instead of per block. This prepares for a next CL that can make the
predictors alter RGB value behind transparent pixels for denser
encoding. Some predictors depend on the top-right pixel, and it must
have been already processed to know its new RGB value, so requires per
scanline instead of per block.
Running the encode speed test on 1000 PNGs 10 times with default
settings:
Before:
Compression (output/input): 2.3745/3.2667 bpp, Encode rate (raw data): 1.497 MP/s
After:
Compression (output/input): 2.3745/3.2667 bpp, Encode rate (raw data): 1.501 MP/s
Same but with quality 0, method 0 and 30 iterations:
Before:
Compression (output/input): 2.9120/3.2667 bpp, Encode rate (raw data): 36.379 MP/s
After:
Compression (output/input): 2.9120/3.2667 bpp, Encode rate (raw data): 36.462 MP/s
No effect on compressed size, this produces exactly same files. No
significant measured effect on speed. Expected faster speed from better
memory layout with scanline processing but slower speed due to needing
to get predictor mode per pixel, may compensate each other.
Change-Id: I40f766f1c1c19f87b62c1e2a1c4cd7627a2c3334
the problem was the incorporation of the extra constant 1<<16 in the kC1
constant, to emulate the addition. It's now removed and the addition is
performed explicitly.
No real speed difference observed.
cf. issue #278
Change-Id: I2c6499031571d98afff392fb5ebe21a5fa60722d
global effect is ~2% faster encoding from JPG source
and ~8% faster lossless-webp source decoding to PGM (e.g.)
Also revamped the YUVA case to first accumulate R/G/B value into 16b
temporary buffer, and then doing the UV conversion.
-> New function: WebPConvertRGBA32ToUV
Change-Id: I1d7d0c4003aa02966ad33490ce0fcdc7925cf9f5
Just for RGB24/BGR24 for now, which are the hard-to-optimize ones.
SSE2 implementation coming next.
ConvertRowToY() should go into dsp/ too, at some point.
Change-Id: Ibc705ede5cbf674deefd0d9332cd82f618bc2425
also switch to using ExtractAlpha() instead of hard-coding the loop.
The ARGBToY/UV functions are rather easy to port to SSE2 / NEON.
Change-Id: I8f1346a9ca427a36ce2d6c848369ca7964d8b3c7
use 'u' rather than the unnecessary 'l' as a suffix. this prevents a
conversion warning with some toolchains
Change-Id: I21c33ce08819b3c839c75e03a8f7f3a6041d0695
same functionality, but better code layout.
What changed:
* don't trash the palette_[] in EncodePalette(), so it can be re-used
* split generation of image from bit-stream coding
* move all the delta-palette code to delta_palettization.c, and only have 1 entry point there WebPSearchOptimalDeltaPalette()
* minimize the number of "#ifdef WEBP_EXPERIMENTAL_FEATURES" in vp8l.c
* clarify the TransformBuffer stuff. more clean-up to come here...
This should make experimenting with delta-palettization easier and more compartimentalized.
Change-Id: Iadaa90e6c5b9dabc7791aec2530e18c973a94610
some limitations: only for RGBA output,
and if reduction factor is not too small (dst_width > src_width / 128)
20-25% faster, ~4-6% global improvement total decoding.
Change-Id: I95366ddaa4a38e0a96bed754dfe790126f7bb84a
It's better to stay with a 32b fixed-point precision overall, otherwise
the C-version on ARM gets *slower*.
Actually, gcc ARM compiler optimizes some instructions pretty
well when WEBP_RESCALER_FIX is exactly 32, even in C.
Change-Id: I0eea97f7db5947470f5af355dee098eca81e178d
The rounding and arithmetic is not the same as previously, to prevent overflow cases for large upscale factors.
We still rely on 32b x 32b -> 64b multiplies. Raised the fixed-point precision to 32b
so that we have some nice shifts from epi64 to epi32.
Changed rescaler_t type to 'uint32_t' in order to squeeze in all the precision required.
The MIPS code has been disabled because it's now out-of-sync. Will be fixed in
a subsequent CL when the dust settles.
~30-35% faster
Change-Id: I32e4ddc00933f1b1aa3463403086199fd5dad07b
* vertical expansion now uses bilinear interpolation
* heavily assumes that the alpha plane is decoded in full, not row-by-row
* split the RescalerExportRow and RescalerImportRow methods into Shrink
and Expand variants.
* MIPS implementation of ExportRowExpand is missing.
There's room for extra speed optim and code re-org, but let's keep that for later patches.
addresses https://code.google.com/p/webp/issues/detail?id=254
Change-Id: I8f12b855342bf07dd467fe85e4fde5fd814effdb
the x_add/x_sub increments were wrong for u/v in the upscaling case.
They shouldn't be left to the caller's discretion, but set up by
WebPRescalerInit to their exact necessary values.
-> Cleaned-up WebPRescalerInit() param list.
-> added safety asserts
-> removed the mips32/mips_r2 variant of "ImportRow" which were buggy prior
Change-Id: I347c75804d835811e7025de92a0758d7929dfc09
this moves the function outside the WEBP_USE_INTRINSICS check.
there's no alternative version and it's ~54% faster at the
function level and mildly faster overall
Change-Id: Ibc648e9ee35021d48901e05aa596aa01067796a2
share the computation between different modes
3-5 % speedup for lossless alpha
1 % for lossy alpha
no change in compression density
Change-Id: I5e31413b3efcd4319121587da8320ac4f14550b2
After several trials at re-organizing the main loop and accumulation scheme,
this is apparently the faster variant.
removed the SSE41 version, which is no longer faster now.
For some reason, the AVX variant seems to benefit most for the change.
Change-Id: Ib11ee18dbb69596cee1a3a289af8e2b4253de7b5
this moves the function outside the WEBP_USE_INTRINSICS check.
there's no alternative version and it's ~70% faster at the
function level and 1-2% faster overall
Change-Id: I59fb4918ec86b1ac3a47cbd5d05ce62f007461cb
Changed the code (again) to process 4 pixels at a time. Loop is more
involved, but overall it's faster.
Removed the SSE4.1 implementation which is now slower than SSE2.
Change-Id: I7734e371033ad8929ace7f7e1373ba930d9bb5f1
New implementations: SubtractGreenFromBlueAndRed and TransformColor
around 1-2% faster lossless encoding.
Change-Id: I1668e36fdc316ba55b3b798b91b4a3e36ce62861
DispatchAlpha* functions are hard to speed up, compared to SSE2.
ExtractAlpha sees a ~15% speed-up though.
Change-Id: I8715c2defecbc832f469eed7e6ffd012146b52de
over a 1000 image corpus
Single photograph benchmark:
Before:
Q=20: 2.560 MP/s
Q=40: 2.593 MP/s
Q=60: 1.795 MP/s
Q=80: 1.603 MP/s
Q=99: 1.122 MP/s
After:
Q=20: 3.334 MP/s
Q=40: 2.464 MP/s
Q=60: 2.009 MP/s
Q=80: 1.871 MP/s
Q=99: 1.163 MP/s
This CL allows for some further improvements that would not be possible
otherwise.
Change-Id: I61ba154beca2266cb96469281cf96e84a4412586
use vld1_dup_u8() rather than a separate ld+dup after the values were
zero extended; mildly faster at the function level
Change-Id: I1b3666a6aeb465722a1214dbc6d71c27689a7f89
VP8EncPredChroma8 improvements over ~20M pixels
left/top: ~67%
left-only: ~52%
top-only: ~57%
none: ~61%
based on dec_sse2 versions with minor changes to benefit from the linear
storage of the left boundary
Change-Id: Iee7e387fb2570b4eb5af5bfd123e9c2e9ea49c76
VP8EncPredLuma16 improvements over ~20M pixels
left/top: ~75%
left-only: ~47%
top-only: ~59%
none: ~63%
based on dec_sse2 versions with minor changes to benefit from the linear
storage of the left boundary
Change-Id: I7548be7214fa85c38fd11d30f5b8b271f437657d
structured extended feature flags require eax = 7; avoids incorrectly
detecting avx2 on some older processors that support avx.
for completeness also check for value=1 support used by the other
checks.
from [1]:
INPUT EAX = 0: Returns CPUID’s Highest Value for Basic Processor
Information and the Vendor Identification String
[1]
http://www.intel.com/content/www/us/en/processors/processor-identification-cpuid-instruction-note.html
Change-Id: I60b20d661a978d551614dbf7acdc25db19cb6046
use psadbw to perform top row summation; left remains in C as repacking
it into a vector to apply the same operation is too costly.
DC8uv: ~19% faster
DC8uvNoLeft: ~12% faster
Change-Id: I707c4f6177a65b5d1f2d3deeca87d2bb740185e2
use psadbw to perform top row summation; left remains in C as repacking
it into a vector to apply the same operation is too costly.
DC16: ~20% faster
DC16NoLeft: ~14% faster
Change-Id: I7ec3f8a6e5923f88a530f79fceb88d5001bef691
generates a stub function when the specific architecture is not enabled,
exposing a symbol in the module, avoiding a compiler warning
Change-Id: Ia9336e57466a9b5241b85c1c95838e91c9283147
Visible speed-up, thanks to pshufb and pabsw and psignw use.
had to tweak configure.ac to make "smmintri.h" presence correctly
detected (we need to set the CPPFLAGS instead of the CFLAGS!)
Change-Id: I2ab99e16a27a64fdf1f09b2b4e30a5e74ccca080
allows the former to be inlined; negligible speed-up in most cases,
however this is structure is consistent with the rest of the optimized
modules
Change-Id: Ib080240b06f7a995b47f1906627850c355b82901
the standard vtbl functions are available there [1][2].
based on a patch from: aaroncrespo
fixes issue #243.
[1]
http://adcdownload.apple.com//Developer_Tools/Xcode_6.3_beta/Xcode_6.3_beta_Release_Notes.pdf
[2] Apple LLVM Compiler Version 6.1
- Xcode 6.3 updates the Apple LLVM compiler to version 6.1.0.
[...]
Support for the arm64 architecture has been significantly revised to
align with ARM's implementation, where the most visible impact is that a
few of the vector intrinsics have changed to match ARM's specifications.
Change-Id: I79a0016f44b9dbe36d0373f7f00a50ab3c2ca447
The MIPS code for cost is not updated yet, that's why i keep Residual::*cost
around for now. Should be removed in favor of *costs later.
Change-Id: Id1d09a8c37ea8c5b34ad5eb8811d6a3ec6c4d89f
affected functions: SimpleVFilter16, SimpleHFilter16,
SimpleVFilter16i and SimpleHFilter16i
noticed bug in FilterLoop26 (fix included in this patch)
Change-Id: I72d9c1e45cbac6393eba52bb549b04924d463e30
removes circular dependency between dsp and enc.
since:
a987fae MIPS: dspr2: added optimization for function GetResidualCost
Change-Id: Ifeb8fc02de89e2ba982ed7ffacd925d649bfec3c
set/get residual C functions moved to new file in src/dsp
mips32 version of GetResidualCost moved to new file
Change-Id: I7cebb7933a89820ff28c187249a9181f281081d2
similar to:
1ba61b0 enable NEON intrinsics in aarch64 builds
vtbl1_u8 is available everywhere but Xcode-based iOS arm64 builds, use
vtbl1q_u8 there.
performance varies based on the input, 1-3% on encode was observed
Change-Id: Ifec35b37eb856acfcf69ed7f16fa078cd40b7034
The 'inverse' variants are harder to parallelize, since
the result of filtering is used for prediction.
The 'direct' way is relatively easier.
The heavy bottleneck left for optimization is still GradientUnfilter()
Change-Id: I358008f492a887e8fff6600cb27857b18dee86e9
and only use it on x86 / x64 where it's available.
has the side-effect of quieting a msvs /analyze warning:
C6001: Using uninitialized memory 'cpu_info'.
Change-Id: Iae51be3b22b2ee949cfc473eeea9fd9fb6b3c2cb
added new function CollectColorRedTransforms to C, which calls
TransformColorRed and it is realized via pointer to function
Change-Id: Ia68d73bfcf1ca2cb443dc2825910946221f87835
explicitly add immintrin.h instead of transitively picking it up via
windows.h presumably. makes the code easier to move around.
Change-Id: If70d5143ac94fc331da763ce034358858e460e06
added new function CollectColorBlueTransforms to C, which calls
TransformColorBlue and it is realized via pointer to function
Change-Id: Ia488b7a7a689223b5d33aae9724afab89b97fced
we compare the current VP8GetCPUInfo pointer to the last used.
This is less code overall and each implementation is still
testable separately (by just changing VP8GetCPUInfo, but not
a separate threads!)
Change-Id: Ia13fa8ffc4561a884508f6ab71ed0d1b9f1ce59b
inline function MakeARGB32 calls changed to call
via pointers to functions which make (a)rgb for
entire row
Change-Id: Ia4bd4be171a46c1e1821e408b073ff5791c587a9
and apply Paeth predictor (predictor#11) for the low effort (m=0) mode.
For 1000 image PNG corpus (m=0), this change yields speedup of 25% at lower quality
range and about 10% for higher quality range.
Change-Id: I0f036b8ffe45c241e63a067cbf01527b13d8de93
check for __apple_build_version__ to distinguish the two; a version
check could work as Apple bumped Xcode's to 5.x/6.x, but it's unclear
how upstream will deal with their versioning as they go 3.6+, so avoid
it for now.
Change-Id: I67cda67c4f68e262a92d805a63cc1496374be063
we don't need to store the whole distribution in order to compute the alpha
Later, we can incorporate the max_value / last_non_zero bookkeeping
in SSE2 directly.
Change-Id: I748ccea4ac17965d7afcab91845ef01be3aa3e15
this is a first step to unifying encoding/decoding cache stride
and possibly sharing the prediction functions in dsp/
With this layout, there's a little (~7%) space lost with unused samples.
But no speed change was observed.
Change-Id: I016df8cad41bde5088df3579e6ad65d884ee711e
~68% faster
reuses TM4() adding support for the additional rows, the columns were
already being done.
Change-Id: I6eac17e58cd1c636082bf7281f70f884ec399a6b
Move all the Entropy evaluation methods to lossless.c (from histogram.c).
There's slight difference in the way entropy is computed for evaluating
entropy in prediction methods and histogram (literal) for huffman trees.
Plan (later) to merge few (static) methods and reduce the code size.
This change has no impact on the compression speed/density.
Change-Id: Ife3d96a3c4a8d78a91723d9e0a8d1b78c0256a15
set WEBP_EXTERN to visibility=default
+ explicitly mark VP8GetCPUInfo as it's referenced within the examples
Change-Id: Ie3d2b15088e888f0b55203b205993eba75899d99
move the attribute to the front of the function to quiet clang warning:
GCC does not allow no_sanitize_thread attribute in this position on a
function definition
Change-Id: Ie4cc6e35a07bd00eab67d9cd6801bd2be9cfe676
avoids an ICE with NDK r10b + NDK_TOOLCHAIN_VERSION=4.9
In function 'SSE16x16':
enc_mips32.c (684) internal compiler error: Segmentation fault
Change-Id: I1a3d33c0a9534c97633ab93bcdf9bf59d3a7e473
got rid of the |a-b|^|b-a| method and went back
to just (a-b)^2 instead.
quality | size(bytes) after/before | time (ms) after/before
Change-Id: Ia3e0e6507b3f903deb1e182f78dad6df07380fd0
SSE2 version is 2.1x faster
This is used to transfer the alpha plane to green channel before lossless compression.
Change-Id: I01d9df0051c183b1ff5d6eb69961d4f43e33141a
* We were re-doing most of the work in plain-C as 'left-over'.
* we were always returning has_alpha = true because of a bad mask all_0xff
These bugs were conservative and silent, in the sense that we were 'just' doing
more work than necessary.
Now, the SSE2 version is really 2x faster than the C version.
