On 18 Feb 2005 18:18:53 -0500
"Nathan J. Williams" <nathanw@wasabisystems.com> wrote:

> > 0xd999ef40: at cpu_switchto+0x44
[...]
> In lines like "0xd999ef40: at cpu_switchto+0x44", the 0xd999ef40 is
> the frame address (on the stack), not a numeric representation of the
> instruction location.
OK, stupid user error.

But why are part of the addresses in the 0x00425xxx range and part in
0xd999exxx?

> > There is some other VM interaction with ddb(4).=20
[...]
> No particular ideas here, I'm afraid. I'm suspicious of mapiodev() not
> setting any guard or cache flags, but it seems to work for other OEA
> ports...
Hmmm. port-prep uses the same code but there it seems to work. The
43P-140, supported by port-prep, uses the same chip set / PCI host
bridge as the 43P-150 I am workin on. But port port-prep does some BAT
mapping also. Hmmmmm. This is subject to further investigation.
--=20


tsch=FC=DF,
       Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/