.. zephyr:board:: rzv2n_evk

Overview

The RZ/V2N is a vision AI microprocessor (MPU) with Renesas proprietary AI accelerator (DRP-AI3) supporting up to 15TOPS AI performance. Its CPUs are quad Arm Cortex-A55 (1.8GHz) and Arm Cortex-M33 (200MHz). The RZ/V2N is equipped with an ISP (Image Signal Processor) and dual-channel MIPI CSI-2 camera interfaces for supporting dual camera signal processing, which is crucial for realizing vision systems. It is also equipped with high-speed interfaces such as PCIe and USB 3.2, allowing for the expansion of external devices. The RZ/V2N is an ideal microprocessor for applications requiring both low power consumption and advanced AI inference, such as DMS (Driver Monitoring System), monitoring cameras, mobile robots, and more.

RZ/V2N Evaluation Board (CPU Board):
- CPU: RZ/V2N
- PMIC: RAA215300
- Clock generator: 5L35023B
- Main memory: LPDDR4X 8GB x1
- xSPI Flash memory: 64MB
- External memory: microSD x1
- Board-to-board connector for sub boards
- High-speed interface:
  - Gigabit Ethernet x2 ports
  - USB 3.2 Gen2 x1ch (Host only)
  - USB 2.0 x1ch (OTG)
  - PCIe Gen3 x1ch (2 lanes max)
  - MIPI CSI-2 x2ch
  - MIPI DSI x 1ch
  - GPIO: 86 pins
RZ/V2H-EVK Expansion Board (EXP Board): Shared with RZ/V2H
- HDMI Tx x1ch
- Audio auxiliary input connector x1ch
- Audio microphone input connector x1ch
- Audio headphone output connector x1ch
- Pmod x4

Hardware

The Renesas RZ/V2N documentation can be found at RZ/V2N Group Website_

.. figure:: rzv2n_block_diagram.webp :width: 600px :align: center :alt: RZ/V2N group feature

    RZ/V2N block diagram (Credit: Renesas Electronics Corporation)

Detailed hardware features for the board can be found at RZV2N-EVK Website_

Supported Features

.. zephyr:board-supported-hw::

Connections and IOs

By default, the board is configured for use with:

SCI_UART

SCI_UART0:
- TX = Pmod 3A pin 2
- RX = Pmod 3A pin 3

LED

LED5 (yelow) = P0_0
LED6 (yelow) = P0_1

Programming and Debugging (M33)

.. zephyr:board-supported-runners::

Applications for the rzv2n_evk board can be built in the usual way as documented in :ref:build_an_application.

Console

The UART port for Cortex-M33 System Core can be accessed by connecting Pmod USBUART <https://store.digilentinc.com/pmod-usbuart-usb-to-uart-interface/>_ to the upper side of PMOD Type 3A.

Debugging

It is possible to load and execute a Zephyr application binary on this board on the Cortex-M33 System Core from the internal SRAM, using JLink debugger (:ref:jlink-debug-host-tools).

Here is an example for building and debugging with the :zephyr:code-sample:hello_world application.

.. zephyr-app-commands:: :zephyr-app: samples/hello_world :board: rzv2n_evk/r9a09g056n48gbg/cm33 :goals: build debug

Flashing

RZ/V2N-EVK is designed to start different systems on different cores. It uses Yocto as the build system to build Linux system and boot loaders to run Zephyr on Cortex-M33 with u-boot. The minimal steps are described below.

Download Multi-OS Package from the RZ/V2N Group Handbook_
Unzip Multi-OS Package then open "r01qs0078ej****-rzv2n-multi-os-pkg" PDF file.
Follow "3.2 Integration of OpenAMP related stuff" to build RZ/V2N AI SDK Source Code. Uncomment the following lines in meta-rz-features/meta-rz-multi-os/meta-rzv2n/conf/layer.conf.

.. code-block:: console

#MACHINE_FEATURES_append = " RZV2N_CM33_BOOT" MACHINE_FEATURES_append = " SRAM_REGION_ACCESS" #MACHINE_FEATURES_append = " CM33_FIRMWARE_LOAD" #MACHINE_FEATURES_append = " CA55_CPU_CLOCKUP"

The below necessary artifacts will be located in the build/tmp/deploy/image

Carry out 1-9 of Step "1-B Write the Linux files to SD card" at Step 7 Deploy AI Application_ to prepare SD card.
Run the following commands to write zephyr.bin to SD card.

.. code-block:: console

$ sudo mkdir /mnt/sd -p $ sudo mount /dev/sdb2 /mnt/sd $ sudo cp /path/to/zephyr.bin /mnt/sd/boot $ sync $ sudo umount /mnt/sd

.. warning::

Change /dev/sdb to your microSD card device name. Use dh -h to check.

Insert the SD card into SD1 of RZ/V2N EVK, set up board at eSD mode (DSW1[4:5] = ON, OFF).
Turn on the board carefully, see "3.3.2.4 Power Supply" in Getting Started with Flexible Software Package_ Then, you should see the following message on the console associated with CN12 of RZ/V2N EVK.

.. code-block:: console

U-Boot 2021.10 (Sep 19 2024 - 10:13:55 +0000) CPU: Renesas Electronics CPU rev 1.0 Model: Renesas EVK based on r9a09g056 DRAM: 7.9 GiB MMC: mmc@15c00000: 0, mmc@15c10000: 1 Loading Environment from SPIFlash... SF: Detected mt25qu512a with page size 256 Bytes, erase size 64 KiB, total 64 MiB OK In: serial@11c01400 Out: serial@11c01400 Err: serial@11c01400 Net: eth0: ethernet@15c30000 Hit any key to stop autoboot: 3

Hit any key within 3 sec to stop autoboot.
Carry out the following setup of u-boot to kick CM33 Core.

.. code-block:: console

=> setenv cm33start 'dcache off => mw.l 0x10420D2C 0x02000000 => mw.l 0x1043080c 0x08003000 => mw.l 0x10430810 0x18003000 => mw.l 0x10420604 0x00040004 => mw.l 0x10420C1C 0x00003100 => mw.l 0x10420C0C 0x00000001 => mw.l 0x10420904 0x00380008 => mw.l 0x10420904 0x00380038 => ext4load mmc 0:2 0x08003000 boot/zephyr.bin => mw.l 0x10420C0C 0x00000000 => dcache on' => saveenv => run cm33start

References

.. target-notes::

.. _RZ/V2N Group Handbook: https://www.renesas.com/en/document/gde/rzv2n-group-handbook

.. _Step 7 Deploy AI Application: https://renesas-rz.github.io/rzv_ai_sdk/5.10/getting_started_v2n.html#step7

.. _Getting Started with Flexible Software Package: https://www.renesas.com/en/document/apn/rzv-getting-started-flexible-software-package

.. _RZV2N-EVK Website: https://www.renesas.com/en/products/microcontrollers-microprocessors/rz-mpus/rzv2n-evk-rzv2n-quad-core-vision-ai-mpu-evaluation-kit

.. _RZ/V2N Group Website: https://www.renesas.com/en/products/microcontrollers-microprocessors/rz-mpus/rzv2n-15tops-quad-core-vision-ai-mpu-2-camera-connection-and-excellent-power-efficiency