- avoids intrinsics API changes
- build options for rvv not needed for inline instructions
- declare vl and vtype clobbers instead of cc
- tune for SiFive X280 using 16/32-bit loads and stores instead of
vlseg/vsseg where more efficient
- tune LMUL for SpacemiT K3 (256-bit vectors) with max m4 (m2 where
constrained by register pressure or vlseg/vsseg field rules)
- YUVTORGB use vmulhu
NV12ToARGB performance estimates:
CPU VLEN C Intrinsics Inline ASM Speedup (Inline)
------------------------------------------------------------------------
SiFive X390 1024-bit 360 ms 29 ms 9.5 ms 37.89x (3.05x)
SiFive X280 512-bit 980 ms 115 ms 38.0 ms 25.79x (3.03x)
SpacemiT K3 256-bit 420 ms 68 ms 28.0 ms 15.00x (2.43x)
Google Coral 256-bit 1120 ms 190 ms 58.0 ms 19.31x (3.28x)
T-Head C920 128-bit 510 ms 88 ms 39.0 ms 13.08x (2.26x)
T-Head C908 128-bit 860 ms 145 ms 62.0 ms 13.87x (2.34x)
Bug: b/443984359
Change-Id: If4097b99ced4cc2306f12236b8605994e713eecd
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8426529
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
- AVX2 uses vpmulhrsw to compute ((128 * a * b) + 16384) >> 15.
- RVV uses e8, m2 (widening to e16, m4) with vwmulu.vv, vwaddu.wx 128,
and vnsrl.wi by 8.
CPU C Intrinsics Inline ASM Speedup (Inline)
------------------------------------------------------------------------
SiFive X390 gen 2 (est.) 2150 ms 29 ms 27 ms 79.63x (1.07x)
SpacemiT K3 (est.) 980 ms 41 ms 38 ms 25.79x (1.08x)
T-Head C908 (est.) 2060 ms 100 ms 94 ms 21.91x (1.06x)
Google Coral (est.) 2680 ms 94 ms 88 ms 30.45x (1.07x)
Test: libyuv_unittest --gtest_filter=*ARGBMultiply*
Bug: None
Change-Id: Iff65794623a6f29e5cb6bda7bd3084c8a5f676d1
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8422278
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
Apple processors (such as A18/A19) may report
hw.optional.arm.FEAT_SME = 1 even when the XNU kernel on iOS disables
EL0 (user-mode) SME execution (hw.optional.arm.sme_max_svl_b == 0),
which causes SMSTART or RDSVL to fault with EXC_BAD_INSTRUCTION.
Gate SME/SME2 flags on have_feature("hw.optional.arm.sme_max_svl_b")
being nonzero. See https://github.com/pytorch/cpuinfo/issues/432 and
https://github.com/pytorch/cpuinfo/pull/433.
Also add SME2 detection test in cpu_test.cc and LIBYUV_DISABLE_SME2
environment variable support in unit_test.cc.
Test: libyuv_unittest --gtest_filter=*TestLinuxAArch64*
Bug: 8339302
Change-Id: If3e4d425ebaa89bf27965295ebdb43b844797ca1
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8428226
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
Add SwapUVRow_RVV using RVV intrinsics by default and unit-stride
e16, m4 inline assembly (vle16.v, vsrl.vi, vsll.vi, vor.vv, vse16.v)
under #ifdef RVV_ASM.
CPU C Intrinsics Inline ASM Speedup (Inline)
--------------------------------------------------------------------
SiFive X280 (est.) 980 ms 115 ms 38 ms 25.79x (3.03x)
SpacemiT K3 (est.) 420 ms 68 ms 28 ms 15.00x (2.43x)
T-Head C908 (est.) 860 ms 145 ms 62 ms 13.87x (2.34x)
Google Coral (est.) 1120 ms 190 ms 58 ms 19.31x (3.28x)
Original author: Matteo Forzan <matteo.forzan95@gmail.com>
R=wtc@google.comCC=matteo.forzan95@gmail.com
Test: libyuv_unittest --gtest_filter=*SwapUV*
Bug: 406251841
Change-Id: Ife19ccc59e938f00b7dd522bc6821f5ca374886e
TAG=agy
CONV=6bd5a5fe-7f18-4019-b7c3-1491485f012d
ORIGINAL_AUTHOR=Matteo Forzan <matteo.forzan95@gmail.com>
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8339302
Reviewed-by: David Gao <davidgao@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
The existing SVE2 path only computes the average for a pair of input
values instead of groups of four as done in the Neon implementation.
The average of two values must be retained for the final odd element if
present. Existing code just handles this by falling back to the C
implementation, so mirror that for SVE2 as well.
Reduction in time taken relative to the existing Neon code, the SVE2
code remains faster even with the correct averaging:
Cortex-A510: -12.66%
Lumex C1-Nano: -6.38%
Cortex-A720: -9.73%
Lumex C1-Pro: +0.99%
Cortex-X4: -3.05%
Lumex C1-Ultra: -18.98%
Bug: 546962730
Change-Id: I7ab348e64a9d079b81d070cea5fd575243b037d2
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8252281
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Justin Green <greenjustin@google.com>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Update BUILD.gn, CMakeLists.txt, and BUILD.bazel to specify Armv8.5-A
baseline architecture flags rather than Armv9:
- BUILD.gn: update :libyuv_sve to -march=armv8.5-a+i8mm+sve2 and
:libyuv_sme to -march=armv8.5-a+i8mm+sme.
- CMakeLists.txt: update SME check and compile options to
-march=armv8.5-a+i8mm+sme (SVE target already uses armv8.5-a).
- BUILD.bazel: update ENABLE_SVE_COPTS to -march=armv8.5-a+i8mm+sve2.
Using -march=armv9-a enabled baseline SVE in the compiler, causing
Clang to emit non-streaming SVE instructions (such as cntd in function
prologues prior to smstart) in SME translation units. On processors
such as Apple M4/M5 that support SME but not SVE, executing these
prologue instructions resulted in SIGILL crashes.
Test: ninja -C out/arm64.release libyuv libyuv_unittest
Bug: 558654023
Change-Id: Ic7f62842bc9bd71608cc78ad11fd5e2214ef5e8e
TAG=agy
CONV=a70f0e53-caf2-4fbe-a024-1616ed25ba9a
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8376021
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Restructure RISC-V Vector (RVV) configuration to follow the pattern
used by NEON and SVE:
- Move RVV_COPTS immediately after SVE_COPTS.
- Define ENABLE_RVV_COPTS (-march=rv32gcv) and ENABLE_RVV64_COPTS
(-march=rv64gcv) with SUPPORTS_RVV_COPTS (-DLIBYUV_RVV).
- Revert ALL_COPTS to PLATFORM_COPTS so RVV vector flags are not
applied to generic sources.
- Add ALL_RVV_COPTS, RVV_DEPS, and libyuv_rvv_impl target.
- Update libyuv.bzl to add libyuv_rvv_srcs and exclude source/*rvv*.cc
from libyuv_srcs.
R=wtc@google.com
Test: cmake -B out/Release -DUNIT_TEST=ON && cmake --build out/Release
Bug: none
Change-Id: Ia60a8bc1da9db46f0c5295064d7b9d3d5609a38f
TAG=agy
CONV=06856819-10a7-4815-93d3-cdcd32da7495
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8357976
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Enable RISC-V Vector (RVV) by default when building natively on RISC-V
platforms (such as SpacemiT K1/K3) using CMake and Bazel:
- CMakeLists.txt: Detect riscv64 and riscv32, probe compiler support
for -march=rv64gcv / -march=rv32gcv, and append architecture flags
with auto-vectorization disabled (-fno-vectorize -fno-slp-vectorize).
Support custom architecture flags via -DRISCV_COMPILER_FLAGS and
provide -DUSE_RVV=OFF to disable.
- BUILD.bazel: Add RVV_COPTS for @platforms//cpu:riscv64 and
@platforms//cpu:riscv32 with -fno-vectorize.
- docs/getting_started.md: Document native SBC builds, custom targets
(SiFive X280, Coral 32-bit Zve32x), Bazel, and Blaze commands.
