/****************************************************************************** * * Copyright (C) 2022 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 * isvcd_residual_resamp_neonintr.c * * @brief * Contains routines that resample for SVC resampling * * @author * Kishore * * @par List of Functions: * - isvcd_pred_residual_recon_4x4_neonintr() * - isvcd_pred_residual_recon_8x8_neonintr() * - isvcd_pred_residual_recon_16x16_neonintr() * - isvcd_pred_residual_recon_chroma_4x4_neonintr() * - isvcd_pred_residual_recon_chroma_8x8_neonintr() * - isvcd_residual_luma_4x4_neonintr() * - isvcd_residual_luma_8x8_neonintr() * - isvcd_residual_luma_16x16_neonintr() * - isvcd_residual_chroma_cb_cr_8x8_neonintr() * * @remarks * ******************************************************************************* */ /*! ************************************************************************** * \file isvcd_residual_resamp_neonintr.c * * \brief * Contains routines that resample for SVC resampling * * Detailed_description * * \date * * * \author : kishore ************************************************************************** */ #include #include #include #include "ih264_typedefs.h" #include "ih264_macros.h" #include "ih264_platform_macros.h" #include "isvcd_structs.h" #include "ih264_debug.h" /*****************************************************************************/ /* */ /* Function Name : isvcd_residual_luma_dyadic_neonintr */ /* */ /* Description : this fucntion does the upsampling of luma residuals for */ /* Dyadic cases */ /* */ /* Inputs : pv_residual_samp_ctxt : Residual upsampling context */ /* pu1_inp_data : input 8 bit data pointer */ /* i4_inp_data_stride : input buffer stride */ /* pi2_out_res : output 16 bit buffer pointer */ /* i4_out_res_stride : Output buffer stride */ /* pu1_inp_bitmap : input packed sign bit data pointer */ /* i4_inp_bitmap_stride : sign bit buffer stride */ /* ps_ref_mb_mode : reference mb mode pointer of base layer */ /* ps_coord : mb co-ordinate pointer */ /* Globals : none */ /* Processing : it does the upsampling with fixed phase values and */ /* reference layer transform size */ /* Outputs : Upsampled residuals for luma */ /* Returns : none */ /* */ /* Issues : none */ /* */ /* Revision History: */ /* */ /* DD MM YYYY Author(s) Changes (Describe the changes made) */ /* 26 05 2021 Dolan creation */ /* */ /*****************************************************************************/ void isvcd_residual_luma_dyadic_neonintr(void *pv_residual_samp_ctxt, WORD16 *pi2_inp_data, WORD32 i4_inp_data_stride, WORD16 *pi2_out_res, WORD32 i4_out_res_stride, mem_element_t *ps_ref_mb_mode, UWORD16 u2_mb_x, UWORD16 u2_mb_y, WORD32 i4_ref_nnz, WORD32 i4_ref_tx_size) { WORD16 *pi2_refarray_buffer; WORD32 i4_blk_ctr; residual_sampling_ctxt_t *ps_ctxt; UNUSED(ps_ref_mb_mode); UNUSED(u2_mb_x); UNUSED(u2_mb_y); ps_ctxt = (residual_sampling_ctxt_t *) pv_residual_samp_ctxt; pi2_refarray_buffer = ps_ctxt->pi2_refarray_buffer; /* based on transform size the counter and interpolation width and */ /* height are intialised as follows */ if((i4_ref_tx_size) && (0 != i4_ref_nnz)) { WORD16 *pi2_ref_data_byte; WORD32 *pi4_ref_array; WORD32 i4_i, i4_j; /* ----------- Horizontal Interpolation ---------------- */ int16x8_t i2_coeff_add_16x8_r0; int16x8_t i2_coeff_16x8_r0_0, i2_coeff_16x8_r0_1; int16x8_t i2_coeff_16x8_sl_r0_0, i2_coeff_16x8_sl_r0_1; int16x8_t result_16x8_r0_0, result_16x8_r0_1; int16x8_t final_result_16x8_r0_0, final_result_16x8_r0_1; int16x8_t i2_coeff_add_16x8_r1; int16x8_t i2_coeff_16x8_r1_0, i2_coeff_16x8_r1_1; int16x8_t i2_coeff_16x8_sl_r1_0, i2_coeff_16x8_sl_r1_1; int16x8_t result_16x8_r1_0, result_16x8_r1_1; int16x8_t final_result_16x8_r1_0, final_result_16x8_r1_1; int16x8x2_t result_16x8x2_t_0; pi2_ref_data_byte = pi2_inp_data; /* ----------- Horizontal Interpolation ---------------- */ pi4_ref_array = (WORD32 *) pi2_refarray_buffer; for(i4_i = 0; i4_i < BLOCK_HEIGHT; i4_i += 2) { i2_coeff_16x8_r0_0 = vld1q_s16(pi2_ref_data_byte); i2_coeff_16x8_r0_1 = vld1q_s16((pi2_ref_data_byte + 1)); i2_coeff_16x8_r1_0 = vld1q_s16(pi2_ref_data_byte + i4_inp_data_stride); i2_coeff_16x8_r1_1 = vld1q_s16((pi2_ref_data_byte + i4_inp_data_stride + 1)); i2_coeff_add_16x8_r0 = vaddq_s16(i2_coeff_16x8_r0_0, i2_coeff_16x8_r0_1); i2_coeff_16x8_sl_r0_0 = vshlq_n_s16(i2_coeff_16x8_r0_0, 1); i2_coeff_16x8_sl_r0_1 = vshlq_n_s16(i2_coeff_16x8_r0_1, 1); i2_coeff_add_16x8_r1 = vaddq_s16(i2_coeff_16x8_r1_0, i2_coeff_16x8_r1_1); i2_coeff_16x8_sl_r1_0 = vshlq_n_s16(i2_coeff_16x8_r1_0, 1); i2_coeff_16x8_sl_r1_1 = vshlq_n_s16(i2_coeff_16x8_r1_1, 1); result_16x8_r0_0 = vaddq_s16(i2_coeff_16x8_sl_r0_0, i2_coeff_add_16x8_r0); result_16x8_r0_1 = vaddq_s16(i2_coeff_16x8_sl_r0_1, i2_coeff_add_16x8_r0); result_16x8_r1_0 = vaddq_s16(i2_coeff_16x8_sl_r1_0, i2_coeff_add_16x8_r1); result_16x8_r1_1 = vaddq_s16(i2_coeff_16x8_sl_r1_1, i2_coeff_add_16x8_r1); result_16x8x2_t_0 = vzipq_s16(result_16x8_r0_0, result_16x8_r0_1); final_result_16x8_r0_0 = result_16x8x2_t_0.val[0]; final_result_16x8_r0_1 = result_16x8x2_t_0.val[1]; result_16x8x2_t_0 = vzipq_s16(result_16x8_r1_0, result_16x8_r1_1); final_result_16x8_r1_0 = result_16x8x2_t_0.val[0]; final_result_16x8_r1_1 = result_16x8x2_t_0.val[1]; vst1q_s32(pi4_ref_array + 1, vmovl_s16(vget_low_s16(final_result_16x8_r0_0))); vst1q_s32(pi4_ref_array + 5, vmovl_s16(vget_high_s16(final_result_16x8_r0_0))); vst1q_s32(pi4_ref_array + 9, vmovl_s16(vget_low_s16(final_result_16x8_r0_1))); vst1q_s32(pi4_ref_array + 13, vmovl_s16(vget_high_s16(final_result_16x8_r0_1))); pi4_ref_array[0] = pi2_ref_data_byte[0] << 2; pi4_ref_array[15] = pi2_ref_data_byte[7] << 2; pi4_ref_array += 16; pi2_ref_data_byte += i4_inp_data_stride; vst1q_s32(pi4_ref_array + 1, vmovl_s16(vget_low_s16(final_result_16x8_r1_0))); vst1q_s32(pi4_ref_array + 5, vmovl_s16(vget_high_s16(final_result_16x8_r1_0))); vst1q_s32(pi4_ref_array + 9, vmovl_s16(vget_low_s16(final_result_16x8_r1_1))); vst1q_s32(pi4_ref_array + 13, vmovl_s16(vget_high_s16(final_result_16x8_r1_1))); pi4_ref_array[0] = pi2_ref_data_byte[0] << 2; pi4_ref_array[15] = pi2_ref_data_byte[7] << 2; pi4_ref_array += 16; /* vertical loop uopdates */ pi2_ref_data_byte = pi2_inp_data + ((i4_i + 2) * i4_inp_data_stride); } /* ----------- Vertical Interpolation ---------------- */ pi4_ref_array = (WORD32 *) pi2_refarray_buffer; { WORD32 *pi4_ref_array_temp; WORD16 *pi2_out; int32x4_t i4_horz_samp_32x4_r1_1, i4_horz_samp_32x4_r1_2, i4_horz_samp_32x4_r1_3, i4_horz_samp_32x4_r1_4; int32x4_t