GitLab

When video starts playing, HTMLMediaElement no longer issues a seek(0) call.
As a result, decoders needs to be prerolled so that video and audio will start outputting frames at the same time.
This change also fixes an issue that when EOS is reached, the decoder does not clean up the prerolling state

internal b/17812723

BUG=

Review URL: https://codereview.chromium.org/638233002

Cr-Commit-Position: refs/heads/master@{#299006}

This fixes stat, read and write methods when dealing
with windows tty file descriptors. We were not catching
this is our tests because we run with stdout/stderr
that are not ttys.

Also, fail early and hard during nacl_io_init and ps_main
if anything goes wrong setting things up.

Review URL: https://codereview.chromium.org/635113002

Cr-Commit-Position: refs/heads/master@{#299005}

This change adds a new python testing framework that will replace
the various adhoc variants of this we use in Chromium and Blink,
at least temporarily until if/when we can switch to infra's
expect_tests framework.

Using this library will allow us to consolidate a bunch of duplicated
code used for telemetry_unittests, webkit_python_tests,
mojo_python_tests, webkit_tests, and elsewhere.

TBR=cpu@chromium.org, dtu@chromium.org
BUG=402172

Review URL: https://codereview.chromium.org/627763002

Cr-Commit-Position: refs/heads/master@{#299004}

Background: For Athena we have an intelligent applciation management which allows us to load and unload applications / activities at any time. In some cases we "pre-load" applications to short loading time. In that case we do not want that the application can show its window activated (thus bringing it to front). Instead we have to override that request.

At the same time several things have changed in the application window framework which required. The Browsertests are testing the activation override as well as the activity order upon (re-)creation.

I added miket as OWNER for the extensions/* changes.

BUG=419379
TEST=AppActivityBrowserTest.CreatedAppGetsFocus, AppActivityBrowserTest.UnloadReloadApplicationInPlace, AppActivityBrowserTest.ReloadActivatedApplication, AppActivityBrowserTest.ReloadMovedApplication

Review URL: https://codereview.chromium.org/631333003

Cr-Commit-Position: refs/heads/master@{#299003}

Revert of Make it easier to debug failed symbolization (patchset #4 id:130001 of https://codereview.chromium.org/646683002/)

Reason for revert:
"Android Tests" bot is failing in "run stack tool with logcat dump" step.

http://build.chromium.org/p/chromium.linux/builders/Android%20Tests

Suspecting this patch.  Reverting.

@@@BUILD_STEP Run stack tool with logcat dump@@@
> /b/build/slave/Android_Tests/build/src/third_party/android_platform/development/scripts/stack --more-info /b/build/slave/Android_Tests/build/src/out/Release/full_log.txt
/b/build/slave/Android_Tests/build/src/third_party/android_platform/development/scripts/../../../../third_party/android_tools/ndk/toolchains/arm-linux-androideabi-4.6/prebuilt/linux-x86_64/bin/arm-linux-androideabi-addr2line: /b/build/slave/Android_Tests/build/src/third_party/android_platform/development/scripts/../../../../out/Release/icudtl.dat: File format not recognized

See bug https://code.google.com/p/chromium/issues/detail?id=422090

Original issue's description:
> Make it easier to debug failed symbolization
>
> This just adds a --verbose argument and some debug logging sprinkled in
> useful places.
>
> Adds '.' as a candidate directory for libraries (gn puts shared
> libraries at the root build directory).
>
> Committed: https://crrev.com/17ffc51dfc9bd6dfe0d38ce8f6f8313544ea25af
> Cr-Commit-Position: refs/heads/master@{#298952}

TBR=rmcilroy@chromium.org,cjhopman@chromium.org
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/642843003

Cr-Commit-Position: refs/heads/master@{#299002}

Use the ApplicationImpl::args() for InitGoogleTest.
(plumbs GTEST args from the mojo_shell commandline)

BUG=392646
TEST=GTEST args work like this> mojo_shell "mojo:mojo_example_apptests --gtest_filter=ExampleApptest.RunCallbackViaService --gtest_repeat=2"
R=hansmuller@chromium.org,davemoore@chromium.org

Review URL: https://codereview.chromium.org/626033003

Cr-Commit-Position: refs/heads/master@{#299001}

Revert of Enable debug fission for Release build (patchset #1 id:1 of https://codereview.chromium.org/605623002/)

Reason for revert:
Speculatively reverting.  Suspect it may be cause of leaks on linux asan lsan bot:

http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/6561

Original issue's description:
> Enable debug fission for Release build
>
> This enables debug fission for Release build. If the target compiler
> satisfies fission requirement, the fission is enabled for target no
> matter how old the host compiler is.
>
> BUG=None
> TEST=simple chrome for x64 platform passes with fission.
>
> Committed: https://crrev.com/d872dd066eb644f2d85e25429f55451cd2dd20d4
> Cr-Commit-Position: refs/heads/master@{#298925}

TBR=tansell@chromium.org,mithro@mithis.com,llozano@chromium.org,llozano@google.com,yunlian@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=None

Review URL: https://codereview.chromium.org/649433002

Cr-Commit-Position: refs/heads/master@{#299000}

Revert of Revert of ui::PlatformWindowDelegate::OnClosed() isn't called when a ui::X11Window is closed (patchset #1 id:1 of https://codereview.chromium.org/647573002/)

Reason for revert:
This was not the cause of the regression. See http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/6569

Original issue's description:
> Revert of ui::PlatformWindowDelegate::OnClosed() isn't called when a ui::X11Window is closed (patchset #1 id:1 of https://codereview.chromium.org/644443002/)
>
> Reason for revert:
> Only cl that seems related to http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/6561 falures.
>
> Original issue's description:
> > ui::PlatformWindowDelegate::OnClosed() isn't called when a ui::X11Window is closed
> >
> > This problem manifests itself in simple Mojo apps that use NativeViewportService on X11. When the NativeViewport window is closed, the NativeViewportClient OnDestroyed() method does not run.
> >
> > BUG=421536
> >
> > Committed: https://crrev.com/b37606e1c2a6703e58a61a978e8e32be72acae0c
> > Cr-Commit-Position: refs/heads/master@{#298926}
>
> TBR=ben@chromium.org,hansmuller@chromium.org
> NOTREECHECKS=true
> NOTRY=true
> BUG=421536
>
> Committed: https://crrev.com/b93c31988a55a9cb2a328e39520f56788585b542
> Cr-Commit-Position: refs/heads/master@{#298985}

TBR=ben@chromium.org,hansmuller@chromium.org,jwd@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=421536

Review URL: https://codereview.chromium.org/642953003

Cr-Commit-Position: refs/heads/master@{#298999}

BUG=422083,422087
TBR=jdduke@chromium.org,perkj@chromium.org

Review URL: https://codereview.chromium.org/644843002

Cr-Commit-Position: refs/heads/master@{#298998}

Far and away the top crasher.

BUG=421641

Review URL: https://codereview.chromium.org/641213002

Cr-Commit-Position: refs/heads/master@{#298997}

This CL is a prerequisite for:
https://codereview.chromium.org/473153002/

BUG=341408
TBR=benwells@chromium.org

Review URL: https://codereview.chromium.org/617183003

Cr-Commit-Position: refs/heads/master@{#298996}

This should fix the current failure to build the
naclports bundle as it includes the change to
the default install path for PNaCl libs and headers.

This brings in the most resent 32 changes from naclports:

873ca49 Update PNaCl toolchain paths le32-nacl/local -> le32-nacl/usr
6d5fdd0 Don't include fonts.dir and fonts.scale in font packages.
8663b87 Fix SDl2 build when libiconv is present.
abc951c Fix gcc mac build by adding required argument to 'find' command.
062c105 Cleanup bash scripts
13e8dd0 Update openal-soft port
f24c45d When patching gcc specs file use existing specs if found
9141902 Fix httpd import in chrome_test
040f937 Remove temporary PNaCl workaround.
3262ba1 Switching back to __builtin_bswap32.
58e9aee Adding make test to devenv, changing layout.
b6bf444 Use sha1 rather than timestamps when checking build directories are up-to-date.
cadf6bf Fix coreutils and git for pnacl and enabling devenv-latest there.
2d2887c Add missing glibc-compat dependency to boost
5c633b4 Fix use of undefined symbol in PRESUBMIT.py
afc2def Add demo for libgit2 that runs in chrome
dcdf280 Update boost nacl.patch
3c26495 Fix compiler used for boost build.
f7d79c9 Update tinyxml and add tests
c023941 Remove libmodplug from the list of prebuild ports.
45e3d9f Fix mmap bug in libgit2
41b62db Fix jpeg6b package
75d83f8 Add tests for libgmp
27c048f Fix ncurses build.
87538f7 Add openjpeg
3cf4d95 Remove -Wenum-compare flag from grep build.
5ad247e Set ARFLAGS when running configure.
7095f09 Add missing glibc-compat dependency to libmodplug
849dcf0 Fix typo in __byte_swap_32 in glibc-compat
d527604 Add mp4v2 library
68aa657 Fix libmodplug build
47c1c17 Add random/urandom to glibc-compat

TEST=build_tools/build_sdk.py --build-ports  --build-app-engine
R=binji@chromium.org

Review URL: https://codereview.chromium.org/641233002

Cr-Commit-Position: refs/heads/master@{#298995}

(Just before I muck with it further, of course.)

R=brettw@chromium.org

Review URL: https://codereview.chromium.org/638343003

Cr-Commit-Position: refs/heads/master@{#298994}

BUG=

Review URL: https://codereview.chromium.org/643693003

Cr-Commit-Position: refs/heads/master@{#298993}

The bug 421097 was caused by the refactoring to disable trace
nesting call feature. Since trace-profiler makes the call to
StartTracing earlier than measurement and makes the call top
StopTracing after measurement has called StopTracing, it
raises TracingNotRunningException().

This fixes the bug by modifying the behavior of
StopTracing() slightly to return the trace data of the last
run. Note that trace-profiler  itself does not provide a
good way to get useful trace data (the measurement uses
tracing with different overhead level without knowing it).
The long term fix should be moving all trace-based
measurement to timeline_based_measurement.

BUG=421097

Review URL: https://codereview.chromium.org/642453004

Cr-Commit-Position: refs/heads/master@{#298992}

The C++ spec says the compiler may instantiate inline template
function definitions (e.g., member functions defined inline within a
templated class, even without an explicit "inline" keyword) even if
there's an "extern template class Foo<Bar>;" declaration.

In the case of scoped_refptr, if we merely have a forward declaration
for a class like "class Opaque;", then "extern template class
scoped_refptr<Opaque>;" will still cause compiler errors because of
scoped_refptr's inline member functions that try to call AddRef() and
Release().

To solve this, this CL introduces static helper methods that are
defined out-of-line to provide access to the AddRef() and Release()
methods.

BUG=414363

Review URL: https://codereview.chromium.org/623633002

Cr-Commit-Position: refs/heads/master@{#298991}

BUG=419850

Review URL: https://codereview.chromium.org/617103008

Cr-Commit-Position: refs/heads/master@{#298990}

From OS X 10.10, NSApplicationDidChangeScreenParametersNotification
will be sent when window enters fullscreen, and [NSScreen screens]
will be updated with new NSScreen instances equal to the previous
ones but with different pointer values. As a result, using == to
compare pointers of desiredScreen and current menu bar/dock screen
will break as the containsObject: check in -update does isEqual:
comparison.

This doesn't affect Chrome as Chrome is using EmbedsFullscreenWidget().

Review URL: https://codereview.chromium.org/635363002

Cr-Commit-Position: refs/heads/master@{#298989}

1. Move the text next to the checkbox down by 1px to make it look aligned.
2. Add 2 more px to the butom of the uma opt-in view.

BUG=293702

Review URL: https://codereview.chromium.org/585473004

Cr-Commit-Position: refs/heads/master@{#298988}

TBR=nasko@chromium.org
BUG=420557
TEST=Close a tab with an infinite unload event, then navigate right away.

Review URL: https://codereview.chromium.org/633323003

Cr-Commit-Position: refs/heads/master@{#298987}

BUG=421085

Review URL: https://codereview.chromium.org/642623005

Cr-Commit-Position: refs/heads/master@{#298986}

Revert of ui::PlatformWindowDelegate::OnClosed() isn't called when a ui::X11Window is closed (patchset #1 id:1 of https://codereview.chromium.org/644443002/)

Reason for revert:
Only cl that seems related to http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/6561 falures.

Original issue's description:
> ui::PlatformWindowDelegate::OnClosed() isn't called when a ui::X11Window is closed
>
> This problem manifests itself in simple Mojo apps that use NativeViewportService on X11. When the NativeViewport window is closed, the NativeViewportClient OnDestroyed() method does not run.
>
> BUG=421536
>
> Committed: https://crrev.com/b37606e1c2a6703e58a61a978e8e32be72acae0c
> Cr-Commit-Position: refs/heads/master@{#298926}

TBR=ben@chromium.org,hansmuller@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=421536

Review URL: https://codereview.chromium.org/647573002

Cr-Commit-Position: refs/heads/master@{#298985}

Review URL: https://codereview.chromium.org/640753003

Cr-Commit-Position: refs/heads/master@{#298984}

Use the versions on TestRenderFrameHost instead.

BUG=304341
TEST=unit_tests, content_unittests
TBR=phajdan.jr@chromium.org, mvanouwerkerk@chromium.org

Review URL: https://codereview.chromium.org/636673002

Cr-Commit-Position: refs/heads/master@{#298983}

This is another issue with HLS streams produced by VLC. Even though it's recommended that each PES packet in MPEG2TS contain a timestamp, we have seen that in the real-world content streams that's not always the case. In order to be able to play such streams, instead of failing stream parsing when we don't have timing info at the beginning of playback, we will just skip some frames until we eventually  get a frame with timing info and from that point we should be able to proceed playing normally. Previous change by Damien fixed this issue for video streams. This change fixes it for AAC ADTS and MPEG1 audio streams in mpeg2 ts containers.

BUG=420227

Review URL: https://codereview.chromium.org/646673002

Cr-Commit-Position: refs/heads/master@{#298982}

Chrome support for binding media stream audio track to speech recognition - renderer implementation.

  WebKit CL     : http://crrev.com/448163003
[ Renderer CL ]
  Browser CL    : http://crrev.com/549373003

SpeechRecognitionAudioSink is an audio sink which delivers audio data
via SharedMemory and SyncSocket to the peer (speech recognizer impl on the browser).
Audio resampling is done here rather than on the browser process.

Synchronization is done by managing two buffer indices, one in SpeechRecognitionAudioSink
incremented locally and the other in shared memory incremented by the peer.
Both buffer indices must match before a new signal is issued to the peer via |SyncSocket::Send()|.
That 4B signal is complemented by a |SyncSocket::Receive()| and used to trigger consumption on the peer.

WebMediaStreamTrack object is checked to be sourced from an audio device, avoiding possible WebAudio injection.
Unsupported tracks get notified to WebKit and session fails to start.

Issues with signaling over the SyncSocket are logged.

Design doc: http://goo.gl/9Ot3PC

BUG=408940

Review URL: https://codereview.chromium.org/499233003

Cr-Commit-Position: refs/heads/master@{#298981}

This ultimate consumers of this want a float anyway, so changing to be a float
all the way through is less misleading.  This also avoids some "value possibly
truncated" warnings (currently disabled) on MSVC.

This also removes a couple of scale-related functions in metafile_skia_wrapper.*
that seem to be entirely unused.

BUG=none
TEST=none

Review URL: https://codereview.chromium.org/641923002

Cr-Commit-Position: refs/heads/master@{#298980}

When extension manifests contain icon paths with a leading ./, we
were failing to process them properly

BUG=410666

Review URL: https://codereview.chromium.org/630243002

Cr-Commit-Position: refs/heads/master@{#298979}

BUG=421971
NOTRY=true

Review URL: https://codereview.chromium.org/640733004

Cr-Commit-Position: refs/heads/master@{#298978}

The caller RenderViewHostManager::SwapInRenderViewHost()
was deleted long time ago in
http://codereview.chromium.org/8081019

BUG=None
Test=No visible change, code cleanup.

Review URL: https://codereview.chromium.org/633313004

Cr-Commit-Position: refs/heads/master@{#298977}

This step is a giant search and replace for OVERRIDE and FINAL to
replace them with their lowercase versions.

BUG=417463

Review URL: https://codereview.chromium.org/637933002

Cr-Commit-Position: refs/heads/master@{#298976}

BUG=417463

Review URL: https://codereview.chromium.org/642083002

Cr-Commit-Position: refs/heads/master@{#298975}

Revert of Remove the RenderProcessHost observer and attach the WebContentsObserver earlier to the GuestView. (patchset #4 id:60001 of https://codereview.chromium.org/642573002/)

Reason for revert:
Speculative revert for double frees on mac gpu bots.

http://build.chromium.org/p/tryserver.chromium.gpu/builders/mac_gpu_retina_triggered_tests/builds/56935
http://build.chromium.org/p/chromium.gpu/builders/Mac%20Debug%20%28Intel%29/builds/38058

Original issue's description:
> Remove the RenderProcessHost observer and attach the WebContentsObserver earlier to the GuestView.
>
> This fixes an issue where if the embedder WebContents is destroyed during a GuestView creation we may end up leaking the guest WebContents.
>
> BUG=419020
>
> Committed: https://crrev.com/52fce9455ffd688ef3752a816192f9e76555e7d1
> Cr-Commit-Position: refs/heads/master@{#298934}

TBR=fsamuel@chromium.org,lfg@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=419020

Review URL: https://codereview.chromium.org/640293005

Cr-Commit-Position: refs/heads/master@{#298974}

Tahoma is the most important, the others are also necessary for
less common languages.

(Added to bottom of list so that it doesn't look like a scary diff; this
will need to be merged to 38.)

R=cpu@chromium.org, eae@chromium.org, shrikant@chromium.org
BUG=421856

Review URL: https://codereview.chromium.org/636063007

Cr-Commit-Position: refs/heads/master@{#298973}

This also makes it unnecessary to signal a sync point
callback for swap completion for browser
compositor surfaces which is skipped now.

BUG=417945

Review URL: https://codereview.chromium.org/628703005

Cr-Commit-Position: refs/heads/master@{#298972}

BUG=356475

Review URL: https://codereview.chromium.org/647513002

Cr-Commit-Position: refs/heads/master@{#298971}

This is necessary for link time validation. The old APIs do not
provide struct name, field order, etc.

BUG=401311
TEST=gpu_unittests,webgl_conformance
R=bajones@chromium.org,kbr@chromium.org

Review URL: https://codereview.chromium.org/619723008

Cr-Commit-Position: refs/heads/master@{#298970}

Revert of Use a single CGLContextObj to draw and update the IOSurface (patchset #4 id:60001 of https://codereview.chromium.org/627003002/)

Reason for revert:
I hear reports of instability on Mac try servers, with crashes in the GL driver.

Original issue's description:
> Use a single CGLContextObj to draw and update the IOSurface
>
> There is no reason that this can't be done, based on documentation,
> but it may be that this will cause crashes during texture destruction
> or context tear-down (see crbug.com/411782).
>
> I have 3 theories as to why these crashes are occurring
> 1. Using an IOSurface in a CGLContextObj that is also used by a
>    CAOpenGLLayer causes bugs
> 2. The crashes observed were because context destruction just crashes
>    at a certain frequency, and previous patches made context destruction
>    more frequent.
> 3. IOSurface-backed textures should be immutable -- no GL texture
>    should be bound to one IOSurface then the other.
>
> If theory 1 is correct, this will need to be reverted because the crashes
> will re-occur. If theories 2 or 3 are correct, then this patch is safe.
>
> BUG=420253
>
> Committed: https://crrev.com/0ea6e165293b7d24deec0d29fb9d6333d3a71b6d
> Cr-Commit-Position: refs/heads/master@{#298916}

TBR=kbr@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=420253

Review URL: https://codereview.chromium.org/647553002

Cr-Commit-Position: refs/heads/master@{#298969}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/e37ecd7..e099dd7

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=reed@google.com

Review URL: https://codereview.chromium.org/637323004

Cr-Commit-Position: refs/heads/master@{#298968}

Currently we fail to insert SPS/PPS if the key frame starts with AUD NALU and has one or more clear-text NALUs after that (I've seen this issue with a stream where key frames start with AUD+SEI NALU).

BUG=417562

Review URL: https://codereview.chromium.org/591713002

Cr-Commit-Position: refs/heads/master@{#298967}