This is the mail archive of the ecos-discuss@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: redboot block size on IQ80310


I got it to burn the big (5M) images by giving it a 'fis create -b
0xa0200000 -l 0x510000 -f 0x00080000 rootfs.img', once I gave it a -f
and a address it no longer complained about not having enough space. I'm
not really sure what causes this behavior and I'm not sure if it can be
classified as a bug, non-intuitive but not necessarily a bug.

On a more technical note it seems like this behavior might be related to
'fis free' not showing matching the values from 'fis list'. In the code
it seems like fis_free looks at the beginning of a flash_block and no
where else in that block, so if you start at some misaligned address you
won't find all the free space you have. I'm not sure about all that
though cause I just gave it a quick once over. :)

Thanks for the replies Andrew.

-Travis

On Tue, 2004-06-22 at 03:13, Andrew Lunn wrote:
> > Humm, Redboot[backup] starts at 0x00040000 and is 0x00040000 in
> > length. That means it finishes at 0x0007ffff. So why is fis free
> > showing you 0x00060000 as being the first free address?
> 
> Reading the source code explains this. RedBoot decided which areas of
> flash are free by looking at the contents of the flash, not what the
> FIS table says is in use. ie if the flash block conatins all 0xff its
> free.
> 
>         Andrew


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]