Regression on U-boot without batery [was: u-boot low-battery handling]

Joerg Reisenweber joerg at
Tue Aug 26 05:34:57 CEST 2008

Am Di  26. August 2008 schrieb Mike Montour:
> Joerg Reisenweber wrote:
> >> Hmm, usually it's the other way around - a recent u-boot in NAND is
> >> better equipped to survive than the older u-boot in NOR. But it can't
> >> hurt to try. All older hardware has a high degree of randomness
> >> coming from the small SYS capacitor anyway.
> > 
> > This is on a rather new device 20080626, probably A6 (have no torx 
atm :-/ )
> > The effect is 100% reproduceable for both booting on old and not booting 
> > new U_Boot. Only randomness is AUX-LED flashing very short (<0.1sec) when 
> > fail on normal boot w/o bat, whereas it flashes slow (~0.5s on) for 4 
> > to "4.5" times with a dead Nokia BL-5C. Result no boot on new U_Boot 
> You might want to try the test with a USB cable to a PC (capable of 
> 500mA) instead of the charger,
Powered via  (this particular) 100/500mA USB-Host, none of both U_Boot would 
boot without battery. Though it's happily working on linux without GSM just 
on USB-power, after boot with battery. New (NAND) will do nothing but a short 
flash (same as with USB-wall-powersupply), Old (NOR) will happily startup to 
menu  and boot kernel from NAND, but kernel breaks down after showing first 
line of console for 0.1sec.

> or to apply the patch from my first email  
> to your NAND u-boot.
If you point me to a kernel-image I might flash with dfu-util, I for sure will 
try what I find with this one.

> My FR is a GTA02v5 with datecode 20080619 and it will not power up at 
> all without a battery (due to the "small SYS capacitor" mentioned 
> above?) so I can't test this scenario.

You tried with OM USB power supply, and (factory) NOR-boot?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: This is a digitally signed message part.
Url : 

More information about the openmoko-kernel mailing list