| /external/llvm/test/MC/AArch64/ |
| D | neon-2velem.s | 10 mla v0.2s, v1.2s, v22.s[2]
12 mla v3.4s, v8.4s, v22.s[3]
30 mls v0.2s, v1.2s, v22.s[2]
32 mls v3.4s, v8.4s, v22.s[3]
52 fmla v0.2s, v1.2s, v22.s[2]
54 fmla v3.4s, v8.4s, v22.s[3]
56 fmla v0.2d, v1.2d, v22.d[1]
70 fmls v0.2s, v1.2s, v22.s[2]
72 fmls v3.4s, v8.4s, v22.s[3]
74 fmls v0.2d, v1.2d, v22.d[1]
[all …]
|
| /external/libavc/common/armv8/ |
| D | ih264_inter_pred_luma_horz_qpel_vert_hpel_av8.s | 178 uaddl v22.8h, v2.8b, v8.8b
185 mls v18.8h, v22.8h , v30.8h
186 uaddl v22.8h, v12.8b, v17.8b
189 mla v22.8h, v24.8h , v28.8h
190 mls v22.8h, v26.8h , v30.8h
195 st1 {v22.4s}, [x9]
196 ext v22.16b, v18.16b , v20.16b , #10
202 saddl v26.4s, v18.4h, v22.4h
206 saddl2 v22.4s, v18.8h, v22.8h
207 smlal2 v22.4s, v0.8h, v28.8h
[all …]
|
| D | ih264_inter_pred_luma_horz_hpel_vert_hpel_av8.s | 116 uaddl v22.8h, v2.8b, v8.8b
123 mls v18.8h, v22.8h , v30.8h
124 uaddl v22.8h, v12.8b, v17.8b
127 mla v22.8h, v24.8h , v28.8h
128 mls v22.8h, v26.8h , v30.8h
155 ext v24.16b, v20.16b , v22.16b , #4
156 ext v26.16b, v20.16b , v22.16b , #6
157 ext v0.16b, v20.16b , v22.16b , #10
160 ext v24.16b, v20.16b , v22.16b , #2
161 ext v26.16b, v20.16b , v22.16b , #8
[all …]
|
| D | ih264_inter_pred_chroma_av8.s | 167 umull v22.8h, v1.8b, v28.8b
169 umlal v22.8h, v4.8b, v29.8b
171 umlal v22.8h, v6.8b, v30.8b
173 umlal v22.8h, v9.8b, v31.8b
174 sqrshrun v27.8b, v22.8h, #6
208 umull v22.8h, v1.8b, v28.8b
209 umlal v22.8h, v4.8b, v29.8b
210 umlal v22.8h, v6.8b, v30.8b
212 umlal v22.8h, v9.8b, v31.8b
214 sqrshrun v27.8b, v22.8h, #6
[all …]
|
| D | ih264_inter_pred_filters_luma_vert_av8.s | 119 movi v22.8h, #20 // Filter coeff 0x14 into Q11
142 mla v14.8h, v12.8h, v22.8h // temp += temp1 * 20
145 mla v20.8h, v18.8h , v22.8h // temp4 += temp3 * 20
152 mla v16.8h, v12.8h , v22.8h
159 mla v14.8h, v12.8h , v22.8h
166 mla v18.8h, v12.8h , v22.8h
173 mla v16.8h, v12.8h , v22.8h
181 mla v14.8h, v12.8h , v22.8h
187 mla v18.8h, v12.8h , v22.8h
201 mla v14.8h, v12.8h , v22.8h // temp += temp1 * 20
[all …]
|
| D | ih264_inter_pred_luma_vert_qpel_av8.s | 126 movi v22.8h, #20 // Filter coeff 0x14 into Q11
149 mla v14.8h, v12.8h , v22.8h // temp += temp1 * 20
152 mla v20.8h, v18.8h , v22.8h // temp4 += temp3 * 20
159 mla v16.8h, v12.8h , v22.8h
166 mla v14.8h, v12.8h , v22.8h
175 mla v18.8h, v12.8h , v22.8h
182 mla v16.8h, v12.8h , v22.8h
192 mla v14.8h, v12.8h , v22.8h
198 mla v18.8h, v12.8h , v22.8h
216 mla v14.8h, v12.8h , v22.8h // temp += temp1 * 20
[all …]
|
| /external/libavc/encoder/armv8/ |
| D | ih264e_half_pel_av8.s | 176 …sqrshrun v22.8b, v12.8h, #5 //// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column3…
182 st1 {v22.h}[0], [x1], x3
315 uaddl v22.8h, v3.8b, v18.8b //// a0 + a5 (column2,row0)
316 umlal v22.8h, v9.8b, v1.8b //// a0 + a5 + 20a2 (column2,row0)
317 umlal v22.8h, v12.8b, v1.8b //// a0 + a5 + 20a2 + 20a3 (column2,row0)
318 umlsl v22.8h, v6.8b, v31.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 (column2,row0)
319 umlsl v22.8h, v15.8b, v31.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 (column2,row0)
321 mov v23.d[0], v22.d[1]
333 ext v31.8b, v21.8b , v22.8b , #2
334 …sqrshrun v3.8b, v22.8h, #5 //// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column2…
[all …]
|
| /external/libhevc/decoder/arm64/ |
| D | ihevcd_fmt_conv_420sp_to_rgba8888.s | 204 sMULL2 v22.4s, v6.8h, v0.h[0] ////(V-128)*C1 FOR R
218 sqshrn2 v7.8h, v22.4s,#13 ////D11 = (V-128)*C1>>13 4 16-BIT VALUES
231 UADDW v22.8h, v7.8h , v31.8b ////Q11 - HAS Y + R
241 sqxtun v22.8b, v22.8h
254 ZIP1 v27.8b, v22.8b, v23.8b
255 ZIP2 v23.8b, v22.8b, v23.8b
256 mov v22.d[0], v27.d[0]
261 mov v22.d[1], v23.d[0]
266 ZIP1 v25.8h, v20.8h, v22.8h
267 ZIP2 v19.8h, v20.8h, v22.8h
[all …]
|
| /external/libhevc/common/arm64/ |
| D | ihevc_intra_pred_chroma_ver.s | 121 ld2 {v22.8b, v23.8b}, [x6] //16 loads (col 16:31)
132 st2 {v22.8b, v23.8b}, [x2], x11
133 st2 {v22.8b, v23.8b}, [x5], x11
134 st2 {v22.8b, v23.8b}, [x8], x11
135 st2 {v22.8b, v23.8b}, [x10], x11
145 st2 {v22.8b, v23.8b}, [x2], x11
146 st2 {v22.8b, v23.8b}, [x5], x11
147 st2 {v22.8b, v23.8b}, [x8], x11
148 st2 {v22.8b, v23.8b}, [x10], x11
158 st2 {v22.8b, v23.8b}, [x2], x11
[all …]
|
| D | ihevc_sao_edge_offset_class1_chroma.s | 190 cmhi v22.16b, v18.16b , v30.16b //II vcgtq_u8(pu1_cur_row, pu1_top_row)
197 …SUB v28.16b, v24.16b , v22.16b //II sign_down = vreinterpretq_s8_u8(vsubq_u8(cmp_lt, cmp…
199 ADD v22.16b, v0.16b , v16.16b //II edge_idx = vaddq_s8(const_2, sign_up)
207 ADD v22.16b, v22.16b , v28.16b //II edge_idx = vaddq_s8(edge_idx, sign_down)
216 TBL v22.16b, {v6.16b},v22.16b //II vtbl1_s8(edge_idx_tbl, vget_low_s8(edge_idx))
222 mov v23.d[0], v22.d[1]
223 UZP1 v27.8b, v22.8b, v23.8b
224 UZP2 v23.8b, v22.8b, v23.8b
225 mov v22.8b,v27.8b
232 TBL v24.8b, {v7.16b},v22.8b //offset = vtbl1_s8(offset_tbl, vget_low_s8(edge_idx))
[all …]
|
| D | ihevc_sao_edge_offset_class1.s | 161 cmhi v22.16b, v18.16b , v30.16b //II vcgtq_u8(pu1_cur_row, pu1_top_row)
167 …SUB v1.16b, v24.16b , v22.16b //II sign_down = vreinterpretq_s8_u8(vsubq_u8(cmp_lt, cmp_…
169 ADD v22.16b, v0.16b , v16.16b //II edge_idx = vaddq_s8(const_2, sign_up)
174 ADD v22.16b, v22.16b , v1.16b //II edge_idx = vaddq_s8(edge_idx, sign_down)
178 TBL v22.16b, {v6.16b},v22.16b //II vtbl1_s8(edge_idx_tbl, vget_low_s8(edge_idx))
191 … TBL v24.16b, {v7.16b},v22.16b //II offset = vtbl1_s8(offset_tbl, vget_low_s8(edge_idx))
228 ADD v22.16b, v0.16b , v16.16b //edge_idx = vaddq_s8(const_2, sign_up)
229 ADD v22.16b, v22.16b , v20.16b //edge_idx = vaddq_s8(edge_idx, sign_down)
230 TBL v22.16b, {v6.16b},v22.16b //vtbl1_s8(edge_idx_tbl, vget_low_s8(edge_idx))
233 TBL v24.16b, {v7.16b},v22.16b //offset = vtbl1_s8(offset_tbl, vget_low_s8(edge_idx))
[all …]
|
| D | ihevc_inter_pred_chroma_copy_w16out.s | 146 ld1 {v22.8b},[x5],x2 //vld1_u8(pu1_src_tmp)
150 uxtl v22.8h, v22.8b //vmovl_u8(vld1_u8(pu1_src_tmp)
152 shl v22.2d, v22.2d,#6 //vshlq_n_s64(temp, 6)
154 st1 {v22.1d},[x10],x6 //vst1q_lane_s64(pi2_dst_tmp, temp, 0)
190 ld1 {v22.8b},[x5],x2 //vld1_u8(pu1_src_tmp)
194 uxtl v22.8h, v22.8b //vmovl_u8(vld1_u8(pu1_src_tmp)
196 shl v22.2d, v22.2d,#6 //vshlq_n_s64(temp, 6)
198 st1 {v22.1d},[x10],x6 //vst1q_lane_s64(pi2_dst_tmp, temp, 0)
229 uxtl v22.8h, v7.8b //vmovl_u8(vld1_u8(pu1_src_tmp)
234 shl v6.8h, v22.8h,#6 //vshlq_n_s16(tmp, 6)
[all …]
|
| D | ihevc_intra_pred_luma_vert.s | 122 ld1 {v22.8b, v23.8b}, [x6] //16 loads (col 16:31)
131 st1 {v22.8b, v23.8b}, [x2], x11
132 st1 {v22.8b, v23.8b}, [x5], x11
133 st1 {v22.8b, v23.8b}, [x8], x11
134 st1 {v22.8b, v23.8b}, [x10], x11
144 st1 {v22.8b, v23.8b}, [x2], x11
145 st1 {v22.8b, v23.8b}, [x5], x11
146 st1 {v22.8b, v23.8b}, [x8], x11
147 st1 {v22.8b, v23.8b}, [x10], x11
156 st1 {v22.8b, v23.8b}, [x2], x11
[all …]
|
| D | ihevc_intra_pred_luma_mode_3_to_9.s | 139 smull v22.8h, v30.8b, v31.8b //(col+1)*intra_pred_angle [0:7](col)
157 xtn v6.8b, v22.8h
164 sshr v22.8h, v22.8h,#5
170 sqxtn v1.8b, v22.8h
200 umull v22.8h, v16.8b, v7.8b //mul (row 1)
201 umlal v22.8h, v17.8b, v6.8b //mul (row 1)
208 rshrn v22.8b, v22.8h,#5 //round shft (row 1)
218 st1 {v22.8b},[x2], x3 //st (row 1)
244 umull v22.8h, v16.8b, v7.8b //mul (row 5)
245 umlal v22.8h, v17.8b, v6.8b //mul (row 5)
[all …]
|
| D | ihevc_inter_pred_chroma_horz.s | 240 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//
247 umlsl v22.8h, v29.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
265 umlal v22.8h, v12.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//
272 umlsl v22.8h, v14.8b, v27.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//
308 sqrshrun v22.8b, v22.8h,#6
321 st1 { v22.4h},[x1],x6 //store the result pu1_dst
354 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//
359 umlsl v22.8h, v29.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
361 umlal v22.8h, v12.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//
364 umlsl v22.8h, v14.8b, v27.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//
[all …]
|
| D | ihevc_itrans_recon_8x8.s | 208 smull v22.4s, v10.4h, v0.h[0] //// y4 * cos4(part of c0 and c1)
242 add v10.4s, v20.4s , v22.4s //// c0 = y0 * cos4 + y4 * cos4(part of a0 and a1)
243 sub v20.4s, v20.4s , v22.4s //// c1 = y0 * cos4 - y4 * cos4(part of a0 and a1)
252 sub v22.4s, v20.4s , v18.4s //// a2 = c1 - d1(part of x2,x5)
