Age | Commit message (Collapse) | Author |
|
Seen with arm-linux-gnueabihf-gcc-8 (8.3.0 & 8.4.0)
Without reworking the code or adding an additional branch this warning
cannot be silenced otherwise. The loopfilter is only called when needed
for a block so these output pixels will be set.
BUG=b/176822719
Change-Id: I9cf6e59bd5de901e168867ccbe021d28d0c04933
|
|
Fix two bugs reported by clang when enable msa optimizatons:
1. clang dose not support uld instruction.
2. ulw instruction will result in unit cases coredump.
Change-Id: I171bed11d18b58252cbc8853428c039e2549cb95
|
|
1. Adjust variable type to match clang compiler.
Clang is more strict on the type of asm operands, float or double
type variable should use constraint 'f', integer variable should
use constraint 'r'.
2. Fix prob of using r-value in output operands.
clang report error: 'invalid use of a cast in a inline asm context
requiring an l-value: remove the cast or build with -fheinous-gnu-extensions'.
Change-Id: Iae9e08f55f249059066c391534013e320812463e
|
|
1.'xor,or,and' to 'pxor,por,pand'. In the case of operating FPR,
gcc supports both of them, clang only supports the second type.
2.'dsrl,srl' to 'ssrld,ssrlw'. In the case of operating FPR, gcc
supports both of them, clang only supports the second type.
Change-Id: I93b47348e7c6580d99f57dc11165b4645236533c
|
|
Define LIBVPX_{ELF,MACHO} to simplify blocks.
Create new globalsym macro and include logic for PRIVATE.
BUG=webm:1679
Change-Id: I303ba1492a2813f685de51155ccef7e4831e1881
|
|
|
|
Change-Id: Idae90838012c78605f20f1d7a3125b71683f6f44
|
|
Chromium needs :function hidden and the space between the symbol and the
colon removed, at least for nasm. This matches x86inc.asm.
BUG=webm:1679
Change-Id: Ie47bb75d44d3130791639cbf4e2ebe019e2d686e
|
|
BUG=b/148271109
Change-Id: I620e26ff1233fcd34ebe0723cb913e82eb58271c
|
|
this allows calls to use better versions (e.g., avx2) if available. in
most other cases the function pointer will be defined to the sse2
variant if another isn't available. this improves performance at 1080P
by ~2% on a Xeon E5-2690.
Change-Id: Ie9da3a567021f8416651a29b8c9ab9238dc4bdf1
|
|
This helps to improve some benchmarks by 10%, e.g. decode_time
PCFullStackTest.VP9SVC_3SL_Low
Bug: 1020220, 977230
Change-Id: Ic992f1eec369f46a08e19eb33bc3a7c15c1e7c87
|
|
Change-Id: Ia2be91a49dfa95906fa2ce232ff9d3a69deda4ad
|
|
|
|
Unit Test: VP9/AqSegmentTest. VP9/CpuSpeedTest, AVX2/Loop8Test6Param
implicit conversion from type 'int' of value 59741 (32-bit, signed) to
type 'int16_t' (aka 'short') changed the value to -5795 (16-bit, signed)
BUG=webm:1615
Change-Id: I2e5b688a97c3caa29d4b8a817b95a4986b81a562
|
|
this fixes a segfault when scaling is enabled; in some cases depending
on the ratio offsets may become odd.
vpx_int_pro_row_sse2 was updated previously, though the reason wasn't
listed:
54eda13f8 Apply fast motion search to golden reference frame
BUG=webm:1600
Change-Id: I8d5e105d876d8cf917919da301fce362adffab95
|
|
|
|
with vp9-highbitdepth off.
Unit Test: SSE2/Trans16x16DCT , VP9/LevelTest.TestTargetLevel20Large, VP9/CpuSpeedTest
implicit conversion from type 'int32_t' (aka 'int') of value -32851
(32-bit, signed) to type 'tran_low_t' (aka 'short') changed the value to
32685 (16-bit, signed)
BUG=webm:1615
BUG=webm:1647
Change-Id: I9ef064dc9ac734379628565ff6505b0876984123
|
|
From unit test: AVX/VP9QuantizeTest; SSSE3/VP9QuantizeTest ...
implicit conversion from type 'int' of value -139812 (32-bit, signed)
to type 'tran_low_t' (aka 'short') changed the value to -8740 (16-bit,
signed)
BUG=webm:1615
Change-Id: I730946ac6c7a250dcbcfd8a2712c0f1150ddb4fd
|
|
this prevents redefinition warnings if a toolchain sets one
BUG=b/117240165
Change-Id: Ib5d8c303cd05b4dbcc8d42c71ecfcba8f6d7b90c
|
|
implicit conversion from type 'int' of value 49161 (32-bit, signed) to
type 'int16_t' (aka 'short') changed the value to -16375 (16-bit,
signed)
BUG=webm:1615
Change-Id: I3f18283609ac2ce365202a63ef61a47eb00c155b
|
|
implicit conversion from type 'int' of value -139 (32-bit, signed)
to type 'int8_t' (aka 'signed char') changed the value to 117 (8-bit, signed)
BUG=webm:1615
Change-Id: Ic64959759f4a188087aa24bedbae5f9fa60674ad
|
|
implicit conversion from type 'int' of value 32768 (32-bit, signed)
to type 'short' changed the value to -32768 (16-bit, signed)
BUG=webm:1615
Change-Id: I7cdba7f7e550f62fd3ac31574e49b1909b6ab054
|
|
implicit conversion from type 'unsigned int' of value 256 (32-bit, unsigned)
to type 'uint8_t' (aka 'unsigned char') changed the value to 0 (8-bit, unsigned)
BUG=webm:1615
Change-Id: Ia9ac3772021ae492368c650a73846e7d22c8fdfc
|
|
|
|
Fix comment typos in transpose_s16_4x4q() and transpose_u16_4x4q().
