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?
How much slower is softfloat - a factor of 5, 15 or 500? Another thing we have to take into account is that new versions of Cairo/Pixman are rather unlikely to find their way onto legacy ARM hardware, because it consists mostly of embedded or mobile systems where updating the software is either quite a task or not possible. So the question we should be answering is: what proportion of *new* ARM systems which Cairo will run on will not have an FPU? I also note that most of the Cairo API uses double precision floats. Regards, Krzysztof _______________________________________________ Pixman mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/pixman
