| /external/libvpx/libvpx/vpx_dsp/arm/ |
| idct4x4_1_add_neon.c | 22 uint16x8_t q8u16; local
40 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q0s16),
42 d6u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
|
| vpx_convolve8_avg_neon.c | 67 uint16x8_t q1u16, q2u16, q8u16, q9u16, q10u16, q11u16, q12u16, q13u16; local
105 q8u16 = vmovl_u8(d0x2u8.val[0]);
111 d16u16 = vget_low_u16(q8u16);
112 d17u16 = vget_high_u16(q8u16);
115 q8u16 = vcombine_u16(d16u16, d18u16); // vswp 17 18
166 d16s16 = vreinterpret_s16_u16(vget_low_u16(q8u16));
167 d17s16 = vreinterpret_s16_u16(vget_high_u16(q8u16));
222 q8u16 = q9u16;
256 uint16x8_t q1u16, q2u16, q8u16, q9u16, q10u16, q11u16, q12u16, q13u16; local
280 q8u16 = vmovl_u8(vreinterpret_u8_u32(d16u32))
[all...] |
| vpx_convolve8_neon.c | 67 uint16x8_t q1u16, q2u16, q8u16, q9u16, q10u16, q11u16, q12u16, q13u16; local
108 q8u16 = vmovl_u8(d0x2u8.val[0]);
113 d16u16 = vget_low_u16(q8u16);
114 d17u16 = vget_high_u16(q8u16);
117 q8u16 = vcombine_u16(d16u16, d18u16); // vswp 17 18
159 d16s16 = vreinterpret_s16_u16(vget_low_u16(q8u16));
160 d17s16 = vreinterpret_s16_u16(vget_high_u16(q8u16));
210 q8u16 = q9u16;
240 uint16x8_t q1u16, q2u16, q8u16, q9u16, q10u16, q11u16, q12u16, q13u16; local
264 q8u16 = vmovl_u8(vreinterpret_u8_u32(d16u32))
[all...] |
| idct4x4_add_neon.c | 19 uint16x8_t q8u16, q9u16; local
134 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q8s16),
139 d26u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
|
| idct8x8_add_neon.c | 249 uint16x8_t q8u16, q9u16, q10u16, q11u16; local
292 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q8s16),
301 d0u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
329 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q8s16),
338 d0u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
365 uint16x8_t q8u16, q9u16, q10u16, q11u16; local
480 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q8s16),
489 d0u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
517 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q8s16),
526 d0u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
[all...] |
| intrapred_neon.c | 663 uint16x8_t q0u16, q2u16, q3u16, q8u16, q10u16; local
679 q8u16 = vdupq_lane_u16(d20u16, 1);
684 q11s16 = vaddq_s16(vreinterpretq_s16_u16(q8u16),
686 q8s16 = vaddq_s16(vreinterpretq_s16_u16(q8u16),
700 q8u16 = vdupq_lane_u16(d20u16, 3);
705 q11s16 = vaddq_s16(vreinterpretq_s16_u16(q8u16),
707 q8s16 = vaddq_s16(vreinterpretq_s16_u16(q8u16),
726 uint16x8_t q0u16, q3u16, q8u16, q9u16, q10u16, q11u16; local
735 q8u16 = vsubl_u8(vget_low_u8(q1u8), vget_low_u8(q0u8));
746 vreinterpretq_s16_u16(q8u16));
[all...] |
| idct16x16_add_neon.c | 338 uint16x8_t q2u16, q3u16, q4u16, q5u16, q8u16; local
696 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q8s16),
698 d12u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
[all...] |
| /external/libvpx/libvpx/vp8/common/arm/neon/ |
| bilinearpredict_neon.c | 35 uint16x8_t q7u16, q8u16, q9u16; local
84 q8u16 = vmull_u8(vreinterpret_u8_u32(d1u32x2.val[0]), d0u8);
88 q8u16 = vmlal_u8(q8u16, vreinterpret_u8_u32(d3u32x2.val[0]), d1u8);
92 d29u8 = vqrshrn_n_u16(q8u16, 7);
143 uint16x8_t q6u16, q7u16, q8u16, q9u16, q10u16; local
163 q8u16 = vmull_u8(vget_low_u8(q3u8), d0u8);
175 q8u16 = vmlal_u8(q8u16, d7u8, d1u8);
181 d24u8 = vqrshrn_n_u16(q8u16, 7)
230 uint16x8_t q6u16, q7u16, q8u16, q9u16, q10u16; local
371 uint16x8_t q1u16, q2u16, q3u16, q4u16, q5u16, q6u16, q7u16, q8u16; local
[all...] |
| sixtappredict_neon.c | 37 uint16x8_t q8u16, q9u16, q10u16, q11u16, q12u16; local
178 q8u16 = vmull_u8(vreinterpret_u8_u32(d1u32x2.val[0]), d5u8);
191 q8u16 = vmlal_u8(q8u16, vreinterpret_u8_u32(d1u32x2.val[0]), d0u8);
200 q8u16 = vmlsl_u8(q8u16, vreinterpret_u8_u32(d1u32x2.val[0]), d1u8);
209 q8u16 = vmlsl_u8(q8u16, vreinterpret_u8_u32(d1u32x2.val[0]), d4u8);
218 q8u16 = vmlal_u8(q8u16, vreinterpret_u8_u32(d1u32x2.val[0]), d2u8)
415 uint16x8_t q8u16, q9u16, q10u16, q11u16, q12u16; local
818 uint16x8_t q8u16, q9u16, q10u16, q11u16, q12u16; local
1293 uint16x8_t q3u16, q4u16, q5u16, q6u16, q7u16, q8u16, q9u16, q10u16; local
[all...] |
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
| bilinearpredict_neon.c | 36 uint16x8_t q7u16, q8u16, q9u16; local
81 q8u16 = vmull_u8(vreinterpret_u8_u32(d1u32x2.val[0]), d0u8);
85 q8u16 = vmlal_u8(q8u16, vreinterpret_u8_u32(d3u32x2.val[0]), d1u8);
89 d29u8 = vqrshrn_n_u16(q8u16, 7);
140 uint16x8_t q6u16, q7u16, q8u16, q9u16, q10u16; local
160 q8u16 = vmull_u8(vget_low_u8(q3u8), d0u8);
172 q8u16 = vmlal_u8(q8u16, d7u8, d1u8);
178 d24u8 = vqrshrn_n_u16(q8u16, 7)
227 uint16x8_t q6u16, q7u16, q8u16, q9u16, q10u16; local
368 uint16x8_t q1u16, q2u16, q3u16, q4u16, q5u16, q6u16, q7u16, q8u16; local
[all...] |
| /external/libvpx/libvpx/vp9/common/arm/neon/ |
| vp9_iht4x4_add_neon.c | 163 uint16x8_t q8u16, q9u16; local
234 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q8s16), vreinterpret_u8_u32(d26u32));
237 d26u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
|
| vp9_iht8x8_add_neon.c | 496 uint16x8_t q8u16, q9u16, q10u16, q11u16; local
600 q8u16 = vaddw_u8(vreinterpretq_u16_s16(q8s16),
609 d0u8 = vqmovun_s16(vreinterpretq_s16_u16(q8u16));
|
