RESOLVED FIXED 238887
Have ImageBuffer store the RenderingPurpose, and send it to the GPU process 
https://bugs.webkit.org/show_bug.cgi?id=238887
Summary Have ImageBuffer store the RenderingPurpose, and send it to the GPU process
Simon Fraser (smfr)
Reported 2022-04-06 13:47:36 PDT
Have ImageBuffer store the RenderingPurpose, and send it to the GPU process
Attachments
Patch (37.13 KB, patch)
2022-04-06 13:50 PDT, Simon Fraser (smfr) 		sabouhallawa: review+
ews-feeder: commit-queue-
DetailsFormatted DiffDiff
View All Add attachment proposed patch, testcase, etc.
Simon Fraser (smfr)
Comment 1 2022-04-06 13:50:23 PDT
Created attachment 456857 [details] Patch
Said Abou-Hallawa
Comment 2 2022-04-06 14:21:01 PDT
Comment on attachment 456857 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=456857&action=review > Source/WebKit/WebProcess/GPU/graphics/RemoteDisplayListRecorderProxy.cpp:452 > + RenderingPurpose purpose = m_imageBuffer ? m_imageBuffer->renderingPurpose() : RenderingPurpose::DOM; I could not understand what this statement means.When can m_imageBuffer be null? And why do we use m_imageBuffer->renderingPurpose() or RenderingPurpose::DOM as the renderingPurpose of the new compatible ImageBuffer? Suppose RemoteDisplayListRecorderProxy was created for RenderingPurpose::Canvas and we are creating a compatible ImageBuffer for SVG clipper, why is this ImageBuffer created with RenderingPurpose::Canvas?
Simon Fraser (smfr)
Comment 3 2022-04-07 13:15:32 PDT
I tried plumbing RenderingPurpose through more but it got out of hand. Landing the patch with a FIXME. https://trac.webkit.org/changeset/292557/webkit
Radar WebKit Bug Importer
Comment 4 2022-04-07 13:16:17 PDT
<rdar://problem/91440114>
Note You need to log in before you can comment on or make changes to this bug.