Re: Servers vs. Libs

On Tue, 13 Feb 2001, Sandro Magi wrote:

> >
> >Except that almost every "serious" graphics program does the inevitable
> >genesis to the point where it wants memory mapped access to the frame
> >buffer. And I'm not really arguing much, since the MGR port could call the
> >main mgr process the "graphics server", and then all clients are in their
> >own address spaces. But, then, MGR is linked with svgalib to actually
> >control the graphics card, so that's adopting the other rule.
> >

well, I don't see anything that wouldn't let the graphics server negotiate
a shared memory frame buffer with a graphics intensitive program running
on top of it. That frame buffer wouldn't use the acceleration in the card
though...
But the graphics server would of course have line drawing commands and
such and use the card acceleration for them. If there is any that is,
otherwise the server should emulate them.

And I suppose MGR isn't going the be the only windowing system for VSTa
forever. and I would be nice to have different windowing systems use the
same graphics drivers. and you wouldn't need to update the windowing
system just because the driver for the card is updated.

> >The other big question is state management. Multiplexing is one (thus, mgr
> >has its own process). But even in a single client case, the complexity of
> >state and requirements for cleanup (i.e., on a segv hardware needs to be
> >un-programmed) might argue for a distinct process.
> >
Yes, that's very true

> >Any rule can be applied in such a way that it doesn't make sense. I don't
> >see any problem with your arguments.
>
> If graphics were in server, would that enable network transparent graphics?
> Since the graphics server would be replying to messages on it's port, it
> doesn't matter where those messages are coming from. Couldn't those messages
> simply be sent to a graphics server across a network instead of locally?
> Would that require much extra effort/coding, or would it happen
> automatically? Am I off base? Some serious security issues also come to
> mind.
>
Yes, and that would basically happen automatically. Or, just as
automatically as with all other filesystems.

> With more thought, the whole distributed graphics server idea seems dubious.
> What would be the point of allowing access to your graphics hardware to
> another computer on the network? This is a driver issue right? (I was
> thinking on a higher level ala X-Windows)

well, you could multihead X usign two computers :)

/Erik
Received on Sun Feb 18 10:38:04 2001