GitLab

- Fixes race on state_ in CompositeFilter detected by TSAN
- Moves us one step closer to removing FilterHost::SetError().

BUG=109875
TEST=AudioRendererBaseTest.* VideoRendererBaseTest.*, FFmpegAudioDecoderTest.*, CompositeFilterTest.*


Review URL: http://codereview.chromium.org/9310028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120135 0039d316-1c4b-4281-b951-d872f2087c98

This is to make it clear that a file stream is closed there,
hence file access is performed. Add ScopedIOAllowIO as well.

BUG=72001
TEST=net_unittests


Review URL: http://codereview.chromium.org/9316048

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120134 0039d316-1c4b-4281-b951-d872f2087c98

Field trial to investigate higher weightings for hits when calculating Omnibox prerender confidence.

BUG=111069


Review URL: http://codereview.chromium.org/9298032

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120133 0039d316-1c4b-4281-b951-d872f2087c98

dump-static-initializers.py: Add '-d' flag for listing SI's in an easily "diff-able" way, for use in the 'sizes' script.

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/9169074

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120131 0039d316-1c4b-4281-b951-d872f2087c98

In accelerated compositing mode we can't start rendering until the
native window (HWND/gtk widget) are created on Windows and GTK without
aura. For most render_widgets the code today is racy, but for
window.open() and similar creation paths this is a race that we pretty
much always lose. This defers the first DoDeferredUpdate on a render_widget
until its host_window_ is available.

Also defers querying for GL_CHROMIUM_supports_swapbuffers_callback
extension until the first draw, since the context isn't ready until the
host window is available.

BUG=106815
TEST=start with --force-compositing-mode and --show-fps-counter, load up
a page that uses window.open() to open a popup, and confirm that the
popup is in compositing mode


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=119977

Review URL: http://codereview.chromium.org/9225050

TBR=jamesr@chromium.org
Review URL: https://chromiumcodereview.appspot.com/9307043

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120130 0039d316-1c4b-4281-b951-d872f2087c98

TEST=NONE
BUG=100893
Review URL: https://chromiumcodereview.appspot.com/9310041

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120126 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/9121064

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120125 0039d316-1c4b-4281-b951-d872f2087c98

R=asargent@chromium.org
BUG=105179


Review URL: http://codereview.chromium.org/9314046

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120124 0039d316-1c4b-4281-b951-d872f2087c98

This crash occurs when an unpacked extension changes IDs by adding a key to its manifest. The extension preferences are stored under the old key. This started causing problems because now we record granted permissions for unpacked extensions.

BUG=112301
TEST=ExtensionServiceTest.UnpackedExtensionCanChangeID

Review URL: https://chromiumcodereview.appspot.com/9320021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120123 0039d316-1c4b-4281-b951-d872f2087c98

BUG=112314
TEST=ExtensionApiTest.Tabs2

Review URL: https://chromiumcodereview.appspot.com/9320016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120121 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/9320020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120120 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/9293036

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120118 0039d316-1c4b-4281-b951-d872f2087c98

This CL added two new examples for the NaCl SDK.  This only affects SDK bots since
the code is completely contained within native_client_sdk subtree.

Fix incorrectly placed ppapi headers.
Remove TODOs and cleanup create_nmf.py.
Add hello_world_glibc example
Add dlopen example.

BUG= 111224
Review URL: https://chromiumcodereview.appspot.com/9234043

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120117 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/9316045

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120116 0039d316-1c4b-4281-b951-d872f2087c98

There is no corresponding "global" setting, only a command-line flag.

BUG=105537

TBR=pkasting

Review URL: https://chromiumcodereview.appspot.com/9264071

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120115 0039d316-1c4b-4281-b951-d872f2087c98

[net/dns] Refactoring of job dispatch in HostResolverImpl in preparation for DnsTransactionFactory. 

- Replaces JobPool with PrioritizedDispatcher.
- Separates Job which aggregates Requests from ProcTask which executes on 
WorkerPool. 
- Queues Jobs rather than Requests before dispatching them. 
(So that we can queue before DnsTransactionFactory and then before ProcTask.) 

This is the second CL in series to merge AsyncHostResolver into 
HostResolverImpl.

BUG=107880
TEST=waterfall

Review URL: http://codereview.chromium.org/9101011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120114 0039d316-1c4b-4281-b951-d872f2087c98

For some reason, the error reporting page doesn't work over HTTPS yet.
That's a separate bug, and when it is fixed we'll fix the last http: URL in
this file.

BUG=112236
TEST=browser_tests: TranslateManagerTest.*; translation still works, and reporting a translation error still works

Review URL: http://codereview.chromium.org/9320015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120113 0039d316-1c4b-4281-b951-d872f2087c98

Change ChildProcess::set_main_thread registration to occur inside the RenderThreadImpl::Init method.

Background: The Init method is currently called from the RenderThreadImpl constructor. If set_main_thread is not called from Init then ContentRendererClient::RenderThreadStarted, which is also called from Init, will not be able to retrieve the current child thread using ChildThread::current. Retrieving the current child thread from inside RenderThreadStarted is necessary for performing actions that must occur before message processing starts such as adding additional MessageFilters to the child thread's channel.

BUG=112335
TEST=ChildThread::current() returns non-NULL in ContentRendererClient::RenderThreadStarted

Review URL: http://codereview.chromium.org/9297022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120112 0039d316-1c4b-4281-b951-d872f2087c98

In accelerated compositing mode we can't start rendering until the
native window (HWND/gtk widget) are created on Windows and GTK without
aura. For most render_widgets the code today is racy, but for
window.open() and similar creation paths this is a race that we pretty
much always lose. This defers the first DoDeferredUpdate on a render_widget
until its host_window_ is available.

Also defers querying for GL_CHROMIUM_supports_swapbuffers_callback
extension until the first draw, since the context isn't ready until the
host window is available.

BUG=106815
TEST=start with --force-compositing-mode and --show-fps-counter, load up
a page that uses window.open() to open a popup, and confirm that the
popup is in compositing mode


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=119977

Review URL: http://codereview.chromium.org/9225050

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120111 0039d316-1c4b-4281-b951-d872f2087c98

When running as runhooks, it's not run via cygwin like it was when run in the build, and isn't able to find git. I believe the test that checked for both cygwin and win32 before was incorrect (but was always run via cygwin before so didn't matter).

Additionally, when runhooks runs it, it's from above src/, so the cwd isn't in the repo. So, pass in the directory of the script itself so that git can find its repo.

BUG=112264

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=120089

Review URL: https://chromiumcodereview.appspot.com/9317037

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120110 0039d316-1c4b-4281-b951-d872f2087c98

BUG=100898
R=pkasting@chromium.org
TBR=sky@chromium.org

Review URL: https://chromiumcodereview.appspot.com/9021046

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120109 0039d316-1c4b-4281-b951-d872f2087c98

enterprise.EnterpriseTestReverse.testEnableSPDY - Linux
autofill.AutofillTest.testDisplayLineItemForEntriesWithNoCCNum - Linux
extensions.ExtensionsTest.testAllowAccessFileURLs - Win

TEST=NONE
BUG=100893, 100365,  111289
Review URL: https://chromiumcodereview.appspot.com/9310030

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120108 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/9307019

TBR=thestig@chromium.org
Review URL: https://chromiumcodereview.appspot.com/9310035

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120107 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/9307019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120106 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120105 0039d316-1c4b-4281-b951-d872f2087c98

This pulls in the following Native Client changes:

r7665: (elijahtaylor) Fix dlopen hang on main thread.
r7666: (bsy) fixed range checking error that prevented objects to be at the end of addr sp
r7667: (jvoung) Pass along -pthread option in pnacl driver to define _REENTRANT, link in libpthread.
r7668: (ncbray) Change the x86-64 decoder to store jump offsets rather than target addresses.
r7669: (kschimpf) Clean up code handling indirect jumps in the x86-64 validator.
r7670: (dschuff) Use pexe+translate steps rather than bc+clang to compile torture tests
r7671: (bsy) syscall and irt hooks to expose the clock_getres and clock_gettime functionality
r7672: (ncbray) Strengthen another x86-64 validator test.
r7673: (bsy) update copyright headers missed in r7671
r7674: (kschimpf) Repair damage to generated pre/post conditions for testing of the x86-64
r7675: (bradnelson) Adding exception_tests test suite.
r7676: (ncbray) Change the x86-64 validator to work in offsets rather than addresses.
r7677: (ncbray) Force expression nodes in the x86-64 validator to have specified sizes.
r7678: (dschuff) Make JIT build work again
r7679: (pdox) run.py:
r7680: (mcgrathr) Update toolchain rev
r7682: (jvoung) Add support for generating nmfs for pnacl pexes with glibc (vs pnacl nexes).
r7683: (elijahtaylor) Add irt_core.nexe to SDK install_bin target
r7684: (jvoung) update llvm rev for data-layout revert: http://codereview.chromium.org/9290056/
r7685: (jvoung) Address some comments from old CLs:
r7686: (mseaborn) Exception handling: Test multiple exception stack alignments
r7687: (pdox) PNaCl: Bitcode linker modification. Handle symbols from .pso files
r7688: (ncbray) Simplify the x86-64 validator's internal representation of constants.
r7689: (ncbray) Change DEBUG printing in the validator to print offsets instead of addresses.
r7690: (dschuff) new testsuite passes/failures
r7691: (bsy) libnacl changes to enable clock_get{res,time}.  moves timespec defn.
r7692: (pdox) PNaCl: Fix typo in driver (env.append instead of env.set).
r7693: (mseaborn) download_toolchains: Get tarball hashes by downloading '.sha1hash' files
r7694: (kschimpf) Add ULDI-disable flag for projects that use .asm files
r7695: (pdox) PNaCl driver fixes:
r7696: (pdox) Augment libnosys w/ raise, execle, execvp, execl, execv, execlp.
r7697: (ncbray) Prevent nacl_dyncode_create from marking memory as used when validation fails.
r7698: (pdox) Add -ansi to pnacl driver.
r7699: (bradchen) Enable platform_qual_test on Win64. Add blacklist unit tests.
r7700: (robertm) Update qemu to v1.0.
r7701: (sehr) A slightly more visual liveness indication during translation.
r7702: (dschuff) Run LLVM toolchain tests with qemu 1.0
r7703: (pdox) Roll Chrome DEPS in NaCl
r7705: (bsy) revert 7703, which broke mac10.6-newlib-dbg-clang gyp due to clang warnings
r7707: (bsy) enable direct syscalls for clock_get{res,time}. used for irt=0 (tests).
r7708: (mseaborn) Windows: Wait for SuspendThread() to take effect using GetThreadContext()
r7709: (dschuff) Test sel_universal streaming interface by sending bitcode data over SRPC
r7710: (mseaborn) Exception handling: Add tests to check that invalid handlers are rejected
r7711: (mseaborn) Exception handling: Test the exception handler's %esp exactly
r7712: (ncbray) Ensure super instructions are marked during dynamic code modification.
r7713: (noelallen) Add NACL_SECURITY_DISABLE environment variable to allow PPAPI developers
r7714: (bradnelson) Add OSX hardware exception handling to sel_ldr.
r7715: (dschuff) Revert "Run LLVM toolchain tests with qemu 1.0"
r7716: (pdox) Roll toolchain DEPS to 7712.
r7718: (pdox) Revert PNaCl DEPS roll (partial revert of r7716)
r7720: (mcgrathr) Make IRT load failure fatal under Chrome

BUG=none
TEST=nacl_integration

Review URL: https://chromiumcodereview.appspot.com/9309040

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120104 0039d316-1c4b-4281-b951-d872f2087c98

BUG=110402
TEST=Launching the default YouTube and Search apps should not show the "<app name> is unreacheable" page when Chrome is offline
R=miket@chromium.org


Review URL: http://codereview.chromium.org/9316019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120103 0039d316-1c4b-4281-b951-d872f2087c98

This is just to get my feet wet before getting more serious about adding
pnacl support.
Mostly adding comments and a tiny bit of cleanup

Two questions:
* where does this get  invoked from?
  I saw some mentioning here:
  tools/build/scripts/master/factory/gyp_factory.py
  bot this is not part of my chrome client

* Where do the toolchain tarballs get downloaded?

Review URL: http://codereview.chromium.org/9159067

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120100 0039d316-1c4b-4281-b951-d872f2087c98

BUG=112109
TEST=Crash the browser and verify that screenlocker observers have been registered (i.e. short power press works).

Review URL: https://chromiumcodereview.appspot.com/9310002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120099 0039d316-1c4b-4281-b951-d872f2087c98

PluginLoaderPosix: Fix the case where the utility process crashes after all plugins have been loaded.

This was causing a crash on load on my system.

BUG=111935
TEST=none
R=rsesek

Review URL: https://chromiumcodereview.appspot.com/9297049

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120098 0039d316-1c4b-4281-b951-d872f2087c98

This causes all animations scheduled during a test to complete immediately.

After making this change, I noticed some code assumed that animations would not complete synchronously. Some of this code used animation observers, and I while fixing the code I have updated it to use the preferred ImplicitAnimationObserver.

BUG=None
TEST=aura_shell_unittests,aura_unittests,compositor_unittests

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=120074

Review URL: https://chromiumcodereview.appspot.com/9222018

TBR=vollick@google.com
Review URL: https://chromiumcodereview.appspot.com/9320018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120097 0039d316-1c4b-4281-b951-d872f2087c98

Removing unnecessary includes of the content\browser\child_process_security_policy.h file in preparation for
creating a public content API around the ChildProcessSecurityPolicy class.

BUG=98716
Review URL: https://chromiumcodereview.appspot.com/9307023

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120096 0039d316-1c4b-4281-b951-d872f2087c98

BUG=109959
TEST=none
(covered by existing tests and tested locally with built-in audio,
 Metric Halo 2882, and Stanton FinalScratch Firewire devices to
 verify that crashes are fixed)

Incidentally, the stream latency was always reported as zero on the devices I tested,
so removing this code should have no practical impact.
Review URL: https://chromiumcodereview.appspot.com/9310004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120095 0039d316-1c4b-4281-b951-d872f2087c98

Change tool-bar behaviour to click-to-show. This is based on user feedback the auto-show behaviour makes it hard to access the top of the host screen. By auto-expanding only the stub, the inaccessible part of the host screen is reduced.

BUG=110211
TEST=Manual

Review URL: https://chromiumcodereview.appspot.com/9129009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120093 0039d316-1c4b-4281-b951-d872f2087c98

This causes all animations scheduled during a test to complete immediately.

After making this change, I noticed some code assumed that animations would not complete synchronously. Some of this code used animation observers, and I while fixing the code I have updated it to use the preferred ImplicitAnimationObserver.

BUG=None
TEST=aura_shell_unittests,aura_unittests,compositor_unittests

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=120074

Review URL: https://chromiumcodereview.appspot.com/9222018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120092 0039d316-1c4b-4281-b951-d872f2087c98

Flaky test is flaky.

TBR=thakis

BUG=86494
Review URL: http://codereview.chromium.org/9310032

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120091 0039d316-1c4b-4281-b951-d872f2087c98

This will be very useful for those of us working
on improving the scoring/ranking of autocomplete
suggestions.

The presentation method could probably be improved, but I'd prefer
to check this in as-is, and see how us omnibox developers use it in
practice.  Then we'll have a better idea of how chrome://omnibox/
should convey the information we most need.

BUG=111791
TEST=


Review URL: http://codereview.chromium.org/9289028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120090 0039d316-1c4b-4281-b951-d872f2087c98

When running as runhooks, it's not run via cygwin like it was when run in the build, and isn't able to find git. I believe the test that checked for both cygwin and win32 before was incorrect (but was always run via cygwin before so didn't matter).

Additionally, when runhooks runs it, it's from above src/, so the cwd isn't in the repo. So, pass in the directory of the script itself so that git can find its repo.

BUG=112264

Review URL: https://chromiumcodereview.appspot.com/9317037

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120089 0039d316-1c4b-4281-b951-d872f2087c98

BUG=112328

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120088 0039d316-1c4b-4281-b951-d872f2087c98

BUG=112006


Review URL: http://codereview.chromium.org/9301043

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120087 0039d316-1c4b-4281-b951-d872f2087c98