NAME

cflow− generate C flow graph

SYNOPSIS

cflow [−r] [−ix] [−i_ ] [−dnum] files

HP-UX COMPATIBILITY

Level: HP-UX/STANDARD

Origin: UCB

DESCRIPTION

Cflow analyzes a collection of C, YACC, LEX, assembler, and object files and attempts to build a graph charting the external references.  Files suffixed in .y, .l, .c, and .i are YACC’d, LEX’d, and C-preprocessed (bypassed for .i files) as appropriate and then run through the first pass of lint(1). (The −Y option of cc(1) and the −I, −D, and −U options of the C-preprocessor are also understood.)  Files suffixed with .s are assembled and information is extracted (as in .o files) from the symbol table.  The output of all this non-trivial processing is collected and turned into a graph of external references which is displayed upon the standard output. 

Each line of output begins with a reference (i.e., line) number, followed by a suitable number of tabs indicating the level.  Then the name of the global (normally only a function not defined as an external or beginning with an underscore; see below for the −i inclusion option) a colon and its definition.  For information extracted from C source, the definition consists of an abstract type declaration (e.g., char ∗), and, delimited by angle brackets, the name of the source file and the line number where the definition was found.  Definitions extracted from object files indicate the file name and location counter under which the symbol appeared (e.g., text). Leading underscores in C-style external names are deleted.

Once a definition of a name has been printed, subsequent references to that name contain only the reference number of the line where the definition may be found.  For undefined references, only <> is printed. 

As an example, given the following in file.c:

inti;
 main()
{
f();
g();
f();
}
 f()
{
i = h();
}

the command

cflow −ix file.c

produces the output

1main: int(), <file.c 4>
2f: int(), <file.c 11>
3h: <>
4i: int, <file.c 1>
5g: <>

When the nesting level becomes too deep, the −e option of pr(1) can be used to compress the tab expansion to something less than every eight spaces.

The following options are interpreted by cflow:

−r Reverse the “caller:callee” relationship producing an inverted listing showing the callers of each function.  The listing is also sorted in lexicographical order by callee. 

−ix Include external and static data symbols. The default is to include only functions in the flowgraph. 

−i_ Include names that begin with an underscore. The default is to exclude these functions (and data if −ix is used). 

−dnum The num decimal integer indicates the depth at which the flowgraph is cut off.  By default this is a very large number.  Attempts to set the cutoff depth to a nonpositive integer will be met with contempt. 

HARDWARE DEPENDENCIES

Series 200:

The size of the internal compiler tables, used by the first pass of lint(1), can be adjusted by using the -N option.  The syntax for this option is described in the HARDWARE DEPENDENCIES section for Series 200 in the manual page for cc(1).

DIAGNOSTICS

Complains about bad options.  Complains about multiple definitions and only believes the first.  Other messages may come from the various programs used (e.g., the C-preprocessor). 

SEE ALSO

as(1), cc(1), cpp(1), lex(1), lint(1), nm(1), pr(1), yacc(1). 

BUGS

Files produced by lex(1) and yacc(1) cause the reordering of line number declarations which can confuse cflow. To get proper results, feed cflow the yacc or lex input. 

Hewlett-Packard  —  last mod. May 11, 2021