This article was previously published under Q326219
This article has been archived. It is offered "as is" and will no longer be updated.
When you change the floating-point control register in managed applications, you may receive exception error messages such as the following in the Microsoft .NET Framework common language runtime:
An unhandled exception of type 'System.ArithmeticException' occurred in system.windows.forms.dll Additional information: Overflow or underflow in the arithmetic operation.
The common language runtime and the .NET Framework assume that the floating-point control register is set to the default state.
If a managed application calls into unmanaged code that must modify the floating-point control register, the unmanaged code must return thefloating-point register to its default state before it returns program control to the managed application.
For example, if you use an Intel system, you can use the following statement before you return program control to the managed application to reset the floating-point control register:
_controlfp(_CW_DEFAULT, 0xfffff); //See the "Steps to Reproduce the Behavior" section of this article.
This behavior is by design.
Steps to Reproduce the Behavior
Start Microsoft Visual Studio .NET.
On the File menu, click New, and then click Project.
In the New Project dialog box, click either Visual Basic Projects or Visual C# Projects under ProjectTypes, and then click Windows Application under Templates. By default, Form1 is created.
Open the .cs or .vb file to see the code for the Form1.Main function.