Frank Barchard
0c07032182
clang format applies to git repo
...
Bug: None
Change-Id: Ida65a0033e8c783230cadf6912416ffd9bbf90e1
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6393515
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2025-03-25 11:49:25 -07:00
Jordan
0fd4581a51
Updating license id for libyuv
...
Bug: b/358504615
Change-Id: I93fecd22c16df8949a8ebe85aabe539c0231985e
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6275535
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2025-03-18 16:37:39 -07:00
Hang Nguyen
4ed75166cf
Enable cfi in sanitizers for library
...
This CL enables the CFI checks for libyuv to be used as a
shared library.
Bug: 400789169
Change-Id: I8c71df235ad6962d02740c976972d8f9dcea6c52
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6353950
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: James Zern <jzern@google.com>
Commit-Queue: Hang Nguyen <hnt@chromium.org>
2025-03-14 11:34:37 -07:00
Frank Barchard
918329caee
Make constant 0x0101 using vpcmpeqb+vpabsb
...
Was
vpcmpeqb %%ymm4,%%ymm4,%%ymm4
vpsrlw $0xf,%%ymm4,%%ymm4
vpackuswb %%ymm4,%%ymm4,%%ymm4
Now
vpcmpeqb %%ymm4,%%ymm4,%%ymm4
vpabsb %%ymm4,%%ymm4
Bug: 381138208
Change-Id: Ib70c24ac636fff95a10c7f06ed8f0a3bc7514906
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6312925
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Justin Green <greenjustin@google.com>
2025-03-10 13:25:16 -07:00
Frank Barchard
c060118bea
ARGBToJ444 use 256 for fixed point scale UV
...
- use negative coefficients for UV to allow -128
- change shift to truncate instead of round for UV
- adapt all row_gcc RGB to UV into matrix functions
- add -DLIBYUV_ENABLE_ROWWIN to allow clang on Windows to use row_win.cc
Bug: 381138208
Change-Id: I6016062c859faf147a8a2cdea6c09976cbf2963c
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6277710
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Reviewed-by: James Zern <jzern@google.com>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2025-02-27 13:04:15 -08:00
Frank Barchard
5257ba4db0
Apply clang format
...
Bug: None
Change-Id: Ibd694d0351966a2b5812445de74bbced9c881a79
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6302317
Reviewed-by: James Zern <jzern@google.com>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2025-02-25 11:39:19 -08:00
Frank Barchard
3a7e0ba671
Apply format with no code changes
...
Bug: None
Change-Id: I8923bacb9af7e7d4f13e210c8b3d7ea6b81568a5
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6301086
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2025-02-24 23:57:01 -08:00
Frank Barchard
61354d2671
ARGBToUV Matrix for AVX2 and SSSE3
...
- Round before shifting to 8 bit to match NEON
- RAWToARGB use unaligned loads and port to AVX2
Was C/SSSE/AVX2
ARGBToI444_Opt (343 ms)
ARGBToJ444_Opt (677 ms)
RAWToI444_Opt (405 ms)
RAWToJ444_Opt (803 ms)
Now AVX2
ARGBToI444_Opt (283 ms)
ARGBToJ444_Opt (284 ms)
RAWToI444_Opt (316 ms)
RAWToJ444_Opt (339 ms)
Profile Now AVX2
38.31% ARGBToUVJ444Row_AVX2
32.31% RAWToARGBRow_AVX2
23.99% ARGBToYJRow_AVX2
Profile Was C/SSSE/AVX2
73.15% ARGBToUVJ444Row_C
15.74% RAWToARGBRow_SSSE3
8.87% ARGBToYJRow_AVX2
Bug: 381138208
Change-Id: I696b2d83435bc985aa38df831e01ff1a658da56e
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6231592
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Ben Weiss <bweiss@google.com>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2025-02-10 18:36:18 -08:00
Frank Barchard
d32d19ccf2
UV subsample on ARM use rounding average of 4 pixels
...
Performance on Samsung S22 Exynos (SVE2+I8MM+DOTPROD+Neon)
AArch64
ARGBToI400_Opt (168 ms)
ARGBToJ400_Opt (103 ms)
ABGRToJ400_Opt (81 ms)
RGBAToJ400_Opt (82 ms)
RGB24ToJ400_Opt (176 ms)
RAWToJ400_Opt (176 ms)
ABGRToI420_Opt (258 ms)
ARGBToI420_Opt (259 ms)
ARGBToI422_Opt (403 ms)
ARGBToI444_Opt (213 ms)
ARGBToJ420_Opt (257 ms)
ARGBToJ422_Opt (403 ms)
ARGBToJ444_Opt (214 ms)
ABGRToJ420_Opt (255 ms)
ABGRToJ422_Opt (399 ms)
ARGB4444ToI420_Opt (285 ms)
RGB565ToI420_Opt (316 ms)
ARGB1555ToI420_Opt (324 ms)
BGRAToI420_Opt (260 ms)
RAWToI420_Opt (303 ms)
RAWToI444_Opt (303 ms)
RAWToJ420_Opt (335 ms)
RAWToJ444_Opt (308 ms)
RGB24ToI420_Opt (372 ms)
RGB24ToJ420_Opt (365 ms)
RGBAToI420_Opt (259 ms)
AArch32 (Neon)
ARGBToI400_Opt (496 ms)
ARGBToJ400_Opt (478 ms)
ABGRToJ400_Opt (483 ms)
RGBAToJ400_Opt (493 ms)
RGB24ToJ400_Opt (343 ms)
RAWToJ400_Opt (341 ms)
ABGRToI420_Opt (993 ms)
ARGBToI420_Opt (992 ms)
ARGBToI422_Opt (1503 ms)
ARGBToI444_Opt (1257 ms)
ARGBToJ420_Opt (1006 ms)
ARGBToJ422_Opt (1521 ms)
ARGBToJ444_Opt (1267 ms)
ABGRToJ420_Opt (1002 ms)
ABGRToJ422_Opt (1504 ms)
ARGB4444ToI420_Opt (1180 ms)
RGB565ToI420_Opt (1112 ms)
ARGB1555ToI420_Opt (1115 ms)
BGRAToI420_Opt (993 ms)
RAWToI420_Opt (703 ms)
RAWToI444_Opt (1717 ms)
RAWToJ420_Opt (704 ms)
RAWToJ444_Opt (1739 ms)
RGB24ToI420_Opt (703 ms)
RGB24ToJ420_Opt (703 ms)
RGBAToI420_Opt (993 ms)
Bug: 381138208
Change-Id: I33728d5237f357362b0bfc509a9ebe6fe46f45d4
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6228987
Reviewed-by: Ben Weiss <bweiss@google.com>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2025-02-04 15:19:19 -08:00
George Steed
ccdf870348
[AArch64] Fix up inline asm name in Convert8To8Row_SVE_SC
...
The existing implementation mistakenly refers to the parameter %2. This
works fine however the parameter is already named %[width], and using
the name should be preferred.
Change-Id: Ifaf8fc83cdfc9b15c79d52e7e47cb72b53270a12
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6225753
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2025-02-04 10:28:17 -08:00
Frank Barchard
5a9a6ea936
Add RAWToI444
...
Skylake Xeon
RAWToI444_Opt (433 ms)
RAWToJ444_Opt (1781 ms)
ARGBToI444_Opt (352 ms)
ARGBToJ444_Opt (1577 ms)
Samsung S22 Exynos
ARGBToI444_Opt (283 ms)
ARGBToJ444_Opt (209 ms)
RAWToI444_Opt (294 ms)
RAWToJ444_Opt (293 ms)
Profiling on Samsung S22 Exynos
37.62%, ARGBToUV444Row_NEON_I8MM
29.42%, RAWToARGBRow_SVE2
19.61%, ARGBToYRow_NEON_DotProd
Passing different --libyuv_cpu_info=N etc we can compare each ISA
C 1 RAWToI444_Opt (781 ms)
NEON 511 RAWToI444_Opt (757 ms)
NEONDOT 1023 RAWToI444_Opt (571 ms)
NEONI8MM 2047 RAWToI444_Opt (334 ms)
SVE2 8191 RAWToI444_Opt (307 ms)
Bug: 390247964
Change-Id: I0316fedd32222588455afa751f5b854f46bce024
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6223658
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2025-02-03 16:13:03 -08:00
Frank Barchard
b3fd3f3f3b
Fix ARGBToUV444Row_NEON
...
- constants passed in are signed and need to be negated to positive.
Bug: 394127527
Change-Id: I531e475d2ddd4583922d4abef13b9282d002dd7a
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6226854
Reviewed-by: Ben Weiss <bweiss@google.com>
2025-02-03 13:33:39 -08:00
Frank Barchard
96f98f6915
ARGBToJ444 and RAWToJ444 NEON
...
- Pass JPEG matrix to ARGBToUV444MatrixRow_NEON
- Remove NEON unsigned constants in favor of DOTPROD signed constants
Samsung S23:
Was C for UV
ARGBToJ444_Opt (320 ms)
RAWToJ444_Opt (411 ms)
Now I8MM
ARGBToJ444_Opt (196 ms)
RAWToJ444_Opt (301 ms)
NEON
ARGBToJ444_Opt (505 ms)
RAWToJ444_Opt (596 ms)
32 bit ARM NEON
ARGBToJ444_Opt (1135 ms)
RAWToJ444_Opt (1546 ms)
Profile of RAWToJ444
37.72% ARGBToUVJ444Row_NEON_I8MM
34.48% RAWToARGBRow_NEON
14.65% ARGBToYJRow_NEON_DotProd
Bug: 390247964
Change-Id: Ia26240bee974a0baf502548f2fc896b193c3006c
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6220890
Reviewed-by: Ben Weiss <bweiss@google.com>
2025-01-31 16:46:29 -08:00
Björn Terelius
1c938f342d
Try removing some unused DEPS
...
Bug: None
Change-Id: I8a0b4ffefbc8ef82627c4f3a5ef5e3fb226e80bb
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6159066
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Commit-Queue: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2025-01-31 01:10:17 -08:00
Frank Barchard
c1bac9e6a5
RAWToJ444 and ARGBToJ444
...
- ARGBToJ444 implements ARGBToUVJ444Row_C
- RAWToJ444 implemented as 2 steps - RAWToARGB and ARGBToJ444
libyuv_test '--gunit_filter=*R*To?444_Opt' --libyuv_width=1280 --libyuv_height=720 --libyuv_repeat=1000 --libyuv_flags=-1 --libyuv_cpu_info=-1
(with bit exact off)
Samsung S23
RAWToJ444_Opt (437 ms)
ARGBToJ444_Opt (337 ms)
ARGBToI444_Opt (196 ms)
Skylake Xeon
RAWToJ444_Opt (1699 ms)
ARGBToJ444_Opt (1559 ms)
ARGBToI444_Opt (346 ms)
Bug: 390247964
Change-Id: Id1b1b45a5e4512ab50830aadf62f780fbe631575
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6207845
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2025-01-29 15:18:38 -08:00
George Steed
c4a0c8d34a
[AArch64] Add SVE2 and SME implementations for Convert8To8Row
...
SVE can make use of the UMULH instruction to avoid needing separate
widening multiply and narrowing steps for the scale application.
Reduction in runtime for Convert8To8Row_SVE2 observed compared to the
existing Neon implementation:
Cortex-A510: -13.2%
Cortex-A520: -16.4%
Cortex-A710: -37.1%
Cortex-A715: -38.5%
Cortex-A720: -38.4%
Cortex-X2: -33.2%
Cortex-X3: -31.8%
Cortex-X4: -31.8%
Cortex-X925: -13.9%
Change-Id: I17c0cb81661c5fbce786b47cdf481549cfdcbfc7
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6207692
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2025-01-28 15:53:26 -08:00
Wan-Teh Chang
eacb08c83e
Update "TEST" to "UNIT_TEST" in a comment
...
Change-Id: I53588bc027c1e75b2ee7f7fc155fd25d75e932bf
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6203445
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
2025-01-27 16:21:09 -08:00
Frank Barchard
6c2415bfab
J420ToI420 AVX2
...
libyuv_test '--gunit_filter=*J420ToI420*' --libyuv_width=1280 --libyuv_height=720 --libyuv_repeat=1000 --libyuv_flags=-1 --libyuv_cpu_info=-1
Skylake Xeon
AVX2 J420ToI420_Opt (114 ms)
C J420ToI420_Opt (596 ms)
Sapphire Rapids
AVX2 J420ToI420_Opt (126 ms)
C J420ToI420_Opt (717 ms)
Samsung S23
NEON J420ToI420_Opt (46 ms)
C J420ToI420_Opt (95 ms)
Bug: 381327032
Change-Id: I2b551507c2a8b1da4f04651b622fc9247a75050d
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6201239
Reviewed-by: Justin Green <greenjustin@google.com>
2025-01-27 11:23:44 -08:00
Frank Barchard
67f3f17d9a
aarch32 J420ToI420
...
