On Fri, 2011-10-07 at 09:14 +0300, Heikki Lindholm wrote: > 7.10.2011 8.01, Ben Hutchings kirjoitti: > > On Mon, 2011-10-03 at 20:05 +0300, Heikki Lindholm wrote: > >> Shouldn't this bug rather be filed against the kernel. The pata_macio > >> module doesn't autoload because the pci id of the keylargo controller > >> isn't listed in the pci match table of the module. The generic macio > >> code matches anything Apple (no idea why, too painful to find out all > >> ids?) and maybe pata_macio should, too. > > > > ide_pmac and pata_macio have exactly the same table of PCI device IDs. > > Probably irrelevant as ide_pmac was =Y intead of =M, no? > > > Further, if a device's ID isn't in the table then not only will the > > module not be auto-loaded but loading it manually will have no effect > > because the driver won't be matched to that device. > > Some chips attach through pci, some through the macio platform driver. Oh, duh, sorry. > I think listing the pci id of the macio chip would allow the autodetection > to work, but would probably not be the correct solution. Maybe the > problem is in udev or whatever, I don't know anything about those > things, or most other things for that matter, but I still don't think > force-loading the module in initramfs is the cleanest solution. I'm also > pretty sure that if the config had PATA_MACIO=Y, it would work. Yes, that seems like the right thing to do. Ben. -- Ben Hutchings For every action, there is an equal and opposite criticism. - Harrison
Attachment:
signature.asc
Description: This is a digitally signed message part