#12573 HIGH 13.1.0: mwifiex/8787 resume failure with deluge of "mwifiex_sdio_host_to_card: no wr_port available" error messages

[Zarro Boogs per Child]

#12573: mwifiex/8787 resume failure with deluge of "mwifiex_sdio_host_to_card: no
wr_port available" error messages
---------------------------------+------------------------------------------
           Reporter:  shep       |       Owner:  shep                              
               Type:  defect     |      Status:  new                               
           Priority:  high       |   Milestone:  13.1.0                            
          Component:  kernel     |     Version:  Development source as of this date
         Resolution:             |    Keywords:                                    
        Next_action:  never set  |    Verified:  0                                 
Deployment_affected:             |   Blockedby:                                    
           Blocking:             |  
---------------------------------+------------------------------------------

Comment(by shep):

 I spent some time yesterday trying to understand this bug, reading the
 mwifiex code in our olpc-kernel/arm-3.5 branch and also reading mwifiex in
 kernel.org 3.9-rc4.   I did find one significant change upstream in
 601216e12c65e131453941962f2a8f6e49a2a379 which changes significantly the
 scenario which (in my not yet perfect understanding of this bug) seemed
 relevant to the sort of race condition I'm looking for to explain this
 bug.

 I tried "git cherry-pick -x 601216e12c65e131453941962f2a8f6e49a2a379" and
 am now testing.

 So far so good: over 350 wake-on-lan wakeups without any problems, which
 is more than I think I've seen before.

 If this keeps up I will push that later this evening.

-- 
Ticket URL: <http://dev.laptop.org/ticket/12573#comment:10>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system