258 add v24.4s, v22.4s , v28.4s //// a2 + b2(part of x2)
259 sub v22.4s, v22.4s , v28.4s //// a2 - b2(part of x5)
270 sqrshrn v14.4h, v22.4s,#shift_stage1_idct //// x5 = (a2 - b2 + rnd) >> 7(shift_stage1_idct)
322 sub v22.4s, v20.4s , v18.4s //// a2 = c1 - d1(part of x2,x5)
328 add v24.4s, v22.4s , v28.4s //// a2 + b2(part of x2)
329 sub v22.4s, v22.4s , v28.4s //// a2 - b2(part of x5)
[all …]
|
| D | ihevc_intra_pred_filters_luma_mode_11_to_17.s | 255 smull v22.8h, v30.8b, v31.8b //(col+1)*intra_pred_angle [0:7](col)
277 xtn v6.8b, v22.8h
284 sshr v22.8h, v22.8h,#5
292 sqxtn v19.8b, v22.8h
320 umull v22.8h, v16.8b, v7.8b //mul (row 1)
321 umlal v22.8h, v17.8b, v6.8b //mul (row 1)
328 rshrn v22.8b, v22.8h,#5 //round shft (row 1)
338 st1 {v22.8b},[x2], x3 //st (row 1)
364 umull v22.8h, v16.8b, v7.8b //mul (row 5)
365 umlal v22.8h, v17.8b, v6.8b //mul (row 5)
[all …]
|
| D | ihevc_intra_pred_luma_horz.s | 215 sqadd v22.8h, v26.8h , v24.8h
218 sqxtun v22.8b, v22.8h
220 st1 {v22.8b},[x2],#8
229 sqadd v22.8h, v26.8h , v24.8h
232 sqxtun v22.8b, v22.8h
234 st1 {v22.8b},[x2],x3
297 sqadd v22.8h, v26.8h , v24.8h
300 sqxtun v22.8b, v22.8h
302 st1 {v22.8b},[x2],x3
341 sqadd v22.8h, v26.8h , v24.8h
[all …]
|
| D | ihevc_inter_pred_luma_copy_w16out.s | 115 ld1 {v22.8b},[x5],x2 //vld1_u8(pu1_src_tmp)
119 uxtl v22.8h, v22.8b //vmovl_u8(vld1_u8(pu1_src_tmp)
121 shl v22.2d, v22.2d,#6 //vshlq_n_s64(temp, 6)
123 st1 {v22.d}[0],[x10],x6 //vst1q_lane_s64(pi2_dst_tmp, temp, 0)
169 uxtl v22.8h, v7.8b //vmovl_u8(vld1_u8(pu1_src_tmp)
174 shl v6.8h, v22.8h,#6 //vshlq_n_s16(tmp, 6)
207 uxtl v22.8h, v7.8b //vmovl_u8(vld1_u8(pu1_src_tmp)
227 shl v6.8h, v22.8h,#6 //vshlq_n_s16(tmp, 6)
249 uxtl v22.8h, v7.8b //vmovl_u8(vld1_u8(pu1_src_tmp)
256 shl v6.8h, v22.8h,#6 //vshlq_n_s16(tmp, 6)
|
| D | ihevc_intra_pred_chroma_mode_27_to_33.s | 212 umull v22.8h, v20.8b, v24.8b //(iv)vmull_u8(ref_main_idx, dup_const_32_fract)
216 umlal v22.8h, v21.8b, v25.8b //(iv)vmull_u8(ref_main_idx_1, dup_const_fract)
236 rshrn v22.8b, v22.8h,#5 //(iv)shift_res = vrshrn_n_u16(add_res, 5)
252 st1 {v22.8b},[x0],x3 //(iv)
294 umull v22.8h, v20.8b, v24.8b //(viii)vmull_u8(ref_main_idx, dup_const_32_fract)
298 umlal v22.8h, v21.8b, v25.8b //(viii)vmull_u8(ref_main_idx_1, dup_const_fract)
318 rshrn v22.8b, v22.8h,#5 //(viii)shift_res = vrshrn_n_u16(add_res, 5)
337 st1 {v22.8b},[x0] //(viii)
357 umull v22.8h, v20.8b, v24.8b //(iv)vmull_u8(ref_main_idx, dup_const_32_fract)
361 umlal v22.8h, v21.8b, v25.8b //(iv)vmull_u8(ref_main_idx_1, dup_const_fract)
[all …]
|
| D | ihevc_intra_pred_luma_mode_27_to_33.s | 217 umull v22.8h, v20.8b, v24.8b //(iv)vmull_u8(ref_main_idx, dup_const_32_fract)
221 umlal v22.8h, v21.8b, v25.8b //(iv)vmull_u8(ref_main_idx_1, dup_const_fract)
241 rshrn v22.8b, v22.8h,#5 //(iv)shift_res = vrshrn_n_u16(add_res, 5)
257 st1 {v22.8b},[x0],x3 //(iv)
299 umull v22.8h, v20.8b, v24.8b //(viii)vmull_u8(ref_main_idx, dup_const_32_fract)
303 umlal v22.8h, v21.8b, v25.8b //(viii)vmull_u8(ref_main_idx_1, dup_const_fract)
324 rshrn v22.8b, v22.8h,#5 //(viii)shift_res = vrshrn_n_u16(add_res, 5)
343 st1 {v22.8b},[x0] //(viii)
363 umull v22.8h, v20.8b, v24.8b //(iv)vmull_u8(ref_main_idx, dup_const_32_fract)
367 umlal v22.8h, v21.8b, v25.8b //(iv)vmull_u8(ref_main_idx_1, dup_const_fract)
[all …]
|
| D | ihevc_inter_pred_chroma_horz_w16out.s | 252 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//
267 umlsl v22.8h, v29.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
275 umlal v22.8h, v12.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//
282 umlsl v22.8h, v14.8b, v27.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//
314 st1 { v22.8h},[x1],#16 //store the result pu1_dst
347 umull v22.8h, v10.8b, v25.8b //mul_res = vmull_u8(src[0_3], coeffabs_3)//
352 umlsl v22.8h, v29.8b, v24.8b //mul_res = vmlsl_u8(src[0_2], coeffabs_2)//
354 umlal v22.8h, v12.8b, v26.8b //mul_res = vmlsl_u8(src[0_0], coeffabs_0)//
359 umlsl v22.8h, v14.8b, v27.8b //mul_res = vmlal_u8(src[0_1], coeffabs_1)//
405 st1 { v22.8h},[x1],#16 //store the result pu1_dst
[all …]
|
| D | ihevc_intra_pred_chroma_mode_3_to_9.s | 133 smull v22.8h, v30.8b, v31.8b //(col+1)*intra_pred_angle [0:7](col)
150 xtn v6.8b, v22.8h
157 sshr v22.8h, v22.8h,#5
163 sqxtn v2.8b, v22.8h
199 umull v22.8h, v16.8b, v7.8b //mul (row 1)
200 umlal v22.8h, v17.8b, v6.8b //mul (row 1)
207 rshrn v22.8b, v22.8h,#5 //round shft (row 1)
217 st1 {v22.8b},[x2], x3 //st (row 1)
245 umull v22.8h, v16.8b, v7.8b //mul (row 5)
246 umlal v22.8h, v17.8b, v6.8b //mul (row 5)
[all …]
|
| /external/llvm/test/MC/AMDGPU/ |
| D | vop3.s | 333 v_div_scale_f32 v24, vcc, v22, v22, v20
337 v_div_scale_f32 v24, vcc, s[10:11], v22, v20
341 v_div_scale_f32 v24, s[10:11], v22, v22, v20
345 v_div_scale_f32 v24, vcc, v22, 1.0, v22
349 v_div_scale_f32 v24, vcc, v22, v22, -2.0
353 v_div_scale_f32 v24, vcc, v22, v22, 0xc0000000
|
| /external/libmpeg2/common/armv8/ |
| D | impeg2_idct.s | 196 uaddw v22.8h, v30.8h , v7.8b
200 sqxtun v7.8b, v22.8h
408 smull v22.4s, v10.4h, v0.h[0] //// y4 * cos4(part of c0 and c1)
442 add v10.4s, v20.4s , v22.4s //// c0 = y0 * cos4 + y4 * cos4(part of a0 and a1)
443 sub v20.4s, v20.4s , v22.4s //// c1 = y0 * cos4 - y4 * cos4(part of a0 and a1)
452 sub v22.4s, v20.4s , v18.4s //// a2 = c1 - d1(part of x2,x5)
458 add v24.4s, v22.4s , v28.4s //// a2 + b2(part of x2)
459 sub v22.4s, v22.4s , v28.4s //// a2 - b2(part of x5)
470 sqrshrn v14.4h, v22.4s, #idct_stg1_shift //// x5 = (a2 - b2 + rnd) >> 7(IDCT_STG1_SHIFT)
523 sub v22.4s, v20.4s , v18.4s //// a2 = c1 - d1(part of x2,x5)
[all …]
|