#11416 BLOC 1.75-so: C1 XO 1.75s may not wake up due to RTC from suspend

[Zarro Boogs per Child]

#11416: C1 XO 1.75s may not wake up due to RTC from suspend
---------------------------------+------------------------------------------
           Reporter:  greenfeld  |       Owner:                                   
               Type:  defect     |      Status:  new                              
           Priority:  blocker    |   Milestone:  1.75-software                    
          Component:  kernel     |     Version:  Development build as of this date
         Resolution:             |    Keywords:                                   
        Next_action:  diagnose   |    Verified:  0                                
Deployment_affected:             |   Blockedby:                                   
           Blocking:  11395      |  
---------------------------------+------------------------------------------

Comment(by pgf):

 martin asked in the meeting if the "new infrastructure" would prevent
 this.

 this "enhanced" rtcwake loop will ensure that wakeups that arrive before
 the system is actually sleeping will cause the attempted sleep to be
 aborted:
 {{{
 #!/bin/sh

 set -x

 WC=/sys/power/wakeup_count

 loops=0

 while :
 do
         # this inner loop guarantees there are no currently pending
 wakeups,
         # and also indicates to the kernel that we don't want to sleep if
 any
         # more arrive before sleep commences.
         while :
         do
                 echo $(cat $WC) > $WC && break;
                 sleep 1
         done


         rtcwake -s1 -mmem

         echo $(( loops++ )) =================

         sleep 1
 done
 }}}

 the theory behind this is explained here (and in other places):

 {{{
 http://lwn.net/Articles/392897/
 }}}

-- 
Ticket URL: <http://dev.laptop.org/ticket/11416#comment:3>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system