[Webkit-unassigned] [Bug 5861] feConvolveMatrix filter is not implemented
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Thu Jul 1 01:19:57 PDT 2010
https://bugs.webkit.org/show_bug.cgi?id=5861
--- Comment #33 from Nikolas Zimmermann <zimmermann at kde.org> 2010-07-01 01:19:56 PST ---
(In reply to comment #32)
> I did your suggested changes. Thanks for the review.
>
> > WebCore/svg/graphics/filters/SVGFEConvolveMatrix.cpp:231
> > + totals[0] += m_kernelMatrix[kernelValue] * static_cast<float>(paintingData.srcPixelArray->get(kernelPixel++));
> > Would it help to store m_kernelMatrix[kernelValue] in a local variable? Not sure.
>
> You mean copy the values to a local array? Not sure that will help.
Oh, I meant just m_kernelMatrix[kernelValue], as you're calling it multiple times, but it's not worth it - just fine as it is.
>
> > WebCore/svg/graphics/filters/SVGFEConvolveMatrix.cpp:244
> > + paintingData.dstPixelArray->set(pixel++, clampRGBAValue(totals[0] / m_divisor + paintingData.bias));
> > Is m_divisor guarenteed to be non null? If so please add an assertion on top of this function.
>
> Added the assert and a comment. m_divisor cannot be 0, otherwise the filter will be not created.
Thanks.
>
> > WebCore/svg/graphics/filters/SVGFEConvolveMatrix.cpp:355
> > + fastSetInteriorPixels<true>(paintingData, clipRight, clipBottom);
> > Just wondering, is it possible to use m_preserveAlpha directly as template parameter? I guess not :-)
>
> Template arguments must be compile time constants.
Ah right, forgot.
>
> I still not have expected pixel test for the generic mac, only for mac-leopard :(
Same here, but I don't care about that, as long as we don't have pixel test bots, where I could grab the right results from. Let someone else update them if he/she notices diffs.
I'm just generating mac-leopard results, and put them into mac. If there are already mac-lepard specific results, I'm just regenerating them, and copy the same file into mac/ - until someone else can fix them for me...
Waiting for EWS before giving the final review.
--
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
More information about the webkit-unassigned
mailing list