Make SkAutoTMalloc's interface look more like SkAutoMalloc:
- add free(), which does what you expect
- make reset() return a pointer fPtr
No public API changes (SkAutoTMalloc is in include/private)
BUG=skia:2148
Review URL: https://codereview.chromium.org/1516833003
- Carmack rsqrt uses an int where it wants a uint32_t.
- turn off all santizers (including signed-integer-overflow) in third_party/externals/sftntly.
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Debug-ASAN-Trybot
BUG=skia:4635
Review URL: https://codereview.chromium.org/1511643002
This tries to cut things down to the very minimum you might want to know:
$ out/Release/nanobench --match nytimes --config 8888 gpu -q
Timer overhead: 29.6ns
! -> high variance, ? -> moderate variance
micros bench
2479.05 ! desk_nytimes.skp_1_mpd 8888
1313.92 desk_nytimes.skp_1_mpd gpu
3617.65 desk_nytimes.skp_1 8888
1158.34 desk_nytimes.skp_1 gpu
1368.99 ! keymobi_nytimes_com_.skp_1_mpd 8888
393.40 keymobi_nytimes_com_.skp_1_mpd gpu
1179.68 ! keymobi_nytimes_com_.skp_1 8888
342.74 keymobi_nytimes_com_.skp_1 gpu
All times are printed in microseconds, and high variance runs are marked.
BUG=skia:
Review URL: https://codereview.chromium.org/1493313003
Reason for revert:
BUG=skia:4609
skbug.com/4609
Seems like GrContextFactory needs to fail when the NVPR option is requested but the driver version isn't sufficiently high.
Original issue's description:
> Make NVPR a GL context option instead of a GL context
>
> Make NVPR a GL context option instead of a GL context.
> This may enable NVPR to be run with command buffer
> interface.
>
> No functionality change in DM or nanobench. NVPR can
> only be run with normal GL APIs.
>
> BUG=skia:2992
>
> Committed: https://skia.googlesource.com/skia/+/eeebdb538d476c1bfc8b63a946094ca1b505ecd1TBR=mtklein@google.com,jvanverth@google.com,kkinnunen@nvidia.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:2992
Review URL: https://codereview.chromium.org/1486153002
eeebdb538d did not update a Config that
only builds if SK_BUILD_FOR_ANDROID_FRAMEWORK. This appears to be the
necessary parameter that is missing.
BUG=skia:2992
Review URL: https://codereview.chromium.org/1491563003
Make NVPR a GL context option instead of a GL context.
This may enable NVPR to be run with command buffer
interface.
No functionality change in DM or nanobench. NVPR can
only be run with normal GL APIs.
BUG=skia:2992
Review URL: https://codereview.chromium.org/1448883002
Skia'a nanobench can hold onto an image, that causes the lifetime
of GrGLGpu to extend past that of the GL interface.
Kill reference to surface and image before killing the interface
Review URL: https://codereview.chromium.org/1472433002
Remove GrContextFactory::getGLContext, it is problematic:
It has the bug of not checking for the context type.
It also is error-prone, since the GL context is not made
current, but it is callers may assume it is current.
It is also not used very much.
Clients can use GrContextFactory::getContextInfo.
BUG=skia:
Review URL: https://codereview.chromium.org/1455093003
SkChecksum::Compute is a very, very poorly distributed hash function.
This replaces all remaining uses with Murmur3.
The only interesting stuff is in src/gpu.
BUG=skia:
Review URL: https://codereview.chromium.org/1436973003
Rename SkCodecTools.h to SkBitmapRegionDecoderPriv.h
Move BRD code to its own directory in tools. This
allows us to not need to expose the entire tools
directory in Android.
BUG=skia:
Review URL: https://codereview.chromium.org/1417393004
We no longer need to worry about namespace
conflicts SkBitmapRegionDecoder in Android (which
we are replacing).
Additionally, the static Create() function does not
need to repeat the name BitmapRegionDecoder.
