/src/libvpx/vp9/common/vp9_idct.c

Source
/*
 *  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
 *
 *  Use of this source code is governed by a BSD-style license
 *  that can be found in the LICENSE file in the root of the source
 *  tree. An additional intellectual property rights grant can be found
 *  in the file PATENTS.  All contributing project authors may
 *  be found in the AUTHORS file in the root of the source tree.
 */

#include <math.h>

#include "./vp9_rtcd.h"
#include "./vpx_dsp_rtcd.h"
#include "vp9/common/vp9_blockd.h"
#include "vp9/common/vp9_idct.h"
#include "vpx_dsp/inv_txfm.h"
#include "vpx_ports/mem.h"

void vp9_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride,
                         int tx_type) {
  const transform_2d IHT_4[] = {
    { idct4_c, idct4_c },   // DCT_DCT  = 0
    { iadst4_c, idct4_c },  // ADST_DCT = 1
    { idct4_c, iadst4_c },  // DCT_ADST = 2
    { iadst4_c, iadst4_c }  // ADST_ADST = 3
  };

  int i, j;
  tran_low_t out[4 * 4];
  tran_low_t *outptr = out;
  tran_low_t temp_in[4], temp_out[4];

  // inverse transform row vectors
  for (i = 0; i < 4; ++i) {
    IHT_4[tx_type].rows(input, outptr);
    input += 4;
    outptr += 4;
  }

  // inverse transform column vectors
  for (i = 0; i < 4; ++i) {
    for (j = 0; j < 4; ++j) temp_in[j] = out[j * 4 + i];
    IHT_4[tx_type].cols(temp_in, temp_out);
    for (j = 0; j < 4; ++j) {
      dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
                                            ROUND_POWER_OF_TWO(temp_out[j], 4));
    }
  }
}

static const transform_2d IHT_8[] = {
  { idct8_c, idct8_c },   // DCT_DCT  = 0
  { iadst8_c, idct8_c },  // ADST_DCT = 1
  { idct8_c, iadst8_c },  // DCT_ADST = 2
  { iadst8_c, iadst8_c }  // ADST_ADST = 3
};

void vp9_iht8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride,
                         int tx_type) {
  int i, j;
  tran_low_t out[8 * 8];
  tran_low_t *outptr = out;
  tran_low_t temp_in[8], temp_out[8];
  const transform_2d ht = IHT_8[tx_type];

  // inverse transform row vectors
  for (i = 0; i < 8; ++i) {
    ht.rows(input, outptr);
    input += 8;
    outptr += 8;
  }

  // inverse transform column vectors
  for (i = 0; i < 8; ++i) {
    for (j = 0; j < 8; ++j) temp_in[j] = out[j * 8 + i];
    ht.cols(temp_in, temp_out);
    for (j = 0; j < 8; ++j) {
      dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
                                            ROUND_POWER_OF_TWO(temp_out[j], 5));
    }
  }
}

static const transform_2d IHT_16[] = {
  { idct16_c, idct16_c },   // DCT_DCT  = 0
  { iadst16_c, idct16_c },  // ADST_DCT = 1
  { idct16_c, iadst16_c },  // DCT_ADST = 2
  { iadst16_c, iadst16_c }  // ADST_ADST = 3
};

void vp9_iht16x16_256_add_c(const tran_low_t *input, uint8_t *dest, int stride,
                            int tx_type) {
  int i, j;
  tran_low_t out[16 * 16];
  tran_low_t *outptr = out;
  tran_low_t temp_in[16], temp_out[16];
  const transform_2d ht = IHT_16[tx_type];

  // Rows
  for (i = 0; i < 16; ++i) {
    ht.rows(input, outptr);
    input += 16;
    outptr += 16;
  }

  // Columns
  for (i = 0; i < 16; ++i) {
    for (j = 0; j < 16; ++j) temp_in[j] = out[j * 16 + i];
    ht.cols(temp_in, temp_out);
    for (j = 0; j < 16; ++j) {
      dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
                                            ROUND_POWER_OF_TWO(temp_out[j], 6));
    }
  }
}

// idct
void vp9_idct4x4_add(const tran_low_t *input, uint8_t *dest, int stride,
                     int eob) {
  if (eob > 1)
    vpx_idct4x4_16_add(input, dest, stride);
  else
    vpx_idct4x4_1_add(input, dest, stride);
}

void vp9_iwht4x4_add(const tran_low_t *input, uint8_t *dest, int stride,
                     int eob) {
  if (eob > 1)
    vpx_iwht4x4_16_add(input, dest, stride);
  else
    vpx_iwht4x4_1_add(input, dest, stride);
}

void vp9_idct8x8_add(const tran_low_t *input, uint8_t *dest, int stride,
                     int eob) {
  // If dc is 1, then input[0] is the reconstructed value, do not need
  // dequantization. Also, when dc is 1, dc is counted in eobs, namely eobs >=1.

  // The calculation can be simplified if there are not many non-zero dct
  // coefficients. Use eobs to decide what to do.
  if (eob == 1)
    // DC only DCT coefficient
    vpx_idct8x8_1_add(input, dest, stride);
  else if (eob <= 12)
    vpx_idct8x8_12_add(input, dest, stride);
  else
    vpx_idct8x8_64_add(input, dest, stride);
}

void vp9_idct16x16_add(const tran_low_t *input, uint8_t *dest, int stride,
                       int eob) {
  assert(((intptr_t)input) % 32 == 0);
  /* The calculation can be simplified if there are not many non-zero dct
   * coefficients. Use eobs to separate different cases. */
  if (eob == 1) /* DC only DCT coefficient. */
    vpx_idct16x16_1_add(input, dest, stride);
  else if (eob <= 10)
    vpx_idct16x16_10_add(input, dest, stride);
  else if (eob <= 38)
    vpx_idct16x16_38_add(input, dest, stride);
  else
    vpx_idct16x16_256_add(input, dest, stride);
}

void vp9_idct32x32_add(const tran_low_t *input, uint8_t *dest, int stride,
                       int eob) {
  assert(((intptr_t)input) % 32 == 0);
  if (eob == 1)
    vpx_idct32x32_1_add(input, dest, stride);
  else if (eob <= 34)
    // non-zero coeff only in upper-left 8x8
    vpx_idct32x32_34_add(input, dest, stride);
  else if (eob <= 135)
    // non-zero coeff only in upper-left 16x16
    vpx_idct32x32_135_add(input, dest, stride);
  else
    vpx_idct32x32_1024_add(input, dest, stride);
}

// iht
void vp9_iht4x4_add(TX_TYPE tx_type, const tran_low_t *input, uint8_t *dest,
                    int stride, int eob) {
  if (tx_type == DCT_DCT)
    vp9_idct4x4_add(input, dest, stride, eob);
  else
    vp9_iht4x4_16_add(input, dest, stride, tx_type);
}

void vp9_iht8x8_add(TX_TYPE tx_type, const tran_low_t *input, uint8_t *dest,
                    int stride, int eob) {
  if (tx_type == DCT_DCT) {
    vp9_idct8x8_add(input, dest, stride, eob);
  } else {
    vp9_iht8x8_64_add(input, dest, stride, tx_type);
  }
}

void vp9_iht16x16_add(TX_TYPE tx_type, const tran_low_t *input, uint8_t *dest,
                      int stride, int eob) {
  if (tx_type == DCT_DCT) {
    vp9_idct16x16_add(input, dest, stride, eob);
  } else {
    vp9_iht16x16_256_add(input, dest, stride, tx_type);
  }
}

#if CONFIG_VP9_HIGHBITDEPTH

void vp9_highbd_iht4x4_16_add_c(const tran_low_t *input, uint16_t *dest,
                                int stride, int tx_type, int bd) {
  const highbd_transform_2d IHT_4[] = {
    { vpx_highbd_idct4_c, vpx_highbd_idct4_c },   // DCT_DCT  = 0
    { vpx_highbd_iadst4_c, vpx_highbd_idct4_c },  // ADST_DCT = 1
    { vpx_highbd_idct4_c, vpx_highbd_iadst4_c },  // DCT_ADST = 2
    { vpx_highbd_iadst4_c, vpx_highbd_iadst4_c }  // ADST_ADST = 3
  };

  int i, j;
  tran_low_t out[4 * 4];
  tran_low_t *outptr = out;
  tran_low_t temp_in[4], temp_out[4];

  // Inverse transform row vectors.
  for (i = 0; i < 4; ++i) {
    IHT_4[tx_type].rows(input, outptr, bd);
    input += 4;
    outptr += 4;
  }

  // Inverse transform column vectors.
  for (i = 0; i < 4; ++i) {
    for (j = 0; j < 4; ++j) temp_in[j] = out[j * 4 + i];
    IHT_4[tx_type].cols(temp_in, temp_out, bd);
    for (j = 0; j < 4; ++j) {
      dest[j * stride + i] = highbd_clip_pixel_add(
          dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 4), bd);
    }
  }
}

static const highbd_transform_2d HIGH_IHT_8[] = {
  { vpx_highbd_idct8_c, vpx_highbd_idct8_c },   // DCT_DCT  = 0
  { vpx_highbd_iadst8_c, vpx_highbd_idct8_c },  // ADST_DCT = 1
  { vpx_highbd_idct8_c, vpx_highbd_iadst8_c },  // DCT_ADST = 2
  { vpx_highbd_iadst8_c, vpx_highbd_iadst8_c }  // ADST_ADST = 3
};

void vp9_highbd_iht8x8_64_add_c(const tran_low_t *input, uint16_t *dest,
                                int stride, int tx_type, int bd) {
  int i, j;
  tran_low_t out[8 * 8];
  tran_low_t *outptr = out;
  tran_low_t temp_in[8], temp_out[8];
  const highbd_transform_2d ht = HIGH_IHT_8[tx_type];

  // Inverse transform row vectors.
  for (i = 0; i < 8; ++i) {
    ht.rows(input, outptr, bd);
    input += 8;
    outptr += 8;
  }

  // Inverse transform column vectors.
  for (i = 0; i < 8; ++i) {
    for (j = 0; j < 8; ++j) temp_in[j] = out[j * 8 + i];
    ht.cols(temp_in, temp_out, bd);
    for (j = 0; j < 8; ++j) {
      dest[j * stride + i] = highbd_clip_pixel_add(
          dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 5), bd);
    }
  }
}

static const highbd_transform_2d HIGH_IHT_16[] = {
  { vpx_highbd_idct16_c, vpx_highbd_idct16_c },   // DCT_DCT  = 0
  { vpx_highbd_iadst16_c, vpx_highbd_idct16_c },  // ADST_DCT = 1
  { vpx_highbd_idct16_c, vpx_highbd_iadst16_c },  // DCT_ADST = 2
  { vpx_highbd_iadst16_c, vpx_highbd_iadst16_c }  // ADST_ADST = 3
};

void vp9_highbd_iht16x16_256_add_c(const tran_low_t *input, uint16_t *dest,
                                   int stride, int tx_type, int bd) {
  int i, j;
  tran_low_t out[16 * 16];
  tran_low_t *outptr = out;
  tran_low_t temp_in[16], temp_out[16];
  const highbd_transform_2d ht = HIGH_IHT_16[tx_type];

  // Rows
  for (i = 0; i < 16; ++i) {
    ht.rows(input, outptr, bd);
    input += 16;
    outptr += 16;
  }

  // Columns
  for (i = 0; i < 16; ++i) {
    for (j = 0; j < 16; ++j) temp_in[j] = out[j * 16 + i];
    ht.cols(temp_in, temp_out, bd);
    for (j = 0; j < 16; ++j) {
      dest[j * stride + i] = highbd_clip_pixel_add(
          dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 6), bd);
    }
  }
}

// idct
void vp9_highbd_idct4x4_add(const tran_low_t *input, uint16_t *dest, int stride,
                            int eob, int bd) {
  if (eob > 1)
    vpx_highbd_idct4x4_16_add(input, dest, stride, bd);
  else
    vpx_highbd_idct4x4_1_add(input, dest, stride, bd);
}

void vp9_highbd_iwht4x4_add(const tran_low_t *input, uint16_t *dest, int stride,
                            int eob, int bd) {
  if (eob > 1)
    vpx_highbd_iwht4x4_16_add(input, dest, stride, bd);
  else
    vpx_highbd_iwht4x4_1_add(input, dest, stride, bd);
}

void vp9_highbd_idct8x8_add(const tran_low_t *input, uint16_t *dest, int stride,
                            int eob, int bd) {
  // If dc is 1, then input[0] is the reconstructed value, do not need
  // dequantization. Also, when dc is 1, dc is counted in eobs, namely eobs >=1.

  // The calculation can be simplified if there are not many non-zero dct
  // coefficients. Use eobs to decide what to do.
  // DC only DCT coefficient
  if (eob == 1) {
    vpx_highbd_idct8x8_1_add(input, dest, stride, bd);
  } else if (eob <= 12) {
    vpx_highbd_idct8x8_12_add(input, dest, stride, bd);
  } else {
    vpx_highbd_idct8x8_64_add(input, dest, stride, bd);
  }
}

void vp9_highbd_idct16x16_add(const tran_low_t *input, uint16_t *dest,
                              int stride, int eob, int bd) {
  // The calculation can be simplified if there are not many non-zero dct
  // coefficients. Use eobs to separate different cases.
  // DC only DCT coefficient.
  if (eob == 1) {
    vpx_highbd_idct16x16_1_add(input, dest, stride, bd);
  } else if (eob <= 10) {
    vpx_highbd_idct16x16_10_add(input, dest, stride, bd);
  } else if (eob <= 38) {
    vpx_highbd_idct16x16_38_add(input, dest, stride, bd);
  } else {
    vpx_highbd_idct16x16_256_add(input, dest, stride, bd);
  }
}

void vp9_highbd_idct32x32_add(const tran_low_t *input, uint16_t *dest,
                              int stride, int eob, int bd) {
  // Non-zero coeff only in upper-left 8x8
  if (eob == 1) {
    vpx_highbd_idct32x32_1_add(input, dest, stride, bd);
  } else if (eob <= 34) {
    vpx_highbd_idct32x32_34_add(input, dest, stride, bd);
  } else if (eob <= 135) {
    vpx_highbd_idct32x32_135_add(input, dest, stride, bd);
  } else {
    vpx_highbd_idct32x32_1024_add(input, dest, stride, bd);
  }
}

// iht
void vp9_highbd_iht4x4_add(TX_TYPE tx_type, const tran_low_t *input,
                           uint16_t *dest, int stride, int eob, int bd) {
  if (tx_type == DCT_DCT)
    vp9_highbd_idct4x4_add(input, dest, stride, eob, bd);
  else
    vp9_highbd_iht4x4_16_add(input, dest, stride, tx_type, bd);
}

void vp9_highbd_iht8x8_add(TX_TYPE tx_type, const tran_low_t *input,
                           uint16_t *dest, int stride, int eob, int bd) {
  if (tx_type == DCT_DCT) {
    vp9_highbd_idct8x8_add(input, dest, stride, eob, bd);
  } else {
    vp9_highbd_iht8x8_64_add(input, dest, stride, tx_type, bd);
  }
}

void vp9_highbd_iht16x16_add(TX_TYPE tx_type, const tran_low_t *input,
                             uint16_t *dest, int stride, int eob, int bd) {
  if (tx_type == DCT_DCT) {
    vp9_highbd_idct16x16_add(input, dest, stride, eob, bd);
  } else {
    vp9_highbd_iht16x16_256_add(input, dest, stride, tx_type, bd);
  }
}
#endif  // CONFIG_VP9_HIGHBITDEPTH

Coverage Report

Created: 2025-11-16 07:20

Line	Count	Source
1		/*
2		* Copyright (c) 2010 The WebM project authors. All Rights Reserved.
3		*
4		* Use of this source code is governed by a BSD-style license
5		* that can be found in the LICENSE file in the root of the source
6		* tree. An additional intellectual property rights grant can be found
7		* in the file PATENTS. All contributing project authors may
8		* be found in the AUTHORS file in the root of the source tree.
9		*/
10
11		#include <math.h>
12
13		#include "./vp9_rtcd.h"
14		#include "./vpx_dsp_rtcd.h"
15		#include "vp9/common/vp9_blockd.h"
16		#include "vp9/common/vp9_idct.h"
17		#include "vpx_dsp/inv_txfm.h"
18		#include "vpx_ports/mem.h"
19
20		void vp9_iht4x4_16_add_c(const tran_low_t input, uint8_t dest, int stride,
21	0	int tx_type) {
22	0	const transform_2d IHT_4[] = {
23	0	{ idct4_c, idct4_c }, // DCT_DCT = 0
24	0	{ iadst4_c, idct4_c }, // ADST_DCT = 1
25	0	{ idct4_c, iadst4_c }, // DCT_ADST = 2
26	0	{ iadst4_c, iadst4_c } // ADST_ADST = 3
27	0	};
28
29	0	int i, j;
30	0	tran_low_t out[4 * 4];
31	0	tran_low_t *outptr = out;
32	0	tran_low_t temp_in[4], temp_out[4];
33
34		// inverse transform row vectors
35	0	for (i = 0; i < 4; ++i) {
36	0	IHT_4[tx_type].rows(input, outptr);
37	0	input += 4;
38	0	outptr += 4;
39	0	}
40
41		// inverse transform column vectors
42	0	for (i = 0; i < 4; ++i) {
43	0	for (j = 0; j < 4; ++j) temp_in[j] = out[j * 4 + i];
44	0	IHT_4[tx_type].cols(temp_in, temp_out);
45	0	for (j = 0; j < 4; ++j) {
46	0	dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
47	0	ROUND_POWER_OF_TWO(temp_out[j], 4));
48	0	}
49	0	}
50	0	}
51
52		static const transform_2d IHT_8[] = {
53		{ idct8_c, idct8_c }, // DCT_DCT = 0
54		{ iadst8_c, idct8_c }, // ADST_DCT = 1
55		{ idct8_c, iadst8_c }, // DCT_ADST = 2
56		{ iadst8_c, iadst8_c } // ADST_ADST = 3
57		};
58
59		void vp9_iht8x8_64_add_c(const tran_low_t input, uint8_t dest, int stride,
60	0	int tx_type) {
61	0	int i, j;
62	0	tran_low_t out[8 * 8];
63	0	tran_low_t *outptr = out;
64	0	tran_low_t temp_in[8], temp_out[8];
65	0	const transform_2d ht = IHT_8[tx_type];
66
67		// inverse transform row vectors
68	0	for (i = 0; i < 8; ++i) {
69	0	ht.rows(input, outptr);
70	0	input += 8;
71	0	outptr += 8;
72	0	}
73
74		// inverse transform column vectors
75	0	for (i = 0; i < 8; ++i) {
76	0	for (j = 0; j < 8; ++j) temp_in[j] = out[j * 8 + i];
77	0	ht.cols(temp_in, temp_out);
78	0	for (j = 0; j < 8; ++j) {
79	0	dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
80	0	ROUND_POWER_OF_TWO(temp_out[j], 5));
81	0	}
82	0	}
83	0	}
84
85		static const transform_2d IHT_16[] = {
86		{ idct16_c, idct16_c }, // DCT_DCT = 0
87		{ iadst16_c, idct16_c }, // ADST_DCT = 1
88		{ idct16_c, iadst16_c }, // DCT_ADST = 2
89		{ iadst16_c, iadst16_c } // ADST_ADST = 3
90		};
91
92		void vp9_iht16x16_256_add_c(const tran_low_t input, uint8_t dest, int stride,
93	0	int tx_type) {
94	0	int i, j;
95	0	tran_low_t out[16 * 16];
96	0	tran_low_t *outptr = out;
97	0	tran_low_t temp_in[16], temp_out[16];
98	0	const transform_2d ht = IHT_16[tx_type];
99
100		// Rows
101	0	for (i = 0; i < 16; ++i) {
102	0	ht.rows(input, outptr);
103	0	input += 16;
104	0	outptr += 16;
105	0	}
106
107		// Columns
108	0	for (i = 0; i < 16; ++i) {
109	0	for (j = 0; j < 16; ++j) temp_in[j] = out[j * 16 + i];
110	0	ht.cols(temp_in, temp_out);
111	0	for (j = 0; j < 16; ++j) {
112	0	dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
113	0	ROUND_POWER_OF_TWO(temp_out[j], 6));
114	0	}
115	0	}
116	0	}
117
118		// idct
119		void vp9_idct4x4_add(const tran_low_t input, uint8_t dest, int stride,
120	177M	int eob) {
121	177M	if (eob > 1)
122	155M	vpx_idct4x4_16_add(input, dest, stride);
123	22.0M	else
124	22.0M	vpx_idct4x4_1_add(input, dest, stride);
125	177M	}
126
127		void vp9_iwht4x4_add(const tran_low_t input, uint8_t dest, int stride,
128	49.8M	int eob) {
129	49.8M	if (eob > 1)
130	47.7M	vpx_iwht4x4_16_add(input, dest, stride);
131	2.07M	else
132	2.07M	vpx_iwht4x4_1_add(input, dest, stride);
133	49.8M	}
134
135		void vp9_idct8x8_add(const tran_low_t input, uint8_t dest, int stride,
136	30.7M	int eob) {
137		// If dc is 1, then input[0] is the reconstructed value, do not need
138		// dequantization. Also, when dc is 1, dc is counted in eobs, namely eobs >=1.
139
140		// The calculation can be simplified if there are not many non-zero dct
141		// coefficients. Use eobs to decide what to do.
142	30.7M	if (eob == 1)
143		// DC only DCT coefficient
144	2.26M	vpx_idct8x8_1_add(input, dest, stride);
145	28.4M	else if (eob <= 12)
146	1.48M	vpx_idct8x8_12_add(input, dest, stride);
147	26.9M	else
148	26.9M	vpx_idct8x8_64_add(input, dest, stride);
149	30.7M	}
150
151		void vp9_idct16x16_add(const tran_low_t input, uint8_t dest, int stride,
152	5.89M	int eob) {
153	5.89M	assert(((intptr_t)input) % 32 == 0);
154		/* The calculation can be simplified if there are not many non-zero dct
155		* coefficients. Use eobs to separate different cases. */
156	5.89M	if (eob == 1) /* DC only DCT coefficient. */
157	576k	vpx_idct16x16_1_add(input, dest, stride);
158	5.31M	else if (eob <= 10)
159	187k	vpx_idct16x16_10_add(input, dest, stride);
160	5.13M	else if (eob <= 38)
161	214k	vpx_idct16x16_38_add(input, dest, stride);
162	4.91M	else
163	4.91M	vpx_idct16x16_256_add(input, dest, stride);
164	5.89M	}
165
166		void vp9_idct32x32_add(const tran_low_t input, uint8_t dest, int stride,
167	3.19M	int eob) {
168	3.19M	assert(((intptr_t)input) % 32 == 0);
169	3.19M	if (eob == 1)
170	320k	vpx_idct32x32_1_add(input, dest, stride);
171	2.86M	else if (eob <= 34)
172		// non-zero coeff only in upper-left 8x8
173	116k	vpx_idct32x32_34_add(input, dest, stride);
174	2.75M	else if (eob <= 135)
175		// non-zero coeff only in upper-left 16x16
176	88.4k	vpx_idct32x32_135_add(input, dest, stride);
177	2.66M	else
178	2.66M	vpx_idct32x32_1024_add(input, dest, stride);
179	3.19M	}
180
181		// iht
182		void vp9_iht4x4_add(TX_TYPE tx_type, const tran_low_t input, uint8_t dest,
183	86.0M	int stride, int eob) {
184	86.0M	if (tx_type == DCT_DCT)
185	24.1M	vp9_idct4x4_add(input, dest, stride, eob);
186	61.8M	else
187	61.8M	vp9_iht4x4_16_add(input, dest, stride, tx_type);
188	86.0M	}
189
190		void vp9_iht8x8_add(TX_TYPE tx_type, const tran_low_t input, uint8_t dest,
191	59.5M	int stride, int eob) {
192	59.5M	if (tx_type == DCT_DCT) {
193	26.8M	vp9_idct8x8_add(input, dest, stride, eob);
194	32.6M	} else {
195	32.6M	vp9_iht8x8_64_add(input, dest, stride, tx_type);
196	32.6M	}
197	59.5M	}
198
199		void vp9_iht16x16_add(TX_TYPE tx_type, const tran_low_t input, uint8_t dest,
200	12.9M	int stride, int eob) {
201	12.9M	if (tx_type == DCT_DCT) {
202	4.68M	vp9_idct16x16_add(input, dest, stride, eob);
203	8.30M	} else {
204	8.30M	vp9_iht16x16_256_add(input, dest, stride, tx_type);
205	8.30M	}
206	12.9M	}
207
208		#if CONFIG_VP9_HIGHBITDEPTH
209
210		void vp9_highbd_iht4x4_16_add_c(const tran_low_t input, uint16_t dest,
211	0	int stride, int tx_type, int bd) {
212	0	const highbd_transform_2d IHT_4[] = {
213	0	{ vpx_highbd_idct4_c, vpx_highbd_idct4_c }, // DCT_DCT = 0
214	0	{ vpx_highbd_iadst4_c, vpx_highbd_idct4_c }, // ADST_DCT = 1
215	0	{ vpx_highbd_idct4_c, vpx_highbd_iadst4_c }, // DCT_ADST = 2
216	0	{ vpx_highbd_iadst4_c, vpx_highbd_iadst4_c } // ADST_ADST = 3
217	0	};
218
219	0	int i, j;
220	0	tran_low_t out[4 * 4];
221	0	tran_low_t *outptr = out;
222	0	tran_low_t temp_in[4], temp_out[4];
223
224		// Inverse transform row vectors.
225	0	for (i = 0; i < 4; ++i) {
226	0	IHT_4[tx_type].rows(input, outptr, bd);
227	0	input += 4;
228	0	outptr += 4;
229	0	}
230
231		// Inverse transform column vectors.
232	0	for (i = 0; i < 4; ++i) {
233	0	for (j = 0; j < 4; ++j) temp_in[j] = out[j * 4 + i];
234	0	IHT_4[tx_type].cols(temp_in, temp_out, bd);
235	0	for (j = 0; j < 4; ++j) {
236	0	dest[j * stride + i] = highbd_clip_pixel_add(
237	0	dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 4), bd);
238	0	}
239	0	}
240	0	}
241
242		static const highbd_transform_2d HIGH_IHT_8[] = {
243		{ vpx_highbd_idct8_c, vpx_highbd_idct8_c }, // DCT_DCT = 0
244		{ vpx_highbd_iadst8_c, vpx_highbd_idct8_c }, // ADST_DCT = 1
245		{ vpx_highbd_idct8_c, vpx_highbd_iadst8_c }, // DCT_ADST = 2
246		{ vpx_highbd_iadst8_c, vpx_highbd_iadst8_c } // ADST_ADST = 3
247		};
248
249		void vp9_highbd_iht8x8_64_add_c(const tran_low_t input, uint16_t dest,
250	0	int stride, int tx_type, int bd) {
251	0	int i, j;
252	0	tran_low_t out[8 * 8];
253	0	tran_low_t *outptr = out;
254	0	tran_low_t temp_in[8], temp_out[8];
255	0	const highbd_transform_2d ht = HIGH_IHT_8[tx_type];
256
257		// Inverse transform row vectors.
258	0	for (i = 0; i < 8; ++i) {
259	0	ht.rows(input, outptr, bd);
260	0	input += 8;
261	0	outptr += 8;
262	0	}
263
264		// Inverse transform column vectors.
265	0	for (i = 0; i < 8; ++i) {
266	0	for (j = 0; j < 8; ++j) temp_in[j] = out[j * 8 + i];
267	0	ht.cols(temp_in, temp_out, bd);
268	0	for (j = 0; j < 8; ++j) {
269	0	dest[j * stride + i] = highbd_clip_pixel_add(
270	0	dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 5), bd);
271	0	}
272	0	}
273	0	}
274
275		static const highbd_transform_2d HIGH_IHT_16[] = {
276		{ vpx_highbd_idct16_c, vpx_highbd_idct16_c }, // DCT_DCT = 0
277		{ vpx_highbd_iadst16_c, vpx_highbd_idct16_c }, // ADST_DCT = 1
278		{ vpx_highbd_idct16_c, vpx_highbd_iadst16_c }, // DCT_ADST = 2
279		{ vpx_highbd_iadst16_c, vpx_highbd_iadst16_c } // ADST_ADST = 3
280		};
281
282		void vp9_highbd_iht16x16_256_add_c(const tran_low_t input, uint16_t dest,
283	0	int stride, int tx_type, int bd) {
284	0	int i, j;
285	0	tran_low_t out[16 * 16];
286	0	tran_low_t *outptr = out;
287	0	tran_low_t temp_in[16], temp_out[16];
288	0	const highbd_transform_2d ht = HIGH_IHT_16[tx_type];
289
290		// Rows
291	0	for (i = 0; i < 16; ++i) {
292	0	ht.rows(input, outptr, bd);
293	0	input += 16;
294	0	outptr += 16;
295	0	}
296
297		// Columns
298	0	for (i = 0; i < 16; ++i) {
299	0	for (j = 0; j < 16; ++j) temp_in[j] = out[j * 16 + i];
300	0	ht.cols(temp_in, temp_out, bd);
301	0	for (j = 0; j < 16; ++j) {
302	0	dest[j * stride + i] = highbd_clip_pixel_add(
303	0	dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 6), bd);
304	0	}
305	0	}
306	0	}
307
308		// idct
309		void vp9_highbd_idct4x4_add(const tran_low_t input, uint16_t dest, int stride,
310	1.26M	int eob, int bd) {
311	1.26M	if (eob > 1)
