| /external/libavc/common/armv8/ |
| ih264_inter_pred_luma_horz_hpel_vert_hpel_av8.s | 79 movi v24.8h, #0x5 // Filter coeff 5 into Q12
114 uaddl v24.8h, v5.8b, v7.8b
117 mla v20.8h, v24.8h , v28.8h
118 uaddl v24.8h, v14.8b, v15.8b
123 mla v22.8h, v24.8h , v28.8h
126 ext v24.16b, v18.16b , v20.16b , #4
130 add v0.8h, v24.8h , v26.8h
131 ext v24.16b, v18.16b , v20.16b , #2
133 add v24.8h, v24.8h , v26.8
[all...] |
| ih264_inter_pred_luma_horz_qpel_vert_qpel_av8.s | 153 uaddl v24.8h, v0.8b, v10.8b
154 umlal v24.8h, v4.8b, v30.8b
155 umlal v24.8h, v6.8b, v30.8b
156 umlsl v24.8h, v2.8b, v31.8b
157 umlsl v24.8h, v8.8b, v31.8b
163 sqrshrun v26.8b, v24.8h, #5
170 uaddl v24.8h, v1.8b, v11.8b
171 umlal v24.8h, v5.8b, v30.8b
172 umlal v24.8h, v7.8b, v30.8b
173 umlsl v24.8h, v3.8b, v31.8
[all...] |
| ih264_inter_pred_luma_horz_qpel_vert_hpel_av8.s | 140 movi v24.8h, #0x5 // Filter coeff 5 into Q12
176 uaddl v24.8h, v5.8b, v7.8b
179 mla v20.8h, v24.8h , v28.8h
180 uaddl v24.8h, v14.8b, v15.8b
185 mla v22.8h, v24.8h , v28.8h
189 ext v24.16b, v18.16b , v20.16b , #4
193 add v0.8h, v24.8h , v26.8h
194 ext v24.16b, v18.16b , v20.16b , #2
196 add v24.8h, v24.8h , v26.8
[all...] |
| ih264_inter_pred_chroma_av8.s | 169 umull v24.8h, v5.8b, v28.8b
171 umlal v24.8h, v8.8b, v29.8b
173 umlal v24.8h, v10.8b, v30.8b
175 umlal v24.8h, v13.8b, v31.8b
178 sqrshrun v18.8b, v24.8h, #6
189 umull v24.8h, v11.8b, v28.8b
191 umlal v24.8h, v14.8b, v29.8b
193 umlal v24.8h, v1.8b, v30.8b
195 umlal v24.8h, v4.8b, v31.8b
197 sqrshrun v27.8b, v24.8h, #
[all...] |
| ih264_resi_trans_quant_av8.s | 118 ld1 {v24.8b}, [x0] //load first 8 pix src row 4
124 usubl v6.8h, v24.8b, v25.8b //find residue row 4
172 add v24.4h, v18.4h , v19.4h //x5 = x0 + x1;
179 st1 {v24.h}[0], [x10] //store the dc value to alternate dc sddress
183 abs v0.4h, v24.4h //abs val of row 1
188 cmgt v4.4h, v24.4h, #0
203 dup v24.4s, w7
205 sshl v20.4s, v20.4s, v24.4s //shift row 1
206 sshl v21.4s, v21.4s, v24.4s //shift row 2
207 sshl v22.4s, v22.4s, v24.4s //shift row
[all...] |
| ih264_inter_pred_filters_luma_vert_av8.s | 118 movi v24.8h, #5 // Filter coeff 0x4 into Q12
145 mls v14.8h, v16.8h , v24.8h // temp -= temp2 * 5
149 mls v20.8h, v26.8h , v24.8h // temp4 -= temp5 * 5
156 mls v16.8h, v18.8h , v24.8h
164 mls v14.8h, v26.8h , v24.8h
170 mls v18.8h, v20.8h , v24.8h
178 mls v16.8h, v26.8h , v24.8h
185 mls v14.8h, v20.8h , v24.8h
188 mls v18.8h, v26.8h , v24.8h
205 mls v14.8h, v16.8h , v24.8h // temp -= temp2 *
[all...] |
| ih264_inter_pred_luma_vert_qpel_av8.s | 125 movi v24.8h, #5 // Filter coeff 0x4 into Q12
152 mls v14.8h, v16.8h , v24.8h // temp -= temp2 * 5
156 mls v20.8h, v26.8h , v24.8h // temp4 -= temp5 * 5
163 mls v16.8h, v18.8h , v24.8h
173 mls v14.8h, v26.8h , v24.8h
179 mls v18.8h, v20.8h , v24.8h
189 mls v16.8h, v26.8h , v24.8h
196 mls v14.8h, v20.8h , v24.8h
200 mls v18.8h, v26.8h , v24.8h
222 mls v14.8h, v16.8h , v24.8h // temp -= temp2 *
[all...] |
| ih264_weighted_pred_av8.s | 224 uxtl v24.8h, v10.8b //converting row 4L to 16-bit
228 mul v24.8h, v24.8h , v2.h[0] //weight mult. for row 4L
243 srshl v24.8h, v24.8h , v0.8h //rounds off the weighted samples from row 4L
249 saddw v24.8h, v24.8h , v3.8b //adding offset for row 4L
252 sqxtun v10.8b, v24.8h //saturating row 4L to unsigned 8-bit
419 uxtl v24.8h, v11.8b //converting row 3H to 16-bit
422 mul v24.8h, v24.8h , v2.8h //weight mult. for row 3
[all...] |
| /external/libmpeg2/common/armv8/ |
| icv_variance_av8.s | 88 umull v24.8h, v2.8b, v2.8b
92 uaddl v25.4s, v24.4h, v26.4h
94 uaddl2 v24.4s, v24.8h, v26.8h
97 add v22.4s, v24.4s, v25.4s
|
| /external/libhevc/common/arm64/ |
| ihevc_intra_pred_luma_vert.s | 188 dup v24.16b,w12 //src[2nt+1]
213 sqxtun v24.8b, v28.8h
214 sqxtun2 v24.16b, v0.8h
218 rev64 v24.16b, v24.16b
219 mov v25.d[0], v24.d[1]
223 bsl v18.8b, v24.8b , v16.8b //only select row values from q12(predpixel)
239 bsl v1.8b, v24.8b , v16.8b
264 bsl v18.8b, v24.8b , v16.8b //only select row values from q12(predpixel)
277 bsl v1.8b, v24.8b , v16.8
[all...] |
| ihevc_inter_pred_chroma_vert_w16inp.s | 224 smull v24.4s, v3.4h, v16.4h //vmull_s16(src_tmp2, coeff_0)
226 smlal v24.4s, v4.4h, v17.4h
228 smlal v24.4s, v5.4h, v18.4h
230 smlal v24.4s, v6.4h, v19.4h
248 sqshrn v24.4h, v24.4s,#6 //right shift
263 sqrshrun v24.8b, v24.8h,#6 //rounding shift
269 st1 {v24.s}[0],[x9] //stores the loaded value
279 smull v24.4s, v3.4h, v16.4h //vmull_s16(src_tmp2, coeff_0
[all...] |
| ihevc_intra_pred_luma_horz.s | 209 usubl v24.8h, v30.8b, v28.8b
212 sshr v24.8h, v24.8h,#1
215 sqadd v22.8h, v26.8h , v24.8h
223 usubl v24.8h, v31.8b, v28.8b
226 sshr v24.8h, v24.8h,#1
229 sqadd v22.8h, v26.8h , v24.8h
291 usubl v24.8h, v30.8b, v28.8b
294 sshr v24.8h, v24.8h,#
[all...] |
| ihevc_itrans_recon_8x8.s | 192 smull v24.4s, v6.4h, v0.h[1] //// y1 * cos1(part of b0)
200 smlal v24.4s, v7.4h, v0.h[3] //// y1 * cos1 + y3 * cos3(part of b0)
234 smlal v24.4s, v14.4h, v1.h[1] //// y1 * cos1 + y3 * cos3 + y5 * sin3(part of b0)
245 smlal v24.4s, v15.4h, v1.h[3] //// b0 = y1 * cos1 + y3 * cos3 + y5 * sin3 + y7 * sin1(part of x0,x7)
255 add v20.4s, v14.4s , v24.4s //// a0 + b0(part of x0)
256 sub v6.4s, v14.4s , v24.4s //// a0 - b0(part of x7)
258 add v24.4s, v22.4s , v28.4s //// a2 + b2(part of x2)
269 sqrshrn v3.4h, v24.4s,#shift_stage1_idct //// x2 = (a2 + b2 + rnd) >> 7(shift_stage1_idct)
304 smull v24.4s, v6.4h, v0.h[1] //// y1 * cos1(part of b0)
309 smlal v24.4s, v7.4h, v0.h[3] //// y1 * cos1 + y3 * cos3(part of b0
[all...] |
| ihevc_itrans_recon_32x32.s | 216 smull v24.4s, v8.4h, v0.h[1] //// y1 * cos1(part of b0)
221 smlal v24.4s, v9.4h, v0.h[3] //// y1 * cos1 + y3 * cos3(part of b0)
256 smlal v24.4s, v14.4h, v1.h[1]
262 smlal v24.4s, v15.4h, v1.h[3]
286 smlal v24.4s, v8.4h, v2.h[1] //// y1 * cos1(part of b0)
291 smlal v24.4s, v9.4h, v2.h[3] //// y1 * cos1 + y3 * cos3(part of b0)
330 smlal v24.4s, v14.4h, v3.h[1]
336 smlal v24.4s, v15.4h, v3.h[3]
362 smlal v24.4s, v8.4h, v4.h[1] //// y1 * cos1(part of b0)
367 smlal v24.4s, v9.4h, v4.h[3] //// y1 * cos1 + y3 * cos3(part of b0
[all...] |
| ihevc_inter_pred_chroma_vert.s | 268 umull v24.8h, v16.8b, v1.8b
270 umlsl v24.8h, v7.8b, v0.8b
273 umlal v24.8h, v17.8b, v2.8b
275 umlsl v24.8h, v18.8b, v3.8b
300 sqrshrun v24.8b, v24.8h,#6
308 st1 {v24.8b},[x7],x3 //stores the loaded value
337 umull v24.8h, v16.8b, v1.8b
348 umlsl v24.8h, v7.8b, v0.8b
351 umlal v24.8h, v17.8b, v2.8
[all...] |
| ihevc_inter_pred_chroma_vert_w16inp_w16out.s | 222 smull v24.4s, v3.4h, v16.4h //vmull_s16(src_tmp2, coeff_0)
224 smlal v24.4s, v4.4h, v17.4h
226 smlal v24.4s, v5.4h, v18.4h
228 smlal v24.4s, v6.4h, v19.4h
245 sqshrn v24.4h, v24.4s,#6 //right shift
264 st1 {v24.2s},[x9] //stores the loaded value
273 smull v24.4s, v3.4h, v16.4h //vmull_s16(src_tmp2, coeff_0)
275 smlal v24.4s, v4.4h, v17.4h
278 smlal v24.4s, v5.4h, v18.4
[all...] |
| ihevc_intra_pred_luma_mode_3_to_9.s | 190 umull v24.8h, v12.8b, v7.8b //mul (row 0)
191 umlal v24.8h, v13.8b, v6.8b //mul (row 0)
197 rshrn v24.8b, v24.8h,#5 //round shft (row 0)
207 st1 {v24.8b},[x2], x3 //st (row 0)
233 umull v24.8h, v12.8b, v7.8b //mul (row 4)
234 umlal v24.8h, v13.8b, v6.8b //mul (row 4)
241 rshrn v24.8b, v24.8h,#5 //round shft (row 4)
251 st1 {v24.8b},[x2], x3 //st (row 4
[all...] |
| ihevc_sao_edge_offset_class0_chroma.s | 198 cmhi v24.16b, v28.16b , v30.16b //II vcltq_u8(pu1_cur_row, pu1_cur_row_tmp)
223 SUB v20.16b, v24.16b , v26.16b //II sign_left = vreinterpretq_s8_u8(vsubq_u8(cmp_lt, cmp_gt))
227 cmhi v24.16b, v28.16b , v30.16b //II vcltq_u8(pu1_cur_row, pu1_cur_row_tmp)
237 SUB v22.16b, v24.16b , v26.16b //II sign_right = vreinterpretq_s8_u8(vsubq_u8(cmp_lt, cmp_gt))
239 ADD v24.16b, v2.16b , v20.16b //II edge_idx = vaddq_s8(const_2, sign_left)
243 ADD v24.16b, v24.16b , v22.16b //II edge_idx = vaddq_s8(edge_idx, sign_right)
249 TBL v24.16b, {v5.16b},v24.16b //II vtbl1_s8(edge_idx_tbl, vget_low_s8(edge_idx))
257 AND v24.16b, v24.16b , v3.16b //II edge_idx = vandq_s8(edge_idx, au1_mask
[all...] |
| ihevc_inter_pred_chroma_horz_w16out.s | 131 dup v24.8b, v2.b[0] //coeffabs_0 = vdup_lane_u8(coeffabs, 0)
210 umlsl v30.8h, v0.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
229 umlsl v28.8h, v1.8b, v24.8b
267 umlsl v22.8h, v29.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
290 umlsl v20.8h, v31.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
308 umlsl v30.8h, v0.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
322 umlsl v28.8h, v1.8b, v24.8b
352 umlsl v22.8h, v29.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
371 umlsl v20.8h, v31.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
389 umlsl v30.8h, v0.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)/
[all...] |
| ihevc_intra_pred_filters_chroma_mode_11_to_17.s | 311 umull v24.8h, v12.8b, v7.8b //mul (row 0)
312 umlal v24.8h, v13.8b, v6.8b //mul (row 0)
318 rshrn v24.8b, v24.8h,#5 //round shft (row 0)
328 st1 {v24.8b},[x2], x3 //st (row 0)
354 umull v24.8h, v12.8b, v7.8b //mul (row 4)
355 umlal v24.8h, v13.8b, v6.8b //mul (row 4)
364 rshrn v24.8b, v24.8h,#5 //round shft (row 4)
374 st1 {v24.8b},[x2], x3 //st (row 4
[all...] |
| /external/libavc/encoder/armv8/ |
| ih264e_half_pel_av8.s | 176 sqrshrun v24.8b, v16.8h, #5 //// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column2,row1)
181 st1 {v23.8b, v24.8b}, [x1], #16 ////Store dest row1
320 uaddl v24.8h, v4.8b, v19.8b //// a0 + a5 (column3,row0)
322 umlal v24.8h, v10.8b, v1.8b //// a0 + a5 + 20a2 (column3,row0)
323 umlal v24.8h, v13.8b, v1.8b //// a0 + a5 + 20a2 + 20a3 (column3,row0)
324 umlsl v24.8h, v7.8b, v31.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 (column3,row0)
325 umlsl v24.8h, v16.8b, v31.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 (column3,row0)
326 mov v25.d[0], v24.d[1]
341 sqrshrun v4.8b, v24.8h, #5 //// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column3,row0)
350 ext v31.8b, v23.8b , v24.8b , #
[all...] |
| ih264e_evaluate_intra_chroma_modes_av8.s | 213 uabdl v24.8h, v1.8b, v29.8b
231 uabal v24.8h, v3.8b, v29.8b
247 uabal v24.8h, v5.8b, v29.8b
262 uabal v24.8h, v7.8b, v29.8b
283 uabal v24.8h, v1.8b, v31.8b
300 uabal v24.8h, v3.8b, v31.8b
317 uabal v24.8h, v5.8b, v31.8b
332 uabal v24.8h, v7.8b, v31.8b
359 add v24.8h, v22.8h , v24.8h ///D
[all...] |
| ih264e_evaluate_intra16x16_modes_av8.s | 181 uabdl v24.8h, v1.8b, v31.8b
202 uabal v24.8h, v3.8b, v31.8b
218 uabal v24.8h, v5.8b, v31.8b
234 uabal v24.8h, v7.8b, v31.8b
251 uabal v24.8h, v1.8b, v31.8b
267 uabal v24.8h, v3.8b, v31.8b
284 uabal v24.8h, v5.8b, v31.8b
301 uabal v24.8h, v7.8b, v31.8b
321 uabal v24.8h, v1.8b, v31.8b
338 uabal v24.8h, v3.8b, v31.8
[all...] |
| /external/llvm/test/MC/Hexagon/ |
| v60-permute.s | 20 #CHECK: 1ff1d81f { v31.h = vpack(v24.w{{ *}},{{ *}}v17.w):sat }
21 v31.h=vpack(v24.w,v17.w):sat
50 #CHECK: 1e01d8fe { v30.h = vshuff(v24.h) }
51 v30.h=vshuff(v24.h)
|
| v60-shift.s | 5 #CHECK: 198fd829 { v9.uw = vlsr(v24.uw,{{ *}}r15) }
6 v9.uw=vlsr(v24.uw,r15)
38 #CHECK: 1e02c2d8 { v24.h = vpopcount(v2.h) }
39 v24.h=vpopcount(v2.h)
|
