Content deleted Content added
→Integer-overflow prevention: Attempt to rewrite section which in its totality didn't make sense before. Also made clear that the language is C++; other languages might limit the size of an "unsigned int" only to available system memory and/or throw an exception on overflow, in which case this problem doesn't exist in that language. Comparison to MAX also changed from < to <=, as the name MAX (="maximum") implies that the value MAX is a valid value. |
→Integer-overflow prevention: Missed a correction. |
||
Line 58:
The problem with the code is it does not check for integer overflow on the addition operation. If the sum of x and y is greater than the maximum possible value of an <code>unsigned int</code>, the addition operation will overflow and perhaps<!-- Note that an overflow will not always result in the calculated sum being less than MAX; MAX might be relatively small and both x and y relatively big, so even an overflow might still be greater than MAX. Example: x=y=UINT_MAX, MAX=1000000. --> result in a value less than or equal to MAX, even though the sum of x and y is greater than MAX.
Below is a
<syntaxhighlight lang="c++">
bool sumIsValid_secure(unsigned int x, unsigned int y) {
| |||