summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Revert 43086 - 1. Create a new sandbox type which allows access to Unix ↵dmaclach@chromium.org2010-03-308-97/+12
| | | | | | | | | | | | | | | | | | sockets in the Mac renderer sandbox to support running Native Client. 2. Put the Native Client sel_ldr (which contains the user's untrusted code into a new Mac sandbox type. 3. Open /dev/random in SandboxWarmup(). 4. Remove the "nosandbox" flag when running Mac tests. BUG=http://code.google.com/p/nativeclient/issues/detail?id=327 TEST=nacl_ui_tests still pass while running in the sandbox. Review URL: http://codereview.chromium.org/1234003 TBR=msneck@google.com git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43093 0039d316-1c4b-4281-b951-d872f2087c98
* Turn on RTTI for the Mac dump_syms. Breakpad r550 requires it, and aftermark@chromium.org2010-03-301-0/+2
| | | | | | | | | | | considering the alternatives, I've decided that it's not terrible. This code does not make it into the Chrome build, it's only used for the symbol dumper. BUG=none TEST=Mac official build Review URL: http://codereview.chromium.org/1535005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43092 0039d316-1c4b-4281-b951-d872f2087c98
* InfoBarController needs to be an NSTextViewDelegate after r42985.mark@chromium.org2010-03-301-1/+2
| | | | | | | | BUG=none TEST=10.6 SDK build Review URL: http://codereview.chromium.org/1552004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43091 0039d316-1c4b-4281-b951-d872f2087c98
* Ignore more gyp/make spew, and add a couple missing svn directories.mmoss@google.com2010-03-301-0/+3
| | | | | | Review URL: http://codereview.chromium.org/1587004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43090 0039d316-1c4b-4281-b951-d872f2087c98
* Add UMA counters for non-frame cross site requests.japhet@chromium.org2010-03-301-0/+73
| | | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/668085 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43089 0039d316-1c4b-4281-b951-d872f2087c98
* Change interface for site-isolation logging from WebKit in advance of ↵johnnyg@chromium.org2010-03-302-3/+4
| | | | | | | | | | | landing WebKit side. BUG=none TEST=none Review URL: http://codereview.chromium.org/1516004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43088 0039d316-1c4b-4281-b951-d872f2087c98
* SPDY: Improve logging messages for SpdySession.willchan@chromium.org2010-03-301-3/+9
| | | | | | Review URL: http://codereview.chromium.org/1582001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43087 0039d316-1c4b-4281-b951-d872f2087c98
* 1. Create a new sandbox type which allows access to Unix sockets in the Macmsneck@google.com2010-03-308-12/+97
| | | | | | | | | | | | | | renderer sandbox to support running Native Client. 2. Put the Native Client sel_ldr (which contains the user's untrusted code into a new Mac sandbox type. 3. Open /dev/random in SandboxWarmup(). 4. Remove the "--nosandbox" flag when running Mac tests. BUG=http://code.google.com/p/nativeclient/issues/detail?id=327 TEST=nacl_ui_tests still pass while running in the sandbox. Review URL: http://codereview.chromium.org/1234003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43086 0039d316-1c4b-4281-b951-d872f2087c98
* Added logic that exits message loop upon test ending even if no browseravayvod@google.com2010-03-301-3/+8
| | | | | | | | | | | | window exists. BUG=None. TEST=Run in process browser test that doesn't create browser window: it should quit without timeout error. Review URL: http://codereview.chromium.org/1508007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43085 0039d316-1c4b-4281-b951-d872f2087c98
* [Mac] If a popup for an extension is shown and then the extension is killed ↵andybons@chromium.org2010-03-302-1/+13
| | | | | | | | | | or crashes, the popup should close for that extension. BUG=35940 TEST=See bug for repro steps. Review URL: http://codereview.chromium.org/1528006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43084 0039d316-1c4b-4281-b951-d872f2087c98
* -[AutocompleteTextFieldEditor delegate] cleanup.mark@chromium.org2010-03-302-10/+21
| | | | | | | | | | | | | | | | | | | | | | AutocompleteTextFieldEditor* requires its delegate to be an AutocompleteTextField*. Override the return type of -delegate and the parameter type of -setDelegate:. This is safe because NSTextView, AutocompleteTextFieldEditor's superclass, uses id<NSTextViewDelegate> as the delegate type (as of the 10.6 SDK, and just id prior), and AutocompleteTextField* implements NSTextViewDelegate (and of course is descended from id). This change allows some casts and scattered type checks in this file to be removed in favor of the cast and type check in one central location. It also fixes another 10.6 SDK build error caused because -delegate is more specific than just id in that SDK. It provides future-proofing against further errors of this sort. BUG=none TEST=10.6 SDK build Review URL: http://codereview.chromium.org/1550002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43083 0039d316-1c4b-4281-b951-d872f2087c98
* Bump breakpad version to version 557, which improves how invalid argument ↵siggi@chromium.org2010-03-302-1/+5
| | | | | | | | | | | CRT reports are generated to allow sorting those dumps by stack signature. BUG=none TEST=none Review URL: http://codereview.chromium.org/1418001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43081 0039d316-1c4b-4281-b951-d872f2087c98
* Add myself to chromeos/login watchlist.nkostylev@google.com2010-03-301-0/+4
| | | | | | | | | | BUG=none TEST=none TBR=dpolukhin Review URL: http://codereview.chromium.org/1588002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43080 0039d316-1c4b-4281-b951-d872f2087c98
* Update V8 version to 2.2.0.2. This fixes a bug in time zone handling.whesse@chromium.org2010-03-301-1/+1
| | | | | | Review URL: http://codereview.chromium.org/1508009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43079 0039d316-1c4b-4281-b951-d872f2087c98
* Fix CoreGraphics and Core Animation drawing for QuickTime pluginstuartmorgan@chromium.org2010-03-302-15/+11
| | | | | | | | | | | Conditionalizes a SetWindow call that was added to fix a crash in QuickDraw mode, but isn't necessary (and causes QuickTime not to draw) in other drawing modes. BUG=38336,36896,35081 TEST=Play QuickTime movies on 10.6. The ones with floating controllers (36896) should draw again, and the other should not only work, but have smoother playback and use less CPU. The test case for bug 35081 should still not crash. Review URL: http://codereview.chromium.org/1569005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43078 0039d316-1c4b-4281-b951-d872f2087c98
* Disabling the DiskCacheBackendTest.NewEvictionDisableFailure2 as it crashes ↵timurrrr@chromium.org2010-03-301-1/+2
| | | | | | | | | | on Linux/shlib BUG=38562 TBR=amit Review URL: http://codereview.chromium.org/1593001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43077 0039d316-1c4b-4281-b951-d872f2087c98
* Add win32 gtest filter files shared between Wine and TSan/Windowstimurrrr@chromium.org2010-03-305-39/+30
| | | | | | | BUG=28363 Review URL: http://codereview.chromium.org/1377003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43076 0039d316-1c4b-4281-b951-d872f2087c98
* Don't run UnloadTest.CrossSiteInfiniteUnloadSync under TSantimurrrr@chromium.org2010-03-301-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | Here's an example output of the failure: [----------] 1 test from UnloadTest [ RUN ] UnloadTest.CrossSiteInfiniteUnloadSync [5442:5442:0330/051539:6569997191168:INFO:chrome/test/ui/ui_test.cc(1199)] BROWSER_WRAPPER was set, prefixing command_line with /b/slave/chromium-rel-linux-tsan-ui/build/valgrind.tmp/browser_wrapper.KXl4L_ Extracting ThreadSanitizer to /tmp/valgrind.de5700 chrome/test/ui/ui_test.cc:595: Failure Value of: tab_proxy->NavigateToURLBlockUntilNavigationsComplete( url, number_of_navigations) Actual: 0 Expected: AUTOMATION_MSG_NAVIGATION_SUCCESS Which is: 1 http://mock.http/title2.html chrome/browser/unload_uitest.cc:127: Failure Value of: GetActiveTabTitle() Actual: L"" Expected: expected_title Which is: L"Title Of Awesomeness" [ FAILED ] UnloadTest.CrossSiteInfiniteUnloadSync (556591 ms) [----------] 1 test from UnloadTest (556591 ms total) BUG=32529 TBR=glider Review URL: http://codereview.chromium.org/1547007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43075 0039d316-1c4b-4281-b951-d872f2087c98
* Disabling the DiskCacheBackendTest.DisableFailure2 as it crashes at times on ↵timurrrr@chromium.org2010-03-301-1/+2
| | | | | | | | | | Linux TBR=amit BUG=38562 Review URL: http://codereview.chromium.org/1592002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43074 0039d316-1c4b-4281-b951-d872f2087c98
* Suppress two benign races in sqlitetimurrrr@chromium.org2010-03-301-0/+14
| | | | | | | | TEST=TSan/UI should become greener TBR=glider Review URL: http://codereview.chromium.org/1552003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43071 0039d316-1c4b-4281-b951-d872f2087c98
* [GTTF] Miscellanous UI tests cleanups:phajdan.jr@chromium.org2010-03-3016-146/+122
| | | | | | | | | | | | | | | - use built-in timeouts instead of "inventing" them in each test case - avoid unneeded checks and operations - use automation calls more effectively - use FLAKY mark instead of DISABLED to maintain test coverage - split some tests to make the above possible TEST=UI test based BUG=39785 Review URL: http://codereview.chromium.org/1547003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43070 0039d316-1c4b-4281-b951-d872f2087c98
* Updating trunk VERSION from 365.0 to 366.0chrome-release@google.com2010-03-301-1/+1
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43068 0039d316-1c4b-4281-b951-d872f2087c98
* Update V8 to version 2.2.0.1.kmillikin@chromium.org2010-03-301-1/+1
| | | | | | | | | | | This fixes a bug in String.prototype.replace. BUG=v8:667 (http://code.google.com/p/v8/issues/detail?id=667) TEST=none Review URL: http://codereview.chromium.org/1529003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43066 0039d316-1c4b-4281-b951-d872f2087c98
* Remove suppression for bug 32892 as the race tim@chromium.org2010-03-301-7/+0
| | | | | | | | | | | has been annotated. BUG=32892 Review URL: http://codereview.chromium.org/1276002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43065 0039d316-1c4b-4281-b951-d872f2087c98
* [Mac] display appcaches in cookie treedanno@chromium.org2010-03-308-67/+707
| | | | | | | | | | | | | xib changes: - Added a new NSView for showing detailed information about appcaches in the cookie tree. The new view uses the same pattern as the existing views. - Added bindings for the new view to control its visibility and the values displayed, also using the same style as the existing detail information subviews. BUG=37459 TEST=new unit test and manual tests Review URL: http://codereview.chromium.org/1555001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43064 0039d316-1c4b-4281-b951-d872f2087c98
* Switch AutomationProxyVisibleTest.AutocompleteMatchesTest from DISABLEDphajdan.jr@chromium.org2010-03-301-2/+8
| | | | | | | | | | | | | | | | to FLAKY on Windows and Linux (required infrastructure is still not available on Mac). There is a slight possiblity that the flakiness has disappeared, but I don't think so. We're going to get more accurate data from the flakiness report. TEST=AutomationProxyVisibleTest.AutocompleteMatchesTest BUG=19876 Review URL: http://codereview.chromium.org/1530004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43063 0039d316-1c4b-4281-b951-d872f2087c98
* [GTTF] Unify polling interval in some of the polling functionsphajdan.jr@chromium.org2010-03-3018-73/+49
| | | | | | | | | | | | | | | | | | | | | | in the UI test framework: - WaitUntilCookieValue - WaitUntilCookieNonEmpty - WaitUntilJavaScriptCondition This helps reduce number of arbitrary "magic timeouts" which are different in each test. We don't remove the test timeout parameter, because it varies depending on test case, and for now we don't have a better solution. Trying a "naive" way leads to hangs on flaky tests. TEST=UI-based tests BUG=none Review URL: http://codereview.chromium.org/1557001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43062 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a bug in LanguageConfigView where it didn't update the view when a ↵satorux@chromium.org2010-03-301-0/+3
| | | | | | | | | | | language is removed. BUG=crosbug.com/2297 TEST=manually Review URL: http://codereview.chromium.org/1585002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43061 0039d316-1c4b-4281-b951-d872f2087c98
* Roll WebKit DEPS 56640:56771.levin@chromium.org2010-03-301-1/+1
| | | | | | | | TBR=hclam@chromium.org Review URL: http://codereview.chromium.org/1545005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43060 0039d316-1c4b-4281-b951-d872f2087c98
* Add support for multiple input methods per language in LanguageConfigView.satorux@chromium.org2010-03-305-37/+96
| | | | | | | | | | | | | Along the way, refactor the code by adding SetLanguageActivated(), This is simpler than having ActivateLanguage() and DeactivateLanguage(). The idea comes from View::SetEnabled(bool enabled). BUG=crosbug.com/2001 TEST=manually Review URL: http://codereview.chromium.org/1546002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43059 0039d316-1c4b-4281-b951-d872f2087c98
* Making the translate context menu reset the blacklisting.jcivelli@chromium.org2010-03-302-7/+23
| | | | | | | | | | | | | BUG=None TEST=Select a site in a language different than the Chrome language, select never translate this site/language. Visit that site again, use the context menu to translate. The site and language should not be blacklisted for translation anymore. Review URL: http://codereview.chromium.org/1551001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43058 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 43009 - This change isn't as bad as it looks, basically it's ↵tim@chromium.org2010-03-308-311/+205
| | | | | | | | | | | templating the AutofillModelAssociator code so we can use it for both formfill (what I'm calling autofillV1 in this patch), and "profiles". Since the diffs didn't line up so well from having to move some implementation to the .h (template), I'm breaking up the review here with just the formfill code rearranged and no autofill++ impl. Review URL: http://codereview.chromium.org/1521002 TBR=tim@chromium.org Review URL: http://codereview.chromium.org/1559005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43055 0039d316-1c4b-4281-b951-d872f2087c98
* Fix WebSocketJob to pass /key_3/ after handshake request header.ukai@chromium.org2010-03-302-5/+54
| | | | | | | | | BUG=none TEST=WebSocketJob::SimpleHandshakeDraft76 passes Review URL: http://codereview.chromium.org/1256001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43054 0039d316-1c4b-4281-b951-d872f2087c98
* webkit/ ifdefs for BSD port; based on sprewell@jaggeri.com patchpvalchev@google.com2010-03-306-15/+15
| | | | | | Review URL: http://codereview.chromium.org/1483001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43053 0039d316-1c4b-4281-b951-d872f2087c98
* Make the infobars.show callback optional.finnur@chromium.org2010-03-302-8/+8
| | | | | | | | | | TBR=asargent TEST=None BUG=None Review URL: http://codereview.chromium.org/1547006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43052 0039d316-1c4b-4281-b951-d872f2087c98
* CL fixing a race-condition during experimental popup extension API pop-up ↵twiz@chromium.org2010-03-301-7/+7
| | | | | | | | | | closure. The dismissal notification is sent asynchronously, after which the pop-up is entirely destructed. The ExtensionPopupHost now holds a reference to the pop-up, so that it can safely make use of it during ExtensionPopupHost::DispatchPopupClosedEvent. BUG=None TEST=None Review URL: http://codereview.chromium.org/1512002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43051 0039d316-1c4b-4281-b951-d872f2087c98
* Makes deletion of ExtensionInstallUI happen on ui thread.sky@chromium.org2010-03-302-6/+12
| | | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/1578003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43050 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a crash when creating tabs from an Infobar by making sure finnur@chromium.org2010-03-304-0/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | the Infobar is not set as the Container when the Infobar dies or its delegate dies (so that we don't get calls from the ExtensionView after our destruction). Some background: Our ExtensionInfobarDelegate owns the Infobar state, including the ExtensionView/Host. This is because Infobars in Chrome are deleted and recreated when you switch away from the tab and come back (and we don't want to lose state doing so). We also have an ExtensionInfobar object, which is set as a "Container" for the ExtensionView, so that it can responds to resize events from the ExtensionView. The problem is that once chrome.tabs.create is called, a new tab is created, which deletes the Infobar object that has yet to receive an OnMouseLeave message from the View. Hence the crash. We therefore null the Container in the destructor for the ExtensionInfobar, to make sure the ExtensionView doesn't report more events to us after we are deleted. The only problem with this is that using the delegate_ pointer in the destructor for ExtensionInfobar is not safe, because the delegate may be deleted before us. Therefore, we also now observe when the delegate is destroyed, so we can null the Container through it. BUG=http://crbug.com/39755 TEST=Using tabs.create should not crash the extension. Also, opening two tabs, both with an infobar should not crash if you switch to the first tab and close it. Review URL: http://codereview.chromium.org/1583002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43049 0039d316-1c4b-4281-b951-d872f2087c98
* [accessibility] Extension toolbar and infobar container views should not ↵ctguil@chromium.org2010-03-309-8/+75
| | | | | | | | | | | | | | | | | | | have state visible if they are not rendered on screen. Add the ROLE_PANE role to viewsaccessibility types. Provide accessible role and name for BrowserActionsContainer, InfoBarContainer, and InfoBar. Set BrowserActionsContainer view visibility to true only when it contains browser actions. Set ExtensionShelf vew visibility to true only when it has non-zero height. BUG=36224 TEST=Verify extension toolbar and infobar container accessibility info within AccExplore32.exe Review URL: http://codereview.chromium.org/669275 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43044 0039d316-1c4b-4281-b951-d872f2087c98
* Have the background for the URL text field in the bookmark editor present as ↵mrossetti@chromium.org2010-03-304-5/+57
| | | | | | | | | | | red when the URL is not properly formed. BUG=17612 TEST=Type an unacceptable character in the URL field or leave the URL field blank and observe a red background. Review URL: http://codereview.chromium.org/837010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43043 0039d316-1c4b-4281-b951-d872f2087c98
* New autofill_specifics for autofill++ support.tim@chromium.org2010-03-301-0/+72
| | | | | | Review URL: http://codereview.chromium.org/1581001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43042 0039d316-1c4b-4281-b951-d872f2087c98
* Remove the themeing from the folder menu background, text in the folder menu ↵mrossetti@chromium.org2010-03-305-54/+5
| | | | | | | | | | | items, and no longer needed calls. BUG=39428 TEST=Pop up a folder menu and insure that the background does not adopt any theme settings. Review URL: http://codereview.chromium.org/1548003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43041 0039d316-1c4b-4281-b951-d872f2087c98
* Dropping use of enivronment variable to select branding (the old way),bradnelson@google.com2010-03-301-51/+60
| | | | | | | | | | | | | in favor of a gyp variable. Not sure how this managed to stick around this long. BUG=39744 TEST=None Review URL: http://codereview.chromium.org/1567002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43040 0039d316-1c4b-4281-b951-d872f2087c98
* views: [accessibility] Add support for GetAccessibleValue in the combobox ↵ctguil@chromium.org2010-03-302-0/+8
| | | | | | | | | | | | | control. BUG=9614 TEST=open AccExplorer, open Report broken website dialog, point AccExplorer to the combobox, see if the string value contained in the 'Bug type'combobox is the currently selected on it. Patch from Thiago Farina <thiago.farina@gmail.com> Review URL: http://codereview.chromium.org/1521001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43039 0039d316-1c4b-4281-b951-d872f2087c98
* Added --enable-video-logging, which will log the presentation timestamp of ↵scherkus@chromium.org2010-03-307-15/+49
| | | | | | | | | | | | | video frames. In conjunction with a log parser we can detect dropped frames and our display jitter. BUG=none TEST=none Review URL: http://codereview.chromium.org/1398001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43038 0039d316-1c4b-4281-b951-d872f2087c98
* Temporarily disable a failing piece of a ChromeOS test. My location bar ↵pkasting@chromium.org2010-03-301-1/+1
| | | | | | | | | | | changes changed how things get focus, so it's not clear to me what the desired behavior here is. I'll get this worked out tomorrow. TBR=sky BUG=none TEST=Unittests don't fail Review URL: http://codereview.chromium.org/1587001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43037 0039d316-1c4b-4281-b951-d872f2087c98
* Fix build error on Chromium Linuxananta@chromium.org2010-03-301-2/+2
| | | | | | | | TBR=inferno Review URL: http://codereview.chromium.org/1568006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43033 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a UMR reported on the valgrind linux build bots while running the ↵ananta@chromium.org2010-03-301-1/+2
| | | | | | | | | | DownloadRequestManagerTest.Allow test TBR=inferno Review URL: http://codereview.chromium.org/1545004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43031 0039d316-1c4b-4281-b951-d872f2087c98
* Fix use_system_ffmpeg and clean up ffmpeg.gypscherkus@chromium.org2010-03-304-206/+184
| | | | | | | | | | | | | To make use_system_ffmpeg work more correctly, we now use the system include path for headers and make use of the version number macros to determine the shared library names. Also removed a bunch of cruft without breaking anything on Windows, Mac or Linux. BUG=23602, 39430 TEST=ffmpeg should still work both with and without use_system_ffmpeg Review URL: http://codereview.chromium.org/1503002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43030 0039d316-1c4b-4281-b951-d872f2087c98
* Split GpuProcessHost into GpuProcessHostUIShim, which runs on the UIkbr@google.com2010-03-3024-240/+377
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | thread, and GpuProcessHost, which now runs on the IO thread and derives from ChildProcessHost. This split was necessary in order to service synchronous messages from the renderer process. Moved message handlers for GPU messages from renderer to browser from BrowserRenderProcessHost to ResourceMessageFilter. Stopped sending multiple ViewHostMsg_EstablishGpuChannel messages from the same renderer if the connection was already established. Resetting the channel was causing failures in Send, and every other page reload containing WebGL content to fail. This cleanup will allow further simplification in the GPU process, but this is being left for a subsequent CL. Fixed bug in sandboxing of GPU process. Fixed latent bugs in cleanup code in GpuChannel and GpuChannelHost. Fixed crashes in ChildProcessHost if resource_dispatcher_host_ was NULL. Fixed apparent latent race conditions in creation of BackingStoreProxy and VideoLayerProxy. With these changes, WebGL content is running in the sandbox on both Mac and Windows. Linux support will be added in a following CL. BUG=29120 TEST=ran WebGL demos on Mac and Windows Review URL: http://codereview.chromium.org/1546001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@43029 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-02-11 23:56:23 +0000