Appeal No. 2004-0098
Application No. 09/393,082
BACKGROUND
The invention is directed to a method and system for detecting and counting
bytecode (e.g., JAVA intermediate code) sequences for optimization of the code. The
method “convolves” a sequence tree into a bytecode sequence occurrence tree to
extract the frequency of occurrences of common subsequences. “Convolving” a tree is
defined in appellants’ specification (at 7) as traversing a first tree to create a second
tree which contains the union of all subtrees of the first tree. Representative claim 6 is
reproduced below.
6. A method in a data processing system, the method comprising the
computer-implemented steps of:
generating a bytecode sequence tree data structure from a set of
bytecode sequences; and
convolving the bytecode sequence tree data structure into a bytecode
subsequence occurrence tree data structure.
The examiner relies on the following references:
Simon S. Lam et al. (Lam),“A Tree Convolution Algorithm for the Solution of Queueing
Networks,” Communications of the ACM, Vol. 26, No. 3, pp. 203-215 (Mar. 1983).
Justiani Hendren et al. (Hendren), “Supporting Array Dependence Testing for an
Optimizing/Parallelizing C Compiler,” Compiler Construction, 5th Int’l Conference, CC
‘94 Edinburgh, U.K. Proceedings, pp. 309-323 (Apr. 7-9, 1994).
H. Dan Lambright (Lambright), “Java Bytecode Optimizations,” IEEE1 pp. 206-210
(1997).
1 The reference appears to be a reprint from a paginated, unidentified IEEE journal.
-2-
Page: Previous 1 2 3 4 5 6 7 Next
Last modified: November 3, 2007