Home
last modified time | relevance | path

Searched refs:v12 (Results 1 – 25 of 116) sorted by relevance

12345

/external/libavc/common/armv8/
Dih264_weighted_bi_pred_av8.s201 ld1 {v12.8b}, [x0], x3 //load row 3 in source 1
210 uxtl v12.8h, v12.8b //converting row 3 in source 1 to 16-bit
216 mul v12.8h, v12.8h , v2.h[0] //weight 1 mult. for row 3
217 mla v12.8h, v14.8h , v2.h[2] //weight 2 mult. for row 3
222 srshl v12.8h, v12.8h , v0.8h //rounds off the weighted samples from row 3
226 saddw v12.8h, v12.8h , v3.8b //adding offset for row 3
230 sqxtun v12.8b, v12.8h //saturating row 3 to unsigned 8-bit
235 st1 {v12.8b}, [x2], x5 //store row 3 in destination
247 ld1 {v12.8b, v13.8b}, [x0], x3 //load row 3 in source 1
264 uxtl v28.8h, v12.8b //converting row 3L in source 1 to 16-bit
[all …]
Dih264_inter_pred_luma_horz_qpel_vert_qpel_av8.s183 ld1 {v12.2s, v13.2s}, [x7], x2 // src[6_0]
191 uaddl v16.8h, v2.8b, v12.8b
244 umlsl v16.8h, v12.8b, v31.8b
289 umlal v0.8h, v12.8b, v30.8b
341 mov v4.16b, v12.16b
387 ld1 {v12.2s, v13.2s}, [x7], x2 // src[10_0]
393 uaddl v16.8h, v2.8b, v12.8b
444 umlsl v16.8h, v12.8b, v31.8b
489 umlal v0.8h, v12.8b, v30.8b
541 mov v4.16b, v12.16b
[all …]
Dih264_inter_pred_filters_luma_vert_av8.s131 uaddl v12.8h, v4.8b, v6.8b // temp1 = src[2_0] + src[3_0]
138 mla v14.8h, v12.8h, v22.8h // temp += temp1 * 20
144 uaddl v12.8h, v6.8b, v8.8b
148 mla v16.8h, v12.8h , v22.8h
151 uaddl v12.8h, v7.8b, v9.8b
155 mla v14.8h, v12.8h , v22.8h
159 uaddl v12.8h, v8.8b, v10.8b
162 mla v18.8h, v12.8h , v22.8h
166 uaddl v12.8h, v9.8b, v11.8b
169 mla v16.8h, v12.8h , v22.8h
[all …]
Dih264_inter_pred_luma_vert_qpel_av8.s138 uaddl v12.8h, v4.8b, v6.8b // temp1 = src[2_0] + src[3_0]
145 mla v14.8h, v12.8h , v22.8h // temp += temp1 * 20
151 uaddl v12.8h, v6.8b, v8.8b
155 mla v16.8h, v12.8h , v22.8h
158 uaddl v12.8h, v7.8b, v9.8b
162 mla v14.8h, v12.8h , v22.8h
169 uaddl v12.8h, v8.8b, v10.8b
171 mla v18.8h, v12.8h , v22.8h
175 uaddl v12.8h, v9.8b, v11.8b
178 mla v16.8h, v12.8h , v22.8h
[all …]
Dih264_inter_pred_luma_horz_qpel_vert_hpel_av8.s156 ld1 {v12.2s}, [x0], x2 // Vector load from src[0_0]
182 uaddl v22.8h, v12.8b, v17.8b
251 ld1 {v12.2s}, [x0], x2 // Vector load from src[6_0]
269 uaddl v22.8h, v13.8b, v12.8b
355 uaddl v26.8h, v15.8b, v12.8b
438 uaddl v24.8h, v17.8b, v12.8b
503 mov v14.16b, v12.16b
523 mov v12.16b, v16.16b
548 uaddl v12.8h, v0.8b, v10.8b
550 mla v12.8h, v14.8h , v26.8h
[all …]
Dih264_inter_pred_luma_horz_hpel_vert_hpel_av8.s95 ld1 {v12.2s}, [x0], x2 // Vector load from src[0_0]
120 uaddl v22.8h, v12.8b, v17.8b
183 ld1 {v12.2s}, [x0], x2 // Vector load from src[6_0]
201 uaddl v22.8h, v13.8b, v12.8b
282 uaddl v26.8h, v15.8b, v12.8b
360 uaddl v24.8h, v17.8b, v12.8b
424 mov v14.16b, v12.16b
440 mov v12.16b, v16.16b
463 uaddl v12.8h, v0.8b, v10.8b
465 mla v12.8h, v14.8h , v26.8h
[all …]
Dih264_deblk_luma_av8.s100 mov v12.s[0], w4 //d12[0] = ui_Bs
105 uxtl v12.8h, v12.8b //q6 = uc_Bs in each 16 bt scalar
118 tbl v14.8b, {v16.16b}, v12.8b //
122 uxtl v12.4s, v12.4h //
126 cmgt v12.4s, v12.4s, #0
142 bic v12.16b, v12.16b , v18.16b //final condition
152 and v20.16b, v20.16b , v12.16b //
153 and v22.16b, v22.16b , v12.16b //
162 … and v18.16b, v18.16b , v12.16b //Making delta zero in places where values shouldn be filterd
263 uabd v12.16b , v4.16b, v6.16b
[all …]
Dih264_deblk_chroma_av8.s226 uabd v12.16b, v0.16b , v2.16b //|p1-p0|
231 cmhi v12.16b, v24.16b , v12.16b //|p1-p0| < beta ?
