As already noted by Matthew Orgass, reverting  Reverting
arch/mips/mipsX_subr.S to version 1.15 make it works like a charm.

If you want to try, just to to cvs here:

http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/arch/mips/mips/mipsX_subr.S?rev=
=3D1.15&content-type=3Dtext/x-cvsweb-markup

And hit the download link. Then copy the file over
sys/arch/mips/mips/mipsX_subr.S and  build the kernel again.

Matthew, do you know exactly what is the problem? If so, it would be
cool explaining on the PR so they can fix it. Thanks for finding this
out.

Regards,
Claudio Neves

On 7/24/05, Andy Ruhl <acruhl@gmail.com> wrote:
> On 7/24/05, Alexander Bochmann <ab@lists.gxis.de> wrote:
> > Hi,
> >
> > ...on Sat, Jun 18, 2005 at 10:17:28AM -0700, Andy Ruhl wrote:
> >
> >  > Here's exactly what it says before it hangs and doesn't continue:
> >  > mem_cluster_cnt =3D 2
> >  > mem_clusters[0] =3D {0x0,0x37d000}
> >  > mem_clusters[1] =3D {0x37d000,0x1c83000}
> >  > loading 0x37d000,0x1c83000
> >  > And that's it. Not even as far as the 3.0_BETA kernel goes.
> >
> > Did anything happen here? I just tried booting
> > an cross-compiled 3.0_BETA Kernel from last week's
> > sources on my Z50, and get about one line further...
> >
> > mem_cluster_cnt =3D 3
> > mem_clusters[0] =3D {0x0,0x39f000}
> > mem_clusters[1] =3D {0x39f000,0xc61000}
> > mem_clusters[2] =3D {0x2000000,0x2000000}
> > loading 0x39f000,0xc61000
> > loading 0x2000000,0x2000000
> >
> > ...and that's it.
>=20
> No, nothing yet as far as I know.
>=20
> The pr is 30590, but I didn't copy port-mips when I opened it like I was =
asked.
>=20
> Hopefully this will put more eyes on it. There is sort of a fix in the
> pr, but I'm not sure if that is ideal. That was from a post in this
> thread, not me.
>=20
> Andy
>