[Webkit-unassigned] [Bug 284281] New: WebGL2 context stays in memory

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Dec 9 01:07:02 PST 2024


https://bugs.webkit.org/show_bug.cgi?id=284281

            Bug ID: 284281
           Summary: WebGL2 context stays in memory
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Canvas
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: jvdheijden90 at gmail.com
                CC: sabouhallawa at apple.com

Created attachment 473508

  --> https://bugs.webkit.org/attachment.cgi?id=473508&action=review

Graphics tab recording of Threejs examples

When debugging an issue within my own project I noticed a lot of canvasses stayed in memory in Safari. I clean up most of the canvasses used in this project by setting them to a size of 1 by 1 so they won't allocate a lot of memory, eventually these also will get cleaned up by the garbage collector. Like described in this article: https://pqina.nl/blog/total-canvas-memory-use-exceeds-the-maximum-limit/

I noticed the canvas that was attached to the DOM using WebGL2 is also staying in memory. It stays visible in the Graphics tab in Safari and it won't be garbage collected. We have cases in our website where iOS users get a crash, meaning the current browser tab will refresh. This seem to only happen on an iPhone, not on a Macbook and not on an iPhone simulator as far as I could reproduce.

I attached a video of the Graphic tab recording some interactions with the threejs examples on https://threejs.org/examples/

Related threejs github issue: https://github.com/mrdoob/three.js/issues/30047

-- 
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/20241209/80bbfe97/attachment.htm>


More information about the webkit-unassigned mailing list