Commit Graph

17798 Commits

Author SHA1 Message Date
joshualitt
01f38915e1 Fixup android launch script to launch VisualBench as well as SampleApp
BUG=skia:

Review URL: https://codereview.chromium.org/1185473003
2015-06-22 05:39:32 -07:00
kkinnunen
78cff136a4 Make GrGLProgramDataManager not refcounted
Make GrGLProgramDataManager not refcounted, refcounting is not used for
the instances.

Review URL: https://codereview.chromium.org/1195573002
2015-06-21 22:55:12 -07:00
skia.buildbots
a45e6d2d79 Update SKP version
Automatic commit by the RecreateSKPs bot.

TBR=
NO_MERGE_BUILDS

Review URL: https://codereview.chromium.org/1194233002
2015-06-21 00:17:25 -07:00
reed
dab1f60d12 remove SK_SUPPORT_LEGACY_OPTIONLESS_GET_PIXELS (now in SkUserConfig.h)
BUG=skia:
TBR=
NOTRY=True

Review URL: https://codereview.chromium.org/1200623003
2015-06-20 08:45:59 -07:00
robertphillips
7b05ff11eb Rename all things "leaky" in SkDevice
Review URL: https://codereview.chromium.org/1198603002
2015-06-19 14:14:55 -07:00
joshualitt
b7134b64c6 Fix for assert in VisualBench
BUG=skia:

Review URL: https://codereview.chromium.org/1198433004
2015-06-19 12:46:33 -07:00
robertphillips
fcf7829b67 remove SkDeviceProperties
There is a lot more clean up to do here but this is probably a big enough bite.

Review URL: https://codereview.chromium.org/1196683003
2015-06-19 11:49:52 -07:00
jvanverth
97c595f304 Switch to glyphs as paths at a higher point size on Android
The regression occurred when we dropped the maximum DF size from 192
to 162, which meant that any glyph > 324 ended up being rendered as paths
rather than the previous > 384. This pushes the threshold for
rendering paths up to 384. Quality looks fine on high-res devices
which is why this is restricted to Android-only (low-res Android devices
should only rarely have text that large).

BUG=chromium:467569

Committed: https://skia.googlesource.com/skia/+/932d413e69845989fadaecf5bcb8686ec8c05032

Review URL: https://codereview.chromium.org/1183053005
2015-06-19 11:06:28 -07:00
egdaniel
393551e338 Move rect_memcopy from helper to global static.
BUG=skia:

Review URL: https://codereview.chromium.org/1197713003
2015-06-19 10:52:25 -07:00
msarett
7054257de9 Do not fail on images that are too small to subset decode.
Specifically (0x0) images being produced by webp scaled
decodes are causing problems.

BUG=skia:

Review URL: https://codereview.chromium.org/1192373003
2015-06-19 07:44:05 -07:00
msarett
a0c414d7b4 Prevent webp from producing 0 dimensional images
BUG=skia:

Review URL: https://codereview.chromium.org/1196643002
2015-06-19 07:34:30 -07:00
joshualitt
b8e0b4a328 Revert of Disable failing GPU tests on Nexus 6. (patchset #1 id:1 of https://codereview.chromium.org/1185753002/)
Reason for revert:
this should hopefully be fixed now.

Original issue's description:
> Disable failing GPU tests on Nexus 6.
>
> BUG=skia:3942
>
> Committed: https://skia.googlesource.com/skia/+/8ed08e54f4868d3ef88225bf873ce678e7716bd9

TBR=bsalomon@google.com,djsollen@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:3942

Review URL: https://codereview.chromium.org/1196653002
2015-06-19 07:28:15 -07:00
robertphillips
8b1fa6dd8d Fix precision error in https://codereview.chromium.org/1188433011/
TBR=reed@google.com

Review URL: https://codereview.chromium.org/1192853002
2015-06-19 06:40:02 -07:00
robertphillips
20eee3f047 Added check for ill-conditioned invert
sk_inv_determinant has a guard that the determinant can't get too big so this CL only checks if the determinant gets too small.

