GitLab

- The profile creator was using integer indexing into the tab list, which
non-deterministically fails. The new code uses dictionary lookup.
- The profile creator was not robust against several types of exceptions that
occur more frequently with developer builds of Chrome.

BUG=442546

Review URL: https://codereview.chromium.org/939143004

Cr-Commit-Position: refs/heads/master@{#317624}

This is a hot fix to stop the builder from failing the CQ while it is
being debugged.

NOTRY=true
TBR=smut@chromium.org
BUG=460930

Review URL: https://codereview.chromium.org/940473003

Cr-Commit-Position: refs/heads/master@{#317623}

Removes GetCompositorMessageLoop from the ContextFactory interface,
and drops the compositor_thread_ members from the ContextFactory
implementations.

There is no compositor thread in the ui compositor so these are now
excess complication.

R=sievers
BUG=329552

Review URL: https://codereview.chromium.org/943763004

Cr-Commit-Position: refs/heads/master@{#317622}

BUG=379378

Review URL: https://codereview.chromium.org/902763002

Cr-Commit-Position: refs/heads/master@{#317621}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/14df835..477bc41

TBR=kbr@chromium.org,hiroshige@chromium.org

Review URL: https://codereview.chromium.org/952613002

Cr-Commit-Position: refs/heads/master@{#317620}

These files declare/define mojo TypeConverters, so the file names
should reflect that.

R=jamesr@chromium.org,msw@chromium.org
BUG=457280

Review URL: https://codereview.chromium.org/950573002

Cr-Commit-Position: refs/heads/master@{#317619}

Add TRACE_EVENT_FLAG_EXPLICIT_TIMESTAMP which is set when an explicit
timestamp is specified for the trace event. Only re-read clock if the
flag is set.

BUG=457441

Review URL: https://codereview.chromium.org/928563003

Cr-Commit-Position: refs/heads/master@{#317618}

Fixes warnings in the msvs-gyp generator.

This is a followup to
https://codereview.chromium.org/615893003 (gesture_event_type_list.h,
    popup_item_type_list.h, result_codes_list.h and
    top_controls_state_list.h)
https://codereview.chromium.org/861373002 (navigator_connect_service.h and
    navigator_connect_servive_factory.h)
https://codereview.chromium.org/865413002 (layer_tree_build_helper.h)
https://codereview.chromium.org/774573003 (platform_notification_service.h)

R=jam@chromium.org,avi@chromium.org,mek@chromium.org,dtrainor@chromium.org,changwan@chromium.org,peter@chromium.org
BUG=459058,405532,351558,426458,451313,439950

Review URL: https://codereview.chromium.org/952443004

Cr-Commit-Position: refs/heads/master@{#317617}

The ICU 54.1 upgrade added a call to uloc_getDefault() to
base/test/test_suite.cc, but did not add a libicuuc dependency
to the corresponding gyp/gn targets.  This makes it possible to
mix system and bundled ICU headers (and will fail to build if
they differ in version).

BUG=428145

Review URL: https://codereview.chromium.org/943233002

Cr-Commit-Position: refs/heads/master@{#317616}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/953df44..6bc1b5f

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=reed@google.com

Review URL: https://codereview.chromium.org/942113003

Cr-Commit-Position: refs/heads/master@{#317615}

Increase max number of copy operations from 16 to 32. This
improves the throughput on systems where raster performance
is relatively high.

BUG=454898

Review URL: https://codereview.chromium.org/937843006

Cr-Commit-Position: refs/heads/master@{#317614}

This Android-specific file isn't used anywhere anymore. Came up while
reviewing Daniel's CL: https://codereview.chromium.org/952443004/

BUG=

Review URL: https://codereview.chromium.org/953543002

Cr-Commit-Position: refs/heads/master@{#317613}

PowerSaveBlocker objects are constructed in response to web
audio and video to keep the system from suspending or
turning its display off. Historically, this was not used on
Chrome OS, as Chrome OS has alternate mechanisms for
detecting audio and video activity and using
PowerSaveBlocker would interfere with enterprise
power-management-policy prefs to ignore audio and video
activity.

However, low-framerate video, such as that used by WebRTC
for Hangouts screen-sharing, can fail to trigger Chrome OS's
video detection code, resulting in the screen being turned
off while in use. This change renames PowerSaveBlocker's
human-readable "reason" field to "description" and adds a
new Reason enum that Chrome OS uses to identify and ignore
audio- or video-triggered PowerSaveBlockers if needed.

BUG=354723,176405
TEST=watch /var/log/power_manager/powerd.LATEST on chrome os
     while on the receiving end of a screen-sharing hangout
     and observe that a policy is sent forcing the screen on

Review URL: https://codereview.chromium.org/946643002

Cr-Commit-Position: refs/heads/master@{#317612}

Browser needs to allow nested tasks to update backgroud tabs.
If rendered requests new system dialog, asan build crashes in COMDLG32.dll

NOTRY=true
BUG=451753

Review URL: https://codereview.chromium.org/934393003

Cr-Commit-Position: refs/heads/master@{#317611}

This determines if sync is initialized and syncing URLs with the
appropriate keystore passphrase type to support contextual searches.

BUG=460786

Review URL: https://codereview.chromium.org/946273002

Cr-Commit-Position: refs/heads/master@{#317610}

This avoids a spurious log error.

BUG=460889

Review URL: https://codereview.chromium.org/952533002

Cr-Commit-Position: refs/heads/master@{#317609}

Catch the error raised by 'check_call' rather than the error emitted by
the called process.

BUG=chromium:460693,chromium:459681
TEST=local
  - Ran without fix, setting CHROME_HEADLESS=1. Script failed with the exception
    in the bug.
  - Ran with fix, 'gsutil' failed with AccessDeniedException output
    and the calling script successfully emitted a warning message.

R=hinoka@chromium.org, navabi@chromium.org

Review URL: https://codereview.chromium.org/949813002

Cr-Commit-Position: refs/heads/master@{#317608}

TBR=mtomasz
NOTRY=true

Review URL: https://codereview.chromium.org/949823005

Cr-Commit-Position: refs/heads/master@{#317607}

BUG=459646

Review URL: https://codereview.chromium.org/940283007

Cr-Commit-Position: refs/heads/master@{#317606}

This addresses two issues:

- NSS clamps client_version to the session version. This means that a
  successful fallback connection is effectively cached, despite our fallback
  being stateless. This causing our metrics to be under-reported and, more
  problematic, makes spurious fallbacks stick.

- BoringSSL does not clamp, but many versions of OpenSSL on the server will
  happily resume older sessions at newer protocol versions, rather than doing a
  full handshake at the newer protocol version. This means a successful
  spurious fallback causes us later resume with a weaker cipher than we should.
  Moreover, this mismatch is forbidden by every other client implementation.
  The metrics are reporting 0.06% of connections on beta channel hit this case.
  I expect it to go down after this change.

Note: this will also increase traffic to version-intolerant servers on NSS
ports. But that's only Linux/CrOS/iOS now and the BoringSSL switch did the same
thing by losing the version clamp.

BUG=459690,441456

Review URL: https://codereview.chromium.org/947603002

Cr-Commit-Position: refs/heads/master@{#317605}

BUG=webrtc:4241

Review URL: https://codereview.chromium.org/943033004

Cr-Commit-Position: refs/heads/master@{#317604}

d:\src\cr2\src\net\base\mime_util.cc(1286): warning C4456: declaration of 'extension' hides previous local declaration
d:\src\cr2\src\net\base\mime_util.cc(1278): note: see declaration of 'extension'

d:\src\cr2\src\net\base\upload_file_element_reader.cc(122): warning C4457: declaration of 'result' hides function parameter
d:\src\cr2\src\net\base\upload_file_element_reader.cc(106): note: see declaration of 'result'

d:\src\cr2\src\net\base\url_util.cc(47): warning C4457: declaration of 'value' hides function parameter
d:\src\cr2\src\net\base\url_util.cc(33): note: see declaration of 'value'

R=rdsmith@chromium.org
BUG=440500

Review URL: https://codereview.chromium.org/945713002

Cr-Commit-Position: refs/heads/master@{#317603}

Component extensions, when added/removed, trigger the following calls to ExtensionRegistryObserver:
Add:
OnExtensionWillBeInstalled
OnExtensionInstalled
OnExtensionLoaded
Remove:
OnExtensionUnloaded
OnExtensionUninstalled
Add (2nd call)
OnExtensionLoaded
Removed (2nd call):
OnExtensionUnloaded
OnExtensionUninstalled

In the non-component case, the 2nd calls are identical to the 1st. To work around this issue but still support whatever observers depend on this behavior, an uninstall reason component has been added and checked for explicitly in CommandService::OnExtensionUninstalled.

This has the positive side effect of removing the special casing in OnExtensionWillBeInstalled for component extensions in CommandService.

BUG=457350,458612
TEST=interactive_ui_tests --gtest_filter=CommandsApiTest.AddRemoveAddComponentExtension

Review URL: https://codereview.chromium.org/904943003

Cr-Commit-Position: refs/heads/master@{#317602}

DCHECK is removed in release builds, so we should move PostTask out of DCHECK.

Review URL: https://codereview.chromium.org/947203002

Cr-Commit-Position: refs/heads/master@{#317601}

Revert of Restore FileSystemProviderApiTest.BigFile. (patchset #1 id:1 of https://codereview.chromium.org/925213002/)

Reason for revert:
FileSystemProviderApiTest.BigFile is still failing, about 98% of runs on cros_trunk bot.

Original issue's description:
> Restore FileSystemProviderApiTest.BigFile.
>
> This test is not more flaky than other tests. There seem to be some test
> launching issue. Sometimes api tests do not start from time to time. It's
> not common though.
>
> Enabling the test, as it's better to have flaky test than no test. The issue
> with launching tests is tracked separately.
>
> TEST=browser_tests: *FileSystemProviderApiTest.BigFile*
> BUG=458122
>
> Committed: https://crrev.com/fae5f10535ecd3215d1e924505301562ed49744a
> Cr-Commit-Position: refs/heads/master@{#316429}

TBR=hirono@chromium.org,mtomasz@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=458122

Review URL: https://codereview.chromium.org/952623002

Cr-Commit-Position: refs/heads/master@{#317600}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/abcfab4..953df44

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=reed@google.com

Review URL: https://codereview.chromium.org/946413002

Cr-Commit-Position: refs/heads/master@{#317599}

BUG=455057
TEST=none

Review URL: https://codereview.chromium.org/946623002

Cr-Commit-Position: refs/heads/master@{#317598}

PRESUBMIT.py used to contain a handwritten list of bots to trigger on
`git cl try`. It was always out of sync with what the cq did.
https://codereview.chromium.org/823823002 awesomely made `git cl try` do
the same thing as the commit queue, but it kept a few special-case rules around.

It's better to not have these special cases: Once the developer hits cq, all
the "intelligently" skipped bots need to be run anyways.  More importantly,
there should be only one system to decide what to run and it should be
identical for cq and `git cl try`.  Currently, we think this system is the
`analyze` step, and it does a decent job of that already.

BUG=none (but made possible by 443613)

Review URL: https://codereview.chromium.org/921323003

Cr-Commit-Position: refs/heads/master@{#317597}

Sorta-follow-up to https://codereview.chromium.org/911923002/

BUG=459121

Review URL: https://codereview.chromium.org/950873002

Cr-Commit-Position: refs/heads/master@{#317596}

The state of the iframes is encoded in the page state, so we need to
decode the page state and recursively check it.

BUG=422871
R=marja@chromium.org

Review URL: https://codereview.chromium.org/948013002

Cr-Commit-Position: refs/heads/master@{#317595}

BUG=460948
TBR=cmumford@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/951533002

Cr-Commit-Position: refs/heads/master@{#317594}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/fa01745..14df835

TBR=kbr@chromium.org,hiroshige@chromium.org

Review URL: https://codereview.chromium.org/950843002

Cr-Commit-Position: refs/heads/master@{#317593}

If a symbol appears more than once in the binary, return the lowest offset for
that symbol, rather than the last one.  Also warn about duplicated symbols.
There are a couple hundred in libchrome.so.  Further investigation is needed to
determine why this is even possible, but in the mean time this should fix the
orderfile bot redness.

Review URL: https://codereview.chromium.org/947073003

Cr-Commit-Position: refs/heads/master@{#317592}

This is already the case on the other desktop platforms.

BUG=460893

Review URL: https://codereview.chromium.org/945363005

Cr-Commit-Position: refs/heads/master@{#317591}

This replaces the old CL (https://codereview.chromium.org/340543004/). This runs faster and doesn't require the label caching CL. (for the example in 370226, this is 20x faster on arm. ~60ms total vs ~3ms total)

I'll look into switching to label once I get the performance data with label after label caching is enabled.

This depends on two CLs
1) https://codereview.chromium.org/916423002/  Tooltip Cleanup
2) https://codereview.chromium.org/915383003/  RTL format problem.

but this can be reviewed independently.

BUG=370226,451853

Review URL: https://codereview.chromium.org/924433002

Cr-Commit-Position: refs/heads/master@{#317590}

Followup to
https://codereview.chromium.org/658383003 (constrained_window)
https://codereview.chromium.org/808773005 (component_updater and
    update_client)

This fixes warnings in gyp's msvs-ninja generator.

R=blundell@chromium.org,oshima@chromium.org,sorin@chromium.org
BUG=459058,410499

Review URL: https://codereview.chromium.org/952543002

Cr-Commit-Position: refs/heads/master@{#317589}

Some observers will need more information than just the url.

BUG=443773

Review URL: https://codereview.chromium.org/930933006

Cr-Commit-Position: refs/heads/master@{#317588}

The title can actually change when switching to a different translation, which would trigger a NOTREACHED in BookmarkModel::SetTitle.

BUG=459448

Review URL: https://codereview.chromium.org/946363002

Cr-Commit-Position: refs/heads/master@{#317587}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/792c80f..abcfab4

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=reed@google.com

Review URL: https://codereview.chromium.org/947933002

Cr-Commit-Position: refs/heads/master@{#317586}

This is a little icky but probably better than just putting it in Tab.
It seems like MediaCaptureDevicesDispatcher could have its own JNI wrapper but
it would be quite thin. At least this way, the accesses are co-located.

BUG=443773

Review URL: https://codereview.chromium.org/942873002

Cr-Commit-Position: refs/heads/master@{#317585}