Rdkemw 9893 enable new dtags 2.38 #1604
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…after M120 upgrade https://bugs.webkit.org/show_bug.cgi?id=267281 rdar://problem/120725671 Reviewed by Eric Carlson. We were sometimes deadlocking when: - main thread asks something to the webrtc stack by hopping to the signalling thread. - signaling thread is destroying the LibWebRTCAudioModule which was calling callOnMainThreadAndWait to do its cleanup. Instead of cleaning things in LibWebRTCAudioModule destructor synchronsouly, we are doing the clean-up asynchronously in LibWebRTCAudioModule::Terminate. Setting m_logTimer to nullptr there is fine since m_logTimer is used in Start/Stop via callOnMainThread. * Source/WebCore/platform/mediastream/libwebrtc/LibWebRTCAudioModule.cpp: (WebCore::LibWebRTCAudioModule::~LibWebRTCAudioModule): (WebCore::LibWebRTCAudioModule::Terminate): * Source/WebCore/platform/mediastream/libwebrtc/LibWebRTCAudioModule.h: Canonical link: https://commits.webkit.org/272828@main
https://bugs.webkit.org/show_bug.cgi?id=265864 Reviewed by Philippe Normand. Also update sources for 271602@main, 271603@main and 271605@main. * Source/ThirdParty/libwebrtc/CMakeLists.txt: * Source/ThirdParty/libwebrtc/Source/webrtc/modules/congestion_controller/goog_cc/loss_based_bwe_v2.cc: Canonical link: https://commits.webkit.org/272175@main
https://bugs.webkit.org/show_bug.cgi?id=265864 Reviewed by Philippe Normand. WebKitGTK and WPE build was broken after the update, it's necessary to update the CMake sources as well as other source files due to API changes. * Source/ThirdParty/libwebrtc/CMakeLists.txt: * Source/WebCore/platform/mediastream/libwebrtc/gstreamer/GStreamerVideoDecoderFactory.cpp: * Source/WebCore/platform/mediastream/libwebrtc/gstreamer/GStreamerVideoEncoderFactory.cpp: (WebCore::GStreamerVideoEncoderFactory::CreateVideoEncoder): Canonical link: https://commits.webkit.org/272416@main
…ase" This reverts commit f0545c4.
Imported graphene version (1.6.0) was from 2017. Library is available through system dependencies.
https://bugs.webkit.org/show_bug.cgi?id=247873 Reviewed by Michael Catanzaro. This fixes several warnings that appear when building for ARMv7 (32-bits): - deprecated-enum-enum-conversion - unused-variable - unused-result - overflow - sign-compare It also disables two warnings for old versions of GCC (old versions give false positives). * PerformanceTests/MallocBench/MallocBench/stress_aligned.cpp: (benchmark_stress_aligned): * Source/JavaScriptCore/assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::revertJumpTo_movT3movtcmpT2): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp12Reg4FourFours): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp12Reg4Reg4Imm12): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp12Reg4Reg4Reg4Imm8): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp12Reg40Imm3Reg4Imm20Imm5): * Source/WebCore/accessibility/atspi/AccessibilityObjectTextAtspi.cpp: (WebCore::AccessibilityObjectAtspi::characterIndex const): (WebCore::AccessibilityObjectAtspi::selectionChanged): * Source/WebCore/platform/graphics/filters/software/FECompositeSoftwareApplier.cpp: (WebCore::FECompositeSoftwareApplier::applyArithmetic const): * Source/WebCore/platform/graphics/gbm/DMABufReleaseFlag.h: (WebCore::DMABufReleaseFlag::release): * Source/cmake/WebKitCompilerFlags.cmake: Canonical link: https://commits.webkit.org/256792@main
Fix error while building Cog: Run-time dependency wayland-protocols found: YES 1.12 Tools/wpe/cog/platform/wayland/meson.build:94:12: ERROR: Problem encountered: Cannot find protocol text-input-unstable-v3, file does not exist: WPEWebKit/WebKitBuild/DependenciesWPE/Root/share/wayland-protocols/unstable/text-input/text-input-unstable-v3.xml
