ML405 failed reboot

ML405 failed reboot

[Guillaume Dargaud]

Now that my Xilinx card is operational with Kernel 2.6.24 and a 
buildroot/busybox/uClibc config, I'm getting on the application side.

First hitch: I can't do a soft reboot of the card:

# reboot
The system is going down NOW!
Sending SIGTERM to all processes
Requesting system reboot
[  434.713999] Restarting system.
[  434.7[  434.718123]   Signal: 4
[  434.718133]   Code:   30001
[  434.718144]   Addr:   c74cbdc8

And then it just hangs.
I also tried the additional -n (no sync) and -f (force) without difference.

Design problem ?
Any other way to force a reboot ?
-- 
Guillaume Dargaud
http://www.gdargaud.net/

Re: ML405 failed reboot

[Guillaume Dargaud]

I'm posting a little more info 'cause no-one took the bait:

# tail -f /tmp/messages  &
# reboot
The system is going down NOW!
Sending SIGTERM to all processes
Requesting system reboot
[ 2021.926957] Restarting system.
[ 2021.9[ 2021.931063] Kernel stack overflow in process c7c52c10, 
r1=c01fc25b
[ 2021.935935] NIP: c7503dc8 LR: 00000001 CTR: 00000000
[ 2021.940871] REGS: c7503d40 TRAP: c000b3b8   Not tainted 
(2.6.24-rc8-xlnx)
[ 2021.947679] MSR: fee1dead <EE,PR,ME,IR>  CR: c01d0000  XER: 00000000
[ 2021.953996] TASK = c7c52c10[261] 'init' THREAD: c7502000
[ 2021.959084] GPR00: 00001001 c01fc25b 00000012 c7503d72 c7503db0 c0017020 
00000000 000007e5
[ 2021.967376] GPR08: 3c343e5b 20323032 312e3932 38373530 5d2000d0 c0017020 
00000000 000007e5
[ 2021.975669] GPR16: 3c303e5b 20323032 312e3932 36393537 5d2000b0 00000000 
00000000 0000004d
[ 2021.983963] GPR24: 00000000 fee1dead 28121969 00000000 c7503df0 c00170a8 
02002222 c7503df8
[ 2021.992430] NIP [c7503dc8] 0xc7503dc8
[ 2021.996059] LR [00000001] 0x1
[ 2021.998997] Call Trace:
[ 2022.001423] Kernel panic - not syncing: kernel stack overflow
[ 2022.007142] Rebooting in 180 seconds..

[ 2201.477426]   Code:   30001
[ 2201.477437]   Addr:   c01e5f28
[ 2201.484367] Oops: Exception in kernel mode, sig: 4 [#1]
[ 2201.489548] NIP: c01e5f28 LR: 000001d6 CTR: 00000000
[ 2201.494480] REGS: c01e5ea0 TRAP: c000b3b8   Not tainted 
(2.6.24-rc8-xlnx)
[ 2201.501293] MSR: 20323032 <FP,ME,IR,DR>  CR: 000001d6  XER: d07870f8
[ 2201.507609] TASK = c7c52c10[261] 'init' THREAD: c7502000
[ 2201.512698] GPR00: c01e5eb0 c01fc25b c01b1634 c01e5f18 c01e5f10 c0017020 
00000012 000019e7
[ 2201.520990] GPR08: c01e5ee0 c0016820 00000030 00021030 c01e5ee0 c01fc271 
00000012 c01e5f02
[ 2201.529284] GPR16: c01e5f40 c0017020 00000000 000007e6 3c303e5b 20323032 
322e3030 0000004d
[ 2201.537578] GPR24: 0002bf20 00000001 28121969 00000000 c01e5f50 c00170a8 
02000000 c01e5f58
[ 2201.546044] NIP [c01e5f28] 0xc01e5f28
[ 2201.549673] LR [000001d6] 0x1d6
[ 2201.552783] Call Trace:
[ 2201.555202] Instruction dump:
[ 2201.558142] XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX 
XXXXXXXX XXXXXXXX
[ 2201.565826] XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX 
XXXXXXXX XXXXXXXX

And now it's truly well hung.
Problem with init ? The wait reboot is compiled ? The reset vector ?
-- 
Guillaume Dargaud
http://www.gdargaud.net/