static void kf_bfly3(
kiss_fft_cpx * Fout,
const size_t fstride,
const kiss_fft_state *st,
int m,
int N,
int mm
)
{
int i;
size_t k;
const size_t m2 = 2*m;
const kiss_twiddle_cpx *tw1,*tw2;
kiss_fft_cpx scratch[5];
kiss_twiddle_cpx epi3;
kiss_fft_cpx * Fout_beg = Fout;
epi3 = st->twiddles[fstride*m];
for (i=0;i<N;i++)
{
Fout = Fout_beg + i*mm;
tw1=tw2=st->twiddles;
k=m;
do {
C_FIXDIV(*Fout,3); C_FIXDIV(Fout[m],3); C_FIXDIV(Fout[m2],3);
C_MUL(scratch[1],Fout[m] , *tw1);
C_MUL(scratch[2],Fout[m2] , *tw2);
C_ADD(scratch[3],scratch[1],scratch[2]);
C_SUB(scratch[0],scratch[1],scratch[2]);
tw1 += fstride;
tw2 += fstride*2;
Fout[m].r = Fout->r - HALF_OF(scratch[3].r);
Fout[m].i = Fout->i - HALF_OF(scratch[3].i);
C_MULBYSCALAR( scratch[0] , epi3.i );
C_ADDTO(*Fout,scratch[3]);
Fout[m2].r = Fout[m].r + scratch[0].i;
Fout[m2].i = Fout[m].i - scratch[0].r;
Fout[m].r -= scratch[0].i;
Fout[m].i += scratch[0].r;
++Fout;
} while(--k);
}
}
static void kf_bfly3(
kiss_fft_cpx * Fout,
const size_t fstride,
const kiss_fft_cfg st,
size_t m
)
{
size_t k=m;
const size_t m2 = 2*m;
kiss_fft_cpx *tw1,*tw2;
kiss_fft_cpx scratch[5];
kiss_fft_cpx epi3;
epi3 = st->twiddles[fstride*m];
tw1=tw2=st->twiddles;
do {
if (!st->inverse) {
C_FIXDIV(*Fout,3);
C_FIXDIV(Fout[m],3);
C_FIXDIV(Fout[m2],3);
}
C_MUL(scratch[1],Fout[m] , *tw1);
C_MUL(scratch[2],Fout[m2] , *tw2);
C_ADD(scratch[3],scratch[1],scratch[2]);
C_SUB(scratch[0],scratch[1],scratch[2]);
tw1 += fstride;
tw2 += fstride*2;
Fout[m].r = Fout->r - HALF_OF(scratch[3].r);
Fout[m].i = Fout->i - HALF_OF(scratch[3].i);
C_MULBYSCALAR( scratch[0] , epi3.i );
C_ADDTO(*Fout,scratch[3]);
Fout[m2].r = Fout[m].r + scratch[0].i;
Fout[m2].i = Fout[m].i - scratch[0].r;
Fout[m].r -= scratch[0].i;
Fout[m].i += scratch[0].r;
++Fout;
} while(--k);
}