Test: cmake build on host and simulated riscv64; clang cross-compile tests
Bug: none
Change-Id: Iecb20b73b2934b6477e7e0c71b3539a8251e4e8b
TAG=agy
CONV=2f28dd8a-c892-4c29-8aed-606afd090398
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8349493
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
Add SwapUVRow_AVX2 implementation in row_win.cc for MSVC and enable
HAS_SWAPUVROW_AVX2 under USE_ROW_WIN.
Share kShuffleUVToVU shuffle table between SSSE3 and AVX2 in row_gcc.cc.
Make NV12ToNV21 the primary implementation and alias NV21ToNV12 to it.
Add NV12ToNV21 test in convert_test.cc and benchmark dispatch for
SSSE3 and AVX2 in planar_test.cc.
CPU Before After Speedup
----------------------------------------------------------------
Intel Granite Rapids 282 ms 125 ms 2.26x
Intel Emerald Rapids 264 ms 108 ms 2.44x
Intel Sapphire Rapids 276 ms 122 ms 2.26x
Intel Ice Lake 310 ms 122 ms 2.54x
Intel Skylake-X 573 ms 157 ms 3.65x
Intel Haswell 620 ms 162 ms 3.83x
AMD Zen 4 (Genoa) 266 ms 52 ms 5.12x
AMD Zen 6 (Venice) 229 ms 43 ms 5.33x
ARM Neoverse-N1 140 ms 71 ms 1.97x
NVIDIA Grace (Neoverse-V2) 62 ms 69 ms 0.90x
R=wtc@google.com
Test: libyuv_unittest --gtest_filter=*SwapUV*:*NV12ToNV21*:*NV21ToNV12*
Bug: 406251841
Change-Id: I9bed5522aedadea13cdf69c27e3cbe633e31dd5b
TAG=agy
CONV=bb50ac62-0c55-4969-8fff-c71ab3d9a79f
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8356968
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
- instead of 64 bit load, use vlseg2 with 32 bit
Test build with:
-fsanitize=alignment -fsanitize-trap=alignment
Test: libyuv_unittest --gtest_filter=*UVScale*:*ScalePlane*:*ARGBScale*
Bug: 553156658
Change-Id: Ib6febe49f085f5b3deb33adb6b6f82f7d836c3fa
TAG=agy
CONV=b29f62f9-15a4-4f38-8414-86c8330e61fb
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8315435
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
- Also remove redundant neon check for I8MM kernels
======================================================================
CPU Before After Speedup
----------------------------------------------------------------------
Intel Granite Rapids (AVX10.1) 251 ms 141 ms 1.78x (+43.8%)
Intel Emerald Rapids (AVX512) 199 ms 108 ms 1.84x (+45.7%)
Intel Sapphire Rapids (AVX512) 230 ms 112 ms 2.05x (+51.3%)
AMD Turin Zen 5 (AVX512) 156 ms 99 ms 1.58x (+36.5%)
AMD Venice Zen 6 (AVX512) 204 ms 106 ms 1.92x (+48.0%)
Intel Skylake-X (AVX512) 320 ms 156 ms 2.05x (+51.2%)
AMD Genoa Zen 4 (AVX512) 236 ms 346 ms 0.68x (-46.6%)
Intel Ice Lake (AVX512) 242 ms 316 ms 0.77x (-30.6%)
AMD Milan Zen 3 (AVX2) 252 ms 230 ms 1.10x (+8.7%)
Intel Haswell (AVX2) 369 ms 456 ms 0.81x (-23.6%)
NVIDIA Grace (SVE2) 336 ms 337 ms 1.00x (-0.3%)
ARM Neoverse-N1 (NEON) 1081 ms 1105 ms 0.98x (-2.2%)
======================================================================
Bug: 537818862
Change-Id: I3ddeddb040fc52a30dabe7ae2c71d16dda021568
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8343467
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
- NV16/24 FOURCC and document
- Rotate take uint8* for ubsan
Bug: 537818862
Change-Id: I2bc7109abaffe2464e031e33c22296c007077d82
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8339427
Commit-Queue: Frank Barchard <fbarchard@google.com>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Add matrix and BT.601 wrapper APIs for converting ARGB to NV16 and
NV24. Reuse the optimized ARGB Y and UV row functions, then merge
the planar chroma rows into the interleaved destination plane.
NV16 performs horizontal 2:1 chroma subsampling on every source row.
