NANDIO.PIO

Accelerating NAND Flash Communication using PIO (Programmable IO).

Features

• High-speed NAND flash communication using PIO and payload builder
• Simulation environment for verification
• async/await friendly API for MicroPython
• Supports Raspberry Pi Pico and JISC-SSD (Jisaku In-Storage Computation SSD) board

Waveforms

Online Simulation

https://wipeseals.github.io/nandio.pio/

Performance

JISC-SSD board with RP2040 and NAND Flash (TC58NVG0S3HTA00) performance comparison.

CPU Clock: 125MHz, PIO Clock: 125MHz

MPY: soft reboot
CPU frequency: 125.0 MHz
# `Fw` commander results:
- Read ID time      : 3391 us
- Erase block time  : 4785 us
- Program page time : 8560594 us
- Read page time    : 3692423 us

# `Pio` commander results:
- Read ID time      : 3527 us
- Erase block time  : 6357 us
- Program page time : 21405 us
- Read page time    : 5491 us

MicroPython v1.25.0 on 2025-04-15; Raspberry Pi Pico with RP2040
Type "help()" for more information.
>>>

CPU Clock: 250MHz, PIO Clock: 125MHz

MPY: soft reboot
CPU frequency: 250.0 MHz
# `Fw` commander results:
- Read ID time      : 1771 us
- Erase block time  : 2353 us
- Program page time : 5421321 us
- Read page time    : 1845875 us

# `Pio` commander results:
- Read ID time      : 1735 us
- Erase block time  : 3733 us
- Program page time : 11197 us
- Read page time    : 3293 us

MicroPython v1.25.0 on 2025-04-15; Raspberry Pi Pico with RP2040
Type "help()" for more information.

Installation

To install the project, you can use the uv tool, which is a Python package manager that simplifies the installation of Python projects with multiple dependencies. You can install uv and the project dependencies using the following commands:

pip install uv
uv sync --all-extras --dev

# run unit tests
uv run pytest

Usage

Local Simulation

To simulate the NAND Flash communication, you can use the provided simulation script. This will run the simulation using the uv command.

# All scenarios
uv run sim/cli.py sim --all
Simulating scenario... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% 0:00:00
Simulation completed for scenarios: reset, read_id, read, program, erase, status_read
All simulations completed successfully. output saved to </path/to/output>

# Specific scenario
uv run sim/cli.py sim --scenario reset

You can view the simulation results on GitHub Pages:
https://wipeseals.github.io/nandio.pio/

Assembly

To assemble the PIO program, you can use the uv command with the asm option. This will compile the PIO assembly code into a binary format that can be used by the Raspberry Pi Pico.

uv run sim/cli.py asm
PIO program assembled successfully!
Output binary saved to: </path/to/output/nandio.pio.bin>

JISC-SSD Board (RP2040 + NAND Flash)

Install MicroPython

• Download the MicroPython UF2 file for RP2040 from the official Raspberry Pi Pico website and flash it to the board.
  • See MicroPython - Raspberry Pi for details.
  • The confirmed working version of MicroPython is v1.24.1 (other versions may not work properly).

Run MicroPython scripts

• Transfer the project to the Raspberry Pi Pico and run.

Using vscode MicroPico Extension

Run the MicroPico: Upload project to Pico (@command:micropico.upload) command to upload the project to the Raspberry Pi Pico.

Using mpremote

uvx mpremote connect COM13 + fs --recursive --force cp main.py :/main.py + cp  nandio.py :/nandio.py + cp  sim/nandio_pio.py :/sim/nandio_pio.py  + cp  mpy/driver.py :/mpy/driver.py + fs ls + soft-reset + run main.py + repl
cp main.py :/main.py
cp sim/nandio_pio.py :/sim/nandio_pio.py
Up to date: /sim/nandio_pio.py
cp mpy/driver.py :/mpy/driver.py
ls :
         607 main.py
           0 mpy/
           0 sim/
(snip)

References

• [VOL-28]JISC-SSD(Jisaku In-Storage Computation SSD 学習ボード)
• TC58NVG0S3HTA00 Datasheet
• RP2040 Datasheet
• GitHub - crane-elec/rawnand_test
• ゼロから学ぶ SSD：構造、動作、耐久性のポイント - キオクシア株式会社

License

This project is licensed under the MIT License - see the LICENSE file for details.