#11982 BLOC 12.1.0: XO-1.5 os16 runin hang

[Zarro Boogs per Child]

#11982: XO-1.5 os16 runin hang
--------------------------------+-------------------------------------------
           Reporter:  Quozl     |       Owner:  dsd                              
               Type:  defect    |      Status:  assigned                         
           Priority:  blocker   |   Milestone:  12.1.0                           
          Component:  kernel    |     Version:  Development build as of this date
         Resolution:            |    Keywords:                                   
        Next_action:  diagnose  |    Verified:  0                                
Deployment_affected:            |   Blockedby:                                   
           Blocking:            |  
--------------------------------+-------------------------------------------

Comment(by dsd):

 Progress today:

 Trying to reproduce on v3.0 is difficult as it has a frequent
 suspend/resume crash in mmc stuff.

 But Sam has completed one of the above tests where the v3.1 kernel is run
 on top of 11.3.1 and reports that there is no crash in that configuration.
 This suggests that the problem isn't a direct regression in the kernel -
 more likely it is a crash triggered by a change in userspace. I'm
 repeating this test locally overnight for confirmation.

 We have started adding more debug messages to the kernel around the crash
 site. We are fairly certain the crash happens in thaw_processes(), after
 thaw_workqueues() and before usermodehelper_enable (confirmed on two
 systems).

 Then through adding more messages, it seems that the hang occurs after all
 the processes are thawed, in the call to read_unlock(&tasklist_lock).
 We've seen this once but have yet to confirm a 2nd case.

-- 
Ticket URL: <http://dev.laptop.org/ticket/11982#comment:21>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system