#5240 NORM Never A: RTC & power button can still cause #1835?

[Zarro Boogs per Child]

#5240: RTC & power button can still cause #1835?
---------------------------------+------------------------------------------
 Reporter:  dilinger             |       Owner:  rsmith        
     Type:  defect               |      Status:  new           
 Priority:  normal               |   Milestone:  Never Assigned
Component:  embedded controller  |     Version:                
 Keywords:                       |    Verified:  0             
---------------------------------+------------------------------------------
 Richard fixed a race in the EC, where an SCI would be raised right before
 or after MAIN_ON went low (ie, when suspending) would hang the machine
 (#1835).  The fix was to add EC cmd 0x32 which told the EC to inhibit
 SCIs.  The kernel runs 0x32 shortly prior to suspend, so we're all good.

 However, there are two additional SCI sources that do not go through the
 EC the way most SCIs go through the EC; PWRBTN and RTC.  The power button
 *does* go through the EC, but the EC just asserts PWRBTN line; we should
 probably inhibit that as well in order to prevent this race.  The RTC race
 should be inhibited by the kernel; on the way down, check if an RTC event
 is scheduled for <= 1s, and if so refuse to suspend.

-- 
Ticket URL: <http://dev.laptop.org/ticket/5240>
One Laptop Per Child <http://dev.laptop.org>
OLPC bug tracking system