LCD, pan, randr1.2, and pan-to-downscale

[Keith Packard]

On Sun, 2007-04-15 at 10:04 -0700, Eric Anholt wrote:

> I think most people would agree that that should really be a compositing
> manager feature.

You could say that rotation should also be a compositing manager
feature; it's just another transformation of the screen contents, after
all. To a great extent, these screen->crtc transformations are just
built-in compositing support. Do any of them deserve to be built-in?

> I'm personally pretty convinced that even basic pan-and-scan ought to be
> a compositing manager feature, as if I was to ever use pan-and-scan (not
> that I do), it seems like I'd surely want a more interesting policy for
> when and where to move the crtc in the framebuffer than what that the
> server used to have.

pan-and-scan needn't be a compositing manager feature to have better
policy; we've got the ability to move the CRTC around the screen, an
application is free to provide whatever policy they want at this point.
I agree that the mouse-based pan-and-scan policy is fairly useless as
pop-up windows often appear off-screen.

Of course, we should fix the driver so setting the position of the CRTC
doesn't cause the server to reset the whole output chain...

-- 
keith.packard at intel.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.x.org/archives/xorg/attachments/20070416/608a4dce/attachment.pgp>