Contents

Silicon on the MCU-based ECU

An automotive microcontroller combines real-time processing, local memory, safety monitoring and protected cryptography to control functions such as braking, steering or battery management.

3 min read · Discipline stake: Hardware Safety Security BSW Update

What runs here

This cell is the physical foundation beneath the ECU’s software. A microcontroller (MCU) integrates processor cores, memory and control peripherals on one chip. The exact combination depends on the device; the cell label describes a representative safety-oriented configuration.

Component Role
Lockstep processing Executes matching instruction streams in paired processing logic and compares results to detect certain hardware faults.
Embedded flash and SRAM Store executable software and persistent data, and hold working data during operation.
Error-correcting code (ECC) Detects and, for supported error patterns, corrects memory errors; coverage depends on the implementation.
Hardware security module (HSM) Provides an isolated environment for protected key use and cryptographic operations.
Timers, ADCs and communication controllers Capture inputs, trigger measurements, generate outputs and exchange network frames.
Protection and monitoring hardware Restricts memory access and detects faults involving execution, clocks, supplies or other monitored resources.

Lockstep is a detection mechanism: its paired execution does not provide two independent application processors or automatically preserve operation after a fault. Memory protection, self-tests and fault reporting complement it. Their availability and required configuration are device-specific. Infineon safety concept

Why it looks like this

Brake, steering and battery-control functions need bounded response times and close coordination with physical inputs and outputs. Local memory, hardware timers and integrated peripherals support predictable execution. Engineers must still account for flash wait states, caches where present, DMA and contention between bus masters.

Safety requirements drive diagnostic coverage and fault reactions. A device intended to support ASIL D development does not make the complete ECU ASIL D compliant. The hardware architecture, software mechanisms and external circuitry must satisfy the allocated safety requirements and the semiconductor supplier’s assumptions of use. ISO 26262-5:2018 addresses development at the hardware level. ISO 26262-5:2018

Integration can reduce board components and external memory interfaces, while embedded flash imposes capacity, endurance and programming constraints. Software updates therefore influence silicon selection: image storage, recovery space, flash-bank organisation and execution restrictions during programming must be checked early. Concrete flash architectures differ between devices. Example MCU memory documentation

An HSM supports protected cryptographic processing, but available algorithms, key storage and boot integration vary. Secure boot requires a correctly provisioned and configured boot chain around that hardware. Example HSM capabilities

How it connects

Immediately above the silicon, AUTOSAR Microcontroller Abstraction Layer (MCAL) drivers or equivalent low-level drivers configure registers, interrupts and peripherals. Examples include Mcu, Port, Adc, Pwm, Wdg and Can. Memory protection hardware supports operating-system isolation; HSM services typically reach platform software through device-specific firmware and a cryptographic driver.

Below and beside the MCU, board circuitry supplies power, clocking, reset supervision and physical interfaces. A CAN controller typically connects through an external transceiver. Sensor interfaces and actuator power stages likewise depend on external circuitry; an MCU output pin does not directly drive a high-power load.

Other nodes communicate through the implemented network interfaces. Silicon detects some local faults, while software must provide message freshness checks, communication supervision and coordinated fault responses across ECUs.

Compared with other node classes

  • HPC application processor: Application cores, GPU/NPU accelerators and LPDDR support larger workloads; memory management, bandwidth and shared-resource interference become more prominent concerns.
  • HPC safety island: Its MCU-like cores and ECC SRAM can supervise or execute safety functions, but shared power, reset and interconnect dependencies require analysis.
  • Zone controller: MCU integration also serves concentrated networking and local I/O. Smart switches may be separate board devices controlling load power.
  • Smart node / LIN slave: A small MCU, LIN system basis chip or ASIC serves a narrower function with tighter resource constraints; lockstep and an HSM are not universal features.

Best practices

  • Select the exact device variant against timing, memory, peripheral, environmental and safety requirements.
  • Translate the supplier’s safety manual assumptions into explicit hardware and software requirements.
  • Measure worst-case execution under representative interrupt, DMA, network and memory contention.
  • Design key provisioning, debug access control and boot recovery before production programming.
  • Budget flash endurance and update storage, including interrupted programming and recovery behaviour.
  • Verify fault reporting and the complete path from detection to the required safe or degraded state.

Common pitfalls

  • Treating lockstep as protection against software design errors or every common-cause hardware fault.
  • Assuming ECC covers every memory, transfer path and error pattern equally.
  • Assuming an HSM automatically authenticates all software and network traffic.
  • Choosing a pin-compatible replacement without rechecking timing, errata, safety assumptions and security provisioning.

What it means for your discipline

  • Hardware & silicon Core

    Selects the MCU variant and designs its power, reset, clock, memory and peripheral connections. Checks electrical limits, errata and external monitoring requirements.

  • Functional safety Core

    Allocates hardware safety requirements and evaluates diagnostic coverage, dependent failures and fault reactions. Converts supplier assumptions of use into verified ECU requirements.

  • Cybersecurity Partial

    Defines requirements for HSM capabilities, key provisioning, boot trust and debug access. Verifies that the selected silicon can support the ECU’s security lifecycle.

  • Base software Partial

    Configures drivers, memory protection and hardware diagnostics for the exact device. Accounts for interrupt behaviour, flash programming restrictions and hardware fault reporting.

  • Software update & OTA Partial

    Checks flash capacity, bank organisation and programming restrictions when defining installation and recovery strategies. Verifies that image storage, cryptographic support and interrupted-programming recovery fit the selected MCU variant.

No direct stake: E/E architecture · HPC & application software · Networks & communication · Diagnostics & lifecycle

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)

Not directly involved on this node type.

MCU-based ECU · AUTOSAR Classic
  • Microcontroller · HSM — Integrates processing cores, flash, peripherals, fault detection and protected cryptography; device capabilities constrain timing, safety mechanisms and update recovery.