On Wed, 30 Jan 2002 [EMAIL PROTECTED] wrote: > >> Here you can see the problem when you are using a card in the second bus > >> and are mapping the IO for the first. To solve this there needs to be a > >> bit of abstraction involved. I suppose one could map all the IO's, and > >> then set a flag in the driver which tells which bus this particular driver > >> or IO operation is dealing with.
> >I wonder if we could use RAC for this, anyone with more knowledge of how RAC > >works cares to comment ? > It's my understanding that the domain patches (which are now in the trunk > afaik) will handle this. I haven't had a chance to try them out yet on > PPC though. The Domain-branch changes are but a first step, and thus not sufficient, to fix multi-head for PowerPC's. Kernel changes are/were a second large piece of this puzzle. Beyond these two, there are a few relatively minor XFree86 changes here and there (e.g, compiler.h, os-support/bus/Imakefile, Pci.h, to name but a few) that are still needed. All this is documented in Domain.note and/or linuxPci.c comments. But, by far, the most difficult part of implementing multi-head is in dealing with the system chipsets found on these motherboards. This difficulty isn't technical at all, but is rooted in humanware. What takes the lllooonnnggg-est time is extracting the needed information from manufacturers and/or other relevent parties, a process I have found, and still find, akin to pulling hens teeth. To be fair, this situation isn't PowerPC-specific. On the technical side, when looking at the various CPU architectures out there, the main problem is that quite a number of radically different PCI schemes have been married to those CPUs and XFree86's model of what a host-bridge/chipset is isn't yet general enough to handle them all. XFree86 still has a rather simplistic picture gleaned from its ix86 origins. Anyway, I'll keep plugging away at this, as I have for almost two years now. Marc. +----------------------------------+-----------------------------------+ | Marc Aurele La France | work: 1-780-492-9310 | | Computing and Network Services | fax: 1-780-492-1729 | | 352 General Services Building | email: [EMAIL PROTECTED] | | University of Alberta +-----------------------------------+ | Edmonton, Alberta | | | T6G 2H1 | Standard disclaimers apply | | CANADA | | +----------------------------------+-----------------------------------+ XFree86 Core Team member. ATI driver and X server internals. _______________________________________________ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
