1. Provide script to build qemu & clang locally
2. Enable risc-v cmake build with clang
3. Enabl...

under gcc 12.2.0 using -Wall:

source/row_any.cc: In function ‘void libyuv::DetileRow_16_Any_SSE...

* Run on SiFive internal FPGA:

RAWToARGB_Opt (~2x vs scalar)

RAWToRGBA_Opt (~2x vs scalar)

RA...

Signed-off-by: Bruce Lai <[email protected]>
Change-Id: Ibdc742040940ee6c6402de103759f979f942...

* Supports:

* The standard single-letter Vector detection.

* Vector fp16 detection.

Signed-...

The Clang static analyzer (scan-build) in LLVM 14 warns about
array index out of bounds in scale...

- Allows code to be optimized with clang 17 -flto-thin
- Bump version number to 1864 to allow de...

- YUV to RGB AArch32 kRGBCoeffBias rewind pointer
- ARGBToRGB565Dither declare width and source ...

- Fix redundent assignment compile warning in GCC
- Apply clang-format
- Bump version to 1863

B...

Fix the algorithm for unpacking the lower 2 bits of M2T2 pixels.

Bug: b:258474032
Change-Id: Ie...

Skylake Xeon
AVX2 Transpose4x4_Opt (290 ms)
SSE2 Transpose4x4_Opt (302 ms)
C Transpose4x4_Opt...

Change log: https://chromium.googlesource.com/chromium/src/+log/504c069755..d4adab061f
Full diff...

- Convert MergeUVRow_AVX512BW to assembly
- Enable MergeUVRow_AVX512BW for Windows with clangcl
...

On Skylake Xeon 640x360 100000 iterations
AVX512 MergeUVPlane_Opt (1196 ms)
AVX2 MergeUVPl...

Bug: libyuv:950
Change-Id: Ic9a094463af875aefd927023f730b5f35f8551de
Reviewed-on: https://chromi...

Add ARGBToYMatrixRow_LSX/LASX, RGBAToYMatrixRow_LSX/LASX and
RGBToYMatrixRow_LSX/LASX functions ...

- was dup of 8h but mul of 4s. now use umull

Bug: libyuv:951
Change-Id: If6cb01f5f006c2235886b...

Bug: libyuv:951
Change-Id: Id323656cb6f99b1be0be7aaa854d3cc15feeba69
Reviewed-on: https://chromi...

- Convert 10 and 12 bit biplanar formats to planar.
- Shift 10 MSB to 10 LSB
- P010 is similar t...

This should fix:
gclient.py: error: Wrong dep format: src/third_party/auto/src@ should be of the...

- Minor variable name changes first/last to top/bottom
- Comments explaining rotate temporary bu...

This initial implementation is based on current unoptimized code in webrtc using just plain for ...

Bug: libyuv:950
Change-Id: I5a77bca9a0230fe00abd810939e217833a14683f
Reviewed-on: https://chromi...

The I410To420 implementation does a two step approach for scaling down and 10-to-8 bit conversio...

- MT2T support for source strides added, but only works for positive values.
- Reduced casting i...

Bug: b/258474032, b/257266635
Change-Id: Ic5cbbc60e2e1463361e359a2fe3e97976c1ea929
Reviewed-on: ...

Bug: libyuv:948
Change-Id: Ibdbe7461f8aa909c3d6e91eb0b90194de5fa1064
Reviewed-on: https://chromi...

- show command line help for yuvconstants util
- fix for jpeg link on OSX with CMake

Bug: libyu...

When `libyuv` is built with `jpeg` support it does not directly link
`libjpeg` in any form. As a...

Change log: https://chromium.googlesource.com/chromium/src/+log/1c174f8519..504c069755
Full diff...

- fix ifdefs for DetilePlane_16 to use 16 bit versions, not 8 bit. (no functional change)

Bug:...

- move power of two macro to planar functions source
- revert row.h IS_ALIGNED change

Bug: b/25...

- Neon and SSE2
- Any for odd widths

Pixel 2 little core AArch32 build
C
TestDetilePlane_16 (12...

Change log: https://chromium.googlesource.com/chromium/src/+log/6974809249..1c174f8519
Full diff...

Change log: https://chromium.googlesource.com/chromium/src/+log/219d891495..6974809249
Full diff...

This should remove the problematic Code-Review+1 label that prevents
the upload to complete.

Bu...

"role/swarming.taskTriggerer" gives "swarming.tasks.createInRealm" permission.
https://source.ch...

libyuv project will use the same RBE backend with webrtc project.

Bug: b:239908030
Change-Id: I...

libyuv builders will also migrate from goma to reclient.

Bug: b:239908030
Change-Id: I3719b3703...

Bug: None
Change-Id: I5c829bb52b8e2c5f99b1b6b926fcaeab4ddca115
Reviewed-on: https://chromium-re...

- Preserve xmm7 in ScaleRowUp2_Bilinear_12_SSSE3
- Previously xmm7 was used in ScaleRowUp2_Bilin...

Change log: https://chromium.googlesource.com/chromium/src/+log/a58ed3145a..219d891495
Full diff...

- add {} around loaded register

Bug: libyuv:944
Change-Id: I0d916e37beb50bda0838e4867742eb7afa5...

- move RGB to UV into BIT_EXACT ifdefs for each compiler
- move RGB to Y to always enabled

Bug:...

- Previously was C for both Y and UV.

Was BGRAToI420_Opt (17780 ms)
Now BGRAToI420_Opt (9546 ms...

Change log: https://chromium.googlesource.com/chromium/src/+log/cbf7ac7473..a58ed3145a
Full diff...

Change log: https://chromium.googlesource.com/chromium/src/+log/1b7024f18f..cbf7ac7473
Full diff...

Change log: https://chromium.googlesource.com/chromium/src/+log/28dca358ed..1b7024f18f
Full diff...

Bug: None
Change-Id: I5658addc2c44e6cabfcc8d349e66e13da6c3c0e6
Reviewed-on: https://chromium-rev...

The former allows the package to be overridden with a local build by
`FetchContent`.

Also inclu...

See chromium/src's CL https://crrev.com/c/3869740 for the context.

Bug: chromium:1340825
Change...

- Optimized YUY2ToNV12 that reduces it from 3 steps to 2 steps
- Was SplitUV, memcpy Y, Interp...

This is not needed anymore, Chromium removed this in 2018 [1].

[1] - https://chromium-review.go...

Change log: https://chromium.googlesource.com/chromium/src/+log/004bde16df..28dca358ed
Full diff...

Change log: https://chromium.googlesource.com/chromium/src/+log/7d683aeda8..004bde16df
Full diff...

This also fixes a presubmit warning by formatting a gni file.

No-Try: True
Bug: 942720
Change-I...

- Provide LIBYUV_LIMITED_DATA macro for backwards compatiblity

Bug: b/474156256
Change-Id: I5d5...

Change log: https://chromium.googlesource.com/chromium/src/+log/829c6df33d..7d683aeda8
Full diff...

Bug: libyuv:938, libyuv:942
Change-Id: I6fb6e7e17fa941785e398bc630f465baf72fcabd
Reviewed-on: ht...

- shift high bits of 10 and 12 bit into lower bits

Bug: libyuv:941, libyuv:942,
Change-Id: I143...

Bug: libyuv:942
Change-Id: I7d2dc87a44cc1cc5c79c37f407583e0c907dc2de
Reviewed-on: https://chromi...

- YUV to RGB use linear for first and last row.
- add assert(yuvconstants)
- rename pointers to ...

Bug: None
Change-Id: I3e0748d3aebdc8bc1ff8ddfc79d9cdb751c03e0f
Reviewed-on: https://chromium-rev...

- Implemented as 3 steps: Upsample UV to 4:4:4, I444ToARGB, ARGBToRGB24
- Fix some build warning...

Add SSE2 optimization for MM21ToYUY2 conversion.

Bug: b/238137982
Change-Id: I189f712514308322f...

MM21 to YUY2 use zip1 for performance

Cortex A510
Was MM21ToYUY2 (612 ms)
Now MM21ToYUY2 (573 m...

- add tests for all single plane formats that reduce or stay same in size

Bug: b/242233673
Chan...

