Todd Vierling <tv@NetBSD.ORG> writes:

> 
> On 15 Jan 1998, Charles M. Hannum wrote:
> 
> : > The problem with the minor-split proposal you had was that it would require
> : > devices, once renumbered, to handle two different kinds of minor numbers. 
> : 
> : Excuse me?  It would require a conversion step for at least every
> : major number that has its minor number format changes -- just like the
> : plan you're talking about.  In this respect, there's no difference.
> 
> Why?  I'm using 16 bit nodes with compatibility translation.  I'm writing
> this message using that kernel.
> 
> Only devices which do not already have named nodes would need new 32 bit
> nodes.  /dev/sd2d will still function as /dev/sd2d as long as the
> compatibility is there.  That's the point of the compatibility code.

I'm talking about inside the kernel.  That should be bloody obvious,
given the context of everything else.  *sigh*