[U-Boot-Users] Re: Kernel Panic

[Will Haines <will.haines@embedded-sys.com>]

Wolfgang Denk <wd <at> denx.de> writes:

> 
> Sounds like a memory probem to me. Are you absolutely sure  that  the
> kernel  sees  the  correct  size  of  your RAM, and that all of it is
> actually working, without any gaps or overlaps or mirroring?
> 

If you mean the UPM RAM words, my setup perfectly matches the table in the
manual.  Also, I have 64 Mb of ram, and u-boot recognizes it as such.  As 
for all of it working, I am going to run the u-boot memory test on the whole 
thing to see if there are any problems.  Do you have any other suggestions 
that might make it more apparent if my memory is not correct?

> 
> It may have to do  with  U-Boot,  more  precisely  with  your  memory
> initialization.
> 

Am I right in saying that the following problem is indicative of a memory
problem?  When I plugged the BDI into the board in an attempt to figure out
where the problem occured exactly, I got the following output:

...

VFS: Mounted root (ext2 filesystem).
Oops: Kernel Mode Software FPU Emulation, sig: 8
NIP: C0048000 XER: 20000000 LR: C0048224 SP: C3A25E30 REGS: c3a25d80 
TRAP: 1000    Not tainted
MSR: 00009032 EE: 1 PR: 0 FP: 0 ME: 1 IR/DR: 11
TASK = c3a24000[1] 'swapper' Last syscall: 6
last math 00000000 last altivec 00000000
GPR00: 00000000 C3A25E30 C3A24000 C3A15960 C3A15960 C00FE4CC 00000000 
C3BEB1A8
GPR08: C3A30580 00000000 C3A15970 C3A15970 22008082 00000000 03FFE000 
00000001
GPR16: 007FFF40 00000000 00000001 007FFF00 03FF8B5C FFFFFFFF 00000000 
03BCF308
GPR24: 00000000 C3A25F60 C00FE4CC C0110000 C3A306D0 C0111314 C3A1B800 
C3A15960
Call backtrace:
C3BEC5D0 C0065DC8 C0045BBC C00360A0 C0036758 C00367D0 C004AAC0
C004AE38 C011DF84 C00026A0 C0002280 C0006B60
Kernel panic: Attempted to kill init!
 <0>Rebooting in 180 seconds..

I read in the FAQ that memory misconfigured PowerPCs tend to have floating 
point exceptions.  I am not able to look at the memory indicated in the call
backtrace without killing the board.  Is there some other way to better
understand this backtrace?

I am starting to be convinced that somehow to SDRAM is not initialized
correctly, but past the UPM RAM words and writing the MAR and MCR, I am not 
sure what else has to be done to initialize it correctly.  Any guidance 
would be greatly appreciated.  I'm very new to this, and your help has
been invaluable.

Thanks again,
Will Haines