/external/libmpeg2/common/armv8/ |
H A D | icv_variance_av8.s | 86 umull v20.8h, v0.8b, v0.8b 91 uaddl v21.4s, v20.4h, v22.4h 93 uaddl2 v20.4s, v20.8h, v22.8h 96 add v20.4s, v20.4s, v21.4s 98 add v20.4s, v20.4s, v22.4s 99 addp v20.4s, v20 [all...] |
H A D | ideint_spatial_filter_av8.s | 70 movi v20.8h, #0 115 uabal v20.8h, v2.8b, v4.8b 127 addp v20.8h, v20.8h, v20.8h 131 uaddlp v20.2s, v20.4h 140 mul v20.2s, v20.2s, v31.2s 149 smov x7, v20 [all...] |
H A D | ideint_cac_av8.s | 142 cmhs v20.8h, v16.8h, v18.8h 143 and v20.16b, v16.16b, v20.16b 145 // v20 now contains 8 absolute diff of sums above the threshold 148 mov v21.d[0], v20.d[1] 149 add v20.4h, v20.4h, v21.4h 151 // v20 has four adj values for two sub-blocks 186 add v20.4h, v0.4h, v20 [all...] |
/external/libhevc/common/arm64/ |
H A D | ihevc_intra_pred_chroma_ver.s | 117 ld2 {v20.8b, v21.8b}, [x6],#16 //16 loads (col 0:15) 127 st2 {v20.8b, v21.8b}, [x2],#16 128 st2 {v20.8b, v21.8b}, [x5],#16 129 st2 {v20.8b, v21.8b}, [x8],#16 130 st2 {v20.8b, v21.8b}, [x10],#16 140 st2 {v20.8b, v21.8b}, [x2],#16 141 st2 {v20.8b, v21.8b}, [x5],#16 142 st2 {v20.8b, v21.8b}, [x8],#16 143 st2 {v20.8b, v21.8b}, [x10],#16 153 st2 {v20 [all...] |
H A D | ihevc_intra_pred_luma_dc.s | 237 uxtl v20.8h, v0.8b 240 add v20.8h, v20.8h , v24.8h //col 1::7 add 3dc+2 (prol) 243 sqshrun v2.8b, v20.8h,#2 //columns shx2 movn (prol) 263 bsl v20.8b, v3.8b , v16.8b //row 1 (prol) 268 st1 {v20.8b},[x2], x3 //store row 1 (prol) 280 bsl v20.8b, v3.8b , v16.8b //row 3 (prol) 285 st1 {v20.8b},[x2], x3 //store row 3 (prol) 296 bsl v20.8b, v3.8b , v16.8b //row 5 (prol) 301 st1 {v20 [all...] |
H A D | ihevc_inter_pred_filters_luma_vert_w16inp.s | 171 smull v20.4s, v2.4h, v23.4h //mul_res2 = vmull_u8(src_tmp3, coeffabs_1)// 174 smlal v20.4s, v1.4h, v22.4h //mul_res2 = vmlal_u8(mul_res2, src_tmp2, coeffabs_0)// 176 smlal v20.4s, v3.4h, v24.4h //mul_res2 = vmlal_u8(mul_res2, src_tmp4, coeffabs_2)// 178 smlal v20.4s, v4.4h, v25.4h //mul_res2 = vmlal_u8(mul_res2, src_tmp1, coeffabs_3)// 180 smlal v20.4s, v5.4h, v26.4h //mul_res2 = vmlal_u8(mul_res2, src_tmp2, coeffabs_4)// 182 smlal v20.4s, v6.4h, v27.4h //mul_res2 = vmlal_u8(mul_res2, src_tmp3, coeffabs_5)// 183 smlal v20.4s, v7.4h, v28.4h //mul_res2 = vmlal_u8(mul_res2, src_tmp4, coeffabs_6)// 184 smlal v20.4s, v16.4h, v29.4h //mul_res2 = vmlal_u8(mul_res2, src_tmp1, coeffabs_7)// 199 sqshrn v20.4h, v20 [all...] |
H A D | ihevc_inter_pred_luma_vert_w16inp_w16out.s | 182 smull v20.4s,v2.4h,v23.4h //mul_res2 = smull_u8(src_tmp3, coeffabs_1)// 185 smlal v20.4s,v1.4h,v22.4h //mul_res2 = smlal_u8(mul_res2, src_tmp2, coeffabs_0)// 187 smlal v20.4s,v3.4h,v24.4h //mul_res2 = smlal_u8(mul_res2, src_tmp4, coeffabs_2)// 189 smlal v20.4s,v4.4h,v25.4h //mul_res2 = smlal_u8(mul_res2, src_tmp1, coeffabs_3)// 191 smlal v20.4s,v5.4h,v26.4h //mul_res2 = smlal_u8(mul_res2, src_tmp2, coeffabs_4)// 193 smlal v20.4s,v6.4h,v27.4h //mul_res2 = smlal_u8(mul_res2, src_tmp3, coeffabs_5)// 194 smlal v20.4s,v7.4h,v28.4h //mul_res2 = smlal_u8(mul_res2, src_tmp4, coeffabs_6)// 195 smlal v20.4s,v16.4h,v29.4h //mul_res2 = smlal_u8(mul_res2, src_tmp1, coeffabs_7)// 210 sub v20.4s, v20 [all...] |
H A D | ihevc_sao_edge_offset_class1.s | 154 SUB v20.16b, v17.16b , v5.16b //sign_down = vreinterpretq_s8_u8(vsubq_u8(cmp_lt, cmp_gt)) 160 ADD v5.16b, v5.16b , v20.16b //edge_idx = vaddq_s8(edge_idx, sign_down) 163 NEG v16.16b, v20.16b //sign_up = vnegq_s8(sign_down) 177 Uxtl v20.8h, v3.8b //pi2_tmp_cur_row.val[0] = vreinterpretq_s16_u16(vmovl_u8(vget_low_u8(pu1_cur_row))) 179 SADDW v20.8h, v20.8h , v5.8b //pi2_tmp_cur_row.val[0] = vaddw_s8(pi2_tmp_cur_row.val[0], offset) 181 SMAX v20.8h, v20.8h , v2.8h //pi2_tmp_cur_row.val[0] = vmaxq_s16(pi2_tmp_cur_row.val[0], const_min_clip) 183 UMIN v20.8h, v20 [all...] |
H A D | ihevc_sao_edge_offset_class1_chroma.s | 183 SUB v20.16b, v19.16b , v5.16b //sign_down = vreinterpretq_s8_u8(vsubq_u8(cmp_lt, cmp_gt)) 189 ADD v5.16b, v5.16b , v20.16b //edge_idx = vaddq_s8(edge_idx, sign_down) 193 NEG v16.16b, v20.16b //sign_up = vnegq_s8(sign_down) 209 Uxtl v20.8h, v3.8b //pi2_tmp_cur_row.val[0] = vreinterpretq_s16_u16(vmovl_u8(vget_low_u8(pu1_cur_row))) 215 SADDW v20.8h, v20.8h , v5.8b //pi2_tmp_cur_row.val[0] = vaddw_s8(pi2_tmp_cur_row.val[0], offset) 217 SMAX v20.8h, v20.8h , v2.8h //pi2_tmp_cur_row.val[0] = vmaxq_s16(pi2_tmp_cur_row.val[0], const_min_clip) 219 UMIN v20.8h, v20 [all...] |
H A D | ihevc_inter_pred_filters_luma_vert.s | 182 umull v20.8h, v2.8b, v23.8b //mul_res2 = vmull_u8(src_tmp3, coeffabs_1)// 186 umlsl v20.8h, v1.8b, v22.8b //mul_res2 = vmlsl_u8(mul_res2, src_tmp2, coeffabs_0)// 190 umlsl v20.8h, v3.8b, v24.8b //mul_res2 = vmlsl_u8(mul_res2, src_tmp4, coeffabs_2)// 193 umlal v20.8h, v4.8b, v25.8b //mul_res2 = vmlal_u8(mul_res2, src_tmp1, coeffabs_3)// 196 umlal v20.8h, v5.8b, v26.8b //mul_res2 = vmlal_u8(mul_res2, src_tmp2, coeffabs_4)// 199 umlsl v20.8h, v6.8b, v27.8b //mul_res2 = vmlsl_u8(mul_res2, src_tmp3, coeffabs_5)// 202 umlal v20.8h, v7.8b, v28.8b //mul_res2 = vmlal_u8(mul_res2, src_tmp4, coeffabs_6)// 206 umlsl v20.8h, v16.8b, v29.8b //mul_res2 = vmlsl_u8(mul_res2, src_tmp1, coeffabs_7)// 224 sqrshrun v20.8b, v20 [all...] |
H A D | ihevc_itrans_recon_4x4_ttype1.s | 154 smull v20.4s, v2.4h, v4.h[1] // 55 * pi2_src[2] 155 smlsl v20.4s, v1.4h, v4.h[2] // 55 * pi2_src[2] - 74 * pi2_src[1] 156 smlsl v20.4s, v3.4h, v4.h[0] // - 74 * pi2_src[1] + 55 * pi2_src[2] - 29 * pi2_src[3] 157 smlal v20.4s, v0.4h, v4.h[3] //pi2_out[3] = 84 * pi2_src[0] - 74 * pi2_src[1] + 55 * pi2_src[2] - 29 * pi2_src[3] 162 sqrshrn v31.4h, v20.4s,#shift_stage1_idct // (pi2_out[3] + rounding ) >> shift_stage1_idct 198 smull v20.4s, v16.4h, v4.h[1] // 55 * pi2_src[2] 199 smlsl v20.4s, v22.4h, v4.h[2] // - 74 * pi2_src[1] + 55 * pi2_src[2] 200 smlsl v20.4s, v17.4h, v4.h[0] // - 74 * pi2_src[1] + 55 * pi2_src[2] - 29 * pi2_src[3] 201 smlal v20.4s, v21.4h, v4.h[3] //pi2_out[3] = 84 * pi2_src[0] - 74 * pi2_src[1] + 55 * pi2_src[2] - 29 * pi2_src[3] 206 sqrshrn v31.4h, v20 [all...] |
H A D | ihevc_intra_pred_luma_mode_3_to_9.s | 211 umull v20.8h, v14.8b, v7.8b //mul (row 2) 212 umlal v20.8h, v15.8b, v6.8b //mul (row 2) 219 rshrn v20.8b, v20.8h,#5 //round shft (row 2) 229 st1 {v20.8b},[x2], x3 //st (row 2) 255 umull v20.8h, v14.8b, v7.8b //mul (row 6) 256 umlal v20.8h, v15.8b, v6.8b //mul (row 6) 263 rshrn v20.8b, v20.8h,#5 //round shft (row 6) 266 st1 {v20 [all...] |
/external/llvm/test/MC/AArch64/ |
H A D | noneon-diagnostics.s | 5 fmla v1.2d, v30.2d, v20.2d 11 // CHECK-ERROR-NEXT: fmla v1.2d, v30.2d, v20.2d 18 fmls v1.2d, v30.2d, v20.2d 25 // CHECK-ERROR-NEXT: fmls v1.2d, v30.2d, v20.2d 33 fmls.2d v1, v30, v20 40 // CHECK-ERROR-NEXT: fmls.2d v1, v30, v20
|
H A D | neon-simd-copy.s | 11 ins v20.s[0], w30 16 mov v20.s[0], w30 21 // CHECK: {{mov|ins}} v20.s[0], w30 // encoding: [0xd4,0x1f,0x04,0x4e] 26 // CHECK: {{mov|ins}} v20.s[0], w30 // encoding: [0xd4,0x1f,0x04,0x4e] 94 dup v17.2s, v20.s[0] 97 dup v17.4s, v20.s[0] 102 // CHECK: {{mov|dup}} v17.2s, v20.s[0] // encoding: [0x91,0x06,0x04,0x0e] 105 // CHECK: {{mov|dup}} v17.4s, v20.s[0] // encoding: [0x91,0x06,0x04,0x4e]
|
/external/libavc/common/armv8/ |
H A D | ih264_inter_pred_luma_horz_hpel_vert_hpel_av8.s | 114 uaddl v20.8h, v4.8b, v6.8b 117 mla v18.8h, v20.8h , v28.8h 119 uaddl v20.8h, v1.8b, v11.8b 121 mla v20.8h, v24.8h , v28.8h 125 mls v20.8h, v26.8h , v30.8h 130 ext v24.16b, v18.16b , v20.16b , #4 131 ext v26.16b, v18.16b , v20.16b , #6 133 ext v23.16b, v18.16b , v20.16b , #10 135 ext v24.16b, v18.16b , v20.16b , #2 136 ext v26.16b, v18.16b , v20 [all...] |
H A D | ih264_inter_pred_luma_horz_qpel_vert_hpel_av8.s | 176 uaddl v20.8h, v4.8b, v6.8b 179 mla v18.8h, v20.8h , v28.8h 181 uaddl v20.8h, v1.8b, v11.8b 183 mla v20.8h, v24.8h , v28.8h 187 mls v20.8h, v26.8h , v30.8h 192 st1 {v20.4s}, [x9], #16 193 ext v24.16b, v18.16b , v20.16b , #4 194 ext v26.16b, v18.16b , v20.16b , #6 196 ext v22.16b, v18.16b , v20.16b , #10 198 ext v24.16b, v18.16b , v20 [all...] |
H A D | ih264_inter_pred_chroma_av8.s | 159 umull v20.8h, v0.8b, v28.8b 161 umlal v20.8h, v3.8b, v29.8b 163 umlal v20.8h, v5.8b, v30.8b 165 umlal v20.8h, v8.8b, v31.8b 166 sqrshrun v26.8b, v20.8h, #6 190 umull v20.8h, v10.8b, v28.8b 191 umlal v20.8h, v13.8b, v29.8b 192 umlal v20.8h, v0.8b, v30.8b 193 umlal v20.8h, v3.8b, v31.8b 194 sqrshrun v26.8b, v20 [all...] |
H A D | ih264_intra_pred_luma_16x16_av8.s | 219 dup v20.16b, v0.b[5] 226 st1 {v20.16b}, [x1], x3 323 dup v20.16b, w15 330 dup v20.8h, w11 337 uqshl v0.8h, v0.8h, v20.8h 339 dup v20.16b, v0.b[0] 343 st1 { v20.16b}, [x1], x3 344 st1 { v20.16b}, [x1], x3 345 st1 { v20.16b}, [x1], x3 346 st1 { v20 [all...] |
H A D | ih264_resi_trans_quant_av8.s | 142 sub v20.4h, v15.4h , v16.4h //x2 = x5-x6 145 shl v22.4h, v20.4h, #1 //u_shift(x2,1,shft) 152 add v25.4h, v23.4h , v20.4h //x6 = u_shift(x3,1,shft) + x2; 176 add v20.4s, v0.4s, v23.4s 183 sshl v20.4s, v20.4s, v24.4s //shift row 1 188 xtn v20.4h, v20.4s //narrow row 1 193 neg v24.8h, v20.8h //get negative 199 cmeq v0.4h, v20 [all...] |
H A D | ih264_inter_pred_filters_luma_vert_av8.s | 143 uaddl v20.8h, v1.8b, v11.8b // temp4 = src[0_8] + src[5_8] 145 mla v20.8h, v18.8h , v22.8h // temp4 += temp3 * 20 153 mls v20.8h, v26.8h , v24.8h // temp4 -= temp5 * 5 161 sqrshrun v31.8b, v20.8h, #5 // dst[0_8] = CLIP_U8((temp4 +16) >> 5) 167 uaddl v20.8h, v6.8b, v0.8b 174 mls v18.8h, v20.8h , v24.8h 180 uaddl v20.8h, v8.8b, v2.8b 189 mls v14.8h, v20.8h , v24.8h 196 uaddl v20.8h, v9.8b, v7.8b // temp4 = src[0_8] + src[5_8] 205 mla v20 [all...] |
/external/libavc/encoder/armv8/ |
H A D | ih264e_half_pel_av8.s | 174 sqrshrun v20.8b, v8.8h, #5 //// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column1,row0) 181 st1 {v20.8b, v21.8b}, [x1], #16 ////Store dest row0 307 uaddl v20.8h, v2.8b, v17.8b //// a0 + a5 (column1,row0) 309 umlal v20.8h, v8.8b, v1.8b //// a0 + a5 + 20a2 (column1,row0) 310 umlal v20.8h, v11.8b, v1.8b //// a0 + a5 + 20a2 + 20a3 (column1,row0) 311 umlsl v20.8h, v5.8b, v31.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 (column1,row0) 312 umlsl v20.8h, v14.8b, v31.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 (column1,row0) 313 mov v21.d[0], v20.d[1] 320 ext v30.8b, v20.8b , v21.8b , #4 325 ext v29.8b, v20 [all...] |
H A D | ih264e_evaluate_intra16x16_modes_av8.s | 135 dup v20.8h, w11 142 uqshl v0.8h, v0.8h, v20.8h 168 dup v20.8b, v9.b[15] ///HORIZONTAL VALUE ROW=0// 178 uabdl v26.8h, v0.8b, v20.8b 189 dup v20.8b, v9.b[14] ///HORIZONTAL VALUE ROW=1// 199 uabal v26.8h, v2.8b, v20.8b 206 dup v20.8b, v9.b[13] ///HORIZONTAL VALUE ROW=2// 215 uabal v26.8h, v4.8b, v20.8b 222 dup v20.8b, v9.b[12] ///HORIZONTAL VALUE ROW=3// 231 uabal v26.8h, v6.8b, v20 [all...] |
H A D | ih264e_evaluate_intra_chroma_modes_av8.s | 132 dup v20.8h, v6.h[0] 136 mov v20.d[0], v23.d[0] 141 mov v28.d[0], v20.d[0] 142 mov v29.d[0], v20.d[1] 195 dup v20.8h, v27.h[7] ///HORIZONTAL VALUE ROW=0// 206 uabdl v26.8h, v0.8b, v20.8b 218 dup v20.8h, v27.h[6] 228 uabal v26.8h, v2.8b, v20.8b 235 dup v20.8h, v27.h[5] 244 uabal v26.8h, v4.8b, v20 [all...] |
/external/libhevc/decoder/arm64/ |
H A D | ihevcd_fmt_conv_420sp_to_rgba8888.s | 203 sMULL v20.4s, v6.4h, v0.h[0] ////(V-128)*C1 FOR R 217 sqshrn v7.4h, v20.4s,#13 ////D10 = (V-128)*C1>>13 4 16-BIT VALUES 230 UADDW v20.8h, v5.8h , v31.8b ////Q10 - HAS Y + B 239 sqxtun v20.8b, v20.8h 251 ZIP1 v27.8b, v20.8b, v21.8b 252 ZIP2 v21.8b, v20.8b, v21.8b 253 mov v20.d[0], v27.d[0] 259 mov v20.d[1], v21.d[0] 266 ZIP1 v25.8h, v20 [all...] |
/external/valgrind/none/tests/arm64/ |
H A D | memory.stdout.exp | 126 0000000000000000 v20.d[0] (xor, xfer vecreg #3) 127 0000000000000000 v20.d[1] (xor, xfer vecreg #3) 156 0000000000000000 v20.d[0] (xor, xfer vecreg #3) 157 0000000000000000 v20.d[1] (xor, xfer vecreg #3) 186 0000000000000000 v20.d[0] (xor, xfer vecreg #3) 187 0000000000000000 v20.d[1] (xor, xfer vecreg #3) 216 0000000000000000 v20.d[0] (xor, xfer vecreg #3) 217 0000000000000000 v20.d[1] (xor, xfer vecreg #3) 246 0000000000000000 v20.d[0] (xor, xfer vecreg #3) 247 0000000000000000 v20 [all...] |