The following reply was made to PR kern/37063; it has been noted by GNATS.

From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/37063: sleepq assertion fails
Date: Fri, 12 Oct 2007 21:44:36 +0200

 I found a few bugs in my instrumentation causing secondary fallout, fixed
 them and retested.
 
 I caught the system at wrong IPL in two places:
 
 IPL_HIGH (instead of IPL_SCHED) at
    kern_synch.c, sched_pstats(), right after  mutex_exit(&proclist_mutex);
 
 IPL_SCHED (instead of IPL_SOFTCLOCK) at
    kern_timeout.c, callout_softclock, right at the start after ci = l->l_cpu;
 
 (the first one multiple times, the second only once).
 
 Martin