benchmark on medium core
adbrun -- taskset 10 blaze-bin/third_party/libyuv/libyuv_test '--gunit_filter=*J420ToI420*' --gunit_also_run_disabled_tests --libyuv_width=1280 --libyuv_height=720 --libyuv_repeat=1000 --libyuv_flags=-1 --libyuv_cpu_info=-1
Now Neon
J420ToI420_Opt (159 ms)
Was C
J420ToI420_Opt (215 ms)
AArch64
J420ToI420_Opt (93 ms)
C version does this:
vld1.8 {d20, d21}, [r6]!
vorr q12, q8, q8
subs r4, #16
vmovl.u8 q11, d21
vmovl.u8 q10, d20
vmul.i16 q11, q9, q11
vmul.i16 q10, q9, q10
vsra.u16 q12, q11, #8
vorr q11, q8, q8
vsra.u16 q11, q10, #8
vmovn.i16 d21, q12
vmovn.i16 d20, q11
vst1.8 {d20, d21}, [r5]!
bne 0x3d9078 <Convert8To8Row_C+0x36> @ imm = #-54
Explanation of above C code
vorr moves 16 into register
vsra does shift + accumulate to that register
Compared to aarch64
instead of mull, C uses movl+mul
instead of uzp2, C uses sra #8 + movn. takes 2 movn vs 1 uzp2
instead of add, C does vorr + sra
Change-Id: I9648f06e52ccbafaecf07bd89f8ffff27565d025
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6189497
Reviewed-by: Justin Green <greenjustin@google.com>
2025-01-22 13:47:09 -08:00
Frank Barchard
26277baf96
J420ToI420 using planar 8 bit scaling
...
- Add Convert8To8Plane which scale and add 8 bit values allowing full range
YUV to be converted to limited range YUV
libyuv_test '--gunit_filter=*J420ToI420*' --gunit_also_run_disabled_tests --libyuv_width=1280 --libyuv_height=720 --libyuv_repeat=1000 --libyuv_flags=-1 --libyuv_cpu_info=-1
Samsung S23
J420ToI420_Opt (45 ms)
I420ToI420_Opt (37 ms)
Skylake
J420ToI420_Opt (596 ms)
I420ToI420_Opt (99 ms)
Bug: 381327032
Change-Id: I380c3fa783491f2e3727af28b0ea9ce16d2bb8a4
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6182631
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2025-01-22 02:50:24 -08:00
Frank Barchard
ef52c1658a
avx10_2 detect
...
Run with sde only -dmr reports AVX10.2
emr:Has AVX10_2 0x0
adl:Has AVX10_2 0x0
icx:Has AVX10_2 0x0
snb:Has AVX10_2 0x0
tnt:Has AVX10_2 0x0
icl:Has AVX10_2 0x0
slm:Has AVX10_2 0x0
dmr:Has AVX10_2 0x2000000
cwf:Has AVX10_2 0x0
mrm:Has AVX10_2 0x0
skx:Has AVX10_2 0x0
wsm:Has AVX10_2 0x0
gnr:Has AVX10_2 0x0
gnr256:Has AVX10_2 0x0
bdw:Has AVX10_2 0x0
cpx:Has AVX10_2 0x0
rpl:Has AVX10_2 0x0
snr:Has AVX10_2 0x0
ptl:Has AVX10_2 0x0
slt:Has AVX10_2 0x0
ivb:Has AVX10_2 0x0
spr:Has AVX10_2 0x0
tgl:Has AVX10_2 0x0
arl:Has AVX10_2 0x0
srf:Has AVX10_2 0x0
nhm:Has AVX10_2 0x0
skl:Has AVX10_2 0x0
mtl:Has AVX10_2 0x0
pnr:Has AVX10_2 0x0
glp:Has AVX10_2 0x0
lnl:Has AVX10_2 0x0
cnl:Has AVX10_2 0x0
hsw:Has AVX10_2 0x0
clx:Has AVX10_2 0x0
glm:Has AVX10_2 0x0
sde -dmr -- libyuv_test --gunit_filter=*Cpu*
[ RUN ] LibYUVBaseTest.TestCpuId
Cpu Vendor: GenuineIntel 0x756e6547 0x49656e69 0x6c65746e
Cpu Family 6 (0x6), Model 214 (0xd6)
[ OK ] LibYUVBaseTest.TestCpuId (34 ms)
[ RUN ] LibYUVBaseTest.TestCpuHas
Kernel Version 6.10
Has X86 0x8
Has SSE2 0x100
Has SSSE3 0x200
Has SSE4.1 0x400
Has SSE4.2 0x800
Has AVX 0x1000
Has AVX2 0x2000
Has ERMS 0x4000
Has FSMR 0x8000
Has FMA3 0x10000
Has F16C 0x20000
Has AVX512BW 0x40000
Has AVX512VL 0x80000
Has AVX512VNNI 0x100000
Has AVX512VBMI 0x200000
Has AVX512VBMI2 0x400000
Has AVX512VBITALG 0x800000
Has AVX10 0x1000000
Has AVX10_2 0x2000000
HAS AVXVNNI 0x4000000
Has AVXVNNIINT8 0x8000000
Has AMXINT8 0x10000000
[ OK ] LibYUVBaseTest.TestCpuHas (10 ms)
This is how oneDNN does avx10 version:
e15d2c220f/src/cpu/x64/xbyak/xbyak_util.h (L698-L701)https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6172267
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
2025-01-21 13:53:19 -08:00
Frank Barchard
10592b60c0
Add required 'Security Critical' field to README.chromium
...
Marked as yes to match webrtc
Bug: b/365319755
Change-Id: I92ba17e3215b6290211519e2b671087ec1386270
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6170587
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Mirko Bonadei <mbonadei@chromium.org>
2025-01-12 03:45:13 -08:00
Jeremy Leconte
e2c92b560c
Update protobuf version in .vpython3.
...
Bug: None
Change-Id: Ib02151731d93c460ad23ff93d7ff4370010bc7b3
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6162534
Commit-Queue: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2025-01-09 00:32:24 -08:00
Jeremy Leconte
2caaf190bb
Revert "Remove linux_tsan2 bot from CQ."
...
This reverts commit 4115e118e880c7089babeb96d449ca30a8a91a2b.
Reason for revert: fixed with the chromium roll
Original change's description:
> Remove linux_tsan2 bot from CQ.
>
> No-Try: True
> Bug: b/384912633
> Change-Id: Ic8a0f50dfd328df8770fbadfe5be51cc40232ee9
> Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105834
> Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
> Commit-Queue: Jeremy Leconte <jleconte@google.com>
Bug: b/384912633
Change-Id: I1b40e659af1d3253c89b20f18385777189812e56
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6162535
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2025-01-09 00:26:58 -08:00
Björn Terelius
84186e163a
Remove libyuv dependency on base/
...
Bug:webrtc:42223878
Change-Id: I3940d2b481e9bbcbbba120353c136ae8f229bacc
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6159065
Reviewed-by: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
2025-01-08 09:04:21 -08:00
Björn Terelius
8a93618f98
Roll chromium_revision f2539aa88a..908f3898af (1398488:1403569)
...
Change log: f2539aa88a..908f3898aff2539aa88a..908f3898afaefc825509..f9db0ed7ca1923aa3e32..f3e95cc9a0b8ff6f5a95..dc7418832615caecbcad..6e4e345fbb0305a85a20..4341e4d7a2d9ea6d4e7d..f25a92da84https://boringssl.googlesource.com/boringssl.git/+log/c361e27940..d3f61eeacb
* src/third_party/catapult: https://chromium.googlesource.com/catapult.git/+log/e7c18388ac..8491e07230
* src/third_party/depot_tools: fcb6178914..423f1e191438272bf853..afc7000cacf3c355f9dd..7d76a231b0d0ddad5b79..74dd76082602cb512292..7681005c624f9b6ad674..2019a9e40bfc2c4ea571..09973d22d8f2539aa88a..908f3898aff2539aa88a..908f3898afhttps://chromium-review.googlesource.com/c/libyuv/libyuv/+/6159064
Commit-Queue: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2025-01-08 08:25:57 -08:00
Björn Terelius
dd216e765c
Remove linux_msan from CQ
...
Bug:libyuv:388428508
No-Try: true
Change-Id: I808653d40e1d3cc8ed94358c633bd97e98441c9d
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6158985
Reviewed-by: Jeremy Leconte <jleconte@google.com>
Commit-Queue: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2025-01-08 08:13:33 -08:00
Björn Terelius
80d19e4906
Roll chromium_revision af3d01376b..f2539aa88a (1174635:1398488)
...
NOTE: The revision range is not quite correct due to the large number of manual changes.
* Update third_party/android_sdk/public/ packages in DEPS
* Update DEPS to cipd as required
* Add support for 'non_git_source'
* Update third_party/instrumented_libs
* Remove "terminal".qemu-x64 fuchsia boot image
* Use a normal dependency on clang-format instead of pulling based on hashes
* Add re2 and llvm libc to DEPS
* Remove dependency on //testing/android/native_test:native_test_native_code (this is to avoid getting an indirect dependency on rust through //base.)
* Add buildtools and third_party/instrumented_libs to recursedeps
* Allow for mac cross compile on linux
Change log: af3d01376b..f2539aa88aaf3d01376b..f2539aa88ad407b7061b..aefc8255095885d3c248..1923aa3e3279ab87fa54..b8ff6f5a95ddd58e86cf..15caecbcad51e9a02297..0305a85a202dc4b18abd..d9ea6d4e7dhttps://boringssl.googlesource.com/boringssl.git/+log/20a06474c0..c361e27940
* src/third_party/catapult: https://chromium.googlesource.com/catapult.git/+log/fa05d995e1..e7c18388ac
* src/third_party/clang-format/script: f97059df7f..37f6e68a10d3e43dd431..fcb61789149e3c5d7e18..38272bf853af29db7ec2..f3c355f9dddb700b5670..1c249be96ee8c3bc9ea9..bbccc2f6ef05fe2a64f5..0eb5ce728484fb809dd6..d0ddad5b798d21803b90..02cb51229230bdb85e30..927aabfcd2f1c687e0aa..d1e95b102f4dbfa0e8c8..215bddfd8e7fc833e889..f477acb104a76c0dbb64..fc2c4ea571af3d01376b..f2539aa88aaf3d01376b..f2539aa88ahttps://chromium-review.googlesource.com/c/libyuv/libyuv/+/6111088
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2025-01-08 06:40:17 -08:00
Björn Terelius
cacaf42e97
Revert "Update third_party/android_sdk/public/ packages in DEPS"
...
This reverts commit 9dfbd0e861e8a01da4bea47cbbb4494ea60719e7.
Reason for revert: This is merged into https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6111088
Original change's description:
> Update third_party/android_sdk/public/ packages in DEPS
>
> Bug: libyuv:358992053
> Change-Id: I0672979f79c395c55a5236fff9deb829624397c0
> Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105835
> Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
> Reviewed-by: Jeremy Leconte <jleconte@google.com>
Bug: libyuv:358992053
Change-Id: If25f69d2378a80948c60691b30691eb962ca1d54
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6135003
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
2025-01-03 04:23:13 -08:00
Björn Terelius
9dfbd0e861
Update third_party/android_sdk/public/ packages in DEPS
...
Bug: libyuv:358992053
Change-Id: I0672979f79c395c55a5236fff9deb829624397c0
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105835
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
2025-01-02 14:21:28 -08:00
Frank Barchard
47ddac2996
Sub sampling conversions use CopyPlane for Y channel
...
- Replace ScalePlane with CopyPlane for Y channel
- Vertical mirroring is supported, but not horizontal mirroring.
- Check src_y is not null when dst_y is not null for all libyuv functions that allow a null dst_y.
- Apply clang-format
- Bump version to 1899
Bug: None
Change-Id: Id1805b52b8024ba95a7f1b098dabf45af48670eb
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6128599
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2025-01-02 13:34:11 -08:00
Frank Barchard
e0040eb318
Apply clang format
...
Bug: None
Change-Id: I0d9db4b384144523e61ae32b6ab3f72e93a0c265
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6138934
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2025-01-02 13:31:20 -08:00
Darren Hsieh
b5a18f9d93
[RVV] Optimize ScaleARGBFilterCols with RVV
...
* Run on SiFive internal FPGA:
Test Case Speedup
ARGBScaleDownBy3by8_Linear x2.05
ARGBScaleDownBy3by8_Bilinear x1.76
ARGBScaleDownBy3by8_Box x1.76
Bug: 42280924
Co-Developed-by: Bruce Lai <bruce.lai@sifive.com>
Change-Id: Ib9979b1f2ca92d2ef5aa373f9b2459c246ded6c8
Signed-off-by: Darren Hsieh <darren.hsieh@sifive.com>
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/5103572
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Bruce Lai <bruce.lai@sifive.com>
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2024-12-29 17:32:00 -08:00
George Steed
cce8950816
[AArch64] Remove unused SVE INDEX instrs from NV{12,21} kernels
...
When reading subsampled UV data in NV{12,21} we previously needed to
permute the data to both (a) duplicate each element into the
corresponding pair of lanes for the Y elements; and (b) arrange the UV
components in the correct lanes. This was done in a vector-length
agnostic way by generating the permute indices dynamically at runtime
through an SVE INDEX instruction.
