array – Create a code with machine instructions

Question:

I have the following problem and I don't know where and how to start. The worst of all is that it is difficult for me to program with machine instructions:

There is an arrangement A = [a1; a2; …; an] of n positive integers with n an even integer.

I need to implement a code in IASSIM (von neumann architecture simulator) that from an array A, computes an array B defined as B = {| log2 a1 |, | log2 a2 |, …, | log2 an |} where | | is the floor function that rounds to the nearest smallest integer. It is recommended that I implement the | log2 x | and then see a way to loop through an array.

Attached a link with the simulator page and with the machine instructions that are used http://www.cs.colby.edu/djskrien/IASSim/ (the simulator is made in java that's why I put the label)

; calcular logaritmo base 2 de un número n el cual entregará la parte entera 
;del resultado

loop:   S(x)->Ac+ n ;carga n al acumulador
        Cc->S(x) log ;si el acumulador es >= 0 salta a log
        halt            ; en otro caso termina
        .empty

log:   S(x)->R resm ;copia el número 2 al AR
       S(x)*R->A dos;se multiplica 2*2 
       At->S(x) resm ; se guarda en resm
       S(x)->Ah+ uno ;se aumenta el contador
       At->S(x) cont ;se guarda el contador
       S(x)->Ac+ n ;se carga n al AC
       S(x)->Ah-  uno;se decrementa n en 1
       At->S(x)   n ;se guarda el n 
       Cu->S(x)   loop; vuelve al principio y hace el proceso de nuevo

      n:    .data 4 ;número a calcular logaritmo
      dos:  .data 2 ;base del logaritmo
      uno:  .data 1 ;para ir aumentando el contador
      resm: .data 2 ;aqui se guardara el resultado de la multiplicación
      cont: .data 0 ;se guarda el resultado del logaritmo

This code is tested in the simulator but it does nothing because it is not right.

Answer:

To implement the | log2 x | I do not think it is difficult, the idea is to simply take the number x can be in int (32 bits) format, then extract from the most significant bit and check if it is 1 . If it is not, it continues to the next bit, and so on until it finds a one or ends with all the bits. The result of | log2 x | It will be the number of bits left from the first bit in one.

Some time ago I used assembler for micro-controllers, in those terms. I could say that it would be something like this in pseudo-code:

 1. B = 32
 2. Si B > 0 avanza al paso 3, en caso contrario sigue al paso 7
 3. x << 1  ;corrimiento de bits a la izquierda
 4. acarreo == 0 ? avanza al paso 5  en caso contrario avanza al paso 7
 5. B-- ; resta uno a B
 6. vuelve al paso 2
 7. |Log2 x|= B
 8. Fin
Scroll to Top