#9836 NORM 1.5-sof: BUG() in if_sdio_handle_cmd()

[Zarro Boogs per Child]

#9836: BUG() in if_sdio_handle_cmd()
-------------------------------+--------------------------------------------
           Reporter:  dsaxena  |       Owner:  dsaxena           
               Type:  defect   |      Status:  new               
           Priority:  normal   |   Milestone:  1.5-software-later
          Component:  kernel   |     Version:  1.5-B3            
         Resolution:           |    Keywords:                    
        Next_action:  code     |    Verified:  0                 
Deployment_affected:           |   Blockedby:                    
           Blocking:           |  
-------------------------------+--------------------------------------------

Comment(by pgf):

 it would be really useful to include test scripts used to reproduce bugs
 and verify fixes in trac along with the bug.

 i've built without our "ugly workaround" and with the marvell patch, and
 things seem to be working.  i can recreate (i think) the original
 symptom(s) with the following:

 from another machine, do a ping flood of the laptop while it runs this:

 {{{
 [root at xo-a7-2a-32 olpc]# /etc/init.d/NetworkManager stop
 Stopping NetworkManager daemon: [  OK  ]
 [root at xo-a7-2a-32 olpc]# stop powerd
 powerd (stop) running, process 2284
 powerd (stop) pre-stop, (main) process 2284
 powerd (stop) stopping, process 2284
 powerd (stop) killed, process 2284
 [  150.053173] dcon_freeze_store: 0
 powerd (stop) post-stop
 powerd (stop) waiting
 [root at xo-a7-2a-32 olpc]# cat netsetup
 ifconfig eth0 down
 dhclient -r eth0
 ifconfig eth0 up
 iwconfig eth0 essid "ml1cc"
 iwconfig eth0 mode Managed
 dhclient eth0
 [root at xo-a7-2a-32 olpc]# while :; do echo awake;sleep 2;./netsetup; sleep
 3; echo asleep; ethtool -s eth0 wol u; rtcwake -s 4 -m mem; done
 awake
 asleep
 rtcwake: assuming RTC uses UTC ...
 rtcwake: wakeup from "mem" using /dev/rtc0 at Wed Jun 23 17:52:19 2010

 +rawake
 asleep
 rtcwake: assuming RTC uses UTC ...
 rtcwake: wakeup from "mem" using /dev/rtc0 at Wed Jun 23 17:52:35 2010

 +rawake
 asleep
 rtcwake: assuming RTC uses UTC ...
 rtcwake: wakeup from "mem" using /dev/rtc0 at Wed Jun 23 17:52:48 2010

 þ+rawake
 asleep
 rtcwake: assuming RTC uses UTC ...

 }}}

 i'm seeing what's possibly a new symptom, though, and that is that after
 running this for a while on XO-1.5, the wireless capabilities of eth0
 altogether.  eth0 doesn't go away -- just its wireless extensions.

-- 
Ticket URL: <http://dev.laptop.org/ticket/9836#comment:28>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system