#11054 NORM Not Tri: EC SCI mask not written correctly when going into suspend

[Zarro Boogs per Child]

#11054: EC SCI mask not written correctly when going into suspend
-------------------------------------------+--------------------------------
           Reporter:  dsd                  |       Owner:  pgf          
               Type:  defect               |      Status:  new          
           Priority:  normal               |   Milestone:  Not Triaged  
          Component:  embedded controller  |     Version:  not specified
         Resolution:                       |    Keywords:               
        Next_action:  never set            |    Verified:  0            
Deployment_affected:                       |   Blockedby:               
           Blocking:                       |  
-------------------------------------------+--------------------------------

Comment(by dsd):

 Could this historic relic be of relevance? found in olpc_fixup_sleep() in
 olpc-pm.c

 {{{
         /*
          * Cmd 0x32 tells the EC that we're going into suspend; this was
          * added to work around hardware races related to SCI events.
 This
          * should cause the EC to inhibit further SCIs while MAIN_ON is
          * transitioning low.
          *
          * There's also some sort of EC race whereby the EC gets its
          * IBF/OBF flags confused and all future communication (after
          * resuming) fails if we suspend too soon after updating
          * the EC SCI mask.  Having this command after updating the
          * SCI mask allows the EC enough time to finish doing what it's
          * doing.
          */
         return olpc_ec_cmd(EC_SET_SCI_INHIBIT, NULL, 0, NULL, 0);
 }}}

-- 
Ticket URL: <http://dev.laptop.org/ticket/11054#comment:5>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system