Hi Marcelo,
I am lacking some understanding concerning the setup() routine of Linux
>=2.4.19.
If I understand it right, then with kernel version 2.4.19, some novelties
concerning the handling of initrd and rootdevs has entered the kernel.
The same "linuxrc" program, e.g., can work with a 2.4.18, but not with a
2.4.19/20.
I have thought (and the initrd documentation in the Documentation/
directory states it, too) that the pivot_root mechanism is the "new one",
whereas the change_root machanism which automatically takes place if the
linuxrc programs exits for some reason, is the "old one".
Now have a look a the following example linuxrc sniplet, taken from
SuSE-7.3/8.0.
--- mk_initrd Tue Mar 4 14:13:43 2003
+++ mk_initrd Tue Mar 4 14:14:43 2003
...
mount -n -t proc proc /proc
+echo 0x0100 > /proc/sys/kernel/real-root-dev
mount -n -t $rootfstype $rootdev /mnt
rm -f /mnt/.initrd 2>/dev/null
mkdir -p /mnt/.initrd
..
The point is that without the "echo ..." line, the mounting of the
__real__ root device (harddisk partition) won't work for kernels >=2.4.19.
On the other hand, this should actually not be necessary if using a
pivot_root call.
Recent distributions now handle the problem by just having linuxrc loading
the necessary modules and exiting all of a sudden, and everything works.
My question is: is pivot_root deprecated by now? I just am quite dazzled
and want to know how to __cleanly__ handle the mounting of a new root
device.
Best regards
Oliver
Dr. Oliver Tennert
+49 -7071 -9457-598
e-mail: O.Tennert@science-computing.de
science + computing AG
Hagellocher Weg 71
D-72070 Tuebingen
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Mar 15 2003 - 22:00:29 EST