This website uses cookies to manage authentication, navigation, and other functions. By using our website, you agree that we can place these types of cookies on your device.

View e-Privacy Directive Documents

Cookies disallowed, click button to change.

Cookies allowed, click button to disallow.


Amber RISC Core

Amber RISC Core

Couple of years ago OpenCores website published open source implementation of ARM7 compatible processor. The proejct was quickly taken down due to ARMs enforcement of patents on certain instructions from the ARMv4 ISA. Here is the take two on open source ARM core which implements older ARMv2A ISA not protected by patents.


"Amber 2 Core Specification" - PDF document link

"1 Introduction

The Amber processor core is an ARM-compatible 32-bit RISC processor. The Amber core is fully compatible with the ARM® v2a instruction set architecture (ISA) and is therefore supported by the GNU toolset. This older version of the ARM instruction set is supported because it is not covered by patents so can be implemented without a license from ARM. The Amber project provides a complete embedded system incorporating the Amber core and a number of peripherals, including UARTs, timers and an Ethernet MAC.

There are two versions of the core provided in the Amber project. The Amber 23 has a 3-stage pipeline, a unified instruction & data cache, a Wishbone interface, and is capable of 0.8 DMIPS per MHz. The Amber 25 has a 5 stage pipeline, seperate data and instruction caches, a Wishbone interface, and is capable of 1.0 DMIPS per Mhz. Both cores implement exactly the same ISA and are 100% software compatible.

The Amber 23 core is a very small 32-bit core that provides good performance. Register based instructions execute in a single cycle, except for instructions involving multiplication. Load and store instructions require three cycles. The core's pipeline is stalled either when a cache miss occurs, or when the core performs a wishbone access.

The Amber 25 core provides 15% to 20% better performance that the 23 core but is 20% to 25% larger. Register based instructions execute in a single cycle, except for instructions involving multiplication. Load and store instructions also execute in a single cycle unless there is a register conflict with a following instruction. The core's pipeline is stalled when a cache miss occurs in either cache, when an instruction conflict is detected, or when the core performs a wishbone access."

Amber Open Source Project - web page at OpenCores link