bsalomon [Thu, 30 Apr 2015 21:18:54 +0000 (14:18 -0700)]
This replaces the texture creation/caching functions on GrContext with a GrTextureProvider interface. The goal is to pass this narrowly focused object in places that currently take a GrContext but don't need and shouldn't use its other methods. It also has an extended private interface for interacting with non-texture resource types.
Review URL: https://codereview.chromium.org/
1107973004
bungeman [Thu, 30 Apr 2015 21:12:58 +0000 (17:12 -0400)]
Move resource fonts to common location.
Move resource fonts to resources/fonts, add a destortable font for
testing, and clean up how the tests create fonts from resources.
R=joshualitt@google.com
Review URL: https://codereview.chromium.org/
1120823002
joshualitt [Thu, 30 Apr 2015 21:03:02 +0000 (14:03 -0700)]
Simple CL to Compute bounds in GrContext::drawVertices
BUG=skia:
Review URL: https://codereview.chromium.org/
1117953003
joshualitt [Thu, 30 Apr 2015 20:49:27 +0000 (13:49 -0700)]
Add batch unit tests for ovals
BUG=skia:
Review URL: https://codereview.chromium.org/
1108403005
reed [Thu, 30 Apr 2015 20:09:24 +0000 (13:09 -0700)]
add heuristic to pour small pictures into recordings, rather than ref'ing
BUG=skia:
Review URL: https://codereview.chromium.org/
1118693003
mtklein [Thu, 30 Apr 2015 19:06:19 +0000 (12:06 -0700)]
Make SkAutoTDelete's operator T*() const, like all the others.
BUG=skia:
Review URL: https://codereview.chromium.org/
1114043005
senorblanco [Thu, 30 Apr 2015 19:06:10 +0000 (12:06 -0700)]
Fix gpu memory accounting for auto-resolving MSAA.
BUG=skia:
Review URL: https://codereview.chromium.org/
1115233002
skia.buildbots [Thu, 30 Apr 2015 17:59:13 +0000 (10:59 -0700)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
NOTREECHECKS=true
Review URL: https://codereview.chromium.org/
1118053002
mtklein [Thu, 30 Apr 2015 16:15:44 +0000 (09:15 -0700)]
Add missing include to SkPictureContentInfo.
BUG=skia:
Review URL: https://codereview.chromium.org/
1113023003
joshualitt [Thu, 30 Apr 2015 16:05:54 +0000 (09:05 -0700)]
simple cleanup in GrTargetCommands
BUG=skia:
Review URL: https://codereview.chromium.org/
1116923003
robertphillips [Thu, 30 Apr 2015 15:55:06 +0000 (08:55 -0700)]
Fix for texture allocation failure in 'preAbandonGpuContext' mode
Review URL: https://codereview.chromium.org/
1118723003
reed [Thu, 30 Apr 2015 14:47:13 +0000 (07:47 -0700)]
use pathops utils to improve precision of cubic chopping in scan converter
BUG=skia:
Review URL: https://codereview.chromium.org/
1113963002
dongseong.hwang [Thu, 30 Apr 2015 14:47:05 +0000 (07:47 -0700)]
Fix build warning: 'this' pointer cannot be null in well-defined C++ code; comparison may be assumed to always evaluate to true [-Wtautological-undefined-compare]
Review URL: https://codereview.chromium.org/
1109933003
mtklein [Thu, 30 Apr 2015 14:35:27 +0000 (07:35 -0700)]
Looks like rssPages * pageSize is overflowing on 32-bit Linux...
... resulting in apparently negative memory usage:
(-2040/2058MB 5867) 1.81s serialize-8888 skp top25desk_google_com_calendar_.skp
Turns out long is 32-bit there, long long is 64-bit.
NOTREECHECKS=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1117923002
mtklein [Thu, 30 Apr 2015 14:11:22 +0000 (07:11 -0700)]
DM+nanobench: print both current and max RSS.
out/Debug/dm:
( 360/380 MB 2112) 396ms gpu gm xfermodes
out/Release/nanobench:
32/33 MB 1659 2.46µs 2.55µs 2.54µs 2.67µs 2% ▄▄▄▅▅▃▂▁▂█ gpu gradient_conicalZero_clamp_hicolor
out/Debug/nanobench:
42/42 MB desk_css3gradients.skp_1 8888
BUG=skia:
NOTREECHECKS=true
Review URL: https://codereview.chromium.org/
1115203002
dongseong.hwang [Thu, 30 Apr 2015 12:53:27 +0000 (05:53 -0700)]
Fix build warning: reference cannot be bound to dereferenced null pointer in well-defined C++ code; comparison may be assumed to always evaluate to true [-Wtautological-undefined-compare]
Review URL: https://codereview.chromium.org/
1110353003
skia.buildbots [Thu, 30 Apr 2015 12:19:38 +0000 (05:19 -0700)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
Review URL: https://codereview.chromium.org/
1118683002
bungeman [Wed, 29 Apr 2015 21:35:49 +0000 (14:35 -0700)]
Check for NULL familyName in SkFontMgr_fontconfig.
If the familyName is NULL, don't bother spending time trying to add
it to the FcPattern.
Review URL: https://codereview.chromium.org/
1115073002
cdalton [Wed, 29 Apr 2015 21:17:00 +0000 (14:17 -0700)]
Use texture barriers to read directly from the RT
Updates GrXferProcessor to read directly from the RT texture when
texture barriers are supported and it needs to know the dst color.
Also adds the notion of an Xfer barrier and uses it to issue texture
barriers when the XP will read the RT.
BUG=skia:
Review URL: https://codereview.chromium.org/
1040303002
reed [Wed, 29 Apr 2015 19:18:57 +0000 (12:18 -0700)]
Use colorfilter instead of imagefilter when possible
needs https://codereview.chromium.org/
1115713003 to land first in chrome
BUG=skia:
Review URL: https://codereview.chromium.org/
986353002
egdaniel [Wed, 29 Apr 2015 18:54:42 +0000 (11:54 -0700)]
Make XPFragmentBuilder only Builder with access to DstCopy.
Plus a bunch of renaming.
BUG=skia:
Review URL: https://codereview.chromium.org/
1110033004
bungeman [Wed, 29 Apr 2015 18:32:57 +0000 (11:32 -0700)]
FontConfig character/lang match requires weak family.
In FontConfig the normal priority order of matches is character, family,
language; however family can be marked as weak which changes the order to
character, language, family. This is what is desired in
matchFamilyStyleCharacter.
Review URL: https://codereview.chromium.org/
1118553002
joshualitt [Wed, 29 Apr 2015 18:32:05 +0000 (11:32 -0700)]
Add dash lines batch unit test
BUG=skia:
Review URL: https://codereview.chromium.org/
1108053004
jvanverth [Wed, 29 Apr 2015 18:18:05 +0000 (11:18 -0700)]
Pull out shader-specific caps into GrShaderCaps and GrGLSLCaps
BUG=skia:
Review URL: https://codereview.chromium.org/
1116713002
rmistry [Wed, 29 Apr 2015 17:15:07 +0000 (10:15 -0700)]
Handpick 16 Skia page sets to create SP enabled SKPs from
Context is in: https://code.google.com/p/skia/issues/detail?id=3763#c18
NOTRY=true
BUG=skia:3763
Review URL: https://codereview.chromium.org/
1115003002
cdalton [Wed, 29 Apr 2015 17:14:03 +0000 (10:14 -0700)]
Fix double blend in GrAAConvexPathRenderer
Updates GrAAConvexPathRenderer to not draw the interior fan of a
convex path when it does not exist (i.e when segment count <= 2).
TODO: We should also detect and combine colinear segments in order to
make sure we catch every case.
BUG=skia:
Review URL: https://codereview.chromium.org/
1094293002
joshualitt [Wed, 29 Apr 2015 17:01:22 +0000 (10:01 -0700)]
Initial CL to create GrBatchTest infrastructure
BUG=skia:
Review URL: https://codereview.chromium.org/
1109153004
reed [Wed, 29 Apr 2015 15:34:00 +0000 (08:34 -0700)]
onCreateDevice -> NULL now means the caller should create its own (bitmap) device
BUG=skia:
Review URL: https://codereview.chromium.org/
1116453002
caryclark [Wed, 29 Apr 2015 15:28:30 +0000 (08:28 -0700)]
minor fixes to cubics code and overall alignment of how bounds and tops are computed for all curve types
All but 17 extended tests work.
A helper function is privately added to SkPath.h to permit a test to modify a given point in a path.
BUG=skia:3588
Review URL: https://codereview.chromium.org/
1107353004
msarett [Wed, 29 Apr 2015 15:17:15 +0000 (08:17 -0700)]
Implementing a scanline decoder for jpeg
BUG=skia:3257
Review URL: https://codereview.chromium.org/
1092303003
mtklein [Wed, 29 Apr 2015 15:16:50 +0000 (08:16 -0700)]
Straighten out SkPicture::AccelData APIs.
Must have been we needed them to be weird (mutable, const setter) before.
It doesn't look like that's necessary now... we can just pass it to the
constructor.
BUG=skia:
Review URL: https://codereview.chromium.org/
1112833003
bungeman [Wed, 29 Apr 2015 15:15:31 +0000 (08:15 -0700)]
Remove system call in DW for default font.
The last user of the GDI default font in DW is
onMatchFamilyStyleCharacter and it doesn't appear to actually need
to use it. This change removes a system call, making things a bit
faster, as well as eliminating a call blocked in the sandbox.
BUG=chromium:459056
Review URL: https://codereview.chromium.org/
1107283003
msarett [Wed, 29 Apr 2015 14:38:43 +0000 (07:38 -0700)]
Requires a minor API change to pass the color table to getScanlineDecoder, as we do with getPixels().
BUG=skia:3722
Review URL: https://codereview.chromium.org/
1061713007
rmistry [Wed, 29 Apr 2015 14:26:33 +0000 (07:26 -0700)]
Create SP enabled SKPs from only the top_25_smooth page set.
Context is here: https://code.google.com/p/skia/issues/detail?id=3763#c14
BUG=skia:3763
NOTRY=true
Review URL: https://codereview.chromium.org/
1111273003
bsalomon [Wed, 29 Apr 2015 11:33:52 +0000 (04:33 -0700)]
Remove unneeded features from GrBufferAllocPool
Review URL: https://codereview.chromium.org/
1103423004
kkinnunen [Wed, 29 Apr 2015 05:19:25 +0000 (22:19 -0700)]
Pass dashing info from drawPath to Gr side
Pass dashing info also from drawPath calls to gpu path rendering code.
Previously dashing info was preserved only for fast-pathed drawRects and
such.
This enables path renderers to see dashing info for arbitrary paths.
Review URL: https://codereview.chromium.org/
1114463002
reed [Wed, 29 Apr 2015 00:50:31 +0000 (17:50 -0700)]
SkRWBuffer for thread-safe 'stream' sharing
WIP
- Can accumulate (write) data in one thread, and share snapshots of it in other threads
... e.g. network accumulates image data, and periodically we want to decode/draw it
- If this sort of thing sticks, should we promote SkData to have the same generality as
SkRBuffer?
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1106113002
bsalomon [Tue, 28 Apr 2015 20:35:17 +0000 (13:35 -0700)]
Remove GrFlushToGpuDrawTarget and move functionality up to GrDrawTarget.
Review URL: https://codereview.chromium.org/
1117433002
joshualitt [Tue, 28 Apr 2015 20:31:18 +0000 (13:31 -0700)]
create GrTestUtils.h, move some common functions into it
BUG=skia:
Review URL: https://codereview.chromium.org/
1117443002
halcanary [Tue, 28 Apr 2015 20:06:54 +0000 (13:06 -0700)]
doc: add quickstart to linux and macos
also, better install_deps
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=
1112663002
Review URL: https://codereview.chromium.org/
1112663002
dongseong.hwang [Tue, 28 Apr 2015 19:47:48 +0000 (12:47 -0700)]
Remove useless SkASSERT in SkRect.h and SkRegion.h
The pointer of reference is always not-null. It causes >50k lines of warnning when chromium is compiled by clang.
Warning looks like
../../third_party/skia/include/core/SkRect.h:284:19: warning: reference cannot be bound to dereferenced null pointer in well-defined C++ code; pointer may be assumed to always convert to true [-Wundefined-bool-conversion]
do { if (&r) break; SkNO_RETURN_HINT(); SkDebugf_FileLine("../../third_party/skia/include/core/SkRect.h", 284, false, "%s:%d: failed assertion \"%s\"\n", "../../third_party/skia/include/core/SkRect.h", 284, "&r"); SkDebugf_FileLine("../../third_party/skia/include/core/SkRect.h", 284, true, "SK_CRASH"); } while (false);
Review URL: https://codereview.chromium.org/
1111013002
bsalomon [Tue, 28 Apr 2015 19:43:01 +0000 (12:43 -0700)]
Do circular clipping in normalized space
BUG=skia:426217
Review URL: https://codereview.chromium.org/
1110173002
dongseong.hwang [Tue, 28 Apr 2015 19:26:52 +0000 (12:26 -0700)]
Fix SkTDynamicHash.h:183:29: warning: equality comparison with extraneous parentheses [-Wparentheses-equality]
Review URL: https://codereview.chromium.org/
1115433004
rmistry [Tue, 28 Apr 2015 17:28:13 +0000 (10:28 -0700)]
Update documentation for "COMMIT=false".
The meaning of "COMMIT=false" was changed in https://code.google.com/p/chromium/issues/detail?id=470609.
I have also updated the corresponding Chromium documentation in https://sites.google.com/a/chromium.org/dev/developers/testing/commit-queue
BUG=skia:
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=
1107343004
Review URL: https://codereview.chromium.org/
1107343004
borenet [Tue, 28 Apr 2015 16:54:55 +0000 (09:54 -0700)]
Run DM on GalaxyS* in one thread
Drive-by cleanup of Xoom
BUG=skia:3214
Review URL: https://codereview.chromium.org/
1106173004
joshualitt [Tue, 28 Apr 2015 16:17:05 +0000 (09:17 -0700)]
Revert of removing equality / compute invariant loops from GrGeometryProcessors (patchset #2 id:20001 of https://codereview.chromium.org/
1111603004/)
Reason for revert:
breaks gl programs
Original issue's description:
> removing equality / compute invariant loops from GrGeometryProcessors
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
ef292a0901205b9785a30daae2c036aa34a970ca
TBR=bsalomon@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1110993002
joshualitt [Tue, 28 Apr 2015 16:08:28 +0000 (09:08 -0700)]
removing equality / compute invariant loops from GrGeometryProcessors
BUG=skia:
Review URL: https://codereview.chromium.org/
1111603004
jvanverth [Tue, 28 Apr 2015 15:48:20 +0000 (08:48 -0700)]
Use GLSLCaps for creating processor keys and GLSL-specific programs
Effectively all this does is future-proof any GLSL-specific code, as
GLSLCaps is just a typedef of GLCaps.
BUG=skia:
Review URL: https://codereview.chromium.org/
1109863004
rmistry [Tue, 28 Apr 2015 15:27:19 +0000 (08:27 -0700)]
Update RecreateSKPs bot to also generate Slimming Paint SKPs with a separate prefix.
Will only be committed after I get the go head in skia:3763.
BUG=skia:3763
NOTRY=true
Review URL: https://codereview.chromium.org/
1065503007
robertphillips [Tue, 28 Apr 2015 15:13:51 +0000 (08:13 -0700)]
Fix memory deletion error in "convex-lineonly-paths" GM
TBR=humper@google.com
Review URL: https://codereview.chromium.org/
1110153002
fmalita [Tue, 28 Apr 2015 15:08:57 +0000 (08:08 -0700)]
Decrement SkCanvas::fDeferredSaveCount in doSave()
We're currently decrementing early in checkForDeferredSave(), but that
means we break the getSaveCount() invariant during willSave() dispatch.
Instead, we can defer to doSave(), until after willSave() is called.
This allows overriders to call getSaveCount().
R=reed@google.com,robertphillips@google.com
Review URL: https://codereview.chromium.org/
1113603002
bsalomon [Tue, 28 Apr 2015 15:08:46 +0000 (08:08 -0700)]
Remove vertex/index buffer factilities from GrDrawTarget.
Review URL: https://codereview.chromium.org/
1105263002
robertphillips [Tue, 28 Apr 2015 14:42:04 +0000 (07:42 -0700)]
Add new GM/bench for line-only convex paths
BUG=472723
Review URL: https://codereview.chromium.org/
1112603002
joshualitt [Tue, 28 Apr 2015 14:04:11 +0000 (07:04 -0700)]
remove old text contexts and fontcache
BUG=skia:
Review URL: https://codereview.chromium.org/
1104343003
halcanary [Tue, 28 Apr 2015 13:21:01 +0000 (06:21 -0700)]
SkPDF: clean up uses of deprecated calls in SkPDFDevice.
Clean up reference count handling to avoid churn and follow pattern used elsewhere in SkPDF.
BUG=skia:3585
Review URL: https://codereview.chromium.org/
1103163002
tfarina [Tue, 28 Apr 2015 00:18:28 +0000 (17:18 -0700)]
Update more directories under src/ to follow C++11 style rule for {virtual,override}.
The Google style guide states that only one of {virtual,override,final}
should be used for each declaration, since override implies virtual and
final implies both virtual and override.
The entries were found using the following command line:
$ find src/ -iname "*.h" -o -iname "*.cpp" | xargs pcregrep -M
"[^\n/]+virtual\ [^;{]+\ [a-zA-Z0-9_]+\([^;{]+\ override[ \n]*[;{]"
The regex was a courtesy of nick@chromium.org
BUG=None
R=mtklein@google.com
Review URL: https://codereview.chromium.org/
1110013002
mtklein [Mon, 27 Apr 2015 22:11:01 +0000 (15:11 -0700)]
De-proc Color32
Also strips SK_SUPPORT_LEGACY_COLOR32_MATH,
which is no longer needed.
Seems handy to have SkTypes include the relevant intrinsics when
we know we've got them, but I'm not married to it.
Locally this looks like a pointlessly small perf win, but I'm mostly
keen to get all the code together.
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
376e9bc206b69d9190f38dfebb132a8769bbd72b
Committed: https://skia.googlesource.com/skia/+/
d65dc0cedd5b50dd407b6ff8fdc39123f11511cc
CQ_EXTRA_TRYBOTS=client.skia.compile:Build-Ubuntu-GCC-Mips-Debug-Android-Trybot
Review URL: https://codereview.chromium.org/
1104183004
mtklein [Mon, 27 Apr 2015 21:51:16 +0000 (14:51 -0700)]
Revert of De-proc Color32 (patchset #5 id:80001 of https://codereview.chromium.org/
1104183004/)
Reason for revert:
duh
Original issue's description:
> De-proc Color32
>
> Also strips SK_SUPPORT_LEGACY_COLOR32_MATH,
> which is no longer needed.
>
> Seems handy to have SkTypes include the relevant intrinsics when
> we know we've got them, but I'm not married to it.
>
> Locally this looks like a pointlessly small perf win, but I'm mostly
> keen to get all the code together.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
376e9bc206b69d9190f38dfebb132a8769bbd72b
>
> Committed: https://skia.googlesource.com/skia/+/
d65dc0cedd5b50dd407b6ff8fdc39123f11511cc
TBR=reed@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1102363006
mtklein [Mon, 27 Apr 2015 21:47:22 +0000 (14:47 -0700)]
De-proc Color32
Also strips SK_SUPPORT_LEGACY_COLOR32_MATH,
which is no longer needed.
Seems handy to have SkTypes include the relevant intrinsics when
we know we've got them, but I'm not married to it.
Locally this looks like a pointlessly small perf win, but I'm mostly
keen to get all the code together.
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
376e9bc206b69d9190f38dfebb132a8769bbd72b
Review URL: https://codereview.chromium.org/
1104183004
mtklein [Mon, 27 Apr 2015 21:22:32 +0000 (14:22 -0700)]
Split rsqrt into rsqrt{0,1,2}, with increasing cost and precision on ARM
This is a logical no-op. Everything was using the equivalent of rsqrt1() before, and is now after.
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
9de16283fdc8cc0d31a84f503578d0ecea4e8297
CQ_EXTRA_TRYBOTS=client.skia.compile:Build-Ubuntu-GCC-Arm64-Debug-Android-Trybot
Review URL: https://codereview.chromium.org/
1109913002
mtklein [Mon, 27 Apr 2015 21:17:37 +0000 (14:17 -0700)]
Revert of De-proc Color32 (patchset #4 id:60001 of https://codereview.chromium.org/
1104183004/)
Reason for revert:
MIPS
Original issue's description:
> De-proc Color32
>
> Also strips SK_SUPPORT_LEGACY_COLOR32_MATH,
> which is no longer needed.
>
> Seems handy to have SkTypes include the relevant intrinsics when
> we know we've got them, but I'm not married to it.
>
> Locally this looks like a pointlessly small perf win, but I'm mostly
> keen to get all the code together.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
376e9bc206b69d9190f38dfebb132a8769bbd72b
TBR=reed@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1108163002
mtklein [Mon, 27 Apr 2015 21:14:16 +0000 (14:14 -0700)]
De-proc Color32
Also strips SK_SUPPORT_LEGACY_COLOR32_MATH,
which is no longer needed.
Seems handy to have SkTypes include the relevant intrinsics when
we know we've got them, but I'm not married to it.
Locally this looks like a pointlessly small perf win, but I'm mostly
keen to get all the code together.
BUG=skia:
Review URL: https://codereview.chromium.org/
1104183004
halcanary [Mon, 27 Apr 2015 21:00:09 +0000 (14:00 -0700)]
SkPDF: clean up uses of deprecated calls in other SkPDF classes
BUG=skia:3585
Review URL: https://codereview.chromium.org/
1107923002
mtklein [Mon, 27 Apr 2015 20:55:53 +0000 (13:55 -0700)]
Revert of Split rsqrt into rsqrt{0,1,2}, with increasing cost and precision on ARM (patchset #2 id:20001 of https://codereview.chromium.org/
1109913002/)
Reason for revert:
arm64 typos
Original issue's description:
> Split rsqrt into rsqrt{0,1,2}, with increasing cost and precision on ARM
>
> This is a logical no-op. Everything was using the equivalent of rsqrt1() before, and is now after.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
9de16283fdc8cc0d31a84f503578d0ecea4e8297
TBR=reed@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1105233003
mtklein [Mon, 27 Apr 2015 20:51:28 +0000 (13:51 -0700)]
Split rsqrt into rsqrt{0,1,2}, with increasing cost and precision on ARM
This is a logical no-op. Everything was using the equivalent of rsqrt1() before, and is now after.
BUG=skia:
Review URL: https://codereview.chromium.org/
1109913002
hcm [Mon, 27 Apr 2015 19:44:40 +0000 (12:44 -0700)]
Add presentations section and pathops file for test
BUG=skia:
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=
1111453002
Review URL: https://codereview.chromium.org/
1111453002
mtklein [Mon, 27 Apr 2015 19:08:01 +0000 (12:08 -0700)]
Mike's radial gradient CL with better float -> int.
patch from issue
1072303005 at patchset 40001 (http://crrev.com/
1072303005#ps40001)
This looks quite launchable. radial_gradient3, min of 100 samples:
N5: 985µs -> 946µs
MBP: 395µs -> 279µs
On my MBP, most of the meat looks like it's now in reading the cache and writing to dst one color at a time. Is that something we could do in float math rather than with a lookup table?
BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia.compile:Build-Mac10.8-Clang-Arm7-Debug-Android-Trybot,Build-Ubuntu-GCC-Arm7-Release-Android_NoNeon-Trybot
Committed: https://skia.googlesource.com/skia/+/
abf6c5cf95e921fae59efb487480e5b5081cf0ec
Review URL: https://codereview.chromium.org/
1109643002
joshualitt [Mon, 27 Apr 2015 19:03:05 +0000 (12:03 -0700)]
In this CL we stop closing batches prematurely, and only close immediately before a flush.
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
db3ce12c810ead7b76faa784e7293197aca0d9f1
Committed: https://skia.googlesource.com/skia/+/
9e830ead85e544924344b01436c44cfb09fcf177
Review URL: https://codereview.chromium.org/
1092403002
joshualitt [Mon, 27 Apr 2015 18:42:30 +0000 (11:42 -0700)]
Revert of Stop closing batch prematurely (patchset #4 id:60001 of https://codereview.chromium.org/
1092403002/)
Reason for revert:
really does break windows bots
Original issue's description:
> In this CL we stop closing batches prematurely, and only close immediately before a flush.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
db3ce12c810ead7b76faa784e7293197aca0d9f1
>
> Committed: https://skia.googlesource.com/skia/+/
9e830ead85e544924344b01436c44cfb09fcf177
TBR=bsalomon@google.com,robertphillips@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1105203002
joshualitt [Mon, 27 Apr 2015 18:29:09 +0000 (11:29 -0700)]
In this CL we stop closing batches prematurely, and only close immediately before a flush.
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
db3ce12c810ead7b76faa784e7293197aca0d9f1
Review URL: https://codereview.chromium.org/
1092403002
mtklein [Mon, 27 Apr 2015 18:21:16 +0000 (11:21 -0700)]
Revert of Mike's radial gradient CL with better float -> int. (patchset #7 id:120001 of https://codereview.chromium.org/
1109643002/)
Reason for revert:
compile failures.
Original issue's description:
> Mike's radial gradient CL with better float -> int.
>
> patch from issue
1072303005 at patchset 40001 (http://crrev.com/
1072303005#ps40001)
>
> This looks quite launchable. radial_gradient3, min of 100 samples:
> N5: 985µs -> 946µs
> MBP: 395µs -> 279µs
>
> On my MBP, most of the meat looks like it's now in reading the cache and writing to dst one color at a time. Is that something we could do in float math rather than with a lookup table?
>
> BUG=skia:
>
> CQ_EXTRA_TRYBOTS=client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Debug-Trybot,Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot
>
> Committed: https://skia.googlesource.com/skia/+/
abf6c5cf95e921fae59efb487480e5b5081cf0ec
TBR=reed@google.com,robertphillips@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1109883003
joshualitt [Mon, 27 Apr 2015 18:19:33 +0000 (11:19 -0700)]
Revert of Stop closing batch prematurely (patchset #4 id:60001 of https://codereview.chromium.org/
1092403002/)
Reason for revert:
breaking windows bots
Original issue's description:
> In this CL we stop closing batches prematurely, and only close immediately before a flush.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
db3ce12c810ead7b76faa784e7293197aca0d9f1
TBR=bsalomon@google.com,robertphillips@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1109773003
mtklein [Mon, 27 Apr 2015 18:13:53 +0000 (11:13 -0700)]
Mike's radial gradient CL with better float -> int.
patch from issue
1072303005 at patchset 40001 (http://crrev.com/
1072303005#ps40001)
This looks quite launchable. radial_gradient3, min of 100 samples:
N5: 985µs -> 946µs
MBP: 395µs -> 279µs
On my MBP, most of the meat looks like it's now in reading the cache and writing to dst one color at a time. Is that something we could do in float math rather than with a lookup table?
BUG=skia:
CQ_EXTRA_TRYBOTS=client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Debug-Trybot,Test-Android-GCC-Nexus9-CPU-Denver-Arm64-Debug-Trybot
Review URL: https://codereview.chromium.org/
1109643002
msarett [Mon, 27 Apr 2015 17:57:22 +0000 (10:57 -0700)]
Make Skia for Android libjpeg use ashmem
BUG=skia:1282
Review URL: https://codereview.chromium.org/
1108013002
joshualitt [Mon, 27 Apr 2015 17:43:19 +0000 (10:43 -0700)]
In this CL we stop closing batches prematurely, and only close immediately before a flush.
BUG=skia:
Review URL: https://codereview.chromium.org/
1092403002
jvanverth [Mon, 27 Apr 2015 17:36:27 +0000 (10:36 -0700)]
Remove legacy NVPR support
BUG=skia:
Review URL: https://codereview.chromium.org/
1110553003
bsalomon [Mon, 27 Apr 2015 17:07:04 +0000 (10:07 -0700)]
Make non-AA hairline stroke rects snap to pixels centers so they close.
BUG=skia:3717
Review URL: https://codereview.chromium.org/
1101663007
joshualitt [Mon, 27 Apr 2015 16:36:55 +0000 (09:36 -0700)]
fix a bug in scrolling BitmapText gpu cached textblobs
BUG=skia:
Review URL: https://codereview.chromium.org/
1103243002
joshualitt [Mon, 27 Apr 2015 16:16:57 +0000 (09:16 -0700)]
Change to add zoom animations to nanobench
BUG=skia:
Review URL: https://codereview.chromium.org/
1061323003
mtklein [Mon, 27 Apr 2015 15:54:23 +0000 (08:54 -0700)]
Reduce ref counting in SkPictureRecorder.
This may be a small help to slimming paint:
picture_overhead_draw 1.25us -> 1.22us 0.98x
picture_overhead_nodraw 318ns -> 276ns 0.87x
It certainly cannot hurt performance.
BUG=chromium:470553
TBR=reed@google.com
No public API changes.
Review URL: https://codereview.chromium.org/
1098183003
mtklein [Mon, 27 Apr 2015 15:45:01 +0000 (08:45 -0700)]
Add source-data sharding to DM.
E.g. out/Debug/dm --shards 4 --shard 1
This shards all inputs, GMs, SKPs, tests, etc.
BUG=skia:
Review URL: https://codereview.chromium.org/
1109813002
jbroman [Mon, 27 Apr 2015 14:55:33 +0000 (07:55 -0700)]
Reduce sizeof(SkPicture::Analysis) from 24 bytes to 2 bytes.
This saves "up to" 22 bytes per SkPicture.
Due to alignment, this reduces sizeof(SkPicture):
- from 88 to 72 bytes on Linux x64
- from 68 to 48 bytes on Android
(with Chrome's build settings)
It also somewhat simplifies the GPU veto logic.
BUG=skia:
[mtklein] No public API changes.
TBR=reed@google.com
Review URL: https://codereview.chromium.org/
1060863004
tfarina [Mon, 27 Apr 2015 14:53:07 +0000 (07:53 -0700)]
Update more directories under src/ to follow C++11 style rule for {virtual,override}.
The Google style guide states that only one of {virtual,override,final}
should be used for each declaration, since override implies virtual and
final implies both virtual and override.
The entries were found using the following command line:
$ find src/ -iname "*.h" -o -iname "*.cpp" | xargs pcregrep -M
"[^\n/]+virtual\ [^;{]+\ [a-zA-Z0-9_]+\([^;{]+\ override[ \n]*[;{]"
The regex was a courtesy of nick@chromium.org
BUG=None
R=mtklein@google.com
NOPRESUBMIT=true
Review URL: https://codereview.chromium.org/
1086143003
tfarina [Mon, 27 Apr 2015 14:01:44 +0000 (07:01 -0700)]
Cleanup: Remove unnecessary double-semicolons.
The entries were found by the following command line:
$ find . -regex ".*\.[cChH]\(pp\)?" | xargs git grep -e ';;' --and --not
-e 'for *(.*;;'
Which is a combination of http://stackoverflow.com/a/3858879 and
http://gitster.livejournal.com/27674.html
BUG=None
R=mtklein@google.com
Review URL: https://codereview.chromium.org/
1088763005
halcanary [Mon, 27 Apr 2015 13:41:35 +0000 (06:41 -0700)]
dm||nanobench --help looks nice when $COLUMNS == 80.
Also, add to --config help message
Review URL: https://codereview.chromium.org/
1105593002
caryclark [Mon, 27 Apr 2015 12:39:33 +0000 (05:39 -0700)]
fix valgrind
TBR=mtklein@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/
1102273003
kkinnunen [Mon, 27 Apr 2015 06:18:49 +0000 (23:18 -0700)]
Extract gpu line dashing to GrDashLinePathRenderer
Move line dashing logic from GrContext::drawPath to
GrDashLinePathRenderer. This makes it possible to let path renderers render arbitrary dashed paths.
End goal is to implement dashing in GrStencilAndCoverPathRenderer.
Review URL: https://codereview.chromium.org/
1100073003
skia.buildbots [Sun, 26 Apr 2015 08:10:31 +0000 (01:10 -0700)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
Review URL: https://codereview.chromium.org/
1060063009
tfarina [Sat, 25 Apr 2015 19:59:44 +0000 (12:59 -0700)]
Update {virtual,override} to follow C++11 style in src/utils.
The Google style guide states that only one of {virtual,override,final}
should be used for each declaration, since override implies virtual
and final implies both virtual and override.
The entries were found using the following command line:
$ find src/utils -iname "*.h" -o -iname "*.cpp" | xargs pcregrep -M
"[^\n/]+virtual\ [^;{]+\ [a-zA-Z0-9_]+\([^;{]+\ override[ \n]*[;{]"
The regex was a courtesy of nick@chromium.org
BUG=None
R=mtklein@google.com
Review URL: https://codereview.chromium.org/
1099333004
halcanary [Sat, 25 Apr 2015 13:45:07 +0000 (06:45 -0700)]
SkPDF: Refactor SkPDFObject heiararchy.
Flatten and use a tagged union type
Δmemory ~= -2.4% ± 0.5%
Δtime ~= -1.2% ± 0.2%
BUG=skia:3585
Review URL: https://codereview.chromium.org/
1069103003
joshualitt [Sat, 25 Apr 2015 01:47:11 +0000 (18:47 -0700)]
fix for GrBatchAtlas valgrind error
TBR=bsalomon@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/
1056383005
halcanary [Fri, 24 Apr 2015 21:33:06 +0000 (14:33 -0700)]
sync-and-gyp script: fix to work in MSYS shell
NOTRY=true
TBR=scroggo@google.com
Review URL: https://codereview.chromium.org/
1104613003
tfarina [Fri, 24 Apr 2015 21:25:33 +0000 (14:25 -0700)]
Update {virtual,override} to follow C++11 style in src/fonts.
The Google style guide states that only one of {virtual,override,final}
should be used for each declaration, since override implies virtual and
final implies both virtual and override.
The entries were found using the following command line:
$ find src/fonts -iname "*.h" -o -iname "*.cpp" | xargs pcregrep -M
"[^\n/]+virtual\ [^;{]+\ [a-zA-Z0-9_]+\([^;{]+\ override[ \n]*[;{]"
The regex was a courtesy of nick@chromium.org
BUG=None
R=mtklein@google.com
Review URL: https://codereview.chromium.org/
1101343002
tfarina [Fri, 24 Apr 2015 20:12:44 +0000 (13:12 -0700)]
Update {virtual,override} to follow C++11 style in src/pipe.
The Google style guide states that only one of {virtual,override,final}
should be used for each declaration, since override implies virtual and
final implies both virtual and override.
The entries were found using the following command line:
$ find src/pipe -iname "*.h" -or -iname "*.cpp" | xargs pcregrep -M
"[^\n/]+virtual\ [^;{]+\ [a-zA-Z0-9_]+\([^;{]+\ override[ \n]*[;{]"
The regex was a courtesy of nick@chromium.org
BUG=None
R=mtklein@google.com
Review URL: https://codereview.chromium.org/
1104863003
msarett [Fri, 24 Apr 2015 18:41:55 +0000 (11:41 -0700)]
Do not crash nanobench in debug mode
Instead print an error message
BUG=skia:
Review URL: https://codereview.chromium.org/
1102083002
robertphillips [Fri, 24 Apr 2015 18:10:51 +0000 (11:10 -0700)]
Fix some lingering typos
Review URL: https://codereview.chromium.org/
1101293003
bsalomon [Fri, 24 Apr 2015 18:08:25 +0000 (11:08 -0700)]
Revert of Remove validation exceptions in GrGLInterface. These are all now set by Chromium. (patchset #1 id:1 of https://codereview.chromium.org/
1100413003/)
Reason for revert:
Crashing bots
Original issue's description:
> Remove validation exceptions in GrGLInterface. These are all now set by Chromium.
>
> Committed: https://skia.googlesource.com/skia/+/
42d9f35a8daf7f24ba542ba3783ec46fd965ef2f
TBR=joshualitt@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1093323005
joshualitt [Fri, 24 Apr 2015 16:57:12 +0000 (09:57 -0700)]
simple fix for preabandon valgrind error
TBR=bsalomon@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/
1106663003
caryclark [Fri, 24 Apr 2015 16:08:57 +0000 (09:08 -0700)]
fix multiple intersection logic
When three or more curves intersect at the same point, ensure that
each curve records the intersections of the others. This fixes a
number of cubic tests.
TBR=reed@google.com
BUG=skia:3588
Review URL: https://codereview.chromium.org/
1105943002
joshualitt [Fri, 24 Apr 2015 15:33:21 +0000 (08:33 -0700)]
fix for using too much memory in GrBatchAtlas
BUG=skia:
Review URL: https://codereview.chromium.org/
1105913002