Re: bus_space(9) overrides & resource reservations

[Eduardo Horvath <eeh%NetBSD.org@localhost>]

On Thu, 27 May 2010, David Young wrote:

> On Fri, May 28, 2010 at 01:29:52AM +0900, Izumi Tsutsui wrote:
> > > You could use bus_space_tag_create() in MD code such as if_ne_mb.c and
> > > dev/sbus/stp4020.c, but I don't know if it will be worth it.  The code
> > > would look more consistent with MI code.
> > 
> > How can you switch multiple MD bus_space_tag_t types in MI API?
> 
> I am not sure that I understand your question.
> 
> To support bus_space_tag_create(9), an architecture's MD
> bus_space_tag_t type will change in an MD way.  That is why I say that
> bus_space_tag_create(9) is an MI API with an MD implementation.
> 
> It may help for you to have a look at the documentation for
> pci_chipset_tag_create(9), since bus_space_tag_create(9) works
> analogously.  I have not written the corresponding documentation for
> bus_space_tag_create(9), sorry.

The pci_chipset_tag has always been a problem with sparc64 as well.  The 
number of hoops we needed to jump through to get the PCI code to work was 
huge.  I think you're using a poor example and trying to generalize it.  
Look at the sparc64 PCI code first, then think about it, then try to apply 
it to generic bus_space.

Eduardo