[cairo] Pixman - refactoring of fbFetchTransformed - performance issue
frederic.plourde at polymtl.ca
Thu Feb 21 13:56:50 PST 2008
Hi ! I've run the perf suite on paint, paint-with-alpha and fill with
"min" and "mag" sources.... And my results are closer to what Bertram got.
you can see them at : http://pastebin.mozilla.org/339355
in the best case, we have 20% speedup
in the worst case, we have 3X slowdown
I ran the tests on a intel Duo Core 2 CPU, 1.6 GHz
and compiled both versions using Visual C++ on WinXP in release mode (-O2)
> As an aside, it's odd that the difference between *_source and *_over
> is so large in your tests, and that *_source turns out to be slower.
> I wonder why.
> My numbers are different, for example:
> [ 9] image-rgba paint_image_rgba_mag_source-256 12614763 6.919 7.006 0.81% 8
> [ 10] image-rgba paint_image_rgba_min_over-256 14194059 7.785 7.788 0.09% 4
> [ 9] image-rgba paint_image_rgba_mag_source-256 16889333 9.264 9.267 0.02% 5
> [ 10] image-rgba paint_image_rgba_min_over-256 18351320 10.066 10.086 0.51% 5
> with hardly any difference between _source and _over.
>> Could a third person run the perf before and after the refactoring to
>> check the numbers?
> Yes, that would help - the more numbers the better.
> The relevant commits are in the pixman repository,
> (before the refactoring)
> Of particular interest are the *_image_rgba_min_* and
> *_image_rgba_mag_* tests in the performance testsuite (now in
> cairo's git repository).
> cairo mailing list
> cairo at cairographics.org
More information about the cairo