[Webkit-unassigned] [Bug 234008] New: WebGL framebuffer performance regression

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Dec 8 07:46:09 PST 2021


https://bugs.webkit.org/show_bug.cgi?id=234008

            Bug ID: 234008
           Summary: WebGL framebuffer performance regression
           Product: WebKit
           Version: Safari 15
          Hardware: iPhone / iPad
                OS: iOS 15
            Status: NEW
          Severity: Major
          Priority: P2
         Component: WebGL
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: jekfer.bichon at 3ds.com
                CC: dino at apple.com, kbr at google.com, kkinnunen at apple.com

Created attachment 446365

  --> https://bugs.webkit.org/attachment.cgi?id=446365&action=review

Sample to reproduce

Hello,

We have detected a performance regression while manipulating framebuffers. I have attached a sample where I bind and clear framebuffers that are created once. The regression has been detected on an iPad Gen 8.

We were getting a comfortable 60 fps on the sample on Safari 14 while we are barely hitting 21 fps on Safari 15. Disabling the experimental option "WebGL via Metal" on Safari 15 puts the performances back at Safari 14 level.

Thank you.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20211208/6f249e8a/attachment.htm>


More information about the webkit-unassigned mailing list