blob: 84a7ec862cbb767a70d88aa668ae84847a559f73 [file] [log] [blame]
/* ------------------------------------------------------------------
* Copyright (C) 1998-2009 PacketVideo
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either
* express or implied.
* See the License for the specific language governing permissions
* and limitations under the License.
* -------------------------------------------------------------------
*/
/*
Pathname: .inv_long_complex_rot.c
Funtions: inv_long_complex_rot
------------------------------------------------------------------------------
REVISION HISTORY
Description: Change the input argument, no shifts information from long fft_rx4
, do not have to check for shifts.
Date: 10/18/2002
Description:
(1) Change the input argument, only a single max is passed.
(2) Eliminate search for max, a fixed shift has replaced the
search for max with minimal loss of precision.
(3) Eliminated unused variables
Date: 10/28/2002
Description:
(1) Added comments per code review
Description:
------------------------------------------------------------------------------
INPUT AND OUTPUT DEFINITIONS
Inputs:
Data_in = Input vector (sized for long windows
TWICE_INV_LONG_CX_ROT_LENGTH), with time domain samples
type Int32 *
Data_out = Output vector with a post-rotation by exp(j(2pi/N)(k+1/8)),
(sized for long windows TWICE_INV_LONG_CX_ROT_LENGTH)
type Int32 *
max = Input, carries the maximum value of the input vector
"Data_in"
type Int32
Local Stores/Buffers/Pointers Needed:
None
Global Stores/Buffers/Pointers Needed:
None
Outputs:
exp = shift factor to reflect signal scaling
Pointers and Buffers Modified:
Results are return in "Data_out"
Local Stores Modified:
None
Global Stores Modified:
None
------------------------------------------------------------------------------
FUNCTION DESCRIPTION
inv_long_complex_rot() performs the complex rotation for the inverse MDCT
for the case of long windows. It also performs digit reverse ordering of
the first and second halves of the input vector "Data_in", as well as
reordering of the two half vectors (following radix-2 decomposition)
Word normalization is also done to ensure 16 by 16 bit multiplications.
------------------------------------------------------------------------------
REQUIREMENTS
inv_long_complex_rot() should execute a post-rotation by
exp(-j(2pi/N)(k+1/8)), digit reverse ordering and word normalization
------------------------------------------------------------------------------
REFERENCES
------------------------------------------------------------------------------
RESOURCES USED
When the code is written for a specific target processor the
the resources used should be documented below.
STACK USAGE: [stack count for this module] + [variable to represent
stack usage for each subroutine called]
where: [stack usage variable] = stack usage for [subroutine
name] (see [filename].ext)
DATA MEMORY USED: x words
PROGRAM MEMORY USED: x words
CLOCK CYCLES: [cycle count equation for this module] + [variable
used to represent cycle count for each subroutine
called]
where: [cycle count variable] = cycle count for [subroutine
name] (see [filename].ext)
------------------------------------------------------------------------------
*/
/*----------------------------------------------------------------------------
; INCLUDES
----------------------------------------------------------------------------*/
#include "digit_reversal_tables.h"
#include "inv_long_complex_rot.h"
#include "imdct_fxp.h"
#include "inv_long_complex_rot.h"
#include "pv_normalize.h"
#include "fxp_mul32.h"
#include "aac_mem_funcs.h"
/*----------------------------------------------------------------------------
; MACROS
; Define module specific macros here
----------------------------------------------------------------------------*/
/*----------------------------------------------------------------------------
; DEFINES
; Include all pre-processor statements here. Include conditional
; compile variables also.
----------------------------------------------------------------------------*/
/*----------------------------------------------------------------------------
; LOCAL FUNCTION DEFINITIONS
; Function Prototype declaration
----------------------------------------------------------------------------*/
/*----------------------------------------------------------------------------
; LOCAL VARIABLE DEFINITIONS
; Variable declaration - defined here and used outside this module
----------------------------------------------------------------------------*/
/*----------------------------------------------------------------------------
; EXTERNAL FUNCTION REFERENCES
; Declare functions defined elsewhere and referenced in this module
----------------------------------------------------------------------------*/
/*----------------------------------------------------------------------------
; EXTERNAL VARIABLES REFERENCES
; Declare variables used in this module but defined elsewhere
----------------------------------------------------------------------------*/
Int inv_long_complex_rot(
Int32 *Data,
Int32 max)
{
Int i;
Int16 I;
const Int32 *p_rotate;
Int32 temp_re;
Int32 temp_im;
Int32 exp_jw;
Int32 *pData_in_1;
Int32 *pData_in_2;
Int exp;
Int32 *pData_in_ref1;
Int32 *pData_in_ref2;
Int16 temp_re_0;
Int16 temp_im_0;
Int16 temp_re_1;
Int16 temp_im_1;
Int16 *p_Data_Int_precision;
Int n = 2048;
Int n_2 = n >> 1;
Int n_4 = n >> 2;
Int n_3_4 = n_2 + n_4;
Int16 *px_1;
Int16 *px_2;
Int16 *px_3;
Int16 *px_4;
Int16 J;
const Int32 *p_rotate2;
p_rotate = &exp_rotation_N_2048[255];
p_rotate2 = &exp_rotation_N_2048[256];
pData_in_ref1 = Data;
pData_in_ref2 = &Data[TWICE_INV_LONG_CX_ROT_LENGTH];
/*
* Apply A/2^(diff) + B
*/
p_Data_Int_precision = (Int16 *)Data;
exp = 16 - pv_normalize(max);
/*
* px2--> <--px1 px4--> <--px3
*
* | | |
* |+++++++++++++|+++++++++++++|+++++++++++++|+++++++++++++|
* | | | |
* n/4 n/2 3n/4
*/
I = 255;
J = 256;
pData_in_1 = pData_in_ref2 + I;
px_1 = (Int16 *)pData_in_1;
px_1++;
pData_in_2 = pData_in_ref2 + J;
px_4 = (Int16 *)pData_in_2;
exp -= 1;
for (i = INV_LONG_CX_ROT_LENGTH >> 1; i != 0; i--)
{
pData_in_2 = pData_in_ref1 + J;
temp_im = *(pData_in_2++);
temp_re = *(pData_in_2);
/*
* cos_n + j*sin_n == exp(j(2pi/N)(k+1/8))
*/
exp_jw = *p_rotate2++;
/*
* Post-rotation
*/
temp_re_0 = (Int16)(cmplx_mul32_by_16(temp_re, -temp_im, exp_jw) >> exp);
temp_im_0 = (Int16)(cmplx_mul32_by_16(temp_im, temp_re, exp_jw) >> exp);
pData_in_1 = pData_in_ref2 + I;
/*
* Use auxiliary variables to avoid double accesses to memory.
* Data in is scaled to use only lower 16 bits.
*/
temp_re = *(pData_in_1--);
temp_im = *(pData_in_1);
/*
* cos_n + j*sin_n == exp(j(2pi/N)(k+1/8))
*/
exp_jw = *p_rotate--;
/*
* Post-rotation
*/
temp_re_1 = (Int16)(cmplx_mul32_by_16(temp_re, -temp_im, exp_jw) >> exp);
temp_im_1 = (Int16)(cmplx_mul32_by_16(temp_im, temp_re, exp_jw) >> exp);
/*
* Repeat procedure for odd index at the output
*/
pData_in_2 = pData_in_ref2 + J;
J += 2;
temp_im = *(pData_in_2++);
temp_re = *(pData_in_2);
*(px_1--) = temp_re_0;
*(px_1--) = temp_im_1;
*(px_4++) = temp_im_0;
*(px_4++) = temp_re_1;
exp_jw = *p_rotate2++;
*(px_1--) = (Int16)(cmplx_mul32_by_16(temp_re, -temp_im, exp_jw) >> exp);
*(px_4++) = (Int16)(cmplx_mul32_by_16(temp_im, temp_re, exp_jw) >> exp);
/*
* Repeat procedure for odd index at the output
*/
pData_in_1 = pData_in_ref1 + I;
I -= 2;
temp_re = *(pData_in_1--);
temp_im = *(pData_in_1);
exp_jw = *p_rotate--;
*(px_4++) = (Int16)(cmplx_mul32_by_16(temp_re, -temp_im, exp_jw) >> exp);
*(px_1--) = (Int16)(cmplx_mul32_by_16(temp_im, temp_re, exp_jw) >> exp);
}
/*
* <--px1 px4-->
*
* | | |
* |-------------|-------------|/////////////|\\\\\\\\\\\\\|
* | | | |
* n/4 n/2 3n/4
*/
px_1 = p_Data_Int_precision + n_2 - 1;
px_2 = p_Data_Int_precision;
px_4 = p_Data_Int_precision + n_3_4 - 1;
for (i = 0; i<INV_LONG_CX_ROT_LENGTH >> 1; i++)
{
Int16 temp_re_0 = *(px_4--);
Int16 temp_im_1 = *(px_4--);
Int16 temp_re_2 = *(px_4--);
Int16 temp_im_3 = *(px_4--);
*(px_1--) = temp_re_0;
*(px_1--) = temp_im_1;
*(px_1--) = temp_re_2;
*(px_1--) = temp_im_3;
*(px_2++) = (-temp_re_0);
*(px_2++) = (-temp_im_1);
*(px_2++) = (-temp_re_2);
*(px_2++) = (-temp_im_3);
}
px_4 = p_Data_Int_precision + n_2;
pv_memcpy(px_4, pData_in_ref2 + 256, TWICE_INV_LONG_CX_ROT_LENGTH*sizeof(*px_4));
/*
* px2--> <--px1 px4--> <--px3
*
* | | |
* |+++++++++++++|+++++++++++++|+++++++++++++|+++++++++++++|
* | | | |
* n/4 n/2 3n/4
*/
px_3 = p_Data_Int_precision + n - 1;
for (i = 0; i<INV_LONG_CX_ROT_LENGTH >> 1; i++)
{
Int16 temp_im_0 = *(px_4++);
Int16 temp_re_1 = *(px_4++);
Int16 temp_im_2 = *(px_4++);
Int16 temp_re_3 = *(px_4++);
*(px_3--) = temp_im_0;
*(px_3--) = temp_re_1;
*(px_3--) = temp_im_2;
*(px_3--) = temp_re_3;
}
return (exp + 1);
}