RESOLVED FIXED 208370
LocalSampleBufferDisplayLayer is not computing the correct bounds/position in GPUProcess 
https://bugs.webkit.org/show_bug.cgi?id=208370
Summary LocalSampleBufferDisplayLayer is not computing the correct bounds/position in...
youenn fablet
Reported 2020-02-28 03:20:59 PST
LocalSampleBufferDisplayLayer is not computing the correct bounds/position in GPUProcess
Attachments
Patch (17.66 KB, patch)
2020-02-28 04:17 PST, youenn fablet 		no flags
DetailsFormatted DiffDiff
Patch (16.43 KB, patch)
2020-02-28 05:16 PST, youenn fablet 		no flags
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
youenn fablet
Comment 1 2020-02-28 04:17:41 PST
Created attachment 391972 [details] Patch
youenn fablet
Comment 2 2020-02-28 05:16:21 PST
Created attachment 391974 [details] Patch
Eric Carlson
Comment 3 2020-02-28 09:12:44 PST
Comment on attachment 391974 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=391974&action=review > Source/WebKit/WebProcess/GPU/webrtc/SampleBufferDisplayLayer.cpp:59 > + if (!m_connection->sendSync(Messages::RemoteSampleBufferDisplayLayerManager::CreateLayer { m_identifier, hideRootLayer, size }, Messages::RemoteSampleBufferDisplayLayerManager::CreateLayer::Reply { contextId }, 0)) Could this be made async by allocating a local CALayer for m_videoLayer and inserting the CALayerHost layer as a child once you have the remote context ID?
youenn fablet
Comment 4 2020-02-28 10:07:44 PST
(In reply to Eric Carlson from comment #3) > Comment on attachment 391974 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=391974&action=review > > > Source/WebKit/WebProcess/GPU/webrtc/SampleBufferDisplayLayer.cpp:59 > > + if (!m_connection->sendSync(Messages::RemoteSampleBufferDisplayLayerManager::CreateLayer { m_identifier, hideRootLayer, size }, Messages::RemoteSampleBufferDisplayLayerManager::CreateLayer::Reply { contextId }, 0)) > > Could this be made async by allocating a local CALayer for m_videoLayer and > inserting the CALayerHost layer as a child once you have the remote context > ID? We should fix this at some point. Your approach sounds good. We could also make this async at WebCore level.
WebKit Commit Bot
Comment 5 2020-03-02 07:34:11 PST
Comment on attachment 391974 [details] Patch Clearing flags on attachment: 391974 Committed r257709: <https://trac.webkit.org/changeset/257709>
WebKit Commit Bot
Comment 6 2020-03-02 07:34:13 PST
All reviewed patches have been landed. Closing bug.
Radar WebKit Bug Importer
Comment 7 2020-03-02 07:35:16 PST
<rdar://problem/59950161>
Note You need to log in before you can comment on or make changes to this bug.