hostapd/wpa_supplicant - stable release v0.4.11

[Jouni Malinen]

New versions of wpa_supplicant and hostapd were just released and are
now available from http://w1.fi/

This release is a bug fix version from the old stable (0.4.x) branch
and it replaces 0.4.10 as the recommended stable version if you have
not yet updated to the current stable branch (0.5.x).

If you are already using 0.5.10, you have all the fixes included in
this 0.4.11 release. If you are still using 0.3.x versions, I would
recommend upgrading to either 0.5.10 or this 0.4.11 release. There may
be one more 0.3.x bug fix release, but the 0.3.x branch is getting close
to reaching its end.


hostapd:
* updated driver_madwifi to work with changes in madwifi-ng; this is
  needed, e.g., with madwifi 0.9.3
* fixed debugging code not to use potentially unaligned read to fetch
  IPv4 addresses
* fixed EAP-SIM and EAP-AKA message parser to validate attribute
  lengths properly to avoid potential crash caused by invalid messages
* fixed Reassociation Response callback processing when using internal
  MLME (driver_{hostap,devicescape,test}.c)

wpa_supplicant:
* updated driver_madwifi to work with changes in madwifi-ng
* fixed EAP-AKA Notification processing to allow Notification to be
  processed after AKA Challenge response has been sent
* fixed OpenSSL TLS wrapper to clear trusted CA list to allow
  network blocks to use different trusted CA configurations
* fixed a potential EAP state machine loop when moving from PSK to EAP
  configuration without restarting wpa_supplicant
* fixed EAP-SIM and EAP-AKA message parser to validate attribute
  lengths properly to avoid potential crash caused by invalid messages
* added driver_wext workaround for race condition between scanning and
  association with drivers that take very long time to scan all
  channels (e.g., madwifi with dual-band cards); wpa_supplicant is now
  using a longer hardcoded timeout for the scan if the driver supports
  notifications for scan completion (SIOCGIWSCAN event); this helps,
  e.g., in cases where wpa_supplicant and madwifi driver ended up in
  loop where the driver did not even try to associate
* fixed EAP-SIM not to include AT_NONCE_MT and AT_SELECTED_VERSION
  attributes in EAP-SIM Start/Response when using fast reauthentication

-- 
Jouni Malinen                                            PGP id EFC895FA