[Webkit-unassigned] [Bug 76225] Release WebGL contexts eagerly

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=76225





--- Comment #8 from Kenneth Russell <kbr at google.com>  2012-01-20 18:15:10 PST ---
(In reply to comment #7)
> > There was an idea to watch for calls to the detach() method of the owning Document of the Canvas (rather than the Canvas itself), and use those as heuristics for when to lose the WebGLRenderingContext.
> > 
> > Another idea, possibly the most principled, would be to watch for destruction of the ScriptExecutionContext (see ContextDestructionObserver in ActiveDOMObject.h) and force loss of the WebGL context when that happens. From what I have heard, it may still be possible for another ScriptExecutionContext to have a reference to the same DOM object, but this is probably not a common case.
> emm... seems two better ideas than previous ones. I'll try to investigate them and find out whether they works.
> 
> > In summary, though, making correct WebGL test execution depend on eager destruction of the WebGL contexts is not going to be a reliable solution. At this point, I firmly believe that fixing Bug 76654 (reallocate the WebGL back buffer at a smaller size if allocation fails) is the best solution and should be the top priority.
> > 
> okay, i'm gonna fix that bug firstly if there is no owner working on it.

Nobody is working on it, so great, thanks. Do you have the ability to assign bugs to yourself? If not let me know.

> > > Another question, is it enough to do the test 'fast/canvas/webgl' of layout tests for this issue?
> > 
> > A full layout test run would definitely be required. Doing full layout test runs should be a habit.
> Yes, of course. I ran the full layout tests on Ubuntu for the webkit chromium port without any code changes. However, it reports many regressions. Below are some failures:
>   fast/canvas/arc360.html = IMAGE
>   fast/canvas/canvas-before-css.html = IMAGE
>   fast/canvas/canvas-composite.html = IMAGE
>   fast/canvas/canvas-resize-reset.html = IMAGE
> Have you met this kind of problem before?

Yes. The first step is to follow the instructions on http://code.google.com/p/chromium/wiki/LayoutTestsLinux if you haven't already. If you have, then you need to use your best judgment about whether the failure is related to your changes. I don't know how to precisely reproduce the configuration on the testing bots, nor why layout test runs seem to frequently have many failures.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.