Para que tengas una idea, es un ejemplo que viene con Emu8086.
name "calc2"
; command prompt based simple calculator (+,-,*,/) for 8086.
; example of calculation:
; input 1 <- number: 10
; input 2 <- operator: -
; input 3 <- number: 5
; -------------------
; 10 - 5 = 5
; output -> number: 5
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;; this maro is copied from emu8086.inc ;;;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; this macro prints a char in AL and advances
; the current cursor position:
PUTC MACRO char
PUSH AX
MOV AL, char
MOV AH, 0Eh
INT 10h
POP AX
ENDM
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
org 100h
jmp start
; define variables:
msg0 db "note: calculator works with integer values only.",0Dh,0Ah
db "to learn how to output the result of a float division see float.asm in examples",0Dh,0Ah,'$'
msg1 db 0Dh,0Ah, 0Dh,0Ah, 'enter first number: $'
msg2 db "enter the operator: + - * / : $"
msg3 db "enter second number: $"
msg4 db 0dh,0ah , 'the approximate result of my calculations is : $'
msg5 db 0dh,0ah ,'thank you for using the calculator! press any key... ', 0Dh,0Ah, '$'
err1 db "wrong operator!", 0Dh,0Ah , '$'
smth db " and something.... $"
; operator can be: '+','-','*','/' or 'q' to exit in the middle.
opr db '?'
; first and second number:
num1 dw ?
num2 dw ?
start:
mov dx, offset msg0
mov ah, 9
int 21h
lea dx, msg1
mov ah, 09h ; output string at ds:dx
int 21h
; get the multi-digit signed number
; from the keyboard, and store
; the result in cx register:
call scan_num
; store first number:
mov num1, cx
; new line:
putc 0Dh
putc 0Ah
lea dx, msg2
mov ah, 09h ; output string at ds:dx
int 21h
; get operator:
mov ah, 1 ; single char input to AL.
int 21h
mov opr, al
; new line:
putc 0Dh
putc 0Ah
cmp opr, 'q' ; q - exit in the middle.
je exit
cmp opr, '*'
jb wrong_opr
cmp opr, '/'
ja wrong_opr
; output of a string at ds:dx
lea dx, msg3
mov ah, 09h
int 21h
; get the multi-digit signed number
; from the keyboard, and store
; the result in cx register:
call scan_num
; store second number:
mov num2, cx
lea dx, msg4
mov ah, 09h ; output string at ds:dx
int 21h
; calculate:
cmp opr, '+'
je do_plus
cmp opr, '-'
je do_minus
cmp opr, '*'
je do_mult
cmp opr, '/'
je do_div
; none of the above....
wrong_opr:
lea dx, err1
mov ah, 09h ; output string at ds:dx
int 21h
exit:
; output of a string at ds:dx
lea dx, msg5
mov ah, 09h
int 21h
; wait for any key...
mov ah, 0
int 16h
ret ; return back to os.
do_plus:
mov ax, num1
add ax, num2
call print_num ; print ax value.
jmp exit
do_minus:
mov ax, num1
sub ax, num2
call print_num ; print ax value.
jmp exit
do_mult:
mov ax, num1
imul num2 ; (dx ax) = ax * num2.
call print_num ; print ax value.
; dx is ignored (calc works with tiny numbers only).
jmp exit
do_div:
; dx is ignored (calc works with tiny integer numbers only).
mov dx, 0
mov ax, num1
idiv num2 ; ax = (dx ax) / num2.
cmp dx, 0
jnz approx
call print_num ; print ax value.
jmp exit
approx:
call print_num ; print ax value.
lea dx, smth
mov ah, 09h ; output string at ds:dx
int 21h
jmp exit
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;; these functions are copied from emu8086.inc ;;;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; gets the multi-digit SIGNED number from the keyboard,
; and stores the result in CX register:
SCAN_NUM PROC NEAR
PUSH DX
PUSH AX
PUSH SI
MOV CX, 0
; reset flag:
MOV CS:make_minus, 0
next_digit:
; get char from keyboard
; into AL:
MOV AH, 00h
INT 16h
; and print it:
MOV AH, 0Eh
INT 10h
; check for MINUS:
CMP AL, '-'
JE set_minus
; check for ENTER key:
CMP AL, 0Dh ; carriage return?
JNE not_cr
JMP stop_input
not_cr:
CMP AL, 8 ; 'BACKSPACE' pressed?
JNE backspace_checked
MOV DX, 0 ; remove last digit by
MOV AX, CX ; division:
DIV CS:ten ; AX = DX:AX / 10 (DX-rem).
MOV CX, AX
PUTC ' ' ; clear position.
PUTC 8 ; backspace again.
JMP next_digit
backspace_checked:
; allow only digits:
CMP AL, '0'
JAE ok_AE_0
JMP remove_not_digit
ok_AE_0:
CMP AL, '9'
JBE ok_digit
remove_not_digit:
PUTC 8 ; backspace.
PUTC ' ' ; clear last entered not digit.
PUTC 8 ; backspace again.
JMP next_digit ; wait for next input.
ok_digit:
; multiply CX by 10 (first time the result is zero)
PUSH AX
MOV AX, CX
MUL CS:ten ; DX:AX = AX*10
MOV CX, AX
POP AX
; check if the number is too big
; (result should be 16 bits)
CMP DX, 0
JNE too_big
; convert from ASCII code:
SUB AL, 30h
; add AL to CX:
MOV AH, 0
MOV DX, CX ; backup, in case the result will be too big.
ADD CX, AX
JC too_big2 ; jump if the number is too big.
JMP next_digit
Continua