[microblaze-uclinux] Fig Out - Now Kernel Panic (DDR is corrupt problem... )

[Greg Miller <apco25@xxxxxxxxx>]

All,

 

OK I figured out the issue. I had the DDR Data bits reversed (duh!). Now I have a different problem. I can now boot with the minimal Kernel from Steve... However when I did my own, I get the following Kernel Panic: (Please help)

 

Linux version 2.4.27-uc1 (root@localhost) (gcc version 2.95.3-4 Xilinx EDK 6.2.1                                                                               
 Build EDK_Gm.12.3) #2 Sun Jan 30 13:20:24 MST 2005                                                  
On node 0 totalpages: 8192                         
zone(0): 8192 pages.                   
zone(1): 0 pages.                
zone(2): 0 pages.                
CPU: MICROBLAZE              
Console: xmbserial on UARTLite
Kernel command line:
Calibrating delay loop... 4.33 BogoMIPS
Memory: 32MB = 32MB total
Memory: 31388KB available (580K code, 435K data, 32K init)
Dentry cache hash table entries: 4096 (order: 3, 32768 bytes)
Inode cache hash table entries: 2048 (order: 2, 16384 bytes)
Mount cache hash table entrie! s: 512 (order: 0, 4096 bytes)
Buffer cache hash table entries: 1024 (order: 0, 4096 bytes)
Page-cache hash table entries: 8192 (order: 3, 32768 bytes)
POSIX conformance testing by UNIFIX
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
Microblaze UARTlite serial driver version 1.00
ttyS0 at 0xffff2000 (irq = 1) is a Microblaze UARTlite
ttyS1 at 0xffff4000 (irq = 2) is a Microblaze UARTlite
Starting kswapd
xgpio #0 at 0xFFFF5000 mapped to 0xFFFF5000
Xilinx GPIO registered
uclinux[mtd]: RAM probe address=0x800c3888 size=0x3b000
uclinux[mtd]: root filesystem index=0
Kernel panic: VFS: Unable to mount root fs on 1f:00

Greg Miller <apco25@yahoo.com> wrote:

Hello,

I am using an Avnet board running MB. It is a similar setup the MB_Vanilla. I also have SRAM on the board that works fine with the Kernel Steve did (min). I am now trying to run out of DDR. When I do It will start to run, but I get a corrupted output:

4<i>uL nexsvor i.n.274u21-(cs nseasddrb@ae-ibn)r(rc  gecsvor i.n52394.X-l n                                                                               

i iDx E.K.6 2u1lB iDd_EmK1G..)2#36  1a2 Sat J9n12: 1445 :T4 U0C52n0nOd  o ezone(

5):zone(5): 5 pages<

>4

.4<o>ez5n:(5)p g sa

e4<.><

>4<6>CPU: MICROBLAZEConsole: xmbserial on UARTLit<

>4deKernel command line:<

>4n Calibrating delay loop... RMemory:P5 B6rM 5MBBB BM 5MBBB BM 5MBBB BM 5MBB7 M

18B MM 8MBMB MM 0284M 4MBMB MM 0284M 0284M 0284M 0284M 0284M 0284M 0284M 0284M 0

284M 0284M 0284M 0284M 0284M 0284M 0284M 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MB

MB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4M

BMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4

MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM

4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM

 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB M

M 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB

MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB

 MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBM

B MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MB

MB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4M

BMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4

MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBMB MM 4MBM M<m>re:o4y4 B0aKa lvbi

ea(l5 K4c6d ,o4e9  6aKad t2, 8nKti

i4<)>.k<r>ee nUl BtGsaa .l:b1c10

34<!>

***- Another Clue. If I take a regular test program and target the DDR space, it seems to run except the commands are 'reversed' :

ADS> help

Unrecognized command "ehpl".

ADS> cls

Unrecognized command "lc".

ADS>

It seems to reverse the commands as shown above. Also, if there is only a three letter command, it does not pick up the first letter. This same program runs out of BRAM just fine.

Now, I have tested the DDR and it passes in software with a random data test. I have also reduced the clock by half and it still does the same thing. Am I missing something simple? Has anyone else seen this?

 

Thanks,

Apco25