GitLab

A reference to capturing_net_log_ was passed into the base class constructor
and used before it was fully initialized. For the derived class the initialization
of the members happens stronly after the base constructors completed.

This fixes a bug found by UBSan Vptr.

BUG=617206

Review-Url: https://codereview.chromium.org/2042493002
Cr-Commit-Position: refs/heads/master@{#397846}

The `window` object is a more natural place for the visualViewport API. The
original concern was causing breaking changes since this would clash with any
variable in the global scrope called `visualViewport` but a search through
httpArchive found that this is unlikely to be an issue.

BUG=595826

Review-Url: https://codereview.chromium.org/2025393003
Cr-Commit-Position: refs/heads/master@{#397845}

BUG=598021
R=agl@chromium.org, estark@chromium.org, lgarron@chromium.org

Review URL: https://codereview.chromium.org/2034843003 .

Cr-Commit-Position: refs/heads/master@{#397844}

Most uses of "hover" become "highlight", but SetHovered still refers to
the logical state of mouse hover so that remains unaltered.

BUG=none

Review-Url: https://codereview.chromium.org/2034963002
Cr-Commit-Position: refs/heads/master@{#397843}

BUG=614277, 543635
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2031833003
Cr-Commit-Position: refs/heads/master@{#397842}

Remove kSyncInputForSyncCompositor switch and all code behind it. M51
shipped without it and looks fine, so remove all the code.

BUG=545628

Review-Url: https://codereview.chromium.org/1974133002
Cr-Commit-Position: refs/heads/master@{#397841}

BUG=458193

Review-Url: https://codereview.chromium.org/2033883002
Cr-Commit-Position: refs/heads/master@{#397840}

Move GL one-off initialization from //ui/gl to //ui/gl/init. Create a
new class GLInitializer to contain the platform specific initialization
methods and add platform specific implementation in a conditionally
included file per platform. Remove the existing one-off initialization
code in GLSurface and from platform specific GLSurface files.

This change is part of step two from the associated bug.

BUG=611142

Review-Url: https://codereview.chromium.org/2024953002
Cr-Commit-Position: refs/heads/master@{#397839}

(Counterpart to https://codereview.chromium.org/1995823002)

BUG=539572

Review-Url: https://codereview.chromium.org/1995833002
Cr-Commit-Position: refs/heads/master@{#397838}

This way a clean built of mash:all and chrome results in opening
chrome.

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

Review-Url: https://codereview.chromium.org/2034263002
Cr-Commit-Position: refs/heads/master@{#397837}

https://chromium.googlesource.com/chromium/src/+/28307938a

BUG=465096
TBR=robhogan@gmail.com

Review URL: https://codereview.chromium.org/2037933004 .

Cr-Commit-Position: refs/heads/master@{#397836}

https://chromium.googlesource.com/chromium/src/+/91e2b1e30

BUG=333011
TBR=kulshin@chromium.org

Review URL: https://codereview.chromium.org/2036203006 .

Cr-Commit-Position: refs/heads/master@{#397835}

This is optional for running leak detector on browser process, but later
it will be necessary to run on renderer process. By initializing the
pthread TLS system early, we avoid initializing it from within an alloc
hook function, in which case it would go into a recursive call loop.

BUG=chromium:615223

Review-Url: https://codereview.chromium.org/2023133003
Cr-Commit-Position: refs/heads/master@{#397834}

To match it with the API. Android and mojo API use 32bit int for this property.

BUG=none

Review-Url: https://codereview.chromium.org/2031973003
Cr-Commit-Position: refs/heads/master@{#397833}

This switches HistoryController and HistoryEntry to look at the last
committed WebHistoryItem for a frame rather than assuming that the
whole HistoryEntry accurately reflects the state of the page.  The
previous logic led to NC_IN_PAGE_NAVIGATION renderer kills when going
back cross-site in ways that looked in-page to the HistoryEntry.

This CL does not yet fix the kills that happen in FrameNavigationEntry
modes, which are covered by the BackTwiceToIframeWithContent test.
Those will be fixed in a follow-up CL.

BUG=612713
TEST=Fewer NC_IN_PAGE_NAVIGATION kills.
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2023013002
Cr-Commit-Position: refs/heads/master@{#397832}

BUG=internal b/26496758

Review-Url: https://codereview.chromium.org/2032813002
Cr-Commit-Position: refs/heads/master@{#397831}

BUG=615486
R=xhwang@chromium.org

Review-Url: https://codereview.chromium.org/2033923002
Cr-Commit-Position: refs/heads/master@{#397830}

ResumableTCPSocket sets a read_callback_ that points to
TCPSocketEventDispatcher::ReadCallback. This callback gets called
from ~TCPSocket, when ~ResumableTCPSocket already completed,
and it calls ResumableTCPSocket methods (which are already illegal to call).

The fix just calls Disconnect from ~ResumableTCPSocket which clears the
callback and all the state that relies on ResumableTCPSocket members.

This fixes a bug found with UBSan Vptr.

BUG=617199

Review-Url: https://codereview.chromium.org/2034233002
Cr-Commit-Position: refs/heads/master@{#397829}

BUG=612342
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2029703002
Cr-Commit-Position: refs/heads/master@{#397828}

The recent refactoring from KeySystemsInfo meant we lost the option
to control whether persistent license support is enabled (see linked
bug).  Add the option back into the new system.

BUG=internal b/28844432

Review-Url: https://codereview.chromium.org/2032923002
Cr-Commit-Position: refs/heads/master@{#397827}

BUG=none

Review-Url: https://codereview.chromium.org/2025153002
Cr-Commit-Position: refs/heads/master@{#397826}

I'd like to make more of the focus ring layers animate, so it will be
simpler if the layers are animation observers and keep track of their own
compositor, rather than storing them all in AccessibilityFocusRingController.

I'm assuming it's not too expensive to have OnAnimationStep() called for a
layer that doesn't need to animate all the time. If that's not true, this
refactoring would make it easier to add StartAnimating and StopAnimating
methods to FocusRingLayer for better efficiency.

BUG=314889

Review-Url: https://codereview.chromium.org/2038093003
Cr-Commit-Position: refs/heads/master@{#397825}

This was just a matter of plumbing the text to the right place.

BUG=609932
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2041493002
Cr-Commit-Position: refs/heads/master@{#397824}

BUG=617318
R=hans@chromium.org

Review URL: https://codereview.chromium.org/2040533003 .

Cr-Commit-Position: refs/heads/master@{#397823}

BUG=617056

Review-Url: https://codereview.chromium.org/2034893003
Cr-Commit-Position: refs/heads/master@{#397822}

BUG=617288
NOTRY=true

Review-Url: https://codereview.chromium.org/2037773004
Cr-Commit-Position: refs/heads/master@{#397821}

WmGlobals->WmShell
A bunch of classes that were in wm have moved to ash/common. Nuked the
wm:: workspace.
I initially wanted to keep this small, but it's easier to move move of
the classes together as less namespace collisions then.

BUG=616265
TEST=none
R=jamescook@chromium.org
TBR=reveman@chromium.org

Review-Url: https://codereview.chromium.org/2035543004
Cr-Commit-Position: refs/heads/master@{#397820}

Automatic downloads
Background sync
Key generation
Plugins
Unsandboxed plugin access

Also changed it so that, in JS land, the Category variable is a string, because the enum value of the ContentSettingsTypes is variable depending on platform, so we don't want to be hardcoding those in JS. This is similar to the Allow/Block values (the PermissionValues enum), which used to be ints, but are now converted int <-> string on the JS/C++ boundary.

TBR=sky
BUG=614277, 543635
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2021343003
Cr-Commit-Position: refs/heads/master@{#397819}

R=jialiul@chromium.org

Review-Url: https://codereview.chromium.org/2036133003
Cr-Commit-Position: refs/heads/master@{#397818}

BUG=431177

Review-Url: https://codereview.chromium.org/2033893006
Cr-Commit-Position: refs/heads/master@{#397817}

Overriding ShouldAddContentSecurityPolicy() to disable CSP is considered
a bug. Enabling CSP while overriding GetContentSecurityPolicyScriptSrc()
to have relaxed rule is better.

BUG=81636

Review-Url: https://codereview.chromium.org/2003963004
Cr-Commit-Position: refs/heads/master@{#397816}

Dangerous ones still don't show the button.

This was an oversight concerning the difference between malicious and
dangerous.

BUG=616615

Review-Url: https://codereview.chromium.org/2029363002
Cr-Commit-Position: refs/heads/master@{#397815}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/9cdf7a6f..a352c80b

Please follow these instructions for assigning/CC'ing issues:
https://github.com/v8/v8/wiki/Triaging%20issues

Please close rolling in case of a roll revert:
https://v8-roll.appspot.com/
This only works with a Google account.

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

TBR=hablich@chromium.org,machenbach@chromium.org,yangguo@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/2038093004
Cr-Commit-Position: refs/heads/master@{#397814}

Has only one libfuzzer-related change:
    ffa4b67 [libfuzzer] adding //base to more targets.

No changes to the library.

BUG=

Review-Url: https://codereview.chromium.org/2030763003
Cr-Commit-Position: refs/heads/master@{#397813}

BUG=611802
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review-Url: https://codereview.chromium.org/2032643002
Cr-Commit-Position: refs/heads/master@{#397812}

BUG=None

Review-Url: https://codereview.chromium.org/2037793003
Cr-Commit-Position: refs/heads/master@{#397811}

..\..\courgette\third_party\bsdiff\paged_array_unittest.cc(20,11):
  error: unused variable 'kIterations' [-Werror,-Wunused-const-variable]
const int kIterations = 20;
          ^
..\..\courgette\third_party\bsdiff\paged_array_unittest.cc(21,11):
  error: unused variable 'kSizeBig' [-Werror,-Wunused-const-variable]
const int kSizeBig = 200 * 1024 * 1024 / sizeof(int);  // 200MB
          ^

BUG=82385
TBR=huangs

Review-Url: https://codereview.chromium.org/2036393002
Cr-Commit-Position: refs/heads/master@{#397810}

BUG=617290
TEST=gpu bots
TBR=kbr@chromium.org
NOTRY=true
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2032063004
Cr-Commit-Position: refs/heads/master@{#397809}

The default is to strip https:// URLs before submitting them to PAC scripts.

This CL introduces the policy "PacHttpsUrlStrippingEnabled" for disabling this security feature.

Setting the policy to "false" causes Chrome to no longer strip https:// URLs before sending them to PAC scripts. This applies to all profiles, and all PAC scripts (including those discovered through WPAD, and those delivered over an insecure transport).

The intent of this policy is to help enterprises with a compatibility problem transition.

BUG=616396
TBR=isherman@chromium.org

Review-Url: https://codereview.chromium.org/2030193004
Cr-Commit-Position: refs/heads/master@{#397808}

This CL bridges the native WebContents::DownloadImage to Java. We are planning to use the method for fetching artwork images (WIP) in MediaMetadata and show them in Android media notification.

Related CLs:
https://codereview.chromium.org/2013813002/
https://codereview.chromium.org/2015433003/
https://codereview.chromium.org/2009243002/ (most relevant one)

BUG=616411

Review-Url: https://codereview.chromium.org/2014553002
Cr-Commit-Position: refs/heads/master@{#397807}