Re: DEVFSv50 and /dev/fb? (or /dev/fb/? ???)

Richard Gooch (Richard.Gooch@atnf.CSIRO.AU)
Fri, 7 Aug 1998 14:06:45 +1000


Anthony Barbachan writes:
> >> naming sceme, change only what needs to be changed like SCSI.
> >
> >By that argument you should never introduce something new because
> >people will be "forced" to use the new thing.
>
> No I am not saying that. But it doesn't make sense to change everything,
> even the devices which do not suffer from large amounts of sub devices.
> Besides I am agruing against the gratuitous use of the new naming sceme not
> dev_fs in general. SCSI may and probably does need it but most other
> devices do not. Why change /dev/hd[a-h] when EIDE devices are already
> location based?

Actually, you have said that in time the "old" names will be thrown
out. Now, I can't promise that Linus wouldn't do that, but I'm sure
he'd only do it with wide support.
Furthermore, even if the "old" names were removed from devfs, you
could still have them by not running devfs in the first place: the
major&minor scheme isn't going to go away (POSIX mandates them).

As I've said before: the IDE situtation is going to get more
SCSI-like, according to the ex-IDE guys. The new IDE naming scheme in
devfs is for future-proofing.

> >There *is* a need for a naming scheme like the new SCSI names, at
> >least for big systems. People with small systems or who don't like the
> >new names can use the existing names. But that should stop people with
> >big systems being allowed to have a location-based naming scheme.
>
> I am not against the name change where it is needed, like SCSI, but EIDE
> devices definately do not need it, nor do other devices.

Like I said, IDE is moving towards SCSI. Furthermore, the new names
make /dev less cluttered and more organised. Administrators should be
given that choice.

Regards,

Richard....

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html