Re: Boot failure on port-amd64 5.99.29

[Paul Goyette <paul%whooppee.com@localhost>]

On Sun, 2 May 2010, Paul Goyette wrote:

> Just tried booting up one of my systems on a -current kernel built from 
> sources only a couple hours old, and got a crash during boot.  I don't have 
> the actual rash message (because the kernel is built with
>
>         DDB_COMMANDONENTER="call ddb_vgapost; bt"
>
> but the back-trace looks like this:
>
> bus_space_map + x20
> itesio_isa_attach + x192
> config_attach_loc + x16e
> isa_search + x18b
> mapply + x23
> config_search_loc + x105
> ...
>
> If I boot -c and disable itesio* then it goes to completion, so I'd guess 
> that there's something wrong with itesio_isa_attach().
>
> I can take a look at it later on, but in the meantime if anyone wants to dig 
> into it, please feel free.

and Jonathan A. Kollasch replied:

> Two bus_space_tag_ts are in the softc, one of them is never 
> initialized.

Yes, that is definitely the problem.  The failure occurs on the first 
reference to sc->sc_ec_iot which has not been initialized.

Is it safe to assume that the ec's bus_space_tag_t can share the "main" 
one?


-------------------------------------------------------------------------
| Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:       |
| Customer Service | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com    |
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette at juniper.net |
| Kernel Developer |                          | pgoyette at netbsd.org  |
-------------------------------------------------------------------------