On Wed, 2003-06-11 at 17:26, Guido Landra wrote: > > -- About PCI/AGP issue, I understood that radeon driver can send > commands to card in two ways: > 1) MMIO, available on both AGP and PCI. > 2) system memory block seen by the GPU (GART), available on all AGP > cards and on some PCI cards. > The first mode is supported on all platforms (x86, alpha, etc.), an > all cards (7xxx, 8xxx, ...), on all buses (PCI, AGP) but is able only > to accelerate 2D. > The second mode is mandatory for accelerating 3D and can optionally > accelerate 2D: it is supported on all AGP cards and on radeon > PCI-based cards (but at driver-level only on alpha platform it is > implemented). > > My original idea was to enable DRI without using agpgart.o (because I > was having problems with nforce north bridge not supported). > It is theoretically possible using PCI GART but it is currently not > implemented on x86 based systems, only on alpha (I have a pc....). > > Is this correct?
No, PCI GART (which is a feature specific to ATI cards as of Rage128) should work everywhere now, as Mike pointed out: > Mike A. Harris wrote: > > > > > The issue is that the Radeon driver for PCI cards relies on the > > > existence of a PCI-GART (it does the same as the AGP GART, but > > > in the PCI bus) which I don't know if it's present on AGP cards. > > > > > It is, and should work on AGP cards just like a PCI card, with > > "ForcePCIMode". -- Earthling Michel Dänzer \ Debian (powerpc), XFree86 and DRI developer Software libre enthusiast \ http://svcs.affero.net/rm.php?r=daenzer ------------------------------------------------------- This SF.NET email is sponsored by: eBay Great deals on office technology -- on eBay now! Click here: http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5 _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
