#10881 NORM Not Tri: Possible EC reading issue with Q3B03, 11.2.0 os19

[Zarro Boogs per Child]

#10881: Possible EC reading issue with Q3B03, 11.2.0 os19
---------------------------------+------------------------------------------
 Reporter:  greenfeld            |                 Owner:  rsmith                           
     Type:  defect               |                Status:  new                              
 Priority:  normal               |             Milestone:  Not Triaged                      
Component:  embedded controller  |               Version:  Development build as of this date
 Keywords:                       |           Next_action:  diagnose                         
 Verified:  0                    |   Deployment_affected:                                   
Blockedby:                       |              Blocking:                                   
---------------------------------+------------------------------------------
 With Q3B03 on a D6 XO-1.5, we are seeing potential issues reading the EC,
 with kernel log messages such as:

 {{{
 May 17 17:43:32 localhost kernel: [11156.099298] olpc-ec:  timeout waiting
 for EC to provide data!
 May 17 17:43:32 localhost kernel: [11156.491049] olpc-ec:  timeout waiting
 for EC to read command!
 May 17 17:43:33 localhost kernel: [11156.511540] power_supply olpc-
 battery: driver failed to report `current_now' property
 May 17 17:43:36 localhost kernel: [11159.551589] olpc-ec:  timeout waiting
 for EC to provide data!
 May 17 17:44:27 localhost kernel: [11210.592896] olpc-ec:  timeout waiting
 for EC to read command!
 May 17 17:44:27 localhost kernel: [11210.609365] power_supply olpc-
 battery: driver failed to report `charge_now' property
 May 17 17:45:21 localhost kernel: [11265.170557] olpc-ec:  timeout waiting
 for EC to read command!
 May 17 17:45:21 localhost kernel: [11265.186869] power_supply olpc-
 battery: driver failed to report `current_avg' property
 May 17 17:46:17 localhost kernel: [11320.902404] olpc-ec:  timeout waiting
 for EC to read command!
 May 17 17:46:17 localhost kernel: [11320.922574] power_supply olpc-
 battery: driver failed to report `charge_type' property
 }}}

 Sometimes it's only one or two; sometimes it is a whole bunch; one time
 the kernel seems to be in an infinite loop of backtraces I couldn't catch
 complaining about something in this area.

 Daniel seems to think this is a possible regression, and independent of
 the Libertas hang issue, as it has been seen on occasion without it.  But
 if it wasn't for the ~8000 second delay, the attached logs from a noisy
 case of this almost make it look like the libertas issue is a potential
 cause.

-- 
Ticket URL: <http://dev.laptop.org/ticket/10881>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system