Kernel: page fault trap, code=0
> Stopped at _gx_release+0x21: movl 0x8(%eax),%eax
> ddb>
>
> I did:
> ddb> print %eax
> e03f931d
>
> ddb> examine
> _gx_release+0x21: ff08408b
>
> ddb> trace
> _gx_release (e16ee000,0,4,2300117) at _gx_release+0x21
> _gx_attach (e16c1b00, ......snip.......) at _gx_attach+0x74c
> _config_attach (e16c1b00, ......snip.......) at _config_attach+0x13e
> _config_found_sm (e16c1b00, ......snip.......) at _config_found_sm+0x2f
> _pciattach (e1693c00, ......snip.......) at _pciattach+0x1df
> _config_attach (e1693c00, ......snip.......) at _config_attach+0x13e
> _config_found_sm (e1693c00, ......snip.......) at _config_found_sm+0x2f
> _pchbttach(e1692f40, .........snip.........) at _pchbattach+0x336
> _config_attach (e1692f40, ......snip.......) at _config_attach+0x13e

oh yeah, that's right, i think it was the pchb that was giving problems.
the truly curious will want to see
http://www.openbsd.org/cgi-bin/cvsweb/src/sys/arch/i386/pci/pchb.c