GitLab

Allow user to define pool of limited size to limit the number of
concurrent tasks that are executed for a given set of tools. The
pool object directly correspond to a ninja pool and can be shared
by multiple targets.

Design document:
https://docs.google.com/document/d/1b598i4WmeFOe3_iRBrjPKSXdcR5R32UPtgxqKmoY2-M/view

BUG=612786

Review-Url: https://codereview.chromium.org/2006923004
Cr-Commit-Position: refs/heads/master@{#397917}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/9ceb2d30..b94ab6fc

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/2038263003
Cr-Commit-Position: refs/heads/master@{#397916}

FF, IE and Edge all repeat a table header group at the top of each printed page.

This area is underspecified but after some discussion in the bug we agreed to
repeat the header group if it has break-inside:avoid. We make this the default
style for theads when printing.

BUG=24826

Review-Url: https://codereview.chromium.org/2021703002
Cr-Commit-Position: refs/heads/master@{#397915}

https://chromium.googlesource.com/chromium/src/+/59fd991c4

BUG=416535
TBR=chrishtr@chromium.org

Review URL: https://codereview.chromium.org/2037263002 .

Cr-Commit-Position: refs/heads/master@{#397914}

1) Unify the validation code into a helper function.
2) In ES3, for sized formats, the read_format and internal_format's color bits
   have to match exactly.

BUG=429053
TEST=gpu_unittests,webgl{2}_conformance
R=piman@chromium.org
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/2029803002
Cr-Commit-Position: refs/heads/master@{#397913}

BUG=416535

Review-Url: https://codereview.chromium.org/2039503002
Cr-Commit-Position: refs/heads/master@{#397912}

Fixes a MacViewsBrowser link error after r397320, starting at
http://go/macviewsbuilder/builds/11045, like

Undefined symbols for architecture x86_64:
  "ChromeExtensionChooserDialog::ShowDialog(ChooserController*) const", referenced from:
      ChromeExtensionBluetoothChooser::ChromeExtensionBluetoothChooser

MacViewsBrowser needs a ShowDialog() symbol, which will provide the
toolkit-views implementation when it's ready.

BUG=425229, 613638
TBR=juncai@chromium.org

Review-Url: https://codereview.chromium.org/2036033003
Cr-Commit-Position: refs/heads/master@{#397911}

This CL introduces a BENCHMARK_MEMORY_LIGHT mode for background field
trials to enable memory-infra category in BACKGROUND mode. The
background tracing controller disables the periodic memory dumps and
takes a single memory snapshot when a light preset is passed.
Since the BACKGROUND mode in memory-infra is restricted only to
whitelisted dump providers (currently empty), this tracing mode causes
very low overhead (currently none) on the user experience.

BUG=613198

Review-Url: https://codereview.chromium.org/2034553002
Cr-Commit-Position: refs/heads/master@{#397910}

Cleanup from https://codereview.chromium.org/2036353002/ done separately to
preserve history.

BUG=615155
TEST=compiles

Review-Url: https://codereview.chromium.org/2032253003
Cr-Commit-Position: refs/heads/master@{#397909}

Previously it used string::swap, but std::move is more effective in this case.

BUG=367418
TEST=
R=abarth@chromium.org,brettw@chromium.org

Review-Url: https://codereview.chromium.org/1360533002
Cr-Commit-Position: refs/heads/master@{#397908}

* Collapse wm_shelf_constants.h with shelf_constants.h
* Start collapsing wm_shelf_types.h with shelf_types.h. This will be done in
two steps to preserve rename history.
* Move the other files as-is

No functional changes.

BUG=615155
TEST=compiles

Review-Url: https://codereview.chromium.org/2036353002
Cr-Commit-Position: refs/heads/master@{#397907}

This avoids a NOTREACHED() in the previous code. Verified that addition
of new constants without updating the switch statement will cause a
compile failure.

BUG=615113
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/2029293002
Cr-Commit-Position: refs/heads/master@{#397906}

BUG=617004

Review-Url: https://codereview.chromium.org/2033813005
Cr-Commit-Position: refs/heads/master@{#397905}

In the event a thread gets to participate in more than
one GC before it manages to leave its safe point, clear
out its weak callback stack before initiating a new
GC. Otherwise we risk keeping around weak callbacks to
duplicate or dead objects.

R=
BUG=611702

Review-Url: https://codereview.chromium.org/2036803004
Cr-Commit-Position: refs/heads/master@{#397904}

BUG=None
TEST=None
R=bbudge@chromium.org,isherman@chromium.org

Review-Url: https://codereview.chromium.org/1858123002
Cr-Commit-Position: refs/heads/master@{#397903}

Replaced wtf/Assertions.h macros in favor of
base/logging.h macros in mediasource module.

WTF_LOG -> DVLOG
ASSERT -> DCHECK
ASSERT_NOT_REACHED -> NOTREACHED

BUG=596522

Review-Url: https://codereview.chromium.org/2002073002
Cr-Commit-Position: refs/heads/master@{#397902}

This fixes crash of webview when its content is scrolled using
2-fingers gesture via touchpad or using touchscreen.

BUG=615512
TEST=Manually on device using Arc++ OptIn UI, ToS is scrolled
     as expected using touchscreen or touchpad, no crashes.
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2015373002
Cr-Commit-Position: refs/heads/master@{#397901}

* Add ability to display additional text above the PaymentOption
  lists.

* Shipping address section now has the option to display text
  saying that the user needs to select a shipping address to
  update the shipping options.

* Optional shipping address section text disappears once the
  user selects an option.

BUG=603635

Review-Url: https://codereview.chromium.org/2040623002
Cr-Commit-Position: refs/heads/master@{#397900}

The CertIssuerSource interface will provide an abstraction for the path builder to
lookup issuers without needing to know the details of how.

CertIssuerSourceStatic is a simple implementation useful for providing a fixed set,
for example the certs received during the TLS handshake.

BUG=410574

Review-Url: https://codereview.chromium.org/2030693002
Cr-Commit-Position: refs/heads/master@{#397899}

https://chromium.googlesource.com/skia.git/+log/862dae548a8e..7062d01a338d

$ git log 862dae548..7062d01a3 --date=short --no-merges --format='%ad %ae %s'
2016-06-03 bungeman More Unref GrXPFactory.

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=herb@google.com

Review-Url: https://codereview.chromium.org/2043463003
Cr-Commit-Position: refs/heads/master@{#397898}

Previously Textures could have an unowned service id, which is used in
place of their service id, except that its lifetime is managed
externally. The use case this is AVDACodecImages, which use a single
SurfaceTexture to back multiple Textures.

This change more explicitly ties the unowned service id concept to
StreamTextureImages, and makes it possible to have StreamTextureImages
own the lifetime of the unowned service id (now called the stream
texture service id), by resetting the service id back to its original
value if a non-StreamTextureImage is bound. If this were not done and a
new image was bound to a level that previously had a
StreamTextureImage, the service id would still be the unowned stream
texture service id, but the StreamTextureImage would be deleted. That
meant it was previously not possible for StreamTextureImages to own the
lifetime of the unowned service id without possibly leaving the
Texture in an invalid state.

Now we can have multiple AVDACodecImages, bound to different Textures,
share a SurfaceTexture service id and safely delete the texture when all
of the AVDACodecImages are destructed.

This CL also adds Texture::SetLevelImageState for setting only the
ImageState and not changing the bound image.

BUG=614090
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/2014313002
Cr-Commit-Position: refs/heads/master@{#397897}

Cr-Commit-Position: refs/heads/master@{#397896}

BUG=595206
R=dgozman@chromium.org

Review-Url: https://codereview.chromium.org/2043453002
Cr-Commit-Position: refs/heads/master@{#397895}

BUG=617177

Review-Url: https://codereview.chromium.org/2031373002
Cr-Commit-Position: refs/heads/master@{#397894}

Recombine the two XML layouts and add logic for adjusting them,
depending on what we're displaying.

BUG=603635

Review-Url: https://codereview.chromium.org/2041483002
Cr-Commit-Position: refs/heads/master@{#397893}

BUG=611802
TBR=piman@chromium.org
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review-Url: https://codereview.chromium.org/2041503002
Cr-Commit-Position: refs/heads/master@{#397892}

The first argument to Response constructor should be the body while
currently passing the header. This patch fixed it and updated the
pageset.

BUG=614985

Review-Url: https://codereview.chromium.org/2032893003
Cr-Commit-Position: refs/heads/master@{#397891}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/71d985b3c851..be4159a15b3e

$ git log 71d985b3c..be4159a15 --date=short --no-merges --format='%ad %ae %s'

BUG=613706

TBR=catapult-sheriff@chromium.org

Review-Url: https://codereview.chromium.org/2035133004
Cr-Commit-Position: refs/heads/master@{#397890}

https://chromium.googlesource.com/skia.git/+log/69df0206252c..862dae548a8e

$ git log 69df02062..862dae548 --date=short --no-merges --format='%ad %ae %s'
2016-06-03 robertphillips Add ability to rotate light direction in drawLitAtlas Sample slide

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=herb@google.com

Review-Url: https://codereview.chromium.org/2034843005
Cr-Commit-Position: refs/heads/master@{#397889}

When there are nested flexboxes, the current code does some O(N^2)
work to update scrolling information after flexing has finished.  This
change streamlines the process for performing flex layout and updating
scrollbars into three distinct phases, controlled by the highest-level
flexbox in the layout tree:

1. Perform flex layout on descendants; any blocks with overflow:auto
scrollbars may add/remove scrollbars, but they will not run the normal
second-pass layout after changing scrollbars, and they will not clamp
their existing scroll positions.

2. If, during the first pass, any descendant added or removed
scrollbars, run a second flex layout pass, but don't allow any
descendants to add or remove scrollbars.

3. After the second pass, go through and clamp the scroll positions
on all scrolling descendants.

BUG=593209,600036

Review-Url: https://codereview.chromium.org/1930183002
Cr-Commit-Position: refs/heads/master@{#397888}

Use context->Global() instead of CreationContext in MainThreadDebugger::memoryInfo.

BUG=616225
R=adamk@chromium.org,dgozman@chromium.org

Review-Url: https://codereview.chromium.org/2034203002
Cr-Commit-Position: refs/heads/master@{#397887}

MessagePumpForGpu has been on Canary for a few days and I
am still assessing whether it is better than standard
MessagePumpForUI in terms of reducing [GPU hang] crashes.
While it has eliminated hangs with some signatures the
rate of hangs in WaitForWork has increased.

When experimenting with MsgWaitForMultipleObjectsEx I
realized that it favors messages over events - when an
event is signaled and a message is available in the queue
it returns a result that indicates a message even when
the message doesn't belong to a current thread (e.g. a
browser UI message). With a high volume of messages it
might be possible for WaitForWork to keep looping and
starve the work processing. It is possible with Direct
Composition where some of UI input messages have to go
through the GPU process message loop because its D3D
surface window covers the entire browser window.

The fix is to do an additional check for the event. Please
note that this is a much less common code path so this
shouldn't affect performance much in scenarios like
video playback. Also I've added some diagnostics code to
help with crash dump analysis.

BUG=588798,596190

Review-Url: https://codereview.chromium.org/2036603002
Cr-Commit-Position: refs/heads/master@{#397886}

Also rolls breakpad to pick up an NDK fix.

BUG=599327

Review-Url: https://codereview.chromium.org/1991343002
Cr-Commit-Position: refs/heads/master@{#397885}

Revert of Enable per thread heap for database thread (patchset #12 id:220001 of https://codereview.chromium.org/1909813002/ )

Reason for revert:
This CL is causing a top #1 crasher.

https://bugs.chromium.org/p/chromium/issues/detail?id=617141

Original issue's description:
> Enable per thread heap for database thread
>
> BUG=591606
>
> Committed: https://crrev.com/993e3a74102e0843bebb2f52de1a904f151843d4
> Cr-Commit-Position: refs/heads/master@{#397383}

TBR=oilpan-reviews@chromium.org,keishi@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=591606

Review-Url: https://codereview.chromium.org/2037233002
Cr-Commit-Position: refs/heads/master@{#397884}

BUG=580337
NOTRY=true

Review-Url: https://codereview.chromium.org/2035653005
Cr-Commit-Position: refs/heads/master@{#397883}

There are pages that attach click handlers to elements that are found inside links, and screen readers might try to click such elements. Our existing doDefaultAction logic wasn't handling this possibility.
For example, there might be a graphic inside a link and the screen reader might invoke the default action on the graphic.
BUG=615904
TBR=dmazzoni@chromium.org

Review-Url: https://codereview.chromium.org/2043433002
Cr-Commit-Position: refs/heads/master@{#397882}

BUG=None

Review-Url: https://codereview.chromium.org/2037763002
Cr-Commit-Position: refs/heads/master@{#397881}

This creates a new script called the linker driver, which runs all three
steps of linking: the image link, debug info link, and stripping. In GYP,
the last two steps were handled as postbuilds, but GN lacks those. Instead,
the linker driver performs all three operations in one build step.

BUG=330301,431177
R=mark@chromium.org,dpranke@chromium.org

Review-Url: https://codereview.chromium.org/1999513002
Cr-Commit-Position: refs/heads/master@{#397880}

This wasn't caught because it's not compiled by default.

BUG=

Review-Url: https://codereview.chromium.org/2037213003
Cr-Commit-Position: refs/heads/master@{#397879}

Review-Url: https://codereview.chromium.org/2034303003
Cr-Commit-Position: refs/heads/master@{#397878}