Open source SoC development kit plugs into a USB port

QuickLogic Corp, has introduced its new Qomu development kit, a small form factor Arm Cortex M4F microcontroller (MCU) + embedded FPGA (eFPGA) combination that fits into a USB Type A port, eliminating the need for hardware setup and cabling. Together with its small size, measuring about 9 x 13 mm, the Qomu board enables users to develop anywhere. The open source SoC development kit targets edge IoT applications that require ultra-low-power consumption and artificial intelligence (AI) or machine learning (ML) capabilities. However, it can be used in a wide range of applications.

The kit, which is optimized for the QuickLogic Open Reconfigurable Computing (QORC) initiative, is supported by a variety of vendor-supported open-source development tools, including SymbiFlow synthesis, place & route, and bitstream generation, and nMigen for a Python-to-FPGA design flow. Other open-source tools include Zephyr, FreeRTOS, Renode, and TensorFlow Lite.

Image: QuickLogic (Click for a larger image.)

The Qomu development kit contains QuickLogic’s EOS S3 MCU + eFPGA SoC, providing a lot of processing power for a tiny development kit.  The Arm Cortex-M4F MCU features up to 80-MHz operating frequency and up to 512 KB SRAM. The eFPGA features 2400 effective logic cells, 64 Kb RAM, up to eight RAM/FIFO controllers, two dedicated multipliers to offload math-intensive functions, a 16-channel DMA, configurable SPI and I2C interfaces, and ultra-low-power consumption in µW.

“The MCU enables seamless software development while the embedded FPGA (eFPGA) can be used to accelerate or offload algorithms from the MCU, or to simply implement custom IP or provide glue logic,” said QuickLogic.

Image: QuickLogic (Click for a larger image.)

There is a big push now to use heterogenous architectures for implementing edge AI/ML applications, commented QuickLogic. “This usually involves integrating general-purpose processors (in this case, Arm Cortex-M4F) with more domain specific cores (in this case, the eFPGA). Until Qomu, people needed big boards to do the architecture exploration to partition AI/ML implementations between general-purpose processors and FPGAs. These boards are cumbersome, and typically connected with lots of cables, meaning people cannot do this type of development very easily outside of their office.”

Qomu literally plugs into a computer so people can develop AI/ML implementations anywhere, said the company.  “Also, since the software tools are 100% open source, there are no limitations to the actual computer being used. It could be a laptop, a desktop, a Raspberry Pi, or even a smartphone could be used to do the FPGA design.”

Example applications and gateware are readily available for free.  The Qomu development kit is available exclusively through Crowd Supply ( via Mouser Electronics).