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>
This reverts commit e24f7f3de7.
Reason for revert: Build-Win-MSVC-x86_64-Debug
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>
TBR=mtklein@google.com,herb@google.com
Change-Id: Ib3ecbe93aa9d14b10dd87e8aa247f275c2c3eb67
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317545
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
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>
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>
Implement min and max using if_then_else(y<x,...) on vectors
rather than recursing to std::min/std::max applied to scalars.
But actually, factor out and use naive_if_then_else(), which Clang can
reason through better than it can our specialized if_then_else(). This
lets every min() or max() I've looked at compile down to ideal codegen,
vmaxps, vpminsw, etc, where if you use if_then_else() you'd see the
literal comparison and blend as written.
I've been looking at q14x2 codegen in the interpreter, and most things
were already good, unexpectedly even uavg_q14x2. The biggest surprise
was how bad the min/max codegen was, and looking back, even the min_f32
and max_f32 codegen is super bad. This CL fixes all that, leaving us
with the ideal codegen using the specific instruction you'd want,
replacing a giant mess of code that recursed down to scalars.
mul_q14x2 is still bad, but an easy follow up.
Change-Id: I77b5d7c9aa20a9a2f5ceb3e40f1e18ace2a1b5c1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317310
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Change-Id: Ib10215e1e5a86bf78cc34f9dca670417bb217b73
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317271
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
The default implementation of if_then_else is logically bitwise,
(cond & true_val) | (~cond & false_val)
The existing skvx specializations work only for 32-bit lanes, but we can
easily make them work for any type where the whole vector is the right
size by reducing the granularity down to byte level.
Existing code using 32-bit values and 0xffff'ffff or 0x0000'0000 masks
will continue to work the same. But this now lets us use, e.g. 16-bit
values with 0xffff and 0x0000 masks, or even things like 32-bit values
and a mask like 0xff00ff00, selecting byte by byte.
We can't go any lower without falling back on the generic bitwise
implementation, so we'll have to settle for not getting to use a mask
like 0x0f0f0f0f.
Change-Id: I8518cb3cafc7f6e1480b4ae8af50daad2d28c5df
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/317170
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This is also the final wire to connect, so with this CL we will try
using input attachments for dst blends in available.
Bug: skia:10409
Change-Id: I8bd953ea5eb056a55d8bf36d91008a9d7298d84d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/315650
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Greg Daniel <egdaniel@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>
This CL adds a new type GrDstSampleType to say how we will sample the dst.
We add tracking of the GrDstSampleType in the recording of GrOps and
then during execution passing the information along to the GrPipeline.
In general the tracking of GrDstSampleType is a global state of a GrOpsTask
so it is kept separate fro the DstProxyView which is more specific to a
single Op on the GrOpsTask.
Bug: skia:10409
Change-Id: Ie843c31f2e48a887daf96cee99ed159b196cb545
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/315645
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Mechanically updated via Xcode "Replace Regular Expression":
typedef (.*) INHERITED;
-->
using INHERITED = $1;
The ClangTidy approach generated an even larger CL which would have
required a significant amount of hand-tweaking to be usable.
Change-Id: I671dc9d9efdf6d60151325c8d4d13fad7e10a15b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/314999
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
There's an _mm_movemask_ps() intrinsic that gets at the movmskps
instruction, which grabs the top (sign) bit of each float directly
without needing to reinterpret them as bytes.
I wouldn't really have done this but I think Chrome's clang is
miscompiling the version at head that uses _mm_movemask_epi8(). The
SkNx<2,float> `!(a+b == a*b).anyTrue()` test case fails when I use that
compiler, and spooky things like adding SkDebugf() make it pass again.
Change-Id: Idd0698d46ccfe9a00909faca1c6693a70e91157a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/314860
Auto-Submit: Mike Klein <mtklein@google.com>
Commit-Queue: Herb Derby <herb@google.com>
Reviewed-by: Herb Derby <herb@google.com>
- remove some workarounds
- more SI/SIN/SIT/SINT use
- rewrap a lot of code to 100 cols
- etc. misc.
Change-Id: I78b7ff272afcbb8658cf147aad8af85d0e2acf42
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/314676
Auto-Submit: Mike Klein <mtklein@google.com>
Commit-Queue: Herb Derby <herb@google.com>
Reviewed-by: Herb Derby <herb@google.com>
Also, move first-direction into SkPathRef.h so it can be referenced
by name in SkPath (instead of using uint8_t)
No functional change expected.
Change-Id: Ica4a8357a8156fd9a516118f23599a965b0fdd47
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/313980
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
This flag was added recently since I noticed it was missing and should
be checked. However, Dawn fails this check in chrome since it doesn't
report this flag correctly to DDLs. Going back to the previous status
quo for now.
Bug: skia:10672
Change-Id: Ib825fe5a69bff7af0d9893b95cd4df80289be7b2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/313905
Commit-Queue: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Auto-Submit: Greg Daniel <egdaniel@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This has always been a potential source of a bug. If the same texture is
used twice in a shader with different swizzles we would overwrite the
swizzle for the first use by that of the second use since there is
only one fixed function swizzle per texture. It's not part of the
sampler state.
We set the swizzle when it is a feature, but always to RGBA.
Also, highly speculative that this may improve ANGLE D3D11 ES3
performance compared to ES2.
Bug: skia:10644
Change-Id: I8877afc3043c5ddaafd26ea9f9bd372303328c71
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/313682
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This also includes all the plumbing of this flag throughout our proxy
and surface system.
Bug: skia:10409
Change-Id: I48d40012049240cfa80e045ea090f68ce2d2ff0d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/313676
Reviewed-by: Robert Phillips <robertphillips@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Other code outside this namespace will want to use this method soon.
Bug: skia:10419
Change-Id: Ib155f224866fd333b8f9a4b78e6c9e51ac0600df
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/311936
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Hopefully, this will let Chrome track down promiseImage fulfillment
mismatches.
Bug: 1116848
Change-Id: Ia1e5d6f7af4e2808ae4adfad85f4e96c1ea4fbd2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/311096
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
To my surprise, this even works with homegrown smart pointers (such as
SkTLazy).
https://clang.llvm.org/extra/clang-tidy/checks/readability-redundant-smartptr-get.html
Find and remove redundant calls to smart pointer’s .get() method.
Examples:
ptr.get()->Foo() ==> ptr->Foo()
*ptr.get() ==> *ptr
*ptr->get() ==> **ptr
if (ptr.get() == nullptr) ... => if (ptr == nullptr) ...
Change-Id: I8ff541e0229656b4d8e875c8053a7e6138302547
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/310976
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Follow-on CLs will push higher up in SkDraw, so that everywhere today
we have to cons-up (with the associated mallocs) a temp SkPath we can
replace it with a stack-based SPath...
- drawRect
- drawOval
- drawRRect
- drawLine(s)
(similar to how this CL already handled quads and triangles)
Bug: skia:10566
Change-Id: I882b4f4c60e80235ca83c86c926e905b269a7afd
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/307784
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Reed <reed@google.com>
https://clang.llvm.org/extra/clang-tidy/checks/readability-const-return-type.html
`const` on a non-pointer/reference return type typically doesn't add
value and can have negative side effects. (i.e., returning a
`const std::string` isn't meaningfully different from returning a
`std::string`, but can sometimes inhibit move-related optimizations.)
In Skia's case, the priv() functions are a notable exception where const
return types are intentional and valuable. These calls have been marked
with NOLINT to exclude them from the check.
This check does not affect pointer and reference returns, where
constness is important.
Change-Id: I86cab92332f164e5ab710b4127182eec99831d7d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308564
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
This fixes a large number of SkSL namespaces which were labeled as if
they were anonymous, and also a handful of other mislabeled namespaces.
Missing namespace-end comments have been added throughout.
A number of diffs are just indentation-related (adjusting 1- or 3-
space indents to 2-space).
Change-Id: I6c62052a0d3aea4ae12ca07e0c2a8587b2fce4ec
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308503
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
The majority of existing call sites were automatically updated using
clang-tidy -fix. A small handful required a manual update,
e.g. CppCodeGen.
This check is a bit lenient, and in particular will not flag cases like
`std::unique_ptr<Base>(new Derived())` which is still pretty common
throughout our codebase. This CL does not attempt to replace all the
cases that ClangTidy does not flag.
Change-Id: I5eba48ef880e25d22de80f321a68c389ba769e36
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/307459
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Need to notify the underlying pathref if we've made oval or rrect
Bug: skia:9000
Change-Id: I57a801f1fb446b99634d7b028249a812a5a978f1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/307516
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
GCC intrisics type validation is stricter than the one in Clang, so
passing a uint16x4_t to a function expected to accept float16x4_t
is not valid.
Bug: chromium:819294
Change-Id: I6d68e5458345e78bdb05dd028481fe9cae36c5ff
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/307276
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
The ClangTidy check `bugprone-unused-raii` has been enabled at
review.skia.org/306838; this check provides equivalent protection.
Change-Id: I9f3858bfd2bede107d509a5a206a08293d5f914c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/306953
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Bug: chromium:1101491
Bug: b/161896447
Found using
git grep -wiEIl \ '(he)|(she)|(his)|(hers)|(him)|(her)|(guy)|(guys)'
Change-Id: I6b91853de067fd4c2e84f7ec70275522ce6c8bfc
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/306186
Commit-Queue: Leon Scroggins <scroggo@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Auto-Submit: Leon Scroggins <scroggo@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Change-Id: Ic44e24057b95bb014504f02a736fb4341afc8971
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/304856
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
fSubRunPtr is currently mutable, but that will change in future CLs.
Change-Id: Ia3ab40855d7ea7c42eadf8889688fefb064f1bc9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/304696
Commit-Queue: Herb Derby <herb@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
For consistency with other enums and public APIs.
Change-Id: I026da5529f11051693cae5691c7ad92fad5ed446
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/304597
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: I016de62543b5ba16a7193262cea343a77a71ba3b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/304201
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
- add f32<->f16 functions to skvx
- add f32<->f16 x86 instructions to skvm::Assembler
- add f32<->f16 ops to skvm,
using the skvx functions in the interpreter
Still TODO:
use the new x86 instructions in the JIT
(For now like in many other ways, the aarch64 JIT
continues to languish. Will pick that back up one day.)
Change-Id: Ib8dc1ccdc75ecb23769ea4947d66d3ab22520f23
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/302942
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Herb Derby <herb@google.com>
Although it appeared that the experimental_simd CanvasKit build was
working, the build was not producing actual wasm SIMD operations. This
CL fixes that issue by changing the build arguments.
This issue also fixes an incorrect type issue with the SkVx wasm SIMD
implementation.
Bug: skia:10453
Change-Id: If26f84b09e4d84df36be589245878c821972dffc
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/302669
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
As soon as I started migrating Chrome, this function showed its
public usefulness. It'll keep coming up over and over again.
Change-Id: Ic6fd08af9b64a4c3287e7fedc9cd0e29bbaf837d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/302259
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>
First move if_then_else() specializations inline using a
quasi-constexpr-if approach, letting them operate on any types of the
right vector and lane size. We can't use constexpr-if per se because
this header is sometimes used in C++14 contexts.
Then, add AVX specialization for 8x32-bit types.
SkVM's interpreter uses if_then_else() on three i32x16, and these
changes allow that to vectorize ideally, as two vblendvps instructions.
Change-Id: I8355c47975c736c1fbc32b1f8ceddb772978d271
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/302080
Auto-Submit: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This is part of a larger effort to force users to provide a context
when manipulating GPU images which may be shared, instead of having
images themselves retain powerful context pointers.
Chrome flag landed in Chrome CL 2292800
Bug: skia:10466
Change-Id: Ic530a2c5eb1f4399db899d243ea944760fdf2055
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/300707
Commit-Queue: Adlai Holler <adlai@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
