The following reply was made to PR kern/32266; it has been noted by GNATS.

From: Sean Davis <dive-nb@endersgame.net>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/32266: deadlocks with 3.99.13/i386
Date: Fri, 9 Dec 2005 21:20:55 -0500

 On Fri, Dec 09, 2005 at 10:27:05AM +0100, Martin Husemann wrote:
 > On Fri, Dec 09, 2005 at 02:10:03AM +0000, Sean Davis wrote:
 > >  Okay, tried it. The following message is scrolled on console non-stop, and
 > >  the machine is totally unusable:
 > >  
 > >  pool_get(PR_WAITOK) with held simple_lock 0xcc6146d8 CPU 0 ../../../../kern/kern_lock.c:561
 > 
 > Ok, one more thing, please:
 > 
 > Could you boot with the -d flag (to drop into the kernel debugger early),
 > then do
 > 
 >   w simple_lock_debugger=1
 >   c
 
 With a USB keyboard, boot -d gives me a dead console, as USB hasn't been
 initialized yet and I do not use USB emulation in the BIOS. However,
 simple_lock_debugger is already set to 1, presumbly by the LOCKDEBUG code.
 
 > and then reproduce it? This time it should not only print the above
 > warning, but also drop into DDB right away. Then please do
 > 
 >   bt
 > 
 > and show us the output.
 
 It dropped into DDB right away and gave a backtrace... I didn't issue a bt
 command, and I wrote this down by hand, so I'm not sure how helpful it will
 be.
 
 Anyhow, after reproducing the crash (same method):
 
 simple_lock: uninitialized lock
 lock: 0xcb35e0d8, currently at: ../../../../kern/kern_lock.c:561
 last locked: :2314
 last unlocked: :26217
 _lockmgr
 vfs_busy
 lookup
 namei
 sys___stat30
 syscall_plain
 --- syscall (number 387) ---
 0xbbba6e53:
 Stopped in pid 436.1 (cvs) at netbsd:cpu_Debugger+0x4: leave
 
 At this point I rebooted the machine, and rolled the kernel back to 3.99.10,
 as every time I reproduce this bug, I have to wait for approx. 477GB to go
 through fsck, which is less than pleasant.
 
 HTH,
 Sean