Now that we are using the READNV_SVE_2X macro everywhere these
instructions are now redundant: the multiplications are done on the
subsampled UV data before the duplication and the conversion macro takes
arguments that adjust whether we need to operate on the even or odd
lanes of the vector.
Since the permute indices generated by these INDEX instructions are now
unused, remove them.
Change-Id: I3298a83aadfda52c4cc89bc4fd6518b06765a187
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6089957
Reviewed-by: Justin Green <greenjustin@google.com>
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2024-12-26 14:47:00 -08:00
Jeremy Leconte
533dc5866b
Use PIE when compiling tests on Linux.
...
This is an attempt to make linux_msan bot less flaky which is suggested by the error message:
FATAL: Make sure to compile with -fPIE and to link with -pie.
Bug: None
Change-Id: I43f4e5c10d9a4c42a8ef5a67bf77806f0c09bf98
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6111093
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
2024-12-19 09:16:10 -08:00
Jeremy Leconte
f54aac5279
Remove .vpython.
...
No-Try: True
Bug: None
Change-Id: I4c1b901831d25acd33e766b9f4cbbe1db9d38ab2
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6111091
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
2024-12-19 08:35:12 -08:00
Jeremy Leconte
e01109dd9b
Add constraints to the luci buckets.
...
Goal is to be able to run led jobs, current error is:
constraints for libyuv:try not found
Bug: b/384862261
Change-Id: Ia88f2de7a4068baf151f94ea8cf708983939d11a
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105837
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2024-12-18 12:09:19 -08:00
Jeremy Leconte
4115e118e8
Remove linux_tsan2 bot from CQ.
...
No-Try: True
Bug: b/384912633
Change-Id: Ic8a0f50dfd328df8770fbadfe5be51cc40232ee9
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105834
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
2024-12-18 08:16:20 -08:00
Björn Terelius
13d5e11d38
Bring in changes from WebRTC autoroller (2)
...
This adds support for GCS deps
No-Try: True
Bug: libyuv:358992053
Change-Id: I22fd87cf783faf2b0b35df108df6cfa1fb181657
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105833
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
2024-12-18 07:44:35 -08:00
Björn Terelius
318a1ad595
Bring in changes from WebRTC autoroller (1)
...
This mirrors https://webrtc-review.googlesource.com/c/src/+/361680
No-Try: True
Bug: libyuv:358992053
Change-Id: I38852169a95d55d87ec10c9804b0759cda25a466
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105832
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
2024-12-18 07:16:17 -08:00
Björn Terelius
427a2c5c76
Fix autoroller find_depot_tools import
...
The import statement was incorrectly moved in https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105550
No-Try: True
Bug: libyuv:358992053
Change-Id: I1dc58b2bc2717ec9acac0f77d89a1e24f8158c6b
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105631
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
2024-12-18 06:36:49 -08:00
Björn Terelius
37e88ecc68
Update pylintrc to a pep-8 like style
...
In particular, this requires reformatting
- autoroller/
- PRESUBMIT.py
- infra/config/PRESUBMIT.py
- get_landmines.py
- download_vs_toolchain.py
as well as some manual fixes.
This CL is preparation to updating the autoroller so that we can roll DEPS from chromium.
No-Try: True
Bug: libyuv:358992053
Change-Id: I090a09c6fa61beff2427a0537bca371f0839fa3e
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6105550
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
Commit-Queue: Jeremy Leconte <jleconte@google.com>
2024-12-18 05:38:56 -08:00
Jeremy Leconte
2c682f8d59
Update vpython3 urllib3 version and add brotli.
...
This is to avoid 'brotli' has no attribute 'error' when running tests on Android bots.
No-Try: true
Bug: b/384862261
Change-Id: I62776cf53765d88a167a76d4675488aa37d8809e
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6097483
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
2024-12-18 05:18:50 -08:00
Wan-Teh Chang
86b09b2450
Free dst_c, dst_opt buffers in TestHalfFloatPlane
...
Fix memory leaks introduced in commit 5951464:
https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6043392
Bug: 377693555
Change-Id: I20311bdc33e98c0b8e79679a2db42c7439ca89df
No-Try: True
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6093137
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: George Steed <george.steed@arm.com>
Commit-Queue: Mirko Bonadei <mbonadei@chromium.org>
2024-12-13 08:28:17 -08:00
George Steed
45c7107f95
[AArch64] Fix compilation when SME is not supported
...
The STREAMING_COMPATIBLE macro is designed to enable use of the
__arm_streaming_compatible attribute with the intent that this macro
expanded to empty if SME is not supported by the compiler or platform
being compiled for, however in reality this macro remained undefined
causing compilation to fail. Fix this by defining the macro to empty as
originally intended.
No-Try: True
Change-Id: I8f5a8a606289b7c045fa1cce609f5a6d644891ac
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6087913
Reviewed-by: Justin Green <greenjustin@google.com>
Commit-Queue: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: Wan-Teh Chang <wtc@google.com>
2024-12-13 08:16:50 -08:00
George Steed
db5a71c528
[AArch64] Remove unused variables in HalfRow_{16To8,16}_SME
...
The HalfRow kernels assume that the fraction is exactly half, so there
is no need to calculate it.
No-Try: True
Change-Id: I2319d55ba99f202aa22c9693ec44c9891e7f72d5
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6087914
Reviewed-by: Wan-Teh Chang <wtc@google.com>
Reviewed-by: Justin Green <greenjustin@google.com>
Commit-Queue: Mirko Bonadei <mbonadei@chromium.org>
2024-12-13 08:00:58 -08:00
George Steed
7fd0bd197e
[AArch64] Port YUVToRGB color conversions to SME
...
Some of the color conversion kernels already have Streaming-SVE
implementations however many do not. We can re-use the existing SVE
implementation by moving it to a new shared row_sve.h header and marking
it with a "streaming-compatible" attribute to ensure it can be called
from both streaming and non-streaming execution modes.
As part of this move to a common header we also add duplicated
streaming-mode implementations of the following kernels that did not
previously have an SME implementation:
- I210AlphaToARGBRow_SME
- I210ToAR30Row_SME
- I210ToARGBRow_SME
- I212ToAR30Row_SME
- I212ToARGBRow_SME
- I400ToARGBRow_SME
- I410AlphaToARGBRow_SME
- I410ToAR30Row_SME
- I410ToARGBRow_SME
- I422AlphaToARGBRow_SME
- I422ToARGB1555Row_SME
- I422ToARGB4444Row_SME
- I422ToRGB24Row_SME
- I422ToRGB565Row_SME
- I422ToRGBARow_SME
- I444AlphaToARGBRow_SME
- NV12ToARGBRow_SME
- NV12ToRGB24Row_SME
- NV21ToARGBRow_SME
- NV21ToRGB24Row_SME
- P210ToAR30Row_SME
- P210ToARGBRow_SME
- P410ToAR30Row_SME
- P410ToARGBRow_SME
- UYVYToARGBRow_SME
- YUY2ToARGBRow_SME
Change-Id: I84583478e465351cbe6fc0ec65254c3009922e84
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6087804
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2024-12-12 03:07:54 -08:00
George Steed
c2e7f8389a
[AArch64] Add SME implementations of InterpolateRow{,_16,_16To8}
...
InterpolateRow_SME and InterpolateRow_16_SME need special cases to
handle if source_y_fraction is 256 since this would overflow a byte and
can just be a call to memcpy instead.
InterpolateRow_16To8_SME is never called with a source_y_fraction value
of 256 so there is no need for a special case here.
Change-Id: I67805b5db2c411acb93ada626cf414b35620f467
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6074375
Reviewed-by: Justin Green <greenjustin@google.com>
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2024-12-12 03:03:41 -08:00
George Steed
2d8652f3e7
[AArch64] Add SME implementation of CopyRow
...
Add a streaming-SVE implementation of CopyRow using normal vector
load/store instructions.
Change-Id: Ia551413f9740a96473fa2e8a0958953be2f4b04e
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6074374
Reviewed-by: Justin Green <greenjustin@google.com>
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2024-12-12 03:02:07 -08:00
George Steed
418b6df0de
[AArch64] Add SME implementation of Convert16To8Row
...
Mostly just a straightforward copy of the Neon code ported to
Streaming-SVE, we can use predication to avoid needing an `Any` kernel.
SVE has a "widening multiply get high half" instruction in UMULH,
however using the same technique as the Neon code to avoid the need for
a widening multiply at all is more performant here.
These is no benefit from this kernel when the SVE vector length is only
128 bits, so skip writing a non-streaming SVE implementation.
Change-Id: Ib12699c5b8b168d004ebc74c0281ea3772ca8d32
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/6070786
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Justin Green <greenjustin@google.com>
2024-12-12 03:01:55 -08:00
