michael@0: /*Copyright (c) 2003-2004, Mark Borgerding
michael@0:   Lots of modifications by Jean-Marc Valin
michael@0:   Copyright (c) 2005-2007, Xiph.Org Foundation
michael@0:   Copyright (c) 2008,      Xiph.Org Foundation, CSIRO
michael@0: 
michael@0:   All rights reserved.
michael@0: 
michael@0:   Redistribution and use in source and binary forms, with or without
michael@0:    modification, are permitted provided that the following conditions are met:
michael@0: 
michael@0:     * Redistributions of source code must retain the above copyright notice,
michael@0:        this list of conditions and the following disclaimer.
michael@0:     * Redistributions in binary form must reproduce the above copyright notice,
michael@0:        this list of conditions and the following disclaimer in the
michael@0:        documentation and/or other materials provided with the distribution.
michael@0: 
michael@0:   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
michael@0:   AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
michael@0:   IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
michael@0:   ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
michael@0:   LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
michael@0:   CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
michael@0:   SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
michael@0:   INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
michael@0:   CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
michael@0:   ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
michael@0:   POSSIBILITY OF SUCH DAMAGE.*/
michael@0: 
michael@0: /* This code is originally from Mark Borgerding's KISS-FFT but has been
michael@0:    heavily modified to better suit Opus */
michael@0: 
michael@0: #ifndef SKIP_CONFIG_H
michael@0: #  ifdef HAVE_CONFIG_H
michael@0: #    include "config.h"
michael@0: #  endif
michael@0: #endif
michael@0: 
michael@0: #include "_kiss_fft_guts.h"
michael@0: #include "arch.h"
michael@0: #include "os_support.h"
michael@0: #include "mathops.h"
michael@0: #include "stack_alloc.h"
michael@0: 
michael@0: /* The guts header contains all the multiplication and addition macros that are defined for
michael@0:    complex numbers.  It also delares the kf_ internal functions.
michael@0: */
michael@0: 
michael@0: static void kf_bfly2(
michael@0:                      kiss_fft_cpx * Fout,
michael@0:                      const size_t fstride,
michael@0:                      const kiss_fft_state *st,
michael@0:                      int m,
michael@0:                      int N,
michael@0:                      int mm
michael@0:                     )
michael@0: {
michael@0:    kiss_fft_cpx * Fout2;
michael@0:    const kiss_twiddle_cpx * tw1;
michael@0:    int i,j;
michael@0:    kiss_fft_cpx * Fout_beg = Fout;
michael@0:    for (i=0;i<N;i++)
michael@0:    {
michael@0:       Fout = Fout_beg + i*mm;
michael@0:       Fout2 = Fout + m;
michael@0:       tw1 = st->twiddles;
michael@0:       for(j=0;j<m;j++)
michael@0:       {
michael@0:          kiss_fft_cpx t;
michael@0:          Fout->r = SHR32(Fout->r, 1);Fout->i = SHR32(Fout->i, 1);
michael@0:          Fout2->r = SHR32(Fout2->r, 1);Fout2->i = SHR32(Fout2->i, 1);
michael@0:          C_MUL (t,  *Fout2 , *tw1);
michael@0:          tw1 += fstride;
michael@0:          C_SUB( *Fout2 ,  *Fout , t );
michael@0:          C_ADDTO( *Fout ,  t );
michael@0:          ++Fout2;
michael@0:          ++Fout;
michael@0:       }
michael@0:    }
michael@0: }
michael@0: 
michael@0: static void ki_bfly2(
michael@0:                      kiss_fft_cpx * Fout,
michael@0:                      const size_t fstride,
michael@0:                      const kiss_fft_state *st,
michael@0:                      int m,
michael@0:                      int N,
michael@0:                      int mm
michael@0:                     )
michael@0: {
michael@0:    kiss_fft_cpx * Fout2;
michael@0:    const kiss_twiddle_cpx * tw1;
michael@0:    kiss_fft_cpx t;
michael@0:    int i,j;
michael@0:    kiss_fft_cpx * Fout_beg = Fout;
michael@0:    for (i=0;i<N;i++)
michael@0:    {
michael@0:       Fout = Fout_beg + i*mm;
michael@0:       Fout2 = Fout + m;
michael@0:       tw1 = st->twiddles;
michael@0:       for(j=0;j<m;j++)
michael@0:       {
michael@0:          C_MULC (t,  *Fout2 , *tw1);
michael@0:          tw1 += fstride;
michael@0:          C_SUB( *Fout2 ,  *Fout , t );
michael@0:          C_ADDTO( *Fout ,  t );
michael@0:          ++Fout2;
michael@0:          ++Fout;
michael@0:       }
michael@0:    }
michael@0: }
michael@0: 
michael@0: static void kf_bfly4(
michael@0:                      kiss_fft_cpx * Fout,
michael@0:                      const size_t fstride,
michael@0:                      const kiss_fft_state *st,
michael@0:                      int m,
michael@0:                      int N,
michael@0:                      int mm
michael@0:                     )
michael@0: {
michael@0:    const kiss_twiddle_cpx *tw1,*tw2,*tw3;
michael@0:    kiss_fft_cpx scratch[6];
michael@0:    const size_t m2=2*m;
michael@0:    const size_t m3=3*m;
michael@0:    int i, j;
michael@0: 
michael@0:    kiss_fft_cpx * Fout_beg = Fout;
michael@0:    for (i=0;i<N;i++)
michael@0:    {
michael@0:       Fout = Fout_beg + i*mm;
michael@0:       tw3 = tw2 = tw1 = st->twiddles;
michael@0:       for (j=0;j<m;j++)
michael@0:       {
michael@0:          C_MUL4(scratch[0],Fout[m] , *tw1 );
michael@0:          C_MUL4(scratch[1],Fout[m2] , *tw2 );
michael@0:          C_MUL4(scratch[2],Fout[m3] , *tw3 );
michael@0: 
michael@0:          Fout->r = PSHR32(Fout->r, 2);
michael@0:          Fout->i = PSHR32(Fout->i, 2);
michael@0:          C_SUB( scratch[5] , *Fout, scratch[1] );
michael@0:          C_ADDTO(*Fout, scratch[1]);
michael@0:          C_ADD( scratch[3] , scratch[0] , scratch[2] );
michael@0:          C_SUB( scratch[4] , scratch[0] , scratch[2] );
michael@0:          C_SUB( Fout[m2], *Fout, scratch[3] );
michael@0:          tw1 += fstride;
michael@0:          tw2 += fstride*2;
michael@0:          tw3 += fstride*3;
michael@0:          C_ADDTO( *Fout , scratch[3] );
michael@0: 
michael@0:          Fout[m].r = scratch[5].r + scratch[4].i;
michael@0:          Fout[m].i = scratch[5].i - scratch[4].r;
michael@0:          Fout[m3].r = scratch[5].r - scratch[4].i;
michael@0:          Fout[m3].i = scratch[5].i + scratch[4].r;
michael@0:          ++Fout;
michael@0:       }
michael@0:    }
michael@0: }
michael@0: 
michael@0: static void ki_bfly4(
michael@0:                      kiss_fft_cpx * Fout,
michael@0:                      const size_t fstride,
michael@0:                      const kiss_fft_state *st,
michael@0:                      int m,
michael@0:                      int N,
michael@0:                      int mm
michael@0:                     )
michael@0: {
michael@0:    const kiss_twiddle_cpx *tw1,*tw2,*tw3;
michael@0:    kiss_fft_cpx scratch[6];
michael@0:    const size_t m2=2*m;
michael@0:    const size_t m3=3*m;
michael@0:    int i, j;
michael@0: 
michael@0:    kiss_fft_cpx * Fout_beg = Fout;
michael@0:    for (i=0;i<N;i++)
michael@0:    {
michael@0:       Fout = Fout_beg + i*mm;
michael@0:       tw3 = tw2 = tw1 = st->twiddles;
michael@0:       for (j=0;j<m;j++)
michael@0:       {
michael@0:          C_MULC(scratch[0],Fout[m] , *tw1 );
michael@0:          C_MULC(scratch[1],Fout[m2] , *tw2 );
michael@0:          C_MULC(scratch[2],Fout[m3] , *tw3 );
michael@0: 
michael@0:          C_SUB( scratch[5] , *Fout, scratch[1] );
michael@0:          C_ADDTO(*Fout, scratch[1]);
michael@0:          C_ADD( scratch[3] , scratch[0] , scratch[2] );
michael@0:          C_SUB( scratch[4] , scratch[0] , scratch[2] );
michael@0:          C_SUB( Fout[m2], *Fout, scratch[3] );
michael@0:          tw1 += fstride;
michael@0:          tw2 += fstride*2;
michael@0:          tw3 += fstride*3;
michael@0:          C_ADDTO( *Fout , scratch[3] );
michael@0: 
michael@0:          Fout[m].r = scratch[5].r - scratch[4].i;
michael@0:          Fout[m].i = scratch[5].i + scratch[4].r;
michael@0:          Fout[m3].r = scratch[5].r + scratch[4].i;
michael@0:          Fout[m3].i = scratch[5].i - scratch[4].r;
michael@0:          ++Fout;
michael@0:       }
michael@0:    }
michael@0: }
michael@0: 
michael@0: #ifndef RADIX_TWO_ONLY
michael@0: 
michael@0: static void kf_bfly3(
michael@0:                      kiss_fft_cpx * Fout,
michael@0:                      const size_t fstride,
michael@0:                      const kiss_fft_state *st,
michael@0:                      int m,
michael@0:                      int N,
michael@0:                      int mm
michael@0:                     )
michael@0: {
michael@0:    int i;
michael@0:    size_t k;
michael@0:    const size_t m2 = 2*m;
michael@0:    const kiss_twiddle_cpx *tw1,*tw2;
michael@0:    kiss_fft_cpx scratch[5];
michael@0:    kiss_twiddle_cpx epi3;
michael@0: 
michael@0:    kiss_fft_cpx * Fout_beg = Fout;
michael@0:    epi3 = st->twiddles[fstride*m];
michael@0:    for (i=0;i<N;i++)
michael@0:    {
michael@0:       Fout = Fout_beg + i*mm;
michael@0:       tw1=tw2=st->twiddles;
michael@0:       k=m;
michael@0:       do {
michael@0:          C_FIXDIV(*Fout,3); C_FIXDIV(Fout[m],3); C_FIXDIV(Fout[m2],3);
michael@0: 
michael@0:          C_MUL(scratch[1],Fout[m] , *tw1);
michael@0:          C_MUL(scratch[2],Fout[m2] , *tw2);
michael@0: 
michael@0:          C_ADD(scratch[3],scratch[1],scratch[2]);
michael@0:          C_SUB(scratch[0],scratch[1],scratch[2]);
michael@0:          tw1 += fstride;
michael@0:          tw2 += fstride*2;
michael@0: 
michael@0:          Fout[m].r = Fout->r - HALF_OF(scratch[3].r);
michael@0:          Fout[m].i = Fout->i - HALF_OF(scratch[3].i);
michael@0: 
michael@0:          C_MULBYSCALAR( scratch[0] , epi3.i );
michael@0: 
michael@0:          C_ADDTO(*Fout,scratch[3]);
michael@0: 
michael@0:          Fout[m2].r = Fout[m].r + scratch[0].i;
michael@0:          Fout[m2].i = Fout[m].i - scratch[0].r;
michael@0: 
michael@0:          Fout[m].r -= scratch[0].i;
michael@0:          Fout[m].i += scratch[0].r;
michael@0: 
michael@0:          ++Fout;
michael@0:       } while(--k);
michael@0:    }
michael@0: }
michael@0: 
michael@0: static void ki_bfly3(
michael@0:                      kiss_fft_cpx * Fout,
michael@0:                      const size_t fstride,
michael@0:                      const kiss_fft_state *st,
michael@0:                      int m,
michael@0:                      int N,
michael@0:                      int mm
michael@0:                     )
michael@0: {
michael@0:    int i, k;
michael@0:    const size_t m2 = 2*m;
michael@0:    const kiss_twiddle_cpx *tw1,*tw2;
michael@0:    kiss_fft_cpx scratch[5];
michael@0:    kiss_twiddle_cpx epi3;
michael@0: 
michael@0:    kiss_fft_cpx * Fout_beg = Fout;
michael@0:    epi3 = st->twiddles[fstride*m];
michael@0:    for (i=0;i<N;i++)
michael@0:    {
michael@0:       Fout = Fout_beg + i*mm;
michael@0:       tw1=tw2=st->twiddles;
michael@0:       k=m;
michael@0:       do{
michael@0: 
michael@0:          C_MULC(scratch[1],Fout[m] , *tw1);
michael@0:          C_MULC(scratch[2],Fout[m2] , *tw2);
michael@0: 
michael@0:          C_ADD(scratch[3],scratch[1],scratch[2]);
michael@0:          C_SUB(scratch[0],scratch[1],scratch[2]);
michael@0:          tw1 += fstride;
michael@0:          tw2 += fstride*2;
michael@0: 
michael@0:          Fout[m].r = Fout->r - HALF_OF(scratch[3].r);
michael@0:          Fout[m].i = Fout->i - HALF_OF(scratch[3].i);
michael@0: 
michael@0:          C_MULBYSCALAR( scratch[0] , -epi3.i );
michael@0: 
michael@0:          C_ADDTO(*Fout,scratch[3]);
michael@0: 
michael@0:          Fout[m2].r = Fout[m].r + scratch[0].i;
michael@0:          Fout[m2].i = Fout[m].i - scratch[0].r;
michael@0: 
michael@0:          Fout[m].r -= scratch[0].i;
michael@0:          Fout[m].i += scratch[0].r;
michael@0: 
michael@0:          ++Fout;
michael@0:       }while(--k);
michael@0:    }
michael@0: }
michael@0: 
michael@0: static void kf_bfly5(
michael@0:                      kiss_fft_cpx * Fout,
michael@0:                      const size_t fstride,
michael@0:                      const kiss_fft_state *st,
michael@0:                      int m,
michael@0:                      int N,
michael@0:                      int mm
michael@0:                     )
michael@0: {
michael@0:    kiss_fft_cpx *Fout0,*Fout1,*Fout2,*Fout3,*Fout4;
michael@0:    int i, u;
michael@0:    kiss_fft_cpx scratch[13];
michael@0:    const kiss_twiddle_cpx * twiddles = st->twiddles;
michael@0:    const kiss_twiddle_cpx *tw;
michael@0:    kiss_twiddle_cpx ya,yb;
michael@0:    kiss_fft_cpx * Fout_beg = Fout;
michael@0: 
michael@0:    ya = twiddles[fstride*m];
michael@0:    yb = twiddles[fstride*2*m];
michael@0:    tw=st->twiddles;
michael@0: 
michael@0:    for (i=0;i<N;i++)
michael@0:    {
michael@0:       Fout = Fout_beg + i*mm;
michael@0:       Fout0=Fout;
michael@0:       Fout1=Fout0+m;
michael@0:       Fout2=Fout0+2*m;
michael@0:       Fout3=Fout0+3*m;
michael@0:       Fout4=Fout0+4*m;
michael@0: 
michael@0:       for ( u=0; u<m; ++u ) {
michael@0:          C_FIXDIV( *Fout0,5); C_FIXDIV( *Fout1,5); C_FIXDIV( *Fout2,5); C_FIXDIV( *Fout3,5); C_FIXDIV( *Fout4,5);
michael@0:          scratch[0] = *Fout0;
michael@0: 
michael@0:          C_MUL(scratch[1] ,*Fout1, tw[u*fstride]);
michael@0:          C_MUL(scratch[2] ,*Fout2, tw[2*u*fstride]);
michael@0:          C_MUL(scratch[3] ,*Fout3, tw[3*u*fstride]);
michael@0:          C_MUL(scratch[4] ,*Fout4, tw[4*u*fstride]);
michael@0: 
michael@0:          C_ADD( scratch[7],scratch[1],scratch[4]);
michael@0:          C_SUB( scratch[10],scratch[1],scratch[4]);
michael@0:          C_ADD( scratch[8],scratch[2],scratch[3]);
michael@0:          C_SUB( scratch[9],scratch[2],scratch[3]);
michael@0: 
michael@0:          Fout0->r += scratch[7].r + scratch[8].r;
michael@0:          Fout0->i += scratch[7].i + scratch[8].i;
michael@0: 
michael@0:          scratch[5].r = scratch[0].r + S_MUL(scratch[7].r,ya.r) + S_MUL(scratch[8].r,yb.r);
michael@0:          scratch[5].i = scratch[0].i + S_MUL(scratch[7].i,ya.r) + S_MUL(scratch[8].i,yb.r);
michael@0: 
michael@0:          scratch[6].r =  S_MUL(scratch[10].i,ya.i) + S_MUL(scratch[9].i,yb.i);
michael@0:          scratch[6].i = -S_MUL(scratch[10].r,ya.i) - S_MUL(scratch[9].r,yb.i);
michael@0: 
michael@0:          C_SUB(*Fout1,scratch[5],scratch[6]);
michael@0:          C_ADD(*Fout4,scratch[5],scratch[6]);
michael@0: 
michael@0:          scratch[11].r = scratch[0].r + S_MUL(scratch[7].r,yb.r) + S_MUL(scratch[8].r,ya.r);
michael@0:          scratch[11].i = scratch[0].i + S_MUL(scratch[7].i,yb.r) + S_MUL(scratch[8].i,ya.r);
michael@0:          scratch[12].r = - S_MUL(scratch[10].i,yb.i) + S_MUL(scratch[9].i,ya.i);
michael@0:          scratch[12].i = S_MUL(scratch[10].r,yb.i) - S_MUL(scratch[9].r,ya.i);
michael@0: 
michael@0:          C_ADD(*Fout2,scratch[11],scratch[12]);
michael@0:          C_SUB(*Fout3,scratch[11],scratch[12]);
michael@0: 
michael@0:          ++Fout0;++Fout1;++Fout2;++Fout3;++Fout4;
michael@0:       }
michael@0:    }
michael@0: }
michael@0: 
michael@0: static void ki_bfly5(
michael@0:                      kiss_fft_cpx * Fout,
michael@0:                      const size_t fstride,
michael@0:                      const kiss_fft_state *st,
michael@0:                      int m,
michael@0:                      int N,
michael@0:                      int mm
michael@0:                     )
michael@0: {
michael@0:    kiss_fft_cpx *Fout0,*Fout1,*Fout2,*Fout3,*Fout4;
michael@0:    int i, u;
michael@0:    kiss_fft_cpx scratch[13];
michael@0:    const kiss_twiddle_cpx * twiddles = st->twiddles;
michael@0:    const kiss_twiddle_cpx *tw;
michael@0:    kiss_twiddle_cpx ya,yb;
michael@0:    kiss_fft_cpx * Fout_beg = Fout;
michael@0: 
michael@0:    ya = twiddles[fstride*m];
michael@0:    yb = twiddles[fstride*2*m];
michael@0:    tw=st->twiddles;
michael@0: 
michael@0:    for (i=0;i<N;i++)
michael@0:    {
michael@0:       Fout = Fout_beg + i*mm;
michael@0:       Fout0=Fout;
michael@0:       Fout1=Fout0+m;
michael@0:       Fout2=Fout0+2*m;
michael@0:       Fout3=Fout0+3*m;
michael@0:       Fout4=Fout0+4*m;
michael@0: 
michael@0:       for ( u=0; u<m; ++u ) {
michael@0:          scratch[0] = *Fout0;
michael@0: 
michael@0:          C_MULC(scratch[1] ,*Fout1, tw[u*fstride]);
michael@0:          C_MULC(scratch[2] ,*Fout2, tw[2*u*fstride]);
michael@0:          C_MULC(scratch[3] ,*Fout3, tw[3*u*fstride]);
michael@0:          C_MULC(scratch[4] ,*Fout4, tw[4*u*fstride]);
michael@0: 
michael@0:          C_ADD( scratch[7],scratch[1],scratch[4]);
michael@0:          C_SUB( scratch[10],scratch[1],scratch[4]);
michael@0:          C_ADD( scratch[8],scratch[2],scratch[3]);
michael@0:          C_SUB( scratch[9],scratch[2],scratch[3]);
michael@0: 
michael@0:          Fout0->r += scratch[7].r + scratch[8].r;
michael@0:          Fout0->i += scratch[7].i + scratch[8].i;
michael@0: 
michael@0:          scratch[5].r = scratch[0].r + S_MUL(scratch[7].r,ya.r) + S_MUL(scratch[8].r,yb.r);
michael@0:          scratch[5].i = scratch[0].i + S_MUL(scratch[7].i,ya.r) + S_MUL(scratch[8].i,yb.r);
michael@0: 
michael@0:          scratch[6].r = -S_MUL(scratch[10].i,ya.i) - S_MUL(scratch[9].i,yb.i);
michael@0:          scratch[6].i =  S_MUL(scratch[10].r,ya.i) + S_MUL(scratch[9].r,yb.i);
michael@0: 
michael@0:          C_SUB(*Fout1,scratch[5],scratch[6]);
michael@0:          C_ADD(*Fout4,scratch[5],scratch[6]);
michael@0: 
michael@0:          scratch[11].r = scratch[0].r + S_MUL(scratch[7].r,yb.r) + S_MUL(scratch[8].r,ya.r);
michael@0:          scratch[11].i = scratch[0].i + S_MUL(scratch[7].i,yb.r) + S_MUL(scratch[8].i,ya.r);
michael@0:          scratch[12].r =  S_MUL(scratch[10].i,yb.i) - S_MUL(scratch[9].i,ya.i);
michael@0:          scratch[12].i = -S_MUL(scratch[10].r,yb.i) + S_MUL(scratch[9].r,ya.i);
michael@0: 
michael@0:          C_ADD(*Fout2,scratch[11],scratch[12]);
michael@0:          C_SUB(*Fout3,scratch[11],scratch[12]);
michael@0: 
michael@0:          ++Fout0;++Fout1;++Fout2;++Fout3;++Fout4;
michael@0:       }
michael@0:    }
michael@0: }
michael@0: 
michael@0: #endif
michael@0: 
michael@0: 
michael@0: #ifdef CUSTOM_MODES
michael@0: 
michael@0: static
michael@0: void compute_bitrev_table(
michael@0:          int Fout,
michael@0:          opus_int16 *f,
michael@0:          const size_t fstride,
michael@0:          int in_stride,
michael@0:          opus_int16 * factors,
michael@0:          const kiss_fft_state *st
michael@0:             )
michael@0: {
michael@0:    const int p=*factors++; /* the radix  */
michael@0:    const int m=*factors++; /* stage's fft length/p */
michael@0: 
michael@0:     /*printf ("fft %d %d %d %d %d %d\n", p*m, m, p, s2, fstride*in_stride, N);*/
michael@0:    if (m==1)
michael@0:    {
michael@0:       int j;
michael@0:       for (j=0;j<p;j++)
michael@0:       {
michael@0:          *f = Fout+j;
michael@0:          f += fstride*in_stride;
michael@0:       }
michael@0:    } else {
michael@0:       int j;
michael@0:       for (j=0;j<p;j++)
michael@0:       {
michael@0:          compute_bitrev_table( Fout , f, fstride*p, in_stride, factors,st);
michael@0:          f += fstride*in_stride;
michael@0:          Fout += m;
michael@0:       }
michael@0:    }
michael@0: }
michael@0: 
michael@0: /*  facbuf is populated by p1,m1,p2,m2, ...
michael@0:     where
michael@0:     p[i] * m[i] = m[i-1]
michael@0:     m0 = n                  */
michael@0: static
michael@0: int kf_factor(int n,opus_int16 * facbuf)
michael@0: {
michael@0:     int p=4;
michael@0: 
michael@0:     /*factor out powers of 4, powers of 2, then any remaining primes */
michael@0:     do {
michael@0:         while (n % p) {
michael@0:             switch (p) {
michael@0:                 case 4: p = 2; break;
michael@0:                 case 2: p = 3; break;
michael@0:                 default: p += 2; break;
michael@0:             }
michael@0:             if (p>32000 || (opus_int32)p*(opus_int32)p > n)
michael@0:                 p = n;          /* no more factors, skip to end */
michael@0:         }
michael@0:         n /= p;
michael@0: #ifdef RADIX_TWO_ONLY
michael@0:         if (p!=2 && p != 4)
michael@0: #else
michael@0:         if (p>5)
michael@0: #endif
michael@0:         {
michael@0:            return 0;
michael@0:         }
michael@0:         *facbuf++ = p;
michael@0:         *facbuf++ = n;
michael@0:     } while (n > 1);
michael@0:     return 1;
michael@0: }
michael@0: 
michael@0: static void compute_twiddles(kiss_twiddle_cpx *twiddles, int nfft)
michael@0: {
michael@0:    int i;
michael@0: #ifdef FIXED_POINT
michael@0:    for (i=0;i<nfft;++i) {
michael@0:       opus_val32 phase = -i;
michael@0:       kf_cexp2(twiddles+i, DIV32(SHL32(phase,17),nfft));
michael@0:    }
michael@0: #else
michael@0:    for (i=0;i<nfft;++i) {
michael@0:       const double pi=3.14159265358979323846264338327;
michael@0:       double phase = ( -2*pi /nfft ) * i;
michael@0:       kf_cexp(twiddles+i, phase );
michael@0:    }
michael@0: #endif
michael@0: }
michael@0: 
michael@0: /*
michael@0:  *
michael@0:  * Allocates all necessary storage space for the fft and ifft.
michael@0:  * The return value is a contiguous block of memory.  As such,
michael@0:  * It can be freed with free().
michael@0:  * */
michael@0: kiss_fft_state *opus_fft_alloc_twiddles(int nfft,void * mem,size_t * lenmem,  const kiss_fft_state *base)
michael@0: {
michael@0:     kiss_fft_state *st=NULL;
michael@0:     size_t memneeded = sizeof(struct kiss_fft_state); /* twiddle factors*/
michael@0: 
michael@0:     if ( lenmem==NULL ) {
michael@0:         st = ( kiss_fft_state*)KISS_FFT_MALLOC( memneeded );
michael@0:     }else{
michael@0:         if (mem != NULL && *lenmem >= memneeded)
michael@0:             st = (kiss_fft_state*)mem;
michael@0:         *lenmem = memneeded;
michael@0:     }
michael@0:     if (st) {
michael@0:         opus_int16 *bitrev;
michael@0:         kiss_twiddle_cpx *twiddles;
michael@0: 
michael@0:         st->nfft=nfft;
michael@0: #ifndef FIXED_POINT
michael@0:         st->scale = 1.f/nfft;
michael@0: #endif
michael@0:         if (base != NULL)
michael@0:         {
michael@0:            st->twiddles = base->twiddles;
michael@0:            st->shift = 0;
michael@0:            while (nfft<<st->shift != base->nfft && st->shift < 32)
michael@0:               st->shift++;
michael@0:            if (st->shift>=32)
michael@0:               goto fail;
michael@0:         } else {
michael@0:            st->twiddles = twiddles = (kiss_twiddle_cpx*)KISS_FFT_MALLOC(sizeof(kiss_twiddle_cpx)*nfft);
michael@0:            compute_twiddles(twiddles, nfft);
michael@0:            st->shift = -1;
michael@0:         }
michael@0:         if (!kf_factor(nfft,st->factors))
michael@0:         {
michael@0:            goto fail;
michael@0:         }
michael@0: 
michael@0:         /* bitrev */
michael@0:         st->bitrev = bitrev = (opus_int16*)KISS_FFT_MALLOC(sizeof(opus_int16)*nfft);
michael@0:         if (st->bitrev==NULL)
michael@0:             goto fail;
michael@0:         compute_bitrev_table(0, bitrev, 1,1, st->factors,st);
michael@0:     }
michael@0:     return st;
michael@0: fail:
michael@0:     opus_fft_free(st);
michael@0:     return NULL;
michael@0: }
michael@0: 
michael@0: kiss_fft_state *opus_fft_alloc(int nfft,void * mem,size_t * lenmem )
michael@0: {
michael@0:    return opus_fft_alloc_twiddles(nfft, mem, lenmem, NULL);
michael@0: }
michael@0: 
michael@0: void opus_fft_free(const kiss_fft_state *cfg)
michael@0: {
michael@0:    if (cfg)
michael@0:    {
michael@0:       opus_free((opus_int16*)cfg->bitrev);
michael@0:       if (cfg->shift < 0)
michael@0:          opus_free((kiss_twiddle_cpx*)cfg->twiddles);
michael@0:       opus_free((kiss_fft_state*)cfg);
michael@0:    }
michael@0: }
michael@0: 
michael@0: #endif /* CUSTOM_MODES */
michael@0: 
michael@0: void opus_fft(const kiss_fft_state *st,const kiss_fft_cpx *fin,kiss_fft_cpx *fout)
michael@0: {
michael@0:     int m2, m;
michael@0:     int p;
michael@0:     int L;
michael@0:     int fstride[MAXFACTORS];
michael@0:     int i;
michael@0:     int shift;
michael@0: 
michael@0:     /* st->shift can be -1 */
michael@0:     shift = st->shift>0 ? st->shift : 0;
michael@0: 
michael@0:     celt_assert2 (fin != fout, "In-place FFT not supported");
michael@0:     /* Bit-reverse the input */
michael@0:     for (i=0;i<st->nfft;i++)
michael@0:     {
michael@0:        fout[st->bitrev[i]] = fin[i];
michael@0: #ifndef FIXED_POINT
michael@0:        fout[st->bitrev[i]].r *= st->scale;
michael@0:        fout[st->bitrev[i]].i *= st->scale;
michael@0: #endif
michael@0:     }
michael@0: 
michael@0:     fstride[0] = 1;
michael@0:     L=0;
michael@0:     do {
michael@0:        p = st->factors[2*L];
michael@0:        m = st->factors[2*L+1];
michael@0:        fstride[L+1] = fstride[L]*p;
michael@0:        L++;
michael@0:     } while(m!=1);
michael@0:     m = st->factors[2*L-1];
michael@0:     for (i=L-1;i>=0;i--)
michael@0:     {
michael@0:        if (i!=0)
michael@0:           m2 = st->factors[2*i-1];
michael@0:        else
michael@0:           m2 = 1;
michael@0:        switch (st->factors[2*i])
michael@0:        {
michael@0:        case 2:
michael@0:           kf_bfly2(fout,fstride[i]<<shift,st,m, fstride[i], m2);
michael@0:           break;
michael@0:        case 4:
michael@0:           kf_bfly4(fout,fstride[i]<<shift,st,m, fstride[i], m2);
michael@0:           break;
michael@0:  #ifndef RADIX_TWO_ONLY
michael@0:        case 3:
michael@0:           kf_bfly3(fout,fstride[i]<<shift,st,m, fstride[i], m2);
michael@0:           break;
michael@0:        case 5:
michael@0:           kf_bfly5(fout,fstride[i]<<shift,st,m, fstride[i], m2);
michael@0:           break;
michael@0:  #endif
michael@0:        }
michael@0:        m = m2;
michael@0:     }
michael@0: }
michael@0: 
michael@0: void opus_ifft(const kiss_fft_state *st,const kiss_fft_cpx *fin,kiss_fft_cpx *fout)
michael@0: {
michael@0:    int m2, m;
michael@0:    int p;
michael@0:    int L;
michael@0:    int fstride[MAXFACTORS];
michael@0:    int i;
michael@0:    int shift;
michael@0: 
michael@0:    /* st->shift can be -1 */
michael@0:    shift = st->shift>0 ? st->shift : 0;
michael@0:    celt_assert2 (fin != fout, "In-place FFT not supported");
michael@0:    /* Bit-reverse the input */
michael@0:    for (i=0;i<st->nfft;i++)
michael@0:       fout[st->bitrev[i]] = fin[i];
michael@0: 
michael@0:    fstride[0] = 1;
michael@0:    L=0;
michael@0:    do {
michael@0:       p = st->factors[2*L];
michael@0:       m = st->factors[2*L+1];
michael@0:       fstride[L+1] = fstride[L]*p;
michael@0:       L++;
michael@0:    } while(m!=1);
michael@0:    m = st->factors[2*L-1];
michael@0:    for (i=L-1;i>=0;i--)
michael@0:    {
michael@0:       if (i!=0)
michael@0:          m2 = st->factors[2*i-1];
michael@0:       else
michael@0:          m2 = 1;
michael@0:       switch (st->factors[2*i])
michael@0:       {
michael@0:       case 2:
michael@0:          ki_bfly2(fout,fstride[i]<<shift,st,m, fstride[i], m2);
michael@0:          break;
michael@0:       case 4:
michael@0:          ki_bfly4(fout,fstride[i]<<shift,st,m, fstride[i], m2);
michael@0:          break;
michael@0: #ifndef RADIX_TWO_ONLY
michael@0:       case 3:
michael@0:          ki_bfly3(fout,fstride[i]<<shift,st,m, fstride[i], m2);
michael@0:          break;
michael@0:       case 5:
michael@0:          ki_bfly5(fout,fstride[i]<<shift,st,m, fstride[i], m2);
michael@0:          break;
michael@0: #endif
michael@0:       }
michael@0:       m = m2;
michael@0:    }
michael@0: }
michael@0: