Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
Commits on Jul 9, 2015
  1. supplicant: convert wpa_printf to wpa_dbg in ctrl_iface file.

    authored
    This allows us to print the station name in the log messages where
    appropriate.  Also, fix one case where NULL wpa_s might have been
    passed to wpa_dbg (end of wpas_global_ctrl_iface_save_config).
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  2. wpa_supplicant: Allow to use external scan results.

    authored
    Helps with how I like to do roaming.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
Commits on May 26, 2015
  1. hs20-client: Fix hostname extraction from url.

    authored
    It was not properly handling cases like this:
    
    https://foo.local:443
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
Commits on May 19, 2015
  1. hs20-server: Fix compile error.

    authored
    Needed to add a new -I path to get it to compile.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  2. spp-client: Warn user if xml file cannot be found.

    authored
    Otherwise, all you get is a cryptic xml validation error out
    of the spp-server.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  3. hs20: Add specific example that worked on Fedora20

    authored
    Hopefully this will make it easier for the next person
    that tries to get this working.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  4. hs20-spp-html: apache + systemd hides the log file.

    authored
    Add note so that the user has a chance to find it more easily.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  5. hs20-client: use configured ca-fname instead of cwd.

    authored
    This gives more flexibility to the hs20 client user.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  6. hwsim: support creating hwsim radio w/out vdev.

    authored
    This way, you can create wlanX interfaces manually with
    the desired name and don't have to worry about what the
    kernel might call the auto-created one.   Should be helpful
    for automated testing.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  7. hwsim: Support creating radio with specific name.

    authored
    Makes it easier to automate tests.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  8. hwsim: Support deleting hwsim by name.

    authored
    More convenient than by index.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  9. supplicant: Fix crash when cleaning up stations.

    authored
    This might have been caused by my earlier changes to
    the scanning related code.
    
        type=0x5f5daa "scan", remove_all=0) at wpa_supplicant.c:3382
        at scan.c:1307
        at wpa_supplicant.c:426
        notify=0, terminate=0) at wpa_supplicant.c:3849
        iface=0x199f2d0) at wpa_supplicant.c:3933
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  10. hostap: Log setting regulatory domain.

    authored
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  11. scan: Properly cancel scans.

    authored
    You need to remove work items as well as cancel events.
    Also, reorder some freq selection logic.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  12. scan: Make scan-on-channel and min-scan-gap work more often.

    authored
    Optimizing this breaks ath10k though, so keep it disabled on
    when system is not configured for concurrent connections, even
    though this logic should only increase the speed at which we can
    connect, not actually do concurrent connections.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  13. supplicant: Allow concurrent association by sharing scan results.

    authored
    This allows one to share scan results and have multiple stations
    associate at the same time.  May cause bugs in some drivers
    so it remains disabled by default.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  14. supplicant: Assume external scans complete within 30 seconds.

    authored
    When testing relatively buggy ath10k driver and firmware,
    I saw cases where supplicant thought external scans were
    active when in fact none were.  This effectively hangs supplicant
    because it will never again scan to start the association logic.
    
    This patch times out external scans after 30 seconds and continues
    on as if the external scan had completed.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  15. eap-sim: Improve derived-key error message.

    authored
    This error is caused by a crypto library that does not have
    proper support for the requested feature, so warn loudly
    about it.
    
    Signed-hostap: Ben Greear <greearb@candelatech.com>
  16. Increase EAPOL timers.

    authored
    This allows 128 stations to associate & authenticate
    all at once.  Without this, 4-way handshake doesn't
    work right.
    
    From: Bjorn Smedman <bjorn.smedman@venatech.se>
    
    NOTE:  This should probably not be applied upstream as is..but
    should be a configuration option in the hostapd.conf file.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  17. hostapd: Use high-priority queue for management packets.

    authored
    Without this patch, wpa_supplicant EAPOL packets (at least)
    are sent on normal best-effort TX queue when using the new
    Linux tx-status-completion API.  I believe they should be on
    the VO high-priority queue instead.
    
    Signed-hostap: Ben Greear <greearb@candelatech.com>
  18. supplicant: Use high-priority queue for management packets.

    authored
    Without this patch, wpa_supplicant EAPOL packets (at least)
    are sent on normal best-effort TX queue.  I believe they
    should be on the VO high-priority queue instead.
    
    Signed-hostap: Ben Greear <greearb@candelatech.com>
  19. nl80211: Remove MLME retry code that causes trouble.

    authored
    When testing 130+ virtual stations, using no encryption
    (no WPA, etc), some number of them will get into endless
    loops of authenticating and de-authenticating.  It appears
    that the code that is supposed to ignore the spurious de-auth
    event created by this retry logic does not work in all cases.
    
