Buffalo NAS-Central Forums

Welcome to the Linkstation Wiki community
It is currently Tue Jan 23, 2018 4:52 am

All times are UTC+01:00




Post new topic  Reply to topic  [ 58 posts ]  Go to page Previous 1 2 3 4
Author Message
PostPosted: Mon Apr 06, 2009 9:36 am 
Offline
Site Admin
User avatar

Joined: Sun Jul 17, 2005 4:34 pm
Posts: 5332
You mean business :)

0. Understood, will use your comands

1. Mounting the HDD from uboot works. And as I can grab everything from my outside uboot aswell:

# cat /etc/default/avr_evtd | grep =
LOG=/var/log
DEBUG=ON
EMMODE=NO
TIMER=OFF
#DEVICE=/dev/ttyS1
SUN-SAT=OFF=01:15,ON=06:20
SHUTDOWN=
POWERON=
DISKCHECK=90
ROOT=hda1
WORK=hda3
REFRESH=40
HOLD=3
DISKNAG=ON
FANSTOP=15

3. Again, from the running 2.6.25.20: /proc/mtd is identical to yours.

7. You can try the blitzer on your LS1 any time, it will d'l and fake the installation of Bauldrick's version. To actually make it flash, uncomment the dd line (line 84)

http://hvkls.dyndns.org/downloads/tmp/flashinstaller.sh


Will need some time for 2, 4-6. I'll start with 4-5 (flashing the precompiled uboot and erasing the variables). I'm somewhat afraid of the Christmas Tree Mode.


Top
   
PostPosted: Mon Apr 06, 2009 4:29 pm 
Offline
Site Admin
User avatar

Joined: Mon Aug 28, 2006 1:15 am
Posts: 2606
hmmm... there were apparently some improvements in code between uboot 1.1.4 and 1.2.0. 1.1.4 seems sooooo slllowwww, in terms of netconsole speed. 1.2.0 is just about instantaneous.

BTW, maybe another few things we could look at:

3.5 While in uboot, try running writeok and writeng, to see if they have any effect. I'm thinking that they only have an effect on running the 2.4 kernels, but I could be wrong on that.

You could run writeok, and then attempt to boot. Same for writeng (I'm thinking that writeng is used to get into EM mode, writeok to get out of it...)

Code:
=> run writeok
run writeok
Un-Protected 1 sectors

Flash erase: first = 55 @ 0xfff70000
             last  = 55 @ 0xfff70000
Flash erase: Done
Erased 1 sectors
Copy to Flash... done


3.6 We can check to see if there is something amiss in terms of uboot and how it is reading the board and flash info:

While in uboot, try running flinfo and bdinfo:
Code:
flinfo
bdinfo


Quote:
=> bdinfo
bdinfo
memstart = 0x00000000
memsize = 0x04000000
flashstart = 0xFFC00000
flashsize = 0x00400000
flashoffset = 0x00000000
sramstart = 0x00000000
sramsize = 0x00000000
bootflags = 0x00000001
intfreq = 196.608 MHz
busfreq = 98.304 MHz
ethaddr = 00:07:40:A4:XX:XX
IP addr = 192.168.11.150
baudrate = 57600 bps
=> flinfo
flinfo

Bank # 1: Flash chip: FUJITSU MBM29PL32TM

CFI Query Results:
Primary Vendor Command Set: 0x0002 (AMD/Fujitsu Standard)
Primary Algorithm Table at 0x0040
Alternate Vendor Command Set: 0x0000 (None)
No Alternate Algorithm Table
Vcc Min.: 2.7 V
Vcc Max.: 3.6 V
No Vpp line
Typical byte/word write timeout: 128 us
Maximum byte/word write timeout: 256 us
Typical full buffer write timeout: 128 us
Maximum full buffer write timeout: 4096 us
Typical block erase timeout: 1024 ms
Maximum block erase timeout: 16384 ms
Chip erase not supported
Device size: 0x400000 bytes (4 MiB)
Flash Device Interface description: 0x0002
- x8 / x16 via BYTE# with asynchronous interface
Max. bytes in buffer write: 32
Number of Erase Block Regions: 2

Actual values used by U-Boot:
Word write timeout: 2 ms
Buffer write timeout: 5 ms
Sector erase timeout: 16384 ms

Size: 4 MiB in 71 Sectors
Sector Start Addresses:
FFC00000 (RO) FFC10000 (RO) FFC20000 (RO) FFC30000 (RO) FFC40000 (RO)
FFC50000 (RO) FFC60000 (RO) FFC70000 (RO) FFC80000 (RO) FFC90000 (RO)
FFCA0000 (RO) FFCB0000 (RO) FFCC0000 (RO) FFCD0000 (RO) FFCE0000 (RO)
FFCF0000 (RO) FFD00000 (RO) FFD10000 (RO) FFD20000 (RO) FFD30000 (RO)
FFD40000 (RO) FFD50000 (RO) FFD60000 (RO) FFD70000 (RO) FFD80000 (RO)
FFD90000 (RO) FFDA0000 (RO) FFDB0000 (RO) FFDC0000 (RO) FFDD0000 (RO)
FFDE0000 (RO) FFDF0000 (RO) FFE00000 (RO) FFE10000 (RO) FFE20000 (RO)
FFE30000 (RO) FFE40000 (RO) FFE50000 (RO) FFE60000 (RO) FFE70000 (RO)
FFE80000 (RO) FFE90000 (RO) FFEA0000 (RO) FFEB0000 (RO) FFEC0000 (RO)
FFED0000 (RO) FFEE0000 (RO) FFEF0000 (RO) FFF00000 (RO) FFF10000 (RO)
FFF20000 (RO) FFF30000 (RO) FFF40000 (RO) FFF50000 (RO) FFF60000 (RO)
FFF70000 (RO) FFF80000 (RO) FFF90000 (RO) FFFA0000 (RO) FFFB0000 (RO)
FFFC0000 (RO) FFFD0000 (RO) FFFE0000 (RO) FFFF0000 (RO) FFFF2000 (RO)
FFFF4000 (RO) FFFF6000 (RO) FFFF8000 (RO) FFFFA000 (RO) FFFFC000 (RO)
FFFFE000 (RO)


Just curious, is your ethaddr value correct? ie., does it match the MAC/etheradd on the stamp of your LS1's box (on back of box)?

_________________
LS1 (foonas, nfs, Tranmission & p910nd print server, Firefly for my Roku)
LS-HG500 (Lenny)
Various LS-Pros v1,v2 (unbricked w/ serial & jtag)
KuroPro, LS2 & KuroHG (foonas)
Working on davysweather.dyndns.org lately...

=> wooohooo!
wooohooo!
Unknown command 'wooohooo!' - try 'help'


Top
   
PostPosted: Tue Apr 07, 2009 5:35 am 
Offline
Site Admin
User avatar

Joined: Sun Jul 17, 2005 4:34 pm
Posts: 5332
3.5 "run writeok" has no visible effect.

3.6 will deal with that tomorrow

4. reflashing uboot itself didn't cure it.

5. erasing the env vars gives me the same output as you, but things still don't work

6. I'll try the old uboot tomorrow.

This is sooo strange. I wonder if something needs to be executed after the bootm command, and if so, how to do it?


Top
   
PostPosted: Tue Apr 07, 2009 6:12 am 
Offline
Site Admin
User avatar

Joined: Mon Aug 28, 2006 1:15 am
Posts: 2606
I'm beginning to think the older uboot is the only way w/ the older kernels to function really correctly.

I'm stumped :shock: , but hopeful that the older uboot and 2.4.33.3v3 firmimg.bin will put you in business. ;)

_________________
LS1 (foonas, nfs, Tranmission & p910nd print server, Firefly for my Roku)
LS-HG500 (Lenny)
Various LS-Pros v1,v2 (unbricked w/ serial & jtag)
KuroPro, LS2 & KuroHG (foonas)
Working on davysweather.dyndns.org lately...

=> wooohooo!
wooohooo!
Unknown command 'wooohooo!' - try 'help'


Top
   
PostPosted: Tue Apr 07, 2009 5:00 pm 
Offline
Site Admin
User avatar

Joined: Sun Jul 17, 2005 4:34 pm
Posts: 5332
OK, I've flashed the old uboot. But I can't load a kernel other than 2.4/EM Mode. It loops.

Loading 0:1:boot/vmlinux.UBoot

3510470 bytes read

2387 bytes read
## Booting image at 00800000 ...
Image Name: 2.6.18-ls-v52-uboot
Image Type: PowerPC Linux Kernel Image (uncompressed)
Data Size: 3510406 Bytes = 3.3 MB
Load Address: 00000000
Entry Point: 00000000
Verifying Checksum ... OK
OK
## Loading RAMDisk Image at 00000000 ...
Bad Magic Number
COMET#0
IDE: Bus 0: OK
Device 0: Model: Hitachi HDT725050VLAT80 Firm: V56OA4NA Ser#: VF1400R4CWK25A
Type: Hard Disk
Supports 48-bit addressing
Capacity: 476940.0 MB = 465.7 GB (976773168 x 512)
Boot in 09 seconds ('s' to stop)...s


It took me a while to figure it out, but I'm back on the foonas-em uboot.

As I seem to have tried everything, and I really want kernel 2.6.18, I would like to get back to the Buffalo flash ROM contents. I'm stuck in the old uboot/EM Mode for now.

but how can I apply http://downloads.buffalo.nas-central.or ... hContents/ ? Just dd as ususal? What about /dev/mtdblock5?


Top
   
PostPosted: Tue Apr 07, 2009 8:19 pm 
Offline
Site Admin
User avatar

Joined: Mon Aug 28, 2006 1:15 am
Posts: 2606
andre wrote:
OK, I've flashed the old uboot. But I can't load a kernel other than 2.4/EM Mode. It loops.

Loading 0:1:boot/vmlinux.UBoot

3510470 bytes read

2387 bytes read
## Booting image at 00800000 ...
Image Name: 2.6.18-ls-v52-uboot
Image Type: PowerPC Linux Kernel Image (uncompressed)
Data Size: 3510406 Bytes = 3.3 MB
Load Address: 00000000
Entry Point: 00000000
Verifying Checksum ... OK
OK
## Loading RAMDisk Image at 00000000 ...
Bad Magic Number
COMET#0
IDE: Bus 0: OK
Device 0: Model: Hitachi HDT725050VLAT80 Firm: V56OA4NA Ser#: VF1400R4CWK25A
Type: Hard Disk
Supports 48-bit addressing
Capacity: 476940.0 MB = 465.7 GB (976773168 x 512)
Boot in 09 seconds ('s' to stop)...s


It took me a while to figure it out, but I'm back on the foonas-em uboot.

As I seem to have tried everything, and I really want kernel 2.6.18, I would like to get back to the Buffalo flash ROM contents. I'm stuck in the old uboot/EM Mode for now.

but how can I apply http://downloads.buffalo.nas-central.or ... hContents/ ? Just dd as ususal? What about /dev/mtdblock5?


If you still have the 1.1.4 uboot + the 2.4.33.3v3 firmimg.bin in, and you experience that continuous looping, that probably means that you need to try running
Code:
run writeok

or
Code:
run writeng

while in uboot.

Darn, it sounds like you are close.

If that doesn't work, I'll peek at how to go back to stock ROM. I've done it a few times for repair clients here in the US that have had bad LSUpdater experiences.

_________________
LS1 (foonas, nfs, Tranmission & p910nd print server, Firefly for my Roku)
LS-HG500 (Lenny)
Various LS-Pros v1,v2 (unbricked w/ serial & jtag)
KuroPro, LS2 & KuroHG (foonas)
Working on davysweather.dyndns.org lately...

=> wooohooo!
wooohooo!
Unknown command 'wooohooo!' - try 'help'


Top
   
PostPosted: Tue Apr 07, 2009 8:40 pm 
Offline
Site Admin
User avatar

Joined: Sun Jul 17, 2005 4:34 pm
Posts: 5332
I tried "run writeok", it didn't seem to make a difference. The box seemed to accept setenv commands only every fifth time or so, much different from uboot 1.2.0.

While I would really like to keep uboot, I sort of depend on kernel 2.6.18, and I wonder why they haven't noticed resp. fixed those NFS speeds upstream. I'm not the only on NFS!

I guess just dd'ing the four mtd dumps will be OK, but I don't want to take unnecessary risks here.


Top
   
PostPosted: Tue Apr 07, 2009 8:55 pm 
Offline
Site Admin
User avatar

Joined: Mon Aug 28, 2006 1:15 am
Posts: 2606
yeah, the stock firmimg.bin and stock bootloader.bin go to the same locations as the 2.4.33.3v4 firmimg and uboot did, in that order.

Be careful that you save doing the bootloader flash as the very last step. And confirm that the stockfirmimg.bin is in ROM before doing the bootloader.bin, using a cmp of the mtdX contents. You may have to dump it out and then ftp it to your desktop computer to do that cmp, since the bender's firmimg.bin doesn't have cmp.

Having the uboot still in there is good insurance in case of a problem w/ the firmimg.bin. With the uboot you can still tftp boot or boot from the hdd.

I know what you mean by the old uboot being very "unsteady" interms of responsiveness. Sometimes it seems to just lose connection. To be honest, the UBoot on the HG & ARM boxes functions much better.

I'm sorry this didn't work out ... :roll: we certainly tried.

_________________
LS1 (foonas, nfs, Tranmission & p910nd print server, Firefly for my Roku)
LS-HG500 (Lenny)
Various LS-Pros v1,v2 (unbricked w/ serial & jtag)
KuroPro, LS2 & KuroHG (foonas)
Working on davysweather.dyndns.org lately...

=> wooohooo!
wooohooo!
Unknown command 'wooohooo!' - try 'help'


Top
   
PostPosted: Tue Apr 07, 2009 9:08 pm 
Offline
Site Admin
User avatar

Joined: Sun Jul 17, 2005 4:34 pm
Posts: 5332
Could you confirm this please?

Given

# cat /proc/mtd
dev: size erasesize name
mtd0: 00300000 00010000 "mtd_firmimg"
mtd1: 00070000 00010000 "mtd_bootcode"
mtd2: 00010000 00010000 "mtd_status"
mtd3: 00080000 00010000 "mtd_conf"
mtd4: 00400000 00010000 "mtd_allflash"
mtd5: 000f0000 00010000 "mtd_data"

and mtd[1-4] from http://downloads.buffalo.nas-central.or ... hContents/ , do I just ignore mtd0 and mtd5, and flash the mtd dumps 1 thru 4 to the matching mtdblocks 1 thru 4?


Top
   
PostPosted: Tue Apr 07, 2009 9:51 pm 
Offline
Site Admin
User avatar

Joined: Mon Aug 28, 2006 1:15 am
Posts: 2606
andre wrote:
Could you confirm this please?

Given

# cat /proc/mtd
dev: size erasesize name
mtd0: 00300000 00010000 "mtd_firmimg"
mtd1: 00070000 00010000 "mtd_bootcode"
mtd2: 00010000 00010000 "mtd_status"
mtd3: 00080000 00010000 "mtd_conf"
mtd4: 00400000 00010000 "mtd_allflash"
mtd5: 000f0000 00010000 "mtd_data"

and mtd[1-4] from http://downloads.buffalo.nas-central.or ... hContents/ , do I just ignore mtd0 and mtd5, and flash the mtd dumps 1 thru 4 to the matching mtdblocks 1 thru 4?


hmm... It is strange that no mtd0 contents are listed on that page of downloads.

mtd0 is where the firmimg.bin goes.
mtd1 for the bootloader.bin

You will want to make absolutely sure that your hda1 contents (especially /boot with the correct vmlinux ,etc) are right where they were before you flashed. The stock bootloader gives no control over the boot process, so if it fails to boot once you return to the stock format, I'm not sure how to fix things other than reinstalling via the LS Updater app.

I don't think I've ever flashed the other blocks 2 and 3 (except perhaps via the writeng or writeok commands). This part is a bit unknown territory for me - I've flashed uboot and then back to stock several times, but always with a blank hdd in the box (never with a hdd filled up w/ important stuff).

You might be able to flash mtd4 as one contiguous image, but I've never done that.

I wonder if kuroguy is around? mindbender?

_________________
LS1 (foonas, nfs, Tranmission & p910nd print server, Firefly for my Roku)
LS-HG500 (Lenny)
Various LS-Pros v1,v2 (unbricked w/ serial & jtag)
KuroPro, LS2 & KuroHG (foonas)
Working on davysweather.dyndns.org lately...

=> wooohooo!
wooohooo!
Unknown command 'wooohooo!' - try 'help'


Top
   
PostPosted: Wed Apr 08, 2009 1:11 am 
Offline
Site Admin
User avatar

Joined: Tue Mar 21, 2006 9:11 pm
Posts: 658
skip that mtd# stuff and flash those images from uboot. It is important to remember that those mtd devices are defined in the kernel and as a result can vary by name and address range from kernel to kernel. I bricked my Kuro HD because the mtd devices vary by kernel. Easiest is to use the "upload" script in uboot. These are the pertinent lines from the uboot environment:
Code:
ubload=echo Loading ${ubpart}:${ubfile};ext2load ide ${ubpart} ${ldaddr} ${ubfile}
ubsaddr=fff00000
ubeaddr=fff2ffff
ubflash=protect off ${ubsaddr} ${ubeaddr};era ${ubsaddr} ${ubeaddr};cp.b ${ldaddr} ${ubsaddr} ${filesize};cmp.b ${ldaddr} ${ubsaddr} ${filesize}
upgrade=run ubload ubflash

To make this work you need to set a couple of variables. Let's assume you have a 4 Mb flash image file named flash.bin and it needs to be loaded starting at ffc00000. you need to change the starting address to ffc00000 you do that with
Code:
setenv ubsaddr ffc00000

You need to change ubeaddr. it needs to be calculated. just add the file size (4Mb = 40000 hex) to the start address to get ffffffff:
Code:
setenv ubeaddr ffffffff

next you need to change the filename, partition for the flash image file. we'll assume it is on the first partition in /boot.
Code:
setenv ubpart 0:1
setenv ubfile /boot/flash.bin

Next you just need to run the upload script:
Code:
run upgrade


good luck

_________________
My completely scratch built 832 push-pull tube amplifier with Padauk frame. It matches my custom mini-ITX machine.


Top
   
PostPosted: Wed Apr 08, 2009 7:49 am 
Offline
Site Admin
User avatar

Joined: Mon Aug 28, 2006 1:15 am
Posts: 2606
Kuroguy has a good point. Safety is a good thing.

Anyway, being stubborn and a rebel I went ahead and did it (I'd just have to jtag it).

Code:
# cat /proc/mtd 
dev:    size   erasesize  name
mtd0: 00300000 00010000 "mtd0: kernel+ramdisk"
mtd1: 00070000 00010000 "mtd1: bootloader"
mtd2: 00010000 00010000 "mtd2: configuration"
mtd3: 00080000 00010000 "mtd3: user space"
mtd4: 00400000 00010000 "mtd4: all flash"
# dd if=mtd4 of=/dev/mtdblock4 bs=1k
4096+0 records in
4096+0 records out
# reboot

(took a loooooong time) :shock:

I was not sure of whether or not I bricked it, but quickly figured out that these http://downloads.buffalo.nas-central.org/LS1_PPC/OriginalFlashContents/ don't look to be stock contents, but someone's backup. It also looks like the 1.2.0 uboot binary in it was compiled for the 192.168.1.x subnet, with an an IPADDR value of 192.168.1.10. I used WireShark to sniff it out and it looks like SERVERIP and/or NCIP are set for 192.168.1.100.

Do you know who those mtd files belong to?

It was not a brick... so now I'm seeing if I can convert it back to stock ... had a hard time finding a stock bootcode.bin (linkstation.yi.org)

_________________
LS1 (foonas, nfs, Tranmission & p910nd print server, Firefly for my Roku)
LS-HG500 (Lenny)
Various LS-Pros v1,v2 (unbricked w/ serial & jtag)
KuroPro, LS2 & KuroHG (foonas)
Working on davysweather.dyndns.org lately...

=> wooohooo!
wooohooo!
Unknown command 'wooohooo!' - try 'help'


Top
   
PostPosted: Wed Apr 08, 2009 7:54 pm 
Offline
Site Admin
User avatar

Joined: Tue Mar 21, 2006 9:11 pm
Posts: 658
it was sounding familiar. I think those are dumps of my old box, but I never had a LS1 PPC, so I'm not sure how they might have wound up in the LS1 PPC directory.

If you want to turn it into a stock Kuro HD the flash contents are at http://downloads.buffalo.nas-central.or ... lFirmware/ . you would just need to flash the correct bits to the correct locations. If you want to stay with the uboot installed skip flashing the bootloader.

_________________
My completely scratch built 832 push-pull tube amplifier with Padauk frame. It matches my custom mini-ITX machine.


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 58 posts ]  Go to page Previous 1 2 3 4

All times are UTC+01:00


Who is online

Users browsing this forum: No registered users and 12 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Powered by phpBB® Forum Software © phpBB Limited