GitLab

Revert of Now that the cq runs tests on rel, the compile-only bug should be dbg. (patchset #1 id:1 of https://codereview.chromium.org/937853002/)

Reason for revert:
Too many red builds on the dbg not, need to investigate tomorrow. Also, it does a static build for some reason.

Original issue's description:
> Now that the cq runs tests on rel, the compile-only bug should be dbg.
>
> Depends on https://codereview.chromium.org/913223007
>
> BUG=456396
> TBR=sergiyb@chromium.org
>
> Committed: https://crrev.com/3d0efdf9afa56e55b8d8c62fd883a1992f35cf45
> Cr-Commit-Position: refs/heads/master@{#316998}

TBR=jam@chromium.org,sergiyb@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=456396

Review URL: https://codereview.chromium.org/914023007

Cr-Commit-Position: refs/heads/master@{#316999}

Depends on https://codereview.chromium.org/913223007

BUG=456396
TBR=sergiyb@chromium.org

Review URL: https://codereview.chromium.org/937853002

Cr-Commit-Position: refs/heads/master@{#316998}

Revert of Re-land: base: Enable browser-wide discardable memory on Linux, CrOS and Windows. (patchset #1 id:1 of https://codereview.chromium.org/793443003/)

Reason for revert:
Caused random crashes on GPU bots with this assertion failure:
WebglConformance.conformance_ogles_GL_log_log_001_to_008:
Error installing extension 'v8/gc'.
[3060:3996:0218/193241:FATAL:discardable_shared_memory_heap.cc(205)] Check failed: spans_.find(span->start_ + span->length_ - 1) != spans_.end().

Original issue's description:
> Re-land: base: Enable browser-wide discardable memory on Linux, CrOS and Windows.
>
> This makes SHMEM implementation of discardable memory preferred
> over EMULATED implementation. This effectively makes SHMEM the
> implementation used by default on Linux, CrOS and Windows.
>
> SHMEM implementation of discardable memory gives the browser
> process control over the total amount of discardable memory used
> and allows us to enforce a global limit of 512MB across all
> renderers.
>
> BUG=429415,429416
>
> Committed: https://crrev.com/2f7a8e80b83aa581aa6de86117a692093d16c217
> Cr-Commit-Position: refs/heads/master@{#316977}

TBR=avi@chromium.org,danakj@chromium.org,reveman@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=429415,429416

Review URL: https://codereview.chromium.org/917023003

Cr-Commit-Position: refs/heads/master@{#316997}

BUG=None
TEST=None
TBR=kenobi@chromium.org

Review URL: https://codereview.chromium.org/942453003

Cr-Commit-Position: refs/heads/master@{#316996}

BUG=430635

Review URL: https://codereview.chromium.org/917213003

Cr-Commit-Position: refs/heads/master@{#316995}

Now that volumes other than Google Drive can be expanded in the directory tree,
there is a case that subdirectories are queried befor the displayroot of volume is
obtained. In this case, we should skip updating subdirectories. After the displayroot
is obtained, updateSubDirectories will be called.

BUG=459534
TEST=manually check the steps described in the issue.

Review URL: https://codereview.chromium.org/940443003

Cr-Commit-Position: refs/heads/master@{#316994}

2) Don't show details header panel once user has completed import.
3) Introduce a local storage wrapper and make all existing importer uses of chrome.storage.local use this...much shorter and safer.
4) Auto-hide details panel when user clicks out.
5) Fix jank on the panel transition (and switch to vertical swoosh).

BUG=420680
TEST=browser_test: FileManagerJsTest.*

Review URL: https://codereview.chromium.org/925233004

Cr-Commit-Position: refs/heads/master@{#316993}

Undo and redo are mapped in a similar way to cut/copy/paste on Mac, via
the mainMenu's performKeyEquivalent, and the Cocoa responder chain.

This CL adds responders for undo: and redo:

Single-character editing commands were not being merged into a single
edit. For example, typing {'c', 'a', 't', Cmd+Z} would only undo the 't'
not the entire 'cat'. This CL fixes by detecting single-character
inserts in response to a keyDown event.

Then, for "redo", some of the textfield unittests would send both Ctrl+Y
and Ctrl+Shift+Z. Cmd+Y for undo isn't supported on Mac (it's mapped to
History -> Show Full History). So a wrapper is added that just sends
Cmd+Shfit+Z again.