…edded/wpe-2.38-libwebrtc-m120 Wpe 2.38 libwebrtc m120
…hen not handled https://bugs.webkit.org/show_bug.cgi?id=271041 Reviewed by Philippe Normand. We may wish to use GST_EVENT_CUSTOM_DOWNSTREAM_OOB event for use cases beyond signaling an "attempt-to-decrypt". Patch by Filipe Norte <filipe.norte@sky.uk>. * Source/WebCore/platform/graphics/gstreamer/eme/WebKitCommonEncryptionDecryptorGStreamer.cpp: (sinkEventHandler): Canonical link: https://commits.webkit.org/276144@main
…p2-build-fix Fix build with glib < 2.68
…8-fix-hole-punching Disable blending of hole punching buffers with rounded rect clip
https://bugs.webkit.org/show_bug.cgi?id=271696 Reviewed by Adrian Perez de Castro. Based on original patch by: Eugene Mutavchi <Ievgen_Mutavchi@comcast.com> * Source/ThirdParty/libwebrtc/CMakeLists.txt: * Source/ThirdParty/libwebrtc/cmake/FindLibPulse.cmake: Added. Canonical link: https://commits.webkit.org/276708@main
https://bugs.webkit.org/show_bug.cgi?id=271696 CMake was failing because there was an attempt to set 'webrtc' when it was not yet defined. Move faulting lines below, once 'webrtc' is already defined. * Source/ThirdParty/libwebrtc/CMakeLists.txt: Canonical link: https://commits.webkit.org/276724@main
https://bugs.webkit.org/show_bug.cgi?id=269287 Reviewed by NOBODY (OOPS!). Compile-time quirks prevent re-usability of the same binaries across different platforms (using containers). The proposed solution is to always compile those, they don't depend on any specific APIs so that shouldn't be an issue. Then at runtime quirks can be tweaked using the WEBKIT_GST_QUIRKS and WEBKIT_GST_HOLE_PUNCH environment variables. * Source/WebCore/platform/SourcesGStreamer.txt: * Source/WebCore/platform/audio/gstreamer/AudioDestinationGStreamer.cpp: (WebCore::AudioDestinationGStreamer::AudioDestinationGStreamer): * Source/WebCore/platform/audio/gstreamer/AudioFileReaderGStreamer.cpp: (WebCore::decodebinAutoplugSelectCallback): (WebCore::AudioFileReader::~AudioFileReader): (WebCore::AudioFileReader::decodeAudioForBusCreation): * Source/WebCore/platform/graphics/gstreamer/GStreamerCommon.cpp: (WebCore::ensureGStreamerInitialized): (WebCore::registerWebKitGStreamerElements): * Source/WebCore/platform/graphics/gstreamer/GStreamerRegistryScanner.cpp: (WebCore::GStreamerRegistryScanner::ElementFactories::ElementFactories): (WebCore::GStreamerRegistryScanner::ElementFactories::hasElementForCaps const): * Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp: (WebCore::MediaPlayerPrivateGStreamer::MediaPlayerPrivateGStreamer): (WebCore::MediaPlayerPrivateGStreamer::~MediaPlayerPrivateGStreamer): (WebCore::setSyncOnClock): (WebCore::MediaPlayerPrivateGStreamer::createAudioSink): (WebCore::MediaPlayerPrivateGStreamer::gstreamerPositionFromSinks const): (WebCore::MediaPlayerPrivateGStreamer::naturalSize const): (WebCore::MediaPlayerPrivateGStreamer::handleMessage): (WebCore::MediaPlayerPrivateGStreamer::configureElement): (WebCore::MediaPlayerPrivateGStreamer::configureElementPlatformQuirks): (WebCore::MediaPlayerPrivateGStreamer::createGSTPlayBin): (WebCore::MediaPlayerPrivateGStreamer::swapBuffersIfNeeded): (WebCore::setRectangleToVideoSink): (WebCore::MediaPlayerPrivateGStreamer::isHolePunchRenderingEnabled const): (WebCore::MediaPlayerPrivateGStreamer::createHolePunchVideoSink): (WebCore::MediaPlayerPrivateGStreamer::shouldIgnoreIntrinsicSize): (WebCore::MediaPlayerPrivateGStreamer::createVideoSink): * Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h: * Source/WebCore/platform/gstreamer/GStreamerHolePunchQuirkBcmNexus.cpp: Added. (WebCore::GStreamerHolePunchQuirkBcmNexus::setHolePunchVideoRectangle): * Source/WebCore/platform/gstreamer/GStreamerHolePunchQuirkBcmNexus.h: Added. * Source/WebCore/platform/gstreamer/GStreamerHolePunchQuirkWesteros.cpp: Added. (WebCore::GStreamerHolePunchQuirkWesteros::createHolePunchVideoSink): (WebCore::GStreamerHolePunchQuirkWesteros::setHolePunchVideoRectangle): * Source/WebCore/platform/gstreamer/GStreamerHolePunchQuirkWesteros.h: Added. * Source/WebCore/platform/gstreamer/GStreamerQuirkAmLogic.cpp: Added. (WebCore::GStreamerQuirkAmLogic::GStreamerQuirkAmLogic): (WebCore::GStreamerQuirkAmLogic::createWebAudioSink): (WebCore::GStreamerQuirkAmLogic::configureElement): * Source/WebCore/platform/gstreamer/GStreamerQuirkAmLogic.h: Added. * Source/WebCore/platform/gstreamer/GStreamerQuirkBcmNexus.cpp: Added. (WebCore::GStreamerQuirkBcmNexus::GStreamerQuirkBcmNexus): (WebCore::GStreamerQuirkBcmNexus::isHardwareAccelerated): * Source/WebCore/platform/gstreamer/GStreamerQuirkBcmNexus.h: Added. * Source/WebCore/platform/gstreamer/GStreamerQuirkBroadcom.cpp: Added. (WebCore::GStreamerQuirkBroadcom::GStreamerQuirkBroadcom): (WebCore::GStreamerQuirkBroadcom::configureElement): (WebCore::GStreamerQuirkBroadcom::isHardwareAccelerated): * Source/WebCore/platform/gstreamer/GStreamerQuirkBroadcom.h: Added. * Source/WebCore/platform/gstreamer/GStreamerQuirkRealtek.cpp: Added. (WebCore::GStreamerQuirkRealtek::GStreamerQuirkRealtek): (WebCore::GStreamerQuirkRealtek::createWebAudioSink): (WebCore::GStreamerQuirkRealtek::configureElement): (WebCore::GStreamerQuirkRealtek::isHardwareAccelerated): * Source/WebCore/platform/gstreamer/GStreamerQuirkRealtek.h: Added. * Source/WebCore/platform/gstreamer/GStreamerQuirkWesteros.cpp: Added. (WebCore::GStreamerQuirkWesteros::GStreamerQuirkWesteros): (WebCore::GStreamerQuirkWesteros::configureElement): (WebCore::GStreamerQuirkWesteros::isHardwareAccelerated): * Source/WebCore/platform/gstreamer/GStreamerQuirkWesteros.h: Added. * Source/WebCore/platform/gstreamer/GStreamerQuirks.cpp: Added. (WebCore::GStreamerQuirksManager::singleton): (WebCore::GStreamerQuirksManager::GStreamerQuirksManager): (WebCore::GStreamerQuirksManager::isEnabled const): (WebCore::GStreamerQuirksManager::createWebAudioSink): (WebCore::GStreamerQuirksManager::createHolePunchVideoSink): (WebCore::GStreamerQuirksManager::setHolePunchVideoRectangle): (WebCore::GStreamerQuirksManager::configureElement): (WebCore::GStreamerQuirksManager::isHardwareAccelerated const): (WebCore::GStreamerQuirksManager::supportsVideoHolePunchRendering const): (WebCore::GStreamerQuirksManager::audioVideoDecoderFactoryListType const): (WebCore::GStreamerQuirksManager::disallowedWebAudioDecoders const): * Source/WebCore/platform/gstreamer/GStreamerQuirks.h: Added. (WebCore::GStreamerQuirk::isPlatformSupported const): (WebCore::GStreamerQuirk::createWebAudioSink): (WebCore::GStreamerQuirk::configureElement): (WebCore::GStreamerQuirk::isHardwareAccelerated): (WebCore::GStreamerQuirk::audioVideoDecoderFactoryListType const): (WebCore::GStreamerQuirk::disallowedWebAudioDecoders const): (WebCore::GStreamerHolePunchQuirk::createHolePunchVideoSink): (WebCore::GStreamerHolePunchQuirk::setHolePunchVideoRectangle): * Source/cmake/OptionsWPE.cmake:
…rceDataMap after content eviction https://bugs.webkit.org/show_bug.cgi?id=290161 Reviewed by Devin Rousso. After eviction of the content of the resource data due to exceeding the threshold for maximum memory usage for gathered resources, WebKit still keeps the structure of resource data in `m_requestIdToResourceDataMap` because it is the primary structure used to track all other metadata used by Web Inspector. We can minimize the data kept there after eviction by resetting the decoder which is not needed if we remove the content of resource data. Additionally in case of platforms different than Cocoa we can skip storing the certificate info because it is not used. The certificate info is used only if supportsShowCertificate(), which is true only in case of Cocoa platform. This change adds a web preference `InspectorSupportsShowingCertificate` which is true only in case of Cocoa and Win platform and enabling storing of the certificate info for resource data in Web Inspector. * Source/WTF/Scripts/Preferences/UnifiedWebPreferences.yaml: * Source/WebCore/inspector/InspectorFrontendHost.cpp: (WebCore::InspectorFrontendHost::supportsShowCertificate const): * Source/WebCore/inspector/NetworkResourcesData.cpp: (WebCore::NetworkResourcesData::ResourceData::evictContent): (WebCore::NetworkResourcesData::NetworkResourcesData): (WebCore::NetworkResourcesData::responseReceived): (WebCore::NetworkResourcesData::setResourceContent): (WebCore::NetworkResourcesData::maybeAddResourceData): (WebCore::NetworkResourcesData::maybeDecodeDataToContent): (WebCore::NetworkResourcesData::ensureFreeSpace): * Source/WebCore/inspector/NetworkResourcesData.h: (WebCore::NetworkResourcesData::Settings::Settings): * Source/WebCore/inspector/agents/InspectorNetworkAgent.cpp: (WebCore::InspectorNetworkAgent::InspectorNetworkAgent): * Source/WebCore/inspector/agents/InspectorNetworkAgent.h: * Source/WebCore/inspector/agents/page/PageNetworkAgent.cpp: (WebCore::PageNetworkAgent::PageNetworkAgent): * Source/WebCore/inspector/agents/worker/WorkerNetworkAgent.cpp: (WebCore::WorkerNetworkAgent::WorkerNetworkAgent): Canonical link: https://commits.webkit.org/293535@main
…://bugs.webkit.org/show_bug.cgi?id=290750 Reviewed by Sihui Liu. Dispatch an error event "canceled" for queued utterances. Specification: https://dvcs.w3.org/hg/speech-api/raw-file/tip/webspeechapi, point: 5.2.5.1. Original author: Andrzej Surdej <Andrzej_Surdej@comcast.com> See: WebPlatformForEmbedded#1474 * LayoutTests/fast/speechsynthesis/speech-synthesis-cancel-expected.txt: * LayoutTests/fast/speechsynthesis/speech-synthesis-cancel.html: * Source/WebCore/Modules/speech/SpeechSynthesis.cpp: (WebCore::SpeechSynthesis::cancel): Canonical link: https://commits.webkit.org/293536@main
…edded/pgorszkowski/2.38/WebInspector-minimize-the-number-of-data-kept-in-m_requestIdToResourceDataMap-after-content-eviction [WebInspector] minimize the number of data kept in m_requestIdToResou…
…edded/pgorszkowski/2.38/SpeechSynthesis-Emit-error-event-for-all-cancelled-utterances [SpeechSynthesis] Emit error event for all cancelled utterances
https://bugs.webkit.org/show_bug.cgi?id=290446 Reviewed by Michael Catanzaro. To cancel all pending requests before network process exits, we have to abort the current soup session. This will cancell all pending cancellable task in glib-network backends (e.g.: OpenSSL, GnuTLS). Original author: Volodymyr Ogorodnik() See: WebPlatformForEmbedded#1482 * Source/WebCore/platform/network/soup/SoupNetworkSession.cpp: (WebCore::SoupNetworkSession::~SoupNetworkSession): Canonical link: https://commits.webkit.org/294000@main
…edded/pgorszkowski/2.38/SOUP-WPENetworkProcess-crashed-on-process-exit [SOUP] WPENetworkProcess crashed on process exit
This is to save the memory allocated for swapchain surfaces when wpe is going invisible. Only active with RESIZE_WAYLAND_SURFACES_ON_SUSPEND_PAINTING compilation var enabled, and with non-composite webgl disabled.
…rczewski/2.38/ThreadedCompositor_resize_wayland_surfaces_at_suspend_painting ThreadedCompositor: resize wayland surfaces at suspend painting
https://bugs.webkit.org/show_bug.cgi?id=291414 Reviewed by Alicia Boya Garcia. If an MP3 with ID3 tags is appended to a SourceBuffer, AppendPipeline's typefind detects it as application/x-id3 instead of audio/mpeg and can't process it. This commit takes care of that case by (optionally) creating an id3demux that takes care of the ID3 tags (just ignores them) and produces audio/mpeg at its output, so the processing can continue normally. We can't know if id3demux can be created or not at AppendPipeline instantiation time. In case the SourceBuffer mimetype is audio/mpeg, we need to listen to the have-type signal from typefind and defer the creation, configuration and connection of the demuxer (or identity, if id3demux is not needed) until the signal happens. The code to configure the demuxer has been refactored into its own method, so it can be reused from the regular and the deferred demuxer configuration codepaths. * LayoutTests/media/content/silence_with_id3_tags.mp3: Added. * LayoutTests/media/media-source/media-source-play-mp3-id3-tag-expected.txt: Added. * LayoutTests/media/media-source/media-source-play-mp3-id3-tag.html: Added. * Source/WebCore/platform/graphics/gstreamer/mse/AppendPipeline.cpp: (WebCore::AppendPipeline::configureOptionalDemuxerFromAnyThread): Refactored code to configure the demuxer. The configuration code has been enhanced to detect the case of demuxers with static pads (such as id3demux) and treat them in the same immediate configuration way as identity. (WebCore::AppendPipeline::AppendPipeline): If the SourceBuffer mimetype is audio/mpeg, listen to have-type and create and configure the demuxer/identity there (just the first time the signal is triggered). * Source/WebCore/platform/graphics/gstreamer/mse/AppendPipeline.h: Added new configureOptionalDemuxerFromAnyThread() private method. Canonical link: https://commits.webkit.org/294106@main
Adopt the created cairo pattern so that it is properly managed.
…edded/spena/wpe-2.38-adopt-pattern [Cairo] Fix cairo pattern memory leak
…kit.org/show_bug.cgi?id=266973 Reviewed by Michael Catanzaro, Carlos Garcia Campos and Xabier Rodriguez-Calvar. Store the WebKitWebSrc pointer as a weak reference in the CachedResourceStreamingClient. Using a const raw pointer was not sufficient, because nothing prevented other code to dispose the element, thus making the pointer point to garbage. We can't use a strong reference because it would introduce a reference cycle between WebKitWebSrc and the CachedResourceStreamingClient. * Source/WebCore/platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp: (webKitWebSrcMakeRequest): (CachedResourceStreamingClient::CachedResourceStreamingClient): (CachedResourceStreamingClient::~CachedResourceStreamingClient): (CachedResourceStreamingClient::checkUpdateBlocksize): (CachedResourceStreamingClient::responseReceived): (CachedResourceStreamingClient::dataReceived): (CachedResourceStreamingClient::accessControlCheckFailed): (CachedResourceStreamingClient::loadFailed): (CachedResourceStreamingClient::loadFinished): Canonical link: https://commits.webkit.org/272807@main Signed-off-by: GlebNovodranPE <gnovodran@productengine.com>
https://bugs.webkit.org/show_bug.cgi?id=296690 Reviewed by Philippe Normand. OpenCDM/Thunder can answer with a null message when loading a wrong session. We need to deal with that. * Source/WebCore/platform/graphics/gstreamer/eme/CDMThunder.cpp: (WebCore::CDMInstanceSessionThunder::loadSession): Canonical link: https://commits.webkit.org/298032@main
https://bugs.webkit.org/show_bug.cgi?id=243796 <rdar://98471033> Reviewed by Eric Carlson. We should be using the `detachMediaSource` method in `setSrcObject` in order to be consistent with the rest of the `HTMLMediaElement` implementation. * Source/WebCore/html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::setSrcObject): Canonical link: https://commits.webkit.org/253314@main
https://bugs.webkit.org/show_bug.cgi?id=300041 Reviewed by Philippe Normand. Sometimes, in a downstream multimedia player, regular video playback remains paused after seek instead of resuming automatically. It will play normally when play() or another seek operation are manually triggered. Log analys shows that in the failing case the browser missed the state transition to 'HaveCurrentData', and application can't resume playback after seek. It was missed because it happened the gstreamer pipeline remains in state transition for too long and can't process buffering events. Normally it goes: seek -> Paused pipeline -> buffering (HaveCurrentData) -> finish_buffering(HaveEnoughData) -> start_playback. However, in the failing case, seek takes too long (~2sec) to set the PAUSED state on the pipeline and we are missing the buffering(HaveCurrentData) stage. See: WebPlatformForEmbedded#1561 This patch returns the player readyState to HaveCurrentData and the networkState to Loading when a seek is being processed. The states will reach their final values as soon as the async state transition caused by the seek finishes. Original author: Volodymyr Ogorodnik (https://github.com/volodymyr-ogorodnik-red) * Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp: (WebCore::MediaPlayerPrivateGStreamer::updateStates): Reset readyState and networkState to HaveCurrentData and Loading while the player is in an async state transition (in a seek) and buffering. Canonical link: https://commits.webkit.org/300933@main
…g on 32 bits. https://bugs.webkit.org/show_bug.cgi?id=293720 Reviewed by Yusuke Suzuki. On 32-bit armv7/linux we run into an issue where the environment strings located at the bottom of the stack, as well as some random bits inside the libc portion of the stack are interpreted as cells pointing into the heap, keeping actually dead objects alive. Yusuke previously attempted to fix this by excluding environ and before, but there are additional fake roots only a few hundred bytes below that. This patch excludes the caller of main entirely. * Source/JavaScriptCore/jsc.cpp: (jscmain): * Source/JavaScriptCore/runtime/JSLock.cpp: (JSC::JSLock::didAcquireLock): * Source/WTF/wtf/StackBounds.cpp: (WTF::StackBounds::setBottomOfMainThreadMain): (WTF::StackBounds::currentThreadStackBoundsInternal): * Source/WTF/wtf/StackBounds.h: (WTF::StackBounds::checkConsistency const): * Source/WebKit/WebProcess/gtk/WebProcessMainGtk.cpp: (WebKit::WebProcessMain): * Source/WebKit/WebProcess/wpe/WebProcessMainWPE.cpp: (WebKit::WebProcessMain):
…/stack-scanning-backport Exclude non-user portions of the main thread stack from stack scanning on 32 bits.
https://bugs.webkit.org/show_bug.cgi?id=303463 Reviewed by Philippe Normand. We were missing to clear up the session and update and clear the keys. Original patch by Santhosh-Kumar-infosys <sakumar.ext@libertyglobal.com>. * Source/WebCore/platform/graphics/gstreamer/eme/CDMThunder.cpp: (WebCore::CDMInstanceSessionThunder::sessionChanged): (WebCore::CDMInstanceSessionThunder::updateLicense): (WebCore::CDMInstanceSessionThunder::loadSession): (WebCore::CDMInstanceSessionThunder::removeSessionData): (WebCore::CDMInstanceSessionThunder::sessionFailure): Deleted. * Source/WebCore/platform/graphics/gstreamer/eme/CDMThunder.h: Canonical link: https://commits.webkit.org/303826@main This fixes WebPlatformForEmbedded#1585.
https://bugs.webkit.org/show_bug.cgi?id=303326 Reviewed by Philippe Normand. GStreamerQuirksManager requires GStreamer to be initialized, which does not happen soon enough if neither canPlayType() nor isTypeSupported() has been called before, so in those cases no quirks are applied. However, the quirks manager may be used in other processes, e.g. NetworkProcess when determining support for a MIME type. To avoid issues, we should only fully init GStreamer if on the WebProcess, and otherwise go for a minimal initialization. * Source/WebCore/platform/graphics/gstreamer/GStreamerCommon.cpp: (WebCore::ensureGStreamerInitializedNonWebProcess): Added as a counterpart to ensureGStreamerInitialized, only does minimal init. * Source/WebCore/platform/graphics/gstreamer/GStreamerCommon.h: * Source/WebCore/platform/graphics/gstreamer/GStreamerRegistryScanner.cpp: (WebCore::GStreamerRegistryScanner::GStreamerRegistryScanner): Use the new function to initialize gst if in the UIProcess * Source/WebCore/platform/gstreamer/GStreamerQuirks.cpp: (WebCore::GStreamerQuirksManager::GStreamerQuirksManager): Init GStreamer using either of the two methods Original author: Andrzej Surdej <Andrzej_Surdej@comcast.com> See: WebPlatformForEmbedded#1584 Canonical link: https://commits.webkit.org/303982@main
…agehide https://bugs.webkit.org/show_bug.cgi?id=303619 Reviewed by Alexey Proskuryakov. When the WebSocket is suspending and it was closed, or it is in the middle of closing, it does not need to be closed separately (and dispatch the error event). There is already WPT test for that case. Now the test passes. * LayoutTests/imported/w3c/web-platform-tests/websockets/back-forward-cache-with-open-websocket-connection-and-close-it-in-pagehide.window-expected.txt: * Source/WebCore/Modules/websockets/WebSocket.cpp: (WebCore::WebSocket::suspend): Canonical link: https://commits.webkit.org/304082@main
…Quirks" This reverts commit 4bdbd3c.
https://bugs.webkit.org/show_bug.cgi?id=303326 Reviewed by Philippe Normand. GStreamerQuirksManager requires GStreamer to be initialized, which does not happen soon enough if neither canPlayType() nor isTypeSupported() has been called before, so in those cases no quirks are applied. However, the quirks manager may be used in other processes, e.g. NetworkProcess when determining support for a MIME type. To avoid issues, we should only fully init GStreamer if on the WebProcess, and otherwise go for a minimal initialization. * Source/WebCore/platform/graphics/gstreamer/GStreamerCommon.cpp: (WebCore::ensureGStreamerInitializedNonWebProcess): Added as a counterpart to ensureGStreamerInitialized, only does minimal init. * Source/WebCore/platform/graphics/gstreamer/GStreamerCommon.h: * Source/WebCore/platform/graphics/gstreamer/GStreamerRegistryScanner.cpp: (WebCore::GStreamerRegistryScanner::GStreamerRegistryScanner): Use the new function to initialize gst if in the UIProcess * Source/WebCore/platform/gstreamer/GStreamerQuirks.cpp: (WebCore::GStreamerQuirksManager::GStreamerQuirksManager): Init GStreamer using either of the two methods Original author: Andrzej Surdej <Andrzej_Surdej@comcast.com> See: WebPlatformForEmbedded#1584 Canonical link: https://commits.webkit.org/303982@main
https://bugs.webkit.org/show_bug.cgi?id=305473 Reviewed by Xabier Rodriguez-Calvar. Sometimes applications create a <video> tag with 'preload="none"' and then remove the element from the DOM without ever starting playback. This results in the video element being leaked, because HTMLMediaElement::virtualHasPendingActivity() retains the element while its network state is still "Loading". See: WebPlatformForEmbedded#1594 This patch keeps the default behaviour of changing the NetworkState to "Loading", except in the case of a delayed load (the case triggered by preload="none"), in which it's set to Idle and therefore allowing the full destruction of the video element. Original author: Eugene Mutavchi <Ievgen_Mutavchi@comcast.com> * Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp: (WebCore::MediaPlayerPrivateGStreamer::load): Reset NetworkState to Idle in case of delayed load. Set to Loading otherwise. Canonical link: https://commits.webkit.org/305578@main
|
@lulicdarko , do you maybe have an isolated patch or applied on 2.38 branch? This PR seems bringing all 2.46 changes to 2.38. If you do not have isolated environment, no problem the other PR is good enough to introduce the change on 2.38 as well. |
Author
Hello @modeveci you can find isolated pull request here #1605 |
|
Clearly this is an invalid PR so I'm closing it. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Added option to enable new dtags.
Explanation:
This change introduces a Cmake option ENABLE_NEW_DTAGS to allow to use --enable-new-dtags for WPE Webkit build.
Primary goal is to force the linker to use DT_RUNPATH instead of legacy DT_RPATH. This change is important because:
Changes: