GitLab

As per request from lpromero, reverting since it might be failing a
test - see

http://build.chromium.org/p/chromium.memory.fyi/builders/Windows%20Tests%20%28tsan%29/builds/17153/steps/memory%20test%3A%20base_unittests/logs/stdio

[ RUN      ] HistogramTest.BasicTest
[3824:2780:1220/111332:4747750:FATAL:statistics_recorder.cc(270)] Check failed: !histograms_.

TBR=lpromero@chromium.org

> Reland "Add a HistogramRecorder class and use cases."
> 
> This CL adds a utility class to streamline the task of testing that metrics have been updated as expected. Specifically, the HistogramRecorder class allows a client to obtain the differential value of a given histogram in the interval since the given HistogramRecorder instance was created.
> 
> This reverts commit b957026f.
> 
> BUG=232414
> TBR=rsleevi, mark, groby, thakis
> 
> Review URL: https://codereview.chromium.org/19866004

TBR=lpromero@chromium.org

Review URL: https://codereview.chromium.org/120273002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242176 0039d316-1c4b-4281-b951-d872f2087c98

BUG=315922
R=fischman@chromium.org

Review URL: https://codereview.chromium.org/65803002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242175 0039d316-1c4b-4281-b951-d872f2087c98

BUG=330105
TBR=oleg@chromium.org

Review URL: https://codereview.chromium.org/103343008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242174 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
R=abarth@chromium.org

Review URL: https://codereview.chromium.org/120083002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242171 0039d316-1c4b-4281-b951-d872f2087c98

Make ExtensionPopup a TabStripModelObserver.
Close the ExtensionPopup's Widget on ActiveTabChanged.
Add an interactive_ui_test for this behavior.

BUG=303479
TEST=Extension popups close on switching tabs via CTRL+TAB, CTRL+1, CTRL+2, etc.
R=sky@chromium.org,jyasskin@chromium.org

Review URL: https://codereview.chromium.org/119633002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242170 0039d316-1c4b-4281-b951-d872f2087c98

Getting closer to get rid of GetDefaultProfile. This is now removing the temporary instrumentation from stevenjb which was adding a check to see that GetDefaultProfile does not get called before a Profile was created. I discussed with dpolukhin and nkostylev and instead of renaming this to something like "WasProfileCreated" (or similar) the instrumentation should get removed (since it was temporary to begin with).

BUG=322682
TEST=covered by existing unit tests

Review URL: https://codereview.chromium.org/119593002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242169 0039d316-1c4b-4281-b951-d872f2087c98

In RenderWidgetHostViewAura::UpdateCursorIfOverSelf(),
the point used for the hit test against the root window
needs to be in the root window's coordinate space.

This CL also removes some Windows-specific code
originally added for issue 292379 which is no longer
needed.

BUG=295125
TEST=RenderWidgetHostViewAuraTest.UpdateCursorIfOverSelf

Review URL: https://codereview.chromium.org/25336003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242168 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: https://codereview.chromium.org/117323003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242164 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242163 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=164232:164240&mode=html
TBR=
BUG=

Review URL: https://codereview.chromium.org/115693006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242162 0039d316-1c4b-4281-b951-d872f2087c98

BUG=261628

Review URL: https://codereview.chromium.org/118713004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242161 0039d316-1c4b-4281-b951-d872f2087c98

BUG=329951

Review URL: https://codereview.chromium.org/113413003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242159 0039d316-1c4b-4281-b951-d872f2087c98

- Refactor the Write code to avoid recursion
- Refactor the read code to avoid recursion
- With read code refactored, hook it up appropriately with connection flow
  to avoid re-entrancy of DoConnectLoop

Review URL: https://codereview.chromium.org/79673003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242158 0039d316-1c4b-4281-b951-d872f2087c98

Also added a unit test for incognito / guest mode.

BUG=chromium:328539
TEST=unit

Review URL: https://codereview.chromium.org/106373003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242157 0039d316-1c4b-4281-b951-d872f2087c98

Rather than passing a code/message pair through the entire open call
sequence, let the data loss status be assigned to the callbacks object
which is already passed along.

R=dgrogan
BUG=329931

Review URL: https://codereview.chromium.org/119343004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242155 0039d316-1c4b-4281-b951-d872f2087c98

There's no need to check whether the parent is accessible to the current context.

BUG=305000

Review URL: https://codereview.chromium.org/117843003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242154 0039d316-1c4b-4281-b951-d872f2087c98

Complete chrome.isolate to work on all platforms by extracting relevant data
from browser_tests.isolate. Then make browser_tests.isolate 'includes'
chrome.isolate and remove the duplicate entries.

The story for bundling lib32 and lib64 NaCl libraries will have to cleared up
since it depends on the OS running the test and on the build itself.

Note: interactive_ui_tests.isolate doesn't actually start the chrome executable
so it is not bundled.

R=csharp@chromium.org
NOTRY=true
BUG=

Review URL: https://codereview.chromium.org/106823003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242152 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/115503003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242150 0039d316-1c4b-4281-b951-d872f2087c98

 * Wire up HMAC SHA224, HMAC SHA384, HMAC SHA512 for NSS.
 * When generating an HMAC key and the key length was not specified use
   the underlying block size for the key length (as this is the maximum
   length of the key). The OpenSSL version had a bug where it would not
   select the block size.
 * Fix a test which intended to check randomness.

BUG=245025

Review URL: https://codereview.chromium.org/108653015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242149 0039d316-1c4b-4281-b951-d872f2087c98

TBR=wad@chromium.org
BUG=244472

Review URL: https://codereview.chromium.org/101283003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242148 0039d316-1c4b-4281-b951-d872f2087c98

BUG=321643

Review URL: https://codereview.chromium.org/93733005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242147 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242146 0039d316-1c4b-4281-b951-d872f2087c98

This change pulls in the w3c's test repos so that we can start running
the tests in Blink. The initial change just pulls in versions that
are empty, so we will clone the repos but not actually run any tests;
this can be used for local testing and development and is a minor
incremental step as the process for importing tests is worked out and
polished. A subsequent change will start pulling in revisions
with selected subdirectories populated.

R=ojan@chromium.org, abarth@chromium.org
BUG=242329

Review URL: https://codereview.chromium.org/107503004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242145 0039d316-1c4b-4281-b951-d872f2087c98

The typedef confused me when I was reading the code, so here I am, fixing it.

BUG=None
R=jar@chromium.org

Review URL: https://codereview.chromium.org/120013002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242143 0039d316-1c4b-4281-b951-d872f2087c98

TBR=joi@chromium.org
BUG=329295
NOTRY=true

Review URL: https://codereview.chromium.org/119933003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242142 0039d316-1c4b-4281-b951-d872f2087c98

On Android, the sandbox logic is put into a shared library and the linker errors
out on having a TEXTREL.

The LDR pseudo-instruction directs the assembler to place the address of the
label in a "literal pool"[1], and then use a PC-relative load from that pool to
get the address during execution. The pool is responsible for generating
the R_ARM_RELATIVE TEXTREL.

Using the ADR instruction[2] does not produce the TEXTREL. This instruction
directs the assembler to calculate the PC-relative address to the label using an
immediate.

The text makes this difference more clear:

...
   8:   4805            ldr     r0, [pc, #20]   ; (20 <SyscallAsm+0x20>)
   a:   e007            b.n     1c <SyscallAsm+0x1c>
...
  1c:   bd80            pop     {r7, pc}
  1e:   0000            .short  0x0000
  20:   0000001c        .word   0x0000001c

Versus:

...
   8:   4804            ldr     r0, [pc, #16]   ; (1c <SyscallAsm+0x1c>)
   a:   e007            b.n     1c <SyscallAsm+0x1c>
...
  1c:   bd80            pop     {r7, pc}
  1e:   bf00            nop


[1] http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0041c/Babbfdih.html
[2] http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0040d/Cihdhgbe.html

BUG=308763

Review URL: https://codereview.chromium.org/119453002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242141 0039d316-1c4b-4281-b951-d872f2087c98

A follow on to crrev.com/112293002 - replace holding an extra
pointer to determine state with querying an existing member object.

R=dgrogan

Review URL: https://codereview.chromium.org/110983004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242140 0039d316-1c4b-4281-b951-d872f2087c98

BUG=330212
TEST=Manual, see bug
R=jamescook@chromium.org

Review URL: https://codereview.chromium.org/101193007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242139 0039d316-1c4b-4281-b951-d872f2087c98

R=caitkp@chromium.org
BUG=329295

Review URL: https://codereview.chromium.org/119433004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242136 0039d316-1c4b-4281-b951-d872f2087c98

BUG=308318,309131,308681
TEST=manual, added MergeSessionTest.MergeSessionThrottle

Review URL: https://codereview.chromium.org/118733002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242135 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/118343003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242134 0039d316-1c4b-4281-b951-d872f2087c98

BUG=158004

Review URL: https://codereview.chromium.org/116093007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242133 0039d316-1c4b-4281-b951-d872f2087c98

The referrer policy is not ignored, but already read out of the extra
data before willSendRequest()

BUG=none
R=dmichael@chromium.org

Review URL: https://codereview.chromium.org/112373003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242132 0039d316-1c4b-4281-b951-d872f2087c98

Add GetNSSCertDatabaseForResourceContext to nss_context to get one.

BUG=218643

Review URL: https://codereview.chromium.org/111273002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242131 0039d316-1c4b-4281-b951-d872f2087c98

This does NOT cause Chrome to use QUIC 13, that will come later.
Refactors the mechanims that QUIC tests use to to create packets.

Review URL: https://codereview.chromium.org/110853010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242130 0039d316-1c4b-4281-b951-d872f2087c98

If we fail to create the java MediaCodec, we should early return in VideoCodecBridge::CreateDecoder(). Otherwise, we will hit NPE.

BUG=329742

Review URL: https://codereview.chromium.org/99423010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242129 0039d316-1c4b-4281-b951-d872f2087c98

Requests using all idempotent HTTP methods (See RFC 2616) should be
retried after receiving a bypass message from the data reduction
proxy. Before this CL, only GET was being retried.

BUG=327740

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=241714

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=241927

Review URL: https://codereview.chromium.org/99283006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242128 0039d316-1c4b-4281-b951-d872f2087c98

I'm going to need to install a another singleton client for
transients. Rather than touch all places to explicitly set n clients
I'm creating Environment to centralize common setup and teardown.

BUG=none
TEST=none
R=ben@chromium.org

Review URL: https://codereview.chromium.org/105813013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242126 0039d316-1c4b-4281-b951-d872f2087c98

Unify the code that dispatches touch events to native, making it easier to
follow touch event flow through the pipeline.  Also guard against nested loops
for touch dispatch, triggered by synchronous acks when there are multiply
enqueud touch events.

Review URL: https://codereview.chromium.org/119333002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242124 0039d316-1c4b-4281-b951-d872f2087c98

consistency

Review URL: https://codereview.chromium.org/119043002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@242123 0039d316-1c4b-4281-b951-d872f2087c98