[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: devfs support in ofpath: criteria for patches



On Wed, May 23, 2001 at 06:12:46PM -0700, Tom Rini wrote:
> 
> Well, the problem is that my fstab has /dev/hda10, but df shows the
> devfs name.  I've (almost) got something done which will turn
> /dev/ide/host0/bus0/target0/lun0/partX into /dev/hdaX and so on (at least
> as much as we have OF aliases for).

i am not opposed to ofpath supporting devfs, there are some criteria
that MUST be met for me to accept a patch:

1) it must NOT damage, cripple, or remove standard, sane short
non-devfs name support.  

2) it must be pure posix /bin/sh code.  read me lips, absolutly no bashisms. 

3) it must not add any new dependencies to ofpath, that is to say,
ofpath currently lives without nice stuff like sed, awk, cut etc.  the
only things out of /usr that is uses is find, head and tail and of
those it only uses what busybox provides.  ofpath MUST work on the
debian boot-floppies, they are quite stripped and will not have new
utilities added.  you can use what utilities ofpath already uses, but
no new ones.  

4) persuant to 3, all calls to external utilities must be busybox compatible.  

5) roughly follows my coding style, mainly i don't want to see patches
with rediculous variables like MY_LONG_ASS_COMPLETLY_UNREADBLE_VARIABLE.

6) it must be a shell function say unfuckdevfs() that is called with
one argument: /dev/retarded/devfs/name and outputs a normal, unix
compliant terse name (eg: standard linux device name). i will almost
certainly not accept a huge patch reworking all of ofpath solely to
cater to devfs.

> > if you can configure your system to never show devfs names (df is the
> > problem usually) then yabootconfig and ofpath will work fine with
> > devfs.
> 
> Right.  But that's not the default config. :)

correct, the default config is CONFIG_FS_DEVFS=n as it should be ;-)

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

Attachment: pgp9OLp5cZH2U.pgp
Description: PGP signature


Reply to: