| /****************************************************************************** |
| * |
| * Copyright (C) 2015 The Android Open Source Project |
| * |
| * 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. |
| * |
| ***************************************************************************** |
| * Originally developed and contributed by Ittiam Systems Pvt. Ltd, Bangalore |
| */ |
| |
| /** |
| ****************************************************************************** |
| * @file ih264e_distortion_metrics.c |
| * |
| * @brief |
| * This file contains definitions of routines that compute distortion |
| * between two macro/sub blocks of identical dimensions |
| * |
| * @author |
| * Ittiam |
| * |
| * @par List of Functions: |
| * - ime_sub_pel_compute_sad_16x16() |
| * - ime_calculate_sad4_prog() |
| * - ime_calculate_sad3_prog() |
| * - ime_calculate_sad2_prog() |
| * - ime_compute_sad_16x16() |
| * - ime_compute_sad_16x16_fast() |
| * - ime_compute_sad_16x16_ea8() |
| * - ime_compute_sad_8x8() |
| * - ime_compute_sad_4x4() |
| * - ime_compute_sad_16x8() |
| * - ime_compute_satqd_16x16_lumainter() |
| * - ime_compute_satqd_8x16_chroma() |
| * - ime_compute_satqd_16x16_lumaintra() |
| * |
| * |
| * @remarks |
| * None |
| * |
| ******************************************************************************* |
| */ |
| |
| /*****************************************************************************/ |
| /* File Includes */ |
| /*****************************************************************************/ |
| |
| /* System include files */ |
| #include <stdio.h> |
| #include <stdlib.h> |
| #include <string.h> |
| |
| /* User include files */ |
| #include "ime_typedefs.h" |
| #include "ime_defs.h" |
| #include "ime_macros.h" |
| #include "ime_statistics.h" |
| #include "ime_platform_macros.h" |
| #include "ime_distortion_metrics.h" |
| |
| |
| /*****************************************************************************/ |
| /* Function Definitions */ |
| /*****************************************************************************/ |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD) at all subpel points about the src location |
| * |
| * @par Description |
| * This functions computes SAD at all points at a subpel distance from the |
| * current source location. |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_ref_half_x |
| * UWORD8 pointer to half pel buffer |
| * |
| * @param[out] pu1_ref_half_y |
| * UWORD8 pointer to half pel buffer |
| * |
| * @param[out] pu1_ref_half_xy |
| * UWORD8 pointer to half pel buffer |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] ref_strd |
| * integer ref stride |
| * |
| * @param[out] pi4_sad |
| * integer evaluated sad |
| * pi4_sad[0] - half x |
| * pi4_sad[1] - half x - 1 |
| * pi4_sad[2] - half y |
| * pi4_sad[3] - half y - 1 |
| * pi4_sad[4] - half xy |
| * pi4_sad[5] - half xy - 1 |
| * pi4_sad[6] - half xy - strd |
| * pi4_sad[7] - half xy - 1 - strd |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| void ime_sub_pel_compute_sad_16x16(UWORD8 *pu1_src, |
| UWORD8 *pu1_ref_half_x, |
| UWORD8 *pu1_ref_half_y, |
| UWORD8 *pu1_ref_half_xy, |
| WORD32 src_strd, |
| WORD32 ref_strd, |
| WORD32 *pi4_sad) |
| { |
| UWORD8 *pu1_ref_half_x_left = pu1_ref_half_x - 1; |
| UWORD8 *pu1_ref_half_y_top = pu1_ref_half_y - ref_strd; |
| UWORD8 *pu1_ref_half_xy_left = pu1_ref_half_xy - 1; |
| UWORD8 *pu1_ref_half_xy_top = pu1_ref_half_xy - ref_strd; |
| UWORD8 *pu1_ref_half_xy_top_left = pu1_ref_half_xy - ref_strd - 1; |
| |
| WORD32 row, col; |
| |
| memset(pi4_sad, 0, 8 * sizeof(WORD32)); |
| |
| for(row = 0; row < MB_SIZE; row++) |
| { |
| for(col = 0; col < MB_SIZE; col++) |
| { |
| WORD32 src; |
| WORD32 diff; |
| |
| src = pu1_src[col]; |
| |
| diff = src - pu1_ref_half_x[col]; |
| pi4_sad[0] += ABS(diff); |
| |
| diff = src - pu1_ref_half_x_left[col]; |
| pi4_sad[1] += ABS(diff); |
| |
| diff = src - pu1_ref_half_y[col]; |
| pi4_sad[2] += ABS(diff); |
| |
| diff = src - pu1_ref_half_y_top[col]; |
| pi4_sad[3] += ABS(diff); |
| |
| diff = src - pu1_ref_half_xy[col]; |
| pi4_sad[4] += ABS(diff); |
| |
| diff = src - pu1_ref_half_xy_left[col]; |
| pi4_sad[5] += ABS(diff); |
| |
| diff = src - pu1_ref_half_xy_top[col]; |
| pi4_sad[6] += ABS(diff); |
| |
| diff = src - pu1_ref_half_xy_top_left[col]; |
| pi4_sad[7] += ABS(diff); |
| } |
| |
| pu1_src += src_strd; |
| |
| pu1_ref_half_x += ref_strd; |
| pu1_ref_half_x_left += ref_strd; |
| |
| pu1_ref_half_y += ref_strd; |
| pu1_ref_half_y_top += ref_strd; |
| |
| pu1_ref_half_xy += ref_strd; |
| pu1_ref_half_xy_left += ref_strd; |
| pu1_ref_half_xy_top += ref_strd; |
| pu1_ref_half_xy_top_left += ref_strd; |
| } |
| } |
| |
| /** |
| ******************************************************************************* |
| * |
| * @brief compute sad |
| * |
| * @par Description: This function computes the sad at vertices of diamond grid |
| * centered at reference pointer and at unit distance from it. |
| * |
| * @param[in] pu1_ref |
| * UWORD8 pointer to the reference |
| * |
| * @param[out] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[in] ref_strd |
| * integer reference stride |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[out] pi4_sad |
| * pointer to integer array evaluated sad |
| * |
| * @returns sad at all evaluated vertexes |
| * |
| * @remarks none |
| * |
| ******************************************************************************* |
| */ |
| void ime_calculate_sad4_prog(UWORD8 *pu1_ref, |
| UWORD8 *pu1_src, |
| WORD32 ref_strd, |
| WORD32 src_strd, |
| WORD32 *pi4_sad) |
| { |
| |
| /* reference ptrs at unit 1 distance in diamond pattern centered at pu1_ref */ |
| UWORD8 *left_ptr = pu1_ref - 1; |
| UWORD8 *right_ptr = pu1_ref + 1; |
| UWORD8 *top_ptr = pu1_ref - ref_strd; |
| UWORD8 *bot_ptr = pu1_ref + ref_strd; |
| |
| /* temp var */ |
| WORD32 count2, count3; |
| UWORD32 u4_ref_buf_offset = ref_strd - MB_SIZE; |
| UWORD32 u4_cur_buf_offset = src_strd - MB_SIZE; |
| |
| memset(pi4_sad, 0, 4 * sizeof(WORD32)); |
| |
| for(count2 = MB_SIZE; count2 > 0; count2--) |
| { |
| for(count3 = MB_SIZE; count3 > 0 ; count3--) |
| { |
| WORD32 src; |
| WORD32 diff; |
| |
| src = *pu1_src++; |
| |
| diff = src - *left_ptr++; |
| pi4_sad[0] += ABS(diff); |
| |
| diff = src - *right_ptr++; |
| pi4_sad[1] += ABS(diff); |
| |
| diff = src - *top_ptr++; |
| pi4_sad[2] += ABS(diff); |
| |
| diff = src - *bot_ptr++; |
| pi4_sad[3] += ABS(diff); |
| } |
| |
| bot_ptr += u4_ref_buf_offset; |
| left_ptr += u4_ref_buf_offset; |
| right_ptr += u4_ref_buf_offset; |
| top_ptr += u4_ref_buf_offset; |
| |
| pu1_src += u4_cur_buf_offset; |
| } |
| |
| } |
| |
| /** |
| ******************************************************************************* |
| * |
| * @brief compute sad |
| * |
| * @par Description: This function computes the sad at vertices of diamond grid |
| * centered at reference pointer and at unit distance from it. |
| * |
| * @param[in] pu1_ref1, pu1_ref2, pu1_ref3 |
| * UWORD8 pointer to the reference |
| * |
| * @param[out] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[in] ref_strd |
| * integer reference stride |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[out] pi4_sad |
| * pointer to integer array evaluated sad |
| * |
| * @returns sad at all evaluated vertexes |
| * |
| * @remarks none |
| * |
| ******************************************************************************* |
| */ |
| void ime_calculate_sad3_prog(UWORD8 *pu1_ref1, |
| UWORD8 *pu1_ref2, |
| UWORD8 *pu1_ref3, |
| UWORD8 *pu1_src, |
| WORD32 ref_strd, |
| WORD32 src_strd, |
| WORD32 *pi4_sad) |
| { |
| /* temp var */ |
| WORD32 i; |
| UWORD32 u4_ref_buf_offset = ref_strd - MB_SIZE; |
| UWORD32 u4_cur_buf_offset = src_strd - MB_SIZE; |
| |
| for(i = 16; i > 0; i--) |
| { |
| USADA8(pu1_src, pu1_ref1, pi4_sad[0]); |
| USADA8(pu1_src, pu1_ref2, pi4_sad[1]); |
| USADA8(pu1_src, pu1_ref3, pi4_sad[2]); |
| pu1_src += 4; |
| pu1_ref1 += 4; |
| pu1_ref2 += 4; |
| pu1_ref3 += 4; |
| |
| USADA8(pu1_src, pu1_ref1, pi4_sad[0]); |
| USADA8(pu1_src, pu1_ref2, pi4_sad[1]); |
| USADA8(pu1_src, pu1_ref3, pi4_sad[2]); |
| pu1_src += 4; |
| pu1_ref1 += 4; |
| pu1_ref2 += 4; |
| pu1_ref3 += 4; |
| |
| USADA8(pu1_src, pu1_ref1, pi4_sad[0]); |
| USADA8(pu1_src, pu1_ref2, pi4_sad[1]); |
| USADA8(pu1_src, pu1_ref3, pi4_sad[2]); |
| pu1_src += 4; |
| pu1_ref1 += 4; |
| pu1_ref2 += 4; |
| pu1_ref3 += 4; |
| |
| USADA8(pu1_src, pu1_ref1, pi4_sad[0]); |
| USADA8(pu1_src, pu1_ref2, pi4_sad[1]); |
| USADA8(pu1_src, pu1_ref3, pi4_sad[2]); |
| pu1_src += 4; |
| pu1_ref1 += 4; |
| pu1_ref2 += 4; |
| pu1_ref3 += 4; |
| |
| pu1_src += u4_cur_buf_offset; |
| pu1_ref1 += u4_ref_buf_offset; |
| pu1_ref2 += u4_ref_buf_offset; |
| pu1_ref3 += u4_ref_buf_offset; |
| } |
| |
| } |
| |
| /** |
| ******************************************************************************* |
| * |
| * @brief compute sad |
| * |
| * @par Description: This function computes the sad at vertices of diamond grid |
| * centered at reference pointer and at unit distance from it. |
| * |
| * @param[in] pu1_ref1, pu1_ref2 |
| * UWORD8 pointer to the reference |
| * |
| * @param[out] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[in] ref_strd |
| * integer reference stride |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[out] pi4_sad |
| * pointer to integer array evaluated sad |
| * |
| * @returns sad at all evaluated vertexes |
| * |
| * @remarks none |
| * |
| ******************************************************************************* |
| */ |
| void ime_calculate_sad2_prog(UWORD8 *pu1_ref1, |
| UWORD8 *pu1_ref2, |
| UWORD8 *pu1_src, |
| WORD32 ref_strd, |
| WORD32 src_strd, |
| WORD32 *pi4_sad) |
| { |
| /* temp var */ |
| WORD32 i; |
| UWORD32 u4_ref_buf_offset = ref_strd - MB_SIZE; |
| UWORD32 u4_cur_buf_offset = src_strd - MB_SIZE; |
| |
| for(i = 16; i > 0; i--) |
| { |
| USADA8(pu1_src, pu1_ref1, pi4_sad[0]); |
| USADA8(pu1_src, pu1_ref2, pi4_sad[1]); |
| pu1_src += 4; |
| pu1_ref1 += 4; |
| pu1_ref2 += 4; |
| |
| USADA8(pu1_src, pu1_ref1, pi4_sad[0]); |
| USADA8(pu1_src, pu1_ref2, pi4_sad[1]); |
| pu1_src += 4; |
| pu1_ref1 += 4; |
| pu1_ref2 += 4; |
| |
| USADA8(pu1_src, pu1_ref1, pi4_sad[0]); |
| USADA8(pu1_src, pu1_ref2, pi4_sad[1]); |
| pu1_src += 4; |
| pu1_ref1 += 4; |
| pu1_ref2 += 4; |
| |
| USADA8(pu1_src, pu1_ref1, pi4_sad[0]); |
| USADA8(pu1_src, pu1_ref2, pi4_sad[1]); |
| pu1_src += 4; |
| pu1_ref1 += 4; |
| pu1_ref2 += 4; |
| |
| pu1_src += u4_cur_buf_offset; |
| pu1_ref1 += u4_ref_buf_offset; |
| pu1_ref2 += u4_ref_buf_offset; |
| } |
| |
| } |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD) between 2 16x16 blocks |
| * |
| * @par Description |
| * This functions computes SAD between 2 16x16 blocks. There is a provision |
| * for early exit if the up-to computed SAD exceeds maximum allowed SAD. To |
| * compute the distortion of the entire block set u4_max_sad to USHRT_MAX. |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] i4_max_sad |
| * integer maximum allowed distortion |
| * |
| * @param[out] pi4_mb_distortion |
| * integer evaluated sad |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| void ime_compute_sad_16x16(UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| WORD32 i4_max_sad, |
| WORD32 *pi4_mb_distortion) |
| { |
| WORD32 i4_sad = 0; |
| UWORD32 u4_src_offset = src_strd - 16; |
| UWORD32 u4_est_offset = est_strd - 16; |
| UWORD32 i; |
| |
| GATHER_16x16_SAD_EE_STATS(gu4_16x16_sad_ee_stats, 16); |
| |
| for(i = 16; i > 0; i--) |
| { |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += 4; |
| pu1_est += 4; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += 4; |
| pu1_est += 4; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += 4; |
| pu1_est += 4; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += 4; |
| pu1_est += 4; |
| |
| /* early exit */ |
| if(i4_max_sad < i4_sad) |
| { |
| |
| GATHER_16x16_SAD_EE_STATS(gu4_16x16_sad_ee_stats, 16-i); |
| |
| *pi4_mb_distortion = i4_sad; |
| return ; |
| } |
| pu1_src += u4_src_offset; |
| pu1_est += u4_est_offset; |
| } |
| |
| *pi4_mb_distortion = i4_sad; |
| return ; |
| } |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD) between 2 16x16 blocks (fast mode) |
| * |
| * @par Description |
| * This functions computes SAD between 2 16x16 blocks. There is a provision |
| * for early exit if the up-to computed SAD exceeds maximum allowed SAD. To |
| * compute the distortion of the entire block set u4_max_sad to USHRT_MAX. |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] i4_max_sad |
| * integer maximum allowed distortion |
| * |
| * @param[out] pi4_mb_distortion |
| * integer evaluated sad |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| void ime_compute_sad_16x16_fast(UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| WORD32 i4_max_sad, |
| WORD32 *pi4_mb_distortion) |
| { |
| |
| WORD32 i4_sad = 0; |
| UWORD32 u4_src_offset = 2 * src_strd - 16; |
| UWORD32 u4_est_offset = 2 * est_strd - 16; |
| UWORD32 i; |
| |
| UNUSED(i4_max_sad); |
| |
| for(i = 16; i > 0; i-= 2) |
| { |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += 4; |
| pu1_est += 4; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += 4; |
| pu1_est += 4; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += 4; |
| pu1_est += 4; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += 4; |
| pu1_est += 4; |
| |
| pu1_src += u4_src_offset; |
| pu1_est += u4_est_offset; |
| } |
| |
| *pi4_mb_distortion = (i4_sad << 1); |
| return ; |
| } |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD) between 2 8x8 blocks |
| * |
| * @par Description |
| * This functions computes SAD between 2 8x8 blocks. There is a provision |
| * for early exit if the up-to computed SAD exceeds maximum allowed SAD. To |
| * compute the distortion of the entire block set u4_max_sad to USHRT_MAX. |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] u4_max_sad |
| * integer maximum allowed distortion |
| * |
| * @param[out] i4_sad |
| * integer evaluated sad |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| |
| void ime_compute_sad_8x8(UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| WORD32 i4_max_sad, |
| WORD32 *pi4_mb_distortion) |
| { |
| WORD32 i4_sad = 0; |
| UWORD32 u4_src_offset = src_strd - 8; |
| UWORD32 u4_est_offset = est_strd - 8; |
| UWORD32 i, j; |
| WORD16 temp; |
| |
| for(i = 8; i > 0; i--) |
| { |
| for(j = 8; j > 0; j--) |
| { |
| /* SAD */ |
| temp = *pu1_src++ - *pu1_est++; |
| i4_sad += ABS(temp); |
| } |
| /* early exit */ |
| if(i4_max_sad < i4_sad) |
| { |
| *pi4_mb_distortion = i4_sad; |
| return; |
| } |
| pu1_src += u4_src_offset; |
| pu1_est += u4_est_offset; |
| } |
| *pi4_mb_distortion = i4_sad; |
| } |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD) between 2 4x4 blocks |
| * |
| * @par Description |
| * This functions computes SAD between 2 4x4 blocks. There is a provision |
| * for early exit if the up-to computed SAD exceeds maximum allowed SAD. To |
| * compute the distortion of the entire block set u4_max_sad to USHRT_MAX. |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] u4_max_sad |
| * integer maximum allowed distortion |
| * |
| * @param[out] pi4_mb_distortion |
| * integer evaluated sad |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| void ime_compute_sad_4x4 |
| ( |
| UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| WORD32 i4_max_sad, |
| WORD32 *pi4_mb_distortion |
| ) |
| { |
| WORD32 i4_sad = 0; |
| |
| UNUSED(i4_max_sad); |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += src_strd; |
| pu1_est += est_strd; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += src_strd; |
| pu1_est += est_strd; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| pu1_src += src_strd; |
| pu1_est += est_strd; |
| |
| USADA8(pu1_src, pu1_est, i4_sad); |
| *pi4_mb_distortion = i4_sad; |
| } |
| |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD) between 2 16x8 blocks |
| * |
| * |
| * @par Description |
| * This functions computes SAD between 2 16x8 blocks. There is a provision |
| * for early exit if the up-to computed SAD exceeds maximum allowed SAD. To |
| * compute the distortion of the entire block set u4_max_sad to USHRT_MAX. |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] u4_max_sad |
| * integer maximum allowed distortion |
| * |
| * @param[out] pi4_mb_distortion |
| * integer evaluated sad |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| void ime_compute_sad_16x8 |
| ( |
| UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| WORD32 i4_max_sad, |
| WORD32 *pi4_mb_distortion |
| ) |
| { |
| WORD32 i4_sad = 0; |
| UWORD32 u4_src_offset = src_strd - 16; |
| UWORD32 u4_est_offset = est_strd - 16; |
| UWORD32 i, j; |
| WORD16 temp; |
| |
| GATHER_16x8_SAD_EE_STATS(gu4_16x8_sad_ee_stats, 8); |
| |
| for(i = 8; i > 0; i--) |
| { |
| for(j = 16; j > 0; j--) |
| { |
| /* SAD */ |
| temp = *pu1_src++ - *pu1_est++; |
| i4_sad += ABS(temp); |
| } |
| /* early exit */ |
| if(i4_max_sad < i4_sad) |
| { |
| |
| GATHER_16x8_SAD_EE_STATS(gu4_16x8_sad_ee_stats, 8-i); |
| |
| *pi4_mb_distortion = i4_sad; |
| |
| return; |
| } |
| pu1_src += u4_src_offset; |
| pu1_est += u4_est_offset; |
| } |
| |
| *pi4_mb_distortion = i4_sad; |
| return; |
| |
| } |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD) between 2 16x16 blocks |
| * |
| * @par Description |
| * This functions computes SAD between 2 16x16 blocks. There is a provision |
| * for early exit if the up-to computed SAD exceeds maximum allowed SAD. To |
| * compute the distortion of the entire block set u4_max_sad to USHRT_MAX. |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] i4_max_sad |
| * integer maximum allowed distortion |
| * |
| * @param[out] pi4_mb_distortion |
| * integer evaluated sad |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| void ime_compute_sad_16x16_ea8(UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| WORD32 i4_max_sad, |
| WORD32 *pi4_mb_distortion) |
| { |
| WORD32 i4_sad = 0; |
| UWORD32 u4_src_offset = src_strd - 16; |
| UWORD32 u4_est_offset = est_strd - 16; |
| UWORD32 i, j; |
| WORD16 temp; |
| UWORD8 *pu1_src_temp = pu1_src + src_strd; |
| UWORD8 *pu1_est_temp = pu1_est + est_strd; |
| |
| for(i = 16; i > 0; i -= 2) |
| { |
| for(j = 16; j > 0; j--) |
| { |
| /* SAD */ |
| temp = *pu1_src++ - *pu1_est++; |
| i4_sad += ABS(temp); |
| } |
| |
| pu1_src += (u4_src_offset + src_strd); |
| pu1_est += (u4_est_offset + est_strd); |
| |
| } |
| |
| /* early exit */ |
| if(i4_max_sad < i4_sad) |
| { |
| *pi4_mb_distortion = i4_sad; |
| return; |
| } |
| |
| pu1_src = pu1_src_temp; |
| pu1_est = pu1_est_temp; |
| |
| for(i = 16; i > 0; i -= 2) |
| { |
| for(j = 16; j > 0; j--) |
| { |
| /* SAD */ |
| temp = *pu1_src++ - *pu1_est++; |
| i4_sad += ABS(temp); |
| } |
| |
| pu1_src += u4_src_offset + src_strd; |
| pu1_est += u4_est_offset + est_strd; |
| } |
| |
| *pi4_mb_distortion = i4_sad; |
| return; |
| } |
| |
| |
| /** |
| ******************************************************************************* |
| * |
| * @brief This function computes SAD between two 16x16 blocks |
| * It also computes if the block will be zero after H264 transform and quant for |
| * Intra 16x16 blocks |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] pu2_thrsh |
| * Threshold for each element of transofrmed quantized block |
| * |
| * @param[out] pi4_mb_distortion |
| * integer evaluated sad |
| * |
| * @param[out] pu4_is_zero |
| * Poitner to store if the block is zero after transform and quantization |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| void ime_compute_satqd_16x16_lumainter(UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| UWORD16 *pu2_thrsh, |
| WORD32 *pi4_mb_distortion, |
| UWORD32 *pu4_is_non_zero) |
| { |
| UWORD32 i,j; |
| WORD16 s1,s2,s3,s4,sad_1,sad_2,ls1,ls2,ls3,ls4,ls5,ls6,ls7,ls8; |
| UWORD8 *pu1_src_lp,*pu1_est_lp; |
| UWORD32 sad = 0; |
| |
| (*pi4_mb_distortion) = 0; |
| for(i=0;i<4;i++) |
| { |
| for(j=0;j<4;j++) |
| { |
| pu1_src_lp = pu1_src + 4*j; |
| pu1_est_lp = pu1_est + 4*j; |
| |
| s1 = ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[3] - (WORD16)pu1_est_lp[3]); |
| s4 = ABS((WORD16)pu1_src_lp[1] - (WORD16)pu1_est_lp[1])+ ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s2 = ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[3] - (WORD16)pu1_est_lp[3]); |
| s3 = ABS((WORD16)pu1_src_lp[1] - (WORD16)pu1_est_lp[1])+ ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s2 += ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[3] - (WORD16)pu1_est_lp[3]); |
| s3 += ABS((WORD16)pu1_src_lp[1] - (WORD16)pu1_est_lp[1])+ ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s1 += ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[3] - (WORD16)pu1_est_lp[3]); |
| s4 += ABS((WORD16)pu1_src_lp[1] - (WORD16)pu1_est_lp[1])+ ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2]); |
| |
| sad_1 = s1+s2+s3+s4; |
| |
| if(sad == 0) |
| { |
| sad_2 = sad_1<<1; |
| |
| ls1 = sad_2 -(s2 + s3); |
| ls2 = sad_2 -(s1 + s4); |
| ls3 = sad_2 -(s3 + s4); |
| ls4 = sad_2 -(s3 - (s1<<1)); |
| ls5 = sad_2 -(s4 - (s2<<1)); |
| ls6 = sad_2 -(s1 + s2); |
| ls7 = sad_2 -(s2 - (s4<<1)); |
| ls8 = sad_2 -(s1 - (s3<<1)); |
| |
| if( |
| pu2_thrsh[8] <= sad_1 || |
| pu2_thrsh[0] <= ls2 || |
| pu2_thrsh[1] <= ls1 || |
| pu2_thrsh[2] <= ls8 || |
| pu2_thrsh[3] <= ls5 || |
| |
| pu2_thrsh[4] <= ls6 || |
| pu2_thrsh[5] <= ls3 || |
| pu2_thrsh[6] <= ls7 || |
| pu2_thrsh[7] <= ls4 |
| |
| )sad = 1; |
| } |
| (*pi4_mb_distortion) += sad_1; |
| } |
| pu1_src += (src_strd *4); |
| pu1_est += (est_strd *4); |
| } |
| *pu4_is_non_zero = sad; |
| } |
| |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD and SAQTD) between 2 16x8 (interleaved) chroma blocks |
| * |
| * |
| * @par Description |
| * This functions computes SAD between2 16x8 chroma blocks(interleaved) |
| * It also checks if the SATDD(Sum of absolute transformed wuqntized differnce beteern the blocks |
| * If SAQTD is zero, it gives back zero |
| * Other wise sad is retrned |
| * There is no provison for early exit |
| * |
| * The transform done here is the transform for chroma blocks in H264 |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] pu2_thrsh |
| * Threshold for each element of transofrmed quantized block |
| * |
| * @param[out] pi4_mb_distortion |
| * integer evaluated sad |
| * |
| * @remarks |
| * Fucntion code is nit updated. |
| * Will require debugging and minor modifications |
| * |
| ****************************************************************************** |
| */ |
| void ime_compute_satqd_8x16_chroma(UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| WORD32 max_sad, |
| UWORD16 *thrsh) |
| { |
| WORD32 i,j,plane; |
| WORD16 s1,s2,s3,s4,sad_1,sad_2,ls1,ls2,ls3,ls4,ls5,ls6,ls7,ls8; |
| UWORD8 *pu1_src_lp,*pu1_est_lp,*pu1_src_plane,*pu1_est_plane; |
| WORD32 sad =0; |
| UNUSED(max_sad); |
| |
| pu1_src_plane = pu1_src; |
| pu1_est_plane = pu1_est; |
| |
| for(plane =0;plane<2;plane++) |
| { |
| for(i=0;i<4;i++) |
| { |
| for(j=0;j<4;j++) |
| { |
| pu1_src_lp = pu1_src + 8*j; |
| pu1_est_lp = pu1_est + 8*j; |
| |
| s1 = ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[6] - (WORD16)pu1_est_lp[6]); |
| s4 = ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2])+ ABS((WORD16)pu1_src_lp[4] - (WORD16)pu1_est_lp[4]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s2 = ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[6] - (WORD16)pu1_est_lp[6]); |
| s3 = ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2])+ ABS((WORD16)pu1_src_lp[4] - (WORD16)pu1_est_lp[4]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s2 += ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[6] - (WORD16)pu1_est_lp[6]); |
| s3 += ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2])+ ABS((WORD16)pu1_src_lp[4] - (WORD16)pu1_est_lp[4]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s1 += ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[6] - (WORD16)pu1_est_lp[6]); |
| s4 += ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2])+ ABS((WORD16)pu1_src_lp[4] - (WORD16)pu1_est_lp[4]); |
| |
| sad_1 = s1+s2+s3+s4; |
| sad_2 = sad_1<<1; |
| |
| ls1 = sad_2 -(s2 + s3); |
| ls2 = sad_2 -(s1 + s4); |
| ls3 = sad_2 -(s3 + s4); |
| ls4 = sad_2 -(s3 - (s1<<1)); |
| ls5 = sad_2 -(s4 - (s2<<1)); |
| ls6 = sad_2 -(s1 + s2); |
| ls7 = sad_2 -(s2 - (s4<<1)); |
| ls8 = sad_2 -(s1 - (s3<<1)); |
| |
| if( |
| //thrsh[0] > sad_1 && Chroma Dc is checked later |
| thrsh[1] > ls1 && |
| thrsh[2] > sad_1 && |
| thrsh[3] > ls2 && |
| |
| thrsh[4] > ls3 && |
| thrsh[5] > ls4 && |
| thrsh[6] > ls3 && |
| thrsh[7] > ls5 && |
| |
| thrsh[8] > sad_1 && |
| thrsh[9] > ls1 && |
| thrsh[10]> sad_1 && |
| thrsh[11]> ls2 && |
| |
| thrsh[12]> ls6 && |
| thrsh[13]> ls7 && |
| thrsh[14]> ls6 && |
| thrsh[15]> ls8 |
| ) |
| { |
| /*set current sad to be zero*/ |
| } |
| else |
| return ; |
| |
| sad += sad_1; |
| } |
| pu1_src += (src_strd *4); |
| pu1_est += (est_strd *4); |
| } |
| if(sad < (thrsh[0]<<1))sad = 0; |
| else return ; |
| |
| pu1_src = pu1_src_plane+1; |
| pu1_est = pu1_est_plane+1; |
| } |
| return ; |
| } |
| |
| |
| /** |
| ****************************************************************************** |
| * |
| * @brief computes distortion (SAD and SAQTD) between 2 16x16 blocks |
| * |
| * @par Description |
| * This functions computes SAD between 2 16x16 blocks. |
| * It also checks if the SATDD(Sum of absolute transformed wuqntized differnce beteern the blocks |
| * If SAQTD is zero, it gives back zero |
| * Other wise sad is retrned |
| * There is no provison for early exit |
| * |
| * The transform done here is the transform for inter 16x16 blocks in H264 |
| * |
| * @param[in] pu1_src |
| * UWORD8 pointer to the source |
| * |
| * @param[out] pu1_dst |
| * UWORD8 pointer to the destination |
| * |
| * @param[in] src_strd |
| * integer source stride |
| * |
| * @param[in] dst_strd |
| * integer destination stride |
| * |
| * @param[in] pu2_thrsh |
| * Threshold for each element of transofrmed quantized block |
| * |
| * @param[out] pi4_mb_distortion |
| * integer evaluated sad |
| * |
| * @remarks |
| * |
| ****************************************************************************** |
| */ |
| void ime_compute_satqd_16x16_lumaintra(UWORD8 *pu1_src, |
| UWORD8 *pu1_est, |
| WORD32 src_strd, |
| WORD32 est_strd, |
| WORD32 max_sad, |
| UWORD16 *thrsh, |
| WORD32 *pi4_mb_distortion, |
| UWORD8 *sig_nz_sad) |
| { |
| UWORD32 i,j; |
| WORD16 s1[4],s2[4],s3[4],s4[4],sad[4]; |
| UWORD8 *pu1_src_lp,*pu1_est_lp; |
| UWORD8 *sig_sad_dc; |
| UWORD32 nz_sad_sig = 0; |
| UNUSED(max_sad); |
| *pi4_mb_distortion =0; |
| |
| sig_sad_dc = sig_nz_sad; |
| sig_nz_sad++; |
| |
| for(i=0;i<4;i++) |
| { |
| for(j=0;j<4;j++) |
| { |
| pu1_src_lp = pu1_src + 4*j; |
| pu1_est_lp = pu1_est + 4*j; |
| |
| s1[j] = ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[3] - (WORD16)pu1_est_lp[3]); |
| s4[j] = ABS((WORD16)pu1_src_lp[1] - (WORD16)pu1_est_lp[1])+ ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s2[j] = ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[3] - (WORD16)pu1_est_lp[3]); |
| s3[j] = ABS((WORD16)pu1_src_lp[1] - (WORD16)pu1_est_lp[1])+ ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s2[j] += ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[3] - (WORD16)pu1_est_lp[3]); |
| s3[j] += ABS((WORD16)pu1_src_lp[1] - (WORD16)pu1_est_lp[1])+ ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2]); |
| |
| pu1_src_lp += src_strd; |
| pu1_est_lp += est_strd; |
| |
| s1[j] += ABS((WORD16)pu1_src_lp[0] - (WORD16)pu1_est_lp[0])+ ABS((WORD16)pu1_src_lp[3] - (WORD16)pu1_est_lp[3]); |
| s4[j] += ABS((WORD16)pu1_src_lp[1] - (WORD16)pu1_est_lp[1])+ ABS((WORD16)pu1_src_lp[2] - (WORD16)pu1_est_lp[2]); |
| |
| sad[j] = ((s1[j]+s2[j]+s3[j]+s4[j])<<1); |
| } |
| |
| for(j=0;j<4;j++) |
| { |
| |
| if( |
| //thrsh[0] > (sad[j] >> 1) &&Dc goes in the other part |
| thrsh[1] > (sad[j] -(s2[j] + s3[j])) && |
| thrsh[2] > (sad[j]>>1) && |
| thrsh[3] > (sad[j] -(s1[j] + s4[j])) && |
| |
| thrsh[4] > (sad[j] -(s3[j] + s4[j])) && |
| thrsh[5] > (sad[j] -(s3[j] - (s1[j]<<1))) && |
| thrsh[6] > (sad[j] -(s3[j] + s4[j])) && |
| thrsh[7] > (sad[j] -(s4[j] - (s2[j]<<1))) && |
| |
| thrsh[8] > (sad[j]>>1) && |
| thrsh[9] > (sad[j] -(s2[j] + s3[j])) && |
| thrsh[10]> (sad[j]>>1) && |
| thrsh[11]> (sad[j] -(s1[j] + s4[j])) && |
| |
| thrsh[12]> (sad[j] -(s1[j] + s2[j])) && |
| thrsh[13]> (sad[j] -(s2[j] - (s4[j]<<1))) && |
| thrsh[14]> (sad[j] -(s1[j] + s2[j])) && |
| thrsh[15]> (sad[j] -(s1[j] - (s3[j]<<1))) |
| ) |
| { |
| //sad[j] = 0; /*set current sad to be zero*/ |
| sig_nz_sad[j] = 0;/*Signal that the sad is zero*/ |
| } |
| else |
| { |
| sig_nz_sad[j] = 1;/*signal that sad is non zero*/ |
| nz_sad_sig = 1; |
| } |
| |
| (*pi4_mb_distortion) += (sad[j]>>1); |
| //if((*pi4_mb_distortion) >= max_sad)return; /*return or some thing*/ |
| } |
| |
| sig_nz_sad += 4; |
| pu1_src += (src_strd *4); |
| pu1_est += (est_strd *4); |
| } |
| |
| if((*pi4_mb_distortion) < thrsh[0]<<2) |
| { |
| *sig_sad_dc = 0; |
| if(nz_sad_sig == 0)(*pi4_mb_distortion) = 0; |
| } |
| else *sig_sad_dc = 1; |
| } |
| |
| |