Frank Barchard
37480f12c6
Add BT.709 Full Range yuv constants.
...
MAKEYUVCONSTANTS macro to generate struct for YUV to RGB
Fix I444AlphaToARGB unit test for ARM by adjusting C version to match Neon implementation.
Bug: libyuv:879, libyuv:878, libyuv:877, libyuv:862, b/178283356
Change-Id: Iedb171fbf668316e7d45ab9e3481de6205ed31e2
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2646472
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2021-01-26 18:36:56 +00:00
Frank Barchard
93b1b332cd
NV12 Bilinear upsampling bug fix
...
Reenable InterpolateRow_AVX2
Bug: libyuv:838, b/68638384, b/176195584
Change-Id: I990fcc204d89ee9b8f5264184558a08aa21d6a9f
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2626067
Reviewed-by: Eugene Zemtsov <eugene@chromium.org>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2021-01-12 23:10:42 +00:00
Frank Barchard
b7a1c5ee5d
Scale by even factor low level row function
...
Bug: b/171884264
Change-Id: I6a94bde0aa05e681bb4590ea8beec33a61ddbfc9
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2518361
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2020-11-03 21:25:18 +00:00
Frank Barchard
a4ec5cf9c2
UVScale down use AVX2 and Neon for aarch32
...
Intel SkylakeX
Was SSSE3 UVScaleDownBy4_Box (2496 ms)
Now AVX2 UVScaleDownBy4_Box (1983 ms)
Was SSSE3 UVScaleDownBy2_Box (380 ms)
Now AVX2 UVScaleDownBy2_Box (360 ms)
Pixel 4 aarch32
Was UVScaleDownBy4_Box (4295 ms)
Now UVScaleDownBy4_Box (3307 ms)
Was UVScaleDownBy2_Box (1022 ms)
Now UVScaleDownBy2_Box (778 ms)
Bug: libuyv:838
Change-Id: Ic823fa15e5761c1b9a897da27341adbf1ed39883
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2470196
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2020-10-14 06:23:26 +00:00
Frank Barchard
725c64015d
UVScale down by 4 use SSSE3/NEON
...
Intel SkylakeX
Was UVScaleDownBy4_Box (7421 ms)
Now UVScaleDownBy4_Box (2496 ms)
Pixel4
Was UVScaleDownBy4_Box (3510 ms)
Now UVScaleDownBy4_Box (2797 ms)
Bug: libuyv:838
Change-Id: Ibbde56e497b0706fbcb7b5ec4a991d40ca17f861
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2469050
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2020-10-13 22:32:25 +00:00
Frank Barchard
d730dc2f18
2x down sample for UV planes ported to SSSE3 / NEON
...
Bug: libuyv:838
Change-Id: Id9fb3282a3e86143d76b5e0cb557f0523a88b3c8
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2465578
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2020-10-13 21:42:15 +00:00
Frank Barchard
e647902212
NV12Scale function and ScaleUV for packed UV plane bilinear scaling
...
Bug: libyuv:718, libyuv:838, b/168918847
Change-Id: I3300c1e7d51407b9c3201cf52b68e2e11346ff5f
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2427868
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
2020-09-29 23:49:05 +00:00
