platform/upstream/libSkiaSharp.git
9 years agoskimage, die die die
mtklein [Mon, 15 Dec 2014 20:47:51 +0000 (12:47 -0800)]
skimage, die die die

DM's running with --images on all the bots now.  They look happy.

BUG=skia:3235

Review URL: https://codereview.chromium.org/804133003

9 years agoAdd xp optimization for RGB coverage.
egdaniel [Mon, 15 Dec 2014 20:38:53 +0000 (12:38 -0800)]
Add xp optimization for RGB coverage.

This is needed since clearColorStages is being changed to ignore color input. For RGB coverage,
we want to clear all the color stages (since we know the final output color), but we don't want
to ignore the color input since it is needed.

In future we will change this so the XP stores the color internally and thus can tell the GP to
simiply ignore color.

BUG=skia:

Review URL: https://codereview.chromium.org/791933006

9 years agoFix rrects that are large enough that we lose/gain a bit when we add the radius to...
reed [Mon, 15 Dec 2014 20:28:33 +0000 (12:28 -0800)]
Fix rrects that are large enough that we lose/gain a bit when we add the radius to a bounds coordinate.

add test that triggers assert in addRRect

BUG=skia:3239

Review URL: https://codereview.chromium.org/803153003

9 years agoSilence warnings from libnanomsg on Mac like we do on Linux.
mtklein [Mon, 15 Dec 2014 20:24:46 +0000 (12:24 -0800)]
Silence warnings from libnanomsg on Mac like we do on Linux.

This will silence these warnings:

../../third_party/externals/nanomsg/src/core/global.c:162:34: warning: missing field 'unused' initializer [-Wmissing-field-initializers]
static struct nn_global self = {0};

../../third_party/externals/nanomsg/src/utils/clock.c:44:61: warning: missing field 'denom' initializer [-Wmissing-field-initializers]
static mach_timebase_info_data_t nn_clock_timebase_info = {0};

BUG=skia:

Review URL: https://codereview.chromium.org/803113003

9 years agoRevert of Roll libwebp to v0.4.2 (latest stable) to fix annoying build warning. ...
mtklein [Mon, 15 Dec 2014 20:23:00 +0000 (12:23 -0800)]
Revert of Roll libwebp to v0.4.2 (latest stable) to fix annoying build warning. (patchset #1 id:1 of https://codereview.chromium.org/807553002/)

Reason for revert:
uses __builtins not available on all our compilers

Original issue's description:
> Roll libwebp to v0.4.2 (latest stable) to fix annoying build warning.
>
> This warning should now go away:
>
> ../../third_party/externals/libwebp/src/enc/quant.c:105:23: warning: unused variable 'kCoeffThresh' [-Wunused-const-variable]
> static const uint16_t kCoeffThresh[16] = {
>
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/e655557a551c545f4153b0764269298cd01cd0c0

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

Review URL: https://codereview.chromium.org/795823005

9 years agoRoll libwebp to v0.4.2 (latest stable) to fix annoying build warning.
mtklein [Mon, 15 Dec 2014 20:11:48 +0000 (12:11 -0800)]
Roll libwebp to v0.4.2 (latest stable) to fix annoying build warning.

This warning should now go away:

../../third_party/externals/libwebp/src/enc/quant.c:105:23: warning: unused variable 'kCoeffThresh' [-Wunused-const-variable]
static const uint16_t kCoeffThresh[16] = {

BUG=skia:

Review URL: https://codereview.chromium.org/807553002

9 years agoRewrite NaN checks in terms of SkScalarIsNaN()
ehsan.akhgari [Mon, 15 Dec 2014 20:08:47 +0000 (12:08 -0800)]
Rewrite NaN checks in terms of SkScalarIsNaN()

We are trying to replace Skia's NaN checker with our own in Mozilla,
so it would be nice to have to patch a single place by making sure
these NaN checks used SkScalarIsNaN().

BUG=skia:

Review URL: https://codereview.chromium.org/809443002

9 years agoRemove SkCanvas::isDrawingToLayer
junov [Mon, 15 Dec 2014 19:34:06 +0000 (11:34 -0800)]
Remove SkCanvas::isDrawingToLayer

BUG=3245

Review URL: https://codereview.chromium.org/803913005

9 years agoUpdate SKP version
skia.buildbots [Mon, 15 Dec 2014 19:09:05 +0000 (11:09 -0800)]
Update SKP version

Automatic commit by the RecreateSKPs bot.

TBR=

Review URL: https://codereview.chromium.org/803183002

9 years agofinal circles?
Brian Salomon [Mon, 15 Dec 2014 18:50:26 +0000 (13:50 -0500)]
final circles?

Review URL: https://codereview.chromium.org/796333005

9 years agoWAE on Macs too. That leaves only Android framework builds.
mtklein [Mon, 15 Dec 2014 18:38:42 +0000 (10:38 -0800)]
WAE on Macs too.  That leaves only Android framework builds.

Will rebase after yours lands.

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/161b6dee11a67e877ef5b339b17775b8e44c5927

Review URL: https://codereview.chromium.org/804063002

9 years agoRevert of WAE on Macs too. That leaves only Android framework builds. (patchset...
mtklein [Mon, 15 Dec 2014 17:49:12 +0000 (09:49 -0800)]
Revert of WAE on Macs too.  That leaves only Android framework builds. (patchset #3 id:40001 of https://codereview.chromium.org/804063002/)

Reason for revert:
Forgot CrOS and NaCl builds.

Original issue's description:
> WAE on Macs too.  That leaves only Android framework builds.
>
> Will rebase after yours lands.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/161b6dee11a67e877ef5b339b17775b8e44c5927

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

Review URL: https://codereview.chromium.org/805043002

9 years agoWAE on Macs too. That leaves only Android framework builds.
mtklein [Mon, 15 Dec 2014 17:45:00 +0000 (09:45 -0800)]
WAE on Macs too.  That leaves only Android framework builds.

Will rebase after yours lands.

BUG=skia:

Review URL: https://codereview.chromium.org/804063002

9 years agocircles circles circles
Brian Salomon [Mon, 15 Dec 2014 17:14:59 +0000 (12:14 -0500)]
circles circles circles

TBR=bensong@google.com

Review URL: https://codereview.chromium.org/805793003

9 years agofix last warnings on w64 and turn on w.a.e.
bsalomon [Mon, 15 Dec 2014 17:13:35 +0000 (09:13 -0800)]
fix last warnings on w64 and turn on w.a.e.

Review URL: https://codereview.chromium.org/801413002

9 years agoAdd SK_OVERRIDE to a few places that are missing it.
mtklein [Mon, 15 Dec 2014 17:08:33 +0000 (09:08 -0800)]
Add SK_OVERRIDE to a few places that are missing it.

This was done automatically by running clang-modernize -add-override -override-macros.

It's not clear to me whether include/ is already clean or I just can't get the tool to work for headers.

BUG=skia:

Review URL: https://codereview.chromium.org/808463002

9 years agoAdd more filetypes to DM's image tests.
scroggo [Mon, 15 Dec 2014 16:53:35 +0000 (08:53 -0800)]
Add more filetypes to DM's image tests.

ico, wbmp, plus the alternate suffix jpeg.

Also check for capitalized versions, since files sometimes use
capitalized suffixes.

BUG=skia:3235

Review URL: https://codereview.chromium.org/798383003

9 years agoMore circle rebaselines on gpu
Brian Salomon [Mon, 15 Dec 2014 16:44:14 +0000 (11:44 -0500)]
More circle rebaselines on gpu

TBR=bensong@google.com

Review URL: https://codereview.chromium.org/803123002

9 years agoRebaseline after circle normalization change.
Brian Salomon [Mon, 15 Dec 2014 16:34:37 +0000 (11:34 -0500)]
Rebaseline after circle normalization change.

TBR=bensong@google.com

Review URL: https://codereview.chromium.org/806893002

9 years agoadd dumpHex option to rect and rrect, to match path
reed [Mon, 15 Dec 2014 15:59:53 +0000 (07:59 -0800)]
add dumpHex option to rect and rrect, to match path

BUG=skia:

Review URL: https://codereview.chromium.org/801383002

9 years agoDo circle anti-aliasing in normalized space to avoid precision issues with half-float...
bsalomon [Mon, 15 Dec 2014 15:11:22 +0000 (07:11 -0800)]
Do circle anti-aliasing in normalized space to avoid precision issues with half-floats on Adreno.

Review URL: https://codereview.chromium.org/761593006

9 years agoensure 4-byte alignment for storage, as SkWriterBuffer requires it
reed [Mon, 15 Dec 2014 14:41:02 +0000 (06:41 -0800)]
ensure 4-byte alignment for storage, as SkWriterBuffer requires it

BUG=skia:3244

Review URL: https://codereview.chromium.org/805803002

9 years agoCleanup: Mark some overridden methods with 'SK_OVERRIDE'.
tfarina [Sun, 14 Dec 2014 23:20:10 +0000 (15:20 -0800)]
Cleanup: Mark some overridden methods with 'SK_OVERRIDE'.

This fixes errors like this:

../../include/gpu/effects/GrPorterDuffXferProcessor.h:27:25: error:
'name' overrides a member function but is not marked 'override'
[-Werror,-Winconsistent-missing-override]

BUG=skia:3075
TEST=ninja -C out/Debug skia_lib
TBR=bsalomon@google.com

Review URL: https://codereview.chromium.org/804813002

9 years agoadd get9 and set9 to matrix, to aid in making keys
reed [Sat, 13 Dec 2014 16:46:48 +0000 (08:46 -0800)]
add get9 and set9 to matrix, to aid in making keys

BUG=skia:
TBR=
NOTREECHECKS=True

Review URL: https://codereview.chromium.org/806543002

9 years agoGLPrograms leak fix
joshualitt [Sat, 13 Dec 2014 14:46:13 +0000 (06:46 -0800)]
GLPrograms leak fix

NOTREECHECKS=True
TBR=
BUG=skia:

Review URL: https://codereview.chromium.org/799063005

9 years agoDM warning-free on win64
bsalomon [Sat, 13 Dec 2014 00:41:46 +0000 (16:41 -0800)]
DM warning-free on win64

Review URL: https://codereview.chromium.org/805543002

9 years agoAdd image decoding mode to DM.
mtklein [Sat, 13 Dec 2014 00:41:12 +0000 (16:41 -0800)]
Add image decoding mode to DM.

This is meant to supplant skimage.

BUG=skia:3235

Review URL: https://codereview.chromium.org/802793002

9 years agoEven more win64 warning fixes
bsalomon [Fri, 12 Dec 2014 23:11:17 +0000 (15:11 -0800)]
Even more win64 warning fixes

Review URL: https://codereview.chromium.org/800993002

9 years agoRebaselines.
scroggo [Fri, 12 Dec 2014 23:10:55 +0000 (15:10 -0800)]
Rebaselines.

TBR=borenet

Review URL: https://codereview.chromium.org/799063004

9 years agoCheck in skimage expectations for nexus player.
scroggo [Fri, 12 Dec 2014 22:26:41 +0000 (14:26 -0800)]
Check in skimage expectations for nexus player.

TBR=borenet

Review URL: https://codereview.chromium.org/799113002

9 years agoimprove cull, so we don't create empty segments
reed [Fri, 12 Dec 2014 20:48:50 +0000 (12:48 -0800)]
improve cull, so we don't create empty segments

BUG=skia:3224

Review URL: https://codereview.chromium.org/798093002

9 years agoHarden validation for SkPathRef
robertphillips [Fri, 12 Dec 2014 20:47:59 +0000 (12:47 -0800)]
Harden validation for SkPathRef

Review URL: https://codereview.chromium.org/801003003

9 years agoAdd Herb as a Skia committer
hcm [Fri, 12 Dec 2014 20:37:28 +0000 (12:37 -0800)]
Add Herb as a Skia committer

TBR=hcm@google.com

NOTRY=true

Review URL: https://codereview.chromium.org/803633002

9 years agoHandle SkBitmaps which have no pixels in SkPDFImage.
jbroman [Fri, 12 Dec 2014 19:28:16 +0000 (11:28 -0800)]
Handle SkBitmaps which have no pixels in SkPDFImage.

A test that ensures this no longer crashes has been added.

BUG=skia:3232

Review URL: https://codereview.chromium.org/802713002

9 years agoUpdate expectations for Test-Mac10.9-MacMini6.2-HD4000-x86_64-Debug
scroggo [Fri, 12 Dec 2014 19:13:35 +0000 (11:13 -0800)]
Update expectations for Test-Mac10.9-MacMini6.2-HD4000-x86_64-Debug

TBR=borenet

Review URL: https://codereview.chromium.org/797273004

9 years agoRemove gms from ignore-tests.
egdaniel [Fri, 12 Dec 2014 19:01:29 +0000 (11:01 -0800)]
Remove gms from ignore-tests.

TBR=

BUG=skia:

Review URL: https://codereview.chromium.org/800033002

9 years agoBlindly check in GM and skimage expectations for new bots
borenet [Fri, 12 Dec 2014 17:50:23 +0000 (09:50 -0800)]
Blindly check in GM and skimage expectations for new bots

BUG=skia:

Review URL: https://codereview.chromium.org/798053002

9 years agoExtract cubic classification from gpu into geometry
caryclark [Fri, 12 Dec 2014 17:11:23 +0000 (09:11 -0800)]
Extract cubic classification from gpu into geometry

Move code from the GPU path utilities into geometry so that
path ops can share.

Review URL: https://codereview.chromium.org/787763007

9 years agoPass gyp_source_dir to generate_tool.
scroggo [Fri, 12 Dec 2014 17:01:55 +0000 (09:01 -0800)]
Pass gyp_source_dir to generate_tool.

When generating the tool makefiles, which also calls
android_framework_gyp, we also need to pass the gyp directory.

TBR=borenet
NOTRY=True
NOTREECHECKS=True

BUG=skia:3233

Review URL: https://codereview.chromium.org/804473005

9 years agoUpdate SKP version
skia.buildbots [Fri, 12 Dec 2014 16:56:53 +0000 (08:56 -0800)]
Update SKP version

Automatic commit by the RecreateSKPs bot.

TBR=

Review URL: https://codereview.chromium.org/797983002

9 years agoUpdate some Galaxy S4 expectations.
jvanverth [Fri, 12 Dec 2014 16:56:48 +0000 (08:56 -0800)]
Update some Galaxy S4 expectations.

TBR=djsollen@google.com

Review URL: https://codereview.chromium.org/804493002

9 years agoIgnore some expectations for Nexus 10.
jvanverth [Fri, 12 Dec 2014 16:51:46 +0000 (08:51 -0800)]
Ignore some expectations for Nexus 10.

BUG=skia:3205
TBR=bsalomon@google.com

Review URL: https://codereview.chromium.org/800483004

9 years agoMore win64 warning fixes
bsalomon [Fri, 12 Dec 2014 16:51:38 +0000 (08:51 -0800)]
More win64 warning fixes

Review URL: https://codereview.chromium.org/798723002

9 years agoRoll ANGLE
bsalomon [Fri, 12 Dec 2014 16:46:38 +0000 (08:46 -0800)]
Roll ANGLE

BUG=skia:3192

Review URL: https://codereview.chromium.org/788373002

9 years agoCull pushCull and popCull from Skia.
mtklein [Fri, 12 Dec 2014 16:46:25 +0000 (08:46 -0800)]
Cull pushCull and popCull from Skia.

These calls are unused and going away.  Waiting on crrev.com/796083002.

BUG=skia:

Review URL: https://codereview.chromium.org/794263002

9 years agoSkRecord: increase min block to 512B, remove max.
mtklein [Fri, 12 Dec 2014 16:41:23 +0000 (08:41 -0800)]
SkRecord: increase min block to 512B, remove max.

When we added the 64K allocation cap, the bots showed we took a perf hit
on some large .skps like desk_pokemonwiki.skp, despite not seeing a local
effect.  I'm still not seeing that locally, but I'd like to try removing the cap on
the bots to see what happens.  For big monolithic pictures, really packing into
memory tightly is probably not as important as it is for tiny ones.

Similarly, we're probably being too cautious about making tiny allocations.
Today we start at 16 bytes, which isn't really enough to record anything.
Even the smallest picture, say,
    save
        clipRect
        drawRect
    restore
requires ~200 bytes, so we might as well move our minimum block size up
near there.

I don't know if 16 bytes is too small to start for GrTextStrikes, so I've left the
behavior the same (though the max is still gone).

Local recording performance is neutral-to-positive:
             tabl_deviantart.skp  126us ->  129us 1.02x
                tabl_nytimes.skp  110us ->  112us 1.02x
           tabl_cuteoverload.skp  521us ->  530us 1.02x
             desk_mobilenews.skp  673us ->  682us 1.01x
             desk_chalkboard.skp  843us ->  854us 1.01x
                 desk_sfgate.skp  528us ->  535us 1.01x
            desk_silkfinance.skp 68.2us ->   69us 1.01x
                desk_youtube.skp  623us ->  629us 1.01x
                desk_blogger.skp  472us ->  475us 1.01x
     desk_jsfiddlehumperclip.skp 42.2us -> 42.5us 1.01x
                   desk_espn.skp  255us ->  256us 1.01x
                   desk_ebay.skp  174us ->  174us 1x
                desk_twitter.skp  454us ->  455us 1x
                 tabl_pravda.skp  200us ->  201us 1x
              desk_wordpress.skp  782us ->  784us 1x
               desk_samoasvg.skp  762us ->  761us 1x
                tabl_mozilla.skp 1.58ms -> 1.58ms 1x
               tabl_slashdot.skp  107us ->  107us 1x
               tabl_techmeme.skp  102us ->  102us 0.99x
            tabl_gamedeksiam.skp  729us ->  724us 0.99x
                 tabl_nofolo.skp 65.3us -> 64.7us 0.99x
            desk_gmailthread.skp  339us ->  336us 0.99x
                tabl_sahadan.skp   91us ->   90us 0.99x
           desk_yahooanswers.skp  144us ->  142us 0.99x
                   tabl_cnet.skp  143us ->  141us 0.99x
             tabl_googleblog.skp  206us ->  203us 0.99x
                    tabl_cnn.skp  160us ->  158us 0.99x
               tabl_frantzen.skp 50.5us -> 49.6us 0.98x
               desk_linkedin.skp  328us ->  323us 0.98x
                   tabl_digg.skp  790us ->  769us 0.97x
         desk_jsfiddlebigcar.skp 40.6us -> 39.5us 0.97x
                 desk_mapsvg.skp 1.57ms -> 1.52ms 0.97x
                  tabl_gmail.skp 19.4us -> 18.6us 0.96x
                   tabl_hsfi.skp 9.81us -> 9.11us 0.93x

BUG=skia:

Review URL: https://codereview.chromium.org/793033002

9 years agovalgrind static uninit fix
joshualitt [Fri, 12 Dec 2014 16:41:15 +0000 (08:41 -0800)]
valgrind static uninit fix

TBR=
BUG=skia:

Review URL: https://codereview.chromium.org/800873002

9 years agoAdd some debugging information to gyp_to_android.
scroggo [Fri, 12 Dec 2014 15:51:37 +0000 (07:51 -0800)]
Add some debugging information to gyp_to_android.

Temporary, to help diagnose skbug.com/3233

TBR=borenet
NOTRY=True
NOTREECHECKS=True
BUG=skia:3233

Review URL: https://codereview.chromium.org/802703002

9 years agoFix float and half float support on mobile.
jvanverth [Fri, 12 Dec 2014 13:58:06 +0000 (05:58 -0800)]
Fix float and half float support on mobile.

It's unclear what params should be used for half float alpha
rendertargets, so they are disabled for the moment.

Review URL: https://codereview.chromium.org/799593002

9 years agoFix GpuLayerCacheTest compile error
kkinnunen [Fri, 12 Dec 2014 13:29:12 +0000 (05:29 -0800)]
Fix GpuLayerCacheTest compile error

Fix GpuLayerCacheTest compile error:
GpuLayerCacheTest.cpp:120:34: error: comparison between signed and
unsigned integer expressions [-Werror=sign-compare]

gcc --version
gcc (Ubuntu/Linaro 4.6.4-1ubuntu1~12.04) 4.6.4

Review URL: https://codereview.chromium.org/794003003

9 years agouse SkScalar instead of float
reed [Fri, 12 Dec 2014 04:19:58 +0000 (20:19 -0800)]
use SkScalar instead of float

BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/796803005

9 years agoadding some flaky gms to ignore
joshualitt [Fri, 12 Dec 2014 02:44:22 +0000 (18:44 -0800)]
adding some flaky gms to ignore

TBR=
NOTRY=True
BUG=skia:

Review URL: https://codereview.chromium.org/789963005

9 years agorebase and remove ignores
joshualitt [Fri, 12 Dec 2014 01:26:44 +0000 (17:26 -0800)]
rebase and remove ignores

NOTRY=True
TBR=
BUG=skia:

Review URL: https://codereview.chromium.org/788193005

9 years agoRemove GP from drawstate, revision of invariant output for GP
joshualitt [Thu, 11 Dec 2014 23:44:02 +0000 (15:44 -0800)]
Remove GP from drawstate, revision of invariant output for GP

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/c6bc58eded89b0c0a36b8e20e193c200f297a0da

Review URL: https://codereview.chromium.org/791743003

9 years agoAdding a custom data field to GrGpuResource
junov [Thu, 11 Dec 2014 22:59:31 +0000 (14:59 -0800)]
Adding a custom data field to GrGpuResource

Motivation: allow Blink to track pre-existing mailbox names for
textures that are recycled by the ganesh scratch texture pool.

BUG=440462
TBR=bsalomon@google.com

Review URL: https://codereview.chromium.org/796163002

9 years agoUpdate SKP version
skia.buildbots [Thu, 11 Dec 2014 22:42:58 +0000 (14:42 -0800)]
Update SKP version

Automatic commit by the RecreateSKPs bot.

TBR=

Review URL: https://codereview.chromium.org/799653002

9 years agoone more ignore for hairlines
joshualitt [Thu, 11 Dec 2014 22:27:54 +0000 (14:27 -0800)]
one more ignore for hairlines

TBR=
NOTRY=True
BUG=skia:

Review URL: https://codereview.chromium.org/795293002

9 years agoFix some win64 warnings.
bsalomon [Thu, 11 Dec 2014 21:31:06 +0000 (13:31 -0800)]
Fix some win64 warnings.

Review URL: https://codereview.chromium.org/795223002

9 years agoIncrease timeout on sfgate pageset
rmistry [Thu, 11 Dec 2014 21:23:28 +0000 (13:23 -0800)]
Increase timeout on sfgate pageset

TBR=scroggo
NOTRY=true
BUG=skia:3196

Review URL: https://codereview.chromium.org/800563002

9 years agoCreate xfer processor backend.
egdaniel [Thu, 11 Dec 2014 21:15:13 +0000 (13:15 -0800)]
Create xfer processor backend.

This includes:
-Having an actual XP stage at the end of the gl pipeline.
-All Blending work is handled by XP until actually setting GL blend states
-GLPrograms test to test XP

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/4dffc940c430eec66d4707490eace19c9b3f7904

Review URL: https://codereview.chromium.org/764643004

9 years agoUpdate GM to permit correctness testing of MPD on the GMs
robertphillips [Thu, 11 Dec 2014 21:10:23 +0000 (13:10 -0800)]
Update GM to permit correctness testing of MPD on the GMs

Review URL: https://codereview.chromium.org/794273003

9 years agonew hairlines gm unstable on hd2000
joshualitt [Thu, 11 Dec 2014 21:10:05 +0000 (13:10 -0800)]
new hairlines gm unstable on hd2000

NOTRY=True
TBR=
BUG=skia:

Review URL: https://codereview.chromium.org/800543002

9 years agoAdd parameter to specify gyp dir to gyp_to_android.
scroggo [Thu, 11 Dec 2014 21:05:14 +0000 (13:05 -0800)]
Add parameter to specify gyp dir to gyp_to_android.

NOTRY=true
NOTREECHECKS=True
BUG=skia:3233

Review URL: https://codereview.chromium.org/800573002

9 years agoEnforce thread-safety of bitmaps in pictures via the type.
mtklein [Thu, 11 Dec 2014 20:43:04 +0000 (12:43 -0800)]
Enforce thread-safety of bitmaps in pictures via the type.

No runtime difference here, but it makes it impossible to forget to make
a shallow copy; you can't get at the full bitmap without it.

NOTREECHECKS=true

BUG=skia:

Review URL: https://codereview.chromium.org/799603002

9 years agoForce embedding full font when serializing pictures.
mtklein [Thu, 11 Dec 2014 19:06:00 +0000 (11:06 -0800)]
Force embedding full font when serializing pictures.

We can't do this unconditionally or pipe will become stupidly slow.

DM's serialize mode fails subtly on Mac when we force embedding, so I've
#ifdef'd that away.  Other platforms look fine.

BUG=skia:

Review URL: https://codereview.chromium.org/796523002

9 years agoReplace EncodeBitmap with an interface.
scroggo [Thu, 11 Dec 2014 18:53:58 +0000 (10:53 -0800)]
Replace EncodeBitmap with an interface.

Gives more flexibility to the caller to decide whether to use the
encoded data returned by refEncodedData().

Provides an implementation that supports the old version of
SkPicture::serialize().

TODO: Update Chrome, so we can remove SK_LEGACY_ENCODE_BITMAP entirely

BUG=skia:3190

Review URL: https://codereview.chromium.org/784643002

9 years agoMaking GrGpuResource::release public for special case in Blink
junov [Thu, 11 Dec 2014 18:32:32 +0000 (10:32 -0800)]
Making GrGpuResource::release public for special case in Blink

BUG=440462

Review URL: https://codereview.chromium.org/739383006

9 years agoRemove action_runner steps for worldjournal pageset to prevent crashes
rmistry [Thu, 11 Dec 2014 18:24:27 +0000 (10:24 -0800)]
Remove action_runner steps for worldjournal pageset to prevent crashes

BUG=skia:3196
NOTRY=true

Review URL: https://codereview.chromium.org/795173002

9 years agohmm one more hairlines 'straggler' might be flipflopping
joshualitt [Thu, 11 Dec 2014 18:02:30 +0000 (10:02 -0800)]
hmm one more hairlines 'straggler' might be flipflopping

TBR=
NOTRY=
BUG=skia:

Review URL: https://codereview.chromium.org/798533005

9 years agostraggler hairline
joshualitt [Thu, 11 Dec 2014 16:58:37 +0000 (08:58 -0800)]
straggler hairline

NOTRY=True
TBR=
BUG=skia:

Review URL: https://codereview.chromium.org/796893002

9 years agoPoppler rebaseline for clipped-bitmap-shaders-clamp
fmalita [Thu, 11 Dec 2014 16:20:38 +0000 (08:20 -0800)]
Poppler rebaseline for clipped-bitmap-shaders-clamp

Follow-up for https://codereview.chromium.org/787043006.

TBR=scroggo@google.com

Review URL: https://codereview.chromium.org/798533003

9 years agoApply the layer's image filter to the hoisted image
robertphillips [Thu, 11 Dec 2014 16:20:31 +0000 (08:20 -0800)]
Apply the layer's image filter to the hoisted image

Exposing SkSurface_Gpu makes me sad and I would welcome alternatives.

This change is desireable since it greatly decreases the render target swaps.

Review URL: https://codereview.chromium.org/792923002

9 years agorebase cpu hairlines
joshualitt [Thu, 11 Dec 2014 15:37:48 +0000 (07:37 -0800)]
rebase cpu hairlines

NOTRY=True
NOTREECHECKS=True
TBR=
BUG=skia:

Review URL: https://codereview.chromium.org/793393002

9 years agoupdate test after I added a clip call, making the count odd
reed [Thu, 11 Dec 2014 15:30:58 +0000 (07:30 -0800)]
update test after I added a clip call, making the count odd

BUG=skia:
TBR=
NOTREECHECKS=True
NOTRY=True

Review URL: https://codereview.chromium.org/797633002

9 years agoDefer saves() until they're needed
reed [Thu, 11 Dec 2014 15:07:37 +0000 (07:07 -0800)]
Defer saves() until they're needed

patch from issue 759443006 at patchset 40001 (http://crrev.com/759443006#ps40001)

BUG=skia:

Review URL: https://codereview.chromium.org/767333002

9 years agorebase hairlines
joshualitt [Thu, 11 Dec 2014 14:53:34 +0000 (06:53 -0800)]
rebase hairlines

TBR=
BUG=skia:

Review URL: https://codereview.chromium.org/795133002

9 years agoReally fix valgrind issue
joshualitt [Thu, 11 Dec 2014 14:52:01 +0000 (06:52 -0800)]
Really fix valgrind issue

TBR=
BUG=skia:

Review URL: https://codereview.chromium.org/795123002

9 years agoMore rebaselines.
scroggo [Thu, 11 Dec 2014 14:22:32 +0000 (06:22 -0800)]
More rebaselines.

TBR=robertphillips
BUG=skia:

Review URL: https://codereview.chromium.org/793353002

9 years agoMark nexus10 images as flaky.
scroggo [Thu, 11 Dec 2014 14:21:50 +0000 (06:21 -0800)]
Mark nexus10 images as flaky.

TBR=robertphillips
BUG=skia:3226

Review URL: https://codereview.chromium.org/785323003

9 years agoAdditional rebaselines for 3490263287e9.
bungeman [Thu, 11 Dec 2014 08:37:37 +0000 (03:37 -0500)]
Additional rebaselines for 3490263287e9.

TBR=bungeman1

Review URL: https://codereview.chromium.org/794863004

9 years agoReplace use of deprecated CG methods.
bungeman [Thu, 11 Dec 2014 05:43:27 +0000 (21:43 -0800)]
Replace use of deprecated CG methods.

OSX10.9 and iOS7.0 deprecated CGContextShowGlyphsAtPoint so a new API
should be used. OSX10.7 and iOS4.2 replace CGContextShowGlyphsAtPoint with
CTFontDrawGlyphs. OSX10.5 and iOS2.0 have CGContextShowGlyphsAtPositions
which works similarly to CTFontDrawGlyphs and has not yet been deprecated.
This change allows the use of CTFontDrawGlyphs when it is available,
falling back to CGContextShowGlyphsAtPositions when it isn't.

Review URL: https://codereview.chromium.org/770383002

9 years agoRevert of Create xfer processor backend. (patchset #6 id:100001 of https://codereview...
joshualitt [Wed, 10 Dec 2014 23:03:00 +0000 (15:03 -0800)]
Revert of Create xfer processor backend. (patchset #6 id:100001 of https://codereview.chromium.org/764643004/)

Reason for revert:
CL breaks windows DM

Original issue's description:
> Create xfer processor backend.
>
> This includes:
> -Having an actual XP stage at the end of the gl pipeline.
> -All Blending work is handled by XP until actually setting GL blend states
> -GLPrograms test to test XP
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/4dffc940c430eec66d4707490eace19c9b3f7904

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

Review URL: https://codereview.chromium.org/789343002

9 years agoRevert of Add more ignore gm's for recent Xfer Processor CL (patchset #1 id:1 of...
joshualitt [Wed, 10 Dec 2014 22:57:56 +0000 (14:57 -0800)]
Revert of Add more ignore gm's for recent Xfer Processor CL (patchset #1 id:1 of https://codereview.chromium.org/786083004/)

Reason for revert:
Preventing revert of cl breaking windows DM

Original issue's description:
> Add more ignore gm's for recent Xfer Processor CL
>
> TBR=bsalomon@google.com
> NOTRY=true
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/125251133a62a07225b7119ab3ac90153ba4c58a

TBR=bsalomon@google.com,egdaniel@google.com
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/794883002

9 years agoRevert of Fix to set correct output type when blending when we've read dst (patchset...
joshualitt [Wed, 10 Dec 2014 22:57:50 +0000 (14:57 -0800)]
Revert of Fix to set correct output type when blending when we've read dst (patchset #4 id:60001 of https://codereview.chromium.org/791143002/)

Reason for revert:
Breaks DM on windows

Original issue's description:
> Fix to set correct output type when blending when we've read dst
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/733fc2054044377e914559fa4f46db66ab7d9cea

TBR=bsalomon@google.com,egdaniel@google.com
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/795783002

9 years agofix for chrome bug with sub pixel hairlines
joshualitt [Wed, 10 Dec 2014 22:23:40 +0000 (14:23 -0800)]
fix for chrome bug with sub pixel hairlines

BUG=skia:

Review URL: https://codereview.chromium.org/796573002

9 years agoInitial baselines for tall_stretched_bitmaps GM
Brian Salomon [Wed, 10 Dec 2014 22:20:15 +0000 (17:20 -0500)]
Initial baselines for tall_stretched_bitmaps GM

BUG=skia:

Review URL: https://codereview.chromium.org/789293003

9 years agoRevert of Remove GP from drawstate, revision of invariant output for GP (patchset...
joshualitt [Wed, 10 Dec 2014 22:12:22 +0000 (14:12 -0800)]
Revert of Remove GP from drawstate, revision of invariant output for GP (patchset #9 id:160001 of https://codereview.chromium.org/791743003/)

Reason for revert:
breaks mac

Original issue's description:
> Remove GP from drawstate, revision of invariant output for GP
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/c6bc58eded89b0c0a36b8e20e193c200f297a0da

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

Review URL: https://codereview.chromium.org/794843002

9 years agoRemove GP from drawstate, revision of invariant output for GP
joshualitt [Wed, 10 Dec 2014 21:48:57 +0000 (13:48 -0800)]
Remove GP from drawstate, revision of invariant output for GP

BUG=skia:

Review URL: https://codereview.chromium.org/791743003

9 years agoDon't store an SkMatrix in BitmapShaderKey.
fmalita [Wed, 10 Dec 2014 21:01:43 +0000 (13:01 -0800)]
Don't store an SkMatrix in BitmapShaderKey.

SkMatrix makes for a poor key component due to mutable/cache fields.

Use its canonical scalars instead.

R=reed@google.com,mtklein@google.com

Review URL: https://codereview.chromium.org/792123002

9 years agoFix to set correct output type when blending when we've read dst
egdaniel [Wed, 10 Dec 2014 20:45:20 +0000 (12:45 -0800)]
Fix to set correct output type when blending when we've read dst

BUG=skia:

Review URL: https://codereview.chromium.org/791143002

9 years agofix for valgrind uninit variables
joshualitt [Wed, 10 Dec 2014 20:28:08 +0000 (12:28 -0800)]
fix for valgrind uninit variables

TBR=bsalomon@google.com
BUG=skia:

Review URL: https://codereview.chromium.org/793773002

9 years agoRebaseline imagefiltersscaled images (and rm from ignored-tests file)
robertphillips [Wed, 10 Dec 2014 20:23:09 +0000 (12:23 -0800)]
Rebaseline imagefiltersscaled images (and rm from ignored-tests file)

Rebaselines for https://codereview.chromium.org/785893004/ (Fix MatrixImageFilter computeFastBounds)

TBR=scroggo@google.com

Review URL: https://codereview.chromium.org/794673004

9 years agoMore rebaselines.
scroggo [Wed, 10 Dec 2014 20:23:03 +0000 (12:23 -0800)]
More rebaselines.

TBR=bsalomon
BUG=skia:

Review URL: https://codereview.chromium.org/790263002

9 years agoRemove MULTILINE, ^, and $ from re finding actual_results.json.
bungeman [Wed, 10 Dec 2014 20:18:03 +0000 (12:18 -0800)]
Remove MULTILINE, ^, and $ from re finding actual_results.json.

MULTILINE, ^, and $ work based on the platform, but the files being
parsed are from other platforms. As a result, the current code for
extracting the actual_results.json version will not find it in
logfiles produced on Windows when run on Mac.

The code for extracting the exact actual_results.json file to use
from a logfile is itself something of a hack, as this information
should be provided ina more structured manner. This proposed
method of finding the exact file is no worse than the old one, and
in cases like above, better.

Review URL: https://codereview.chromium.org/789253002

9 years agoDetect discarded SkPictureShader pixel refs.
fmalita [Wed, 10 Dec 2014 20:17:58 +0000 (12:17 -0800)]
Detect discarded SkPictureShader pixel refs.

This doesn't address the more general problem, but reduces the race
window significantly.

BUG=440282
R=reed@google.com

Review URL: https://codereview.chromium.org/787043006

9 years agoRebaseline.
scroggo [Wed, 10 Dec 2014 18:53:28 +0000 (10:53 -0800)]
Rebaseline.

All these images barely differ.

TBR=bsalomon

BUG=skia:

Review URL: https://codereview.chromium.org/795673002

9 years agoInclude SkTypes.h from SkDynamicAnnotations.h
mtklein [Wed, 10 Dec 2014 18:49:42 +0000 (10:49 -0800)]
Include SkTypes.h from SkDynamicAnnotations.h

This allows us to control SK_DYNAMIC_ANNOTATIONS_ENABLED by Sk*Config.h files.
This is a no-op today, because we control it from the compiler command line.

BUG=430815

Review URL: https://codereview.chromium.org/787003003

9 years agoGradient shaders: make fPtsToUnit const, pre-cache getType().
mtklein [Wed, 10 Dec 2014 18:29:19 +0000 (10:29 -0800)]
Gradient shaders: make fPtsToUnit const, pre-cache getType().

This prevents races when calling fPtsToUnit.getType() from multiple threads.

This introduces a small amount of redundant code in SkTwoPointRadialGradient,
but it's probably optimized together into no extra run-time work.

BUG=437511

Review URL: https://codereview.chromium.org/793763003

9 years agoChange how SkDebugf is sent to stdout on Android.
scroggo [Wed, 10 Dec 2014 18:23:04 +0000 (10:23 -0800)]
Change how SkDebugf is sent to stdout on Android.

Previously, a function was called using dlsym in skia_launcher.

Add a static initializer that changes the setting, and include that for
the tools we automate for testing.

Also only do va_copy if we actually use it.

BUG=skia:2454

Review URL: https://codereview.chromium.org/753543003

9 years agodd readPixels to SkImage
reed [Wed, 10 Dec 2014 17:53:42 +0000 (09:53 -0800)]
dd readPixels to SkImage

patch from issue 789673007 at patchset 1 (http://crrev.com/789673007#ps1)

BUG=skia:

Review URL: https://codereview.chromium.org/793723002