summaryrefslogtreecommitdiffstats
path: root/chrome/renderer
Commit message (Collapse)AuthorAgeFilesLines
* Test server cleanup patch of death:phajdan.jr@chromium.org2010-08-171-24/+9
| | | | | | | | | | | | | | - reduce the number of different classes - clean up the internal code - modify the interface to expose less internal details (this will allow more flexibility with port numbers) TEST=many BUG=49680 Review URL: http://codereview.chromium.org/3080029 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56405 0039d316-1c4b-4281-b951-d872f2087c98
* Pepper2 Graphics3D implementation - browser side.neb@chromium.org2010-08-172-0/+234
| | | | | | | | | BUG=46374 TEST=opengl demos work on Pepper2 Review URL: http://codereview.chromium.org/3020049 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56401 0039d316-1c4b-4281-b951-d872f2087c98
* Re-land part of r55416, minus actually moving the test_suite.h contents to a ↵brettw@chromium.org2010-08-171-4/+4
| | | | | | | | | | | | | | | | | | | | | | | .cc file. Cleanup in base. This moves the implementation (and a bunch of header file dependencies) from the multiprocess test and the test_suite headers to .cc files. Moves multiprocess_test to the test directory, and all of this stuff to the existing base_test_support project. I also used the base namespace. Previously other projects included this functionality just by #include because it was all inline, so I had to add dependencies on base_test_support in a few places. Moves and renames the command line switch this was using to base_switches. Move the base switch for process type to chrome switches. TEST=none BUG=none Review URL: http://codereview.chromium.org/3026055 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56399 0039d316-1c4b-4281-b951-d872f2087c98
* Include ICU headers in a way that will also work with use_system_icu=1phajdan.jr@chromium.org2010-08-172-6/+8
| | | | | | | | | TEST=none BUG=28294 Review URL: http://codereview.chromium.org/3145014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56396 0039d316-1c4b-4281-b951-d872f2087c98
* Reapplies all the IPC system work (reverts the revert r56272).erg@google.com2010-08-1711-1/+14
| | | | | | | | | | | That patch wasn't what caused the regression in the page cycler. BUG=51411,52103 TEST=still compiles Review URL: http://codereview.chromium.org/3106018 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56390 0039d316-1c4b-4281-b951-d872f2087c98
* Remove remaining deprecated wstring methods from base/values.{cc,h}.viettrungluu@chromium.org2010-08-171-4/+4
| | | | | | | | | BUG=23581 TEST=builds and passes tests Review URL: http://codereview.chromium.org/3109025 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56362 0039d316-1c4b-4281-b951-d872f2087c98
* media: use VideoDecodeEngine interface in gpu_video_decoder abstraction.jiesun@google.com2010-08-171-0/+3
| | | | | | | | some logics are moved into system memory path only. ( such as allocation of output buffer, and usage of FTBDoneACK messages. Review URL: http://codereview.chromium.org/3141014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56358 0039d316-1c4b-4281-b951-d872f2087c98
* Forbid NaN and infinite numbers in extensions API type validator.skerner@chromium.org2010-08-171-0/+10
| | | | | | | | | BUG=51449 TEST=JsonSchemaTest.TestNumber Review URL: http://codereview.chromium.org/3158015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56350 0039d316-1c4b-4281-b951-d872f2087c98
* DevTools: Support runtime property name/value pairs instead of feature names ↵pfeldman@chromium.org2010-08-172-10/+27
| | | | | | | | as navigation state. Review URL: http://codereview.chromium.org/3110018 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56348 0039d316-1c4b-4281-b951-d872f2087c98
* Support for multiple speech enabled elements in same page.satish@chromium.org2010-08-172-18/+18
| | | | | | | | | | | | | | | This is the chromium side of https://bugs.webkit.org/show_bug.cgi?id=43922. For all speech input requests WebKit now sends a 'requestId' which identifies the input element within the page. This requestId is returned back to WebKit in all the speech input callbacks/events. Within chromium, the pair (render view id, request id) uniquely identifies a caller and we introduce the SpeechInputCallerId to hold this data. All code which was previously using render view id now uses this caller id. Once this is working fully on both ends I'll add relevant browser tests. BUG=none TEST=manually Review URL: http://codereview.chromium.org/3130018 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56333 0039d316-1c4b-4281-b951-d872f2087c98
* - Registered service types to make Services menu availableyutak@chromium.org2010-08-171-1/+3
| | | | | | | | | | | | | | | - implement NSServiesRequests protocol on BrowserWindowCotroller to provide selection text to the services framework. See http://developer.apple.com/mac/library/documentation/cocoa/conceptual/SysServices/Articles/using.html for services implementation. Patch contributed by Hajime Morita <morrita@google.com>. Original review URL: http://codereview.chromium.org/2883062 BUG=20868 TEST=manual Review URL: http://codereview.chromium.org/3174012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56313 0039d316-1c4b-4281-b951-d872f2087c98
* Fix CRLF in extension_process_bindings.ccmpcomplete@chromium.org2010-08-171-700/+700
| | | | | | | | | | TBR=erikkay TEST=no BUG=no Review URL: http://codereview.chromium.org/3152022 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56280 0039d316-1c4b-4281-b951-d872f2087c98
* Before the notification provider goes away, detach it from all the active ↵johnnyg@chromium.org2010-08-172-1/+5
| | | | | | | | | | | notifications. BUG=50386 TEST=see bug Review URL: http://codereview.chromium.org/3095007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56279 0039d316-1c4b-4281-b951-d872f2087c98
* Completely revert all my IPC work to see if this was what regressed the page ↵erg@google.com2010-08-1711-14/+1
| | | | | | | | | | | cycler. BUG=51411,52103 TEST=page cycler Review URL: http://codereview.chromium.org/3170020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56272 0039d316-1c4b-4281-b951-d872f2087c98
* Relanding r55510:apatrick@chromium.org2010-08-161-4/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | This CL originally caused crashes on the bots. I tracked the problem down to a bug in ANGLE that I fixed upstream: http://codereview.appspot.com/1956042/show This fix was committed upstream and I rolled the Chrome DEPS so the bot crash issue should be resolved. Original comment for r55510: Prelaunch GPU process after 10 seconds. This will happen if the --prelaunch-gpu-process command line switch is present or for 5% of users in a field trial. This is so we can collect some stats of GPU capabilities. TEST=try BUG=none Review URL: http://codereview.chromium.org/3179010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56239 0039d316-1c4b-4281-b951-d872f2087c98
* DevTools: fix compilation on Linux: initialize fields in the same order as ↵yurys@chromium.org2010-08-161-1/+1
| | | | | | | | they are declared. Review URL: http://codereview.chromium.org/3156015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56185 0039d316-1c4b-4281-b951-d872f2087c98
* DevTools: support dispatching devtools commands when there is a tight loop ↵yurys@chromium.org2010-08-164-11/+41
| | | | | | | | in JavaScript Review URL: http://codereview.chromium.org/3152011 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56184 0039d316-1c4b-4281-b951-d872f2087c98
* FBTF: Remove "obviously" unneeded standard C++ library #includes.viettrungluu@chromium.org2010-08-159-15/+4
| | | | | | | | | BUG=none TEST=builds Review URL: http://codereview.chromium.org/3179017 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56171 0039d316-1c4b-4281-b951-d872f2087c98
* Remove (deprecated) wstring version of Value::CreateStringValue().viettrungluu@chromium.org2010-08-151-2/+4
| | | | | | | | | BUG=23581 TEST=builds and passes tests Review URL: http://codereview.chromium.org/3136012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56170 0039d316-1c4b-4281-b951-d872f2087c98
* Clean up PLT histograms for a move to navigator_state.ccjar@chromium.org2010-08-152-107/+166
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Clean up PLT gathering in preparation for moving it to take place in navigation_state.cc, during the instance destructor. This CL should have almost no semantic impact, and I'm submitting it separately so that when I move large blocks, I can mostly move them unchanged. Add a bunch of DCHECKs to be sure that finish and finish_doc times are only set once, and that they are set only in the correct order. In a future CL, I expect to push a lot of the time-gathering code into navigation_state.cc, along with the generation of the final histograms. The challenge is to mark the NavigationState instances that are worthy of recording (which looks like it MAY be done by marking the state with the boolean to indicate it is part of an HTTP or HTTPS connection), and make sure we have good numbers throughout. bug=48970 r=mbelshe Review URL: http://codereview.chromium.org/2993004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56164 0039d316-1c4b-4281-b951-d872f2087c98
* Plumb the find progress's "we're done" value to the browser.avi@chromium.org2010-08-134-39/+23
| | | | | | | | | BUG=51433 TEST=as in bug Review URL: http://codereview.chromium.org/3157004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56069 0039d316-1c4b-4281-b951-d872f2087c98
* Fix building with enable_gpu=0neb@chromium.org2010-08-133-4/+11
| | | | | | | | | BUG=crbug.com/51459 TEST=passed compilation Review URL: http://codereview.chromium.org/3168008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@56041 0039d316-1c4b-4281-b951-d872f2087c98
* remove toolstripserikkay@chromium.org2010-08-126-868/+701
| | | | | | | | | | | BUG=51703 BUG=30178 BUG=25106 TEST=none Review URL: http://codereview.chromium.org/3129003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55921 0039d316-1c4b-4281-b951-d872f2087c98
* Check for the GL_EXT_read_format_bgra extension, now that the command buffer ↵senorblanco@google.com2010-08-121-4/+2
| | | | | | | | | | | supports it. BUG=none TEST=--use-accelerated-2d-canvas, any canvas content w/images Review URL: http://codereview.chromium.org/3125007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55914 0039d316-1c4b-4281-b951-d872f2087c98
* FBTF: Remove all the easy headers from *_messages.h.erg@google.com2010-08-1211-1/+12
| | | | | | | | | BUG=51411 TEST=compiles Review URL: http://codereview.chromium.org/3174002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55902 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 55888: Allow per-plugin content settings.bauerb@chromium.org2010-08-123-26/+15
| | | | | | | | | | | | | | | | | | | | | | | Run with --enable-resource-content-settings and --enable-click-to-play to enable. TabSpecificContentSettings now keeps track of which resources were blocked. The content setting bubble displays those resource, and selecting the "allow radio buttons adds exceptions for these resources. The infobar for non-sandboxed plug-ins now also shows a button to always allow the blocked plug-in on that site. Screenshot of the plugin bubble: http://imgur.com/6npqv.png Because we need to change the title and radio button labels depending on whether we track specific plug-ins, we copy them from the ContentSettingBubbleModel. XIB changes: Add a text field for the blocked plug-ins to ContentBlockedPlugins.xib, hooked up to the |blockedResourcesField_| outlet. BUG=39252,38432 TEST=manual Review URL: http://codereview.chromium.org/2873104 TBR: bauerb@chromium.org git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55891 0039d316-1c4b-4281-b951-d872f2087c98
* Allow per-plugin content settings.bauerb@chromium.org2010-08-123-15/+26
| | | | | | | | | | | | | | | | | | | | | Run with --enable-resource-content-settings and --enable-click-to-play to enable. TabSpecificContentSettings now keeps track of which resources were blocked. The content setting bubble displays those resource, and selecting the "allow radio buttons adds exceptions for these resources. The infobar for non-sandboxed plug-ins now also shows a button to always allow the blocked plug-in on that site. Screenshot of the plugin bubble: http://imgur.com/6npqv.png Because we need to change the title and radio button labels depending on whether we track specific plug-ins, we copy them from the ContentSettingBubbleModel. XIB changes: Add a text field for the blocked plug-ins to ContentBlockedPlugins.xib, hooked up to the |blockedResourcesField_| outlet. BUG=39252,38432 TEST=manual Review URL: http://codereview.chromium.org/2873104 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55888 0039d316-1c4b-4281-b951-d872f2087c98
* Initialize the members and avoid inlining the constructors.finnur@chromium.org2010-08-125-12/+25
| | | | | | | | | | BUG=None TEST=None CID=12377, 12379, 12380, 12381 Review URL: http://codereview.chromium.org/3176004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55863 0039d316-1c4b-4281-b951-d872f2087c98
* A fix for Issue 46278.hbono@chromium.org2010-08-122-9/+31
| | | | | | | | | | When an ICU ruleset defines a dictionary value "$dictionary = [:LineBreak = Complex_Context:]", ICU expects we use this value in breaking text. On the other hand, when we do not use this value, ubrk_next() gets stuck in an infinite loop and it causes Issue 46278. To fix this issue, we define |$dictionary| only for the languages we need the value, e.g. Korean and Thai. BUG=46278 TEST=unit_test.exe --gtest_filter=SpellcheckWordIteratorTest.RuleConsistency Review URL: http://codereview.chromium.org/3108003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55856 0039d316-1c4b-4281-b951-d872f2087c98
* Moving the autofill code from WebKit EditorClientImpl to Chromium.jcivelli@google.com2010-08-126-43/+151
| | | | | | | | | | | This is still behind an #ifdef at this point. BUG=None TEST=None Review URL: http://codereview.chromium.org/3060010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55824 0039d316-1c4b-4281-b951-d872f2087c98
* Adding tts experimental extensions apichaitanyag@chromium.org2010-08-123-1/+17
| | | | | | Review URL: http://codereview.chromium.org/2969005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55822 0039d316-1c4b-4281-b951-d872f2087c98
* Autocomplete doesn't honor maxlength attribute.dhollowa@chromium.org2010-08-121-2/+4
| | | | | | | | | | | When filling Autocomplete suggestions limit the filling of the suggestion to the maxlength attribute of the filled element. BUG=46899 TEST=Manual test as per bug steps. Review URL: http://codereview.chromium.org/3179005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55814 0039d316-1c4b-4281-b951-d872f2087c98
* AutoFill: Release the cached frame when we receive the frameDestroyed() messagejhawkins@chromium.org2010-08-114-5/+20
| | | | | | | | | | | from WebKit. BUG=48857 TEST=none Review URL: http://codereview.chromium.org/3173005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55789 0039d316-1c4b-4281-b951-d872f2087c98
* Add support for BGRA pixel format to the command buffer implementation of ↵senorblanco@google.com2010-08-112-0/+22
| | | | | | | | | | | WebGraphicsContext3D. This will not be called until the WebKit-side change lands. BUG=none TEST=none Review URL: http://codereview.chromium.org/3170005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55786 0039d316-1c4b-4281-b951-d872f2087c98
* Enable GLSL shader translator by default. With this change, Chrome'skbr@google.com2010-08-111-1/+1
| | | | | | | | | | | | WebGL implementation requires GLSL ES compliant shaders. Now pass --disable-glsl-translator on command line to disable the translator. BUG=none TEST=ran WebGL demos Review URL: http://codereview.chromium.org/3121005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55764 0039d316-1c4b-4281-b951-d872f2087c98
* Reapplies r55735, fixing some message code that was checked in that I ↵erg@google.com2010-08-111-0/+2
| | | | | | | | | | | | collided with. BUG=51411 TEST=none TBR=mpcomplete Review URL: http://codereview.chromium.org/3118009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55749 0039d316-1c4b-4281-b951-d872f2087c98
* Revert "FBTF: Allow forward declaration of classes passed to sync IPC messages."erg@google.com2010-08-111-2/+0
| | | | | | | | | | | | This reverts commit r55735. BUG=none TEST=none TBR=mirandac Review URL: http://codereview.chromium.org/3152007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55738 0039d316-1c4b-4281-b951-d872f2087c98
* FBTF: Allow forward declaration of classes passed to sync IPC messages.erg@google.com2010-08-111-0/+2
| | | | | | | | | | | | | This is a continuation of r55259, where we convert the remaining IPC messages to the new system. At this point, all the messages should be able to accept forward declared classes/structs as parameters. BUG=51411 TEST=none Review URL: http://codereview.chromium.org/3018045 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55735 0039d316-1c4b-4281-b951-d872f2087c98
* Chromium plumbing for Device Orientation.hans@chromium.org2010-08-114-0/+127
| | | | | | | | | | | | | | | Add the plumbing needed for communicating with the Device Orientation code in WebKit. RenderView provides an implementation of WebKit::WebDeviceOrientationClient: DeviceOrientationDispatcher. This communicates with the browser-side class device_orientation::DispatcherHost. device_orientation::Provider, responsible for providing the orientation data, is just an empty shell for now. BUG=44654 TEST=browser_tests --gtest_filter=DeviceOrientationBrowserTest.BasicTest Review URL: http://codereview.chromium.org/2858049 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55724 0039d316-1c4b-4281-b951-d872f2087c98
* Initialize member struct in Pepper plugin.finnur@chromium.org2010-08-111-0/+1
| | | | | | | | | | BUG=None TEST=None CID=12352 Review URL: http://codereview.chromium.org/3037055 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55722 0039d316-1c4b-4281-b951-d872f2087c98
* Implements initial plumbing for IDBCursor.bulach@chromium.org2010-08-116-1/+143
| | | | | | | | | This is the Chromium side of https://bugs.webkit.org/show_bug.cgi?id=41888 It's almost purely plumbing: actual logic and cursor implementation will follow soon. Review URL: http://codereview.chromium.org/2918002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55693 0039d316-1c4b-4281-b951-d872f2087c98
* When running in single process mode, the sandbox is off. Tell the file ↵jorlow@chromium.org2010-08-111-0/+1
| | | | | | | | | | | system object this. TEST=IndexedDB works with --single-process BUG=none Review URL: http://codereview.chromium.org/3116003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55691 0039d316-1c4b-4281-b951-d872f2087c98
* DevTools: remove usages of InjectDispatch.jsyurys@chromium.org2010-08-112-7/+0
| | | | | | Review URL: http://codereview.chromium.org/3125004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55687 0039d316-1c4b-4281-b951-d872f2087c98
* Making window.focus() work in Chrome.jcivelli@chromium.org2010-08-111-7/+1
| | | | | | | | | | | | | | | | A patch on the WebKit side https://bugs.webkit.org/show_bug.cgi?id=43542 ensures we get the correct notification when window.focus()/blur() is called. This CL makes the browser active/unactive in response to the focus/blur message. BUG=29643 TEST=Visit gmail. Open an IM conversation window and pop-it out. Focus the browser (so the IM window is in the back). Click on the person name in the IM section of gmail. The conversation window should be brought back to the front. Repeat but this time minimize the conversation window. Review URL: http://codereview.chromium.org/3060045 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55678 0039d316-1c4b-4281-b951-d872f2087c98
* Added event traces at various points during Chrome and ChromeFrame startup ↵ananta@chromium.org2010-08-112-0/+11
| | | | | | | | | | to help measure and track performance. Bug=51638 Review URL: http://codereview.chromium.org/3152001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55671 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 55510 - Prelaunch GPU process after 10 seconds.mirandac@chromium.org2010-08-101-17/+4
| | | | | | | | | | | | | | | | This will happen if the --prelaunch-gpu-process command line switch is present or for 5% of users in a field trial. This is so we can collect some stats of GPU capabilities. TEST=try, check --prelaunch-gpu-process works BUG=none Review URL: http://codereview.chromium.org/3119002 TBR=apatrick@chromium.org Review URL: http://codereview.chromium.org/3162003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55570 0039d316-1c4b-4281-b951-d872f2087c98
* 1. ipc_video_decoder.cc/h is media pipeline filter which use the gpu decoder ↵jiesun@google.com2010-08-1011-1/+928
