Page 284 - Introduction to Microcontrollers Architecture, Programming, and Interfacing of The Motorola 68HC12

9.2 Expressions and Assignment Statements 261

One can complement accumulator D by complementing accumulator A and then
complementing accumulator B. We might want to implement NEGD in the same way
with a pair of instructions NEGA NEGB, but 1 is always added to A whether a carry is
generated by adding 1 to B or not. Because NEGB sets C if the contents of B is nonzero,
we can cancel the addition of 1 to A by NEGA, except when the contents of B is 0, by

NEGA
NEGB
SBCA #0

Multiplication uses a multiply instruction EMUL or EMULS, depending on
whether the operations is signed; for instance the expression in Figure 9.3, lui = Isc
* guc, is
LDAA 2, SP ; get global variable Isc
SEX A, Y ; upcast from char to int, copy to register Y
LDAB $0800 ; get guc
CLRA ; upcast from unsigned char to int
EMUL ; multiply unsigned
STD 0, SP ; put 16-bit result into local lui

Similarly, division can be implemented using a divide instruction IDIV, EDIV, or
EDIVS, depending on whether the operation is signed; for instance in Figure 9.3 the
expression lui = lui / guc; (or equivalently, lui /= guc;) is implemented

STD 2, -SP ; save numerator from previous statement
LDAB $ 8 0 0 ; get denominator guc
CLRA ; upcast from unsigned char to unsigned int
TFR D, X ; put in X for IDIV
LDD 2 f SP+ ; get global variable lui
IDIV ; divide X into D, putting quotient in X
STX 0, SP ; put into local variable lui
To discuss how increment and decrement operators are handled, we use main in
Figure 9.4 as an example; it merely increments and decrements some variables.
The simple increment and decrement instructions can be used directly with char or
unsigned char variables .The statement in Figure 9.4, guc++;, is implemented

INC $ 0 8 0 0 ; increment global variable guc

and the statement Isc—; in Figure 9.4 is implemented
DEC 0, SP ; decrement local variable Isc

However, the increment and decrement instructions on int or unsigned int variables
use a longer sequence. The first problem with this operator is that there is no 16-bit
memory operand version of it, and the second problem is that the carry bit is not affected
by the 8-bit version of this operator. The statement gsi++; in Figure 9.4 is coded as