Re: [Gumstix-users] USB Network access, Kernel upgrading, misc. getting started

[Craig H. <cr...@hu...>]

The symptoms you describe are pretty odd -- if 1:0 in flash hasn't been 
overwritten, then it doesn't much matter what you put in 1:2-32...  It 
should at least boot to u-boot (which is in 1:0), unless there's a 
hardware problem.  I assume you've done the "standard" recovery stuff, 
like unplug the gumstix, restart your term program, etc, and that 
didn't clear the problem up?
C

On Sep 12, 2004, at 3:02 PM, Heilpern, Mark wrote:

> I won't swear I didn't typo along the way, but I'm certain I never
> issued a "protect off" command at all; however (details in the previous
> email I sent) currently my stix is dead following an attempt at 
> flashing
> a new root fs. Any advice for me, or do I need to either purchase a 
> JTAG
> device or send my stix back?
>
> -----Original Message-----
> From: gum...@li...
> [mailto:gum...@li...] On Behalf Of Craig
> Hughes
> Sent: Sunday, September 12, 2004 5:49 PM
> To: gum...@li...
> Subject: [SPAM] - Re: [Gumstix-users] USB Network access, Kernel
> upgrading, misc. getting started - Email found in subject
>
> On Sep 12, 2004, at 9:11 AM, David Farrell wrote:
>
>>> experiment on my own because I don't want to turn my gumstix into a
>>> small brick ;)
>>
>> You can only do this if you mess with u-boot.
>> Stay away from u-boot sector 1:0 and you won't brick you gumstix.
>> Hint about u-boot, you always need the 1:x when specifing flash.  ie
>> prot off 1:2-4, era 1:2-31 etc. Some emails in this list do not make
>> this clear.
>
> There's one caveat about u-boot: even without doing a protect off on 
> 1:0
> you can *still* accidentally nuke it (and brick your gumstik) if you
> loadb or fatload to flash instead of to RAM.  both fatload and loadb
> just poke the bytestream they're copying directly to the target address
> space, which will break flash badly.  I'm going to put a patch together
> to have both commands check that you're not actually writing to flash,
> but that's not done yet.  So in summary: be careful with loadb/fatload,
> and don't protect off 1:0 and you'll be brick-safe.
>
> C
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> Project Admins to receive an Apple iPod Mini FREE for your judgement on
> who ports your project to Linux PPC the best. Sponsored by IBM.
> Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php
> _______________________________________________
> gumstix-users mailing list
> gum...@li...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users