Change-Id: I6c8132a267fe3c7a3d1fa70e7a5fcd10719543fa
Split initialization of YUV444Converters[] out of Upsamplers init.
update test for NULL function pointers
Change-Id: I9603f54250f90c85a12ffbecfd6c59e9b06c47e0
vtbl4_u8 is available everywhere except iOS arm64: use vtbl2q_u8 there
with a corresponding change in the load.
Change-Id: Ib84212dda3c7875348282726c29e3b79b78b0eac
_M_IX86 will be defined in mingw builds after including windows.h. as
the gcc inline asm is first, this missing check would only have caused
an error if the code was reorganized.
Change-Id: I395679bcfc43e94d308d1ceb0c0fbf932b2c378c
this defines WORDS_BIGENDIAN, replacing uses of
__BIG_ENDIAN__/__BYTE_ORDER__ with it
+ fixes lossless BGRA output with big-endian toolchains
that do not define __BIG_ENDIAN__ (codesourcery mips gcc)
Change-Id: Ieaccd623292d235343b5e34b7a720fc251c432d7
this will remove a warning about the shift amount not being
an immediate (=constant).
Change-Id: Ie9a00fefdb9a07ec8994fb113f24234518bc878a
Also: fix the NULL sharpen argument mismatch.
Another store to load forward block was detected coming from the function
FTransform.
FTransform save the output data 4 times 8 bytes each. when this data is
later being loaded by the QuantizeBlock function in one chunk of 16 bytes
that caused a store to load forward block.
The fix was done in the FTransform function where each two consecutive 8 bytes
were merged into one 16 bytes register and saved into the memory.
This fix gives ~21% function level gain and 1.6% user level gain.
Change-Id: Idc27c307d5083f3ebe206d3ca19059e5bd465992
only 1 of <lib>_CPPFLAGS and AM_CPPFLAGS is used, with the former
getting precedence when it's defined. configure's DEFAULT_INCLUDES is
covering what's necessary given the include paths are all source
relative.
Change-Id: I7d14076acd266b28a88a3d92bcc3d7165284d5f3
this change has the side-effect of using directory names in the
include, silencing a lint warning.
Change-Id: Ib91cf63a90534e32fadfa5c2372bfdb29f854d02
if res->first = 1, coeffs[0]=0 because of quant.c:749 and line
added at quant.c:744
So, no need for the extra case.
Going forward, TrellisQuantizeBlock() should also be calling
a variant of VP8SetResidualCoeffs() to set the 'last' field.
also: fixes a warning for win64
+ slight speed-up
Change-Id: Ib24b611f7396d24aeb5b56dc74d5c39160f048f0
+ add a WEBP_HAVE_SSE2 to dsp.h
not all 32-bit toolchain configurations will have sse2 enabled by
default
Change-Id: I7c675e511581f93cf55c79f960fa7efa2df4987e
this is used to set WEBP_USE_AVX2 in files where the build flag won't be
used, i.e., dsp/enc.c, which enables VP8EncDspInitAVX2() to be called
Change-Id: I362f4ba39ca40d3e07a081292d5f743c649d9d7f
Extract loop invariant and avoid storing/loading samples
if they can be re-used. This is particularly interesting when
a transpose is involved (HFilter16i).
Change-Id: I93274620f6da220a35025ff8708ff0c9ee8c4139
The luminance needs to be pre- and post- multiplied by
the alpha value in case of rescaling, for proper averaging.
Also:
- removed util/alpha_processing and moved it to dsp/
- removed WebPInitPremultiply() which was mostly useless
and merged it with the new function WebPInitAlphaProcessing()
Change-Id: If089cefd4ec53f6880a791c476fb1c7f7c5a8e60
VP8EncDspInitAVX2 is included in sse2 builds for now, later a configure
flag should be added to avoid the stub when avx2 is unavailable/disabled
Change-Id: I6127b687c273f46f41652aaf8e3b86ae3cfb8108
* remove some sign-bit flipping
* turn some macro into inline functions
* fix some 'const' in signatures
* clarify the int8/uint8 usage
Change-Id: Ib04459ac34cb280c57579c5d79a5efd2f8d5e99d
The predictors based on Average2 are tad slower.
Following is the performance data for these predictors normalized to
number of instruction cycles (as per valgrind) per operation:
- Predictor6 & Predictor7 now takes 15 instruction cycles compared to 11
instruction cycles for the C version.
- Predictor8 & Predictor9 now takes 15 instruction cycles compared to 12
instruction cycles for the C version.
The predictors based on Average4 is faster and Average3 is tad slower:
- Predictor10 (Average4) now takes 23 instruction cycles compared to 25
instruction cycles for the C version.
- Predictor5 (Average3) now takes 20 instruction cycles compared to 18
instruction cycles for the C version.
Maybe SSE2 version of Average2 can be improved further. Otherwise, we can
remove the SSE2 version and always fallback to the C version.
Change-Id: I388b2871919985bc28faaad37c1d4beeb20ba029
* merged the two HistogramAdd/AddEval() into a single call
(with detection of special case when b==out)
* added a SSE2 variant
* harmonize the histogram type to 'uint32_t' instead
of just 'int'. This has a lot of ripples on signatures.
* 1-2% faster
Change-Id: I10299ff300f36cdbca5a560df1ae4d4df149d306
move simple loop filter defines closer to their use and LOAD* to a
location common with the intrinsics
Change-Id: Iaec506d27bbc9a01be20936e30b68a4b0e690ee3
the complex loop filter has no inline equivalent; the simple loop filter
remains conditional on USE_INTRINSICS: it's left undefined for now.
Change-Id: I4f258e10458df53a7a1819707c8f46b450e9d9d2
CollectHistogram / SSE* / QuantizeBlock have no inline equivalents,
enable them where possible and use USE_INTRINSICS to control borderline
cases: it's left undefined for now.
Change-Id: I62235bc4ddb8aa0769d1ce18a90e0d7da1e18155
using this in Load4x16 was slightly slower and didn't help mitigate any
of the remaining build issues with 4.6.x.
Change-Id: Idabfe1b528842a514d14a85f4cefeb90abe08e51
HuffmanCost and HuffmanCostCombined optimized and added
'const' to some variables from ExtraCost functions.
Change-Id: I28b2b357a06766bee78bdab294b5fc8c05ac120d
Some versions of compiler in debug build can't find
a register in class 'GR_REGS' while reloading 'asm'
Number of used registers is decreased in this fix.
Change-Id: I7d7b8172b8f37f1de4db3d8534a346d7a72c5065
This is to help further optimizations.
(like in https://gerrit.chromium.org/gerrit/#/c/69787/)
There's a small slowdown (~0.5% at -z 9 quality) due to
function pointer usage. Note that, for speed, it's important
to return VP8LStreaks by value, and not pass a pointer.
Change-Id: Id4167366765fb7fc5dff89c1fd75dee456737000
avoids:
src/dsp/enc_mips32.c: In function 'ITransformOne':
src/dsp/enc_mips32.c:123:3: can't find a register in class 'GR_REGS' while reloading 'asm'
src/dsp/enc_mips32.c:123:3: 'asm' operand has impossible constraints
Change-Id: Ic469667ee572f25e502c9873c913643cf7bbe89d
apparently faster, but we might save some load/store to/from memory
once we settle for the intrinsics-based FTransform()
(also: fixed some #ifdef USE_INTRINSICS problems)
Change-Id: I426dea299cea0c64eb21c4d81a04a960e0c263c7
Functions VP8LFastLog2Slow and VP8LFastSLog2Slow
also: replaced some "% y" by "& (y-1)" in the C-version
(since y is a power-of-two)
Change-Id: I875170384e3c333812ca42d6ce7278aecabd60f0
Verified OK, but right now they don't seem faster.
So they are disabled behind a USE_INTRINSICS flag (off for now)
Change-Id: I72a1c4fa3798f98c1e034f7ca781914c36d3392c
+ reorganize the cost-evaluation code by moving some functions
to cost.h/cost.c and exposing VP8Residual
Change-Id: Id976299b5d4484e65da8bed31b3d2eb9cb4c1f7d
slightly faster than the inline asm
in practice not much faster than the C-code in a full NEON build, but
still better overall in an Android-like one that only enables NEON for
certain files.
Change-Id: I69534016186064fd92476d5eabc0f53462d53146
* inverse transform is actually slower with intrinsics + gcc-4.6,
so is left disabled for now.
With gcc-4.8, it's a bit faster than inlined assembly.
* Sum of Square error function provide a 2-3% speed up
There's enabled by default (since there's no inlined-asm equivalent)
Change-Id: I361b3f0497bc935da4cf5b35e330e379e71f498a
+ misc cosmetics
* seems 4% slower than inlined-asm with gcc-4.6
* is a tad faster (<1%) with gcc-4.8
(disabled for now)
Change-Id: Iea6cd00053a2e9c1b1ccfdad1378be26584f1095
The nice trick is to pack 8 u + 8 v samples into a single uint8x16x_t
register, and re-use the previous (luma) functions
Change-Id: Idf50ed2d6b7137ea080d603062bc9e0c66d79f38
+ added some work-around gcc-4.6 to make it compile (except one function).
+ lots of revamping
All variants tested ok.
Speed-up is ~5-7%
Change-Id: I5ceda2ee5debfada090907fe3696889eb66269c3
vertical only currently, 2.5-3% faster
placed under USE_INTRINSICS as this change depends on the simple
loopfilter
improves the simple loopfilter slightly thanks to some reorganization
Change-Id: I6611441fa54228549b21ea74c013cb78d53c7155
When 4 pixels are left, they should be processed with SSE2.
Decoding is marginally faster (~0.4%).
Encoding speed: No observable difference.
Change-Id: I3cf21c07145a560ff795451e65e64faf148d5c3e
new file: lossless_neon.c
speedup is ~5%
gcc 4.6.3 seems to be doing some sub-optimal things here,
storing register on stack using 'vstmia' and such.
Looks similar to gcc.gnu.org/bugzilla/show_bug.cgi?id=51509
I've tried adding -fno-split-wide-types and it does help
the generated assembly. But the overall speed gets worse with
this flag. We should only compile lossless_neon.c with it -> urk.
Change-Id: I2ccc0929f5ef9dfb0105960e65c0b79b5f18d3b0
It's disable for now, because it crashes gcc-4.6.3 during compilation
with -O2 or -O3. It's been tested OK with -O1.
Code is still globally disabled with USE_INTRINSICS, though.
Change-Id: I3ca6cf83f3b9545ad8909556f700758b3cefa61c
disabled for now (but tested OK), thanks to the USE_INTRINSICS #define
We'll activate the code when we're on par with non-intrinsics
Change-Id: Idbfb9cb01f4c7c9f5131b270f8c11b70d0d485ff
expose the predictor array as function pointers instead
of each individual sub-function
+ merged Average2() into ClampedAddSubtractHalf directly
+ unified the signature as "VP8LProcessBlueAndRedFunc"
no speed diff observed
Change-Id: Ic3c45dff11884a8330a9ad38c2c8e82491c6e044
Get back some of the compression gains by extending the search space for
GetBestGreenRedToBlue. Also removed the SkipRepeatedPixels call, as it was not
helping much in yielding better compression density.
Before:
1000 files, 63530337 pixels, 1 loops => 45.0s (45.0 ms/file/iterations)
Compression (output/input): 2.463/3.268 bpp, Encode rate (raw data): 1.347 MP/s
After:
1000 files, 63530337 pixels, 1 loops => 45.9s (45.9 ms/file/iterations)
Compression (output/input): 2.461/3.268 bpp, Encode rate (raw data): 1.321 MP/s
Change-Id: I044ba9d3f5bec088305e94a7c40c053ca237fd9d
Restructure PredictorInverseTransform & ColorSpaceInverseTransform to remove
one if condition inside the main/critial loop. Also separated TransformColor &
TransformColorInverse into separate functions and avoid one 'if condition'
inside this critical method.
This change speeds up lossless decoding for Lenna image about 5% and 1000 image
corpus by 3-4%.
Change-Id: I4bd390ffa4d3bcf70ca37ef2ff2e81bedbba197d
