GitLab

Method evaluateOnCallFrame was moved to V8DebuggerAgentImpl.
While evaluatiom on call frame for command line API support, method based on V8WindowCustom binding is used instead existing scope_extension argument. In consequent callFrame was removed from CommandLineLineAPI constructor.
Introduced InjectScopeExtensionByName helper class.

BUG=595206
R=dgozman@chromium.org
TBR=pfeldman@chromium.org

Review URL: https://codereview.chromium.org/1811853002

Cr-Commit-Position: refs/heads/master@{#381834}

This makes it possible for a client to prevent surface contents
from appearing in screenshots or from being visible on
non-secure outputs.

This also removes some oom checks from server.cc that should have
been removed by: https://codereview.chromium.org/1717773002

BUG=549781

Review URL: https://codereview.chromium.org/1802993003

Cr-Commit-Position: refs/heads/master@{#381833}

The resampler can partially consume audio samples (since when resampling,
one output sample can correspond to fractional input samples). By keeping
track of the partially-consumed samples, we can improve the accuracy of
the estimated rendering delay.

BUG=

Review URL: https://codereview.chromium.org/1809203003

Cr-Commit-Position: refs/heads/master@{#381832}

R=dpapad@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/1801293005

Cr-Commit-Position: refs/heads/master@{#381831}

The comment claims it only fails on 10.6, which we don't test anymore.

R=kbr@chromium.org, piman@chromium.org

Review URL: https://codereview.chromium.org/1813883003

Cr-Commit-Position: refs/heads/master@{#381830}

Previously, GN builds would not properly include the Chromium version
build and patch numbers in the Chrome-Proxy header sent to the Data
Reduction Proxy. This CL fixes this, and adds assertions and tests to
ensure that the build and patch numbers are always present.

BUG=595471

Review URL: https://codereview.chromium.org/1808333002

Cr-Commit-Position: refs/heads/master@{#381829}

Autofill wallet metadata notification identifier should be
"WALLET_METADATA" instead of "AUTOFILL_WALLET_METADATA" on the client
side to match the server side sync. The server uses the name of the
protobuf WalletMetadataSpecifics to derive the identifier. This affects
only wallet metadata update notifications.

BUG=593546

Review URL: https://codereview.chromium.org/1806853006

Cr-Commit-Position: refs/heads/master@{#381828}

ESC should be handled when pressed independently from other keys. Otherwise, we'll be overriding combinations such as shift+esc, which are keyboard shortcuts.

BUG=592422

Review URL: https://codereview.chromium.org/1800213002

Cr-Commit-Position: refs/heads/master@{#381827}

We can expose test_runner::WebTask as blink::WebTaskRunner::Task.
Doing this means that components/test_runner/web_task.h header
no longer needs to be included outside of components/test_runner.

BUG=595089

Review URL: https://codereview.chromium.org/1805753002

Cr-Commit-Position: refs/heads/master@{#381826}

The spinner is takes up more height than the "Device missing?" text, so there's some seemingly janky behavior with the dialog a few seconds after the dialog is opened if there are no devices found. This change ensures the dialog doesn't resize immediately after opening when transitioning to showing "Device missing?".

BUG=595021

Review URL: https://codereview.chromium.org/1805533004

Cr-Commit-Position: refs/heads/master@{#381825}

BUG=592869

Review URL: https://codereview.chromium.org/1814703002

Cr-Commit-Position: refs/heads/master@{#381824}

Change discoverNow to fire the deviceList event with the latest
device list even if the list had not changed. The patch ensures that
by incrementing the registry generation if discovery was started
during discoverNow.

BUG=595204

Review URL: https://codereview.chromium.org/1810433002

Cr-Commit-Position: refs/heads/master@{#381823}

Open all links from the dialog (JS side) in a new tab. This prevents navigating the user away from whatever they're currently doing on the current active tab. The dialog remains open on the original tab.

URLs that are opened on the C++ side already open in a new tab.

BUG=594653

Review URL: https://codereview.chromium.org/1809463002

Cr-Commit-Position: refs/heads/master@{#381822}

A problem with the design in that blocking touch events could get
ahead of the non-blocking events was revealed in discussions. Change
the queue so that it processes all events going to the main thread.
It may elect to put blocking events in queues that it maintains so that
events are not re-ordered when dispatched to the main thread.

Force the ack behavior in the renderer to match the disposition
of the event. The benefit of this is uncancelable touch events
now can be coalesced if the main thread is behind.

BUG=489802

Review URL: https://codereview.chromium.org/1780953003

Cr-Commit-Position: refs/heads/master@{#381821}

BUG=580337

Review URL: https://codereview.chromium.org/1810843002

Cr-Commit-Position: refs/heads/master@{#381820}

To simplify public API surface of components/test_runner we can hide
LayoutDumpFlags behind WebTestRunner interface.  For example - instead
of returning the whole LayoutDumpFlags struct, we can only expose
IsRecursiveLayoutDumpRequested method - this is consistent with how
other similar methods are exposed via WebTestRunner (i.e. we already
have ShouldDumpAsAudio and/or ShouldGeneratePixelResults).

After doing this components/test_runner/layout_dump_flags.h and
components/test_runner/layout_dump.h headers no longer needs to be
included outside of components/test_runner.

BUG=595089

Review URL: https://codereview.chromium.org/1805243002

Cr-Commit-Position: refs/heads/master@{#381819}

MB was originally intended to be used to manage non-bot configs
as well, but we ended up not using this. This CL removes a bunch
of associated code that was hence no longer needed.

R=jbudorick@chromium.org, kjellander@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1808093002

Cr-Commit-Position: refs/heads/master@{#381818}

Revert of Use PhantomReference for CleanupReference (patchset #1 id:1 of https://codereview.chromium.org/1802233002/ )

Reason for revert:
Found out that PhantomReference can survive (ie not get enqueued) a GC pass. Tested on Android M.

This would cause unacceptable memory regressions for apps that relies on gc to clean up webview.

Original issue's description:
> android: Use PhantomReference for CleanupReference
>
> PhantomReference is less of a burden on dalvik/art gc systems.
>
> Committed: https://crrev.com/9b1b3599500eca2b8d664714edf6b6771a789123
> Cr-Commit-Position: refs/heads/master@{#381173}

TBR=dtrainor@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.

Review URL: https://codereview.chromium.org/1815443002

Cr-Commit-Position: refs/heads/master@{#381817}

https://chromium.googlesource.com/native_client/src/native_client.git/+log/fc8541e88a9b..4bac1a1e26d2

$ git log fc8541e88..4bac1a1e2 --date=short --no-merges --format='%ad %ae %s'
2016-03-17 petarj [MIPS] Add missing -L flag to mips newlib nexe action

CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_chromium_dbg_32_ng,linux_nacl_sdk_build
TBR=mseaborn@chromium.org

Review URL: https://codereview.chromium.org/1815483002

Cr-Commit-Position: refs/heads/master@{#381816}

A few flakes in the test have been resolved and it is passing
everywhere. Re-enable it for the Win8 TSan build in hopes that
it won't occur.

BUG=588199

Review URL: https://codereview.chromium.org/1806853007

Cr-Commit-Position: refs/heads/master@{#381815}

Ownership of MockScreenOrientationClient can be moved to TestRunner
(which holds ownership of other things that can be influenced by
javascript bindings).

Furthermore, TestRunner doesn't need to take a detour via
WebTestDelegate (aka BlinkTestRunner) to interact with
MockScreenOrientationClient.  This means that we can remove three
screen-orientation-related methods from WebTestDelegate.  This in turn
means that components/test_runner/mock_screen_orientation_client.h
header no longer needs to be included outside of components/test_runner.

BUG=595089

Review URL: https://codereview.chromium.org/1807733002

Cr-Commit-Position: refs/heads/master@{#381814}

+ Also fixes a bug where the proxy, firewall suggestion link opens an invalid URL.

BUG=591987,584615

Review URL: https://codereview.chromium.org/1786113002

Cr-Commit-Position: refs/heads/master@{#381813}

In order to get BoringSSL to build on iOS we need to make a couple of initial changes:

* We disable unittests due to the current inability to run new processes on iOS.
* We disable the use of custom assembly since we are unable to build the assembly for the iOS architecture and would otherwise be unable to test it.

BUG=591545

Review URL: https://codereview.chromium.org/1809153002

Cr-Commit-Position: refs/heads/master@{#381812}

But not on Chrome OS.
This reverts commit 7812fb8b.

BUG=62400

Review URL: https://codereview.chromium.org/1811613005

Cr-Commit-Position: refs/heads/master@{#381811}

https://chromium.googlesource.com/skia.git/+log/852f15da7ceb..c571c001cee4

$ git log 852f15da7..c571c001c --date=short --no-merges --format='%ad %ae %s'
2016-03-17 brianosman Revert of sRGB support in Ganesh. Several pieces: (patchset #12 id:220001 of https://codereview.chromium.org/1789663002/ )
2016-03-17 cdalton Add asRRect method to SkClipStack::Element
2016-03-17 mtklein DM: make --verbose send the vlog to stderr.
2016-03-17 brianosman sRGB support in Ganesh. Several pieces:
2016-03-17 jbroman Fix pointer aliasing bug in SkImageFilter::computeFastBounds.
2016-03-17 mtklein just write the verbose log file any time we have a --writePath
2016-03-17 egdaniel Update how we send draws to gpu backend to reduce state setting.

