GitLab

Keep track of the display configuration callbacks and trigger them when the
GPU process responds.

BUG=429746
NOTRY=true
TBR=kenrb@chromium.org

Review URL: https://codereview.chromium.org/800743002

Cr-Commit-Position: refs/heads/master@{#308255}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/1fdbfa4..d8c8e9e

TBR=lushnikov@chromium.org,vsevik@chromium.org,dglazkov@chromium.org

Review URL: https://codereview.chromium.org/797833004

Cr-Commit-Position: refs/heads/master@{#308254}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/1629118..1607156

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

Review URL: https://codereview.chromium.org/801963002

Cr-Commit-Position: refs/heads/master@{#308253}

woops, when I was testing this, I hadn't noticed the mistake because
I was testing over a white background, and the shadow looks exactly
like the bubble in that case.

I had used the wrong color for the shadow color, and as a result I
didn't notice that the shadow was in the wrong spot.

I did a bit more extensive testing, if you have chrome fullscreen on the
left side, and you show the bubble on the right, it does extend to right
screen by 1 pixel.  This happened before and after my change.

BUG=435910

Review URL: https://codereview.chromium.org/792243002

Cr-Commit-Position: refs/heads/master@{#308252}

Use this for platforms with no control over input devices.

BUG=441856
TEST=build

Review URL: https://codereview.chromium.org/802813002

Cr-Commit-Position: refs/heads/master@{#308251}

Cleaning up since the entry points should not be used/needed anymore.
Also removing the Configure() calls from TakeControl() and replace them
with a call to probe and configure displays.

BUG=429746

Review URL: https://codereview.chromium.org/796263002

Cr-Commit-Position: refs/heads/master@{#308250}

Frankly I'm tired of all the inconsistencies and <!DOCTYPE> looks odd being the
only tag-like thing on the page that's UPPERCASE.

It Officially Doesn't Matter according to the HTML5 spec:
http://www.w3.org/TR/html5/syntax.html#the-doctype

R=estade@chromium.org
BUG=none
TEST=less annoyance

Review URL: https://codereview.chromium.org/802863002

Cr-Commit-Position: refs/heads/master@{#308249}

Not only does this do less work in the browser process, but it also
sidesteps issues with SkBitmap pickling for IPC, resolving a crash
when the user attempts to search for an invalid image.

BUG=431418

Review URL: https://codereview.chromium.org/792903002

Cr-Commit-Position: refs/heads/master@{#308248}

Depends on blink side change in https://codereview.chromium.org/782463002/
and on the blink side change in https://codereview.chromium.org/799703003/

BUG=426458

Review URL: https://codereview.chromium.org/781723002

Cr-Commit-Position: refs/heads/master@{#308247}

BUG=302704

Review URL: https://codereview.chromium.org/798173002

Cr-Commit-Position: refs/heads/master@{#308246}

DXVAVideoDecodeAccelerator::GetState uses InterlockedAdd(0) to get the recent state.

BUG=441000

Review URL: https://codereview.chromium.org/793993002

Cr-Commit-Position: refs/heads/master@{#308245}

In case actions.xml is included in the changelist the chrome/browser/resources/PRESUBMIT.py shouldn't run because the presubmit for actions.xml will do more complete job.

Fixing the bug to look into all changed files in the CL where previously it was only looking at subdirectories of current presubmit script to find actions.xml

BUG=439978

Review URL: https://codereview.chromium.org/760163004

Cr-Commit-Position: refs/heads/master@{#308244}

The proxy config service for android does not allow
one to configure an exception for localhost be made.
The implementation for Linux did allow this in:
   proxy_config_service_linux.cc
it just seems it was overlooked for android.

Instead of adding in the ability to configure
localhost as an exclusion, it makes more sense
to just force the exception; it seems silly to
send traffic for localhost through a proxy.

This patch addresses the following bug in android:
   https://code.google.com/p/android/issues/detail?id=37031

BUG=

Review URL: https://codereview.chromium.org/796003002

Cr-Commit-Position: refs/heads/master@{#308243}

Now since clank supports Enhanced Bookmark feature on tablet, this flag
should be turned on.

BUG=NONE

Review URL: https://codereview.chromium.org/803703002

Cr-Commit-Position: refs/heads/master@{#308242}

The problematic sequence of events is as follows:
1. ShouldRequestOnNonUiThread, which posts a closure
(request_draw_gl_closure_) to UI thread
2. That closure gets run on UI thread, and it schedules the invokeFunctor
with Android framework
3. Before the corresponding invokeFunctor actually happens on RT (which
is DrawGL process mode),  ShouldRequestOnUiTdread is called on the UI
thread. At this point, pending_non_ui_ is not null, we cancel the
callback, which does nothing, because WebView has already scheduled an
invokeFunctor with the Android framework in Step 2. Then we schedule
another invokeFunctor immediately on the UI thread. So there are 2
invokeFunctors queued in Android framework in this case.

This CL tries keep track of whether or not we've
queued an invokeFunctor in Android framework already.

BUG=442013

Review URL: https://codereview.chromium.org/801923003

Cr-Commit-Position: refs/heads/master@{#308241}

Review URL: https://codereview.chromium.org/804663002

Cr-Commit-Position: refs/heads/master@{#308240}

Part of change to collect, upload and expose histogram data
for Cronet.

BUG=441466

Review URL: https://codereview.chromium.org/794493002

Cr-Commit-Position: refs/heads/master@{#308239}

BUG=441966
TEST=Upload a previously selected file from subframe form after restore.

Review URL: https://codereview.chromium.org/805563003

Cr-Commit-Position: refs/heads/master@{#308238}

Add a method to EventDeviceInfo to determine if the device has a
keyboard on it based on the key bitmap. We'll require the first 31 keys
to be considered a keyboard; this is compatible with udev.

TEST=chrome on link_freon & rush_ryu
BUG=437539

Review URL: https://codereview.chromium.org/797483002

Cr-Commit-Position: refs/heads/master@{#308237}

The SurfaceAggregator will never need the resources for a frame after a new frame has been submitted for that Surface, so return the resources early.

BUG=441698,441751,440213

Review URL: https://codereview.chromium.org/800133002

Cr-Commit-Position: refs/heads/master@{#308236}

BUG=437700
TEST=everything builds as before

Review URL: https://codereview.chromium.org/790603006

Cr-Commit-Position: refs/heads/master@{#308235}

BUG=406439
TEST=none
R=tengs@chromium.org

Review URL: https://codereview.chromium.org/802823002

Cr-Commit-Position: refs/heads/master@{#308234}

The code currently excludes paste-and-go operations.  It should probably
also exclude paste+enter operations; these are effectively equivalent.

TODO before submitting: test interactively

BUG=
TBR=bauerb@chromium.org

Review URL: https://codereview.chromium.org/795343002

Cr-Commit-Position: refs/heads/master@{#308233}

We previously were not properly expanding the free list when
dup2 results in growing the descriptor range.

Added of test of this.

BUG=None
TEST=Dup2Allocate
R=binji@chromium.org

Review URL: https://codereview.chromium.org/795213003

Cr-Commit-Position: refs/heads/master@{#308232}

Fullscreen exit bubble is actually used to control both mouse lock and fullscreen. Addition of keyboard lock would make it the third behavior requiring some clean-up. This is part 1 of the refactor and contains just renames.

Also the difference from the original CL is that I have reverted rename of fullscreen_controller in lieu of doing it when I actually move the functionality and using exclusive_access_* instead of lock_exit.

BUG=166928

Review URL: https://codereview.chromium.org/789403002

Cr-Commit-Position: refs/heads/master@{#308231}

BUG=441955

Review URL: https://codereview.chromium.org/805553002

Cr-Commit-Position: refs/heads/master@{#308230}

APIPermission::kUnlimitedStorage is used in the storage implementation in
//extensions so the permission should live there.

This allows app_shell to load apps with unlimitedStorage in the manifest.

BUG=none
TEST=existing browser_tests

Review URL: https://codereview.chromium.org/805563002

Cr-Commit-Position: refs/heads/master@{#308229}

Original review: https://codereview.chromium.org/801493002/

BUG=429746
TESTS=display_unittests with ASAN
TBR=derat@chromium.org

Review URL: https://codereview.chromium.org/800143002

Cr-Commit-Position: refs/heads/master@{#308228}

They've been supplanted by ffmpeg_regression_tests. They also crash
when you run them.

TBR=glider

Review URL: https://codereview.chromium.org/801993002

Cr-Commit-Position: refs/heads/master@{#308227}

Stop using HTTP/2 in case TLS 1.2 is not supported, connection has been
downgraded to below TLS 1.2, or AES-GCM cipher required by HTTP/2 draft
specification is not supported.

BUG=436835

Review URL: https://codereview.chromium.org/757033004

Cr-Commit-Position: refs/heads/master@{#308226}

BUG=285151
R=guohui@chromium.org

Review URL: https://codereview.chromium.org/797303002

Cr-Commit-Position: refs/heads/master@{#308225}

BUG=428919

Review URL: https://codereview.chromium.org/804443002

Cr-Commit-Position: refs/heads/master@{#308224}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/095dcae..1fdbfa4

TBR=lushnikov@chromium.org,vsevik@chromium.org,dglazkov@chromium.org

Review URL: https://codereview.chromium.org/788073005

Cr-Commit-Position: refs/heads/master@{#308223}

This is temporary instrumentation to try and narrow down the area
and mode where Chrome is (likely) hanging on logoff/restart.

TBR=mpearson@chromium.org,erikwright@chromium.org

BUG=412384

Review URL: https://codereview.chromium.org/796963002

Cr-Commit-Position: refs/heads/master@{#308222}

When tasks are posted to a message loop, we may need to schedule the
message loop to wake up if it was sleeping so it processes the message.
The incoming task queue checks whether the incoming queue was already
empty as a way to avoid scheduling unnecessarily, but we can be more
precise. Once a message loop is scheduled, we know that it will sooner
or later wake up and enter its run loop. In its run loop it will
repeatedly drain the work queue and attempt to reload from the incoming
queue until there is no work immediately available to do. This means that
before waiting for more work the loop will always reload from an empty
incoming queue. Thus, once we schedule a loop we do not need to schedule
the same loop again until after an empty reload occurs.

This adds a bool to IncomingTaskQueue protected by the incoming task
queue lock that keeps track of if the loop has been scheduled since the
last empty reload.

Local testing on Linux desktop and a ChromeShell Android build shows
that this avoids roughly 40% of the ScheduleWork calls when browsing
around to random websites.

This also has the very nice consequence that when running a task and
posting another task to the same thread we *never* need to invoke
ScheduleWork which avoids the cost of the ScheduleWork call and avoids
a spurious wakeup when going to sleep.

BUG=412137

Review URL: https://codereview.chromium.org/614723003

Cr-Commit-Position: refs/heads/master@{#308221}

and -> &

BUG=441480

Review URL: https://codereview.chromium.org/799123002

Cr-Commit-Position: refs/heads/master@{#308220}

Fixes a null pointer dereference in RenderFrameProxyHost::RenderFrameProxyHost,
when a RenderFrameProxyHost is created as a side effect of destroying the
FrameTreeNode and RenderFrameHostManager. The crash occurs because of the
operation below:

  if (!frame_tree_node_->IsMainFrame() &&
      frame_tree_node_->parent()
              ->render_manager()
              ->current_frame_host()
              ->GetSiteInstance() == site_instance) {

When a FrameTreeNode is being detached from the tree, it is not the main
frame (because it is not equal to the root), but it also does not have a
parent (because, for reasons explained in FrameTreeNode::RemoveChild, it
is trimmed from the tree before destruction). So the parent() call above
returns NULL, resulting in great misfortune.

Add a test that failed without the fix.

BUG=441357

Review URL: https://codereview.chromium.org/799593004

Cr-Commit-Position: refs/heads/master@{#308219}

While in a nested loop, Chrome is not blocked, keeps updating the client area and is responding to user actions.
Subtracting the time in nested loops from the task in which it happens, so that we don't diagnose this task as a source of jank.

BUG=401560, 440919

Review URL: https://codereview.chromium.org/788883004

Cr-Commit-Position: refs/heads/master@{#308218}

The file descriptors aren't needed anymore once the file is mapped, and we want to avoid running out.

BUG=364272,362048,434890,425766,412136,376441,436677

Review URL: https://codereview.chromium.org/799133003

Cr-Commit-Position: refs/heads/master@{#308217}

Doing so forces the library to load and triggers our FMA3-disabling
code.

BUG=440892

Review URL: https://codereview.chromium.org/797313003

Cr-Commit-Position: refs/heads/master@{#308216}