#8843 HIGH 9.1.0: Tell userspace whether an RTC alarm woke us up.

[Zarro Boogs per Child]

#8843: Tell userspace whether an RTC alarm woke us up.
----------------------+-----------------------------------------------------
   Reporter:  cjb     |       Owner:  dsaxena      
       Type:  defect  |      Status:  new          
   Priority:  high    |   Milestone:  9.1.0        
  Component:  kernel  |     Version:  not specified
 Resolution:          |    Keywords:               
Next_action:  review  |    Verified:  0            
  Blockedby:          |    Blocking:               
----------------------+-----------------------------------------------------
Changes (by cjb):

 * cc: cosmicpenguin (removed)
 * cc: JordanCrouse (added)


Comment:

 Replying to [comment:3 dsaxena]:
 > Did you open with O_NONBLOCK? It shouldn't hang it that case but just
 return 0 if an alarm did not happen. If you're programming an alarm and it
 is not triggering, then something else
 > is going on.

 I used "rtcwake -m on -s 10" and then ran echo mem > /sys/power/state.
 After a successful wakeup in ten seconds (minus the time it took to
 suspend), rtcwake was still running, and strace showed that it got
 -ERESTARTSYS to read(2) after the suspend, and then retried the read with
 a hang.  (Should be easy to reproduce.)

 > Jordan, if we get woken up by RTC, will we still receive the RTC
 interrupt?

 My patch tests for the RTC wakeup bit on the southbridge being set -- is
 that different to the RTC interrupt?

-- 
Ticket URL: <http://dev.laptop.org/ticket/8843#comment:4>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system