Motorola M68000 Betriebsanweisung PDF herunterladen (Seite 165)

Advanced Arithmetic 151

JSR GETC

GET A CHARCTER

SOB.B # ' 0'/DO

MAKE INTO DECIMAL

LSL.B #4,DO

SHIFT INTO HIGH NIBBLE

MOVE.L DO/-(SP) SAVE DO

JSR

GETC

GET A CHARACTER

SUB.B #'0',D0 MAKE INTO DECIMAL

ADD.L

(SP)+/DO

ADD HIGH NIBBLE

This routine does not have any error checking, so if a character other

than 0 through 9 is entered, an error will result. It is straightforward to

add instructions to test the range of each digit as it enters to ensure that

it is valid.

BCD numbers are unsigned. The high-order bit of the high-order byte

is not interpreted as a sign bit. How then can we represent signed BCD

numbers? One method is to reserve a single byte as the sign byte. We can

store the ASCII character + or — in this sign byte and the absolute value

of the BCD number in the remaining bytes. Unfortunately we can’t just

add or subtract these numbers. The sign byte can’t enter into the actual

arithmetic. Before using a signed BCD number, the sign byte must first

be checked. If the BCD number is negative, tens complement must first

be taken prior to its use in a calculation. The ten’s complement is used

in a manner very similar to the two’s complement of binary numbers.

The ten’s complement is the nine’s complement plus one. To compute

the nine’s complement, simply subtract each BCD digit from the number

nine, then add in the one. For example: the nine’s complement of the

four-digit BCD number 1234 is 8765. The ten’s complement is then 8766.

This adjusted number used in a calculation will now give a correct result.

From the above example, the number 8766 can be thought of as a negative

1234. If we add a positive 1234, we should get a result of zero. This is

true if we ignore the carry.

8766 -1234

1234 +1234

10000 0000

The same rules apply as for two’s complement binary numbers. Unfor

tunately, we must keep track of the signs separately, or use more com

plicated schemes. One such scheme would be to use the high-order digit

for the sign, with a positive number being represented by a 0 and a neg

ative number by a 9. This is equivalent to the use of the high-order bit

as a sign bit with two’s complement binary numbers.

The NBCD instruction is used to compute the ten’s complement. The

action of this instruction is to perform the ten’s complement of a pair

of BCD bytes and then subtract the extend bit. If the extend bit is 0,

then the result is truly the ten’s complement. If the extend bit is a 1, then

1 2 ... 160 161 162 163 164 165 166 167 168 169 170 ... 255 256

Kommentare zu diesen Handbüchern

Keine Kommentare

Motorola M68000 Betriebsanweisung Seite 165

Kommentare zu diesen Handbüchern

Verwandte Produkte und Handbücher für Nein Motorola M68000