> Have you tried giving lilo the 'root=xxxx' command? If hda1 is now
> being detected at hde1, I suggest typing
> 'linux root=/dev/hde1' at the LILO prompt.
This will break because of the contents of the /etc/fstab.
> The change hda1->hde1 may be intentional, as support for up to 8 idea
This was to clean up the additions for IDE-devices.
This is the second case that I know about where the on-board ide-chipset
is given a higher pci-bus number than the pci-slots.
The ASUS P55T2P4D (Intel 430HX), some 486 (UMC UM8881F), and maybe other
mainboards may do this oddity. This blows the plan for using the
addition to the lilo.conf append line of "pci=reverse".
I wanted to avoid hard-setting of bootablity of add-on cards to have
some flexablity. The goal was to prevent the silly game of card/slot
order (I recall some discusion about this with SCSI crowd).
The second option that just came to mind is to allow one to only reverse
the IDE-controller scan order. This would allow "pci=reverse" to still be
called with the constraints of imposed by my first patch against my
changes added into 2.1.122. This sounds better and retains to orginal
flexablity.
> devices was added, I think, in 2.1.122. However, you may not have /dev/
> files for them! So you may have to poke around in patch-2.1.122 for the
> proper major/minor device numbers for /dev/hde and /dev/hdf. I think I
> remember seeing this in the patch ... just search for 'hde' :)
Not need, unless there is something that I truly missed.
> Does that help? I don't have a Promise card or anything, so I don't
> know. But it sound like it would make sense.
Cheers,
AMH
Sometime late tonight or in the morning will post a second method
against pci-bus-number-order for IDE controllers.
The first was "pci_ide_order-2.1.122.patch"
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/