Subject: kern/30283: atu(4) needs serious work (Crash, and stall)
To: None <kern-bug-people@netbsd.org, gnats-admin@netbsd.org,>
From: Martijn van Buul <martijnb@balthasar.ipv6.stack.nl>
List: netbsd-bugs
Date: 05/19/2005 18:01:00
>Number:         30283
>Category:       kern
>Synopsis:       atu(4) needs serious work (Crash, and stall)
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu May 19 18:01:00 +0000 2005
>Originator:     Martijn van Buul
>Release:        NetBSD 3.0_BETA
>Organization:
	
>Environment:
	
	
System: NetBSD balthasar.ipv6.stack.nl 3.0_BETA NetBSD 3.0_BETA (BALTHASAR) #1: Thu May 19 18:46:24 CEST 2005 martijnb@balthasar.ipv6.stack.nl:/usr/src/sys/arch/i386/compile/BALTHASAR i386
Architecture: i386
Machine: i386
>Description:
	The atu(4) device driver needs some work:

	1) Scanning for APs using 'wicontrol atu0 -D' yields "No APs found",
          even when they are available. In some cases, it will lock up the
          machine

	2) Kernel panic in ltsleep (in usbd_transfer; I can compile a kernel
          with less optimisations and jot down a complete stackframe if 
	  desired; the bug is reproducable) while trying to do IPv6 work.

>How-To-Repeat:
	
	1) 'wicontrol atu0 -D', and watch it fail.
	2) Cause some ipv6 traffic. It often fails on the first router
           solicitation (using 'rtsol atu0'), but sometimes it'll live for
           a few minutes. 
>Fix:
	Don't know :(

>Unformatted: