Module unloading still broken on amd64?

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

With a kernel built from sources updated just a couple hours ago 
(13:40:07 UTC) I got the following panic:

panic: kernel diagnostic assertion "me->flags & UVM_MAP_QUANTUM" failed: 
file "/usr/src/sys/uvm/uvm_map.c", line 820
fatal breakpoint trap in supervisor mode
trap type 1 code 0 rip ffffffff80150af5 cs 8 rflags 246 cr2 7f7ffd958d30 
cpl 0 rsp ffff80004beef9c0
stopped in pid 0.8 (system) at netbsd:breakpoint+0x5:  leave

The backtrace looks like this (transcribed manually):

breakpoint+0x5
panic+0x24d
__kernassert_0x2d
uvm_mapent_allocsplit+0x107
uvm_map_clip_start+0x2b
uvm_unmap_remove+0x2f7
uvm_unmap1+0x68
kobj_unload+0x58
module_do_unload+0x10f
module_thread+0xe9

This is a totally modular kernel, with NO built-in modules.

This was actually the second crash I had (the first one happened while 
X was running, so I didn't get the console info).  I had logged in on 
the console and copied a file (approximately 5MB) from an nfs-mounted 
filesystem to a local file, and logged out.  Some time later, I logged 
back in and tried to cd to a directory on the nfs-mounted fs, and got 
the panic immediately - before the shell-prompt came back.

The system is totally wedged now, so I can't grab any more debug info 
from this event.  Is there something useful that I can add to my 
ddd.commandonenter for the next occurrence?



-------------------------------------------------------------------------
|   Paul Goyette   | PGP DSS 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 |
|                  |                          | pgoyette at netbsd.org  |
-------------------------------------------------------------------------