Efficient code generation for ASIPs with different word sizes
نویسندگان
چکیده
We propose a complete methodology for extending our automatic ASIP (Architecture Specific Instruction set Processor) synthesis framework to a much wider target architecture space. In this new architecture space the width of the integer data word and of any hardware resource data path is user-definable and application specific. This methodology, developed on the basis of a retargetable C compiler (the SUIF compiler), requires the definition of a new intermediate format and a complete set of non-ambiguous rules determining the bit precision of each integer operation depending on the size of its operands. We also assess the implementation requirements and the complexity increases of the scheduler and the hardware template. Examples illustrate the methodology.
منابع مشابه
A Retargetable Compiler of VLIW ASIP for Media Signal Processing
In the last decade extensive researches have been carried out in ASIP (Application Specific Instruction Processor) design field. One of the key steps in ASIP design is code generation by a retargetable compiler. In this paper we describe our experience in implementing a retargetable compiler for VLIW ASIP based on ORC (Open Research Compiler) framework. Orienting towards a new register file acc...
متن کاملInstruction Set De nition and Instruction Selection for ASIPsJohan
Application Speciic Instruction set Processors (ASIPs) are eld or mask programmable processors of which the architecture and instruction set are opti-mised to a speciic application domain. ASIPs ooer a high degree of exibility and are therefore increasingly being used in competitive markets like telecommunications. However, adequate CAD techniques for the design and programming of ASIPs are mis...
متن کاملEfficient OVSF code assignment strategy in UMTS with Multiple Codes
In order to provide variable data rates in a flexible manner to support applications with different bandwidth requirements, the orthogonal variable spreading factor (OVSF) codes, adopted by the third-generation partnership project technical specifications, are used as channelization codes. This paper considers the environment of using multiple OVSF codes to support a request with any data rate,...
متن کاملCode optimizations for a VLIW-style network processing unit
The explosive growth in network bandwidth and Internet services such as QoS (quality of service) and SLA (service level agreement) monitoring have created the need for new networking hardware called a Network Processing Unit (NPU). In order to rapidly reconfigure the NPU for frequently varying Internet services and technologies, a high-performance C compiler is urgently needed. Several code gen...
متن کاملAn Automatic System for Application-Specific Instruction Format Design and Code Generation for VLIW and EPIC processors
Introduction. Whereas the workstation and personal computer markets are rapidly converging on a small number of similar architectures, the embedded systems market is enjoying an explosion of architectural diversity. This diversity is driven by demands for higher performance at a lower cost and power consumption, and is propelled by the possibility of designing application-specific instruction-s...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1997