Procedure Multiplication Division.pdf

Procedure,
Multiplication & Division
5/2/2016
EEE 315: MICROPROCESSOR & INTERFACING
Md.Ayaz Masud
Lecturer,Dept of EEE, BUET

Procedures
Decompose the original problem into a
series of sub-problems that are easier to
solve
Main procedure contains the entry to the
program
Main procedure calls one of the other
procedures
These procedures can call each other and a
procedure can also call itself
When one procedure calls another, control
transfers to the called procedure
5/2/2016
Md.Ayaz Masud

Procedures
 Syntax:
name PROC type
;body of the procedure
RET
name ENDP
 Type (near or far) is optional
 Near: the statement that calls the procedure is in
the same segment as the procedure itself. NEAR
is assumed if type is omitted.
 Far: the statement that calls the procedure is in a
different segment.
5/2/2016
Md.Ayaz Masud

Procedures
5/2/2016
Md.Ayaz Masud
Main PROC
CALL PROC1
next instruction
PROC1 PROC
first instruction
RET

Procedures
The RET instruction causes control to
transfer back to the calling procedure.
Every procedure should have a RET
someplace (except the main procedure)
Usually it is the last statement in the
procedure
5/2/2016
Md. Ayaz Masud

CALL and RET
To invoke a procedure, the CALL instruction
is used.
There are two kinds of procedure calls,
direct and indirect.
Direct: CALL name
Indirect: CALL address_expression
address_expression specifies a register or
memory location containing the address of
a procedure.
5/2/2016
Md. Ayaz Masud

Executing CALL instruction
The return address of the calling
program is saved on the stack. This is the
offset of the next instruction after the
CALL statement. CS:IP has segment:
offset of this instruction at the time when
the call is executed
 IP gets the offset address of the first
instruction of the procedure. This
transfers control to the procedure.
5/2/2016
Md. Ayaz Masud

Before CALL
CODE SEGMENT STACK SEGMENT
5/2/2016
Md. Ayaz Masud
OFFSET
address
Content
Main PROC
0010H CALL PROC1
0012H next instruction
…….
PROC PROC1
0200H first instruction
……..
0300H RET
IP
OFFSET
address
Content
00FCH
00FDH
00FE H
00FF H
0100 H
SP

After CALL
5/2/2016
Md. Ayaz Masud
OFFSET
address
Content
Main PROC
0010H CALL PROC1
…….
PROC PROC1
……..
0300H RET
IP
SP
OFFSET
address
Content
00FCH
00FDH
00FE H 12H
00FF H 00H
0100 H

Executing RET instruction
Syntax: RET pop_value
The integer argument pop_value is
optional.
For a NEAR procedure, execution of RET
causes the stack to be popped into IP.
If a pop value N is specified, it is added to
SP and thus has the effect of removing N
additional bytes from the stack.
CS:IP now contains the segment:offset of the
return address and control returns to the
calling program.
5/2/2016
Md. Ayaz Masud

Before RET
5/2/2016
Md. Ayaz Masud
OFFSET
address
Content
Main PROC
0010H CALL PROC1
…….
PROC PROC1
……..
0300H RET
IP
SP
OFFSET
address
Content
00FCH
00FDH
00FE H 12H
00FF H 00H
0100 H

After RET
5/2/2016
Md. Ayaz Masud
OFFSET
address
Content
Main PROC
0010H CALL PROC1
…….
PROC PROC1
……..
0300H RET
IP
SP
OFFSET
address
Content
00FCH
00FDH
00FE H XX
00FF H XX
0100 H

Multiplication Instruction
Syntax:
MUL source (for unsigned multiplication)
IMUL source (for signed multiplication)
If two bytes are multiplied, the product is a
word (16 bits).
If two words are multiplied, the product is a
doubleword (32 bits).
For multiplication of positive numbers both
MUL and IMUL give the same result.
5/2/2016
Md. Ayaz Masud

Multiplication Instruction:
BYTE form
Multiplier in source byte
Multiplicand in AL
16 bit product in AX
The source may be a byte register, a
memory byte, but not a constant
MUL BL
MUL MYBYTE
5/2/2016
Md. Ayaz Masud

WORD form
Multiplier in source word
Multiplicand in AX
Most significant 16 bits of product in DX
Least significant 16 bits of product in AX
The source may be a 16 bit register, a
memory word but not a constant
MUL BX
MUL MYWORD
5/2/2016
Md. Ayaz Masud

Effect on Flags
SF, ZF, AF, PF : Undefined
CF/ OF:
5/2/2016
Md. Ayaz Masud
Instruction CF/OF
MUL
0, if the upper half of the result is zero
1, otherwise
IMUL
0, if the upper half of the result is the sign
extension of the lower half
1, otherwise

Examples
AX=FFFF H, BX=FFFF H
AL=80 H, BL=FF H
5/2/2016
Md. Ayaz Masud
Instruction Decimal
Product
Hex
Product
DX AX CF/OF
MUL BX 4294836225 FFFE0001 FFFE 0001 1
IMUL BX 1 00000001 0000 0001 0
Instruction Decimal
Product
Hex
Product
AH AL CF/OF
MUL BL 32640 7F80 7F 80 1
IMUL BL 128 0080 00 80 1

Division Instruction
 Syntax:
DIV divisor (for unsigned multiplication)
IDIV divisor (for signed multiplication)
 The quotient and the remainder have the same
size as the divisor.
 All the flags are undefined
 Divide Overflow: If the quotient is too long to fit
in the destination, program terminates and the
system displays the message“Divide Overflow”
5/2/2016
Md. Ayaz Masud

Division Instruction:
BYTE form
Divisor is an 8 bit register or memory
byte
16 bit dividend is in AX
8 bit quotient is in AL
8 bit remainder is in AH
5/2/2016
Md. Ayaz Masud

Division Instruction:
WORD form
Divisor is a 16 bit register or memory
word
32 bit dividend is in DX:AX
16 bit quotient is in AX
16 bit remainder is in DX
5/2/2016
Md. Ayaz Masud

Sign Extension of the Dividend
 Word division: If the dividend fits in AX only
– For DIV, DX should be cleared
– For IDIV, DX should be made sign extension of
AX. The instruction CWD (Convert Word to
Doubleword)will do the extension
 Word division: If the dividend fits in AL only
– For DIV, AH should be cleared
– For IDIV, AH should be made sign extension
of AL.The instruction CBW (Convert Byte to
Word) will do the extension
 CBW and CWD has no operands and has no effect
on flags
5/2/2016
Md. Ayaz Masud

Procedure Multiplication Division.pdf

More Related Content

Similar to Procedure Multiplication Division.pdf

Recently uploaded

Procedure Multiplication Division.pdf