Creates a new path renderer, GrAtlasPathRenderer, that handles all the
atlasing. Managing the atlas in its own path renderer gives us more
control over when atlasing happens in the chain, will allow us to more
easily use the atlas in kCoverage mode, and makes the clipping code
cleaner.
Bug: skia:12258
Change-Id: Ie0b669974936c23895c8ab794e2d97206ed140f8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/431896
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
This should hopefully make the code safer in C++14 by not violating
the ODR.
(bungeman@: "I think in C++14 we will in fact get multiple definitions
of this symbol, though in C++17 we won't. Maybe we should mark this as
'static' so each TU gets its own zero byte copy in C++14 but in C++17
the 'static' will also be implicitly 'static inline' and we'll only get
one?)
Change-Id: I3570421fcf2be314e1baccff0783ded4a1078666
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/427037
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Change-Id: I3f2e927001da10652b2965687e31aa8452c9dfc6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425997
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Bug: skia:11296
Change-Id: I7d41614957d6fa535faadebbeca890b54b6977ac
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425996
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Change-Id: Id2061ebe7873aa8b9480a2d8b0133c2fb79e79bb
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/424098
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Add full compliment of operator*, have them assert in debug,
but not throw in release as per c++ spec.
Add full compliment of value() have them crash if there is
no value as per the c++ spec.
Add value_or.
Add test for value_or.
Change-Id: Ib0d9a01fba367768e66d00fbc61b471397e8582c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425476
Commit-Queue: Herb Derby <herb@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This reverts commit 1c467774e5.
Change-Id: Iad409002ddd71f0c26117e5a0814bf021fba8178
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425186
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This reverts commit 4a77813008.
Reason for revert: Memory regression in Chrome.
Original change's description:
> Convert GrConfigConversionEffect to a runtime FP
>
> Change-Id: I7f22447cf3356b1558d73665ff3e9b61639ebe83
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/423576
> Auto-Submit: Brian Osman <brianosman@google.com>
> Commit-Queue: Brian Salomon <bsalomon@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
TBR=bsalomon@google.com,brianosman@google.com,ethannicholas@google.com
Change-Id: I05497295b88b378f0aa236f18118fe676bbf05c9
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/424256
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: I7f22447cf3356b1558d73665ff3e9b61639ebe83
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/423576
Auto-Submit: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This is a reland of e58831cd95
Original change's description:
> Add format-specifier warnings to SkDebugf.
>
> This CL fixes up many existing format-specifier violations in Skia.
> Note that GCC has a warning for formatting nothing, so existing calls to
> `SkDebugf("")` have been removed, or replaced with `SkDebugf("%s", "")`.
> These were apparently meant to be used as a place to set a breakpoint.
>
> Some of our clients also use SkDebug with bad format specifiers, so this
> check is currently only enabled when SKIA_IMPLEMENTATION is true.
>
> Change-Id: I8177a1298a624c6936adc24e0d8f481362a356d0
> Bug: skia:12143
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/420902
> Auto-Submit: John Stiles <johnstiles@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Brian Osman <brianosman@google.com>
Bug: skia:12143
Change-Id: Id3c0c21436ebd13899908d5ed5d44c42a0e23921
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/421918
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit e58831cd95.
Reason for revert: looks like breaking a few build bots
Original change's description:
> Add format-specifier warnings to SkDebugf.
>
> This CL fixes up many existing format-specifier violations in Skia.
> Note that GCC has a warning for formatting nothing, so existing calls to
> `SkDebugf("")` have been removed, or replaced with `SkDebugf("%s", "")`.
> These were apparently meant to be used as a place to set a breakpoint.
>
> Some of our clients also use SkDebug with bad format specifiers, so this
> check is currently only enabled when SKIA_IMPLEMENTATION is true.
>
> Change-Id: I8177a1298a624c6936adc24e0d8f481362a356d0
> Bug: skia:12143
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/420902
> Auto-Submit: John Stiles <johnstiles@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Brian Osman <brianosman@google.com>
TBR=brianosman@google.com,ethannicholas@google.com,johnstiles@google.com
Change-Id: I07848c1bf8992925c9498e916744d0840355a077
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:12143
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/421917
Reviewed-by: Tyler Denniston <tdenniston@google.com>
Commit-Queue: Tyler Denniston <tdenniston@google.com>
This CL fixes up many existing format-specifier violations in Skia.
Note that GCC has a warning for formatting nothing, so existing calls to
`SkDebugf("")` have been removed, or replaced with `SkDebugf("%s", "")`.
These were apparently meant to be used as a place to set a breakpoint.
Some of our clients also use SkDebug with bad format specifiers, so this
check is currently only enabled when SKIA_IMPLEMENTATION is true.
Change-Id: I8177a1298a624c6936adc24e0d8f481362a356d0
Bug: skia:12143
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/420902
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This involves a significant number of overloads, as optionals can be
compared against other optionals, or nullopt, or a value; either side
can mix and match as desired.
Change-Id: I26b8420e417300d32569fc76a55bcfd8e01eb322
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/420576
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
This reverts commit 7bf6bc0d06.
Reason for revert: Android build references kCoverageCounting
Original change's description:
> Purge ccpr
>
> Now that the clip atlas has been successfully migrated to
> tessellation, we don't need this code anymore!
>
> Change-Id: Ic97f50cff7c4ee59f4476f8410f0b30a32df4e90
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/419857
> Reviewed-by: Robert Phillips <robertphillips@google.com>
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
TBR=robertphillips@google.com,brianosman@google.com,csmartdalton@google.com,michaelludwig@google.com
Change-Id: I01d99287978f848eb8bf900c07cba90ceb3b6edc
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/419898
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Now that the clip atlas has been successfully migrated to
tessellation, we don't need this code anymore!
Change-Id: Ic97f50cff7c4ee59f4476f8410f0b30a32df4e90
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/419857
Reviewed-by: Robert Phillips <robertphillips@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
This is almost purely a query/replace CL (% the #include juggling).
The VisitProxyFunc is used by more than just the Ops and will probably
still be required in the NGA.
This is a prelude to making all the remaining GrOp-derived classes OGA-only.
Bug: skia:11837
Change-Id: If1c127e5c126c676be529ed2a61dd7953abb03d0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/419162
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Change-Id: I2de7d90504c2030f9d2ce2694899d350362ebb5c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/419056
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Like SkColorFilter, SkShader, etc., this has a public-facing component
(SkBlender) and a private subclass (SkBlenderBase) which can be
obtained via a helper function (as_BB). At present there are no public-
facing methods, but the type needs to be exposed to be usable by the
outside world.
These classes exist for SkRuntimeEffect to subclass. The blender base
provides a `program` method with the parameters that blending will use.
Change-Id: I75c772fd4108a9c21fbda84201a8b23d3750a0df
Bug: skia:12080
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/416916
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: I17c624e9145d7152f65695ca3ce4592a11bb09a2
Bug: skia:12080
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/418637
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This CL preserves the "StringFragment" name as an alias for
string_view to reduce the impact. The StringFragment alias
will be removed in a followup CL.
Change-Id: I89209bc626b0be0d0190823b6217f4c83cafe1bc
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/416736
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Runtime Blend effects always take two input colors--source and
destination--instead of one. This CL adds a new ProgramKind for blend
effects, a new program module (empty for now), and adds a test to
confirm that the signature for blend functions is checked. Currently
these are only accessible via skslc; there's no Runtime Effect API to
create one and the dest color isn't hooked up to anything.
Change-Id: I5272a811d2d76b878cfdf3429efa78c9c8b3fd97
Bug: skia:12080
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/416798
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Instead of having an enum that says how to sample the dst in the shader
(or lack of dst sampling), we now rely on whether or not we have a valid
proxy in the GrDstProxyView to know whether we are sampling the dst at
all. Then if so we additionally have GrDstSampleFlags to say whether we
need to use a texture barrier and whether we are sampling from an input
attachment.
Change-Id: Id0390a8ad57ec52674922807f6c050d59b7e75a2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/416416
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Chris Dalton <csmartdalton@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Change-Id: Id02ed33b4fe60663327cc50d79db2bf876ad6df7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/415458
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
This adds copy constructors / assignment operator, reset(), value(), and
has_value() to SkTOptional.
Change-Id: I564552a75a4c612685cdaa8e80e4359b394b64a4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/414338
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This is needed by the upcoming DSLParser.
Change-Id: I54a0714e55feeb78894df766b14c795970f2c2d4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/411308
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Previously we would completely disable hardware tessellation for a path
if there was any chance of a curve requiring more segments than
supported by the hardware. This CL updates the tessellators to simply
chop paths until they fit in patches, allowing us to finally draw any
path using hardware tessellation.
Bug: skia:10419
Change-Id: I5c9f78cda3e30b8810aff3cb908235965706f2d8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/410977
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Other than GMs, there is exactly one use of matrix-sampling
(GrMatrixEffect). It is always uniform (not literal or an
expression containing a uniform). The uniform always has the
same name. Bake these simplifications in, which also shrinks
SampleUsage quite a bit.
Change-Id: I0d5e32069d710af475ccc1030e2988c5fc965a98
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/411296
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
- Remove ctypes that were entirely unused
- Remove explicit selection of default ctypes
- After that, only two ctype tokens are needed (SkPMColor4f and SkV4)
... remove all of the others from the parser
Change-Id: I2322aab73a19127b3b26850aefdad6140ea0f7e7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/410057
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This was only used in the context of sk_SampleMask, which was removed
recently.
Change-Id: Id70d7af8b3a100ff157c2984bad4131f1b92f317
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/410056
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: I6019418526def09c6c9f4b22567a2c76542d043c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/409876
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Want to also use this to use this for texture formats so remove
"ColorType" from name.
Also class rather than struct.
Bug: chromium:1113801
Change-Id: Ieb08a3d81c465b92b956d9bc04b39c5783715ea8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/409476
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Both SkColorData.h and SkNx_sse.h are in include/private, but not
(currently) included by any public headers. Adding them to a public
header reveals warnings about implicit conversion changing signedness
(-Wsign-conversion), enforced in our public headers warnings check.
Bug: skia:11995
Change-Id: Iee44b24a6df031431113e2d25e42f0a8dd55a5c0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/408056
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Herb Derby <herb@google.com>
This layout qualifier is not actually used anywhere.
Change-Id: I817c9affdd00e492c70f251eb52680644b7ff3f7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/406141
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
SkSpan is a convenient type to use on API boundaries - callers can have
their collection in a stack-allocated array, or any contiguous container
(std::vector, etc). Those will all implicitly convert to SkSpan, making
call-sites easy to read/write.
Before making it part of Skia's public API, this CL removes parts of the
API that aren't present in std::span -- when Skia moves to C++20, this
should allow us to use std::span instead, and remove SkSpan entirely.
The most disruptive change is the removal of `int count()`. That's
replaced by a new free function `SkCount` that works on any container,
and incorporates the overflow check from SkTo<int>.
Change-Id: I86cdfad1700f341da3b81a965b396ca9d8b79df9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/405816
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Add an assert in SkTHashTable::uncheckedSet that each added key is equal
to itself. This is an implied precondition which is easy to get wrong if
the key contains floating point values which may become NaN.
Motivated by https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=30314
which would have been much easier to track down with this assert, since
the issue would have been discovered at insertion instead of much later
on removal.
Change-Id: I4eabb6892d2bff1e7bc33c04fd6b297b189a3b02
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/405695
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
All internal usage has migrated to MakeFor..., this removes the old
program kind, and updates some tests.
Bug: skia:11813
Change-Id: I56733b071270e1ae3fab5d851e23acf6c02e3361
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/402536
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Simplifies SampleUsage quite a bit (no need to track multiple kinds of
sampling, variable matrices don't exist any more, etc...).
Change-Id: I58b8de7218d00c4d882d2650672e5fe01892a062
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/402177
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
They're only used for the legacy implementation of sk_float_rsqrt,
which is going away.
SkFloatingPoint.h is part of ca 10,000 translation units when building
Chrome, and include graph analysis shows that including xmmintrin.h
adds a total of 3.5 GB to the build.
Avoiding these includes shaves ca 7 CPU-minutes off a local Chrome
build.
Bug: skia:11861 chromium:242216
Change-Id: Ib0fd53c4fb3ae254814fb6ae5afbb298efb0eb83
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/400437
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
The instructions used to implement it don't deliever
the same results across processors, even on the same
architecture.
Minor diffs on lighting image filter GMs;
this will likely require staged rollout.
Bug: skia:11861
Cq-Include-Trybots: luci.chromium.try:linux-blink-rel
Change-Id: I4c0b0eab221a051f863f4949051f50652466651e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/396723
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Mixed samples is no longer relevant for Ganesh. DMSAA and the new
Ganesh architecture both rely on full MSAA, and any platform where
mixed samples is supported will ultimately not use the old
architecture.
Change-Id: I5acc745010e090ef26310d92ec6240be2cd494cf
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/399837
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
This is another strange, experimental feature that clutters the
implementation and isn't used by anyone (to my knowledge).
Change-Id: I538b7eca0cd28aab32f4739b23459731ade9105e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/398226
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This was an experimental feature. It worked (but only the GPU backend).
It was never adopted or used by anyone, to my knowledge. It's a large
amount of code, and a strange corner of SkSL for users to stumble into.
Bug: skia:10680
Change-Id: I0dda0364bce7dbffa58c32de4c7801ec2a6bc42e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/398222
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
These enforce stricter rules about the signature of main, and each one
uses a separate pre-include module. That prevents color filters from
being able to reference sk_FragCoord (or coords passed to main) at all.
It also limits the versions of sample() that are exposed.
In the new world, an effect created for a specific stage of the Skia
pipeline can only be used to create instances of that stage (SkShader or
SkColorFilter). For now, SkRuntimeEffect::Make uses kRuntimeEffect,
which continues to be more lenient and allow creation of either shaders
or color filters from a single effect. After we migrate all clients, we
can deprecate and then delete that mode.
Bug: skia:11813
Change-Id: I0afd79a72beeec84da42c86146e8fcd8d0e4c09f
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/395716
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
The instructions used to implement it don't deliever
the same results across processors, even on the same
architecture.
Bug: skia:11861
Cq-Include-Trybots: luci.chromium.try:linux-blink-rel
Change-Id: I7831640fa6c0f01540619fce3f509c8969aaaaff
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/396722
Reviewed-by: Chris Dalton <csmartdalton@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
The instructions used to implement it don't deliever
the same results across processors, even on the same
architecture.
Bug: skia:11861
Change-Id: Ifb87fc04a4ec866f84d7c2f2472c7f4b65661d28
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/396721
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Bug: skia:11803
Change-Id: If4c2280585e192eec1b1588dfe022149543584bb
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/391917
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Prevents introduction of a bug when modifying a type that is used
with one of these containers in such way that makes the use
unsafe.
Change-Id: I959e6f3bfbc9664a6b0ced636f5ae6cbe72eee1a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/390676
Reviewed-by: Herb Derby <herb@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
std::result_of is deprecated and removed in C++20, std::invoke_result is
in C++17 the library so not ready yet.
In the case of SkMakeArray there is no need for the full power of
std::invoke_result as SFINAE is not needed, nor is calling a method
supported. As a result a simple decltype is sufficient.
Bug: skia:11811
Change-Id: I6c07c380970cc2a991ab8bd457997320ba634e6c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/390476
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
This reverts commit 8b6788079f.
Reason for revert: Fixed variable type for chrome & flutter
Basically, you can't use objc_externally_retained with `auto` type
variables because the compiler isn't smart enough to deduce the type
before checking that it's compatible with the attribute.
Original change's description:
> Revert "Use objc_externally_retained to cut down retain/release in Metal"
>
> This reverts commit 78381ee5e0.
>
> Reason for revert: breaking chromium and flutter mac builds
>
> Original change's description:
> > Use objc_externally_retained to cut down retain/release in Metal
> >
> > Until we get off of ARC, these extra retains and releases can really
> > add up in previous profiling of Metal, especially on iOS.
> >
> > Change-Id: Ib3aaa0b863c4469bdb6ad930395a170719c0293d
> > Reviewed-on: https://skia-review.googlesource.com/c/skia/+/381110
> > Commit-Queue: Adlai Holler <adlai@google.com>
> > Reviewed-by: Jim Van Verth <jvanverth@google.com>
> > Reviewed-by: Brian Salomon <bsalomon@google.com>
>
> TBR=jvanverth@google.com,bsalomon@google.com,adlai@google.com
>
> Change-Id: I36aff7698a70052a42b91a910b92fd512b3eecd4
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/388717
> Reviewed-by: Michael Ludwig <michaelludwig@google.com>
> Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Change-Id: I5cad02fc9d8c7f21e22021858b2e6e9b038a3de6
Cq-Include-Trybots: luci.skia.skia.primary:Canary-Flutter
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/389157
Commit-Queue: Adlai Holler <adlai@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
This reverts commit 78381ee5e0.
Reason for revert: breaking chromium and flutter mac builds
Original change's description:
> Use objc_externally_retained to cut down retain/release in Metal
>
> Until we get off of ARC, these extra retains and releases can really
> add up in previous profiling of Metal, especially on iOS.
>
> Change-Id: Ib3aaa0b863c4469bdb6ad930395a170719c0293d
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/381110
> Commit-Queue: Adlai Holler <adlai@google.com>
> Reviewed-by: Jim Van Verth <jvanverth@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
TBR=jvanverth@google.com,bsalomon@google.com,adlai@google.com
Change-Id: I36aff7698a70052a42b91a910b92fd512b3eecd4
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/388717
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Until we get off of ARC, these extra retains and releases can really
add up in previous profiling of Metal, especially on iOS.
Change-Id: Ib3aaa0b863c4469bdb6ad930395a170719c0293d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/381110
Commit-Queue: Adlai Holler <adlai@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Using the dedicated function is slightly faster than calling snprintf
and is documented to behave the same.
Change-Id: I9bc64066b55cf74d2369c531283c68e05bcc402c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/386676
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Bug: skia:11760
Change-Id: I1baaec529b47954018d000856912f121b4f1454a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/386597
Reviewed-by: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Eric Boren <borenet@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
As you might expect, a function tagged with `noinline` will never be
considered as a candidate for inlining.
Change-Id: Ia098f8974e6de251d78bb2a76cd71db8a86bc19c
Bug: skia:11362
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/382337
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
It turned out that everywhere we were using or testing DSL code either
directly or indirectly imported big chunks of the SkSL library. These
imports turned out to be necessary; code written using just DSL.h would
fail with various template instantiation errors.
Change-Id: Iae72d15b0d6ef14614ac1a4ff08c36bc1876cd4d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/381638
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
In addition to the unsurprising changes to eliminate references to
src/, we also had to tighten up some C++17-isms as they are not
permitted in public headers.
Change-Id: Ie5005a33d7a135e69fb66beca5e7a5f960dbd453
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/378496
Reviewed-by: Brian Osman <brianosman@google.com>
All layout(key) fields include the field name meta-data, and use as few
bits as possible.
Bug: skia:11372
Change-Id: Ie12b3e0d01148457e5ea078cbf7d0a4bff35302e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/378596
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
The automatic copy and move constructors and such are not appropriate
for this type. It could be refactored to be movable but not copyable,
but as written it's neither.
Change-Id: I398a1f91238530c7a6ae2aaf3e7e8455691655fa
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/377536
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
A given object may be unique, but its owner may not be unique
and another thread may ref the object via the owner at any time.
This happens e.g. when sharing SkPictures between threads.
Bug: skia:10286
Change-Id: I51b5239338a81aaa4d67db05f01f2c7c24182096
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/376619
Reviewed-by: Ben Wagner <bungeman@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Adlai Holler <adlai@google.com>
This detaches PromiseImages from any specific context, just to
a certain family.
Next up is to remove the tileSpecificSKP code from the DDLTileHelper.
Currently we have this janky PromiseImageDummy GrImageContext that
we make for each promise image. It's not ideal but it'll tide us over.
Bug: skia:10286
Change-Id: I12ab0bb7df9360a08af594da80de9df14cc2a44f
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/372516
Commit-Queue: Adlai Holler <adlai@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
This enforces write-only access to the mapped buffers, will enable
chaining of indirect strokes, and gives us the ability to reorder the
fields for Metal.
Bug: chromium:1172543
Bug: skia:11291
Bug: skia:10419
Change-Id: I4449ff85dd0019f6d6d6781ede52bcf26dee8b02
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/367416
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
This is a reland of fa87f9c7b9
Went back to original approach of checking isFramebufferOnly outside
GrCopyRenderTask::Make().
Original change's description:
> Add backend texture and backend render target versions snapshot GMs
>
> GPU may behave differently depending on whether the SkSurface backing
> store is wrapped and whether it is a texture or not.
>
> Bug: skia:11208
> Change-Id: I5e9921d56c0840cfe34ed2926a55be7460409b23
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/364639
> Reviewed-by: Adlai Holler <adlai@google.com>
> Commit-Queue: Brian Salomon <bsalomon@google.com>
Bug: skia:11208
Change-Id: I7af6078e904fdd772b97ad3a27e26cab577fc6a3
Cq-Include-Trybots: luci.skia.skia.primary:Test-Ubuntu18-Clang-Golo-GPU-QuadroP400-x86_64-Debug-All-PreAbandonGpuContext
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/365702
Reviewed-by: Adlai Holler <adlai@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This reverts commit fa87f9c7b9.
Reason for revert: assert on n5x/n7
Original change's description:
> Add backend texture and backend render target versions snapshot GMs
>
> GPU may behave differently depending on whether the SkSurface backing
> store is wrapped and whether it is a texture or not.
>
> Bug: skia:11208
> Change-Id: I5e9921d56c0840cfe34ed2926a55be7460409b23
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/364639
> Reviewed-by: Adlai Holler <adlai@google.com>
> Commit-Queue: Brian Salomon <bsalomon@google.com>
TBR=bsalomon@google.com,robertphillips@google.com,adlai@google.com
Change-Id: Id1488586e2d17cdce35f95ea6aa9b2bffd9218d0
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:11208
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/365698
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
GPU may behave differently depending on whether the SkSurface backing
store is wrapped and whether it is a texture or not.
Bug: skia:11208
Change-Id: I5e9921d56c0840cfe34ed2926a55be7460409b23
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/364639
Reviewed-by: Adlai Holler <adlai@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
The goal here is remove the need for all the specialized vulkan buffer
subclasses and to not have the new class use GrVkResources for tracking
lifetime on command buffers.
This CL just makes the new class and it is not actually used by anything
yet.
Bug: skia:11226
Change-Id: I5f8d8d112af773ba1e8da17e07e75f6f4100e927
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/364617
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
These are basic vector types, required by GLSL ES2, but we could not
create helper functions using them because they were missing from our
GrSLType enum. (This also prevented Runtime Effects from using these
types in helper functions.)
Change-Id: I78c328499e8ed90cb29c641b90ee59460a5a45de
Bug: skia:11246
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/364036
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
These types are supported by SkSL but were mysteriously absent from
GrSL.
Change-Id: Id3479a23b1ddee0604362ed8c12da0eea7c6fa56
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/355981
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
This is a reland of 6ea387e7c7
Original change's description:
> Enable GrTessellationPathRenderer by default
>
> Moves GrTessellationPathRenderer to the end of the chain and enables
> it by default.
>
> Also updates nvpr to not draw volatile paths. The tessellator is much
> faster at these.
>
> Bug: skia:10419
> Change-Id: I97ca7d4d1dff65fc9d4040c267f9808c8c33b548
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/344377
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
Bug: skia:10419
Change-Id: I53b50d6aa01ae1bea697e05c3f60d57d132d6aad
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/345759
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
To unblock https://skia-review.googlesource.com/c/skia/+/347616:
In file included from ../../../../../../skia/tools/public_headers_warnings_check.cpp:8:
In file included from gen\skia.h:19:
In file included from ../../../../../../skia\include/core/SkContourMeasure.h:11:
In file included from ../../../../../../skia\include\core\SkPath.h:13:
In file included from ../../../../../../skia\include/private/SkPathRef.h:16:
In file included from ../../../../../../skia\include/private/SkIDChangeListener.h:13:
../../../../../../skia\include/private/SkTDArray.h(76,48): error: implicit conversion changes signedness: 'const int' to 'unsigned long long' [-Werror,-Wsign-conversion]
!memcmp(a.fArray, b.fArray, a.fCount * sizeof(T)));
~~^~~~~~ ~
...
Change-Id: I1b55580ae5e73bde41cecb20d137295bbaf8cbb6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/347860
Reviewed-by: Joe Gregorio <jcgregorio@google.com>
Commit-Queue: Florin Malita <fmalita@google.com>
Updated Pair type in SkTHashMap to derive from std::pair to fix C++14
issues with structured bindings.
Original change's description:
> Add support for range-based for loops to SkTHashSet/Map.
>
> This allows loops over SkTHashes to break in the middle, and also
> removes the need to use lambda captures to bring variables inside the
> loop's scope.
>
> Change-Id: Ief55d776b2c57a44b24cfe1c94493a5d514791c8
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/346496
> Reviewed-by: Mike Klein <mtklein@google.com>
> Commit-Queue: John Stiles <johnstiles@google.com>
Change-Id: I2ac5b2c59e70ed0ec3b42b32e7994d6bcdf56b40
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/346502
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
This reverts commit 5d00e15625.
Reason for revert: tree breakage on Chromebook standard lib
Original change's description:
> Add support for range-based for loops to SkTHashSet/Map.
>
> This allows loops over SkTHashes to break in the middle, and also
> removes the need to use lambda captures to bring variables inside the
> loop's scope.
>
> Change-Id: Ief55d776b2c57a44b24cfe1c94493a5d514791c8
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/346496
> Reviewed-by: Mike Klein <mtklein@google.com>
> Commit-Queue: John Stiles <johnstiles@google.com>
TBR=mtklein@google.com,johnstiles@google.com
Change-Id: I165872ac41f66f3b3255cf8970626392e5283412
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/346500
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
This allows loops over SkTHashes to break in the middle, and also
removes the need to use lambda captures to bring variables inside the
loop's scope.
Change-Id: Ief55d776b2c57a44b24cfe1c94493a5d514791c8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/346496
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
This reverts commit 6ea387e7c7.
Reason for revert: valgrind failing
Original change's description:
> Enable GrTessellationPathRenderer by default
>
> Moves GrTessellationPathRenderer to the end of the chain and enables
> it by default.
>
> Also updates nvpr to not draw volatile paths. The tessellator is much
> faster at these.
>
> Bug: skia:10419
> Change-Id: I97ca7d4d1dff65fc9d4040c267f9808c8c33b548
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/344377
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
TBR=bsalomon@google.com,csmartdalton@google.com
Change-Id: I3718ae210ebc403959d187160214cf0df895e4f5
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:10419
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/345718
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Moves GrTessellationPathRenderer to the end of the chain and enables
it by default.
Also updates nvpr to not draw volatile paths. The tessellator is much
faster at these.
Bug: skia:10419
Change-Id: I97ca7d4d1dff65fc9d4040c267f9808c8c33b548
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/344377
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Adds a new kDirectionLight_ShadowFlag which indicates that lightPosition
is a vector pointing towards the light, and lightRadius is the blur
radius when the occluder is at z == 1. For other values of z it will
scale linearly. The direction is specified in device space.
Bug: skia:10781
Change-Id: I14d530f006e5541ed6e22bfbeb29c7441c21fb8e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/321792
Commit-Queue: Jim Van Verth <jvanverth@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Just the class/files. variable names and additional comments to follow.
Change-Id: Ic03d07fd5009eaf3d706c2536486a117328963fc
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/342617
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Initially we restricted tessellated stroking to opaque solid colors.
This CL adds support for transparency by enabling a stencil test. The
stencil test also allows us to use mixed samples.
Bug: skia:10419
Change-Id: Ie40f3099d2b009f92ed49f7f43e5f269b1a479af
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/340798
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Follow-on to https://skia-review.googlesource.com/c/skia/+/342616
Assuming this survives, we can probably simplify other code in SkPath
(e.g. Iterators) that today worry about consing up moves after close.
Change-Id: I913d6be9c20547267230c287d394fd300b985d31
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/342619
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
These macros are not meant to leak out of the file.
Change-Id: I7e24f65a3053785410c7fac760fd3af46c5c1f1c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/337739
Auto-Submit: Mike Klein <mtklein@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Chris Dalton <csmartdalton@google.com>
It's a bad idea, difficult to reason about, and may be causing
deadlocks.
Change-Id: Id9749661f4f3f942ee983e9c1fdab2bd7f287edb
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/335242
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
Previously, `checkRealloc` would always size up the requested buffer to
allow for 50% extra growth. In some cases (e.g. repeated push_back),
this is a clear win for performance. In other cases (e.g. assigning one
array to another, reserve_back), this extra padding is typically
unwanted and goes against most caller's expectations.
Change-Id: I2d2b5cf81268026822dc5ea08396771bbf06b25a
Bug: skia:10930
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/333797
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Lots of this stuff can be delegated to each other,
cutting the protected SkTArray constructors to two.
Change-Id: Ie35b7a5ceb0ffef5a9548afccc546e076bd668cc
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/333256
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This decreases alignment for things like float
and increases alignment for things like __m128.
The other users of SkAlignedSStorage looked simple enough to port to
aligned char[]. I haven't changed either of their alignments---still
the old max(void*,double)---but we can now if we want.
All that together lets us delete SkAlignedSStorage.
Change-Id: I6b5957a26f42ad859de383054573fb58d5cd0576
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/333196
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Using private inheritance is similar to having a class member,
except it's initialized before the next base class, SkTArray.
This lets us pass it to SkTArray's constructors.
I think we can make related changes (updating the various SkAlignedFoo,
not using them here, or not using them anywhere) independently.
... storage constructors made explicit at suggestion of GCC's -Wextra.
... now with explicit static_cast<STORAGE*>(this)
Change-Id: I665cf840e111da68f039416c9649ce328cc308d6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/333146
Reviewed-by: John Stiles <johnstiles@google.com>
f01a9d9020
is the culprit
Revert "GrRefCntedCallback has Make function."
This reverts commit b2c42140ea.
Revert "Add SkImage::MakeFromYUVATexturesCopyToExternal"
This reverts commit f01a9d9020.
Bug: skia:10632
Change-Id: Ief076f168b63ff8ca15b607163a13d5f52a733d2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/331798
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Each caller does not need to check for null function.
Also inherit from SkNVRefCnt.
Change-Id: I9a53c3c8e9dec1361db6cfb21b97d178173b23ac
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/331490
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Also added unit tests for each of SkTArray's various constructors, and
added `SkTArray::value_type` which allows calling code to refer to the
array's value-type. These unit tests exposed some preexisting strict-
aliasing issues in SkSTArray when compiled on GCC 6+ with optimizations
enabled, which are being investigated separately at skia:10891.
Change-Id: Ia0fb18830cfbbdcb1545fe7f7ac51d8e768a3f94
Bug: skia:10891
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/330279
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
With the addition of vertex data to the thread safe cache we also have
to handle the case where a given SkPath becomes inaccessible and
proactively invalidate the matching entry.
Bug: 1108408
Change-Id: Id11ce2aa10517f7c0772a253634d3c0d13e13460
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/330261
Reviewed-by: Adlai Holler <adlai@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This reverts commit 375721d7bb.
Some manual edits needed due to time elapsed.
Bug: chromium:1141332, skia:10566
Change-Id: Iadb15d3f5334d9eed4e7053e9c19d75a0bbeb9de
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/330196
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Adds an SkChopCubicAt overload that performs two chops at once in
SIMD. Also updates SkChopCubicAt to accept T values of 0 and 1. This
has been the source of bugs in the past.
Bug: skia:10419
Change-Id: Ic8a482a69192fb1685f3766411cbdceed830f9b7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/327436
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Bug: skia:10866
Change-Id: I06c675a423380d30017ebcc5485d556211d403fa
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/329390
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Auto-Submit: Mike Klein <mtklein@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
It's just much easier to remember and think about
max(lo, min(x, hi)) than max(min(hi, x), lo), and
both pin NaN to one of the two limits. I'm not sure
if anything in Skia depends on which limit we pin to.
Change-Id: Iceca36a8fffd7072180e82b8b6eb81cbdb5ac97f
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/327788
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Auto-Submit: Mike Klein <mtklein@google.com>
Reviewed-by: Chris Dalton <csmartdalton@google.com>
Change-Id: I70b2fdea570a9091afc81a1455fa61f90b0357a1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/327786
Reviewed-by: Chris Dalton <csmartdalton@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
The semantics of `vector::reserve` and `SkTArray::reserve` were not the
same. SkTArray::reserve takes a delta over the current array size,
whereas vector takes a total array size. This could lead to subtle
errors with over- or under-reservation, hurting performance.
This CL renames `SkTArray::reserve` to `SkTArray::reserve_back` to give
the SkTArray behavior a distinct (hopefully easily understandable) name,
leaving its functionality as-is.
Change-Id: Icbd3114bb317fd5f307f393c02ae6fb6f83764e9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326956
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Instead of looking at point/verb/weight counts, add an
SkPathRef::approximateBytesUsed() using their reserve counts.
This shows SkPathBuilder::snapshot() can return more memory-efficient
SkPaths than SkPathBuilder::detach(), at the cost of a copy.
Change-Id: I4e208c41643480d7682daba6ac674ffa63c74de2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326608
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@google.com>
Now that we have skvx::map(), anyone can write this sort of
scalar-to-vector code. There are no vector instructions for these, so
they'll never going to be particularly interesting for SkVx to provide.
We did work out _approximate_ versions of each of these for SkVM, and
that's what we use to evaluate these programs there. So if this stuff
really matters we could port that logic back over to SkVx.h.
But in terms of pure refactoring, I think this is where we want to sit
until we decide to use those approximations. I don't really want to
invest much time in the SkSLByteCode interpreter any more.
Change-Id: I4e595dee5fd9e608905305e46b2aebcab986c561
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326277
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: Mike Klein <mtklein@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Cq-Include-Trybots: luci.chromium.try:linux_chromium_cfi_rel_ng
Bug: chromium:1137652, chromium:1137958
Change-Id: I8575b588f9a1ba89740b95382b2462338e34bec5
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326478
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Cut & paste with some reorganization of includes & forward-decls.
Next up is GrContextPriv.
Change-Id: I72d2d95c62692e3b37608517b796c0041ffedea3
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326157
Reviewed-by: Robert Phillips <robertphillips@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Adlai Holler <adlai@google.com>
This is a reland of b3c42efd25
with a few more signed/unsigned mismatches fixed up.
Original change's description:
> pack SkTArray
>
> Same idea as http://review.skia.org/325857; I just wanted to feel out
> the options myself. A couple key ideas:
>
> Prefer SkTo32() to separate range assertions and casts.
>
> Keep using int indices, counting to this->count() where
> fCount would have warned about signed/unsigned mismatch.
>
> I've kept new comments and assertions to minimum. In the end we won't
> change the max size of SkTArray, and I don't see much need to call out
> how much it hasn't changed. Reading back over the accumulated changes,
> I don't really see much that's newly error-prone.
>
> Change-Id: I86a8a161b9ae44f24fc25093741945b75fbfa770
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326106
> Commit-Queue: Mike Klein <mtklein@google.com>
> Reviewed-by: John Stiles <johnstiles@google.com>
Change-Id: I04d928ecc185fb5c7b9d32f60e94df9f8d137898
Cq-Include-Trybots: luci.skia.skia.primary:Build-Win-MSVC-x86_64-Debug-Vulkan,Build-Win-MSVC-x86-Debug
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326297
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Cut & paste job, but I did replace some cases of just calling
inherited implementations with using-statements. There are
other methods on this class that are using-statements and
that pattern is cleaner.
Change-Id: Ie369c643e44bdb8f82dfffcf45c1f65d48606899
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/325660
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Auto-Submit: Adlai Holler <adlai@google.com>
This reverts commit b3c42efd25.
Reason for revert: Build-Win-MSVC-x86_64-Debug-Vulkan
Original change's description:
> pack SkTArray
>
> Same idea as http://review.skia.org/325857; I just wanted to feel out
> the options myself. A couple key ideas:
>
> Prefer SkTo32() to separate range assertions and casts.
>
> Keep using int indices, counting to this->count() where
> fCount would have warned about signed/unsigned mismatch.
>
> I've kept new comments and assertions to minimum. In the end we won't
> change the max size of SkTArray, and I don't see much need to call out
> how much it hasn't changed. Reading back over the accumulated changes,
> I don't really see much that's newly error-prone.
>
> Change-Id: I86a8a161b9ae44f24fc25093741945b75fbfa770
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326106
> Commit-Queue: Mike Klein <mtklein@google.com>
> Reviewed-by: John Stiles <johnstiles@google.com>
TBR=mtklein@google.com,johnstiles@google.com
Change-Id: I8267a05e1ce075bfa7893c60f7c77dbf4f8cf061
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326296
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
These don't return reliable portable results, so I don't want to promote
them as good ideas to use. You can get at least 5 different results
from these across the four main architectures we support, and they've
been the root cause of bugs uncovered only in production on undertested
platforms.
Luckily, unused outside of tests.
Change-Id: I532731fe4cddf127253341e5ace8d9c5c9ebb0f1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326108
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Same idea as http://review.skia.org/325857; I just wanted to feel out
the options myself. A couple key ideas:
Prefer SkTo32() to separate range assertions and casts.
Keep using int indices, counting to this->count() where
fCount would have warned about signed/unsigned mismatch.
I've kept new comments and assertions to minimum. In the end we won't
change the max size of SkTArray, and I don't see much need to call out
how much it hasn't changed. Reading back over the accumulated changes,
I don't really see much that's newly error-prone.
Change-Id: I86a8a161b9ae44f24fc25093741945b75fbfa770
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/326106
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This basically adds the functionality to GrResourceProvider to handle
scrate MSAA attachments. There are not current users of this, but
users will be added in follow on changes.
Bug: skia:10727
Change-Id: Ieb8d247e034fb22ac9ff4fc549935310329a1c1e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/320267
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Just a little refactor no-op.
Change-Id: I1842a0190cd96c60da2fe3c7f88fa56c9f73af81
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/325681
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Rewrite map() to allow any number of arguments,
now also used for 2-argument (pow) and 3-argument (fma) operations.
I left a note about fma()... I can't understand why, but calling as
map(fmaf, x,y,z) ends up with scalar calls to fmaf(), but with the
lambda indirection we see perfect vector codegen.
I had to break map() back into two parts. I don't see any way to pass
both a variadic number of arguments and play our trick with the default
std::index_sequence parameter. The lane lambda similarly exists only to
split up the expansion of the Rest... type pack from the I... index
pack; you can't use two pack expansions in the same expression.
Change-Id: Ia156a7fd846237f687d6018a7f95550c9fd4a56d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/325736
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Herb Derby <herb@google.com>
A change made in the spirit of not being weird. C++ containers typically
expose their capacities to non-test code, with a function "capacity."
Change-Id: Icc7e175a20aff53ef9e144ac9620ced29ef5e95a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/325657
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
The call order here is a little weird – it could use some refactoring
but for now it's just a cut-and-paste job.
Change-Id: I931493a421f7fd1df69f0c47055b90a2996c7dc3
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/325459
Reviewed-by: Robert Phillips <robertphillips@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Commit-Queue: Adlai Holler <adlai@google.com>
Auto-Submit: Adlai Holler <adlai@google.com>
This is pretty much just cutting & pasting the code naïvely,
but it's the safe way to go so that we don't impact the
order of events during this housekeeping.
Change-Id: I8d9f2ffb560b34fd1a8d9603174ff1f17803e447
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/324619
Commit-Queue: Adlai Holler <adlai@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Auto-Submit: Adlai Holler <adlai@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This reverts commit 48d8724097.
Reason for revert: Breaking MSVC bots
Original change's description:
> Merge GrContext::init into GrDirectContext ...
>
> And minor cleanups associated with the initialization of these classes.
>
> Change-Id: Ida0372d0b1a0b8bf5b309814de5418e47ea34fdb
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/324122
> Commit-Queue: Adlai Holler <adlai@google.com>
> Reviewed-by: Robert Phillips <robertphillips@google.com>
> Auto-Submit: Adlai Holler <adlai@google.com>
TBR=robertphillips@google.com,adlai@google.com
Change-Id: I00b2ddaeaef53e3fe05a338c88158cf235f325c4
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/324132
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
And minor cleanups associated with the initialization of these classes.
Change-Id: Ida0372d0b1a0b8bf5b309814de5418e47ea34fdb
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/324122
Commit-Queue: Adlai Holler <adlai@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Auto-Submit: Adlai Holler <adlai@google.com>
Like the destructors, most of these can just go unspoken.
SkTArray only knows its size in Debug builds, so it can't easily grow a
copy constructor. Instead just keep the explicit copies and moves on
SkTHashTable, and mark the others as explicitly default. I inverted the
pattern from assign-by-copy to copy-by-assign, but it's all basically
the same.
Change-Id: I0df552ef547ce9c7e232267c072a5f3b09197847
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/324056
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This will enable us to use SkTHashMap to store our definition maps.
Change-Id: I6017dfa71e1c5e68a20c97e955bb3d3abf347f0d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/323891
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Currently there is nothing using these formats, but this change just adds
them along with a quaries about their properties. They will be used in a
follow up change.
Bug: skia:10727
Change-Id: Iaaf13baf372799d47c65bd974fd204a32be57617
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/322622
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
and begin using it for cached SW-generated blur masks.
This is needed to begin mixing and matching HW & SW-generated blur
masks since they have different draw-rects.
It will also be useful if/when we add support for triangulated paths
to the thread-safe cache.
Bug: 1108408
Change-Id: I085ad1127dc2deb98b35d704b06e50b27c72fd1c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/322657
Reviewed-by: Adlai Holler <adlai@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
I wrote code that called SkAutoTArray::data() and discovered that it
was broken, but not generating compile errors because it's part of a
template and never instantiated anywhere else. I fixed the
implementation and added it to our container unit test to prevent later
regression. This revealed another issue, that "containers in
SkTemplates.h [should] all have a consistent api", according to
test_container_apis. However, data() was never added to the non-array
container APIs. So I added data() to the other containers as well.
Change-Id: I52532c91fdab3fc8c4539053ba8420815b7b0ee5
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/323276
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This is a reland of e24f7f3de7
... with fix for ~0 constants for the pedantic MSVC.
Original change's description:
> refactor any()/all(), ptest for all()
>
> Part of this is a simple refactor, adapting any() and all() to the new
> style of specialization.
>
> And with that refactor in place, add AVX2/SSE4.1 for all() using ptest.
> This isn't terribly important, but it does help make Op::asserts run
> faster in the SkVM interpreter. I like to run with asserts enabled, and
> this makes passing asserts much cheaper---failing asserts are expensive
> still of course, printing to SkDebugf(), etc.
>
> Change-Id: Iebdeee701fab7c50cce8e457674b565f7dd2ec21
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317422
> Reviewed-by: Herb Derby <herb@google.com>
> Commit-Queue: Mike Klein <mtklein@google.com>
Cq-Include-Trybots: luci.skia.skia.primary:Build-Win-MSVC-x86_64-Debug
Change-Id: I93f08177ef3439e65e4383cc517dba60c0c4ef3e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317638
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Herb Derby <herb@google.com>
This is a reland of 4985db413d
...with a better implementation of map(). I don't understand
why we had to revert, but it had something with calling the
function pointer in map_(), so maybe this will help.
I've flattened the map_() / map() merge CL into this one,
and marked the resulting map() as no_sanitize("cfi"). I
don't see anything wrong, so I think it's a false positive.
Original change's description:
> update skvx scalar-fallback strategy
>
> Turns out Clang's a lot better at auto-vectorizing "obvious" scalar code
> into obvious vector code when it's written out the long way, e.g.
>
> F32x4 x = ...;
> x = { sqrtf(x[0]), sqrtf(x[1]), sqrtf(x[2]), sqrtf(x[3]) };
>
> vectorizes into sqrtps a lot more reliably than our recurse-onto-scalars
> strategy, and also better than the other naive approach,
>
> F32x4 x = ...;
> for (int i = 0; i < 4; i++) { x[i] = sqrtf(x[i]); }
>
> So here I've added a map(V, fn) -> V' using C++14 tricks to let the
> compiler handle the expansion of x = { fn(x[0]), fn(x[1]), ...
> fn(x[N-1]) } for any N, and implemented most skvx scalar fallback code
> using that.
>
> With these now vectorizing well at any N, we can remove any
> specializations we'd written for particular N, really tidying up.
>
> Over in the SkVM interpreter, this is a big improvement for ceil and
> floor, which were being done 2 floats at a time instead of 8. They're
> now slimmed way down to
>
> shlq $6, %r13
> vroundps $K, (%r12,%r13), %ymm0
> vroundps $K, 32(%r12,%r13), %ymm1
> jmp ...
>
> where K is 9 or 10 depending on the op.
>
> I haven't found a scalar function that Clang will vectorize to vcvtps2pd
> (the rounding one, not truncating vcvttps2pd), so I've kept lrint()
> written the long way, updated to the style I've been using lately with
> specializations inline.
>
> Change-Id: Ia97abe3c876008228bf62b1daacd6f6140408fc4
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317375
> Reviewed-by: Herb Derby <herb@google.com>
> Commit-Queue: Mike Klein <mtklein@google.com>
Cq-Include-Trybots: luci.chromium.try:linux_chromium_cfi_rel_ng
Bug: chromium:1129408
Change-Id: Ia9c14074b9a14a67dd221f4925894d35a551f9d7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317551
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Herb Derby <herb@google.com>
This CL is also imperfect and incomplete but, although currently unused, it sketches in how the threadSafeProxyCache will be plumbed through the GrContexts and GrResourceCache.
Bug: 1108408
Change-Id: Idb012b6efd49291de69bd88e4b4c531458a3e553
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317360
Reviewed-by: Adlai Holler <adlai@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This reverts commit 4985db413d.
Reason for revert:
../../third_party/skia/include/private/SkVx.h:491:14: runtime error: control flow integrity check for type 'float (float)' failed during indirect function call
(/lib/x86_64-linux-gnu/libm.so.6+0x36460): note: (unknown) defined here
../../third_party/skia/include/private/SkVx.h:491:14: note: check failed in /b/s/w/ir/out/Release/viz_unittests, destination function located in /lib/x86_64-linux-gnu/libm.so.6
#0 0x55e964d3c1f9 in skvx::Vec<4, float> skvx::map_<4, float, float, 0ul, 1ul, 2ul, 3ul>(skvx::Vec<4, float> const&, float (*)(float), std::__1::integer_sequence<unsigned long, 0ul, 1ul, 2ul, 3ul>) ./../../third_party/skia/include/private/SkVx.h:491
I don't understand what's wrong here, but I have a better map() coming up anyway.
Original change's description:
> update skvx scalar-fallback strategy
>
> Turns out Clang's a lot better at auto-vectorizing "obvious" scalar code
> into obvious vector code when it's written out the long way, e.g.
>
> F32x4 x = ...;
> x = { sqrtf(x[0]), sqrtf(x[1]), sqrtf(x[2]), sqrtf(x[3]) };
>
> vectorizes into sqrtps a lot more reliably than our recurse-onto-scalars
> strategy, and also better than the other naive approach,
>
> F32x4 x = ...;
> for (int i = 0; i < 4; i++) { x[i] = sqrtf(x[i]); }
>
> So here I've added a map(V, fn) -> V' using C++14 tricks to let the
> compiler handle the expansion of x = { fn(x[0]), fn(x[1]), ...
> fn(x[N-1]) } for any N, and implemented most skvx scalar fallback code
> using that.
>
> With these now vectorizing well at any N, we can remove any
> specializations we'd written for particular N, really tidying up.
>
> Over in the SkVM interpreter, this is a big improvement for ceil and
> floor, which were being done 2 floats at a time instead of 8. They're
> now slimmed way down to
>
> shlq $6, %r13
> vroundps $K, (%r12,%r13), %ymm0
> vroundps $K, 32(%r12,%r13), %ymm1
> jmp ...
>
> where K is 9 or 10 depending on the op.
>
> I haven't found a scalar function that Clang will vectorize to vcvtps2pd
> (the rounding one, not truncating vcvttps2pd), so I've kept lrint()
> written the long way, updated to the style I've been using lately with
> specializations inline.
>
> Change-Id: Ia97abe3c876008228bf62b1daacd6f6140408fc4
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317375
> Reviewed-by: Herb Derby <herb@google.com>
> Commit-Queue: Mike Klein <mtklein@google.com>
TBR=mtklein@google.com,herb@google.com
Change-Id: I27b5eff3328bf2ddf7063ee0dee14a378ff23b89
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317546
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
