DocumentCode
293674
Title
Variable tracking technique: a single-pass method to determine data dependence
Author
Wo, David ; Forward, Kevin
Author_Institution
Dept. of Electr. & Electron. Eng., Melbourne Univ., Parkville, Vic., Australia
Volume
1
fYear
1995
fDate
19-21 Apr 1995
Firstpage
177
Abstract
This paper presents a new data dependence checking technique called the variable tracking technique (VTT). It is a single-pass data dependence checking method which locates dependent statements in a serial computer program. VTT produces a schedule which lists the operations in the source code in groups. The list of operations in a particular group can be executed concurrently. The user is not required to provide a profile of the program to the compiler, hence VTT is suitable for applications which automate the process of exploiting parallelism. Here we describe the use of this technique in gacc, a parallelising compiler, which compiles C functions to field programmable gate array (FPGA) circuits. The results presented in this paper show that VTT has been instrumental in gaining improved performance from a parallelising compiler which automates the process of executing the computational intensive portion of the program in hardware
Keywords
PLD programming; field programmable gate arrays; logic design; parallel architectures; parallelising compilers; program compilers; C functions; dependent statements; field programmable gate array circuits; parallelising compiler; serial computer program; single-pass data dependence checking method; single-pass method; variable tracking technique; Application software; Concurrent computing; Coprocessors; Data engineering; Field programmable gate arrays; Hardware design languages; High performance computing; Logic circuits; Parallel processing; Programming profession;
fLanguage
English
Publisher
ieee
Conference_Titel
Algorithms and Architectures for Parallel Processing, 1995. ICAPP 95. IEEE First ICA/sup 3/PP., IEEE First International Conference on
Conference_Location
Brisbane, Qld.
Print_ISBN
0-7803-2018-2
Type
conf
DOI
10.1109/ICAPP.1995.472183
Filename
472183
Link To Document