Editing Software bug (section)

=== Arithmetic ===

Numerical operations can result in unexpected output, slow processing, or crashing.<ref>{{cite conference
|author1=Anthony Di Franco
|author2=Hui Guo
|author3=Cindy Rubio-González
|title=A comprehensive study of real-world numerical bug characteristics
|conference=2017 32nd [[IEEE]]/[[Association for Computing Machinery |ACM]] International Conference on Automated Software Engineering (ASE)
|date=November 23, 2017
|publisher=[[IEEE]]
|doi=10.1109/ASE.2017.8115662}}</ref>
Such a bug can be from a lack of awareness of the qualities of the data storage such as a [[arithmetic precision|loss of precision]] due to [[rounding]], [[numerically unstable]] algorithms, [[arithmetic overflow]] and [[underflow]], or from lack of awareness of how calculations are handled by different software coding languages such as [[Division by zero#Computer arithmetic|division by zero]] which in some languages may throw an exception, and in others may return a special value such as [[NaN]] or [[Infinity#Computing|infinity]].