How is the % processed?

[Bones3]

After code with the modulus operator (%) is compiled how does the CPU process it? I mean does it have some sort of efficient way of makeing x % y take up less processing power than if we just put it in a loop that calcualtes it? Same thing with powers. If we could (I don't believe we can) do this: 4^50 would it be processed the same as

for(i = 0; i<50; i++)
{
4 * 4;
}
?

-Bones

 

[apatterno]

There is no x86 instruction for modulus. However, the time needed for this is insignificant. A modulus is probably implemented as three operations: integer divide, integer multiply, and subtract. Think of it this way:

int operator% (int a, int b)
{ return (a/b)*b-a; )

There is no x86 instruction for general exponents, i.e. calculating a^b. However, the most common exponential operation in computers is to multiply a number by 2^x. In C/C++ code we use the shift operators << and >> to do this. For general exponentiation we use pow(a,b).

_{I REALLY hope that helps.}
Will

 

[dlldelhi]

Might have been with ALU(Arithematical Logic Unit) in any x86 microprocessor. You can look for more details in Liu Gibson book or in some book of microprocessor architecture

 

[Clairvoyant1332]

The x86 DIV instruction returns both a quotient and a remainder.

Dennis

 

[Bones3]

Thanks everyone, it was just one of those things I wonder about. Is there a place where you can find some x86 instruction documentation online?

-Bones

 

[JOLESEN]

I think you can download it from Intel.

/JOlesen