Re: Boot-time debug messages

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

On Sat, 13 Dec 2008, Paul Goyette wrote:

> > > speedy:paul {101} dmesg | grep sysc
> > > sysctl_createv: sysctl_create(kern) returned 1
> > > sysctl_createv: sysctl_locate(boottime) returned 2
> > > sysctl_createv: sysctl_create(vm) returned 1
> > > sysctl_createv: sysctl_locate(loadavg) returned 2
> > > sysctl_createv: sysctl_create(hw) returned 1
> > > sysctl_createv: sysctl_locate(machine) returned 2
> > > sysctl_createv: sysctl_locate(machine_arch) returned 2
> > > speedy:paul {102}

A little bit of detective work with ddb shows that these are being 
generated during initialization of compat_netbsd32 code.  With a 
breakpoint set at kern_sysctl.c:sysctl_createv() line 2010 (the printf 
call for "sysctl_createv: sysctl_locate...") I get the following back 
trace:

        netbsd32_sysctl_init + 0xe9
        compat_netbsd32_modcmd + 0x25
        module_do_builtin + 0x19f
        module_init_class + 0x62

Since the first of each pair of messages is a return of EPERM from 
sysctl_create(), it seems that perhaps the netbsd32 code is trying to 
add a node with CTLFLAG_PERMANENT after the flag has been set on the 
tree's root?


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