GitLab

Chrome browser opens Files app as a file saving dialog.  The caller of saving
dialog has a special handling for drive paths, but it does not handle other
non-native paths (e.g. FSP, MTP)

The CL lets Files app know which types of paths are handled by caller and hide
volumes that are not supported by caller.

BUG=581984
TEST=Check with standalone Files app, save dialog, open dialog

Review URL: https://codereview.chromium.org/1701163003

Cr-Commit-Position: refs/heads/master@{#376073}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/c152be3c..24a43392

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/1708023002

Cr-Commit-Position: refs/heads/master@{#376072}

R=michaelpg@chromium.org
BUG=586579

Review URL: https://codereview.chromium.org/1700273003

Cr-Commit-Position: refs/heads/master@{#376071}

BUG=582688

Review URL: https://codereview.chromium.org/1705613002

Cr-Commit-Position: refs/heads/master@{#376070}

https://pdfium.googlesource.com/pdfium.git/+log/53520a1..1e1d3b0

BUG=587620
TBR=thestig@chromium.org

Review URL: https://codereview.chromium.org/1711573002

Cr-Commit-Position: refs/heads/master@{#376069}

This CL enables SurfaceView fullscreen video with the unified media pipeline
in cases where we use the AVDA deferred rendering strategy today.

* There's a new media interface called SurfaceManager with two methods:
  CreateFullscreenSurface() and NaturalSizeChanged().
* SurfaceManager is implemented by RendererSurfaceViewManager, which is a
  RenderFrameObserver, and BrowserSurfaceViewManager, which is created for
  each RenderFrame by MediaWebContentsObserverAndroid.
* BrowserSurfaceViewManager creates a ContentVideoView in response to a
  CreateFullscreenSurface message, and registers it in a surface map that the
  decoder can look up in the GPU process.
* WMPI interacts with SurfaceManager on behalf of the decoder. It passes
  GpuVideoDecoder a callback for requesting surfaces which it calls before
  initializing VDAs.
* In response to the callback, if the player is in fullscreen WMPI will pass
  the request to the SurfaceManager which will return the surface id to GVD. If
  the player is not in fullscreen WMPI will return a null surface id.
* When GVD gets a surface id back it completes its initialization by calling
  VDA::Initialize with the surface id and the VDA will render to the given
  surface.
* WMPI records that it got a surface request callback so that it knows that
  future fullscreen transitions will require a restart of the decoder in order
  to switch surfaces.
* A future CL will allow us to switch the surface dynamically without
  requiring a restart.

BUG=533630

Review URL: https://codereview.chromium.org/1655083002

Cr-Commit-Position: refs/heads/master@{#376068}

This CL has no tests.  We're not sure if this CL makes user-visible behavior
changes.  The purpose of this CL is to reduce ClusterFuzz crashes in the future.

This CL also updates comments in EditingState.h.  We can't remove
NoEditingAbortChecker and ASSERT_NO_EDITING_ABORT.

BUG=586846

Review URL: https://codereview.chromium.org/1703903002

Cr-Commit-Position: refs/heads/master@{#376067}

Also reimplement mojo_wait_for_incoming_message using mojo.support.asyncWait.

Review URL: https://codereview.chromium.org/1698933002

Cr-Commit-Position: refs/heads/master@{#376066}

Presence of OOPIFs introduces a delay into loading of cross-origin media and
exposes a race in the
http/tests/security/mixedContent/insecure-audio-video-in-main-frame.html
test.  To get rid of the race, we should wait until media has actually
started to load, before posting / reporting that the test is "done".

BUG=582541
TEST=The affected test passes when run 100 times with and without --site-per-process flag.

Review URL: https://codereview.chromium.org/1703333002

Cr-Commit-Position: refs/heads/master@{#376065}

BUG=

Review URL: https://codereview.chromium.org/1701933004

Cr-Commit-Position: refs/heads/master@{#376064}

This change is a subset of https://codereview.chromium.org/1698183005/

BUG=553459

Review URL: https://codereview.chromium.org/1702813002

Cr-Commit-Position: refs/heads/master@{#376063}

Revert of Add check so WIN_FASTLINK will be ignored on VS 2013 (patchset #4 id:60001 of https://codereview.chromium.org/1707813002/ )

Reason for revert:
Speculative revert: https://code.google.com/p/chromium/issues/detail?id=587687

Original issue's description:
> Add check so WIN_FASTLINK will be ignored on VS 2013
>
> win_fastlink is helpful for improving build speeds with VS 2015 but is
> not supported on VS 2013. To make it easier to enable this flag it needs
> to be silently ignored by VS 2013. This makes that happen.
>
> BUG=440500
>
> Committed: https://crrev.com/6b851bfe96519bca59237b79a003df41dcf98c3b
> Cr-Commit-Position: refs/heads/master@{#375999}

TBR=scottmg@chromium.org,brucedawson@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=440500

Review URL: https://codereview.chromium.org/1708073002

Cr-Commit-Position: refs/heads/master@{#376062}

This a followup to https://codereview.chromium.org/1693913002
which added the -m32 unnecessarily (since :compiler_cpu_abi
already adds it).

Review URL: https://codereview.chromium.org/1710663002

Cr-Commit-Position: refs/heads/master@{#376061}

As explained in crbug.com/585952#c7, lack of synchronization between updates
of allocators' free slots and the pending queue can lead a race resulting
in a pending spawn being stuck in the queue forewer.

This issue has been revealed by WebView's testCreateLoadDestroyManyTimes in
the sandboxed renderer mode.

BUG=429657,585952

Review URL: https://codereview.chromium.org/1695723004

Cr-Commit-Position: refs/heads/master@{#376060}

After we changed most of the oilpan bots to be named
"non-oilpan", we started getting an error because the
experimental linux_blink_oilpan_rel bot on
tryserver.chromium.linux did not get renamed on the
waterfall but did get renamed in mb_config.pyl .

Changing the bot name on the waterfall is somewhat annoying,
so for the moment we'll change the entry in MB instead.

TBR=haraken@chromium.org
BUG=587322
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_blink_oilpan_rel

Review URL: https://codereview.chromium.org/1701303003

Cr-Commit-Position: refs/heads/master@{#376059}

BUG=561584

Review URL: https://codereview.chromium.org/1703323002

Cr-Commit-Position: refs/heads/master@{#376058}

Set the target API for the class to 21 (lollipop) which is what is supported
by updatable webview. This prevents us individually suppress lint messages.

BUG=

Review URL: https://codereview.chromium.org/1704263002

Cr-Commit-Position: refs/heads/master@{#376057}

BUG= internal b/27226258

Review URL: https://codereview.chromium.org/1711563002

Cr-Commit-Position: refs/heads/master@{#376056}

BUG=384723
R=dmazzoni@chromium.org
TESTED=browser tests, using Jaws

Review URL: https://codereview.chromium.org/1695673004

Cr-Commit-Position: refs/heads/master@{#376055}

Calling IOSurfaceLock() followed by IOSurfaceUnlock() appears sufficient.

BUG=584760

Review URL: https://codereview.chromium.org/1709443002

Cr-Commit-Position: refs/heads/master@{#376054}

Use the non-linearized PDF code path if the server doesn't respond to range
requests with a partial response. Previously, Chrome entered a request loop if
the PDF was linearized and the server advertized byte range support but
responded with a 200 code instead of 206.

BUG=587502
R=thestig@chromium.org
TEST=Load https://www.usenix.org/legacy/publications/library/proceedings/sec96/full_papers/goldberg/goldberg.pdf

Review URL: https://codereview.chromium.org/1705353002

Cr-Commit-Position: refs/heads/master@{#376053}

Also, make sure that test objects inherit the versions from the parent.

BUG=578890

Review URL: https://codereview.chromium.org/1698043005

Cr-Commit-Position: refs/heads/master@{#376052}

This patch adds CSS parser support for the snap-height property, behind
the CSSSnapSize runtime flag.

Supporting layout will be in following patches.

A test was imported from csswg-test and was marked as failure. This
patch makes the test pass.

Spec: https://drafts.csswg.org/css-snap-size/
Dashboard: https://www.chromestatus.com/features/5734273533345792

BUG=586413
TEST=imported/csswg-test/css-snap-size-1/snap-height-parsing-001.html

Review URL: https://codereview.chromium.org/1696373003

Cr-Commit-Position: refs/heads/master@{#376051}

BUG=587669
TBR=dcheng@chromium.org, reillyg@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1705353003

Cr-Commit-Position: refs/heads/master@{#376050}

Certain drivers, particularly Mac drivers and ANGLE when running on the
D3D11 backend do not support timestamp queries despite implementing an
extension that has them in the specification (ARB_timer_query and
EXT_disjoint_timer_query). To indicate this, the driver returns 0 when
queried for the size of the timestamps they return using glGetQueryiv.
Additional logic was added to the GPUTimingImpl class to check for that
when using those extensions and to fall back to the time elapsed query
implementation of timestamps if the driver does not have native support.

BUG=587173
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel

Review URL: https://codereview.chromium.org/1687353002

Cr-Commit-Position: refs/heads/master@{#376049}

When onStartCommand() is called, it is possible that the intent is null.
According to the SDK page, service must take care of the null check.

BUG=587399

Review URL: https://codereview.chromium.org/1708913002

Cr-Commit-Position: refs/heads/master@{#376048}

BUG=587664
TBR=tbansai@chromium.org, thestig@chromium.org, reillyg@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1708043002

Cr-Commit-Position: refs/heads/master@{#376047}

Previously, the panel caused some flakeyness in the interactive ui tests. Try re-enabling again.

Review URL: https://codereview.chromium.org/1711483003

Cr-Commit-Position: refs/heads/master@{#376046}

BUG=586056
R=mdempsky@chromium.org

Review URL: https://codereview.chromium.org/1705243002

Cr-Commit-Position: refs/heads/master@{#376045}

The automation functions exponentialRampToValueAtTime,
setTargetAtTime, and setValueCurveAtTime were not correctly updating
the .value attribute of the AudioParam.  For the first two methods,
the computed value was one frame too far into the future.  For last
method, the return value was always the last curve element, which is
wrong.

