This needs to be done both in the more normal location in
p2p_timeout_connect_listen() (internal timeout after driver event) and
in p2p_listen_end() as a workaround for the case where the driver event
is delayed to happen after the internal timeout.
Previously, both NULL and ff:ff:ff:ff:ff:ff addr were used in various
places to indicate default/broadcast keys. Make this more consistent
and useful by defining NULL to mean default key (i.e., used both for
unicast and broadcast) and ff:ff:ff:ff:ff:ff to indicate broadcast
key (i.e., used only with broadcast).
When hostapd is removing a virtual BSS interface, the loop here was
incorrectly not updating the iterator during list traversal and
ended up in an infinite loop in some cases.
Use NULL instead of (u8 *) "" as the seq value and make sure the
driver wrapper implementations can handle NULL value. This was
previously already done in number of places, but not everywhere.
wpa_supplicant seems to crash from time to time on a NetBSD 4.0 MIPS
platform. The root cause turned out to be a MIPS alignment issue.
In my wpa_supplicant crash case, in function
wpa_driver_bsd_event_receive (from driver_bsd.c), the buf[2048] address
is started from i.e. 0x7fffd546, which is not 4 bytes aligned. Later
when it is casted to (struct if_msghdr *), and rtm->rtm_flags is used.
rtm->rtm_flags is "int" type, but its address is not 4 bytes aligned.
This is because the start address of rtm is not 4 bytes aligned.
Unfortunately in NetBSD MIPS kernel (unlike Linux MIPS kernel emulates
unaligned access in its exception handler), the default behavior is to
generate a memory fault to the application that accesses unaligned
memory address. Thus comes the early mentioned wpa_supplicant crash. An
interesting note is when I'm using the wpa_supplicant version 0.4.9, I
never saw this problem. Maybe the stack layout is different. But I
didn't look into details.
I used below patch to resolve this problem. Now it runs correctly for at
least several hours. But you might have a better fix (maybe we can use
malloc/free so that it is at least cache line aligned?). I'm also not
sure if other drivers should have the same problem.
This adds partial callbacks and events to allow P2P management to be
implemented in a driver/firmware. This is not yet complete and is
very much subject to change in the future.
CONFIG_WPS_REG_DISABLE_OPEN=y can be used to configure wpa_supplicant
to disable open networks by default when wps_reg command is used to
learn the current AP settings. When this is enabled, there will be a
WPS-OPEN-NETWORK ctrl_iface event and the user will need to explicitly
enable the network (e.g., with "select_network <id>") to connect to
the open network.
If the underlying driver supports off-channel TX, it will now be used by
the nl80211 driver wrapper, setting WPA_DRIVER_FLAGS_OFFCHANNEL_TX
accordingly.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
With the new kernel functionality coming to Linux to allow off-channel
TX, we can take advantage of that in the P2P code that currently uses
remain-on-channel. If a driver advertises support for it, it will be
asked to handle off-channel TX by itself.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
The nl80211 driver can report low ACK condition (in fact it reports
complete loss right now only). Use that, along with a config option, to
disconnect stations when the data connection is not working properly,
e.g., due to the STA having went outside the range of the AP. This is
disabled by default and can be enabled with disassoc_low_ack=1 in
hostapd or wpa_supplicant configuration file.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
ClientTimeout changes from EAP peer methods were not supposed to
change behavior for other EAP peer methods or even other sessions
of the same method. Re-initialize ClientTimeout whenever an EAP
peer method is initialized to avoid this. This addresses problems
where WPS (EAP-WSC) reduces the timeout and consecutive EAP runs
may fail due to too small timeout.
If the peer you want to connect to is no longer available (does not
acknowledge frames) when wpa_supplicant sends GO Negotition Request
frames, retransmission of this frame is done until the associated
p2p_device structure is removed on timeout. In that case, no signal
is emitted to inform the GO Negotiation has failed.
When sending an Invitation Request frame, the same retransmission
mechanism is in place but limit the transmission to 100 and hitting
the limit generates an event.
This patch adds the same mechanism as the one in place for Invitation
Request, but with limit of 120 to match the existing wait_count for
for GO Negotiation.
If a station received unprotected Deauthentication or Disassociation
frame with reason code 6 or 7 from the current AP, there may be a
mismatch in association state between the AP and STA. Verify whether
this is the case by using SA Query procedure. If not response is
received from the AP, deauthenticate.
This implementation is only for user space SME with
driver_nl80211.c.
Some new code will require access to P2P group members, so add API to
retrieve the number of members and iterate the members themselves.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
If there is a pending GO Negotiation when p2p_cancel is used,
unauthorize the peer to avoid immediate reconnection from being
accepted without a new p2p_connect command.
This can be used by external programs (e.g., wlantest_cli) to inject
raw frames (hex dump of the frame header and body). The data can be
requested to be sent as-is or protected with the current key.
Previously, pairwise and group cipher suites were configured only
when kernel SME (nl80211 connect API) was used. However, mac80211
needs this information even in the user space SME case for one
thing: to disable HT when TKIP/WEP is used. Add
NL80211_ATTR_CIPHER_SUITES_PAIRWISE to fix this special case with
user space SME. This allows mac80211 to disable HT properly when
the AP is configured with configuration that is not allowed.
This add preliminary code for setting the per-STA RX GTK for
RSN IBSS when nl80211 drivers. For some reason, this does not
seem to fully work, but at least driver_nl80211.c is now aware of
what kind of key is being set and the whatever is missing from
making this key configuration go through should be specific to
nl80211/cfg80211.
The frame needs to be sent from an individual (non-group) address,
so drop invalid frames before sending Deauth/Disassoc frames to
not associated STAs.
One of the pointers to the PAC buffer was not updated after realloc
and if the realloc ended up returning new pointer, the *pos pointer
was still pointing at the old location (i.e., freed memory at
this point).
When controlling multiple virtual interfaces on the same physical
radio, share the scan results events with sibling interfaces. This
decreases the time it takes to connect many virtual interfaces.
This is currently only supported on Linux with cfg80211-based
drivers when using nl80211 or wext driver interface.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Some drivers are not providing exactly reliable error codes (e.g.,
with WEXT), but others may actually indicate reliable information.
Allow driver wrappers to indicate if that is the case and use
optimizations if so. For now, this improves nl80211 with
NL80211_CMD_CONNECT for a case where connection request fails.
driver_param=use_p2p_group_interface=1 can now be used to test
nl80211-drivers with separate P2P group interface. In other words,
the main interface (e.g., wlan0) is reserved for P2P management
operations and non-P2P connections and a new group interface (e.g.,
p2p-wlan0-0) is created for the P2P group.
This implementation is very minimal, i.e., it only support address
allocation for a single P2P group interface (if the driver does not
handle this internally). In addition, not all functionality has yet
been tested, so for now, this is disabled by default and needs that
special driver_param to enable.
WPA_DRIVER_FLAGS_P2P_MGMT_AND_NON_P2P flag can now be used to
indicate that the initial interface (e.g., wlan0) is used for
P2P management operations and potentially non-P2P connections.
This is otherwise identical to
WPA_DRIVER_FLAGS_P2P_DEDICATED_INTERFACE, but the possibility of
non-P2P connections makes some operations differ.
Getting rid of these inline functions seems to reduce the code size
quite a bit, so convert the most commonly used hostapd driver ops to
function calls.
This is not needed anymore and just makes things more difficult
to understand, so move the remaining function pointers to direct
function calls and get rid of the struct hostapd_driver_ops.
send_eapol, set_key, read_sta_data, sta_clear_stats,
set_radius_acl_auth, set_radius_acl_expire, and set_beacon
to use inline functions instead of extra abstraction.
Commit bf65bc638f started the path to
add this new abstraction for driver operations in AP mode to allow
wpa_supplicant to control AP mode operations. At that point, the
extra abstraction was needed, but it is not needed anymore since
hostapd and wpa_supplicant share the same struct wpa_driver_ops.
Start removing the unneeded abstraction by converting
send_mgmt_frame() to an inline function, hostapd_drv_send_mlme().
This is similar to the design that is used in wpa_supplicant and
that was used in hostapd in the past (hostapd_send_mgmt_frame()
inline function).
driver.h defines these functions to return 0 on success, not
number of bytes transmitted. Most callers are checking "< 0" for
error condition, but not all. Address this by following the driver
API specification on 0 meaning success.
On Linux, verify that the kernel entropy pool is capable of providing
strong random data before allowing WPA/WPA2 connection to be
established. If 20 bytes of data cannot be read from /dev/random,
force first two 4-way handshakes to fail while collecting entropy
into the internal pool in hostapd. After that, give up on /dev/random
and allow the AP to function based on the combination of /dev/urandom
and whatever data has been collected into the internal entropy pool.
wlan0: RADIUS No authentication server configured
MEMLEAK[0x999feb8]: len 1040
WPA_TRACE: memleak - START
[3]: ./hostapd(radius_msg_new+0x33) [0x8074f43]
radius_msg_new() ../src/radius/radius.c:117
[4]: ./hostapd() [0x806095e]
ieee802_1x_encapsulate_radius() ../src/ap/ieee802_1x.c:439
ieee802_1x_aaa_send() ../src/ap/ieee802_1x.c:1496
For example, this error occured when I used WPS hostapd without
"eap_server=1" definition in configuration file.
By default, make hostapd and wpa_supplicant maintain an internal
entropy pool that is fed with following information:
hostapd:
- Probe Request frames (timing, RSSI)
- Association events (timing)
- SNonce from Supplicants
wpa_supplicant:
- Scan results (timing, signal/noise)
- Association events (timing)
The internal pool is used to augment the random numbers generated
with the OS mechanism (os_get_random()). While the internal
implementation is not expected to be very strong due to limited
amount of generic (non-platform specific) information to feed the
pool, this may strengthen key derivation on some devices that are
not configured to provide strong random numbers through
os_get_random() (e.g., /dev/urandom on Linux/BSD).
This new mechanism is not supposed to replace proper OS provided
random number generation mechanism. The OS mechanism needs to be
initialized properly (e.g., hw random number generator,
maintaining entropy pool over reboots, etc.) for any of the
security assumptions to hold.
If the os_get_random() is known to provide strong ramdom data (e.g., on
Linux/BSD, the board in question is known to have reliable source of
random data from /dev/urandom), the internal hostapd random pool can be
disabled. This will save some in binary size and CPU use. However, this
should only be considered for builds that are known to be used on
devices that meet the requirements described above. The internal pool
is disabled by adding CONFIG_NO_RANDOM_POOL=y to the .config file.
This commit adds a new wrapper, random_get_bytes(), that is currently
defined to use os_get_random() as is. The places using
random_get_bytes() depend on the returned value being strong random
number, i.e., something that is infeasible for external device to
figure out. These values are used either directly as a key or as
nonces/challenges that are used as input for key derivation or
authentication.
The remaining direct uses of os_get_random() do not need as strong
random numbers to function correctly.
This adds more time for the system entropy pool to be filled before
requesting random data for generating the WPA/WPA2 encryption keys.
This can be helpful especially on embedded devices that do not have
hardware random number generator and may lack good sources of
randomness especially early in the bootup sequence when hostapd is
likely to be started.
GMK and Key Counter are still initialized once in the beginning to
match the RSN Authenticator state machine behavior and to make sure
that the driver does not transmit broadcast frames unencrypted.
However, both GMK (and GTK derived from it) and Key Counter will be
re-initialized when the first station connects and is about to
enter 4-way handshake.
The example GMK-to-GTK derivation described in the IEEE 802.11 standard
is marked informative and there is no protocol reason for following it
since this derivation is done only on the AP/Authenticator and does not
need to match with the Supplicant. Mix in more data into the derivation
process to get more separation from GMK.
The previous commit broke completion in various places. The proper
way of handling the completion of full word is to verify whether
there are more than one possible match at that point.
Completion needs to be done even if the full word has been entered.
In addition, fix the space-after-full-word to properly allocate room
for the extra character when completion is used in the middle of the
string.
This adds new commands for wlantest_cli to request wlantest to
inject frames. This version can only send out Authentication
frames and unprotected SA Query Request frames, but there is
now place to add more frames and encryption with future commits.
ap_setup_locked=2 can now be used to enable a special mode where
WPS ER can learn the current AP settings, but cannot change then.
In other words, the protocol is allowed to continue past M2, but
is stopped at M7 when AP is in this mode. WPS IE does not
advertise AP Setup Locked in this case to avoid interoperability
issues.
In wpa_supplicant, use ap_setup_locked=2 by default. Since the AP PIN
is disabled by default, this does not enable any new functionality
automatically. To allow the read-only ER to go through the protocol,
wps_ap_pin command needs to be used to enable the AP PIN.
There is no point in maintaining two almost identical versions
of this parser. Move WPA IE parser into wpa_common.c similarly
to what was already the case with RSN IE parse.
When multiple wireless interfaces are used with WPS, the UPnP
subscriptions need to be removed whenever a matching Registrar
instance gets removed. This avoids a segfault due to access to
freed memory during hostapd shutdown.
In addition, the UPnP interface instance structure needs to be
freed to avoid memory leak.
WPS external Registrars can get confused about multiple UPnP
instances (one per radio) on a dual-concurrent APs. Simplify the
design by sharing a single UPnP state machine for all wireless
interfaces controlled by hostapd. This matches with the previous
changes that made a single command enable WPS functionality on
all interfaces.
This is relatively minimal change to address the sharing of the
state among multiple struct hostapd_data instances. More cleanup
can be done separately to remove unnecessary copies of information.
When the peer does not include our initial preference in the Channel
List attribute during GO Negotiation, try to use the best channel of
the other band as the new preference instead of falling back to the
first channel in the intersection.
The driver wrapper may now indicate the preferred channel (e.g., based
on scan results) on both 2.4 GHz and 5 GHz bands (and an overall best
frequency). When setting up a GO, this preference information is used
to select the operating channel if configuration does not include
hardcoded channel. Similarly, this information can be used during
GO Negotiation to indicate preference for a specific channel based
on current channel conditions.
p2p_group_add command can now use special values (freq=2 and freq=5)
to indicate that the GO is to be started on the specified band.
Commit d8d940b746 introduced a regression
that prevented TSN APs from being used with WEP since the AP was
rejected if it advertised WPA or RSN IE when we were configured to use
WEP. Resolve this by checking whether the AP is advertising a TSN, i.e.,
whether the AP allows WEP to be used as a group cipher. If so, allow
the AP to be selected if we are configured to use static WEP or
IEEE 802.1X (non-WPA).
It should be noted that this is still somewhat more restricted in AP
selection than earlier wpa_supplicant branches (0.7.x or older) that
ignore the WPA/RSN IE completely when configured for non-WPA.
This tool can be used to capture IEEE 802.11 frames either from a
monitor interface for realtime capturing or from pcap files for
offline analysis. This version is only adding basic infrastructure for
going through the frames and parsing their headers.
Commit c3fc47ea8e fixed EAP passthrough
server to allow Logoff/Re-authentication to be used. However, it
broke EAP standalone server while doing that. Fix this by reverting
the earlier fix and by clearing the EAP Identity information in the
EAP server code whenever an EAPOL-Start or EAPOL-Logoff packet is
received.
We need to drop the pending frame to avoid issues with the new GO
Negotiation, e.g., when the pending frame was from a previous attempt at
starting a GO Negotiation.
These are not used by any driver wrapper, i.e., only the four
data queues (BK, BE, VI, VO) are configurable. Better remove these
so that there is no confusion about being able to configure
something additional.
Pass data frames from unknown STAs to hostapd in order to reply with
a Deauthentication or Disassociation frame. This fixes compliance
with IEEE Std 802.11-2007, 11.3.
Furthermore, this does not cause a lot of overhead (at least with
mac80211 drivers) since mac80211 does not pass all data frames (but
at least from unauthenticated and unassociated STAs) to cooked monitor
interfaces.
Tested with rt2800pci on a MIPS board.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
drivers.mak is shared between hostapd and wpa_supplicant and
compiling them together may create "wrong" flags.
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
We are not actually interested in the PD Response in join-a-group
case, so there is no point in trying to send PD Request until the
response is received. This avoids an extra PD getting started after
a join-a-group operation in some cases.
This is needed to avoid problems with other applications setting and
leaving the interface to IF_OPER_DORMANT state. In AP mode, the interface
is ready immediately after the keys are set, so we better make sure the
DORMANT state does not prevent normal operations after that.
This is needed to avoid issues with APs that restart their UPnP,
e.g., when ER reconfigures them. The previously known settings are
now cached and taken into use if an AP is detected to leave
(ssdp:byebye) and then return.
Verify that the UUID given to wps_er_pbc command is known and only
activate PBC mode on the matching AP. The UUID can be that of the
AP or the station/Enrollee.
Add a new driver event, EVENT_INTERFACE_UNAVAILABLE, for indicating
that the driver is not able to continue operating the virtual
interface in its current mode anymore, e.g., due to operating
channel for GO interface forced to a DFS channel by another virtual
interface.
When this happens for a P2P group interface, the P2P group will
be terminated and P2P-GROUP-REMOVED event shows the reason for
this as follows:
P2P-GROUP-REMOVED wlan0 GO reason=UNAVAILABLE
A new configuration parameter, p2p_group_idle, can now be used to set
idle timeout value for P2P groups in seconds (0 = no timeout). If set,
this values is used to remove P2P group (both GO and P2P client)
interfaces after the group has been idle (no clients/GO seen) for the
configuration duration.
The P2P-GROUP-REMOVED event is now indicating the reason for group
removal when known. For example:
P2P-GROUP-REMOVED wlan0 GO reason=REQUESTED
P2P-GROUP-REMOVED wlan1 client reason=IDLE
Some driver wrappers may end up indicating a disassociation or
deauthentication event without the address of the station, e.g.,
based on a previous non-AP mode event. Avoid crashing hostapd
by verifying that the require address parameter is available in
the event before processing it.
In order to avoid confusing the driver with a new remain-on-channel
request, delay sending of a new Action frame if the driver indicates
Action frame RX diromg a scan.
This is needed to optimize response to GO Negotiation Request frames.
The extra remain-on-channel cancel followed by new remain-on-channel for
the same channel takes too much time with some driver/firmware
designs for the response to go out quickly enough to avoid peer
timing out while waiting for our response.
This command can be used to enroll a network based on a local
network configuration block instead of having to (re-)learn the
current AP settings with wps_er_learn.
This test command is supposed to change the WPS version number in all
places immediately, so make sure that the IEs used in management
frames get updated immediately.
The P2P group component is now tracking of associated stations
in the group and the Group Limit bit in the Group Capabilities
is updated based on whether there is room for new clients in
the group.
The workaround to ignore no ctrl::ack received for GO Negotiation
Confirmation frame was only supposed to be used when the frame was
actually transmitted and just the ack was not received. However, due
to the way the driver failure on transmitting the frame were reported,
this ended up getting applied for all failures in sending the GO
Negotiation Confirmation frame.
Improve this by providing a mechanism to indicate whether send_action
operations fail locally before the frame was actually transmitted or
because of not receiving ack frame after having transmitted the frame.
UPnP event subscriptions are not supposed to be dropped based on
whether events can be delivered quickly enough. Leave dropping to
happen only based on failed deliveries to avoid issues with a burst
of events kicking out still active subscribers.
Do not send more than five Probe Request WLANEvent notifications
per second. Even though the limit should in theory apply to all
WLANEvents, it is better not to drop EAP notifications because
of Probe Request frames and really, the only real reason for
event bursts is Probe Request frames.
Instead of queuing all events for a subscriber, only queue more
important events if delivery of event notifications starts failing.
This allows more time for the subscriber to recover since the maximum
queue length if not reached because of Probe Request frames only.
Instead of dropping the subscription on first failure, allow up to
10 failures before dropping. In addition, drop the callback URLs
one by one instead of full subscription if only one URL is failing.
