[Pixman] [PATCH 08/15] pixman-filter: Don't recurse unnecessarily.

[Oded Gabbay]

On Tue, Dec 22, 2015 at 9:38 PM, Bill Spitzak <spitzak at gmail.com> wrote:
> On Tue, Dec 22, 2015 at 1:38 AM, Oded Gabbay <oded.gabbay at gmail.com> wrote:
>>
>> On Sat, Dec 12, 2015 at 8:06 PM,  <spitzak at gmail.com> wrote:
>> > From: Bill Spitzak <spitzak at gmail.com>
>> >
>> > Only LINEAR is not differentiable at zero,
>>
>> I'm sorry, I don't understand this sentence. Could you please give me
>> a more detailed explanation + reference ?
>
>
> All the other filter functions have a continuous first derivative over their
> entire range.
>
> The LINEAR function is a triangle and the first derivative changes from +1
> to -1 at 0.0.

ok, so add it to the commit msg and this patch is:

Acked-by: Oded Gabbay <oded.gabbay at gmail.com>


>
> I believe Søren was concerned that the Simpson's integration would not work
> at this point. He solved this by splitting the integral into 2 or 3 at the
> zeros, so the integration is not done across these points.
>
> I figured I should keep this, though I suspect the error is really invisibly
> tiny. Apparently Simpsons integration will have some error for any function
> where the third derivative is not constant, which is true for all the
> filters here except box. But the error is really really tiny and was being
> ignored for all the other filters, and it may be ok to ignore it here too.
>
>> Thanks,
>>
>>         Oded
>>
>> > so only do the recursive split of the integral for it.
>> > ---
>> >  pixman/pixman-filter.c | 34 +++++++++++++++++-----------------
>> >  1 file changed, 17 insertions(+), 17 deletions(-)
>> >
>> > diff --git a/pixman/pixman-filter.c b/pixman/pixman-filter.c
>> > index 782f73d..0cd4a68 100644
>> > --- a/pixman/pixman-filter.c
>> > +++ b/pixman/pixman-filter.c
>> > @@ -160,38 +160,38 @@ integral (pixman_kernel_t reconstruct, double x1,
>> >           pixman_kernel_t sample, double scale, double x2,
>> >           double width)
>> >  {
>> > +    if (reconstruct == PIXMAN_KERNEL_IMPULSE)
>> > +    {
>> > +       assert (width == 0.0);
>> > +       return filters[sample].func (x2 / scale);
>> > +    }
>> > +    else if (reconstruct == PIXMAN_KERNEL_BOX && sample ==
>> > PIXMAN_KERNEL_BOX)
>> > +    {
>> > +       assert (width <= 1.0);
>> > +       return width;
>> > +    }
>> > +    else if (sample == PIXMAN_KERNEL_IMPULSE)
>> > +    {
>> > +       assert (width == 0.0);
>> > +       return filters[reconstruct].func (x1);
>> > +    }
>> >      /* If the integration interval crosses zero, break it into
>> >       * two separate integrals. This ensures that filters such
>> >       * as LINEAR that are not differentiable at 0 will still
>> >       * integrate properly.
>> >       */
>> > -    if (x1 < 0 && x1 + width > 0)
>> > +    else if (reconstruct == PIXMAN_KERNEL_LINEAR && x1 < 0 && x1 +
>> > width > 0)
>> >      {
>> >         return
>> >             integral (reconstruct, x1, sample, scale, x2, - x1) +
>> >             integral (reconstruct, 0, sample, scale, x2 - x1, width +
>> > x1);
>> >      }
>> > -    else if (x2 < 0 && x2 + width > 0)
>> > +    else if (sample == PIXMAN_KERNEL_LINEAR && x2 < 0 && x2 + width >
>> > 0)
>> >      {
>> >         return
>> >             integral (reconstruct, x1, sample, scale, x2, - x2) +
>> >             integral (reconstruct, x1 - x2, sample, scale, 0, width +
>> > x2);
>> >      }
>> > -    else if (reconstruct == PIXMAN_KERNEL_IMPULSE)
>> > -    {
>> > -       assert (width == 0.0);
>> > -       return filters[sample].func (x2 / scale);
>> > -    }
>> > -    else if (reconstruct == PIXMAN_KERNEL_BOX && sample ==
>> > PIXMAN_KERNEL_BOX)
>> > -    {
>> > -       assert (width <= 1.0);
>> > -       return width;
>> > -    }
>> > -    else if (sample == PIXMAN_KERNEL_IMPULSE)
>> > -    {
>> > -       assert (width == 0.0);
>> > -       return filters[reconstruct].func (x1);
>> > -    }
>> >      else
>> >      {
>> >         /* Integration via Simpson's rule */
>> > --
>> > 1.9.1
>> >
>> > _______________________________________________
>> > Pixman mailing list
>> > Pixman at lists.freedesktop.org
>> > http://lists.freedesktop.org/mailman/listinfo/pixman
>
>