i4_horz_samp_32x4_r2_1, i4_horz_samp_32x4_r2_2, i4_horz_samp_32x4_r2_3, i4_horz_samp_32x4_r2_4; int32x4_t i4_horz_res_32x4_r1_1, i4_horz_res_32x4_r1_2, i4_horz_res_32x4_r1_3, i4_horz_res_32x4_r1_4; int32x4_t i4_horz_res_32x4_r2_1, i4_horz_res_32x4_r2_2, i4_horz_res_32x4_r2_3, i4_horz_res_32x4_r2_4; int32x4_t i4_horz_res_32x4_r3_1, i4_horz_res_32x4_r3_2, i4_horz_res_32x4_r3_3, i4_horz_res_32x4_r3_4; int32x4_t horz_add_32x4_r2_1, horz_add_32x4_r2_2, horz_add_32x4_r2_3, horz_add_32x4_r2_4; int16x8_t comb_horz_16x8_1, comb_horz_16x8_2, comb_horz_16x8_3, comb_horz_16x8_4; pi4_ref_array_temp = pi4_ref_array; pi2_out = pi2_out_res; i4_horz_samp_32x4_r1_1 = vld1q_s32(pi4_ref_array_temp); i4_horz_samp_32x4_r1_2 = vld1q_s32(pi4_ref_array_temp + 4); i4_horz_samp_32x4_r1_3 = vld1q_s32(pi4_ref_array_temp + 8); i4_horz_samp_32x4_r1_4 = vld1q_s32(pi4_ref_array_temp + 12); /* populate the first inter sample */ i4_horz_res_32x4_r1_1 = vrshrq_n_s32(i4_horz_samp_32x4_r1_1, 2); i4_horz_res_32x4_r1_2 = vrshrq_n_s32(i4_horz_samp_32x4_r1_2, 2); i4_horz_res_32x4_r1_3 = vrshrq_n_s32(i4_horz_samp_32x4_r1_3, 2); i4_horz_res_32x4_r1_4 = vrshrq_n_s32(i4_horz_samp_32x4_r1_4, 2); comb_horz_16x8_1 = vcombine_s16(vmovn_s32(i4_horz_res_32x4_r1_1), vmovn_s32(i4_horz_res_32x4_r1_2)); comb_horz_16x8_2 = vcombine_s16(vmovn_s32(i4_horz_res_32x4_r1_3), vmovn_s32(i4_horz_res_32x4_r1_4)); vst1q_s16(pi2_out, comb_horz_16x8_1); vst1q_s16(pi2_out + 8, comb_horz_16x8_2); pi2_out += i4_out_res_stride; for(i4_j = 0; i4_j < 14; i4_j += 2) { pi4_ref_array_temp += MB_WIDTH; i4_horz_samp_32x4_r2_1 = vld1q_s32(pi4_ref_array_temp); i4_horz_samp_32x4_r2_2 = vld1q_s32(pi4_ref_array_temp + 4); i4_horz_samp_32x4_r2_3 = vld1q_s32(pi4_ref_array_temp + 8); i4_horz_samp_32x4_r2_4 = vld1q_s32(pi4_ref_array_temp + 12); horz_add_32x4_r2_1 = vaddq_s32(i4_horz_samp_32x4_r1_1, i4_horz_samp_32x4_r2_1); horz_add_32x4_r2_2 = vaddq_s32(i4_horz_samp_32x4_r1_2, i4_horz_samp_32x4_r2_2); horz_add_32x4_r2_3 = vaddq_s32(i4_horz_samp_32x4_r1_3, i4_horz_samp_32x4_r2_3); horz_add_32x4_r2_4 = vaddq_s32(i4_horz_samp_32x4_r1_4, i4_horz_samp_32x4_r2_4); i4_horz_res_32x4_r2_1 = vaddq_s32(vshlq_n_s32(i4_horz_samp_32x4_r1_1, 1), horz_add_32x4_r2_1); i4_horz_res_32x4_r2_2 = vaddq_s32(vshlq_n_s32(i4_horz_samp_32x4_r1_2, 1), horz_add_32x4_r2_2); i4_horz_res_32x4_r2_3 = vaddq_s32(vshlq_n_s32(i4_horz_samp_32x4_r1_3, 1), horz_add_32x4_r2_3); i4_horz_res_32x4_r2_4 = vaddq_s32(vshlq_n_s32(i4_horz_samp_32x4_r1_4, 1), horz_add_32x4_r2_4); i4_horz_res_32x4_r3_1 = vaddq_s32(vshlq_n_s32(i4_horz_samp_32x4_r2_1, 1), horz_add_32x4_r2_1); i4_horz_res_32x4_r3_2 = vaddq_s32(vshlq_n_s32(i4_horz_samp_32x4_r2_2, 1), horz_add_32x4_r2_2); i4_horz_res_32x4_r3_3 = vaddq_s32(vshlq_n_s32(i4_horz_samp_32x4_r2_3, 1), horz_add_32x4_r2_3); i4_horz_res_32x4_r3_4 = vaddq_s32(vshlq_n_s32(i4_horz_samp_32x4_r2_4, 1), horz_add_32x4_r2_4); i4_horz_res_32x4_r2_1 = vrshrq_n_s32(i4_horz_res_32x4_r2_1, 4); i4_horz_res_32x4_r2_2 = vrshrq_n_s32(i4_horz_res_32x4_r2_2, 4); i4_horz_res_32x4_r2_3 = vrshrq_n_s32(i4_horz_res_32x4_r2_3, 4); i4_horz_res_32x4_r2_4 = vrshrq_n_s32(i4_horz_res_32x4_r2_4, 4); i4_horz_res_32x4_r3_1 = vrshrq_n_s32(i4_horz_res_32x4_r3_1, 4); i4_horz_res_32x4_r3_2 = vrshrq_n_s32(i4_horz_res_32x4_r3_2, 4); i4_horz_res_32x4_r3_3 = vrshrq_n_s32(i4_horz_res_32x4_r3_3, 4); i4_horz_res_32x4_r3_4 = vrshrq_n_s32(i4_horz_res_32x4_r3_4, 4); comb_horz_16x8_1 = vcombine_s16(vmovn_s32(i4_horz_res_32x4_r2_1), vmovn_s32(i4_horz_res_32x4_r2_2)); comb_horz_16x8_2 = vcombine_s16(vmovn_s32(i4_horz_res_32x4_r2_3), vmovn_s32(i4_horz_res_32x4_r2_4)); comb_horz_16x8_3 = vcombine_s16(vmovn_s32(i4_horz_res_32x4_r3_1), vmovn_s32(i4_horz_res_32x4_r3_2)); comb_horz_16x8_4 = vcombine_s16(vmovn_s32(i4_horz_res_32x4_r3_3), vmovn_s32(i4_horz_res_32x4_r3_4)); /* populate 2 samples based on current coeffs */ vst1q_s16(pi2_out, comb_horz_16x8_1); vst1q_s16(pi2_out + 8, comb_horz_16x8_2); pi2_out += i4_out_res_stride; vst1q_s16(pi2_out, comb_horz_16x8_3); vst1q_s16(pi2_out + 8, comb_horz_16x8_4); pi2_out += i4_out_res_stride; /* store the coeff 2 to coeff 1 */ /* (used in next iteration) */ i4_horz_samp_32x4_r1_1 = i4_horz_samp_32x4_r2_1; i4_horz_samp_32x4_r1_2 = i4_horz_samp_32x4_r2_2; i4_horz_samp_32x4_r1_3 = i4_horz_samp_32x4_r2_3; i4_horz_samp_32x4_r1_4 = i4_horz_samp_32x4_r2_4; } /* populate the first inter sample */ i4_horz_res_32x4_r1_1 = vrshrq_n_s32(i4_horz_samp_32x4_r1_1, 2); i4_horz_res_32x4_r1_2 = vrshrq_n_s32(i4_horz_samp_32x4_r1_2, 2); i4_horz_res_32x4_r1_3 = vrshrq_n_s32(i4_horz_samp_32x4_r1_3, 2); i4_horz_res_32x4_r1_4 = vrshrq_n_s32(i4_horz_samp_32x4_r1_4, 2); comb_horz_16x8_1 = vcombine_s16(vmovn_s32(i4_horz_res_32x4_r1_1), vmovn_s32(i4_horz_res_32x4_r1_2)); comb_horz_16x8_2 = vcombine_s16(vmovn_s32(i4_horz_res_32x4_r1_3), vmovn_s32(i4_horz_res_32x4_r1_4)); vst1q_s16(pi2_out, comb_horz_16x8_1); vst1q_s16(pi2_out + 8, comb_horz_16x8_2); /* horizontal loop updates */ pi4_ref_array++; pi2_out_res++; } } else { /* ----------------------------------------------------------------- */ /* LOOP over number of blocks */ /* ----------------------------------------------------------------- */ for(i4_blk_ctr = 0; i4_blk_ctr < 4; i4_blk_ctr++) { /* if reference layer is not coded then no processing */ if(0 != (i4_ref_nnz & 0x1)) { int16x8_t i2_coeff1_16x8_r0_0, i2_coeff1_16x8_r0_1; int16x8_t i2_coeff1_16x8_r1_0, i2_coeff1_16x8_r1_1; int16x8_t i2_coeff1_16x8_r2_0, i2_coeff1_16x8_r2_1; int16x8_t i2_coeff1_16x8_r3_0, i2_coeff1_16x8_r3_1; int16x8_t i2_add_16x8_r0_0; int16x8_t i2_add_16x8_r1_0; int16x8_t i2_add_16x8_r2_0; int16x8_t i2_add_16x8_r3_0; int16x8_t i2_res_16x8_r0_0, i2_res_16x8_r0_1; int16x8_t i2_res_16x8_r1_0, i2_res_16x8_r1_1; int16x8_t i2_res_16x8_r2_0, i2_res_16x8_r2_1; int16x8_t i2_res_16x8_r3_0, i2_res_16x8_r3_1; i2_coeff1_16x8_r0_0 = vld1q_s16(pi2_inp_data); i2_coeff1_16x8_r1_0 = vld1q_s16(pi2_inp_data + i4_inp_data_stride); i2_coeff1_16x8_r2_0 = vld1q_s16(pi2_inp_data + (i4_inp_data_stride << 1)); i2_coeff1_16x8_r3_0 = vld1q_s16(pi2_inp_data + (i4_inp_data_stride << 1) + i4_inp_data_stride); i2_coeff1_16x8_r0_1 = vextq_s16(i2_coeff1_16x8_r0_0, i2_coeff1_16x8_r0_0, 1); i2_coeff1_16x8_r1_1 = vextq_s16(i2_coeff1_16x8_r1_0, i2_coeff1_16x8_r1_0, 1); i2_coeff1_16x8_r2_1 = vextq_s16(i2_coeff1_16x8_r2_0, i2_coeff1_16x8_r2_0, 1); i2_coeff1_16x8_r3_1 = vextq_s16(i2_coeff1_16x8_r3_0, i2_coeff1_16x8_r3_0, 1); i2_add_16x8_r0_0 = vaddq_s16(i2_coeff1_16x8_r0_1, i2_coeff1_16x8_r0_0); i2_add_16x8_r1_0 = vaddq_s16(i2_coeff1_16x8_r1_1, i2_coeff1_16x8_r1_0); i2_add_16x8_r2_0 = vaddq_s16(i2_coeff1_16x8_r2_1, i2_coeff1_16x8_r2_0); i2_add_16x8_r3_0 = vaddq_s16(i2_coeff1_16x8_r3_1, i2_coeff1_16x8_r3_0); i2_coeff1_16x8_r0_0 = vshlq_n_s16(i2_coeff1_16x8_r0_0, 1); i2_coeff1_16x8_r1_0 = vshlq_n_s16(i2_coeff1_16x8_r1_0, 1); i2_coeff1_16x8_r2_0 = vshlq_n_s16(i2_coeff1_16x8_r2_0, 1); i2_coeff1_16x8_r3_0 = vshlq_n_s16(i2_coeff1_16x8_r3_0, 1); i2_coeff1_16x8_r0_1 = vshlq_n_s16(i2_coeff1_16x8_r0_1, 1); i2_coeff1_16x8_r1_1 = vshlq_n_s16(i2_coeff1_16x8_r1_1, 1); i2_coeff1_16x8_r2_1 = vshlq_n_s16(i2_coeff1_16x8_r2_1, 1); i2_coeff1_16x8_r3_1 = vshlq_n_s16(i2_coeff1_16x8_r3_1, 1); i2_res_16x8_r0_0 = vaddq_s16(i2_coeff1_16x8_r0_0, i2_add_16x8_r0_0); i2_res_16x8_r1_0 = vaddq_s16(i2_coeff1_16x8_r1_0, i2_add_16x8_r1_0); i2_res_16x8_r2_0 = vaddq_s16(i2_coeff1_16x8_r2_0, i2_add_16x8_r2_0); i2_res_16x8_r3_0 = vaddq_s16(i2_coeff1_16x8_r3_0, i2_add_16x8_r3_0); i2_res_16x8_r0_1 = vaddq_s16(i2_coeff1_16x8_r0_1, i2_add_16x8_r0_0); i2_res_16x8_r1_1 = vaddq_s16(i2_coeff1_16x8_r1_1, i2_add_16x8_r1_0); i2_res_16x8_r2_1 = vaddq_s16(i2_coeff1_16x8_r2_1, i2_add_16x8_r2_0); i2_res_16x8_r3_1 = vaddq_s16(i2_coeff1_16x8_r3_1, i2_add_16x8_r3_0); i2_res_16x8_r0_0 = vzipq_s16(i2_res_16x8_r0_0, i2_res_16x8_r0_1).val[0]; i2_res_16x8_r1_0 = vzipq_s16(i2_res_16x8_r1_0, i2_res_16x8_r1_1).val[0]; i2_res_16x8_r2_0 = vzipq_s16(i2_res_16x8_r2_0, i2_res_16x8_r2_1).val[0]; i2_res_16x8_r3_0 = vzipq_s16(i2_res_16x8_r3_0, i2_res_16x8_r3_1).val[0]; i2_coeff1_16x8_r0_0 = vshlq_n_s16(i2_coeff1_16x8_r0_0, 1); i2_coeff1_16x8_r1_0 = vshlq_n_s16(i2_coeff1_16x8_r1_0, 1); i2_coeff1_16x8_r2_0 = vshlq_n_s16(i2_coeff1_16x8_r2_0, 1); i2_coeff1_16x8_r3_0 = vshlq_n_s16(i2_coeff1_16x8_r3_0, 1); vst1q_s16(pi2_refarray_buffer + 1, i2_res_16x8_r0_0); vst1q_lane_s16(pi2_refarray_buffer, i2_coeff1_16x8_r0_0, 0); vst1q_lane_s16(pi2_refarray_buffer + 7, i2_coeff1_16x8_r0_0, 3); vst1q_s16(pi2_refarray_buffer + 9, i2_res_16x8_r1_0); vst1q_lane_s16(pi2_refarray_buffer + 8, i2_coeff1_16x8_r1_0, 0); vst1q_lane_s16(pi2_refarray_buffer + 15, i2_coeff1_16x8_r1_0, 3); vst1q_s16(pi2_refarray_buffer + 17, i2_res_16x8_r2_0); vst1q_lane_s16(pi2_refarray_buffer + 16, i2_coeff1_16x8_r2_0, 0); vst1q_lane_s16(pi2_refarray_buffer + 23, i2_coeff1_16x8_r2_0, 3); vst1q_s16(pi2_refarray_buffer + 25, i2_res_16x8_r3_0); vst1q_lane_s16(pi2_refarray_buffer + 24, i2_coeff1_16x8_r3_0, 0); vst1q_lane_s16(pi2_refarray_buffer + 31, i2_coeff1_16x8_r3_0, 3); { int16x4_t i4_horz_samp_16x4_r0_1, i4_horz_samp_16x4_r0_2; int16x4_t i4_horz_samp_16x4_r1_1, i4_horz_samp_16x4_r1_2; int16x4_t i4_horz_samp_16x4_r2_1, i4_horz_samp_16x4_r2_2; int16x4_t i4_horz_samp_16x4_r3_1, i4_horz_samp_16x4_r3_2; int32x4_t i4_horz_samp_32x4_r0_1, i4_horz_samp_32x4_r0_2; int32x4_t i4_horz_samp_32x4_r1_1, i4_horz_samp_32x4_r1_2; int32x4_t i4_horz_samp_32x4_r2_1, i4_horz_samp_32x4_r2_2; int32x4_t i4_horz_samp_32x4_r3_1, i4_horz_samp_32x4_r3_2; int32x4_t i4_horz_add_32x4_r1_1, i4_horz_add_32x4_r1_2; int32x4_t i4_horz_add_32x4_r2_1, i4_horz_add_32x4_r2_2; int32x4_t i4_horz_add_32x4_r3_1, i4_horz_add_32x4_r3_2; int16x4_t i4_horz_res_16x4_r0_1, i4_horz_res_16x4_r0_2; int16x4_t i4_horz_res_16x4_r1_1, i4_horz_res_16x4_r1_2; int16x4_t i4_horz_res_16x4_r2_1, i4_horz_res_16x4_r2_2; int16x4_t i4_horz_res_16x4_r3_1, i4_horz_res_16x4_r3_2; int16x4_t i4_horz_res_16x4_r4_1, i4_horz_res_16x4_r4_2; int16x4_t i4_horz_res_16x4_r5_1, i4_horz_res_16x4_r5_2; int16x4_t i4_horz_res_16x4_r6_1, i4_horz_res_16x4_r6_2; int16x4_t i4_horz_res_16x4_r7_1, i4_horz_res_16x4_r7_2; int32x4_t i4_horz_res_32x4_r1_1, i4_horz_res_32x4_r1_2; int32x4_t i4_horz_res_32x4_r2_1, i4_horz_res_32x4_r2_2; int32x4_t i4_horz_res_32x4_r3_1, i4_horz_res_32x4_r3_2; int32x4_t i4_horz_res_32x4_r4_1, i4_horz_res_32x4_r4_2; int32x4_t i4_horz_res_32x4_r5_1, i4_horz_res_32x4_r5_2; int32x4_t i4_horz_res_32x4_r6_1, i4_horz_res_32x4_r6_2; i4_horz_samp_16x4_r0_1 = vld1_s16(pi2_refarray_buffer); i4_horz_samp_16x4_r0_2 = vld1_s16(pi2_refarray_buffer + 4); i4_horz_samp_16x4_r1_1 = vld1_s16(pi2_refarray_buffer + 8); i4_horz_samp_16x4_r1_2 = vld1_s16(pi2_refarray_buffer + 12); i4_horz_samp_16x4_r2_1 = vld1_s16(pi2_refarray_buffer + 16); i4_horz_samp_16x4_r2_2 = vld1_s16(pi2_refarray_buffer + 20); i4_horz_samp_16x4_r3_1 = vld1_s16(pi2_refarray_buffer + 24); i4_horz_samp_16x4_r3_2 = vld1_s16(pi2_refarray_buffer + 28); i4_horz_res_16x4_r0_1 = vrshr_n_s16(i4_horz_samp_16x4_r0_1, 2); i4_horz_res_16x4_r0_2 = vrshr_n_s16(i4_horz_samp_16x4_r0_2, 2); i4_horz_add_32x4_r1_1 = vaddl_s16(i4_horz_samp_16x4_r0_1, i4_horz_samp_16x4_r1_1); i4_horz_add_32x4_r1_2 = vaddl_s16(i4_horz_samp_16x4_r0_2, i4_horz_samp_16x4_r1_2); i4_horz_add_32x4_r2_1 = vaddl_s16(i4_horz_samp_16x4_r1_1, i4_horz_samp_16x4_r2_1); i4_horz_add_32x4_r2_2 = vaddl_s16(i4_horz_samp_16x4_r1_2, i4_horz_samp_16x4_r2_2); i4_horz_add_32x4_r3_1 = vaddl_s16(i4_horz_samp_16x4_r2_1, i4_horz_samp_16x4_r3_1); i4_horz_add_32x4_r3_2 = vaddl_s16(i4_horz_samp_16x4_r2_2, i4_horz_samp_16x4_r3_2); i4_horz_samp_32x4_r0_1 = vshll_n_s16(i4_horz_samp_16x4_r0_1, 1); i4_horz_samp_32x4_r0_2 = vshll_n_s16(i4_horz_samp_16x4_r0_2, 1); i4_horz_samp_32x4_r1_1 = vshll_n_s16(i4_horz_samp_16x4_r1_1, 1); i4_horz_samp_32x4_r1_2 = vshll_n_s16(i4_horz_samp_16x4_r1_2, 1); i4_horz_samp_32x4_r2_1 = vshll_n_s16(i4_horz_samp_16x4_r2_1, 1); i4_horz_samp_32x4_r2_2 = vshll_n_s16(i4_horz_samp_16x4_r2_2, 1); i4_horz_samp_32x4_r3_1 = vshll_n_s16(i4_horz_samp_16x4_r3_1, 1); i4_horz_samp_32x4_r3_2 = vshll_n_s16(i4_horz_samp_16x4_r3_2, 1); i4_horz_res_32x4_r1_1 = vaddq_s32(i4_horz_samp_32x4_r0_1, i4_horz_add_32x4_r1_1); i4_horz_res_32x4_r1_2 = vaddq_s32(i4_horz_samp_32x4_r0_2, i4_horz_add_32x4_r1_2); i4_horz_res_32x4_r2_1 = vaddq_s32(i4_horz_samp_32x4_r1_1, i4_horz_add_32x4_r1_1); i4_horz_res_32x4_r2_2 = vaddq_s32(i4_horz_samp_32x4_r1_2, i4_horz_add_32x4_r1_2); i4_horz_res_32x4_r3_1 = vaddq_s32(i4_horz_samp_32x4_r1_1, i4_horz_add_32x4_r2_1); i4_horz_res_32x4_r3_2 = vaddq_s32(i4_horz_samp_32x4_r1_2, i4_horz_add_32x4_r2_2); i4_horz_res_32x4_r4_1 = vaddq_s32(i4_horz_samp_32x4_r2_1, i4_horz_add_32x4_r2_1); i4_horz_res_32x4_r4_2 = vaddq_s32(i4_horz_samp_32x4_r2_2, i4_horz_add_32x4_r2_2); i4_horz_res_32x4_r5_1 = vaddq_s32(i4_horz_samp_32x4_r2_1, i4_horz_add_32x4_r3_1); i4_horz_res_32x4_r5_2 = vaddq_s32(i4_horz_samp_32x4_r2_2, i4_horz_add_32x4_r3_2); i4_horz_res_32x4_r6_1 = vaddq_s32(i4_horz_samp_32x4_r3_1, i4_horz_add_32x4_r3_1); i4_horz_res_32x4_r6_2 = vaddq_s32(i4_horz_samp_32x4_r3_2, i4_horz_add_32x4_r3_2); i4_horz_res_16x4_r1_1 = vqrshrn_n_s32(i4_horz_res_32x4_r1_1, 4); i4_horz_res_16x4_r1_2 = vqrshrn_n_s32(i4_horz_res_32x4_r1_2, 4); i4_horz_res_16x4_r2_1 = vqrshrn_n_s32(i4_horz_res_32x4_r2_1, 4); i4_horz_res_16x4_r2_2 = vqrshrn_n_s32(i4_horz_res_32x4_r2_2, 4); i4_horz_res_16x4_r3_1 = vqrshrn_n_s32(i4_horz_res_32x4_r3_1, 4); i4_horz_res_16x4_r3_2 = vqrshrn_n_s32(i4_horz_res_32x4_r3_2, 4); i4_horz_res_16x4_r4_1 = vqrshrn_n_s32(i4_horz_res_32x4_r4_1, 4); i4_horz_res_16x4_r4_2 = vqrshrn_n_s32(i4_horz_res_32x4_r4_2, 4); i4_horz_res_16x4_r5_1 = vqrshrn_n_s32(i4_horz_res_32x4_r5_1, 4); i4_horz_res_16x4_r5_2 = vqrshrn_n_s32(i4_horz_res_32x4_r5_2, 4); i4_horz_res_16x4_r6_1 = vqrshrn_n_s32(i4_horz_res_32x4_r6_1, 4); i4_horz_res_16x4_r6_2 = vqrshrn_n_s32(i4_horz_res_32x4_r6_2, 4); i4_horz_res_16x4_r7_1 = vrshr_n_s16(i4_horz_samp_16x4_r3_1, 2); i4_horz_res_16x4_r7_2 = vrshr_n_s16(i4_horz_samp_16x4_r3_2, 2); vst1_s16(pi2_out_res, i4_horz_res_16x4_r0_1); vst1_s16(pi2_out_res + 4, i4_horz_res_16x4_r0_2); vst1_s16(pi2_out_res + i4_out_res_stride, i4_horz_res_16x4_r1_1); vst1_s16(pi2_out_res + i4_out_res_stride + 4, i4_horz_res_16x4_r1_2); vst1_s16(pi2_out_res + (i4_out_res_stride << 1), i4_horz_res_16x4_r2_1); vst1_s16(pi2_out_res + (i4_out_res_stride << 1) + 4, i4_horz_res_16x4_r2_2); vst1_s16(pi2_out_res + (i4_out_res_stride * 3), i4_horz_res_16x4_r3_1); vst1_s16(pi2_out_res + (i4_out_res_stride * 3) + 4, i4_horz_res_16x4_r3_2); vst1_s16(pi2_out_res + (i4_out_res_stride << 2), i4_horz_res_16x4_r4_1); vst1_s16(pi2_out_res + (i4_out_res_stride << 2) + 4, i4_horz_res_16x4_r4_2); vst1_s16(pi2_out_res + (i4_out_res_stride * 5), i4_horz_res_16x4_r5_1); vst1_s16(pi2_out_res + (i4_out_res_stride * 5) + 4, i4_horz_res_16x4_r5_2); vst1_s16(pi2_out_res + (i4_out_res_stride * 6), i4_horz_res_16x4_r6_1); vst1_s16(pi2_out_res + (i4_out_res_stride * 6) + 4, i4_horz_res_16x4_r6_2); vst1_s16(pi2_out_res + (i4_out_res_stride * 7), i4_horz_res_16x4_r7_1); vst1_s16(pi2_out_res + (i4_out_res_stride * 7) + 4, i4_horz_res_16x4_r7_2); pi2_out_res += BLOCK_WIDTH; } } else { pi2_out_res += BLOCK_WIDTH; } /* Block level loop updates */ if(1 == i4_blk_ctr) { pi2_inp_data -= SUB_BLOCK_WIDTH; pi2_inp_data += (i4_inp_data_stride * SUB_BLOCK_HEIGHT); pi2_out_res -= MB_WIDTH; pi2_out_res += (i4_out_res_stride * BLOCK_HEIGHT); i4_ref_nnz >>= 2; } else { pi2_inp_data += SUB_BLOCK_WIDTH; } i4_ref_nnz >>= 1; } /* end of loop over all the blocks */ } return; } /*****************************************************************************/ /* */ /* Function Name : isvcd_interpolate_residual_neonintr */ /* */ /* Description : this fucntion does the upsampling of residuals. */ /* */ /* Inputs : pv_residual_samp_ctxt : Residual upsampling context */ /* pu1_inp_data : input 8 bit data pointer */ /* i4_inp_data_stride : input buffer stride */ /* pi2_out_res : output 16 bit buffer pointer */ /* i4_out_res_stride : Output buffer stride */ /* pu1_inp_bitmap : input packed sign bit data pointer */ /* i4_inp_bitmap_stride : sign bit buffer stride */ /* ps_ref_mb_mode : reference mb mode pointer of base layer */ /* ps_coord : mb co-ordinate pointer */ /* Globals : none */ /* Processing : it does the upsampling with fixed phase values and */ /* reference layer transform size */ /* Outputs : Upsampled residuals. */ /* Returns : none */ /* */ /* Issues : none */ /* */ /* Revision History: */ /* */ /* DD MM YYYY Author(s) Changes (Describe the changes made) */ /* 26 05 2021 Dolan creation */ /* */ /*****************************************************************************/ void isvcd_interpolate_residual_neonintr(void *pv_residual_samp_ctxt, WORD16 *pi2_out, WORD32 i4_out_stride, WORD32 i4_refarray_wd, UWORD16 u2_mb_x, UWORD16 u2_mb_y, WORD32 i4_chroma_flag) { residual_sampling_ctxt_t *ps_ctxt; residual_samp_map_ctxt_t *ps_map_ctxt; res_lyr_ctxt *ps_lyr_ctxt; ref_pixel_map_t *ps_x_pos_phase; ref_pixel_map_t *ps_y_pos_phase; WORD32 i4_x, i4_y; WORD32 i4_frm_mb_x, i4_frm_mb_y; WORD32 i4_temp_array_ht; WORD32 i4_mb_wd; WORD32 i4_mb_ht; WORD16 *pi2_ref_array; UWORD8 *pu1_ref_x_ptr_incr, *pu1_ref_y_ptr_incr; UWORD8 arr_y_ref_pos[16] = {0}; UWORD8 arr_x_ref_pos[16] = {0}; UWORD8 arr_x_ref_pos_low[16] = {0}; UWORD8 arr_x_phase[16] = {0}; UWORD8 arr_y_phase[16] = {0}; UWORD8 *pi1_y_ref_pos; UWORD8 *pi1_x_ref_pos; UWORD8 *pi1_x_ref_pos_low; UWORD8 *pi1_y_phase; UWORD8 *pi1_x_phase; ps_ctxt = (residual_sampling_ctxt_t *) pv_residual_samp_ctxt; ps_lyr_ctxt = &ps_ctxt->as_res_lyrs[ps_ctxt->i4_res_lyr_id]; pi2_ref_array = ps_ctxt->pi2_refarray_buffer; pu1_ref_x_ptr_incr = ps_ctxt->pu1_ref_x_ptr_incr; pu1_ref_y_ptr_incr = ps_ctxt->pu1_ref_y_ptr_incr; if(1 == i4_chroma_flag) ps_map_ctxt = &ps_lyr_ctxt->s_chroma_map_ctxt; else ps_map_ctxt = &ps_lyr_ctxt->s_luma_map_ctxt; i4_mb_wd = MB_WIDTH >> i4_chroma_flag; i4_mb_ht = MB_HEIGHT >> i4_chroma_flag; ps_x_pos_phase = ps_map_ctxt->ps_x_pos_phase; ps_y_pos_phase = ps_map_ctxt->ps_y_pos_phase; i4_temp_array_ht = i4_mb_ht; i4_frm_mb_y = u2_mb_y * i4_mb_ht; i4_frm_mb_x = u2_mb_x * i4_mb_wd; /* --------------------------------------------------------------------- */ /* Loop for interpolation */ /* --------------------------------------------------------------------- */ if(i4_chroma_flag == 0) { int16x8_t ref_arr_16x8_r0_0, ref_arr_16x8_r0_1; int16x8_t ref_arr_16x8_r1_0, ref_arr_16x8_r1_1; uint8x16_t x_ref_pos_mask_r0_0, x_ref_rnd_mask_r0_0; uint8x16_t u1_incr_8x16_r0_0, x_ref_pos_mask_temp_r0_0, u1_incr_not_8x16_r0_0, u1_y_incr_8x16_r0_0, phs_mask_8x16_0; uint8x16_t u1_incr_8x16_r1_0, x_ref_pos_mask_temp_r1_0, u1_incr_not_8x16_r1_0; int16x8_t ref_arr_temp0_16x8_r0_0, res_16x8_r0_0, vert_res_16x8_r0_0; int16x8_t ref_arr_temp0_16x8_r1_0, res_16x8_r1_0, vert_res_16x8_r1_0; int16x8_t ref_arr_temp1_16x8_r0_0; int16x8_t ref_arr_temp1_16x8_r1_0; uint8x16_t x_ref_pos_mask_temp_r0_1, u1_incr_not_8x16_r0_1; uint8x16_t x_ref_pos_mask_temp_r1_1, u1_incr_not_8x16_r1_1; int16x8_t ref_arr_temp0_16x8_r0_1, res_16x8_r0_1, vert_res_16x8_r0_1; int16x8_t ref_arr_temp0_16x8_r1_1, res_16x8_r1_1, vert_res_16x8_r1_1; int16x8_t ref_arr_temp1_16x8_r0_1; int16x8_t ref_arr_temp1_16x8_r1_1; uint16x8_t u1_y_incr_16x8_r0_0, u1_y_incr_16x8_r0_1; uint8x16_t u1_incr_not_8x16_r0_0_even, u1_incr_not_8x16_r1_0_even, x_ref_pos_mask_temp_r0_0_even, x_ref_pos_mask_temp_r1_0_even; uint8x16_t u1_incr_not_8x16_r0_0_odd, u1_incr_not_8x16_r1_0_odd, x_ref_pos_mask_temp_r0_0_odd, x_ref_pos_mask_temp_r1_0_odd; uint8x16x2_t u1_incr_not_8x16_2, x_ref_pos_mask_temp; int16x8_t prev_res_16x8_r0_0; int16x8_t prev_res_16x8_r1_0; int16x8_t prev_res_16x8_r0_1; int16x8_t prev_res_16x8_r1_1; uint8x8x2_t u1_incr_8x8x2_t; uint8x8_t u1_incr_8x8_t0, u1_incr_8x8_t1; uint16x8_t u1_prev_y_incr_16x8_r0_0; uint16x8_t u1_prev_y_incr_16x8_r0_1; WORD32 zero_r0_r1 = 0; int32x4_t res_32x4_l_0, res_32x4_h_0; int32x4_t res_32x4_l_1, res_32x4_h_1; int16x8_t res_16x8_l, res_16x8_h; uint16x8_t phs_mask_16x8_0, phs_mask_16x8_1; int16x8_t const_16_16x8, phs_mask_16min_16x8_0; int16x8_t dup_val_1, dup_val_2, dup_val_3, dup_val_4, dup_val_5, dup_abs; uint8x16_t phs_mask_div8_8x16_0, mid_indx_8x16; int16x8_t phs_mask_16min_16x8_1; uint16x8_t ones = vdupq_n_u16(0xFFFF); uint8x16_t const_ones = vdupq_n_u8(1); uint8x8x2_t u1_temp_8x8x2_t; uint8x8_t u1_temp_8x8_t0, u1_temp_8x8_t1; WORD16 *pi2_ref_array_temp; UWORD8 *pu1_ref_x_ptr_incr_temp, *pu1_ref_y_ptr_incr_temp; WORD32 i4_y_phase; WORD32 out_stride_temp; WORD32 strt_indx_h = 0; for(i4_y = 0; i4_y < (i4_temp_array_ht); i4_y++) { arr_y_phase[i4_y] = (UWORD8) ps_y_pos_phase[i4_y + i4_frm_mb_y].i2_phase; arr_y_ref_pos[i4_y] = (UWORD8) (ps_y_pos_phase[i4_y + i4_frm_mb_y].i2_ref_pos); } pi1_y_ref_pos = arr_y_ref_pos; pi1_y_phase = arr_y_phase; strt_indx_h = (ps_x_pos_phase[8 + i4_frm_mb_x].i2_ref_pos); for(i4_x = 0; i4_x < i4_mb_wd; i4_x++) { arr_x_ref_pos[i4_x] = (WORD8) ps_x_pos_phase[i4_x + i4_frm_mb_x].i2_ref_pos; arr_x_phase[i4_x] = (WORD8) ps_x_pos_phase[i4_x + i4_frm_mb_x].i2_phase; } pi1_x_ref_pos = arr_x_ref_pos; pi1_x_phase = arr_x_phase; phs_mask_8x16_0 = vld1q_u8((pi1_x_phase)); phs_mask_16x8_0 = vmovl_u8(vget_low_u8(phs_mask_8x16_0)); phs_mask_16x8_1 = vmovl_u8(vld1_u8((pi1_x_phase + 8))); x_ref_pos_mask_r0_0 = vld1q_u8((pi1_x_ref_pos)); const_16_16x8 = vdupq_n_s16(16); phs_mask_div8_8x16_0 = vshrq_n_u8(phs_mask_8x16_0, 3); phs_mask_16min_16x8_0 = vsubq_s16(const_16_16x8, vreinterpretq_s16_u16(phs_mask_16x8_0)); phs_mask_16min_16x8_1 = vsubq_s16(const_16_16x8, vreinterpretq_s16_u16(phs_mask_16x8_1)); x_ref_rnd_mask_r0_0 = vaddq_u8(x_ref_pos_mask_r0_0, phs_mask_div8_8x16_0); mid_indx_8x16 = vdupq_n_u8((strt_indx_h << 1)); for(i4_y = 0; i4_y < (i4_temp_array_ht); i4_y++) { if((i4_y > 0) && (pi1_y_ref_pos[i4_y] == pi1_y_ref_pos[i4_y - 1])) { if(!zero_r0_r1) { res_16x8_l = vdupq_n_s16(0); res_16x8_h = vdupq_n_s16(0); out_stride_temp = (i4_y * i4_out_stride); vst1q_s16((pi2_out + out_stride_temp), res_16x8_l); vst1q_s16((pi2_out + out_stride_temp + 8), res_16x8_h); continue; } res_16x8_r0_0 = prev_res_16x8_r0_0; res_16x8_r1_0 = prev_res_16x8_r1_0; res_16x8_r0_1 = prev_res_16x8_r0_1; res_16x8_r1_1 = prev_res_16x8_r1_1; u1_y_incr_16x8_r0_0 = u1_prev_y_incr_16x8_r0_0; u1_y_incr_16x8_r0_1 = u1_prev_y_incr_16x8_r0_1; } else { pi2_ref_array_temp = pi2_ref_array + ((pi1_y_ref_pos[i4_y]) * i4_refarray_wd); pu1_ref_x_ptr_incr_temp = pu1_ref_x_ptr_incr + ((pi1_y_ref_pos[i4_y]) * i4_refarray_wd); ref_arr_16x8_r0_0 = vld1q_s16((pi2_ref_array_temp)); ref_arr_16x8_r1_0 = vld1q_s16((pi2_ref_array_temp + i4_refarray_wd)); ref_arr_16x8_r0_1 = vld1q_s16((pi2_ref_array_temp + strt_indx_h)); ref_arr_16x8_r1_1 = vld1q_s16((pi2_ref_array_temp + i4_refarray_wd + strt_indx_h)); dup_val_1 = vabsq_s16(ref_arr_16x8_r0_0); dup_val_2 = vabsq_s16(ref_arr_16x8_r1_0); dup_val_3 = vabsq_s16(ref_arr_16x8_r0_1); dup_val_4 = vabsq_s16(ref_arr_16x8_r1_1); dup_val_5 = vqaddq_s16(dup_val_1, dup_val_2); dup_val_1 = vqaddq_s16(dup_val_3, dup_val_4); dup_abs = vqaddq_s16(dup_val_1, dup_val_5); zero_r0_r1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || dup_abs[6] || dup_abs[7]; if(zero_r0_r1) { u1_incr_8x16_r0_0 = vld1q_u8((pu1_ref_x_ptr_incr_temp)); u1_incr_8x16_r1_0 = vld1q_u8((pu1_ref_x_ptr_incr_temp + i4_refarray_wd)); u1_incr_8x8x2_t.val[0] = vget_low_u8(u1_incr_8x16_r0_0); u1_incr_8x8x2_t.val[1] = vget_high_u8(u1_incr_8x16_r0_0); u1_incr_8x8_t0 = vtbl2_u8(u1_incr_8x8x2_t, vget_low_u8(x_ref_pos_mask_r0_0)); u1_incr_8x8_t1 = vtbl2_u8(u1_incr_8x8x2_t, vget_high_u8(x_ref_pos_mask_r0_0)); u1_incr_8x16_r0_0 = vcombine_u8(u1_incr_8x8_t0, u1_incr_8x8_t1); u1_incr_8x8x2_t.val[0] = vget_low_u8(u1_incr_8x16_r1_0); u1_incr_8x8x2_t.val[1] = vget_high_u8(u1_incr_8x16_r1_0); u1_incr_8x8_t0 = vtbl2_u8(u1_incr_8x8x2_t, vget_low_u8(x_ref_pos_mask_r0_0)); u1_incr_8x8_t1 = vtbl2_u8(u1_incr_8x8x2_t, vget_high_u8(x_ref_pos_mask_r0_0)); u1_incr_8x16_r1_0 = vcombine_u8(u1_incr_8x8_t0, u1_incr_8x8_t1); u1_incr_not_8x16_r0_0 = vbicq_u8(phs_mask_div8_8x16_0, u1_incr_8x16_r0_0); u1_incr_not_8x16_r1_0 = vbicq_u8(phs_mask_div8_8x16_0, u1_incr_8x16_r1_0); u1_incr_not_8x16_r0_0 = vaddq_u8(u1_incr_not_8x16_r0_0, x_ref_pos_mask_r0_0); u1_incr_not_8x16_r1_0 = vaddq_u8(u1_incr_not_8x16_r1_0, x_ref_pos_mask_r0_0); x_ref_pos_mask_temp_r0_0 = vaddq_u8(u1_incr_not_8x16_r0_0, u1_incr_8x16_r0_0); x_ref_pos_mask_temp_r1_0 = vaddq_u8(u1_incr_not_8x16_r1_0, u1_incr_8x16_r1_0); u1_incr_not_8x16_r0_0_even = vshlq_n_u8(u1_incr_not_8x16_r0_0, 1); u1_incr_not_8x16_r1_0_even = vshlq_n_u8(u1_incr_not_8x16_r1_0, 1); x_ref_pos_mask_temp_r0_0_even = vshlq_n_u8(x_ref_pos_mask_temp_r0_0, 1); x_ref_pos_mask_temp_r1_0_even = vshlq_n_u8(x_ref_pos_mask_temp_r1_0, 1); u1_incr_not_8x16_r0_0_odd = vaddq_u8(u1_incr_not_8x16_r0_0_even, const_ones); u1_incr_not_8x16_r1_0_odd = vaddq_u8(u1_incr_not_8x16_r1_0_even, const_ones); x_ref_pos_mask_temp_r0_0_odd = vaddq_u8(x_ref_pos_mask_temp_r0_0_even, const_ones); x_ref_pos_mask_temp_r1_0_odd = vaddq_u8(x_ref_pos_mask_temp_r1_0_even, const_ones); u1_incr_not_8x16_2 = vzipq_u8(u1_incr_not_8x16_r0_0_even, u1_incr_not_8x16_r0_0_odd); u1_incr_not_8x16_r0_0 = u1_incr_not_8x16_2.val[0]; u1_incr_not_8x16_r0_1 = u1_incr_not_8x16_2.val[1]; u1_incr_not_8x16_2 = vzipq_u8(u1_incr_not_8x16_r1_0_even, u1_incr_not_8x16_r1_0_odd); u1_incr_not_8x16_r1_0 = u1_incr_not_8x16_2.val[0]; u1_incr_not_8x16_r1_1 = u1_incr_not_8x16_2.val[1]; x_ref_pos_mask_temp = vzipq_u8(x_ref_pos_mask_temp_r0_0_even, x_ref_pos_mask_temp_r0_0_odd); x_ref_pos_mask_temp_r0_0 = x_ref_pos_mask_temp.val[0]; x_ref_pos_mask_temp_r0_1 = x_ref_pos_mask_temp.val[1]; x_ref_pos_mask_temp = vzipq_u8(x_ref_pos_mask_temp_r1_0_even, x_ref_pos_mask_temp_r1_0_odd); x_ref_pos_mask_temp_r1_0 = x_ref_pos_mask_temp.val[0]; x_ref_pos_mask_temp_r1_1 = x_ref_pos_mask_temp.val[1]; u1_incr_not_8x16_r0_1 = vsubq_u8(u1_incr_not_8x16_r0_1, mid_indx_8x16); u1_incr_not_8x16_r1_1 = vsubq_u8(u1_incr_not_8x16_r1_1, mid_indx_8x16); x_ref_pos_mask_temp_r0_1 = vsubq_u8(x_ref_pos_mask_temp_r0_1, mid_indx_8x16); x_ref_pos_mask_temp_r1_1 = vsubq_u8(x_ref_pos_mask_temp_r1_1, mid_indx_8x16); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_0))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_0))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(u1_incr_not_8x16_r0_0)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(u1_incr_not_8x16_r0_0)); ref_arr_temp0_16x8_r0_0 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_0))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_0))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(u1_incr_not_8x16_r1_0)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(u1_incr_not_8x16_r1_0)); ref_arr_temp0_16x8_r1_0 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_0))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_0))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(x_ref_pos_mask_temp_r0_0)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(x_ref_pos_mask_temp_r0_0)); ref_arr_temp1_16x8_r0_0 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_0))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_0))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(x_ref_pos_mask_temp_r1_0)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(x_ref_pos_mask_temp_r1_0)); ref_arr_temp1_16x8_r1_0 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_1))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_1))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(u1_incr_not_8x16_r0_1)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(u1_incr_not_8x16_r0_1)); ref_arr_temp0_16x8_r0_1 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_1))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_1))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(u1_incr_not_8x16_r1_1)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(u1_incr_not_8x16_r1_1)); ref_arr_temp0_16x8_r1_1 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_1))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_1))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(x_ref_pos_mask_temp_r0_1)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(x_ref_pos_mask_temp_r0_1)); ref_arr_temp1_16x8_r0_1 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_1))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_1))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(x_ref_pos_mask_temp_r1_1)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(x_ref_pos_mask_temp_r1_1)); ref_arr_temp1_16x8_r1_1 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); res_16x8_r0_0 = vmulq_s16(ref_arr_temp0_16x8_r0_0, phs_mask_16min_16x8_0); res_16x8_r1_0 = vmulq_s16(ref_arr_temp0_16x8_r1_0, phs_mask_16min_16x8_0); res_16x8_r0_0 = vmlaq_s16(res_16x8_r0_0, ref_arr_temp1_16x8_r0_0, vreinterpretq_s16_u16(phs_mask_16x8_0)); res_16x8_r1_0 = vmlaq_s16(res_16x8_r1_0, ref_arr_temp1_16x8_r1_0, vreinterpretq_s16_u16(phs_mask_16x8_0)); res_16x8_r0_1 = vmulq_s16(ref_arr_temp0_16x8_r0_1, phs_mask_16min_16x8_1); res_16x8_r1_1 = vmulq_s16(ref_arr_temp0_16x8_r1_1, phs_mask_16min_16x8_1); res_16x8_r0_1 = vmlaq_s16(res_16x8_r0_1, ref_arr_temp1_16x8_r0_1, vreinterpretq_s16_u16(phs_mask_16x8_1)); res_16x8_r1_1 = vmlaq_s16(res_16x8_r1_1, ref_arr_temp1_16x8_r1_1, vreinterpretq_s16_u16(phs_mask_16x8_1)); pu1_ref_y_ptr_incr_temp = pu1_ref_y_ptr_incr + (pi1_y_ref_pos[i4_y] * i4_refarray_wd); u1_y_incr_8x16_r0_0 = vld1q_u8((pu1_ref_y_ptr_incr_temp)); u1_incr_8x8x2_t.val[0] = vget_low_u8(u1_y_incr_8x16_r0_0); u1_incr_8x8x2_t.val[1] = vget_high_u8(u1_y_incr_8x16_r0_0); u1_incr_8x8_t0 = vtbl2_u8(u1_incr_8x8x2_t, vget_low_u8(x_ref_rnd_mask_r0_0)); u1_incr_8x8_t1 = vtbl2_u8(u1_incr_8x8x2_t, vget_high_u8(x_ref_rnd_mask_r0_0)); u1_y_incr_8x16_r0_0 = vcombine_u8(u1_incr_8x8_t0, u1_incr_8x8_t1); u1_y_incr_16x8_r0_0 = vmovl_u8(vget_low_u8(u1_y_incr_8x16_r0_0)); u1_y_incr_16x8_r0_1 = vmovl_u8(vget_high_u8(u1_y_incr_8x16_r0_0)); u1_y_incr_16x8_r0_0 = vtstq_u16(u1_y_incr_16x8_r0_0, ones); u1_y_incr_16x8_r0_1 = vtstq_u16(u1_y_incr_16x8_r0_1, ones); prev_res_16x8_r0_0 = res_16x8_r0_0; prev_res_16x8_r1_0 = res_16x8_r1_0; prev_res_16x8_r0_1 = res_16x8_r0_1; prev_res_16x8_r1_1 = res_16x8_r1_1; u1_prev_y_incr_16x8_r0_0 = u1_y_incr_16x8_r0_0; u1_prev_y_incr_16x8_r0_1 = u1_y_incr_16x8_r0_1; } } if(!zero_r0_r1) { res_16x8_l = vdupq_n_s16(0); res_16x8_h = vdupq_n_s16(0); } else { i4_y_phase = pi1_y_phase[i4_y]; if((i4_y_phase) >> 3) { vert_res_16x8_r0_0 = vbslq_s16(u1_y_incr_16x8_r0_0, res_16x8_r0_0, res_16x8_r1_0); vert_res_16x8_r1_0 = vbslq_s16(u1_y_incr_16x8_r0_0, res_16x8_r1_0, res_16x8_r1_0); vert_res_16x8_r0_1 = vbslq_s16(u1_y_incr_16x8_r0_1, res_16x8_r0_1, res_16x8_r1_1); vert_res_16x8_r1_1 = vbslq_s16(u1_y_incr_16x8_r0_1, res_16x8_r1_1, res_16x8_r1_1); } else { vert_res_16x8_r0_0 = vbslq_s16(u1_y_incr_16x8_r0_0, res_16x8_r0_0, res_16x8_r0_0); vert_res_16x8_r1_0 = vbslq_s16(u1_y_incr_16x8_r0_0, res_16x8_r1_0, res_16x8_r0_0); vert_res_16x8_r0_1 = vbslq_s16(u1_y_incr_16x8_r0_1, res_16x8_r0_1, res_16x8_r0_1); vert_res_16x8_r1_1 = vbslq_s16(u1_y_incr_16x8_r0_1, res_16x8_r1_1, res_16x8_r0_1); } res_32x4_l_0 = vmull_n_s16(vget_low_s16(vert_res_16x8_r0_0), 16 - i4_y_phase); res_32x4_l_0 = vmlal_n_s16(res_32x4_l_0, vget_low_s16(vert_res_16x8_r1_0), i4_y_phase); res_32x4_l_1 = vmull_n_s16(vget_high_s16(vert_res_16x8_r0_0), 16 - i4_y_phase); res_32x4_l_1 = vmlal_n_s16(res_32x4_l_1, vget_high_s16(vert_res_16x8_r1_0), i4_y_phase); res_32x4_h_0 = vmull_n_s16(vget_low_s16(vert_res_16x8_r0_1), 16 - i4_y_phase); res_32x4_h_0 = vmlal_n_s16(res_32x4_h_0, vget_low_s16(vert_res_16x8_r1_1), i4_y_phase); res_32x4_h_1 = vmull_n_s16(vget_high_s16(vert_res_16x8_r0_1), 16 - i4_y_phase); res_32x4_h_1 = vmlal_n_s16(res_32x4_h_1, vget_high_s16(vert_res_16x8_r1_1), i4_y_phase); res_32x4_l_0 = vrshrq_n_s32(res_32x4_l_0, 8); res_32x4_l_1 = vrshrq_n_s32(res_32x4_l_1, 8); res_32x4_h_0 = vrshrq_n_s32(res_32x4_h_0, 8); res_32x4_h_1 = vrshrq_n_s32(res_32x4_h_1, 8); res_16x8_l = vcombine_s16(vmovn_s32(res_32x4_l_0), vmovn_s32(res_32x4_l_1)); res_16x8_h = vcombine_s16(vmovn_s32(res_32x4_h_0), vmovn_s32(res_32x4_h_1)); } out_stride_temp = (i4_y * i4_out_stride); vst1q_s16((pi2_out + out_stride_temp), res_16x8_l); vst1q_s16((pi2_out + out_stride_temp + 8), res_16x8_h); } } else { int16x8_t ref_arr_16x8_r0_0; int16x8_t ref_arr_16x8_r1_0; uint8x16_t x_ref_pos_mask_r0_0, x_ref_rnd_mask_r0_0; uint16x8_t u1_incr_16x8_r0_0, u1_incr_mask_16x8_r0_0, phs_mask_16x8_0, u1_incr_not_16x8_r0_0, u1_y_incr_16x8_r0_0; uint16x8_t u1_incr_16x8_r1_0, u1_incr_mask_16x8_r1_0, u1_incr_not_16x8_r1_0; uint8x16_t u1_incr_8x16_r0_0, x_ref_pos_mask_temp_r0_0, u1_incr_mask_8x16_r0_0, u1_incr_not_8x16_r0_0, u1_y_incr_8x16_r0_0; uint8x16_t u1_incr_8x16_r1_0, x_ref_pos_mask_temp_r1_0, u1_incr_mask_8x16_r1_0, u1_incr_not_8x16_r1_0; int16x8_t ref_arr_temp0_16x8_r0_0, res_16x8_r0_0, vert_res_16x8_r0_0; int16x8_t ref_arr_temp0_16x8_r1_0, res_16x8_r1_0, vert_res_16x8_r1_0; int16x8_t ref_arr_temp1_16x8_r0_0; int16x8_t ref_arr_temp1_16x8_r1_0; int32x4_t res_32x4_l_0; int32x4_t res_32x4_l_1; int16x8_t out_16x8_0, out_16x8_1; uint16x8_t phs_mask_div8_16x8_0, phs_mask_div8_msb_16x8_0; int16x8_t const_16_16x8, phs_mask_16min_16x8_0; uint8x8x2_t u1_temp_8x8x2_t; uint8x8_t u1_temp_8x8_t0, u1_temp_8x8_t1; uint16x8_t chroma_mask_16x8 = vreinterpretq_u16_u32(vdupq_n_u32(0x0000ffff)); uint16x8_t ones = vdupq_n_u16(0xFFFF); WORD16 *pi2_ref_array_temp; UWORD8 *pu1_ref_x_ptr_incr_temp, *pu1_ref_y_ptr_incr_temp; WORD32 i4_y_phase; uint8x8x2_t u1_incr_8x8x2_t; uint8x8_t u1_incr_8x8_t0, u1_incr_8x8_t1; int16x8_t prev_res_16x8_r0_0; int16x8_t prev_res_16x8_r1_0; int16x8_t dup_val_1, dup_val_2, dup_abs; uint16x8_t u1_prev_y_incr_16x8_r0_0; WORD32 out_stride_temp; WORD32 zero_r0_r1 = 0; WORD32 i4_x2 = 0; for(i4_y = 0; i4_y < (i4_temp_array_ht); i4_y++) { arr_y_phase[i4_y] = (WORD8) ps_y_pos_phase[i4_y + i4_frm_mb_y].i2_phase; arr_y_ref_pos[i4_y] = (WORD8) (ps_y_pos_phase[i4_y + i4_frm_mb_y].i2_ref_pos); } pi1_y_ref_pos = arr_y_ref_pos; pi1_y_phase = arr_y_phase; for(i4_x = 0; i4_x < i4_mb_wd; i4_x++) { arr_x_ref_pos[i4_x] = (WORD8) ps_x_pos_phase[i4_x + i4_frm_mb_x].i2_ref_pos; arr_x_phase[i4_x] = (WORD8) ps_x_pos_phase[i4_x + i4_frm_mb_x].i2_phase; i4_x2 = i4_x << 1; arr_x_ref_pos_low[i4_x2] = (arr_x_ref_pos[i4_x]) << 1; arr_x_ref_pos_low[i4_x2 + 1] = arr_x_ref_pos_low[i4_x2] + 1; } pi1_x_ref_pos_low = arr_x_ref_pos_low; pi1_x_phase = arr_x_phase; phs_mask_16x8_0 = vmovl_u8(vld1_u8((pi1_x_phase))); x_ref_pos_mask_r0_0 = vld1q_u8((pi1_x_ref_pos_low)); const_16_16x8 = vdupq_n_s16(16); phs_mask_div8_16x8_0 = vshrq_n_u16(phs_mask_16x8_0, 3); phs_mask_div8_msb_16x8_0 = vsliq_n_u16(phs_mask_div8_16x8_0, phs_mask_div8_16x8_0, 8); phs_mask_16min_16x8_0 = vsubq_s16(const_16_16x8, vreinterpretq_s16_u16(phs_mask_16x8_0)); x_ref_rnd_mask_r0_0 = vaddq_u8( x_ref_pos_mask_r0_0, vreinterpretq_u8_u16(vshlq_n_u16(phs_mask_div8_msb_16x8_0, 1))); for(i4_y = 0; i4_y < (i4_temp_array_ht); i4_y++) { if((i4_y > 0) && (pi1_y_ref_pos[i4_y] == pi1_y_ref_pos[i4_y - 1])) { if(!zero_r0_r1) { res_32x4_l_0 = vdupq_n_s32(0); res_32x4_l_1 = vdupq_n_s32(0); out_stride_temp = (i4_y * i4_out_stride); out_16x8_0 = vld1q_s16(pi2_out + out_stride_temp); out_16x8_1 = vld1q_s16(pi2_out + out_stride_temp + 8); out_16x8_0 = vbslq_s16(chroma_mask_16x8, vreinterpretq_s16_s32(res_32x4_l_0), out_16x8_0); out_16x8_1 = vbslq_s16(chroma_mask_16x8, vreinterpretq_s16_s32(res_32x4_l_1), out_16x8_1); vst1q_s16((pi2_out + out_stride_temp), out_16x8_0); vst1q_s16((pi2_out + out_stride_temp + 8), out_16x8_1); continue; } res_16x8_r0_0 = prev_res_16x8_r0_0; res_16x8_r1_0 = prev_res_16x8_r1_0; u1_y_incr_16x8_r0_0 = u1_prev_y_incr_16x8_r0_0; } else { pi2_ref_array_temp = pi2_ref_array + ((pi1_y_ref_pos[i4_y]) * i4_refarray_wd); pu1_ref_x_ptr_incr_temp = pu1_ref_x_ptr_incr + ((pi1_y_ref_pos[i4_y]) * i4_refarray_wd); ref_arr_16x8_r0_0 = vld1q_s16((pi2_ref_array_temp)); ref_arr_16x8_r1_0 = vld1q_s16((pi2_ref_array_temp + i4_refarray_wd)); dup_val_1 = vabsq_s16(ref_arr_16x8_r0_0); dup_val_2 = vabsq_s16(ref_arr_16x8_r1_0); dup_abs = vqaddq_s16(dup_val_1, dup_val_2); zero_r0_r1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || dup_abs[6] || dup_abs[7]; if(zero_r0_r1) { u1_incr_16x8_r0_0 = (vmovl_u8(vld1_u8((pu1_ref_x_ptr_incr_temp)))); u1_incr_16x8_r1_0 = (vmovl_u8(vld1_u8((pu1_ref_x_ptr_incr_temp + i4_refarray_wd)))); u1_incr_8x8x2_t.val[0] = vget_low_u8(vreinterpretq_u8_u16(u1_incr_16x8_r0_0)); u1_incr_8x8x2_t.val[1] = vget_high_u8(vreinterpretq_u8_u16(u1_incr_16x8_r0_0)); u1_incr_8x8_t0 = vtbl2_u8(u1_incr_8x8x2_t, vget_low_u8(x_ref_pos_mask_r0_0)); u1_incr_8x8_t1 = vtbl2_u8(u1_incr_8x8x2_t, vget_high_u8(x_ref_pos_mask_r0_0)); u1_incr_8x16_r0_0 = vcombine_u8(u1_incr_8x8_t0, u1_incr_8x8_t1); u1_incr_8x8x2_t.val[0] = vget_low_u8(vreinterpretq_u8_u16(u1_incr_16x8_r1_0)); u1_incr_8x8x2_t.val[1] = vget_high_u8(vreinterpretq_u8_u16(u1_incr_16x8_r1_0)); u1_incr_8x8_t0 = vtbl2_u8(u1_incr_8x8x2_t, vget_low_u8(x_ref_pos_mask_r0_0)); u1_incr_8x8_t1 = vtbl2_u8(u1_incr_8x8x2_t, vget_high_u8(x_ref_pos_mask_r0_0)); u1_incr_8x16_r1_0 = vcombine_u8(u1_incr_8x8_t0, u1_incr_8x8_t1); u1_incr_16x8_r0_0 = vreinterpretq_u16_u8(u1_incr_8x16_r0_0); u1_incr_16x8_r1_0 = vreinterpretq_u16_u8(u1_incr_8x16_r1_0); u1_incr_mask_16x8_r0_0 = vsliq_n_u16(u1_incr_16x8_r0_0, u1_incr_16x8_r0_0, 8); u1_incr_mask_16x8_r1_0 = vsliq_n_u16(u1_incr_16x8_r1_0, u1_incr_16x8_r1_0, 8); u1_incr_not_16x8_r0_0 = vbicq_u16(phs_mask_div8_msb_16x8_0, u1_incr_mask_16x8_r0_0); u1_incr_not_16x8_r1_0 = vbicq_u16(phs_mask_div8_msb_16x8_0, u1_incr_mask_16x8_r1_0); u1_incr_mask_8x16_r0_0 = vreinterpretq_u8_u16(vshlq_n_u16(u1_incr_mask_16x8_r0_0, 1)); u1_incr_mask_8x16_r1_0 = vreinterpretq_u8_u16(vshlq_n_u16(u1_incr_mask_16x8_r1_0, 1)); u1_incr_not_8x16_r0_0 = vreinterpretq_u8_u16(vshlq_n_u16(u1_incr_not_16x8_r0_0, 1)); u1_incr_not_8x16_r1_0 = vreinterpretq_u8_u16(vshlq_n_u16(u1_incr_not_16x8_r1_0, 1)); u1_incr_not_8x16_r0_0 = vaddq_u8(u1_incr_not_8x16_r0_0, x_ref_pos_mask_r0_0); u1_incr_not_8x16_r1_0 = vaddq_u8(u1_incr_not_8x16_r1_0, x_ref_pos_mask_r0_0); x_ref_pos_mask_temp_r0_0 = vaddq_u8(u1_incr_not_8x16_r0_0, u1_incr_mask_8x16_r0_0); x_ref_pos_mask_temp_r1_0 = vaddq_u8(u1_incr_not_8x16_r1_0, u1_incr_mask_8x16_r1_0); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_0))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_0))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(u1_incr_not_8x16_r0_0)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(u1_incr_not_8x16_r0_0)); ref_arr_temp0_16x8_r0_0 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_0))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_0))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(u1_incr_not_8x16_r1_0)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(u1_incr_not_8x16_r1_0)); ref_arr_temp0_16x8_r1_0 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_0))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r0_0))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(x_ref_pos_mask_temp_r0_0)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(x_ref_pos_mask_temp_r0_0)); ref_arr_temp1_16x8_r0_0 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); u1_temp_8x8x2_t.val[0] = vreinterpret_u8_s8(vget_low_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_0))); u1_temp_8x8x2_t.val[1] = vreinterpret_u8_s8(vget_high_s8(vreinterpretq_s8_s16(ref_arr_16x8_r1_0))); u1_temp_8x8_t0 = vtbl2_u8(u1_temp_8x8x2_t, vget_low_u8(x_ref_pos_mask_temp_r1_0)); u1_temp_8x8_t1 = vtbl2_u8(u1_temp_8x8x2_t, vget_high_u8(x_ref_pos_mask_temp_r1_0)); ref_arr_temp1_16x8_r1_0 = vreinterpretq_s16_s8(vcombine_s8( vreinterpret_s8_u8(u1_temp_8x8_t0), vreinterpret_s8_u8(u1_temp_8x8_t1))); res_16x8_r0_0 = vmulq_s16(ref_arr_temp0_16x8_r0_0, phs_mask_16min_16x8_0); res_16x8_r1_0 = vmulq_s16(ref_arr_temp0_16x8_r1_0, phs_mask_16min_16x8_0); res_16x8_r0_0 = vmlaq_s16(res_16x8_r0_0, ref_arr_temp1_16x8_r0_0, vreinterpretq_s16_u16(phs_mask_16x8_0)); res_16x8_r1_0 = vmlaq_s16(res_16x8_r1_0, ref_arr_temp1_16x8_r1_0, vreinterpretq_s16_u16(phs_mask_16x8_0)); pu1_ref_y_ptr_incr_temp = pu1_ref_y_ptr_incr + (pi1_y_ref_pos[i4_y] * i4_refarray_wd); u1_y_incr_16x8_r0_0 = vmovl_u8(vld1_u8((pu1_ref_y_ptr_incr_temp))); u1_incr_8x8x2_t.val[0] = vget_low_u8(vreinterpretq_u8_u16(u1_y_incr_16x8_r0_0)); u1_incr_8x8x2_t.val[1] = vget_high_u8(vreinterpretq_u8_u16(u1_y_incr_16x8_r0_0)); u1_incr_8x8_t0 = vtbl2_u8(u1_incr_8x8x2_t, vget_low_u8(x_ref_rnd_mask_r0_0)); u1_incr_8x8_t1 = vtbl2_u8(u1_incr_8x8x2_t, vget_high_u8(x_ref_rnd_mask_r0_0)); u1_y_incr_8x16_r0_0 = vcombine_u8(u1_incr_8x8_t0, u1_incr_8x8_t1); u1_y_incr_16x8_r0_0 = vreinterpretq_u16_u8(u1_y_incr_8x16_r0_0); u1_y_incr_16x8_r0_0 = vtstq_u16(u1_y_incr_16x8_r0_0, ones); prev_res_16x8_r0_0 = res_16x8_r0_0; prev_res_16x8_r1_0 = res_16x8_r1_0; u1_prev_y_incr_16x8_r0_0 = u1_y_incr_16x8_r0_0; } } if(!zero_r0_r1) { res_32x4_l_0 = vdupq_n_s32(0); res_32x4_l_1 = vdupq_n_s32(0); } else { i4_y_phase = pi1_y_phase[i4_y]; if((i4_y_phase) >> 3) { vert_res_16x8_r0_0 = vbslq_s16(u1_y_incr_16x8_r0_0, res_16x8_r0_0, res_16x8_r1_0); vert_res_16x8_r1_0 = vbslq_s16(u1_y_incr_16x8_r0_0, res_16x8_r1_0, res_16x8_r1_0); } else { vert_res_16x8_r0_0 = vbslq_s16(u1_y_incr_16x8_r0_0, res_16x8_r0_0, res_16x8_r0_0); vert_res_16x8_r1_0 = vbslq_s16(u1_y_incr_16x8_r0_0, res_16x8_r1_0, res_16x8_r0_0); } res_32x4_l_0 = vmull_n_s16(vget_low_s16(vert_res_16x8_r0_0), 16 - i4_y_phase); res_32x4_l_0 = vmlal_n_s16(res_32x4_l_0, vget_low_s16(vert_res_16x8_r1_0), i4_y_phase); res_32x4_l_1 = vmull_n_s16(vget_high_s16(vert_res_16x8_r0_0), 16 - i4_y_phase); res_32x4_l_1 = vmlal_n_s16(res_32x4_l_1, vget_high_s16(vert_res_16x8_r1_0), i4_y_phase); res_32x4_l_0 = vrshrq_n_s32(res_32x4_l_0, 8); res_32x4_l_1 = vrshrq_n_s32(res_32x4_l_1, 8); } out_stride_temp = (i4_y * i4_out_stride); out_16x8_0 = vld1q_s16(pi2_out + out_stride_temp); out_16x8_1 = vld1q_s16(pi2_out + out_stride_temp + 8); out_16x8_0 = vbslq_s16(chroma_mask_16x8, vreinterpretq_s16_s32(res_32x4_l_0), out_16x8_0); out_16x8_1 = vbslq_s16(chroma_mask_16x8, vreinterpretq_s16_s32(res_32x4_l_1), out_16x8_1); vst1q_s16((pi2_out + out_stride_temp), out_16x8_0); vst1q_s16((pi2_out + out_stride_temp + 8), out_16x8_1); } } return; } /* End of Interpolation Function */ /*****************************************************************************/ /* */ /* Function Name : isvcd_residual_reflayer_const_non_boundary_mb_neonintr */ /* */ /* Description : */ /* */ /* Inputs : */ /* Globals : none */ /* Processing : */ /* */ /* Outputs : none */ /* Returns : none */ /* */ /* Issues : none */ /* */ /* Revision History: */ /* */ /* DD MM YYYY Author(s) Changes (Describe the changes made) */ /* 25 11 2021 Dolan creation */ /* */ /*****************************************************************************/ void isvcd_residual_reflayer_const_non_boundary_mb_neonintr( WORD16 *pi2_inp_data, WORD32 i4_inp_data_stride, WORD16 *pi2_ref_array, WORD32 i4_refarray_wd, WORD32 i4_refarray_ht, WORD32 i4_ref_mb_type_q0, WORD32 i4_ref_mb_type_q1, WORD32 i4_ref_mb_type_q2, WORD32 i4_ref_mb_type_q3, WORD32 i4_mb_quard1_part_x, WORD32 i4_mb_quard1_part_y, WORD32 i4_chroma_flag) { WORD32 i4_y; WORD16 *pi2_ref_data_byte; WORD16 *pi2_ref_array_temp; if(i4_chroma_flag == 0) { WORD16 index_0[8] = {0, 1, 2, 3, 4, 5, 6, 7}; int16x8_t ref_mb_type_16x8_q0, ref_mb_type_16x8_q1, ref_mb_type_16x8_q2, ref_mb_type_16x8_q3, mb_quard1_part_x_16x8; int16x8_t ref_mb_type_16x8_0, ref_mb_type_16x8_1; int16x8_t ref_mb_type_16x8_low_0, ref_mb_type_16x8_low_1; uint16x8_t mb_type_mask_16x8_0, mb_type_mask_16x8_1; uint16x8_t mb_type_mask_16x8_low_0, mb_type_mask_16x8_low_1; uint16x8_t mask_16x8_0; int16x8_t index_arr_0; int16x8_t inp_data_16x8_0, inp_data_16x8_1; int16x8_t res_16x8_0, res_16x8_1; int16x8_t one_16x8 = vdupq_n_s16(1); int16x8_t zero_16x8 = vdupq_n_s16(0); index_arr_0 = vld1q_s16(&index_0[0]); ref_mb_type_16x8_q0 = vdupq_n_s16(i4_ref_mb_type_q0); ref_mb_type_16x8_q1 = vdupq_n_s16(i4_ref_mb_type_q1); ref_mb_type_16x8_q2 = vdupq_n_s16(i4_ref_mb_type_q2); ref_mb_type_16x8_q3 = vdupq_n_s16(i4_ref_mb_type_q3); if((i4_mb_quard1_part_x >= i4_refarray_wd) && (i4_mb_quard1_part_y >= i4_refarray_ht)) { // Quard 0 ref_mb_type_16x8_0 = ref_mb_type_16x8_q0; ref_mb_type_16x8_1 = ref_mb_type_16x8_q0; mb_type_mask_16x8_0 = vceqq_s16(ref_mb_type_16x8_0, one_16x8); mb_type_mask_16x8_1 = mb_type_mask_16x8_0; } else if((i4_mb_quard1_part_y >= (i4_refarray_ht - 1)) && (i4_mb_quard1_part_x < i4_refarray_wd)) { // Quard 0 & 1 if(i4_mb_quard1_part_x == 8) { ref_mb_type_16x8_0 = ref_mb_type_16x8_q0; ref_mb_type_16x8_1 = ref_mb_type_16x8_q1; } else if(i4_mb_quard1_part_x < 8) { mb_quard1_part_x_16x8 = vdupq_n_s16((i4_mb_quard1_part_x)); mask_16x8_0 = vcltq_s16(index_arr_0, mb_quard1_part_x_16x8); // return 1 if a= i4_refarray_wd) { ref_mb_type_16x8_0 = ref_mb_type_16x8_q0; ref_mb_type_16x8_1 = ref_mb_type_16x8_q0; ref_mb_type_16x8_low_0 = ref_mb_type_16x8_q2; ref_mb_type_16x8_low_1 = ref_mb_type_16x8_q2; } else { // Quard 0, 1, 2, 3 if(i4_mb_quard1_part_x == 8) { ref_mb_type_16x8_0 = ref_mb_type_16x8_q0; ref_mb_type_16x8_1 = ref_mb_type_16x8_q1; ref_mb_type_16x8_low_0 = ref_mb_type_16x8_q2; ref_mb_type_16x8_low_1 = ref_mb_type_16x8_q3; } else if(i4_mb_quard1_part_x < 8) { mb_quard1_part_x_16x8 = vdupq_n_s16((i4_mb_quard1_part_x)); mask_16x8_0 = vcltq_s16(index_arr_0, mb_quard1_part_x_16x8); // return 1 if a= i4_refarray_wd) && (i4_mb_quard1_part_y >= i4_refarray_ht)) { // Quard 0 ref_mb_type_16x8_0 = ref_mb_type_16x8_q0; mb_type_mask_16x8_0 = vceqq_s16(ref_mb_type_16x8_0, one_16x8); } else if((i4_mb_quard1_part_y >= (i4_refarray_ht - 1)) && (i4_mb_quard1_part_x < i4_refarray_wd)) { // Quard 0 & 1 mb_quard1_part_x_16x8 = vdupq_n_s16((i4_mb_quard1_part_x)); mask_16x8_0 = vcltq_s16(index_arr_0, mb_quard1_part_x_16x8); // return 1 if a= i4_refarray_wd) { ref_mb_type_16x8_0 = ref_mb_type_16x8_q0; ref_mb_type_16x8_low_0 = ref_mb_type_16x8_q2; } else { mb_quard1_part_x_16x8 = vdupq_n_s16((i4_mb_quard1_part_x)); mask_16x8_0 = vcltq_s16(index_arr_0, mb_quard1_part_x_16x8); // return 1 if a