GitLab

https://skia.googlesource.com/skia.git/+log/114a3c0b2b26..b365cf590f87

$ git log 114a3c0b2..b365cf590 --date=short --no-merges --format='%ad %ae %s'
2017-05-30 fmalita Speculative fix for SkPerlinNoiseShaderImpl::asFragmentProcessor leak
2017-05-30 herb Add aarch64 tail code.
2017-05-30 fmalita [Reland] Relocate shaders to own dir
2017-05-30 fmalita Delete SkGaussianEdgeShader
2017-05-30 reed move perlinenoise2 into effects
2017-05-30 mtklein simplify sse41::srcover_srgb_srgb
2017-05-30 herb Add arm tail code.
2017-05-30 mtklein remove sse2::srcover_srgb_srgb
2017-05-30 fmalita Don't pull private headers in SkOverdrawColorFilter.h

Created with:
  roll-dep src/third_party/skia


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.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;master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=herb@chromium.org

Change-Id: I27e6535e4fc5894b10a8ebdd63bd2e6f962a0db9
Reviewed-on: https://chromium-review.googlesource.com/518425Reviewed-by: Skia Deps Roller <skia-deps-roller@chromium.org>
Commit-Queue: Skia Deps Roller <skia-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475787}

This CL flips `use_once_callback` flag on the Mojo code generator, and
fixes all compile errors after it. After this CL, Mojo interfaces in
//services/device starts using base::OnceCallback instead of
base::Callback on its return value handling.

The migration recipe was:
 - Convert pass-by-ref callback objects to pass-by-value.
 - Use std::move() to forward it to other consumer, or to invoke it
   with Callback::Run().
 - Handle wherever copies are required manually.
 - Check if the conversion doesn't change the semantics. As the transfer
   and invocation clobber the callback object, care about use-after-move.
   It's considered safe to consume almost scoped-out callback.

Bug: 714018
Change-Id: Ida11b9eaa331cce7f72ed0f1554731fd26659f2f
Reviewed-on: https://chromium-review.googlesource.com/517586Reviewed-by: Ken Rockot <rockot@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Commit-Queue: Taiju Tsuiki <tzik@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475786}

Add a "color spin" set of primaries to be used by layout tests. These
primaries are rotated RGB->GBR and have a ginormous gamut, so any
places where we fail to do color conversion should be very noticeable.

BUG=713891

Review-Url: https://codereview.chromium.org/2912903003
Cr-Commit-Position: refs/heads/master@{#475785}

Begin/FinishProcessingOrderNumber can release fence syncs and run their
callbacks. Those callbacks can call into the scheduler, in particular,
Enable/DisableSequence. So move the order processing to outside the lock
and make it safe to reenter the scheduler.

The included test deadlocks without this change. I haven't figured out
what happens after gpu process crash that causes problematic sync token
waits. I suspect it's sync tokens submitted by renderers that are left
over from the previous gpu process instance.

R=piman
BUG=514813
TEST=gpu_unittests SchedulerTest.ReentrantEnableSequenceShouldNotDeadlock

Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_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
Change-Id: I0a1d4154bb5f7b83b32a197fa8cb47a0f0d8e188
Reviewed-on: https://chromium-review.googlesource.com/518583Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Sunny Sachanandani <sunnyps@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475784}

Single-line messages in the network detailed view are updated to follow
Material Design specs. Also, fixes the padding for single-line messages
in Bluetooth detailed view from 22dip to 18dip.

BUG=687778
TEST=none

Change-Id: I82c377ad401026c9bdac595f614006584bd74336
Reviewed-on: https://chromium-review.googlesource.com/517967Reviewed-by: Terry Anderson <tdanderson@chromium.org>
Commit-Queue: Mohsen Izadi <mohsen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475783}

Now it lives in a search result container, which can be selected,
highlighted, navigated with keyboard etc.

Opening the result (i.e. what happens upon clicking on it) is not yet
implemented.

Unit test for the new result container isn't yet implemented since both
result opening and accessibility behavior are not yet implemented.

Bug=712331

Review-Url: https://codereview.chromium.org/2905523004
Cr-Commit-Position: refs/heads/master@{#475782}

Currently, we only construct instances of TextMatchMarker (previously
RenderedDocumentMarker) inside TextMatchMarkerListImpl, by calling
TextMatchMarker::Create() on an instance of DocumentMarker. This CL changes this
so callers directly construct instances of TextMatchMarker. This sets us up to
eventually move the storage and accessor methods for whether or not the marker
is active out of DocumentMarker and into TextMatchMarker.

BUG=707867

Review-Url: https://codereview.chromium.org/2905753002
Cr-Original-Commit-Position: refs/heads/master@{#475724}
Committed: https://chromium.googlesource.com/chromium/src/+/6b0a494122ebe2fc48099a9549b0acbf0b28571d
Review-Url: https://codereview.chromium.org/2905753002
Cr-Commit-Position: refs/heads/master@{#475781}

Bug: 
Change-Id: Id45f9ef44b25aac984b1f444911a233457a630f5
Reviewed-on: https://chromium-review.googlesource.com/517550Reviewed-by: Gabriel Charette <gab@chromium.org>
Commit-Queue: Takeshi Yoshino <tyoshino@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475780}

Active LayerImpl properties are never pushed. This patch enforces this
by making LayerImpl::SetNeedsPushProperties only work on the pending tree.

The existing VerifyLayerChangesAreTrackedProperly test has been modified
to work on the pending tree instead of the active tree. The existing
mask_to_bounds tests have been updated as they were intended for
another era when SetBounds did depend on mask_to_bounds.

An additional test has been added for active layer change tracking.

Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I711109df86e6c644333a8e14a5540a4b22675894
Reviewed-on: https://chromium-review.googlesource.com/516826
Commit-Queue: Philip Rogers <pdr@chromium.org>
Reviewed-by: enne <enne@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475779}

This CL makes it so editing commands will keep touch selections if they were
present before the command.

BUG=719890

Review-Url: https://codereview.chromium.org/2890313003
Cr-Commit-Position: refs/heads/master@{#475778}

SortableDataGrid.create may return null in some cases, subsequent call
to setStriped in such cases results in an exception logged.

Change-Id: I3f125c5747aeeaf54fce6596bd8451e2eb61b124
Reviewed-on: https://chromium-review.googlesource.com/518505Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Reviewed-by: Alexei Filippov <alph@chromium.org>
Reviewed-by: Andrey Kosyakov <caseq@chromium.org>
Commit-Queue: Eugene Ostroukhov <eostroukhov@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475777}

Bug: 726439
Change-Id: Ia688e6b35c33e48353d371476991c35559eb1b7d
Reviewed-on: https://chromium-review.googlesource.com/516865Reviewed-by: Ned Nguyen <nednguyen@google.com>
Commit-Queue: Matt Giuca <mgiuca@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475776}

On the x86 architecture on Android, there is an ordering issue for when
an std::move(...) happens. The call is on the form:
std::unique_ptr<Foo> foo;
...
foo->Init(std::move(foo));

On arm and other platform, this causes no issue, but for Android x86,
this pattern leads to |this| being nullptr inside Foo::Init(...).

This CL fixes that by first creating a copy of the raw pointer, and
invoking the method on that raw pointer instead of the std::unique_ptr.

BUG=727291, 727677

Change-Id: I617a1a3e41e486aeb1ede4d61bf3b35677f617b5
Reviewed-on: https://chromium-review.googlesource.com/518423Reviewed-by: David Trainor <dtrainor@chromium.org>
Reviewed-by: Bo Liu <boliu@chromium.org>
Commit-Queue: Tommy Nyquist <nyquist@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475775}

This patch removes unnecessary LayerById lookups in DidUpdateScrollOffset
in an attempt to improve both understanding and performance.

In addition:
1) An call to InnerViewportContainerLayer has been removed from
DidUpdateScrollState. This contains a LayerById lookup and is false in the
common case. This null-check removal is safe because it will be null-checked
later in DidUpdateScrollState.
2) The scrollbar vertical adjustment code has been moved so it only applies
to viewport scrolls. Non-viewport scrolls do not have ViewportBoundsDelta.

Bug: 724193
Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel
Change-Id: Ia1bc76769781243b2e486d0f1bbbae59bbe07ad3
Reviewed-on: https://chromium-review.googlesource.com/517902Reviewed-by: enne <enne@chromium.org>
Commit-Queue: Philip Rogers <pdr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475774}

In our vain attempt to re-enable performance_browser_tests on the Mac
bots, a number of breakages seem to have happened over the months where
we were not even aware the tests weren't running.

This change fixes all of the following:

1. Adds a BUILD dependency on //chrome:chrome_app. Without this, the
test binary would be missing many runtime dependencies (e.g., the whole
"Chrome Framework" bundle).

2. Migrate MachPortsTest to use BrowserTestBase::embedded_test_server()
instead of its own server of "fake content" for its browser tests.

3. Add base::ScopedAllowIO for a test utility class that creates and
tears down its own set of threads. It's not clear why the thread
restriction is only checked on Mac (perhaps it's in the way the
browser test's main thread is set up?). Nevertheless, it's perfectly
reasonable for the main thread to block on stopping these other threads:
After the test has run, the threads are shut down before the data
analysis begins, since we want to be sure the data sets are no longer
mutating. Also, added PRESUBMIT.py exception for this use case.

BUG=697444,722367
TBR=nduca

Review-Url: https://codereview.chromium.org/2901113003
Cr-Commit-Position: refs/heads/master@{#475773}

TextIterator::Advance() used to contain some code before its main loop
to handle the remembered progress, if any. This patch wraps the code
into a new function TextIterator::HandleRememberedProgress() for better
code health.

Follow-up patches will utilize this function to make the control flow
in TextIterator cleaner.

BUG=721957
TEST=n/a; no behavioral change

Review-Url: https://codereview.chromium.org/2911373003
Cr-Commit-Position: refs/heads/master@{#475772}

 - Fixing almost all existing violations
   (remaining ones are fixed at crrev.com/2903063007).
 - Enabling http://eslint.org/docs/rules/semi check during PRESUBMIT

This enforces the following section from the styleguide:
https://google.github.io/styleguide/jsguide.html#formatting-semicolons-are-required

BUG=720034
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2909503003
Cr-Commit-Position: refs/heads/master@{#475771}

This CL stops adding preloaded RawResources to MemoryCache. This is needed
to match preloaded resources with requests with UseStreamOnResponse specified.

From the following reasons, I believe the performance impact is negligible.

 - Preloading matching is now decoupled from MemoryCache in general.
 - Loading resources can be shareable only in the same frame.
 - There is no speculative RawResource preload.

Bug: 652228
Change-Id: Icc5f7fc893702233cc45bc130eb13e6fcae919df
Reviewed-on: https://chromium-review.googlesource.com/516926Reviewed-by: Hiroshige Hayashizaki <hiroshige@chromium.org>
Commit-Queue: Yutaka Hirano <yhirano@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475770}

R=dalecurtis@chromium.org

BUG=667532
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_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

Review-Url: https://codereview.chromium.org/2902313003
Cr-Commit-Position: refs/heads/master@{#475769}

-Adds support for switch fields in the PaymentRequestEditViewController.
-Refactors PaymentRequestEditViewController subclasses into their parent.

BUG=602666

TBR=jdonnelly@

Review-Url: https://codereview.chromium.org/2908033002
Cr-Commit-Position: refs/heads/master@{#475768}

- introduce TimelineHistoryManager;
- keep a list of at most 5 recorded/loaded timelines, discard LRU ones;
- present recording history as a drop-down list of overview & screenshot thumbnails,
    accessible via a toolbar button or shortcut;
- (drive-by) make ListControl focusable, properly disable Reload action button.
- (drive-by) fix initial state of action button if action is disabled;

Bug: 726381
Change-Id: I8e154530da6c60744c7f5192cae1914ab3623e4f
Reviewed-on: https://chromium-review.googlesource.com/517295
Commit-Queue: Andrey Kosyakov <caseq@chromium.org>
Reviewed-by: Alexei Filippov <alph@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475767}

Give an indication in the documentation where this structure is
defined/specified. So others readers studing this code can read
more information about it.

BUG=None

Change-Id: Iedf747d27231be4982010519365a6caa6dfc1c6d
Reviewed-on: https://chromium-review.googlesource.com/518046
Commit-Queue: Thiago Farina <tfarina@chromium.org>
Reviewed-by: Eric Roman <eroman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475766}

Record the source of PPDs and the success or failure of a printer setup
request from the setup flow in settings.

BUG=725741

Review-Url: https://codereview.chromium.org/2911523002
Cr-Commit-Position: refs/heads/master@{#475765}

This CL uses /media/test/data/ as the seed_corpus for
media_pipeline_integration_fuzzer.

BUG=584119

Change-Id: Ic0f237f4326b8dc4ab41fd9c5fced6c2a9c5ac05
Reviewed-on: https://chromium-review.googlesource.com/518256
Commit-Queue: Xiaohan Wang <xhwang@chromium.org>
Commit-Queue: Max Moroz <mmoroz@chromium.org>
Reviewed-by: Max Moroz <mmoroz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475764}

Bug: 718063
Change-Id: Ibe1fd15363646bc7132975548c6ef7944508c5e1
Reviewed-on: https://chromium-review.googlesource.com/518482Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Commit-Queue: Alexei Filippov <alph@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475763}

Cr-Commit-Position: refs/heads/master@{#475762}

https://pdfium.googlesource.com/pdfium.git/+log/2c200030b55f..26cb2fa42b1a

$ git log 2c200030b..26cb2fa42 --date=short --no-merges --format='%ad %ae %s'
2017-05-30 tsepez Use unowned ptr back to delegates in codec modules

Created with:
  roll-dep src/third_party/pdfium


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


TBR=dsinclair@chromium.org

Change-Id: Icda5cd5cde972aa253bd9af91b8832f1d02a1396
Reviewed-on: https://chromium-review.googlesource.com/518547
Reviewed-by: <pdfium-deps-roller@chromium.org>
Commit-Queue: <pdfium-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475761}

BUG=727599

Review-Url: https://codereview.chromium.org/2908323002
Cr-Commit-Position: refs/heads/master@{#475760}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/e18c0f9c7950..32a3f0b8f49b

$ git log e18c0f9c7..32a3f0b8f --date=short --no-merges --format='%ad %ae %s'
2017-05-30 rnephew [BattOr] Update BattOr agent Binary.
2017-05-30 simonhatch Pinpoint - Remove some unneeded files.
2017-05-30 ccraik Switch from printf to echo for tgid fixup
2017-05-30 ccraik Fix viewer html generation on windows
2017-05-30 dtu [pinpoint] Make Change.deps a frozenset.
2017-05-30 jbudorick [devil] Make devil.utils.cmd_helper work on Windows?
2017-05-30 dproy Fix markdown typo in cpu_time tree doc
2017-05-30 phsilva Implement the Ownership Diagnostic Class

Created with:
  roll-dep src/third_party/catapult
BUG=699581


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=sullivan@chromium.org

Change-Id: I4502fff3bec2c407c5f82649a68b86d24a35f000
Reviewed-on: https://chromium-review.googlesource.com/518507
Reviewed-by: <catapult-deps-roller@chromium.org>
Commit-Queue: <catapult-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475759}

Review-Url: https://codereview.chromium.org/2910243002
Cr-Commit-Position: refs/heads/master@{#475758}

This is another migration from BrowserAccessibility to AXPlatformNodeWin.
In this installment, we're converting get_accRole.  I have updated the
AXPlatformNodeWin::MSAARole() to include many of the missing roles that
the BrowserAccessibility code had.  I also introduced a new method that
allows get_accRole to return a string (overriding the default type value
that MSAARole() returns).  I have not dug into why we need to return the
various values instead of just using the role number or html element
name. Instead, I just added todos for me to follow up on.

BUG=703369

Review-Url: https://codereview.chromium.org/2909853002
Cr-Commit-Position: refs/heads/master@{#475757}

To prevent excessive IPC messages, a parent frame of an OOPIF only
updates its remote viewport intersection rect when the intersection
changes. However, a frame can navigate without the parent frame being
aware of it, in which case the rect is lost without being resent.

To address that problem, this CL caches the rect on
CrossProcessFrameConnector in the browser process, and sends it to the
renderer whenever RenderWidgetHostImpl::SendScreenRects() is called.

This CL also moves the rect's storage in Blink from LocalFrameView to
LocalFrame so that it persists across same-process navigations.

BUG=720342
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2910023002
Cr-Commit-Position: refs/heads/master@{#475756}

code that creates tutorial UI is never referenced anywhere.

Review-Url: https://codereview.chromium.org/2899083003
Cr-Commit-Position: refs/heads/master@{#475755}

BUG=723975
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2913083002
Cr-Commit-Position: refs/heads/master@{#475754}

This CL also fixes a BUILD file error so that JSCompiler can
be run.

BUG=727904

Review-Url: https://codereview.chromium.org/2913073002
Cr-Commit-Position: refs/heads/master@{#475753}

BUG=723406
R=eae@chromium.org,pdr@chromium.org

Change-Id: Ia0fd51944a549f4ef6b8e704fc5101071ae22bf2
Reviewed-on: https://chromium-review.googlesource.com/518252
Commit-Queue: Stefan Zager <szager@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#475752}

TEST=Tested manually with apps with popups out of task bounds.
BUG=b/62088045

Review-Url: https://codereview.chromium.org/2909213003
Cr-Commit-Position: refs/heads/master@{#475751}

Added DownloadStore which is a layer around a leveldb ProtoDatabase that
stores download params, scheduling params, request headers and various
metadata as protos. Callers of DownloadStore expect a list of Entry's,
hence conversion utility methods were added to convert between Entry
and protodb::Entry.

Other details:
1- Moved NoopStore to download/internal/test/
2- Replaced usage of OnceCallback with repeated Callback, since the
proto database expects repeated Callback. In order to change ProtoDatabase
to use OnceCallback, it probably requires a larger change invloving
changing the client sites of ProtoDatabase.

BUG=722705

Review-Url: https://codereview.chromium.org/2881173003
Cr-Commit-Position: refs/heads/master@{#475750}

In general there shouldn't really be any need for both. We could just remove
AvoidsFloats() and keep CreatesNewFormattingContext(). But then again, it might
be considered weird to say that replaced content creates a block formatting
context. So let's keep the both of them for now.

However, the following rule should always apply: If the object creates a new
formatting context, it implies that it also avoids floats. That's the reason for
declaring LayoutBlockFlow::AvoidsFloats() as final.

Furthermore, the only kind of LayoutBlock-type objects that DOESN'T create a
new formatting context, are certain LayoutBlockFlow (block container) objects.
So move the checks over from LayoutBlock to LayoutBlockFlow. Keep a virtual
true-returning CreatesNewFormattingContext() in LayoutBlock. We could actually
consider removing this, but there are 2-3 call sites that currently need it.

This CL causes a couple of minor rendering differences: Previously, HR elements
were told to avoid floats, but not to establish block formatting contexts.
Let's be consistent and return true for both. This makes us more compatible
with Edge. Test included. Turning HR elements into true block formatting
contexts also affects margin collapsing. Its top and bottom margins no longer
collapse. Had to make a change to the default style sheet for HR inside
SELECT because of this. Inside SELECT, HR loses its borders, so that its top
and bottom margins would previously collapse, but not anymore, now that it
establishes a block formatting context. To cancel out any rendering difference,
reduce margins in this particular case.

Similarly for RT (ruby text). LayoutRubyText objects are repositioned after
layout, which essentially requires them to contain all child floats, i.e.
establish a new block formatting context. Previously they achieved a
quasi-formatting context by returning true from AvoidFloats(). The screenshot
test fast/ruby/float-overhang-from-ruby-text.html would fail spectacularly if
they exposed child floats to their sibling ruby base. That test expectation
still requires a tiny update to its layout tree dump, though.

This CL also makes CreatesNewFormattingContext() return true for tables,
flexboxes, grids, and anything blocky that isn't LayoutBlockFlow. This
shouldn't be a web-exposable change, though, since no non-LayoutBlockFlow types
can contain float children (e.g. a float inside a table would have to be
wrapped inside either a table-cell or a table-caption (which both establish a
new block formatting context), and a float inside a flexbox would have to be
wrapped inside a flex item (which establishes a new block formatting context)).

BUG=460704

Change-Id: If60c1fc636db73a7ff241471ea7bf95adf996512
Reviewed-on: https://chromium-review.googlesource.com/512824Reviewed-by: Robert Hogan <robhogan@gmail.com>
Commit-Queue: Morten Stenshorne <mstensho@opera.com>
Cr-Commit-Position: refs/heads/master@{#475749}

This bug was discovered by the Kitchen Sync integration test
BothClientsEnabledEncryptionAndChangedMultipleTimes_E2ETest. What was
happening is that for a given client, it was mid PostTask "bounce" in
TrySyncCycleJob, when a second nudge was scheduled. This was allowed to
get through because the pending_wakeup_timer_ wasn't running.

However, when TrySyncCycleJobImpl contacted the server, we hit an error,
and the scheduler tried to call RestartWaiting(). This didn't have any
effect, because now the pending_wakeup_timer_ has a imminent nudge,
with a smaller delay time, and we had no way of knowing the priority of
the pending wakeup tasks.

The above is arguably all okay, but when the now pending nudge was run,
it has low priority, but we're in an error state, and this no-ops. And
at the end of PerformDelayedNudge we explicitly don't call
RestartWaiting(), which seems to be wrong.

BUG=721850

Review-Url: https://codereview.chromium.org/2911603003
Cr-Commit-Position: refs/heads/master@{#475748}