#9690 NORM 1.5-F11: suspend/resume from VT sometimes causes spurious linefreeds

[Zarro Boogs per Child]

#9690: suspend/resume from VT sometimes causes spurious linefreeds
------------------------------------+---------------------------------------
           Reporter:  pgf           |       Owner:                                   
               Type:  defect        |      Status:  new                              
           Priority:  normal        |   Milestone:  1.5-F11                          
          Component:  not assigned  |     Version:  Development build as of this date
         Resolution:                |    Keywords:                                   
        Next_action:  diagnose      |    Verified:  0                                
Deployment_affected:                |   Blockedby:                                   
           Blocking:                |  
------------------------------------+---------------------------------------
Changes (by Quozl):

  * next_action:  reproduce => diagnose


Old description:

> suspend/resume itself seems to work correctly, via /sys/power/state, from
> either Terminal or a VT. however, in the case of using a VT to commence
> the suspend, on resume a) the screen often flashes and loses data, and b)
> sometimes many 30 or more) extra shell prompts are produced.  it looks as
> if the enter key which was used to finish the "echo mem" may be
> repeating.
>
> this was os41 (the second version) and q3a16 on a B3.

New description:

 suspend/resume itself seems to work correctly, via /sys/power/state, from
 either Terminal or a VT. however, in the case of using a VT to commence
 the suspend, on resume a) the screen often flashes and loses data, and b)
 sometimes many 30 or more) extra shell prompts are produced.  it looks as
 if the enter key which was used to finish the "echo mem" may be repeating.

 this was os41 (the second version) and q3a16 on a B3.

--

Comment:

 Triage.

 Reproduced the screen flash and loss of data on os40; the data is then
 repainted.

 I've seen the excess enter keystrokes with XO-1 with manual suspend as
 well, the key up event occurs after the system has suspended, and is not
 delivered until some time after resume.

 It is hard to reproduce on XO-1.5 with SD because of how long it takes to
 begin to suspend.  In my tests just now, with os40, it would do it one out
 of every five attempts.

 On os40 with a B2, resuming from VT1 while X is on VT3 also causes a mouse
 cursor to be placed on VT1 (and VT2), which can't be removed by scrolling,
 though it does disappear if you switch to X and back.

-- 
Ticket URL: <http://dev.laptop.org/ticket/9690#comment:1>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system