (or its hash) stored in hardware.
  • A chain: each stage verifies the signature of the next before starting it.
  • Variants: secure boot (stop if verification fails) and measured/authenticated boot (record measurements; HSM verifies in parallel and may withhold keys).

How it works

Software images are signed in a protected signing environment at the manufacturer. At startup, the boot ROM verifies the bootloader, the bootloader verifies the application, and so on. On HPCs the chain continues through firmware, hypervisor and operating systems.

ECU boot chain of trust
Boot ROMImmutable root of trust
BootloaderRuns only after successful verification
ApplicationRuns only after successful verification

Example chain; actual stages vary by ECU, and a failed verification prevents the affected stage from starting.

Best practices

  • Protect signing keys in an HSM with strict access control.
  • Plan key rotation and revocation for the vehicle’s lifetime.
  • Disable or lock debug interfaces in production.
  • Balance verification time against startup requirements (e.g. parallel verification by the HSM).

Common pitfalls

  • Verifying only a checksum, not a cryptographic signature.
  • Leaving a downgrade path to older, vulnerable software versions.

What it means for your discipline

  • Cybersecurity Core

    Defines the chain of trust, signing infrastructure, key hierarchy and response to verification failures.

  • Hardware & silicon Core

    Boot ROM, HSM, one-time-programmable keys and debug protection are the hardware anchors.

  • Base software Core

    Bootloaders and startup code implement verification and handle failures.

  • HPC & application software Partial

    On HPCs, each stage (firmware, hypervisor, OS, containers) must verify the next.

  • Functional safety Partial

    Boot verification time affects startup time; failure handling must lead to a safe state.

  • Diagnostics & lifecycle Partial

    Verification failures should be reported so that tampering and corrupted flash can be distinguished.

  • Software update & OTA Partial

    Ensures update images satisfy the ECU's signature and trust-chain checks before activation. Coordinates anti-downgrade rules with rollback and recovery so a failed update has an authorised bootable fallback.

No direct stake: E/E architecture · Networks & communication

Where it sits in the software stack

Generic layered software of a central HPC (application processor with hypervisor, plus a safety island) and of a classic MCU-based ECU (AUTOSAR Classic). Highlighted modules matter for this topic. Click any module to open its page.

HPC · application processor (MPU) and safety island (MCU)
  • Secure boot chain · TEE — Verifies successive boot stages and enforces signing trust, revocation and anti-downgrade rules before execution.
  • Boot ROM · HSM — Anchors the safety island's boot verification in immutable ROM and HSM-protected trust material.
MCU-based ECU · AUTOSAR Classic
  • Flash bootloader · secure boot — Verifies application signatures before execution and rejects images that violate trust or anti-downgrade policy.
  • Microcontroller · HSM — Provides immutable boot ROM, protected trust anchors and debug restrictions that underpin the chain of trust.