[gentoo-embedded] Re: baselayout-2, busybox and uclibc

[Marc Blumentritt]

Natanael Copa schrieb:
> On Sun, 2007-10-14 at 15:44 +0200, Marc Blumentritt wrote:
>>> clock: There is the following message:
>>> /sbin/hwclock: unrecognized option '--noadjfile'
>> This is done by setting of /etc/TZ, so clock is not needed in my setup.
>
> This should be fixed in busybox. IIRC its fairly simple.

I'm not a dev, so I have no clue how best to fix this.

>
>>> -> busybox problem or other configuration of clock is required.
>>>
>>> localmount: there is just thise message:
>>> Segmentation fault
>>> I don't know, where this comes from. Further investigation is required.
>>>
>>> keymap and consolefont:
>>> /lib/rcscripts/sh/runscript.sh: line 91:kbd_mode: not found
>>> /bin/loadkeys not found
>> I will do this with a small script and busybox tools, so these two
>> scripts are not needed anymore, too.
>
> I use loadkmap from busybox. It would actually be neat if baselayout-2
> supported native loadkmap as a fallback.

 I plan to use loadkmap from bb, too. And if baselayout-2 would support
this, it would be great.
>
>
> Have a look at https://bugs.gentoo.org/show_bug.cgi?id=175179 for a
> patch and info on how to generate the bb kmaps.
Why not using dumpkmap from bb?

>
>>> net.eth0: I want to use udhcpc from busybox (set modules="udhcpc" in
>>> net), but I get the following error:
>>> /sbin/udhcpc: unrecognized option '--env'
>>>
>>> -> busybox problem or other configuration of net is required.
>> Still a problem. Would be cool, if either baselayout-2 supports
>> bb-udhcpc or bb-udhcpc gets more conform with udhcpc
>
> what is this --env option supposed to do?

>From what I see, this option results from a Gentoo patch to udhcpc. I
found this in /lib/rcscripts/net/udhcpc.sh :
[...]
udhcpc_start() {
        [...]
        # Setup options for the udhcpc script
        # This requires a specfic Gentoo patch to udhcp which will not
        # be accepted upstream.
        if [[ " ${!d} " == *" nogateway "* ]] ; then
                opts="${opts} --env PEER_ROUTERS=no"
        else
                opts="${opts} --env PEER_ROUTERS=yes"
        fi
	[...]

So this seems to be a "Gentoo made" problem, which should be solved
inside of gentoo?

Regards,
Marc

-- 
gentoo-embedded@g.o mailing list