#12197 NORM 13.1.0: xo-4 page faults on boot. mwifiex?

[Zarro Boogs per Child]

#12197: xo-4 page faults on boot.  mwifiex?
---------------------------------+------------------------------------------
           Reporter:  pgf        |       Owner:  shep                             
               Type:  defect     |      Status:  new                              
           Priority:  normal     |   Milestone:  13.1.0                           
          Component:  wireless   |     Version:  Development build as of this date
         Resolution:             |    Keywords:                                   
        Next_action:  reproduce  |    Verified:  0                                
Deployment_affected:             |   Blockedby:                                   
           Blocking:             |  
---------------------------------+------------------------------------------

Comment(by shep):

 Summary of the  one_hundred_boots_4b1_8787_os13_console.log.gz  I just
 attached:

 After removing my USB flash drive from my 4b1 laptop with os13, I used the
 P1
 command on the EC console to turn the power on 101 times.


 One power on resulted in:

 {{{
 CForth built 2012-11-15 22:27 from commit bbdf8572...
 Matrix mode on
 releasing
 Data Abort
 ok
 }}}

 and the other 100 times it got linux booting.   Of the 100 boots of linux,
 all got at least to the point of printing out:

 {{{
           mwifiex_sdio mmc0:0001:1: WLAN FW is active

 }}}

 Two of those (the 51st and 64th boot) resulted in serial console spewage
 after that.  One of them I powered off with P0 after about 30 seconds,
 the other stopped the spewage on its own after about 35 seconds.

 Five of these (the 14th, 43rd, 59th, 74th, and 90th boot) resulted in a
 complete wedgeup with no more console output.   In each of these cases,
 I used the P0 command on the EC console to force power off and then the
 P1 command to start the next power up and boot and in each case the next
 boot was succesful.

 The last 3 lines of console output before the wedgeup were always:


 {{{
                mwifiex_sdio mmc0:0001:1: WLAN FW already running! Skip FW
 dnld
                mwifiex_sdio mmc0:0001:1: WLAN FW is active
                mwifiex_sdio mmc0:0001:1: driver_version = mwifiex 1.0
 (14.66.9.p96)
 }}}

 and the next line of console output on a succesful boot would normally be:

 {{{
                systemd-udevd[261]: renamed network interface mlan0 to eth0

 }}}

 The remaining 93 linux boots were succesful and I was able to get a shell
 prompt and issue a "shutdown -h now" command.   Lots of random craziness
 happened on shutdown if you watch the serial console output.  In two cases
 the shutdown took a long time.  In one of those cases I lost patience and
 forced power off with P0 on the EC console.   In the other case I was
 distracted (and hence more patient) and it managed to shut itself off
 after a couple of minutes.

-- 
Ticket URL: <http://dev.laptop.org/ticket/12197#comment:5>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system