GitLab

The button defaults to FocusBehavior::ACCESSIBLE_ONLY which (apart from
on Mac) only affects buttons in AccessiblePaneViews (which the close
button is not). Make it FocusBehavior::NEVER so the behaviour is
consistent across platforms.

Note this does not affect screen readers' ability to focus the element.
Keyboard access to this element when not using a screen reader is done
via the ESC key handler in DialogClientView.

BUG=741251

Review-Url: https://codereview.chromium.org/2982533002
Cr-Commit-Position: refs/heads/master@{#486228}

Bug: 287726
Change-Id: I521fad3c7ca7b77aff23b4b4d1a609b8efefcd40
Reviewed-on: https://chromium-review.googlesource.com/568793
Commit-Queue: Shimi Zhang <ctzsm@chromium.org>
Reviewed-by: Selim Gurun <sgurun@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486227}

This cl moves folderSelector, folderEditor and editViewController to
a common super class from bookmark handset and table view controllers.

Bug: 705339
Change-Id: I0ed26a3a6dac8d89ff8498f2753995e4772e4a8d
Reviewed-on: https://chromium-review.googlesource.com/566778
Commit-Queue: Ramya Sharma <ramyasharma@chromium.org>
Reviewed-by: Eric Noyau <noyau@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486226}

With the new ChromeOS settings page, accessing the lock screen settings requires
the user to reauth with their password. This obsoletes the previous behaviour of
reauthing specifically for EasyUnlock.

BUG=740290

Change-Id: I4788e81a750fb36217a4de3bff8e1e5505985bc2
Reviewed-on: https://chromium-review.googlesource.com/565778
Commit-Queue: Tim Song <tengs@chromium.org>
Reviewed-by: Toni Barzic <tbarzic@chromium.org>
Reviewed-by: Gustavo Sacomoto <sacomoto@chromium.org>
Reviewed-by: Sammie Quon <sammiequon@chromium.org>
Reviewed-by: Jacob Dufault <jdufault@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486225}

Bug: 740794
Change-Id: I7e4de04dfd2230b1eef2d7e85d7874055122b6dd
Reviewed-on: https://chromium-review.googlesource.com/567902Reviewed-by: Xiaohan Wang <xhwang@chromium.org>
Reviewed-by: Miguel Casas <mcasas@chromium.org>
Commit-Queue: Taiju Tsuiki <tzik@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486224}

This patch changes where we keep the lcd state flag to be on the layer
instead of the raster source. This cleans up a bunch of code that was
there since the days of tile sharing. Since tiles are no longer shared,
we can afford to make the code cleaner and eliminate a bunch of lcd text
references throughout.

This also allows us to remove the pseudo copy constructor from the
raster source, meaning that we can now store truly uncopyable types on
it.

R=enne@chromium.org, khushalsagar@chromium.org

Bug: 741775
Change-Id: I376e34f547de03cc5914686eaac60825ccbf2aeb
Reviewed-on: https://chromium-review.googlesource.com/568660Reviewed-by: Khushal <khushalsagar@chromium.org>
Commit-Queue: Vladimir Levin <vmpstr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486223}

After landing crrev/2956553002 which removes the metrics, these two
elapsed timers are no longer needed.

Bug: 738126
Change-Id: I04b29e4728cedb9e7f1fc740036d1a6106cbf0a5
Reviewed-on: https://chromium-review.googlesource.com/566990Reviewed-by: Patrick Monette <pmonette@chromium.org>
Commit-Queue: Xi Cheng <chengx@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486222}

BUG=741764

Review-Url: https://codereview.chromium.org/2976913002
Cr-Commit-Position: refs/heads/master@{#486221}

Removes a CHECK which was being hit in the wild. Poorly muxed media
can trigger the CHECK condition. Replace check with early return
in splicing code (no splice).

BUG=736740

Change-Id: Icf8f64d7283a6ff41ab98333676ba18341523819
Reviewed-on: https://chromium-review.googlesource.com/567680
Commit-Queue: Chrome Cunningham <chcunningham@chromium.org>
Reviewed-by: Matthew Wolenetz <wolenetz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486220}

Blink generates many files from scripts and templates. Only a few of
the generated files listed the source files used to create their output.

Clearly and consistently documenting the input for generated files
makes it easier to understand code and how to change it.

A new macro is introduced in templates/macros.tml and is consistently
used in all templates.

The parameters needed are added to all generating scripts.

Example output in generated files:
grep -r -C 10 "Generated from template" out/Release/gen
https://drive.google.com/file/d/0Bx8PJXs-5E92eTdkZnU0bHBYekk

Change-Id: I50b43d4257568118b63194859c06325d1ff1ead2
Reviewed-on: https://chromium-review.googlesource.com/568668
Commit-Queue: Vincent Scheib <scheib@chromium.org>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486219}

So that this class could be reused by bookmark_collection_view for
keeping track of the syncing status. (and in later CL, it will show
a loading spinner when bookmark is syncing.)

Bug: 696893
Change-Id: I370c91ad1f9815c6d3c70021939140478beaebcc
Reviewed-on: https://chromium-review.googlesource.com/567786
Commit-Queue: Marti Wong <martiw@chromium.org>
Reviewed-by: Mihai Sardarescu <msarda@chromium.org>
Reviewed-by: Jean-François Geyelin <jif@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486218}

This is a reland of c3dec2b9
Original change's description:
> gpu: Fix FD leak when importing DMA-bufs with multiple FDs.
> 
> CreateFromHandle is expected to take ownership of all FDs so
> when passed multiple FDs we need to make sure they are closed
> if not used.
>     
> BUG=b/63409814
> TEST=exo_unittests --gtest_also_run_disabled_tests --gtest_filter=DisplayTest.DISABLED_CreateLinuxDMABufBuffer
> 
> Change-Id: I2a8a7b0149a08f08890a725c6ce9aff8aeaab97f
> Reviewed-on: https://chromium-review.googlesource.com/566879
> Reviewed-by: Daniele Castagna <dcastagna@chromium.org>
> Commit-Queue: David Reveman <reveman@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#485761}

Bug: b/63409814
Change-Id: I3ff7bc12f17be39b721f75610e336a322d909502
Reviewed-on: https://chromium-review.googlesource.com/566960
Commit-Queue: Kazuhiro Inaba <kinaba@chromium.org>
Reviewed-by: Kazuhiro Inaba <kinaba@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486217}

Previously the function was declared, but not defined.

Bug: 734791
Change-Id: I848bca19b1e79f33c8f4b796f5b340fdade315f9
Reviewed-on: https://chromium-review.googlesource.com/565796Reviewed-by: Ken Rockot <rockot@chromium.org>
Commit-Queue: Sergey Ulanov <sergeyu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486216}

provider is not Google.

Bug:736038
TEST:unit tested, device tested with --enable-features=EnableFullcreenAppList

Change-Id: Ib305fbbdfea06b7f21fad2aa717af91e0f20aec0
Reviewed-on: https://chromium-review.googlesource.com/557505
Commit-Queue: Maajid <maajid@chromium.org>
Reviewed-by: Yury Khmel <khmel@chromium.org>
Reviewed-by: Qiang(Joe) Xu <warx@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486215}

Change QuicChromiumPacketWriter's socket to specify DatagramClientSocket, which it already is.  No functional change.

R=rch@chromium.org

Bug: 
Change-Id: I2ecd1abf335c29bee1a1791ebeab2d31ca393dbb
Reviewed-on: https://chromium-review.googlesource.com/568841Reviewed-by: Ryan Hamilton <rch@chromium.org>
Commit-Queue: Ian Swett <ianswett@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486214}

This CL ensures that Isolate is set to V8 parameters (e.g., HeapLimitMode) only
one time.

Isolate can be shared by WorkerThreads for worklets. Before this CL, Isolate is
set to the parameters every time a new WorkerThread is created. This is harmless
because Isolate is shared only by homogenous global scopes and given parameters
are always the same, but it shouldn't be a good design.

Bug: 710364
Change-Id: Ic44f1d5d2faebcc8b44c15d91e6afaa715045343
Reviewed-on: https://chromium-review.googlesource.com/567794Reviewed-by: Yuki Shiino <yukishiino@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Matt Falkenhagen <falken@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Commit-Queue: Hiroki Nakagawa <nhiroki@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486213}

After all the clients have been converted to use wake lock mojom interfaces,
the power_save_blocker should be hidden inside Device Service.

One exception is //remoting which is the only client that still directly uses
the power save blocker instead of wake lock mojo interfaces. The crbug.com/689423
records the reason.

BUG=689429

Change-Id: I485369f1d7b3984ecaabb667b15562702cc9b5b1
Reviewed-on: https://chromium-review.googlesource.com/563159Reviewed-by: Joe Downing <joedow@chromium.org>
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Reviewed-by: Ryo Hashimoto <hashimoto@chromium.org>
Reviewed-by: Colin Blundell <blundell@chromium.org>
Commit-Queue: Ke He <ke.he@intel.com>
Cr-Commit-Position: refs/heads/master@{#486212}

This came from https://crrev.com/c/559933. The cast was added to satisfy
a Windows compile error as it couldn't resolve the arguments to
operator!=.

A less ugly fix is to just get the raw pointer out of Member.

Bug: 
Change-Id: I14ed38a936aaf0457c2001a5e2c75b1a5ed91ac5
Reviewed-on: https://chromium-review.googlesource.com/567171Reviewed-by: Steve Kobes <skobes@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: David Bokan <bokan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486211}

After https://skia-review.googlesource.com/21821, Skia performs the same
optimization internally.

Change-Id: Ibf228865a4dfc1e0b9e3a851e4867c808862d67e
Reviewed-on: https://chromium-review.googlesource.com/568855Reviewed-by: Mike Reed <reed@chromium.org>
Reviewed-by: Fredrik Söderquist <fs@opera.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486210}

BUG=672263

Review-Url: https://codereview.chromium.org/2949343002
Cr-Commit-Position: refs/heads/master@{#486209}

Ref: https://codereview.chromium.org/2450963002/

The reason for this CL is that when I tried to turn PVer4 as default
on bots, the bots started failing. The reason for those failures is
that the tests in SafeBrowsingServiceTest instantiate PVer3-like
DB manager explicitly, but the base::feature list says that PVer4
is enabled.

The fix is to check the method on the DB manager, instead of the
feature list, inside chrome_subresource_filter_client.cc


Bug: 543161
Change-Id: I4c7772248c36f78fcac530ea15cd90b9e24ea382
Reviewed-on: https://chromium-review.googlesource.com/567394Reviewed-by: Charlie Harrison <csharrison@chromium.org>
Reviewed-by: Jialiu Lin <jialiul@chromium.org>
Commit-Queue: Varun Khaneja <vakh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486208}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/6c40c273a7fe..1286055c7baf

$ git log 6c40c273a..1286055c7 --date=short --no-merges --format='%ad %ae %s'
2017-07-12 adithyas Add table for Blink RuntimeCallStats
2017-07-12 bpastene devil: Add a device_utils method to mv-then-rm a device file.
2017-07-12 nednguyen Rename webpagereplay_go.py to webpagereplay_go_server.py
2017-07-12 nednguyen Add WPR go binary for Mac platform
2017-07-12 bpastene devil: When pushing data deps to device, don't mkdir redundantly.
2017-07-12 simonhatch Dashboard - Fix test-picker add button never un-greying
2017-07-12 simonhatch Dashboard - Add api_request_handler
2017-07-12 etienneb Fix symbolisation for address out-of-range

Created with:
  roll-dep src/third_party/catapult
BUG=724543,739899,739899,739783


Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls


CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=sullivan@chromium.org

Change-Id: I114d56bf15f445060eb48951a120fd99819e0024
Reviewed-on: https://chromium-review.googlesource.com/569138
Reviewed-by: <catapult-deps-roller@chromium.org>
Commit-Queue: <catapult-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486207}

This makes UKM able to be enabled on Android when no Incognito windows
are open.

BUG=739971

Change-Id: I310d540d59594bc3377adf9eedb8c05502c5f5bb
Reviewed-on: https://chromium-review.googlesource.com/566241
Commit-Queue: Steven Holte <holte@chromium.org>
Reviewed-by: Robert Kaplow <rkaplow@chromium.org>
Reviewed-by: David Trainor <dtrainor@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486206}

Previously, almost all usages of OffsetAdjuster applied a post-processing
step to clamp the resulting offsets to a "sane" range.

This refactor moves the clamping into the base function, but as an
optional parameter.

This also reduces the API surface by removing the LimitOffset class.

BUG=732582

Review-Url: https://codereview.chromium.org/2953943003
Cr-Commit-Position: refs/heads/master@{#486205}

WallpaperController in ash needs access to local state for color caching.
Whether or not we combine WallpaperManager and WallpaperController, we'll
need local state to be accessible to ash just as user pref is.

Bug: NONE
Change-Id: I8360402864d0cdb173f3a84731d6abdeaf12a4ce
Reviewed-on: https://chromium-review.googlesource.com/566026
Commit-Queue: Wenzhao (Colin) Zang <wzang@chromium.org>
Reviewed-by: James Cook <jamescook@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486204}

Currently, MediaPlayerRender has no way of knowing whether or not the
RenderFrameHost that created it has been destroyed. MediaPlayerRender
only gets notified when the RenderFrame gets destroyed, and
MojoRendererService detects that the other end of the connection has
been closed. It is possible for MediaPlayerRenderer to still be
initializing when the RenderFrameHost has been destroyed. This leads to
crashes when trying to access RenderProcessHost resources when trying to
create a MediaResourceGetter.

This CL removes the unsafe passing of a RenderFrameHost*, adds a check
to see whether the creating RFH is still alive, and gets the
RenderProcessHost directly from its ID.

Bug: 675584
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_site_isolation
Change-Id: Ie8c9936dc0f3e132cc63cadf3bb3ebcf464621c1
Reviewed-on: https://chromium-review.googlesource.com/525133Reviewed-by: Chris Watkins <watk@chromium.org>
Reviewed-by: Nick Carter <nick@chromium.org>
Commit-Queue: Thomas Guilbert <tguilbert@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486203}

This is ongoing refactoring of offline pages, in order to remove/isolate
Android-specific dependdencies. In this case, it's just a result
of:
tools/git/move_source_file.py chrome/browser/android/offline_pages/offline_page_bookmark_observer.* chrome/browser/offline_pages/

Bug: 722467
Change-Id: I35ecde79bee6b6fba7d7c27d75c5fb7be1799c00
Reviewed-on: https://chromium-review.googlesource.com/568565Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Yafei Duan <romax@chromium.org>
Commit-Queue: Dmitry Titov <dimich@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486202}

This fixes a test failure caused by https://chromium-review.googlesource.com/c/567235.

Bug: 672263
Change-Id: I861f8170827f60e3aab7abdf56c1da35d8b6bb92
Reviewed-on: https://chromium-review.googlesource.com/569068Reviewed-by: Kyle Horimoto <khorimoto@chromium.org>
Commit-Queue: Kyle Horimoto <khorimoto@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486201}

Bug: 740871
Change-Id: Ica5bcd76a3f2995a8f414c80698e468f8780734c
Reviewed-on: https://chromium-review.googlesource.com/568595Reviewed-by: Kentaro Hara <haraken@chromium.org>
Commit-Queue: Eric Roman <eroman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486200}

https://skia.googlesource.com/skia.git/+log/1c8bb8a4c63c..d425a1d013b6

$ git log 1c8bb8a4c..d425a1d01 --date=short --no-merges --format='%ad %ae %s'
2017-07-12 halcanary SkPDF:  start to support AlphaOnly Images as masks.
2017-07-12 egdaniel Add arc support to gpu Obj c++ code
2017-07-12 halcanary SkPDF: simplify drawImage/Bitmap/Sprite code.
2017-07-12 fmalita Add some raster pipeline perspective asserts
2017-07-12 herb Experimental blur code with 32 bit fix.
2017-07-12 brianosman On Windows, always re-create window for GL or Vulkan
2017-07-12 brianosman Move expensive initialization to onOnceBeforeDraw
2017-07-12 bsalomon Revert "Revert "Make DashOp a non-legacy GrMeshDrawOp.""
2017-07-12 reed remove references to kIndex_8_SkColorType from our tools/tests
2017-07-12 liyuqian Implement Sk4i's abs, min, max
2017-07-12 brianosman Extend final point in closed degenerate hairline segments
2017-07-12 bsalomon Fix assert in lattice op triggered by GLProgramsTest
2017-07-12 fmalita Clarify adjustMatrixAndAppendStages semantics
2017-07-12 bsalomon Revert "Make DashOp a non-legacy GrMeshDrawOp."

Created with:
  roll-dep src/third_party/skia


Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls


CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=ethannicholas@chromium.org

Change-Id: I832b49480e40cf659d748c618f7d2ee687a4bc0c
Reviewed-on: https://chromium-review.googlesource.com/569010Reviewed-by: Skia Deps Roller <skia-deps-roller@chromium.org>
Commit-Queue: Skia Deps Roller <skia-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486199}

Change-Id: I73a4bca08180684a447a1e41c4e268e322a8492d
Reviewed-on: https://chromium-review.googlesource.com/568714Reviewed-by: Andrey Lushnikov <lushnikov@chromium.org>
Commit-Queue: Pavel Feldman <pfeldman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486198}

ImageQualityController used heuristics to draw resizing images at a lower
quality during resizes, then snap them to high quality once resizing stops.
This was needed for performance of high-quality image scales in the past, but
is less useful today because images are never scaled with a high quality filter.

This patch completely removes the ImageQualityController which has the
following benefits:
1) One less virtual on ImageResourceObserver.
2) No more visual snap as image quality changes (https://crbug.com/503040).
3) A call removed from LayoutBoxModelObject::WillBeDestroyed.
4) Much complexity removed.

[1] https://chromium-review.googlesource.com/c/541609/

Bug: 503040, 734215
Change-Id: I6552767925887b3154a99339976a4b624e64bac5
Reviewed-on: https://chromium-review.googlesource.com/567564Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Commit-Queue: Philip Rogers <pdr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486197}

Bug: 740993
Change-Id: I742b2aa94048e55204d204e0ce6fcaee7b0af6d8
Reviewed-on: https://chromium-review.googlesource.com/567776
Commit-Queue: Aleks Totic <atotic@chromium.org>
Reviewed-by: Christian Biesinger <cbiesinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486196}

After some offline discussion, we felt that we shouldn't have files
directly in the common directory so that we can better control DEPS.

This CL moves SurfaceId, FrameSinkId, LocalSurfaceId, FrameSinkIdAllocator,
and LocalSurfaceIdAllocator to the components/viz/common/surfaces directory
and updates includes.

TBR=tsepez@chromium.org

Bug: 722935
Change-Id: I7d301b277dbf07ead10ec96771ee64469d2bcbac
Reviewed-on: https://chromium-review.googlesource.com/568372Reviewed-by: Fady Samuel <fsamuel@chromium.org>
Reviewed-by: Bo Liu <boliu@chromium.org>
Reviewed-by: Justin Novosad <junov@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Commit-Queue: Fady Samuel <fsamuel@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486195}

Bug: 689520
Change-Id: I65b372857d44c52f97986d02598c74db8d423312
Reviewed-on: https://chromium-review.googlesource.com/567783Reviewed-by: Avi Drissman <avi@chromium.org>
Commit-Queue: Sam McNally <sammc@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486194}

Change-Id: I84cc5a2fb6b09daa12fe4560664773cfaf635d54
Reviewed-on: https://chromium-review.googlesource.com/569078Reviewed-by: Yuwei Huang <yuweih@chromium.org>
Commit-Queue: Scott Nichols <nicholss@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486193}

to avoid file io.

Bug: 740801
Change-Id: I64c21aabb75bdc8cd2926a665fad9ab2bccfe1d9
Reviewed-on: https://chromium-review.googlesource.com/567396Reviewed-by: Robert Sesek <rsesek@chromium.org>
Commit-Queue: Michael Nordman <michaeln@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486192}

This verifies the potential for use-after-free that can occur if the
WaitableEventWatcher is deleted after the event is signaled but before the
callback is run.

Bug: 681167
Bug: 741256
Change-Id: Ic5ca2bc48cb51d5f50be7e67773f8213fbbc80d3
Reviewed-on: https://chromium-review.googlesource.com/568371
Commit-Queue: Robert Sesek <rsesek@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486191}

Bug: none
Change-Id: I1121e7b64cceb1c70e9a8115d9f9cc9d63716133
Reviewed-on: https://chromium-review.googlesource.com/568903Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Commit-Queue: Evan Stade <estade@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486190}

1. Added a dummy FontMgr for Fuchsia
2. Fixed libjpeg_turbo linking issues (ELF wasn't defined)

Bug: 737802
Change-Id: I3a48e448fd03a350130680ced5fafc309fc00ae9
Reviewed-on: https://chromium-review.googlesource.com/567032
Commit-Queue: Sergey Ulanov <sergeyu@chromium.org>
Reviewed-by: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#486189}