DS14928 · DatasheetArm® Cortex®-M33 32-bit MCU with FPU, 144 MHz, 593 CoreMark®, up to 512‑Kbyte dual‑bank flash memory, 128-Kbyte RAM
Download .pdfOpen in the full document

Functional overview

Functional overview

Arm Cortex-M33 with FPU

The Cortex®-M33 is a highly energy-efficient processor designed for microcontrollers and deeply embedded applications, especially those requiring efficient security.

The Cortex® processor delivers a high‑computational performance with low‑power consumption and an advanced response to interrupts.

It features:

  • Memory protection units (MPUs) supporting eight regions
  • Floating-point arithmetic functionality with support for single-precision arithmetic

The processor supports a set of DSP instructions that allows an efficient signal processing and a complex algorithm execution.

The Cortex®-M33 processor features:

  • System AHB bus:

    The system AHB (S-AHB) bus interface is used for any instruction fetch and data access to the memory‑mapped SRAM, peripheral, or Vendor_SYS regions of the Armv8‑M memory map.

  • Code AHB bus:

    The code AHB (C-AHB) bus interface is used for any instruction fetch and data access to the code region of the Armv8‑M memory map.

Refer to Figure 1 STM32C55xxx block diagram for more details.

Instruction cache (ICACHE)

The instruction cache (ICACHE) is introduced on the C-AHB code bus of the Cortex®‑M33 processor to improve performance when fetching instructions and data from internal memories. Some specific features, like hit‑under‑miss and critical-word‑first refill policy, allow close to zero‑wait‑state performance in most use cases.

The ICACHE main features are:
  • Bus interface:
    • One 32-bit AHB slave port, the execution port (input from Cortex®‑M33 C‑AHB code interface)
    • One 128-bit AHB master port: master1 port (output to Fast bus of the main AHB bus matrix)
    • One 32-bit AHB slave port for control (input from AHB peripherals interconnect, for ICACHE registers access)
  • Cache access:
    • Zero wait-state on hits
    • Hit-under-miss capability: ability to serve processor requests (access to cached data) during an ongoing line refill due to a previous cache miss
    • Optimal cache line refill thanks to WRAPw bursts of the size of the cache line (32-bit word size, w, aligned on cache line size)
    • n-way set-associative default configuration with the possibility to configure as 1‑way, meaning direct‑mapped cache, for applications needing a very‑low‑power consumption profile
  • Memory address remap:
    • Possibility to remap input addresses falling into up to four memory regions (used to remap aliased code in SRAM memories to the code region, for execution from the C‑AHB code interface)
  • Replacement and refill:
    • pLRU-t replacement policy (pseudo-least-recently-used, based on binary tree), algorithm with the best complexity/performance balance
    • Critical-word-first refill policy, minimizing processor stalls
    • Possibility to configure burst type of AHB memory transaction for remapped regions: INCRw or WRAPw (size w aligned on cache line size)
  • Performance counters:

    The ICACHE implements two performance counters:

    • Hit monitor counter (32-bit)
    • Miss monitor counter (16-bit)
  • Error management:
    • Possibility to detect an unexpected cacheable write access, to flag an error, and optionally to raise an interrupt
  • Maintenance operation:
    • Cache invalidate: full cache invalidation, fast command, noninterruptible

Memory protection unit

The memory protection unit (MPU) is used to manage the CPU accesses to the memory. It also prevents one task to accidentally corrupt the memory or the resources used by any other active task. This memory area is organized into up to eight protected areas.

The MPU is especially helpful for applications where some critical or certified code must be protected against the misbehavior of other tasks. An RTOS (real-time operating system) usually manages the MPU.

If a program accesses a memory location that is prohibited by the MPU, the RTOS can detect it and take action. In an RTOS environment, the kernel can dynamically update the MPU area setting based on the process to be executed.

Memories

Embedded flash memory

The devices feature up to 512 Kbytes embedded flash memory that is available for storing programs and data.

The flash memory interface features:
  • Dual-bank operating modes
  • Read-while-write (RWW)

This allows a read operation to be performed from one bank while an erase or program operation is performed to the other bank. Each bank contains 32 pages of 8 Kbytes.

The flash memory embeds 4.5-Kbyte OTP (one-time programmable) for user data.

Enhanced flash memory protection mechanisms are available. These mechanisms can be activated by option bytes:
  • RDP states for protecting memory content from debug access
  • Page group write-protection (WRPG)
  • One hide protection area (HDP) per bank that provides temporal isolation for startup code
The whole nonvolatile memory embeds the error correction code (ECC) feature supporting:
  • Single-error detection and correction
  • Double-error detection
  • ECC fail address report
User data flash memory

The device features 64 Kbytes of user data flash memory split in two banks of 16 × 2‑Kbyte sectors offering perfect space for storing EEPROM emulation data.

Embedded SRAMs

Two SRAMs are embedded in the STM32C55xxx devices.

These SRAMs are made of several blocks that can be powered down in Stop mode to reduce consumption:
  • SRAM1: 64 Kbytes
  • SRAM2: 64 Kbytes with optional ECC

Boot modes

At startup, the BOOT0 pin allows the system to boot either from the user Flash or from the bootloader.

When boot from user Flash is selected, BOOTADD defines the boot address. This address can be locked thanks to BOOT_LOCK.

The embedded bootloader is located in the system memory, programmed by STMicroelectronics during production. It is used to reprogram the flash memory by using USART, SPI, FDCAN (available only in STM32C552x devices), or USB in device mode through the device firmware upgrade (DFU).

Refer to the application note STM32 microcontroller system memory boot mode (AN2606) for more details.

Power supply schemes

The devices require a 2.7 V to 3.6 V operating voltage supply VDD.

  • VDD = 2.7 V to 3.6 V

    VDD is the external power supply for the I/Os, the internal regulator, and the system analog such as reset, power management, and internal clocks. It is provided externally through the VDD pins.

  • VDDA = VDD

    VDDA is the analog power supply for ADCs, DACs, and comparator.

  • VREF-, VREF+

    VREF+ is the input reference voltage for ADCs and DAC.

    VREF+ can be grounded when ADCs and DAC are not active.

    VREF- and VREF+ pins are not available on all packages. When not available, they are bonded to VSSA and VDDA, respectively.

The STM32C55xxx devices embed a LDO regulator to provide the VCORE supply for digital peripherals, SRAM1, SRAM2, and embedded flash memory. The LDO generates this voltage on VCAP pin connected to an external capacitor of 2.2 μF typical.

The LDO regulator can operate in Stop modes where it may provide two different voltages (voltage scaling).

Figure 2. STM32C55xxx power supply overview


Power supply supervisor

The devices have an integrated power‑on reset (POR) / power‑down reset (PDR) circuitry:
  • Power-on reset (POR)

    The POR supervisor monitors the VDD power supply and compares it to a fixed threshold. The devices remain in reset mode when VDD is below this threshold.

  • Power-down reset (PDR)

    The PDR supervisor monitors the VDD power supply. A reset is generated when VDD drops below a fixed threshold.

  • Programmable voltage detector (PVD)

    The PVD monitors the VDD power supply by comparing it with a threshold fixed by hardware.

    An interrupt can be generated when VDD drops below the VPVD threshold and/or when VDD is higher than the VPVD threshold. The interrupt service routine can then generate a warning message and/or put the device into a safe state. The software enables the PVD.

Low-power modes

By default, the microcontroller is in Run mode after a system or a power reset. It is up to the user to select one of the low‑power modes described below:

  • Sleep mode

    In Sleep mode, only the CPU is stopped. All peripherals continue to operate and can wake up the CPU when an interrupt/event occurs.

  • Stop mode

    Stop mode achieves the lowest power consumption while retaining the content of SRAM and registers. All clocks in the VCORE domain are stopped, the PSI, the HSI, and the HSE crystal oscillators are disabled. The LSE or LSI is still running.

    The RTC can remain active (Stop mode with RTC, Stop mode without RTC).

    The system clock is HSI 144 MHz when exiting stop mode.

    Stop 0 mode maintains the regulator output at a nominal voltage of 1.2 V.

    Stop 1 mode reduces power consumption by lowering VCORE to 0.95 V, which results in a longer wake-up time and fewer wake-up sources compared to Stop 0 mode.

  • Standby mode

    The Standby mode is used to achieve the lowest power consumption with BOR.

    The PSI, the HSI, and the HSE crystal oscillators are also switched off.

    The RTC can remain active (Standby mode with RTC, Standby mode without RTC).

    The state of each I/O during Standby mode can be retained.

    After entering Standby mode, SRAMs and register contents are lost except for registers in the RTC domain and Standby circuitry.

    The device exits Standby mode in the following cases:
    • An external reset with NRST pin
    • An IWDG reset
    • A WKUP pin event (configurable rising or falling edge)
    • An RTC event occurs (alarm, periodic wake-up, timestamp), or in a tamper detection. The tamper detection can be raised either due to external pins or due to an internal failure detection.

    The system clock after wake-up is HSI 144 MHz.

Reset mode

To improve the consumption under reset, the I/O state under and after reset is “analog state” (the I/O Schmitt trigger is disabled).