GitLab

To verify that ServiceWorkerProcessManager doesn't prevent render
process hosts from shutting down.

BUG=639193,636377

Review-Url: https://codereview.chromium.org/2289003002
Cr-Commit-Position: refs/heads/master@{#415193}

This is just some clean up before I make InProcessContextFactory also
use unique ptr of a struct.

Review-Url: https://codereview.chromium.org/2297473002
Cr-Commit-Position: refs/heads/master@{#415192}

This patch converts "editing/inserting/insert-thai-characters-001.html" to
use w3c test harness to simplify test script for improving readability.

BUG=n/a
TEST=n/a; no behavior changes

Review-Url: https://codereview.chromium.org/2288703002
Cr-Commit-Position: refs/heads/master@{#415191}

I want to remove SkImage::DeferredTextureImageUsageParams' default ctor.
Before I make that change in Skia, I want to update Chrome to no longer call it.

The default ctor did not initialize all of the members, which lead to surprises.
We could simply initialize all the values to some weird state.
But a better option would be to not allow the object to be created unless it
can be in a valid state.

BUG=2242883004
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2285013003
Cr-Commit-Position: refs/heads/master@{#415190}

This DCHECK contradicts the DCHECK above it, so it's obsolete.

Review-Url: https://codereview.chromium.org/2286243002
Cr-Commit-Position: refs/heads/master@{#415189}

Recently, lint wanrnings were made fatal for Android builds:
  https://codereview.chromium.org/2241233002

A number of internal files for the Chromecast Android build have lint
errors, breaking the build at compile time. b/31140436 has been assigned
to clean up these files. In the meantime, we need to supress these
errors without leaking internal details about Cast code. Use the
following regexes where necessary to suppress errors:
  * "chromecast/internal"
  * "com/android/tv"
  * "com/google/android/apps/cast"
  * "com/google/android/apps/mediashell"

The changes introduced in this CL will be reverted as they are
cleaned up.

BUG= internal b/31140436

Review-Url: https://codereview.chromium.org/2285083003
Cr-Commit-Position: refs/heads/master@{#415188}

Without these fixes DCHECKs are triggered indicated that the
resources are missing. Likely this also means the mobile view is
not quite the same as if it was actually on mobile.

See style sheets used in LayoutThemeMobile.cpp for details:
https://cs.chromium.org/chromium/src/third_party/WebKit/Source/core/layout/LayoutThemeMobile.cpp

BUG=none
TEST=manual

Review-Url: https://codereview.chromium.org/2291883002
Cr-Commit-Position: refs/heads/master@{#415187}

Don't use SSLStatus from FrameHostMsg_DidCommitProvisionalLoad and instead cache it on the browser side.

This is along the earlier patchsets of stopping to send SSLStatus to the renderer. This will help network service (since currently it's what sends this data, and it shouldn't know about UI state) and PlzNavigate (since this is a cursor for removing cert_ids which depend on having a renderer process during navigate). After this change, there's only one use left in the renderer that (devtools).

BUG=598073, 504347
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2239273002
Cr-Commit-Position: refs/heads/master@{#415186}

...when reporting policy enforced.

R=robertshield@chromium.org
BUG=641540

Review-Url: https://codereview.chromium.org/2282143002
Cr-Commit-Position: refs/heads/master@{#415185}

BUG=none

Review-Url: https://codereview.chromium.org/2285003002
Cr-Commit-Position: refs/heads/master@{#415184}

Inlines the sources from ppapi_sources.gypi into the respective BUILD.gn files.

Previously the cpp sources were compiled into two targets, one in //ppapi:ppapi_cpp_sources (referenced only from nacl), and one in //ppapi/cpp (referenced only from trusted). This patch merges them into the cpp directory, which now morphs as required for the build type.

R=bbudge
TBR=bradnelson (chrome/test/data/nacl)

Review-Url: https://codereview.chromium.org/2289983002
Cr-Commit-Position: refs/heads/master@{#415183}

And a base/message_loop/message_loop.h include as well.
Also fix lint issues.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
TBR=aboxhall@chromium.org,asanka@chromium.org,finnur@chromium.org,jochen@chromium.org

Review-Url: https://codereview.chromium.org/2231533003
Cr-Commit-Position: refs/heads/master@{#415182}

BUG=641454

Review-Url: https://codereview.chromium.org/2290713002
Cr-Commit-Position: refs/heads/master@{#415181}

Due to issues with the MemoryCache backend QueryCache wasn't allowed to read
cache entries as it iterated them. That's been fixed, so now it's safe to
filter entries as they're iterated, which is more resource friendly.

BUG=634428

Review-Url: https://codereview.chromium.org/2281983003
Cr-Commit-Position: refs/heads/master@{#415180}

BUG=609747, 641035

Review-Url: https://codereview.chromium.org/2243863002
Cr-Commit-Position: refs/heads/master@{#415179}

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/9b6db57c0d56330b02420d5a2f771d68ddd47aa1 Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
depot_tools:
  https://crrev.com/81665dcf76d0e2cb0497e64ab23ab885e96f3bf6 Add more debugging in presubmit owners check for Gerrit. (tandrii@chromium.org)

TBR=martiniss@chromium.org,phajdan.jr@chromium.org
BUG=641384

Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Review-Url: https://codereview.chromium.org/2287423002
Cr-Commit-Position: refs/heads/master@{#415178}

This CL makes the SiteEngagementService available in incognito. When in
incognito mode, the service will initialize values from the original
profile and then subsequently maintain its own independent copy.

Incognito profiles will not record periodic metrics.

BUG=640863
TBR=dbeam@chromium.org,rdevlin.cronin@chromium.org,dtrainor@chromium.org

Review-Url: https://codereview.chromium.org/2274973003
Cr-Commit-Position: refs/heads/master@{#415177}

Based on UI reviewer's comments, the setting button on opt-in IME menu should be update as only text when there's only one button.

BUG=641300
TEST=Verified on local build.

Review-Url: https://codereview.chromium.org/2287953002
Cr-Commit-Position: refs/heads/master@{#415176}

It's possible hardware_gpu_enabled_ is true (because the GPU is
blacklisted and hasn't crashed at all) even when using Swiftshader. In
this case, the code incorrectly tried to launch the GPU process even
if it had already crashed too much.

Review-Url: https://codereview.chromium.org/2281253002
Cr-Commit-Position: refs/heads/master@{#415175}

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2288613004
Cr-Commit-Position: refs/heads/master@{#415174}

Review-Url: https://codereview.chromium.org/2287223003
Cr-Commit-Position: refs/heads/master@{#415173}

When converting arrays to keys, only own properties should be
considered, much like clones. Otherwise, this opens up the behavior to
side effects via getters on the Object/Array prototype even if the
value being converted to a key is a clone.

BUG=641101
R=haraken@chromium.org

Review-Url: https://codereview.chromium.org/2279343002
Cr-Commit-Position: refs/heads/master@{#415172}

The image is converted into an SkBitmap, which has no concept of retina-ness. In
the previous behavior, Chrome would maintain the image size but lose half the
pixels. In the new behavior, Chrome keeps all the pixels and doubles the size of
the SkBitmap.

BUG=355477

Review-Url: https://codereview.chromium.org/2276183002
Cr-Commit-Position: refs/heads/master@{#415171}

SyncExtensionFunctions, AsyncExtensionFunctions,
ChromeSyncExtensionFunctions, and ChromeAsyncExtensionFunctions are
deprecated.

Remove SyncExtensionFunctions from:
- easy_unlock_private
- input_api

Note: input_api had some tests which were incorrectly passing,
even though an error was being thrown. See crbug.com/641425

BUG=634140

Review-Url: https://codereview.chromium.org/2285573002
Cr-Commit-Position: refs/heads/master@{#415170}

Chrome browser compositor sends the primary buffer as a
normal overlay, therefore the optimization to replace the
primary plane can be done earlier in the browser compositor (crrev.com/2281033003).

This CL removes the optimization in Ozone.

BUG=b/29430506

Review-Url: https://codereview.chromium.org/2284893002
Cr-Commit-Position: refs/heads/master@{#415169}

BUG=642057

Review-Url: https://codereview.chromium.org/2285373002
Cr-Commit-Position: refs/heads/master@{#415168}

BUG=641934
TBR=petrcermak@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2293553002
Cr-Commit-Position: refs/heads/master@{#415167}

Revert of Reland [Presubmit tests] Warn when new source files are not added to GN/GYPi files. (patchset #1 id:1 of https://codereview.chromium.org/2278093004/ )

Reason for revert:
This causes warning if files are referenced in gyp file. https://codereview.chromium.org/2285703003/ triggered presubmit when it shouln't have.

Original issue's description:
> Reland [Presubmit tests] Warn when new source files are not added to GN/GYPi files.
>
> Add a presubmit check that warns when new source files are
> not added to GN or gypi files. Check is done on upload only before gni includes are fixed.
>
> BUG=112371
>
> Committed: https://crrev.com/1efc2f5f650a355056a98b6c99d4d0260742a750
> Cr-Commit-Position: refs/heads/master@{#414975}

TBR=dpranke@chromium.org,maksim.sisov@intel.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=112371

Review-Url: https://codereview.chromium.org/2293763002
Cr-Commit-Position: refs/heads/master@{#415166}

Use for-range style loops and get rid of the local 'size' variable.
Also move the sort() call down after the early-out when there are no
animations to apply.

BUG=641437

Review-Url: https://codereview.chromium.org/2287973002
Cr-Commit-Position: refs/heads/master@{#415165}

new uses of the flag and thus reduce the cost when it's finally removed.

BUG=516499
R=mmenke@chromium.org
TBR=megjablon@chromium.org
TBR=pkasting@chromium.org
TBR=droger@chromium.org

Review-Url: https://codereview.chromium.org/2278823003
Cr-Commit-Position: refs/heads/master@{#415164}

This will eliminate some wasted cycles during the rename dance for the
majority of installs, thereby leading to a better update experience for
users.

Also included in this change: removed some duplicated code and fixed up
use of the registry. More on the latter: all integration points between
Chrome and Google Update unconditionally take place in the 32-bit view
of the Windows registry. As such, we use KEY_WOW64_32KEY everywhere even
when it isn't strictly needed. It seems that this spot was missed during
the great 64-bit-ification of Chrome. It just so happens that it's
harmless for per-user installs since HKCU\Software is not redirected for
64-bit apps in the way that HKLM\Software is.

BUG=610305
R=brucedawson@chromium.org

Review-Url: https://codereview.chromium.org/2283363005
Cr-Commit-Position: refs/heads/master@{#415163}

https://chromium.googlesource.com/skia.git/+log/839d3ad854c0..6f5df6acb7de

$ git log 839d3ad85..6f5df6acb --date=short --no-merges --format='%ad %ae %s'
2016-08-29 mtklein Disable sktexttopdf when building for mipsel.
2016-08-29 mtklein GN: fontmgr_android
2016-08-29 msarett Skip spinlock overhead on Android
2016-08-29 msarett Add Gray support to SkPNGImageEncoder
2016-08-29 mtklein remove dead code
2016-08-29 jvanverth Fix stroked rrects with batching.
2016-08-29 egdaniel Remove unneeded pipeline barrier in vulkan
2016-08-29 jcgregorio benchmarks: Avoid brackets and commas in test names.
2016-08-29 mtklein GN: ft and fc as optionals
2016-08-29 mtklein add gn_android_flavor
2016-08-29 bsalomon Skip batch iteration in GrDrawTarget::forwardCombine when the max look ahead is 0
2016-08-29 mtklein Add docs for GN/Android
2016-08-29 mtklein GN: release -> -O3, to match GYP
2016-08-29 vjiaoblack Moved ambient lights out of SkLight's light array
2016-08-29 jvanverth Batch all circular rrects together
2016-08-29 halcanary SkPDF: Remove some unused glyphs from fallback fonts
2016-08-29 bsalomon Add support for glDrawRangeElements GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2287003002
2016-08-29 msarett Add width, height, unique ID, alpha to debugger DrawImageCommand
2016-08-29 vjiaoblack Revert of Moved ambient lights out of SkLight's light array (patchset #7 id:120001 of https://codereview.chromium.org/2287553002/ )
2016-08-29 rmistry Add Gerrit API support to Skia's PRESUBMIT.py
2016-08-29 reed dedup images/blobs/pictures by ID in old serialization format
2016-08-29 rmistry Whitespace change to test Gerrit Submit
2016-08-29 jvanverth Use stroked rrects for Android shadow sample
2016-08-29 mtklein Roll GN
2016-08-29 vjiaoblack Moved ambient lights out of SkLight's light array
2016-08-29 reed change readTypeface to return sk_sp
2016-08-29 brianosman Remove FP16 support from ANGLE. Previous workaround was incomplete.
2016-08-29 fmalita [SVGDom] Don't truncate opacity scalars

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
TBR=bsalomon@google.com

Review-Url: https://codereview.chromium.org/2290013003
Cr-Commit-Position: refs/heads/master@{#415162}

Have the NGConstraintSpace constructors take a NGPhysicalConstraintSpace
and make it mandatory. Also add accessors to NGConstraintSpace and allow
a new NGPhysicalConstraintSpace to be constructed from a NGPhysicalSize.

R=ikilpatrick@chromium.org,cbiesinger@chromium.org
BUG=635619
NOTRY=true

Review-Url: https://codereview.chromium.org/2283233002
Cr-Commit-Position: refs/heads/master@{#415161}

BUG=641597

Review-Url: https://codereview.chromium.org/2284143002
Cr-Commit-Position: refs/heads/master@{#415160}

This way GpuProcessHost::RecordProcessCrash will be called, allowing
Chrome to fall back to SwiftShader or just give up on launching the GPU
after a little while.

Review-Url: https://codereview.chromium.org/2281243002
Cr-Commit-Position: refs/heads/master@{#415159}

Since Chrome can do window reparenting, DisplayMetrics should
be extracted from the Context that is associated with the window.

In this CL we pass the context to RenderCoordinates that obtains
the display density (device scale factor) and wheel scroll factor
from it.

BUG=620929

Review-Url: https://codereview.chromium.org/2268103002
Cr-Commit-Position: refs/heads/master@{#415158}

first sent packet after receiving a packet, even if the sent packet is
not a retransmission. Protected by default enabled
--quic_better_last_send_for_timeout

Merge internal change: 131623416

BUG=642017

Review-Url: https://codereview.chromium.org/2288843002
Cr-Commit-Position: refs/heads/master@{#415157}

Textfield::SetTextEditCommandForNextKeyEvent can be used to schedule a text
editing command to be executed on the next key press. However, Textfield
subclasses like OmniboxViewViews are currently oblivious to the scheduled text
edit command accounting. When a key press event is handled in
OmnboxViewViews::OnKeyPressed itself, the scheduled_text_edit_command_ property
in Textfield is not cleared, leading to a failed DCHECK when
Textfield::SetTextEditCommandForNextKeyEvent is invoked next.

This CL resolves the crash by prohibiting Textfield subclasses from overriding
OnKeyPressed/Released. This allows Textfield to handle the accounting related to
the scheduled text edit command. This necessiates moving all the
OmniboxViewViews::OnKeyPressed and OmniboxViewViews::OnKeyReleased logic to
OmniboxViewViews::HandleKeyEvent. A test is also added which fails on the
current master.

BUG=613948
TEST=Build with mac_views_browser=1. In the browser, go to the omnibox and type
a partial url such that there are atleast 2 autocomplete suggestions. Press
arrow down key twice. Ensure the browser does not crash.

Review-Url: https://codereview.chromium.org/2273263002
Cr-Commit-Position: refs/heads/master@{#415156}

The pin keyboard now hides its input element and passes its data to the user pod password element.

BUG=632383
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2260653002
Cr-Commit-Position: refs/heads/master@{#415155}

Update them so that they return strings instead of calling Node::showNode(),
which uses WTFLogAlways().

- Rename showTreeForThis() to toTreeStringForThis()
- Rename showTreeForThisInFlatTree() to toFlatTreeStringForThis()
- Rename showTreeAndMark() to toMarkedTreeString().
- Rename showTreeAndMarkInFlatTree() to toMarkedFlatTreeString()
- Rename traverseTreeAndMark() to appendMarkedTree()
- Rename traverseTreeAndMarkInFlatTree() to appendMarkedFlatTree()
- Add Node::toString() to help these functions.

We'd like to return a string rather than taking a std::ostream argument because
they are used with LOG macros. For example,
  VLOG(1) << node.toMarkedTreeString().utf8().data() << "...";

If we took a std::ostream argument, the code would be:
  std::stringstream buffer;
  node.printMarkedTreeTo(buffer);
  VLOG(1) << buffer.str() << "...";

The former is simpler.

This CL doesn't affect production.

BUG=638849

Review-Url: https://codereview.chromium.org/2293703002
Cr-Commit-Position: refs/heads/master@{#415154}