#12039 NORM Not Tri: Botched resume +X on XO-1.5

[Zarro Boogs per Child]

#12039: Botched resume +X on XO-1.5
---------------------------------+------------------------------------------
 Reporter:  dsd                  |                 Owner:  wmb at firmworks.com
     Type:  defect               |                Status:  new              
 Priority:  normal               |             Milestone:  Not Triaged      
Component:  ofw - open firmware  |               Version:  not specified    
 Keywords:                       |           Next_action:  never set        
 Verified:  0                    |   Deployment_affected:                   
Blockedby:                       |              Blocking:                   
---------------------------------+------------------------------------------
 Looking at old IRC logs it seems like the bug where the XO-1.5
 occasionally performs a full reboot on resume after printing "+X" is not
 an unknown issue.

 But this issue is still present today, and we have a feeling that it might
 be more common than before (with new kernel, firmware, motherboard
 revision?).

 Here's how it appears:

 {{{
 screenlog.3-4ddc646-reboots-[ 2007.229908] libertas_sdio mmc1:0001:1:
 Suspend without wake params -- powering down card
 screenlog.3-4ddc646-reboots-[ 2007.250703] mmc1: card 0001 removed
 screenlog.3-4ddc646-reboots-[ 2007.320139] PM: suspend of devices complete
 after 224.434 msecs
 screenlog.3-4ddc646-reboots-[ 2007.326607] ehci_hcd 0000:00:10.4: wake-up
 capability enabled by ACPI
 screenlog.3-4ddc646-reboots-[ 2007.350165] uhci_hcd 0000:00:10.2: wake-up
 capability enabled by ACPI
 screenlog.3-4ddc646-reboots-[ 2007.370170] uhci_hcd 0000:00:10.1: wake-up
 capability enabled by ACPI
 screenlog.3-4ddc646-reboots-[ 2007.390159] uhci_hcd 0000:00:10.0: wake-up
 capability enabled by ACPI
 screenlog.3-4ddc646-reboots-[ 2007.410391] PM: late suspend of devices
 complete after 84.203 msecs
 screenlog.3-4ddc646-reboots-[ 2007.416905] ACPI: Preparing to enter system
 sleep state S3
 screenlog.3-4ddc646-reboots-[ 2007.520175] PM: Saving platform NVS memory
 screenlog.3-4ddc646-reboots:+r+r+X+
 screenlog.3-4ddc646-reboots-Forthmacs
 screenlog.3-4ddc646-reboots-Type 'i' to interrupt stand-init sequence
 screenlog.3-4ddc646-reboots-Unknown value in TS tag
 screenlog.3-4ddc646-reboots-USB2 devices:
 screenlog.3-4ddc646-reboots-USB1 devices:
 screenlog.3-4ddc646-reboots-OLPC D5, 1 GiB memory installed, 4 GB internal
 storage, S/N SHC01901E0B
 screenlog.3-4ddc646-reboots-OpenFirmware  Q3C07   EC Firmware Ver:2.2.10
 2012-08-02 16:36:31 UTC
 screenlog.3-4ddc646-reboots-
 --
 screenlog.3-4ddc646-reboots-[ 2009.314240] libertas_sdio mmc1:0001:1:
 Suspend without wake params -- powering down card
 screenlog.3-4ddc646-reboots-[ 2009.330793] mmc1: card 0001 removed
 screenlog.3-4ddc646-reboots-[ 2009.400142] PM: suspend of devices complete
 after 224.617 msecs
 screenlog.3-4ddc646-reboots-[ 2009.406586] ehci_hcd 0000:00:10.4: wake-up
 capability enabled by ACPI
 screenlog.3-4ddc646-reboots-[ 2009.430167] uhci_hcd 0000:00:10.2: wake-up
 capability enabled by ACPI
 screenlog.3-4ddc646-reboots-[ 2009.450174] uhci_hcd 0000:00:10.1: wake-up
 capability enabled by ACPI
 screenlog.3-4ddc646-reboots-[ 2009.470161] uhci_hcd 0000:00:10.0: wake-up
 capability enabled by ACPI
 screenlog.3-4ddc646-reboots-[ 2009.490395] PM: late suspend of devices
 complete after 84.229 msecs
 screenlog.3-4ddc646-reboots-[ 2009.496925] ACPI: Preparing to enter system
 sleep state S3
 screenlog.3-4ddc646-reboots-[ 2009.600176] PM: Saving platform NVS memory
 screenlog.3-4ddc646-reboots:+r+X+
 screenlog.3-4ddc646-reboots-Forthmacs
 screenlog.3-4ddc646-reboots-Type 'i' to interrupt stand-init sequence
 screenlog.3-4ddc646-reboots-Unknown value in TS tag
 screenlog.3-4ddc646-reboots-USB2 devices:
 screenlog.3-4ddc646-reboots-USB1 devices:
 screenlog.3-4ddc646-reboots-OLPC D5, 1 GiB memory installed, 4 GB internal
 storage, S/N SHC01901E0B
 screenlog.3-4ddc646-reboots-OpenFirmware  Q3C07   EC Firmware Ver:2.2.10
 2012-08-02 17:24:59 UTC
 }}}

 Sam has managed to reproduce this several times in the last few hours over
 a group of 5 laptops. I also hit it once this morning with 1 laptop. My
 test case (which is practically the same as what sam is doing) is:
  1. Boot 12.1.0 build 18
  2. Update to latest x86-3.3 kernel
  3. Modify runin code to disable the X part
  4. touch /runin/{soiled,aggressive,no-camera,force}
  5. reboot

 Mitch, when back from bringup and when you aren't flooded with high-
 priority tasks, perhaps we could run a debugging session to get a better
 understanding of this issue and maybe a fix.

-- 
Ticket URL: <http://dev.laptop.org/ticket/12039>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system