[Webkit-unassigned] [Bug 99829] OpenCL version of SourceAlpha, SourceGraphics and FETurbulence filter effects

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=99829





--- Comment #23 from Zoltan Herczeg <zherczeg at webkit.org>  2012-11-23 03:58:10 PST ---
> The kernel might be correct, but the compiler might not. My experience with
> OpenCL is taking one kernel that builds on one implementation won't work on
> another. I wouldn't put so much faith in the compiler.

Any example for that? We does not use anything which is not specified in the standard. Btw this patch contains a feature, which should be tested before enabled, so in general the only thing which could happen here is the out-of-resource condition.

> Such is the way with OpenCL. On the Nvidia platform in particular most problems end up manifesting themselves as an error from clFinish.

In general WK does not have a mechanism to detect out-of-resource conditions. It calls CRASH(), or just simply crashes.

> > > Where do you wait for the kernel to complete? The only clFinish/clWaitForEvents I see is before a clEnqueueNDRangeKernel
> > 
> > Opencl supports async execution, so we don't need to wait here.
> 
> You do when you'll need the result (which I guess I'm missing)

That is performed by clEnqueueReadImage() (and ideally only happens for the last image). The standard says: All commands that use this image object have finished execution before the read command begins execution. The general idea is that we just enque OpenCL commands, and let WK and OCL do their work, and when the result is needed we read it. Up to that point they can run parallelly.

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