After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 357253 - Optimized processing methods.
Optimized processing methods.
Status: RESOLVED FIXED
Product: GEGL
Classification: Other
Component: general
unspecified
Other All
: Low normal
: ---
Assigned To: Default Gegl Component Owner
Default Gegl Component Owner
Depends on: 357251 427532
Blocks:
 
 
Reported: 2006-09-22 22:09 UTC by Øyvind Kolås (pippin)
Modified: 2008-05-17 12:33 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Øyvind Kolås (pippin) 2006-09-22 22:09:29 UTC 
Right now all computation is done in 32bit floating point RGBA.
Allowing to register optimized versions of different pixel formats
(that can be regression tested against the float implementation) is
and important step. This would also allow to plug-in external code,
like snippets of cairo code to do evaluation in 8bit.
Comment 1 Øyvind Kolås (pippin) 2007-01-21 23:56:27 UTC 
This should also allow computation on the GPU and use of MMX and similar extensions.
Comment 2 Øyvind Kolås (pippin) 2007-01-22 00:19:04 UTC 
In some respects this depends on bug #357251
Comment 3 Øyvind Kolås (pippin) 2008-05-17 12:33:49 UTC 
2008-04-18  Øyvind Kolås  <pippin@gimp.org>

        * gegl/operation/gegl-operation-processors.c:
        (gegl_class_register_alternate_vfunc): added new more general function
        for runtime overriding of perclass vfuncs, that takes a pointer to the
        class structure and the vfunc to operate on.
        (gegl_operation_class_add_processor): use the new internal function
        internally (still uses hard-coded logic for the GEGL op base classes
        to look up the vfunc pointer.)


(and more commits after this)