#12537 NORM Not Tri: mwifiex/8787 failure to get packets transmitted effectively after fresh boot, rmmod/modprobe of mwifiex_sdio fixed it.

[Zarro Boogs per Child]

#12537: mwifiex/8787 failure to get packets transmitted effectively after fresh
boot, rmmod/modprobe of mwifiex_sdio fixed it.
--------------------------+-------------------------------------------------
 Reporter:  shep          |                 Owner:  shep                              
     Type:  defect        |                Status:  new                               
 Priority:  normal        |             Milestone:  Not Triaged                       
Component:  not assigned  |               Version:  Development source as of this date
 Keywords:                |           Next_action:  never set                         
 Verified:  0             |   Deployment_affected:                                    
Blockedby:                |              Blocking:                                    
--------------------------+-------------------------------------------------
 On an XO-4 with mwifiex/8787 wireless module with
 {{{
 /etc/powerd/flags/inhibit-suspend
 }}}
 after a fresh boot network appeared not to work at all, but it had managed
 to obtain an IPv4 address via DHCP. (Maybe it was using a cached value?
 I'm not sure how reliable the presence of the IPv4 address is as an
 indication of packet transmit having worked.)

 Further investigation with ping, tcpdump and looking at the arp cache
 showed that packets were being received OK  by mwifiex/8787 but I could
 find no evidence of any packets getting out.

 No error messages appeared on the console, and dmesg log showed nothing
 unusual.

 I tried:

 {{{

 rmmod mwifiex_sdio
 modprobe mwifiex_sdio

 }}}

 and that restored connectivity.

 Note that this sort of failure is unlikely to be noticed for long on an
 XO-4 with suspend/resume enabled and
 {{{
 /sys/module/mwifiex_sdio/parameters/mwifiex_always_poweroff_on_sleep
 }}}
 turned on because the card gets power-cycled and reset each time the
 laptop suspends and resumes.

 I think I may have seen a similar incident a week or so ago, but didn't
 investigate much then.

-- 
Ticket URL: <http://dev.laptop.org/ticket/12537>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system