| /external/libhevc/common/arm/ |
| D | ihevc_inter_pred_chroma_copy_w16out.s | 139 vld1.8 {d0},[r0] @vld1_u8(pu1_src_tmp)
141 vmovl.u8 q0,d0 @vmovl_u8(vld1_u8(pu1_src_tmp)
145 vld1.8 {d22},[r5],r2 @vld1_u8(pu1_src_tmp)
149 vmovl.u8 q11,d22 @vmovl_u8(vld1_u8(pu1_src_tmp)
150 vld1.8 {d24},[r5],r2 @vld1_u8(pu1_src_tmp)
152 vmovl.u8 q12,d24 @vmovl_u8(vld1_u8(pu1_src_tmp)
155 vld1.8 {d26},[r5],r2 @vld1_u8(pu1_src_tmp)
157 vmovl.u8 q13,d26 @vmovl_u8(vld1_u8(pu1_src_tmp)
181 vld1.8 {d0},[r0] @vld1_u8(pu1_src_tmp)
183 vmovl.u8 q0,d0 @vmovl_u8(vld1_u8(pu1_src_tmp)
[all …]
|
| D | ihevc_inter_pred_luma_copy_w16out.s | 107 vld1.8 {d0},[r0] @vld1_u8(pu1_src_tmp)
109 vmovl.u8 q0,d0 @vmovl_u8(vld1_u8(pu1_src_tmp)
113 vld1.8 {d22},[r5],r2 @vld1_u8(pu1_src_tmp)
117 vmovl.u8 q11,d22 @vmovl_u8(vld1_u8(pu1_src_tmp)
118 vld1.8 {d24},[r5],r2 @vld1_u8(pu1_src_tmp)
120 vmovl.u8 q12,d24 @vmovl_u8(vld1_u8(pu1_src_tmp)
123 vld1.8 {d26},[r5],r2 @vld1_u8(pu1_src_tmp)
125 vmovl.u8 q13,d26 @vmovl_u8(vld1_u8(pu1_src_tmp)
154 vld1.8 {d8},[r0]! @vld1_u8(pu1_src_tmp)
155 vld1.8 {d10},[r6],r2 @vld1_u8(pu1_src_tmp)
[all …]
|
| D | ihevc_inter_pred_filters_luma_vert_w16inp.s | 149 vld1.16 {d1},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
150 vld1.16 {d0},[r0]! @src_tmp1 = vld1_u8(pu1_src_tmp)@
152 vld1.16 {d2},[r3],r2 @src_tmp3 = vld1_u8(pu1_src_tmp)@
154 vld1.16 {d3},[r3],r2 @src_tmp4 = vld1_u8(pu1_src_tmp)@
156 vld1.16 {d4},[r3],r2 @src_tmp1 = vld1_u8(pu1_src_tmp)@
158 vld1.16 {d5},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
160 vld1.16 {d6},[r3],r2 @src_tmp3 = vld1_u8(pu1_src_tmp)@
162 vld1.16 {d7},[r3],r2 @src_tmp4 = vld1_u8(pu1_src_tmp)@
167 vld1.16 {d16},[r3],r2 @src_tmp1 = vld1_u8(pu1_src_tmp)@
174 vld1.16 {d17},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
[all …]
|
| D | ihevc_inter_pred_luma_vert_w16inp_w16out.s | 159 vld1.16 {d1},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
160 vld1.16 {d0},[r0]! @src_tmp1 = vld1_u8(pu1_src_tmp)@
162 vld1.16 {d2},[r3],r2 @src_tmp3 = vld1_u8(pu1_src_tmp)@
164 vld1.16 {d3},[r3],r2 @src_tmp4 = vld1_u8(pu1_src_tmp)@
166 vld1.16 {d4},[r3],r2 @src_tmp1 = vld1_u8(pu1_src_tmp)@
168 vld1.16 {d5},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
170 vld1.16 {d6},[r3],r2 @src_tmp3 = vld1_u8(pu1_src_tmp)@
172 vld1.16 {d7},[r3],r2 @src_tmp4 = vld1_u8(pu1_src_tmp)@
177 vld1.16 {d16},[r3],r2 @src_tmp1 = vld1_u8(pu1_src_tmp)@
184 vld1.16 {d17},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
[all …]
|
| D | ihevc_inter_pred_filters_luma_vert.s | 160 vld1.u8 {d1},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
161 vld1.u8 {d0},[r0]! @src_tmp1 = vld1_u8(pu1_src_tmp)@
163 vld1.u8 {d2},[r3],r2 @src_tmp3 = vld1_u8(pu1_src_tmp)@
165 vld1.u8 {d3},[r3],r2 @src_tmp4 = vld1_u8(pu1_src_tmp)@
167 vld1.u8 {d4},[r3],r2 @src_tmp1 = vld1_u8(pu1_src_tmp)@
169 vld1.u8 {d5},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
171 vld1.u8 {d6},[r3],r2 @src_tmp3 = vld1_u8(pu1_src_tmp)@
173 vld1.u8 {d7},[r3],r2 @src_tmp4 = vld1_u8(pu1_src_tmp)@
175 vld1.u8 {d16},[r3],r2 @src_tmp1 = vld1_u8(pu1_src_tmp)@
177 vld1.u8 {d17},[r3],r2 @src_tmp2 = vld1_u8(pu1_src_tmp)@
[all …]
|
| /external/libvpx/libvpx/vp8/common/arm/neon/ |
| D | vp8_loopfilter_neon.c | 186 d6 = vld1_u8(u); in vp8_loop_filter_horizontal_edge_uv_neon()
188 d7 = vld1_u8(v); in vp8_loop_filter_horizontal_edge_uv_neon()
190 d8 = vld1_u8(u); in vp8_loop_filter_horizontal_edge_uv_neon()
192 d9 = vld1_u8(v); in vp8_loop_filter_horizontal_edge_uv_neon()
194 d10 = vld1_u8(u); in vp8_loop_filter_horizontal_edge_uv_neon()
196 d11 = vld1_u8(v); in vp8_loop_filter_horizontal_edge_uv_neon()
198 d12 = vld1_u8(u); in vp8_loop_filter_horizontal_edge_uv_neon()
200 d13 = vld1_u8(v); in vp8_loop_filter_horizontal_edge_uv_neon()
202 d14 = vld1_u8(u); in vp8_loop_filter_horizontal_edge_uv_neon()
204 d15 = vld1_u8(v); in vp8_loop_filter_horizontal_edge_uv_neon()
[all …]
|
| D | bilinearpredict_neon.c | 24 return vreinterpret_u8_u64(vshl_n_u64(vreinterpret_u64_u8(vld1_u8(a)), 32)); in load_and_shift()
38 a1 = vld1_u8(src_ptr); in vp8_bilinear_predict4x4_neon()
42 a3 = vld1_u8(src_ptr); in vp8_bilinear_predict4x4_neon()
44 a4 = vld1_u8(src_ptr); in vp8_bilinear_predict4x4_neon()
58 a0 = vld1_u8(src_ptr); in vp8_bilinear_predict4x4_neon()
60 a1 = vld1_u8(src_ptr); in vp8_bilinear_predict4x4_neon()
62 a2 = vld1_u8(src_ptr); in vp8_bilinear_predict4x4_neon()
64 a3 = vld1_u8(src_ptr); in vp8_bilinear_predict4x4_neon()
66 a4 = vld1_u8(src_ptr); in vp8_bilinear_predict4x4_neon()
132 d22u8 = vld1_u8(src_ptr); in vp8_bilinear_predict8x4_neon()
[all …]
|
| D | mbloopfilter_neon.c | 219 d6 = vld1_u8(u); in vp8_mbloop_filter_horizontal_edge_uv_neon()
221 d7 = vld1_u8(v); in vp8_mbloop_filter_horizontal_edge_uv_neon()
223 d8 = vld1_u8(u); in vp8_mbloop_filter_horizontal_edge_uv_neon()
225 d9 = vld1_u8(v); in vp8_mbloop_filter_horizontal_edge_uv_neon()
227 d10 = vld1_u8(u); in vp8_mbloop_filter_horizontal_edge_uv_neon()
229 d11 = vld1_u8(v); in vp8_mbloop_filter_horizontal_edge_uv_neon()
231 d12 = vld1_u8(u); in vp8_mbloop_filter_horizontal_edge_uv_neon()
233 d13 = vld1_u8(v); in vp8_mbloop_filter_horizontal_edge_uv_neon()
235 d14 = vld1_u8(u); in vp8_mbloop_filter_horizontal_edge_uv_neon()
237 d15 = vld1_u8(v); in vp8_mbloop_filter_horizontal_edge_uv_neon()
[all …]
|
| /external/XNNPACK/src/qu8-dwconv/ |
| D | up8x9-minmax-neon.c | 80 const uint8x8_t vk0 = vld1_u8(w); w = (void*) ((uintptr_t) w + sizeof(uint8x8_t)); in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
81 const uint8x8_t vi0 = vld1_u8(i0); i0 += 8; in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
87 const uint8x8_t vk1 = vld1_u8(w); w = (void*) ((uintptr_t) w + sizeof(uint8x8_t)); in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
88 const uint8x8_t vi1 = vld1_u8(i1); i1 += 8; in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
94 const uint8x8_t vk2 = vld1_u8(w); w = (void*) ((uintptr_t) w + sizeof(uint8x8_t)); in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
95 const uint8x8_t vi2 = vld1_u8(i2); i2 += 8; in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
101 const uint8x8_t vk3 = vld1_u8(w); w = (void*) ((uintptr_t) w + sizeof(uint8x8_t)); in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
102 const uint8x8_t vi3 = vld1_u8(i3); i3 += 8; in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
108 const uint8x8_t vk4 = vld1_u8(w); w = (void*) ((uintptr_t) w + sizeof(uint8x8_t)); in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
109 const uint8x8_t vi4 = vld1_u8(i4); i4 += 8; in xnn_qu8_dwconv_minmax_ukernel_up8x9__neon()
[all …]
|
| /external/XNNPACK/src/qu8-gavgpool/ |
| D | 7p7x-minmax-neon-c8.c | 44 const uint8x8_t vi0 = vld1_u8(i0); i0 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
45 const uint8x8_t vi1 = vld1_u8(i1); i1 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
46 const uint8x8_t vi2 = vld1_u8(i2); i2 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
47 const uint8x8_t vi3 = vld1_u8(i3); i3 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
48 const uint8x8_t vi4 = vld1_u8(i4); i4 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
49 const uint8x8_t vi5 = vld1_u8(i5); i5 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
50 const uint8x8_t vi6 = vld1_u8(i6); i6 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
79 const uint8x8_t vi0 = vld1_u8(i0); i0 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
80 const uint8x8_t vi1 = vld1_u8(i1); i1 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
81 const uint8x8_t vi2 = vld1_u8(i2); i2 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7p7x__neon_c8()
[all …]
|
| D | 7x-minmax-neon-c8.c | 67 const uint8x8_t vi0 = vld1_u8(i0); i0 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
68 const uint8x8_t vi1 = vld1_u8(i1); i1 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
69 const uint8x8_t vi2 = vld1_u8(i2); i2 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
70 const uint8x8_t vi3 = vld1_u8(i3); i3 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
71 const uint8x8_t vi4 = vld1_u8(i4); i4 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
72 const uint8x8_t vi5 = vld1_u8(i5); i5 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
73 const uint8x8_t vi6 = vld1_u8(i6); i6 += 8; in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
137 const uint8x8_t vi0 = vld1_u8(i0); in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
138 const uint8x8_t vi1 = vld1_u8(i1); in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
139 const uint8x8_t vi2 = vld1_u8(i2); in xnn_qu8_gavgpool_minmax_ukernel_7x__neon_c8()
[all …]
|
| /external/XNNPACK/src/qu8-avgpool/ |
| D | 9p8x-minmax-neon-c8.c | 95 const uint8x8_t vi0 = vld1_u8(i0); i0 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
96 const uint8x8_t vi1 = vld1_u8(i1); i1 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
97 const uint8x8_t vi2 = vld1_u8(i2); i2 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
98 const uint8x8_t vi3 = vld1_u8(i3); i3 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
99 const uint8x8_t vi4 = vld1_u8(i4); i4 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
100 const uint8x8_t vi5 = vld1_u8(i5); i5 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
101 const uint8x8_t vi6 = vld1_u8(i6); i6 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
102 const uint8x8_t vi7 = vld1_u8(i7); i7 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
103 const uint8x8_t vi8 = vld1_u8(i8); i8 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
167 const uint8x8_t vi0 = vld1_u8(i0); i0 += 8; in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8()
[all …]
|
| D | 9x-minmax-neon-c8.c | 119 const uint8x8_t vi0 = vld1_u8(i0); i0 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
120 const uint8x8_t vi1 = vld1_u8(i1); i1 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
121 const uint8x8_t vi2 = vld1_u8(i2); i2 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
122 const uint8x8_t vi3 = vld1_u8(i3); i3 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
123 const uint8x8_t vi4 = vld1_u8(i4); i4 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
124 const uint8x8_t vi5 = vld1_u8(i5); i5 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
125 const uint8x8_t vi6 = vld1_u8(i6); i6 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
126 const uint8x8_t vi7 = vld1_u8(i7); i7 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
127 const uint8x8_t vi8 = vld1_u8(i8); i8 += 8; in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
192 const uint8x8_t vi0 = vld1_u8(i0); in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
[all …]
|
| /external/libvpx/libvpx/vpx_dsp/arm/ |
| D | avg_neon.c | 34 b = vld1_u8(a); in vpx_avg_8x8_neon()
36 c = vld1_u8(a); in vpx_avg_8x8_neon()
41 const uint8x8_t d = vld1_u8(a); in vpx_avg_8x8_neon()
191 const uint8x16_t a01 = vcombine_u8(vld1_u8(a), vld1_u8(a + a_stride)); in vpx_minmax_8x8_neon()
193 vcombine_u8(vld1_u8(a + 2 * a_stride), vld1_u8(a + 3 * a_stride)); in vpx_minmax_8x8_neon()
195 vcombine_u8(vld1_u8(a + 4 * a_stride), vld1_u8(a + 5 * a_stride)); in vpx_minmax_8x8_neon()
197 vcombine_u8(vld1_u8(a + 6 * a_stride), vld1_u8(a + 7 * a_stride)); in vpx_minmax_8x8_neon()
199 const uint8x16_t b01 = vcombine_u8(vld1_u8(b), vld1_u8(b + b_stride)); in vpx_minmax_8x8_neon()
201 vcombine_u8(vld1_u8(b + 2 * b_stride), vld1_u8(b + 3 * b_stride)); in vpx_minmax_8x8_neon()
203 vcombine_u8(vld1_u8(b + 4 * b_stride), vld1_u8(b + 5 * b_stride)); in vpx_minmax_8x8_neon()
[all …]
|
| D | vpx_convolve8_neon.h | 22 *s0 = vld1_u8(s); in load_u8_8x4()
24 *s1 = vld1_u8(s); in load_u8_8x4()
26 *s2 = vld1_u8(s); in load_u8_8x4()
28 *s3 = vld1_u8(s); in load_u8_8x4()
36 *s0 = vld1_u8(s); in load_u8_8x8()
38 *s1 = vld1_u8(s); in load_u8_8x8()
40 *s2 = vld1_u8(s); in load_u8_8x8()
42 *s3 = vld1_u8(s); in load_u8_8x8()
44 *s4 = vld1_u8(s); in load_u8_8x8()
46 *s5 = vld1_u8(s); in load_u8_8x8()
[all …]
|
| D | vpx_convolve8_neon.c | 562 d01 = vcombine_u8(vld1_u8(d + 0 * dst_stride), in vpx_convolve8_avg_horiz_neon()
563 vld1_u8(d + 1 * dst_stride)); in vpx_convolve8_avg_horiz_neon()
564 d23 = vcombine_u8(vld1_u8(d + 2 * dst_stride), in vpx_convolve8_avg_horiz_neon()
565 vld1_u8(d + 3 * dst_stride)); in vpx_convolve8_avg_horiz_neon()
566 d45 = vcombine_u8(vld1_u8(d + 4 * dst_stride), in vpx_convolve8_avg_horiz_neon()
567 vld1_u8(d + 5 * dst_stride)); in vpx_convolve8_avg_horiz_neon()
568 d67 = vcombine_u8(vld1_u8(d + 6 * dst_stride), in vpx_convolve8_avg_horiz_neon()
569 vld1_u8(d + 7 * dst_stride)); in vpx_convolve8_avg_horiz_neon()
621 s0 = vreinterpret_s16_u16(vget_low_u16(vmovl_u8(vld1_u8(src)))); in vpx_convolve8_vert_neon()
623 s1 = vreinterpret_s16_u16(vget_low_u16(vmovl_u8(vld1_u8(src)))); in vpx_convolve8_vert_neon()
[all …]
|
| D | vpx_convolve8_avg_vert_filter_type2_neon.asm | 79 vld1.u8 {d1}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
80 vld1.u8 {d0}, [r0]! ;src_tmp1 = vld1_u8(pu1_src_tmp);
82 vld1.u8 {d2}, [r3], r2 ;src_tmp3 = vld1_u8(pu1_src_tmp);
85 vld1.u8 {d3}, [r3], r2 ;src_tmp4 = vld1_u8(pu1_src_tmp);
88 vld1.u8 {d4}, [r3], r2 ;src_tmp1 = vld1_u8(pu1_src_tmp);
91 vld1.u8 {d5}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
94 vld1.u8 {d6}, [r3], r2 ;src_tmp3 = vld1_u8(pu1_src_tmp);
97 vld1.u8 {d7}, [r3], r2 ;src_tmp4 = vld1_u8(pu1_src_tmp);
100 vld1.u8 {d16}, [r3], r2 ;src_tmp1 = vld1_u8(pu1_src_tmp);
103 vld1.u8 {d17}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
[all …]
|
| D | vpx_convolve8_vert_filter_type1_neon.asm | 79 vld1.u8 {d1}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
80 vld1.u8 {d0}, [r0]! ;src_tmp1 = vld1_u8(pu1_src_tmp);
82 vld1.u8 {d2}, [r3], r2 ;src_tmp3 = vld1_u8(pu1_src_tmp);
85 vld1.u8 {d3}, [r3], r2 ;src_tmp4 = vld1_u8(pu1_src_tmp);
88 vld1.u8 {d4}, [r3], r2 ;src_tmp1 = vld1_u8(pu1_src_tmp);
91 vld1.u8 {d5}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
94 vld1.u8 {d6}, [r3], r2 ;src_tmp3 = vld1_u8(pu1_src_tmp);
97 vld1.u8 {d7}, [r3], r2 ;src_tmp4 = vld1_u8(pu1_src_tmp);
100 vld1.u8 {d16}, [r3], r2 ;src_tmp1 = vld1_u8(pu1_src_tmp);
103 vld1.u8 {d17}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
[all …]
|
| D | vpx_convolve8_vert_filter_type2_neon.asm | 79 vld1.u8 {d1}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
80 vld1.u8 {d0}, [r0]! ;src_tmp1 = vld1_u8(pu1_src_tmp);
82 vld1.u8 {d2}, [r3], r2 ;src_tmp3 = vld1_u8(pu1_src_tmp);
85 vld1.u8 {d3}, [r3], r2 ;src_tmp4 = vld1_u8(pu1_src_tmp);
88 vld1.u8 {d4}, [r3], r2 ;src_tmp1 = vld1_u8(pu1_src_tmp);
91 vld1.u8 {d5}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
94 vld1.u8 {d6}, [r3], r2 ;src_tmp3 = vld1_u8(pu1_src_tmp);
97 vld1.u8 {d7}, [r3], r2 ;src_tmp4 = vld1_u8(pu1_src_tmp);
100 vld1.u8 {d16}, [r3], r2 ;src_tmp1 = vld1_u8(pu1_src_tmp);
103 vld1.u8 {d17}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
[all …]
|
| D | vpx_convolve8_avg_vert_filter_type1_neon.asm | 78 vld1.u8 {d1}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
79 vld1.u8 {d0}, [r0]! ;src_tmp1 = vld1_u8(pu1_src_tmp);
81 vld1.u8 {d2}, [r3], r2 ;src_tmp3 = vld1_u8(pu1_src_tmp);
84 vld1.u8 {d3}, [r3], r2 ;src_tmp4 = vld1_u8(pu1_src_tmp);
87 vld1.u8 {d4}, [r3], r2 ;src_tmp1 = vld1_u8(pu1_src_tmp);
90 vld1.u8 {d5}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
93 vld1.u8 {d6}, [r3], r2 ;src_tmp3 = vld1_u8(pu1_src_tmp);
96 vld1.u8 {d7}, [r3], r2 ;src_tmp4 = vld1_u8(pu1_src_tmp);
99 vld1.u8 {d16}, [r3], r2 ;src_tmp1 = vld1_u8(pu1_src_tmp);
102 vld1.u8 {d17}, [r3], r2 ;src_tmp2 = vld1_u8(pu1_src_tmp);
[all …]
|
| /external/XNNPACK/src/x8-zip/ |
| D | x4-neon.c | 28 vxyzw.val[0] = vld1_u8(x); x += 8; in xnn_x8_zip_x4_ukernel__neon()
29 vxyzw.val[1] = vld1_u8(y); y += 8; in xnn_x8_zip_x4_ukernel__neon()
30 vxyzw.val[2] = vld1_u8(z); z += 8; in xnn_x8_zip_x4_ukernel__neon()
31 vxyzw.val[3] = vld1_u8(w); w += 8; in xnn_x8_zip_x4_ukernel__neon()
38 vxyzw.val[0] = vld1_u8(x + address_increment); in xnn_x8_zip_x4_ukernel__neon()
39 vxyzw.val[1] = vld1_u8(y + address_increment); in xnn_x8_zip_x4_ukernel__neon()
40 vxyzw.val[2] = vld1_u8(z + address_increment); in xnn_x8_zip_x4_ukernel__neon()
41 vxyzw.val[3] = vld1_u8(w + address_increment); in xnn_x8_zip_x4_ukernel__neon()
|
| D | x3-neon.c | 27 vxyz.val[0] = vld1_u8(x); x += 8; in xnn_x8_zip_x3_ukernel__neon()
28 vxyz.val[1] = vld1_u8(y); y += 8; in xnn_x8_zip_x3_ukernel__neon()
29 vxyz.val[2] = vld1_u8(z); z += 8; in xnn_x8_zip_x3_ukernel__neon()
36 vxyz.val[0] = vld1_u8(x + address_increment); in xnn_x8_zip_x3_ukernel__neon()
37 vxyz.val[1] = vld1_u8(y + address_increment); in xnn_x8_zip_x3_ukernel__neon()
38 vxyz.val[2] = vld1_u8(z + address_increment); in xnn_x8_zip_x3_ukernel__neon()
|
| /external/libvpx/config/arm-neon/vpx_dsp/arm/ |
| D | vpx_convolve8_avg_vert_filter_type1_neon.asm.S | 84 vld1.u8 {d1}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
85 vld1.u8 {d0}, [r0]! @src_tmp1 = vld1_u8(pu1_src_tmp);
87 vld1.u8 {d2}, [r3], r2 @src_tmp3 = vld1_u8(pu1_src_tmp);
90 vld1.u8 {d3}, [r3], r2 @src_tmp4 = vld1_u8(pu1_src_tmp);
93 vld1.u8 {d4}, [r3], r2 @src_tmp1 = vld1_u8(pu1_src_tmp);
96 vld1.u8 {d5}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
99 vld1.u8 {d6}, [r3], r2 @src_tmp3 = vld1_u8(pu1_src_tmp);
102 vld1.u8 {d7}, [r3], r2 @src_tmp4 = vld1_u8(pu1_src_tmp);
105 vld1.u8 {d16}, [r3], r2 @src_tmp1 = vld1_u8(pu1_src_tmp);
108 vld1.u8 {d17}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
[all …]
|
| D | vpx_convolve8_avg_vert_filter_type2_neon.asm.S | 85 vld1.u8 {d1}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
86 vld1.u8 {d0}, [r0]! @src_tmp1 = vld1_u8(pu1_src_tmp);
88 vld1.u8 {d2}, [r3], r2 @src_tmp3 = vld1_u8(pu1_src_tmp);
91 vld1.u8 {d3}, [r3], r2 @src_tmp4 = vld1_u8(pu1_src_tmp);
94 vld1.u8 {d4}, [r3], r2 @src_tmp1 = vld1_u8(pu1_src_tmp);
97 vld1.u8 {d5}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
100 vld1.u8 {d6}, [r3], r2 @src_tmp3 = vld1_u8(pu1_src_tmp);
103 vld1.u8 {d7}, [r3], r2 @src_tmp4 = vld1_u8(pu1_src_tmp);
106 vld1.u8 {d16}, [r3], r2 @src_tmp1 = vld1_u8(pu1_src_tmp);
109 vld1.u8 {d17}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
[all …]
|
| D | vpx_convolve8_vert_filter_type1_neon.asm.S | 85 vld1.u8 {d1}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
86 vld1.u8 {d0}, [r0]! @src_tmp1 = vld1_u8(pu1_src_tmp);
88 vld1.u8 {d2}, [r3], r2 @src_tmp3 = vld1_u8(pu1_src_tmp);
91 vld1.u8 {d3}, [r3], r2 @src_tmp4 = vld1_u8(pu1_src_tmp);
94 vld1.u8 {d4}, [r3], r2 @src_tmp1 = vld1_u8(pu1_src_tmp);
97 vld1.u8 {d5}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
100 vld1.u8 {d6}, [r3], r2 @src_tmp3 = vld1_u8(pu1_src_tmp);
103 vld1.u8 {d7}, [r3], r2 @src_tmp4 = vld1_u8(pu1_src_tmp);
106 vld1.u8 {d16}, [r3], r2 @src_tmp1 = vld1_u8(pu1_src_tmp);
109 vld1.u8 {d17}, [r3], r2 @src_tmp2 = vld1_u8(pu1_src_tmp);
[all …]
|