Metal layer was using linear filter which was introducing blur.
Changed to nearest filter.
Bug: skia:10172
Change-Id: Ia69d6c9a56f9a2f69b5e98fd9941be52ec01b7d3
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/286616
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
IMPORTANT LESSON: when bringing in node (and possibly other
executables) via CIPD, add them to the path in gen_tasks_logic
so the parent executable (the task driver itself) has the right
PATH set. Otherwise, the subprocesses it spawns might grab the
wrong version because of how golang handles environments of
subprocesses.
This is starting as a fork of Skottie WASM. I hope to have a more unified
system for creating and running benchmarks.
Overall overview:
gen_tasks_logic.go creates a task in task.json that compiles
CanvasKit and the task drivers and then executes our task
(i.e. perf_puppeteer.go)
perf_puppeteer runs a node program (perf-with-puppeteer.js)
that uses puppeteer to execute benchmarking code on an
html page (canvaskit-skottie-frames-load.html).
I needed to update the node package so npm could be updated from
3.x to 6.14.4 so it knew about `npm ci`. This may not have been
entirely necessary, given the problems of executing the correct
npm (see important lesson above), but it hasn't broken things
further, so more up-to-date is probably a good thing.
Suggested Review Order:
- canvaskit-skottie-frames-load.html (note it is similar to
skottie-wasm-perf.html, but it waits for a button click
to start animating and records times from the main JS thread
itself)
- perf-with-puppeteer.js (similar to skottie-wasm-perf.js, but
has some things made optional [e.g. tracing])
- perf_puppeteer_test.go (shows the inputs/outputs of various steps)
- perf_puppeteer.go
- Everything else.
Change-Id: I380e81b825f36682c257664d488267edaf36369e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/285783
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Eric Boren <borenet@google.com>
- Lots of skstd::foo is now std::foo since C++14.
- Get rid of SK_WHEN(cond,T); std::enable_if_t<cond,T> is pithy enough.
- Move SkBitmaskEnum.h contents into sknonstd.
Change-Id: Ie5dc459405b1ff55e5b3ac57e70df7edd7cf38c0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/286315
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Change-Id: Id8065e4ff7299c12b1469468dab278b771c0382d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/286277
Auto-Submit: Adlai Holler <adlai@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Change-Id: Ibc0a47258009a24bf7d9b378c9e21729b13d0216
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/286027
Commit-Queue: Mike Klein <mtklein@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Auto-Submit: Mike Klein <mtklein@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
This reverts commit 44fc53b7f5.
Reason for revert: Test to see if this is causing the linux-rel MediaColorTest.Yuv420pHighBitDepth failure on the Chrome roll
Original change's description:
> Add BGR_10A2 support to Ganesh
>
> Bug: 1068416
> Change-Id: I40aa84b7f3f770ba550b7bea44c10173ae9a7ddf
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/285356
> Reviewed-by: Greg Daniel <egdaniel@google.com>
> Reviewed-by: Jim Van Verth <jvanverth@google.com>
> Commit-Queue: Robert Phillips <robertphillips@google.com>
TBR=egdaniel@google.com,jvanverth@google.com,robertphillips@google.com
Change-Id: I0ad0197ebd8de9b8761f84ba808c9f90891b9238
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 1068416
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/285958
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This JIT mode helps debugging and profiling by shelling out to an
external assembler then loading its results back in via dlopen(),
so you can see coherent function profiles and not just every
instruction as its own line in the profile.
It's very slow, so viewer will stutter for a second or two before
drawing goes smooth again. We can paper over this by using the
interpreter while these compiles are in progress, but I haven't hooked
that up yet.
Change-Id: I23e74d65a1a3a6d89649733296db8217be306438
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/285864
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@google.com>
Bug: 1068416
Change-Id: I40aa84b7f3f770ba550b7bea44c10173ae9a7ddf
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/285356
Reviewed-by: Greg Daniel <egdaniel@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Besides better matching Viz's behavior this also reduces a lot of choppiness in the composition RenderTask DAG.
In the previous approach DDL draws and compositing draws would be interleaved resulting in a lot of render target swaps.
This necessitated some reorganization bc I wanted to reuse PromiseImageCallbackContext to manage the tiles' promiseImages.
Change-Id: I513bf060a69ff2bfe0e7b82ae72f149dfede632e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/285056
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This fixes several issues:
* sysopen.py was moved to bin/sysopen in previous CL. Change to
use os.system to run it
* "adb push" doesn't follow symlinks.
* SkQP added a new commandline flag for rendertests file.
* The report path was wrong in old code.
Bug: skia:10156
Change-Id: I821a49c49bffe588f34d1216d9406c97ae5a903b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/284524
Reviewed-by: Tyler Denniston <tdenniston@google.com>
Auto-Submit: Lepton Wu <lepton@chromium.org>
Commit-Queue: Tyler Denniston <tdenniston@google.com>
Change-Id: I65cec59749f0e7f5fb13675293720afecffa6a80
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/284321
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Reed <reed@google.com>
SkMipMap::Build fails surprisingly often. Fall back to unmipped promise images in that case.
The prior approach created the mipMaps too late (i.e., at backendTexture creation time) for the test harness to change its mind about the mip status of a promise image. This approach moves the mipmap creation earlier - to when the PromiseImageInfos are created (i.e., at SKP deflation time).
Change-Id: Id3c67a44cc84da7ee76d02f4d44d7f27ce8e39d6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/284136
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Bug: skia:10149
Change-Id: Ia9b0bf2ed2042ba75088409612f6b091c5052ea9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/284220
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This better matches Chrome's use of DDLs.
With path, image, and text draws stripped out, here is the perf impact of this change:
before CL after CL
w/ DDLs 7.792 1.038
w/o DDLs 0.800 0.876
This perf improvement (in the DDL case) is from backend texture wrapping SkSurfaces being created w/o initialization. The prior method of SkSurface creation was resulting in double clearing of all the surfaces.
This perf improvement won't be seen by Chrome since they've always being using wrapped backend texture SkSurfaces.
TBR=bsalomon@google.com
Bug: 1056730
Change-Id: Ic04d322cad96df845e75437211208495862c6555
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/283866
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
When timing DDL performance with SKPs, if the mipmapping requirements of a given SkImage are gotten wrong at record time, the mipmaps will keep getting regenerated over and over again. This CL works around the problem by just creating all promise images as mipmapped. A better (but longer term solution) would be to examine the actual draw ops w/in an SKP.
Even more aggressively, we may want to disable mipmap regeneration w/in DDLs.
For desk_nytimes.skp on Windows/gl we have:
before CL after CL
w/ DDLs 7.999 3.136
w/o DDLs 1.953 1.863
This perf improvement won't be seen by Chrome since, presumably, they're creating their backend textures w/ the correct mipmappedness. Additionally, they, hopefully, aren't recording the same DDL over and over again (with incorrect mipmappedness).
Bug: 1056730
Change-Id: I8bf9dc9e64bc77159a04d89e5e3ac398e98beaa7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/283677
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Change-Id: Ifa25a44a75c6c193a8bd0780bda0f016fe3cbc31
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/283675
Commit-Queue: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: Mike Klein <mtklein@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit 7ae9d2fca6.
Reason for revert: Triggering Vulkan Debug layer errors
Original change's description:
> Update DDL test harness to use backendTextures to back tiles
>
> This better matches Chrome's use of DDLs.
>
> With path, image, and text draws stripped out, here is the perf impact of this change:
>
> before CL after CL
> w/ DDLs 7.792 1.038
> w/o DDLs 0.800 0.876
>
> This perf improvement (in the DDL case) is from backend texture wrapping SkSurfaces being created w/o initialization. The prior method of SkSurface creation was resulting in double clearing of all the surfaces.
>
> This perf improvement won't be seen by Chrome since they've always being using wrapped backend texture SkSurfaces.
>
> TBR=bsalomon@google.com
> Change-Id: Ice3993ca125fce37804e58c353c265cf659dbe2f
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/283456
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Robert Phillips <robertphillips@google.com>
TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com
Change-Id: Ife023ede0774ec2cce4c0d6e7708c036347ebf54
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/283648
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This better matches Chrome's use of DDLs.
With path, image, and text draws stripped out, here is the perf impact of this change:
before CL after CL
w/ DDLs 7.792 1.038
w/o DDLs 0.800 0.876
This perf improvement (in the DDL case) is from backend texture wrapping SkSurfaces being created w/o initialization. The prior method of SkSurface creation was resulting in double clearing of all the surfaces.
This perf improvement won't be seen by Chrome since they've always being using wrapped backend texture SkSurfaces.
TBR=bsalomon@google.com
Change-Id: Ice3993ca125fce37804e58c353c265cf659dbe2f
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/283456
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
These are neat but mostly just a distraction for now.
I've left all the assembly in place and unit tested
to make putting these back easy when we want to.
Change-Id: Id2bd05eca363baf9c4e31125ee79e722ded54cb7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/283307
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Herb Derby <herb@google.com>
Previously, in cases where the backbuffer was null, we would call layer->onPrePaint (which starts a new frame with ImGui::NewFrame) without a matching call to layer->onPaint (which calls ImGui::Render). This causes the ImGui frame counter to go out of sync, which leads to an assertion when NewFrame is next called.
(The backbuffer can become null if unsupported settings are used, e.g. a MSAA setting above the maximum supported by your GPU.)
Change-Id: I1b8fbaa1bf78ad5e1aa41fd04eda6be206c952b4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/283196
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Change-Id: I2cb6255a553852a292427d6dc9ef8c5ed7f8286d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/252926
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Mike Reed <reed@google.com>
New SDK is out and fixes the underlying issue. Removing workaround as
anyone can just update to the newest SDK.
Change-Id: I75ebde05edc75d98a8b4b1228e75a39bb22670a1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/282850
Commit-Queue: Greg Daniel <egdaniel@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Auto-Submit: Greg Daniel <egdaniel@google.com>
Reviewed-by: Chris Dalton <csmartdalton@google.com>
This is not part of the public API of Dawn and will be unnecessary once
Dawn move to "surface based swapchains". DawnD3D12WindowContext didn't
need the include so all that's needed is copying the helper in
DawnMTLWindowContext.
Change-Id: I8d0682b6b0801dc311ad4c7d73d4c07e575ecf4e
Bug: chromium:1064305
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/282496
Reviewed-by: Zhenyao Mo <zmo@google.com>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Main motivation for this was convenience when rendering directories of
external files (e.g. --svgs) as readdir() does not guarantee any
ordering.
Change-Id: I78dced834e5a3edde4a5e8e0e65ad946c7d3a3fa
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/282617
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Tyler Denniston <tdenniston@google.com>
It's nice to be able to resize the SVG container on the fly.
Change-Id: I6624e6151aaff7e2a7f6eb07e84aee88c4a2e408
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/282418
Auto-Submit: Florin Malita <fmalita@chromium.org>
Commit-Queue: Tyler Denniston <tdenniston@google.com>
Reviewed-by: Tyler Denniston <tdenniston@google.com>
We think this is the driver's problem and not ours, right?
Change-Id: I2c94ee904c7ca4844a5e91291b10b0efa8ce3806
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/282298
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
I can't reproduce the leak locally on my workstation.
Change-Id: I54b23b9d040467fabe99af9a3a2d2a0615aac6e9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/282072
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
SK_SUPPORT_LEGACY_DIDCONCAT44
Change-Id: Ie208c245d7121fddc693a7d3c5d6866441c27433
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/281864
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Reed <reed@google.com>
Change-Id: Idcc598af6b3859d33b674402c973438ddfdd632b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/282038
Commit-Queue: Greg Daniel <egdaniel@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
Auto-Submit: Greg Daniel <egdaniel@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Having GrAtlasManager.h in GrContextPriv.h was needlessly propagating dependence on that header.
Change-Id: Idf5836f1e217ecd2da91f751b488a63a884c02ce
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/281739
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
bit_clear is at least useful as a special case for select(),
which helps with code readability.
Add is_NaN() and use these all together in sweep gradient.
Change-Id: I57a54f8956f85e0db0662b33f8446b8dc7342d8d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/281685
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
bit_clear() is just another bit_and(),
and bytes() is a way of expression pshufb
that we never really use (yet).
Can always add them back later, but there's
some extra complexity to think about for each
that I'd like to not think about now:
- common sub-expression elimination between bit_and and bit_clear
- large constant management JIT'ing bytes
Change-Id: I3a54afa963231fec1d5de949acc647e3430ed0d8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/281557
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This is a reland of d84b691950
Original change's description:
> Reland "Use glDraw.*BaseInstance calls to avoid deferred buffer binding"
>
> This is a reland of e8c963d474
>
> Original change's description:
> > Use glDraw.*BaseInstance calls to avoid deferred buffer binding
> >
> > Change-Id: I968dab317673051acc65f87ea76a0d657d89b3d2
> > Reviewed-on: https://skia-review.googlesource.com/c/skia/+/279538
> > Commit-Queue: Chris Dalton <csmartdalton@google.com>
> > Reviewed-by: Michael Ludwig <michaelludwig@google.com>
> > Reviewed-by: Brian Salomon <bsalomon@google.com>
>
> Change-Id: I79b2d23e5e66d47214898a9068079b6fe2269599
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/280806
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
Change-Id: I9240c5cc6e58ce196be393c649bbdd86560602f0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/281077
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Change-Id: I2d19c4f0ff1439dcd923a3064eb3ba78432a5113
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/281043
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
This is a reland of e8c963d474
Original change's description:
> Use glDraw.*BaseInstance calls to avoid deferred buffer binding
>
> Change-Id: I968dab317673051acc65f87ea76a0d657d89b3d2
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/279538
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
> Reviewed-by: Michael Ludwig <michaelludwig@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
Change-Id: I79b2d23e5e66d47214898a9068079b6fe2269599
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/280806
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Bug: skia:9935
Change-Id: I4c2d70f69e30f78caca0f49629880565f178f495
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/280609
Commit-Queue: Jim Van Verth <jvanverth@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This reverts commit e8c963d474.
Reason for revert: assert failures: https://logs.chromium.org/logs/skia/4b4b7fcb2ff3fe11/+/steps/dm/0/stdout
Original change's description:
> Use glDraw.*BaseInstance calls to avoid deferred buffer binding
>
> Change-Id: I968dab317673051acc65f87ea76a0d657d89b3d2
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/279538
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
> Reviewed-by: Michael Ludwig <michaelludwig@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
TBR=bsalomon@google.com,csmartdalton@google.com,michaelludwig@google.com
Change-Id: Ic760a56ca9d112e924baf7e833adb09b371928b0
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/280817
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
This reverts commit e990fcc4b0.
Reason for revert: Build-Win-Clang-x86_64-Release-Shared
Original change's description:
> Enable deprecated-copy-dtor warning.
>
> In C++11 a user declared destructor still requires the compiler to
> implicitly default the copy constructor and copy assignment operator,
> but this is deprecated. Note that a user declared destructor suppresses
> the move constructor and move assignment operator; a user declared
> destructor exists if any '~Foo' method declaration appears inside
> 'class Foo' (even if defaulted); if the copy and move operations are the
> same then copy operations that take 'const Foo&' will do fine double
> duty as move operations.
>
> Clang seems to have an issue with this warning, in that it does not
> appear to distinguish between compiler defaulted and user defaulted
> destructors. As a result, it does not always warn when it should.
> There may yet be places in the code where a move operation is desired
> but may be suppressed because the implicitly defaulted moves are not
> declared because a destructor has been declared.
>
> This wraps dawn and shaderc configs in 'third_party' so that their
> headers will be included through '-isystem' in order to avoid the
> warnings generated by including their headers.
>
> Change-Id: I681524cd890d86305aa99b6b765a52113b4dfa4b
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/280406
> Reviewed-by: Mike Klein <mtklein@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Ben Wagner <bungeman@google.com>
TBR=mtklein@google.com,bsalomon@google.com,bungeman@google.com
Change-Id: Icd6a2487637d21fcf7c4c7ab7cba7a8adfda5afd
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/280836
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
