By Topic

Software implications of virtual instruction set computers

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
V. Adve ; Dept. of Comput. Sci., Illinois Univ., Urbana, IL, USA ; M. Brukman ; A. Evlogimenos ; B. Gaeke

Summary form only given. Virtual instruction set computer (VISC) architectures define two separate instruction sets (one to serve as the representation of stored programs, and another to control the hardware), and use a hardware-specific translation layer to optimize and execute code on the hardware. This approach lends great flexibility to hardware and compiler design, enabling sophisticated hardware-specific optimizations at all stages of a program's lifetime. Building on our experience with the LLVM compiler infrastructure, we have proposed (1) a novel virtual instruction set for VISC architectures, and (2) a translation strategy that permits both offline and online translation, as well as dynamic optimization. Our V-ISA design is low-level and language-independent, yet retains enough high-level information to support sophisticated, "lifelong" optimization of software. The VISC strategy and V-ISA design also have important implications for high-level language virtual machines and for operating system kernels. We describe our instruction set design and translation strategy, and our ongoing work on dynamic optimization for VISC systems, and briefly discuss our work in the software implications of VISC systems.

Published in:

Parallel and Distributed Processing Symposium, 2004. Proceedings. 18th International

Date of Conference:

26-30 April 2004