summaryrefslogtreecommitdiffstats
path: root/content/browser
Commit message (Collapse)AuthorAgeFilesLines
* Add conversion rule of WebAXRoleBlockquoteje_julie.kim2014-10-252-0/+8
| | | | | | | | | | | | The related to CL: https://codereview.chromium.org/654303004/ BUG=347854 TBR=rockot Review URL: https://codereview.chromium.org/656683004 Cr-Commit-Position: refs/heads/master@{#301272}
* Fix Flash fullscreen focus regressions, and add interactive_ui_tests.miu2014-10-257-26/+44
| | | | | | | | | | | | | | | | | | | | | | | This revisits focus handling logic relating to Flash fullscreen in WebContentsImpl. Before this change, calls to WebContentsImpl::Focus() would directly call RenderWidgetHostView::Focus(), but this prevented WebContentsView and its delegate from "shadowing" these focus changes in the browser UI. In other words, on Windows/Linux/CrOS, the focus state for widgets in the ui::views toolkit were not necessarily synchronized with the render widget's focus state (see bug for details). 90% of this change adds regression testing for everything described above. The tests run a simulated fullscreened Flash widget, and then invoke user keyboard and mouse actions on the browser UI to confirm both the Flash widget and ui::views are in agreement and respond correctly to focus changes; also, including focus store/restore when switching tabs. BUG=403641 TEST=Repro steps in bug 403641; and also that focus remains with the omnibox or tab contents when switching back and forth to other tabs. New automated interactive_ui_tests. Review URL: https://codereview.chromium.org/670653005 Cr-Commit-Position: refs/heads/master@{#301269}
* Re-land: Add browser-wide discardable memory implementation.reveman2014-10-252-0/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | This provides a platform agnostic implementation of discardable memory based on shared memory. Managment of discardable memory is moved to the browser process instead of being local to each renderer. This provides much better control over chromium's total discardable memory usage. A round-trip to the browser IO thread is necessary for a renderer to allocate a new discardable memory segment but the cost of this can be reduced by maintaining free lists on the renderer side and have discardable memory instances share the same discardable memory segment. The implementation of this optimization is left as a follow up. Note: while the general implementation is platform agnostic, it takes advantage of the ftruncate syscall available on Posix platforms. This syscall makes it possible to release discardable memory segments immediately to the OS from the browser process. BUG=381178,400423,422953 TEST=base_unittests --gtest_filter=DiscardableMemoryTests*, base_unittests --gtest_filter=DiscardableSharedMemory.*, content_unittests --gtest_filter=HostDiscardableSharedMemoryManagerTest.* Review URL: https://codereview.chromium.org/531343002 Cr-Commit-Position: refs/heads/master@{#301247}
* Enable SitePerProcessBrowserTest cases on Android.nasko2014-10-251-19/+5
| | | | | | | | BUG=399775 Review URL: https://codereview.chromium.org/673243003 Cr-Commit-Position: refs/heads/master@{#301244}
* [OSX] Remove RWHV::SetTakesFocusOnlyOnMouseDowngroby2014-10-257-64/+1
| | | | | | | | | | | SetTakesFocusOnlyOnMouseDown is no longer used. BUG=424766 R=avi@chromium.org Review URL: https://codereview.chromium.org/679503002 Cr-Commit-Position: refs/heads/master@{#301237}
* Make OnGpuSwitching available in the GPU processccameron2014-10-246-19/+33
| | | | | | | | | | | | | | | | | | | | | The motivation for this is that the GPU process on Mac will need to force its CAOpenGLLayers (among other things) to re-create themselves when the GPU switches. Move the OnGpuSwitching observer method to a new class, ui::GpuSwitchingObserver. Inform the GPU process of OnGpuSwitching events via an IPC from the browser. This is so that - any blacklisting or filtering of events can be done in one location - it is possible to reason about the order in which notifications occur - the existing interactions with GpuDataManager remain unchanged BUG=424433 Review URL: https://codereview.chromium.org/649933005 Cr-Commit-Position: refs/heads/master@{#301207}
* Fix race condition in surfaceless partial swap.achaulk2014-10-246-11/+34
| | | | | | | | | | | Existing GLHelper uses a different command buffer, so the copies race the bind & swap. BUG=380861 Review URL: https://codereview.chromium.org/655103004 Cr-Commit-Position: refs/heads/master@{#301190}
* [ServiceWorkerCache] Clean up the service worker cache protobufjkarlin2014-10-244-86/+87
| | | | | | | | | | | | | Remove unnecessary values (cache size) and organize the request/response metadata into separate messages. This is cleaner in the long run and now is the time to do it before the cache bit is flipped and people start using it. BUG=426878 Review URL: https://codereview.chromium.org/650993005 Cr-Commit-Position: refs/heads/master@{#301180}
* Revert of Add browser-wide discardable memory implementation. (patchset #23 ↵avi2014-10-242-18/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | id:910001 of https://codereview.chromium.org/531343002/) Reason for revert: Causes compile error. http://build.chromium.org/p/chromium.linux/builders/Android%20Builder/builds/39232 ../../base/memory/discardable_shared_memory.cc:208:5:error: ignoring return value of function declared with warn_unused_result attribute [-Werror,-Wunused-result] ftruncate(handle.fd, sizeof(SharedState)); ^~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1 error generated. Original issue's description: > Add browser-wide discardable memory implementation. > > This provides a platform agnostic implementation of discardable > memory based on shared memory. Managment of discardable memory is > moved to the browser process instead of being local to each renderer. > This provides much better control over chromium's total discardable > memory usage. > > A round-trip to the browser IO thread is necessary for a renderer > to allocate a new discardable memory segment but the cost of this > can be reduced by maintaining free lists on the renderer side and > have discardable memory instances share the same discardable > memory segment. The implementation of this optimization is left as > a follow up. > > Note: while the general implementation is platform agnostic, it > takes advantage of the ftruncate syscall available on Posix > platforms. This syscall makes it possible to release discardable > memory segments immediately to the OS from the browser process. > > BUG=381178,400423,422953 > TEST=base_unittests --gtest_filter=DiscardableMemoryTests*, base_unittests --gtest_filter=DiscardableSharedMemory.*, content_unittests --gtest_filter=HostDiscardableSharedMemoryManagerTest.* > > Committed: https://crrev.com/2ac2d83d86b0206b908eea0d6e1002850d238a88 > Cr-Commit-Position: refs/heads/master@{#301152} TBR=skuhne@chromium.org,rmcilroy@chromium.org,danakj@chromium.org,rsesek@chromium.org,nasko@chromium.org,reveman@chromium.org NOTREECHECKS=true NOTRY=true BUG=381178,400423,422953 Review URL: https://codereview.chromium.org/678603003 Cr-Commit-Position: refs/heads/master@{#301156}
* Add browser-wide discardable memory implementation.reveman2014-10-242-0/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | This provides a platform agnostic implementation of discardable memory based on shared memory. Managment of discardable memory is moved to the browser process instead of being local to each renderer. This provides much better control over chromium's total discardable memory usage. A round-trip to the browser IO thread is necessary for a renderer to allocate a new discardable memory segment but the cost of this can be reduced by maintaining free lists on the renderer side and have discardable memory instances share the same discardable memory segment. The implementation of this optimization is left as a follow up. Note: while the general implementation is platform agnostic, it takes advantage of the ftruncate syscall available on Posix platforms. This syscall makes it possible to release discardable memory segments immediately to the OS from the browser process. BUG=381178,400423,422953 TEST=base_unittests --gtest_filter=DiscardableMemoryTests*, base_unittests --gtest_filter=DiscardableSharedMemory.*, content_unittests --gtest_filter=HostDiscardableSharedMemoryManagerTest.* Review URL: https://codereview.chromium.org/531343002 Cr-Commit-Position: refs/heads/master@{#301152}
* Revert "Get rid of net::LOAD_ENABLE_UPLOAD_PROGRESS."Mitsuru Oshima2014-10-245-11/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | > Get rid of net::LOAD_ENABLE_UPLOAD_PROGRESS. > It's only used by Loader code. > BUG=426442 > > Review URL: https://codereview.chromium.org/669073003 > > Cr-Commit-Position: refs/heads/master@{#301110} TBR=ljagielski@opera.com Reason for revert: This is causing use-of-unititialized-value on msan bots. http://build.chromium.org/p/chromium.memory.fyi/builders/Linux%20MSan%20Browser%20%281%29/builds/1746/steps/browser_tests/logs/stdio ==29276== WARNING: MemorySanitizer: use-of-uninitialized-value #0 0x7fcb9a0f0d17 in content::ResourceLoader::ReportUploadProgress() content/browser/loader/resource_loader.cc:163:9 #1 0x7fcb9a0f3241 in content::ResourceLoader::OnResponseStarted(net::URLRequest*) content/browser/loader/resource_loader.cc:341:3 #2 0x7fcb980bb80f in net::URLRequestJob::NotifyHeadersComplete() net/url_request/url_request_job.cc:376:3 #3 0x7fcb98304f13 in net::URLRequestHttpJob::NotifyHeadersComplete() net/url_request/url_request_http_job.cc:355:3 #4 0x7fcb983080de in net::URLRequestHttpJob::SaveNextCookie() net/url_request/url_request_http_job.cc:692:5 #5 0x7fcb98301b30 in net::URLRequestHttpJob::OnStartCompleted(int) net/url_request/url_request_http_job.cc:848:5 #6 0x7fcb97f1f140 in Run base/callback.h:441:12 #7 0x7fcb97f1f140 in net::HttpCache::Transaction::DoCallback(int) net/http/http_cache_transaction.cc:727:0 #8 0x7fcb97f1ae2f in HandleResult net/http/http_cache_transaction.cc:733:5 #9 0x7fcb97f1ae2f in net::HttpCache::Transaction::DoLoop(int) net/http/http_cache_transaction.cc:994:0 #10 0x7fcb97f185f4 in net::HttpCache::Transaction::OnIOComplete(int) net/http/http_cache_transaction.cc:2955:3 Review URL: https://codereview.chromium.org/680583002 Cr-Commit-Position: refs/heads/master@{#301139}
* Remove abstract Clone and Cancel methods from MotionEventjdduke2014-10-2410-417/+187
| | | | | | | | | | | | | | | | Create a single implementation of MotionEvent's |Clone()| and |Cancel()| methods using MotionEventGeneric. Expand MotionEventGeneric to support historical events, also removing CompoundMotionEvent from the MotionEventBuffer implementation. By avoiding JNI and garbage creation, this reduces the cost of both methods by ~10x on Android, from ~25us call to ~2us. BUG=501503003 Review URL: https://codereview.chromium.org/502993004 Cr-Commit-Position: refs/heads/master@{#301136}
* Revert of Close the select box when the parent RWHVA loses focus (patchset ↵yzshen2014-10-243-50/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | #2 id:170001 of https://codereview.chromium.org/652793003/) Reason for revert: Broke Linux ChromiumOS Tests (dbg)(1): RenderWidgetHostViewAuraTest.PopupClosesWhenParentLosesFocus (run #1): [ RUN ] RenderWidgetHostViewAuraTest.PopupClosesWhenParentLosesFocus Xlib: extension "RANDR" missing on display ":9". ../../content/browser/renderer_host/render_widget_host_view_aura_unittest.cc:875: Failure Value of: observer.destroyed() Actual: false Expected: true [ FAILED ] RenderWidgetHostViewAuraTest.PopupClosesWhenParentLosesFocus (10 ms) Original issue's description: > Close the select box when the parent RWHVA loses focus > > BUG=422264,414550 > TEST=RenderWidgetHostViewAuraTest.PopupClosesWhenParentLosesFocus > > Committed: https://crrev.com/1f909ee8c8ae5cbc92416482374d2228f087684c > Cr-Commit-Position: refs/heads/master@{#301108} TBR=tkent@chromium.org,ben@chromium.org,pkotwicz@chromium.org NOTREECHECKS=true NOTRY=true BUG=422264,414550 Review URL: https://codereview.chromium.org/663483005 Cr-Commit-Position: refs/heads/master@{#301127}
* Add DumpAccessibilityTree tests (5 of 20).shreeram.k2014-10-241-0/+24
| | | | | | | | | | | | | | | | | HTML Tags: head, input type tel, noscript ARIA Role: listbox, separator ARIA state/property: aria-required(true/false) Following HTML tags are undergoing/done in some other CL: HTML Tag - input type num, pre, header, blockquote. HTML ruby tag is not exposed correctly. Will submit it with the fix in new CL. BUG=170577 Review URL: https://codereview.chromium.org/680483002 Cr-Commit-Position: refs/heads/master@{#301115}
* Get rid of net::LOAD_ENABLE_UPLOAD_PROGRESS.ljagielski2014-10-245-1/+11
| | | | | | | | | | It's only used by Loader code. BUG=426442 Review URL: https://codereview.chromium.org/669073003 Cr-Commit-Position: refs/heads/master@{#301110}
* Close the select box when the parent RWHVA loses focuspkotwicz2014-10-243-8/+50
| | | | | | | | | BUG=422264,414550 TEST=RenderWidgetHostViewAuraTest.PopupClosesWhenParentLosesFocus Review URL: https://codereview.chromium.org/652793003 Cr-Commit-Position: refs/heads/master@{#301108}
* Reset accessibility if it gets out of sync.dmazzoni2014-10-243-10/+257
| | | | | | | | | | | | | | | | | | | | Previously if the browser got an accessibility IPC it was unable to interpret, it killed the renderer. This change makes it reset the accessibility state instead so we're tolerant of corner cases that are very difficult to completely eliminate in practice. The accessibility reset is designed to be safe, it throws away all accessibility state on the browser side and waits for an IPC from the renderer acknowledging the renderer reset as well. BUG=372478 Committed: https://crrev.com/0c5e8d16661df7e08c708997f24c0a3070e58080 Cr-Commit-Position: refs/heads/master@{#298297} Review URL: https://codereview.chromium.org/625443002 Cr-Commit-Position: refs/heads/master@{#301106}
* [ServiceWorkerCache] Return the real size of the cache in GetOriginUsage.jkarlin2014-10-248-6/+134
| | | | | | | | | | | | The size of the cache on disk is used if the backend is persistent, else the size of the entries in memory is returned. Downstream of: https://codereview.chromium.org/674873002/ BUG=420159 Review URL: https://codereview.chromium.org/672943002 Cr-Commit-Position: refs/heads/master@{#301097}
* [ServiceWorkerCache] Call QuotaManager::NotifyStorageModified from Cache.jkarlin2014-10-249-74/+220
| | | | | | | | | | | | | | | | | | | With this call the quota manager can delete origins modified in the last X hours. Changes: * Plumb origin and QuotaManagerProxy through to ServiceWorkerCache. * Call NotifyStorageModified on SWCache::Put() and SWCache::Delete() with the delta being the number of content bytes written (not including caching structure overhead). * Update the tests to use a MockQuotaManagerProxy instead of passing a nullptr. Upstream of: https://codereview.chromium.org/672943002 BUG=420159 Review URL: https://codereview.chromium.org/674873002 Cr-Commit-Position: refs/heads/master@{#301086}
* Remove old WebSocket implementation from content/ricea2014-10-245-567/+0
| | | | | | | | | | | | | | | | | | Remove the old SocketStream-based WebSocket implementation from content/. Large parts of the implementation still exist in net/. They will be removed in a subsequent CL. There are also a few traces in chrome/, and some interfaces in Blink that cannot be removed until this CL is submitted. BUG=423201 TEST=layout tests, browser_tests Review URL: https://codereview.chromium.org/655253006 Cr-Commit-Position: refs/heads/master@{#301075}
* content: Cleanup, remove unused SkPorterDuff.h include.tfarina2014-10-241-1/+0
| | | | | | | | | | | | | This skia API is marked as deprecated, so we need to remove all usages of it, to be able to remove it in Skia. BUG=None TEST=None TBR=skyostil@chromium.org Review URL: https://codereview.chromium.org/669303005 Cr-Commit-Position: refs/heads/master@{#301045}
* Cleanup: Update the paths to gfx vector2d_conversions.h include.tfarina2014-10-241-1/+1
| | | | | | | | | | | | | The geometry headers were moved from /ui/gfx/ to ui/gfx/geometry, so we are updating the include paths to fix this. BUG=395370 TEST=None TBR=ben@chromium.org,danakj@chromium.org Review URL: https://codereview.chromium.org/669153008 Cr-Commit-Position: refs/heads/master@{#301035}
* Disable ServiceWorkerBlackBoxBrowserTest.Registration on ChromeOSksakamoto2014-10-241-1/+7
| | | | | | | | | | This partially reverts r300403. The test was still flaking on ChromeOS. BUG=387045 Review URL: https://codereview.chromium.org/669333002 Cr-Commit-Position: refs/heads/master@{#301034}
* Chrome side of passing on the service worker registration with geofencing ↵mek2014-10-245-14/+71
| | | | | | | | | | API calls. BUG=383125 Review URL: https://codereview.chromium.org/623823002 Cr-Commit-Position: refs/heads/master@{#301010}
* Cleanup: Merge device/udev_linux/udev.{cc,h} into ↵thestig2014-10-244-39/+36
| | | | | | | | | | | | | device/udev_linux/scoped_udev.h. - Update header files elsewhere. - Use ScopedUdev*Ptr classes in more places. BUG=415212 Review URL: https://codereview.chromium.org/666653002 Cr-Commit-Position: refs/heads/master@{#300999}
* [ServiceWorker] Check ignore_certificate_errors flag in ↵horo2014-10-231-3/+8
| | | | | | | | | | | | | | | | | ServiceWorkerWriteToCacheJob. https://codereview.chromium.org/643773004 broke LayoutTest fetch-mixed-content.html. While executing LayoutTest, ignore_certificate_errors is set in LayoutTestBrowserContext. So every cert errors are ignored in HttpStreamFactoryImpl::Job::HandleCertificateError. At this case we should also ignore the cert errors in ServiceWorkerWriteToCacheJob. BUG=426047,425396 Review URL: https://codereview.chromium.org/664343005 Cr-Commit-Position: refs/heads/master@{#300989}
* Android: add support for Device Orientation on devices where ROTATION_VECTOR ↵timvolodine2014-10-232-2/+14
| | | | | | | | | | is not available. BUG=347507 Review URL: https://codereview.chromium.org/661413003 Cr-Commit-Position: refs/heads/master@{#300985}
* Sets the default background color of inline signin and user manager to greyguohui2014-10-2314-34/+51
| | | | | | | | | | | | | | | The purpose is to avoid the white flash before the web content is loaded. This CL only handles the view implementation, mac implementation will be handled in the next cl. BUG=368225 Committed: https://crrev.com/fe517ffc1a9550ff4557b5718b01b103bd134616 Cr-Commit-Position: refs/heads/master@{#299594} Review URL: https://codereview.chromium.org/637083002 Cr-Commit-Position: refs/heads/master@{#300961}
* Refactor GeofencingManager to have one instance per StoragePartition.mek2014-10-2313-396/+939
| | | | | | | | | | | Add a new GeofencingService class for the few global tasks GeofencingManager used to do. BUG=383125 Review URL: https://codereview.chromium.org/645763003 Cr-Commit-Position: refs/heads/master@{#300960}
* Use manifest for streamlined hosted app creation for title and URL.benwells2014-10-231-4/+7
| | | | | | | | | | This uses the manifest as defined in https://w3c.github.io/manifest/. BUG=425837 Review URL: https://codereview.chromium.org/667053007 Cr-Commit-Position: refs/heads/master@{#300956}
* Always destroy SpeechRecognitionDispatcherHost on IO thread.lazyboy2014-10-232-0/+8
| | | | | | | | | | | | | SRDH can be destroyed from ~PendingTask on mac (non-IO) Deleting SRDH on non-IO results in a DCHECK failure while calling: SpeechRecognitionManagerImpl::AbortAllSessionsForRenderProcess(). BUG=426495 Test=See bug description for repro, requires a debug build (or DCHECK enabled somehow). Review URL: https://codereview.chromium.org/662033003 Cr-Commit-Position: refs/heads/master@{#300952}
* Make enable override disable impl-side paintingenne2014-10-231-6/+3
| | | | | | | | | | | | | | | In order to turn on impl-side painting for layout tests incrementally, it'd be convenient to have a positive flag that can be added to each suite as it is ready. Rather than adding some override force flag, just make the presence of enable override disable. The only user of disable is currently content shell. R=piman@chromium.org BUG=381919 Review URL: https://codereview.chromium.org/670253002 Cr-Commit-Position: refs/heads/master@{#300946}
* Enable --site-per-process browser tests on ChromeOS.nasko2014-10-231-4/+4
| | | | | | | | BUG=399775 Review URL: https://codereview.chromium.org/658753005 Cr-Commit-Position: refs/heads/master@{#300913}
* Add DumpAccessibilityTree tests (4 of 20)shreeram.k2014-10-231-0/+28
| | | | | | | | | | | | HTML Tag: input type submit, title, textarea ARIA Role: menubar, definition, tree ARIA state/property: aria-checked BUG=170576 Review URL: https://codereview.chromium.org/661083002 Cr-Commit-Position: refs/heads/master@{#300910}
* Revert "Ensure RenderFrameHostImpl's RenderFrameSetupPtr doesn't die too early"yzshen2014-10-232-13/+5
| | | | | | | | | | | | | | | | This reverts commit 235d012e9d23fbcc456a06496b246fb4e05d98d6. Reason: caused compilation failure when compiling content_browser_test_utils: ../../content/browser/frame_host/render_frame_host_impl.h:22:10: fatal error: 'content/common/render_frame_setup.mojom.h' file not found TBR=blundell@chromium.org BUG= Review URL: https://codereview.chromium.org/672263002 Cr-Commit-Position: refs/heads/master@{#300907}
* Remove IsDIPEnabled from Screenscottmg2014-10-232-10/+7
| | | | | | | | | | | | Appears to be unused. TBR=ben@chromium.org R=sky@chromium.org BUG=143619 Review URL: https://codereview.chromium.org/672823002 Cr-Commit-Position: refs/heads/master@{#300906}
* Web MIDI: distributes MIDIPort information asynchronouslytoyoshim2014-10-232-11/+18
| | | | | | | | | | | | | | | | | | | MidiPortInfoList for input and output devices were accessible from MidiHost, and MidiHost distributed it via IPC on sending MidiMsg_SessionStarted. To update these device lists asynchronously and keep them consistent in all renderers, stop exposing them and invoke a client method, AddInputPort() or AddOutputPort() insteads. These client methods are called for all existing devices on opening a session at once. If MidiManager supports, these can be called for a device on each connection and disconnection. BUG=422333 TEST=media_unittests Review URL: https://codereview.chromium.org/664843002 Cr-Commit-Position: refs/heads/master@{#300894}
* Prepare to enable win32K lockdown mode by default for renderer processes on ↵ananta2014-10-231-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Windows 8+ Win32K lockdown mode is currently enabled by a browser side command line switch "enable_win32k_lockdown_mode". Regular renderers and extension processes work well in the win32k lockdown mode on Windows 8+. The pepper PDF plugin crashes because of the following:- 1. Uses GDI for text rendering. 2. It uses image data functionality which eventually ends up creating DIB sections which fail in the win32k lockdown mode. We already have work underway to run the PDF plugin out of process. This works nicely for us as these processes run as pepper processes and thus don't have the lockdown mode enabled. Changes in this patch are as below:- 1. Propagate the --enable_win32k_lockdown_mode command line switch to renderer processes. This is to enable them to check if they are running in lockdown mode. There is a Windows API to query this. However the switch alleviates the need for that. 2. If the renderer is in win32k lockdown mode, we always use the SIMPLE image data type which creates a SkCanvas backed by memory rather than actual DIB sections. This works well in my testing. I added TODO's in the relevant places to investigate if there is a loss of functionality here. 3. If the --enable_win32k_lockdown_mode switch is specified in the browser we force PDF plugins to run out of process with the --out-of-process-pdf command line switch. This works well in most cases with subtle bugs like pdfs in iframes, not working. This bug is being worked on at the moment. Once the PDF out of process feature is complete we can enable the win32k lockdown mode by default on Windows 8+. BUG=365160 Review URL: https://codereview.chromium.org/669623005 Cr-Commit-Position: refs/heads/master@{#300875}
* [ServiceWorker] Add unittests for ServiceWorkerWriteToCacheJob.horo2014-10-233-0/+302
| | | | | | | | BUG=425396, 414026 Review URL: https://codereview.chromium.org/660253004 Cr-Commit-Position: refs/heads/master@{#300867}
* Ensure RenderFrameHostImpl's RenderFrameSetupPtr doesn't die too earlyblundell2014-10-232-5/+13
| | | | | | | | | | | | | | | | | | | | | | | RenderFrameHostImpl is creating a RenderFrameSetupPtr, binding it via ConnectToRemoteService(), and then using it to bootstrap the per-frame Mojo connection. However, it holds this instance in a local variable, meaning that the pipe is closed when the local variable goes out of scope. The closing of the pipe races with the processing of the message write that is the result of calling RenderFrameSetupPtr->GetServiceProviderForFrame(), as that processing occurs on a different thread. The net outcome is that the per-frame Mojo connection will flakily fail to be set up. To eliminate this race, this CL changes the RenderFrameSetupPtr instance that the RFHI holds from a local variable to a member variable. BUG=424069 TEST=Visit about://omnibox in multiple tabs. In each tab, check that submitting text causes output to appear. Review URL: https://codereview.chromium.org/667683002 Cr-Commit-Position: refs/heads/master@{#300864}
* [DevTools] Migrate DevToolsTracingHandler to generated handler.dgozman2014-10-2311-380/+274
| | | | | | | | | | | This patch also adds SendRawMessage, which is used for performance in tracing handler. BUG=405566 Review URL: https://codereview.chromium.org/635733003 Cr-Commit-Position: refs/heads/master@{#300861}
* Manage MIDI related objects and sessions' lifecycles correctlytoyoshim2014-10-232-12/+24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Following changes are needed to maintain MIDIPortInfoList contents consistent in each MidiMessageFilter instances when MidiManager update the list asynchronously. Essential changes: - Call EndSession iff a session is open. - Do not use client_id to communicate between browser and renderer. Multiple clients are managed in a MidiMessageFilter existing in each renderer. Minor changes: - Remove midi_manager_ checks in MIDIHost, and add CHECK insteads. - Rename MidiManagerFilter::StartSession to MidiManagerFilter::AddClient to be aligned with RemoveClient. A session can be shared with multiple clients. After this change, MidiManager::EndSession is called from each Renderer and RendererHost at the right timing, e.g., GC collects the last MIDIAccess reference in JavaScript. See, also http://crbug.com/424859. BUG=424859 Review URL: https://codereview.chromium.org/662853003 Cr-Commit-Position: refs/heads/master@{#300851}
* Exposes the text input flags to ChromeOS IME extensions.shuchen2014-10-2315-20/+49
| | | | | | | | | BUG=419127 TEST=Verified on linux_chromeos. Review URL: https://codereview.chromium.org/659323003 Cr-Commit-Position: refs/heads/master@{#300820}
* [ServiceWorkerCache] Don't handle response blobs if they don't existjkarlin2014-10-231-1/+2
| | | | | | | | | | Need to add some guards. Clearly the dispatcher needs to have tests which I hope to get to asap. BUG=426150 Review URL: https://codereview.chromium.org/672733002 Cr-Commit-Position: refs/heads/master@{#300817}
* Reenable EncryptedMediaTests on Arm64 Android.xhwang2014-10-221-4/+0
| | | | | | | | | | | | The bug was fixed in Android L release and test devices are updated. This partially reverts r297046. BUG=418039 Review URL: https://codereview.chromium.org/676463002 Cr-Commit-Position: refs/heads/master@{#300792}
* Revert RoutingIDIssuer.morrita2014-10-227-167/+4
| | | | | | | | | | | | | | As the crash is gone, these diagnosis tools are no longer needed. I'd keep a CHECK() in AddRouter() for preventing regressions. TEST=none BUG=381990, 415059 R=nasko@chromium.org TBR=mattm@chromium.org Review URL: https://codereview.chromium.org/662493004 Cr-Commit-Position: refs/heads/master@{#300788}
* Revert of [DevTools] Migrate DevToolsTracingHandler to generated handler. ↵rdevlin.cronin2014-10-2211-274/+380
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | (patchset #6 id:130001 of https://codereview.chromium.org/635733003/) Reason for revert: Broke compile, e.g. obj/content/gen/content/browser/devtools/protocol/content_browser.devtools_protocol_handler_impl.o gen/content/browser/devtools/protocol/devtools_protocol_handler_impl.cc:951:7: error: no viable conversion from 'std::string' (aka 'basic_string<char>') to 'const std::string *' (aka 'const basic_string<char> *') in_categories, ^~~~~~~~~~~~~ ../../content/browser/devtools/protocol/tracing_handler.h:40:26: note: passing argument to parameter 'categories' here const std::string* categories, ^ 1 error generated. ninja: build stopped: subcommand failed. Original issue's description: > [DevTools] Migrate DevToolsTracingHandler to generated handler. > > This patch also adds SendRawMessage, which is used for performance > in tracing handler. > > BUG=405566 > > Committed: https://crrev.com/f7762cb9261c8e2e15f2cf2b36f5408c75d72bdd > Cr-Commit-Position: refs/heads/master@{#300711} TBR=caseq@chromium.org,vkuzkokov@chromium.org,dgozman@chromium.org NOTREECHECKS=true NOTRY=true BUG=405566 Review URL: https://codereview.chromium.org/675473005 Cr-Commit-Position: refs/heads/master@{#300726}
* [ServiceWorkerCache] QuotaClientMask should handle kServiceWorkerCacheIDjkarlin2014-10-221-1/+4
| | | | | | | | | | When removing ServiceWorker data with the QuotaClient we should remove both registrations and cache data. This CL adds the cache data id. BUG=420159 Review URL: https://codereview.chromium.org/672433002 Cr-Commit-Position: refs/heads/master@{#300712}
* [DevTools] Migrate DevToolsTracingHandler to generated handler.dgozman2014-10-2211-380/+274
| | | | | | | | | | | This patch also adds SendRawMessage, which is used for performance in tracing handler. BUG=405566 Review URL: https://codereview.chromium.org/635733003 Cr-Commit-Position: refs/heads/master@{#300711}
* Support the placeholder attribute on Androiddmazzoni2014-10-221-0/+4
| | | | | | | | | | | | | Note: I still plan to refactor these so that more logic is shared between the platforms, but it's nontrivial. In the meantime I think this is worth it as long as it keeps improving the tests, because that will make the refactoring that much less error-prone. BUG=417104 Review URL: https://codereview.chromium.org/665333004 Cr-Commit-Position: refs/heads/master@{#300706}
generated by cgit v1.1 at 2025-03-22 09:14:13 +0000