| /external/llvm/test/MC/AArch64/ |
| neon-2velem.s | 10 mla v0.2s, v1.2s, v22.s[2]
12 mla v3.4s, v8.4s, v22.s[3]
15 // CHECK: mla v0.2s, v1.2s, v22.s[2] // encoding: [0x20,0x08,0x96,0x2f]
17 // CHECK: mla v3.4s, v8.4s, v22.s[3] // encoding: [0x03,0x09,0xb6,0x6f]
30 mls v0.2s, v1.2s, v22.s[2]
32 mls v3.4s, v8.4s, v22.s[3]
35 // CHECK: mls v0.2s, v1.2s, v22.s[2] // encoding: [0x20,0x48,0x96,0x2f]
37 // CHECK: mls v3.4s, v8.4s, v22.s[3] // encoding: [0x03,0x49,0xb6,0x6f]
52 fmla v0.2s, v1.2s, v22.s[2]
54 fmla v3.4s, v8.4s, v22.s[3
[all...] |
| neon-max-min.s | 85 fmin v10.4h, v15.4h, v22.4h
86 fmin v10.8h, v15.8h, v22.8h
87 fmin v10.2s, v15.2s, v22.2s
91 // CHECK: fmin v10.4h, v15.4h, v22.4h // encoding: [0xea,0x35,0xd6,0x0e]
92 // CHECK: fmin v10.8h, v15.8h, v22.8h // encoding: [0xea,0x35,0xd6,0x4e]
93 // CHECK: fmin v10.2s, v15.2s, v22.2s // encoding: [0xea,0xf5,0xb6,0x0e]
115 fminnm v10.4h, v15.4h, v22.4h
116 fminnm v10.8h, v15.8h, v22.8h
117 fminnm v10.2s, v15.2s, v22.2s
121 // CHECK: fminnm v10.4h, v15.4h, v22.4h // encoding: [0xea,0x05,0xd6,0x0e
[all...] |
| /cts/tools/dasm/test/ |
| all_opcodes.d | 77 cmpl-float v11, v22, v33
78 cmpg-float v11, v22, v33
79 cmpl-double v11, v22, v33
80 cmpg-double v11, v22, v33
81 cmp-long v11, v22, v33
93 aget v11, v22, v33
94 aget-wide v11, v22, v33
95 aget-object v11, v22, v33
96 aget-boolean v11, v22, v33
97 aget-byte v11, v22, v33
[all...] |
| /external/libhevc/common/arm64/ |
| 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], x1
[all...] |
| 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_gt))
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
[all...] |
| 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_gt))
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, v23.8
[all...] |
| 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
[all...] |
| 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
320 //mov v22.s[1],v23.s[0]
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)/
[all...] |
| 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
[all...] |
| 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, v22.8
[all...] |
| 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
[all...] |
| 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], x1
[all...] |
| 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
[all...] |
| 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_ds
[all...] |
| 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
[all...] |
| /external/libavc/common/armv8/ |
| ih264_inter_pred_luma_horz_hpel_vert_hpel_av8.s | 112 uaddl v22.8h, v2.8b, v8.8b
119 mls v18.8h, v22.8h , v30.8h
120 uaddl v22.8h, v12.8b, v17.8b
123 mla v22.8h, v24.8h , v28.8h
124 mls v22.8h, v26.8h , v30.8h
151 ext v24.16b, v20.16b , v22.16b , #4
152 ext v26.16b, v20.16b , v22.16b , #6
153 ext v0.16b, v20.16b , v22.16b , #10
156 ext v24.16b, v20.16b , v22.16b , #2
157 ext v26.16b, v20.16b , v22.16b , #
[all...] |
| ih264_inter_pred_luma_horz_qpel_vert_hpel_av8.s | 174 uaddl v22.8h, v2.8b, v8.8b
181 mls v18.8h, v22.8h , v30.8h
182 uaddl v22.8h, v12.8b, v17.8b
185 mla v22.8h, v24.8h , v28.8h
186 mls v22.8h, v26.8h , v30.8h
191 st1 {v22.4s}, [x9]
192 ext v22.16b, v18.16b , v20.16b , #10
198 saddl v26.4s, v18.4h, v22.4h
202 saddl2 v22.4s, v18.8h, v22.8
[all...] |
| ih264_inter_pred_filters_luma_vert_av8.s | 115 movi v22.8h, #20 // Filter coeff 0x14 into Q11
138 mla v14.8h, v12.8h, v22.8h // temp += temp1 * 20
141 mla v20.8h, v18.8h , v22.8h // temp4 += temp3 * 20
148 mla v16.8h, v12.8h , v22.8h
155 mla v14.8h, v12.8h , v22.8h
162 mla v18.8h, v12.8h , v22.8h
169 mla v16.8h, v12.8h , v22.8h
177 mla v14.8h, v12.8h , v22.8h
183 mla v18.8h, v12.8h , v22.8h
197 mla v14.8h, v12.8h , v22.8h // temp += temp1 * 2
[all...] |
| ih264_inter_pred_chroma_av8.s | 161 umull v22.8h, v1.8b, v28.8b
163 umlal v22.8h, v4.8b, v29.8b
165 umlal v22.8h, v6.8b, v30.8b
167 umlal v22.8h, v9.8b, v31.8b
168 sqrshrun v27.8b, v22.8h, #6
202 umull v22.8h, v1.8b, v28.8b
203 umlal v22.8h, v4.8b, v29.8b
204 umlal v22.8h, v6.8b, v30.8b
206 umlal v22.8h, v9.8b, v31.8b
208 sqrshrun v27.8b, v22.8h, #
[all...] |
| ih264_resi_trans_quant_av8.s | 167 shl v22.4h, v20.4h, #1 //u_shift(x2,1,shft)
175 sub v27.4h, v21.4h , v22.4h //x8 = x3 - u_shift(x2,1,shft);
200 add v22.4s, v2.4s, v23.4s
207 sshl v22.4s, v22.4s, v24.4s //shift row 3
212 xtn v22.4h, v22.4s //narrow row 3
217 neg v26.8h, v22.8h //get negative
223 cmeq v2.4h, v22.4h, #0
228 bsl v6.8b, v22.8b, v26.8b //restore sign of row 1 and
[all...] |
| ih264_inter_pred_luma_horz_hpel_vert_qpel_av8.s | 150 movi v22.8h, #20 // Filter coeff 0x14 into Q11
166 mla v6.8h, v8.8h , v22.8h
180 mla v8.8h, v10.8h , v22.8h
194 mla v10.8h, v12.8h , v22.8h
208 mla v12.8h, v14.8h , v22.8h
222 mla v14.8h, v16.8h , v22.8h
239 mla v16.8h, v18.8h , v22.8h
261 smlal v18.4s, v30.4h, v22.4h
263 smlal2 v6.4s, v30.8h, v22.8h
267 mla v20.8h, v2.8h , v22.8
[all...] |
| /external/libavc/encoder/armv8/ |
| ih264e_half_pel_av8.s | 174 sqrshrun v22.8b, v12.8h, #5 //// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column3,row0)
180 st1 {v22.h}[0], [x1], x3
311 uaddl v22.8h, v3.8b, v18.8b //// a0 + a5 (column2,row0)
312 umlal v22.8h, v9.8b, v1.8b //// a0 + a5 + 20a2 (column2,row0)
313 umlal v22.8h, v12.8b, v1.8b //// a0 + a5 + 20a2 + 20a3 (column2,row0)
314 umlsl v22.8h, v6.8b, v31.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 (column2,row0)
315 umlsl v22.8h, v15.8b, v31.8b //// a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 (column2,row0)
317 mov v23.d[0], v22.d[1]
329 ext v31.8b, v21.8b , v22.8b , #2
330 sqrshrun v3.8b, v22.8h, #5 //// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column2,row0
[all...] |
| /external/libhevc/decoder/arm64/ |
| 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, v22.8
[all...] |
| /external/libmpeg2/common/armv8/ |
| icv_variance_av8.s | 87 umull v22.8h, v1.8b, v1.8b
91 uaddl v21.4s, v20.4h, v22.4h
93 uaddl2 v20.4s, v20.8h, v22.8h
97 add v22.4s, v24.4s, v25.4s
98 add v20.4s, v20.4s, v22.4s
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/ppc/ |
| filter_bilinear_altivec.asm | 70 ;# v22 tmp
97 lvx v22, r10, r3
102 vperm v21, v21, v22, v17
114 lvx v22, r10, r3
120 vperm \V, v21, v22, v17
132 vmuleub v22, \P0, v20 ;# 64 + 4 positive taps
133 vadduhm v22, v18, v22
138 vadduhm v22, v22, v24 ;# Re = evens, saturation unnecessar
[all...] |