312	1.06M	vpx_highbd_idct4x4_16_add(input, dest, stride, bd);
313	193k	else
314	193k	vpx_highbd_idct4x4_1_add(input, dest, stride, bd);
315	1.26M	}
316
317		void vp9_highbd_iwht4x4_add(const tran_low_t input, uint16_t dest, int stride,
318	322k	int eob, int bd) {
319	322k	if (eob > 1)
320	282k	vpx_highbd_iwht4x4_16_add(input, dest, stride, bd);
321	40.5k	else
322	40.5k	vpx_highbd_iwht4x4_1_add(input, dest, stride, bd);
323	322k	}
324
325		void vp9_highbd_idct8x8_add(const tran_low_t input, uint16_t dest, int stride,
326	324k	int eob, int bd) {
327		// If dc is 1, then input[0] is the reconstructed value, do not need
328		// dequantization. Also, when dc is 1, dc is counted in eobs, namely eobs >=1.
329
330		// The calculation can be simplified if there are not many non-zero dct
331		// coefficients. Use eobs to decide what to do.
332		// DC only DCT coefficient
333	324k	if (eob == 1) {
334	59.6k	vpx_highbd_idct8x8_1_add(input, dest, stride, bd);
335	264k	} else if (eob <= 12) {
336	102k	vpx_highbd_idct8x8_12_add(input, dest, stride, bd);
337	162k	} else {
338	162k	vpx_highbd_idct8x8_64_add(input, dest, stride, bd);
339	162k	}
340	324k	}
341
342		void vp9_highbd_idct16x16_add(const tran_low_t input, uint16_t dest,
343	501k	int stride, int eob, int bd) {
344		// The calculation can be simplified if there are not many non-zero dct
345		// coefficients. Use eobs to separate different cases.
346		// DC only DCT coefficient.
347	501k	if (eob == 1) {
348	90.8k	vpx_highbd_idct16x16_1_add(input, dest, stride, bd);
349	410k	} else if (eob <= 10) {
350	136k	vpx_highbd_idct16x16_10_add(input, dest, stride, bd);
351	273k	} else if (eob <= 38) {
352	118k	vpx_highbd_idct16x16_38_add(input, dest, stride, bd);
353	155k	} else {
354	155k	vpx_highbd_idct16x16_256_add(input, dest, stride, bd);
355	155k	}
356	501k	}
357
358		void vp9_highbd_idct32x32_add(const tran_low_t input, uint16_t dest,
359	247k	int stride, int eob, int bd) {
360		// Non-zero coeff only in upper-left 8x8
361	247k	if (eob == 1) {
362	90.9k	vpx_highbd_idct32x32_1_add(input, dest, stride, bd);
363	156k	} else if (eob <= 34) {
364	111k	vpx_highbd_idct32x32_34_add(input, dest, stride, bd);
365	111k	} else if (eob <= 135) {
366	27.7k	vpx_highbd_idct32x32_135_add(input, dest, stride, bd);
367	27.7k	} else {
368	17.2k	vpx_highbd_idct32x32_1024_add(input, dest, stride, bd);
369	17.2k	}
370	247k	}
371
372		// iht
373		void vp9_highbd_iht4x4_add(TX_TYPE tx_type, const tran_low_t *input,
374	677k	uint16_t *dest, int stride, int eob, int bd) {
375	677k	if (tx_type == DCT_DCT)
376	569k	vp9_highbd_idct4x4_add(input, dest, stride, eob, bd);
377	107k	else
378	107k	vp9_highbd_iht4x4_16_add(input, dest, stride, tx_type, bd);
379	677k	}
380
381		void vp9_highbd_iht8x8_add(TX_TYPE tx_type, const tran_low_t *input,
382	156k	uint16_t *dest, int stride, int eob, int bd) {
383	156k	if (tx_type == DCT_DCT) {
384	103k	vp9_highbd_idct8x8_add(input, dest, stride, eob, bd);
385	103k	} else {
386	53.5k	vp9_highbd_iht8x8_64_add(input, dest, stride, tx_type, bd);
387	53.5k	}
388	156k	}
389
390		void vp9_highbd_iht16x16_add(TX_TYPE tx_type, const tran_low_t *input,
391	391k	uint16_t *dest, int stride, int eob, int bd) {
392	391k	if (tx_type == DCT_DCT) {
393	317k	vp9_highbd_idct16x16_add(input, dest, stride, eob, bd);
394	317k	} else {
395	74.1k	vp9_highbd_iht16x16_256_add(input, dest, stride, tx_type, bd);
396	74.1k	}
397	391k	}
398		#endif // CONFIG_VP9_HIGHBITDEPTH