2. Memory and bus architecture

2.1 System architecture

In STM32F401xB/C and STM32F401xD/E, the main system consists of 32-bit multilayer AHB bus matrix that interconnects:

• • Six masters:
  • – Cortex ^® -M4 with FPU core I-bus, D-bus and S-bus
  • – DMA1 memory bus
  • – DMA2 memory bus
  • – DMA2 peripheral bus
• • Five slaves:
  • – Internal Flash memory ICode bus
  • – Internal Flash memory DCode bus
  • – Main internal SRAM
  • – AHB1 peripherals including AHB to APB bridges and APB peripherals
  • – AHB2 peripherals

The bus matrix provides access from a master to a slave, enabling concurrent access and efficient operation even when several high-speed peripherals work simultaneously. This architecture is shown in Figure 1 .

Figure 1. System architecture

1. 1. STM32F401xB/C: 128 KBytes / 256 KBytes Flash with 64 KBytes SRAM.
   STM32F401xD/E: 384 KBytes / 512KBytes Flash with 96 KBytes SRAM.

2.1.1 I-bus

This bus connects the Instruction bus of the Cortex ^® -M4 with FPU core to the BusMatrix. This bus is used by the core to fetch instructions. The target of this bus is a memory containing code (internal flash memory/SRAM).

2.1.2 D-bus

This bus connects the databus of the Cortex ^® -M4 with FPU to the BusMatrix. This bus is used by the core for literal load and debug access. The target of this bus is a memory containing code or data (internal flash memory/SRAM).

2.1.3 S-bus

This bus connects the system bus of the Cortex ^® -M4 with FPU core to a BusMatrix. This bus is used to access data located in a peripheral or in SRAM. Instructions may also be fetched on this bus (less efficient than ICode). The targets of this bus are the internal SRAM, the AHB1 peripherals including the APB peripherals and the AHB2 peripherals.

2.1.4 DMA memory bus

This bus connects the DMA memory bus master interface to the BusMatrix. It is used by the DMA to perform transfer to/from memories. The targets of this bus are data memories: internal Flash memory, internal SRAM and additionally for S4 the AHB1/AHB2 peripherals including the APB peripherals.

2.1.5 DMA peripheral bus

This bus connects the DMA peripheral master bus interface to the BusMatrix. This bus is used by the DMA to access AHB peripherals or to perform memory-to-memory transfers. The targets of this bus are the AHB and APB peripherals plus data memories: Flash memory and internal SRAM.

2.1.6 BusMatrix

The BusMatrix manages the access arbitration between masters. The arbitration uses a round-robin algorithm.

2.1.7 AHB/APB bridges (APB)

The two AHB/APB bridges, APB1 and APB2, provide full synchronous connections between the AHB and the two APB buses, allowing flexible selection of the peripheral frequency.

Refer to the device datasheets for more details on APB1 and APB2 maximum frequencies, and to Table 1 for the address mapping of AHB and APB peripherals.

After each device reset, all peripheral clocks are disabled (except for the SRAM and flash memory interface). Before using a peripheral you have to enable its clock in the RCC_AHBxENR or RCC_APBxENR register.

Note: When a 16- or an 8-bit access is performed on an APB register, the access is transformed into a 32-bit access: the bridge duplicates the 16- or 8-bit data to feed the 32-bit vector.

2.2 Memory organization

Program memory, data memory, registers and I/O ports are organized within the same linear 4 Gbyte address space.

The bytes are coded in memory in little endian format. The lowest numbered byte in a word is considered the word's least significant byte and the highest numbered byte, the word's most significant.

For the detailed mapping of peripheral registers, please refer to the related chapters.

The addressable memory space is divided into 8 main blocks, each of 512 MB.

All the memory areas that are not allocated to on-chip memories and peripherals are considered "Reserved"). Refer to the memory map figure in the product datasheet.

2.3 Memory map

See the datasheet corresponding to your device for a comprehensive diagram of the memory map. Table 1 gives the boundary addresses of the peripherals available in STM32F401xB/C and STM32F401xD/E devices.

Table 1. STM32F401xB/C and STM32F401xD/E register boundary addresses

Table 1. STM32F401xB/C and STM32F401xD/E register boundary addresses (continued)

2.3.1 Embedded SRAM

STM32F401xB/C devices feature 64 Kbytes and STM32F401xD/E devices feature 96 Kbytes of system SRAM.

The embedded SRAM can be accessed as bytes, half-words (16 bits) or full words (32 bits). Read and write operations are performed at CPU speed with 0 wait state.

The CPU can access the embedded SRAM through the System bus or through the I-Code/D-Code buses when boot from SRAM is selected or when physical remap is selected ( Section 7.2.1: SYSCFG memory remap register (SYSCFG_MEMRMP) in the SYSCFG controller). To get the max performance on SRAM execution, physical remap should be selected (boot or software selection).

2.3.2 Flash memory overview

The flash memory interface manages CPU AHB I-Code and D-Code accesses to the flash memory. It implements the erase and program flash memory operations and the read and write protection mechanisms. It accelerates code execution with a system of instruction prefetch and cache lines.

The flash memory is organized as follows:

• • A main memory block divided into sectors.
• • System memory from which the device boots in System memory boot mode
• • 512 OTP (one-time programmable) bytes for user data.
• • Option bytes to configure read and write protection, BOR level, watchdog software/hardware and reset when the device is in Standby or Stop mode.

Refer to Section 3: Embedded flash memory interface for more details.

2.3.3 Bit banding

The Cortex®-M4 with FPU memory map includes two bit-band regions. These regions map each word in an alias region of memory to a bit in a bit-band region of memory. Writing to a word in the alias region has the same effect as a read-modify-write operation on the targeted bit in the bit-band region.

In the STM32F4xx devices both the peripheral registers and the SRAM are mapped to a bit-band region, so that single bit-band write and read operations are allowed. The operations are only available for Cortex®-M4 with FPU accesses, and not from other bus masters (e.g. DMA).

A mapping formula shows how to reference each word in the alias region to a corresponding bit in the bit-band region. The mapping formula is:

where:

• – bit_word_addr is the address of the word in the alias memory region that maps to the targeted bit
• – bit_band_base is the starting address of the alias region
• – byte_offset is the number of the byte in the bit-band region that contains the targeted bit
• – bit_number is the bit position (0-7) of the targeted bit

Example

The following example shows how to map bit 2 of the byte located at SRAM address 0x20000300 to the alias region:

Writing to address 0x22006008 has the same effect as a read-modify-write operation on bit 2 of the byte at SRAM address 0x20000300.

Reading address 0x22006008 returns the value (0x01 or 0x00) of bit 2 of the byte at SRAM address 0x20000300 (0x01: bit set; 0x00: bit reset).

For more information on bit-banding, please refer to the Cortex®-M4 with FPU programming manual (see Related documents on page 1 ).

2.4 Boot configuration

Due to its fixed memory map, the code area starts from address 0x0000 0000 (accessed through the ICode/DCode buses) while the data area (SRAM) starts from address 0x2000 0000 (accessed through the system bus). The Cortex®-M4 with FPU CPU always fetches the reset vector on the ICode bus, which implies to have the boot space available only in the code area (typically, flash memory). STM32F4xx microcontrollers implement a special mechanism to be able to boot from other memories (like the internal SRAM).

In the STM32F4xx, three different boot modes can be selected through the BOOT[1:0] pins as shown in Table 2 .

Table 2. Boot modes

The values on the BOOT pins are latched on the 4th rising edge of SYSCLK after a reset. It is up to the user to set the BOOT1 and BOOT0 pins after reset to select the required boot mode.

BOOT0 is a dedicated pin while BOOT1 is shared with a GPIO pin. Once BOOT1 has been sampled, the corresponding GPIO pin is free and can be used for other purposes.

The BOOT pins are also resampled when the device exits the Standby mode. Consequently, they must be kept in the required Boot mode configuration when the device is in the Standby mode. After this startup delay is over, the CPU fetches the top-of-stack value from address 0x0000 0000, then starts code execution from the boot memory starting from 0x0000 0004.

Note: When the device boots from SRAM, in the application initialization code, you have to relocate the vector table in SRAM using the NVIC exception table and the offset register.

Embedded bootloader

The embedded bootloader mode is used to reprogram the flash memory using one of the following serial interfaces:

• • USART1 (PA9/PA10)
• • USART2 (PD5/PD6)
• • I2C1 (PB6/PB7)
• • I2C2 (PB10/PB3)
• • I2C3 (PA8/PB4)
• • SPI1 (PA4/PA5/PA6/PA7)
• • SPI2 (PB12/PB13/PB14/PB15)
• • SPI3 (PA15/PC10/PC11/PC12)
• • USB OTG FS (PA11/12) in Device mode (DFU: device firmware upgrade).

The USART peripherals operate at the internal 16 MHz oscillator (HSI) frequency, while the USB OTG FS require an external clock (HSE) multiple of 1 MHz (ranging from 4 to 26 MHz).

The embedded bootloader code is located in system memory. It is programmed by ST during production. For additional information, refer to application note AN2606.

Physical remap in STM32F401xB/C and STM32F401xD/E

Once the boot pins are selected, the application software can modify the memory accessible in the code area (in this way the code can be executed through the ICode bus in place of the System bus). This modification is performed by programming the Section 7.2.1: SYSCFG memory remap register (SYSCFG_MEMRMP) in the SYSCFG controller.

The following memories can thus be remapped:

• • Main Flash memory
• • System memory
• • Embedded SRAM1

Table 3. Memory mapping vs. Boot mode/physical remap in STM32F401xB/C

1. 1. Even when aliased in the boot memory space, the related memory is still accessible at its original memory space.

1. 1. Even when aliased in the boot memory space, the related memory is still accessible at its original memory space.