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | facilities in video stack. it is only enabled when (a) hardware composition is on (b) hardware decoding command line is on (c) h264 codec is specified. 2. gpu_video_service.cc/h is a singleton in gpu process which provide video services for renderer process, through it we could create decoder. ( in my imagination, in the future, we could create encoder or capturer too) 3. gpu_video_decoder.cc/h. abstract interface for hardware decoder. 4. gpu_video_service_host.cc/h is singleton in renderer process which provide proxy for gpu_video_service. 5. gpu_video_decoder_host.cc/h is proxy for gpu_video_decoder. (1 to 1 map).basically there is one global GpuVideoService in GPU process, one GpuVideoServiceHost in Renderer process. for each renderer process, there are could be multiple renderer view, each could had multiple GpuVideoDecoderHost the connect to GpuVideoDeocder through GPUCHannelHOst/GpuChannel. 6. gpu_video_common.cc/h: IPC message definition and pickle/marshaling support. ISSUES: 1. in media pipeline, we need let decoder to determine if bit stream filter should be used instead of let command line to determine it. 2. stop readback from D3D surface use ANGLE. 3. Flush logic still need fine tuning. 4. CreateThread in GpuVideoDecoder, and post message in message handler, and derived classs handle message loop. ? 5. Error handling. 6. Input ring buffer implementation. Current impl is naive. 7.Add output queue for MFT decoder. 8. Query Capabilities at GetVideoServices()... BUG=None TEST=Windows7 Review URL: http://codereview.chromium.org/2873089 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55516 0039d316-1c4b-4281-b951-d872f2087c98
* Prelaunch GPU process after 10 seconds.apatrick@chromium.org2010-08-101-4/+17
| | | | | | | | | | | | | This will happen if the --prelaunch-gpu-process command line switch is present or for 5% of users in a field trial. This is so we can collect some stats of GPU capabilities. TEST=try, check --prelaunch-gpu-process works BUG=none Review URL: http://codereview.chromium.org/3119002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55510 0039d316-1c4b-4281-b951-d872f2087c98
* Histograms & Field Trails about prefetchinggavinp@google.com2010-08-092-6/+111
| | | | | | | | | | | | This CL lands some histograms to track prefetching, and a 10% field trial enabling the feature to track performance. BUG=none TEST=none Review URL: http://codereview.chromium.org/3036050 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55431 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 55405 - Special thanks for in-ming cheng's MFT hardware decodering code.ananta@chromium.org2010-08-0911-928/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1. ipc_video_decoder.cc/h is media pipeline filter which use the gpu decoder facilities in video stack. it is only enabled when (a) hardware composition is on (b) hardware decoding command line is on (c) h264 codec is specified. 2. gpu_video_service.cc/h is a singleton in gpu process which provide video services for renderer process, through it we could create decoder. ( in my imagination, in the future, we could create encoder or capturer too) 3. gpu_video_decoder.cc/h. abstract interface for hardware decoder. 4. gpu_video_decoder_mft.cc/h media foundation transform hardware decoder which run on windows 7 only. 5. gpu_video_service_host.cc/h is singleton in renderer process which provide proxy for gpu_video_service. 6. gpu_video_decoder_host.cc/h is proxy for gpu_video_decoder. (1 to 1 map).basically there is one global GpuVideoService in GPU process, one GpuVideoServiceHost in Renderer process. for each renderer process, there are could be multiple renderer view, each could had multiple GpuVideoDecoderHost the connect to GpuVideoDeocder through GPUCHannelHOst/GpuChannel. 7. gpu_video_common.cc/h: IPC message definition and pickle/marshaling support. ISSUES: 1. in media pipeline, we need let decoder to determine if bit stream filter should be used instead of let command line to determine it. 2. stop readback from D3D surface use ANGLE. 3. Flush logic still need fine tuning. 4. CreateThread in GpuVideoDecoder, and post message in message handler, and derived classs handle message loop. ? 5. Error handling. 6. Input ring buffer implementation. Current impl is naive. 7.Add output queue for MFT decoder. 8. Query Capabilities at GetVideoServices()... BUG=None TEST=Windows7 Review URL: http://codereview.chromium.org/2873089 TBR=jiesun@google.com Review URL: http://codereview.chromium.org/3020077 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@55419 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-01-18 23:57:36 +0000