BUG=492263

Review URL: https://codereview.chromium.org/1188433011
2015-06-19 05:14:26 -07:00
bungeman
13a007d9df Expose SkFaceRec less.
The SkFaceRec class is some data which needs to remain alive for the
life of an FT_Face. ref_ft_face returns SkFaceRec, but confusingly
unref_ft_face takes an FT_Face. Since no one was using the SkFaceRec
for anything other than accessing the FT_Face, have ref_ft_face
return FT_Face instead and remove the unused SkFaceRec pointer from
the scaler context.

Review URL: https://codereview.chromium.org/1180223005
2015-06-19 05:09:39 -07:00
robertphillips
9fc8275a9b Remove gamma field from SkDeviceProperties
Review URL: https://codereview.chromium.org/1189223002
2015-06-19 04:46:45 -07:00
reed
ce777c9ea3 remove deprecated NewRasterPMColor
BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1192353002
2015-06-19 03:42:56 -07:00
fmalita
9aa9e716ec Revert of Switch to glyphs as paths at a higher point size on Android (patchset #1 id:1 of https://codereview.chromium.org/1183053005/)
Reason for revert:
GM:mixedtextblobs looks significantly worse on Android.

Before: https://gold.skia.org/img/images/19d171dd72ba68df9501736dbbc2aa44.png

After: https://gold.skia.org/img/images/d4c38072d2836356c00fdc1a7b537350.png

Original issue's description:
> Switch to glyphs as paths at a higher point size on Android
>
> The regression occurred when we dropped the maximum DF size from 192
> to 162, which meant that any glyph > 324 ended up being rendered as paths
> rather than the previous > 384. This pushes the threshold for
> rendering paths up to 384. Quality looks fine on high-res devices
> which is why this is restricted to Android-only (low-res Android devices
> should only rarely have text that large).
>
> BUG=chromium:467569
>
> Committed: https://skia.googlesource.com/skia/+/932d413e69845989fadaecf5bcb8686ec8c05032

TBR=joshualitt@google.com,jvanverth@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:467569

Review URL: https://codereview.chromium.org/1192203002
2015-06-18 14:34:53 -07:00
joshualitt
83bc229703 Work around for nexus 6 TexSubImage issue
BUG=skia:

Committed: https://skia.googlesource.com/skia/+/c69fe20ef35df0776b179cbd5ce4d1bdc0ba4090

Review URL: https://codereview.chromium.org/1173203005
2015-06-18 14:18:03 -07:00
reed
35a5261085 privatize
BUG=skia:
TBR=bsalomon@google.com

Review URL: https://codereview.chromium.org/1189173005
2015-06-18 14:05:07 -07:00
joshualitt
5599996b72 disable floating point textures if GLSL version is < 330
BUG=skia:

Review URL: https://codereview.chromium.org/1178723007
2015-06-18 13:47:10 -07:00
jvanverth
932d413e69 Switch to glyphs as paths at a higher point size on Android
The regression occurred when we dropped the maximum DF size from 192
to 162, which meant that any glyph > 324 ended up being rendered as paths
rather than the previous > 384. This pushes the threshold for
rendering paths up to 384. Quality looks fine on high-res devices
which is why this is restricted to Android-only (low-res Android devices
should only rarely have text that large).

BUG=chromium:467569

Review URL: https://codereview.chromium.org/1183053005
2015-06-18 13:46:46 -07:00
reed
7ab04d949b remove dead guarded code for legacy SkData proc
BUG=skia:

Review URL: https://codereview.chromium.org/1173173007
2015-06-18 13:42:03 -07:00
reed
de49988bc2 add callbacks to Images that wrap client-provided content
BUG=skia:

Review URL: https://codereview.chromium.org/1169553003
2015-06-18 13:41:40 -07:00
senorblanco
d7395d82d5 Initialize the sample config for wrapped RTs to kUnified.
BUG=skia:3958

Review URL: https://codereview.chromium.org/1177163005
2015-06-18 13:26:52 -07:00
joshualitt
74953c388c Modify android_install_app to install VisualBench
BUG=skia:

Review URL: https://codereview.chromium.org/1181973008
2015-06-18 13:08:26 -07:00
scroggo
6f5e619b87 Add SkWebpCodec, for decoding .webp images.
Based on SkImageDecoder_libwebp.

TODO:
Support YUV? (Longer term - may influence our API for SkImageGenerator)

BUG=skia:3257

Review URL: https://codereview.chromium.org/1044433002
2015-06-18 12:53:43 -07:00
reed
bedd0e87e6 remove SkData proc guard
BUG=skia:
NOTRY=True

Review URL: https://codereview.chromium.org/1188773008
2015-06-18 12:11:36 -07:00
cdalton
4271765d11 Don't call bindTexture from GrGLProgram
Updates GrGLProgram to tell the gpu object which textures it wants
bound, instead of calling bindTexture directly. This begins to break
its dependence on the specific GrGLGpu object.

BUG=skia:

Review URL: https://codereview.chromium.org/1192463003
2015-06-18 11:54:31 -07:00
bsalomon
241b56db1d Allows windowed apps to be built on the Mac just by writing a SkOSWindow subclass without needing nib/plist files.
Makes visualbench work on the Mac.

Review URL: https://codereview.chromium.org/1184143011
2015-06-18 11:49:42 -07:00
joshualitt
274980cda1 Revert of Work around for nexus 6 TexSubImage issue (patchset #4 id:60001 of https://codereview.chromium.org/1173203005/)
Reason for revert:
breaks gms

Original issue's description:
> Work around for nexus 6 TexSubImage issue
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/c69fe20ef35df0776b179cbd5ce4d1bdc0ba4090

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

Review URL: https://codereview.chromium.org/1190463006
2015-06-18 11:49:18 -07:00
joshualitt
c69fe20ef3 Work around for nexus 6 TexSubImage issue
BUG=skia:

Review URL: https://codereview.chromium.org/1173203005
2015-06-18 11:07:09 -07:00
bungeman
1e23736f59 Remove FT_HAS_COLOR definition.
The intent was to define FT_HAS_COLOR when building with a pre
2.5.1 version of FreeType for forward compatibility. However,
the definition here is wrong and also never used.

Review URL: https://codereview.chromium.org/1178943009
2015-06-18 10:48:08 -07:00
mtklein
cd08effd00 Plumb through out_row byte length so we can assert we stay underneath it.
Sadly, not asserting for me yet.  Can't hurt.

BUG=chromium:491660

Review URL: https://codereview.chromium.org/1187173005
2015-06-18 10:30:32 -07:00
mtklein
525e90ab92 Try again to rename one SkJpegUtility.h to remove ambiguity.
This time rename the new codec one, which is not referenced externally.

BUG=skia:

Review URL: https://codereview.chromium.org/1181093008
2015-06-18 09:58:57 -07:00
bsalomon
6dc6f5f4a1 Add support for creating texture backed images where Skia will delete the texture.
Review URL: https://codereview.chromium.org/1187523005
2015-06-18 09:12:17 -07:00
rmistry
4c58e085b2 Remove separate SP enabled SKP run since it is now on by default
Yesterday's RecreateSKPs_Canary run looks good:
http://build.chromium.org/p/client.skia.fyi/builders/Housekeeper-Nightly-RecreateSKPs_Canary/builds/143

The generated SKPs look very reasonably sized:
gs://chromium-skia-gm/playback_386/skps

BUG=skia:
NOTRY=true

Review URL: https://codereview.chromium.org/1191113002
2015-06-18 07:52:14 -07:00
bungeman
7b7c844bca Remove unused SkTypeface_android.h.
This file had one declaration in it which actually affected the
SkFontMgr and had nothing to do with typefaces specifically. As a
result, the declaration was moved to SkFontMgr_android.h. Now that
the users have been updated, remove this now unused file.

Review URL: https://codereview.chromium.org/1177173004
2015-06-18 07:46:58 -07:00
msarett
b3ec76b34f Update android perf script
BUG=skia:

Review URL: https://codereview.chromium.org/1185293006
2015-06-18 06:29:48 -07:00
scroggo
d49f48ddf7 Add test for SkRegion::writeToMemory.
When calling SkRegion::writeToMemory(NULL), it should return the same
number of bytes that it writes when calling
SkRegion::writeToMemory(buffer). Add a test to confirm this.

BUG=b/21271229

Review URL: https://codereview.chromium.org/1188293002
2015-06-18 06:16:36 -07:00
robertphillips
bcd7ab5c0d Add useDFT field to SkDeviceProperties
Review URL: https://codereview.chromium.org/1191943002
2015-06-18 05:27:18 -07:00
mtklein
00b621cfc0 Add sk_parallel_for()
This should be a drop-in replacement for most for-loops to make them run in parallel:
   for (int i = 0; i < N; i++) { code... }
   ~~~>
   sk_parallel_for(N, [&](int i) { code... });

This is just syntax sugar over SkTaskGroup to make this use case really easy to write.
There's no more overhead that we weren't already forced to add using an interface like batch(),
and no extra heap allocations.

I've replaced 3 uses of SkTaskGroup with sk_parallel_for:
  1) My unit tests for SkOnce.
  2) Cary's path fuzzer.
  3) SkMultiPictureDraw.
Performance should be the same.  Please compare left and right for readability. :)

BUG=skia:

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

Review URL: https://codereview.chromium.org/1184373003
2015-06-17 15:26:15 -07:00
jvanverth
5a9e2994c9 Revert of Bump up point where we switch to distance fields for large glyphs (patchset #1 id:1 of https://codereview.chromium.org/1184153004/)
Reason for revert:
Seeing some Nexus 4 perf regressions in individual tests in Chromium that may be due to this change. This doesn't appear to be the correct fix for the bug in any case.

Original issue's description:
> Bump up point where we switch to distance fields for large glyphs
>
> This is hopefully a temporary fix. It's unclear why distance fields
> are so much slower on the N4 (and N7).
>
> BUG=chromium:467569
>
> Committed: https://skia.googlesource.com/skia/+/0fce1fb02d93e66d42528f322f8aa4ca64ff0fb2

TBR=joshualitt@google.com,bsalomon@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:467569

Review URL: https://codereview.chromium.org/1178713005
2015-06-17 13:06:48 -07:00
bungeman
77f85adadb Move Android SkFontMgr::Factory to its own file.
Review URL: https://codereview.chromium.org/1190993002
2015-06-17 12:07:10 -07:00
mtklein
2a4685f8bb SkSemaphore: use SYNC_POLICY_LIFO on Mach.
Turns out _LIFO does work as we expected, keeping N < M threads active when we
have M live threads but only N units of work to do at a time.  Also as we
expected _FIFO does round-robin through the threads.

Performance doesn't seem to be affected, but let's do it anyway.
If nothing else this makes profiles a little easier to read.

I don't see POSIX or Windows equivalents.

BUG=skia:

Review URL: https://codereview.chromium.org/1192433006
2015-06-17 11:49:04 -07:00
reed
c292ad4c1d add guard for SkData proc change
BUG=skia:
NOTRY=True
TBR=

Review URL: https://codereview.chromium.org/1178023003
2015-06-17 11:48:57 -07:00
mtklein
61fa22be10 Add and use SkSemaphore
This allows a faster implementation of our SkTaskGroup thread pool.
It also means we don't need SkCondVar (which, remember, isn't supported on XP.)

Doing some testing with SampleApp, this really cuts down on the overhead from SkTaskGroup, e.g. 30% to 10%.

BUG=skia:

Review URL: https://codereview.chromium.org/1192573003
2015-06-17 10:50:25 -07:00
msarett
ab80e35fbd Improved subset benchmarks
I think these changes to the subset benchmarks cover what we discussed yesterday.

I removed the divisor benchmarks (2x2, 3x3) and changed the single subset benchmarks.

Also, we will no longer benchmark subset decodes on small images.

BUG=skia:

Review URL: https://codereview.chromium.org/1188223002
2015-06-17 10:28:22 -07:00
reed
6b7f34e34c simplify release-proc for SkData, removing unneeded size param
request https://codereview.chromium.org/1184953005 to land first

BUG=skia:

Review URL: https://codereview.chromium.org/1193553002
2015-06-17 09:58:25 -07:00
sheyang
12555d7efb Add signcla verifier to skia.
This verifier is just informative now.

BUG=457428

Review URL: https://codereview.chromium.org/1193433003
2015-06-17 09:15:52 -07:00