On 2002.04.27 15:33 Benjamin Herrenschmidt wrote: > > > >I don't see how. The function in which Leif fixed the big had a debug > >output statement which would appear in the kmsg if it was the last one. > > > > From what I've read the *_nopage functions are associated with > accessing > >memory maped regions. It could be that the DMA pages aren't being > properly > >maped in the client memory space due to some x86 architecture specific > >assumption in the DRM. > > What type of DMA page are you using ? The AGP ones ? The DRM in my > kernels have hacks to make this work with Apple UniNorth chipset > (though I never managed to have AGP operations stable with neither > r128 or radeon, the card dies after a while). If it's normal > PCI DMA, there should be no difference between x86 and PPC, at > least on pmacs
It's normal PCI DMA. > > What PPC machine is this ? (sorry, I missed the beginning of the It was never specified so far. > thread). On some PPC's like PReP, the mapping between bus addresses > and CPU physical addresses on PCI isn't 1:1 and the system RAM is > not mapped at 0 for bus mastering PCI devices. If you use the PCI > DMA API, things are ok. If you aren't, then some tweaks may be needed. > (See the definition of deprecated virt_to_bus in asm-ppc/io.h) The DRM doesn't use the PCI DMA API. Instead does everything manually. Why do you say virt_to_bus is deprecated? Where can I find more information about these APIs? > > Ben. > José Fonseca _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel