GitLab

where it would be available to all profiles, including Incognito, to
ProfileImplIOData, where it would only be available to non-
Incognito profiles. 

In the process, (1) refactored and simplified DataReductionProxyUsageStats
so that it could live on the IO thread. (2) Added a keyed service called
DataReductionProxyChromeSettings, so chrome browsers on all platforms
could use the same pattern.

BUG=396695

Review URL: https://codereview.chromium.org/412143009

Cr-Commit-Position: refs/heads/master@{#288288}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288288 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=True

Review URL: https://codereview.chromium.org/448823002

Cr-Commit-Position: refs/heads/master@{#288287}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288287 0039d316-1c4b-4281-b951-d872f2087c98

Three related changes:
* Only run idle tasks in the queue, not newly added ones.
* Move g_request_pending logic to SharedRendererState. This
  allows pending tasks to be cancelled correctly.
* Wait for kModeProcess to happen first before trying to
  request another to avoid a queue of pending tasks.

BUG=401712

Review URL: https://codereview.chromium.org/444173003

Cr-Commit-Position: refs/heads/master@{#288285}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288285 0039d316-1c4b-4281-b951-d872f2087c98

Bad media may result in splice frames way in the past.  We need
to avoid crashing in this case.  See the bug for more details.

BUG=400442
TEST=new media test.

Review URL: https://codereview.chromium.org/440803002

Cr-Commit-Position: refs/heads/master@{#288284}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288284 0039d316-1c4b-4281-b951-d872f2087c98

Before this, a task is added to the idle queue each time in flush and
ScheduleMoreIdleWork. This means that as long as there is idle work, the
size of the queue grows by one in each flush.

Fix this by not adding any idle tasks if there is already one pending.

BUG=401712

Review URL: https://codereview.chromium.org/450793002

Cr-Commit-Position: refs/heads/master@{#288283}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288283 0039d316-1c4b-4281-b951-d872f2087c98

It updates several comments and renames member variables to increase clarity and
better match the style guide.

R=wtc@chromium.org

Review URL: https://codereview.chromium.org/448293002

Cr-Commit-Position: refs/heads/master@{#288282}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288282 0039d316-1c4b-4281-b951-d872f2087c98

I didn't move cocoa impl because that's not necessary.

BUG=380438
R=sky@chromium.org

Review URL: https://codereview.chromium.org/447243004

Cr-Commit-Position: refs/heads/master@{#288281}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288281 0039d316-1c4b-4281-b951-d872f2087c98

SyncManager::Init is implemented in several places, takes a whole lot of
args, and in the near future is going to get a few more.

In some cases, many of the args are optional and it is tedious to pass a
bunch of NULLs or empty strings.  Use a struct instead of an arg list.

BUG=

Review URL: https://codereview.chromium.org/451743002

Cr-Commit-Position: refs/heads/master@{#288280}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288280 0039d316-1c4b-4281-b951-d872f2087c98

Cr-Commit-Position: refs/heads/master@{#288279}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288279 0039d316-1c4b-4281-b951-d872f2087c98

https://skia.googlesource.com/skia/+log/0aeea6d344f12e35e29a79f4bbc48af88f913204..acc125860053d0a65bc7b67dd1b886aa7932d292

CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:linux_layout_rel
TBR=bungeman@google.com
BUG=

Review URL: https://codereview.chromium.org/453623002

Cr-Commit-Position: refs/heads/master@{#288278}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288278 0039d316-1c4b-4281-b951-d872f2087c98

It removes the transport connect states from MockSSLCLientSocket Connect()
and renames some member variables for clarity.

R=wtc@chromium.org,rsleevi@chromium.org

BUG=398967

Review URL: https://codereview.chromium.org/454513002

Cr-Commit-Position: refs/heads/master@{#288277}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288277 0039d316-1c4b-4281-b951-d872f2087c98

BUG=398337,401716
NOTRY=true
TBR=zmo@chromium.org

Review URL: https://codereview.chromium.org/455503002

Cr-Commit-Position: refs/heads/master@{#288276}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288276 0039d316-1c4b-4281-b951-d872f2087c98

Previous implementation did the parsing of string to base::Value
on blocking pool, and that of base::Value to specific data types
either on UI thread or on yet another post to blocking pool.

The previous implementation is slightly inefficient and moreover
involves a subtle bug 284244.

BUG=284244, 401843

This CL once landed as r288017 (Patch Set 4) and reverted.
This is the second try.

Review URL: https://codereview.chromium.org/442193002

Cr-Commit-Position: refs/heads/master@{#288275}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288275 0039d316-1c4b-4281-b951-d872f2087c98

BUG=399567

Review URL: https://codereview.chromium.org/382143005

Cr-Commit-Position: refs/heads/master@{#288273}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288273 0039d316-1c4b-4281-b951-d872f2087c98

2. Adds a couple of functions to return different fingerprint value for tamper detection.

3. Adds a function to remove Chrome-Proxy header's fingerprint from its header values, and return the rest of header values.

4. Changes HasDataReductionProxyViaHeader to also tell 
whether data reduction proxy Via header occurs at the last 
or not.


BUG=381907

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=287561

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=287855

Review URL: https://codereview.chromium.org/387353003

Cr-Commit-Position: refs/heads/master@{#288272}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288272 0039d316-1c4b-4281-b951-d872f2087c98

Currently, NaClListener has supports two modes, SFI and non-SFI.
As a preparation of newlib switching of non-SFI mode, this CL splits into
each delegate class, so that, we can easily build non-SFI related code
only by PNaCl toolchain.

BUG=358465
TEST=Ran trybots.
CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:linux_rel_precise32

Review URL: https://codereview.chromium.org/439713002

Cr-Commit-Position: refs/heads/master@{#288271}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288271 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true

Review URL: https://codereview.chromium.org/430943003

Cr-Commit-Position: refs/heads/master@{#288270}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288270 0039d316-1c4b-4281-b951-d872f2087c98

This refactoring CL makes it easier to change NTP designs. The key
variable to control this switch is configData.ntpDesignName, which was
added in
https://chromiumcodereview.appspot.com/447033002/

We also added titleTextAlign and titleTextFade properties. The support
for these fields in Most Visited <iframes> were are added in
https://chromiumcodereview.appspot.com/448793003/

BUG=400346
TBR=jhawkins

Review URL: https://codereview.chromium.org/447243003

Cr-Commit-Position: refs/heads/master@{#288269}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288269 0039d316-1c4b-4281-b951-d872f2087c98

Promote the chrome.hid API to the stable channel.

BUG=305300

Review URL: https://codereview.chromium.org/448143003

Cr-Commit-Position: refs/heads/master@{#288268}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288268 0039d316-1c4b-4281-b951-d872f2087c98

This change detects buggy ARM chips and disables the Poly1305 code only on
those chips.

BUG=341598

Review URL: https://codereview.chromium.org/442863003

Cr-Commit-Position: refs/heads/master@{#288267}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288267 0039d316-1c4b-4281-b951-d872f2087c98

> Enable ServiceDiscoveryClientMdns.
> 
> BUG=349645
> 
> Review URL: https://codereview.chromium.org/287923002

BUG=391907

TBR=vitalybuka@chromium.org

Review URL: https://codereview.chromium.org/452843002

Cr-Commit-Position: refs/heads/master@{#288266}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288266 0039d316-1c4b-4281-b951-d872f2087c98

This implements the same basic logic we use in GTest-based tests to
retry tests that fail during a run. By default, we will not retry
anything (this matches what gtests do when run not in botmode).

If --retry-limit is passed on the command line, we will retry each
failing test up to that number of times. So, to enable retries on
the bots by default they will have to pass this arg.

R=dtu@chromium.org, tonyg@chromium.org
BUG=398027

Review URL: https://codereview.chromium.org/426123002

Cr-Commit-Position: refs/heads/master@{#288265}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288265 0039d316-1c4b-4281-b951-d872f2087c98

Remove leading zero to fix literal that was not meant to be octal in
pwg_encoder.cc.

BUG=401714

Review URL: https://codereview.chromium.org/447233004

Cr-Commit-Position: refs/heads/master@{#288264}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288264 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/449923004

Cr-Commit-Position: refs/heads/master@{#288263}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288263 0039d316-1c4b-4281-b951-d872f2087c98

* Android Checkin server started handling OAuth2 tokens for checkin
  very recently and is using the added scope to do it.

BUG=374969
R=zea@chromium.org

Review URL: https://codereview.chromium.org/452603002

Cr-Commit-Position: refs/heads/master@{#288262}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288262 0039d316-1c4b-4281-b951-d872f2087c98

This patch does three related things:

1. Plumbs an optional IdentityProvider through 
ExtensionUpdater / ExtensionDownloader.

2. Adds support in ExtensionDownloader for OAuth2 token-
based authentication to protected Webstore download URLs,
using the optional IdentityProvider as its account ID and 
token source. Bonus points for general cleanup of the 
credential iteration code in ExtensionDownloader.

3. Wires up Chrome's ExtensionDownloader (via
ExtensionService) with a ProfileIdentityProvider so that
it may authenticate as the Chrome-signed-in user via OAuth2.

BUG=376553,122763

Review URL: https://codereview.chromium.org/434493002

Cr-Commit-Position: refs/heads/master@{#288261}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288261 0039d316-1c4b-4281-b951-d872f2087c98

For future use with favicons, remove reference to "thumbnail". There is no functionality change in this CL, just a rename.

BUG=388747
TEST=ImageManager*

Review URL: https://codereview.chromium.org/445753004

Cr-Commit-Position: refs/heads/master@{#288260}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288260 0039d316-1c4b-4281-b951-d872f2087c98

Moved recvmsg log to VPLOG

BUG=377403

Review URL: https://codereview.chromium.org/414243002

Cr-Commit-Position: refs/heads/master@{#288259}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288259 0039d316-1c4b-4281-b951-d872f2087c98

The gfx_unittests CanvasTestMac.* passes on Mac Views.
(can also run the tests with --enable-harfbuzz-rendertext)

BUG=391843
TEST=CanvasTestMac.* passes, no Mac text rendering changes.
R=asvitkine@chromium.org,ckocagil@chromium.org

Review URL: https://codereview.chromium.org/440213004

Cr-Commit-Position: refs/heads/master@{#288255}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288255 0039d316-1c4b-4281-b951-d872f2087c98

BUG=399922

Review URL: https://codereview.chromium.org/425713003

Cr-Commit-Position: refs/heads/master@{#288254}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288254 0039d316-1c4b-4281-b951-d872f2087c98

I can't actually reproduce the bug as described by the reporter; this is
a blind change that expands the minimum width which should theoretically
fix the problem though.

BUG=400144

Review URL: https://codereview.chromium.org/453573003

Cr-Commit-Position: refs/heads/master@{#288253}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288253 0039d316-1c4b-4281-b951-d872f2087c98

In file mips_linux_syscalls.h are definitions of syscall
numbers for O32 ABI, so this check is needed in order for
Mips architectures with other ABIs to work properly.

BUG=400684
TEST=compile sandbox_linux_unittest for MIPS32 and MIPS64

Review URL: https://codereview.chromium.org/446213003

Cr-Commit-Position: refs/heads/master@{#288252}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288252 0039d316-1c4b-4281-b951-d872f2087c98

Any leak of boringssl.gyp (which included unittests) ends up pulling in base/
and breaks the WebRTC build.

However, switching targets to boringssl_lib.gypi:boringssl breaks GYP because
then there are two targets with the same name (boringssl.gyp:boringssl and
boringssl_lib.gypi:boringssl).

This change gives up and splits the unittests into a different .gyp file.

BUG=none

Review URL: https://codereview.chromium.org/447213004

Cr-Commit-Position: refs/heads/master@{#288251}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288251 0039d316-1c4b-4281-b951-d872f2087c98

Add mojo/example/apptest with a service and client.
Add example_apptest.cc with a GTest Mojo application.

These tests can be run from out/<Debug|Release>/ via:
./mojo_shell --origin=file://`pwd` --disable-cache mojo:mojo_example_apptests

BUG=392646
TEST=The new mojo_example_apptests passes.
R=ben@chromium.org

Review URL: https://codereview.chromium.org/399653004

Cr-Commit-Position: refs/heads/master@{#288250}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288250 0039d316-1c4b-4281-b951-d872f2087c98

Cr-Commit-Position: refs/heads/master@{#288249}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288249 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=179717:179742&mode=html
TBR=jbroman@chromium.org
BUG=

Review URL: https://codereview.chromium.org/443133004

Cr-Commit-Position: refs/heads/master@{#288248}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288248 0039d316-1c4b-4281-b951-d872f2087c98

QUIC: don't keep around state for streams created by a request
w/no body. Waiting for a stream frame w/ FIN that will never arrive, and
the closed stream map grows without bound.  Not flag protected.

Merge internal change: 72338396
https://codereview.chromium.org/447093002/


Default socket receive buffer to 256K bytes and remove unnecessary
CongestionFeedbackFrames from tests.

Merge internal change: 72338299
https://codereview.chromium.org/447083002/


Include the socket receive buffer in the connection handshake.
PRESENCE_OPTIONAL

The goal is to eventually remove the TcpCongestionFeedbackFrame.
Currently this sends the same value for every packet. All it contains is
the tcp_receive_window (now called socket_receive_buffer) which is fixed
at 256000. So it can be sent in the connection establishment and then
never retransmitted.

Because only one type of CongestionFeedbackFrame can be sent with an
ack, this frees up space for a TimestampFrame which is the motivation
for removing the TcpCongestionFeedbackFrame.

Merge internal change: 72334240
https://codereview.chromium.org/446283002/


Improve QUIC's SendAlgorithmSimulator to allow setting the delayed ack
timer and add a new test of a second BBR flow starting after a first
flow has stabilized.

Merge internal change: 72332690
https://codereview.chromium.org/449713002/


QUIC congestion option that forces ICWND to 10.  Not flag protected.

Merge internal change: 72323017
https://codereview.chromium.org/446253002/


Allow receiving a QUIC CongestionFeedbackFrame that is not kTCP.

Currently the only congestion feedback frames that are sent are TCP, but
I want to send timestamp packets in the next version, so the send
algorithms should not assume that the incoming congestion feedback frame
is kTCP.

Merge internal change: 72322097
https://codereview.chromium.org/443313002/


QUIC - minor cleanup to match internal source tree.

* Formatting cleanup and .get() != NULL changes.

Merge internal change: 72204223
https://codereview.chromium.org/447973004/


Bring back InterArrivalReceiver and rename it to TimestampeReceiver.
This is not in production.

Merge internal change: 72170964
https://codereview.chromium.org/449693002/


Move Quic AppendTimestampFrame method out of AppendCongestionFrame.
Also, remove obsolete TODO's and rename InterArrival to Timestamp.
Not used in production.

Merge internal change: 72148809
https://codereview.chromium.org/446063005/

R=rch@chromium.org

Review URL: https://codereview.chromium.org/447093004

Cr-Commit-Position: refs/heads/master@{#288247}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288247 0039d316-1c4b-4281-b951-d872f2087c98

BUG=385623

Review URL: https://codereview.chromium.org/357323002

Cr-Commit-Position: refs/heads/master@{#288245}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288245 0039d316-1c4b-4281-b951-d872f2087c98

These tests use ContentLayer, so change them to use PictureLayer
instead and stop explicitly disabling impl-side painting.

R=enne
BUG=401492

Review URL: https://codereview.chromium.org/443353003

Cr-Commit-Position: refs/heads/master@{#288244}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288244 0039d316-1c4b-4281-b951-d872f2087c98

In this context, when needed means when Screen Orientation API is being
used.

BUG=400158

Review URL: https://codereview.chromium.org/398013002

Cr-Commit-Position: refs/heads/master@{#288243}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@288243 0039d316-1c4b-4281-b951-d872f2087c98