Hello I'm trying to simulate an FFT and an iFFT (with code composer
studio)using functions in DSPlib, in particular i'm using
DSPF_sp_fftSPxSP function. I have made to work with success the
following code:
#define CHIP_6711C
#include <stdio.h>
#include <stdlib.h>
#include <csl.h>
#include <csl_timer.h>
#include <math.h>
#include "twiddleSPxSP4096.h"
#include "brev_table.h"
#define PI (3.141592654)
#define NN (4096)
float x[2 * NN];
float y[2 * NN];
void main(void)
{
int i;
int k;
int j;
short F1 = 10, F2 = 40;
/*
====================================================================
*/
/* Generate input
data */
/*
====================================================================
*/
for(i=0; i<NN; i++)
{
/* real part */
x[2 * i] = (sin(2 * PI * F1 * i / NN) + sin(2 * PI * F2 * i / NN));
/* img part */
x[2 * i + 1] = 0;
}
DSPF_sp_fftSPxSP (NN, x ,w , y , brev ,4 ,0 ,NN);
DSPF_sp_ifftSPxSP (NN, y ,w , x , brev ,4,0 ,NN);
}
In "TMS320C67x DSP Library Programmer's Reference
Guide"(spra657b)I
have read that "the c code of the functions is the equivalent of the
assembly code without restrictions" and that "the assembly code is
hand optimized and restrictions may apply".
What does it means?
If I call the functions as you can see in the program above,am I
using the hand optimized assembler routines or not?
In other words, simply calling the Dsplib functions with the c
language am I using the optimized code?
Questions about the use of the DSPlib FFT functions on TMS320C6711
Started by ●January 26, 2006
Reply by ●January 26, 20062006-01-26
Short answer is - yes. You can call ASM from C and get all the so-called
hand-optimizations.
Restrcitions are probably related to array alignment in memory. The hand
coded ASM assumes particluar alignment so the LDDW instructions can be used.
- Andrew E. dukke3d wrote:
>Hello I'm trying to simulate an FFT and an iFFT (with code composer
>studio)using functions in DSPlib, in particular i'm using
>DSPF_sp_fftSPxSP function. I have made to work with success the
>following code:
>
>#define CHIP_6711C
>#include <stdio.h>
>#include <stdlib.h>
>#include <csl.h>
>#include <csl_timer.h>
>#include <math.h>
>
>#include "twiddleSPxSP4096.h"
>#include "brev_table.h"
>
>#define PI (3.141592654)
>#define NN (4096)
>
>float x[2 * NN];
>float y[2 * NN];
>
>void main(void)
>{
> int i;
> int k;
> int j;
> short F1 = 10, F2 = 40;
> /*
>====================================================================
>*/
> /* Generate input
>data */
> /*
>====================================================================
>*/
> for(i=0; i<NN; i++)
> {
> /* real part */
> x[2 * i] = (sin(2 * PI * F1 * i / NN) + sin(2 * PI * F2 * i / NN));
> /* img part */
> x[2 * i + 1] = 0;
> }
>
>DSPF_sp_fftSPxSP (NN, x ,w , y , brev ,4 ,0 ,NN);
>
>DSPF_sp_ifftSPxSP (NN, y ,w , x , brev ,4,0 ,NN);
>}
>
>In "TMS320C67x DSP Library Programmer's Reference Guide"(spra657b)I
>have read that "the c code of the functions is the equivalent of the
>assembly code without restrictions" and that "the assembly code is
>hand optimized and restrictions may apply".
>What does it means?
>If I call the functions as you can see in the program above,am I
>using the hand optimized assembler routines or not?
>In other words, simply calling the Dsplib functions with the c
>language am I using the optimized code?
