GitLab

The shader cache code is going to move into //gpu/ipc/host. However, the
singleton is content only. So factor out the code that deals with the
singleton instance inside content.

BUG=604369

Review-Url: https://codereview.chromium.org/2565223003
Cr-Commit-Position: refs/heads/master@{#438079}

This also includes DevToolsSession::Dispatch utility method
which offers message to DevToolsManagerDelegate.

BUG=664683
TBR=pfeldman

Review-Url: https://codereview.chromium.org/2548263002
Cr-Commit-Position: refs/heads/master@{#438078}

I've root caused this and the problem was introduced in this CL:
https://codereview.chromium.org/2446053002

That CL moved the logic which reported the worker process launch from the
IPC Channel Connected method to an IO completion port listener.  This works
fine on un-official builds, however it breaks on official builds.  The
reason for the failure is that the remoting_desktop process on official
builds are signed and include 'UiAccess' in their manifest.  When the
launcher process uses ShellExecuteEx to launch the worker process in this
scenario, we actually see two processes get launched sequentially.  The
first starts and exists with error code 'STATUS_ELEVATION_REQUIRED' and
the second launches with the correct permissions.  This behavior worked
fine before as we listened for the connection to the IPC channel which
was done by the second, successful process launch.  With the new code,
we observed the first process launch, set up the Mojo channel for it, and
tried waiting on its process handle which exits immediately.  The second
process then starts and fails to connect to the Mojo channel.

I investigated whether UiAccess is truly required for the desktop binary
and I think that it is.  For the Ctrl+Alt+Del scenario, there are registry
keys that can be set which will require that flag.  For Alt+Tab, it is
possible that some windows might not be accessible if they have a high
high enough integrity level (+ UiAccess themselves).

So instead of removing the UiAccess flag, my approach is to listen for the
worker process creation and exit events.  I store the value of the last seen
worker process id and use that in our process launch detection code once the
launcher process exits.  This allows both un-official builds (which do not
require the extra permissions hop) and official builds will work consistently.

BUG=666992

Review-Url: https://codereview.chromium.org/2568983004
Cr-Commit-Position: refs/heads/master@{#438077}

Refactoring only, no observable change intended.

A test is added that fails if the continue statement were dropped, as
apparently there was no coverage for this.

Note: This CL is also a test for the WPT export process.

BUG=402376

Review-Url: https://codereview.chromium.org/2565183002
Cr-Commit-Position: refs/heads/master@{#438076}

Revert of Restrict transmission of external exp ids to signed in users. (patchset #6 id:140001 of https://codereview.chromium.org/2558913003/ )

Reason for revert:
Speculative revert for recent flakiness in
Contextual Search related test failures on
"Android Tests" bot:

https://build.chromium.org/p/chromium.linux/builders/Android%20Tests?numbuilds=100

e.g. https://build.chromium.org/p/chromium.linux/builders/Android%20Tests/builds/35779

I'm not very confident that this is the real
offender; if it's not, I will reland this CL.

Original issue's description:
> Restrict transmission of external exp ids to signed in users.
>
> Since external experiment ids are not based on Chrome's low
> entropy source, they do not have the same guarantees about
> not identifying a user as Chrome's variations. As such, we
> should only transmit them for signed in users, whose identity
> is already known by Google so there's no risk of identifying
> them through these headers.
>
> Note: The signed-in state checking in this CL is only done for
> web content area requests and not other internal requests,
> like to the suggestion service, where it treats the state as
> "not signed in". This is fine to do because variations service
> ids are still sent, which is what the other call sites are
> interested in.
>
> BUG=672532
> TBR=mpearson@chromium.org,mattm@chromium.org,donnd@chromium.org,afakhry@chromium.org
>
> Committed: https://crrev.com/9ed7b5611a61505c3dba734fe55b92211df3c2f6
> Cr-Commit-Position: refs/heads/master@{#437959}

TBR=jwd@chromium.org,mathp@chromium.org,treib@chromium.org,mmenke@chromium.org,mpearson@chromium.org,mattm@chromium.org,donnd@chromium.org,afakhry@chromium.org,asvitkine@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=672532

Review-Url: https://codereview.chromium.org/2569973002
Cr-Commit-Position: refs/heads/master@{#438075}

Revert of Add ubsan_no_recover_hack to missing ubsan builders. (patchset #1 id:1 of https://codereview.chromium.org/2538103002/ )

Reason for revert:
UBSan is noisy and can't be enabled for explicit crash behavior on builders. On ClusterFuzz, it will be still enabled to create testcases.

BUG=673488

Original issue's description:
> Add ubsan_no_recover_hack to missing ubsan builders.
>
> R=inferno@chromium.org, ochang@chromium.org, phajdan.jr@chromium.org
> BUG=669566
>
> Committed: https://crrev.com/a50c295543d07efaedd8534d117f3860a3df52e8
> Cr-Commit-Position: refs/heads/master@{#435317}

TBR=ochang@chromium.org,phajdan.jr@chromium.org,mmoroz@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=669566

Review-Url: https://codereview.chromium.org/2568883003
Cr-Commit-Position: refs/heads/master@{#438074}

This CL fixes some colors in MD History that were low contrast. It also
makes a bunch of colors consistent and pulls most colors into
shared_vars.html so we can better manage our vast color spectrum.

BUG=670836
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2549133003
Cr-Commit-Position: refs/heads/master@{#438073}

Changed EOverflowAnchor to an enum class and gave it an unsigned
underlying type.

Changing it to an enum class enforces better namespacing and code
practices. Adding the unsigned underlying type is pre-work for when the
class is eventually stored as an enum bitfield (it would be done in this
patch, except a presubmit warning already exists that prevents that. The
presubmit warning needs to be updated before that change can occur.)

This is also pre-work to move EOverflowAnchor to be generated in
ComputedStyleBase.

BUG=628043
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2561973002
Cr-Commit-Position: refs/heads/master@{#438072}

This patch marks a few tests that are being flaky as so.
Will try to find owners for these tests.

BUG=673632

Review-Url: https://codereview.chromium.org/2565233005
Cr-Commit-Position: refs/heads/master@{#438071}

This removes a bunch of calls which end up with a WebString pointing to
the global empty string, which was showing up in profiles.

BUG=348655

Review-Url: https://codereview.chromium.org/2555983003
Cr-Commit-Position: refs/heads/master@{#438070}

Flip quic flags by 12/12/2016 18:30 PST
https://codereview.chromium.org/2569703003/

relnote: Protected by --quic_fix_beta_last_max. Fix conservative decrease for n-connection emulation.
Merge internal change: 141565085
https://codereview.chromium.org/2569723004

relnote: Allow changing initial flow control window via connection options. Protected by --quic_large_ifw_options. Partially reverts internal change: 128225749
Merge internal change: 141507430
https://codereview.chromium.org/2569113002

rel-note: automated rollback of internal change 138108562
Merge internal change: 141469809
https://codereview.chromium.org/2566943003

relnote: Use event_time instead of ApproximateNow() in QUIC cubic code. Protected by --quic_use_event_time
Merge internal change: 141384938
https://codereview.chromium.org/2571643002

relnote: Remove unused QuicClock member of BbrSender.
Merge internal change: 141366593
https://codereview.chromium.org/2566333002

relnote: Non-functional change. Cleanup pg3 support for quic_packet_reader
Bring it more inline with Chromium source but cleanup the definition of
MMSG_MORE macro to exist in the header so that there's no violation of
ODR for QUicPacketReader class based upon inclusion.
Merge internal change: 141357226
https://codereview.chromium.org/2570703002

relnote: Protected by --quic_fix_cubic_bytes_quantization. Fix CubicBytes Quantization. iDue to an unfortunate ordering of operations in the actual "cubing" part of the CWND calculation, we were inadvertantly quantizing the congestion-window calculation.  (Note: The "reno" part of the cubic algorithm wis smooth and not quantized before.)  Switching the order of operations obviates the problem.
Additionally, removed parameterization from cubic_test.cc and
cubic_bytes_test.cc, because the parameterization was getting
unwieldy.  Instead, the parameters are controlled by command-line
flag.
Merge internal change: 141183279
https://codereview.chromium.org/2569873002/

BUG=

Review-Url: https://codereview.chromium.org/2566403002
Cr-Commit-Position: refs/heads/master@{#438069}

Add StyleDifference::needsPaintPropertyUpdate() to indicate a style
change directly needs paint property update, and set the flag on
changes of css clip, transform, etc.

Move two tests from their original directory into paint/invalidation.

BUG=645667
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2570463004
Cr-Commit-Position: refs/heads/master@{#438068}

BUG=665948

Review-Url: https://codereview.chromium.org/2563883002
Cr-Commit-Position: refs/heads/master@{#438067}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/c69690acc34b..e6e0862c8165

$ git log c69690acc..e6e0862c8 --date=short --no-merges --format='%ad %ae %s'
2016-12-12 eakuefner [Telemetry] Increase timeout after sending SIGTERM

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=catapult-sheriff@chromium.org

Review-Url: https://codereview.chromium.org/2571713003
Cr-Commit-Position: refs/heads/master@{#438066}

SSLClientCertificateSelectorCocoaTest.HideShow is failing on macOS Sierra,
blocking 10.12 rollout for Chrome. Fix this using the same refactor as seen in
r400627, which ignores mouse events instead of resizing the certificate selector
window when it's hidden.

BUG=650898

Review-Url: https://codereview.chromium.org/2559763003
Cr-Commit-Position: refs/heads/master@{#438065}

Revert of ash: Use system tray mojo interface to show system update tray icon (patchset #6 id:100001 of https://codereview.chromium.org/2558043006/ )

Reason for revert:
Broke compile on Google Chrome ChromeOS:

https://build.chromium.org/p/chromium.chrome/builders/Google%20Chrome%20ChromeOS/builds/26896

../../chrome/browser/component_updater/pepper_flash_component_installer.cc:94:3: error: no type named 'SystemTrayClient' in namespace 'chromeos'; did you mean simply 'SystemTrayClient'?
  chromeos::SystemTrayClient* tray = chromeos::SystemTrayClient::Get();
  ^~~~~~~~~~~~~~~~~~~~~~~~~~
  SystemTrayClient
../../chrome/browser/ui/ash/system_tray_client.h:27:7: note: 'SystemTrayClient' declared here
class SystemTrayClient : public ash::mojom::SystemTrayClient,
      ^
../../chrome/browser/component_updater/pepper_flash_component_installer.cc:94:38: error: no member named 'SystemTrayClient' in namespace 'chromeos'; did you mean simply 'SystemTrayClient'?
  chromeos::SystemTrayClient* tray = chromeos::SystemTrayClient::Get();
                                     ^~~~~~~~~~~~~~~~~~~~~~~~~~
                                     SystemTrayClient
../../chrome/browser/ui/ash/system_tray_client.h:27:7: note: 'SystemTrayClient' declared here
class SystemTrayClient : public ash::mojom::SystemTrayClient,
      ^
2 errors generated.

Original issue's description:
> ash: Use system tray mojo interface to show system update tray icon
>
> Change the flow so that Chrome explicitly asks ash to show the icon
> rather than having ash ask Chrome whether there is an update available.
>
> * Add ShowUpdateIcon to system_tray.mojom
> * Introduce update.mojom
> * Migrate ash to using ash::mojom::UpdateSeverity internally
> * Migrate update methods from SystemTrayDelegate to SystemTrayClient
> * Add a new SystemTrayClientTest and move existing Flash test there
>
> Also add docs for SystemTrayItem (used by TrayUpdate) since tray view
> vs. default view vs. detailed view always confuses me.
>
> BUG=647412
> TEST=chrome browser_tests, ash_unittests
>
> Committed: https://crrev.com/196ee3b66b3700733cecb8472d63d872d99e3783
> Cr-Commit-Position: refs/heads/master@{#438051}

TBR=msw@chromium.org,kerrnel@chromium.org,tsepez@chromium.org,waffles@chromium.org,sky@chromium.org,jamescook@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=647412

Review-Url: https://codereview.chromium.org/2568413002
Cr-Commit-Position: refs/heads/master@{#438064}

This CL introduces two changes in the restriction of FetchEvent.respondWith().

1. Allow responding to non-navigation requests which redirect mode is 'manual'
   with opaque-redirect responses.
   Ex:
     SW: self.onfetch = evt => { evt.respondWith(fetch(evt.request)); };
     Page: fetch(new Request("/redirect-url", {redirect: 'manual'}));
     Server: Returns a redirect response to somewhere.
     Before this CL: fetch() fails.
     After this CL: fetch() returns the opaque-redirect response.

2. Add a deprecation warning for responding to requests which redirect mode is
   not 'follow' with redirected responses. Not to suddenly break existing sites
   we allow responding to navigation requests with redirected responses and show
   two warning messages. One in the DevTools attached to the service worker from
   RespondWithObserver::responseWasFulfilled(). And one in the DevTools attached
   to the page tab from DocumentLoader::finishedLoading().

BUG=658249

Review-Url: https://codereview.chromium.org/2526343003
Cr-Commit-Position: refs/heads/master@{#438063}

That way, we don't have to consider the same RuleSet multiple times for
invalidation on active stylesheet update. This fixes a regression in
PerformanceTests/CSS/StyleSheetInsert.html which would have been
introduced by https://codereview.chromium.org/2557533005

This works because the same style element source text used multiple
times will make us use the same StyleSheetContents from the cache and
hence the same RuleSet for all 50 sheets added in that test. It's a bit
like cheating, but this will also make sure we don't invalidate for the
same RuleSet twice if we re-order stylesheets by removing/inserting a
style element where the CSSStyleSheet pointer will be different, but
the RuleSet stays the same.

R=sashab@chromium.org,esprehn@chromium.org
BUG=567021

Review-Url: https://codereview.chromium.org/2569733003
Cr-Commit-Position: refs/heads/master@{#438062}

ServiceWorker: Replace RegistrationCallback and GetRegistrationsCallback with CallbackPromiseAdapter

This is a cleanup CL and does not change behavior.

BUG=n/a

Review-Url: https://codereview.chromium.org/2564363002
Cr-Commit-Position: refs/heads/master@{#438061}

BUG=670180

Review-Url: https://codereview.chromium.org/2570503003
Cr-Commit-Position: refs/heads/master@{#438060}

R=yhirano@chromium.org
BUG=none

Review-Url: https://codereview.chromium.org/2572493002
Cr-Commit-Position: refs/heads/master@{#438059}

This CL helps rewrite_to_chrome_style tool, by preventing the tool from
suggesting the following conflicting renames:
- |s_field| -> |field_|
- |m_field| -> |field_|

BUG=641004

Review-Url: https://codereview.chromium.org/2569433002
Cr-Commit-Position: refs/heads/master@{#438058}

Added the CSSPropertyAPI for a single method parseSingleValue. Made
CSSPaddingProperties which implements this API for a few css
properties. Added CSS property descriptor which uses a struct to store
function pointers for the property's functions.

BUG=668012

Review-Url: https://codereview.chromium.org/2533673002
Cr-Commit-Position: refs/heads/master@{#438057}

BUG=570441

Review-Url: https://codereview.chromium.org/2570553003
Cr-Commit-Position: refs/heads/master@{#438056}

If the observer is tracking intersections with the top-level frame
viewport (AKA the "implicit root"), it's not necessary to store a
pointer to the root node on the IntersectionObserver.  Doing so
makes bookkeeping more difficult when DOM modifications happen.
It's also strictly incompatible with OOPIF.

Methods like mapToVisualRectInAncestorSpace and localToAbsolute already
set a precedent by allowing a nullptr "ancestor" argument to indicate
"map all the way to the top frame".  Changing IntersectionObserver to
let nullptr signify the implict root keeps with this convention.

The new trackingDocument() method serves two purposes: it fixes a bug
where only IntersectionObservers registered on the top-level FrameView
run the observation steps (test added); and it is a necessary wart
until OOPIF support lands (https://codereview.chromium.org/2431473003).

Because IntersectionObserver::m_root is a weak pointer, we must track
explicitly whether the IO is tracking the implicit root, to distinguish
that from an IO whose root has been garbage collected; hence the
new m_rootIsImplicit field.

R=ojan@chromium.org,kenrb@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/2553343004
Cr-Commit-Position: refs/heads/master@{#438055}

WebSecurityOrigin forwards temporaries to Origin. We should just use
these, rather than consider them opaque StringPieces and copying data
out of them.

This CL also removes some dead code.

BUG=672877

Review-Url: https://codereview.chromium.org/2561363002
Cr-Commit-Position: refs/heads/master@{#438054}

BUG=673535
TBR=ben@chromium.org

Review-Url: https://codereview.chromium.org/2566353002
Cr-Commit-Position: refs/heads/master@{#438053}

BUG=665160

Review-Url: https://codereview.chromium.org/2550883003
Cr-Commit-Position: refs/heads/master@{#438052}

Change the flow so that Chrome explicitly asks ash to show the icon
rather than having ash ask Chrome whether there is an update available.

* Add ShowUpdateIcon to system_tray.mojom
* Introduce update.mojom
* Migrate ash to using ash::mojom::UpdateSeverity internally
* Migrate update methods from SystemTrayDelegate to SystemTrayClient
* Add a new SystemTrayClientTest and move existing Flash test there

Also add docs for SystemTrayItem (used by TrayUpdate) since tray view
vs. default view vs. detailed view always confuses me.

BUG=647412
TEST=chrome browser_tests, ash_unittests

Review-Url: https://codereview.chromium.org/2558043006
Cr-Commit-Position: refs/heads/master@{#438051}

Cr-Commit-Position: refs/heads/master@{#438050}

HttpResponseHeaders::Get{MimeType,Charset} calls GetMimeTypeAndCharset internally
and discards the other.
As we are going to use both in parseMultipartHeadersFromBody, we should call it directly.

BUG=None

Review-Url: https://codereview.chromium.org/2568343002
Cr-Commit-Position: refs/heads/master@{#438049}

BUG=none
R=tapted@chromium.org

Review-Url: https://codereview.chromium.org/2566513003
Cr-Commit-Position: refs/heads/master@{#438048}

This reverts commit 140c9756.

Reason for revert:
Breaks EarlGrey tests in Release configuration for iOS downstream.

BUG=673495
NOTRY=true

Review-Url: https://codereview.chromium.org/2564403003
Cr-Commit-Position: refs/heads/master@{#438047}

Instead of a MediaDelegate returned by the ShellDelegate, we now have a
MediaController on the ash side, which receives a MediaClient from the
chrome side at startup.

This also makes minor changes to the interface to send information with
a notification instead of having ash try to read the information back
after notification.

BUG=647409

Review-Url: https://codereview.chromium.org/2563643003
Cr-Commit-Position: refs/heads/master@{#438046}

We can keep this logic entirely in AwContents, it doesn't need to be in
ContentViewCore. I'm using the existing GestureStateListener interface to signal
AwZoomControls of the relevant events, but I did have to add a few more events
to that interface. Hopefully they're OK and will work with our future plans to
refactor more input-handling stuff out of ContentViewCore.

Note on testing this: the Android WebView shell works for testing but I had to
knock out the "&& mAwContents.getSettings().shouldDisplayZoomControls()" check
in getZoomController() to enable the zoom controls. Once you do this they appear
when scrolling.

BUG=664339

Review-Url: https://codereview.chromium.org/2561463004
Cr-Commit-Position: refs/heads/master@{#438045}

Do not let ash connect to chrome to get the presenter.

Expose mojom::AppList interface in ash to set the presenter.
Chrome's AppListPresenterService connects to register.
Move instance from Chrome's FactoryImpl to AppListServiceAsh.

Implement app_list::AppList; owned by ash::WmShell.

Add export info to app_list's 'presenter' component.
(lets it add the mojom as a public DEPS; thanks Yuzhu!)

Remove connection and interface pointer from AppListPresenterMus.

TODO: What's not TODO here? It's a mess!

BUG=670775
TEST=No regressions in chrome --mash app list behavior
R=jamescook@chromium.org,tsepez@chromium.org,xiyuan@chromium.org,rockot@chromium.org

Review-Url: https://codereview.chromium.org/2567523002
Cr-Commit-Position: refs/heads/master@{#438044}

Rather than running callbacks immediately, post a task to run the
callbacks, to avoid conflicts while undeferring a page.

R=dcheng@chromium.org
BUG=670092

Review-Url: https://codereview.chromium.org/2565863002
Cr-Commit-Position: refs/heads/master@{#438043}

Revert of android: Randomize initial child service bind order (patchset #3 id:40001 of https://codereview.chromium.org/2557273004/ )

Reason for revert:
Android keeps the intent for a long time, so
randomize doesn't help at all. More details:
crbug.com/664341#c91

Original issue's description:
> android: Randomize initial child service bind order
>
> Android restarts crashed services with the previous intent.
> If browser crashes and is restarted at the same time, then
> it's possible to bind to the restarted service with stale
> intent and bundle data. See crbug.com/664341#c83 for an
> example. Randomizing the start up order is a pure
> workaround to make this case less likely.
>
> BUG=664341
>
> Committed: https://crrev.com/bb725fed9bd8c672178c8a815f4b1a2500eef444
> Cr-Commit-Position: refs/heads/master@{#437367}

TBR=agrieve@chromium.org,jcivelli@chromium.org,mariakhomenko@chromium.org,hanxi@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=664341

Review-Url: https://codereview.chromium.org/2571723002
Cr-Commit-Position: refs/heads/master@{#438042}

Running this command will throw an error:
  gn gen out/Release --check --args="is_debug=false enable_webrtc=false"

Fix this error by adding a nogncheck comment by the errant include.

BUG=

Review-Url: https://codereview.chromium.org/2563773003
Cr-Commit-Position: refs/heads/master@{#438041}

Its not being used, except from unit tests.

BUG=644430

Review-Url: https://codereview.chromium.org/2559283003
Cr-Commit-Position: refs/heads/master@{#438040}