If you check the code in cmd640.c in drivers/block, there is a routine
which checks for the secondary port secondary_port_responding(). Apparently
that routine returns 0xff inspite of device being present on the secondary
IDE interface.
So we got it the cdrom working by just commeting the code in the
routine ide_probe_for_cmd640x where it toggles the bit for secondary
port. But that should not be the case. Can Igor check why the PCI is
behaving like this ?
Amitay.
----- Before you climb a tree you have to start at the bottom. -----
Buji Proverb (Nigeria)