GitLab

In Yosemite, the owner of a nib file is retain/autoreleased during the
initialization of the nib. I wrapped the constructor of the
BookmarkBarController in an @autoreleasepool block to allow the test to control
the destruction timing of the BookmarkBarController.

BUG=390276

Review URL: https://codereview.chromium.org/660063004

Cr-Commit-Position: refs/heads/master@{#300312}

This might-maybe-hope-not crash, and so has debugging statements.

BUG=407376
TEST=no crashing, we hope
TBR=rsesek@chromium.org

Review URL: https://codereview.chromium.org/669613005

Cr-Commit-Position: refs/heads/master@{#300311}

Recently class was refactored https://codereview.chromium.org/651253002 ,
and there was an unexpected functional change.

BUG=424816

Review URL: https://codereview.chromium.org/669443002

Cr-Commit-Position: refs/heads/master@{#300310}

This makes some sense, since as we potentially output more and more files
(instead of a single results.html), we want to store them all in the same
location.

SHERIFFS: If error "This flag is deprecated. Please use --output-dir instead."
arises, this patch can be safely reverted.

Original discussion at https://codereview.chromium.org/616063004/

BUG=

Review URL: https://codereview.chromium.org/651383002

Cr-Commit-Position: refs/heads/master@{#300309}

Servers do not reliably send close_notify. This regressed in
https://codereview.chromium.org/367963007 which had a side effect of making us
sensitive to this.

BUG=422246

Review URL: https://codereview.chromium.org/655813003

Cr-Commit-Position: refs/heads/master@{#300308}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/a71aee6..33e6466

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

Review URL: https://codereview.chromium.org/667013002

Cr-Commit-Position: refs/heads/master@{#300307}

Also move some extension headers to ENABLE_EXTENSIONS sections.

BUG=424746

Review URL: https://codereview.chromium.org/662783002

Cr-Commit-Position: refs/heads/master@{#300306}

Settings overlays are generally cancel/OK, donottrack was OK/cancel.

BUG=425141
R=dbeam@chromium.org

Review URL: https://codereview.chromium.org/665113003

Cr-Commit-Position: refs/heads/master@{#300305}

Navigation transitions (web to native app): Get names and rects of transition elements (Chrome side)

Web to native app navigation transition uses Activity Transitions APIs in Android L. It requires the names and rects of transition elements. This CL gets the names and rects and pass them to TransitionRequestManager.

This is the Chrome side of the CL. Blink side will be done after this CL is landed. https://codereview.chromium.org/654953002/

Design doc: https://docs.google.com/a/chromium.org/document/d/17jg1RRL3RI969cLwbKBIcoGDsPwqaEdBxafGNYGwiY4/edit#
Demo video: https://drive.google.com/a/google.com/file/d/0B3hetueIc91Gd01DU25uT2hWU2M/view?usp=sharing
Activity Transitions in Android L: https://developer.android.com/preview/material/animations.html#transitions

BUG=370696

Review URL: https://codereview.chromium.org/652283002

Cr-Commit-Position: refs/heads/master@{#300304}

This was the 2nd breaking mistake in that one CL. Note: oth mistakes were not caught by unit tests or by running a bisect try job once before submitting.

BUG=425173

Review URL: https://codereview.chromium.org/662403003

Cr-Commit-Position: refs/heads/master@{#300303}

BUG=397984

Review URL: https://codereview.chromium.org/658973003

Cr-Commit-Position: refs/heads/master@{#300302}

This depends on https://codereview.chromium.org/662763002/

BUG=424750

Review URL: https://codereview.chromium.org/651773003

Cr-Commit-Position: refs/heads/master@{#300301}

This cleans up the the interface to BrowserCompositorViewMac to
prepare for moving it closer to ui::views.

BUG=424058

Review URL: https://codereview.chromium.org/638123003

Cr-Commit-Position: refs/heads/master@{#300300}

BUG=424973
TEST=chromeos=1 compile,
TEST=out_chromeos/Debug/sandbox_linux_unittests --gtest_filter=*SIGSYS_InvalidSyscall*
...
0 tests run

R=mdempsky@chromium.org

Review URL: https://codereview.chromium.org/646853005

Cr-Commit-Position: refs/heads/master@{#300299}

BUG=224786

Review URL: https://codereview.chromium.org/611383004

Cr-Commit-Position: refs/heads/master@{#300298}

BVR::DidDestroyCompositor can be called at any point, which resets
the compositor pointer. Which means each use of the pointer
should be checked or DCHECKed. In particular, DidDestroyCompositor
can happen before tearing down hardware mode.

This can happen when javascript calls window.close, which closes the
RenderView before WebContents.

BUG=424614

Review URL: https://codereview.chromium.org/654623002

Cr-Commit-Position: refs/heads/master@{#300297}

This is a preparation for media code to call into content code to create the CDMs, so that we could have EME support without depending on content code.

BUG=422730

Review URL: https://codereview.chromium.org/660673002

Cr-Commit-Position: refs/heads/master@{#300296}

New tests:
- Added new ChromeProxyExplicitBypass integration tests that test "bypass=0", "bypass=<X>", "block=0", and "block=<X>" Chrome-Proxy directives using the test server
- Added new ChromeProxyFallback integration test that gets a "bypass=0" directive from the real data reduction proxy

Cleanup:
- Rewrote the VerifyBadProxies logic, adding support for per-proxy retry times
- Also changed VerifyBadProxies to raise an exception instead of returning false when the actual and expected bad proxy lists were different, which was causing all the existing bypass tests to pass in cases when they should have failed
- Fixed typo in ChromeProxyCorsBypass test, it passes now
- Some general python style cleanup

BUG=374397

Review URL: https://codereview.chromium.org/659333004

Cr-Commit-Position: refs/heads/master@{#300295}

Revert of Temporarily suppress Android API deprecation warnings. (patchset #1 id:1 of https://chromiumcodereview.appspot.com/615163002/)

Reason for revert:
We have switched to L sdk.

Original issue's description:
> Temporarily suppress Android API deprecation warnings.
>
> We will remove this suppression once all of the builders support the L SDK.
>
> BUG=405174,398669,411361,411366,411367,411376,416041
>
> Committed: https://crrev.com/c26c106e7724c400fed9980ce371a1c81d1af849
> Cr-Commit-Position: refs/heads/master@{#297505}

TBR=dmazzoni@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=405174,398669,411361,411366,411367,411376,416041

Review URL: https://codereview.chromium.org/649653004

Cr-Commit-Position: refs/heads/master@{#300294}

Revert of Roll src/third_party/skia a71aee6:958e962 (patchset #1 id:1 of https://codereview.chromium.org/663133002/)

Reason for revert:
Broke all Mac GPU debug bots.

0   libbase.dylib                       0x0cc6b4bf base::debug::StackTrace::StackTrace() + 63
1   libbase.dylib                       0x0cc6b51b base::debug::StackTrace::StackTrace() + 43
2   libbase.dylib                       0x0cd197b2 logging::LogMessage::~LogMessage() + 82
3   libbase.dylib                       0x0cd1852b logging::LogMessage::~LogMessage() + 43
4   libskia.dylib                       0x24256d1b SkDebugf_FileLine(char const*, int, bool, char const*, ...) + 283
5   libskia.dylib                       0x24069f85 Sk4x<float>::store(float*) const + 213
6   libskia.dylib                       0x24068ea0 SkTileGrid::intersectsGrid(Sk4x<float> const&) const + 64
7   libskia.dylib                       0x24069099 SkTileGrid::insert(unsigned int, SkRect const&, bool) + 425
8   libskia.dylib                       0x23fe75a1 SkRecords::FillBounds::FillBounds(SkRecord const&, SkBBoxHierarchy*) + 753
9   libskia.dylib                       0x23fe42d7 SkRecords::FillBounds::FillBounds(SkRecord const&, SkBBoxHierarchy*) + 71
10  libskia.dylib                       0x23fe3ad1 SkRecordFillBounds(SkRecord const&, SkBBoxHierarchy*) + 65
11  libskia.dylib                       0x23faeb86 SkPicture::SkPicture(float, float, SkRecord*, SkBBoxHierarchy*) + 390
12  libskia.dylib                       0x23fadcc1 SkPicture::SkPicture(float, float, SkRecord*, SkBBoxHierarchy*) + 113
13  libskia.dylib                       0x23fd4d12 SkPictureRecorder::endRecording() + 162
14  libcc.dylib                         0x22402cbc cc::Picture::Record(cc::ContentLayerClient*, SkTileGridFactory::TileGridInfo const&, cc::Picture::RecordingMode) + 2188
15  libcc.dylib                         0x2240235e cc::Picture::Create(gfx::Rect const&, cc::ContentLayerClient*, SkTileGridFactory::TileGridInfo const&, bool, cc::Picture::RecordingMode) + 174
16  libcc.dylib                         0x2242bc3f cc::PicturePile::UpdateAndExpandInvalidation(cc::ContentLayerClient*, cc::Region*, unsigned int, bool, bool, gfx::Size const&, gfx::Rect const&, int, cc::Picture::RecordingMode, cc::RenderingStatsInstrumentation*) + 7391
17  libcc.dylib                         0x222e101f cc::PictureLayer::Update(cc::ResourceUpdateQueue*, cc::OcclusionTracker<cc::Layer> const*) + 1887
18  libcc.dylib                         0x22556174 cc::LayerTreeHost::PaintLayerContents(cc::RenderSurfaceLayerList const&, cc::ResourceUpdateQueue*, bool*, bool*) + 820
19  libcc.dylib                         0x225556c9 cc::LayerTreeHost::UpdateLayers(cc::Layer*, cc::ResourceUpdateQueue*) + 2665
20  libcc.dylib                         0x22554bc8 cc::LayerTreeHost::UpdateLayers(cc::ResourceUpdateQueue*) + 648
21  libcc.dylib                         0x225fce07 cc::ThreadProxy::BeginMainFrame(scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >) + 2919
22  libcc.dylib                         0x2260e923 base::internal::RunnableAdapter<void (cc::ThreadProxy::*)(scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >)>::Run(cc::ThreadProxy*, scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >) + 179
23  libcc.dylib                         0x2260e7c7 base::internal::InvokeHelper<true, void, base::internal::RunnableAdapter<void (cc::ThreadProxy::*)(scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >)>, void (base::WeakPtr<cc::ThreadProxy> const&, scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >)>::MakeItSo(base::internal::RunnableAdapter<void (cc::ThreadProxy::*)(scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >)>, base::WeakPtr<cc::ThreadProxy> const&, scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >) + 151
24  libcc.dylib                         0x2260e607 base::internal::Invoker<2, base::internal::BindState<base::internal::RunnableAdapter<void (cc::ThreadProxy::*)(scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >)>, void (cc::ThreadProxy*, scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >), void (base::WeakPtr<cc::ThreadProxy>, base::internal::PassedWrapper<scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> > >)>, void (cc::ThreadProxy*, scoped_ptr<cc::ThreadProxy::BeginMainFrameAndCommitState, base::DefaultDeleter<cc::ThreadProxy::BeginMainFrameAndCommitState> >)>::Run(base::internal::BindStateBase*) + 215
25  libbase.dylib                       0x0cc5474f base::Callback<void ()>::Run() const + 63
26  libbase.dylib                       0x0cc6d9bc base::debug::TaskAnnotator::RunTask(char const*, char const*, base::PendingTask const&) + 1084
27  libbase.dylib                       0x0cd5a999 base::MessageLoop::RunTask(base::PendingTask const&) + 793
28  libbase.dylib                       0x0cd5ab42 base::MessageLoop::DeferOrRunPendingTask(base::PendingTask const&) + 98
29  libbase.dylib                       0x0cd5ae11 base::MessageLoop::DoWork() + 321
30  libbase.dylib                       0x0cc2c715 base::MessagePumpCFRunLoopBase::RunWork() + 101
31  libbase.dylib                       0x0cc2bb31 base::MessagePumpCFRunLoopBase::RunWorkSource(void*) + 49
32  CoreFoundation                      0x9bbd3b5f __CFRUNLOOP_IS_CALLING_OUT_TO_A_SOURCE0_PERFORM_FUNCTION__ + 15
33  CoreFoundation                      0x9bbc495b __CFRunLoopDoSources0 + 235
34  CoreFoundation                      0x9bbc405e __CFRunLoopRun + 1022
35  CoreFoundation                      0x9bbc39ea CFRunLoopRunSpecific + 394
36  CoreFoundation                      0x9bbc384b CFRunLoopRunInMode + 123
37  Foundation                          0x902b85b9 -[NSRunLoop(NSRunLoop) runMode:beforeDate:] + 277
38  libbase.dylib                       0x0cc2d242 base::MessagePumpNSRunLoop::DoRun(base::MessagePump::Delegate*) + 162
39  libbase.dylib                       0x0cc2c328 base::MessagePumpCFRunLoopBase::Run(base::MessagePump::Delegate*) + 104
40  libbase.dylib                       0x0cd5a30c base::MessageLoop::RunHandler() + 300
41  libbase.dylib                       0x0cdcc5f3 base::RunLoop::Run() + 83
42  libbase.dylib                       0x0cd59617 base::MessageLoop::Run() + 55
43  libcontent.dylib                    0x1d021a35 content::RendererMain(content::MainFunctionParams const&) + 1925
44  libcontent.dylib                    0x1b9782ac content::RunNamedProcessTypeMain(std::string const&, content::MainFunctionParams const&, content::ContentMainDelegate*) + 268
45  libcontent.dylib                    0x1b97999b content::ContentMainRunnerImpl::Run() + 779
46  libcontent.dylib                    0x1b977745 content::ContentMain(content::ContentMainParams const&) + 133
47  libchrome_main_dll.dylib            0x0004155c ChromeMain + 92
48  Chromium Helper                     0x00038f7b main + 43
49  Chromium Helper                     0x00038f45 start + 53

Original issue's description:
> Roll src/third_party/skia a71aee6:958e962
>
> Summary of changes available at:
> https://chromium.googlesource.com/skia/+log/a71aee6..958e962
>
> CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
> TBR=mtklein@google.com
>
> Committed: https://crrev.com/fbf0dae317434cf09510db0da0e5b9126b5dfa9b
> Cr-Commit-Position: refs/heads/master@{#300273}

TBR=mtklein@google.com,skia-deps-roller@chromium.org
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/669743002

Cr-Commit-Position: refs/heads/master@{#300293}

As a part of migrating away from NSS, CrOS is going to start
generating the device Owner keypair using Chaps, which is
exposed to the browser via the TPM-backed PK11 token that gets
loaded into each user's private NSS PK11 slot. To support the
current mode of operation as well as the new, make
OwnerSettingsServiceChromeOS check both slots for the key.

BUG=422036
TEST=unit test, ownership autotests
R=stevenjb

Review URL: https://codereview.chromium.org/660343002

Cr-Commit-Position: refs/heads/master@{#300292}

Review URL: https://codereview.chromium.org/667483003

Cr-Commit-Position: refs/heads/master@{#300291}

nacl_helper_nonsfi is actually NOT the untrusted code, but this is built by
the PNaCl toolchain, which is native_client untrusted toolchain in the
context of common.gypi. In order to keep the other _nacl.gyp files simple,
this CL adds disable_nacl_untrusted==0 guard to the nacl_helper_nonsfi
creating condition, so that it does not cause the gyp error due to depending
on missing targets when disable_nacl_untrusted is set to 1.

BUG=358465
TEST=Ran trybots.

