/external/libvpx/libvpx/vp8/encoder/arm/neon/ |
vp8_shortwalsh4x4_neon.c | 29 int16x4_t dEmptys16, d0s16, d1s16, d2s16, d3s16, d4s16, d5s16, d6s16, d7s16; local 42 d1s16 = vld1_s16(input); 50 v2tmp3 = vtrn_s32(vreinterpret_s32_s16(d1s16), 72 d1s16 = vadd_s16(d7s16, d6s16); 78 v2tmp2 = vtrn_s32(vreinterpret_s32_s16(d1s16), 118 d1s16 = vshrn_n_s32(q9s32, 3); 122 q0s16 = vcombine_s16(d0s16, d1s16);
|
shortfdct_neon.c | 17 int16x4_t d0s16, d1s16, d2s16, d3s16, d4s16, d5s16, d6s16, d7s16; local 36 d1s16 = vld1_s16(input); 44 v2tmp3 = vtrn_s32(vreinterpret_s32_s16(d1s16), 69 d1s16 = vshrn_n_s32(q9s32, 12); 75 v2tmp3 = vtrn_s32(vreinterpret_s32_s16(d1s16), 106 d1s16 = vshrn_n_s32(q11s32, 16); 107 d1s16 = vsub_s16(d1s16, vreinterpret_s16_u16(d4u16)); 110 q0s16 = vcombine_s16(d0s16, d1s16); 122 int16x4_t d0s16, d1s16, d2s16, d3s16, d4s16, d5s16, d6s16, d7s16 local [all...] |
/external/libvpx/libvpx/vp9/common/arm/neon/ |
vp9_iht4x4_add_neon.c | 47 int16x4_t *d1s16, 50 *d1s16 = vdup_n_s16(cospi_16_64); 69 int16x4_t *d1s16, 88 q13s32 = vmull_s16(d23s16, *d1s16); 89 q14s32 = vmull_s16(d24s16, *d1s16); 160 int16x4_t d0s16, d1s16, d2s16, d3s16, d4s16, d5s16; local 179 GENERATE_COSINE_CONSTANTS(&d0s16, &d1s16, &d2s16); 183 IDCT4x4_1D(&d0s16, &d1s16, &d2s16, &q8s16, &q9s16); 193 GENERATE_COSINE_CONSTANTS(&d0s16, &d1s16, &d2s16); 203 IDCT4x4_1D(&d0s16, &d1s16, &d2s16, &q8s16, &q9s16) [all...] |
vp9_iht8x8_add_neon.c | 112 int16x4_t d0s16, d1s16, d2s16, d3s16; local 121 d1s16 = vdup_n_s16(cospi_4_64); 147 q2s32 = vmlsl_s16(q2s32, d30s16, d1s16); 148 q3s32 = vmlsl_s16(q3s32, d31s16, d1s16); 159 q2s32 = vmull_s16(d18s16, d1s16); 160 q3s32 = vmull_s16(d19s16, d1s16); 189 d1s16 = vdup_n_s16(cospi_8_64); 200 q8s32 = vmull_s16(d20s16, d1s16); 201 q12s32 = vmull_s16(d21s16, d1s16); 203 q2s32 = vmlsl_s16(q2s32, d28s16, d1s16); 268 int16x4_t d0s16, d1s16, d2s16, d3s16, d4s16, d5s16, d6s16, d7s16; local [all...] |
/external/libvpx/libvpx/vpx_dsp/arm/ |
idct8x8_add_neon.c | 94 int16x4_t d0s16, d1s16, d2s16, d3s16; local 103 d1s16 = vdup_n_s16(cospi_4_64); 129 q2s32 = vmlsl_s16(q2s32, d30s16, d1s16); 130 q3s32 = vmlsl_s16(q3s32, d31s16, d1s16); 141 q2s32 = vmull_s16(d18s16, d1s16); 142 q3s32 = vmull_s16(d19s16, d1s16); 171 d1s16 = vdup_n_s16(cospi_8_64); 182 q8s32 = vmull_s16(d20s16, d1s16); 183 q12s32 = vmull_s16(d21s16, d1s16); 185 q2s32 = vmlsl_s16(q2s32, d28s16, d1s16); [all...] |
vpx_convolve8_avg_neon.c | 30 int16x4_t d0s16, d1s16; local 33 d1s16 = vget_high_s16(q0s16); 39 qdst = vmlal_lane_s16(qdst, dsrc4, d1s16, 0); 40 qdst = vmlal_lane_s16(qdst, dsrc5, d1s16, 1); 41 qdst = vmlal_lane_s16(qdst, dsrc6, d1s16, 2); 42 qdst = vmlal_lane_s16(qdst, dsrc7, d1s16, 3);
|
vpx_convolve8_neon.c | 30 int16x4_t d0s16, d1s16; local 33 d1s16 = vget_high_s16(q0s16); 39 qdst = vmlal_lane_s16(qdst, dsrc4, d1s16, 0); 40 qdst = vmlal_lane_s16(qdst, dsrc5, d1s16, 1); 41 qdst = vmlal_lane_s16(qdst, dsrc6, d1s16, 2); 42 qdst = vmlal_lane_s16(qdst, dsrc7, d1s16, 3);
|
idct16x16_add_neon.c | 89 int16x4_t d0s16, d1s16, d2s16, d3s16; local 147 d1s16 = vdup_n_s16(cospi_4_64); 151 q5s32 = vmull_s16(d18s16, d1s16); 152 q6s32 = vmull_s16(d19s16, d1s16); 154 q2s32 = vmlsl_s16(q2s32, d30s16, d1s16); 155 q3s32 = vmlsl_s16(q3s32, d31s16, d1s16); 332 int16x4_t d0s16, d1s16, d2s16, d3s16, d4s16, d5s16, d6s16, d7s16; local 405 d1s16 = vqrshrn_n_s32(q3s32, 14); 408 q0s16 = vcombine_s16(d0s16, d1s16); 1027 int16x4_t d0s16, d1s16, d2s16, d3s16, d4s16, d5s16, d6s16, d7s16; local [all...] |