Searched refs:v10 (Results 1 – 25 of 216) sorted by relevance
123456789
151 ld1 {v10.2s, v11.2s}, [x7], x2 // Vector load from src[5_0]153 uaddl v24.8h, v0.8b, v10.8b195 umlsl v16.8h, v10.8b, v31.8b242 umlal v16.8h, v10.8b, v30.8b288 umlal v0.8h, v10.8b, v30.8b338 mov v2.16b, v10.16b355 ld1 {v10.2s, v11.2s}, [x7], x2 // Vector load from src[9_0]357 uaddl v24.8h, v0.8b, v10.8b397 umlsl v16.8h, v10.8b, v31.8b442 umlal v16.8h, v10.8b, v30.8b[all …]
101 uaddl v10.8h, v7.8b, v1.8b //Q4,Q5 = q0 + p1110 umlal v10.8h, v3.8b, v31.8b //Q5,Q4 = (X2(q1U) + q0U + p1U)124 rshrn v9.8b, v10.8h, #2 //Q4 = (X2(q1U) + q0U + p1U + 2) >> 2127 rshrn v10.8b, v14.8h, #2 //129 mov v10.d[1], v11.d[0]131 bit v10.16b, v4.16b , v18.16b //133 mov v11.d[0], v10.d[1]135 st2 {v10.8b, v11.8b}, [x4], x1 //206 mov v10.16b, v2.16b209 mov v4.16b, v10.16b[all …]
172 ld1 {v10.s}[0], [x1], x4 //load row 3 in source 2173 ld1 {v10.s}[1], [x1], x4 //load row 4 in source 2175 uxtl v10.8h, v10.8b //converting rows 3,4 in source 2 to 16-bit179 mla v8.8h, v10.8h , v2.h[2] //weight 2 mult. for rows 3,4199 ld1 {v10.8b}, [x1], x4 //load row 2 in source 2207 uxtl v10.8h, v10.8b //converting row 2 in source 2 to 16-bit213 mla v8.8h, v10.8h , v2.h[2] //weight 2 mult. for row 2245 ld1 {v10.8b, v11.8b}, [x1], x4 //load row 2 in source 2257 uxtl v26.8h, v10.8b //converting row 2L in source 2 to 16-bit261 uxtl v10.8h, v11.8b //converting row 2H in source 2 to 16-bit[all …]
174 ld1 {v10.8b}, [x0], x2 //load row 4 in source179 uxtl v10.8h, v10.8b //converting row 4 to 16-bit182 mul v10.8h, v10.8h , v2.h[0] //weight mult. for row 4188 srshl v10.8h, v10.8h , v0.8h //rounds off the weighted samples from row 4193 saddw v10.8h, v10.8h , v3.8b //adding offset for row 4196 sqxtun v10.8b, v10.8h //saturating row 4 to unsigned 8-bit202 st1 {v10.8b}, [x1], x3 //store row 4 in destination215 ld1 {v10.8b, v11.8b}, [x0], x2 //load row 4 in source224 uxtl v24.8h, v10.8b //converting row 4L to 16-bit252 sqxtun v10.8b, v24.8h //saturating row 4L to unsigned 8-bit[all …]
99 ld1 {v10.8b, v11.8b}, [x0], x1 //p2 values are loaded into q5107 mov v10.d[1], v11.d[0]124 uabd v28.16b, v10.16b, v6.16b156 uaddl v10.8h, v16.8b, v10.8b //Q14,Q5 = p2 + (p0+q0+1)>>1164 sub v10.8h, v10.8h , v16.8h //168 sqshrn v28.8b, v10.8h, #1 //Q14 = i_macro_p1256 ld1 {v10.8b, v11.8b}, [x14] //load p1 to Q5260 mov v10.d[1] , v11.d[0]265 uabd v16.16b , v10.16b, v6.16b289 uaddw v16.8h, v16.8h , v10.8b //2*(p0+q0+q1)+q2 +p1 L[all …]
132 ld1 {v10.2s, v11.2s}, [x0], x2 // Vector load from src[5_0]136 uaddl v14.8h, v0.8b, v10.8b // temp = src[0_0] + src[5_0]147 uaddl v18.8h, v4.8b, v10.8b159 uaddl v12.8h, v8.8b, v10.8b174 uaddl v12.8h, v10.8b, v0.8b191 uaddl v16.8h, v10.8b, v4.8b // temp2 = src[1_0] + src[4_0]213 uaddl v14.8h, v8.8b, v10.8b216 ld1 {v10.2s, v11.2s}, [x0], x2230 uaddl v14.8h, v10.8b, v0.8b242 uaddl v16.8h, v10.8b, v4.8b[all …]
175 uaddl v10.8h, v2.8b, v3.8b180 mla v8.8h, v10.8h , v22.8h182 uaddl v10.8h, v1.8b, v4.8b184 mls v8.8h, v10.8h , v24.8h186 uaddl v10.8h, v0.8b, v5.8b194 mla v10.8h, v12.8h , v22.8h198 mls v10.8h, v12.8h , v24.8h205 st1 {v10.4s}, [x9], x6 // store temp buffer 2243 add v30.8h, v10.8h , v12.8h271 add v28.8h, v10.8h , v16.8h[all …]
179 add v10.4h, v4.4h , v7.4h // x0+x3189 trn1 v4.4h, v10.4h, v11.4h190 trn2 v5.4h, v10.4h, v11.4h194 trn1 v10.2s, v4.2s, v6.2s // 0203 add v14.4h, v10.4h, v12.4h // x0 = q0 + q2//204 sub v15.4h, v10.4h, v12.4h // x1 = q0 - q2//366 add v10.4h, v4.4h , v7.4h // x0+x3378 trn1 v4.4h, v10.4h, v11.4h379 trn2 v5.4h, v10.4h, v11.4h383 trn1 v10.2s, v4.2s, v6.2s // 0[all …]
139 ld1 {v10.2s, v11.2s}, [x0], x2 // Vector load from src[5_0]143 uaddl v14.8h, v0.8b, v10.8b // temp = src[0_0] + src[5_0]154 uaddl v18.8h, v4.8b, v10.8b169 uaddl v12.8h, v8.8b, v10.8b183 uaddl v12.8h, v10.8b, v0.8b205 uaddl v16.8h, v10.8b, v4.8b // temp2 = src[1_0] + src[4_0]230 uaddl v14.8h, v8.8b, v10.8b233 ld1 {v10.2s, v11.2s}, [x0], x2252 uaddl v14.8h, v10.8b, v0.8b264 uaddl v16.8h, v10.8b, v4.8b[all …]
262 st1 {v8.8b, v9.8b, v10.8b, v11.8b}, [sp], 32305 st1 {v8.4h, v9.4h, v10.4h, v11.4h}, [sp], 32 /* save NEON registers */330 mov v10.16b, v14.16b339 smlal v10.4s, ROW5L.4h, XFIX_2_053119869_MINUS_2_562915447341 smlsl v10.4s, ROW3L.4h, XFIX_2_562915447351 add v2.4s, v6.4s, v10.4s353 sub v6.4s, v6.4s, v10.4s354 saddl v10.4s, ROW0L.4h, ROW4L.4h360 shl v10.4s, v10.4s, #13363 add v4.4s, v10.4s, v12.4s[all …]
211 ld1 {v10.4h},[x0],x6230 smull v20.4s, v10.4h, v0.h[0]234 smull v22.4s, v10.4h, v0.h[0]237 smull v16.4s, v10.4h, v0.h[0]240 smull v18.4s, v10.4h, v0.h[0]280 ld1 {v10.4h},[x0],x6300 smlal v20.4s, v10.4h, v2.h[0]304 smlal v22.4s, v10.4h, v6.h[0]307 smlsl v16.4s, v10.4h, v6.h[0]310 smlsl v18.4s, v10.4h, v2.h[0][all …]
169 umull v10.8h, v23.8b, v30.8b //(i row)vmull_u8(ref_main_idx, dup_const_32_fract)172 umlal v10.8h, v9.8b, v31.8b //(i row)vmull_u8(ref_main_idx_1, dup_const_fract)187 rshrn v10.8b, v10.8h,#5 //(i row)shift_res = vrshrn_n_u16(add_res, 5)201 st1 {v10.8b},[x2],#8 //(i row)230 umull v10.8h, v23.8b, v30.8b //(v)vmull_u8(ref_main_idx, dup_const_32_fract)233 umlal v10.8h, v9.8b, v31.8b //(v)vmull_u8(ref_main_idx_1, dup_const_fract)253 rshrn v10.8b, v10.8h,#5 //(v)shift_res = vrshrn_n_u16(add_res, 5)267 st1 {v10.8b},[x0],x3 //(v)310 umull v10.8h, v23.8b, v30.8b //(i)vmull_u8(ref_main_idx, dup_const_32_fract)314 umlal v10.8h, v9.8b, v31.8b //(i)vmull_u8(ref_main_idx_1, dup_const_fract)[all …]
174 umull v10.8h, v23.8b, v30.8b //(i row)vmull_u8(ref_main_idx, dup_const_32_fract)177 umlal v10.8h, v9.8b, v31.8b //(i row)vmull_u8(ref_main_idx_1, dup_const_fract)192 rshrn v10.8b, v10.8h,#5 //(i row)shift_res = vrshrn_n_u16(add_res, 5)206 st1 {v10.8b},[x2],#8 //(i row)235 umull v10.8h, v23.8b, v30.8b //(v)vmull_u8(ref_main_idx, dup_const_32_fract)238 umlal v10.8h, v9.8b, v31.8b //(v)vmull_u8(ref_main_idx_1, dup_const_fract)258 rshrn v10.8b, v10.8h,#5 //(v)shift_res = vrshrn_n_u16(add_res, 5)272 st1 {v10.8b},[x0],x3 //(v)315 umull v10.8h, v23.8b, v30.8b //(i)vmull_u8(ref_main_idx, dup_const_32_fract)319 umlal v10.8h, v9.8b, v31.8b //(i)vmull_u8(ref_main_idx_1, dup_const_fract)[all …]
284 umull v10.8h, v23.8b, v30.8b //(i row)vmull_u8(ref_main_idx, dup_const_32_fract)287 umlal v10.8h, v9.8b, v31.8b //(i row)vmull_u8(ref_main_idx_1, dup_const_fract)301 rshrn v10.8b, v10.8h,#5 //(i row)shift_res = vrshrn_n_u16(add_res, 5)315 st1 {v10.8b},[x2],#8 //(i row)342 umull v10.8h, v23.8b, v30.8b //(v)vmull_u8(ref_main_idx, dup_const_32_fract)345 umlal v10.8h, v9.8b, v31.8b //(v)vmull_u8(ref_main_idx_1, dup_const_fract)364 rshrn v10.8b, v10.8h,#5 //(v)shift_res = vrshrn_n_u16(add_res, 5)378 st1 {v10.8b},[x0],x3 //(v)422 umull v10.8h, v23.8b, v30.8b //(i)vmull_u8(ref_main_idx, dup_const_32_fract)425 umlal v10.8h, v9.8b, v31.8b //(i)vmull_u8(ref_main_idx_1, dup_const_fract)[all …]
163 LD1 {v10.8b},[x14],#8 //band_table_v.val[1]223 …ADD v14.8b, v10.8b , v30.8b //band_table_v.val[1] = vadd_u8(band_table_v.val[1], band_…235 …ADD v10.8b, v14.8b , v28.8b //band_table_v.val[1] = vadd_u8(band_table_v.val[1], vdup_…266 cmhs v19.8b, v29.8b , v10.8b //vcle_u8(band_table.val[1], vdup_n_u8(16))268 … ORR v10.8b, v10.8b , v19.8b //band_table.val[1] = vorr_u8(band_table.val[1], au1_cmp)280 … AND v10.8b, v10.8b , v19.8b //band_table.val[1] = vand_u8(band_table.val[1], au1_cmp)296 mov v9.d[1],v10.d[0]297 mov v10.d[0],v11.d[0]298 mov v10.d[1],v12.d[0]324 …TBX v6.8b, {v9.16b- v10.16b},v8.8b //vtbx4_u8(au1_cur_row_deint.val[1], band_table_v, vsub…[all …]
197 ld1 { v10.2s},[x4],x11 //vector load pu1_src240 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//295 ld1 { v10.2s},[x4],x11 //vector load pu1_src354 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//394 ld1 { v10.2s},[x4],x11 //vector load pu1_src425 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//499 umull v10.8h, v5.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//500 umlsl v10.8h, v4.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//504 umlal v10.8h, v6.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//505 umlsl v10.8h, v7.8b, v27.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//[all …]
246 umull v10.8h, v15.8b, v27.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//248 umlsl v10.8h, v14.8b, v26.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//250 umlal v10.8h, v16.8b, v28.8b //mul_res = vmlal_u8(src[0_4], coeffabs_4)//252 umlsl v10.8h, v17.8b, v29.8b //mul_res = vmlsl_u8(src[0_5], coeffabs_5)//254 umlal v10.8h, v18.8b, v30.8b //mul_res = vmlal_u8(src[0_6], coeffabs_6)//255 umlsl v10.8h, v19.8b, v31.8b //mul_res = vmlsl_u8(src[0_7], coeffabs_7)//257 umlsl v10.8h, v12.8b, v24.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//258 umlal v10.8h, v13.8b, v25.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//262 sqrshrun v8.8b, v10.8h,#6 //right shift and saturating narrow result 2376 umull v10.8h, v2.8b, v25.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//[all …]
235 ld1 {v10.4h},[x0],x6270 smull v12.4s, v10.4h, v0.h[0]272 smull v14.4s, v10.4h, v0.h[0]274 smull v16.4s, v10.4h, v0.h[0]276 smull v18.4s, v10.4h, v0.h[0]319 ld1 {v10.4h},[x0],x6358 smlal v12.4s, v10.4h, v0.h[0]366 smlsl v14.4s, v10.4h, v0.h[0]372 smlsl v16.4s, v10.4h, v0.h[0]378 smlal v18.4s, v10.4h, v0.h[0][all …]
212 ld1 { v10.2s},[x4],x11 //vector load pu1_src252 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//302 ld1 { v10.2s},[x4],x11 //vector load pu1_src347 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//391 ld1 { v10.2s},[x4],x11 //vector load pu1_src415 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//482 umull v10.8h, v5.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//483 umlsl v10.8h, v4.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//486 umlal v10.8h, v6.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//487 umlsl v10.8h, v7.8b, v27.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//[all …]
432 umull v10.8h, v15.8b, v27.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//434 umlsl v10.8h, v14.8b, v26.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//436 umlal v10.8h, v16.8b, v28.8b //mul_res = vmlal_u8(src[0_4], coeffabs_4)//438 umlsl v10.8h, v17.8b, v29.8b //mul_res = vmlsl_u8(src[0_5], coeffabs_5)//440 umlal v10.8h, v18.8b, v30.8b //mul_res = vmlal_u8(src[0_6], coeffabs_6)//441 umlsl v10.8h, v19.8b, v31.8b //mul_res = vmlsl_u8(src[0_7], coeffabs_7)//443 umlsl v10.8h, v12.8b, v24.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//444 umlal v10.8h, v13.8b, v25.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//450 st1 {v10.8h},[x10],#16 //store the result pu1_dst558 umull v10.8h, v2.8b, v25.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//[all …]
39 function select(n, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10) { argument50 v9 = v10;51 v10 = tmp;56 function select_while(n, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10) { argument68 v9 = v10;69 v10 = tmp;
173 uaddw v10.8h, v30.8h , v1.8b177 sqxtun v1.8b, v10.8h241 sxtl v10.4s, v3.4h243 raddhn2 v12.8h, v0.4s, v10.4s251 sxtl v10.4s, v3.4h253 raddhn2 v12.8h, v0.4s, v10.4s261 sxtl v10.4s, v3.4h263 raddhn2 v12.8h, v0.4s, v10.4s271 sxtl v10.4s, v3.4h273 raddhn2 v12.8h, v0.4s, v10.4s[all …]
131 dup v10.8h, w10139 add v0.4h, v0.4h, v10.4h160 ld1 {v10.8b}, [x6], #8172 uabdl v16.8h, v0.8b, v10.8b192 uabal v16.8h, v2.8b, v10.8b208 uabal v16.8h, v4.8b, v10.8b224 uabal v16.8h, v6.8b, v10.8b241 uabal v16.8h, v0.8b, v10.8b257 uabal v16.8h, v2.8b, v10.8b273 uabal v16.8h, v4.8b, v10.8b[all …]
85 fmin v10.4h, v15.4h, v22.4h86 fmin v10.8h, v15.8h, v22.8h87 fmin v10.2s, v15.2s, v22.2s115 fminnm v10.4h, v15.4h, v22.4h116 fminnm v10.8h, v15.8h, v22.8h117 fminnm v10.2s, v15.2s, v22.2s
12 ld1.8b {v7, v8, v9, v10}, [x4]103 ; CHECK: ld1.8b { v7, v8, v9, v10 }, [x4] ; encoding: [0x87,0x20,0x40,0x0c]231 ld3.8b {v9, v10, v11}, [x9]247 st3.8b {v10, v11, v12}, [x9]264 ; CHECK: ld3.8b { v9, v10, v11 }, [x9] ; encoding: [0x29,0x41,0x40,0x0c]280 ; CHECK: st3.8b { v10, v11, v12 }, [x9] ; encoding: [0x2a,0x41,0x00,0x0c]1315 ld1 { v7.4h, v8.4h, v9.4h, v10.4h }, [x1]1320 ld1 { v7.8h, v8.8h, v9.8h, v10.8h }, [x1]1325 ld1 { v7.2s, v8.2s, v9.2s, v10.2s }, [x1]1330 ld1 { v7.4s, v8.4s, v9.4s, v10.4s }, [x1][all …]