BUG=574905
TEST=audioparam-update-value-attribute.html

Review URL: https://codereview.chromium.org/1695573002

Cr-Commit-Position: refs/heads/master@{#376044}

Reland of Change assert to release assert for WTF::double_conversion::Vector to prevent OOB memory access. (patchset #1 id:1 of https://codereview.chromium.org/1694093002/ )

Reason for revert:
This revert was not the culprit CL. Reverting the revert.

Original issue's description:
> Revert of Change assert to release assert for WTF::double_conversion::Vector to prevent OOB memory access. (patchset #2 id:20001 of https://codereview.chromium.org/1677363002/ )
>
> Reason for revert:
> Perf failures. Speculative revert to see if it fixes.
>
> BUG=586581,574802
>
> Original issue's description:
> > Change assert to release assert for WTF::double_conversion::Vector to prevent OOB memory access.
> >
> > R=inferno@chromium.org, mbarbella@chromium.org, ochang@chromium.org, tkent@chromium.org
> > BUG=574802
> >
> > Committed: https://crrev.com/ad618c3357e189fb685ea88fe4b8b25ac34f2c75
> > Cr-Commit-Position: refs/heads/master@{#374424}
>
> TBR=mbarbella@chromium.org,ochang@chromium.org,tkent@chromium.org,mmoroz@chromium.org
> # Not skipping CQ checks because original CL landed more than 1 days ago.
> BUG=574802
>
> Committed: https://crrev.com/64cb1f5327522142c658100746f34f8a012a45fa
> Cr-Commit-Position: refs/heads/master@{#375389}

TBR=mbarbella@chromium.org,ochang@chromium.org,tkent@chromium.org,mmoroz@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=586581,574802

Review URL: https://codereview.chromium.org/1708703003

Cr-Commit-Position: refs/heads/master@{#376043}

BUG=

Review URL: https://codereview.chromium.org/1711503002

Cr-Commit-Position: refs/heads/master@{#376042}

tcmalloc was disabled in crrev.com/1668643002 - and from that point on,
the WebKit valgrind bot was continuously red. Reverting to see if that
ameliorates the issue.

Revert "Disable tcmalloc when building for valgrind"

This reverts commit 65aeea27.

BUG=582398
R=thakis@chromium.org
TBR=bratell@opera.com, thestig@chromium.org, reillyg@chromium.org, primiano@chromium.org

Review URL: https://codereview.chromium.org/1707983002

Cr-Commit-Position: refs/heads/master@{#376041}

BUG=488195

Review URL: https://codereview.chromium.org/1708683003

Cr-Commit-Position: refs/heads/master@{#376040}

New functions have been added so a sync token can represent
multiple resources at once. Adding a single sync token at
the end of generating a multitude of resources can be a win.

The most obvious use cases of this is the video layer which
generates a resource for each layer. A single sync token is
now generated at the end of the layer generation.

The layer tree UI resource also generates its own sync token.

BUG=584381
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1707143002

Cr-Commit-Position: refs/heads/master@{#376039}

deqp/functional/gles3/negativeshaderapi.html failing on Linux/Release as
well.

R=zmo@chromium.org,kbr@chromium.org
BUG=483282

Review URL: https://codereview.chromium.org/1707803003

Cr-Commit-Position: refs/heads/master@{#376038}

deqp_functional_gles3_buffercopy is flaky on the Win8 release nvidia bot.

R=zmo@chromium.org,kbr@chromium.org
BUG=587601

Review URL: https://codereview.chromium.org/1704163002

Cr-Commit-Position: refs/heads/master@{#376037}

Adds the new argument, stores the list of origins for which to intercept
requests, and adds origin checks to the interception code. Also updated the
various unit and layout tests.

This implements (parts of) the changes to the spec made in
https://github.com/mkruisselbrink/ServiceWorker/commit/88bbe9324fa8ac38d8af3e347564450d598757aa

BUG=540509

Review URL: https://codereview.chromium.org/1656933003

Cr-Commit-Position: refs/heads/master@{#376036}

BUG=None

Review URL: https://codereview.chromium.org/1641893003

Cr-Commit-Position: refs/heads/master@{#376035}

This CL adds support for always encoding all images within SkPicture to
WebP. On the decoding side it also assumes that the input will be WebP.

The pixels from Skia are premultiplied, but the import functions available
from WebP only supports un-premultiplied data, so the code unpremultiples
the values during the encoding step. Also, the quality and method
configuration is hard coded since any change in it would change the
hash of the data.

Also, the alpha-channel is always used, since the WebP encoder will
optimize the alpha-channel if all values are 255.

BUG=577262

Review URL: https://codereview.chromium.org/1680333004

Cr-Commit-Position: refs/heads/master@{#376034}