PERFORCE change 154573 for review

[M. Warner Losh]

In message: <200812141559.59329.hselasky at c2i.net>
            Hans Petter Selasky <hselasky at c2i.net> writes:
: On Saturday 13 December 2008, Warner Losh wrote:
: > From: Hans Petter Selasky <hselasky at c2i.net>
: > Subject: Re: PERFORCE change 154573 for review
: > Date: Sat, 13 Dec 2008 10:05:32 +0100
: >
: 
: >
: > This is absoultely the wrong way to implement this.  It is so wrong, I
: > don't even know where to begin.  Consider this an 'over my dead body'
: > level of objection to this design.
: >
: 
: We don't want to check a variable in the softc every time we do a htoleXX() ??
: 
: Do you have a better suggestion?

Usually the bus_space stuff is used for situations like this so that
the drivers don't have to care.

Warner