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=bungeman@google.com

Review URL: https://codereview.chromium.org/1810053003

Cr-Commit-Position: refs/heads/master@{#381810}

BUG=595444

Review URL: https://codereview.chromium.org/1807253002

Cr-Commit-Position: refs/heads/master@{#381809}

Previously only desktop screenshots were retried. This works around some flakiness we've observed in recent versions of Chrome for Android. More context (for Googlers only, sorry) is available at b/27644687.

BUG=
TBR=stgao@chromium.org

Review URL: https://codereview.chromium.org/1809833002

Cr-Commit-Position: refs/heads/master@{#381808}

This allows it to be used in various algorithms.

BUG=
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Review URL: https://codereview.chromium.org/1728253003

Cr-Commit-Position: refs/heads/master@{#381807}

BUG=595534
TEST=On device pin Arc app in shelf. Pin persistent across reboots
     and order is kept. App is started once clicked on pinned item.
     Uninstall app and pin is removed.
TEST=Extended unit_tests passes

Review URL: https://codereview.chromium.org/1808963002

Cr-Commit-Position: refs/heads/master@{#381806}

There's no need to build an entire list of codecs and transport that
across JNI when we always want the first one in the list.

Notably this reduces the cost of GpuChildThread::OnInitialize() by
over 400ms on an N5 running Marshmallow (450.208ms vs 39.64ms)! The
reduction comes from VideoEncodeAccelerator no longer creating ~3
MediaCodec instances to determine the codec name during the call to
VideoEncodeAccelerator::GetSupportedProfiles() during child thread
startup. I suspect this will translate into real GPU process startup
improvements.

BUG=594814
TEST=codec lists match before after on KitKat below and above.

Review URL: https://codereview.chromium.org/1805163002

Cr-Commit-Position: refs/heads/master@{#381805}

The Clear Browsing Data dialog UI will provide a notice informing the
user that history.google.com stores other forms of browsing history
(such as the search history).

We will only show the notice to those users for whom it is valid. This
will be determined by Sync status, and by a JSON response from
history.google.com [not yet implemented].

See https://docs.google.com/document/d/1ZMDSAd44KmzKhqXPjobZOf9rZezs6VqiBnqpDD0auCU/
for the background.

To communicate between backends (ProfileSyncService, HistoryService) and
frontends on all platforms, we need to put this logic to a component
(especially because of iOS). We create a new component "browsing_data_ui".

BUG=595332

Review URL: https://codereview.chromium.org/1806873002

Cr-Commit-Position: refs/heads/master@{#381804}

Only used for Metro on Windows, which no longer exists.

R=dpapad@chromium.org
BUG=558054

Review URL: https://codereview.chromium.org/1809213002

Cr-Commit-Position: refs/heads/master@{#381803}

These tests fail to set up correctly on Windows, so running the tests
just crashes. Restore the early outs with an ifdef for windows so we
document where the early out will happen, along with comments.

R=kbr@chromium.org, piman@chromium.org
BUG=595823, 584497

Review URL: https://codereview.chromium.org/1812033002

Cr-Commit-Position: refs/heads/master@{#381802}

Remove options that were valid for pnacl toolchain but are not valid for
mipsel-nacl-clang.
This fixes the build issues:

 mipsel-nacl-clang: error: unsupported option '--pnacl-mips-bias'
 mipsel-nacl-clang: error: unsupported option '--pnacl-allow-translate'

Review URL: https://codereview.chromium.org/1806403002

Cr-Commit-Position: refs/heads/master@{#381801}

Now that VideoFrames are transported across mojo using shared memory,
the original needs to stay around until the other side is no longer
using it, after which the shared memory can be reused for a
subsequent VideoFrame.

BUG=510153
TEST=compiles on Windows

Review URL: https://codereview.chromium.org/1809903002

Cr-Commit-Position: refs/heads/master@{#381800}

This removes the method from the WebGraphicsContext3D API, and makes
callers go directly to the GLES2Interface instead.

The DepthStencilTrackingGLES2Interface is introduced for the
DepthStencilTrackingContext to use in DrawingBufferTest, and code will
migrate to it as methods are replaced in WebGraphicsContext3D.

R=kbr@chromium.org
BUG=584497
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel

Review URL: https://codereview.chromium.org/1808933002

Cr-Commit-Position: refs/heads/master@{#381799}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/873938e2..79f4376c

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.

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

Review URL: https://codereview.chromium.org/1811083002

Cr-Commit-Position: refs/heads/master@{#381798}

WebTestInterfaces::GetAppBannerClient can be removed, because
TestRunner::ResolveBeforeInstallPromptPromise can interact with
AppBannerClient directly, rather than going through WebTestDelegate.
This also means that components/test_runner/app_banner_client.h header
no longer needs to be included outside of components/test_runner.

BUG=595089

Review URL: https://codereview.chromium.org/1807643002

Cr-Commit-Position: refs/heads/master@{#381797}

Previoulsy both token_binding_nss.cc and token_binding_openssl.cc
were compiled when compiling net for NaCl. This was causing linker
warkings about duplicate symbols.

BUG=595790

Review URL: https://codereview.chromium.org/1807313002

Cr-Commit-Position: refs/heads/master@{#381796}

Up to and including internal change 117245736

Fix a type casting bug in quic stream sequencer buffer.

Withough this change, a QUIC frame with offset higher than 2^32 might corrupt StreamSequencerBuffer's missing frame recording logic in a 32-bit machine.

Merge internal change: 117245736
https://codereview.chromium.org/1808013003/

Always use recvmmsg or recvmsg if it's available, and consolidate ReadAndDispatchPackets and ReadAndDispatchSinglePacket into a single API. Not flag protected. No functional change.

Note that this does somewhat reduce the test coverage of QuicServerTest, now
requiring that changes to QuicPacketReader::ReadAndDispatchSinglePacket be run
on platforms that don't support recvmmsg.

Merge internal change: 117245279
https://codereview.chromium.org/1810103002/

Log when client nonce is invalid. No functional change. Not flag protected.

Merge internal change: 116987771
https://codereview.chromium.org/1810073002/

Pass QuicTime, QuicTime::Delta, and QuicBandwidth exclusively by value. Not flag protected. No functional change.

Merge internal change: 116973081
https://codereview.chromium.org/1809123002/

Change QUIC tests that use port 80 to use 443.

Merge internal change: 116970314
https://codereview.chromium.org/1808963003/

Use byte conservation instead of packet conservation in QUIC's Slow Start Large Recovery experiment. Protected by quic_sslr_byte_conservation.

Merge internal change: 116964633
https://codereview.chromium.org/1814873002/

Remove the unused has_retransmittable_data argument from QuicSendAlgorithm::TimeUntilSend.  No functional change.  Not flag protected.

Merge internal change: 116904472
https://codereview.chromium.org/1814843002/

Deprecates check_peer_address_change_after_decryption.

Merge internal change: 116887287
https://codereview.chromium.org/1808613003/

Deprecate quic_respect_send_alarm2.

Merge internal change: 116879870
https://codereview.chromium.org/1813513002/

Deprecate --quic_include_path_id_in_iv.

Merge internal change: 116877752
https://codereview.chromium.org/1810743002/

Don't copy the QuicAckFrame into the QuicPacketGenerator.  Protected by quic_dont_copy_acks.

Merge internal change: 116808648
https://codereview.chromium.org/1807033002/

No longer limit QUIC's max CWND to 200 packets.  Protected by quic_dont_limit_max_cwnd.

This is a fix for the deprecation of a similar flag in cr/113769379.

Merge internal change: 116794514
https://codereview.chromium.org/1812633002/

Remove QuicFrameList, because it's unused with the new QuicStreamSequencer.

Merge internal change: 116768420
https://codereview.chromium.org/1805323003/

Add a QuicConfig option to ask the client not to attempt migration.

Merge internal change: 116712124
https://codereview.chromium.org/1808513003/

Do not disable TLP by default in QuicConnection tests.

Merge internal change: 116698228
https://codereview.chromium.org/1808863002/

Add a new QUIC connection option, 5RTO, which closes the QUIC connection after 5 consecutive RTOs. Protected by quic_enable_rto_timeout.

Merge internal change: 116687954
https://codereview.chromium.org/1806083002/

Change QuicConnection's ScopedPacketBundler to use SEND_QUEUED_ACK instead of NO_ACK, to ensure a queued ack is always sent.  Small functional change.  Not flag protected.

Merge internal change: 116677107
https://codereview.chromium.org/1814483002/

deprecate quic_use_rfc7539

Merge internal change: 116589902
https://codereview.chromium.org/1809703002/

Use cached compressed certs if available to reduce load of compressing certs. Flag protected by quic_use_cached_compressed_certs.

Merge internal change: 116579250.
https://codereview.chromium.org/1808793002/

R=rch@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1811043002

Cr-Commit-Position: refs/heads/master@{#381795}