2.4.23 oops on PowerPC

[Hamish Moffatt]

Hi,

I'm trying to get Tux running on an embedded powerpc system. I have
2.4.23 (including patches from my embedded vendor) to which I applied
http://people.redhat.com/mingo/TUX-patches/tux3-2.4.23-A3

As well I patched two other sources files for PPC;
arch/ppc/kernel/misc.S to add the handler for the syscall, and
include/asm-ppc/fcntl.h to add the #define for ATOMICLOOKUP.

The kernel builds and boots OK. Tux starts. When I fetch a page from it,
it serves the page then has an oops, and refuses any more connections.
I'm serving off NFS, btw.

>From the list archives it looks like others have had this working. Can
anyone help me out with any suggestions as to what I might have missed,
or how to debug this?

My .config | grep TUX says

CONFIG_TUX=y
# CONFIG_TUX_EXTCGI is not set
# CONFIG_TUX_EXTENDED_LOG is not set
# CONFIG_TUX_DEBUG is not set

.. and the ksymoops output is below.

Thanks,
Hamish
 

ksymoops 2.4.9 on ppc 2.4.23-tux3-np.  Options used
     -v /boot/vmlinux-2.4.23-tux3-np (specified)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.23-tux3-np/ (default)
     -m /boot/System.map-2.4.23-tux3-np (default)

No modules in ksyms, skipping objects
Warning (read_lsmod): no symbols in lsmod, is /proc/modules a valid lsmod file?
Warning (compare_maps): mismatch on symbol xchg_u32  , ksyms_base says c000d4f0, vmlinux says c00082d4.  Ignoring ksyms_base entry
Oops: kernel access of bad area, sig: 11
NIP: C012AA1C XER: 00000000 LR: C012A934 SP: CCEFFE20 REGS: cceffd70 TRAP: 0300    Not tainted
Using defaults from ksymoops -t elf32-powerpc -a powerpc:common
MSR: 00001032 EE: 0 PR: 0 FP: 0 ME: 1 IR/DR: 11
TASK = ccefe000[1033] 'tux' Last syscall: 225
last math cf320000 last altivec 00000000
GPR00: 1112BEEF CCEFFE20 CCEFE000 00000001 00009032 00000001 CFE32580 00000001
GPR08: C02075F0 00000024 00000000 00000000 00000001 1002066C 0000FFFE 7FFFFDF4
GPR16: 0000000A 00000000 00000000 00000000 00009032 0CEFFF40 00000000 C0005E68
GPR24: CCEF7004 00000000 C01E0000 00000024 00000001 00000040 CFEBF800 00000000
Call backtrace:
00001032 C012AFB0 C0128F4C C013867C C013AD00 C0005C1C 0FE0C1C4
100028A0 10003824 0FDDD100 00000000
Warning (Oops_read): Code line not seen, dumping what data is available


>>NIP; c012aa1c <__log_request+1e0/39c>   <=====

>>GPR8; c02075f0 <threadinfo+17c/1340>
>>GPR23; c0005e68 <ret_from_except+0/34>
>>GPR26; c01e0000 <malloc_cache+a8/c0>

Trace; 00001032 Before first symbol
Trace; c012afb0 <flush_request+b4/714>
Trace; c0128f4c <process_requests+10c/140>
Trace; c013867c <event_loop+dc/288>
Trace; c013ad00 <__sys_tux+5c4/e08>
Trace; c0005c1c <ret_from_syscall_1+0/b4>
Trace; 0fe0c1c4 Before first symbol
Trace; 100028a0 Before first symbol
Trace; 10003824 Before first symbol
Trace; 0fddd100 Before first symbol
Trace; 00000000 Before first symbol


3 warnings issued.  Results may not be reliable.


-- 
Hamish Moffatt VK3SB <hamish at debian.org> <hamish at cloud.net.au>