GitLab

Scrubbed Chrome OS profiles (well, one at the moment) live in
chrome/android/profiles. Hopefully we won't need to scrub them soon.

In order to do AFDO on Chromium, we need to pull them locally.

size wins are observed as with unscrubbed profiles on Chrome.apk
(assuming the profile actually gets turned on, which is moved to
another patch).

Bug: 794750
Test: ChromePublic.apk builds after `gclient sync`. Similar perf/binary
Change-Id: I9cc034ca21e948b372944b6bcdc8c5055fd71ca8
Reviewed-on: https://chromium-review.googlesource.com/843774Reviewed-by: agrieve <agrieve@chromium.org>
Reviewed-by: Maria Khomenko <mariakhomenko@chromium.org>
Commit-Queue: George Burgess <gbiv@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526794}

Bug: 734095
Change-Id: Id8d1a51014187c0b49726810985b38c67caa9c5f
Reviewed-on: https://chromium-review.googlesource.com/847681
Commit-Queue: Xi Cheng <chengx@chromium.org>
Reviewed-by: Finnur Thorarinsson <finnur@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526793}

https://pdfium.googlesource.com/pdfium.git/+log/05dcbc931eac..cbd4410908e2

$ git log 05dcbc931..cbd441090 --date=short --no-merges --format='%ad %ae %s'
2018-01-03 dsinclair Fold CXFA_MarginData into CXFA_Margin
2018-01-03 dsinclair Fold CXFA_CalculateData into CXFA_Calculate

Created with:
  roll-dep src/third_party/pdfium


The AutoRoll server is located here: https://pdfium-roll.skia.org

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


TBR=dsinclair@chromium.org

Change-Id: Ie9829258472bbfca1bdaf4a0f149ff8a7d511dbc
Reviewed-on: https://chromium-review.googlesource.com/848243
Reviewed-by: <pdfium-deps-roller@chromium.org>
Commit-Queue: <pdfium-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526792}

R=tandrii

Bug: 793347
Change-Id: I348a5e568066b7304cdacd71c79ed203de3fd217
Reviewed-on: https://chromium-review.googlesource.com/849272Reviewed-by: Andrii Shyshkalov <tandrii@chromium.org>
Commit-Queue: Aaron Gable <agable@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526791}

Pepper can destroy a fullscreen object and a dangling ptr to a a lock
target can be left. If an IPC is received after it is destroyed the
dangling pointer could be called.

BUG=759542

Change-Id: I746ef83dac8120559ef82ed772bd8861ab9e79bf
Reviewed-on: https://chromium-review.googlesource.com/848057Reviewed-by: Kentaro Hara <haraken@chromium.org>
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526790}

I inadvertently broke some Windows bots when I landed:
https://chromium-review.googlesource.com/c/chromium/src/+/848322

This sync'd the devtools-node-modules repo for all platforms, which caused
an issue on Windows because some of the node modules were deeply nested
(npm v2 and earlier installs it in a deeply nested manner). I restructured
the repo by installing it with npm v5 which flattens the node modules directory
which should solve these bot issues.

Bug: 798765
Change-Id: I09056dbf6db3e4101918910f9c23d1cab260cc85
Reviewed-on: https://chromium-review.googlesource.com/848510Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Commit-Queue: Will Chen <chenwilliam@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526789}

It's a struct, and there are no subclasses.

Bug: none
Change-Id: Ie3078977c9b59315869752d092f0e2382327fca2
Reviewed-on: https://chromium-review.googlesource.com/848640Reviewed-by: Miriam Gershenson <mgersh@chromium.org>
Commit-Queue: Matt Menke <mmenke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526788}

There's an existing pixel test, LayerWithRealCompositorTest's
CanvasDrawFadedString. However,
 - Nothing calls Canvas::DrawFadedText() except this test
 - The test is only enabled on Windows ("pixel tests are hard")
 - The test's reference image is too big..
 - ..which means it doesn't actually need to elide/fade the string.
 - It requires a lot of plumbing (DrawFadedStringLayerDelegate)
 - Is uses cc::FuzzyPixelComparator with some magic numbers.

http://crrev.com/c/828220 would delete Canvas::DrawFadedText() since
nothing uses it, along with the existing pixel test. However, faded
text still exists and is used in views::Label.

This CL moves the pixel test to views::Label and "seeds" a pixel
testing framework under ui/views for testing UI. Native UI often,
intentionally differs across platforms and OS versions. The approach
for the pixel testing framework attempts to cater for that without
creating a maintenance burden.

Bug: 791391
Change-Id: I6c71f2c98cf7feca07e2a4a10cf37926710b9d9f
Reviewed-on: https://chromium-review.googlesource.com/805516Reviewed-by: danakj <danakj@chromium.org>
Reviewed-by: Michael Wasserman <msw@chromium.org>
Commit-Queue: Trent Apted <tapted@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526787}

Also renames mojo_chromiumos_gtests to mojo_chromiumos_fyi_gtests to
make it clear this only applies to the FYI waterfall.

BUG=776514
TEST=test only changes

Change-Id: I5ae831753a81e1f8eac54c01272c7cdd16017bab
Reviewed-on: https://chromium-review.googlesource.com/848272
Commit-Queue: Scott Violet <sky@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526786}

This temporary hack to get an implicit copy constructor is no longer
needed.

Change-Id: I06d45124f4d629e03be1bbefaa8479ad2060aac7
Reviewed-on: https://chromium-review.googlesource.com/847957Reviewed-by: Bence Béky <bnc@chromium.org>
Commit-Queue: Reilly Grant <reillyg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526785}

To fix the flakiness the tests are using mock time increment.

Bug: 795617
Change-Id: I58ac018638ae33e4cd691bf1ea15a47f3b74d80f
Reviewed-on: https://chromium-review.googlesource.com/847461Reviewed-by: Timothy Dresser <tdresser@chromium.org>
Reviewed-by: Dave Tapuska <dtapuska@chromium.org>
Commit-Queue: Sahel Sharifymoghaddam <sahel@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526784}

Bug: 772697
Change-Id: Ic8f75ef3970aeaa0c7f968957873898fd12c7137
Reviewed-on: https://chromium-review.googlesource.com/846341
Commit-Queue: Dmitry Gozman <dgozman@chromium.org>
Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526783}

The WifiPollingPolicy controls how frequently the network location
provider will initiate Wi-Fi scans. Previously, the policy was attached
to the WifiDataProvider, which is destroyed when there are no active
geolocation API calls. Destruction of the provider causes us to lose track
of our polling policy.

The policy allows the first scan to be performed immediately and schedules
subsequent scans at regular intervals. When the provider is destroyed and
recreated rapidly it considers each scan to be "first", which allows many
Wi-Fi scans to be performed in a short window.

This change moves the policy state to a global instance so it can be saved
when the provider is destroyed and recreated. The new policy allows the
first-ever scan to be performed immediately, and subsequent scans may also
be performed immediately if enough time has lapsed since the previous scan.
In all other situations, the configured polling interval is enforced.

BUG=764954

Change-Id: I00649dde2a707127e53e0330511172fef4a43471
Reviewed-on: https://chromium-review.googlesource.com/834988
Commit-Queue: Matt Reynolds <mattreynolds@chromium.org>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526782}

This CL rewrites two parts of VideoResourceUpdater::CreateForSoftwarePlanes()
to take advantage of the faster libyuv library functions:
- libyuv::Convert16To8Plane() was landed as part of libyuv:751
and is way faster than the C counterpart [1]
- libyuv::CopyPlane() existed already and is also faster than
the C counterpart.

This CL is purely refactoring, functionality is covered by unit
and browser tests.

[1] https://bugs.chromium.org/p/libyuv/issues/detail?id=751#c11

Bug: 798485
Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
Change-Id: Iead82969b13917aabdaacee47799f6816a164171
Reviewed-on: https://chromium-review.googlesource.com/848055Reviewed-by: Miguel Casas <mcasas@chromium.org>
Reviewed-by: danakj <danakj@chromium.org>
Commit-Queue: Miguel Casas <mcasas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526781}

This CL fixes some typos in comments. There are no logic changes or user
visible changes.

TBR=tommycli@chromium.org
(code health)

Bug: None
Change-Id: I7772f14fa473d93e773179bfb5a66352da5aaeef
Reviewed-on: https://chromium-review.googlesource.com/848374
Commit-Queue: Dave Schuyler <dschuyler@chromium.org>
Reviewed-by: Dave Schuyler <dschuyler@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526780}

This CL fixes sign-in behind an authenticated proxy. Two actual changes
were necessary in this context:
- SigninPartitionManager: copy auth data from the system request
  context into new StoragePartitions
  This makes SigninPartitionManager::StartSigninSession async, which
  made small changes necessary in GaiaScreenHandler and
  EnrollmentScreenHandler
- SigninScreenHandler: Don't reload gaia immediately when auth details
  have been supplied. This was unnecessary (the URL Request will
  continue) and actually harmful, as now that we're using a new
  StoragePartition for each sign-in attempt, this discards auth data
  before we can copy them.

To support the browsertest, testserver.py gets an argument to redirect
CONNECT requests to localhost when proxying (in sign-in browsertests
without proxy, we use RuleBasedHostResolverProc::AddRule to achieve
this effect).

Bug: 793524
Test: browser_tests --gtest_filter=WebviewProxyAuthLoginTest.* &&
      unit_tests --gtest_filter=SigninPartitionManagerTest*
  Manual test:
   Setup:
     Setup an proxy server with Basic authentication
     Configure the device to use the proxy server.
     Make sure the user adding screen is shown.
   Test 1:
     Expect proxy auth dialog. Enter correct proxy auth data.
     Expect that sign-in screen is shown and sign-in works.
   Test 2:
     Expect proxy auth dialog. Enter incorrect proxy auth data.
     Expect that proxy-auth dialog is shown again.
   Test 3:
     Expect proxy auth dialog. Press Cancel.
     Expect that a network error dialog is shown.
     Press "try to sign in again".
     Expect proxy auth dialog. Enter correct proxy auth data.
     Expect that sign-in screen is shown and sign-in works.

Change-Id: I78b5988b5ca4c6eaaf4e6f9443e299499d6f0f08
Reviewed-on: https://chromium-review.googlesource.com/837920Reviewed-by: Matt Menke <mmenke@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Commit-Queue: Pavol Marko <pmarko@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526779}

This reverts commit 0346a3a1.

Reason for revert: <INSERT REASONING HERE>
Causing compilation failure here:
https://ci.chromium.org/buildbot/chromium.win/WinMSVC64%20(dbg)/1592

Original change's description:
> Adds predictable multimatch support in PreviewsOptimizationGuide
> 
> Supports multiple hint matches with precendence given to the match
> that occurs first in the configuration. Also, allows matches
> without whitelisted optimization to be able to turn off whitelisting
> for a host pattern (that otherwise would be whitelisted for another
> matching pattern that occurs later in config).
> 
> Bug: 783237
> Change-Id: I9239b74b170a1d5a19b13caccecc732f9715de90
> Reviewed-on: https://chromium-review.googlesource.com/847816
> Reviewed-by: Ryan Sturm <ryansturm@chromium.org>
> Commit-Queue: Doug Arnett <dougarnett@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#526763}

TBR=dougarnett@chromium.org,ryansturm@chromium.org

Change-Id: I92772c5873d06af184449ae6347bb2ebacb5100f
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 783237
Reviewed-on: https://chromium-review.googlesource.com/848535Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526778}

This reverts commit 0346a3a1.

Reason for revert: Causing build failure on WinMSVC64 (dbg):
https://ci.chromium.org/buildbot/chromium.win/WinMSVC64%20(dbg)/1592

Original change's description:
> Adds predictable multimatch support in PreviewsOptimizationGuide
> 
> Supports multiple hint matches with precendence given to the match
> that occurs first in the configuration. Also, allows matches
> without whitelisted optimization to be able to turn off whitelisting
> for a host pattern (that otherwise would be whitelisted for another
> matching pattern that occurs later in config).
> 
> Bug: 783237
> Change-Id: I9239b74b170a1d5a19b13caccecc732f9715de90
> Reviewed-on: https://chromium-review.googlesource.com/847816
> Reviewed-by: Ryan Sturm <ryansturm@chromium.org>
> Commit-Queue: Doug Arnett <dougarnett@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#526763}

TBR=dougarnett@chromium.org,ryansturm@chromium.org

Change-Id: If54e402cac060422cbff45a11ead41ca6c31d542
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 783237
Reviewed-on: https://chromium-review.googlesource.com/849281Reviewed-by: Thomas Anderson <thomasanderson@chromium.org>
Commit-Queue: Thomas Anderson <thomasanderson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526777}

--dry-run was broken after https://crrev.com/525224 because binary_name
was no longer defined in _GetSymbolsMapping(), while still being used
for dry-run.

Change-Id: I76566e3e22007401a5c5d80daa5e7dc30af2c91f
Reviewed-on: https://chromium-review.googlesource.com/848431Reviewed-by: Wez <wez@chromium.org>
Commit-Queue: Wez <wez@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526776}

Make mutableCopy of http_request_headers_ to avoid crashes in
AddHttpRequestHeaders and RemoveHttpRequestHeaderForKey.

Bug: 798070
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: Iaca5eb950f79b5936a75523392322036a37158a5
Reviewed-on: https://chromium-review.googlesource.com/846433
Commit-Queue: Eugene But <eugenebut@chromium.org>
Reviewed-by: Mike Dougherty <michaeldo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526775}

https://skia.googlesource.com/skia.git/+log/aebad197dcaf..d9852cd27a76

$ git log aebad197d..d9852cd27 --date=short --no-merges --format='%ad %ae %s'
2018-01-03 brianosman Remove obsolete comment
2018-01-03 brianosman Remove unused color from all distance field GPs and GrAtlasTextOp
2018-01-03 reed partial revert of https://skia-review.googlesource.com/c/skia/+/90026 for android
2018-01-03 reed update pipe for lattice and shadowrec
2018-01-03 caryclark refresh public includes
2018-01-03 caryclark support srgb flag in bookmaker
2018-01-03 angle-skia-autoroll Roll skia/third_party/externals/angle2/ d8724a947..93b059db4 (1 commit)
2018-01-03 rmistry Add Housekeeper-PerCommit-Bookmaker to cq.cfg with 100% experiment%

Created with:
  roll-dep src/third_party/skia


The AutoRoll server is located here: https://autoroll.skia.org

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


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=jcgregorio@chromium.org

Change-Id: Ie824679dea9d0130e98f51792a0986501bd7aa06
Reviewed-on: https://chromium-review.googlesource.com/849154
Commit-Queue: Skia Deps Roller <skia-deps-roller@chromium.org>
Reviewed-by: Skia Deps Roller <skia-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526774}

Bug: 753594
Change-Id: Ie03bcba5d02c20fafce965cb3f48cb1dd976d211
Reviewed-on: https://chromium-review.googlesource.com/843205
Commit-Queue: Thanh Le <thanhdle@chromium.org>
Reviewed-by: Doug Arnett <dougarnett@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526773}

This instance shows that this test could timeout on android.

TBR=hiroshige@chromium.org
NOTRY=true

Bug: 755405
Change-Id: Ibbdad38e0812c6c162b732fbac200b002d2bb97d
Reviewed-on: https://chromium-review.googlesource.com/848647Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526772}

Noticed that bloaty does this and I think it's a good idea. It enables you to
distinguish templates from non-templates, while still allowing grouping.

Also fixes template stripping of operator>>

Change-Id: I6deddcf01464e191c64f6d4569f8479c628a2c94
Reviewed-on: https://chromium-review.googlesource.com/849133
Commit-Queue: agrieve <agrieve@chromium.org>
Reviewed-by: Samuel Huang <huangs@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526771}

Bug: 782015
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_mojo
Change-Id: I877446e14fb53d0c6c0ea73b5eda8399821668fc
Reviewed-on: https://chromium-review.googlesource.com/801235Reviewed-by: Ken Rockot <rockot@chromium.org>
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Commit-Queue: Chris Mumford <cmumford@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526770}

This instance shows that the
virtual/threaded/fast/scroll-behavior/smooth-scroll/mousewheel-scroll-interrupted.html
could time out on Linux.

TBR=hiroshige@chromium.org
NOTRY=true

Bug: 665577
Change-Id: I65f1e8f70a180cdcc0f6d8e9263fd4897231bfeb
Reviewed-on: https://chromium-review.googlesource.com/848035Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526769}

I think this started appearing when we added AFDO profiles.

Change-Id: Icc17e00411368d3bd51ba6d3844e35228523e2e5
Reviewed-on: https://chromium-review.googlesource.com/848034Reviewed-by: Samuel Huang <huangs@chromium.org>
Commit-Queue: agrieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526768}

Cq-Include-Trybots: master.tryserver.chromium.linux:closure_compilation
Change-Id: I1f2cddd7b8faade4c9913e08010ebe6be0dd12fe
Reviewed-on: https://chromium-review.googlesource.com/842709Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526767}

This regressed in r492972 (switch to a custom window frame view class).

When not using NSWindowStyleMaskFullSizeContentView (macOS <10.12, but
it can probably be turned on back to 10.10), certain circumstances
caused the tab strip background view to partially cover up the toolbar.
The most reliable way to reproduce it, for me, is:

- Be running macOS 10.11 or older.
- Have multiple monitors connected.
- Turn off "Displays have separate Spaces" in System Preferences.
- Enter web content fullscreen (e.g. on YouTube).
- Exit fullscreen by pressing Esc.

Three things came together to make this happen:

1. |-[FramedBrowserWindow browserFrameViewPaintHeight]| returned a
   hard-coded value when not using FullSizeContentView. I believe it
   used to represent the height of the tab strip but no longer does, and
   I believe it was just kept out of caution when the full size content
   view path was added, but it's bigger than it needs to be and makes
   the tab strip background view slightly larger than necessary. (This
   value originally appeared in r40677).

2. |-[TabWindowController
   insertTabStripBackgroundViewIntoWindow:titleBar:]| explicitly
   inserted the background view behind all other views.

3. The immersive fullscreen path, which is only taken when multiple
   monitors are connected and "Displays have separate Spaces" is turned
   off (see |-[BrowserWindowController enterWebContentFullscreen]|)
   inserts the content view behind all other views on exiting fullscreen
   (see |-[BrowserWindowController
   moveViewsForImmersiveFullscreen:regularWindow:fullscreenWindow:]|),
   which means that it's now behind the tab strip background view.

I'm fixing this with three changes:

1. Get rid of the special value in |-browserFrameViewPaintHeight| and
   return |chrome::kTabStripHeight| in both cases.

2. Hide the |visualEffectView_|/|tabStripBackgroundView_| in fullscreen
   always, not just when using FullSizeContentView — othersise, it can be seen
   visible on top of the content in fullscreen.

3. When setting up a window, insert the tab strip background view
   normally instead of behind its siblings. It just made these bugs
   harder to find (the strip would have appeared all the time,
   otherwise).

Bug: 779082
Change-Id: Id9536dc6ff8593b3e7e2b3f38bbf9149418d0ac6
Reviewed-on: https://chromium-review.googlesource.com/848641Reviewed-by: Elly Fong-Jones <ellyjones@chromium.org>
Commit-Queue: Sidney San Martín <sdy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526766}

Previously, all U2F HID device requests would fail if a packet for
different channel was received from hid device during channel
allocation. Changed this behavior so that packets for different channels
will be disregarded without causing request failure.

Bug: 785337
Change-Id: Ibad715d4fd0024dd0d3ca35ac409e3d72ec0eb32
Reviewed-on: https://chromium-review.googlesource.com/770305
Commit-Queue: Jun Choi <hongjunchoi@chromium.org>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526765}

Bug: 488146
Test: existing
Change-Id: I41d59690688ca8dbba01fee9260fd25497061ff5
Reviewed-on: https://chromium-review.googlesource.com/848056Reviewed-by: Robert Sesek <rsesek@chromium.org>
Reviewed-by: Sebastien Seguin-Gagnon <sebsg@chromium.org>
Commit-Queue: Sebastien Seguin-Gagnon <sebsg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526764}

Supports multiple hint matches with precendence given to the match
that occurs first in the configuration. Also, allows matches
without whitelisted optimization to be able to turn off whitelisting
for a host pattern (that otherwise would be whitelisted for another
matching pattern that occurs later in config).

Bug: 783237
Change-Id: I9239b74b170a1d5a19b13caccecc732f9715de90
Reviewed-on: https://chromium-review.googlesource.com/847816Reviewed-by: Ryan Sturm <ryansturm@chromium.org>
Commit-Queue: Doug Arnett <dougarnett@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526763}

Previously, NetworkContext::ClearNetworkingHistorySince would tell
HttpServerProperties to clear data, and then immediately inform the
caller that the data was cleared. The data was deleted from memory by
that point, but hadn't been written to disk, and we didn't even trigger
a flush to disk, at that point.

This CL makes changes that. We now pass in the callback to the
HttpServerProperties, which writes the data to the pref store and
then tells the pref store to flush data to disk. Only once the newly
cleared data has been flushed do we invoke the callback passed to
NetworkContext::ClearNetworkingHistorySince.

Bug: 795877
Cq-Include-Trybots: master.tryserver.chromium.android:android_cronet_tester;master.tryserver.chromium.linux:linux_mojo;master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I4b1ed23eb5bdbab77871c79cdd068b95c0972535
Reviewed-on: https://chromium-review.googlesource.com/834393Reviewed-by: Andrei Kapishnikov <kapishnikov@chromium.org>
Reviewed-by: Rohit Rao <rohitrao@chromium.org>
Reviewed-by: Ryan Hamilton <rch@chromium.org>
Commit-Queue: Matt Menke <mmenke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526762}

TBR=chrishtr@chromium.org, robertma@chromium.org
NOTRY=true

Change-Id: Ieaccbe40de8e60bce9ca24523f8445af71c08101
Reviewed-on: https://chromium-review.googlesource.com/848646Reviewed-by: Stephen Chenney <schenney@chromium.org>
Reviewed-by: Robert Ma <robertma@chromium.org>
Commit-Queue: Stephen Chenney <schenney@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526761}

This is a fix around what looks like a race where `GetSensor` is called
twice and leads to `OnSensorCreated` to also be called twice. However,
the latter method doesn't expects this and has direct and indirects
checks to prevent it.

Bug: 798409
Change-Id: Ic2a3acd72eec2a49334551347cb3fd690377e3ed
Reviewed-on: https://chromium-review.googlesource.com/847003Reviewed-by: Tim Volodine <timvolodine@chromium.org>
Commit-Queue: Mounir Lamouri <mlamouri@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526760}

The equivalent metrics are now uploaded by ProcessMemoryMetricsEmitter, except
for all platforms rather than just macOS.

Bug: 720541
Change-Id: I36ed3a888bb0021744e12df84587c796cfd80865
Reviewed-on: https://chromium-review.googlesource.com/837656
Commit-Queue: Erik Chen <erikchen@chromium.org>
Reviewed-by: Siddhartha S <ssid@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Reviewed-by: Steven Holte <holte@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526759}

Bug: 789577
Change-Id: Idc8ac26283f863b280a50ec3036990630c6e81c3
Reviewed-on: https://chromium-review.googlesource.com/847853
Commit-Queue: Dmitry Gozman <dgozman@chromium.org>
Reviewed-by: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526758}

The only production implementations of the OnTunnelConnectCompleted,
OnBeforeTunnelRequest and OnTunnelHeadersReceived are empty. They are
used however in http_proxy_client_socket_pool_unittest.cc which has had
to have some test expectations removed.

This effectively reverts r294252 as DRP's custom HTTP headers are now
added by the DataReductionProxyNetworkDelegate instead of the
DataReductionProxyDelegate.

Bug: 721403
Change-Id: Ifb683e7fc28e97efc7182df1a33bf46f1b74a7ec
Reviewed-on: https://chromium-review.googlesource.com/846261Reviewed-by: Matt Menke <mmenke@chromium.org>
Reviewed-by: Tarun Bansal <tbansal@chromium.org>
Commit-Queue: Reilly Grant <reillyg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526757}

A recent change(*) switched a shmem buffer Clone() call from READ_WRITE
to READ_ONLY. New VIZ tab capture implementation depends on the special
semantics of the READ_WRITE argument, only in-so-much as it avoids a
crash due to what looks like a design flaw in base::SharedMemory
(details in crbug). This change reverts the call point, until a long-
term solution can be worked out.

Testing: Confirmed that the change works with both tab capture and
USB webcam capture.

* See: https://chromium-review.googlesource.com/820011 and
http://crbug.com/793503

Bug: 797470
Change-Id: I9bffdd0b71cbfa37b6b5bda03ce8863b560d8747
Reviewed-on: https://chromium-review.googlesource.com/848239Reviewed-by: Christian Fremerey <chfremer@chromium.org>
Commit-Queue: Yuri Wiitala <miu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526756}

It's a follow-up CL of https://chromium-review.googlesource.com/c/chromium/src/+/820854.

Bug: 791654
Test: Manual, browser_tests
Change-Id: I7e04a6839d1af6897a560a9875280ea7a2c18004
Reviewed-on: https://chromium-review.googlesource.com/833486
Commit-Queue: Xiaoqian Dai <xdai@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Reviewed-by: Mitsuru Oshima <oshima@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Wenzhao (Colin) Zang <wzang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#526755}