Re: [PATCH] mkinitrd --without option

[Bill Nottingham <notting redhat com>]

Warren Togami (wtogami redhat com) said: 
>> Considering the drivers do their own handoff in this case, how does this
>> actually cause problems/
>
> They do? Hmm...  in both kvm and with a real thin client, removing 8139too 
> from the initrd allowed NFS to work while it would consistently timeout or 
> error out with it.

The drivers check the PCI revision ID and ignore the card if it's the wrong
one. Note that udev would load both drivers as well.

Bill