NV24 preserves full-resolution 4:4:4 chroma. Both paths support odd
widths, negative heights, and the existing matrix constants.
Add tests that compare scalar and optimized paths at an odd width and
height through the matrix API and its BT.601 wrapper.
Bug: 537818862
Change-Id: Iddefc714a800cdec4dd1394ac7a52bf7b51fd350
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8314188
Reviewed-by: Frank Barchard <fbarchard@google.com>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
- Add avx512bw, neon 32 bit and RVV versions
- Adapt SSE2 to psrlw instead of pmulhuw
CPU Before After Improvement
---------------------------------------------------------------
Granite Rapids (AVX10) 77 ms 75 ms +2.6% (1.03x)
Emerald Rapids (AVX512) 75 ms 67 ms +10.7% (1.12x)
Sapphire Rapids (AVX512) 91 ms 85 ms +6.6% (1.07x)
AMD Venice Zen 6 (AVX512) 30 ms 30 ms 0.0% (1.00x)
AMD Turin Zen 5 (AVX512) 32 ms 32 ms 0.0% (1.00x)
AMD Genoa Zen 4 (AVX512) 35 ms 36 ms -2.9% (0.97x)
Skylake-X (AVX512) 119 ms 119 ms 0.0% (1.00x)
Ice Lake (AVX512) 93 ms 104 ms -11.8% (0.89x)
AMD Milan Zen 3 (AVX2) 43 ms 44 ms -2.3% (0.98x)
Haswell (AVX2) 129 ms 138 ms -7.0% (0.93x)
NVIDIA Grace (SVE2) 42 ms 43 ms -2.4% (0.98x)
Neoverse-N1 (NEON) 87 ms 87 ms 0.0% (1.00x)
---------------------------------------------------------------
Bug: None
Change-Id: Ia8f310a70c8b1229927fcdf117fe22e131d70ea9
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8330912
Commit-Queue: Frank Barchard <fbarchard@google.com>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Add missing RGBToYMatrixRow_Any_{LSX,LASX} and kAddY offset fix so
RAW/RGB24 to YUV works on LoongArch.
Change-Id: I4c990a3705d7f521fa6d0c9e7e01f1e36f81c379
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8292090
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Frank Barchard <fbarchard@google.com>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Some headers in the include/libyuv directory are private headers and
should not be visible to the clients of the libyuv library target. List
the public headers explicitly in the hdrs list of the libyuv library
target.
Ideally those private headers should be moved to some other directory
such as the source/ directory.
Change-Id: I04b1e209f6891f06c688fcf0bf5d167e82a27818
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8308801
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Frank Barchard <fbarchard@google.com>
1. source/convert_from_argb.cc should include libyuv/convert.h because
it calls ARGBToI420Matrix().
2. Include <string.h> for memset().
Change-Id: I3f4841fe604639c7357148ef4788fe0415f0b81b
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8255654
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Frank Barchard <fbarchard@google.com>
Reduces include/libyuv/basic_types.h to only define LIBYUV_API.
Directly includes <stdint.h> and <stddef.h> where needed in public
headers.
Replaces LIBYUV_BOOL, LIBYUV_FALSE, and LIBYUV_TRUE with C++ bool,
true, and false in C++ code (MJpegDecoder, unit tests) and int (0, 1)
for C API functions.
Updates documentation and unit tests.
Test: ninja -C out/Release libyuv_unittest && ./out/Release/libyuv_unittest --gtest_filter=*
Change-Id: I2d58e872f0f976263bcda546dcc3b1b5f7e53d58
TAG=agy
CONV=2730bc56-93ff-48ed-b615-dcde1b834394
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8280662
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
libyuv/scale_rgb.h declares an API function, so it is a public header
and should be included by the umbrella header libyuv.h.
libyuv/scale_row.h should be an internal header. It should not be
included by the umbrella header libyuv.h.
Fixed: 553139990, 553143607
Change-Id: I5c1a61296a06a96a36c33ec470513ab7bd1f0e66
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8301033
Reviewed-by: Frank Barchard <fbarchard@google.com>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
CPU / Architecture Before (C) After (NEON) Improvement
----------------------------------------------------------------------
Cortex-A53 (Pixel 2 XL Little) 3248 ms 792 ms 4.10x
Cortex-A73 (Pixel 2 XL Big) 432 ms 392 ms 1.10x
Cortex-A55 (Pixel 7 Little) 780 ms 563 ms 1.39x
Cortex-A78 (Pixel 7 Mid) 254 ms 170 ms 1.49x
Cortex-X1 (Pixel 7 Big) 158 ms 129 ms 1.22x
Cortex-A510 (Galaxy S22 Little) 816 ms 441 ms 1.85x
Cortex-A710 (Galaxy S22 Mid) 270 ms 129 ms 2.09x
Cortex-X2 (Galaxy S22 Big) 116 ms 80 ms 1.45x
Cortex-A510 (Galaxy S23 Little) 723 ms 373 ms 1.94x
Cortex-A715 (Galaxy S23 Mid) 183 ms 119 ms 1.54x
Cortex-X3 (Galaxy S23 Big) 122 ms 72 ms 1.69x
Lumix C1 Pro (PD2509 Little) 246 ms 156 ms 1.58x
Lumix C1 Premium (PD2509 Mid) 211 ms 112 ms 1.88x
Lumix C1 Ultra (PD2509 Big) 82 ms 50 ms 1.64x
Oryon-v3M (OnePlus Mid) 233 ms 150 ms 1.55x
Oryon-v3L (OnePlus Big) 103 ms 85 ms 1.21x
On Icelake
Function (1280x720) C SSSE3 AVX2 AVX-512BW Gain
----------------------------------------------------------------------
BlendPlane 901 ms 165 ms 135 ms 115 ms 1.17x
I420Blend 1398 ms 287 ms 198 ms 187 ms 1.06x
Bug: 546077674
Change-Id: I189c44cc757c0e26838a7b6fdb2db1cbe5f96f20
TAG=agy
CONV=0f82c25c-616e-4bb3-beeb-981decb0f40c
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8297264
Commit-Queue: Frank Barchard <fbarchard@google.com>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
LIBYUV_UNBIASED_DATA is a new configuration macro introduced in
commit ba68378. It fixes the AR30 rounding bias offset (+24 / -24) for
10-bit YUV to RGB conversions.
Bug: 551985141
Change-Id: Id724d8cabaee441147d998f77034bc3e31c25a9d
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8298160
Reviewed-by: Frank Barchard <fbarchard@google.com>
CPU / Arch Before (Kernel) After (Kernel) Delta
----------------------------------------------------------------------
Haswell (x86_64) 263 ms (SSSE3) 263 ms (AVX2) 0.0%
Milan (Zen 3) 115 ms (SSSE3) 138 ms (AVX2) -20.0%
SapphRap (x86_64) 173 ms (SSSE3) 178 ms (AVX512VBMI) -2.9%
GNR (Granite R) 191 ms (SSSE3) 189 ms (AVX512VBMI) +1.0%
Turin (Zen 5) 69 ms (SSSE3) 55 ms (AVX512VBMI) +20.3%
Icelake (x86_64) 204 ms (SSSE3) 186 ms (AVX512VBMI) +8.8%
Cascade Lake 237 ms (SSSE3) 229 ms (AVX512BW) +3.4%
Neoverse (aarch64)* 103 ms (SVE2) 103 ms (SVE2) 0.0%
Bug: None
Change-Id: Ib218632e16b3308a214e9ffcbc745d6c41116c6e
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8280536
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
ScalePlane's same-width vertical fast path used the endpoint-oriented
FixedDiv1 step for all upsampling. That stepping is appropriate for
bilinear interpolation, but not for point sampling, where source rows
should cover equally sized destination cells.
Use the same centered FixedDiv stepping as the generic point sampler.
Allow the point path to select the final source row while passing a
zero interpolation stride so row kernels cannot form a pointer beyond
it. Add 8-bit and 16-bit point regression coverage and verify that
bilinear upsampling remains unchanged.
Bug: 537818862
Change-Id: I7bee21b85913b31c146ecf7fe9f5f08608a8a17d
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8264449
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Frank Barchard <fbarchard@google.com>
The READYUV444, READYUVA444, READYUVA422, and STOREARGB macros are
defined but not used.
Change-Id: Ic1174b5b9c7dc8554f362871de73080c32a58537
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8279223
Reviewed-by: Frank Barchard <fbarchard@google.com>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
- reduced shifts to isolate 10 bits
- permi2w instead of packlwd to combine words (ARG) (GB) into ARGB
- vpternlogd to combine A,R,G into 1 register
Saphhire Rapids 10000 iterations of 1280x720P:
Conversion Kernel │ Was │ Now │ Speedup
──────────────────────┼────────────────────────┼────────────────
I420ToAR30_Opt │ 1675 ms │ 1568 ms │ +7.1%
H420ToAR30_Opt │ 1632 ms │ 1555 ms │ +5.2%
I210ToAR30_Opt │ 2459 ms │ 2167 ms │ +13.5%
P010ToAR30_Opt │ 2054 ms │ 1926 ms │ +6.6%
P016ToAR30_Opt │ 2043 ms │ 1938 ms │ +5.4%
Total Test Suite Time │ 9,866 ms │ 9,155 ms │ +7.8% overall
Bug: None
Change-Id: I7972f98408a65a2bb51c506145e4c0c63824c424
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8259796
Commit-Queue: Frank Barchard <fbarchard@google.com>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Add src/third_party/cpython3 CIPD packages for Linux and host platforms
('\''infra/3pp/tools/cpython3/${platform}'\'' at '\''version:3@3.11.9.chromium.38'\'').
This enables hermetic Python execution in Siso builds where Siso configs
inherited from build/ require the hermetic Python binary at
src/third_party/cpython3/linux-amd64/bin/python3 on remote workers.
Bug: chromium:537680330
Change-Id: I439b2af56dcc5f325ec6f3dbe2837cf550e5c0d2
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8266781
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Commit-Queue: Alex Ovsienko <ovsienko@google.com>
In P010ToARGBMatrixFilterOverflow, P210ToARGBMatrixFilterOverflow,
P010ToAR30MatrixFilterOverflow, and P210ToAR30MatrixFilterOverflow,
the tests set src_stride_uv to width (1039) and allocated
width * src_uv_height elements. For odd widths, the interleaved UV
row requires (width + 1) & ~1 elements (1040). Using width for the
stride and buffer size caused the second row to read past the allocated
buffer and access uninitialized memory under MSan.
Fix this by computing src_stride_uv = (width + 1) & ~1 and sizing the
allocated buffer to src_stride_uv * src_uv_height.
In TestI400LargeSize, the test attempts large allocations totaling
over 10 GB (~2 GB for source and ~8.5 GB for destination). Under
MemorySanitizer (MSan), large allocations can exceed shadow memory
virtual address limits or assert. Skip this test under MSan by checking
__SANITIZE_MEMORY__ and __has_feature(memory_sanitizer).
Test: libyuv_unittest --gtest_filter=*FilterOverflow*:*I400Large*
Bug: 549112862
Change-Id: I679b7c67d93822e83f066b1895924589fc7242e7
TAG=agy
CONV=dd635180-175b-4e35-b7f0-2e38105d3506
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8269585
Commit-Queue: Frank Barchard <fbarchard@google.com>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
The INCP instructions are particularly slow on some CPUs and in this
case can be replaced by a simple ADD instruction, so use that instead.
Reduction in time taken relative to the existing SVE2 code for
{YUY2,UYVY}ToUVRow_SVE2:
Cortex-A510: −12.1%
Lumex C1-Nano: 0.0%
Cortex-A720: 0.0%
Lumex C1-Pro: 0.0%
Cortex-X4: −20.9%
Lumex C1-Ultra: −20.3%
Bug: 546962730
Change-Id: Iea3d75f7b70fe8a29f12afdd877ac02d43e49f04
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8252282
Reviewed-by: Frank Barchard <fbarchard@google.com>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
- improves I420ToRAW performance
I420ToRAW on Skylake
Was AVX2
I420ToRAW_Opt (262 ms)
Now AVX512
I420ToRAW_Opt (212 ms)
I420ToRAW on Icelake with avx512vbmi
Was I420ToRAW_Opt (200 ms)
Now I420ToRAW_Opt (187 ms)
Bug: None
Change-Id: If5b171801ba931557d379696fc128d9b155f7c44
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8267746
Commit-Queue: Frank Barchard <fbarchard@google.com>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
