Some additional info.
This is with a self compiled klibc-2.1-3
If this is is irrelevant to #349857, I will eventually open a new incident.
I am trying this with 2.6.15-4-s390 kernel.. The 2.6.15-3 headers were just
not capable of compiling klibc 1.2-3 for s390 anyway.
Upon IPL (boot), I am getting this :
kjournald starting. Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.
Begin: Running /scripts/log-bottom ...
Done.
Done.
Begin: Running /scripts/init-bottom ...
Done.
Spawning shell within the initramfs
BusyBox v1.01 (Debian 1:1.01-4) Built-in shell (ash)
Enter 'help' for a list of built-in commands.
/bin/sh: can't access tty; job control turned off
/ #
.exit
Kernel panic - not syncing: Attempted to kill init×
HHCCP011I CPU0000: Disabled wait state
PSW=000A0000 800441CE
(note : a set -x was added just so that I could figure out where it was
having a problem).
(note : the HHCCP011I message is related to the S/390 emulator, not linux,
and indicates the machine has gone into a stopped state).
If I have 'break=init' in the boot parm line, the following occurs (if I
manually issue the commands) :
(PS : break=init is a initramfs-tool feature, unrelated to klibc).
+ mount -n -o move /sys /root/sys
+ mount -n -o move /proc /root/proc
+ Ö × -x /root/sbin/init !
+ Ö -n !
+ unset debug
+ maybe_break init
+ Ö xinit = xinit !
+ panic Spawning shell within the initramfs
+ Ö -x /sbin/usplash_write !
+ modprobe -q i8042
+ modprobe -q atkbd
+ echo Spawning shell within the initramfs
Spawning shell within the initramfs
+ FS1=(initramfs) /bin/sh
BusyBox v1.01 (Debian 1:1.01-4) Built-in shell (ash)
Enter 'help' for a list of built-in commands.
/bin/sh: can't access tty; job control turned off
/ # .run-init /root /sbin/init
run-init /root /sbin/init
run-init: //var: error 4294967275
/ #
Note : This *may* be expected since 'run-init' wasn't invoked using an
'exec'
Also tried :
.pivot_root /root /root/mnt
pivot_root /root /root/mnt
pivot_root: error 4294967274
/ #
Apparently, the errnos shown are negative numbers (?) and are resolving to
-21 (for the 1st : -EISDIR)
-22 (for the 2nd : -EINVAL)
The 1st occurence seems to occur in the call to
utils/libruninit[run_init/nuke/nuke_dir/nuke_dirent].. The second seems to
be directly associated with the call to pivot_root(2)
--Ivan
