![]() |
![]() |
FFTLIB provides a collection of C-callable high-performance routines consisting of 1d, 2d, & 3d double precision & single Precision Fast Fourier Transforms. These functions exploit the high performance and multiprocessing capabilities of Texas Instruments C66x DSPs.
All functions in FFTLIB have been developed for a little-endian memory model. Some may work in a big-endian memory model as well, however, their functionality is not guaranteed.
All functions within FFTLIB following the naming conventions described below.
fft_
or ifft_
. This serves to prevent namespace conflicts.omp_
for multiprocessor routines. This descriptor highlights routines designed to run on multiple cores utilizing OpenMP.sp_
or double precision dp_
is designated, followed by the dimensions supported 1d_
, 2d_
, or 3d_
by the fft.r
or complex c
.r2c
c2c
c2r
The following sections provide a detailed reference of all functions within FFTLIB. The functions are organized into the following three categories: