Debian on LaCie CloudBox (V2 Mini) (100GB) (88F6192 Rev 3 LE)

Hello again!

The same error again.

root@debian:~# flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb /dev/mtd0
Erasing blocks: 112/112 (100%)
Writing data: 448k/448k (100%)
Verifying data: 10k/448k (2%)File does not seem to match flash data. First mismatch at 0x00000000-0x00002800

Can I try something else? I still have the original hdd.

I have not rebooted yet.

root@debian:~# uptime
 06:09:06 up 3 days,  9:56,  3 users,  load average: 0.00, 0.00, 0.00

/Kenneth

Kenneth,

It might be a mismatch between stock u-boot and kernel, i.e. how both see SPI flash. Let me post an alternative way of flashing. Also I'll look back at stock-u-boot log and see if I need more info.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Kenneth,

With serial console connected (run picocom/minicom instead of kwboot). Power up and let it boot automatically. Interrupt stock u-boot countdown and

help
printenv
boot

If you can boot into the old HDD system this way, it would be good to see stock kernel booting.

And then post the entire serial console log here.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 1 time(s). Last edit at 10/08/2025 11:47PM by bodhi.

bodhi Wrote:
-------------------------------------------------------
> Kenneth,
>
> With serial console connected (run picocom/minicom
> instead of kwboot). Power up and let it boot
> automatically. Interrupt stock u-boot countdown

picocom doesn't show anything, and the old disk doesn't start.

However, I saved the outputs from help and printenv some time ago when I only had netconsole.

help:

Marvell>> help
help
?       - alias for 'help'
base    - print or set address offset
boot    - boot default, i.e., run 'bootcmd'
bootd   - boot default, i.e., run 'bootcmd'
bootext2    dev:boot_part1,boot_part2 addr boot_image linux_dev_name 
bootm   - boot application image from memory
bootp	- boot image via network using BootP/TFTP protocol
bubt	- Burn an image on the Boot Flash.
chpart	- change active partition
cmp     - memory compare
cmpm	- Compare Memory
coninfo - print console devices and information
cp      - memory copy
cpumap - Display CPU memory mapping settings.
crc32   - checksum calculation
date    - get/set/reset date & time
dclk	- Display the MV device CLKs.
dhcp	- invoke DHCP client to obtain IP/boot params
diskboot- boot from IDE device
echo    - echo args to console
eeprom  - EEPROM sub-system
erase   - erase FLASH memory
exit    - exit script
ext2load- load binary file from a Ext2 filesystem
ext2ls  - list files in a directory (default /)
factory    - reset to factory command
fatinfo - print information about filesystem
fatload - load binary file from a dos filesystem
fatls   - list files in a directory (default /)
fi	- Find value in the memory.
flinfo  - print FLASH memory information
fsinfo	- print information about filesystems
fsload	- load binary file from a filesystem image
g	- start application at cached address 'addr'(default addr 0x40000)
go      - start application at address 'addr'
help    - print online help
icrc32  - checksum calculation
ide     - IDE sub-system
iloop   - infinite loop on address range
imd     - i2c memory display
iminfo  - print header information for application image
imm[.b, .s, .w, .l]     - i2c memory modify (auto-incrementing)
imw     - memory write (fill)
inm     - memory modify (constant address)
iprobe  - probe to discover valid I2C chip addresses
ir	- reading and changing MV internal register values.
loop    - infinite loop on address range
ls	- list files in a directory (default /)
lump    - Wait for a LUMP
map	- Diasplay address decode windows
md      - memory display
me	- PCI master enable
mm      - memory modify (auto-incrementing)
mp	- map PCI BAR
mtdparts- define flash/nand partitions
mtest   - simple RAM test
mw      - memory write (fill)
nm      - memory modify (constant address)
pci     - list and access PCI Configuration Space
phyRead	- Read PCI-E Phy register
pciePhyWrite	- Write PCI-E Phy register
phyRead	- Read Phy register
phyWrite	- Write Phy register
ping	- send ICMP ECHO_REQUEST to network host
printenv- print environment variables
protect - enable or disable FLASH write protection
rarpboot- boot image via network using RARP/TFTP protocol
rcvr	- Satrt recovery process (Distress Beacon with TFTP server)
reset   - Perform RESET of the CPU
resetenv	- Return all environment variable to default.
run     - run commands in an environment variable
saveenv - save environment variables to persistent storage
se	- PCI Slave enable
setenv  - set environment variables
sflash	- read, write or erase the external SPI Flash.
sg	- scanning the PHYs status
sp	- Scan PCI bus.
test    - minimal test like /bin/sh
tftpboot- boot image via network using TFTP protocol
usb     - USB sub-system
usbboot - boot from USB device
version - print monitor version

printenv:

Marvell>> printenv
printenv
baudrate=115200
loads_echo=0
rootpath=/mnt/ARM_FS/
netmask=255.255.255.0
console=console=ttyS0,115200 mtdparts=spi_flash:0x80000@0(uboot)ro,0x300000@0x100000(root)
CASset=min
MALLOC_len=1
ethprime=egiga0
bootargs_root=root=/dev/nfs rw
bootargs_end=:::DB88FXX81:eth0:none
image_name=uImage
standalone=fsload 0x2000000 $(image_name);setenv bootargs $(console) root=/dev/mtdblock0 rw ip=$(ipaddr):$(serverip)$(bootargs_end) $(mvPhoneConfig); bootm 0x2000000;
ethmtu=1500
mvPhoneConfig=mv_phone_config=dev0:fxs,dev1:fxs
mvNetConfig=mv_net_config=(00:11:88:0f:62:81,0:1:2:3),mtu=1500
usb0Mode=host
yuk_ethaddr=00:00:00:EE:51:81
netretry=no
rcvrip=169.254.100.100
loadaddr=0x02000000
autoload=no
ethact=egiga0
stderr=serial
mainlineLinux=no
enaMonExt=no
enaCpuStream=no
enaWrAllo=no
pexMode=RC
disL2Cache=no
setL2CacheWT=yes
disL2Prefetch=yes
enaICPref=yes
enaDCPref=yes
sata_dma_mode=yes
netbsd_en=no
vxworks_en=no
disaMvPnp=no
enaAutoRecovery=yes
start_lump=lump 3
pre_lump=lump 1
resetdisk=ide reset
bootdelay=0
boot_fail=lump
kernel_addr=0x800000
productType_env=ASTON_KW
primaryPart=6
secondaryPart=A
boot_usb=usb start;usbboot 0x800000 0:1;bootm;
resetFlag_env=0
bootargs=console=ttyS0,115200 root=/dev/sda7 ro reset=0 productType=ASTON_KW
bootcmd=run disk_disk
boot_disk4=disk ${kernel_addr} 1:${primaryPart}; setenv rootfs /dev/sdb7;
boot_disk3=if disk ${kernel_addr} 0:${primaryPart}; then setenv rootfs /dev/sda7; else run boot_disk4; fi
boot_disk2=if disk ${kernel_addr} 1:${secondaryPart}; then setenv rootfs /dev/sdb7; else run boot_disk3; fi
boot_disk1=if disk ${kernel_addr} 0:${secondaryPart}; then setenv rootfs /dev/sda7; else run boot_disk2; fi
boot_disk=if test ${resetFlag_env} -eq 0; then run boot_disk1; else run boot_disk3; fi
disk_disk=run boot_disk; setenv bootargs console=ttyS0,115200 root=${rootfs} ro reset=${resetFlag_env} productType=${productType_env}; bootm ${kernel_addr};
ethaddr=00:D0:4B:91:3C:1D
ipaddr=192.168.1.72
ncip=192.168.1.6
serverip=192.168.1.6
stdin=nc
stdout=nc

> If you can boot into the old HDD system this way,
> it would be good to see stock kernel booting.
>
> And then post the entire serial console log here.

I can currently only boot with kwboot.
Are you interested in /var/log/messages ?

/Kenneth

Kenneth,

Quote

Are you interested in /var/log/messages ?

Syslog does not have info we need.

However, this protect command is what I'm looking for.

protect - enable or disable FLASH write protection

In order to test the protect command you need to run stock u-boot. If stock u-boot was saved before, we could kwboot that. My bad, I should have asked you to try dumping the mtd before. It migh be usable now.

For another Kirkwood box (Dell M300), this was needed:

protect off all

Quote

picocom doesn't show anything, and the old disk doesn't start.

Strange, if you can kwboot, then picocom should work. Make sure you use the same paramaters as in your kwboot command. Here is how I connect serial console.

alias myserial='picocom --b 115200 --f n --p n --d 8 /dev/ttyUSB0'

Try this: kwboot again, boot into Debian. And reboot. See if anything coming out of serial console afterward. kwboot can be used as terminal too.

======

I've seen this flashing problem before. Let me see if I can find it in my notes.

I will post another u-boot image for flashing. If it does not work in Debian, we'll try flashing it in u-boot.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

The problem I think is with how Debian SPI flash tool behaves.

Here is another tarball. It's basically the same build. But this time I changed the way the kwb image is packed.

uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251009.tar

sha256sum:
4cc3b3e9b174833426c0ca6044ea9c9118b66ddc44fa1cf6a07cf6bb6c750933

This tarball contains 4 files. Only the image is different from previous post (uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb).

uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment.img
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment
kirkwood-ns2mini.dtb

Back up mtd

dd if=/dev/mtd0 of=mtd0.netspace_mini_v2 bs=448k conv=sync
dd if=/dev/mtd1 of=mtd1.netspace_mini_v2 bs=64k conv=sync

And then flash uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb to mtd0.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 1 time(s). Last edit at 10/09/2025 04:12PM by bodhi.

bodhi Wrote:
-------------------------------------------------------
> Kenneth,
>
>

Quote

Are you interested in /var/log/messages
> ?

>
> Syslog does not have info we need.
>
OK, I was hoping that

Sep 20 18:38:00 CloudBox kernel: [    2.150000] SPI Serial flash detected @ 0xf4000000, 512KB (128sec x 4KB)

and

Sep 20 18:58:53 CloudBox kernel: [    0.230000] SPI_CS ....base f4000000, size  16MB

could help.

> However, this protect command is what I'm
> looking for.
>

> protect - enable or disable FLASH write
> protection
>

> In order to test the protect command you need to
> run stock u-boot. If stock u-boot was saved
> before, we could kwboot that. My bad, I should
> have asked you to try dumping the mtd before. It
> migh be usable now.
>
> For another Kirkwood box (Dell M300), this was
> needed:
>

> protect off all
>

>

LaCie accepted some changes to the envs but other was reverted back, at the next boot I guess.

>

Quote

picocom doesn't show anything, and the old
> disk doesn't start.

>
> Strange, if you can kwboot, then picocom should
> work. Make sure you use the same paramaters as in
> your kwboot command. Here is how I connect serial
> console.
>

> alias myserial='picocom --b 115200 --f n --p n --d
> 8 /dev/ttyUSB0'
>

>
> Try this: kwboot again, boot into Debian. And
> reboot. See if anything coming out of serial
> console afterward. kwboot can be used as terminal
> too.

Since kwboot works, the USB adapter and cables works.
I can have kwboot in 1 tab and picocom in another (the 2 shares the communication)
but after reboot nothing happens in either of them.

root@debian:~# reboot

Broadcast message from root@debian (pts/0) (Fri Oct 10 06:38:34 2025):

The system is going down for reboot NOW!
root@debian:~# Connection to 192.168.1.128 closed by remote host.
Connection to 192.168.1.128 closed.

Will now restart.
[  703.960394][ T1315] sd 0:0:0:0: [sda] Synchronizing SCSI cache
[  703.967231][ T1315] reboot: Restarting system
�	 � ��@

And the system stays off.

 $ ping 192.168.1.128
PING 192.168.1.128 (192.168.1.128) 56(84) bytes of data.
From 192.168.1.127 icmp_seq=9 Destination Host Unreachable
...

>
> I've seen this flashing problem before. Let me see
> if I can find it in my notes.
>
> I will post another u-boot image for flashing. If
> it does not work in Debian, we'll try flashing it
> in u-boot.

Good idea, to me it seems like the original is deleted or damaged.
Or can LaCie have used unexpected addresses?

Or should I, from the raspberry, try somtehing similar to the kwboot man page example?

kwboot -b u-boot-with-spl.kwb -B 5200000 -t /dev/ttyUSB0

The no-filled gave the same error.

root@debian:~# flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb /dev/mtd0
Erasing blocks: 93/93 (100%)
Writing data: 371k/371k (100%)
Verifying data: 10k/371k (2%)File does not seem to match flash data. First mismatch at 0x00000000-0x00002800

/Kenneth



Edited 1 time(s). Last edit at 10/10/2025 03:02AM by Kenneth68.

Kenneth,

> Since kwboot works, the USB adapter and cables
> works.
> I can have kwboot in 1 tab and picocom in another
> (the 2 shares the communication)

Nice setup!

> Good idea, to me it seems like the original is
> deleted or damaged.

It was corrupted during the flashcp. Again, my bad for forgetting to recommend backing up the original u-boot! I guess when I see kwboot working, I know it is always recoverable, so I was a bit careless.

> Or can LaCie have used unexpected addresses?

Not address. The flash is always at the same address, and the MTD layout determines where the parts are in the flash.

Debian kernel:

[    0.000000][    T0] Kernel command line: console=ttyS0,115200 root=LABEL=rootfs rootdelay=10 mtdparts=spi0.0:448K(u-boot),4K@448K(u-boot-env)

[    6.951345][    T1] spi-nor spi0.0: mx25l4005a (512 Kbytes)
[    6.957062][    T1] 2 cmdlinepart partitions found on MTD device spi0.0
[    6.963810][    T1] Creating 2 MTD partitions on "spi0.0":
[    6.969362][    T1] 0x000000000000-0x000000070000 : "u-boot"
[    6.976282][    T1] 0x000000070000-0x000000071000 : "u-boot-env"

Stock Kernel:

Sep 20 18:38:00 CloudBox kernel: [    2.150000] SPI Serial flash detected @ 0xf4000000, 512KB (128sec x 4KB)

The 0xf4000000 is HW address for the SPI flash. This is not printed out by the Debian kernel. And in userspace, we only acces the SPI flash through its MTD device.


> Or should I, from the raspberry, try somtehing
> similar to the kwboot man page example?
>

> kwboot -b u-boot-with-spl.kwb -B 5200000 -t
> /dev/ttyUSB0
>

That's not applicable to Kirkwood SoC boxes such as this one. The u-boot-with-spl.kwb image is built for SoCs that support SPL, e.g. Armada SoCs 370/375, 380/385/388, and many other newer SoCs.

> The no-filled gave the same error.
>

> root@debian:~# flashcp -v
> uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb
> /dev/mtd0
> Erasing blocks: 93/93 (100%)
> Writing data: 371k/371k (100%)
> Verifying data: 10k/371k (2%)File does not seem to
> match flash data. First mismatch at
> 0x00000000-0x00002800

OK. I'll write up instruction how to flash in serial console.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 1 time(s). Last edit at 10/10/2025 02:07PM by bodhi.

One last attempt in Debian.

As a reminder, the DTB in the rootfs must be the new one in the tarball
uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251009.tar.

Quote

1. Prepare the rootfs, check your SPI mtds, backup stock u-boot

1.1. Prepare the rootfs

Copy the DTB kirkwood-ns2mini.dtb from the tarball to the rootfs

cp -a /boot/dts/kirkwood-ns2mini.dtb /boot/dts/kirkwood-ns2mini.dtb.orig
cp -a kirkwood-ns2mini.dtb /boot/dts/

kwboot the same way with the u-boot image from uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251009.tar.

Now, after you logged into Debian do the flash this way:

flash_unlock /dev/mtd0
flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb /dev/mtd0

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Now I got a new error:

root@debian:~# flash_unlock /dev/mtd0
flash_unlock: error!: could not unlock device: /dev/mtd0

              error 95 (Operation not supported)

I guess this explains previous errors.

/Kenneth

Kenneth,

> Now I got a new error:
>
>

> root@debian:~# flash_unlock /dev/mtd0
> flash_unlock: error!: could not unlock device:
> /dev/mtd0
> 
>               error 95 (Operation not supported)
>

>

I meant the flash was already unlocked. Debian flash utils is not very friendly! it should have said "already unlocked".

OK let me post the u-boot flash instrution.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Flashing new u-boot for Netspace Mini v2 in serial console.

In this box, create a directory /uboot_ns2mini (or whatever name you'd like) where you keep the u-boot tarball uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251009.tar and have extracted its content in it.

kwboot the image uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb. Interrupt u-boot count down.

Load u-boot image

sata init
ext2load sata 0:1 0x800000 /uboot_ns2mini/uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb

expected output

380416 bytes read in ...

See the filesize variable

printenv filesize

expected output

filesize=5CE00

(Note that the filesize is displayed in hex instead of decimal like in other commands)

Bring up the SPI flash

sf probe

expected output

SF: Detected m25p32 with page size .....  total 512 KB

Erase the flash

sf erase 0 +$filesize

expected output

SF: 380416 bytes @ 0x0 Erased: OK

Write u-boot image

sf write 0x800000 0 $filesize

expected output

device 0 offset 0x0, size 0x80000
SF: 380416 bytes @ 0x0 Written: OK

If all goes well, go ahead and boot

boot

Once in Debian, reboot. Let it boot automatically. Make sure the u-boot banner showing the build date Oct 8.

Please post the entire flashing session in serial console for future instructional purpose.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 1 time(s). Last edit at 10/11/2025 01:07AM by bodhi.

There were some differences:

instead of m25p32, my board has a mx25l4005a

ns2-mini> sf probe
SF: Detected mx25l4005a with page size 256 Bytes, erase size 4 KiB, total 512 KiB

ns2-mini> sf erase 0 +$filesize
SF: 380928 bytes @ 0x0 Erased: OK

ns2-mini> sf write 0x800000 0 $filesize
device 0 offset 0x0, size 0x5ce00
SF: 380416 bytes @ 0x0 Written: OK

I can boot the system, but at reboot the hdd shuts off and stays off.
Pressing the power button makes no difference.

/Kenneth



Edited 1 time(s). Last edit at 10/11/2025 11:04AM by Kenneth68.

Kenneth,

Please post the entire flashing session log. Don't post excerpts.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Of course, sorry.

 99 % [............................                                          ]
Done
Finishing transfer
[Type Ctrl-\ + c to quit]


U-Boot 2025.07-tld-1 (Oct 08 2025 - 16:08:54 -0700) NS v2 Mini

SoC:   Kirkwood 88F6192_A1
MPP:   Variant 88F6192
DRAM:  128 MiB
Core:  15 devices, 12 uclasses, devicetree: separate
Loading Environment from SPIFlash... SF: Detected mx25l4005a with page size 256 Bytes, erase size 4 KiB, total 512 KiB
*** Warning - bad CRC, using default environment

In:    serial
Out:   serial
Err:   serial
Net:   
Warning: ethernet-controller@72000 (eth0) using random MAC address - 02:23:83:41:82:51
eth0: ethernet-controller@72000
Hit any key to stop autoboot:  0 
ns2-mini> sata init
ns2-mini> ext2load sata 0:1 0x800000 /uboot_ns2mini/uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb
380416 bytes read in 66 ms (5.5 MiB/s)
ns2-mini> printenv filesize
filesize=5ce00
ns2-mini> sf probe
SF: Detected mx25l4005a with page size 256 Bytes, erase size 4 KiB, total 512 KiB
ns2-mini> sf erase 0 +$filesize
SF: 380928 bytes @ 0x0 Erased: OK
ns2-mini> sf write 0x800000 0 $filesize
device 0 offset 0x0, size 0x5ce00
SF: 380416 bytes @ 0x0 Written: OK
ns2-mini> boot

/Kenneth

Kenneth,

> /uboot_ns2mini/uboot.2025.07-rc5-tld-1.netspace_mini_v2-no-filled.kwb
> 380416 bytes read in 66 ms (5.5 MiB/s)

OK.

> ns2-mini> printenv filesize
> filesize=5ce00

OK.

> ns2-mini> sf probe
> SF: Detected mx25l4005a with page size 256 Bytes,
> erase size 4 KiB, total 512 KiB

OK. The flash model is not important, but I will make note to change to the correct one in installation instruction

> ns2-mini> sf erase 0 +$filesize
> SF: 380928 bytes @ 0x0 Erased: OK

Now it is interesting.

> ns2-mini> sf write 0x800000 0 $filesize
> device 0 offset 0x0, size 0x5ce00
> SF: 380416 bytes @ 0x0 Written: OK

OK

====

I think there might be a wrong u-boot config that led to this and problem with flashing in Linux. Will be back.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 1 time(s). Last edit at 10/11/2025 05:58PM by bodhi.

Here is the new build.

Flashing new u-boot for Netspace Mini v2 in serial console.

uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251011.tar

sha256: 
da0a710e8d24b1aae78fa15c2a1d85fcb7c3dda505f90a86fac31a1b7f7ba57c

There are 5 files in this tarball.

uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2-no_filled.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment.img
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment
kirkwood-ns2mini.dtb

In this box, create a directory /uboot_ns2mini (or whatever name you'd like) where you keep the u-boot tarball uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251011.tar and have extracted its content in it.

kwboot the image uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb. Interrupt u-boot count down.

Load u-boot image

sata init
ext2load sata 0:1 0x800000 /uboot_ns2mini/uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb

expected output

458752 bytes read in ...

See the filesize variable

printenv filesize

expected output

filesize=70000

(Note that the filesize is displayed in hex instead of decimal like in other commands)

Bring up the SPI flash

sf probe

expected output

SF: Detected mx25l4005a with page size 256 Bytes, erase size 4 KiB, total 512 KiB

Erase the flash

sf erase 0 +$filesize

expected output

SF: 458752 bytes @ 0x0 Erased: OK

Write u-boot image

sf write 0x800000 0 $filesize

expected output

device 0 offset 0x0, size 0x80000
SF: 458752 bytes @ 0x0 Written: OK

If all goes well, go ahead and boot

boot

Once in Debian, reboot. Let it boot automatically. Make sure the u-boot banner showing the build date Oct 11.

Please post the entire flashing session in serial console for future instructional purpose.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

OK, here is the output from the latest attempt:

 99 % [...........................                                           ]
Done
Finishing transfer
[Type Ctrl-\ + c to quit]


U-Boot 2025.07-tld-1 (Oct 11 2025 - 16:10:44 -0700) NS v2 Mini

SoC:   Kirkwood 88F6192_A1
MPP:   Variant 88F6192
DRAM:  128 MiB
Core:  15 devices, 12 uclasses, devicetree: separate
Loading Environment from SPIFlash... SF: Detected mx25l4005a with page size 256 Bytes, erase size 4 KiB, total 512 KiB
*** Warning - bad CRC, using default environment

In:    serial
Out:   serial
Err:   serial
Net:   
Warning: ethernet-controller@72000 (eth0) using random MAC address - f2:6d:cc:14:2f:0d
eth0: ethernet-controller@72000
Hit any key to stop autoboot:  0 
ns2-mini> sata init
ns2-mini> ext2load sata 0:1 0x800000 /uboot_ns2mini/uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb
458752 bytes read in 77 ms (5.7 MiB/s)
ns2-mini> printenv filesize
filesize=70000
ns2-mini> sf probe
SF: Detected mx25l4005a with page size 256 Bytes, erase size 4 KiB, total 512 KiB
ns2-mini> sf erase 0 +$filesize
SF: 458752 bytes @ 0x0 Erased: OK
ns2-mini> sf write 0x800000 0 $filesize 
device 0 offset 0x0, size 0x70000
SF: 458752 bytes @ 0x0 Written: OK
ns2-mini> boot
Initializing devices...

It looks pretty much like before. Still doesn't reboot, can only be booted with kwboot.
Was it correct no not use the no-filled?

/Kenneth

Kenneth,

> It looks pretty much like before. Still doesn't
> reboot, can only be booted with kwboot.
> Was it correct no not use the no-filled?

Yes. It's intentional. I believe there is some SPI flash bug in u-boot so I thought erasing the whole partition of 448K would help. The no-filled image size is not aligned to the sector boundary (it should work fine though).

Let me search in the u-boot mailing list to see if I can find if there is recent SPI flash patches.

In the mean time, try this. In the steps above, change the erase from

Erase the flash 
sf erase 0 +$filesize

to

Erase the entire 512K flash 
sf erase 0 524288

See if it works!

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

bodhi Wrote:
-------------------------------------------------------

> In the mean time, try this. In the steps above,
> change the erase from
>

> Erase the flash 
> sf erase 0 +$filesize
>

> to
>

> Erase the entire 512K flash 
> sf erase 0 524288
>

That would be

sf erase 0 80000

> See if it works!
I see no difference, but it was worth a try.

But I wonder if the power button has to be configured in the software or firmware?
At the same time, I do not need to push it when kwbooting.

/Kenneth

Kenneth,

> > See if it works!
> I see no difference, but it was worth a try.

OK. Looks like we'll need to try another work around.

>
> But I wonder if the power button has to be
> configured in the software or firmware?

Yes. When you run Debian, the power button needs to be programmed to do what you want.

Wiki thread:

Quote
https://forum.doozan.com/read.php?2,23630

Key daemon (button control)

Multimedia keyboard daemon for Linux
How to use set up button to reboot or shutdown with Key Daemon (esekeyd)
Shutdown with Power button with regret option by esekeyd

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

I think I can see the problem with this flashing. Will try something tomorrow.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Kenneth,

Please try again!

=========

uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251015.tar

sha256: 
bdcca33488ba052c16a131e038ad710382c13ad178c4cddd4eab5230fb7f1272

There are 5 files in this tarball.

uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2.no_filled.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment.img
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment
kirkwood-ns2mini.dtb

Log into Debian, download the uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251015.tar tarball. Extract the files.

Check your SPI mtds.

cat /proc/mtd

Expected output

dev:    size   erasesize  name
mtd0: 00070000 00001000 "u-boot"
mtd1: 00001000 00001000 "u-boot-env"

Erase mtd0

dd if=/dev/zero ibs=1 count=458752 | tr "\000" "\377" > /dev/mtd0

Flash new u-boot image

flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb /dev/mtd0

Flash u-boot envs image

flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment.img  /dev/mtd1

Please post the entire flashing session log.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 1 time(s). Last edit at 10/16/2025 11:47PM by bodhi.

Hello Bodhi!

It's still the same.

root@debian:/uboot_ns2mini# sha256sum *tar
bdcca33488ba052c16a131e038ad710382c13ad178c4cddd4eab5230fb7f1272  uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251015.tar
root@debian:/uboot_ns2mini# tar -xvf uboot.2025.07-rc5-tld-1.netspace_mini_v2.bod
uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2.no_filled.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment.img
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment
kirkwood-ns2mini.dtb
root@debian:/uboot_ns2mini# cat /proc/mtd
dev:    size   erasesize  name
mtd0: 00070000 00001000 "u-boot"
mtd1: 00001000 00001000 "u-boot-env"
root@debian:/uboot_ns2mini# flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2.k
Erasing blocks: 112/112 (100%)
Writing data: 448k/448k (100%)
Verifying data: 10k/448k (2%)File does not seem to match flash data. First mismatch at 0x00000000-0x00002800

/Kenneth

Kenneth,

I've revised the post above to include the step to erase mtd0 before flashing. Please try again!

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Hello again, same thing again:

root@debian:/uboot_ns2mini# dd if=/dev/zero ibs=1 count=458752 | tr "\000" "\377" > /dev/mtd0
458752+0 records in
896+0 records out
458752 bytes (459 kB, 448 KiB) copied, 2.12367 s, 216 kB/s
root@debian:/uboot_ns2mini# flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb /dev/mtd0
Erasing blocks: 112/112 (100%)
Writing data: 448k/448k (100%)
Verifying data: 10k/448k (2%)File does not seem to match flash data. First mismatch at 0x00000000-0x00002800

Can we read the status of the write protection bits BP0, BP1 and BP2 mentioned in the datasheet?

/Kenneth

Kenneth,

The new u-boot unprotect the flash be default. So is the Linux kernel. But perhpaps there is some problem in either u-boot or Debian.

I'll post another way to flash this.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Kenneth,

Please try again this version.

I've found that this flash chip mx25l4005a was not considered when u-boot unprotects all SPI flash upon starting. Just that it was old and lacking maintenance. So I added it to the list. Let's see if it will help.

=========

uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251020.tar

sha256: 
a9dfb1cfb77dfc5465715a12a4202929a3a8dda6dc8715a7346d27d5979d2f49

There are 5 files in this tarball.

uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2_no_filled.kwb
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment.img
uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment
kirkwood-ns2mini.dtb

Download the uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251020.tar tarball.

Extract the files.

Copy the kirkwood-ns2mini.dtb to the rootfs /boot/dts. This is important because the SPI frequency 50Mhz is matching with the new u-boot.

Start kwboot with the new u-boot

Reboot this box.

Note the u-boot startup activity will show the SPI flash is unprotected.

Log into Debian.

Check your SPI mtds.

cat /proc/mtd

Expected output

dev:    size   erasesize  name
mtd0: 00070000 00001000 "u-boot"
mtd1: 00001000 00001000 "u-boot-env"

Erase mtd0

dd if=/dev/zero ibs=1 count=458752 | tr "\000" "\377" > /dev/mtd0

Flash new u-boot image

flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2.kwb /dev/mtd0

Flash u-boot envs image

flashcp -v uboot.2025.07-rc5-tld-1.netspace_mini_v2.environment.img  /dev/mtd1

Please post the entire flashing session log.

=====

Depending on whether your test is successful, I'll look at adding code to print out the protect bits.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

That was it! I can now reboot as expected.
Shut down, cut power and power up again also boots the box.

The output was:

root@debian:~# cat /proc/mtd
dev:    size   erasesize  name
mtd0: 00070000 00001000 "u-boot"
mtd1: 00001000 00001000 "u-boot-env"
root@debian:~# dd if=/dev/zero ibs=1 count=458752 | tr "\000" "\377" > /dev/mtd0
458752+0 records in
896+0 records out
458752 bytes (459 kB, 448 KiB) copied, 2.16658 s, 212 kB/s
root@debian:~# flashcp /uboot_ns2mini/uboot*v2.kwb /dev/mtd0
root@debian:~# flashcp /uboot_ns2mini/uboot*img /dev/mtd1

/Kenneth

Kenneth,

> That was it! I can now reboot as expected.
> Shut down, cut power and power up again also boots
> the box.
>
> The output was:
>

> root@debian:~# cat /proc/mtd
> dev:    size   erasesize  name
> mtd0: 00070000 00001000 "u-boot"
> mtd1: 00001000 00001000 "u-boot-env"
> root@debian:~# dd if=/dev/zero ibs=1 count=458752
> | tr "\000" "\377" > /dev/mtd0
> 458752+0 records in
> 896+0 records out
> 458752 bytes (459 kB, 448 KiB) copied, 2.16658 s,
> 212 kB/s
> root@debian:~# flashcp /uboot_ns2mini/uboot*v2.kwb
> /dev/mtd0
> root@debian:~# flashcp /uboot_ns2mini/uboot*img
> /dev/mtd1
>

>

Cool!

Could you flash it again using -v

flashcp -v /uboot_ns2mini/uboot*v2.kwb /dev/mtd0

I'm trying to see if flashcp is ouptputing bad error. If you see the same verification error, then flash again without -v like you did. And reboot.

And please also post the u-boot boot log (from banner until the kernel booting).

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)