SOS - uboot drowned ;)

[Christian Daniel]

Hi Nils!

My u-boot was drowned as well since I made the same mistake - not seeing the 
file with the crucial "B"...

All the howtos didn't help here either - until I recognized, that my battery 
ran empty. With the wrong bootloader in flash the charger doesn't start and 
the CPU draws power all the time.

For me the solution was to put in the battery from my Nokia phone - but you 
could also recharge yours in an external charger or (as a last resort and not 
without a current limiter) use a power supply at 3.7V as fake battery during 
reflashing.

Hope this helps...
Christian

On Wednesday 28 March 2007, Nils Faerber wrote:
> Harald Welte schrieb:
> > On Tue, Mar 27, 2007 at 12:19:46PM +0200, Nils Faerber wrote:
> >> Hi!
> >> After there were some message about changing uboot's behaviour and
> >> different initialisations on the mailinglists I thought I should give a
> >> new uboot version a try on the phase-0 device. So I flashed
> >> 	u-boot-gta01v3-r3_0_0_1511.bin
> >
> > GTA01v3 != GTA01Bv3 !!!
> >
> :(
>
> Really my fault, sorry. I did not see the file in the list and thus did
> not recognise the missing "b".
> What a difference a letter can make...
>
> > GTA01v3 has completely different PMU voltage configuration, memory size,
> > etc.  There's about six months in hardware development time between
> > them.
>
> Uh!
> Hopefully I cannot toast the NEO then!?
>
> > I guess the only way to resurrect your device is to use OpenOCD to halt
> > the device immediately after the cpu powers on, then use OpenOCD to
> > ram-load a GTA01Bv3 lowlevel_foo/u-boot image, execute that and then
> > flash a GTA01Bv3 u-boot from there.
> >
> > http://wiki.openmoko.org/wiki/Bootloader#Using_JTAG_to_boot_from_RAM
> > http://wiki.openmoko.org/wiki/NAND_bad_blocks#JTAG_.2F_OpenOCD_.2F_u-boot
> >_RAM_based
>
> I will try with Mickeys tips again.
> Anyway, many thanks so far!
> The tools themselves are very cool and impressive! I have seen much
> worse already. Though stability seems to be an issue. The other JTAG
> tools we used so far were more reliable. I am pretty sure that it is not
> a software fault - maybe some resistors/capacities could be added to
> make the signals more stable?
> Well, maybe for the over-next debug board revision ;)
>
> I'll keep you posted!
>
> Cheers
>   nils faerber



-- 
----------------------------------------------
| maintech # Dipl. Inf (FH) Christian Daniel |
|   GmbH ###    Seeweg 60 · D-97204 Höchberg |
----------------------------------------------
| http://www.maintech.de      cd at maintech.de |
----------------------------------------------