This ties the caps to the compiler instance, paving the way for
pre-optimizing the shared code. Most of the time, the compiler is
created and owned the GPU instance, so this is fine. For runtime
effects, we now use the shared (device-agnostic) compiler instance
for the first compile, even on GPU. It's configured with caps that
apply no workarounds. We pass the user's SkSL to the backend as
cleanly as possible, and then apply any workarounds once it's part
of the full program.
Bug: skia:10905
Bug: skia:10868
Change-Id: Ifcf8d7ebda5d43ad8e180f06700a261811da83de
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/331493
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Move the SVG rendering code to modules/svg, and componentize.
Also split into include/src/utils.
As external clients still reference the old header locations,
introduce temporary forwarding headers to facilitate the migration.
This reverts commit d6cf56fd34.
TBR=
Change-Id: Ibadd7c8dc0464ec0c27841530ade0c2098305d20
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/327344
Commit-Queue: Florin Malita <fmalita@google.com>
Reviewed-by: Florin Malita <fmalita@google.com>
This reverts commit 6fc4106a9d.
Reason for revert: Blocking the Android roll
Original change's description:
> [svg] Relocate out of experimental
>
> Move the SVG rendering code to modules/svg, and componentize.
> Also split into include/src/utils.
>
> As external clients still reference the old header locations,
> introduce temporary forwarding headers to facilitate the migration.
>
> Change-Id: Ib289dbdcd80c16a01c47805e7242f2e08bebc165
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326948
> Reviewed-by: Tyler Denniston <tdenniston@google.com>
> Commit-Queue: Florin Malita <fmalita@google.com>
TBR=fmalita@chromium.org,fmalita@google.com,tdenniston@google.com
Change-Id: I386cf77a15a9e1d392029804abaf937dae53f435
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/327342
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Move the SVG rendering code to modules/svg, and componentize.
Also split into include/src/utils.
As external clients still reference the old header locations,
introduce temporary forwarding headers to facilitate the migration.
Change-Id: Ib289dbdcd80c16a01c47805e7242f2e08bebc165
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326948
Reviewed-by: Tyler Denniston <tdenniston@google.com>
Commit-Queue: Florin Malita <fmalita@google.com>
This should fix the chrome roll.
Change-Id: I2de68f972996bf6124cf5cc27dfd538aa1161057
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/316877
Auto-Submit: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Leon Scroggins <scroggo@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
We had several defines around the code base that were not
very descriptive. Additionally, we had a patch of extra
runtime restrictions living in oss-fuzz that were applied
when fuzzing over there for some fuzzers.
This has all be consolidated and controlled via the defines
documented in site/dev/testing/fuzz.md
As such, we can remove one of the patches that is in oss-fuzz,
taking us closer to being able to fuzz in the CI/CQ.
PS 1 renames existing fuzz defines to the new schema.
PS 2-3 backports skia.diff from oss-fuzz and changes those
definitions to have the _GREATLY modifier.
PS 5+ further condenses the defines so that there is one
define for gating the runtime checks.
Change-Id: Ia4ad96f30c1e9620a2123b510e97c6f501a2e257
Docs-Preview: https://skia.org/?cl=316443
Bug: skia:10713
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/316443
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Keep the same default value, but add a (private) API to change it when
compiling SkRuntimeEffect code.
Use the new API to improve fuzzer coverage, by fuzzing with inlining
disabled and (enthusiastically) enabled.
This reverts commit 3e8fae7193, reworked
to avoid the static initializer.
Change-Id: I7e6cd39d4af2daa4b1be41f1c7d99f32df7a51ab
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309664
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Runtime effects previously allowed two kinds of global input variables:
'in' variables could be bool, int, or float. 'uniform' could be float,
vector, or matrix. Uniform variables worked like you'd expect, but 'in'
variables were baked into the program statically. There was a large
amount of machinery to make this work, and it meant that 'in' variables
needed to have values before we could make decisions about program
caching, and before we could catch some errors. It was also essentially
syntactic sugar over the client just inserting the value into their SkSL
as a string. Finally: No one was using the feature.
To simplify the mental model, and make the API much more predictable,
this CL removes 'in' variables entirely. We no longer need to
"specialize" runtime effect programs, which means we can catch more
errors up front (those not detected until optimization). All of the API
that referred to "inputs" (the previous term that unified 'in' and
'uniform') now just refers to "uniforms".
Bug: skia:10593
Change-Id: I971f620d868b259e652b3114f0b497c2620f4b0c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309050
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This reverts commit ad3db40d78.
Reason for revert: Chrome doesn't like static initializers.
Original change's description:
> Add inline threshold to SkSL::Program::Settings
>
> Keep the same default value, but add a (private) API to change it when
> compiling SkRuntimeEffect code.
>
> Use the new API to improve fuzzer coverage, by fuzzing with inlining
> disabled and (enthusiastically) enabled.
>
> Change-Id: I36424bac95144aeb727cfb949754fbe998d5d7de
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308181
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Kevin Lubick <kjlubick@google.com>
> Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
TBR=kjlubick@google.com,brianosman@google.com,ethannicholas@google.com
Change-Id: Ic233203f3728a7285a1958c53567d915e56023af
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308757
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Keep the same default value, but add a (private) API to change it when
compiling SkRuntimeEffect code.
Use the new API to improve fuzzer coverage, by fuzzing with inlining
disabled and (enthusiastically) enabled.
Change-Id: I36424bac95144aeb727cfb949754fbe998d5d7de
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308181
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
SkDescriptor's programatic API is neither
exposed for untrusted use nor harded for it.
Why are we fuzzing it?
Do we need a change in oss-fuzz before deleting this?
Bug: oss-fuzz:19648
Bug: oss-fuzz:24417
Change-Id: Id8d075938d831ec8cad4014c8fe6efaef46edb55
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/307177
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Herb Derby <herb@google.com>
Auto-Submit: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This is an attempt to fuzz the usage of SkSurfaceCharacterization,
SkDeferredDisplayRecorder, and SkDeferredDisplayList.
This fuzzer first makes a surface and characterization from
GrDirectContext and then create a DDL and draw it on the surface.
The code is compiled with ninja and run with AFL at the speed around
600/sec
The future changes will include:
1. An alternative way to create DDL: first create the surface and
extract the characterization from that existing surface.
2.currently we just pass the ownership of the DDL into draw_ddl. In
the future we should add a version that retains ownership of the DDL
in order to fuzz the lifetime of the DDL.
3. Refactorize line 62-119
Change-Id: I9cd9736813be3abc82430bd4eeb559d6993ecbd4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/303600
Commit-Queue: Zepeng Hu <zepenghu@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
The major improvement is that now the fuzzer is able to execute
the sksl code (before it just compiled it). The fuzzer will
reserve 256 bytes for providing uniforms to the shader;
meanwhile, the fuzzer will read the remaining bytes as sksl code
to create SkRuntimeEffect. It then creates a shader and executes
it by painting the shader on a canvas.
The code was tested locally with afl-fuzz, and the execution
speed was around 700/sec.
An alternative implementation would have been using Fuzz.h to
read bytes; I decided to go with sk_sp<SkData> since it has a
comparable format to other binary fuzzer and meets all the
functionality in this fuzzer.
For future changes, there are 2 important improvements to the
implementation:
1) Current shader does not have children shaders; thus,
makeShader() will fail if the SkSL ever tries to use an 'in shader'.
As pointed out in patchset 11, after creating the runtime effect,
effect->children().count() will tell you how many children it's
expecting (how many 'in shader' variables were declared). When you
call makeShader(), the second and third arguments are a
(C-style) array of shader pointers, and
a count (which must match children().count()).
Some helpful examples can be SkRTShader::CreateProc in
SkRuntimeEffect.cpp, make_fuzz_shader in FuzzCanvas.cpp.
2)
In this fuzzer, after creating the paint from a shader, the paint
can be drawn on either GPU canvas or CPU, so a possible way is to
use SkSurface::MakeRenderTarget to create GPU canvas and use a byte
to determine which canvas it will be drawn on.
Change-Id: Ib0385edd0f5ec2f23744aa517135a6955c53ba38
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/300618
Commit-Queue: Zepeng Hu <zepenghu@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
When manipulating svg file, the implementation of SkSVGCanvas will be used instead of the
implementation of SkCanvas, so the api are tested against SkSVGCanvas. In addition, there are
more api need to be covered in the function fuzz_canvas. As a result, the main changes are to
add new DEF_FUZZ for SkSVGCanvas and to modify fuzz_canvas to increase the coverages of api.
Change-Id: Iaf6114bb0e2929c73549ff398c3db5592e736ea2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/298977
Commit-Queue: Zepeng Hu <zepenghu@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Bundling the pipeline stage arguments also simplifies the code in
several spots.
Change-Id: I85e81b436a39378f753cc9404b6eeb27fe055525
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/261778
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
One is an API fuzzer, the other is for deserializing.
Bug: skia:9548
Change-Id: I5923b8fb76f36ec09fca74d5ba82245a8ddb5938
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/249776
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Bug: skia:
Change-Id: I1483cdf7229b7234be41d21407e2b4abf99fff76
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/239925
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Current strategy: everything from the top
Things to look at first are the manual changes:
- added tools/rewrite_includes.py
- removed -Idirectives from BUILD.gn
- various compile.sh simplifications
- tweak tools/embed_resources.py
- update gn/find_headers.py to write paths from the top
- update gn/gn_to_bp.py SkUserConfig.h layout
so that #include "include/config/SkUserConfig.h" always
gets the header we want.
No-Presubmit: true
Change-Id: I73a4b181654e0e38d229bc456c0d0854bae3363e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/209706
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Hal Canary <halcanary@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Turns out lots of tools had two copies of many of these flags.
Some GN and .cpp file refactoring to make sure when flags are
present in a binary, they do something in that binary.
I think this finally finishes the flag refrag.
Change-Id: I01488e37ab73a5c4361786863ddb137a7f1095b1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/203420
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>
sk_tool_utils doesn't really fit the naming convention
the rest of code under tools/ tends to use.
Change-Id: I45326a174101c6eb4b6149e9c742f658f2fd23b1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/202313
Auto-Submit: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: I0326eb9cc1e1e38b0fdc417567987a595f9021d2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/202310
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>
Otherwise, the string constructor can walk off the end
looking for a null terminator that never arrives.
Fix some logging copypasta
Bug: skia:
Change-Id: I0cb1b0b75673f64a5ac647307dbc04253f707686
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/199937
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
This should make reproducing certain fuzzes easier between oss-fuzz
and a typical dev's desktop.
This was the most straight-forward way I could think of to
accomplish this. An ideal solution would "compile" a set of
fonts that was not the test set and embed it, but I lack the
domain knowledge to craft such a set.
If this method works ok, we can explore making the font set
more robust and varied.
Bug: 818769
Change-Id: I03eb2bc316caf7aec3ffa88e59ff29d76c8557ec
Reviewed-on: https://skia-review.googlesource.com/c/177800
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
Fuzz SkAndroidCodec to help to catch errors in both incrementalDecode
and scanlineDecode. Try a variety of sample sizes, but cap it at 64.
Though sometimes larger sample sizes are used, the lower ones tend to
more common. Also draw the resulting bitmap to verify that we
initialized all pixels.
Independently test incrementalDecode to ensure that it initializes
rowsDecoded.
Change-Id: I20d8a408cd280262fdc62f902a6f04f0f57f5ad2
Reviewed-on: https://skia-review.googlesource.com/c/162025
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Auto-Submit: Leon Scroggins <scroggo@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
This CL renames FuzzPath() to FuzzNicePath() to remind us that it's
meant to create paths that a user could reasonably want to create
in good faith, to pass to Skia via its API, etc.
Then, add fuzz_nice_rect(), and have FuzzNicePath() use that to create
its rectangles and use FuzzNiceMatrix() to create its matrices, just
like we already use FuzzNiceRRect() to create rounded rectangles and
FuzzNicePath() itself to create sub-paths.
Using fuzz_nice_rect() should be the fix for the attached bug.
Using FuzzNiceMatrix() is by analogy, more preemptive.
While we're at it, rename BuildPath to FuzzEvilPath, so the contrast
with FuzzNicePath is more clear.
Update the assertions that we create a valid path in FuzzNicePath()
to tell us where things went wrong if they do.
Bug: oss-fuzz:10667, skia:8384
Change-Id: I6d802182a62815cd969c65cf0479609f64b1da55
Reviewed-on: https://skia-review.googlesource.com/156840
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Auto-Submit: Mike Klein <mtklein@google.com>
With the fixes in https://skia-review.googlesource.com/c/skia/+/150465
this should allow us to fuzz Pathop on oss-fuzz.
Bug: skia:
Change-Id: Id5df511f850f23b5aad0bcb39664d18f639ddb69
Reviewed-on: https://skia-review.googlesource.com/150560
Auto-Submit: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Cary Clark <caryclark@google.com>
Commit-Queue: Cary Clark <caryclark@google.com>
Prepare SkRegion.h, SkShader.h, SkStream.h for documentation.
Name params, add trailing commas to enum member list,
move or remove some public SkRegion.h stuff.
SkRegion gets a minor overhaul to move some pieces
to private: or SkRegionPriv. The intent is to preserve the
current code so that the fixes for documentation do not impact
performance or code size.
R=djsollen@google.com,reed@google.com
Docs-Preview: https://skia.org/?cl=141284
Bug: skia:6818
Change-Id: I0d82794081b8739a9e8af0d1cd4a0e5d32d04f04
Reviewed-on: https://skia-review.googlesource.com/141284
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Mike Reed <reed@google.com>
SkTextBlob has a number of untested entry points
to serialize and deserialize. Privitize ones only used
by Skia, and remove ones suspected to be unused
and untested.
R=fmalita@chromium.orgTBR=reed@google.com,bsalomon@google.com
Bug: skia:6818
Change-Id: I6a9982a26a883982af3592f3302029a1bcdf5aa3
Reviewed-on: https://skia-review.googlesource.com/141820
Reviewed-by: Florin Malita <fmalita@chromium.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
Commit-Queue: Cary Clark <caryclark@skia.org>
It's a tiny, core-ish component -- might as well treat as such to
simplify dependencies.
Change-Id: I6f31ce2d151f9a629d88bfc7f15d64891d5150c0
Reviewed-on: https://skia-review.googlesource.com/135780
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Bug: skia:7776
Change-Id: I4f2791375d8be05486d10ce8f7f2e58aa032cfd8
Reviewed-on: https://skia-review.googlesource.com/118166
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Jonathan Metzman <metzman@chromium.org>
This also includes a helper "fuzzer" for making a corpus.
Point it at an image or folder of images and it will
decode those images and write the SkPixmap's bytes to
disk, such that the fuzzer will be able to read in
those bytes as if it had decoded the image (or gotten
it from another source).
Bug: skia:
Change-Id: Iaf223a39078f2b62908fb47929add5d63f22d973
Reviewed-on: https://skia-review.googlesource.com/117367
Reviewed-by: Leon Scroggins <scroggo@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Bug: 827225
Change-Id: Icb30c0c234326340213af0cc402a4124dd0336b3
Reviewed-on: https://skia-review.googlesource.com/117150
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Bug: skia:
Change-Id: I0d34bfff4a53f831986614844bdc955935f28501
Reviewed-on: https://skia-review.googlesource.com/115582
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Bug: skia:
Change-Id: I3e051cefd6861b63bab33a1812674eacf67a35dd
Reviewed-on: https://skia-review.googlesource.com/113748
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Also Remove ScaleToSides, which we hadn't been running for a while.
Bug: skia:
Change-Id: I772dad722c34681392d5b635b3de716f3b00d597
Reviewed-on: https://skia-review.googlesource.com/110443
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
This only changes it for the oss-fuzz executable
which allows our normal fuzz executable to repro
on older versions, if needed.
This CL also accompanies additions to the corpus
of a bunch of v4 paths.
Bug: skia:
Change-Id: I4a1a3b27f48423f2bddc73e1b8bf63b82dfa59ff
Reviewed-on: https://skia-review.googlesource.com/109560
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
This also adds in a few small guards to prevent libfuzzer from frequently
running out of memory when an image claims to have billions of pixels.
Bug: skia:
Change-Id: I47a9daac832c4d85a42000698482b61721c38880
Reviewed-on: https://skia-review.googlesource.com/106264
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Leon Scroggins <scroggo@google.com>
FuzzImageFilterDeserialize is already being used in oss-fuzz
but the target lived there and not here. This moves it here.
Then we can turn on:
- FuzzPathDeserialize
- FuzzTextBlobDeserialize
Bug: skia:
Change-Id: I7baee8386fb7aeebc43a68abfff9a670ba16f82c
Reviewed-on: https://skia-review.googlesource.com/105763
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