BUG=skia:
Review URL: https://codereview.chromium.org/1415243007
In SkSampledCodec, allow the native codec to do its scaling first, then
sample on top of that. Since the only codec which can do native scaling
is JPEG, and we know what it can do, hard-code for JPEG. Check to see
if the sampleSize is something JPEG supports, or a multiple of
something it supports. If so, use JPEG directly or combine them.
BUG=skia:4320
Review URL: https://codereview.chromium.org/1417583009
These are benches similar to the imagefilterscropexpand GM: an
input filter is cropped to a smaller size, then the blur is re-expanded
out to a larger size.
BUG=skbug:4502
Review URL: https://codereview.chromium.org/1412373004
Recent changes to WallTimer broke --samplingTime. In particular, this idiom became nonsensical:
WallTimer timer;
timer.start();
do {
...
timer.end();
} while(timer.fWall < ...);
WallTimer started making private use of fWall between when start() and end() were called, so the second time around the loop we end up with nonsense.
If that makes no sense, don't worry. The code here using now_ms() is just as fast, just as precise, and clearer.
I took the opportunity to simplify --samplingTime <complicated string parsing> to --ms <int>, and to simplify the code that depends on it.
BUG=skia:
Review URL: https://codereview.chromium.org/1419103004
The result SkBitmap, the pixel allocator, and the alpha
preference need to be communicated from the client to
the region decoder.
BUG=skia:
Review URL: https://codereview.chromium.org/1418093006
This change updates a small subset of benchmarks to flush the GrContext
between draw loops (specifically SKP benchmarks, SampleApp, and the
warmup in visualbench). This helps improve timing accuracy by not
allowing the gpu to batch across draw boundaries in the affected
benchmarks.
BUG=skia:
Review URL: https://codereview.chromium.org/1427533002
We've migrated SkHwuiRenderer into the Android Framework as
android::uirenderer::TestWindowContext in response to an internal
bug; we now delete that class and change our build references here.
R=djsollen@google.com
Review URL: https://codereview.chromium.org/1407053009
Remove a call to canFilterImageGPU() / filterImageGPU() from
filterInputGPU(). There's no reason to do this, since
the subsequent filterImage() call will do it for us anyway.
And this call actually defeats caching (as demonstrated by
the attached bench).
BUG=skia:
Review URL: https://codereview.chromium.org/1411013004
I don't really expect this to fix the errors, but I think
it's worth it to try shaking up the valgrind bot overnight.
There's some strange behavior with regard to color type on
the valgrind bot that I can't reproduce and that we aren't
seeing on any of the other bots.
TBR=mtklein,scroggo
BUG=skia:
Review URL: https://codereview.chromium.org/1418723002
- Remove --images '' to renable image benchmarking
- Add a flag to disable testing JPEG's buildTileIndex, since it also leaks memory
- Do not run images on GPU
- Do not run large interlaced images on 32 bit bots
- When buildTileIndex is not being used in the subset benches, do not use it for BRD
BUG=skia:3418
BUG=skia:4469
BUG=skia:4471
BUG=skia:4360
Review URL: https://codereview.chromium.org/1396113002
This CL allows the SkScanlineDecoder to decode partial
scanlines.
This is a first step in efficiently implementing subsetting
in SkScaledCodec.
BUG=skia:4209
Review URL: https://codereview.chromium.org/1390213002
SubsetTranslateBench.cpp:
Unref the color table, so it gets deleted.
SkBitmapRegionDecoderInterface.cpp:
Delete the stream if it is not used.
BUG=skia:3418
Review URL: https://codereview.chromium.org/1396113003
Rather than implementing some sort of "fill" in every
SkCodec subclass for incomplete images, let's make the
parent class handle this situation.
This includes an API change to SkCodec.h
SkCodec::getScanlines() now returns the number of lines it
read successfully, rather than an SkCodec::Result enum.
getScanlines() most often fails on an incomplete input, in
which case it is useful to know how many lines were
successfully decoded - this provides more information than
kIncomplete vs kSuccess. We do lose information when the
API is used improperly, as we are no longer able to return
kInvalidParameter or kScanlineNotStarted.
Known Issues:
Does not work for incomplete fFrameIsSubset gifs.
Does not work for incomplete icos.
BUG=skia:
Review URL: https://codereview.chromium.org/1332053002
Instead of decoding one line at a time, if the ScanlineOrder is kNone,
decode all of the lines in one pass, and then copy the subset into the
output. This will allow us to more realistically test subset decodes
for interlaced png. It also makes running them not take forever.
Do *not* support other modes (besides kTopDown), since they are not
used by the big three we need to replace BitmapRegionDecoder
implementation (skbug.com/4428).
Fix a bug in SubsetTranslateBench and SubsetZoomBench:
When we decode another subset, we need to reset the scanline decode
first. This bug appears to have been present since the introduction of
these tests in crrev.com/1160953002
BUG=skia:4205
BUG=skia:3418
Review URL: https://codereview.chromium.org/1387233002
Reason for revert:
Breaks Chrome with this link error: ../../third_party/skia/include/effects/SkMorphologyImageFilter.h:75: error: undefined reference to 'SkMorphologyImageFilter::SkMorphologyImageFilter(int, int, SkImageFilter*, SkImageFilter::CropRect const*)'
../../third_party/skia/include/effects/SkMorphologyImageFilter.h:104: error: undefined reference to 'SkMorphologyImageFilter::SkMorphologyImageFilter(int, int, SkImageFilter*, SkImageFilter::CropRect const*)'
Presumably due to code in third_party/WebKit/Source/platform/graphics/filters/FEMorphology.cpp that contains:
#include "SkMorphologyImageFilter.h"
...
if (m_type == FEMORPHOLOGY_OPERATOR_DILATE)
return adoptRef(SkDilateImageFilter::Create(radiusX, radiusY, input.get(), &rect));
return adoptRef(SkErodeImageFilter::Create(radiusX, radiusY, input.get(), &rect));
Original issue's description:
> factories should return baseclass, allowing the impl to specialize
>
> waiting on https://codereview.chromium.org/1386163002/# to land
>
> BUG=skia:4424
>
> Committed: https://skia.googlesource.com/skia/+/80a6dcaa1b757826ed7414f64b035d512d9ccbf8TBR=senorblanco@google.com,robertphillips@google.com,reed@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:4424
Review URL: https://codereview.chromium.org/1389063002
GLBenches do not expect gl state to change between onPerCanvasPreDraw and *PostDraw, but we do a clear and sometimes we clear as draw. This causes us to bind vertex objects / programs / etc.
This change creates two new virtual methods which are called right before and immediately after timing.
BUG=skia:
Review URL: https://codereview.chromium.org/1379853003
Benefits:
- This mimics other decoding APIs (including the ones SkCodec relies
on, e.g. a png_struct, which can be used to decode an entire image or
one line at a time).
- It allows a client to ask us to do what we can do efficiently - i.e.
start from encoded data and either decode the whole thing or scanlines.
- It removes the duplicate methods which appeared in both SkCodec and
SkScanlineDecoder (some of which, e.g. in SkJpegScanlineDecoder, just
call fCodec->sameMethod()).
- It simplifies moving more checks into the base class (e.g. the
examples in skbug.com/4284).
BUG=skia:4175
BUG=skia:4284
=====================================================================
SkScanlineDecoder.h/.cpp:
Removed.
SkCodec.h/.cpp:
Add methods, enums, and variables which were previously in
SkScanlineDecoder.
Default fCurrScanline to -1, as a sentinel that start has not been
called.
General changes:
Convert SkScanlineDecoders to SkCodecs.
General changes in SkCodec subclasses:
Merge SkScanlineDecoder implementation into SkCodec. Most (all?) owned
an SkCodec, so they now call this-> instead of fCodec->.
SkBmpCodec.h/.cpp:
Replace the unused rowOrder method with an override for
onGetScanlineOrder.
Make getDstRow const, since it is called by onGetY, which is const.
SkCodec_libpng.h/.cpp:
Make SkPngCodec an abstract class, with two subclasses which handle
scanline decoding separately (they share code for decoding the entire
image). Reimplement onReallyHasAlpha so that it can return the most
recent result (e.g. after a scanline decode which only decoded part
of the image) or a better answer (e.g. if the whole image is known to
be opaque).
Compute fNumberPasses early, so we know which subclass to instantiate.
Make SkPngInterlaceScanlineDecoder use the base class' fCurrScanline
rather than a separate variable.
CodexTest.cpp:
Add tests for the state changes in SkCodec (need to call start before
decoding scanlines; calling getPixels means that start will need to
be called again before decoding more scanlines).
Add a test which decodes in stripes, currently only used for an
interlaced PNG.
TODO: Add tests for onReallyHasAlpha.
Review URL: https://codereview.chromium.org/1365313002
To avoid breaking existing SKPs, add a deserialization stub which
unflattens SkBitmapSource records to SkImageSources.
R=reed@google.com,mtklein@google.com,robertphillips@google.com
Review URL: https://codereview.chromium.org/1363913002
SkBitmapRegionDecoderInterface provides an interface
for multiple implementations of Android's
BitmapRegionDecoder.
We already have correctness tests in DM that will enable us
to compare the quality of our various BRD implementations.
We also need these performance tests to compare the speed
of our various implementations.
BUG=skia:4357
Review URL: https://codereview.chromium.org/1344993003
The command buffer's GetShaderiv and GetProgramiv code checks
that the success value passed in is either -1 or 0.
Review URL: https://codereview.chromium.org/1318143004
This switches over SkXfermodes_opts.h and SkColorMatrixFilter to use Sk4f,
and converts the SkPMFloat benches to Sk4f benches.
No pixels should change here, and no code beyond the Sk4f_ benches should change speed.
The benches are faster than the old versions.
BUG=skia:4117
Review URL: https://codereview.chromium.org/1324743002
Unfortunately, immintrin.h (which is also included by SkTypes)
includes xmmintrin.h which includes mm_malloc.h which includes
stdlib.h for malloc even though, from the implementation, it is
difficult to see why.
Fortunately, arm_neon.h does not seem to be involved in such
shenanigans, so building for Android will keep things sane.
TBR=reed@google.com
Doesn't change Skia API, just moves an include.
Review URL: https://codereview.chromium.org/1313203003
Prior to this CL, if a client wanted to decode scanlines, they had to
create an SkCodec in order to get an SkScanlineDecoder. This introduces
complications if input data is not easily shared between the two
objects.
Instead, add methods to SkScanlineDecoder for creating a new one from
input data, and remove the creation functions from SkCodec.
Update DM and tests.
Review URL: https://codereview.chromium.org/1267583002
(and a couple presubmit fixes)
This allows us to turn back on -Werror for LLVM coverage builds,
and more generally supports building with Clang 3.7.
No public API changes.
TBR=reed@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/1232463006
Make getScanlineDecoder return a new object each time, which is
owned by the caller, and independent from any existing scanline
decoders and the SkCodec itself.
Since the SkCodec already contains the entire state machine, and it
is used by the scanline decoders, simply create a new SkCodec which
is now owned by the scanline decoder.
Move code that cleans up after using a scanline decoder into its
destructor
One side effect is that creating the first scanline decoder requires
a duplication of the stream and re-reading the header. (With some
more complexity/changes, we could pass the state machine to the
scanline decoder and make the SkCodec recreate its own state machine
instead.) The typical client of the scanline decoder (region decoder)
uses an SkMemoryStream, so the duplication is cheap, although we
should consider the extra time to reread the header/recreate the state
machine. (If/when we use the scanline decoder for other purposes,
where the stream may not be cheaply duplicated, we should consider
passing the state machine.)
One (intended) result of this change is that a client can create a
new scanline decoder in a new thread, and decode different pieces of
the image simultaneously.
In SkPngCodec::decodePalette, use fBitDepth rather than a parameter.
Review URL: https://codereview.chromium.org/1230033004
SkImageGenerator makes some assumptions that are not necessarily valid
for SkCodec. For example, SkCodec does not assume that it can always be
rewound.
We also have an ongoing question of what an SkCodec should report as
its default settings (i.e. the return from getInfo). It makes sense for
an SkCodec to report that its pixels are unpremultiplied, if that is
the case for the underlying data, but if a client of SkImageGenerator
uses the default settings (as many do), they will receive
unpremultiplied pixels which cannot (currently) be drawn with Skia. We
may ultimately decide to revisit SkCodec reporting an SkImageInfo, but
I have left it unchanged for now.
Import features of SkImageGenerator used by SkCodec into SkCodec.
I have left SkImageGenerator unchanged for now, but it no longer needs
Result or Options. This will require changes to Chromium.
Manually handle the lifetime of fScanlineDecoder, so SkScanlineDecoder.h
can include SkCodec.h (where Result is), and SkCodec.h does not need
to include it (to delete fScanlineDecoder).
In many places, make the following simple changes:
- Now include SkScanlineDecoder.h, which is no longer included by
SkCodec.h
- Use the enums in SkCodec, rather than SkImageGenerator
- Stop including SkImageGenerator.h where no longer needed
Review URL: https://codereview.chromium.org/1220733013
This makes nanobench picture recording benchmarks somewhat useful again,
as opposed to all taking about 5us to run no matter the content.
ATTN Sheriff: this will probably trigger perf.skia.org alerts.
BUG=skia:
Review URL: https://codereview.chromium.org/1219873002
All of the CodecSubset benches fail when the color type is
kIndex8. We need to pass a color table to allocPixels()
when we want to decode to kIndex8 or it will throw a failure.
BUG=skia:
Review URL: https://codereview.chromium.org/1213983003
Changes verbose mode to print both the table and the individual sample
values. No need to hold back information in verbose mode.
BUG=skia:
Review URL: https://codereview.chromium.org/1208763003
Adds a nanobench mode that takes samples for a fixed amount of time,
rather than taking a fixed amount of samples.
BUG=skia:
Review URL: https://codereview.chromium.org/1204153002
Now that Sk4px exists, there's a lot less sense in eeking out every
cycle of speed from SkPMFloat: if we need to go _really_ fast, we
should use Sk4px. SkPMFloat's going to be used for things that are
already slow: large-range intermediates, divides, sqrts, etc.
A [0,1] range is easier to work with, and can even be faster if we
eliminate enough *255 and *1/255 steps. This is particularly true
on ARM, where NEON can do the *255 and /255 steps for us while
converting float<->int.
We have lots of experimental SkPMFloat <-> SkPMColor APIs that
I'm now removing. Of the existing APIs, roundClamp() is the sanest,
so I've kept only that, now called round(). The 4-at-a-time APIs
never panned out, so they're gone.
There will be small diffs on:
colormatrix coloremoji colorfilterimagefilter fadefilter imagefilters_xfermodes imagefilterscropexpand imagefiltersgraph tileimagefilter
BUG=skia:
Review URL: https://codereview.chromium.org/1201343004
Improves the GPU measuring accuracy of nanobench by using fence syncs.
Fence syncs are very widely supported and available on almost every
platform.
NO_MERGE_BUILDS
BUG=skia:
Review URL: https://codereview.chromium.org/1194783003
I think these changes to the subset benchmarks cover what we discussed yesterday.
I removed the divisor benchmarks (2x2, 3x3) and changed the single subset benchmarks.
Also, we will no longer benchmark subset decodes on small images.
BUG=skia:
Review URL: https://codereview.chromium.org/1188223002
Let's make CPU-bound .SKP benching mimic Chrome's tiles.
Unfortunately, the CPU code also performs a lot better with those big wide tiles...
BUG=skia:
Review URL: https://codereview.chromium.org/1189863002
This makes it easier to benchmark _mpd variants in a profiler.
E.g.,
<profiler> out/Release/nanobench --images --config 8888 --loops -1 --match sp_desk_nytimes
BUG=skia:
Review URL: https://codereview.chromium.org/1184673006
I haven't figured out a pithy way to have these apply to only classes
originating from SkNx, so let's just remove them. There aren't too
many use cases, and it's not really any less readable without them.
Semantically, this is a no-op.
BUG=skia:
Review URL: https://codereview.chromium.org/1167153002
The existing bench only tests the fast path, but we're looking to speed
up the general case. It'd be nice to be able to measure that speedup.
BUG=skia:
Review URL: https://codereview.chromium.org/1146953003
Constructing the gm tests and benches causes many calls to font loads.
This is visible as profiling samples in fontconfig and freetype on Linux
for all profiling runs of nanobench. This complicates analysis of
test-cases that are suspected of being slow due to font-related issues.
Move the font loading to GM::onOnceBeforeDraw and Benchmark::onPreDraw.
This way the code is not executed if the testcase does not match the
nanobench --match filter. This way the samples in font-related code are
more easy to identify as legitimate occurances caused by the testcase.
This should not cause differences in timings, because:
* Benchmark::preDraw / onPreDraw is defined to be run outside the timer
* GM::runAsBench is not enabled for any of the modified testcases. Also
nanobench untimed warmup round should run the onOnceBeforeDraw.
(and there are other GM::runAsBench gms already doing loading in
onOnceBeforeDraw).
Changes the behavior:
In TextBench:
Before, the test would report two different gms with the same name if
the color emoji font was not loaded successfully.
After, the test always reports all tests as individual names.
Generally:
The errors from loading fonts now print inbetween each testcase, as
opposed to printing during construction phase. Sample output:
( 143/145 MB 1872) 14.7ms 8888 gm quadclosepathResource /fonts/Funkster.ttf not a valid font.
( 160/160 MB 1831) 575µs 8888 gm surfacenewResource /fonts/Funkster.ttf not a valid font.
( 163/165 MB 1816) 12.5ms 8888 gm linepathResource /fonts/Funkster.ttf not a valid font.
( 263/411 MB 1493) 118ms 8888 gm typefacestyles_kerningResource /fonts/Funkster.ttf not a valid font.
( 374/411 MB 1231) 7.16ms 565 gm getpostextpathResource /fonts/Funkster.ttf not a valid font.
( 323/411 MB 1179) 4.92ms 565 gm stringartResource /fonts/Funkster.ttf not a valid font.
( 347/493 MB 917) 191ms 565 gm patch_gridResource /fonts/Funkster.ttf not a valid font.
( 375/493 MB 857) 23.9ms gpu gm clipdrawdrawCannot render path (0)
( 393/493 MB 706) 2.91ms unit test ParsePath------ png error IEND: CRC error
( 394/493 MB 584) 166ms gpu gm hairmodesResource /fonts/Funkster.ttf not a valid font.
Resource /fonts/Funkster.ttf not a valid font.
Resource /fonts/Funkster.ttf not a valid font.
...
Review URL: https://codereview.chromium.org/1144023002
Make GrResourceCache performance less sensitive to key length change.
The memcmp in GrResourceKey is called when SkTDynamicHash jumps the
slots to find the hash by a index. Avoid most of the memcmps by
comparing the hash first.
This is important because small changes in key data length can cause
big performance regressions. The theory is that key length change causes
different hash values. These hash values might trigger memcmps that
originally weren't there, causing the regression.
Adds few specialized benches to grresourcecache_add to test different
key lengths. The tests are run only on release, because on debug the
SkTDynamicHash validation takes too long, and adding many such delays
to development test runs would be unproductive. On release the tests
are quite fast.
Effect of this patch to the added tests on amd64:
grresourcecache_find_10 738us -> 768us 1.04x
grresourcecache_find_2 472us -> 476us 1.01x
grresourcecache_find_25 841us -> 845us 1x
grresourcecache_find_4 565us -> 531us 0.94x
grresourcecache_find_54 1.18ms -> 1.1ms 0.93x
grresourcecache_find_5 834us -> 749us 0.9x
grresourcecache_find_3 620us -> 542us 0.87x
grresourcecache_add_25 2.74ms -> 2.24ms 0.82x
grresourcecache_add_56 3.23ms -> 2.56ms 0.79x
grresourcecache_add_54 3.34ms -> 2.62ms 0.78x
grresourcecache_add_5 2.68ms -> 2.1ms 0.78x
grresourcecache_add_10 2.7ms -> 2.11ms 0.78x
grresourcecache_add_2 1.85ms -> 1.41ms 0.76x
grresourcecache_add 1.84ms -> 1.4ms 0.76x
grresourcecache_add_4 1.99ms -> 1.49ms 0.75x
grresourcecache_add_3 2.11ms -> 1.55ms 0.73x
grresourcecache_add_55 39ms -> 13.9ms 0.36x
grresourcecache_find_55 23.2ms -> 6.21ms 0.27x
On arm64 the results are similar.
On arm_v7_neon, the results lack the discontinuity at 55:
grresourcecache_add 4.06ms -> 4.26ms 1.05x
grresourcecache_add_2 4.05ms -> 4.23ms 1.05x
grresourcecache_find 1.28ms -> 1.3ms 1.02x
grresourcecache_find_56 3.35ms -> 3.32ms 0.99x
grresourcecache_find_2 1.31ms -> 1.29ms 0.99x
grresourcecache_find_54 3.28ms -> 3.24ms 0.99x
grresourcecache_add_5 6.38ms -> 6.26ms 0.98x
grresourcecache_add_55 8.44ms -> 8.24ms 0.98x
grresourcecache_add_25 7.03ms -> 6.86ms 0.98x
grresourcecache_find_25 2.7ms -> 2.59ms 0.96x
grresourcecache_find_4 1.45ms -> 1.38ms 0.95x
grresourcecache_find_10 2.52ms -> 2.39ms 0.95x
grresourcecache_find_55 3.54ms -> 3.33ms 0.94x
grresourcecache_find_5 2.5ms -> 2.32ms 0.93x
grresourcecache_find_3 1.57ms -> 1.43ms 0.91x
The extremely slow case, 55, is postulated to be due to the index jump
collisions running the memcmp. This is not visible on arm_v7_neon probably due
to hash function producing different results for 32 bit architectures.
This change is needed for extending path cache key in Gr
NV_path_rendering codepath. Extending is needed in order to add dashed
paths to the path cache.
Review URL: https://codereview.chromium.org/1132723003
I'm thinking of using this in perf with something like:
ratio(fill(filter("test=foo")), fill(filter("test=control")))
Does that make sense to you?
Not sure that this is really a good control bench on all bots,
but I propose we just run it a bit and find out if it needs work.
BUG=skia:
Review URL: https://codereview.chromium.org/1129823003
The benches for N <= 10 get around 2x faster on my N7 and N9. I believe this
is because of the reduced function-call-then-function-pointer-call overhead on
the N7, and additionally because it seems autovectorization beats our NEON code
for small N on the N9.
My desktop is unchanged, though that's probably because N=10 lies well within a
region where memset's performance is essentially constant: N=100 takes only
about 2x as long as N=1 and N=10, which perform nearly identically.
BUG=skia:
Review URL: https://codereview.chromium.org/1073863002
The colorfilter is applied to a single (paint's) color, so the bench does not
measure the filter at all, but simply the blit of a color.
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/1055383002
Now that all SkCodecs can rewind (assuming the stream is rewindable),
we do not need to special case it.
Pointed out by Derek in the code review that added this.
TBR=djsollen
Review URL: https://codereview.chromium.org/1058633002
CodecBench:
Add new class for timing using SkCodec.
DecodingBench:
Include creating a decoder inside the loop. This is to have a better
comparison against SkCodec. SkCodec's factory function does not
necessarily read the same amount as SkImageDecoder's, so in order to
have a meaningful comparison, read the entire stream from the
beginning. Also for comparison, create a new SkStream from the
SkData each time.
Add a debugging check to make sure we have an SkImageDecoder.
Add include guards.
nanobench.cpp:
Decode using SkCodec.
When decoding using SkImageDecoder, exclude benches where we decoded
to a different color type than requested. SkImageDecoder may decide to
decode to a different type, in which case the name is misleading.
TODOs:
Now that we ignore color types that do not match the desired
color type, we should add Index8. This also means calling the more
complex version of getPixels so CodecBench can support kIndex8.
BUG=skia:3257
Review URL: https://codereview.chromium.org/1044363002
The primary feature this delivers is SkNf and SkNd for arbitrary power-of-two N. Non-specialized types or types larger than 128 bits should now Just Work (and we can drop in a specialization to make them faster). Sk4s is now just a typedef for SkNf<4, SkScalar>; Sk4d is SkNf<4, double>, Sk2f SkNf<2, float>, etc.
This also makes implementing new specializations easier and more encapsulated. We're now using template specialization, which means the specialized versions don't have to leak out so much from SkNx_sse.h and SkNx_neon.h.
This design leaves us room to grow up, e.g to SkNf<8, SkScalar> == Sk8s, and to grown down too, to things like SkNi<8, uint16_t> == Sk8h.
To simplify things, I've stripped away most APIs (swizzles, casts, reinterpret_casts) that no one's using yet. I will happily add them back if they seem useful.
You shouldn't feel bad about using any of the typedef Sk4s, Sk4f, Sk4d, Sk2s, Sk2f, Sk2d, Sk4i, etc. Here's how you should feel:
- Sk4f, Sk4s, Sk2d: feel awesome
- Sk2f, Sk2s, Sk4d: feel pretty good
No public API changes.
TBR=reed@google.com
BUG=skia:3592
Review URL: https://codereview.chromium.org/1048593002
Need to land SK_SUPPORT_LEGACY_SCALAR_MAPPOINTS in chrome to suppress Affine
version which causes slight differences (which will need to be rebaselined)
BUG=skia:
Review URL: https://codereview.chromium.org/1045493002
Duplicate code from the HWUI backends for DM and nanobench
moves into a single place, saving a hundred lines or more of
cut-and-paste.
There's some indication that this increases the incidence of
SkCanvas "Unable to find device for layer." warnings, but no
clear degradation in test results.
R=djsollen@google.com,mtklein@google.com
BUG=skia:3589
Review URL: https://codereview.chromium.org/1036303002
Add and test trunc(), which is what get() used to be before rounding.
Using trunc() is a ~40% speedup on our linear gradient bench.
#neon #floats
BUG=skia:3592
#n5
#n9
CQ_INCLUDE_TRYBOTS=client.skia.android:Test-Android-Nexus5-Adreno330-Arm7-Debug-Trybot;client.skia.android:Test-Android-Nexus9-TegraK1-Arm64-Release-Trybot
Review URL: https://codereview.chromium.org/1032243002
Am I going nuts or can we get this down to just adds and converts in the loop?
#floats #n9
BUG=skia:3592
CQ_INCLUDE_TRYBOTS=client.skia.android:Test-Android-Nexus9-TegraK1-Arm64-Release-Trybot
Review URL: https://codereview.chromium.org/1008973004
There is no reason to require the 4 SkPMFloats (registers) to be adjacent.
The only potential win in loads and stores comes from the SkPMColors being adjacent.
Makes no difference to existing bench.
BUG=skia:
Review URL: https://codereview.chromium.org/1035583002
