Krzysztof Kosiński <[email protected]> writes:

> So what is the bottom line? Should we have two versions of all code,
> one using float and one using pixman_fixed_t, or is replacing
> pixman_fixed_t with float sufficient?

As far as I am concerned, using floating point is fine. I have not
heard otherwise from anybody who has done any kind of real work on the
pixman ARM backends.

> How much slower is softfloat - a factor of 5, 15 or 500?

I don't know the number, but it's likely true that if you hit
softfloat at any regularity, the software becomes too slow to be
usable. So when I say we should use floating point, that is implicitly
saying that pixman does not support non-FPU hardware. I'm fine with
that.

If someone shows up with (a) a credible claim that he will ship pixman
to a non-trivial number of people in a non-FPU device, and (b) the
resources to maintain an integer version of all of the pixman code,
then that could be considered.

Until then, floating point is fine.


Soren
_______________________________________________
Pixman mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/pixman

Reply via email to