Integer division by zero

Dividing integer number by zero

Description

This defect occurs when the denominator of a division or modulo operation can be a zero-valued integer.

Risk

A division by zero can result in a program crash.

Fix

The fix depends on the root cause of the defect. Often the result details show a sequence of events that led to the defect. Use this event list to determine how the denominator variable acquires a zero value. You can implement the fix on any event in the sequence. If the result details do not show the event history, you can trace back using right-click options in the source code and see previous related events. See also Interpret Bug Finder Results in Polyspace Desktop User Interface.

It is a good practice to check for zero values of a denominator before division and handle the error. Instead of performing the division directly:

res = num/den;

use a library function that handles zero values of the denominator before performing the division:

res = div(num, den);

See examples of fixes below.

If you do not want to fix the issue, add comments to your result or code to avoid another review. See Address Polyspace Results Through Bug Fixes or Justifications.

Examples

Dividing an Integer by Zero

int fraction(int num)
{
    int denom = 0;
    int result = 0;

    result = num/denom;

    return result;
}

A division by zero error occurs at num/denom because denom is zero.

Correction — Check Before Division

int fraction(int num)
{
    int denom = 0;
    int result = 0;

    if (denom != 0)
        result = num/denom;

    return result;
}

Before dividing, add a test to see if the denominator is zero, checking before division occurs. If denom is always zero, this correction can produce a dead code defect in your Polyspace^® results.

Correction — Change Denominator

One possible correction is to change the denominator value so that denom is not zero.

int fraction(int num)
{
    int denom = 2;
    int result = 0;

    result = num/denom;

    return result;
}

Modulo Operation with Zero

int mod_arr(int input)
{
    int arr[5];
    for(int i = 0; i < 5; i++)
    {
        arr[i] = input % i;
    }

    return arr[0]+arr[1]+arr[2]+arr[3]+arr[4];
}

In this example, Polyspace flags the modulo operation as a division by zero. Because modulo is inherently a division operation, the divisor (right hand argument) cannot be zero. The modulo operation uses the for loop index as the divisor. However, the for loop starts at zero, which cannot be an iterator.

Correction — Check Divisor Before Operation

One possible correction is checking the divisor before the modulo operation. In this example, see if the index i is zero before the modulo operation.

int mod_arr(int input)
{
    int arr[5];
    for(int i = 0; i < 5; i++)
    {
        if(i != 0)
        {
             arr[i] = input % i;
        }
        else
        {
             arr[i] = input;
        }
    }

    return arr[0]+arr[1]+arr[2]+arr[3]+arr[4];
}

Correction — Change Divisor

Another possible correction is changing the divisor to a nonzero integer. In this example, add one to the index before the % operation to avoid dividing by zero.

int mod_arr(int input)
{
    int arr[5];
    for(int i = 0; i < 5; i++)
    {
         arr[i] = input % (i+1);
    }

    return arr[0]+arr[1]+arr[2]+arr[3]+arr[4];
}

Result Information

Group: Numerical

Language: C | C++

Default: On

Command-Line Syntax: INT_ZERO_DIV

Impact: High

CWE ID: 189, 369