Home
last modified time | relevance | path

Searched refs:LD1 (Results 1 – 25 of 110) sorted by relevance

12345

/external/libxaac/decoder/armv8/
Dixheaacd_sbr_qmfsyn64_winadd.s54 LD1 {v0.4h}, [x0], #8
58 LD1 {v1.4h}, [x2], #8
81 LD1 {v2.4h}, [x0], x8
86 LD1 {v3.4h}, [x2], x9
88 LD1 {v4.4h}, [x0], x8
91 LD1 {v5.4h}, [x2], x9
93 LD1 {v6.4h}, [x0], x8
96 LD1 {v7.4h}, [x2], x9
98 LD1 {v8.4h}, [x0], x8
101 LD1 {v9.4h}, [x2], x9
[all …]
Dixheaacd_sbr_qmf_analysis32_neon.s143 LD1 {v0.4h}, [x0], #8
154 LD1 {v2.4h}, [x0], x6
168 LD1 {v4.4h}, [x0], x6
175 LD1 {v6.4h}, [x0], x6
182 LD1 {v8.4h}, [x0], x6
190 LD1 {v10.4h}, [x1], #8
201 LD1 {v12.4h}, [x1], x6
211 LD1 {v14.4h}, [x1], x6
216 LD1 {v16.4h}, [x1], x6
223 LD1 {v18.4h}, [x1], x6
[all …]
Dixheaacd_inv_dit_fft_8pt.s33 LD1 {v1.s}[0], [x0], x5
34 LD1 {v2.s}[0], [x6], x5
35 LD1 {v1.s}[1], [x0], x5
36 LD1 {v2.s}[1], [x6], x5
37 LD1 {v3.s}[0], [x0], x5
38 LD1 {v4.s}[0], [x6], x5
39 LD1 {v3.s}[1], [x0], x5
40 LD1 {v4.s}[1], [x6], x5
41 LD1 {v5.s}[0], [x0], x5
42 LD1 {v6.s}[0], [x6], x5
[all …]
Dixheaacd_cos_sin_mod_loop2.s58 LD1 {v2.s}[0], [x3]
69 LD1 {v3.s}[0], [x3] // im = *psubband1;
77 LD1 {v2.s}[1], [x11] //re = *psubband12;
84 LD1 {v4.s}[0], [x10]
101 LD1 {v3.s}[1], [x11]
132 LD1 {v2.2s}, [x0]
133 LD1 {v3.2s}, [x10]
179 LD1 {v2.s}[0], [x3]
180 LD1 {v2.s}[1], [x11]
Dixheaacd_no_lap1.s61 LD1 {V0.4S}, [X6], X8
64 LD1 {V6.4S}, [X6], X8
78 LD1 {V0.4S}, [X6], X8
92 LD1 {V6.4S}, [X6], X8
/external/XNNPACK/src/f32-dwconv/
Dup4x9-minmax-aarch64-neonfma-cortex-a55.S96 LD1 {v0.2S}, [x17], 8
99 LD1 {v1.2S}, [x17], 8
102 LD1 {v4.2S}, [x7], 8
105 LD1 {v5.2S}, [x17], 8
108 LD1 {v6.2S}, [x7], 8
111 LD1 {v7.2S}, [x17], 8
114 LD1 {v28.2S}, [x8], 8
117 LD1 {v29.2S}, [x17], 8
120 LD1 {v10.2S}, [x8], 8
123 LD1 {v11.2S}, [x17], 8
[all …]
/external/libhevc/common/arm64/
Dihevc_inter_pred_chroma_copy.s121 LD1 {v0.s}[0],[x0] //vld1_lane_u32((uint32_t *)pu1_src_tmp, src_tmp, 0)
125 LD1 {v0.s}[0],[x7],x2 //vld1_lane_u32((uint32_t *)pu1_src_tmp, src_tmp, 0)
128 LD1 {v0.s}[0],[x7],x2 //vld1_lane_u32((uint32_t *)pu1_src_tmp, src_tmp, 0)
131 LD1 {v0.s}[0],[x7],x2 //vld1_lane_u32((uint32_t *)pu1_src_tmp, src_tmp, 0)
152 LD1 {v0.s}[0],[x0] //vld1_lane_u32((uint32_t *)pu1_src_tmp, src_tmp, 0)
156 LD1 {v0.s}[0],[x7],x2 //vld1_lane_u32((uint32_t *)pu1_src_tmp, src_tmp, 0)
176 LD1 {v0.8b},[x0],#8 //vld1_u8(pu1_src_tmp)
179 LD1 {v1.8b},[x7],x2 //vld1_u8(pu1_src_tmp)
182 LD1 {v2.8b},[x7],x2 //vld1_u8(pu1_src_tmp)
184 LD1 {v3.8b},[x7],x2 //vld1_u8(pu1_src_tmp)
[all …]
Dihevc_sao_band_offset_luma.s100 LD1 {v1.8b},[x14],#8 //band_table.val[0]
103 LD1 {v2.8b},[x14],#8 //band_table.val[1]
110 LD1 {v3.8b},[x14],#8 //band_table.val[2]
114 LD1 {v4.8b},[x14],#8 //band_table.val[3]
118 LD1 {v0.8b},[x4],#8 //Load pu1_src[(ht - 1) * src_strd + col]
123 LD1 {v30.8b},[x6] //pi1_sao_offset load
205 LD1 {v13.8b},[x4] //au1_cur_row = vld1_u8(pu1_src_cpy)
208 LD1 {v15.8b},[x5] //au1_cur_row = vld1_u8(pu1_src_cpy)
210 LD1 {v17.8b},[x6] //au1_cur_row = vld1_u8(pu1_src_cpy)
212 LD1 {v19.8b},[x10] //au1_cur_row = vld1_u8(pu1_src_cpy)
Dihevc_sao_edge_offset_class1.s116 LD1 {v6.8b},[x14] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
117 LD1 {v7.8b},[x6] //offset_tbl = vld1_s8(pi1_sao_offset)
131 …LD1 {v1.16b},[x9],#16 //pu1_top_row = vld1q_u8(pu1_src_top_cpy || pu1_src - src_…
132 LD1 {v3.16b},[x0],#16 //pu1_cur_row = vld1q_u8(pu1_src)
134 LD1 {v30.16b},[x12],#16 //vld1q_u8(pu1_src[(ht - 1) * src_strd])
145 LD1 {v18.16b},[x10] //pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
149 LD1 {v30.16b},[x6] //II pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
222 LD1 {v18.16b},[x10] //pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
267 …LD1 {v1.16b},[x9],#16 //pu1_top_row = vld1q_u8(pu1_src_top_cpy || pu1_src - src_…
268 LD1 {v3.16b},[x0],#16 //pu1_cur_row = vld1q_u8(pu1_src)
[all …]
Dihevc_sao_edge_offset_class1_chroma.s136 LD1 {v6.8b},[x14] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
137 LD1 {v7.8b},[x6] //offset_tbl_u = vld1_s8(pi1_sao_offset_u)
138 LD1 {v1.8b},[x7] //offset_tbl_v = vld1_s8(pi1_sao_offset_v)
152 …LD1 {v28.16b},[x11],#16 //pu1_top_row = vld1q_u8(pu1_src_top_cpy || pu1_src - src_…
154 LD1 {v3.16b},[x0],#16 //pu1_cur_row = vld1q_u8(pu1_src)
157 LD1 {v30.16b},[x12],#16 //vld1q_u8(pu1_src[(ht - 1) * src_strd])
169 LD1 {v18.16b},[x10] //pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
176 LD1 {v30.16b},[x6] //II pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
270 LD1 {v18.16b},[x10] //pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
327 …LD1 {v28.16b},[x11] //pu1_top_row = vld1q_u8(pu1_src_top_cpy || pu1_src - src_…
[all …]
Dihevc_sao_band_offset_chroma.s114 LD1 {v1.8b},[x14],#8 //band_table_u.val[0]
119 LD1 {v2.8b},[x14],#8 //band_table_u.val[1]
123 LD1 {v3.8b},[x14],#8 //band_table_u.val[2]
131 LD1 {v4.8b},[x14],#8 //band_table_u.val[3]
135 LD1 {v0.8b},[x4],#8 //Load pu1_src[(ht - 1) * src_strd + col]
140 LD1 {v30.8b},[x7] //pi1_sao_offset_u load
160 LD1 {v9.8b},[x14],#8 //band_table_v.val[0]
163 LD1 {v10.8b},[x14],#8 //band_table_v.val[1]
219 LD1 {v11.8b},[x14],#8 //band_table_v.val[2]
222 LD1 {v12.8b},[x14],#8 //band_table_v.val[3]
[all …]
Dihevc_sao_edge_offset_class2.s108 LD1 {v0.8b},[x11],#8 //pu1_src[(ht - 1) * src_strd + col]
215 LD1 {v7.8b},[x6] //offset_tbl = vld1_s8(pi1_sao_offset)
222 LD1 {v6.8b},[x11] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
264 …LD1 {v3.16b},[x8] //pu1_top_row = vld1q_u8(pu1_src - src_strd - 1) || vld1q_…
268 LD1 {v5.16b},[x0] //pu1_cur_row = vld1q_u8(pu1_src)
294 LD1 {v16.16b},[x8] //I pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
360 LD1 {v16.16b},[x8] //II pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
364 LD1 {v30.16b},[x11] //III pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
476 LD1 {v16.16b},[x8] //pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
581 …LD1 {v3.16b},[x8] //pu1_top_row = vld1q_u8(pu1_src - src_strd - 1) || vld1q_…
[all …]
Dihevc_sao_edge_offset_class2_chroma.s122 LD1 {v0.8b},[x11],#8 //pu1_src[(ht - 1) * src_strd + col]
323 LD1 {v6.8b},[x6] //offset_tbl_u = vld1_s8(pi1_sao_offset_u)
332 LD1 {v7.8b},[x6] //offset_tbl_v = vld1_s8(pi1_sao_offset_v)
365 LD1 {v5.16b},[x0] //pu1_cur_row = vld1q_u8(pu1_src)
381 …LD1 {v3.16b},[x8] //pu1_top_row = vld1q_u8(pu1_src - src_strd - 2) || vld1q_…
406 LD1 {v16.16b},[x8] //I pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
454 LD1 {v30.8b},[x2] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
502 LD1 {v16.16b},[x8] //II pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
505 LD1 {v30.16b},[x11] //III pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
556 LD1 {v22.8b},[x2] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
[all …]
Dihevc_sao_edge_offset_class3_chroma.s119 LD1 {v0.8b},[x11],#8 //pu1_src[(ht - 1) * src_strd + col]
318 LD1 {v6.8b},[x6] //offset_tbl_u = vld1_s8(pi1_sao_offset_u)
320 LD1 {v7.8b},[x6] //offset_tbl_v = vld1_s8(pi1_sao_offset_v)
356 LD1 {v5.16b},[x0] //pu1_cur_row = vld1q_u8(pu1_src)
367 LD1 {v3.16b},[x8] //pu1_top_row = vld1q_u8(pu1_src - src_strd + 2)
398 LD1 {v16.16b},[x11] //I pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
440 LD1 {v28.8b},[x2] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
492 LD1 {v16.16b},[x11] //II pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
502 LD1 {v30.16b},[x4] //III pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
517 LD1 {v21.8b},[x2] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
[all …]
Dihevc_sao_edge_offset_class3.s108 LD1 {v0.8b},[x11],#8 //pu1_src[(ht - 1) * src_strd + col]
227 LD1 {v7.8b},[x6] //offset_tbl = vld1_s8(pi1_sao_offset)
239 LD1 {v6.8b},[x6] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
278 LD1 {v3.16b},[x8] //pu1_top_row = vld1q_u8(pu1_src - src_strd + 1)
282 LD1 {v5.16b},[x0] //pu1_cur_row = vld1q_u8(pu1_src)
304 LD1 {v16.16b},[x8] //I pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
374 LD1 {v16.16b},[x8] //II pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
378 LD1 {v30.16b},[x2] //III pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
500 LD1 {v16.16b},[x8] //pu1_next_row = vld1q_u8(pu1_src_cpy + src_strd)
606 LD1 {v3.16b},[x8] //pu1_top_row = vld1q_u8(pu1_src - src_strd + 1)
[all …]
Dihevc_sao_edge_offset_class0.s100 LD1 {v5.8b},[x14] //edge_idx_tbl = vld1_s8(gi1_table_edge_idx)
104 LD1 {v7.8b},[x8] //offset_tbl = vld1_s8(pi1_sao_offset)
110 LD1 {v0.8b},[x4],#8 //Load pu1_src[(ht - 1) * src_strd + col]
145 LD1 {v17.16b},[x12],x1 //pu1_cur_row = vld1q_u8(pu1_src_cpy)
152 LD1 {v26.16b},[x12] //II Iteration pu1_cur_row = vld1q_u8(pu1_src_cpy)
290 LD1 {v17.16b},[x12] //pu1_cur_row = vld1q_u8(pu1_src_cpy)
/external/libhevc/decoder/arm64/
Dihevcd_fmt_conv_420sp_to_420sp.s114 LD1 {v0.8b},[x0],#8
115 LD1 {v1.8b},[x0],#8
116 LD1 {v2.8b},[x0],#8
117 LD1 {v3.8b},[x0],#8
134 LD1 {v0.8b},[x0],#8
135 LD1 {v1.8b},[x0],#8
136 LD1 {v2.8b},[x0],#8
137 LD1 {v3.8b},[x0],#8
172 LD1 {v0.8b},[x1],#8
173 LD1 {v1.8b},[x1],#8
[all …]
/external/llvm/test/CodeGen/ARM/
Dvcombine.ll7 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
10 ; CHECK-LE-DAG: vmov r2, r3, [[LD1]]
23 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
26 ; CHECK-LE-DAG: vmov r2, r3, [[LD1]]
40 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
43 ; CHECK-LE: vmov r2, r3, [[LD1]]
57 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
60 ; CHECK-LE: vmov r2, r3, [[LD1]]
73 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
76 ; CHECK-LE: vmov r2, r3, [[LD1]]
[all …]
/external/llvm/docs/
DBigEndianNEON.rst52 ``LDR`` and ``LD1``
66 Big endian vector load using ``LD1``. Note that the lanes retain the correct ordering.
69 Because of this, the instruction ``LD1`` performs a vector load but performs byte swapping not on t…
71 It may seem that ``LD1`` should suffice to peform vector loads on a big endian machine. However the…
76 …2. The content of a vector register is the same *as if* it had been loaded with an ``LD1`` instruc…
78 Because ``LD1 == LDR + REV`` and similarly ``LDR == LD1 + REV`` (on a big endian system), we can si…
109 …t one advantage over ``LD1`` and ``ST1``. ``LDR`` and ``STR`` are oblivious to the size of the ind…
124 …``uint16x4_t``, which is equivalent in register content, if we passed as ``LD1`` we'd break this c…
133 …LD1`` only requires it to be as aligned as the lane size. If we canonicalised on using ``LDR``, we…
143 | | ``LDR`` layout | ``LD1`` layout |
[all …]
/external/llvm-project/llvm/docs/
DBigEndianNEON.rst52 ``LDR`` and ``LD1``
66 Big endian vector load using ``LD1``. Note that the lanes retain the correct ordering.
69 Because of this, the instruction ``LD1`` performs a vector load but performs byte swapping not on t…
71 It may seem that ``LD1`` should suffice to peform vector loads on a big endian machine. However the…
76 …2. The content of a vector register is the same *as if* it had been loaded with an ``LD1`` instruc…
78 Because ``LD1 == LDR + REV`` and similarly ``LDR == LD1 + REV`` (on a big endian system), we can si…
109 …t one advantage over ``LD1`` and ``ST1``. ``LDR`` and ``STR`` are oblivious to the size of the ind…
124 …``uint16x4_t``, which is equivalent in register content, if we passed as ``LD1`` we'd break this c…
133 …LD1`` only requires it to be as aligned as the lane size. If we canonicalised on using ``LDR``, we…
143 | | ``LDR`` layout | ``LD1`` layout |
[all …]
/external/llvm-project/llvm/test/CodeGen/ARM/
Dvcombine.ll7 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
10 ; CHECK-LE-DAG: vmov r2, r3, [[LD1]]
23 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
26 ; CHECK-LE-DAG: vmov r2, r3, [[LD1]]
40 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
42 ; CHECK-LE: vmov r2, r3, [[LD1]]
57 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
59 ; CHECK-LE: vmov r2, r3, [[LD1]]
73 ; CHECK-DAG: vldr [[LD1:d[0-9]+]], [r1]
75 ; CHECK-LE: vmov r2, r3, [[LD1]]
[all …]
/external/llvm-project/llvm/test/CodeGen/NVPTX/
Dproxy-reg-erasure-ptx.ll79 ; PTX-DAG: ld.param.v2.b64 {[[LD0:%rd[0-9]+]], [[LD1:%rd[0-9]+]]}, [retval0+0];
83 ; PTX-WITHOUT-DAG: mov.b64 [[PROXY1:%rd[0-9]+]], [[LD1]];
85 ; PTX-WITH-DAG: st.param.v2.b64 [func_retval0+0], {[[LD0]], [[LD1]]};
140 …; PTX-DAG: ld.param.v4.b32 {[[LD0:%r[0-9]+]], [[LD1:%r[0-9]+]], [[LD2:%r[0-9]+]], [[LD3:%r[0-9]+]]…
144 ; PTX-WITHOUT-DAG: mov.b32 [[PROXY1:%r[0-9]+]], [[LD1]];
148 ; PTX-WITH-DAG: st.param.v4.b32 [func_retval0+0], {[[LD0]], [[LD1]], [[LD2]], [[LD3]]};
173 ; PTX-DAG: ld.param.v2.f64 {[[LD0:%fd[0-9]+]], [[LD1:%fd[0-9]+]]}, [retval0+0];
177 ; PTX-WITHOUT-DAG: mov.f64 [[PROXY1:%fd[0-9]+]], [[LD1]];
179 ; PTX-WITH-DAG: st.param.v2.f64 [func_retval0+0], {[[LD0]], [[LD1]]};
/external/llvm/test/CodeGen/X86/
Dmerge-store-partially-alias-loads.ll22 ; DBGDAG-DAG: [[LD1:t[0-9]+]]: i8,ch = load<LD1[%tmp12]> [[ENTRYTOKEN]], [[ADDPTR]], undef:i64
24 ; DBGDAG: [[LOADTOKEN:t[0-9]+]]: ch = TokenFactor [[LD2]]:1, [[LD1]]:1
27 ; DBGDAG-DAG: [[ST1:t[0-9]+]]: ch = store<ST1[%tmp14]> [[ST2]], [[LD1]], t{{[0-9]+}}, undef:i64
/external/llvm-project/llvm/test/Transforms/InstCombine/
Dmultiple-uses-load-bitcast-select.ll8 ; CHECK-NEXT: [[LD1:%.*]] = load double, double* [[Y1]], align 8
10 ; CHECK-NEXT: [[TMP10:%.*]] = fcmp olt double [[LD1]], [[LD2]]
11 ; CHECK-NEXT: [[TMP121:%.*]] = select i1 [[TMP10]], double [[LD1]], double [[LD2]]
/external/llvm-project/llvm/test/CodeGen/X86/
Dmerge-store-partially-alias-loads.ll22 ; DBGDAG-DAG: [[LD1:t[0-9]+]]: i8,ch = load<(load 1 from %ir.tmp12)> [[ENTRYTOKEN]], [[ADDPTR]], un…
24 ; DBGDAG-DAG: [[ST1:t[0-9]+]]: ch = store<(store 1 into %ir.tmp14)> [[ENTRYTOKEN]], [[LD1]], t{{[0-…
25 ; DBGDAG-DAG: [[LOADTOKEN:t[0-9]+]]: ch = TokenFactor [[LD2]]:1, [[LD1]]:1

12345