| 1 |
On Tue, Dec 02, 2003 at 12:04:35PM -0700, AJ Armstrong wrote: |
| 2 |
> -----BEGIN PGP SIGNED MESSAGE----- |
| 3 |
> Hash: SHA1 |
| 4 |
> |
| 5 |
> On 2 Dec 2003 11:44, david@×××××××××.com emitted: |
| 6 |
> |
| 7 |
> > ex: do we really need 500 lines of bash sourced so we can have colored |
| 8 |
> > [ OK ] |
| 9 |
> > I don't think this is really necessary. |
| 10 |
> |
| 11 |
> Not only is it unnecessary, but I think it is _wrong_ if the code is running |
| 12 |
> in an embedded environment. I think we should be starting with what is |
| 13 |
> absolutely required for a system to boot as the baselayout - no frills, no |
| 14 |
> pretification. Those could perhaps be build as optional ebuilds that can be |
| 15 |
> emerged _after_ baselayout for those whose target platforms can support them. |
| 16 |
> |
| 17 |
|
| 18 |
Well I'm not sure where to draw the line... |
| 19 |
do we want to still have /var/run PID functionality for example ? Or should we |
| 20 |
leave that up to the embedded user... it could most likely be |
| 21 |
implemented in a few lines of bash. |
| 22 |
|
| 23 |
or maybe some sort of busybox like funcitonality with symlinks... i.e. |
| 24 |
|
| 25 |
/etc/init.d/busyinit |
| 26 |
/etc/init.d/network -> /etc/init.d/busyinit |
| 27 |
|
| 28 |
and the initscript takes care of the pid info in a very generic way... |
| 29 |
|
| 30 |
just throwing out ideas. |
| 31 |
|
| 32 |
-- |
| 33 |
gentoo-embedded@g.o mailing list |
Archives