/external/libavc/encoder/armv8/ |
D | ime_distortion_metrics_av8.s | 104 movi v30.8h, #0 113 uabal v30.8h, v0.8b, v1.8b 114 uabal2 v30.8h, v0.16b, v1.16b 116 uabal v30.8h, v2.8b, v3.8b 117 uabal2 v30.8h, v2.16b, v3.16b 124 uabal v30.8h, v4.8b, v5.8b 125 uabal2 v30.8h, v4.16b, v5.16b 127 uabal v30.8h, v6.8b, v7.8b 128 uabal2 v30.8h, v6.16b, v7.16b 134 addp v30.8h, v30.8h, v30.8h [all …]
|
D | ih264e_evaluate_intra16x16_modes_av8.s | 127 dup v30.16b, w6 145 dup v30.16b, v0.b[0] 182 uabdl v22.8h, v0.8b, v30.8b 203 uabal v22.8h, v2.8b, v30.8b 219 uabal v22.8h, v4.8b, v30.8b 235 uabal v22.8h, v6.8b, v30.8b 252 uabal v22.8h, v0.8b, v30.8b 268 uabal v22.8h, v2.8b, v30.8b 285 uabal v22.8h, v4.8b, v30.8b 302 uabal v22.8h, v6.8b, v30.8b [all …]
|
D | ih264e_half_pel_av8.s | 108 ext v30.8b, v3.8b , v4.8b , #5 112 uaddl v10.8h, v30.8b, v3.8b //// a0 + a5 (column2,row0) 122 ext v30.8b, v3.8b , v4.8b , #2 125 umlal v10.8h, v30.8b, v1.8b //// a0 + a5 + 20a2 (column2,row0) 135 ext v30.8b, v3.8b , v4.8b , #3 138 umlal v10.8h, v30.8b, v1.8b //// a0 + a5 + 20a2 + 20a3 (column2,row0) 148 ext v30.8b, v3.8b , v4.8b , #1 151 umlsl v10.8h, v30.8b, v0.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 (column2,row0) 161 ext v30.8b, v3.8b , v4.8b , #4 164 umlsl v10.8h, v30.8b, v0.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 (column2,row0) [all …]
|
/external/libhevc/common/arm64/ |
D | ihevc_inter_pred_filters_luma_vert_w16inp.s | 202 smull v30.4s, v4.4h, v23.4h 203 smlal v30.4s, v3.4h, v22.4h 204 smlal v30.4s, v5.4h, v24.4h 205 smlal v30.4s, v6.4h, v25.4h 207 smlal v30.4s, v7.4h, v26.4h 209 smlal v30.4s, v16.4h, v27.4h 211 smlal v30.4s, v17.4h, v28.4h 213 smlal v30.4s, v18.4h, v29.4h 242 sqshrn v30.4h, v30.4s,#6 260 sqrshrun v30.8b, v30.8h,#6 [all …]
|
D | ihevc_inter_pred_chroma_vert_w16inp.s | 159 sqshrn v30.4h, v7.4s,#6 //right shift 161 sqrshrun v30.8b, v30.8h,#6 //rounding shift 163 st1 {v30.s}[0],[x7] //stores the loaded value 192 smull v30.4s, v0.4h, v16.4h //vmull_s16(src_tmp1, coeff_0) 194 smlal v30.4s, v1.4h, v17.4h 195 smlal v30.4s, v2.4h, v18.4h 197 smlal v30.4s, v3.4h, v19.4h 211 sqshrn v30.4h, v30.4s,#6 //right shift 220 sqrshrun v30.8b, v30.8h,#6 //rounding shift 225 st1 {v30.s}[0],[x1],#4 //stores the loaded value [all …]
|
D | ihevc_inter_pred_chroma_vert_w16inp_w16out.s | 159 sqshrn v30.4h, v7.4s,#6 //right shift 161 st1 {v30.2s},[x7] //stores the loaded value 191 smull v30.4s, v0.4h, v16.4h //vmull_s16(src_tmp1, coeff_0) 193 smlal v30.4s, v1.4h, v17.4h 194 smlal v30.4s, v2.4h, v18.4h 196 smlal v30.4s, v3.4h, v19.4h 209 sqshrn v30.4h, v30.4s,#6 //right shift 223 st1 {v30.2s},[x1],#8 //stores the loaded value 238 smull v30.4s, v0.4h, v16.4h //vmull_s16(src_tmp1, coeff_0) 240 smlal v30.4s, v1.4h, v17.4h [all …]
|
D | ihevc_inter_pred_filters_luma_vert.s | 228 umull v30.8h, v4.8b, v23.8b 230 umlsl v30.8h, v3.8b, v22.8b 231 umlsl v30.8h, v5.8b, v24.8b 232 umlal v30.8h, v6.8b, v25.8b 234 umlal v30.8h, v7.8b, v26.8b 236 umlsl v30.8h, v16.8b, v27.8b 238 umlal v30.8h, v17.8b, v28.8b 240 umlsl v30.8h, v18.8b, v29.8b 278 sqrshrun v30.8b, v30.8h,#6 291 st1 {v30.8b},[x14],x6 [all …]
|
D | ihevc_sao_edge_offset_class1.s | 134 LD1 {v30.16b},[x12],#16 //vld1q_u8(pu1_src[(ht - 1) * src_strd]) 137 ST1 { v30.16b},[x3],#16 //vst1q_u8(pu1_src_top[col]) 149 LD1 {v30.16b},[x6] //II pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd) 161 cmhi v22.16b, v18.16b , v30.16b //II vcgtq_u8(pu1_cur_row, pu1_top_row) 165 cmhi v24.16b, v30.16b , v18.16b //II vcltq_u8(pu1_cur_row, pu1_top_row) 188 mov v3.16b, v30.16b //II pu1_cur_row = pu1_next_row 211 xtn v30.8b, v26.8h //II vmovn_s16(pi2_tmp_cur_row.val[0]) 213 xtn2 v30.16b, v28.8h //II vmovn_s16(pi2_tmp_cur_row.val[1]) 215 ST1 { v30.16b},[x10],x1 //II vst1q_u8(pu1_src_cpy, pu1_cur_row) 245 xtn v30.8b, v26.8h //vmovn_s16(pi2_tmp_cur_row.val[0]) [all …]
|
D | ihevc_sao_band_offset_chroma.s | 140 LD1 {v30.8b},[x7] //pi1_sao_offset_u load 143 dup v29.8b, v30.b[1] //vdup_n_u8(pi1_sao_offset_u[1]) 146 dup v28.8b, v30.b[2] //vdup_n_u8(pi1_sao_offset_u[2]) 149 dup v27.8b, v30.b[3] //vdup_n_u8(pi1_sao_offset_u[3]) 153 dup v26.8b, v30.b[4] //vdup_n_u8(pi1_sao_offset_u[4]) 157 movi v30.8b, #16 //vdup_n_u8(16) 173 cmhs v13.8b, v30.8b , v4.8b //vcle_u8(band_table.val[3], vdup_n_u8(16)) 182 cmhs v14.8b, v30.8b , v3.8b //vcle_u8(band_table.val[2], vdup_n_u8(16)) 192 cmhs v15.8b, v30.8b , v2.8b //vcle_u8(band_table.val[1], vdup_n_u8(16)) 202 cmhs v16.8b, v30.8b , v1.8b //vcle_u8(band_table.val[0], vdup_n_u8(16)) [all …]
|
D | ihevc_itrans_recon_32x32.s | 219 smull v30.4s, v8.4h, v1.h[3] //// y1 * sin1(part of b3) 224 smlal v30.4s, v9.4h, v5.h[1] //// y1 * sin1 - y3 * sin3(part of b3) 259 smlsl v30.4s, v14.4h, v7.h[1] 265 smlsl v30.4s, v15.4h, v3.h[3] 289 smlsl v30.4s, v8.4h, v0.h[1] //// y1 * sin1(part of b3) 294 smlsl v30.4s, v9.4h, v3.h[1] //// y1 * sin1 - y3 * sin3(part of b3) 333 smlsl v30.4s, v14.4h, v6.h[3] 339 smlal v30.4s, v15.4h, v5.h[3] 365 smlal v30.4s, v8.4h, v2.h[1] //// y1 * sin1(part of b3) 370 smlal v30.4s, v9.4h, v1.h[1] //// y1 * sin1 - y3 * sin3(part of b3) [all …]
|
D | ihevc_inter_pred_chroma_vert.s | 237 umull v30.8h, v5.8b, v1.8b //mul with coeff 1 239 umlsl v30.8h, v4.8b, v0.8b 241 umlal v30.8h, v6.8b, v2.8b 242 umlsl v30.8h, v7.8b, v3.8b 254 sqrshrun v30.8b, v30.8h,#6 260 st1 {v30.8b},[x1],#8 //stores the loaded value 288 umull v30.8h, v5.8b, v1.8b //mul with coeff 1 290 umlsl v30.8h, v4.8b, v0.8b 293 umlal v30.8h, v6.8b, v2.8b 298 umlsl v30.8h, v7.8b, v3.8b [all …]
|
D | ihevc_sao_edge_offset_class0_chroma.s | 174 LD1 {v30.16b},[x12] //II Iteration pu1_cur_row = vld1q_u8(pu1_src_cpy) 192 …EXT v28.16b, v28.16b , v30.16b,#14 //II pu1_cur_row_tmp = vextq_u8(pu1_cur_row_tmp, pu1_… 195 cmhi v26.16b, v30.16b , v28.16b //II vcgtq_u8(pu1_cur_row, pu1_cur_row_tmp) 198 cmhi v24.16b, v28.16b , v30.16b //II vcltq_u8(pu1_cur_row, pu1_cur_row_tmp) 216 …EXT v28.16b, v30.16b , v28.16b,#2 //II pu1_cur_row_tmp = vextq_u8(pu1_cur_row, pu1_cur_r… 225 cmhi v26.16b, v30.16b , v28.16b //II vcgtq_u8(pu1_cur_row, pu1_cur_row_tmp) 227 cmhi v24.16b, v28.16b , v30.16b //II vcltq_u8(pu1_cur_row, pu1_cur_row_tmp) 281 …Uxtl v28.8h, v30.8b //II pi2_tmp_cur_row.val[0] = vreinterpretq_s16_u16(vmovl_… 295 …Uxtl2 v30.8h, v30.16b //II pi2_tmp_cur_row.val[1] = vreinterpretq_s16_u16(vmovl_… 297 …SADDW v30.8h, v30.8h , v27.8b //II pi2_tmp_cur_row.val[1] = vaddw_s8(pi2_tmp_cur_row.val… [all …]
|
D | ihevc_inter_pred_filters_luma_vert_w16out.s | 182 umull v30.8h, v4.8b, v23.8b 184 umlsl v30.8h, v3.8b, v22.8b 185 umlsl v30.8h, v5.8b, v24.8b 186 umlal v30.8h, v6.8b, v25.8b 188 umlal v30.8h, v7.8b, v26.8b 190 umlsl v30.8h, v16.8b, v27.8b 192 umlal v30.8h, v17.8b, v28.8b 194 umlsl v30.8h, v18.8b, v29.8b 236 st1 {v30.16b},[x14],x6 287 umull v30.8h, v4.8b, v23.8b [all …]
|
D | ihevc_weighted_pred_bi_default.s | 208 sqadd v30.4h,v22.4h,v23.4h 209 sqadd v30.4h,v30.4h,v0.4h //vaddq_s32(i4_tmp1_t1, tmp_lvl_shift_t) iii iteration 214 mov v30.d[1],v31.d[0] 217 sqshrun v30.8b, v30.8h,#7 218 …st1 {v30.s}[0],[x14],x5 //store pu1_dst iii iteration … 220 st1 {v30.s}[1],[x14],x5 //store pu1_dst iv iteration 298 ld1 { v30.8h},[x12],x4 //load and increment the pi2_src2 ii iteration 300 sqadd v22.8h,v28.8h,v30.8h //vaddq_s32(i4_tmp2_t1, i4_tmp2_t2) 305 sqadd v30.8h,v16.8h,v18.8h 308 sqadd v30.8h,v30.8h,v0.8h //vaddq_s32(i4_tmp1_t1, tmp_lvl_shift_t) iii iteration [all …]
|
D | ihevc_itrans_recon_16x16.s | 258 smull v30.4s, v6.4h, v1.h[3] //// y1 * sin1(part of b3) 263 smlsl v30.4s, v7.4h, v2.h[3] //// y1 * sin1 - y3 * sin3(part of b3) 285 smlsl v30.4s, v8.4h, v0.h[3] 291 smlal v30.4s, v9.4h, v3.h[3] 334 smlal v30.4s, v6.4h, v0.h[1] //// y1 * sin1(part of b3) 339 smlal v30.4s, v7.4h, v3.h[1] //// y1 * sin1 - y3 * sin3(part of b3) 346 smlsl v30.4s, v8.4h, v1.h[1] 352 smlsl v30.4s, v9.4h, v2.h[1] 394 add v16.4s, v18.4s , v30.4s 395 sub v28.4s, v18.4s , v30.4s [all …]
|
D | ihevc_sao_edge_offset_class1_chroma.s | 157 LD1 {v30.16b},[x12],#16 //vld1q_u8(pu1_src[(ht - 1) * src_strd]) 161 ST1 { v30.16b},[x3],#16 //vst1q_u8(pu1_src_top[col]) 176 LD1 {v30.16b},[x6] //II pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd) 190 cmhi v22.16b, v18.16b , v30.16b //II vcgtq_u8(pu1_cur_row, pu1_top_row) 195 cmhi v24.16b, v30.16b , v18.16b //II vcltq_u8(pu1_cur_row, pu1_top_row) 229 mov v3.16b, v30.16b //II pu1_cur_row = pu1_next_row 259 xtn v30.8b, v26.8h //II vmovn_s16(pi2_tmp_cur_row.val[0]) 261 xtn2 v30.16b, v28.8h //II vmovn_s16(pi2_tmp_cur_row.val[1]) 263 ST1 { v30.16b},[x10],x1 //II vst1q_u8(pu1_src_cpy, pu1_cur_row) 305 xtn v30.8b, v26.8h //vmovn_s16(pi2_tmp_cur_row.val[0]) [all …]
|
/external/libavc/common/armv8/ |
D | ih264_inter_pred_luma_horz_qpel_vert_qpel_av8.s | 139 movi v30.8b, #20 // Filter coeff 20 158 umlal v24.8h, v4.8b, v30.8b 159 umlal v24.8h, v6.8b, v30.8b 169 umlal v28.8h, v20.8b, v30.8b 170 umlal v28.8h, v21.8b, v30.8b 175 umlal v24.8h, v5.8b, v30.8b 176 umlal v24.8h, v7.8b, v30.8b 190 umlal v24.8h, v20.8b, v30.8b 191 umlal v24.8h, v21.8b, v30.8b 196 umlal v16.8h, v6.8b, v30.8b [all …]
|
D | ih264_inter_pred_filters_luma_horz_av8.s | 134 ext v30.8b, v3.8b , v4.8b, #5 ////extract a[5] (column2,row0) 137 uaddl v10.8h, v30.8b, v3.8b //// a0 + a5 (column2,row0) 142 ext v30.8b, v3.8b , v4.8b, #2 ////extract a[2] (column2,row0) 145 umlal v10.8h, v30.8b, v1.8b //// a0 + a5 + 20a2 (column2,row0) 150 ext v30.8b, v3.8b , v4.8b, #3 ////extract a[3] (column2,row0) 153 umlal v10.8h, v30.8b, v1.8b //// a0 + a5 + 20a2 + 20a3 (column2,row0) 158 ext v30.8b, v3.8b , v4.8b, #1 ////extract a[1] (column2,row0) 161 umlsl v10.8h, v30.8b, v0.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 (column2,row0) 166 ext v30.8b, v3.8b , v4.8b, #4 ////extract a[4] (column2,row0) 169 umlsl v10.8h, v30.8b, v0.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 (column2,row0) [all …]
|
D | ih264_inter_pred_luma_horz_hpel_vert_hpel_av8.s | 96 movi v30.8h, #0x5 // Filter coeff 5 into Q12 123 mls v18.8h, v22.8h , v30.8h 125 mls v20.8h, v26.8h , v30.8h 128 mls v22.8h, v26.8h , v30.8h 141 smlsl v26.4s, v24.4h, v30.4h 145 smlsl2 v23.4s, v24.8h, v30.8h 166 smlsl v26.4s, v24.4h, v30.4h 170 smlsl2 v22.4s, v24.8h, v30.8h 204 mls v18.8h, v22.8h , v30.8h 206 mls v20.8h, v26.8h , v30.8h [all …]
|
D | ih264_deblk_luma_av8.s | 126 uabd v30.16b, v4.16b, v0.16b 133 cmhi v22.16b, v16.16b , v30.16b //Q11=(Aq<Beta) 136 usubl v30.8h, v1.8b, v7.8b // 140 shl v26.8h, v30.8h, #2 //Q13 = (q0 - p0)<<2 142 usubl v30.8h, v9.8b, v3.8b //Q15 = (p1 - q1)H 145 add v26.8h, v26.8h , v30.8h //Q13,Q12 = [ (q0 - p0)<<2 ] + (p1 - q1) 158 uaddl v30.8h, v17.8b, v5.8b // 172 sub v30.8h, v30.8h , v26.8h //Q15,Q2 = [q2 + (p0+q0+1)>>1] - (q1<<1) 176 sqshrn v31.8b, v30.8h, #1 // 177 sqshrn v30.8b, v4.8h, #1 //Q15 = i_macro_q1 [all …]
|
D | ih264_inter_pred_luma_horz_qpel_av8.s | 142 ext v30.8b, v3.8b , v4.8b , #5 145 uaddl v10.8h, v30.8b, v3.8b //// a0 + a5 (column2,row0) 150 ext v30.8b, v3.8b , v4.8b , #2 153 umlal v10.8h, v30.8b, v1.8b //// a0 + a5 + 20a2 (column2,row0) 158 ext v30.8b, v3.8b , v4.8b , #3 161 umlal v10.8h, v30.8b, v1.8b //// a0 + a5 + 20a2 + 20a3 (column2,row0) 166 ext v30.8b, v3.8b , v4.8b , #1 169 umlsl v10.8h, v30.8b, v0.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 (column2,row0) 174 ext v30.8b, v3.8b , v4.8b , #4 177 umlsl v10.8h, v30.8b, v0.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 (column2,row0) [all …]
|
D | ih264_inter_pred_chroma_av8.s | 147 dup v30.8b, w10 163 umlal v20.8h, v5.8b, v30.8b 171 umlal v22.8h, v6.8b, v30.8b 179 umlal v24.8h, v10.8b, v30.8b 186 umlal v16.8h, v11.8b, v30.8b 192 umlal v20.8h, v0.8b, v30.8b 199 umlal v24.8h, v1.8b, v30.8b 206 umlal v20.8h, v5.8b, v30.8b 210 umlal v22.8h, v6.8b, v30.8b 225 umlal v24.8h, v10.8b, v30.8b [all …]
|
D | ih264_inter_pred_luma_horz_qpel_vert_hpel_av8.s | 157 movi v30.8h, #0x5 // Filter coeff 5 into Q12 185 mls v18.8h, v22.8h , v30.8h 187 mls v20.8h, v26.8h , v30.8h 190 mls v22.8h, v26.8h , v30.8h 204 smlsl v26.4s, v24.4h, v30.4h 208 smlsl2 v22.4s, v24.8h, v30.8h 229 smlsl v26.4s, v24.4h, v30.4h 233 smlsl2 v22.4s, v24.8h, v30.8h 272 mls v18.8h, v22.8h , v30.8h 274 mls v20.8h, v26.8h , v30.8h [all …]
|
D | ih264_inter_pred_luma_vert_qpel_av8.s | 163 sqrshrun v30.8b, v14.8h, #5 // dst[0_0] = CLIP_U8((temp +16) >> 5) 170 urhadd v30.16b, v20.16b , v30.16b // Interpolation to obtain qpel value 174 st1 {v30.2s, v31.2s}, [x1], x3 // Vector store to dst[0_0] 178 sqrshrun v30.8b, v16.8h, #5 188 urhadd v30.16b, v14.16b , v30.16b // Interpolation to obtain qpel value 194 st1 {v30.2s, v31.2s}, [x1], x3 //store row 1 195 sqrshrun v30.8b, v18.8h, #5 205 urhadd v30.16b, v16.16b , v30.16b // Interpolation to obtain qpel value 208 st1 {v30.2s, v31.2s}, [x1], x3 //store row 2 211 sqrshrun v30.8b, v14.8h, #5 [all …]
|
/external/libmpeg2/common/armv8/ |
D | impeg2_idct.s | 166 dup v30.8h, w4 171 uaddw v8.8h, v30.8h , v0.8b 173 uaddw v10.8h, v30.8h , v1.8b 176 uaddw v12.8h, v30.8h , v2.8b 180 uaddw v14.8h, v30.8h , v3.8b 184 uaddw v16.8h, v30.8h , v4.8b 188 uaddw v18.8h, v30.8h , v5.8b 192 uaddw v20.8h, v30.8h , v6.8b 196 uaddw v22.8h, v30.8h , v7.8b 239 ld1 {v30.8b}, [x2], x5 [all …]
|