Re: bus_space(9) overrides & resource reservations

[David Young <dyoung%pobox.com@localhost>]

On Fri, May 28, 2010 at 02:31:02AM +0900, Izumi Tsutsui wrote:
> > I think that if there is a difference between what you write and what
> > I have done, it is that bus_space_tag_create() *does* create a new tag
> > from the old before installing MI hooks into the new.  Otherwise, I
> > think that we are writing about the same thing.
> 
> If you call a function "bus_space_tag_create()",
> it should fill whole necessary members in MD bus_space_tag_t.
> But it looks your bus_space_tag_create() just add MI hooks
> and leave all other MD alloc/access details.
> 
> You might claim your MI bus_space_tag_create() can also add them
> in MD implementation, but there is no defined API to specify
> various MD implementation (map/access, stride/byteswap/wordoffset etc.)
> in MI bus_space_tag_create().

It sounds like you may be concerned that I intend bus_space_tag_create()
to replace comparable MD routines such as bus_space_tag_alloc() on
sparc.  I do not.

If the name troubles you, is bus_space_tag_refine() better?

Dave

-- 
David Young             OJC Technologies
dyoung%ojctech.com@localhost      Urbana, IL * (217) 278-3933