Commits · v1.7.0-linphone · BC / public / external / libvpx

26 Oct, 2017 1 commit

vpx: hadamard: use ptrdiff_t instead of int for stride · 3bf02ad7

Scott LaVarnway authored Oct 26, 2017

Eliminates the following instruction for the x86 (64 bit)
intrinsic code:

movslq %esi,%rax

Change-Id: I8f5ebd40726f998708a668b0f52ea7a0576befae

3bf02ad7

30 Sep, 2017 1 commit
- ppc: Add vpx_idct32x32_1024_add_vsx · fb7fc1db
  Alexandra Hájková authored Jul 31, 2017
```
Change-Id: I55cd0a1569ccc47a53d0ecf751aac259d510e10d
```
  fb7fc1db
05 Sep, 2017 1 commit

Remove get_filter_base() and get_filter_offset() in convolve · d331e7a1

Linfeng Zhang authored Aug 28, 2017

so that the convolve functions are independent of table alignment.

Change-Id: Ieab132a30d72c6e75bbe9473544fbe2cf51541ee

d331e7a1

31 Aug, 2017 1 commit
- inv_txfm_vsx: fix loads in high-bitdepth · f8f64c30
  James Zern authored Aug 30, 2017
```
vec_vsx_ld -> load_tran_low

Change-Id: Id3144cdd528d2d406a515e5812e2ea9e4db64bf1
```
  f8f64c30
25 Jul, 2017 1 commit
- ppc: Add vpx_idct16x16_256_add_vsx · 666c543f
  Alexandra Hájková authored Jun 13, 2017
```
Change-Id: Ibc3f7965423fd91179f8d8e77c7ae3e6d7f80572
```
  666c543f
01 Jul, 2017 2 commits
- ppc: Add vpx_idct8x8_64_add_vsx · c757d6dd
  Alexandra Hájková authored Jun 12, 2017
```
Change-Id: I4ed1312f365509e0595dcc09890ecb050f6f2069
```
  c757d6dd
- ppc: Add vpx_idct4x4_16_add_vsx · d8c27703
  Alexandra Hájková authored Jun 10, 2017
```
Change-Id: Id2673eece32027fb245919c7a5c81994a4a19fd8
```
  d8c27703
24 May, 2017 1 commit
- ppc: Add vpx_sadnxmx4d_vsx for n,m = {8, 16, 32 ,64} · 8bf6eaf4
  Alexandra Hájková authored May 24, 2017
```
Change-Id: I547d0099e15591655eae954e3ce65fdf3b003123
```
  8bf6eaf4
13 May, 2017 1 commit
- ppc: Add vpx_sad64/32/16x64/32/16_avg_vsx · bcbc3929
  Alexandra Hájková authored May 13, 2017
```
Change-Id: Ic9639b1331d8c5cbc207c2a036891ff0137fc56f
```
  bcbc3929
12 May, 2017 5 commits
- ppc: Add get_mb_ss_vsx · 143b21e3
  Luca Barbato authored May 11, 2017
```
Change-Id: I1b54a7a5bb642e4b836d786ea1ae506eed025e3f
```
  143b21e3
- ppc: Add get4x4sse_cs_vsx · 6d225eb5
  Luca Barbato authored May 11, 2017
```
Change-Id: I3028bdadf653665d18e781d28e9625f62804b3d8
```
  6d225eb5
- ppc: Add comp_avg_pred_vsx · a7f8bd45
  Luca Barbato authored May 11, 2017
```
Change-Id: I59788cd98231e707239c2ad95ae54f67cfe24e10
```
  a7f8bd45
- ppc: Add vpx_sad64x32/64_vsx · f48532e2
  Alexandra Hájková authored May 08, 2017
```
Change-Id: I84e3705fa52f75cb91b2bab4abf5cc77585ee3e2
```
  f48532e2
- ppc Add vpx_sad32x16/32/64_vsx · 0b15bf1e
  Alexandra Hájková authored May 08, 2017
```
Change-Id: I3c4f9d595275669580413a71b3c3c810e7ddcacd
```
  0b15bf1e
10 May, 2017 1 commit
- ppc: Add vpx_sad16x8/16/32_vsx · cc7f0c0f
  Alexandra Hájková authored May 06, 2017
```
Change-Id: I60619d28fffd9809f93b1af510a50e1aa02519a9
```
  cc7f0c0f
03 May, 2017 5 commits

ppc: Add convolve8_vsx and convolve8_avg_vsx · e2ad8909
Luca Barbato authored May 01, 2017
```
Change-Id: Ia5293d948003a7fff5a7cbad6e83d8a72717c857
```
e2ad8909

ppc: Add convolve8_avg_vert_vsx · e6ca81ee

Luca Barbato authored May 01, 2017

Only the generic one again, speedups for 8x8 and larger blocks to
come later.

Change-Id: I90d481d3a602d1e277ead8f3934eca126b86b72d

e6ca81ee

ppc: Add convolve8_vert · a65f1771

Luca Barbato authored May 01, 2017

Only the generic one again, speedups for 8x8 and larger blocks
to come later.

Change-Id: Ia509d6225984b4930ec03928c9bcbf51486da99f

a65f1771

ppc: Add convolve8_horiz_avg · 77772350

Luca Barbato authored Apr 29, 2017

The 8x8 and larger blocks cases can be sped up further.

Change-Id: I54549b03ac6c7a4e3f485738b100c3cac7ac2e15

77772350

ppc: Add convolve8_horiz · 08edb85b

Luca Barbato authored Apr 29, 2017

The 8x8 and larger blocks cases can be sped up further.

Change-Id: I89b635d6b01c59f523f2d54b1284ed32916c5046

08edb85b

29 Apr, 2017 2 commits
- ppc: Add convolve_avg · d51d3934
  Luca Barbato authored Apr 26, 2017
```
Change-Id: Ib203c444c708f42072e38301ee3db97b5b53d014
```
  d51d3934
- ppc: Add convolve_copy · 63860ba7
  Luca Barbato authored Apr 26, 2017
```
Change-Id: Ie26d6dbe090e711d84bac01ba7da270db983f405
```
  63860ba7
20 Apr, 2017 16 commits

ppc: h predictor 8x8 · 914b160f

Luca Barbato authored Apr 18, 2017

Slightly faster with the current compiler.

Change-Id: Iae225fac08395eb430c97a2abec69c60f5cf5c47

914b160f

ppc: d63 predictor 8x8 · 0b9be932
Luca Barbato authored Apr 12, 2017
```
10x faster.

Change-Id: I7cedbf4df2ce7df5b6f1108b11815d088fdb9ba8
```
0b9be932
ppc: tm predictor 4x4 · ee9325b0
Luca Barbato authored Apr 09, 2017
```
Slightly faster.

Change-Id: I0ca43f309b3d9b50435d69bd5be64b53a99bd191
```
ee9325b0
ppc: h predictor 4x4 · 2904eb58
Luca Barbato authored Apr 09, 2017
```
2x faster.

Change-Id: I0583dec353299c6797401b646099f18db4e0420d
```
2904eb58

ppc: dc predictor 8x8 · 58245d70

Luca Barbato authored Apr 09, 2017

Slightly faster, the other dc predictors cannot be faster since
the computation speedup is overwhelmed by the time spent reading
dst to write just the 8x8 part.

Change-Id: I94a0b50500adf8b7b6bb919dbf5c7adf5b9fba66

58245d70

ppc: d45 predictor 8x8 · 6b4a65e8
Luca Barbato authored Apr 09, 2017
```
11x faster.

Change-Id: I5b8f39213ee1f5260724fc254e3fb5c462435798
```
6b4a65e8
ppc: d63 predictor 32x32 · 92e33c7b
Luca Barbato authored Apr 09, 2017
```
About 10x faster.

Change-Id: If7d0645f75c5d7deb9751edd0bf47e2f9068e9e7
```
92e33c7b
ppc: d63 predictor 16x16 · a5469a00
Luca Barbato authored Apr 09, 2017
```
About 18x faster.

Change-Id: Id043bf76c011e03e992085bb5e20f330d3e98cd4
```
a5469a00
ppc: d45 predictor 32x32 · cc868da5
Luca Barbato authored Apr 08, 2017
```
About 12x faster.

Change-Id: I22c150256aefb4941861ab1f6c17d554fb694bed
```
cc868da5
ppc: d45 predictor 16x16 · 7a7dc9e6
Luca Barbato authored Apr 08, 2017
```
About 16x faster.

Change-Id: Ie5469fb32d5fd11bb6cb06318cea475d8a5b00b9
```
7a7dc9e6
ppc: dc predictor 32x32 · c08baa29
Luca Barbato authored Apr 08, 2017
```
10x and 5x faster.

Change-Id: I7913c58c768334d818f541a5e219f1035791eeaf
```
c08baa29
ppc: dc top and left predictor 32x32 · 22ca468c
Luca Barbato authored Apr 08, 2017
```
6x faster.

Change-Id: I717995b4056e5579c68191d11b495372971fe1ae
```
22ca468c
ppc: dc top and left predictor 16x16 · ad9dea1f
Luca Barbato authored Apr 08, 2017
```
13x faster.

Change-Id: I1771ac39fda599153f933cb3f0506c9f97a6cbe6
```
ad9dea1f
ppc: dc_128 predictor 32x32 · d68d3787
Luca Barbato authored Apr 08, 2017
```
6x faster.

Change-Id: I1da8f51b4262871cb98f0aa03ccda41b0ac2b08b
```
d68d3787
ppc: dc_128 predictor 16x16 · f9d20e6d
Luca Barbato authored Apr 08, 2017
```
20x faster.

Change-Id: I05f0deb2d38ae7966eae6b71fbc0aa51880e5709
```
f9d20e6d
ppc: tm predictor 32x32 · 0d9417de
Luca Barbato authored Apr 07, 2017
```
About 8x faster.

Change-Id: I9bad827ccbdf47ec95406e961c74ac2ff45f80cf
```
0d9417de

18 Apr, 2017 2 commits
- ppc: tm predictor 16x16 · 479443a5
  Luca Barbato authored Apr 07, 2017
```
About 10x faster.

Change-Id: I1f5a3752d346459df3b45f92963208bf3e520f06
```
  479443a5
- ppc: tm predictor 8x8 · c8f5a55d
  Luca Barbato authored Apr 07, 2017
```
About 5x faster.

Change-Id: I951230517f49c0dca9ac9eac2efa8916a303b85a
```
  c8f5a55d