| add_noise_msa.c |
|
1849 |
| avg_msa.c |
|
28993 |
| common_dspr2.c |
|
836 |
| common_dspr2.h |
prefetch data for store |
1458 |
| convolve_common_dspr2.h |
|
2361 |
| convolve2_avg_dspr2.c |
prefetch data to cache memory |
11790 |
| convolve2_avg_horiz_dspr2.c |
prefetch data to cache memory |
46324 |
| convolve2_dspr2.c |
prefetch data to cache memory |
46323 |
| convolve2_horiz_dspr2.c |
prefetch data to cache memory |
36820 |
| convolve2_vert_dspr2.c |
prefetch data to cache memory |
10111 |
| convolve8_avg_dspr2.c |
prefetch data to cache memory |
34229 |
| convolve8_avg_horiz_dspr2.c |
prefetch data to cache memory |
61992 |
| convolve8_dspr2.c |
prefetch data to cache memory |
74543 |
| convolve8_horiz_dspr2.c |
prefetch data to cache memory |
52466 |
| convolve8_vert_dspr2.c |
prefetch data to cache memory |
19259 |
| deblock_msa.c |
|
34315 |
| fwd_dct32x32_msa.c |
1st and 2nd set |
35778 |
| fwd_txfm_msa.c |
stp 1 |
10259 |
| fwd_txfm_msa.h |
FDCT stage1 |
25335 |
| idct4x4_msa.c |
load vector elements of 4x4 block |
3019 |
| idct8x8_msa.c |
load vector elements of 8x8 block |
4562 |
| idct16x16_msa.c |
stage 2 |
17607 |
| idct32x32_msa.c |
1st & 2nd 8x8 |
27220 |
| intrapred_msa.c |
|
24400 |
| intrapred4_dspr2.c |
|
11884 |
| intrapred8_dspr2.c |
|
34598 |
| intrapred16_dspr2.c |
|
19287 |
| inv_txfm_dspr2.h |
out = dct_const_round_shift(dc * cospi_16_64); |
4048 |
| inv_txfm_msa.h |
|
29458 |
| itrans4_dspr2.c |
temp_1 = (input[0] + input[2]) * cospi_16_64;
step_0 = dct_const_round_shift(temp_1);
temp_2 = (input[0] - input[2]) * cospi_16_64;
step_1 = dct_const_round_shift(temp_2);
|
16605 |
| itrans8_dspr2.c |
temp_1 = (input[0] + input[4]) * cospi_16_64;
step2_0 = dct_const_round_shift(temp_1);
temp_2 = (input[0] - input[4]) * cospi_16_64;
step2_1 = dct_const_round_shift(temp_2);
|
33559 |
| itrans16_dspr2.c |
prefetch row |
65772 |
| itrans32_cols_dspr2.c |
prefetch vpx_ff_cropTbl |
69903 |
| itrans32_dspr2.c |
prefetch row |
70496 |
| loopfilter_4_msa.c |
load vector elements |
5949 |
| loopfilter_8_msa.c |
load vector elements |
13677 |
| loopfilter_16_msa.c |
load vector elements |
51488 |
| loopfilter_filters_dspr2.c |
create quad-byte |
12050 |
| loopfilter_filters_dspr2.h |
inputs & outputs are quad-byte vectors |
37566 |
| loopfilter_macros_dspr2.h |
|
29311 |
| loopfilter_masks_dspr2.h |
processing 4 pixels at the same time
compute hev and mask in the same function |
17070 |
| loopfilter_mb_dspr2.c |
create quad-byte |
23804 |
| loopfilter_mb_horiz_dspr2.c |
create quad-byte |
32094 |
| loopfilter_mb_vert_dspr2.c |
create quad-byte |
32384 |
| loopfilter_msa.h |
|
12317 |
| macros_msa.h |
|
99080 |
| sad_mmi.c |
|
41541 |
| sad_msa.c |
|
27830 |
| sub_pixel_variance_msa.c |
|
66652 |
| subtract_mmi.c |
|
19030 |
| subtract_msa.c |
|
9192 |
| sum_squares_msa.c |
|
5074 |
| txfm_macros_msa.h |
|
6089 |
| variance_mmi.c |
Use VARIANCE_SSE_SUM_8_FOR_W64 in vpx_variance64x64,vpx_variance64x32,
vpx_variance32x64. VARIANCE_SSE_SUM_8 will lead to sum overflow. |
69408 |
| variance_msa.c |
|
20275 |
| vpx_convolve_avg_msa.c |
|
8099 |
| vpx_convolve_copy_msa.c |
|
8224 |
| vpx_convolve_msa.h |
|
7488 |
| vpx_convolve8_avg_horiz_msa.c |
rearranging filter |
28948 |
| vpx_convolve8_avg_msa.c |
rearranging filter |
25100 |
| vpx_convolve8_avg_vert_msa.c |
|
27310 |
| vpx_convolve8_horiz_msa.c |
rearranging filter |
25204 |
| vpx_convolve8_mmi.c |
clang-format off |
32431 |
| vpx_convolve8_msa.c |
8 width cases |
47973 |
| vpx_convolve8_vert_msa.c |
|
27090 |