DOC HOME SITE MAP MAN PAGES GNU INFO SEARCH PRINT BOOK
 
Using programming tools

Using cflow

The cflow command presents a list of the functions and function calls within a program. This list can be helpful when debugging a program, especially if there are external references. The information displayed by cflow indicates the source filename and the line number of the function, and lists additional functions that are in turn called by that function. To see a list of the functions called within testcase.c, we can run cflow:

   cflow testcase.c
   1	main: int (), <testcase.c 76>
   2		gets: <>
   3		GetWords: int (), <testcase.c 43>
   4			strlen: <>
   5		PrintWords: void (), <testcase.c 21>
   6			printf: <>
Output is displayed in the following fields:

line number
the number of the line in the output

tabs
the level at which the symbol is defined

symbol name
the name of the symbol under consideration followed by a colon

symbol information
the type of the symbol and the arguments to the symbol, if any

location
the name of the file and the number of the line within the file in which the symbol is defined
In analyzing the output, we see that the function main is the first function called by testcase.c and is defined on line 76 as shown in the angle brackets (<>). Notice the function main returns a type int and it calls the functions gets, GetWords, and PrintWords. The <> next to the gets function are empty because cflow only looks in the source file to resolve function call references. The gets function is not in the source code (testcase.c) and, therefore, cflow does not have any information on it. The function main calls two additional functions, GetWords and PrintWords. Both of these functions are found in the source file and, therefore, their locations are displayed. The two additional functions, strlen and printf, called respectively from GetWords and PrintWords, do not exist within testcase.c; hence, no line number information is available for them.

As a result of analyzing testcase.c with cflow, we now know the names of all of the functions referenced in the source file and the location of their definitions in the source file. If we need to change any of them, they should be easy to find. For a larger more complex program, you can make decisions on where best to optimize code, either by reducing the number of function calls or by identifying obsolete functions and removing them. For more information on using cflow, see cflow(CP).


Next topic: Using cxref
Previous topic: Using lint

© 2003 Caldera International, Inc. All rights reserved.
SCO OpenServer Release 5.0.7 -- 11 February 2003