Page 254 - Introduction to Microcontrollers Architecture, Programming, and Interfacing of The Motorola 68HC12

8,5 Procedures and Their Arguments 231

element of the structs in the table. Index addressing is useful for accessing elements in a
row of a table. If the address register points to the first word of any row, then the
displacement can be used to access words in any desired column. Also, autoincrement
addressing can be used to select consecutive words from a row of the table.
In C, a table tbl is considered a vector whose elements are structures. For instance,
the declaration
struct {char ll;int 12;char 13;} tbl[3];

allocates a table whose rows are similar to the list list above. The dot notation with
indexes can be used to access it, as in
a = tbl[2].11;

In simple compilers, multidimensional arrays and structs are not implemented. They
can be reasonably simulated using one-dimensional vectors. The user becomes
responsible for generating vector index values to access row-column elements or struct
elements.
Finally a comment is anything enclosed by / * and * / . These can be put
anywhere in your program, except within quotation marks. Alternatively, everything
after / / on a source line is considered a comment. However, the latter syntax is not
available on all C compilers.

8.5 Procedures and Their Arguments

A procedure in C may be called by another procedure in C as a procedure. The arguments
may be the data themselves, which is call by value, or the address of the data, which is
call by name. Call by reference is not used in C (it is often used in FORTRAN).
Consider the following example: RaisePower computes i to the power j , returning
the answer in k where i, j, and k are integers. The variable i is passed by value,
while j and k are passed by name. The calling procedure would have
RaisePower (i, & j, &k); and the called procedure would have
void RaisePower (int i, int *j, int *k) { int n;
for(*k =1 , n = 0; n < *j; n++) *k =* k * i;
}
Formal parameters are listed after the procedure name in parentheses, as in (i j,k),
and in the same order they are listed after the procedure name as they would be for a
declaration of local variables. However, they are listed before the curly bracket ({).
Call by value, as i is passed, does not allow data to be output from a procedure,
but any number of call by value input parameters can be used in a procedure. Actual
parameters passed by name in the calling procedure have an ampersand (&) prefixed to
them to designate that the address is put in the parameter. In the called procedure, the
formal parameters generally have an asterisk (*) prefixed to them to designate that the
data at the address are accessed. Observe that call by name formal parameters j or k
used inside the called procedure all have a prefix asterisk. A call by name parameter can
pass data into or out of a procedure, or both. Data can be input into a procedure using
call by name, because the address of the result is passed into the procedure and the