On Fri, 11 Mar 2005, Greg KH wrote: > On Fri, Mar 11, 2005 at 11:03:17AM -0500, Alan Stern wrote: > > Greg: > > > > When the PCI subsystem discovers a device, it calls pci_setup_device to > > initialize various things. It doesn't call pci_disable_device to stop the > > device from doing DMA, or do the equivalent (whatever that might be) to > > stop the device from generating IRQs. Unless I'm mistaken, which is > > certainly possible. > > That is because that caused too much trouble when we tried to do it, > from what I remember. I think the lkml archives has that discussion > somewhere...
It's quite believable that this might cause problems somewhere. And it's unfortunate that the mailing list archives are so large and have such inadequate indexing and search facilities... > > During a normal boot most devices are left in reasonably quiescent and > > safe condition when the BIOS passes control to the OS. But sometimes a > > few of them aren't; that's why we need to have the USB early-handoff quirk > > code. And during a kexec reboot things might be even worse. > > Well, that's a kexec issue then :) > > > Is it feasible to have the PCI device initialization sequence disable DMA > > and IRQs from the device? This could solve the problems we've been seeing > > with non-quiescent devices sharing an IRQ line at startup. > > Have patch? Maybe I'll try to put one together. Can you give me any pointers to examples showing how one prevents a generic PCI device from issuing IRQs? Alan Stern ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
