GitLab

It's common to add --mash to the /etc/chrome_dev.conf when working on a
Chrome OS device. However, that can lead to two copies of --mash if you run
an autotest on the device that also passes --mash. This makes Chrome crash.

Ignore duplicate --mash flags.

BUG=669705
TEST=Run chrome --mash --mash, doesn't crash.

Review-Url: https://codereview.chromium.org/2545633002
Cr-Commit-Position: refs/heads/master@{#435655}

BUG=582315

Review-Url: https://codereview.chromium.org/2256913002
Cr-Commit-Position: refs/heads/master@{#435654}

Some useful subresources could be fetched by webpages via XMLHttpRequest. This
CL adds support of this type of requests to ResourcePrefetchPredictor.

BUG=631966

Review-Url: https://codereview.chromium.org/2539263004
Cr-Commit-Position: refs/heads/master@{#435653}

Skip DevToolsPixelOutputTests.TestLatencyInfoInstrumentation in site_per_process_browser_tests due to flakiness.

Doesn't seem similarly flaky on browser_tests, but further investigation is needed.

BUG=670362
TBR=lukasza@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2538923004
Cr-Commit-Position: refs/heads/master@{#435652}

Due to third-party modifications, some devices access the disk in
Context#bindService(). This call is supposed to be safe to make from
the UI thread, so suppress the StrictMode violation.

BUG=670195

Review-Url: https://codereview.chromium.org/2549473002
Cr-Commit-Position: refs/heads/master@{#435651}

BUG=angleproject:1442
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2542653002
Cr-Commit-Position: refs/heads/master@{#435650}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/852f85163d54..da3aa694af97

$ git log 852f85163..da3aa694a --date=short --no-merges --format='%ad %ae %s'
2016-12-01 simonhatch Dashboard - Also check for "staging_win" when modifying non-telemetry commands.

BUG=668529

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=catapult-sheriff@chromium.org

Review-Url: https://codereview.chromium.org/2546653004
Cr-Commit-Position: refs/heads/master@{#435649}

You can now run an alternative window manager from mash_session, mash_init etc by passing --window-manager=<name_of_wm>

R=sky@chromium.org

Review-Url: https://codereview.chromium.org/2511233002
Cr-Commit-Position: refs/heads/master@{#435648}

Remove WebContentsObserver from CastMediaBlocker.

BUG=660331

Review-Url: https://codereview.chromium.org/2535343003
Cr-Commit-Position: refs/heads/master@{#435647}

This reverts commit fe899c8e.

BUG=669797
TBR=thestig@chromium.org

Review-Url: https://codereview.chromium.org/2541843005
Cr-Commit-Position: refs/heads/master@{#435646}

Revert of Roll src/third_party/skia/ 71b762f2a..1c038af27 (3 commits). (patchset #1 id:1 of https://codereview.chromium.org/2542133002/ )

Reason for revert:
speculative revert to fix
https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Mac10.10/builds/27239

Original issue's description:
> Roll src/third_party/skia/ 71b762f2a..1c038af27 (3 commits).
>
> https://skia.googlesource.com/skia.git/+log/71b762f2aca1..1c038af278a6
>
> $ git log 71b762f2a..1c038af27 --date=short --no-merges --format='%ad %ae %s'
> 2016-12-01 reed Revert "Revert "remove (empty) SkXfermode.h""
> 2016-12-01 halcanary No SkEncodeImageAsKTX for ANDROID_FRAMEWORK
> 2016-11-30 msarett SkColorSpaceXform bug fixes
>
> Documentation for the AutoRoller is here:
> https://skia.googlesource.com/buildbot/+/master/autoroll/README.md
>
> If the roll is causing failures, see:
> http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls
>
> CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel
> TBR=brianosman@google.com
>
> Committed: https://crrev.com/749c5466102850a151448c59d99be6cffffc0bf5
> Cr-Commit-Position: refs/heads/master@{#435626}

TBR=brianosman@google.com,skia-deps-roller@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=670347

Review-Url: https://codereview.chromium.org/2537303006
Cr-Commit-Position: refs/heads/master@{#435645}

"Just ignore popstate events. I got this." - route.js

R=tommycli@chromium.org
BUG=668004
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2538393002
Cr-Commit-Position: refs/heads/master@{#435644}

A recent metrics fix (r432864) also mistakenly broke this code by not
passing in a default argument. This CL fixes that problem and adds an
expectation in unit tests.

BUG=662999,596676

Review-Url: https://codereview.chromium.org/2544583003
Cr-Commit-Position: refs/heads/master@{#435643}

https://pdfium.googlesource.com/pdfium.git/+log/c62980ec7539..02759102cf99

$ git log c62980ec7..02759102c --date=short --no-merges --format='%ad %ae %s'
2016-12-01 tsepez Protect against unexpected library create/destroy

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

TBR=dsinclair@chromium.org

Review-Url: https://codereview.chromium.org/2542803003
Cr-Commit-Position: refs/heads/master@{#435642}

Add ArcAppDialogView for arc app uninstall confirmation. Add Test coverage.

BUG=661076
Test=Add browsertest. Tests passed.
Test=Manual Test.

Review-Url: https://codereview.chromium.org/2529783002
Cr-Commit-Position: refs/heads/master@{#435641}

CastGinRunner is now owned by the gin::PerContextData associated
with a content::RenderFrame. Do not rely on RenderFrameObserver
notifications to manager it's lifetime. Instead, this class will
be deleted when gin::PerContextData is destroyed.

BUG=internal b/32916396

R=slan@chromium.org,alokp@chromium.org

Review-Url: https://codereview.chromium.org/2535313004
Cr-Commit-Position: refs/heads/master@{#435640}

Chrome stores those information in SharedPreferences for all downloads now.
So there is no need to keep those extra fields.

BUG=669702

Review-Url: https://codereview.chromium.org/2544633003
Cr-Commit-Position: refs/heads/master@{#435639}

It used to be exiting the move loop and clearing the move_loop_widget_ before
EndDrag() had a chance to set the window position to manageable again.

This CL adds a maximized browser window waiter, and moves setting the window
position managed logic to ash.

BUG=626761, 331924
TEST=interactive_ui_tests --gtest_filter=TabDragging*

Review-Url: https://codereview.chromium.org/2494713003
Cr-Commit-Position: refs/heads/master@{#435638}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/403eccb0..228237cf

Please follow these instructions for assigning/CC'ing issues:
https://github.com/v8/v8/wiki/Triaging%20issues

Please close rolling in case of a roll revert:
https://v8-roll.appspot.com/
This only works with a Google account.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

TBR=hablich@chromium.org,machenbach@chromium.org,littledan@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/2544003002
Cr-Commit-Position: refs/heads/master@{#435637}

Just a refactor, no changed behavior or new tests.

This is a follow-up on crrev.com/2468233002 (Pull up
clearCacheExceptFrame impl. to ImageDecoder), after a behavior change
for WebP's cache clearing has landed in crrev.com/2499633002 (Always
preserve provided frame in WebP::clearCacheExceptFrame).

Pull up the implementation of clearCacheExceptFrame in WEBPImageDecoder and
GIFImageDecoder to ImageDecoder. This increases code reuse, especially since
PNGImageDecoder can use it for the Animated PNG implementation as well. Child
classes of ImageDecoder only need to override clearFrameBuffer() and / or
frameStatusIsSufficientForSuccessors(), if necessary.

The baseline idea of the implementation is that the provided frame is
always preserved, and that there are two cases in which a previous
frame is also preserved:

1. The provided frame is not yet sufficiently decoded to decode subsequent
frames, as indicated by frameStatusIsSufficientForSuccessors(). We need
to keep the required previous frame to sufficiently decode the provided
frame first.
2. The disposal method of the provided frame is DisposeOverwritePrevious. In
this case, we need to keep the required previous frame in cache to
prevent re-decoding that frame when the provided frame is disposed.

WebP does not know DisposeOverwritePrevious, so the new implementation
is effectively equivalent to the original method, because
frameStatusIsSufficientForSuccessors returns true iff the status is
FrameComplete.

Gif's original implementation is somewhat different, in the sense that
the roles of |clearExceptFrame| and |clearExceptFrame2| are
interchanged. Now, |clearExpectFrame| is always kept and
|clearExceptFrame2| contains any required previous frame. The resulting
behavior is not changed, though, given the following case analysis:

- The provided frame is FrameEmpty. The old implementation would only
  keep the nearest non-empty frame in cache, since the first block is a
  no-op due to the != FrameEmpty condition, and the second block would
  search for the nearest non-empty frame in cache.

  The new patch sets |clearExceptFrame2| to the required previous frame,
  since !frameStatusIsSufficientForSuccessors evaluates to true. The
  second block will find the nearest non-empty frame in cache.
  |clearExceptFrame| is also passed to clearCacheExceptTwoFrames, but
  is ignored since that method checks for FrameEmpty frames.

- The provided frame is FrameComplete or FramePartial, and the disposal
  method is DisposeOverwritePrevious. In the old implementation this
  results in |clearExceptFrame| being set to the required previous
  frame in the first block, and |clearExceptFrame2| is set to the
  provided frame. The second block searches for the nearest non-empty
  frame, starting at |clearExceptFrame|.

  The new implementation stores the required previous frame in
  |clearExceptFrame2| because of the disposal method, and similarly
  searches for the nearest non-empty frame in the second block.

- The provided frame is FrameComplete or FramePartial, and the disposal
  method is *not* DisposeOverwritePrevious. In this case, the old
  implementation only preserves the provided frame since the first block
  is a no-op due to the disposal condition, and the second block is a
  no-op as well since the frames' status is not FrameEmpty.

  The new implementation does not set |clearExceptFrame2| because of the
  disposal condition in the first block, and thus only preserves the
  provided frame.

Review-Url: https://codereview.chromium.org/2516593003
Cr-Commit-Position: refs/heads/master@{#435636}

Rationale:
Current safe browsing code lives in the chrome/ layer, but we
want to make it available in WebView as well (at least the
relevant parts of it). Hence this effort of making a safe
browsing component.

In this patch:
- move IPCs (spellcheck_messages) + add dedicated generator
- isolate and move safe browsing specific constants
- isolate and move safe browsing specific switches
- remove a few unused includes

Note:
To reduce the number of components and keep everything
related in one place the idea is to move the code currently
living in components/safe_browsing_db/ to this safe_browsing/
component in the future.

BUG=488675
TBR=brettw@chromium.org

Review-Url: https://codereview.chromium.org/2450993003
Cr-Commit-Position: refs/heads/master@{#435635}

We have already supported globalCompositeOperation, but globalAlpha is
not exposed yet.

This CL expose this attributes. Similar to the globalCompositeOperation,
the implementation is already there, we just need to expose it in the
.idl file.

The layout test is copied from fast/canvas/2d.composite.globalAlpha.fillPath.html,
with very small tweak.

BUG=659260

Review-Url: https://codereview.chromium.org/2526693003
Cr-Commit-Position: refs/heads/master@{#435634}

1. Created three proguard files:
1.1. cronet_impl_common_proguard.cfg - for apps that depend on cronet_impl_common_java.jar
1.2. cronet_impl_platform_proguard.cfg - for apps that depend on cronet_impl_platform_java.jar
1.3. cronet_impl_native_proguard.cfg - for apps that depend on cronet_impl_native_java.jar

2. Enabled proguarding of cronet_sample_apk even for debug builds.

3. Changed cronet_sample_apk & cronet_perf_test_apk to test with proguard
   configs that we ship.

BUG=665446
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_cronet_tester

Review-Url: https://codereview.chromium.org/2540773003
Cr-Commit-Position: refs/heads/master@{#435633}

Dalvik JIT generated code doesn't always align the stack to a 16 byte
boundary when calling into native, causing crashes in code that expects
16 byte alignment. Force the compiler to realign the stack when entering
native from Java, so that other code can assume 16 byte alignment as
expected by the ABI.

Move the function attributes into a macro so that the generated header
file is less repetitive (this also makes the generator less repetitive
as a bonus).

Also, to stop presubmit complaining about golden_sample_for_tests_jni.h
not being correctly clang-formatted, rename it to .golden like the other
test files.

BUG=655248

Review-Url: https://codereview.chromium.org/2531273002
Cr-Commit-Position: refs/heads/master@{#435632}

This state is not set in the service side at glEnable/glDisable time.
Instead, it is set during framebuffer validation, to accommodate
driver bug workarounds on desktop.

Move the logic for updating framebuffer SRGB in the decoder to
a separate function, GLES2DecoderImpl::UpdateFramebufferSRGB.

BUG=655247
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2519603002
Cr-Commit-Position: refs/heads/master@{#435631}

BUG=520706
R=erikchen@chromium.org

Review-Url: https://codereview.chromium.org/2545823002
Cr-Commit-Position: refs/heads/master@{#435630}

Requires host machines to have jdk8 installed and uses retrolambda to
convert java 8 .jar files into java 7 .jar files.

Future work includes adding Retrolambda into third_party libraries.

BUG=642600

Review-Url: https://codereview.chromium.org/2543463002
Cr-Commit-Position: refs/heads/master@{#435629}

Mark fast/speech/scripted/speechrecognition-restart-onend.html as flakily leaking an ActiveDOMObject.

BUG=670324
TBR=sigbjornf@opera.com
NOTRY=true

Review-Url: https://codereview.chromium.org/2545763003
Cr-Commit-Position: refs/heads/master@{#435628}

Mark virtual/mojo-loading/http/tests/xmlhttprequest/workers/xmlhttprequest-response-type-blob-sync.html as flaky timeout.

https://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=virtual%2Fmojo-loading%2Fhttp%2Ftests%2Fxmlhttprequest%2Fworkers%2Fxmlhttprequest-response-type-blob-sync.html

BUG=659917
TBR=yhirano@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2542923002
Cr-Commit-Position: refs/heads/master@{#435627}

https://skia.googlesource.com/skia.git/+log/71b762f2aca1..1c038af278a6

$ git log 71b762f2a..1c038af27 --date=short --no-merges --format='%ad %ae %s'
2016-12-01 reed Revert "Revert "remove (empty) SkXfermode.h""
2016-12-01 halcanary No SkEncodeImageAsKTX for ANDROID_FRAMEWORK
2016-11-30 msarett SkColorSpaceXform bug fixes

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel
TBR=brianosman@google.com

Review-Url: https://codereview.chromium.org/2542133002
Cr-Commit-Position: refs/heads/master@{#435626}

BUG=626342

Review-Url: https://codereview.chromium.org/2539213002
Cr-Commit-Position: refs/heads/master@{#435625}

Seems to have been erroneously added as part of "Enable Analytic AA Fix":
https://codereview.chromium.org/2538943002

BUG=668907
TBR=liyuqian@google.com,fmalita@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2534313005
Cr-Commit-Position: refs/heads/master@{#435624}

Changes: https://chromium.googlesource.com/external/webrtc/trunk/webrtc.git/+log/61e42da..42ad2d4

$ git log 61e42da..42ad2d4 --date=short --no-merges --format=%ad %ae %s
2016-12-01 sprang@webrtc.org Wire up RTCP XR target bitrate in rtp/rtcp module
2016-12-01 kthelgason@webrtc.org Remove limit on how often quality scaling downscales
2016-12-01 kthelgason@webrtc.org Increase test timeout to combat flakiness.
2016-12-01 mflodman@webrtc.org Remove OnLocalSsrcChanged
2016-12-01 magjed@webrtc.org Move webrtc/api/androidtests to webrtc/sdk/android/instrumentationtests
2016-12-01 kthelgason@webrtc.org Add method on AVFoundation capturer to adapt output format.
2016-12-01 magjed@webrtc.org Move VideoDecoder::Create() logic to separate internal video decoder factory
2016-12-01 brandtr@webrtc.org Rename full_stack.cc to full_stack_tests.cc.
2016-12-01 kthelgason@webrtc.org Enable VideoToolbox encoder on mac
2016-11-30 skvlad@webrtc.org Make SurfaceTextureHelper and I420Frame public in Java.
2016-11-30 terelius@webrtc.org Change assert to RTC_DCHECK in bwe_test_logging.cc
2016-11-30 sakal@webrtc.org Allow custom drawers to be added to framelisteners.
2016-11-30 henrik.lundin@webrtc.org Prep to remove API-related #defines from voice_engine_configurations.h
2016-11-30 michaelt@webrtc.org Reland "Update rtt on audio only calls".
2016-11-30 brandtr@webrtc.org Start gathering perf data for VP8 + FlexFEC.
2016-11-30 nisse@webrtc.org Delete all of the video_processing module but the denoiser code.
2016-11-30 stefan@webrtc.org Wire up x-google-{min,start,max}-bitrate to WebRtcVoiceMediaChannel.
2016-11-30 kthelgason@webrtc.org Delete nalu parser in mediaencoder
2016-11-30 terelius@webrtc.org Templatize percentile_filter.h and move it to base/analytics.
2016-11-30 minyue@webrtc.org Reland "Renaming AudioEncoder::SetTargetBitrate and SetProjectedPacketLossRate."
2016-11-30 ehmaldonado@webrtc.org Replace test_support_main by test_main and get rid of test_support_main_threaded_mac
2016-11-30 kthelgason@webrtc.org Remove warning suppression from VideoToolboxEncoder
2016-11-30 asapersson@webrtc.org Change unit of logged bitrate stats in bytes/s to bits/s.
2016-11-30 phoglund@webrtc.org Greatly reduce number of level controller tests.
2016-11-30 perkj@webrtc.org Make sure GetLastError on a PlatformThread return an error that is relevant to the thread.
2016-11-30 magjed@webrtc.org RtpDataEngine, FindCodecByName: Don't reassign codecs
2016-11-30 magjed@webrtc.org Create webrtc/sdk/android folder
2016-11-30 minyue@webrtc.org Relanding "Pass time constant to bwe smoothing filter."
2016-11-30 ossu@webrtc.org Disabled flaky P2PTestConductor tests on ASAN and MSAN.

TBR=
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_chromium_archive_rel_ng;master.tryserver.chromium.mac:mac_chromium_archive_rel_ng
BUG=

Review-Url: https://codereview.chromium.org/2545763002
Cr-Commit-Position: refs/heads/master@{#435623}

This reverts commit 468ed78e.
https://codereview.chromium.org/2540763002/

Failing on bots.

NotificationImageLoaderTest.SuccessTest (run #1):
[ RUN      ] NotificationImageLoaderTest.SuccessTest
../../base/test/histogram_tester.cc:158: Failure
Value of: actual_count
  Actual: 1
Expected: expected_count
Which is: 0
Histogram "Notifications.LoadFinishTime.Icon" does not have the right number of samples (0) in the expected bucket (0). It has (1).
[  FAILED  ] NotificationImageLoaderTest.SuccessTest (5 ms)

https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20Trusty/builds/20543

BUG=669621,614456
TBR=johnme@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2540423003
Cr-Commit-Position: refs/heads/master@{#435622}

These methods are not used anymore, remove them.

BUG=664924

Review-Url: https://codereview.chromium.org/2541683003
Cr-Commit-Position: refs/heads/master@{#435621}

The IDL for proposed API has been reviewed: https://codereview.chromium.org/2174423002/

This is the actual implementation of the API. The dialog to request PIN is defined in request_pin_view.h

The CertificateProviderApi is receiving the requests from extensions and uses the service object to manage the view. CertificateProviderService object keeps the state of the view and forwards updates to it. And the RequestPinView is notifying the CertificateProviderApi about the user input through callback.

BUG=612886

Review-Url: https://codereview.chromium.org/2094333002
Cr-Commit-Position: refs/heads/master@{#435620}

BUG=542537
TBR=meade@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2541223003
Cr-Commit-Position: refs/heads/master@{#435619}

SchemeRegistry.cpp has bunch of distinct static URLSchemesSet's and they
are getting a bit out of control. This patch:

* Introduces URLSchemesRegisry that bundles all the URLSchemesSet's so
  that they can be initialized at once
* Provides mutable and non-mutable version of getters for the
  URLSchemesRegistry instance so that one can safely call non-mutable
  getters without worrying about thread-safety, while we have  the
  thread assertion in the mutable version of getter.

BUG=n/a

Review-Url: https://codereview.chromium.org/2545443002
Cr-Commit-Position: refs/heads/master@{#435618}

It crashes due to a WeakPtr SequenceChecker failure.

BUG=670307
TBR=ikilpatrick@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2549533002
Cr-Commit-Position: refs/heads/master@{#435617}

This patch implements detection for the loading RAIL mode. This mode
starts when the top-level frame navigates and lasts until we see the
first meaningful paint for the top-level frame (which is currently not
guaranteed to happen on all pages).

We don't currently use this state for scheduling but only pass the
information to v8 which can use it to adjust garbage collection
policies.

As a safeguard against cases where FMP is not detected, we terminate the
load stage if we see user input. The assumption is that if the user is interacting
with the page, there is likely meaningful content on the screen at that point.

BUG=613520
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2397753006
Cr-Commit-Position: refs/heads/master@{#435616}