GitLab

Revert of Improve the ScopedHandle verifier. (patchset #1 of https://codereview.chromium.org/506013004/)

Reason for revert:
All windows perf bots (official builds) crashing.

http://build.chromium.org/p/chromium.perf/builders/Win%20Builder/builds/73062/steps/generate_telemetry_profiles/logs/stdio

	ChildEBP RetAddr
	0022f7c0 6a3ffa19 chrome_69c90000!base::debug::BreakDebugger+0x10
	0022f830 6a400622 chrome_69c90000!CheckIsChromeSxSProcess+0x26
	0022f834 6a400d52 chrome_69c90000!InstallUtil::IsChromeSxSProcess+0x16
	0022f83c 6a400c29 chrome_69c90000!BrowserDistribution::GetSpecificDistribution+0x3b
	0022f844 6a3fe166 chrome_69c90000!BrowserDistribution::GetDistribution+0x7
	0022f930 6a3fe11d chrome_69c90000!`anonymous namespace'::GetChromeChannelInternal+0x2c
	0022f948 6a092c6c chrome_69c90000!GoogleUpdateSettings::GetChromeChannel+0x21
	0022f9b4 6a08f70f chrome_69c90000!chrome::VersionInfo::GetChannel+0x60
	0022f9b8 6a08f65a chrome_69c90000!`anonymous namespace'::UseHooks+0x5
	0022f9dc 6a08eb5a chrome_69c90000!InstallCloseHandleHooks+0x15
	0022fa20 00fb7623 chrome_69c90000!ChromeMain+0x3e
	0022fab0 00fb7026 chrome!MainDllLoader::Launch+0x15f
	0022faf4 00fd936a chrome!wWinMain+0x5a
	0022fb40 766e338a chrome!__tmainCRTStartup+0xfd
	WARNING: Stack unwind information not available. Following frames may be wrong.
	0022fb4c 76f99f72 kernel32!BaseThreadInitThunk+0x12
	0022fb8c 76f99f45 ntdll!RtlInitializeExceptionChain+0x63
	0022fba4 00000000 ntdll!RtlInitializeExceptionChain+0x36

Original issue's description:
> Improve the ScopedHandle verifier.
>
> 1. Automate the selection of the proper channel to enable the verifier.
>    Now the code is enabled at runtime.
>
> 2. Switch to a hash_map to track handles.
>
> 3. Intercept CloseHandle to detect the code that is closing handles owned
>    by ScopedHandles. The initial implementation only covers chrome.exe/dll,
>    but the plan is to extend that in the future to all modules loaded in the
>    process.
>
> BUG=362176
> R=cpu@chromium.org
> R=sky@chromium.org
>
> See https://codereview.chromium.org/490043002/ for the actual review.
>
> TBR=cpu@chromium.org
> TBR=sky@chromium.org
>
> Committed: https://chromium.googlesource.com/chromium/src/+/c928d0383db43f2f4baf8f9b24ed7454bf7eda64

TBR=cpu@chromium.org,sky@chromium.org,rvargas@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=362176

Review URL: https://codereview.chromium.org/507893002

Cr-Commit-Position: refs/heads/master@{#292047}

The tests actually deadlock in practice.

BUG=407452
TBR=xhwang@chromium.org

Review URL: https://codereview.chromium.org/507163002

Cr-Commit-Position: refs/heads/master@{#292046}

This patch was generated by running the rewrite_scoped_refptr clang tool
on a Linux build.

BUG=110610

Review URL: https://codereview.chromium.org/508733002

Cr-Commit-Position: refs/heads/master@{#292045}

AnalysisRaster was needed because we must run the analysis on a separate
thread.  My second attempt placed the analysis in the picturepileimpl,
but, as vmpstr predicted, this caused issues when software rasterizing
since it uses yet another thread.  The way around all of this was to
write another raster function without the:
  DCHECK(raster_thread_checker_.CalledOnValidThread());

We're calling the Analysis before any synchronization occures with impl
so it should be safe to call without the check.

BUG=396908

Review URL: https://codereview.chromium.org/494503002

Cr-Commit-Position: refs/heads/master@{#292044}

It's no longer necessary as I fixed the underlying bug in the recipe
side.

BUG=407419
TEST=none
R=jam@chromium.org
TBR=jam@chromium.org

Review URL: https://codereview.chromium.org/507853002

Cr-Commit-Position: refs/heads/master@{#292043}

BUG=407357
TBR=rsesek@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/510473003

Cr-Commit-Position: refs/heads/master@{#292042}

This patch was generated by running the rewrite_scoped_refptr clang tool
on a Linux build.

BUG=110610

Review URL: https://codereview.chromium.org/488463005

Cr-Commit-Position: refs/heads/master@{#292041}

This version of the dll seems to be causing a disproportionate number of
crashes, so we fall back to software decoding if that is the version of
the DLL on the user's machine.

BUG=403440

Review URL: https://codereview.chromium.org/502183002

Cr-Commit-Position: refs/heads/master@{#292040}

https://skia.googlesource.com/skia/+log/76db31a2ca11da4dd45016e964b2e49f39f3dd49..2a65a238b0e688356aee09d08f3f8a317359b4d2

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=djsollen@google.com

Review URL: https://codereview.chromium.org/504223004

Cr-Commit-Position: refs/heads/master@{#292039}

* AppActivity is now base class. There will be one for chrome
* Moved content_activity_factory.h from public to impl as ther eis no need to expose header.
* Added CreateAppActivity that takes apps::Window. Chrome
 impl will use this to create an activity. This simply returns NULL for app_shell version.

BUG=397167
R=mukai@chromium.org
TBR=jamescook@chromium.org
TEST=no functional change. all tests must pass.

Review URL: https://codereview.chromium.org/505273002

Cr-Commit-Position: refs/heads/master@{#292038}

It was hitting a DCHECK when attempting to copy the CopyOutputRequest whenever navigating between pages.

BUG=

Review URL: https://codereview.chromium.org/511603002

Cr-Commit-Position: refs/heads/master@{#292037}

There's really no reason to have these in the same file, and it's cluttering it
up.

Bonus:
- Convert ExtensionActionStorageManager to use a BrowserContext.
- Remove ExtensionActionStorageManager's ExtensionService dependency.
- Remove ExtensionActionStorageManager's public weak ptr interface
- Consolidate extension_misc:: ExtensionAction icon stuff and ExtensionActionAPI
  icon stuff.

BUG=407290

Review URL: https://codereview.chromium.org/502033003

Cr-Commit-Position: refs/heads/master@{#292036}

After processing a batch of requests, our IDB transaction initiates an
inactivity timer; if the timer fires before any further requests
arrive from the front-end we assume the front-end is wedged and abort
the transaction. The timer is be stopped on any new request, including
a signal to commit or abort. During the introduction of blobs, the
"stop timer on commit signal" ended towards the end of the commit
flow, but it should be towards the start so it doesn't fire if the
back-end is busy.

Speculative fix for frequent timeouts of W3C web-platform-tests seen
on slow bots.

R=cmumford@chromium.org

Review URL: https://codereview.chromium.org/502643003

Cr-Commit-Position: refs/heads/master@{#292035}

The crash was caused by a dangling pointer to a deleted observer
registered by MostVisitedSites.

BUG=406996

Review URL: https://codereview.chromium.org/501973004

Cr-Commit-Position: refs/heads/master@{#292034}

BUG=394352
TEST=NONE

Review URL: https://codereview.chromium.org/433513008

Cr-Commit-Position: refs/heads/master@{#292033}

System dialog shows only properties of selected printers, no system dialog with printers.
Removed global Ctrl+Shift+P shortcut on windows.

BUG=374321

Review URL: https://codereview.chromium.org/480303002

Cr-Commit-Position: refs/heads/master@{#292032}

Take over of https://codereview.chromium.org/232003002/, which
instead utilizes the seemingly new cache timeout logic.

All partial requests which hit net::ERR_IO_PENDING will now be
immediately timed out.

BUG=31014
TEST=layout tests, net_unittests, new unittest, manual.

Review URL: https://codereview.chromium.org/478763004

Cr-Commit-Position: refs/heads/master@{#292031}

One special request for mygadgetguardian.lookout.com and an additional
Google domain.

Cr-Commit-Position: refs/heads/master@{#292030}

This allows us to set the icon title (used for hovertext) independently
of the window title (which is empty for Settings windows).

BUG=406954

Review URL: https://codereview.chromium.org/507003002

Cr-Commit-Position: refs/heads/master@{#292029}

Add cmdline switch::kEnableWebRtcHWH264Encoding to control whether
to report WebRtc HW H264 encoding capability.

BUG=385941
R=xiyuan@chromium.org,kalman@chromium.org,posicak@chromium.org,jamesr@chromium.org
TBR=piman@chromium.org

Review URL: https://codereview.chromium.org/453063002

Cr-Commit-Position: refs/heads/master@{#292028}

BUG=407662
NOTRY=true

Review URL: https://codereview.chromium.org/504223003

Cr-Commit-Position: refs/heads/master@{#292027}

BUG=371463
TBR=blundell

Review URL: https://codereview.chromium.org/493953002

Cr-Commit-Position: refs/heads/master@{#292026}

Causing Android Build (dbg) findbugs to fail.

This reverts commit c4a3f88c.

BUG=none
TBR=hush

Review URL: https://codereview.chromium.org/514443002

Cr-Commit-Position: refs/heads/master@{#292025}

This is analogous to ipc_perftests. It measures roundtrip
time with various message sizes.

TEST=message_pipe_perftest.cc
R=viettrungluu@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/501983003

Cr-Commit-Position: refs/heads/master@{#292024}

This allows better integration of OS X and iOS media frameworks
with chromium code and makes buffer management less problematic,
especially when using CVPixelBufferPools provided by hardware
encoders or decoders.

BUG=401308
R=dalecurtis,hubbe

Review URL: https://codereview.chromium.org/446163003

Cr-Commit-Position: refs/heads/master@{#292023}

BUG=374321

Review URL: https://codereview.chromium.org/506273004

Cr-Commit-Position: refs/heads/master@{#292022}

TEST=manually run through release workflow and verify api endpoints return success.
BUG=none

Review URL: https://codereview.chromium.org/417703002

Cr-Commit-Position: refs/heads/master@{#292021}

until we fix the problem. rjshade is looking into the flakiness in the
internal source tree.

R=rch@chromium.org, rjshade@chromium.org
BUG=407756

Review URL: https://codereview.chromium.org/488233004

Cr-Commit-Position: refs/heads/master@{#292020}

An older implementation of the safe-browsing database code injected
prefixes when receiving full hashes, due to an incorrect reading of how
full-hash matches should happen.  Current code checks the full hashes
and prefixes independently, so the duplicate prefixes are no longer
generated.  This code removed duplicate prefixes found in existing
databases.  99.99% of runs over the past two weeks have not removed any
such duplicate prefixes.

In the current code hitting a full hash has the same end result as
hitting a prefix, so the primary downside of removing this code is
slight excess storage.  Insofar as full hashes are deleted by sub, there
may be stale prefixes left behind.  The stale prefixes will be deleted
when the chunk in question is deleted.  The result may be a small number
of unexpected gethash requests which will return misses (and thus not
affect the browsing experience other than adding a little latency).

BUG=361248

Review URL: https://codereview.chromium.org/507653003

Cr-Commit-Position: refs/heads/master@{#292019}

This patch was generated by running the rewrite_scoped_refptr clang tool
on a Linux build.

BUG=110610

Review URL: https://codereview.chromium.org/503253003

Cr-Commit-Position: refs/heads/master@{#292018}

Add a switch to enable it. This feature would fix the problem when a large notification moves downwards, it is partially obstructed.

asvitkine: Please review histograms.xml

BUG=

Review URL: https://codereview.chromium.org/493303002

Cr-Commit-Position: refs/heads/master@{#292017}

We're seeing the dbg builder have pending jobs so this should help reduce that.

BUG=109173
R=sky@chromium.org

Review URL: https://codereview.chromium.org/506273005

Cr-Commit-Position: refs/heads/master@{#292016}

Clear internal state information about a pending suspend attempt when
the power manager restarts to prevent chrome from sending bogus messages
to the new power manager.

BUG=none
Signed-off-by: Chirantan Ekbote <chirantan@chromium.org>

Review URL: https://codereview.chromium.org/507803002

Cr-Commit-Position: refs/heads/master@{#292015}

Instead of a hardcoded port that could be taken.

BUG=

Review URL: https://codereview.chromium.org/502503003

Cr-Commit-Position: refs/heads/master@{#292014}

This patch was generated by running the rewrite_scoped_refptr clang tool
on a Linux build.

BUG=110610

Review URL: https://codereview.chromium.org/506083002

Cr-Commit-Position: refs/heads/master@{#292013}

This patch was generated by running the rewrite_scoped_refptr clang tool
on a Linux build.

BUG=110610

Review URL: https://codereview.chromium.org/511473002

Cr-Commit-Position: refs/heads/master@{#292012}

This patch was generated by running the rewrite_scoped_refptr clang tool
on a Linux build.

BUG=110610

Review URL: https://codereview.chromium.org/506093002

Cr-Commit-Position: refs/heads/master@{#292011}

Checking for existence of TopSites.

BUG=407527

Review URL: https://codereview.chromium.org/507653002

Cr-Commit-Position: refs/heads/master@{#292010}

It causes problems when running multiple Telemetries in parallel in
certain environments and also causes problems with certain devices.

BUG=

Review URL: https://codereview.chromium.org/499073005

Cr-Commit-Position: refs/heads/master@{#292009}

Allow the browser to Send messages to an embedded worker once the script as been loaded without having to wait for the initial eval to complete.

BUG=405701

Review URL: https://codereview.chromium.org/489253002

Cr-Commit-Position: refs/heads/master@{#292008}