243 and v8.16b, v8.16b , v12.16b //|p0-q0| < alpha && |q1-q0| < beta && |p1-p0| < beta
340 mov v12.s[0], w7 //D12[0] = ui_Bs
344 tbl v14.8b, {v16.16b}, v12.8b //Retreiving cliptab values for U
345 tbl v28.8b, {v17.16b}, v12.8b //Retrieving cliptab values for V
346 uxtl v12.8h, v12.8b //Q6 = uc_Bs in each 16 bit scalar
382 cmgt v12.4h, v12.4h, #0
392 mov v13.8b, v12.8b
393 mov v12.d[1], v13.d[0] //
[all …]
Dih264_iquant_itrans_recon_av8.s181 sub v12.4h, v5.4h , v6.4h // x1-x2
191 trn1 v6.4h, v12.4h, v13.4h
192 trn2 v7.4h, v12.4h, v13.4h
196 trn2 v12.2s, v4.2s, v6.2s // 4
203 add v14.4h, v10.4h, v12.4h // x0 = q0 + q2//
204 sub v15.4h, v10.4h, v12.4h // x1 = q0 - q2//
368 sub v12.4h, v5.4h , v6.4h // x1-x2
380 trn1 v6.4h, v12.4h, v13.4h
381 trn2 v7.4h, v12.4h, v13.4h
385 trn2 v12.2s, v4.2s, v6.2s // 4
[all …]
Dih264_iquant_itrans_recon_dc_av8.s246 ld1 {v12.d}[0], [x2], x4
247 ld1 {v12.d}[1], [x2]
262 bit v12.16b, v2.16b, v31.16b
266 st1 {v12.d}[0], [x0], x4
267 st1 {v12.d}[1], [x0]
374 uaddw v12.8h, v0.8h, v29.8b
383 sqxtun v12.8b, v12.8h
392 st1 {v12.8b}, [x2]
Dih264_intra_pred_luma_8x8_av8.s341 add v12.8h, v8.8h , v10.8h
342 sqrshrun v31.8b, v12.8h, #4
361 uaddlp v12.1d, v13.2s
362 rshrn v4.8b, v12.8h, #3
707 ext v12.16b, v16.16b , v16.16b , #1
712 st1 {v12.h}[5], [x1], #2
730 st1 {v12.s}[2], [x1], #4
825 mov v12.16b, v8.16b
828 trn1 v13.8h, v12.8h, v14.8h
829 trn2 v14.8h, v12.8h, v14.8h
[all …]
Dih264_inter_pred_luma_horz_hpel_vert_qpel_av8.s189 uaddl v12.8h, v2.8b, v3.8b
194 mla v10.8h, v12.8h , v22.8h
196 uaddl v12.8h, v1.8b, v4.8b
198 mls v10.8h, v12.8h , v24.8h
200 uaddl v12.8h, v0.8b, v5.8b
208 mla v12.8h, v14.8h , v22.8h
212 mls v12.8h, v14.8h , v24.8h
219 st1 {v12.4s}, [x9], x6 // store temp buffer 3
243 add v30.8h, v10.8h , v12.8h
273 add v30.8h, v12.8h , v14.8h
[all …]
Dih264_intra_pred_chroma_av8.s270 dup v12.8h, v0.h[5]
276 st1 {v12.8h}, [x1], x3
433 usubl v12.8h, v3.8b, v7.8b
435 mul v16.8h, v12.8h , v8.8h
456 rshrn v12.4h, v24.4s, #6
474 dup v4.8h, v12.h[0]
498 mul v12.8h, v4.8h , v8.8h
500 add v12.8h, v0.8h , v12.8h
508 add v24.8h, v12.8h , v4.8h
510 add v2.8h, v12.8h , v14.8h
[all …]
Dih264_inter_pred_luma_horz_qpel_av8.s179 ld1 {v12.2s, v13.2s}, [x7], x2 //Load value for interpolation (column1,row0)
184 urhadd v20.16b, v12.16b , v20.16b //Interpolation step for qpel calculation
195 ld1 {v12.2s, v13.2s}, [x7], x2 //Load value for interpolation (column1,row1)
197 urhadd v18.16b, v12.16b , v18.16b //Interpolation step for qpel calculation
240 ld1 {v12.2s, v13.2s}, [x7], x2 //Load value for interpolation (column1,row2)
245 urhadd v20.16b, v12.16b , v20.16b //Interpolation step for qpel calculation
252 ld1 {v12.2s, v13.2s}, [x7], x2 //Load value for interpolation (column1,row3)
256 urhadd v18.16b, v12.16b , v18.16b //Interpolation step for qpel calculation
298 ld1 {v12.2s, v13.2s}, [x7], x2 //Load value for interpolation (column1,row4)
303 urhadd v20.16b, v12.16b , v20.16b //Interpolation step for qpel calculation
[all …]
Dih264_weighted_pred_av8.s212 uxtl v12.8h, v4.8b //converting row 1L to 16-bit
217 mul v12.8h, v12.8h , v2.h[0] //weight mult. for row 1L
229 srshl v12.8h, v12.8h , v0.8h //rounds off the weighted samples from row 1L
233 saddw v12.8h, v12.8h , v3.8b //adding offset for row 1L
236 sqxtun v4.8b, v12.8h //saturating row 1L to unsigned 8-bit
410 ld1 {v12.8b, v13.8b}, [x0], x2 //load row 4 in source
421 uxtl v26.8h, v12.8b //converting row 4L to 16-bit
452 sqxtun v12.8b, v26.8h //saturating row 4L to unsigned 8-bit
458 st1 {v12.8b, v13.8b}, [x1], x3 //store row 4 in destination
/external/libhevc/common/arm64/
Dihevc_intra_pred_chroma_planar.s187 umull v12.8h, v5.8b, v0.8b //(row+1) * src[nt-1]
190 umlal v12.8h, v6.8b, v10.8b //(nt-1-row) * src[2nt+1+col]
192 umlal v12.8h, v17.8b, v1.8b //(col+1) * src[3nt+1]
194 umlal v12.8h, v30.8b, v4.8b //(nt-1-col) * src[2nt-1-row]
211 add v12.8h, v12.8h , v16.8h //add (nt)
213 sshl v12.8h, v12.8h, v14.8h //shr
228 xtn v12.8b, v12.8h
239 st1 {v12.2s, v13.2s}, [x2], x3
257 umull v12.8h, v18.8b, v0.8b //(row+1) * src[nt-1]
259 umlal v12.8h, v19.8b, v10.8b //(nt-1-row) * src[2nt+1+col]
[all …]
Dihevc_itrans_recon_32x32.s245 ld1 {v12.4h},[x0],x6
268 smlal v20.4s, v12.4h, v1.h[0]
270 smlal v22.4s, v12.4h, v3.h[0]
272 smlal v16.4s, v12.4h, v5.h[0]
274 smlal v18.4s, v12.4h, v7.h[0]
317 ld1 {v12.4h},[x0],x6
342 smlal v20.4s, v12.4h, v3.h[0]
344 smlsl v22.4s, v12.4h, v7.h[0]
346 smlsl v16.4s, v12.4h, v1.h[0]
348 smlsl v18.4s, v12.4h, v5.h[0]
[all …]
Dihevc_itrans_recon_8x8.s201 ld1 {v12.4h},[x0],x5
296 movi v12.4h, #0
367 smull v22.4s, v12.4h, v0.h[0] //// y4 * cos4(part of c0 and c1)
377 add v12.4s, v20.4s , v22.4s //// c0 = y0 * cos4 + y4 * cos4(part of a0 and a1)
385 add v16.4s, v12.4s , v8.4s //// a0 = c0 + d0(part of e0,e7)
386 sub v12.4s, v12.4s , v8.4s //// a3 = c0 - d0(part of e3,e4)
399 add v26.4s, v12.4s , v30.4s //// a3 + b3(part of e3)
400 sub v30.4s, v12.4s , v30.4s //// a3 - b3(part of x4)
409 sqrshrn v12.4h, v30.4s,#shift_stage1_idct //// x4 = (a3 - b3 + rnd) >> 7(shift_stage1_idct)
561 add v12.4s, v20.4s , v14.4s //// a0 = c0 + d0(part of x0,x7)
[all …]
Dihevc_inter_pred_chroma_horz.s202 ld1 { v12.2s},[x4],x11 //vector load pu1_src
265 umlal v22.8h, v12.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//
299 ld1 { v12.2s},[x4],x11 //vector load pu1_src
361 umlal v22.8h, v12.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//
400 ld1 { v12.2s},[x4],x11 //vector load pu1_src
427 umlal v22.8h, v12.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//
589 umull v12.8h, v15.8b, v25.8b //(3)mul_res = vmull_u8(src[0_3], coeffabs_3)//
592 umlsl v12.8h, v14.8b, v24.8b //(3)mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
595 umlal v12.8h, v16.8b, v26.8b //(3)mul_res = vmlsl_u8(src[0_0], coeffabs_0)//
599 umlsl v12.8h, v17.8b, v27.8b //(3)mul_res = vmlal_u8(src[0_1], coeffabs_1)//
[all …]
Dihevc_itrans_recon_16x16.s270 smull v12.4s, v10.4h, v0.h[0]
271 smlal v12.4s, v11.4h, v0.h[2]
297 smlal v12.4s, v4.4h, v1.h[0]
298 smlal v12.4s, v5.4h, v1.h[2]
358 smlal v12.4s, v10.4h, v0.h[0]
359 smlal v12.4s, v11.4h, v2.h[2]
360 smlal v12.4s, v4.4h, v3.h[0]
361 smlal v12.4s, v5.4h, v3.h[2]
384 add v20.4s, v12.4s , v24.4s
385 sub v22.4s, v12.4s , v24.4s
[all …]
/external/libmpeg2/common/armv8/
Dimpeg2_idct.s176 uaddw v12.8h, v30.8h , v2.8b
181 sqxtun v2.8b, v12.8h
242 raddhn v12.4h, v0.4s, v8.4s
243 raddhn2 v12.8h, v0.4s, v10.4s
244 uaddw v14.8h, v12.8h , v30.8b
252 raddhn v12.4h, v0.4s, v8.4s
253 raddhn2 v12.8h, v0.4s, v10.4s
254 uaddw v14.8h, v12.8h , v30.8b
262 raddhn v12.4h, v0.4s, v8.4s
263 raddhn2 v12.8h, v0.4s, v10.4s
[all …]
/external/libhevc/decoder/arm64/
Dihevcd_fmt_conv_420sp_to_rgba8888.s206 sMULL v12.4s, v4.4h, v0.h[1] ////(U-128)*C2 FOR G
207 sMLAL v12.4s, v6.4h, v0.h[2] ////Q6 = (U-128)*C2 + (V-128)*C3
222 sqshrn v12.4h, v12.4s,#13 ////D12 = [(U-128)*C2 + (V-128)*C3]>>13 4 16-BIT VALUES
223 sqshrn2 v12.8h, v14.4s,#13 ////D13 = [(U-128)*C2 + (V-128)*C3]>>13 4 16-BIT VALUES
228 UADDW v18.8h, v12.8h , v30.8b ////Q9 - HAS Y + G
232 UADDW v24.8h, v12.8h , v31.8b ////Q12 - HAS Y + G
284 UADDW v18.8h, v12.8h , v28.8b ////Q3 - HAS Y + G
288 UADDW v24.8h, v12.8h , v29.8b ////Q12 - HAS Y + G
369 sMULL v12.4s, v4.4h, v0.h[1] ////(U-128)*C2 FOR G
370 sMLAL v12.4s, v6.4h, v0.h[2] ////Q6 = (U-128)*C2 + (V-128)*C3
[all …]
/external/libvpx/libvpx/vp8/common/ppc/
Didctllm_altivec.asm33 load_c v12, shift_16, 0, r9, r10
47 vsraw v4, v4, v12
52 vsraw v5, v5, v12 ;# ip[12] * cos(pi/8) * sqrt(2)
58 vsraw v3, v3, v12
62 vsraw v5, v5, v12 ;# ip[ 4] * cos(pi/8) * sqrt(2)
91 vsraw v4, v4, v12
95 vsraw v5, v5, v12 ;# ip[12] * cos(pi/8) * sqrt(2)
101 vsraw v2, v2, v12
105 vsraw v5, v5, v12 ;# ip[ 4] * cos(pi/8) * sqrt(2)
Dloopfilter_filters_altivec.asm120 Tpair v24,v25, v4,v12
133 Tpair v12,v13, v22,v30
252 Tpair v12, v13, v1, v5
260 Tpair v4, v5, v12, v16
290 vmrghb v12, v22, v30
302 vmrghb v24, v4, v12
303 vmrglb v25, v4, v12
322 vmrghb v12, v22, v30
334 vmrghb v24, v4, v12
335 vmrglb v25, v4, v12
[all …]
/external/llvm/test/CodeGen/AArch64/
Darm64-copy-tuple.ll19 …ffect "", "~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v1…
22 …ffect "", "~{v0},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v1…
36 …ffect "", "~{v0},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v1…
39 …ffect "", "~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v1…
53 …"", "~{v1},~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v1…
56 …ffect "", "~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v1…
70 …ffect "", "~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v1…
73 …"", "~{v1},~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v1…
90 …tail call void asm sideeffect "", "~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},…
93 …tail call void asm sideeffect "", "~{v0},~{v1},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},…
[all …]

12345