[Gegl-developer] gdb messages

[Øyvind Kolås]

On Tue, May 13, 2008 at 5:37 PM, Zhang Junbo <bearscafe2.0 at gmail.com> wrote:
> Hi,
>
> I'm just starting to write my code for gegl, I spent the whole day to
> debug but failed. My code(svn 2287) can be found on
> http://svn.gnome.org/viewvc/gegl/branches/branch_zhangjb/ . Here is
> some messages from gdb.

The reason the code segfaults is that you are passing buffer of 32bit
floating point to a function in the underlying library that expects
64bit floating point. The attached patch creates the needed pixel
formats to represent 64bit buffers since babl doesn't know this format
yet.

The format could have been created with:

babl_format_new (
    "name", "RGB double",
    babl_model("RGB"),
    babl_type ("double"),
    babl_component ("R"),
    babl_component ("G"),
    babl_component ("B"),
    NULL);

and then referred to, I chose to just create it directly without
giving it a name.

If the fftw library has functions working on 32bit buffers it might be
more suited
for integration, if not I guess we just have to live with these larger buffers.

I've attached a patch that makes the operation operate in 64bit like
the function used.

/Øyvind K.

-- 
«The future is already here. It's just not very evenly distributed»
 -- William Gibson
http://pippin.gimp.org/ http://ffii.org/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dftdiff
Type: application/octet-stream
Size: 2659 bytes
Desc: not available
Url : /lists/gegl-developer/attachments/20080516/e32ff787/attachment.obj