Commit a5a1102a added a function to the public ABI. Update the
version number to 1838.

Bug: b/2...

The code already exists to use a specific matrix. This CL simply
adds a function to use a generi...

- fix crash when width is not a multiple of 16
- apply clang format
- bump version

Bug: libyuv:...

Explicitly initialize the 'pad' field of RgbConstants to 0. This
prevents the following warning/...

Fix the following MSVC warnings:
src\source\row_win.cc(117): warning C4309: 'argument': truncati...

Suppress the following MSVC compiler warnings:
src\source\cpu_id.cc(140): warning C4996: 'fopen'...

These changes were landed in the android's snapshot of libyuv.

Pushing those to upstream so tha...

Fix the follwing Ninja warning:
ninja: warning: multiple rules generate yuv.lib. builds involvin...

This file was removed in commit ebbe88ac.

Bug: b/241008246
Change-Id: Id08d9a6e550610c9b00755ed...

- Define HAS_SCALEROWUP2_BILINEAR_16_SSE2: it's now fixed.
- Correct function name to ScaleRowUp...

- Undefine HAS_SCALEROWUP2_BILINEAR_16_SSE2
- Save XMM7 in ScaleRowUp2_Bilinear_16_SSE2().
- Ren...

Bug: b/238765359
No-Try: True
Change-Id: Ic43403f04337625ceb2b3bf69e6d7bf830698088
Reviewed-on: ...

Some hooks were still referring python.

No-Try: True
Bug: b/238765359
Change-Id: I0a547bdc22696...

- This test used to fail on ARM, but is passing now, so re-enable
- Kept behind a flag so it can...

MergeRGB and SplitRGB use a register to point to 9 shuffle tables.

- fixes an out of registers ...

- Avoid stepping to height + 1 for bilinear filter 2nd row for last row of source
- Box filter u...

Bug: b/228605787
Change-Id: Ia8912e4c1599401320ee82882a2593e78bf56582
Reviewed-on: https://chrom...

Bug: libyuv:931, b/228605787, b/233233302, b/233634772, b/234558395, b/234340482
Change-Id: Ib13...

Fixes chromium PaintCanvasVideoRendererTest.HighBitDepth

sqdmulh was creating a 9 bit value wit...

This reverts commit 60254a1d846a93a4d7559009004cdd91bcc04d82.

Reason for revert: breaks PaintCa...

- Add I210ToI420 to convert 10 bit 4:2:2 YUV to 4:2:0 8 bit
- Add NEON InterpolateRow_16 for fas...

If a width, height, and src/dst strides passed in are all 0, height is updated to 1 which means ...

Bug: None
Change-Id: I7fbb7621904dad786b00e853f72ed9dabab7735a
Reviewed-on: https://chromium-re...

Pixel 3
Was C I010ToI420_Opt (749 ms)
Now NEON I010ToI420_Opt (356 ms)

Pixel 4
Was C I010...

This function reads 2 byte values and writes the 2nd byte to the destination.
It turns out this ...

When building for ARM target with "libyuv_use_neon" set to false, the
build fails because there ...

Bug: b/230550621
Change-Id: Ie36d3b8bdadb4300d54611798a4dfd488c30ca8d
Reviewed-on: https://chrom...

Bug: b/228841445
Change-Id: I0342856e1bfcea69851d718459d66926bb170219
Reviewed-on: https://chrom...

On Pixel 4
Was C
AArch64 TestDetileSplitUVPlane_Benchmark (935 ms)
AArch32 TestDetileSplitUVPlan...

Bug: b/228997215
Change-Id: If4f35179c5ef1a0d49e8b597d9ba4ca76fdf0364
Reviewed-on: https://chrom...

Bug: libyuv:928

xed -i scale_gcc.o:
SYM ScaleUVRowUp2_Linear_16_SSE2:
XDIS 0: LOGICAL SSE2 ...

The "vpackuswb %%xmm2,%%xmm0,%%xmm0" and "vmovdqu %%xmm0,(%1)"
instructions in ScaleRowUp2...

Was
[----------] 59 tests from LibYUVScaleTest (223179 ms total)
Now
[----------] 13 tests from ...