This might be easier than building with SK_USE_SKVM_BLITTER defined.
Change-Id: I72c1f361195bacf5a14e6633dd49e9a5ad721cff
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/264383
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
- allow clients to pass in raw-array at public level
- rewrite m44 concat to take raw-array for 2nd input
Extended canvas_matrix bench to also time 44 concats:
Before
207.51 canvas_matrix_4x4 8888
100.75 ? canvas_matrix_3x3 8888
100.79 canvas_matrix_2x3 8888
140.93 canvas_matrix_scale 8888
133.35 canvas_matrix_trans 8888
After
120.80 canvas_matrix_4x4 8888
...
bug: skia:9768
Change-Id: Iaa361b9897a183d930fd31aa67327caed25cd51d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/263209
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Reed <reed@google.com>
This reverts commit dff4c1d5fd.
Reason for revert: Fix WIP
Original change's description:
> Revert "Avoid loop-dependent behavior in GrMemoryPoolBench"
>
> This reverts commit f811fc331a.
>
> Reason for revert: breaks on iOS
>
> Original change's description:
> > Avoid loop-dependent behavior in GrMemoryPoolBench
> >
> > This helps stability of benchmark across repeated runs, and across code
> > changes. Previously, a change to the tuned loop count could radically
> > change the allocation behavior within the loop's iteration and lead to
> > unfair comparisons.
> >
> > In addition, this separates the stack allocation pattern into N allocations
> > followed by N LIFO releases, and a push-pop alternating pattern of N
> > allocates and releases (so still LIFO, but reuses the memory at the start
> > of a block).
> >
> > In later CLs experimenting on the memory pool, I found that there were
> > surprising effects on performance linked to the specific interaction between
> > the allocation size, per-allocation metadata, and per-block metadata. To
> > help differentiate these coincidences, this adds two modes of allocation
> > where one should already be aligned.
> >
> > It also moves away from a global pool, so that it's possible to benchmark
> > on different block sizes and factor in the allocation/release cost of the
> > actual blocks (vs. the cursor management of a larger sized pool). As part
> > of this, the new/delete reference operator is added as an explicit benchmark.
> >
> > Change-Id: I12b8c11cb75db0df70460fe2e8cf6c029db7eb22
> > Reviewed-on: https://skia-review.googlesource.com/c/skia/+/262936
> > Commit-Queue: Michael Ludwig <michaelludwig@google.com>
> > Reviewed-by: Brian Salomon <bsalomon@google.com>
>
> TBR=bsalomon@google.com,michaelludwig@google.com
>
> Change-Id: I16f2810699a378eb5a516ab9fb1834c10b65f01b
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/263029
> Reviewed-by: Michael Ludwig <michaelludwig@google.com>
> Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Change-Id: I51a9db111c26b49572c3fa0928e26ef6e5a7a74c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/263196
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Auto-Submit: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
The reason for the assert was breaking an assert, that if the CTM was scale/translate, that after
a preTranslate, it should still be that.
This is true... unless the new translate values are non-finite. In that case, we might turn a zero
into a NaN, (0 * non_finite --> nan), so we either have to require finite args (which we don't
at the moment) or we can't make this assert. This re-land removes that assert.
This reverts commit 268ed57d71.
Change-Id: I3c48a0aa17649351a246c1fbab5449f2d59aaf84
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/263023
Commit-Queue: Mike Reed <reed@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: Mike Reed <reed@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit f811fc331a.
Reason for revert: breaks on iOS
Original change's description:
> Avoid loop-dependent behavior in GrMemoryPoolBench
>
> This helps stability of benchmark across repeated runs, and across code
> changes. Previously, a change to the tuned loop count could radically
> change the allocation behavior within the loop's iteration and lead to
> unfair comparisons.
>
> In addition, this separates the stack allocation pattern into N allocations
> followed by N LIFO releases, and a push-pop alternating pattern of N
> allocates and releases (so still LIFO, but reuses the memory at the start
> of a block).
>
> In later CLs experimenting on the memory pool, I found that there were
> surprising effects on performance linked to the specific interaction between
> the allocation size, per-allocation metadata, and per-block metadata. To
> help differentiate these coincidences, this adds two modes of allocation
> where one should already be aligned.
>
> It also moves away from a global pool, so that it's possible to benchmark
> on different block sizes and factor in the allocation/release cost of the
> actual blocks (vs. the cursor management of a larger sized pool). As part
> of this, the new/delete reference operator is added as an explicit benchmark.
>
> Change-Id: I12b8c11cb75db0df70460fe2e8cf6c029db7eb22
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/262936
> Commit-Queue: Michael Ludwig <michaelludwig@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
TBR=bsalomon@google.com,michaelludwig@google.com
Change-Id: I16f2810699a378eb5a516ab9fb1834c10b65f01b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/263029
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
This helps stability of benchmark across repeated runs, and across code
changes. Previously, a change to the tuned loop count could radically
change the allocation behavior within the loop's iteration and lead to
unfair comparisons.
In addition, this separates the stack allocation pattern into N allocations
followed by N LIFO releases, and a push-pop alternating pattern of N
allocates and releases (so still LIFO, but reuses the memory at the start
of a block).
In later CLs experimenting on the memory pool, I found that there were
surprising effects on performance linked to the specific interaction between
the allocation size, per-allocation metadata, and per-block metadata. To
help differentiate these coincidences, this adds two modes of allocation
where one should already be aligned.
It also moves away from a global pool, so that it's possible to benchmark
on different block sizes and factor in the allocation/release cost of the
actual blocks (vs. the cursor management of a larger sized pool). As part
of this, the new/delete reference operator is added as an explicit benchmark.
Change-Id: I12b8c11cb75db0df70460fe2e8cf6c029db7eb22
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/262936
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This reverts commit 98bfcc7ff3.
Reason for revert: Flutter hitting assert:
../../third_party/skia/src/core/SkCanvas.cpp:1432: fatal error: "assert(fIsScaleTranslate == fMCRec->fMatrix.isScaleTranslate())"
Original change's description:
> Extend SkCanvas matrix stack to be 4x4, but with (basically) the same public API.
>
> Devices receive the 4x4, but by default they simply downsample it to SkMatrix.
>
> New SkM44 matrix for the impl. It differs from SkMatrix44 in a few ways
> - no tracking of "type"
> - faster for concat, as it does not use doubles for intermediates
> - much simpler API
>
> There are some low-bit differences in some gms, so adding a flag for clients to
> stage this change. (due to faster but lower-precision in SkM44::concat)
>
> Performance: running canvas_matrix bench
>
> 3x3 version:
>
> 167.93 canvas_matrix_3x3 8888
> 209.97 canvas_matrix_2x3 8888
> 174.87 canvas_matrix_scale 8888
> 135.30 canvas_matrix_trans 8888
>
> 4x4 version:
>
> 116.59 canvas_matrix_3x3 8888
> 105.40 canvas_matrix_2x3 8888
> 159.83 ? canvas_matrix_scale 8888
> 113.47 canvas_matrix_trans 8888
>
> Why faster?
> - not tracking matrix_type helps a lot it seems
> - faster full concat (no doubles)
>
> Before adding the specialized preConcats...
>
> 318.11 ? canvas_matrix_3x3 8888
> 339.38 canvas_matrix_2x3 8888
> 383.28 canvas_matrix_scale 8888
> 251.67 canvas_matrix_trans 8888
>
> Change-Id: I68eac942919fa5418081e789f31710a1e2a752da
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/262056
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Reviewed-by: Florin Malita <fmalita@chromium.org>
> Commit-Queue: Mike Reed <reed@google.com>
TBR=mtklein@google.com,bsalomon@google.com,herb@google.com,fmalita@chromium.org,fmalita@google.com,reed@google.com,michaelludwig@google.com
Change-Id: I28c3d69c19ba44ab65ca7c059221b64c7dffef22
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/263021
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This reverts commit 1c16b43033.
Reason for revert: Red on tree
Original change's description:
> Move makeDeferredRenderTargetContext calls to factory on RTC.
>
> Change-Id: Iaa8f5829d9f8650ff27a60f75fb2216f016ab85e
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/262058
> Commit-Queue: Greg Daniel <egdaniel@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
TBR=egdaniel@google.com,bsalomon@google.com
Change-Id: I9e3c9d13c66b5437c87ad7136d283fa4ac81df1f
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/263019
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
Devices receive the 4x4, but by default they simply downsample it to SkMatrix.
New SkM44 matrix for the impl. It differs from SkMatrix44 in a few ways
- no tracking of "type"
- faster for concat, as it does not use doubles for intermediates
- much simpler API
There are some low-bit differences in some gms, so adding a flag for clients to
stage this change. (due to faster but lower-precision in SkM44::concat)
Performance: running canvas_matrix bench
3x3 version:
167.93 canvas_matrix_3x3 8888
209.97 canvas_matrix_2x3 8888
174.87 canvas_matrix_scale 8888
135.30 canvas_matrix_trans 8888
4x4 version:
116.59 canvas_matrix_3x3 8888
105.40 canvas_matrix_2x3 8888
159.83 ? canvas_matrix_scale 8888
113.47 canvas_matrix_trans 8888
Why faster?
- not tracking matrix_type helps a lot it seems
- faster full concat (no doubles)
Before adding the specialized preConcats...
318.11 ? canvas_matrix_3x3 8888
339.38 canvas_matrix_2x3 8888
383.28 canvas_matrix_scale 8888
251.67 canvas_matrix_trans 8888
Change-Id: I68eac942919fa5418081e789f31710a1e2a752da
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/262056
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Reed <reed@google.com>
Change-Id: I0b11d4210c6e663cfb4854fc33e1396fd79fe9a4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/261780
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
The interpreter doesn't (yet) support all GPU intrinsics, so adjust the
code to avoid 'max' and 'saturate'.
The interpreter is more picky about uniform blocks being the correct
size, so don't pass the color matrix data to the "None" effect, which
has no uniforms.
Change-Id: I4609f913eaa762ca171b2875eb25d23141a0646c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/261931
Auto-Submit: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Updated internal usage of SkRuntimeColorFilterFactory to use this
instead. Once this lands, we can update SkiaRenderer in Chrome to
use it, and remove SkRuntimeColorFilterFactory.
This doesn't support a CPU callback function in the runtime color
filter - I don't think we're going to support that in the long term.
Change-Id: I714413bd590cf5cf4416ef62809a6e1d92211688
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/261681
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
- have to commit to a perspective stage (used to only use affine)
- coordinate between updater and shader/stages
Benches:
verts 608 --> 270
atlas 17.9 --> 5.6
Change-Id: I1d1ce9eb27b49d2ddcc610f6ea5c4720901b9e1b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/261136
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Reed <reed@google.com>
Most of these did nothing, and I'd like to remove the feature. Also, the
benchmark was using 'in uniform' which isn't supported (it asserts in
debug build).
Change-Id: I671ca69fdd50811a2090c7a03f5f23e6b38e6f96
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/260043
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Change-Id: I7c672ff6b8eb95ec8c1123a5bfdb202e1644f494
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/259281
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
Saves one heap allocation per DDL recorded.
Change-Id: I9393aedc3b48031cd2ea5f0160b107915077099a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/259419
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
This is a first step towards pulling the creation of the GrProgramInfos explicitly into the Ops' onExecute methods. We need this behavior so programInfo creation can be moved forward to onPrePrepare.
For now, pipeline creation is a static member on GrSimpleMeshDrawOpHelper so GrPipeline creation can be bottle-neckedOps for Ops that don't use the helper. In some future world CreatePipeline could become non-static to reduce some of the duplicate code.
Bug: skia:9455
Change-Id: I2d35dd223db824e84616f5df0f1dca34c1b6e412
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/258003
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Drop unnecessary () in `const char* (*fText)`. Sure, whatever.
Then a couple tweaks where we use strncpy():
In one case in SkICC.cpp, GCC warns us we're passing sizeof(dst) as
the copy count, which can be error prone because you may miss a
terminating nul. I believe the existing strncpy() behavior is what
we want here, so I've stifled this and left a note. We don't really
want a terminating nul... we want a zero fill of the remaining dst.
On the other side of that if-statment in that same function, GCC
warns that we've passed exactly sizeof(src) as the copy count when
writing the calculated description prefix, which again might be
buggy due to a missed terminating nul. Seemed like memcpy() was the
way to go here for that prefix. Ultimately we check to make sure we
wrote every single destination byte.
Over in SkPDFMetadata.cpp, strncpy() is concerned once again that
we're copying strlen(src) and thus omitting the terminating nul. As
usual we don't want that terminating nul, so switch to memcpy() to
make that explicit.
I'm starting to think strncpy() is too helpful to reason about sanely.
Change-Id: I1bcd5f4e406ca388c1c236c5f3ee531bf1737705
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/257408
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
When iterating over the coord transforms or texture samplers of a
FP also have access to the owning FP.
Pass a coord transform range to GPs rather than a pointer to an
iterator.
Change-Id: If7c829a67dce6600d7f49e12d6f49f685dcace3a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/256216
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
This reverts commit 3e7af41224.
Change-Id: Id4f66b3956f4bdbe690db20fc478b7365ee89717
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/256676
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
Auto-Submit: Mike Reed <reed@google.com>
This reverts commit 1792b19485.
Reason for revert: need to update legacy_convexity, still used by google3
Original change's description:
> Revert "Revert "Use flat version of path-direction enum""
>
> This reverts commit 0dacc6b7d3.
>
> Change-Id: Ie103e9f36b07e4ee256a3688a4decf3a6dd74314
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/255832
> Auto-Submit: Mike Reed <reed@google.com>
> Reviewed-by: Mike Reed <reed@google.com>
> Commit-Queue: Mike Reed <reed@google.com>
TBR=reed@google.com
Change-Id: I0ecea0eb8a237298c6b908cc4bfd1cacdfc5b900
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/255976
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
This reverts commit 0dacc6b7d3.
Change-Id: Ie103e9f36b07e4ee256a3688a4decf3a6dd74314
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/255832
Auto-Submit: Mike Reed <reed@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
This reverts commit e0fbe94351.
Reason for revert: need to add guard flag to flutter
Original change's description:
> Use flat version of path-direction enum
>
> Bug: skia:9663
> Change-Id: I00077d9f2b14b3e983e6a46ef6f560cabdb1678d
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/242557
> Commit-Queue: Mike Reed <reed@google.com>
> Reviewed-by: Florin Malita <fmalita@chromium.org>
TBR=fmalita@chromium.org,reed@google.com
Change-Id: If47173d9b203b2d3a175af290a15d986accb4703
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:9663
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/255831
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
This CL is 100% plumbing. We ultimately want all GrPrimitiveProcessor-derived objects to not be refCounted. This will make several helper objects POD and, by putting them into an arena, will make managing their lifetime easier (e.g., for DDL prePreparing).
Note: the CCPR GrGeometryProcessor-derived classes only ever appear on the stack so aren't forced into arenas.
Bug: skia:9455
Change-Id: Ib9be503d2fbf8c2578642df93fc301156629829d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/255304
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Handy if you want to flush, or otherwise understand the underlying buffer(s).
Change-Id: I3d6610695c8603232192b26c687c6c74512165dd
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/254803
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
More cleaning to do if we like this idea...
Change-Id: I608143db085911565dd5f5426f7ee6436ec58cdf
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/254680
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
This further consolidates the information required to compute the program key (esp. for Vulkan). This CL mainly comprises the plumbing portion - a follow up CL will actually use it.
Bug: skia:9455
Change-Id: Iaac716c289916981a1757a333bfa57b3051fd35b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/252161
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Propagating usage of these helpers will, hopefully, improve code reuse and centralize future changes.
This is pulled out of https://skia-review.googlesource.com/c/skia/+/252161/ (Add GrPrimitiveType to GrProgramInfo)
Bug: skia:9455
Change-Id: I9b46e42db8561d98fb724a41a1f7a3bec4bf97b0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/252484
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This reverts commit e63595dd85.
Reason for revert: Breaking the Perf tests:
ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0 [vsyscall]
Command exited with code 139
#######################################
symbolized stacktrace follows
#######################################
build/nanobench BulkRectBench<1000, (RectangleLayout)1, (ImageMode)2, (DrawMode)2>::onPerCanvasPreDraw(SkCanvas*) at skia/bench/BulkRectBench.cpp:218
build/nanobench main at skia/bench/nanobench.cpp:1277
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf1)[0x7f1213dad2e1]
build/nanobench _start at ??:?
Original change's description:
> Add variety of bulk API benchmarks
>
> Covers drawing 1000 anti-aliased rectangles, either sharing 1 image, using
> a unique image, or as a solid color. Tests using the bulk APIs and regular
> one-at-a-time drawing.
>
> Change-Id: Icdf5a1f6df229aed785212d0941b89c11aa785c0
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/251003
> Commit-Queue: Michael Ludwig <michaelludwig@google.com>
> Reviewed-by: Robert Phillips <robertphillips@google.com>
TBR=robertphillips@google.com,egdaniel@chromium.org,michaelludwig@google.com
Change-Id: I3b26d5a04c472993585242a108c2764ea2b96940
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/251213
Reviewed-by: Joe Gregorio <jcgregorio@google.com>
Commit-Queue: Joe Gregorio <jcgregorio@google.com>
Covers drawing 1000 anti-aliased rectangles, either sharing 1 image, using
a unique image, or as a solid color. Tests using the bulk APIs and regular
one-at-a-time drawing.
Change-Id: Icdf5a1f6df229aed785212d0941b89c11aa785c0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/251003
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Use `extra_cflags=["-DSK_CAPTURE_DRAW_TEXT_BLOB"]` to enable.
Change-Id: I1d6db478ee91696cdce090647b889c17a83a2718
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/250259
Commit-Queue: Hal Canary <halcanary@google.com>
Reviewed-by: Herb Derby <herb@google.com>
There is a bug on Pixel and Pixel2 devices where the program
eventually terminates with a non-zero exit code. Closing the
outResultsFile between JSON flushes seems to fix it (for whatever
reason).
Bug: b/143074513
Change-Id: I935e982e88758fda19292129c8031f8501cca615
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/249821
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
This is a reland of 6fc04f88a8
Original change's description:
> Reland "SkSurface asynchronous read APIs allow client to extend pixel lifetime"
>
> This is a reland of ce240cc6fd
>
> Original change's description:
> > SkSurface asynchronous read APIs allow client to extend pixel lifetime
> >
> > Previously the pixel data passed to the client was only valid during
> > the client's callback. This meant if the client wanted to defer
> > processing of the data a copy was necessary.
> >
> > Now we pass an object to the callback and the pixel lifetime is tied
> > to the lifetime of that object.
> >
> > The object may be holding a GPU transfer buffer mapped. We don't assume
> > that the object will be released on the direct GrContext thread. So
> > when the object is destroyed it posts a message to a new type,
> > GrClientMappedBufferManager, hanging off the direct context. The direct
> > context will periodically check for messages and unmap and then unref
> > buffers so that they can be reused. Currently this is done in
> > GrContext::performDeferredCleanup() and GrDrawingManager::flush().
> >
> > The old API is kept around for backwards compatibility but it is
> > reimplemented as a bridge on top of the new mechanism.
> >
> > Also a utility function to SkImageInfo is added to directly make a new
> > info with a specified dimensions rather than passing the width and
> > height separately to makeWH().
> >
> > Bug: chromium:973403
> > Bug: skia:8962
> >
> > Change-Id: Id5cf04235376170142a48e90d3ecd13fd021a2a6
> > Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245457
> > Reviewed-by: Brian Osman <brianosman@google.com>
> > Commit-Queue: Brian Salomon <bsalomon@google.com>
>
> Bug: chromium:973403, skia:8962
> Change-Id: I5cecd36276c8b6dc942cf549c7095db2df88530c
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245678
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Brian Salomon <bsalomon@google.com>
Bug: chromium:973403, skia:8962
Change-Id: Ie584c1c3ef8021c976f71b708e53871c693cc450
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/246057
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This reverts commit 6fc04f88a8.
Reason for revert: Chrome roll failure suspect because of:
* https://chromium-review.googlesource.com/c/chromium/src/+/1837131 (22 commits)
* https://chromium-review.googlesource.com/c/chromium/src/+/1837214 (24 commits)
Original change's description:
> Reland "SkSurface asynchronous read APIs allow client to extend pixel lifetime"
>
> This is a reland of ce240cc6fd
>
> Original change's description:
> > SkSurface asynchronous read APIs allow client to extend pixel lifetime
> >
> > Previously the pixel data passed to the client was only valid during
> > the client's callback. This meant if the client wanted to defer
> > processing of the data a copy was necessary.
> >
> > Now we pass an object to the callback and the pixel lifetime is tied
> > to the lifetime of that object.
> >
> > The object may be holding a GPU transfer buffer mapped. We don't assume
> > that the object will be released on the direct GrContext thread. So
> > when the object is destroyed it posts a message to a new type,
> > GrClientMappedBufferManager, hanging off the direct context. The direct
> > context will periodically check for messages and unmap and then unref
> > buffers so that they can be reused. Currently this is done in
> > GrContext::performDeferredCleanup() and GrDrawingManager::flush().
> >
> > The old API is kept around for backwards compatibility but it is
> > reimplemented as a bridge on top of the new mechanism.
> >
> > Also a utility function to SkImageInfo is added to directly make a new
> > info with a specified dimensions rather than passing the width and
> > height separately to makeWH().
> >
> > Bug: chromium:973403
> > Bug: skia:8962
> >
> > Change-Id: Id5cf04235376170142a48e90d3ecd13fd021a2a6
> > Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245457
> > Reviewed-by: Brian Osman <brianosman@google.com>
> > Commit-Queue: Brian Salomon <bsalomon@google.com>
>
> Bug: chromium:973403, skia:8962
> Change-Id: I5cecd36276c8b6dc942cf549c7095db2df88530c
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245678
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Brian Salomon <bsalomon@google.com>
TBR=bsalomon@google.com,brianosman@google.com
Change-Id: I9e01d1b82fb399b94292441d91da51176bb161d9
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:973403, skia:8962
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245956
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Ravi Mistry <rmistry@google.com>
This is a reland of ce240cc6fd
Original change's description:
> SkSurface asynchronous read APIs allow client to extend pixel lifetime
>
> Previously the pixel data passed to the client was only valid during
> the client's callback. This meant if the client wanted to defer
> processing of the data a copy was necessary.
>
> Now we pass an object to the callback and the pixel lifetime is tied
> to the lifetime of that object.
>
> The object may be holding a GPU transfer buffer mapped. We don't assume
> that the object will be released on the direct GrContext thread. So
> when the object is destroyed it posts a message to a new type,
> GrClientMappedBufferManager, hanging off the direct context. The direct
> context will periodically check for messages and unmap and then unref
> buffers so that they can be reused. Currently this is done in
> GrContext::performDeferredCleanup() and GrDrawingManager::flush().
>
> The old API is kept around for backwards compatibility but it is
> reimplemented as a bridge on top of the new mechanism.
>
> Also a utility function to SkImageInfo is added to directly make a new
> info with a specified dimensions rather than passing the width and
> height separately to makeWH().
>
> Bug: chromium:973403
> Bug: skia:8962
>
> Change-Id: Id5cf04235376170142a48e90d3ecd13fd021a2a6
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245457
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Brian Salomon <bsalomon@google.com>
Bug: chromium:973403, skia:8962
Change-Id: I5cecd36276c8b6dc942cf549c7095db2df88530c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245678
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This reverts commit ce240cc6fd.
Reason for revert: crashing in chrome unit test, abandoned context related?
Original change's description:
> SkSurface asynchronous read APIs allow client to extend pixel lifetime
>
> Previously the pixel data passed to the client was only valid during
> the client's callback. This meant if the client wanted to defer
> processing of the data a copy was necessary.
>
> Now we pass an object to the callback and the pixel lifetime is tied
> to the lifetime of that object.
>
> The object may be holding a GPU transfer buffer mapped. We don't assume
> that the object will be released on the direct GrContext thread. So
> when the object is destroyed it posts a message to a new type,
> GrClientMappedBufferManager, hanging off the direct context. The direct
> context will periodically check for messages and unmap and then unref
> buffers so that they can be reused. Currently this is done in
> GrContext::performDeferredCleanup() and GrDrawingManager::flush().
>
> The old API is kept around for backwards compatibility but it is
> reimplemented as a bridge on top of the new mechanism.
>
> Also a utility function to SkImageInfo is added to directly make a new
> info with a specified dimensions rather than passing the width and
> height separately to makeWH().
>
> Bug: chromium:973403
> Bug: skia:8962
>
> Change-Id: Id5cf04235376170142a48e90d3ecd13fd021a2a6
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245457
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Brian Salomon <bsalomon@google.com>
TBR=bsalomon@google.com,brianosman@google.com
Change-Id: Ic14cf07a7629b167c9f34a651aa87a0326e74207
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:973403, skia:8962
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245721
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Previously the pixel data passed to the client was only valid during
the client's callback. This meant if the client wanted to defer
processing of the data a copy was necessary.
Now we pass an object to the callback and the pixel lifetime is tied
to the lifetime of that object.
The object may be holding a GPU transfer buffer mapped. We don't assume
that the object will be released on the direct GrContext thread. So
when the object is destroyed it posts a message to a new type,
GrClientMappedBufferManager, hanging off the direct context. The direct
context will periodically check for messages and unmap and then unref
buffers so that they can be reused. Currently this is done in
GrContext::performDeferredCleanup() and GrDrawingManager::flush().
The old API is kept around for backwards compatibility but it is
reimplemented as a bridge on top of the new mechanism.
Also a utility function to SkImageInfo is added to directly make a new
info with a specified dimensions rather than passing the width and
height separately to makeWH().
Bug: chromium:973403
Bug: skia:8962
Change-Id: Id5cf04235376170142a48e90d3ecd13fd021a2a6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245457
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Add version SkImageInfo::Make() that takes SkISize instead of separate
width and height.
Change-Id: I42aa79d23b19e22f5405631728c245b04bce0559
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245172
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
- Update the parameter lists to both run and runStriped so
that they're in the same (sane) order, named consistently,
and always take counts with pointer arguments.
- Add the same count-based safety checks to run that were
already in runStriped.
- Remove the N parameter to run, it was only used to run
things one-at-a-time (other than one spot in unit tests),
and it simplifies the code quite a bit. If you want to run
multiple times, use the striped version. I also moved that
functions 'N' earlier in the parameter list, to make the
pattern of the remaining parameters clearer.
- Remove an interpreter benchmark class that was never used.
Change-Id: Ibff0a47bdb2d29d095a0addd27e65ab13cb80fce
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/244716
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
On Windows I'm getting:
curr/maxrss loops min median mean max stddev samples config bench
33/38 MB 1495 2.62us 2.65us 2.66us 2.72us 1% .o.o.oO.OO gl DDLRecorder
Change-Id: I529e6ac612c455915c166472124b08647eb7bffd
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/241039
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Also don't specify redundant width/height to SkGpuDevice.
Change-Id: I389df5c4b073c2c05632ba6b7c95b02a22dfaf98
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/235824
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
- exercise different options
- reduce triangle sizes to emphasize overhead (small triangles is also common)
Change-Id: I24ff889cdeeba136349b3a855c6b5881109e2a88
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/235461
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
Auto-Submit: Mike Reed <reed@google.com>
Touch the memory that the iterator returns, so we're (slightly) more
sure the compiler isn't eliding code we want to be timing.
Change-Id: I62eb36759413d83fb18f4a6e0bf23d4d98f11999
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/235860
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Reed <reed@google.com>
Old version still exists until client code is migrated.
Change-Id: I087b6b977f586d334f9a20954e7ed7b5e4b7ea5e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/234579
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Mostly use unique_ptr more consistently.
Change-Id: I6e11b272a7904eb662dea59b03fbc309a4cfc25d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/233984
Commit-Queue: Ben Wagner <bungeman@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Change-Id: If8a4d3cfdeaef58dd9924af430cfed834519b096
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/232496
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Hal Canary <halcanary@google.com>
We special-case unrotated shapes today (drawing rects instead of quads/paths).
This new bench shows that the rotated case is over 4x slow. Probably should
do something about that.
Change-Id: Ie80a42b981a1841f980c70a5f8f8a32357d764b7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/233976
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
Auto-Submit: Mike Reed <reed@google.com>
Remove return statements immediately following SK_ABORTs.
Change-Id: I6446b6ba3a6fb272d84ea27f38cf366095cf801d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/233565
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Additionally this changes removes the version of the call that that takes
a GrPixelConfig.
As part of this change many call sites that were calling getRTSampleCount
to just check renderability have been changed to call
isFormat[AsColorType]Renderable instead. This change has also started to
move us to just checking format renderability (without GrCT) in classes
that are specifically dealing with GrSurface (GrResourceProvider, GrGpu, etc.).
Bug: skia:6718
Change-Id: Icf4a1a21a81a44e6863a7cd2059d21b9833d581f
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/233039
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
2x faster than calling drawVertices as the impl.
Lots more to do in future CLs
- much of the time is spent in malloc, as we cons-up private shaders.
we *could* create a private shader and wack its data for each draw
(breaking the immutable contract, but that may be ok for a raster-only
internal-use shader...)
- also spend time building the pipeline for each draw, even though
all that has changed is a color payload (and the ctm). A custome
stage(s) that exposed its private data could be reused with new
data...
Bug: skia:
Change-Id: I0ff15155e37c0af7931abd34c0883701a47a048a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/203168
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Reed <reed@google.com>
Also refactors 3 duplicate functions that converted an SkTileMode to
a char*. This will rename the matrix convolution benchmarks that used
"clampToBlack" to the more modern "decal" terminology.
Bug: skia:9280
Change-Id: I43a4f9edeba36eb5af1243f7e4ab7647253b91f2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/230883
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
We can't remove the loopers themselves, as they are still used
by android and chrome (they just don't ever pass them to skia).
Eventually each of those clients will resolve this, but for now
we just keep the classes (and tests) in skia.
Bug: skia:4783
Change-Id: I5f507e6bb82280f2bc7c0b21eebe59c287aa9265
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/230579
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Reed <reed@google.com>
Bug: skia:
Change-Id: I2ae0caf08f8434302cae8151ae1ea0fda8d56928
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/230397
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
PS2 adds a rewrite for Skia #include <...> to #include "...", letting
them be otherwise rewritten and sorted too. (We do need one exception
for the Vulkan headers, which will otherwise be rewritten to always
point to our own.) I don't think it's particularly important to
favor "" or <>, but picking one keeps things consistent.
PS3 adds a missing SkMutex.h include.
PS4 fixes a terrible readability problem.
Change-Id: Id9fe752727ef30e802b1daf755ee2ed15e267577
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/229742
Commit-Queue: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: Mike Klein <mtklein@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Adding cache
Caching shaped results
Base+Index for referencing arrays
The very first and naive version of cache
Cache measurement, lines and picture
Added text blob cache for lines
Removed Run* from Cluster
Removed const char* from Cluster and Run
Few minor changes
Change-Id: I444a1defa950aed5999cfa1c3545fd83ccb54ce9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227840
Commit-Queue: Julia Lavrova <jlavrova@google.com>
Reviewed-by: Julia Lavrova <jlavrova@google.com>
Reviewed-by: Mike Reed <reed@google.com>
This is a reland of f42de9e1e5
Original change's description:
> Interpreter: Bounds check array access, add bool return from run
>
> Out of bounds access with constant indices is a compile error.
> At runtime, causes the interpreter to fail. Made several other
> conditions trigger the same failure logic, and updated all
> uses of the interpreter to validate success.
>
> Change-Id: I3720b3c83903220b010ec574121fc64dbe102378
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228256
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Mike Reed <reed@google.com>
Change-Id: I8849de815f7efb730ac9c55b6edd296cb9ca7599
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228353
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit f42de9e1e5.
Reason for revert: All the SANs
Original change's description:
> Interpreter: Bounds check array access, add bool return from run
>
> Out of bounds access with constant indices is a compile error.
> At runtime, causes the interpreter to fail. Made several other
> conditions trigger the same failure logic, and updated all
> uses of the interpreter to validate success.
>
> Change-Id: I3720b3c83903220b010ec574121fc64dbe102378
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228256
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Mike Reed <reed@google.com>
TBR=mtklein@google.com,brianosman@google.com,ethannicholas@google.com,reed@google.com
Change-Id: I434601960d54fbd7d00e2af2dc6269a83a768c5b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228352
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Out of bounds access with constant indices is a compile error.
At runtime, causes the interpreter to fail. Made several other
conditions trigger the same failure logic, and updated all
uses of the interpreter to validate success.
Change-Id: I3720b3c83903220b010ec574121fc64dbe102378
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/228256
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Only used in three somewhat dubious places.
Change-Id: I7ccd1aef41f826d0eb62606751f4d3f0ceda267d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227065
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
N=15 and N=63 make for nice even looking profiles
on ARM and x86 respectively, with N=15 running 3
body loops and 3 tail loops on ARM, N=63 running
7 body loops and 7 tail loops on x86.
Change-Id: Ie7616bd99c949328bbb7d7048fc6f468ff1e3ad2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/227220
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
These didn't work correctly, and they're extremely tricky to get right
in the vectorized execution model (vs. structured control flow). As a
side effect, determine the maximum stack depth used for the execution
masking - the same idea will be used for the primary stack in a later
CL. Add a unit test to verify the new restriction, and fix two places
that were relying on this feature before.
In addition, boolean external values need to be masks. I may implement
this in the code-gen at some point, but this is already a fringe
feature, so just fix the one unit test for now.
Change-Id: I9607ffaf67c7795dbf42e4009180aea8f3e65c44
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/226849
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
+ fix bug in SkBulkGlyphMetricsAndImages where the underlying SkAutoSTArray
was not reset.
Change-Id: I94882da30e858d012aa02f79c5fdfe4bc7d2e8aa
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/225541
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Herb Derby <herb@google.com>
config is completely redundant. No caller really cares what the backend
format is.
Change-Id: I93f1feb3ee61db6c21b7915bab3ee3fba5656f92
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/225194
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
I'm kind of getting bored of having to pass --loops 0
in optimized+asserts builds.
We'd been defaulting to 1 loop in Debug and ASAN builds,
and manually setting Valgrind builds to 1 loop with 1
sample. Remove the default so all builds auto-tune,
but set Debug and ASAN bots like Valgrind bots.
Change-Id: Ifedd98a11ea74a2c222fe59bd50b96683d12b98c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/224816
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
There are a few places that have been checking whether the *proxy*
needs stencil, in order to determine if the current render target
context needs stencil. This is problematic since a render target
context can not require stencil itself, but wrap an existing proxy
that already has stencil.
Bug: skia:
Change-Id: I2719dd3a9df15fef3d64f991cda4fadea23266bb
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/223970
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Because it mutates the glyph.
Change-Id: Ic7ce320350764454d7a76335828d398f19b149d6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/223797
Commit-Queue: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
This is largely redundant with GrPixelConfig. However, we intend to
remove GrPixelConfig.
Bug: skia:7580
Change-Id: I03d92303be832711f7821f8a97d36387c9b04a9f
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/222883
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Looks like ~50ns overhead for RP vs ~14,000ns for SkVM.
Change-Id: I85ef73d3387657b14615fcfa5cfd9df5c2325343
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/223302
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Converting to glyph() style calls that return SkGlyph*. This is mainly preparation
for removing converting findImage(const SkGlyph&) to prepareImage(SkGlyph*).
+ Misc cleanups mainly fWidth -> width() type things.
Change-Id: Id5c9b0ba5856b3ea54353ece4d05fa495cc5a640
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/223187
Commit-Queue: Herb Derby <herb@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
From now on, sample counts always refer to the number of actual color
samples, and render targets don't have separate color and stencil
sample counts.
If mixed samples support is available when making a
"GrAAType::kCoverage" draw, then an op may attach and use a mixed
sampled stencil buffer internally. But this will all be invisible to
the client.
After this CL, we temporarily won't have a mode to use nvpr with mixed
samples. That will soon be fixed by a follow-on CL that enables nvpr
with mixed samples in the normal "gl" and "gles" configs.
Bug: skia:
Change-Id: I1cb8277f0d2d0d371f24bb9f39cd473ed5c5c83b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/221878
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This new bench lets us measure the overhead of program building,
optimization, and JITting. Surprisingly, at head the optimization in
Builder::done() takes longer than the JIT.
The new bench clocks in around 40µs on my laptop at head,
then 32µs after switching val_to_reg to be an std::vector,
then 27µs after switching deaths to be an std::vector too,
then 22µs after switching fIndex to be an SkTHashMap,
then 20µs after calling program.reserve(fProgram.size()),
then 19µs after switching JIT data maps to SkTHashMap too.
I tried swapping some std::vector for SkTDArray to no benefit, actually
a little detriment. So I think this is roughly all the low-hanging
fruit, with time split now roughly equally between Builder::Done(),
JITting in Program::eval(), and the original calls to Builder
themselves.
Also disable perf dumps on Mac. No real value there until I can dump a
dylib, and it's just one more thing I have to remember to disable before
running this sort of benchmark.
Change-Id: I1c6e58ed00ac94ad622c7d740712634f60787102
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/222984
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This moves the responsibility for allocating executable code out of
Assembler. The pages Xbyak uses are obviously executable, so this is
redundant right now, but it'll let us switch to something simple like
std::vector<uint8_t> as we continue to cut out Xbyak.
Make how Program holds its cached JIT program slightly less of a mess.
Change-Id: I38d6f01006da1da60f4aed675e9ddf97de9aec52
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/222575
Auto-Submit: Mike Klein <mtklein@google.com>
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
- 32x8 i32 add,sub,mul
- add I32_Naive bench/test builder to get better i32 mul coverage
- minor refactoring all over
Change-Id: I13cc19ff37a2da0bcff289ba51baac08f456d6c5
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/222485
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This extends the runStriped to all for return data.
GPU impl not done yet, will be done in a follow-on CL.
Change-Id: Ib107d2945f6fdb34ce1b5405a6c88a5ae7e9f7ac
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/221539
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Reed <reed@google.com>
This reverts commit 59b22f9297.
Reason for revert: ios
Original change's description:
> Reland "Reland "Interpreter: Support striped inputs for less overhead""
>
> This reverts commit 2c59b4e9ea.
>
> Change-Id: I2b06936994430722b8fc3890ff9b4a6f4710db04
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/221998
> Reviewed-by: Brian Osman <brianosman@google.com>
> Reviewed-by: Mike Klein <mtklein@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
TBR=mtklein@google.com,brianosman@google.com,reed@google.com
Change-Id: I321d257e400f5df3c6115dfe9aa18a2794a95910
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/222099
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
