add_noise_sse2.asm |
|
2427 |
avg_intrin_avx2.c |
|
19203 |
avg_intrin_sse2.c |
|
21670 |
avg_pred_avx2.c |
|
4393 |
avg_pred_sse2.c |
comp_pred and pred must be 16 byte aligned. |
2504 |
avg_ssse3_x86_64.asm |
|
3369 |
bitdepth_conversion_avx2.h |
|
1599 |
bitdepth_conversion_sse2.asm |
|
2679 |
bitdepth_conversion_sse2.h |
|
1938 |
convolve.h |
|
20310 |
convolve_avx2.h |
|
7085 |
convolve_sse2.h |
|
4001 |
convolve_ssse3.h |
|
4715 |
deblock_sse2.asm |
|
12487 |
fwd_dct32x32_impl_avx2.h |
|
159250 |
fwd_dct32x32_impl_sse2.h |
|
163689 |
fwd_txfm_avx2.c |
|
16518 |
fwd_txfm_impl_sse2.h |
|
45786 |
fwd_txfm_sse2.c |
|
8832 |
fwd_txfm_sse2.h |
|
16233 |
fwd_txfm_ssse3_x86_64.asm |
|
7216 |
highbd_convolve_avx2.c |
|
55462 |
highbd_idct4x4_add_sse2.c |
|
6863 |
highbd_idct4x4_add_sse4.c |
|
1696 |
highbd_idct8x8_add_sse2.c |
|
7415 |
highbd_idct8x8_add_sse4.c |
|
7286 |
highbd_idct16x16_add_sse2.c |
io[16] |
11914 |
highbd_idct16x16_add_sse4.c |
io[16] |
11580 |
highbd_idct32x32_add_sse2.c |
step1[16] |
28798 |
highbd_idct32x32_add_sse4.c |
step1[16] |
27883 |
highbd_intrapred_intrin_sse2.c |
|
21317 |
highbd_intrapred_intrin_ssse3.c |
; ------------------------------------------
; input: x, y, z, result
;
; trick from pascal
; (x+2y+z+2)>>2 can be calculated as:
; result = avg(x,z)
; result -= xor(x,z) & 1
; result = avg(result,y)
; ------------------------------------------
|
42470 |
highbd_intrapred_sse2.asm |
|
14633 |
highbd_inv_txfm_sse2.h |
out[2] |
15995 |
highbd_inv_txfm_sse4.h |
in[2] |
4364 |
highbd_loopfilter_sse2.c |
|
44958 |
highbd_quantize_intrin_avx2.c |
|
9647 |
highbd_quantize_intrin_sse2.c |
|
5851 |
highbd_sad_avx2.c |
sums_16 will outrange after 2 rows, so add current sums_16 to
sums_32 |
24178 |
highbd_sad_sse2.asm |
|
13813 |
highbd_sad4d_avx2.c |
[4] |
19393 |
highbd_sad4d_sse2.asm |
|
10177 |
highbd_subpel_variance_impl_sse2.asm |
|
31896 |
highbd_variance_impl_sse2.asm |
|
9911 |
highbd_variance_sse2.c |
|
35729 |
intrapred_sse2.asm |
|
26349 |
intrapred_ssse3.asm |
|
30605 |
inv_txfm_avx2.c |
|
22859 |
inv_txfm_sse2.c |
|
44720 |
inv_txfm_sse2.h |
|
28934 |
inv_txfm_ssse3.c |
in[32] |
12644 |
inv_txfm_ssse3.h |
io[8] |
4636 |
inv_wht_sse2.asm |
|
2789 |
loopfilter_avx2.c |
AVX2 |
36427 |
loopfilter_intrin_sse2.c |
(abs(q1 - q0), abs(p1 - p0) |
77419 |
mem_sse2.h |
|
6008 |
post_proc_sse2.c |
|
4789 |
quantize_avx.c |
|
9088 |
quantize_avx2.c |
|
12403 |
quantize_sse2.c |
|
3873 |
quantize_sse2.h |
|
5202 |
quantize_ssse3.c |
|
8185 |
quantize_ssse3.h |
|
1898 |
sad_avx2.c |
|
10353 |
sad_sse2.asm |
|
10149 |
sad4d_avx2.c |
[4] |
7584 |
sad4d_avx512.c |
|
3209 |
sad4d_sse2.asm |
|
8420 |
sse_avx2.c |
|
13904 |
sse_sse4.c |
|
10515 |
ssim_opt_x86_64.asm |
|
6226 |
subpel_variance_sse2.asm |
|
43556 |
subtract_avx2.c |
|
8518 |
subtract_sse2.asm |
|
3780 |
sum_squares_sse2.c |
|
3902 |
transpose_sse2.h |
|
12981 |
txfm_common_sse2.h |
|
1355 |
variance_avx2.c |
clang-format off |
39581 |
variance_sse2.c |
|
23953 |
vpx_convolve_copy_sse2.asm |
|
6649 |
vpx_high_subpixel_8t_sse2.asm |
|
24926 |
vpx_high_subpixel_bilinear_sse2.asm |
|
11795 |
vpx_subpixel_4t_intrin_sse2.c |
The macro used to generate functions shifts the src_ptr up by 3 rows already
|
50509 |
vpx_subpixel_8t_intrin_avx2.c |
|
56170 |
vpx_subpixel_8t_intrin_ssse3.c |
|
42998 |
vpx_subpixel_8t_sse2.asm |
|
23772 |
vpx_subpixel_8t_ssse3.asm |
|
28701 |
vpx_subpixel_bilinear_sse2.asm |
|
10066 |
vpx_subpixel_bilinear_ssse3.asm |
|
9245 |