Re: [madwifi-tickets] [MadWifi] #539: ioctl[IEEE80211_IOCTL_SETMLME]: Invalid argument ERROR :S

#539: ioctl[IEEE80211_IOCTL_SETMLME]: Invalid argument ERROR :S
----------------------------------+-----------------------------------------
      Reporter:  Yansito!         |       Owner:                                
             
          Type:  defect           |      Status:  new                           
             
      Priority:  major            |   Milestone:                                
             
     Component:  madwifi: driver  |     Version:  trunk                         
             
    Resolution:                   |    Keywords:  deauthenticated due to local 
deauth request
Patch_attached:  0                |  
----------------------------------+-----------------------------------------
Comment (by espy@xxxxxxxxxx):

 I also see this error when running wpa_supplicant 0.4.7 ( started with
 ''-Dmadwifi'' ) along with a slightly patched version of madwifi 0.9.0.

 I did a bit of code analysis and it looks to me like the case statement
 inside of ''ieee80211_ioctl_setmlme()'' is incorrect.

 I found that ''vap->iv_nsparams.result'' is only set in two places, the
 ''_setmlme'' function itself ( ''result = 0'' ), and the function
 ''ieee80211_sta_join1_tasklet'' which sets ''result'' to the return value
 from ''ieee80211_new_state()''.

 ''ieee80211_new_state()'' calls ''vap->iv_newstate()'' which gets set to
 ''ieee80211_newstate()'', which it turns out always returns '0' as it's
 return value!

 There's also a function called ''_ _ieee80211_newstate()'' in the call
 stack which only returns '0' as well.

 So, it looks like:
 {{{
 if (vap->iv_nsparams.result)
 }}}

 really should be:
 {{{
 if (!vap->iv_nsparams.result)
 }}}

 ...or since it looks like none of the code in any of the ''_newstate''
 functions ever return any kind of error in the first place, maybe this
 ''if'' check can just be deleted?

 Perhaps at some point, the code should be reviewed in order to make return
 values a little more consistent ( ie. is success always '0', or is it '1'
 )?

-- 
Ticket URL: <http://madwifi.org/ticket/539>
MadWifi <http://madwifi.org/>
Multiband Atheros Driver for Wireless FidelityUsing Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Madwifi-tickets mailing list
Madwifi-tickets@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/madwifi-tickets