Appeal 2007-1599
Application 10/255,748
and 28 under 35 U.S.C. § 103(a) based upon the teachings of Peyton and
Kolodner.
Appellants contend that inter alia that Peyton does not perform any
analysis of the portions of the program to which the compiler lacks access
prior to the compiling operation (Br. 9, 10, 12, and 15 to 19).
ISSUE
Does Peyton perform the claimed operations on portions of the
program to which the compiler lacks access prior to compiling the program?
FINDINGS OF FACT
Peyton describes a source code compiler that breaks up a single
computer source code listing into separate code modules, operates upon the
separate code modules, and converts the modules into executable object
code in a language translator (col. 1, ll. 6 to 27). The separate code modules
are referred to in Peyton as compilation units (CUs) (col. 1, ll. 29 to 31).
During the source code to object code conversion, cross-CU optimization is
performed in the compiler (col. 2, ll. 40 to 43). The compiler method
includes the step of analyzing each CU, and deriving: a global CU table 40
which includes a reference to each analyzed CU; a program symbol table 42
which indicates in which CU each program routine is defined and/or referred
to; a global call graph 44 which notes the routine in each CU, indicates
references therebetween, and further indicates where the routine exists in the
program symbol table; and a CU symbol table 54 which includes
information that includes a reference for each routine defined in a CU to the
intermediate representation for that routine (Figures 2 and 2a; col. 2, ll. 43 to
55). The CU symbol table 54 includes listings of variables in each of the
3
Page: Previous 1 2 3 4 5 6 Next
Last modified: September 9, 2013