On Mon, 2009-07-06 at 18:32 +1000, Greg Ungerer wrote:
> Hi Kris,
>
> On 07/04/2009 02:13 AM, Microbit_P43000 wrote:
> > Thanks for reviewing.
> > I will absorb the submit patches issue in the near future.
> >
> > Typical - with some further effort, I solved the mysterious fsck.jffs2
> > puzzle, it was more
> > a matter of looking deeper again into the scripts. Once the script has
> > decided that e2fsck
> > indeed is executed, I noticed it uses file system info from fstab...
> > I did not know however that fsck can't process jffs2 files (in hindsight it
> > makes perfect
> > sense of course since the mtd layers are needed on top of Flash for JFFS2).
> > To cut a long story short, the root is there as type jffs2 which results in
> > a
> > concatenation of fsck with jffs2, hence the error, as this executable
> > doesn't exist.
> >
> > Well, at least I'm glad I _did_ solve it myself after all - always has been
> > and will be
> > the best way to learn !
> > I just need to figure out now how to inhibit that futile fsck call, the
> > start up error is
> > irritating me.. :-)
> >
> > Finally, I wonder if you could clarify for me Greg : I added a kernel
> > command parameter in
> > (menu)config 'fastboot', but the error still persists ?
> >> From studying the script, I thought fastboot ensured that fsck is not run
> >> - at least not
> > on rootcheck ?
>
> Which "script" are your referring to?
> This sounds like it is completely vendor dependant - in other
> words whatever Olimex put on there.
>
> Regards
> Greg
>
>
>
> >> -----Original Message-----
> >> From: uclinux-dev-boun...@uclinux.org
> >> [mailto:uclinux-dev-boun...@uclinux.org] On Behalf
> > Of
> >> Greg Ungerer
> >> Sent: Friday, 3 July 2009 3:24 PM
> >> To: uClinux development list
> >> Subject: Re: [uClinux-dev] port to Olimex SAM9-L9260 ( boot "fsck:
> >> fsck.jffs2:not found"
> > message)
> >> Hi Kris,
> >>
> >> Microbit_P43000 wrote:
> >>> Hi all,
> >>>
> >>> I'm new to this list and (uC)linux overall.
> >>>
> >>> However, with some persistence (and great interest in uClinux :-) I got
> >>> the
> > uCLinux-dist
> >>> (20080808) ported and running smoothly on the Olimex SAM9-L9260 board
> >>> (similar to Atmel AT91SAM9260-EK).
> >>>
> >>> I don't know yet how to go about patches or implementing something so my
> >>> effort can be
> >>> incorporated into future versions.
> >> The best place to start is to read the documentation in
> >>
> >>
> >> linux-2.6.x/Documentation/SubmittingPatches
> >>
> >> Regards
> >> Greg
> >
> > _______________________________________________
> > uClinux-dev mailing list
> > uClinux-dev@uclinux.org
> > http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
> > This message was resent by uclinux-dev@uclinux.org
> > To unsubscribe see:
> > http://mailman.uclinux.org/mailman/options/uclinux-dev
> >
>
Hi Greg,
> Which "script" are your referring to?
> This sounds like it is completely vendor dependant - in other
> words whatever Olimex put on there.
I did figure that out in the interim. In hindsight a bit dumb of me - I
suppose that's the whole idea behind it, its flexibility with runlevels.
Unfortunately it also (can) make it very confusing for newbies like
me... :-)
I checked again - it's rcS.d-> S10checkroot.sh, but this is what had me
stumped :
>>>>>>>>>>>
FSCKCODE=0
if [ -f /fastboot ]
then
[ "$rootcheck" = yes ] && log_warning_msg "Fast boot enabled, so
skipping file system check."
rootcheck=no
fi
<<<<<<<<<<
I couldn't identify another var in the shell that forces fsck after all
(overriding the decision to NOT execute e2fsk).
But in the interim I wondered about building my own minimal rootfs, so
it's a lot clearer (I prefer the 'hands-on' approach, rather than just
reading it).
I installed buildroot, am close to creatin' a small jffs2 image.
Once I got the loose ends right wrt mapping & uboot handover to the
kernel, then I can think/see clearer :-)
Thanks in any case, Greg.
Best regards,
Kris
_______________________________________________
uClinux-dev mailing list
uClinux-dev@uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev@uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev
