The descriptor field further describes the data associated with the operation. It may contain implied parameters and/or indicate additional required parameters. To minimize the number of code branches in the function and thus reduce potentially unnecessary execution overhead, most of the general functions were split into separate primitive functions, with some of their parameters entering the primitive function name as descriptors.
However, some functions may still have parameters that determine internal operations (for example, ippsThreshold).
The following descriptors are used in signal processing functions:
I |
Operation is in-place (default is not-in-place). |
Sfs |
Saturation and fixed scaling mode (default is saturation and no scaling). |
Dx |
Signal is x-dimensional (default is D1). |
L |
One pointer is used for each row (in D2). |
P |
Operation is performed for the specified number of vectors. |
The abbreviations of descriptors in function names are always present in the alphabetical order.
Not all functions have every abbreviation listed above. For example, in-place mode makes no sense for a copy operation.
Copyright © 2000 - 2010, Intel Corporation. All rights reserved.