Open Source Hardware Development and the OpenRISC Project : A Review of the OpenRISC Architecture and Implementations

• Main Author: Baxter, Julius
• Format: Others
• Language: English
• Published: KTH, Skolan för informations- och kommunikationsteknik (ICT) 2011
• Subjects: TECHNOLOGY; TEKNIKVETENSKAP
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-48819

Advances in the design and manufacture of microelectronic devices since the 1960s have enabled embedded computers that are ubiquitous. Microprocessors, the core component in modern computers, and their architectures have evolved continuously over this time, too. During the 1980s a new architectural approach, favoring a reduction in design complexity, emerged and became known as reduced instruction set computer, or RISC, architectures. The mid-1980s also saw the beginning of a widespread change in the attitudes towards computer software. The Free Software Foundation (FSF) was set up and aimed to foster the development of free (as in freedom) and open source software, as a reaction to increasingly protective measures software vendors were taking to restrict the use of their software. The concept of less restrictive software has proved successful but it took over fifteen years before this philosophy was applied to the discipline of electronic hardware design. One of the earliest and most prominent projects to do so was initiated in the late 1990s by a group of students aiming to develop an open source microprocessor architecture and set of implementations. Their goals were realised in the OpenRISC project, a RISC microprocessor specification and implementations. The initial development team then created OpenCores, an online open source hardware design community focusing on developing register transfer level (RTL) designs of functional cores based on the principles of the open source software movement. The application of open source principles to hardware design gathered pace throughout the decade that followed, but despite good progress early in the OpenRISC project, it slowed as the maintainers decided to pursue commercial interests and ceased development of the publicly released versions. Recent interest in the architecture, and increased uptake in use of open source hardware, has lead to a much-needed rejuvenation of the project. Twelve years on from the inception of the OpenRISC project, this work has lead to questions about the state of the open source hardware development movement, and about a possible successor to the first OpenRISC architecture. This document will discuss the underlying technologies and philosophies of the OpenRISC project, present the recent work on the platform, undertake a critical analysis of the project as a whole, and present a section on the future directions of the OpenRISC 1000 project in particular, and open source hardware development in general. Recommendations of specific work to be done on the project and arguments for the general direction of development are presented. Finally, the proposed successor architecture, OpenRISC 2000, is discussed.