#12636 HIGH 13.2.0: mwifiex/8787 resume failure

[Zarro Boogs per Child]

#12636: mwifiex/8787 resume failure
---------------------------------+------------------------------------------
           Reporter:  wad        |       Owner:  shep                             
               Type:  defect     |      Status:  new                              
           Priority:  high       |   Milestone:  13.2.0                           
          Component:  wireless   |     Version:  Development build as of this date
         Resolution:             |    Keywords:                                   
        Next_action:  reproduce  |    Verified:  0                                
Deployment_affected:             |   Blockedby:                                   
           Blocking:             |  
---------------------------------+------------------------------------------

Comment(by shep):

 Wad, this does not appear likely to be a separate bug.  (But thanks for
 the report anyway.)

 The few isolated "no wr_port e" messages do appear to suggest this is a
 bit different, however it looks more like this is #12573's error messages
 with a bug in how you are logging the serial console output (the e is in
 column 80, like your terminal emulator you are using is not set up to wrap
 properly, so the "availabl" that precedes the e got overwritten and then
 you did something like cut-and-paste in some way that didn't preserve all
 the received characters).

 I have seen the rapid error messages from #12573 eventually stop on their
 own after several minutes, leaving a non-working interface and few or no
 further error messages.

 Also, #12537 can eventually get you to the point eventually where the arp
 cache has timed out and cannot be refreshed, so you start getting
 unreachable errors.  (#12573 probably does that too.)

 #12569, #12573, and #12537 all leave you with non-working wireless network
 (#12537 with an effectively non-working wireless network, though packets
 can still be received, nothing works as there's nothing prepared to do
 anything useful with the remaining one-way connectivity).

 For now I'm not planning on looking into this as as separate bug, but
 leave it hanging here until the others are resolved.

-- 
Ticket URL: <http://dev.laptop.org/ticket/12636#comment:2>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system