    Removing the retry code as this patch does fixes the problem
    for me.
    
    Maybe the entire patch (6d6f4bb) should be reverted?
    
    Signed-hostap: Ben Greear <greearb@candelatech.com>
  20. supplicant: Add state transition timestamps.

    authored
    Helps with understanding how long it takes to do various
    tasks.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  21. Add some documentation relating to EAP-AKA.

    authored
    Signed-hostapd: Ben Greear <greearb@candelatech.com>
  22. rsn: Add debug code for pmksa cache.

    authored
    Dodgy radius server was causing failures.  This code was added
    to debug the problem, but was never used as user figured out RADIUS
    problem first.
    
    Signed-hostap: Ben Greear <greearb@candelatech.com>
  23. supplicant: Allow user-defined defaults for Interworking network blocks.

    authored
    This way users can still configure the HT over-rides and some other
    constraints that Interworking has no interest or ability to configure.
    
    Example config file (to disable HT-40 on an Interworking/HS20 interface):
    
    ctrl_interface=/var/run/wpa_supplicant
    fast_reauth=1
    
    interworking=1
    auto_interworking=1
    access_network_type=0
    
    hs20=1
    
    bss_max_count=2000
    network={
        interworking_defaults=1
        disable_ht=0
        disable_ht40=1
        disable_sgi=0
        ht_mcs=""
        disable_max_amsdu=-1
        ampdu_factor=-1
        ampdu_density=-1
    
    }
    
    cred={
        username="client2"
        password="lanforge"
        ca_cert="/home/lanforge/ca.pem"
        private_key="/home/lanforge/client.p12"
        private_key_passwd="lanforge"
        realm="lanforge.org"
        domain="lanforge.org"
        eap=TLS
    
    }
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  24. supplicant: Update HS20 readme.

    authored
    Mention dependencies needed to properly implement HS20.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  25. nl80211: Increase verbosity for some auth failures.

    authored
    Makes the logs a bit more useful when running in default
    message level (INFO).
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  26. supplicant: Allow limiting number of concurrent associations.

    authored
    When using lots of virtual clients, it may be useful to
    limit the number that attempt to connect to the AP at
    once.  Limitting the associations per scan response is
    an easy way to do this.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  27. wpa_supplicant: Support minimum interval between scan attempts.

    authored
    When running lots of vifs, they can all end up wanting to
    scan right after the other.  This leaves too little time for
    useful work to happen.  Allow users to configure a minimum
    timer interval (in seconds) between scans.  Default is zero,
    which is current behaviour.
    
    Signed-off-by: Ben Greear <greearb@candelatech.com>
  28. hostap: Ensure netlink recv buf size is large enough.

    authored
    I was getting this error on wpa_supplicant startup when configured
    for 200+ stations:
    
    1333514002.074458: netlink: recvfrom failed: No buffer space available
    
    So, increase rcv buffer size to 1MB so we don't drop netlink
    packets.
    
    Signed-hostap: Ben Greear <greearb@candelatech.com>
Commits on May 8, 2015
  1. @jmalinen

    tests: Fix persistent_group_per_sta_psk for P2P Device case

    jmalinen authored
    The dev[1] <--> dev[2] data connectivity test was using incorrect
    function. dev[2] is also using a P2P group and as such, can have a
    different group interface.
    
    Signed-off-by: Jouni Malinen <j@w1.fi>
  2. @jmalinen

    tests: Fix go_neg_with_bss_connected with P2P Device

    jmalinen authored
    Group interface name was fetched from the results of an incorrect group
    formation and because of this, group removal failed in case P2P Device
    is used and dev[1] ends up getting different group ifname for the
    groups.
    
    Signed-off-by: Jouni Malinen <j@w1.fi>
  3. @jmalinen

    tests: Add P2P Device support for persistent_group

    jmalinen authored
    The network operations need to use the global control interface to be
    performed on the interface that stores the network profiles for
    persistent groups.
    
    Signed-off-by: Jouni Malinen <j@w1.fi>
  4. @jmalinen

    tests: Update group_ifname after group start

    Ben Rosenfeld authored jmalinen committed
    After P2P-GROUP-STARTED event, use group_form_result in order to update
    the group_ifname for the device. This is needed when using P2P Device
    for managing P2P operations which results in a separate group interface
    being used.
    
    Signed-off-by: Ben Rosenfeld <ben.rosenfeld@intel.com>
Something went wrong with that request. Please try again.