Subject: Re: kern/11627
To: gabriel rosenkoetter <gr@eclipsed.net>
From: Nathan J. Williams <nathanw@MIT.EDU>
List: current-users
Date: 01/31/2001 17:06:46
gabriel rosenkoetter <gr@eclipsed.net> writes:

> (What about PCI-ISA busses in non-i386 machines, then? Do we treat
> them the same way? I'm pretty sure such things do exist...)

alpha: "sio" PCI-ISA bridge is subtractive, only supports one ISA bus.
       "jensenio" mainbus-EISA/ISA bridge only supports one of itself,
       hence only one ISA bus.
arc:   "isabr" mainbus-ISA bridge assumes only one ISA bus.
arm32: pcib assumes only one ISA bus.
atari: Existence of one isabus is hardcoded in autoconf.c
bebox: pcib assumes only one ISA bus.
hpcmips: "vrisab" mainbus-ISA bridge assumes only one ISA bus.
prep:  pcib assumes only one ISA bus.

> > (I'm in favor of ripping out "isa* at pcib?" and "isa* at pceb?" and
> > just leaving "isa0 at mainbus0", since the former two give the
> > not-very-useful impression that the entire ISA bus is behind them.)
> 
> Uh... would the ISA bus in this ThinkPad laptop still be detected
> properly, then? (It comes up on pcib? right now...)

Yes. "isa0 at mainbus0" will succeed if no other ISA bus has been
found.

        - Nathan