Patent
1996-09-27
1998-11-24
Banankhah, Majid A.
395705, G06F 945
Patent
active
058420223
ABSTRACT:
The entire space of a loop is analyzed for dependencies between target array elements so that dependency ID's are assigned to the target array elements. Optimization is carried out on the basis of the dependency ID's. Dependency ID's covering dependency for an arbitrary turn of the loop are reassigned to the target array elements that have been subjected to optimization, whereupon a further optimization is carried out. An examination for identifying overlappings is carried out on the basis of the dependency ID assigned to the target array elements so that instruction scheduling is carried out.
REFERENCES:
patent: 5109331 (1992-04-01), Ishida et al.
"Advanced Compiler Optimizations For SuperComputers", David A. Padua and Michael J. Wolfe; Comm of The ACM. Dec. 1986, vol. 29.
"Compiler Code Transformations for Supercolor-Based High-performance System", Scott A. Mahlke et al., 1992 IEEE.
"Compilers, Principal, Techniques, and Tool", Alfred V. Aho et al., 1986 Bell Tel. Laboratories, Incorp.
Haraguchi Masatoshi
Nakahira Tadashi
Banankhah Majid A.
Fujitsu Limited
LandOfFree
Loop optimization compile processing method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Loop optimization compile processing method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Loop optimization compile processing method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1713395