#12541 BLOC 13.1.0: XO-4: Wakeup events during suspend hang the machine. 13.1.0 build 29

Zarro Boogs per Child bugtracker at laptop.org
Sat Mar 9 11:00:49 EST 2013


#12541: XO-4: Wakeup events during suspend hang the machine. 13.1.0 build 29
---------------------------------+------------------------------------------
           Reporter:  rsmith     |       Owner:  pgf                              
               Type:  defect     |      Status:  new                              
           Priority:  blocker    |   Milestone:  13.1.0                           
          Component:  kernel     |     Version:  Development build as of this date
         Resolution:             |    Keywords:                                   
        Next_action:  never set  |    Verified:  0                                
Deployment_affected:             |   Blockedby:                                   
           Blocking:             |  
---------------------------------+------------------------------------------

Comment(by pgf):

 actually, we're talking about two different bugs here.

 1) in the original bug, the system was stuck, but could be revived by
 another interrupt -- that's why the EC rescue mechanism worked.

 2) the hang i've reproduced easily in the previous comment is different.
 the system is unrevivable.  EC rescue will not help in this case.

 we think jnettlet's patch in pm-mmp3.c has narrowed the race considerably
 for case 1).  EC rescue should cover the remaining cases.  i'd still like
 to verify with marvell that getting stuck partway into suspend in this way
 is expected.  their state diagram suggests so, but it's rather poorly
 designed if so.  in any case, we should add EC rescue back in.

 for 2), i'm not sure what to do.  that it's easy to hit with the game keys
 makes me think that the SP needs to hold off after a wakeup before sending
 that data to the PJ4.  for the PS/2 data, the holdoff is accomplished by
 the EC's flow-control mechanism.  the game-keys have no such flow-control.

-- 
Ticket URL: <http://dev.laptop.org/ticket/12541#comment:13>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system


More information about the Bugs mailing list