Gets the following tests passing:
     TextfieldTest.DragAndDrop_ToTheLeft
     TextfieldTest.DragAndDrop_ToTheRight
     TextfieldTest.UndoRedoTest

BUG=454353

Review URL: https://codereview.chromium.org/923903002

Cr-Commit-Position: refs/heads/master@{#316992}

Previously Files.app uses @typedef for expressing metadata item.
The CL introduces new class instead, for better compiler check.

BUG=410766
TEST=None

Review URL: https://codereview.chromium.org/939653002

Cr-Commit-Position: refs/heads/master@{#316991}

BUG=

Review URL: https://codereview.chromium.org/923723004

Cr-Commit-Position: refs/heads/master@{#316990}

The current calculation assumes positions[0].x() is zero, a run
may start with certain x-offset. It depends on the font settings.
In that case, the entire run might be shifted horizontally.

This is still really hard to be tested in the current code base,
the x-offset value comes from the font settings. Let me leave
the test for the further CL(s).

BUG=456692
R=msw@chromium.org, ckocagil@chromium.org

Review URL: https://codereview.chromium.org/923973003

Cr-Commit-Position: refs/heads/master@{#316989}

Merge address_input_strings_translations.grdp into
address_input_strings.grd to follow the conventions in other grd
files. The translations were originally separated to use the Java
widget from libaddressinput, which Chrome since stopped using. This
change makes translation scripts more reliable.

BUG=459791

Review URL: https://codereview.chromium.org/941503002

Cr-Commit-Position: refs/heads/master@{#316988}

This restores a string that got accidentally removed in a recent CL.

BUG=420680

Review URL: https://codereview.chromium.org/936963002

Cr-Commit-Position: refs/heads/master@{#316987}

BUG=

Review URL: https://codereview.chromium.org/932003005

Cr-Commit-Position: refs/heads/master@{#316986}

BUG=459820
TBR=zmo@chromium.org

Review URL: https://codereview.chromium.org/938793003

Cr-Commit-Position: refs/heads/master@{#316985}

When GL_VERSION is set to ES3, Skia assumes that various functions exist in its
GL function table. This CL wires up those functions so that Skia can use
gl_bindings_skia_in_process on devices that support ES3.

One of these functions, glGetStringi, was missing from the autogen script, so I
added it.

R=kbr@chromium.org

Review URL: https://codereview.chromium.org/940523003

Cr-Commit-Position: refs/heads/master@{#316984}

Adding a python script which will read in a GYP file, convert it to valid JSON, and then load it as a dictionary object.  The script will take a set of parameters which will allow the caller to target a set of remoting apps and output a space delimited string containing the target property they are interested in.  This output string will be used by other bash/python scripts in a future CL to make generating remoting apps simpler.

BUG=

Review URL: https://codereview.chromium.org/934253004

Cr-Commit-Position: refs/heads/master@{#316983}

This enables JScompile for Chromoting's browsertest code and fixes all
of the reported errors.

This cl also includes a few minor typos and jscompile inconsistencies in the non-test code.

BUG=

Review URL: https://codereview.chromium.org/927373005

Cr-Commit-Position: refs/heads/master@{#316982}

This temporarily restores the original behavior. It'd be nice to use the
more specific type, but some components of Blink are generating long
data: URLs and passing them to WriteBookmark(), which causes the browser
to kill the renderer.

BUG=447094

Review URL: https://codereview.chromium.org/934373002

Cr-Commit-Position: refs/heads/master@{#316981}

BUG=393291

Review URL: https://codereview.chromium.org/929393003

Cr-Commit-Position: refs/heads/master@{#316980}

In the GPU process, GpuMain can return, destroying the process and
thread, before ChildProcessService.onDestroy (which calls
ShutdownThread) runs, leaving g_child_thread invalid.  Set it to null on
ChildThreadImpl destructor and check for that.

BUG=458868

Review URL: https://codereview.chromium.org/913203004

Cr-Commit-Position: refs/heads/master@{#316979}

This adds a new simpler method to ResourceProvider called
CopyToResource and uses that instead. This removes TextureUploader
from the path of most impl-side painting code.
R=enne
BUG=454575

Review URL: https://codereview.chromium.org/938893002

Cr-Commit-Position: refs/heads/master@{#316978}

This makes SHMEM implementation of discardable memory preferred
over EMULATED implementation. This effectively makes SHMEM the
implementation used by default on Linux, CrOS and Windows.

SHMEM implementation of discardable memory gives the browser
process control over the total amount of discardable memory used
and allows us to enforce a global limit of 512MB across all
renderers.

BUG=429415,429416

Review URL: https://codereview.chromium.org/793443003

Cr-Commit-Position: refs/heads/master@{#316977}

BUG=459688

Review URL: https://codereview.chromium.org/940723002

Cr-Commit-Position: refs/heads/master@{#316976}

BUG=430054
R=rockot@chromium.org

Review URL: https://codereview.chromium.org/936123002

Cr-Commit-Position: refs/heads/master@{#316975}

Discover the Zygote's PID properly when using the namespace sandbox.

BUG=459724

Review URL: https://codereview.chromium.org/940603003

Cr-Commit-Position: refs/heads/master@{#316974}

Non-android bots will not have credentials to download SDK packages from Google
Storage buckets. Thus, if a bot does not have credentials, the hook should print
a warning and continue.

BUG=459681

Review URL: https://codereview.chromium.org/940633003

Cr-Commit-Position: refs/heads/master@{#316973}

BUG=459642

Review URL: https://codereview.chromium.org/933343002

Cr-Commit-Position: refs/heads/master@{#316972}

Revert of Reland #4: Ensure WebView notifies desktop automation on creation,  (patchset #1 id:1 of https://codereview.chromium.org/940763002/)

Reason for revert:
Still flakes on bots.

Original issue's description:
> Reland #4: Ensure WebView notifies desktop automation on creation, destruction, and change
>
> Previous attempts:
> https://codereview.chromium.org/880063002
> https://codereview.chromium.org/895623003/
> https://codereview.chromium.org/890013006/
> https://codereview.chromium.org/907973004/
>
> This latest attempt removes the logic that reloads ChromeVox at the beginning of a test.
>
> Locally, under extremely heavy load, tests progressively ran slowe. The same copy of a test ran 1 second slower than a previous copy. Ultimately, this led to a timeout.
>
> With the above change, this no longer occurs and hopefully fixes the flakeyness.
>
> Change is located in chromevox_e2e_test_base.js.
>
> TBR=dmazzoni@chromium.org
>
> Committed: https://crrev.com/47cb3b3798d325f3083565a05a2dcdaee117c296
> Cr-Commit-Position: refs/heads/master@{#316950}

TBR=dmazzoni@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/939853003

Cr-Commit-Position: refs/heads/master@{#316971}

BUG=408152

Review URL: https://codereview.chromium.org/932293002

Cr-Commit-Position: refs/heads/master@{#316970}

BUG=429053
TEST=gpu_unittests
R=piman@chromium.org

Review URL: https://codereview.chromium.org/936183002

Cr-Commit-Position: refs/heads/master@{#316969}

There's a bug in AppKit that causes Chrome to crash when users attempt to save
or open a file. The code path that causes the crash is only hit by applications
linked against OSX 10.7 SDK (or earlier). This CL registers a default value for
the preference NSViewKeepLayersAround, which prevents this code path from being
hit. The functional implications of changing the preference
NSViewKeepLayersAround are minimal.

BUG=428977

Review URL: https://codereview.chromium.org/941453002

Cr-Commit-Position: refs/heads/master@{#316968}

Review URL: https://codereview.chromium.org/924803003

Cr-Commit-Position: refs/heads/master@{#316967}

TBR=cpu
BUG=458571

Review URL: https://codereview.chromium.org/930243007

Cr-Commit-Position: refs/heads/master@{#316966}

This change makes tools/perf/page_sets/tough_animation_cases/resources/perftesthelper.js
consistent with the Chromium Javascript style guide:
 - Fixed indentation
 - Added missing semicolon
 - Renamed file to be underscore separated

Review URL: https://codereview.chromium.org/927583005

Cr-Commit-Position: refs/heads/master@{#316965}

This change ensures that if a QUIC proxy is unreachable,
then it is marked as bad. Allows chrome to retry the
request with the next proxy instead of showing
ERR_QUIC_PROTOCOL to user.

BUG=459426

Review URL: https://codereview.chromium.org/936243002

Cr-Commit-Position: refs/heads/master@{#316964}

BUG=458687, 403800

Review URL: https://codereview.chromium.org/929483004

Cr-Commit-Position: refs/heads/master@{#316963}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/830d38e..37a8479

TBR=yhirano@chromium.org,kbr@chromium.org

Review URL: https://codereview.chromium.org/939843002

Cr-Commit-Position: refs/heads/master@{#316962}

BUG=426953
TEST=See bug for repro steps.

Review URL: https://codereview.chromium.org/913223009

Cr-Commit-Position: refs/heads/master@{#316961}

Also deletes a bunch of comments that no longer apply and need to
be rewritten.

BUG=457413

Review URL: https://codereview.chromium.org/932263002

Cr-Commit-Position: refs/heads/master@{#316960}