The IIS Chip Gallery

Crocodilo (2025)

Additional pictures below, click to see larger versions



by

Giulio Ferraro, Vittorio Bonacina

Main Details

Application	Pulp
Technology	130
Manufacturer	IHP
Type	Teaching
Package	QFN56
Dimensions	2235μm x 2235μm
Gates	350 kGE
Voltage	1.2 V
Power	47.4 mW @80 MHz
Clock	80 MHz

Description

Modern embedded systems often require real-time processing of error-corrected sensor streams. The goal of Crocodilo is to accelerate a computational workload of this kind that consists of three main stages:

• Decoding a signal composed of 64 samples of 32-bit words that uses the Hamming (32, 26) Error Correction Code (ECC).
• Computing a Fast Fourier Transform of the decoded signal.
• Encoding back the processed signal using the same Hamming scheme.

First, we have developed a complete software implementation of this workload that can run on the original Croc design (with some minor modifications) and that will serve as a baseline. Then we have improved the original hardware in four main ways:

• Added a custom memory mapped ECC accelerator to the user domain of the SoC
• Added a fast multiplier in the CVE2 core
• Increased the SRAM capacity to 16 KBytes
• Increased the maximum clock frequency, pushing it to about 94.3 MHz

This project achieved an overall 37x performance speed-up for a real-time signal processing workload on the Croc SoC, reducing total latency from 5.2 ms to 142 us. This result was driven by four targeted hardware enhancements:

• A single-cycle ECC accelerator which boosted encode/decode performance by over 40x.
• A fast multiplier unit which accelerated the FFT stage by nearly 30x.
• An increase in SRAM capacity to 16 KB to support the application's memory footprint.
• Timing optimizations that pushed the clock frequency to 94.3 MHz.

The Crocodilo GitHub page contains the entire design.

This chip was designed as part of the VLSI design course at ETH Zurich which uses a (mostly) open source design flow for its exercises. Students are required to modify a Croc based SoC to improve its capabilities somehow to pass the course. This was one of the top-rated designs from the course and has been sent to manufacturing.

Other chips from this series include:

• Chiffre
• Fluffy
• Pjononcroc
• River

This design has received generous support from Leibniz Institute for High Performance Microelectronics through the BMBF project FMD-QNC (16ME0831).

Related Publication

No related publications yet