[Trac #322] hard crash with DCON test loop.

[Zarro Boogs per Child]

#322: hard crash with DCON test loop.
---------------------+------------------------------------------------------
 Reporter:  dwmw2    |        Owner:  dwmw2  
     Type:  defect   |       Status:  new    
 Priority:  blocker  |    Milestone:  BTest-1
Component:  kernel   |   Resolution:         
 Keywords:           |  
---------------------+------------------------------------------------------
Comment (by dwmw2):

 Preliminary testing seems to suggest that setting the console loglevel to
 1 prevents this.
 And that's even true when the serial port is the only console -- it's not
 fbcon rendering.

 Probably console/klog/dmesg locking -- it's exacerbated by the 'dmesg' in
 the script.

 Will attempt to catch with FS2 in the morning -- can probably be triggered
 by _any_ printk within an IRQ handler; not just my DCON hack.

-- 
Ticket URL: <http://dev.laptop.org/ticket/322#comment:1>
One Laptop Per Child <http://laptop.org/>