Bug 208727

Summary: [GPUP] Move AVSystemController code into the GPU process
Product: WebKit Reporter: Jer Noble <jer.noble>
Component: New BugsAssignee: Jer Noble <jer.noble>
Status: RESOLVED FIXED    
Severity: Normal CC: commit-queue, eric.carlson, ews-watchlist, glenn, jonlee, peng.liu6, philipj, sergio, webkit-bug-importer, youennf
Priority: P2 Keywords: InRadar
Version: WebKit Nightly Build   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Patch
none
Patch
none
Patch
eric.carlson: review+
Patch for landing
none
Patch for landing
none
Patch for landing
none
Patch for landing jer.noble: commit-queue+

Jer Noble
Reported 2020-03-06 11:42:17 PST
[GPUP] Move AVSystemController code into the GPU process
Attachments
Patch (217.67 KB, patch)
2020-03-06 12:10 PST, Jer Noble
no flags
Patch (54.45 KB, patch)
2020-03-06 12:17 PST, Jer Noble
no flags
Patch (104.98 KB, patch)
2020-03-06 12:35 PST, Jer Noble
eric.carlson: review+
Patch for landing (104.85 KB, patch)
2020-03-06 16:32 PST, Jer Noble
no flags
Patch for landing (104.89 KB, patch)
2020-03-07 08:49 PST, Jer Noble
no flags
Patch for landing (105.04 KB, patch)
2020-03-08 10:13 PDT, Jer Noble
no flags
Patch for landing (106.56 KB, patch)
2020-03-08 10:57 PDT, Jer Noble
jer.noble: commit-queue+
Jer Noble
Comment 1 2020-03-06 12:10:17 PST
Jer Noble
Comment 2 2020-03-06 12:17:51 PST
Jer Noble
Comment 3 2020-03-06 12:35:36 PST
Eric Carlson
Comment 4 2020-03-06 15:54:04 PST
Comment on attachment 392755 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=392755&action=review > Source/WebCore/platform/audio/ios/MediaSessionHelperIOS.mm:72 > +#if HAVE(MEDIA_PLAYER) && !PLATFORM(WATCHOS) HAVE_MEDIA_PLAYER is always defined for PLATFORM(COCOA) > Source/WebCore/platform/audio/ios/MediaSessionHelperIOS.mm:88 > +#if HAVE(MEDIA_PLAYER) && !PLATFORM(WATCHOS) Ditto. > Source/WebCore/platform/audio/ios/MediaSessionHelperIOS.mm:183 > + m_objcObserver = nil; Nit: this isn't necessary since it is a RetainPtr > Source/WebCore/platform/audio/ios/MediaSessionHelperIOS.mm:361 > +#if HAVE(MEDIA_PLAYER) && !PLATFORM(WATCHOS) HAVE_MEDIA_PLAYER is always defined for PLATFORM(COCOA) > Source/WebCore/platform/audio/ios/MediaSessionHelperIOS.mm:387 > +#if HAVE(MEDIA_PLAYER) && !PLATFORM(WATCHOS) Ditto
Jer Noble
Comment 5 2020-03-06 16:32:34 PST
Created attachment 392803 [details] Patch for landing
Radar WebKit Bug Importer
Comment 6 2020-03-06 18:01:59 PST
Jer Noble
Comment 7 2020-03-07 08:41:02 PST
Won't land until I fix those test failures.
Jer Noble
Comment 8 2020-03-07 08:49:44 PST
Created attachment 392860 [details] Patch for landing
Jer Noble
Comment 9 2020-03-08 10:13:41 PDT
Created attachment 392958 [details] Patch for landing
Jer Noble
Comment 10 2020-03-08 10:57:16 PDT
Created attachment 392961 [details] Patch for landing
Jer Noble
Comment 11 2020-03-08 12:04:56 PDT
Layout test failure (fast/hidpi/image-srcset-relative-svg-canvas-2x.html) is a failure already in the tree (image diff by 0.01%), and is unrelated. Landing.
Jer Noble
Comment 12 2020-03-08 12:22:48 PDT
youenn fablet
Comment 13 2020-03-09 02:34:49 PDT
We register two applicationWillEnterForeground for get_UIKit_UIApplicationWillEnterForegroundNotification and WebUIApplicationWillEnterForegroundNotification. Ditto for applicationDidBecomeActive, applicationWillResignActive and applicationDidEnterBackground. Given we post a WebUIApplicationWillEnterForegroundNotification notification from WebProcess/WebPage, shouldn't we listen to this notification from WebProcess only? Maybe we do not need WebUIApplicationWillEnterForegroundNotification anymore and can rely on get_UIKit_UIApplicationWillEnterForegroundNotification only? Ditto for others? Also, we are listening for UIKit_UIApplicationWillEnterForegroundNotification in both UIProcess and GPUProcess. Do we need both? If we need both, should we try to have a single code path (GPUProcess -> UIProcess -> WebProcess for instance)?
Note You need to log in before you can comment on or make changes to this bug.