On Sun, 22 Jun 1997, Darren Reed wrote:

> In some mail I received from Bill Studenmund, sie wrote
> > 
> > On Sun, 22 Jun 1997, Darren Reed wrote:
> > 
> > > Ummm, I think we have three things being handled here then (well, on this
> > > subject anyway ;)
> > > 
> > > 1: accessing non-BSD fdisk-partitions.
> > > 2: allowing 16 partitions inside a BSD fdisk-partition.
> > > 3: using 32 bits for minor #'s rather than 16 bits.
> > > 
> > > Doing 3 will help 1 & 2, but doing 2 won't solve 1.
> > 
> > But 2 will help 1. As port i386 only has four "free" partitions (a is
> > root, b swap, c is whole BSD, and d is whole disk), needing one of these 4
> > slots to get at a DOS partition is limiting (you can't have /usr, /tmp,
> > /var, and /home all on seperate partitions w/ a DOS partition around).
> 
> No, 2 doesn't help 1.  If you can have > 16 logical partitions (inside either
> DOS extended partitions, other BSD partitions, SVR4 partitions, etc), then
> even upping the limit to 4 provides little benefit.

"upping the limit to 4?" We're talking about upping the limit to 16,
giving 12 (61-4) freely-asignable partitions.

But re-reading the thread (what a clever concept!), I think we're both
right. I said having more freely-asignable partitions will help, and you
said it won't solve the problem. We're both right. :-)

Take care,

Bill