On Thu, 27 Jan 2005 23:45:58 +0100 (MET), Peter Karlsson <[EMAIL PROTECTED]> wrote: > On Wed, 26 Jan 2005, Timothy Miller wrote: > > > To be honest, I'm not even sure if we'll want to use DRI. We need the X > > server to be able to sleep while waiting on interrupts. We also need > > for the X server to be able to do most DMA without the ioctl overhead. > > If DRI doesn't give us exactly what we need for best performance, we'll > > need our own driver. > > How is the drivers supposed to be shipped? I would very much like to see > them as part of Xorg.
The driver will be "shipped" on CD and downloadable, all in binary and corresponding source form. However, the LKML guys seem very amenable to putting the kernel driver into mainline. I don't know how to get into X.org, but that probably won't be a problem either. We need to deal with both people who want to install in existing systems as well as those who want to upgrade their distro. > > > If the DRI folks want to rip apart our driver and add anything "missing" > > into DRI, all the power to them. But what we should produce for the > > prototypes is what works best for us, not necessarily what works in > > exactly the way everyone expects. > > I guess I can live with that, but perhaps one shouldn't count on people > jumping for joy and starting to code on a driver, especially if that > driver is totally different from what other drivers use (mesa/dri/drm). I > think you would find the dri/drm people much happier and more willing to > work with you if you supply them with some functional starting point > (driver). I hope you can see what I mean... Tech Source will write the prototype driver, if we can't get volunteers to help on that. We can mostly just hack our existing Raptor driver for Linux and release it. > > We use XFree86 for our ATC and medical products, but our kernel driver > > doesn't use DRI. > > These seem to be 2D, mainly. The pre-dri XFree < 4.x was also 2D. Dri was > created to allow a fast path to the graphics card so that 3D could be > fully utilised. Perhaps the mesa/dri/drm/glx/X is not the perfect > "marriage" but so far it's the best we've got. Of course that doesn't mean > I'm against progress. In fact I could see this card becoming a de-facto > standard to test and implement new X extensions/protocols on. We have our own kernel driver because we do a number of things on vertical blank interrupt. That includes colormap and cursor glyph updates (so the hardware doesn't glitch during active video). Additionally, we provide access to other interrupts to the X server. _______________________________________________ Open-graphics mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-graphics List service provided by Duskglow Consulting, LLC (www.duskglow.com)
