Error Reporting

The Intel IPP functions return the status of the performed operation to report errors and warnings to the calling program. Thus, it is up to the application to perform error-related actions and/or recover from the error. The last value of the error status is not stored, and the user is to decide whether to check it or not as the function returns. The status values are of IppStatus type and are global constant integers.

Below you can see a list of status codes and corresponding messages reported by the Intel IPP for small matrices.

ippStsSizeMatchMatrixErr	Unsuitable sizes of the source matrices.
ippStsCountMatrixErr	Count parameter is negative or equal to 0.
ippStsRoiShiftMatrixErr	RoiShift is negative or not divisible by the size of the data type.
ippStsStrideMatrixErr	Stride value is not positive or not divisible by the size of the data type.
ippStsSingularErr	Matrix is singular.
ippStsNotPosDefErr	Not positive-definite matrix.
ippStsSizeErr	Wrong value of the data size.
ippStsNoErr	No error, it's OK.
ippStsDivByZeroErr	An attempt to divide by zero.
ippStsNullPtrErr	Null pointer error.
ippStsConvergeErr	Indicates an error if the algorithm does not converge.
ippStsSizeMatchMatrixErr	Unsuitable sizes of the source matrices.

The status codes ending with Err (except for the ippStsNoErr status) indicate an error; the integer values of these codes are negative. When an error occurs, the function execution is interrupted.

For example, if the source matrix for ippmLUDecomp is singular, the function stops execution and returns with the error status ippStsSingularErr. If the input stride value is 0 or 3, the function stops execution and returns with the error status ippStsStrideMatrixErr.