GitLab

Follow-up to https://codereview.chromium.org/1128153003/. Rather than query all
keys and checking the slot afterwards, just look up keys in the slot directly.

BUG=478777

Review URL: https://codereview.chromium.org/1139533002

Cr-Commit-Position: refs/heads/master@{#329498}

BUG=369661
TEST=No whammies!

Review URL: https://codereview.chromium.org/1136663004

Cr-Commit-Position: refs/heads/master@{#329497}

Chrome Apps cannot switch UI language, so we need to initialize apps after
app_locale is set correctly.

This CL splits ProfileImpl::OnPrefsLoaded into two steps
(the former and OnLocaleReady), adding intermediate call
to switch to correct locale. Also Profile is added as
parameter to InputMethodChanged() observers.
chromeos::locale_util::SwitchLanguage() now updates default
input methods for user profile, not currently active set.

TBR=dmazzoni@chromium.org

BUG=469268
TEST=manual

Review URL: https://codereview.chromium.org/1055863002

Cr-Commit-Position: refs/heads/master@{#329496}

BUG=177163

Review URL: https://codereview.chromium.org/1128253005

Cr-Commit-Position: refs/heads/master@{#329495}

supported in cronet to fix a crash bug in cronet.

BUG=487355
R=mef@chromium.org, rch@chromium.org

Review URL: https://codereview.chromium.org/1138943002

Cr-Commit-Position: refs/heads/master@{#329494}

SPDY's multiplexing is incompatible with the renego + client auth hack. The old
implementation also assumed that the SpdyStream was still alive which isn't
true. We could attempt save a copy of it, but that doesn't resolve the
fundamental incompatibility of the hack. Instead
https://codereview.chromium.org/1131763002/ disables renegotiation, so this
codepath should be impossible.

BUG=462283

Review URL: https://codereview.chromium.org/1128873008

Cr-Commit-Position: refs/heads/master@{#329493}

I think this regressed due to:
https://codereview.chromium.org/1119923002

BUG=486851

Review URL: https://codereview.chromium.org/1141573003

Cr-Commit-Position: refs/heads/master@{#329492}

That state isn't needed. When an operation fails in OpenSSL, SSL_get_error
tells you exactly what event it was waiting for.

BUG=none

Review URL: https://codereview.chromium.org/1136813002

Cr-Commit-Position: refs/heads/master@{#329491}

BUG=477106

Review URL: https://codereview.chromium.org/1126383003

Cr-Commit-Position: refs/heads/master@{#329490}

This allows experimenting with increasing the variations seed fetch
period.

BUG=483057

Review URL: https://codereview.chromium.org/1119893003

Cr-Commit-Position: refs/heads/master@{#329489}

When FastNavigationProfileExtender was first written, it was intended to
navigate to the top 10,000 sites in Alexa. Many of these sites would fail to
load for unexpected reasons. Furthermore, the exceptions raised by Telemetry
when these errors occured were frequently less than useful. As a result,
FastNavigationProfileExtender was catching many fatal exceptions but treating
them as non-fatal exceptions.

When FastNavigationProfileExtender is used on Windows, and Chrome crashes in a
way that invokes the Windows Error dialog, attempts by Telemetry to communicate
with the crashed instance of Chrome can cause Python to stall. I did not
investigate the details, but I suspect this is related to the implementation of
the thirdparty websocket library that Telemetry uses.

This CL makes the FastNavigationProfileExtender treat fatal exceptions
appropriately, instead of ignoring them.

BUG=477375

Review URL: https://codereview.chromium.org/1124543006

Cr-Commit-Position: refs/heads/master@{#329488}

This pulls in a new API that allows chrome to stop including private
PDFium headers.  Chrome is modified to use it, and the inculde rules
are updated to reflect a "public header only" policy.

BUG=486818

Review URL: https://codereview.chromium.org/1129223005

Cr-Commit-Position: refs/heads/master@{#329487}

Update AshPopupAlignmentDelegate calculations of its work area.

When windows are docked this reduces the work area for windows.
AshPopupAlignmentDelegate uses the work area to determine the positioning of
notifications. Update this calculation to include the docked region.

TEST=AshPopupAlignmentDelegateTest.DockedWindow
BUG=284574

Review URL: https://codereview.chromium.org/1121893004

Cr-Commit-Position: refs/heads/master@{#329486}

This hasn't been thoroughly tested yet. Once more pages are ready
and the rest of the structure is in place I will look more closely
at how pages are selected

BUG=485381

Review URL: https://codereview.chromium.org/1141603003

Cr-Commit-Position: refs/heads/master@{#329485}

BUG=467550

Review URL: https://codereview.chromium.org/1126563003

Cr-Commit-Position: refs/heads/master@{#329484}

Exporting a class on Windows forces instantiation of default constructor,
copy constructor, and sundry similar other functions.  The clang plugin then
complains about these not being out-of-line, even if they're not referred to
from anywhere.  Since having this only on Windows is annoying, we're suppressing
these warnings on exported classes.

In this case however, the inner classes STSState and PKPState aren't exported,
and their copy constructors are only instantiated because their outer class
_is_ exported and has STSState and PKPState members.  This cannot be easily
suppressed, and it's also the only place in the whole codebase where this
happens.

So just export these two structs to make the plugin happy in Windows
component builds.

BUG=467287,483986

Review URL: https://codereview.chromium.org/1124233005

Cr-Commit-Position: refs/heads/master@{#329483}

Intent.parseUri can throw other excpetions like NumberFormatException.
So we need to catch Exception to avoid crash.

BUG=484336

Review URL: https://codereview.chromium.org/1134163002

Cr-Commit-Position: refs/heads/master@{#329482}

We recently introduced SchemeIsCryptographic() and IsOriginSecure(),
which are meant to replace SchemeIsSecure().

IsOriginSecure() roughly means "do we trust this content not to be
tampered with before it reaches the user?" [1] This is a higher-level
definition that corresponds to the new "privileged contexts" spec. [2]

SchemeIsCryptographic() [3] is close to the old definition of
SchemeIsSecure(), and literally just checks if the scheme is a
cryptographic scheme (HTTPS or WSS as of right now). The difference is
that SchemeIsCryptographic() will not consider filesystem URLs secure.

[1] https://code.google.com/p/chromium/codesearch#chromium/src/content/public/common/origin_util.h&sq=package:chromium&type=cs&l=19&rcl=143099866
[2] https://www.chromium.org/Home/chromium-security/prefer-secure-origins-for-powerful-new-features and https://w3c.github.io/webappsec/specs/powerfulfeatures/
[3] https://code.google.com/p/chromium/codesearch#chromium/src/url/gurl.h&sq=package:chromium&type=cs&l=250&rcl=1430998666

BUG=362214

Review URL: https://codereview.chromium.org/1135693002

Cr-Commit-Position: refs/heads/master@{#329481}

This patch removes error_console accessor from ExtensionSystem.
It can be its own browser context keyed service. And it can be
built by its new factory.

BUG=

Review URL: https://codereview.chromium.org/1139643005

Cr-Commit-Position: refs/heads/master@{#329480}

Revert of Adding total available size of heap in v8 isolate memory dump provider. (patchset #1 id:1 of https://codereview.chromium.org/1129403003/)

Reason for revert:
The HeapStatistics::total_available_size returned by v8 API doesn't return the correct value right now. It only returns the sum of available size of 4 spaces only. So, this should land after the API gets fixed.

Original issue's description:
> Adding total available size of heap in v8 isolate memory dump provider.
>
> The total available size is now returned by GetHeapStatistics api.
> This CL uses the value in the dump provider to show the value in
> other_spaces segment of the memory dump.
>
> BUG=481504
>
> Committed: https://crrev.com/9ad94e6e06201059c08d67b9d9ab4a30a766b541
> Cr-Commit-Position: refs/heads/master@{#329303}

TBR=jochen@chromium.org,picksi@chromium.org,primiano@chromium.org
BUG=481504

Review URL: https://codereview.chromium.org/1134233002

Cr-Commit-Position: refs/heads/master@{#329479}

BUG=487056

Review URL: https://codereview.chromium.org/1124383005

Cr-Commit-Position: refs/heads/master@{#329478}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/9a048f2..80f95a7

TBR=dpranke@chromium.org,alancutter@chromium.org

Review URL: https://codereview.chromium.org/1140753002

Cr-Commit-Position: refs/heads/master@{#329477}

Note that I did not add the checkbox compiled_resources.gyp to third_party/closure_compiler/compiled_resources.gyp because there are still some errors to dig through.

BUG=486240

Review URL: https://codereview.chromium.org/1112403006

Cr-Commit-Position: refs/heads/master@{#329476}

EasyUnlockTpmKeyManager should not start poking TPM on worker thread
before TPM initialization is finished on IO thread in order to
prevent IO thread being blocked during startup waiting for easy
signin key pair being created.

BUG=464524
TEST=Easy signin works after initializing easy unlock and signin in
once.

Review URL: https://codereview.chromium.org/1138123003

Cr-Commit-Position: refs/heads/master@{#329475}

This brings paper-icon-button v0.8.0 into third_party/polymer/v0_8.

BUG=485381

Review URL: https://codereview.chromium.org/1136293002

Cr-Commit-Position: refs/heads/master@{#329474}

BUG=380227

Review URL: https://codereview.chromium.org/1130793006

Cr-Commit-Position: refs/heads/master@{#329473}

This measures time spent in DistillPageImpl(), which is mostly
ExecuteJavaScript(). This includes JavaScript compilation and execution.

We already have DomDistiller.Time.DistillationTotal, but it is measured
within JavaScript.

Review URL: https://codereview.chromium.org/1131493005

Cr-Commit-Position: refs/heads/master@{#329472}

Add two unittests to test the condition that triggers
ERR_REQUEST_FOR_SECURE_RESOURCE_OVER_INSECURE_QUIC.

BUG=

Review URL: https://codereview.chromium.org/1135103003

Cr-Commit-Position: refs/heads/master@{#329471}

Fix infinite loop bug on javascript.

BUG=486879

Review URL: https://codereview.chromium.org/1132113005

Cr-Commit-Position: refs/heads/master@{#329470}

Slimming paint is now much faster than PicturePile backed recording
for impl-side painting. It's what we're going to ship when we enable
impl-side, so make it the default. Left the disable switch just in case
to try out PicturePile if we need to for some reason until we turn
on impl-side by default.

R=piman@chromium.org, sky
BUG=466426

Review URL: https://codereview.chromium.org/1141583003

Cr-Commit-Position: refs/heads/master@{#329469}

This can be merged to ValidateAndMeasurePage method.

BUG=455391, 470147
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect

Review URL: https://codereview.chromium.org/1124033004

Cr-Commit-Position: refs/heads/master@{#329468}

Revert of Cache IsNonBrowserProcess values, so we only take a loader-lock the first time. (patchset #4 id:60001 of https://codereview.chromium.org/1132473003/)

Reason for revert:
Possibly causing mysterious fails on the Win DBG bots:
https://build.chromium.org/p/chromium.webkit/builders/Win7%20%28dbg%29/builds/16344

http://build.chromium.org/p/chromium.win/builders/Win7%20Tests%20%28dbg%29%281%29/builds/37896

Original issue's description:
> Cache IsNonBrowserProcess values, so we only take a loader-lock the first time.
>
> BUG=485656,477137
>
> Committed: https://crrev.com/78d41b5d4e59dc7d52d7e7d6530fdd0a78b36d33
> Cr-Commit-Position: refs/heads/master@{#329246}

TBR=siggi@chromium.org,grt@chromium.org,csharp@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=485656,477137

Review URL: https://codereview.chromium.org/1134253002

Cr-Commit-Position: refs/heads/master@{#329467}

Only HTTP/1.1 (and below) sockets may renegotiate. This fix a
crash because SpdyHttpStream didn't account for this properly.
(And can't as the renego + client auth hack is inherently
incompatible with multiplexing.)

Tested manually against hacked up Go servers:

- HTTP/1.1 server which renegotiates with client auth before
  sending a response on a fresh socket.

- Same as above but with a reused socket (the server only
  requests renego when fetching /auth).

- HTTP/2 which incorrectly renegotiates with client auth upon
  requesting /auth. Verified that we get ERR_SSL_PROTOCOL_ERROR
  and not crash.

- HTTP/1.1 server which does two handshakes in a row with Finished
  and HelloRequest in the same record. NSS and BoringSSL differ in
  their behavior here, but in neither port should we miss the
  renego.

BUG=484543,462283

Review URL: https://codereview.chromium.org/1131763002

Cr-Commit-Position: refs/heads/master@{#329466}

Error buttons should not be visible in stable.
Error buttons should be visible in dev.

BUG=463245

Review URL: https://codereview.chromium.org/1132493004

Cr-Commit-Position: refs/heads/master@{#329465}

This brings paper-item v0.8.2 into third_party/polymer/v0_8.

I got a presubmit warning about not updating README.chromium.
Will that actually be a problem in the CQ?

BUG=485381

Review URL: https://codereview.chromium.org/1124133004

Cr-Commit-Position: refs/heads/master@{#329464}

Revert of [MemSheriff] Suppress use-after-free in content::RenderFrameImpl::~RenderFrameImpl (patchset #1 id:1 of https://codereview.chromium.org/791053005/)

Reason for revert:
The bug is expected to be fixed quite some time ago, reverting to check.

Original issue's description:
> [MemSheriff] Suppress use-after-free in content::RenderFrameImpl::~RenderFrameImpl
>
> BUG=445836
> TBR=lazyboy@chromium.org
>
> Committed: https://chromium.googlesource.com/chromium/src/+/bb25da2ef63be828fc0ae3303af0862cbca70123

TBR=bruening@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=445836

Review URL: https://codereview.chromium.org/1129973004

Cr-Commit-Position: refs/heads/master@{#329463}

This change decouples the DisplaySnapshot from the DrmDisplay since the
snapshot is only needed to IPC the display information to the browser
process. Also, move all the display configuration calls into the
DrmDisplay object to simplify state and clarify the intent of the
operations.

This change also moves the logic to parse the DRM display state into
DisplaySnapshot objects into a common location intended to be used by
both GPU and Browser processes. The parsing logic will be used in a
subsequent change to allow proper parsing of display configuration on
the browser side during startup.

BUG=484294

Review URL: https://codereview.chromium.org/1129923004

Cr-Commit-Position: refs/heads/master@{#329462}

This CL adds HTTP server and connection classes, and implements lifespan management for them.

The following work will be in future CLs:
- convert net::HttpServer::Delegate events about HTTP/WebSocket requests to HttpConnectionDelegate Mojo interface method calls.
- tests.

BUG=478249
TEST=None

Review URL: https://codereview.chromium.org/1134883002

Cr-Commit-Position: refs/heads/master@{#329461}

BUG=486240

Review URL: https://codereview.chromium.org/1136093004

Cr-Commit-Position: refs/heads/master@{#329460}

Moved the definition of a 10.7+ NSString constant into
mac/sdk_forward_declarations.h, and ensured that the NSString is explicitly
defined regardless of the OSX SDK version being compiled against.

BUG=471823

Review URL: https://codereview.chromium.org/1133523005

Cr-Commit-Position: refs/heads/master@{#329459}