This is distilled from 82161 so we can first change the GMs of our
previous bots and then compare the same GM between the old algorithm
and the new one.
Bug: skia:
Change-Id: I4b13bb1a06d109b60c8e29f5f87adad47d0a811b
Reviewed-on: https://skia-review.googlesource.com/84523
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
4.24% faster in gradient_conical_clamp_shallow_dither
6.43% faster in gradient_conicalOutZero_clamp_hicolor
10.41% faster in gradient_conicalOut_clamp_hicolor
14.85% faster in gradient_conicalOutZero_clamp_3color
16.72% faster in gradient_conicalOut_clamp
24.30% faster in gradient_conicalOut_clamp_3color
CQ_INCLUDE_TRYBOTS=skia.primary:Test-Android-Clang-Nexus7-GPU-Tegra3-arm-Release-All-Android,Test-Android-Clang-Nexus7-GPU-Tegra3-arm-Debug-All-Android
Bug: skia:
Change-Id: I6bc1f24c9463fc9c2acbcba7bd4d55b37ade1613
Reviewed-on: https://skia-review.googlesource.com/82161
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
gm.h includes sk_tool_utils.h but does not use it.
The bulk of this CL makes each gm that uses sk_tool_utils include it.
sk_tool_utils.h also provided SkRandom and SkTDArray,
so a couple GMs add those headers too.
Change-Id: Ieb2a7c542f0ca89c3223f744fc11b0ff37af36c1
Reviewed-on: https://skia-review.googlesource.com/10014
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>
'static const' means, there must be at most one of these, and initialize it at
compile time if possible or runtime if necessary. This leads to unexpected
code execution, and TSAN* will complain about races on the guard variables.
Generally 'constexpr' or 'const' are better choices. Neither can cause races:
they're either intialized at compile time (constexpr) or intialized each time
independently (const).
This CL prefers constexpr where possible, and uses const where not. It even
prefers constexpr over const where they don't make a difference... I want to have
lots of examples of constexpr for people to see and mimic.
The scoped-to-class static has nothing to do with any of this, and is not changed.
* Not yet on the bots, which use an older TSAN.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2300623005
Review-Url: https://codereview.chromium.org/2300623005
Migrating the flag to embedder defines (Chromium already guarded).
Also augment gradient-focused GMs to generate both dithered/undithered
results.
BUG=skia:4436
R=reed@google.com,robertphillips@google.com
Review URL: https://codereview.chromium.org/1400813006
Reason for revert:
breaks many bots with refcnt error
Original issue's description:
> make gm background colors 565 compatible
>
> Change a batch of GM tests to convert their background color
> so that it is representable in 8888 and 565.
>
> Enable portable text in those same tests to minimize platform
> differences.
>
> In a couple of bitmap tests, use portable typefaces instead of
> choosing 'Times' which may or may not be available on the platform.
>
> R=borenet@google.com
>
> Committed: https://skia.googlesource.com/skia/+/be7f768a357aefb39c42d24b81b24d647bb6ab70TBR=borenet@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/1184123002
Change a batch of GM tests to convert their background color
so that it is representable in 8888 and 565.
Enable portable text in those same tests to minimize platform
differences.
In a couple of bitmap tests, use portable typefaces instead of
choosing 'Times' which may or may not be available on the platform.
R=borenet@google.com
Review URL: https://codereview.chromium.org/1176243006
Now that we're drawing tiles threaded like implside painting, remove the checks
that those lock counts are balanced. They're just not right for anyone anymore.
SkBitmaps themselves are not threadsafe (even const ones), so shallow copy them
on playback of an SkRecord. (The underlying SkPixelRefs are threadsafe.)
Simplify quilt drawing by using SkBitmap::extractSubset. No need for locking.
Bump up to 256x256 tiles. 16x16 tiles just murders performance (way too much
contention). This has the nice side effect of letting us enable a bunch more
GMs for quilt mode; they drew wrong with small tiles but exactly right with large.
BUG=171776
R=reed@google.com, mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/371023005
- Rename TileGrid -> Quilt to avoid the name overload.
- Tag all failing GMs with kSkipTiled_Flag.
You may be wondering, do any GMs pass? Yes, some do! And that trends towards all of them as we increase --quiltTile.
Two GMs only fail in --quilt mode in 565. Otherwise all GMs which fail are skipped, and those which don't fail aren't. (The 8888 variants of those two GMs are skipped even though they pass.)
BUG=skia:2477
R=reed@google.com, mtklein@google.com
Author: mtklein@chromium.org
Review URL: https://codereview.chromium.org/256373002
git-svn-id: http://skia.googlecode.com/svn/trunk@14457 2bbb7eff-a529-9590-31e7-b0007b416f81
