| Previous | Next |
| STATUS_FLOAT_DIVIDE_BY_ZERO | STATUS_FLOAT_INVALID_OPERATION |
STATUS_FLOAT_INEXACT_RESULT
The exact mathematical result was not representable
The inexact condition is raised whenever a result must be rounded to fit the destination format. It is extremely common in ordinary floating-point work, which is why runtime libraries normally mask it and retain only the status flag. Unmasking it can make routine arithmetic, conversions, and transcendental functions generate frequent exceptions.
This status does not by itself mean the numerical result is unusable. Determine whether the application intentionally requires trap-based numerical auditing or whether a library accidentally changed the control state. Record the rounding mode and destination precision, because both determine the stored value. Clearing the flag without restoring the intended mask can cause an immediate trap on the next inexact operation.
What to inspect
- Confirm whether inexact exceptions were intentionally unmasked.
- Capture the exact operation, destination precision, and rounding mode.
- Check thread-local floating-point state across third-party library calls.
- Evaluate the numerical error tolerance instead of treating every rounded result as corruption.
References
- Microsoft Learn: _fpieee_flt
- Microsoft Learn: _control87 and _controlfp
- Microsoft Learn: /fp floating-point behavior
- Intel: Floating-Point Reference Sheet
Looking for a different code? Search another status or error code.