[beagleBoard] beagleBoard mmc recovery failed

Hi,

The nand flash is ecc failed when powering on, so I have to
do a mmc recovery according to the steps in the link:

  http://elinux.org/BeagleBoardRecovery#MMC_recovery

But each time doing mmc recovery, my beagle board hangs after
printing the messages:

Texas Instruments X-Loader 1.41
Starting on with MMC
Reading boot sector

717372 Bytes Read from MMC
Starting OS Bootloader from MMC...

I do not know how to recovery my board further, who can
help me to fix it?

Thanks!

The nand flash is ecc failed when powering on, so I have to
do a mmc recovery according to the steps in the link:

  BeagleBoardRecovery - eLinux.org

But each time doing mmc recovery, my beagle board hangs after
printing the messages:

Texas Instruments X-Loader 1.41
Starting on with MMC
Reading boot sector

717372 Bytes Read from MMC
Starting OS Bootloader from MMC...

I do not know how to recovery my board further, who can
help me to fix it?

Seems like you have a problem with uboot on you MMC card - The board as such
is starting fine, since it writes "Starting OS Bootloader from MMC..."

Try to double check the content on you MMC card - Might be you removed it
from your PC before all data was written to it thereby causing a corrupt
uboot?

Best regards - Good luck
  Søren

No, I can confirm the u-boot.bin is correct, also I try another MMC card and
the result is same. I guess nand flash init code stops at some point
because of
the nand ecc failure. If booting from nand flash, the uart message is :

        Texas Instruments X-Loader 1.41
        ECC Failed, page 0x00080700
        30 0 9f e5 0 c0 89 e5 f6 2d 0 eb 8c d0 89 e2 30 80 bd e8 e8 d5
e9 80 cc df e9 80 14
        e0 e9 80 38
       ~~~~ now bb hang.

How can I fix it? Who have found the problem and fixed it?

Thanks!

upgrade MLO to 1.42

Where can I get MLO 1.42 ?
Thanks!

leiming wrote:

>
> Where can I get MLO 1.42 ?
> Thanks!

Google Code Archive - Long-term storage for Google Code Project Hosting.

Which one is 1.42? v1, v2 or v3?
Thanks!

于 Sat, 18 Apr 2009 11:14:55 +0200
Koen Kooi <koen@beagleboard.org> 写道:

>
>>
>>> The nand flash is ecc failed when powering on, so I have to
>>> do a mmc recovery according to the steps in the link:
>>>
>>> BeagleBoardRecovery - eLinux.org
>>>
>>> But each time doing mmc recovery, my beagle board hangs after
>>> printing the messages:
>>>
>>> Texas Instruments X-Loader 1.41
>>> Starting on with MMC
>>> Reading boot sector
>>>
>>> 717372 Bytes Read from MMC
>>> Starting OS Bootloader from MMC...
>>>
>>> I do not know how to recovery my board further, who can
>>> help me to fix it?
>>
>> Seems like you have a problem with uboot on you MMC card - The
>> board as such
>> is starting fine, since it writes "Starting OS Bootloader from
>> MMC..."
>>
>> Try to double check the content on you MMC card - Might be you
>> removed it
>> from your PC before all data was written to it thereby causing a
>> corrupt
>> uboot?
>
> No, I can confirm the u-boot.bin is correct, also I try another
> MMC card and
> the result is same. I guess nand flash init code stops at some
> point because of
> the nand ecc failure. If booting from nand flash, the uart
> message is :
>
> Texas Instruments X-Loader 1.41
> ECC Failed, page 0x00080700
> 30 0 9f e5 0 c0 89 e5 f6 2d 0 eb 8c d0 89 e2 30 80 bd e8 e8
> d5 e9 80 cc df e9 80 14
> e0 e9 80 38
> ~~~~ now bb hang.
>
> How can I fix it?

upgrade MLO to 1.42

After upgrading mlo to 1.42, my board still cat't boot sucessfully,
the uart messages is :

  Texas Instruments X-Loader 1.4.2 (Feb 3 2009 - 15:34:17)
  Reading boot sector
  Loading u-boot.bin from mmc
  =========now the board seems died.

I am going to be crazy for the problem. Who can help me?

Thanks.

leiming wrote:

于 Sat, 18 Apr 2009 11:14:55 +0200
Koen Kooi <koen@beagleboard.org> 写道:

The nand flash is ecc failed when powering on, so I have to
do a mmc recovery according to the steps in the link:

      BeagleBoardRecovery - eLinux.org

But each time doing mmc recovery, my beagle board hangs after
printing the messages:

Texas Instruments X-Loader 1.41
Starting on with MMC
Reading boot sector

717372 Bytes Read from MMC
Starting OS Bootloader from MMC...

I do not know how to recovery my board further, who can
help me to fix it?

Seems like you have a problem with uboot on you MMC card - The board as such
is starting fine, since it writes "Starting OS Bootloader from MMC..."

Try to double check the content on you MMC card - Might be you removed it
from your PC before all data was written to it thereby causing a corrupt
uboot?

No, I can confirm the u-boot.bin is correct, also I try another
MMC card and
the result is same. I guess nand flash init code stops at some
point because of
the nand ecc failure. If booting from nand flash, the uart
message is :

       Texas Instruments X-Loader 1.41
       ECC Failed, page 0x00080700
       30 0 9f e5 0 c0 89 e5 f6 2d 0 eb 8c d0 89 e2 30 80 bd e8 e8
d5 e9 80 cc df e9 80 14
       e0 e9 80 38
      ~~~~ now bb hang.

How can I fix it?

upgrade MLO to 1.42

After upgrading mlo to 1.42, my board still cat't boot sucessfully,
the uart messages is :

  Texas Instruments X-Loader 1.4.2 (Feb 3 2009 - 15:34:17)
  Reading boot sector
  Loading u-boot.bin from mmc
  =========now the board seems died.

I am going to be crazy for the problem. Who can help me?

Take an other MMC card or delete your existing one and (re-) format it *exactly* like the manuals tell and put the binaries on it in the correct order. Try until it works :wink:

See

http://elinux.org/BeagleBoardFAQ#SD_Card

too.

Dirk

leiming said the following on 04/17/2009 08:54 PM:

Hi,

The nand flash is ecc failed when powering on, so I have to

http://groups.google.com/group/beagleboard/browse_thread/thread/d97b2152fe7185dd#
you may want to give this a shot if mmc recovery is tough...
Regards,
Nishanth Menon

Thanks for your reply.

I have tried the usb & uart recovery under the guide of the link above,
and still failed, so I guess hw of the board has been destructed.

I am still a bit surprised about this, after all x-loader can work
correctly, but u-boot have not been booted successfully and does not
print everything. Maybe I need to dig into u-boot source code to locate
what is wrong about the board.

This is the usb & uart recovery information.

[lm@linux-lm beagle_recover]$ sudo ./recover_beagle.sh /dev/ttyUSB0
[sudo] password for lm:
NOTE:
When you see the message "Waiting for USB device..", Connect USB Cable
to the board with the switch S1 pressed Keep S1 pressed for 1 second at
least after connecting usb cable. Waiting for USB device
vendorID=0x451and productID=0xD009: Dev #37: Texas Instruments -
OMAP3430
  - Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
lx
^C
[lm@linux-lm beagle_recover]$
[lm@linux-lm beagle_recover]$ sudo ./recover_beagle.sh /dev/ttyUSB0
NOTE:
When you see the message "Waiting for USB device..", Connect USB Cable
to the board with the switch S1 pressed Keep S1 pressed for 1 second at
least after connecting usb cable. Waiting for USB device
vendorID=0x451and productID=0xD009: Dev #39: Texas Instruments -
OMAP3430
  - Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
loadb -f /dev/ram0

## Ready for binary (kermit) download to 0x00000000 offset on /dev/ram0
device at 115200 bps... Match Found. Operation completed!
Downloading file: 100.000% completed(174632/174632 bytes)
File Download completed
Output:
go 0x80000000

## Starting application at 0x80000000 ...
/* not anything printed any longer */

Ming Lei said the following on 04/18/2009 10:11 AM:

Thanks for your reply.

I have tried the usb & uart recovery under the guide of the link above,
and still failed, so I guess hw of the board has been destructed.

I am still a bit surprised about this, after all x-loader can work
correctly, but u-boot have not been booted successfully and does not
print everything. Maybe I need to dig into u-boot source code to locate
what is wrong about the board.

This is the usb & uart recovery information.

[lm@linux-lm beagle_recover]$ sudo ./recover_beagle.sh /dev/ttyUSB0
[sudo] password for lm:
NOTE:
When you see the message "Waiting for USB device..", Connect USB Cable
to the board with the switch S1 pressed Keep S1 pressed for 1 second at
least after connecting usb cable. Waiting for USB device
vendorID=0x451and productID=0xD009: Dev #37: Texas Instruments -
OMAP3430
  - Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
lx
^C
  

No.. dont hit CTRL+C. wait for it to come back to $ prompt.. it takes a
few seconds to continue.. but have a little patience..

[lm@linux-lm beagle_recover]$
[lm@linux-lm beagle_recover]$ sudo ./recover_beagle.sh /dev/ttyUSB0
NOTE:
When you see the message "Waiting for USB device..", Connect USB Cable
to the board with the switch S1 pressed Keep S1 pressed for 1 second at
least after connecting usb cable. Waiting for USB device
vendorID=0x451and productID=0xD009: Dev #39: Texas Instruments -
OMAP3430
  - Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
loadb -f /dev/ram0

## Ready for binary (kermit) download to 0x00000000 offset on /dev/ram0
device at 115200 bps... Match Found. Operation completed!
Downloading file: 100.000% completed(174632/174632 bytes)
File Download completed
Output:
go 0x80000000

## Starting application at 0x80000000 ...
/* not anything printed any longer */
  

Now, this is weird.. yeah board specific debugging is required.. :frowning: I
have a rev C and a rev A board and it worked out of the box.. for both.
log as an example:
$ ./recover_beagle.sh /dev/ttyUSB0
NOTE:
When you see the message "Waiting for USB device..", Connect USB Cable
to the board with the switch S1 pressed
Keep S1 pressed for 1 second at least after connecting usb cable.
Waiting for USB device vendorID=0x451and productID=0xD009:
Dev #8: Texas Instruments - OMAP3430
  - Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
loadb -f /dev/ram0

## Ready for binary (kermit) download to 0x00000000 offset on /dev/ram0
device at 115200 bps...
Match Found. Operation completed!
Downloading file: 100.000% completed(174632/174632 bytes)
File Download completed
Output:
go 0x80000000

## Starting application at 0x80000000 ...

U-Boot 2009.03-00325-gf75a729 (Apr 09 2009 - 13:36:05)

OMAP3530-GP rev 2, CPU-OPP2 L3-165MHz

OMAP3 Beagle board + LPDDR/NAND

DRAM: 128 MB

NAND: 256 MiB

*** Warning - bad CRC or NAND, using default environment

In: serial
Match Found. Operation completed!
Output:
0
OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand erase

NAND erase: device 0 whole chip

Erasing at 0x0 -- 0% complete.
<snip>
Erasing at 0xfae0000 -- 98% complete.
Erasing at 0xfd60000 -- 99% complete.
Erasing at 0xffe0000 -- 100% complete.

OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand write.i 80000000 80000 80000

NAND write: device 0 offset 0x80000, size 0x80000

524288 bytes written: OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
loadb 80000000

## Ready for binary (kermit) download to 0x80000000 at 115200 bps...
Match Found. Operation completed!
Downloading file: 100.000% completed(20392/20392 bytes)
File Download completed
Output:
nandecc hw

HW ECC selected

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand write.i 80000000 0 80000

NAND write: device 0 offset 0x0, size 0x80000

524288 bytes written: OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
setenv bootcmd;nandecc sw;saveenv

SW ECC selected

Saving Environment to NAND...

Erasing Nand...

Erasing at 0x260000 -- 100% complete.

Writing to Nand... done

OMAP3 beagleboard.org #
Match Found. Operation completed!
$

Regards,
Nishanth Menon

Ming Lei said the following on 04/18/2009 10:11 AM:
> Thanks for your reply.
>
> I have tried the usb & uart recovery under the guide of the link
above,
> and still failed, so I guess hw of the board has been destructed.
>
> I am still a bit surprised about this, after all x-loader can work
> correctly, but u-boot have not been booted successfully and does not
> print everything. Maybe I need to dig into u-boot source code to
locate
> what is wrong about the board.
>
> This is the usb & uart recovery information.
>
> [lm@linux-lm beagle_recover]$ sudo ./recover_beagle.sh /dev/ttyUSB0
> [sudo] password for lm:
> NOTE:
> When you see the message "Waiting for USB device..", Connect USB
Cable
> to the board with the switch S1 pressed Keep S1 pressed for 1 second
at
> least after connecting usb cable. Waiting for USB device
> vendorID=0x451and productID=0xD009: Dev #37: Texas Instruments -
> OMAP3430
> - Serial Number: N/A
> Downloading file: 100.000% completed(38960/38960 bytes)
> ./target_files/u-boot-v2.bin downloaded successfully to target
> Output:
> lx
> ^C
>
No.. dont hit CTRL+C. wait for it to come back to $ prompt.. it takes a
few seconds to continue.. but have a little patience..
> [lm@linux-lm beagle_recover]$
> [lm@linux-lm beagle_recover]$ sudo ./recover_beagle.sh /dev/ttyUSB0
> NOTE:
> When you see the message "Waiting for USB device..", Connect USB
Cable
> to the board with the switch S1 pressed Keep S1 pressed for 1 second
at
> least after connecting usb cable. Waiting for USB device
> vendorID=0x451and productID=0xD009: Dev #39: Texas Instruments -
> OMAP3430
> - Serial Number: N/A
> Downloading file: 100.000% completed(38960/38960 bytes)
> ./target_files/u-boot-v2.bin downloaded successfully to target
> Output:
> loadb -f /dev/ram0
>
>
>
> ## Ready for binary (kermit) download to 0x00000000 offset on
/dev/ram0
> device at 115200 bps... Match Found. Operation completed!
> Downloading file: 100.000% completed(174632/174632 bytes)
> File Download completed
> Output:
> go 0x80000000
>
>
>
> ## Starting application at 0x80000000 ...
> /* not anything printed any longer */
>
Now, this is weird.. yeah board specific debugging is required.. :frowning: I
have a rev C and a rev A board and it worked out of the box.. for both.
log as an example:
$ ./recover_beagle.sh /dev/ttyUSB0
NOTE:
When you see the message "Waiting for USB device..", Connect USB Cable
to the board with the switch S1 pressed
Keep S1 pressed for 1 second at least after connecting usb cable.
Waiting for USB device vendorID=0x451and productID=0xD009:
Dev #8: Texas Instruments - OMAP3430
  - Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
loadb -f /dev/ram0

## Ready for binary (kermit) download to 0x00000000 offset on /dev/ram0
device at 115200 bps...
Match Found. Operation completed!
Downloading file: 100.000% completed(174632/174632 bytes)
File Download completed
Output:
go 0x80000000

## Starting application at 0x80000000 ...

U-Boot 2009.03-00325-gf75a729 (Apr 09 2009 - 13:36:05)

OMAP3530-GP rev 2, CPU-OPP2 L3-165MHz

OMAP3 Beagle board + LPDDR/NAND

DRAM: 128 MB

NAND: 256 MiB

*** Warning - bad CRC or NAND, using default environment

In: serial
Match Found. Operation completed!
Output:
0
OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand erase

NAND erase: device 0 whole chip

Erasing at 0x0 -- 0% complete.
<snip>
Erasing at 0xfae0000 -- 98% complete.
Erasing at 0xfd60000 -- 99% complete.
Erasing at 0xffe0000 -- 100% complete.

OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand write.i 80000000 80000 80000

NAND write: device 0 offset 0x80000, size 0x80000

524288 bytes written: OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
loadb 80000000

## Ready for binary (kermit) download to 0x80000000 at 115200 bps...
Match Found. Operation completed!
Downloading file: 100.000% completed(20392/20392 bytes)
File Download completed
Output:
nandecc hw

HW ECC selected

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand write.i 80000000 0 80000

NAND write: device 0 offset 0x0, size 0x80000

524288 bytes written: OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
setenv bootcmd;nandecc sw;saveenv

SW ECC selected

Saving Environment to NAND...

Erasing Nand...

Erasing at 0x260000 -- 100% complete.

Writing to Nand... done

OMAP3 beagleboard.org #
Match Found. Operation completed!
$

You might have a problem with your external memory (the POP NAND and DDR). I
this case the x-loader will work with no problems (since it's only using
internal memory), while U-boot will/might fail (since it's run from external
memory)...

Does anybody have a validation test for external memory capable of running
from internal memory only?

Best regards
  Søren

Op 18 apr 2009, om 20:10 heeft Søren Steen Christensen het volgende geschreven:

Ming Lei said the following on 04/18/2009 10:11 AM:

Thanks for your reply.

I have tried the usb & uart recovery under the guide of the link

above,

and still failed, so I guess hw of the board has been destructed.

I am still a bit surprised about this, after all x-loader can work
correctly, but u-boot have not been booted successfully and does not
print everything. Maybe I need to dig into u-boot source code to

locate

what is wrong about the board.

This is the usb & uart recovery information.

[lm@linux-lm beagle_recover]$ sudo ./recover_beagle.sh /dev/ttyUSB0
[sudo] password for lm:
NOTE:
When you see the message "Waiting for USB device..", Connect USB

Cable

to the board with the switch S1 pressed Keep S1 pressed for 1 second

at

least after connecting usb cable. Waiting for USB device
vendorID=0x451and productID=0xD009: Dev #37: Texas Instruments -
OMAP3430
- Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
lx
^C

No.. dont hit CTRL+C. wait for it to come back to $ prompt.. it takes a
few seconds to continue.. but have a little patience..

[lm@linux-lm beagle_recover]$
[lm@linux-lm beagle_recover]$ sudo ./recover_beagle.sh /dev/ttyUSB0
NOTE:
When you see the message "Waiting for USB device..", Connect USB

Cable

to the board with the switch S1 pressed Keep S1 pressed for 1 second

at

least after connecting usb cable. Waiting for USB device
vendorID=0x451and productID=0xD009: Dev #39: Texas Instruments -
OMAP3430
- Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
loadb -f /dev/ram0

## Ready for binary (kermit) download to 0x00000000 offset on

/dev/ram0

device at 115200 bps... Match Found. Operation completed!
Downloading file: 100.000% completed(174632/174632 bytes)
File Download completed
Output:
go 0x80000000

## Starting application at 0x80000000 ...
/* not anything printed any longer */

Now, this is weird.. yeah board specific debugging is required.. :frowning: I
have a rev C and a rev A board and it worked out of the box.. for both.
log as an example:
$ ./recover_beagle.sh /dev/ttyUSB0
NOTE:
When you see the message "Waiting for USB device..", Connect USB Cable
to the board with the switch S1 pressed
Keep S1 pressed for 1 second at least after connecting usb cable.
Waiting for USB device vendorID=0x451and productID=0xD009:
Dev #8: Texas Instruments - OMAP3430
- Serial Number: N/A
Downloading file: 100.000% completed(38960/38960 bytes)
./target_files/u-boot-v2.bin downloaded successfully to target
Output:
loadb -f /dev/ram0

## Ready for binary (kermit) download to 0x00000000 offset on /dev/ram0
device at 115200 bps...
Match Found. Operation completed!
Downloading file: 100.000% completed(174632/174632 bytes)
File Download completed
Output:
go 0x80000000

## Starting application at 0x80000000 ...

U-Boot 2009.03-00325-gf75a729 (Apr 09 2009 - 13:36:05)

OMAP3530-GP rev 2, CPU-OPP2 L3-165MHz

OMAP3 Beagle board + LPDDR/NAND

DRAM: 128 MB

NAND: 256 MiB

*** Warning - bad CRC or NAND, using default environment

In: serial
Match Found. Operation completed!
Output:
0
OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand erase

NAND erase: device 0 whole chip

Erasing at 0x0 -- 0% complete.
<snip>
Erasing at 0xfae0000 -- 98% complete.
Erasing at 0xfd60000 -- 99% complete.
Erasing at 0xffe0000 -- 100% complete.

OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand write.i 80000000 80000 80000

NAND write: device 0 offset 0x80000, size 0x80000

524288 bytes written: OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
loadb 80000000

## Ready for binary (kermit) download to 0x80000000 at 115200 bps...
Match Found. Operation completed!
Downloading file: 100.000% completed(20392/20392 bytes)
File Download completed
Output:
nandecc hw

HW ECC selected

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
nand write.i 80000000 0 80000

NAND write: device 0 offset 0x0, size 0x80000

524288 bytes written: OK

OMAP3 beagleboard.org #
Match Found. Operation completed!
Output:
setenv bootcmd;nandecc sw;saveenv

SW ECC selected

Saving Environment to NAND...

Erasing Nand...

Erasing at 0x260000 -- 100% complete.

Writing to Nand... done

OMAP3 beagleboard.org #
Match Found. Operation completed!
$

You might have a problem with your external memory (the POP NAND and DDR). I
this case the x-loader will work with no problems (since it's only using
internal memory), while U-boot will/might fail (since it's run from external
memory)...

Does anybody have a validation test for external memory capable of running
from internal memory only?

'memtest' in uboot2?

regards,

Koen

Koen Kooi said the following on 04/18/2009 01:18 PM:

'memtest' in uboot2?

Should be easy to enable from menuconfig if there is one..
Regards,
Nishanth Menon

Yes , there is memtest in uboot-v2. I try to config ,compile and
download uboot2, but
not succeed to run it: there is anything shown in console after
downloading uboot.bin to
beagle. I guess the config is not valid.

Who can post a good beagle board .config for uboot v2?

Thanks!

"make omap3530_beagle_per_uart_defconfig" does work.

memtest has not passed indeed, maybe I need to replace omap3
chip.

于 Sun, 19 Apr 2009 12:53:34 +0800
Ming Lei <tom.leiming@gmail.com> 写道:

>>
>> Koen Kooi said the following on 04/18/2009 01:18 PM:
>>>
>>>
>>> 'memtest' in uboot2?
>>>
>> Should be easy to enable from menuconfig if there is one..
>
> Yes , there is memtest in uboot-v2. I try to config ,compile and
> download uboot2, but
> not succeed to run it: there is anything shown in console after
> downloading uboot.bin to
> beagle. I guess the config is not valid.
>
> Who can post a good beagle board .config for uboot v2?

"make omap3530_beagle_per_uart_defconfig" does work.

memtest has not passed indeed, maybe I need to replace omap3
chip.

It seems there is something wrong about sdram data line D18 or address
line A18. Is it possible to fix it by choosing other sdram
configuration in initlization code?

Thanks!

X-load Beagle>icache
icache is enabled
X-load Beagle>icache 0
X-load Beagle>icache
icache is disabled
X-load Beagle>mtest 0x80001000 0x80001040
Testing 0x80001000 ... 0x80001040:
Pattern 0xFFFFF00F Writing... Reading...
X-load Beagle>mtest 0x80001000 0x80002000
Testing 0x80001000 ... 0x80002000:
Pattern 0xFFFFFFFF Writing... Reading...
Mem error @ 0x800010A0: found 0xFFFBFFD7, expected 0xFFFFFFD7

Mem error @ 0x80001110: found 0xFFFBFFBB, expected 0xFFFFFFBB

Mem error @ 0x80001174: found 0xFFFBFFA2, expected 0xFFFFFFA2

Mem error @ 0x800011D4: found 0xFFFBFF8A, expected 0xFFFFFF8A

Mem error @ 0x80001248: found 0xFFFBFF6D, expected 0xFFFFFF6D

Mem error @ 0x800012A8: found 0xFFFBFF55, expected 0xFFFFFF55

Mem error @ 0x800012FC: found 0xFFFBFF40, expected 0xFFFFFF40

Mem error @ 0x8000135C: found 0xFFFBFF28, expected 0xFFFFFF28

Mem error @ 0x800013B4: found 0xFFFBFF12, expected 0xFFFFFF12

Mem error @ 0x80001430: found 0xFFFBFEF3, expected 0xFFFFFEF3

Mem error @ 0x800014A8: found 0xFFFBFED5, expected 0xFFFFFED5

Mem error @ 0x80001514: found 0xFFFBFEBA, expected 0xFFFFFEBA

Mem error @ 0x80001578: found 0xFFFBFEA1, expected 0xFFFFFEA1

Mem error @ 0x800015E0: found 0xFFFBFE87, expected 0xFFFFFE87

Mem error @ 0x80001648: found 0xFFFBFE6D, expected 0xFFFFFE6D

Mem error @ 0x800016B0: found 0xFFFBFE53, expected 0xFFFFFE53

Mem error @ 0x80001718: found 0xFFFBFE39, expected 0xFFFFFE39

Mem error @ 0x80001770: found 0xFFFBFE23, expected 0xFFFFFE23

Mem error @ 0x800017D0: found 0xFFFBFE0B, expected 0xFFFFFE0B

Mem error @ 0x80001848: found 0xFFFBFDED, expected 0xFFFFFDED

Mem error @ 0x800018BC: found 0xFFFBFDD0, expected 0xFFFFFDD0

Mem error @ 0x80001930: found 0xFFFBFDB3, expected 0xFFFFFDB3

Mem error @ 0x80001998: found 0xFFFBFD99, expected 0xFFFFFD99

Mem error @ 0x800019FC: found 0xFFFBFD80, expected 0xFFFFFD80

Mem error @ 0x80001A68: found 0xFFFBFD65, expected 0xFFFFFD65

Mem error @ 0x80001AC8: found 0xFFFBFD4D, expected 0xFFFFFD4D

Mem error @ 0x80001B30: found 0xFFFBFD33, expected 0xFFFFFD33

Mem error @ 0x80001B90: found 0xFFFBFD1B, expected 0xFFFFFD1B

Mem error @ 0x80001BE8: found 0xFFFBFD05, expected 0xFFFFFD05

Mem error @ 0x80001C5C: found 0xFFFBFCE8, expected 0xFFFFFCE8

Mem error @ 0x80001CD0: found 0xFFFBFCCB, expected 0xFFFFFCCB

Mem error @ 0x80001D48: found 0xFFFBFCAD, expected 0xFFFFFCAD

Mem error @ 0x80001DAC: found 0xFFFBFC94, expected 0xFFFFFC94

Mem error @ 0x80001E18: found 0xFFFBFC79, expected 0xFFFFFC79

Mem error @ 0x80001E88: found 0xFFFBFC5D, expected 0xFFFFFC5D

Mem error @ 0x80001EE8: found 0xFFFBFC45, expected 0xFFFFFC45

Mem error @ 0x80001F50: found 0xFFFBFC2B, expected 0xFFFFFC2B

Mem error @ 0x80001FB0: found 0xFFFBFC13, expected 0xFFFFFC13
Pattern 0xFFFFFFFE Writing... Reading...
Mem error @ 0x80001098: found 0xFFFBFFD8, expected 0xFFFFFFD8

Mem error @ 0x8000110C: found 0xFFFBFFBB, expected 0xFFFFFFBB

Mem error @ 0x80001174: found 0xFFFBFFA1, expected 0xFFFFFFA1

Mem error @ 0x800011E0: found 0xFFFBFF86, expected 0xFFFFFF86

Mem error @ 0x8000124C: found 0xFFFBFF6B, expected 0xFFFFFF6B

Mem error @ 0x800012AC: found 0xFFFBFF53, expected 0xFFFFFF53

Mem error @ 0x8000130C: found 0xFFFBFF3B, expected 0xFFFFFF3B

Mem error @ 0x80001368: found 0xFFFBFF24, expected 0xFFFFFF24

Mem error @ 0x800013C4: found 0xFFFBFF0D, expected 0xFFFFFF0D

Mem error @ 0x8000143C: found 0xFFFBFEEF, expected 0xFFFFFEEF

Mem error @ 0x800014AC: found 0xFFFBFED3, expected 0xFFFFFED3

Mem error @ 0x80001520: found 0xFFFBFEB6, expected 0xFFFFFEB6

Mem error @ 0x8000158C: found 0xFFFBFE9B, expected 0xFFFFFE9B

Mem error @ 0x800015F0: found 0xFFFBFE82, expected 0xFFFFFE82

Mem error @ 0x80001660: found 0xFFFBFE66, expected 0xFFFFFE66

Mem error @ 0x800016C0: found 0xFFFBFE4E, expected 0xFFFFFE4E

Mem error @ 0x8000171C: found 0xFFFBFE37, expected 0xFFFFFE37

Mem error @ 0x80001774: found 0xFFFBFE21, expected 0xFFFFFE21

Mem error @ 0x800017D0: found 0xFFFBFE0A, expected 0xFFFFFE0A

Mem error @ 0x8000184C: found 0xFFFBFDEB, expected 0xFFFFFDEB

Mem error @ 0x800018C8: found 0xFFFBFDCC, expected 0xFFFFFDCC

Mem error @ 0x80001944: found 0xFFFBFDAD, expected 0xFFFFFDAD

Mem error @ 0x800019AC: found 0xFFFBFD93, expected 0xFFFFFD93

Mem error @ 0x80001A18: found 0xFFFBFD78, expected 0xFFFFFD78

Mem error @ 0x80001A80: found 0xFFFBFD5E, expected 0xFFFFFD5E

Mem error @ 0x80001AD8: found 0xFFFBFD48, expected 0xFFFFFD48

Mem error @ 0x80001B38: found 0xFFFBFD30, expected 0xFFFFFD30

Mem error @ 0x80001B94: found 0xFFFBFD19, expected 0xFFFFFD19

Mem error @ 0x80001BF0: found 0xFFFBFD02, expected 0xFFFFFD02

Mem error @ 0x80001C68: found 0xFFFBFCE4, expected 0xFFFFFCE4

Mem error @ 0x80001CCC: found 0xFFFBFCCB, expected 0xFFFFFCCB

Mem error @ 0x80001D44: found 0xFFFBFCAD, expected 0xFFFFFCAD

Mem error @ 0x80001DAC: found 0xFFFBFC93, expected 0xFFFFFC93

Mem error @ 0x80001E18: found 0xFFFBFC78, expected 0xFFFFFC78

Mem error @ 0x80001E88: found 0xFFFBFC5C, expected 0xFFFFFC5C

Mem error @ 0x80001EEC: found 0xFFFBFC43, expected 0xFFFFFC43

Mem error @ 0x80001F4C: found 0xFFFBFC2B, expected 0xFFFFFC2B

Mem error @ 0x80001FAC: found 0xFFFBFC13, expected 0xFFFFFC13

Ming Lei said the following on 04/19/2009 12:16 AM:

It seems there is something wrong about sdram data line D18 or address
line A18. Is it possible to fix it by choosing other sdram
configuration in initlization code?
  

These are POP(Package On Package). you may need to check on RMA
procedure - the device probably wont work in other configurations
Regards,
Nishanth Menon

What is RMA procedure? Would you give me a detailed introduction?
BTW: the board is B5 revision.

Thanks!