Change-Id: I21bcc1fb3fb880798e5a3927c3dbe81dd518c83b
|
|
Change-Id: I4dbb7b6c8979c39eb6ffb97750e3cca0f4b7921f
|
|
This speed up non_greedy_mv by 4%
Change-Id: I9288c88db56ea4201a7ec4493ca5c567d76af0f1
|
|
These functions cause test failures when running the entire suite.
BUG=webm:1522
Change-Id: I2c1dc4923e9f149464f365ef63dc59621cfabf5a
|
|
There are no sse functions which use these files. Cleans up spurious
warnings when building with --disable-sse2
Change-Id: I04d84b8b7ecfe6da7d5d4df63840796c7b04c085
|
|
|
|
Change-Id: Ifab64a783c205cc79b841a3f77fb77b156b23b23
|
|
Based on the spatial complexity, increase the
bits allocated to key frame.
Change-Id: I4f96990a13bcc3bdb7a22d50e67e2bd622f1ff7b
|
|
Add a macro to to exclude VP9 specific assembly files from build if VP9
is not configured. This would otherwise cause a linking error for VP8
only builds.
BUG=webm:1625
Change-Id: I6d892b7c2837a2574538d18b776fd2b6d706da96
|
|
clang 7 integer sanitizer warns on unsigned->signed conversions when
the highest bit is 1.
BUG=webm:1615
Change-Id: I6381efaff9233254b40cb78f7bcf87090e0ad353
|
|
Change-Id: I339899cff65c7ef563f9411f2d7af9a32a08a705
|
|
Cherry-pick libaom #1362e15990eccab8101305be418528824fb32245 to fix
PSNRHVS calculation.
BUG=webm:1620
Change-Id: Ife7bd8056fcaed06992ad76e8c1ab734c3956ad7
|
|
|
|
added files that are affected by clang-format version 7.
BUG=b/120815481
Change-Id: I40662ce962e4f4b1fcdf183b700f85cc5c0f9f82
|
|
This reverts commit aa04b6f9a7475e9d9457dfc5bf441faf15efc466.
It caused big regression on webrtc VP8 tests.
Change-Id: I937e769d133abeca62ba063e59a58b5c461f5b5e
|
|
VP8 and VP9 have different padding on buffer stride.
VP8 microblock is 16x16 so the buffer stride needs to be divisible by
16. Thus UV buffer stride is divisible by 8.
VP9 microblock is 8x8 so the buffer stride is only extended to be
divisible by 8. Then UV buffer stride isn't divisible by 8.
Change-Id: I6fa953feb951f2fb2e48f72a623786b85e23822f
|
|
Change-Id: Id654a48d2fa40355552d7267e58461e6cc1c6998
|
|
BUG=webm:1606
Change-Id: I661485b860243c95b6450035dbac77b0dd4d9ff4
|
|
BUG=webm:1270
Change-Id: I7d56667d946196bbbe355303de805422e40b0763
|
|
Change-Id: Ib8ca37f1b58e9903e7efa29689a0a49f14b4d73a
|
|
Exclude low bit depth optimizations from high bit depth builds.
BUG=webm:1584
Change-Id: I86a7ebafa557d262257358e1e055a06d52659977
|
|
|
|
BUG=webm:1584
Change-Id: Ifdebf33356abcc6869f695d129165ba17e042dcd
|
|
BUG=webm:1584
Change-Id: I4cbfafe8ea72b3d4523aabcaed4848fa29bb19fe
|
|
Reduces the number of rows calculated for 2D 4-tap interpolation filter
from h+7 rows to h+3 rows.
Also fixes a bug in the avx2 function for 4-tap filters where the last
row is computed incorrectly.
Performance:
| Baseline | Result | Pct Gain |
bitdepth lo| 4.00 fps | 4.02 fps | 0.5% |
bitdepth 10| 1.90 fps | 1.91 fps | 0.5% |
The performance is evaluated on speed 1 on jets.y4m br 500 over 100
frames.
No BDBR loss is observed.
Change-Id: I90b0d4d697319b7bba599f03c5dc01abd85d13b1
|
|
BUG=webm:1584
Change-Id: I596f5f0e1a1c152493cd8177b32d416cc79937e0
|