/*

Copyright (c) 2003-2004, Mark Borgerding

All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

    * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

    * Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

    * Neither the author nor the names of any contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

*/

/* kiss_fft.h

   defines kiss_fft_s32_scalar as either short or a float type

   and defines

   typedef struct { kiss_fft_s32_scalar r; kiss_fft_s32_scalar i; }kiss_fft_s32_cpx; */

#include "kiss_fft_s32.h"

#include <limits.h>

/* The 2*sizeof(size_t) alignment here is borrowed from

 * GNU libc, so it should be good most everywhere.

 * It is more conservative than is needed on some 64-bit

 * platforms, but ia64 does require a 16-byte alignment.

 * The SIMD extensions for x86 and ppc32 would want a

 * larger alignment than this, but we don't need to

 * do better than malloc.

 *

 * Borrowed from GLib's gobject/gtype.c

 */

#define STRUCT_ALIGNMENT (2 * sizeof (size_t))

#define ALIGN_STRUCT(offset) \

      ((offset + (STRUCT_ALIGNMENT - 1)) & -STRUCT_ALIGNMENT)

#define MAXFACTORS 32

/* e.g. an fft of length 128 has 4 factors 

 as far as kissfft is concerned

 4*4*4*2

 */

struct kiss_fft_s32_state{

    int nfft;

    int inverse;

    int factors[2*MAXFACTORS];

    kiss_fft_s32_cpx twiddles[1];

};

/*

  Explanation of macros dealing with complex math:

   C_MUL(m,a,b)         : m = a*b

   C_FIXDIV( c , div )  : if a fixed point impl., c /= div. noop otherwise

   C_SUB( res, a,b)     : res = a - b

   C_SUBFROM( res , a)  : res -= a

   C_ADDTO( res , a)    : res += a

 * */

#define FRACBITS 31

#define SAMPPROD int64_t

#define SAMP_MAX 2147483647

#define SAMP_MIN -SAMP_MAX

#if defined(CHECK_OVERFLOW)

#  define CHECK_OVERFLOW_OP(a,op,b)  \

	if ( (SAMPPROD)(a) op (SAMPPROD)(b) > SAMP_MAX || (SAMPPROD)(a) op (SAMPPROD)(b) < SAMP_MIN ) { \

		fprintf(stderr,"WARNING:overflow @ " __FILE__ "(%d): (%d " #op" %d) = %ld\n",__LINE__,(a),(b),(SAMPPROD)(a) op (SAMPPROD)(b) );  }

#endif

#   define smul(a,b) ( (SAMPPROD)(a)*(b) )

#   define sround( x )  (kiss_fft_s32_scalar)( ( (x) + (1<<(FRACBITS-1)) ) >> FRACBITS )

#   define S_MUL(a,b) sround( smul(a,b) )

#   define C_MUL(m,a,b) \

      do{ (m).r = sround( smul((a).r,(b).r) - smul((a).i,(b).i) ); \

          (m).i = sround( smul((a).r,(b).i) + smul((a).i,(b).r) ); }while(0)

#   define DIVSCALAR(x,k) \

	(x) = sround( smul(  x, SAMP_MAX/k ) )

#   define C_FIXDIV(c,div) \

	do {    DIVSCALAR( (c).r , div);  \

		DIVSCALAR( (c).i  , div); }while (0)

#   define C_MULBYSCALAR( c, s ) \

    do{ (c).r =  sround( smul( (c).r , s ) ) ;\

        (c).i =  sround( smul( (c).i , s ) ) ; }while(0)

#ifndef CHECK_OVERFLOW_OP

#  define CHECK_OVERFLOW_OP(a,op,b) /* noop */

#endif

#define  C_ADD( res, a,b)\

    do { \

	    CHECK_OVERFLOW_OP((a).r,+,(b).r)\

	    CHECK_OVERFLOW_OP((a).i,+,(b).i)\

	    (res).r=(a).r+(b).r;  (res).i=(a).i+(b).i; \

    }while(0)

#define  C_SUB( res, a,b)\

    do { \

	    CHECK_OVERFLOW_OP((a).r,-,(b).r)\

	    CHECK_OVERFLOW_OP((a).i,-,(b).i)\

	    (res).r=(a).r-(b).r;  (res).i=(a).i-(b).i; \

    }while(0)

#define C_ADDTO( res , a)\

    do { \

	    CHECK_OVERFLOW_OP((res).r,+,(a).r)\

	    CHECK_OVERFLOW_OP((res).i,+,(a).i)\

	    (res).r += (a).r;  (res).i += (a).i;\

    }while(0)

#define C_SUBFROM( res , a)\

    do {\

	    CHECK_OVERFLOW_OP((res).r,-,(a).r)\

	    CHECK_OVERFLOW_OP((res).i,-,(a).i)\

	    (res).r -= (a).r;  (res).i -= (a).i; \

    }while(0)

#  define KISS_FFT_S32_COS(phase)  floor(.5+SAMP_MAX * cos (phase))

#  define KISS_FFT_S32_SIN(phase)  floor(.5+SAMP_MAX * sin (phase))

#  define HALF_OF(x) ((x)>>1)

#define  kf_cexp(x,phase) \

	do{ \

		(x)->r = KISS_FFT_S32_COS(phase);\

		(x)->i = KISS_FFT_S32_SIN(phase);\

	}while(0)

/* a debugging function */

#define pcpx(c)\

    fprintf(stderr,"%g + %gi\n",(double)((c)->r),(double)((c)->i) )