Kernel support for graphics cards

[olafBuddenhagen at gmx.net]

Hi,

On Wed, Feb 08, 2006 at 05:51:56PM +0100, Michel Dänzer wrote:
> 
> Please attribute your quotations properly.

Sorry, bad old habit of mine :-(

> On Tue, 2006-02-07 at 18:44 +0100, olafBuddenhagen at gmx.net wrote:
> > 
> > I think there is some misunderstanding here. The idea is *not* to
> > have the kernel put limits what kind of hardware access the
> > userspace drivers are allowed to do, using perfect knowledge of the
> > hardware to decide what is safe and what is not. The idea is rather
> > for the kernel to do the graphics hardware access *itself*, the
> > userspace driver only telling what operation it wants performed.
> > 
> > This doesn't require any more knowledge about the hardware than
> > doing the same access in userspace directly. The difference is that
> > the kernel driver can ensure that only known safe operations can be
> > invoked. Thus, it's actually a great advantage in case of incomplete
> > specs.
> 
> I basically agree; this is more or less what the DRM does.

Indeed, DRM might be a good starting point for part of the necessary
interface.

> Note that this kind of interface does implicitly limit what userspace
> can do, and requires careful design to balance between providing
> functionality, maintaining security and limiting overhead. It's not as
> if simply choosing this kind of interface automagically solves all
> problems.

Right. Therefore it's crucial to come up with a universally accepted
framework, spuring broad participation, not some isolated approach.

So we are asking for input, trying to establish a design all agree on --
a common ground on which we can work together with X developers and
other involved parties.

> The hardware specific DRM code is shared between OSs as well.

Ah, nice: So having portable kernel-level graphics drivers is already a
widely established idea in principle. One trouble less :-)

-antrik-