[cairo] Profiling the slowness of xlib surface

Carl Worth cworth at cworth.org
Thu Apr 20 11:27:04 PDT 2006


On Thu, 20 Apr 2006 10:50:38 -0700, "Vladimir Vukicevic" wrote:
> if you're running under X, in many cases you're better off doing as
> much as you can in a local image surface, and only blitting the final
> result to X.

In the past, this was good advice as X rendering operation were always
a bad fit for what one wanted to do.

In the future, (as I anticipate it), this will be bad advice as the X
operations will be a good fit and will take advantage of
video-hardware for acceleration.

In the present, we're in a transition. With cairo plus the Render
extension we have a fairly good story that connects what application
developers want to do with operations being performed in the X
server. So all that's left is making those operations as fast as
possible. And that work is currently happening with the work being
done on EXA, etc. I'll let the people involved with that mention any
more specifics.

Currently, whether you're living in the world of the past or the world
of the future can depend quite strongly on what video cards/X
servers/driver versions you are using.

> On 4/20/06, Timothée Lecomte <timothee.lecomte at ens.fr> wrote:
> >
> > If I create my temporary surface with cairo_surface_create_similar(), I
> > get the artifacts as reported in a previous post (thread "avoiding
> > seams").
> > If I create the temporary surface with cairo_image_surface_create(), the
> > output is correct.

That kind of thing, (two surfaces giving very different results), is
definitely a bug. I appreciate the test case you sent. We need to get
that into cairo's test suite.

It's probably a bug that needs to be fixed in the X server, (or
perhaps already has in some versions), again I'll let others (== Eric
Anholt) comment on whether rendercheck already includes a SATURATE
test that would cover this case.

-Carl
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 191 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/cairo/attachments/20060420/90f32bf3/attachment.pgp


More information about the cairo mailing list