| /****************************************************************************** |
| * |
| * 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_pred_residual_recon_neonintr.c |
| * |
| * @brief |
| * Contains definition of functions for h264 inverse quantization inverse |
| * transformation and resd comp |
| * |
| * @author |
| * Kishore |
| * |
| * @par List of Functions: |
| * - isvcd_pred_residual_recon_16x16_neonintr() |
| * - isvcd_pred_residual_recon_8x8_neonintr() |
| * - isvcd_pred_residual_recon_4x4_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 Includes */ |
| /*****************************************************************************/ |
| #include <string.h> |
| #include <arm_neon.h> |
| |
| /* User include files */ |
| #include "ih264_typedefs.h" |
| #include "ih264_defs.h" |
| #include "ih264_trans_macros.h" |
| #include "ih264_macros.h" |
| #include "ih264_platform_macros.h" |
| #include "ih264_trans_data.h" |
| #include "ih264_size_defs.h" |
| #include "ih264_structs.h" |
| #include "isvcd_pred_residual_recon.h" |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_pred_residual_recon_4x4_neonintr */ |
| /* */ |
| /* Description : this function computes the recon data from the */ |
| /* pred and residual buffer */ |
| /* */ |
| /* Inputs : */ |
| /* Globals : none */ |
| /* Processing : */ |
| /* */ |
| /* Outputs : none */ |
| /* Returns : nnz */ |
| /* */ |
| /* Issues : none */ |
| /* */ |
| /* Revision History: */ |
| /* */ |
| /* DD MM YYYY Author(s) Changes (Describe the changes made) */ |
| /* 25 11 2021 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| WORD32 isvcd_pred_residual_recon_4x4_neonintr(UWORD8 *pu1_pred, WORD16 *pi2_rsd, UWORD8 *pu1_out, |
| WORD32 pred_strd, WORD32 rsd_strd, WORD32 out_strd) |
| { |
| uint8x8_t pred0_in, pred1_in, pred2_in, pred3_in; |
| int16x8_t pred0, pred1, pred2, pred3; |
| int16x8_t resd0_in, resd1_in, resd2_in, resd3_in; |
| int16x8_t resd01_in, resd23_in; |
| WORD32 i4_nnz; |
| int16x8_t dup_val_1, dup_val_2, dup_abs; |
| |
| pred0_in = vld1_u8((uint8_t *) pu1_pred); |
| pred1_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 1)); |
| pred2_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 2)); |
| pred3_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 3)); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 3)); |
| |
| resd01_in = vreinterpretq_s16_s64(vcombine_s64(vget_low_s64(vreinterpretq_s64_s16(resd0_in)), |
| vget_low_s64(vreinterpretq_s64_s16(resd1_in)))); |
| |
| resd23_in = vreinterpretq_s16_s64(vcombine_s64(vget_low_s64(vreinterpretq_s64_s16(resd2_in)), |
| vget_low_s64(vreinterpretq_s64_s16(resd3_in)))); |
| |
| dup_val_1 = vabsq_s16(resd01_in); |
| dup_val_2 = vabsq_s16(resd23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| i4_nnz = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| |
| vst1_lane_u32((uint32_t *) (pu1_out), vreinterpret_u32_u8(pred0_in), 0); |
| vst1_lane_u32((uint32_t *) (pu1_out + out_strd), vreinterpret_u32_u8(pred1_in), 0); |
| vst1_lane_u32((uint32_t *) (pu1_out + out_strd * 2), vreinterpret_u32_u8(pred2_in), 0); |
| vst1_lane_u32((uint32_t *) (pu1_out + out_strd * 3), vreinterpret_u32_u8(pred3_in), 0); |
| |
| return i4_nnz; |
| } |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_pred_residual_recon_8x8_neonintr */ |
| /* */ |
| /* Description : this function computes the recon data from the */ |
| /* pred and residual buffer */ |
| /* */ |
| /* Inputs : */ |
| /* Globals : none */ |
| /* Processing : */ |
| /* */ |
| /* Outputs : none */ |
| /* Returns : nnz */ |
| /* */ |
| /* Issues : none */ |
| /* */ |
| /* Revision History: */ |
| /* */ |
| /* DD MM YYYY Author(s) Changes (Describe the changes made) */ |
| /* 25 11 2021 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| WORD32 isvcd_pred_residual_recon_8x8_neonintr(UWORD8 *pu1_pred, WORD16 *pi2_rsd, UWORD8 *pu1_out, |
| WORD32 pred_strd, WORD32 rsd_strd, WORD32 out_strd) |
| { |
| uint8x8_t pred0_in, pred1_in, pred2_in, pred3_in; |
| uint8x8_t pred4_in, pred5_in, pred6_in, pred7_in; |
| int16x8_t pred0, pred1, pred2, pred3; |
| int16x8_t resd0_in, resd1_in, resd2_in, resd3_in; |
| int16x8_t pred4, pred5, pred6, pred7; |
| int16x8_t resd4_in, resd5_in, resd6_in, resd7_in; |
| int16x8_t dup_val_1, dup_val_2, dup_abs; |
| int64x2_t resd0_in_64x2, resd1_in_64x2, resd2_in_64x2, resd3_in_64x2, resd4_in_64x2, |
| resd5_in_64x2, resd6_in_64x2, resd7_in_64x2; |
| |
| int16x8_t resd_b0_r01_in; |
| int16x8_t resd_b0_r23_in; |
| int16x8_t resd_b1_r01_in; |
| int16x8_t resd_b1_r23_in; |
| int16x8_t resd_b2_r45_in; |
| int16x8_t resd_b2_r67_in; |
| int16x8_t resd_b3_r45_in; |
| int16x8_t resd_b3_r67_in; |
| |
| WORD32 nnz, nnz_b0, nnz_b1, nnz_b2, nnz_b3; |
| |
| pred0_in = vld1_u8((uint8_t *) pu1_pred); |
| pred1_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd)); |
| pred2_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 2)); |
| pred3_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 3)); |
| pred4_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 4)); |
| pred5_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 5)); |
| pred6_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 6)); |
| pred7_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 7)); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| pred4 = vreinterpretq_s16_u16(vmovl_u8(pred4_in)); |
| pred5 = vreinterpretq_s16_u16(vmovl_u8(pred5_in)); |
| pred6 = vreinterpretq_s16_u16(vmovl_u8(pred6_in)); |
| pred7 = vreinterpretq_s16_u16(vmovl_u8(pred7_in)); |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz = (nnz_b0 | (nnz_b1 << 1) | (nnz_b2 << 4) | (nnz_b3 << 5)); |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| pred4 = vaddq_s16(pred4, resd4_in); |
| pred5 = vaddq_s16(pred5, resd5_in); |
| pred6 = vaddq_s16(pred6, resd6_in); |
| pred7 = vaddq_s16(pred7, resd7_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| pred4_in = vqmovun_s16(pred4); |
| pred5_in = vqmovun_s16(pred5); |
| pred6_in = vqmovun_s16(pred6); |
| pred7_in = vqmovun_s16(pred7); |
| |
| vst1_u8((uint8_t *) (pu1_out), pred0_in); |
| vst1_u8((uint8_t *) (pu1_out + out_strd), pred1_in); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 2), pred2_in); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 3), pred3_in); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 4), pred4_in); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 5), pred5_in); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 6), pred6_in); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 7), pred7_in); |
| |
| return nnz; |
| } |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_pred_residual_recon_16x16_neonintr */ |
| /* */ |
| /* Description : this function computes the recon data from the */ |
| /* pred and residual buffer */ |
| /* */ |
| /* Inputs : */ |
| /* Globals : none */ |
| /* Processing : */ |
| /* */ |
| /* Outputs : none */ |
| /* Returns : nnz */ |
| /* */ |
| /* Issues : none */ |
| /* */ |
| /* Revision History: */ |
| /* */ |
| /* DD MM YYYY Author(s) Changes (Describe the changes made) */ |
| /* 25 11 2021 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| WORD32 isvcd_pred_residual_recon_16x16_neonintr(UWORD8 *pu1_pred, WORD16 *pi2_rsd, UWORD8 *pu1_out, |
| WORD32 pred_strd, WORD32 rsd_strd, WORD32 out_strd) |
| { |
| uint8x8_t pred0_in, pred1_in, pred2_in, pred3_in; |
| uint8x8_t pred4_in, pred5_in, pred6_in, pred7_in; |
| int16x8_t pred0, pred1, pred2, pred3; |
| int16x8_t resd0_in, resd1_in, resd2_in, resd3_in; |
| int16x8_t pred4, pred5, pred6, pred7; |
| int16x8_t resd4_in, resd5_in, resd6_in, resd7_in; |
| int16x8_t dup_val_1, dup_val_2, dup_abs; |
| UWORD8 *pu1_pred_ptr = pu1_pred; |
| WORD16 *pi2_rsd_ptr = pi2_rsd; |
| UWORD8 *pu1_out_ptr = pu1_out; |
| |
| int64x2_t resd0_in_64x2, resd1_in_64x2, resd2_in_64x2, resd3_in_64x2, resd4_in_64x2, |
| resd5_in_64x2, resd6_in_64x2, resd7_in_64x2; |
| |
| int16x8_t resd_b0_r01_in; |
| int16x8_t resd_b0_r23_in; |
| int16x8_t resd_b1_r01_in; |
| int16x8_t resd_b1_r23_in; |
| int16x8_t resd_b2_r45_in; |
| int16x8_t resd_b2_r67_in; |
| int16x8_t resd_b3_r45_in; |
| int16x8_t resd_b3_r67_in; |
| |
| WORD32 nnz, nnz_b0, nnz_b1, nnz_b2, nnz_b3; |
| |
| /* First row of 8, first 8x8 elements */ |
| pred0_in = vld1_u8((uint8_t *) pu1_pred_ptr); |
| pred1_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd)); |
| pred2_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 2)); |
| pred3_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 3)); |
| pred4_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 4)); |
| pred5_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 5)); |
| pred6_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 6)); |
| pred7_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 7)); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| pred4 = vreinterpretq_s16_u16(vmovl_u8(pred4_in)); |
| pred5 = vreinterpretq_s16_u16(vmovl_u8(pred5_in)); |
| pred6 = vreinterpretq_s16_u16(vmovl_u8(pred6_in)); |
| pred7 = vreinterpretq_s16_u16(vmovl_u8(pred7_in)); |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd_ptr); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd_ptr + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz = (nnz_b0 | (nnz_b1 << 1) | (nnz_b2 << 4) | (nnz_b3 << 5)); |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| pred4 = vaddq_s16(pred4, resd4_in); |
| pred5 = vaddq_s16(pred5, resd5_in); |
| pred6 = vaddq_s16(pred6, resd6_in); |
| pred7 = vaddq_s16(pred7, resd7_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| pred4_in = vqmovun_s16(pred4); |
| pred5_in = vqmovun_s16(pred5); |
| pred6_in = vqmovun_s16(pred6); |
| pred7_in = vqmovun_s16(pred7); |
| |
| vst1_u8((uint8_t *) (pu1_out_ptr), pred0_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd), pred1_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 2), pred2_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 3), pred3_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 4), pred4_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 5), pred5_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 6), pred6_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 7), pred7_in); |
| |
| /* first row of 8, sec 8x8 elements */ |
| pu1_out_ptr = pu1_out_ptr + 8; |
| pi2_rsd_ptr = pi2_rsd_ptr + 8; |
| pu1_pred_ptr = pu1_pred_ptr + 8; |
| |
| pred0_in = vld1_u8((uint8_t *) pu1_pred_ptr); |
| pred1_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd)); |
| pred2_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 2)); |
| pred3_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 3)); |
| pred4_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 4)); |
| pred5_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 5)); |
| pred6_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 6)); |
| pred7_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 7)); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| pred4 = vreinterpretq_s16_u16(vmovl_u8(pred4_in)); |
| pred5 = vreinterpretq_s16_u16(vmovl_u8(pred5_in)); |
| pred6 = vreinterpretq_s16_u16(vmovl_u8(pred6_in)); |
| pred7 = vreinterpretq_s16_u16(vmovl_u8(pred7_in)); |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd_ptr); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd_ptr + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz |= (nnz_b0 << 2 | (nnz_b1 << 3) | (nnz_b2 << 6) | (nnz_b3 << 7)); |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| pred4 = vaddq_s16(pred4, resd4_in); |
| pred5 = vaddq_s16(pred5, resd5_in); |
| pred6 = vaddq_s16(pred6, resd6_in); |
| pred7 = vaddq_s16(pred7, resd7_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| pred4_in = vqmovun_s16(pred4); |
| pred5_in = vqmovun_s16(pred5); |
| pred6_in = vqmovun_s16(pred6); |
| pred7_in = vqmovun_s16(pred7); |
| |
| vst1_u8((uint8_t *) (pu1_out_ptr), pred0_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd), pred1_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 2), pred2_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 3), pred3_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 4), pred4_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 5), pred5_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 6), pred6_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 7), pred7_in); |
| |
| pu1_out_ptr = pu1_out + (8 * out_strd); |
| pi2_rsd_ptr = pi2_rsd + (8 * rsd_strd); |
| pu1_pred_ptr = pu1_pred + (8 * pred_strd); |
| |
| /*Sec row of 8, first 8x8*/ |
| pred0_in = vld1_u8((uint8_t *) pu1_pred_ptr); |
| pred1_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd)); |
| pred2_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 2)); |
| pred3_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 3)); |
| pred4_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 4)); |
| pred5_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 5)); |
| pred6_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 6)); |
| pred7_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 7)); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| pred4 = vreinterpretq_s16_u16(vmovl_u8(pred4_in)); |
| pred5 = vreinterpretq_s16_u16(vmovl_u8(pred5_in)); |
| pred6 = vreinterpretq_s16_u16(vmovl_u8(pred6_in)); |
| pred7 = vreinterpretq_s16_u16(vmovl_u8(pred7_in)); |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd_ptr); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd_ptr + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz |= (nnz_b0 << 8 | (nnz_b1 << 9) | (nnz_b2 << 12) | (nnz_b3 << 13)); |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| pred4 = vaddq_s16(pred4, resd4_in); |
| pred5 = vaddq_s16(pred5, resd5_in); |
| pred6 = vaddq_s16(pred6, resd6_in); |
| pred7 = vaddq_s16(pred7, resd7_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| pred4_in = vqmovun_s16(pred4); |
| pred5_in = vqmovun_s16(pred5); |
| pred6_in = vqmovun_s16(pred6); |
| pred7_in = vqmovun_s16(pred7); |
| |
| vst1_u8((uint8_t *) (pu1_out_ptr), pred0_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd), pred1_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 2), pred2_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 3), pred3_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 4), pred4_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 5), pred5_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 6), pred6_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 7), pred7_in); |
| |
| /*Sec row of 8, Sec 8x8*/ |
| pu1_out_ptr = pu1_out_ptr + 8; |
| pi2_rsd_ptr = pi2_rsd_ptr + 8; |
| pu1_pred_ptr = pu1_pred_ptr + 8; |
| |
| pred0_in = vld1_u8((uint8_t *) pu1_pred_ptr); |
| pred1_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd)); |
| pred2_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 2)); |
| pred3_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 3)); |
| pred4_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 4)); |
| pred5_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 5)); |
| pred6_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 6)); |
| pred7_in = vld1_u8((uint8_t *) pu1_pred_ptr + (pred_strd * 7)); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| pred4 = vreinterpretq_s16_u16(vmovl_u8(pred4_in)); |
| pred5 = vreinterpretq_s16_u16(vmovl_u8(pred5_in)); |
| pred6 = vreinterpretq_s16_u16(vmovl_u8(pred6_in)); |
| pred7 = vreinterpretq_s16_u16(vmovl_u8(pred7_in)); |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd_ptr); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd_ptr + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz |= (nnz_b0 << 10 | (nnz_b1 << 11) | (nnz_b2 << 14) | (nnz_b3 << 15)); |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| pred4 = vaddq_s16(pred4, resd4_in); |
| pred5 = vaddq_s16(pred5, resd5_in); |
| pred6 = vaddq_s16(pred6, resd6_in); |
| pred7 = vaddq_s16(pred7, resd7_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| pred4_in = vqmovun_s16(pred4); |
| pred5_in = vqmovun_s16(pred5); |
| pred6_in = vqmovun_s16(pred6); |
| pred7_in = vqmovun_s16(pred7); |
| |
| vst1_u8((uint8_t *) (pu1_out_ptr), pred0_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd), pred1_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 2), pred2_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 3), pred3_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 4), pred4_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 5), pred5_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 6), pred6_in); |
| vst1_u8((uint8_t *) (pu1_out_ptr + out_strd * 7), pred7_in); |
| |
| return nnz; |
| } |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_pred_residual_recon_chroma_8x8_neonintr */ |
| /* */ |
| /* Description : this function computes the recon data from the */ |
| /* pred and residual buffer */ |
| /* */ |
| /* 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 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| void isvcd_pred_residual_recon_chroma_8x8_neonintr(UWORD8 *pu1_pred, WORD16 *pi2_rsd, |
| UWORD8 *pu1_out, WORD32 pred_strd, |
| WORD32 rsd_strd, WORD32 out_strd) |
| { |
| uint8x8_t pred0_in, pred1_in, pred2_in, pred3_in; |
| uint8x8_t pred4_in, pred5_in, pred6_in, pred7_in; |
| int16x8_t pred0, pred1, pred2, pred3; |
| int16x8_t resd0_in, resd1_in, resd2_in, resd3_in; |
| int16x8_t pred4, pred5, pred6, pred7; |
| int16x8_t resd4_in, resd5_in, resd6_in, resd7_in; |
| |
| UWORD8 *pu1_pred_ptr = pu1_pred; |
| WORD16 *pi2_rsd_ptr = pi2_rsd; |
| UWORD8 *pu1_out_ptr = pu1_out; |
| uint8x16_t pred0_inp_full, pred1_inp_full, pred2_inp_full, pred3_inp_full; |
| uint8x16_t pred4_inp_full, pred5_inp_full, pred6_inp_full, pred7_inp_full; |
| uint8x8_t i4_out_horz_8x8_r0, i4_out_horz_8x8_r1, i4_out_horz_8x8_r2, i4_out_horz_8x8_r3; |
| uint8x8_t i4_out_horz_8x8_r4, i4_out_horz_8x8_r5, i4_out_horz_8x8_r6, i4_out_horz_8x8_r7; |
| uint8x8_t chroma_mask_8x8; |
| |
| pred0_inp_full = vld1q_u8((uint8_t *) pu1_pred); |
| pred1_inp_full = vld1q_u8((uint8_t *) pu1_pred + (pred_strd)); |
| pred2_inp_full = vld1q_u8((uint8_t *) pu1_pred + (pred_strd * 2)); |
| pred3_inp_full = vld1q_u8((uint8_t *) pu1_pred + (pred_strd * 3)); |
| pred4_inp_full = vld1q_u8((uint8_t *) pu1_pred + (pred_strd * 4)); |
| pred5_inp_full = vld1q_u8((uint8_t *) pu1_pred + (pred_strd * 5)); |
| pred6_inp_full = vld1q_u8((uint8_t *) pu1_pred + (pred_strd * 6)); |
| pred7_inp_full = vld1q_u8((uint8_t *) pu1_pred + (pred_strd * 7)); |
| |
| pred0_in = vget_low_u8(pred0_inp_full); |
| pred1_in = vget_low_u8(pred1_inp_full); |
| pred2_in = vget_low_u8(pred2_inp_full); |
| pred3_in = vget_low_u8(pred3_inp_full); |
| pred4_in = vget_low_u8(pred4_inp_full); |
| pred5_in = vget_low_u8(pred5_inp_full); |
| pred6_in = vget_low_u8(pred6_inp_full); |
| pred7_in = vget_low_u8(pred7_inp_full); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| pred4 = vreinterpretq_s16_u16(vmovl_u8(pred4_in)); |
| pred5 = vreinterpretq_s16_u16(vmovl_u8(pred5_in)); |
| pred6 = vreinterpretq_s16_u16(vmovl_u8(pred6_in)); |
| pred7 = vreinterpretq_s16_u16(vmovl_u8(pred7_in)); |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 7)); |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| pred4 = vaddq_s16(pred4, resd4_in); |
| pred5 = vaddq_s16(pred5, resd5_in); |
| pred6 = vaddq_s16(pred6, resd6_in); |
| pred7 = vaddq_s16(pred7, resd7_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| pred4_in = vqmovun_s16(pred4); |
| pred5_in = vqmovun_s16(pred5); |
| pred6_in = vqmovun_s16(pred6); |
| pred7_in = vqmovun_s16(pred7); |
| |
| chroma_mask_8x8 = vreinterpret_u8_u16(vdup_n_u16(0x00ff)); |
| |
| i4_out_horz_8x8_r0 = vld1_u8(pu1_out); |
| i4_out_horz_8x8_r1 = vld1_u8(pu1_out + out_strd); |
| i4_out_horz_8x8_r2 = vld1_u8(pu1_out + out_strd * 2); |
| i4_out_horz_8x8_r3 = vld1_u8(pu1_out + out_strd * 3); |
| i4_out_horz_8x8_r4 = vld1_u8(pu1_out + out_strd * 4); |
| i4_out_horz_8x8_r5 = vld1_u8(pu1_out + out_strd * 5); |
| i4_out_horz_8x8_r6 = vld1_u8(pu1_out + out_strd * 6); |
| i4_out_horz_8x8_r7 = vld1_u8(pu1_out + out_strd * 7); |
| |
| i4_out_horz_8x8_r0 = vbsl_u8(chroma_mask_8x8, pred0_in, i4_out_horz_8x8_r0); |
| i4_out_horz_8x8_r1 = vbsl_u8(chroma_mask_8x8, pred1_in, i4_out_horz_8x8_r1); |
| i4_out_horz_8x8_r2 = vbsl_u8(chroma_mask_8x8, pred2_in, i4_out_horz_8x8_r2); |
| i4_out_horz_8x8_r3 = vbsl_u8(chroma_mask_8x8, pred3_in, i4_out_horz_8x8_r3); |
| i4_out_horz_8x8_r4 = vbsl_u8(chroma_mask_8x8, pred4_in, i4_out_horz_8x8_r4); |
| i4_out_horz_8x8_r5 = vbsl_u8(chroma_mask_8x8, pred5_in, i4_out_horz_8x8_r5); |
| i4_out_horz_8x8_r6 = vbsl_u8(chroma_mask_8x8, pred6_in, i4_out_horz_8x8_r6); |
| i4_out_horz_8x8_r7 = vbsl_u8(chroma_mask_8x8, pred7_in, i4_out_horz_8x8_r7); |
| |
| vst1_u8((uint8_t *) (pu1_out), i4_out_horz_8x8_r0); |
| vst1_u8((uint8_t *) (pu1_out + out_strd), i4_out_horz_8x8_r1); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 2), i4_out_horz_8x8_r2); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 3), i4_out_horz_8x8_r3); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 4), i4_out_horz_8x8_r4); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 5), i4_out_horz_8x8_r5); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 6), i4_out_horz_8x8_r6); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 7), i4_out_horz_8x8_r7); |
| |
| /* for the next 4 elements interleaved format */ |
| pred0_in = vget_high_u8(pred0_inp_full); |
| pred1_in = vget_high_u8(pred1_inp_full); |
| pred2_in = vget_high_u8(pred2_inp_full); |
| pred3_in = vget_high_u8(pred3_inp_full); |
| pred4_in = vget_high_u8(pred4_inp_full); |
| pred5_in = vget_high_u8(pred5_inp_full); |
| pred6_in = vget_high_u8(pred6_inp_full); |
| pred7_in = vget_high_u8(pred7_inp_full); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| pred4 = vreinterpretq_s16_u16(vmovl_u8(pred4_in)); |
| pred5 = vreinterpretq_s16_u16(vmovl_u8(pred5_in)); |
| pred6 = vreinterpretq_s16_u16(vmovl_u8(pred6_in)); |
| pred7 = vreinterpretq_s16_u16(vmovl_u8(pred7_in)); |
| |
| pi2_rsd = pi2_rsd + 8; |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 7)); |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| pred4 = vaddq_s16(pred4, resd4_in); |
| pred5 = vaddq_s16(pred5, resd5_in); |
| pred6 = vaddq_s16(pred6, resd6_in); |
| pred7 = vaddq_s16(pred7, resd7_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| pred4_in = vqmovun_s16(pred4); |
| pred5_in = vqmovun_s16(pred5); |
| pred6_in = vqmovun_s16(pred6); |
| pred7_in = vqmovun_s16(pred7); |
| |
| pu1_out = pu1_out + 8; |
| i4_out_horz_8x8_r0 = vld1_u8(pu1_out); |
| i4_out_horz_8x8_r1 = vld1_u8(pu1_out + out_strd); |
| i4_out_horz_8x8_r2 = vld1_u8(pu1_out + out_strd * 2); |
| i4_out_horz_8x8_r3 = vld1_u8(pu1_out + out_strd * 3); |
| i4_out_horz_8x8_r4 = vld1_u8(pu1_out + out_strd * 4); |
| i4_out_horz_8x8_r5 = vld1_u8(pu1_out + out_strd * 5); |
| i4_out_horz_8x8_r6 = vld1_u8(pu1_out + out_strd * 6); |
| i4_out_horz_8x8_r7 = vld1_u8(pu1_out + out_strd * 7); |
| |
| i4_out_horz_8x8_r0 = vbsl_u8(chroma_mask_8x8, pred0_in, i4_out_horz_8x8_r0); |
| i4_out_horz_8x8_r1 = vbsl_u8(chroma_mask_8x8, pred1_in, i4_out_horz_8x8_r1); |
| i4_out_horz_8x8_r2 = vbsl_u8(chroma_mask_8x8, pred2_in, i4_out_horz_8x8_r2); |
| i4_out_horz_8x8_r3 = vbsl_u8(chroma_mask_8x8, pred3_in, i4_out_horz_8x8_r3); |
| i4_out_horz_8x8_r4 = vbsl_u8(chroma_mask_8x8, pred4_in, i4_out_horz_8x8_r4); |
| i4_out_horz_8x8_r5 = vbsl_u8(chroma_mask_8x8, pred5_in, i4_out_horz_8x8_r5); |
| i4_out_horz_8x8_r6 = vbsl_u8(chroma_mask_8x8, pred6_in, i4_out_horz_8x8_r6); |
| i4_out_horz_8x8_r7 = vbsl_u8(chroma_mask_8x8, pred7_in, i4_out_horz_8x8_r7); |
| |
| vst1_u8((uint8_t *) (pu1_out), i4_out_horz_8x8_r0); |
| vst1_u8((uint8_t *) (pu1_out + out_strd), i4_out_horz_8x8_r1); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 2), i4_out_horz_8x8_r2); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 3), i4_out_horz_8x8_r3); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 4), i4_out_horz_8x8_r4); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 5), i4_out_horz_8x8_r5); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 6), i4_out_horz_8x8_r6); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 7), i4_out_horz_8x8_r7); |
| |
| pu1_out = pu1_out_ptr; |
| pu1_pred = pu1_pred_ptr; |
| pi2_rsd = pi2_rsd_ptr; |
| } |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_pred_residual_recon_chroma_4x4_neonintr */ |
| /* */ |
| /* Description : this function computes the recon data from the */ |
| /* pred and residual buffer */ |
| /* */ |
| /* 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 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| void isvcd_pred_residual_recon_chroma_4x4_neonintr(UWORD8 *pu1_pred, WORD16 *pi2_rsd, |
| UWORD8 *pu1_out, WORD32 pred_strd, |
| WORD32 rsd_strd, WORD32 out_strd) |
| { |
| uint8x8_t pred0_in, pred1_in, pred2_in, pred3_in; |
| int16x8_t pred0, pred1, pred2, pred3; |
| int16x8_t resd0_in, resd1_in, resd2_in, resd3_in; |
| |
| uint8x8_t i4_out_horz_8x8_r0, i4_out_horz_8x8_r1, i4_out_horz_8x8_r2, i4_out_horz_8x8_r3; |
| uint8x8_t chroma_mask_8x8 = vreinterpret_u8_u16(vdup_n_u16(0x00ff)); |
| |
| UWORD8 *pu1_pred_ptr = pu1_pred; |
| WORD16 *pi2_rsd_ptr = pi2_rsd; |
| UWORD8 *pu1_out_ptr = pu1_out; |
| |
| pred0_in = vld1_u8((uint8_t *) pu1_pred); |
| pred1_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd)); |
| pred2_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 2)); |
| pred3_in = vld1_u8((uint8_t *) pu1_pred + (pred_strd * 3)); |
| |
| pred0 = vreinterpretq_s16_u16(vmovl_u8(pred0_in)); |
| pred1 = vreinterpretq_s16_u16(vmovl_u8(pred1_in)); |
| pred2 = vreinterpretq_s16_u16(vmovl_u8(pred2_in)); |
| pred3 = vreinterpretq_s16_u16(vmovl_u8(pred3_in)); |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 3)); |
| |
| pred0 = vaddq_s16(pred0, resd0_in); |
| pred1 = vaddq_s16(pred1, resd1_in); |
| pred2 = vaddq_s16(pred2, resd2_in); |
| pred3 = vaddq_s16(pred3, resd3_in); |
| |
| pred0_in = vqmovun_s16(pred0); |
| pred1_in = vqmovun_s16(pred1); |
| pred2_in = vqmovun_s16(pred2); |
| pred3_in = vqmovun_s16(pred3); |
| |
| i4_out_horz_8x8_r0 = vld1_u8(pu1_out); |
| i4_out_horz_8x8_r1 = vld1_u8(pu1_out + out_strd); |
| i4_out_horz_8x8_r2 = vld1_u8(pu1_out + out_strd * 2); |
| i4_out_horz_8x8_r3 = vld1_u8(pu1_out + out_strd * 3); |
| |
| i4_out_horz_8x8_r0 = vbsl_u8(chroma_mask_8x8, pred0_in, i4_out_horz_8x8_r0); |
| i4_out_horz_8x8_r1 = vbsl_u8(chroma_mask_8x8, pred1_in, i4_out_horz_8x8_r1); |
| i4_out_horz_8x8_r2 = vbsl_u8(chroma_mask_8x8, pred2_in, i4_out_horz_8x8_r2); |
| i4_out_horz_8x8_r3 = vbsl_u8(chroma_mask_8x8, pred3_in, i4_out_horz_8x8_r3); |
| |
| vst1_u8((uint8_t *) (pu1_out), i4_out_horz_8x8_r0); |
| vst1_u8((uint8_t *) (pu1_out + out_strd), i4_out_horz_8x8_r1); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 2), i4_out_horz_8x8_r2); |
| vst1_u8((uint8_t *) (pu1_out + out_strd * 3), i4_out_horz_8x8_r3); |
| |
| pu1_out = pu1_out_ptr; |
| pu1_pred = pu1_pred_ptr; |
| pi2_rsd = pi2_rsd_ptr; |
| } |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_residual_luma_4x4_neonintr */ |
| /* */ |
| /* Description : this function computes the nnz from resd */ |
| /* */ |
| /* Inputs : */ |
| /* Globals : none */ |
| /* Processing : */ |
| /* */ |
| /* Outputs : none */ |
| /* Returns : nnz */ |
| /* */ |
| /* Issues : none */ |
| /* */ |
| /* Revision History: */ |
| /* */ |
| /* DD MM YYYY Author(s) Changes (Describe the changes made) */ |
| /* 25 11 2021 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| WORD32 isvcd_residual_luma_4x4_neonintr(WORD16 *pi2_rsd, WORD32 rsd_strd) |
| { |
| int16x8_t resd0_in, resd1_in, resd2_in, resd3_in; |
| int16x8_t resd01_in, resd23_in; |
| WORD32 i4_nnz; |
| int16x8_t dup_val_1, dup_val_2, dup_abs; |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 3)); |
| |
| resd01_in = vreinterpretq_s16_s64(vcombine_s64(vget_low_s64(vreinterpretq_s64_s16(resd0_in)), |
| vget_low_s64(vreinterpretq_s64_s16(resd1_in)))); |
| |
| resd23_in = vreinterpretq_s16_s64(vcombine_s64(vget_low_s64(vreinterpretq_s64_s16(resd2_in)), |
| vget_low_s64(vreinterpretq_s64_s16(resd3_in)))); |
| |
| dup_val_1 = vabsq_s16(resd01_in); |
| dup_val_2 = vabsq_s16(resd23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| i4_nnz = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| return i4_nnz; |
| } |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_residual_luma_8x8_neonintr */ |
| /* */ |
| /* Description : this function computes the nnz from resd */ |
| /* */ |
| /* Inputs : */ |
| /* Globals : none */ |
| /* Processing : */ |
| /* */ |
| /* Outputs : none */ |
| /* Returns : nnz */ |
| /* */ |
| /* Issues : none */ |
| /* */ |
| /* Revision History: */ |
| /* */ |
| /* DD MM YYYY Author(s) Changes (Describe the changes made) */ |
| /* 25 11 2021 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| WORD32 isvcd_residual_luma_8x8_neonintr(WORD16 *pi2_rsd, WORD32 rsd_strd) |
| { |
| int16x8_t resd0_in, resd1_in, resd2_in, resd3_in; |
| int16x8_t resd4_in, resd5_in, resd6_in, resd7_in; |
| |
| int64x2_t resd0_in_64x2, resd1_in_64x2, resd2_in_64x2, resd3_in_64x2, resd4_in_64x2, |
| resd5_in_64x2, resd6_in_64x2, resd7_in_64x2; |
| |
| int16x8_t resd_b0_r01_in; |
| int16x8_t resd_b0_r23_in; |
| int16x8_t resd_b1_r01_in; |
| int16x8_t resd_b1_r23_in; |
| int16x8_t resd_b2_r45_in; |
| int16x8_t resd_b2_r67_in; |
| int16x8_t resd_b3_r45_in; |
| int16x8_t resd_b3_r67_in; |
| |
| int16x8_t dup_val_1, dup_val_2, dup_abs; |
| WORD32 nnz, nnz_b0, nnz_b1, nnz_b2, nnz_b3; |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz = (nnz_b0 | (nnz_b1 << 1) | (nnz_b2 << 4) | (nnz_b3 << 5)); |
| |
| return nnz; |
| } |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_residual_luma_16x16_neonintr */ |
| /* */ |
| /* Description : this function computes the nnz from resd */ |
| /* */ |
| /* Inputs : */ |
| /* Globals : none */ |
| /* Processing : */ |
| /* */ |
| /* Outputs : none */ |
| /* Returns : nnz */ |
| /* */ |
| /* Issues : none */ |
| /* */ |
| /* Revision History: */ |
| /* */ |
| /* DD MM YYYY Author(s) Changes (Describe the changes made) */ |
| /* 25 11 2021 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| WORD32 isvcd_residual_luma_16x16_neonintr(WORD16 *pi2_rsd, WORD32 rsd_strd) |
| { |
| int16x8_t resd0_in, resd1_in, resd2_in, resd3_in; |
| int16x8_t resd4_in, resd5_in, resd6_in, resd7_in; |
| |
| WORD16 *pi2_rsd_ptr = pi2_rsd; |
| int64x2_t resd0_in_64x2, resd1_in_64x2, resd2_in_64x2, resd3_in_64x2, resd4_in_64x2, |
| resd5_in_64x2, resd6_in_64x2, resd7_in_64x2; |
| |
| int16x8_t resd_b0_r01_in; |
| int16x8_t resd_b0_r23_in; |
| int16x8_t resd_b1_r01_in; |
| int16x8_t resd_b1_r23_in; |
| int16x8_t resd_b2_r45_in; |
| int16x8_t resd_b2_r67_in; |
| int16x8_t resd_b3_r45_in; |
| int16x8_t resd_b3_r67_in; |
| |
| int16x8_t dup_val_1, dup_val_2, dup_abs; |
| WORD32 nnz, nnz_b0, nnz_b1, nnz_b2, nnz_b3; |
| |
| /* First row of 8, first 8x8 elements */ |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd_ptr); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd_ptr + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz = (nnz_b0 | (nnz_b1 << 1) | (nnz_b2 << 4) | (nnz_b3 << 5)); |
| |
| /* first row of 8, sec 8x8 elements */ |
| pi2_rsd_ptr = pi2_rsd_ptr + 8; |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd_ptr); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd_ptr + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz |= (nnz_b0 << 2 | (nnz_b1 << 3) | (nnz_b2 << 6) | (nnz_b3 << 7)); |
| |
| pi2_rsd_ptr = pi2_rsd + (8 * rsd_strd); |
| /*Sec row of 8, first 8x8*/ |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd_ptr); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd_ptr + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz |= (nnz_b0 << 8 | (nnz_b1 << 9) | (nnz_b2 << 12) | (nnz_b3 << 13)); |
| |
| /*Sec row of 8, Sec 8x8*/ |
| pi2_rsd_ptr = pi2_rsd_ptr + 8; |
| |
| resd0_in = vld1q_s16((int16_t *) pi2_rsd_ptr); |
| resd1_in = vld1q_s16((int16_t *) pi2_rsd_ptr + rsd_strd); |
| resd2_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 2)); |
| resd3_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 3)); |
| resd4_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 4)); |
| resd5_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 5)); |
| resd6_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 6)); |
| resd7_in = vld1q_s16((int16_t *) pi2_rsd_ptr + (rsd_strd * 7)); |
| |
| resd0_in_64x2 = vreinterpretq_s64_s16(resd0_in); |
| resd1_in_64x2 = vreinterpretq_s64_s16(resd1_in); |
| resd2_in_64x2 = vreinterpretq_s64_s16(resd2_in); |
| resd3_in_64x2 = vreinterpretq_s64_s16(resd3_in); |
| resd4_in_64x2 = vreinterpretq_s64_s16(resd4_in); |
| resd5_in_64x2 = vreinterpretq_s64_s16(resd5_in); |
| resd6_in_64x2 = vreinterpretq_s64_s16(resd6_in); |
| resd7_in_64x2 = vreinterpretq_s64_s16(resd7_in); |
| |
| resd_b0_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_in_64x2), vget_low_s64(resd1_in_64x2))); |
| resd_b0_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_in_64x2), vget_low_s64(resd3_in_64x2))); |
| resd_b1_r01_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_in_64x2), vget_high_s64(resd1_in_64x2))); |
| resd_b1_r23_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_in_64x2), vget_high_s64(resd3_in_64x2))); |
| resd_b2_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_in_64x2), vget_low_s64(resd5_in_64x2))); |
| resd_b2_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_in_64x2), vget_low_s64(resd7_in_64x2))); |
| resd_b3_r45_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_in_64x2), vget_high_s64(resd5_in_64x2))); |
| resd_b3_r67_in = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_in_64x2), vget_high_s64(resd7_in_64x2))); |
| dup_val_1 = vabsq_s16(resd_b0_r01_in); |
| dup_val_2 = vabsq_s16(resd_b0_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_in); |
| dup_val_2 = vabsq_s16(resd_b1_r23_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_in); |
| dup_val_2 = vabsq_s16(resd_b2_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_in); |
| dup_val_2 = vabsq_s16(resd_b3_r67_in); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz |= (nnz_b0 << 10 | (nnz_b1 << 11) | (nnz_b2 << 14) | (nnz_b3 << 15)); |
| return nnz; |
| } |
| |
| /*****************************************************************************/ |
| /* */ |
| /* Function Name : isvcd_residual_chroma_cb_cr_8x8_neonintr */ |
| /* */ |
| /* Description : this function computes the nnz from resd */ |
| /* */ |
| /* Inputs : */ |
| /* Globals : none */ |
| /* Processing : */ |
| /* */ |
| /* Outputs : none */ |
| /* Returns : nnz */ |
| /* */ |
| /* Issues : none */ |
| /* */ |
| /* Revision History: */ |
| /* */ |
| /* DD MM YYYY Author(s) Changes (Describe the changes made) */ |
| /* 25 11 2021 Kishore creation */ |
| /* */ |
| /*****************************************************************************/ |
| |
| WORD32 isvcd_residual_chroma_cb_cr_8x8_neonintr(WORD16 *pi2_rsd, WORD32 rsd_strd) |
| { |
| int16x8x2_t resd0_in, resd1_in, resd2_in, resd3_in; |
| int16x8x2_t resd4_in, resd5_in, resd6_in, resd7_in; |
| |
| int64x2_t resd0_cr_64x2, resd1_cr_64x2, resd2_cr_64x2, resd3_cr_64x2, resd4_cr_64x2, |
| resd5_cr_64x2, resd6_cr_64x2, resd7_cr_64x2; |
| |
| int16x8_t resd_b0_r01_cr; |
| int16x8_t resd_b0_r23_cr; |
| int16x8_t resd_b1_r01_cr; |
| int16x8_t resd_b1_r23_cr; |
| int16x8_t resd_b2_r45_cr; |
| int16x8_t resd_b2_r67_cr; |
| int16x8_t resd_b3_r45_cr; |
| int16x8_t resd_b3_r67_cr; |
| |
| int64x2_t resd0_cb_64x2, resd1_cb_64x2, resd2_cb_64x2, resd3_cb_64x2, resd4_cb_64x2, |
| resd5_cb_64x2, resd6_cb_64x2, resd7_cb_64x2; |
| |
| int16x8_t resd_b0_r01_cb; |
| int16x8_t resd_b0_r23_cb; |
| int16x8_t resd_b1_r01_cb; |
| int16x8_t resd_b1_r23_cb; |
| int16x8_t resd_b2_r45_cb; |
| int16x8_t resd_b2_r67_cb; |
| int16x8_t resd_b3_r45_cb; |
| int16x8_t resd_b3_r67_cb; |
| |
| WORD32 nnz, nnz_b0, nnz_b1, nnz_b2, nnz_b3; |
| int16x8_t dup_val_1, dup_val_2, dup_abs; |
| |
| resd0_in = vld2q_s16((int16_t *) pi2_rsd); |
| resd1_in = vld2q_s16((int16_t *) pi2_rsd + rsd_strd); |
| resd2_in = vld2q_s16((int16_t *) pi2_rsd + (rsd_strd * 2)); |
| resd3_in = vld2q_s16((int16_t *) pi2_rsd + (rsd_strd * 3)); |
| resd4_in = vld2q_s16((int16_t *) pi2_rsd + (rsd_strd * 4)); |
| resd5_in = vld2q_s16((int16_t *) pi2_rsd + (rsd_strd * 5)); |
| resd6_in = vld2q_s16((int16_t *) pi2_rsd + (rsd_strd * 6)); |
| resd7_in = vld2q_s16((int16_t *) pi2_rsd + (rsd_strd * 7)); |
| |
| resd0_cb_64x2 = vreinterpretq_s64_s16(resd0_in.val[0]); |
| resd1_cb_64x2 = vreinterpretq_s64_s16(resd1_in.val[0]); |
| resd2_cb_64x2 = vreinterpretq_s64_s16(resd2_in.val[0]); |
| resd3_cb_64x2 = vreinterpretq_s64_s16(resd3_in.val[0]); |
| resd4_cb_64x2 = vreinterpretq_s64_s16(resd4_in.val[0]); |
| resd5_cb_64x2 = vreinterpretq_s64_s16(resd5_in.val[0]); |
| resd6_cb_64x2 = vreinterpretq_s64_s16(resd6_in.val[0]); |
| resd7_cb_64x2 = vreinterpretq_s64_s16(resd7_in.val[0]); |
| |
| resd_b0_r01_cb = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_cb_64x2), vget_low_s64(resd1_cb_64x2))); |
| resd_b0_r23_cb = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_cb_64x2), vget_low_s64(resd3_cb_64x2))); |
| resd_b1_r01_cb = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_cb_64x2), vget_high_s64(resd1_cb_64x2))); |
| resd_b1_r23_cb = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_cb_64x2), vget_high_s64(resd3_cb_64x2))); |
| resd_b2_r45_cb = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_cb_64x2), vget_low_s64(resd5_cb_64x2))); |
| resd_b2_r67_cb = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_cb_64x2), vget_low_s64(resd7_cb_64x2))); |
| resd_b3_r45_cb = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_cb_64x2), vget_high_s64(resd5_cb_64x2))); |
| resd_b3_r67_cb = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_cb_64x2), vget_high_s64(resd7_cb_64x2))); |
| |
| resd0_cr_64x2 = vreinterpretq_s64_s16(resd0_in.val[1]); |
| resd1_cr_64x2 = vreinterpretq_s64_s16(resd1_in.val[1]); |
| resd2_cr_64x2 = vreinterpretq_s64_s16(resd2_in.val[1]); |
| resd3_cr_64x2 = vreinterpretq_s64_s16(resd3_in.val[1]); |
| resd4_cr_64x2 = vreinterpretq_s64_s16(resd4_in.val[1]); |
| resd5_cr_64x2 = vreinterpretq_s64_s16(resd5_in.val[1]); |
| resd6_cr_64x2 = vreinterpretq_s64_s16(resd6_in.val[1]); |
| resd7_cr_64x2 = vreinterpretq_s64_s16(resd7_in.val[1]); |
| |
| resd_b0_r01_cr = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd0_cr_64x2), vget_low_s64(resd1_cr_64x2))); |
| resd_b0_r23_cr = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd2_cr_64x2), vget_low_s64(resd3_cr_64x2))); |
| resd_b1_r01_cr = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd0_cr_64x2), vget_high_s64(resd1_cr_64x2))); |
| resd_b1_r23_cr = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd2_cr_64x2), vget_high_s64(resd3_cr_64x2))); |
| resd_b2_r45_cr = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd4_cr_64x2), vget_low_s64(resd5_cr_64x2))); |
| resd_b2_r67_cr = vreinterpretq_s16_s64( |
| vcombine_s64(vget_low_s64(resd6_cr_64x2), vget_low_s64(resd7_cr_64x2))); |
| resd_b3_r45_cr = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd4_cr_64x2), vget_high_s64(resd5_cr_64x2))); |
| resd_b3_r67_cr = vreinterpretq_s16_s64( |
| vcombine_s64(vget_high_s64(resd6_cr_64x2), vget_high_s64(resd7_cr_64x2))); |
| |
| dup_val_1 = vabsq_s16(resd_b0_r01_cr); |
| dup_val_2 = vabsq_s16(resd_b0_r23_cr); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_cr); |
| dup_val_2 = vabsq_s16(resd_b1_r23_cr); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_cr); |
| dup_val_2 = vabsq_s16(resd_b2_r67_cr); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_cr); |
| dup_val_2 = vabsq_s16(resd_b3_r67_cr); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz = ((nnz_b0 | (nnz_b1 << 1) | (nnz_b2 << 2) | (nnz_b3 << 3)) << 4); |
| |
| dup_val_1 = vabsq_s16(resd_b0_r01_cb); |
| dup_val_2 = vabsq_s16(resd_b0_r23_cb); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b0 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b1_r01_cb); |
| dup_val_2 = vabsq_s16(resd_b1_r23_cb); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b1 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b2_r45_cb); |
| dup_val_2 = vabsq_s16(resd_b2_r67_cb); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b2 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| dup_val_1 = vabsq_s16(resd_b3_r45_cb); |
| dup_val_2 = vabsq_s16(resd_b3_r67_cb); |
| dup_abs = vqaddq_s16(dup_val_1, dup_val_2); |
| nnz_b3 = dup_abs[0] || dup_abs[1] || dup_abs[2] || dup_abs[3] || dup_abs[4] || dup_abs[5] || |
| dup_abs[6] || dup_abs[7]; |
| |
| nnz |= ((nnz_b0 | (nnz_b1 << 1) | (nnz_b2 << 2) | (nnz_b3 << 3))); |
| return nnz; |
| } |
