Commit Graph

17051 Commits

Author SHA1 Message Date
mtklein
d8afd50b5a Android's Clang toolchain can't assemble the JPEG's .S files.
Only affecting Build-Mac10.8-Clang-Arm7-Debug-Android.

BUG=skia:

Review URL: https://codereview.chromium.org/1093423003
2015-04-21 12:07:38 -07:00
cdalton
fd4167ddf1 Import glTextureBarrier
BUG=skia:

Review URL: https://codereview.chromium.org/1090163002
2015-04-21 11:45:56 -07:00
mtklein
d557462f1f Skip msaa4 on Tegra 2 and 3.
This will fix the N7 and Xoom bots.

TBR=borenet@google.com

BUG=skia:

Review URL: https://codereview.chromium.org/1099943002
2015-04-21 11:25:47 -07:00
senorblanco
42a2c1d231 Revert "One more ASAN fix."
This reverts commit 409fd66a5a.

We don't expect to pass this test until the ASAN bots are
correctly configured, and can run GPU tests.

BUG=skia:
TBR=robertphillips@google.com

Review URL: https://codereview.chromium.org/1080393003
2015-04-21 11:13:17 -07:00
mtklein
05641a5582 DM: failure to canary a sink is now fatal.
This requires we remove NVPR from the default set of configs, as we only find
out at runtime that it's not available.  All the other defaults will either be
compiled in and supported, or not compiled in and non-fatally skipped as
unknown configs.

BUG=skia:

Review URL: https://codereview.chromium.org/1100773003
2015-04-21 10:49:14 -07:00
mtklein
fca5c88873 Ask for NVPR only when we expect it to be available.
BUG=skia:

Review URL: https://codereview.chromium.org/1099023003
2015-04-21 10:34:47 -07:00
Greg Humphreys
1bdc9351ec typo in sherrifing docs
BUG=skia:
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=1097353002

Review URL: https://codereview.chromium.org/1097353002
2015-04-21 10:28:44 -07:00
tomhudson
d7c677104a Support HWUI rename of DisplayListRenderer
R=djsollen@google.com,ccraik@google.com

Review URL: https://codereview.chromium.org/1086083004
2015-04-21 10:09:07 -07:00
joshualitt
010db5346e Some simple optimizations for improving GrAtlasTextContext perf
BUG=skia:

Review URL: https://codereview.chromium.org/1093083003
2015-04-21 10:07:26 -07:00
senorblanco
409fd66a5a One more ASAN fix.
Don't fail the GL validation check if the GL context fails to create.

TBR=robertphillips@google.com
BUG=skia:

Review URL: https://codereview.chromium.org/1096113003
2015-04-21 09:43:08 -07:00
joshualitt
64c99cc90e Add support for rotating / scaling / translating gpu cached distance field textblobs
BUG=skia:

Review URL: https://codereview.chromium.org/1092283002
2015-04-21 09:43:03 -07:00
mtklein
59bcfaed66 Split default and user-supplied event tracers.
Chrome's got a test scenario where setting the user-supplied tracer is racing with a concurrent use of the default tracer.

BUG=chromium:437044

No public API changes.
TBR=reed@google.com

Review URL: https://codereview.chromium.org/1099123002
2015-04-21 09:38:03 -07:00
joshualitt
ae32c102e7 Store pointers to GrGlyph directly in BitmapTextBlob. This patch improves performance by avoiding hashmap lookups under normal use
BUG=skia:

Review URL: https://codereview.chromium.org/1087203004
2015-04-21 09:37:57 -07:00
stephana
cdec56e72b Landing directly for testing the ios buildbot code.
BUG=skia:
TBR=borenet

Review URL: https://codereview.chromium.org/1087653005
2015-04-21 08:31:59 -07:00
mtklein
a4a0aeb748 Revert of Convert Color32 code to perfect blend. (patchset #6 id:100001 of https://codereview.chromium.org/1098913002/)
Reason for revert:
Xfermode_SrcOver not looking encouraging.  Up to 50% regressions.

https://perf.skia.org/#3242

Original issue's description:
> Convert Color32 code to perfect blend.
>
> Before we commit to blend_256_round_alt, let's make sure blend_perfect is
> really slower in practice (i.e. regresses on perf.skia.org).
>
> blend_perfect is really the most desirable algorithm if we can afford it.  Not
> only is it correct, but it's easy to think about and break into correct pieces:
> for instance, its div255() doesn't require any coordination with the multiply.
>
> This looks like a 30% hit according to microbenches.  That said, microbenches
> said my previous change would be a 20-25% perf improvement, but it didn't end
> up showing a significant effect at a high level.
>
> As for correctness, I see a bunch of off-by-1 compared to blend_256_round_alt
> (exactly what we'd expect), and one off-by-3 in a GM that looks like it has a
> bunch of overdraw.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/61221e7f87a99765b0e034020e06bb018e2a08c2

TBR=reed@google.com,fmalita@chromium.org,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1083923006
2015-04-21 08:09:30 -07:00
joshualitt
8672f4dffa Pack BitmapTextBlob a bit tighter
BUG=skia:

Review URL: https://codereview.chromium.org/1092973004
2015-04-21 08:03:04 -07:00
joshualitt
25ba7ea2ed reuse scaler across consecutive textbatch flushes
BUG=skia:

Review URL: https://codereview.chromium.org/1091313002
2015-04-21 07:49:49 -07:00
joshualitt
fcfb9fc4bd A small patch to enable distance field text in textblobs
BUG=skia:

Review URL: https://codereview.chromium.org/1098653005
2015-04-21 07:35:10 -07:00
senorblanco
c8e934007a Yet more ASAN fixes.
More nullptr checks for factories I have added.
Other checks more Yoda-like I have made. (Skia style this is.)

BUG=skia:

Review URL: https://codereview.chromium.org/1086393004
2015-04-21 07:20:36 -07:00
senorblanco
84bfd3958f Another speculative ASAN fix.
Add a GrContext nullptr check in SurfaceTest.

TBR=robertphillips@google.com

Review URL: https://codereview.chromium.org/1098993004
2015-04-21 06:59:17 -07:00
mtklein
33d73c39dd SkTHash: remove()
BUG=skia:

Review URL: https://codereview.chromium.org/1057043003
2015-04-21 06:53:56 -07:00
egdaniel
bc227140ff Undo the bounds mapping of bounds for GrOvalRenderer for certain draws
BUG=skia:

Review URL: https://codereview.chromium.org/1099873002
2015-04-21 06:28:08 -07:00
caryclark
df386c54bd fix valgrind uninitialized var error
R=mtklein@google.com

Review URL: https://codereview.chromium.org/1097293002
2015-04-21 05:27:02 -07:00
senorblanco
0e95b116fc Speculative fix for ASAN build.
Check for NULL GrContext in TessellatingPathRendererTests.
Also fix an upload nag (no #includes before GPU #ifdef).

TBR=bsalomon@google.com

Review URL: https://codereview.chromium.org/1097943004
2015-04-20 15:00:06 -07:00
senorblanco
f3c2c46cf8 Make the GPU dashing effect MSAA-friendly.
Review URL: https://codereview.chromium.org/1092793006
2015-04-20 14:44:26 -07:00
bungeman
e1ddb3b190 Regularize lmp::XXX_element_handler.
This makes all non-XMLCALL element_handler functions uniform by making
their signature

void XXX_element_handler(FamilyData* self, const char** attributes)

This makes the code somewhat easier to follow, as it makes explicit
which functions are actually logically methods of FamilyData (start
with 'FamilyData* self') and which are called directly by expat
(start with 'void* data').

Review URL: https://codereview.chromium.org/1086943007
2015-04-20 13:31:46 -07:00
jvanverth
d1e72875d1 Minor modifications to GrProgramDesc.h.
BUG=skia:

Review URL: https://codereview.chromium.org/1100483003
2015-04-20 12:29:37 -07:00
egdaniel
9ef1bb1375 Return correctly mapped bounds from GrOvalRenderer.
Before we were returning non transformed bounds for the ovals instead of
device bounds from the various draw calls. This was causing us to do dst
copies from the wrong portion of the screen when there was some transform
on the matrix.

This fixes the gpu issues from the new imagefilters-xfermodes gm as well an
old bug from the srcmode GM.

BUG=skia:

Review URL: https://codereview.chromium.org/1090453003
2015-04-20 12:28:57 -07:00
bungeman
f61475e95d Regularize informative messages in Android font parser.
The Android font parser prints a number of helpful messages when
something seems amiss in the configuration file. Give all of these
messages a common and more informative format.

Review URL: https://codereview.chromium.org/1096063003
2015-04-20 11:16:48 -07:00
mtklein
61221e7f87 Convert Color32 code to perfect blend.
Before we commit to blend_256_round_alt, let's make sure blend_perfect is
really slower in practice (i.e. regresses on perf.skia.org).

blend_perfect is really the most desirable algorithm if we can afford it.  Not
only is it correct, but it's easy to think about and break into correct pieces:
for instance, its div255() doesn't require any coordination with the multiply.

This looks like a 30% hit according to microbenches.  That said, microbenches
said my previous change would be a 20-25% perf improvement, but it didn't end
up showing a significant effect at a high level.

As for correctness, I see a bunch of off-by-1 compared to blend_256_round_alt
(exactly what we'd expect), and one off-by-3 in a GM that looks like it has a
bunch of overdraw.

BUG=skia:

Review URL: https://codereview.chromium.org/1098913002
2015-04-20 10:52:26 -07:00
reed
4912437891 adjust GM to have a final opaque bg, to make it easier to triage
BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1085333003
2015-04-20 09:10:31 -07:00
caryclark
1049f1246e Now, path ops natively intersect conics, quads, and cubics in any combination. There are still a class of cubic tests that fail and a handful of undiagnosed failures from skps and fuzz tests, but things are much better overall.
Extended tests (150M+) run to completion in release in about 6 minutes; the standard test suite exceeds 100K and finishes in a few seconds on desktops.

TBR=reed
BUG=skia:3588

Review URL: https://codereview.chromium.org/1037953004
2015-04-20 08:31:59 -07:00
reed
5c476fb277 apply xfermode after imagefilter when we create a tmp layer
BUG=skia:3741

Review URL: https://codereview.chromium.org/1091173003
2015-04-20 08:04:21 -07:00
reed
7e4186acce add GM to test imagefilters + xfermodes
draws incorrectly as is, but will be fixed by https://codereview.chromium.org/1091173003

BUG=skia:3741

Review URL: https://codereview.chromium.org/1088773004
2015-04-20 07:27:15 -07:00
scroggo
29ccd898d1 Update our libjpeg to the version used by Chromium.
We had previously not updated since 2013. This includes the fix
for http://seclists.org/fulldisclosure/2013/Nov/83

BUG=skia:3744

Review URL: https://codereview.chromium.org/1097943002
2015-04-20 06:58:49 -07:00
senorblanco
6d88bdb6f1 Fix point count computation in tessellating path renderer.
The conic, quadratic and cubic computations were using tolerance^2
instead of tolerance when computing maximum point count, causing paths
to be undertessellated when magnifying and overtessellated when minifying.

(Funny story: this bug went unnoticed back when we were tessellating
paths in screen space, when tolerance and its square were both 1.)

BUG=skia:3731

Review URL: https://codereview.chromium.org/1095773003
2015-04-20 05:41:48 -07:00
reed
a8ec4805db remove dead code from SK_SUPPORT_LEGACY_BLITANTIH2V2
no image changes expected in skia or chrome/blink

BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1050953003
2015-04-20 05:18:01 -07:00
skia.buildbots
d13444acf2 Update SKP version
Automatic commit by the RecreateSKPs bot.

TBR=

Review URL: https://codereview.chromium.org/1095953002
2015-04-19 01:15:08 -07:00
mtklein
f3f9440372 Revert of Allow NEON on iOS. (patchset #3 id:40001 of https://codereview.chromium.org/1091823002/)
Reason for revert:
need to coordinate this with chrome gyps/gn

Original issue's description:
> Allow NEON on iOS.
>
> I have nanobench building and running (Color32_arm_neon) on my iPad.
>
> No public API changes.
> TBR=reed@google.com
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/e5043b7ea5170a639367b67c986568907576be4b

TBR=caryclark@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1094843005
2015-04-17 14:02:40 -07:00
bungeman
e85a754a4c Respect declared font style on Android.
Previously the normal/italic style bit was obtained from scanning the
font file. With the new format the style may be stated explicitly, and
this explicit value in the configuration file should override any
information obtained from the font data itself.

This change allows the font element's style attribute to override the
font's style, but retains the default 'auto' setting for backwards
compatibility. Repecting the style bit may become more important with
variation fonts, because it will be up to the configuration writer to
determine what values of the 'slnt' variation should be considered
'normal' or 'italic'.

DOCS_PREVIEW= https://skia.org/?cl=1092093002

Committed: https://skia.googlesource.com/skia/+/673e902c9b9982a167f54f1cc175d8d9cab8bcaf

Review URL: https://codereview.chromium.org/1092093002
2015-04-17 13:51:08 -07:00
jcgregorio
21d3ed52e5 Revert of Respect declared font style on Android. (patchset #6 id:100001 of https://codereview.chromium.org/1092093002/)
Reason for revert:
Failed on the compile bots.

Original issue's description:
> Respect declared font style on Android.
>
> Previously the normal/italic style bit was obtained from scanning the
> font file. With the new format the style may be stated explicitly, and
> this explicit value in the configuration file should override any
> information obtained from the font data itself.
>
> This change allows the font element's style attribute to override the
> font's style, but retains the default 'auto' setting for backwards
> compatibility. Repecting the style bit may become more important with
> variation fonts, because it will be up to the configuration writer to
> determine what values of the 'slnt' variation should be considered
> 'normal' or 'italic'.
>
> DOCS_PREVIEW= https://skia.org/?cl=1092093002
>
> Committed: https://skia.googlesource.com/skia/+/673e902c9b9982a167f54f1cc175d8d9cab8bcaf

TBR=mtklein@google.com,tomhudson@google.com,scroggo@google.com,bungeman@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1082173004
2015-04-17 13:30:35 -07:00
halcanary
a8448bc3df PDF: Correctly embed JPEG images directly into PDF output.
We only embed images with YUV planes.  That should only grab the
subset of color JPEGs supported by PDF.

BUG=skia:3180

Review URL: https://codereview.chromium.org/1025773002
2015-04-17 13:27:24 -07:00
bungeman
673e902c9b Respect declared font style on Android.
Previously the normal/italic style bit was obtained from scanning the
font file. With the new format the style may be stated explicitly, and
this explicit value in the configuration file should override any
information obtained from the font data itself.

This change allows the font element's style attribute to override the
font's style, but retains the default 'auto' setting for backwards
compatibility. Repecting the style bit may become more important with
variation fonts, because it will be up to the configuration writer to
determine what values of the 'slnt' variation should be considered
'normal' or 'italic'.

DOCS_PREVIEW= https://skia.org/?cl=1092093002

Review URL: https://codereview.chromium.org/1092093002
2015-04-17 13:25:03 -07:00
enne
54b666b144 Trace contention for lockPixels mutex
Review URL: https://codereview.chromium.org/1090103002
2015-04-17 12:42:49 -07:00
mtklein
afe2ffb8ba Rework SSE and NEON Color32 algorithms to be more correct and faster.
This algorithm changes the blend math, guarded by SK_LEGACY_COLOR32_MATH.  The new math is more correct: it's never off by more than 1, and correct in all the interesting 0x00 and 0xFF edge cases, where the old math was never off by more than 2, and not always correct on the edges.

If you look at tests/BlendTest.cpp, the old code was using the `blend_256_plus1_trunc` algorithm, while the new code uses `blend_256_round_alt`.  Neither uses `blend_perfect`, which is about ~35% slower than `blend_256_round_alt`.

This will require an unfathomable number of rebaselines, first to Skia, then to Blink when I remove the guard.

I plan to follow up with some integer SIMD abstractions that can unify these two implementations into a single algorithm.  This was originally what I was working on here, but the correctness gains seem to be quite compelling.  The only places these two algorithms really differ greatly now is the kernel function, and even there they can really both be expressed abstractly as:
  - multiply 8-bits and 8-bits producing 16-bits
  - add 16-bits to 16-bits, returning the top 8 bits.
All the constants are the same, except SSE is a little faster to keep 8 16-bit inverse alphas, NEON's a little faster to keep 8 8-bit inverse alphas.  I may need to take this small speed win back to unify the two.

We should expect a ~25% speedup on Intel (mostly from unrolling to 8 pixels) and a ~20% speedup on ARM (mostly from using vaddhn to add `color`, round, and narrow back down to 8-bit all into one instruction.

(I am probably missing several more related bugs here.)
BUG=skia:3738,skia:420,chromium:111470

Review URL: https://codereview.chromium.org/1092433002
2015-04-17 11:00:55 -07:00
bungeman
9d911d5a93 Remove filename from SkFontDescriptor.
No one actually uses it, so remove it. Keep the string id for backward
compatibility.

TBR=reed@google.com
This doesn't change any public API.

Review URL: https://codereview.chromium.org/1057413005
2015-04-17 11:00:06 -07:00
joshualitt
fec19e1751 Small change to allow DistanceField and BMP text to coexist in a blob
BUG=skia:

Review URL: https://codereview.chromium.org/1096753002
2015-04-17 10:32:32 -07:00
bungeman
f4fddfc300 Correct SkFixedToDouble.
SkFixedToDouble uses the decimal floating point string representation
of 1/(2^16) to provide a constant value. This value is exactly
representable by a double but the decimal constant currently used
does not map to this value.

This change modifies this constant to more decimal digits so that
the actual double value desired is created.

Review URL: https://codereview.chromium.org/1097473002
2015-04-17 10:05:43 -07:00
joshualitt
9bd2daff8e Initial CL to add distance field support to GrAtlasTextContext
BUG=skia:

Review URL: https://codereview.chromium.org/1082843002
2015-04-17 09:30:07 -07:00
joshualitt
523ed6c0c1 adding new gm to rotate, translate, and scale textblobs
this is to test distance field caching and regen of texture blobs on scale.

BUG=skia:

Review URL: https://codereview.chromium.org/1097563002
2015-04-17 09:01:05 -07:00