Does modulus use floating point behind the scenes?

Question

In C++ does a modulus use any floating point math behind the scenes?

int x = 1234;
int y = 5678;
int z = y % x;  // any floating point used underneath to calculate the integer result?

As background, I was thinking about this question where he said he couldn't use any floating point without FP emulation. Then I realized that I wasn't sure if the modulus operator used any sort of floating point assembly operations. My guess is it does not, but I would like to be sure.

The result is simply the remainder after integer division... — Oliver Charlesworth, Feb 06 '14 at 21:57
_'Does modulus use floating point behind the scenes?'_ Simply: NO! — πάντα ῥεῖ, Feb 06 '14 at 22:00

score 3 · Accepted Answer · edited May 23 '17 at 10:31

No¹. Refer to some implementations of such an operator.

Assembly Language - How to Do Modulo? - "[In x86] the DIV instruction [..] gives both the quotient and remainder"
Assembly mod algorithm on processor with no division operator

¹ An implementation can do whatever it wants insofar as the observed behavior is within the specification. However, I don't know of any implementation which would choose to use floating point operations, nor can I think of a general justification for doing so.

score 2 · Answer 2 · answered Feb 06 '14 at 21:56

No it does not use the floating point arithmetic. The result can be obtained very simply

z = y - ( y/x ) * x;

Early computers sometimes have no floating point coprocessor. So such operations are performed by using machine commands that operate with integer numbers.

score 1 · Answer 3 · answered Feb 06 '14 at 21:58

1

It's up to the implementation how "C++" calculates the modulus "behind the scenes"

answered Feb 06 '14 at 21:58

twin

1,669
13
11

3

True, but misleading. Do you know of even one example architecture where the answer is 'yes'? – Yakk - Adam Nevraumont Feb 06 '14 at 22:07

Does modulus use floating point behind the scenes?

3 Answers3