GitLab

deploy the clang plugins to the waterfall/trybots.

BUG=none
TEST=compiles

Review URL: http://codereview.chromium.org/6366019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72846 0039d316-1c4b-4281-b951-d872f2087c98

Problems:
   - the SETTINGS frame was the first packet we'd send, uncorked.
     This was wasting the first packet of our cwnd.
   - uncork was overaggressive.  We now only uncork when there are
     no more packets in our queue.
   - rework the packet sizing to fully fill packets better.

Question - should I remove the MSG_MORE code now?  It seems to be an
alternative way to do corking, but it just doesn't work with the SSL layer.

BUG=none
TEST=n/a

Review URL: http://codereview.chromium.org/6327022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72844 0039d316-1c4b-4281-b951-d872f2087c98

This fixes a regression. Originally we listened to click events for middle click. The WebKit was change to not fire click events for middle clicks so I added code to work around that. The WebKit change caused a lot of regressions and was reverted. Now the BMM opened 2 background tabs (one for click and one for the work around). With this change I no longer listen to click and only do the work around.

BUG=70929
TEST=Open bookmark manager.
     Middle click on a bookmark.
     Only one background tab should be opened.

Review URL: http://codereview.chromium.org/6332020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72843 0039d316-1c4b-4281-b951-d872f2087c98

that I introduced where if you run "about:kill" or kill a page first, and then run "about:crash" in the same window,
then it shows the killed page instead of the crashed page, and vice-versa.

BUG=none
TEST=tried "about:kill", then "about:crash", and then "about:kill" again in the same session, and vice versa.

Review URL: http://codereview.chromium.org/6347015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72842 0039d316-1c4b-4281-b951-d872f2087c98

Fix printing PDFs through right click on Mac and Linux.  Looks like they both give the menu closed notification before the notification about which item was clicked.

BUG=70987
Review URL: http://codereview.chromium.org/6336014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72840 0039d316-1c4b-4281-b951-d872f2087c98

Refactor WebGraphicsContext3D to use WGC3D types which match corresponding GL types. (Land a second time to fix a bug caused by auto-merging.)

BUG=none TEST=compiled fine
TBR=kbr@chromium.org

Review URL: http://codereview.chromium.org/6349017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72825 0039d316-1c4b-4281-b951-d872f2087c98

BUG=69155
TEST=Verify font names list does not contain duplicates in WinChrome.
Review URL: http://codereview.chromium.org/6277022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72822 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6403004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72818 0039d316-1c4b-4281-b951-d872f2087c98

TBR=yuzo,yurys
Review URL: http://codereview.chromium.org/6356014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72813 0039d316-1c4b-4281-b951-d872f2087c98

- Use 'raw-button' for X button so that it fix the X image;
- Add a wrapper block to email and name text that fills up the space
  between icon and X button and has a max-width to avoid push X button
  out of view;

BUG=chromium-os:11220,chromium-os:10944
TEST=Verify fix for chromium-os:11220

Review URL: http://codereview.chromium.org/6380015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72810 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/6341012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72809 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=compiled fine

Review URL: http://codereview.chromium.org/6350016

TBR=zmo@google.com
Review URL: http://codereview.chromium.org/6328012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72808 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=compiled fine

Review URL: http://codereview.chromium.org/6350016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72807 0039d316-1c4b-4281-b951-d872f2087c98

BUG=69224
TEST=NewTabUIProcessPerTabTest.NavBeforeNTPCommits

Review URL: http://codereview.chromium.org/6335014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72806 0039d316-1c4b-4281-b951-d872f2087c98

Adds a heapcheck suppression for SyncUIUtilTest.ConstructAboutInformationWithUnrecoverableErrorTest

BUG=71070
TEST=memory bots go green
TBR=akalin@chromium.org

Review URL: http://codereview.chromium.org/6364015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72805 0039d316-1c4b-4281-b951-d872f2087c98

Made code changes to create a single metafile for the entire printing document.

Implemented the following function:
1. CreatePreviewDocument(): Create a single metafile and render all the pages(entire printing document) into the same metafile.

Note: Although the structure of PrintPage() function looks similar to CreatePreviewDocument(), merging of these functions is difficult for the following reasons:
  1. As of now, we are using EMF metafile to create a preview document. In future we are going to use Skia->PDF metafile. So construction and destruction of the metafile will be different.
  2. The message for preview workflow(ViewHostMsg_PagesReadyForPreview) is different from normal printing workflow(ViewHostMsg_DidPrintPage).
  3. The param struct which is sent along with the message for preview workflow is different from printing workflow.
  4. Finally, the semantics of PrintPage() function is to print a page per metafile which is completely different from the semantics of CreatePreviewDocument() function.

Once Skia->Pdf work is completed, Modify CreatePreviewDocument() function to use this PDF metafile.

BUG=64121
TEST=printing works after code changes.

Review URL: http://codereview.chromium.org/6261025

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72804 0039d316-1c4b-4281-b951-d872f2087c98

[SYNC] Tests to catch datatype manager behavior when datatype controller is busy when stop() is called.

BUG=61804
TEST=self

Review URL: http://codereview.chromium.org/6357018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72803 0039d316-1c4b-4281-b951-d872f2087c98

BUG=70928
TEST=./tools/valgrind/waterfall.sh fetch, then match [on both Linux and Mac]
TBR=thakis,glider
Review URL: http://codereview.chromium.org/6376012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72802 0039d316-1c4b-4281-b951-d872f2087c98

This adds an histogram to the HTTP response header parser.  I reviewed
the code, and I've convinced myself that this captures the two main
cases of HTTP header parsing (tunneling, and normal HTTP), and that
double-calls aren't occuring.  I liked having all the code in the one
translation unit.  But, I might have missed one, and we might still
be getting called twice for the same load: so the statistics might be off.

Alternatively, I could put a static function on HttpResponseHeader, to
be called from the site of parse.  But this might miss a parse.

All in all, I wanted to be sure to catch all distinct response codes
we get, so we can be sure not to stomp on anything.  So I chose the
approach using the HttpResponseHeaders constructor.

TEST=none
BUG=70428

Review URL: http://codereview.chromium.org/6317011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72801 0039d316-1c4b-4281-b951-d872f2087c98

BUG=70401
TEST=nope

Review URL: http://codereview.chromium.org/6340013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72800 0039d316-1c4b-4281-b951-d872f2087c98

The communications path for creating a viewable command buffer used to be directly from the renderer (gpu_channel.cc) to the gpu process (gpu_channel.cc).  This patch makes the browser an intermediary:

- renderer (gpu_channel.cc) makes a synchronous request to the browser (picked up in renderer_message_filter.cc and forwarded to gpu_process_host.cc)

- browser (gpu_process_host.cc) makes an asynchronous request to the gpu process (picked up in gpu_thread.cc and forwarded to gpu_channel.cc) for the command buffer

- gpu process (gpu_thread.cc) sends an ACK with the route_id for the command buffer back to the browser (gpu_process_host.cc)

- browser (gpu_process_host.cc) sends a delayed reply back to the renderer
(gpu_channel_host.cc), which had blocked

There are several motivations for this patch:

- creating an onscreen command buffer requires a window to draw into (which is acquired/locked in the browser); by routing through the browser, we can acquire the get the window beforehand (thereby preventing a deadlock in some other work that I'm doing)

- we can eliminate several separate synchronous IPC messages for obtaining and releasing the window associated with drawing (I've tried to unify the different code paths for Linux, Windows, and Mac)

- in the future, we can have the browser allocate SHM for the command buffer and transfer buffers, allowing us to sandbox the gpu process

BUG=none
TEST=by hand on all 3 platforms, trybots

Review URL: http://codereview.chromium.org/6343006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72798 0039d316-1c4b-4281-b951-d872f2087c98

Avoid overlapping class definitions (e.g. MessageLoopQuitListener).

BUG=None
TEST=unit_tests,browser_tests

Review URL: http://codereview.chromium.org/6341013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72793 0039d316-1c4b-4281-b951-d872f2087c98

ADMX policies will have the XML attribute class set to "both", ADM policies will be presented twice, once under CLASS MACHINE, and once under CLASS USER.

BUG=70232
TEST=Add chrome.adm to Group Policy Editor and see if Chrome policies appear both under User Configuration and Machine Configuration, and they can be configured independently. Repeat this with chrome.admx.

Review URL: http://codereview.chromium.org/6327011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72792 0039d316-1c4b-4281-b951-d872f2087c98

2nd attempt with shutdown crash fix (only difference is in browser.cc).

BUG=66413
TEST=None

Review URL: http://codereview.chromium.org/6279012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72791 0039d316-1c4b-4281-b951-d872f2087c98

This CL allows somebody to ask the PrefService for the effective preference values.

BUG=70809
TEST=prefs.PrefsTest.testUnderTheHoodPref and prefs.PrefsTest.testHomepagePrefs of chrome/test/functional/prefs.py will work again.

Review URL: http://codereview.chromium.org/6353015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72790 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=compiles

Review URL: http://codereview.chromium.org/6348007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72789 0039d316-1c4b-4281-b951-d872f2087c98

BUG=70761
TEST=none

Review URL: https://chromereviews.googleplex.com/2402014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72788 0039d316-1c4b-4281-b951-d872f2087c98

Reason for the refactoring is the need to properly handle disappearing policies.
This situation arises when a managed preference is removed that only has
been registered in one pref store (for example local_store) the notification
is sent to all pref stores though but the profile pref store doesn't have
a default value for that pref and therefore ControllingPrefStoreForPref returns
INVALID_STORE which shouldn't be treated as an exceptional situation (DCHECKed).

BUG=71028
TEST=unit_tests --gtest_filter=*Pref* should give pretty good coverage.

Review URL: http://codereview.chromium.org/6357019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72787 0039d316-1c4b-4281-b951-d872f2087c98

Clear prefs that the server reports TYPE_NULL values for and log a warning if the type is off otherwise.

BUG=none
TEST=compiles and passes tests

Review URL: http://codereview.chromium.org/6288016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72786 0039d316-1c4b-4281-b951-d872f2087c98

This is the first step to being able to re-constitute policy from a non-generic protobuf in a policy request in a fully automated way (i.e. the code for the process can be generated). The non-generic protobuf is required for automatic merging of policy on the server-side.

BUG=68309
TEST=existing policy/provider tests

Review URL: http://codereview.chromium.org/6002015

TBR=danno@chromium.org

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72785 0039d316-1c4b-4281-b951-d872f2087c98

This is the first step to being able to re-constitute policy from a non-generic protobuf in a policy request in a fully automated way (i.e. the code for the process can be generated). The non-generic protobuf is required for automatic merging of policy on the server-side.

BUG=68309
TEST=existing policy/provider tests

Review URL: http://codereview.chromium.org/6002015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72784 0039d316-1c4b-4281-b951-d872f2087c98

(there's an exclusion for _nss on non-linux platforms, however we need
this file for mac and windows as well.)
this is a build fix for http://src.chromium.org/viewvc/chrome?view=rev&revision=72782)

BUG=none
TEST=compiles

Review URL: http://codereview.chromium.org/6374015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72783 0039d316-1c4b-4281-b951-d872f2087c98

Stubs out SslServerSocket for OpenSSL.

BUG=none
TEST=Sha256Test.TestContext (and compiles with openssl flag).

Review URL: http://codereview.chromium.org/6276002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72782 0039d316-1c4b-4281-b951-d872f2087c98

The WebKit side code will change to use onSuccess(SerializedScriptValue::nullValue()) instead.

BUG=70221
TEST=no new functionality

Review URL: http://codereview.chromium.org/6265027

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72781 0039d316-1c4b-4281-b951-d872f2087c98

BUG=chromium-os:11087
TEST=manual

Review URL: http://codereview.chromium.org/6287014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72780 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72776 0039d316-1c4b-4281-b951-d872f2087c98

This change should have been included in
http://codereview.chromium.org/5981001/.

BUG=chromium-os:9682
TEST=make chrome passes

Review URL: http://codereview.chromium.org/6279015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72771 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=native_client/tests with --enable-nacl

Review URL: http://codereview.chromium.org/6306015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72768 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/6320018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72767 0039d316-1c4b-4281-b951-d872f2087c98

- Been targeted by mass malware.
- Do not yet have a good sandboxing story.

BUG=60458
TEST=http://java.sun.com/products/plugin/1.4/demos/applets/Blink/example1.html with default plug-in settings.

Review URL: http://codereview.chromium.org/6350010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@72766 0039d316-1c4b-4281-b951-d872f2087c98