Review URL: https://codereview.chromium.org/664153002

Cr-Commit-Position: refs/heads/master@{#300290}

BUG=423699, 390618

Review URL: https://codereview.chromium.org/643923004

Cr-Commit-Position: refs/heads/master@{#300289}

* New ContainerParams
 -  default_parent to specify the default parent when no
    transient parent is specified.
 - modal_container_priority now controls the modal container
   used for a modal window created for the given container.
 - if not specified, it will fallback downwards to find one.
 - if window is specified as always_on_top, it will use top
   most modal container.

* Changed network selector/shutdown dialog to use the new API.

* Other change:
 Separated test windows to athena/test/base/test_windows.h

BUG=410499
TBR=sky@chromium.org,reed@chromium.org
TEST=coverd by unit tests

Committed: https://crrev.com/4903fd36b2b36f00efeb1b7bba81b7de6e9457a3
Cr-Commit-Position: refs/heads/master@{#300192}

Review URL: https://codereview.chromium.org/662763002

Cr-Commit-Position: refs/heads/master@{#300288}

R=estade@chromium.org
BUG=424052

Review URL: https://codereview.chromium.org/669573002

Cr-Commit-Position: refs/heads/master@{#300287}

This used to happen on view initiated destruction, as opposed to
RenderWidgetHostImpl initiated ones.  In RWH initiated destruction,
the views gets deleted properly.

"View initiated destruction" means the view is first deleted before
the host, this is aura's RWHViewAura::OnWindowDestroyed() or mac's
~RenderWidgetHostViewMac().

In view(V1) initiated destruction, RWH::SetView(NULL) call makes
the host(H) drop the pointer to the view(V2) it's holding, in cases
where the view initiated the destruction is not the view of
the host(V1!=V2), it will leak V2.

This scenario happens when there's an interstitial page showing
inside <webview> guest. The platform RWHView gets deleted but
the RWHViewGuest leaks as the RWH's view pointer is cleared while
clearing the platform view.

BUG=321662
Test=No visible change, internal change.
To check leak sanitizer test, my steps are:

gyp:
CC=clang CXX=clang++ builddir_name=./tmp-llvm GYP_PARALLEL=1 \
   PATH=$PWD/third_party/llvm-build/Release+Asserts/bin:$PATH \
   GYP_GENERATORS='ninja' ./build/gyp_chromium -Gconfig=Release \
   -Goutput_dir=out_asan_new -D"asan=1" -D"clang=1" \
   -D"component=static_library" -D"lsan=1" -D"target_arch=x64" \
   -D"use_allocator=none"

build:
PATH=$PWD/third_party/llvm-build/Release+Asserts/bin:$PATH \
     ninja -C out_asan_new/Release -j 2048 browser_tests

run:
ASAN_OPTIONS="detect_leaks=1" \
    LSAN_OPTIONS="suppressions=./tools/lsan/suppressions.txt" \
    ./out_asan_new/Release/browser_tests \
    --gtest_filter=WebViewTest.InterstitialTeardown \
    --disable-seccomp-sandbox

Review URL: https://codereview.chromium.org/647613002

Cr-Commit-Position: refs/heads/master@{#300286}

This shrinks the android ChromeShell binary by 1076 bytes, and makes the
code easier for developers to read.

Performance-wise this change seems neutral (as opposed to the perf wins
from similar changes for Rect and Point classes). Performance numbers
are at http://crbug.com/407444#c19

R=sky
BUG=407444

Review URL: https://codereview.chromium.org/662273002

Cr-Commit-Position: refs/heads/master@{#300285}

While there, remove some error messages because of incomplete test setup or incorrect fake behavior.

BUG=425049

Review URL: https://codereview.chromium.org/648623004

Cr-Commit-Position: refs/heads/master@{#300284}

This shrinks the android ChromeShell APK by 600 bytes, and improves
performance on compositor benchmarks.

Numbers can be found here https://crbug.com/407444#c15 but the summary
is this saves about 0.1ms cpu time on the JS_FULL_SCREEN_INVALIDATION
test in the thread_times.tough_compositor_cases test suite.

This also makes it easier to read and find functions in the Point and
PointF classes for developers.

R=sky
BUG=407444

Review URL: https://codereview.chromium.org/642343003

Cr-Commit-Position: refs/heads/master@{#300283}

Clicking anywhere in an entry already toggles the checkbox and this was
just messing up VoiceOver spoken feedback.

Also add aria-label to satisfy a11y audit and make voice feedback better.

R=dmazzoni@chromium.org
BUG=385344

Review URL: https://codereview.chromium.org/661043006

Cr-Commit-Position: refs/heads/master@{#300282}

The reverse writer for viewsource had an incorrect assumption about the path component of URL which was causing a duplication of query and fragment parameters. This change is to unit test the fix that  went in
https://codereview.chromium.org/655023002/

BUG=336846

Review URL: https://codereview.chromium.org/659153003

Cr-Commit-Position: refs/heads/master@{#300281}

Revert of Requiring confidence in initial regression range before bisecting. (patchset #4 id:110001 of https://codereview.chromium.org/644323002/)

Reason for revert:
Bisect is broken because the ConfidenceScore expects list of list, but here only list is passed.
http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/693/steps/Running%20Bisection/logs/stdio

Original issue's description:
> Requiring confidence in initial regression range before bisecting.
>
> BUG=422727
>
> Committed: https://crrev.com/4436048a948a2010062d3c6a171775d9389fd732
> Cr-Commit-Position: refs/heads/master@{#300196}

TBR=qyearsley@chromium.org,chrisphan@chromium.org,tonyg@chromium.org,robertocn@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=422727

Review URL: https://codereview.chromium.org/661393003

Cr-Commit-Position: refs/heads/master@{#300280}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/6220dd2..7541f2d

TBR=dstockwell@chromium.org,aboxhall@chromium.org

Review URL: https://codereview.chromium.org/646643004

Cr-Commit-Position: refs/heads/master@{#300279}

This message filter is only being used when running layout tests.

BUG=420994
R=mkwst@chromium.org

Review URL: https://codereview.chromium.org/659073004

Cr-Commit-Position: refs/heads/master@{#300278}

This deletes code that froze the animation clock when we skipped
frames because of checkerboarding. This logic was only enabled
for non-impl-side painting (to handle the case of animating very large
layers, where failing to freeze the clock would cause rasterization to
constantly lag behind, resulting in checkerboarding whenever a draw was
forced). Since all shipping renderers use impl-side painting now, this
logic is no longer needed.

BUG=None

Review URL: https://codereview.chromium.org/654203007

Cr-Commit-Position: refs/heads/master@{#300277}

This is the second step to update the Blink APIs.
To remove old Blink APIs that do not have isActive flag,
update all callers to specify the isActive flag.

The first change to add the new Blink APIs is
https://codereview.chromium.org/649683006/

BUG=279097
TBR=scherkus@chromium.org

Review URL: https://codereview.chromium.org/661323002

Cr-Commit-Position: refs/heads/master@{#300276}

This is not the right layer to do this, since it forces all test suites to have
an initialized Cocoa NSApp. Instead, only do it for test suites that require it.

BUG=71686

Review URL: https://codereview.chromium.org/656293003

Cr-Commit-Position: refs/heads/master@{#300275}

When the microphone volume is low, the pre-processed audio data contains energy, while the post-processed audio data has only 0, we report 0 as energy before this patch, which is wrong since the audio flow contains energy.

This patch fix the problem by checking the energy of pre-processed data before reporting 0 as energy.

BUG=424149
TEST=1, use a USB headset.
2, set the microphone volume low, like below 50%.
3, join a hangout call.
4, open chrome://webrtc-internals, and look at the energy of input stream, constantly it is 0 there.

Review URL: https://codereview.chromium.org/661693003

Cr-Commit-Position: refs/heads/master@{#300274}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/a71aee6..958e962

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

Review URL: https://codereview.chromium.org/663133002

Cr-Commit-Position: refs/heads/master@{#300273}