GitLab

Waterfall Builder: iOS Simulator (dbg) is failing in bot_update since

http://build.chromium.org/p/chromium.mac/builders/iOS%20Simulator%20%28dbg%29/builds/16396

Suspecting this. Will unrevert if not.

BUG=377342

> Attempt #4: Change Skia DEPS to pull from googlesource (git)
> 
> - Uploading from pure-svn and intending to commit manually to properly remove the directory from svn.
> - Includes deps_whitelist change for new Skia DEP
> - Attempt #3 failed due to Poppler license issue. This was resolved in https://skia.googlesource.com/skia/+/3056bdd2364fe1aa68f4ec7b0fe07740c3e8d123
> 
> R=mkosiba@chromium.org
> 
> Review URL: https://codereview.chromium.org/296903004

TBR=borenet@google.com

Review URL: https://codereview.chromium.org/297303002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272769 0039d316-1c4b-4281-b951-d872f2087c98

Reason for revert:
Suspected for waterfall failures "gsignal.c:2459: signal `insert' is invalid" also possibly some obscure ASAN/LSAN errors

http://build.chromium.org/p/chromium.linux/buildstatus?builder=Linux%20Tests%20%28dbg%29%282%29&number=46389
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/2531

[15124:15124:0524/063846:FATAL:browser_main_loop.cc(205)] GLib-GObject: /build/buildd/glib2.0-2.32.4/./gobject/gsignal.c:2459: signal `insert' is invalid for instance `0xcb71d63c0b0'
#0 0x7f91f99d3019 base::debug::StackTrace::StackTrace()
#1 0x7f91f9a4f139 logging::LogMessage::~LogMessage()
#2 0x7f920330e20d content::(anonymous namespace)::GLibLogHandler()
#3 0x7f91f3ab0f61 g_logv
#4 0x7f91f3ab1172 g_log
#5 0x7f91f3d7dd47 g_signal_connect_data
#6 0x7f91d9293efb <unknown>
#7 0x7f91d9294e8e <unknown>
#8 0x7f91d92961a8 dbusmenu_gtk_parse_menu_structure
#9 0x7f91d96af8fe app_indicator_set_menu
#10 0x7f91f8ac609b libgtk2ui::AppIndicatorIcon::SetMenu()
#11 0x7f91f8ac5e3a libgtk2ui::AppIndicatorIcon::SetImageFromFile()
#12 0x7f91f8ac92c8 base::internal::RunnableAdapter<>::Run()

==7567==ERROR: AddressSanitizer: heap-use-after-free on address 0x60800001e028 at pc 0x4cfb688 bp 0x7f4facffd2d0 sp 0x7f4facffd2c8
WRITE of size 4 at 0x60800001e028 thread T7 (CompositorRaste)

==14726==ERROR: LeakSanitizer: detected memory leaks
Direct leak of 116 byte(s) in 2 object(s) allocated from:
    #0 0x5758ab in __interceptor_malloc /usr/local/google/work/chromium/src/third_party/llvm/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:75
    #1 0x7f52ed561a38 in g_malloc /build/buildd/glib2.0-2.32.4/./glib/gmem.c:159


Original issue's description:
> linux_aura: Add Trusty libappindicator.
> 
> BUG=374686
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=272718

TBR=pkotwicz@chromium.org,erg@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=374686

Review URL: https://codereview.chromium.org/296393003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272763 0039d316-1c4b-4281-b951-d872f2087c98

Causing compile failures on Windows.  Possible comatability issue
with https://codereview.chromium.org/298703007

Sample errors:

FAILED: ninja -t msvc -e environment.x86 -- C:\b\build\goma/gomacc "C:\b\depot_tools\win_toolchain\vs2013_files\VC\bin\amd64_x86\cl.exe" /nologo /showIncludes /FC @obj\ui\wm\core\wm_unittests.nested_accelerator_controller_unittest.obj.rsp /c ..\..\ui\wm\core\nested_accelerator_controller_unittest.cc /Foobj\ui\wm\core\wm_unittests.nested_accelerator_controller_unittest.obj /Fdobj\ui\wm\wm_unittests.cc.pdb 
c:\b\build\slave\win\build\src\ui\wm\core\nested_accelerator_controller_unittest.cc(83) : error C2039: 'host' : is not a member of 'aura::Window'
        c:\b\build\slave\win\build\src\ui\aura\window.h(62) : see declaration of 'aura::Window'
c:\b\build\slave\win\build\src\ui\wm\core\nested_accelerator_controller_unittest.cc(83) : error C2227: left of '->PostNativeEvent' must point to class/struct/union/generic type
c:\b\build\slave\win\build\src\ui\wm\core\nested_accelerator_controller_unittest.cc(85) : error C2039: 'host' : is not a member of 'aura::Window'
        c:\b\build\slave\win\build\src\ui\aura\window.h(62) : see declaration of 'aura::Window'
c:\b\build\slave\win\build\src\ui\wm\core\nested_accelerator_controller_unittest.cc(85) : error C2227: left of '->PostNativeEvent' must point to class/struct/union/generic type
ninja: build stopped: subcommand failed.


> Refactor and move ash independent accelerator handling code in nested loop to ui/wm/core
> 
> I also renamed classes to NestedAcceleratorXxx. I felt this is a bit more clearer than NestedDispatcher, especially in ui/wm/core. Please let me know if you disagree or have better suggestion. I'm happy to rename them.
> 
> BUG=None
> 
> Review URL: https://codereview.chromium.org/298703007

TBR=oshima@chromium.org

Review URL: https://codereview.chromium.org/300093002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272759 0039d316-1c4b-4281-b951-d872f2087c98

R=zty@chromium.org

Review URL: https://codereview.chromium.org/302433003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272750 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272742 0039d316-1c4b-4281-b951-d872f2087c98

I also renamed classes to NestedAcceleratorXxx. I felt this is a bit more clearer than NestedDispatcher, especially in ui/wm/core. Please let me know if you disagree or have better suggestion. I'm happy to rename them.

BUG=None

Review URL: https://codereview.chromium.org/298703007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272740 0039d316-1c4b-4281-b951-d872f2087c98

This patch converts the following gyp targets to GN:
- favicon_base
- query_parser
- metro_viewer_messages -> metro_viewer
- native_theme

BUG=None
TEST=None
R=scottmg@chromium.org
TBR=ben@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/294423004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272739 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true

Review URL: https://codereview.chromium.org/299313004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272738 0039d316-1c4b-4281-b951-d872f2087c98

BUG=45482
R=thestig@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/293983023

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272737 0039d316-1c4b-4281-b951-d872f2087c98

TBR=jochen@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/300633002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272736 0039d316-1c4b-4281-b951-d872f2087c98

this change:
- changes border-color instead of opacity of checkboxes on hover
- makes checked and hovered checkboxes look more similar
- makes hovered rows more noticeable

And I touched a few more things that weren't really noticeable
before I changed the background color:
- checkbox vertical alignment
- added padding in front of checkboxes

BUG=351565
R=dmazzoni@chromium.org
TEST=visible checkboxes

Review URL: https://codereview.chromium.org/293303002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272735 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=None

Review URL: https://codereview.chromium.org/293963004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272734 0039d316-1c4b-4281-b951-d872f2087c98

This is rudimentary implemenatation of AttachmentDownloader along with fake for tests.

BUG=376073
R=maniscalco@chromium.org

Review URL: https://codereview.chromium.org/296153014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272733 0039d316-1c4b-4281-b951-d872f2087c98

Identical logic change in both AudioSender and VideoSender: Send a sender
report just before each frame, up to the first 100 frames.  Then, they are
sent at ~1 second intervals thereafter.  However, if a receiver report
shows up, we halt the aggressive report sending immediately since we know
the data dependency (receiver side) has been satisfied.

Also, misc clean-ups:

1. RtpTimestampHelper threw away significant precision in its calculations
   (was truncating to millis).
2. Unit tests were using EXPECT with args in the wrong order (per gtest
   requirements).
3. Got rid of useless LocalRtcpXXXXXSenderFeedback class.

BUG=371482

Review URL: https://codereview.chromium.org/295603004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272732 0039d316-1c4b-4281-b951-d872f2087c98

scoped_array is deprecated in favor of scoped_ptr[]
as of https://codereview.chromium.org/14081006

Review URL: https://codereview.chromium.org/291893010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272731 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/291153012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272730 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true
TBR=michaelbai@chromium.org

Review URL: https://codereview.chromium.org/297183002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272729 0039d316-1c4b-4281-b951-d872f2087c98

Also remove cruft and simplify code.

Review URL: https://codereview.chromium.org/297023002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272728 0039d316-1c4b-4281-b951-d872f2087c98

For tests I'm creating a MessagePumpGlib on a thread, destroying it,
then creating another MessagePumpGlib on another thread. The DCHECK
fires in this case because it only remembers the first thread. I made
it so the thread id is cleared when the MessagePumpGlib is destroyed.

BUG=none
TEST=none
R=sadrul@chromium.org
TBR=darin@chromium.org

Review URL: https://codereview.chromium.org/289043010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272727 0039d316-1c4b-4281-b951-d872f2087c98

BUG=368455

Review URL: https://codereview.chromium.org/296933016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272726 0039d316-1c4b-4281-b951-d872f2087c98

I was doing some benchmarking on tests on linux, and noticed that 10-20% of time was
in ThreadCheckerImpl::CalledOnValidThread. I looked into it and found that we're doing
a syscall to get the thread ID, which is not needed on any platform where pthread_t
is already an integer type.  Replacing that syscall with pthread_self() makes this particular
test (End2EndTest.BasicFakeSoftwareVideo) go from 488ms to 374ms, which is a 20% speedup.
While other tests might not use ThreadChecker quite as much, it should still be a pretty
good win across a large set of tests.

Review URL: https://codereview.chromium.org/292873002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272725 0039d316-1c4b-4281-b951-d872f2087c98

The vsync is unnecessary for ratelimiting or tear prevention, and can cause issues when multiple windows are displayed.

BUG=313104

Review URL: https://codereview.chromium.org/299443020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272724 0039d316-1c4b-4281-b951-d872f2087c98

Turn on ENABLE_NEW_GAMEPAD_API flag so that the new gamepad code is used.
Also remove the old code on Chromium side. The ENABLE_NEW_GAMEPAD_API is only
needed until the Blink side #ifdefs are removed.

BUG=344556

Review URL: https://codereview.chromium.org/289423004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272723 0039d316-1c4b-4281-b951-d872f2087c98

This change removes the NPAPI plugin from the webapp and switch Windows to use the Mac workflow where the user is prompted to download and install the host manually.

The removal of the dispatching code and the NPAPI target will be done in later CLs.

BUG=377046

Review URL: https://codereview.chromium.org/299983002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272722 0039d316-1c4b-4281-b951-d872f2087c98

This is in preparation for enabling HiDPI on Windows by default. Most of the changes here are replacing usage of
the ui::GetImageScale function with ui::GetScaleForScaleFactor.

The patch to remove the HiDPI switch and enable it by default will follow. This patch has been incorporated from
girard's patch ( https://codereview.chromium.org/232483005/)

BUG=149881, 362245

Review URL: https://codereview.chromium.org/294863002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272721 0039d316-1c4b-4281-b951-d872f2087c98

BUG=376924

Review URL: https://codereview.chromium.org/295983011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272720 0039d316-1c4b-4281-b951-d872f2087c98

Since we had to change it on Win/Linux, I thought it would be good to be 
consistent and draw the frame on Mac as well.

BUG=372721

Review URL: https://codereview.chromium.org/298973007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272719 0039d316-1c4b-4281-b951-d872f2087c98

BUG=374686

Review URL: https://codereview.chromium.org/297833013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272718 0039d316-1c4b-4281-b951-d872f2087c98

BUG=n/a
NOTRY=true

Review URL: https://codereview.chromium.org/297903005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272717 0039d316-1c4b-4281-b951-d872f2087c98

Crash seems to indicate we're registering twice for a
notification. Perhaps we're calling ScheduleLoad and TabIsLoading for
the same tab. Only way I could see that happening is if ShowBrowser()
activated some other tab. This should help isolate if that is
happening.

BUG=368236
TEST=none
R=marja@chromium.org

Review URL: https://codereview.chromium.org/299053002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272716 0039d316-1c4b-4281-b951-d872f2087c98

To show the information about ServiceWorkers I added iframe of "chrome://serviceworker-internals/#iframe" to chrome://inspect.
And I used postMessage() to achieve the cross-domain messaging between chrome://inspect and chrome://serviceworker-internals.

BUG=358657

Review URL: https://codereview.chromium.org/298993003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272715 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None

Review URL: https://codereview.chromium.org/296133015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272714 0039d316-1c4b-4281-b951-d872f2087c98

After r271640, they no longer need this information, since the browser
process sets it in crash reports.

TBR=jochen@chromium.org

Review URL: https://codereview.chromium.org/296213004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272713 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/292743017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272712 0039d316-1c4b-4281-b951-d872f2087c98

Refactors the StatusChangeChecker class hierarchy to allow for more code
sharing.  Allows children of StatusChangeChecker to share logic around
performing a blocking wait, checking exit conditions, and timing out.

As before, the sharing is acheived through inheritance.  We should move
to a delegation pattern if this becomes too complicated, but I think the
use of inheritance still makes sense for now.

This is a pre-requisite for a future patch that adds a new kind of
StatusChangeChecker that does not listen to ProfileSyncService changes.
I expect that there will be many more kinds of StatusChangeCheckers
defined in the future.  This refactoring will help us share code between
them.

This is attached to issue 97780 and 95742 because it will enable us to
write more StatusChangeCheckers that we can use to refactor tests that
currently listen for snapshots and self-notifications.

BUG=97780,95742

Review URL: https://codereview.chromium.org/299843007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272711 0039d316-1c4b-4281-b951-d872f2087c98

The streamsPrivate API provides the response headers for the original URL request in the stream details. This passes them along to the plugin with the URL.

This also adds the ability for the plugin to cancel the stream, which is necessary if the plugin wants to re-request the original URL.

BUG=303491

Review URL: https://codereview.chromium.org/275333004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272710 0039d316-1c4b-4281-b951-d872f2087c98

Also, add an optimization to not decode frames if we are behind and multiple frames are currently decodable.

Review URL: https://codereview.chromium.org/289483003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272709 0039d316-1c4b-4281-b951-d872f2087c98

Turns MetricsLogChromeOS into a metrics::MetricsProvider. Splits the
ChromeOS-specific unittests out from the MetricsLog test into separate
ChromeOSMetricsProvider tests. Also moves LogChromeOSCrash() from
MetricsService to ChromeOSMetricsProvider.

BUG=374221
R=asvitkine@chromium.org
TBR=derat, pam

Review URL: https://codereview.chromium.org/292433015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272708 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Don't expose --disable-gpu-compositing, only set internally. (https://codereview.chromium.org/300523002/)

Reason for revert:
Makes a bunch of virtual/softwarecompositing tests Timeout on Mac
https://code.google.com/p/chromium/issues/detail?id=377145

Original issue's description:
> Don't expose --disable-gpu-compositing, only set internally.
> 
> The flag isn't meant to be set externally, it doesn't work on most
> configs. It's only meant internally to make sure the renderer doesn't
> try to use GPU when the browser is in software.
> 
> BUG=None
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=272636

TBR=jbauman@chromium.org,piman@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=None

Review URL: https://codereview.chromium.org/299383002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272707 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@272706 0039d316-1c4b-4281-b951-d872f2087c98