On Sun, 16 Sep 2001, Manuel Teira wrote:
> BTW. Could anybody say me what is the AVOID_CPIO define used for ?
Looking at xc/programs/Xserver/hw/xfree86/drivers/ati/Imakefile, this is
defined by default only for Sparc and PPC. Here's the comment:
/*
* The following configuration logic is only meant as a first cut, and is
* therefore incomplete. ...And, no, you do NOT have permission to move this
* into xfree86.cf...
*
* Currently, ATIAvoidCPIO >MUST< be #define'd as YES for those platforms
* (architecture/OS combinations) that neither provide nor emulate a
* little-endian undirected PIO address space of at least 64 kB in size.
*
* "Undirected" means the driver does not need to determine the identity nor
* location of the responding adapter before accessing a particular location in
* the PIO address space.
*
* #define'ing ATIAvoidCPIO to YES generates a driver that will only support
* PCI/AGP Mach64 adapters using a linear aperture and the accelerator CRTC.
* The resulting driver will also require the same of the environment on server
* entry.
*
* For testing purposes, #define'ing ATIAvoidCPIO as YES is also supported on
* platforms that do, in fact, provide or emulate a PIO address space as
* described above, but this should not be the default driver configuration.
*
* Currently, ATIAvoidNonPCI needs to be set to YES for those platforms that
* "drop down" to firmware on accesses to allocated, but disabled, PCI space.
* ATIAvoidNonPCI necessarily implies ATIAvoidCPIO.
*/
> > By the way, gang, Gareth's old code has hacks in it to allow you to run
> > accelerated 3D within Mesa so long as the DRI driver isn't loaded- gets
> > about 115-120 fps in gears on my PII 450 with the code I have in hand from
> > Manuel (His first pass at this...). I do agree with his sentiment that it
> > has no business being in the trunk, but I do believe with the new
> > re-worked, repackaged stuff, we have the makings of a new branch for the
> > CVS tree.
>
> Do you think we need to startup from another code base? I made the Gareth's
> code migration to the XFree4.1.0 CVS trunk, but somebody said before that
> there's a newer trunk for the DRI.
> My opinion is that we need urgently to have a common codebase for all the
> people interested to work in the Mach64 driver. So, if you think we must
> merge with the DRI newest trunk please make me know.
I agree that we need a common codebase, but my understanding is that we
need a working patch before we can start a new branch in CVS. I guess the
question is whether it makes more sense to get a patch working based on
the 4.1.0 release first and then merge in changes from the trunk or to do
the merge now. Of course the trunk is a moving target, but there may be
changes in the 2D driver that would be helpful. Frank?
--Leif Delgass
_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel
