[cairo] pixman regression

Soeren Sandmann sandmann at daimi.au.dk
Fri Mar 7 10:37:35 PST 2008


"Robert O'Callahan" <robert at ocallahan.org> writes:

> On 07 Mar 2008 06:06:21 +0100, Soeren Sandmann <sandmann at daimi.au.dk> wrote:
> 
> > Comments appreciated, both on the math and the attached patch.
> >
> 
> I can't really comment on the math but
> 
> +       v.vector[0] -= (pixman_fixed_e * v.vector[2]) >> 16;
> +       v.vector[1] -= (pixman_fixed_e * v.vector[2]) >> 16;
> +       unit.vector[0] -= (pixman_fixed_e * unit.vector[2]) >> 16;
> +       unit.vector[0] -= (pixman_fixed_e * unit.vector[2]) >> 16;
> 
> Shouldn't the last line be "unit.vector[1]"? Also, I'd put the common
> subexpression in a local variable but maybe that's just me.

Good catch, thanks. This piece of code is not particularly time
critical, so I'll just leave the subexpressions in place.



Soren


More information about the cairo mailing list