GitLab

Suspicious root cause of the last revert was fixed. See crbug.com/670638.
Description from https://crrev.com/dc8303234f47ccfc841185d33c45e4852776b65e:

This CL introduces adaptive webfont display behavior, reducing unnecessary Flash
of Unstyled Text (FOUT) if webfont is already available in disk cache.  This is
available as a experimental Web Platform feature.

In webfont display, fallback font will be used if webfont loading time exceeds
certain timeout values in slow network. However it's observed that disk cache
RTT may also hit the timeout regardless of network speed. In such cases we would
like to block display until disk cache misses and actual network request is
being sent.

BUG=570205
TBR=isherman@chromium.org

Review-Url: https://codereview.chromium.org/2556683003
Cr-Commit-Position: refs/heads/master@{#437522}

As discussed in https://codereview.chromium.org/2556743002/,
formatting is strange "because clang-format tries to align
columns, even if there's only one row with aligned columns.
I agree that this looks weird, and it sounds like it will be
fixed."

Removing the "// clang-format off" section as it has a cost
since it requires developer to manually code format the region.

BUG=None

Review-Url: https://codereview.chromium.org/2563593004
Cr-Commit-Position: refs/heads/master@{#437521}

This intervention only changes code in CC, so We do not need this flag in Webkit,
removing the flag from Webkit's WebRuntimeFeatures.

BUG=595327

Review-Url: https://codereview.chromium.org/2559283002
Cr-Commit-Position: refs/heads/master@{#437520}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/72905fa4..722a4d56

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.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

TBR=hablich@chromium.org,machenbach@chromium.org,littledan@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/2564803002
Cr-Commit-Position: refs/heads/master@{#437519}

This only checks the request priority. Next: attempt to reliably intercept cases
when request priority gets upgraded.

Changes prerender::test_utils::CreatePrefetchOnlyInterceptorOnIO(...) to a more
generic prerender::test_utils::InterceptRequestAndCount(..., callback).

BUG=none

Review-Url: https://codereview.chromium.org/2540203002
Cr-Commit-Position: refs/heads/master@{#437518}

Now, NavigationController has multiple Reload*() methods to trigger
various kinds of reload operations. But this is confusing because it
takes one boolean argument, and in some contexts, one boolean argument
for reload means a flag to bypass caches.
Also, using one method and calling it with ReloadType flag is more
consistent with other navigation related implementation in Blink.

This is the first patch to introduce new Reload method with ReloadType
argument. Once this patch is submitted, other changes will follow to
call the new Reload from several places, and remove Reload* methods.

BUG=670232
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
TBR=jam@chromium.org

Review-Url: https://codereview.chromium.org/2552663002
Cr-Commit-Position: refs/heads/master@{#437517}

INPUT element: Do not use identical ID name for SearchFieldCancelButtonElement and ClearButtonElement.

Using identical ID for different Element subclasses is dangerous.
This CL avoids a runtime CHECK failure.

BUG=669847

Review-Url: https://codereview.chromium.org/2568443002
Cr-Commit-Position: refs/heads/master@{#437516}

BUG=668997

Review-Url: https://codereview.chromium.org/2556543004
Cr-Commit-Position: refs/heads/master@{#437515}

This patch has no functional change, but will allow subsequent patches
to specify a different target color space depending on their application
(e.g, canvas and WebGL can request sRGB, and raster can request
the output device behavior).

BUG=667420

Review-Url: https://codereview.chromium.org/2556973002
Cr-Commit-Position: refs/heads/master@{#437514}

Chrome iOS is moving to Objective-C Automatic Reference Counting. Under
it, arguments are supposed to be autoreleased. This CL adds a new
constructor that can take both a retained and an autoreleased object.
This is a part of a multiple CL setup that will change the Image(UIImage*) constructor from consuming +1 refcount to retaining the argument.

TEST=none
BUG=645125

Review-Url: https://codereview.chromium.org/2504683002
Cr-Commit-Position: refs/heads/master@{#437513}

According to the base/files/file.h, the GetLength() method should return
a negative number on failure.
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.win:win10_chromium_x64_rel_ng

Review-Url: https://codereview.chromium.org/2404823002
Cr-Commit-Position: refs/heads/master@{#437512}

BUG=chromium:671511
TEST=trybot

Review-Url: https://codereview.chromium.org/2559643002
Cr-Commit-Position: refs/heads/master@{#437511}

We used CHECKs to confirm if these checks are valid and can survive
on Canary. Now that we know these are correct, let's make it back
to DCHECKs.

BUG=670638

Review-Url: https://codereview.chromium.org/2558433002
Cr-Commit-Position: refs/heads/master@{#437510}

algorithms to the coordinator rather than run them directly.

(This is step 5 of https://docs.google.com/document/d/1vUmyaZoIqnVyvqAf5lFhnh_BdHYknLqViLISZiy-Iek/edit#)

BUG=636993

Review-Url: https://codereview.chromium.org/2519263002
Cr-Commit-Position: refs/heads/master@{#437509}

This patch adjusts the algorithm of toIntegralType(), which parses an integer
value out of a string. Previously, toIntegralType() used a positive number
as an accumulator regardless of whether we had seen the negative sign or not,
and it inverted the accumulator value if there had been a negative sign. This
causes an integer overflow when INT_MIN is specified as the source string,
because the absolute value of INT_MIN is one larger than that of INT_MAX.

This patch fixes this issue by accumulating a negative value if the result
will be negative. Also, the detection algorithm of overflows is updated so
that it can detect overflows correctly in every possible case (it used to have
a small bug that the function fails to parse INT_MIN in base 16).

Additionally, unit tests are added.

BUG=665110

Review-Url: https://codereview.chromium.org/2563643002
Cr-Commit-Position: refs/heads/master@{#437508}

BUG=661171

Review-Url: https://codereview.chromium.org/2561753002
Cr-Commit-Position: refs/heads/master@{#437507}

Also fix the seen merge logic.

BUG=651041

Review-Url: https://codereview.chromium.org/2558183004
Cr-Commit-Position: refs/heads/master@{#437506}

updateDistribution() is no longer required here since slotname-to-slot hashmap is
dynamically updated at the timing of DOM mutations.

BUG=671311

Review-Url: https://codereview.chromium.org/2559373003
Cr-Commit-Position: refs/heads/master@{#437505}

This patch replaces PropertyInterpolationTypesMapping with CSS and SVG
InterpolationTypesMap classes that are provided to InvalidatableInterpolation
objects via the InterpolationEnvironment.

InvalidatableInterpolations are no longer provided with the list of
InterpolationTypes at construction, now they retrieve the list at effect
application time during applyStack().

This is in preparation for supporting smooth animation of registered
custom properties.

BUG=671904

Committed: https://crrev.com/3ac8b90fcbda4c6e3ef484140751cd0341b31c48
Committed: https://crrev.com/dbf9ff76db17cc97436ee43a53ff269af3a45231
Review-Url: https://codereview.chromium.org/2555923003
Cr-Original-Original-Commit-Position: refs/heads/master@{#437381}
Cr-Original-Commit-Position: refs/heads/master@{#437474}
Cr-Commit-Position: refs/heads/master@{#437504}

Several functions in InProcessWorkerObjectProxy and ThreadedWorkletObjectProxy
are common. This CL introduces ThreadedMessagingProxyBase class, factors out the
functions into the base class and makes the both proxies inherit it.

[Class relationships after this CL]

On MainThread                   / On WorkerThread
ThreadedMessagingProxyBase      / ThreadedObjectProxyBase
- InProcessWorkerMessagingProxy / - InProcessWorkerObjectProxy
- ThreadedWorkletMessagingProxy / - ThreadedWorkletObjectProxy

BUG=667357

Review-Url: https://codereview.chromium.org/2566623002
Cr-Commit-Position: refs/heads/master@{#437503}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/46bd8103..72905fa4

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.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

TBR=hablich@chromium.org,machenbach@chromium.org,littledan@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/2566653002
Cr-Commit-Position: refs/heads/master@{#437502}

This CL addresses a few problems with #ifdefs:

1) DCHECK requires the presence of variables used in it even if DCHECK_IS_ON()
evaluates to 0. Therefore, if debug variable is defined only when DCHECK_IS_ON()
is true, its usage must be guarded the same way.

2) If debug method in base class is defined only when DCHECK_IS_ON() is true,
its descendants must be guarded with the same check.

BUG=666660

Review-Url: https://codereview.chromium.org/2550403002
Cr-Commit-Position: refs/heads/master@{#437501}

When length == 0, it should return a collasped range instead of a range
with distance == 1.
This is pre-CL of https://codereview.chromium.org/2530843003/

BUG=620549

Review-Url: https://codereview.chromium.org/2564763002
Cr-Commit-Position: refs/heads/master@{#437500}

Function enclosingTextControl(Node*) is not supposed to change anything,
and hence, is changed to take a const Node* instead.

BUG=n/a

Review-Url: https://codereview.chromium.org/2561633005
Cr-Commit-Position: refs/heads/master@{#437499}

BUG=672003
TEST=ParameterizedWebFrameTest.FindInPageJavaScriptUpdatesDOMProperOrdinal

Review-Url: https://codereview.chromium.org/2548253005
Cr-Commit-Position: refs/heads/master@{#437498}

Mark http/tests/security/mixedContent/websocket/insecure-websocket-in-secure-page-worker.html as leaky

BUG=672740
NOTRY=true
TBR=asargent@chromium.org, grt@chromium.org, vasilii@chromium.org

Review-Url: https://codereview.chromium.org/2562943002
Cr-Commit-Position: refs/heads/master@{#437497}

Cr-Commit-Position: refs/heads/master@{#437496}

It's an essential part of Timeline which should work in incognito mode.

BUG=490148

Review-Url: https://codereview.chromium.org/2556803003
Cr-Commit-Position: refs/heads/master@{#437495}

BUG=645667
NOTRY=true
TBR=asargent@chromium.org, grt@chromium.org, vasilii@chromium.org

Review-Url: https://codereview.chromium.org/2560143002
Cr-Commit-Position: refs/heads/master@{#437494}

Currently the redirected document URL of ServiceWorkerProviderHost is not
correctly set if skip_service_worker is set. This inconsistency is causing the
renderer kill crashes in ServiceWorkerDispatcherHost::OnGetRegistration() and
OnRegisterServiceWorker().

BUG=630496,630495

Review-Url: https://codereview.chromium.org/2562523003
Cr-Commit-Position: refs/heads/master@{#437493}

Chrome calls EnumDisplayDevices() to get details about display devices.
On some Windows machines EnumDisplayDevices() can only get the mirror
driver "RDPUDD Chained DD" with empty DeviceID and no physical display
devices when Chrome is opened in a Remote Desktop session, which will
cause the failure of launching the GPU process.

This patch fixes this issue by skipping the check of DeviceID when the
primary display device is "RDPUDD Chained DD" according to the result
of EnumDisplayDevices() so that Chrome can obtain GPU information
afterwards (in CollectDriverInfoD3D()).

BUG=671539
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2557513005
Cr-Commit-Position: refs/heads/master@{#437492}

This CL introduces generic functions for parsing variation parameters
into int, double, and bool (internally, variation parameters are
represented by a string).

The CL replaces some functions emulating this functionality in
ntp_snippets by the new ones in variations. More clean-up in
ntp_snippets is needed, left for further CLs.

BUG=671979

Review-Url: https://codereview.chromium.org/2558743003
Cr-Commit-Position: refs/heads/master@{#437491}

Don't run the prerendering tests that verify prerendering is cancelled for cross-process navigations under PlzNavigate.

Prerendering has these tests since it requires only one render process is used for prerendering. Without PlzNavigate, it detects a cross-process navigation and cancels prerendering. However with PlzNavigate, this scenario doesn't exist since we only determine the (one) render process to use for prerendering after redirects have been followed.

This fixes
PrerenderBrowserTest.PrerenderCrossProcessServerRedirect
PrerenderBrowserTest.PrerenderCrossProcessServerRedirectNoHang
with PlzNavigate.

BUG=504347

Review-Url: https://codereview.chromium.org/2560933002
Cr-Commit-Position: refs/heads/master@{#437490}

Split from https://codereview.chromium.org/2553803003/

BUG=671531

Review-Url: https://codereview.chromium.org/2553023002
Cr-Commit-Position: refs/heads/master@{#437489}

BUG=angleproject:1423
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2556033002
Cr-Commit-Position: refs/heads/master@{#437488}

This patch fixes some style errors caused at the Blink style change,
and some long living other style nits.

BUG=670237

Review-Url: https://codereview.chromium.org/2553383002
Cr-Commit-Position: refs/heads/master@{#437487}

Made caption-side property independent. Also updated
independent-inheritance-fast-path.html to include captionSide.

Tested using independent-inheritance-fast-path.html.

BUG=628043

Review-Url: https://codereview.chromium.org/2563443006
Cr-Commit-Position: refs/heads/master@{#437486}

Since our search is looking for found lines, not matches. This
patch changes wording about search results to make this explicit.

As a drive by, this patch starts using proper singular/plural
nouns.

BUG=585406
R=dgozman

Review-Url: https://codereview.chromium.org/2558173004
Cr-Commit-Position: refs/heads/master@{#437485}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/0b7222ff3002..df2363501bd7

$ git log 0b7222ff3..df2363501 --date=short --no-merges --format='%ad %ae %s'
2016-12-08 ksakamoto ChromeRendererHelper treats Chrome_InProcRendererThread as main thread
2016-12-08 benjhayden Reland Replace ValueSetOutputFormatter with HistogramSetJsonOutputFormatter.

BUG=671730,651188

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/2568433002
Cr-Commit-Position: refs/heads/master@{#437484}

Revert of Make InvalidatableInterpolation's InterpolationTypes decided at effect application time (patchset #4 id:60001 of https://codereview.chromium.org/2555923003/ )

Reason for revert:
Speculative revert for mass animation layout tests failures.

Original issue's description:
> Make InvalidatableInterpolation's InterpolationTypes decided at effect application time
>
> This patch replaces PropertyInterpolationTypesMapping with CSS and SVG
> InterpolationTypesMap classes that are provided to InvalidatableInterpolation
> objects via the InterpolationEnvironment.
>
> InvalidatableInterpolations are no longer provided with the list of
> InterpolationTypes at construction, now they retrieve the list at effect
> application time during applyStack().
>
> This is in preparation for supporting smooth animation of registered
> custom properties.
>
> BUG=671904
>
> Committed: https://crrev.com/3ac8b90fcbda4c6e3ef484140751cd0341b31c48
> Committed: https://crrev.com/dbf9ff76db17cc97436ee43a53ff269af3a45231
> Cr-Original-Commit-Position: refs/heads/master@{#437381}
> Cr-Commit-Position: refs/heads/master@{#437474}

TBR=ericwilligers@chromium.org,timloh@chromium.org,alancutter@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=671904

Review-Url: https://codereview.chromium.org/2562933002
Cr-Commit-Position: refs/heads/master@{#437483}