hostapd deinit functions were not ready to handle a case where the data
structures were not fully initialized. Make these more robust to allow
wpa_supplicant mesh implementation to use the current deinit design in
OOM error cases without causing NULL pointer dereferences.
Signed-off-by: Jouni Malinen <j@w1.fi>
Instead of reporting the memory allocation failure and stopping, run the
scan even if the frequency list cannot be created due to allocation
failure. This allows the wpa_s->reattach flag to be cleared and the scan
to be completed even if it takes a bit longer time due to all channels
getting scanned.
Signed-off-by: Jouni Malinen <j@w1.fi>
The code contained some places that used an additional argument for
setup_hw after -R and also contained places where setup_hw cmdline was
passed as a string instead of an argument list. It also contained places
where the ifname was only treated as a single interface and disregarded
the possiblity of multiple interfaces. This commit fixes these issues
and executes setup_hw from a single function for all cases.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
Use a monitor interface given in the command line that is not also a
station or an AP as a monitor running wlantest on the channel used by
the test. This makes all the tests that use wlantest available for
execution on real hardware on remote hosts.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
In monitor.py in the remote tests code there is fucntion create() that
creates standalone monitor interfaces. In this function there is an
iteration of the ifaces of the host by using the ifaces variable but
this variable is non-existing. This patch creates this variable before
its usage.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
In wpaspy.py in the Ctrl object constructor there is a try/except. In
the except part the code references the s attribute of the object. This
attribute is only created later in the try part. If an exception occurs
before the attribute creation then the except part references a non
existing attribute. Fix that by assigning None to the s attribute at the
beginning of the try part.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
This function is used for remote tests when a monitor interface is
needed on the channel on which the AP operates. This change enables us
to also query P2P interfaces for the channel information to use for
monitor interfaces.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
Instead of accessing the logs list member of the remote host directly,
use a function to add logs to the remote host to be collected after the
test. This enables us to later have different implementation of remote
hosts or logs collection without requiring to have this list as the
implementation.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
The regular hwsim tests use both unicast and broadcast frames to test
the connectivity between 2 interfaces. For real hardware (remote hwsim
tests) the broadcast frames will sometimes not be seen by all connected
stations since they can be in low power mode during DTIM or because
broadcast frames are not ACKed. Use 10 retries for broadcast
connectivity tests for real hardware so that the test will pass if we
successfully received at least one of them.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
This avoids internal access of structs and also removes the dependency
on the reimplemented TLS PRF functions when EAP-FAST support is not
enabled. Notably, BoringSSL doesn't support EAP-FAST, so there is no
need to access its internals with openssl_get_keyblock_size().
Signed-Off-By: David Benjamin <davidben@google.com>
This is needed to work with the tls_openssl.c changes that renamed the
function that is used for deriving the EAP-FAST keys.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Most protocols extracting keys from TLS use RFC 5705 exporters which is
commonly implemented in TLS libraries. This is the mechanism used by
EAP-TLS. (EAP-TLS actually predates RFC 5705, but RFC 5705 was defined
to be compatible with it.)
EAP-FAST, however, uses a legacy mechanism. It reuses the TLS internal
key block derivation and derives key material after the key block. This
is uncommon and a misuse of TLS internals, so not all TLS libraries
support this. Instead, we reimplement the PRF for the OpenSSL backend
and don't support it at all in the GnuTLS one.
Since these two are very different operations, split
tls_connection_prf() in two. tls_connection_export_key() implements the
standard RFC 5705 mechanism that we expect most TLS libraries to
support. tls_connection_get_eap_fast_key() implements the
EAP-FAST-specific legacy mechanism which may not be implemented on all
backends but is only used by EAP-FAST.
Signed-Off-By: David Benjamin <davidben@google.com>
Commit 68ae4773a4 ('OpenSSL: Use library
wrapper functions to access cert store') fixed most of these, but missed
a few.
Signed-Off-By: David Benjamin <davidben@google.com>
If preassoc_mac_addr is used and updating the MAC address fails in
wpas_trigger_scan_cb(), the cloned scan parameters were leaked. Fix that
and also send a CTRL-EVENT-SCAN-FAILED event in this and another error
case.
Signed-off-by: Jouni Malinen <j@w1.fi>
There is no need to have a separate if statement to skip the cases where
phase1 is not set. Just check it with the strstr comparison since this
case is not really used in practice.
Signed-off-by: Jouni Malinen <j@w1.fi>
wpa_s->current_ssid is set to a non-NULL ssid pointer value here, so
there is no need for the extra if statement.
Signed-off-by: Jouni Malinen <j@w1.fi>
The P2P Device interface can only send Public Action frames. Non-Public
Action frames must be sent over a group interface. The previous
implementation sometimes tried to send non-Public Action frames such as
GO Discoverability over the P2P Device interface, however, the source
address of the frame was set to the group interface address so the code
in offchannel.c knew to select the correct interface for the TX.
The check breaks when the P2P Device and group interfaces have the same
MAC address. In this case the frame will be sent over the P2P Device
interface and the send will fail.
Fix this problem in two places:
1. In offchannel, route non-Public Action frames to the GO
interface when the above conditions are met.
2. When a TX_STATUS event arrives on such routed frame, it will
arrive on the GO interface but it must be handled by the P2P Device
interface since it has the relevant state logic.
Signed-off-by: Lior David <qca_liord@qca.qualcomm.com>
This makes changes such that hostapd (and wpa_supplicant AP mode)
registers to kernel for specific Action frames instead of generically
registering for all Action frames. This makes it easier for other
programs to register for some Action frames that hostapd does not handle
today without having to somehow coordinate directly with hostapd.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This allows using P2PS config method with or without PIN for connection.
wpa_supplicant should internally handle the default PIN "12345670" and
shall also allow connection irrespective of PIN used in P2P_CONNECT.
For example,
1. P2P_CONNECT 02:2a:fb:22:22:33 p2ps
2. P2P_CONNECT 02:2a:fb:22:22:33 xxxxxxxx p2ps
Where the second one is maintained for backwards compatibility.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Add P2PS config flag only when config_methods are set. This restores the
pre-P2PS behavioer for the cases where Display or Keypad config method
is specified for a peer (i.e., do not add the new P2PS method in that
case).
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This gets rid of a valgrind warning on uninitialized memory read in the
eap_proto_sake_errors test case where the result was used after the
failed eap_sake_compute_mic() call.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This gets rid of a valgrind warning on uninitialized memory read in the
eap_proto_pax_errors test case where the result was used after the
failed eap_pax_mac() call.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This gets rid of a valgrind warning on uninitialized memory read in the
eap_proto_fast_errors test case where the result was used after the
failed sha1_t_prf() call.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This gets rid of a valgrind warning on uninitialized memory read in the
wpas_ctrl_error test case where the result was used after the failed
sha256_vector() call.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This gets rid of a valgrind warning on uninitialized memory read in the
hostapd_oom_wpa2_eap_connect test case where the result is used after
failed md5_vector() call.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This gets rid of a valgrind warning on uninitialized memory read in the
hostapd_oom_wpa2_eap_connect test case where memcmp is used after failed
hmac_md5() call.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This gets rid of a valgrind warning on uninitialized memory read in the
hostapd_oom_wpa2_eap test case where memcmp is used after failed
md5_vector() call.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Check for hmac_sha256() failures and exit from wps_derive_psk() without
printing out the derived keys if anything fails. This removes a valgrind
warning on uninitialized value when running the ap_wps_m3_oom test case.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
work->type was pointing to the allocated work->ctx buffer and the debug
print in radio_work_free() ended up using freed memory if a started
external radio work was removed as part of FLUSH command operations. Fix
this by updating work->type to point to a constant string in case the
dynamic version gets freed.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Add a new parameter wps_disabled to network block (wpa_ssid). This
parameter allows WPS functionality to be disabled in AP mode.
Signed-off-by: Lior David <qca_liord@qca.qualcomm.com>
While this is unlikely to make any practical difference, it is better to
keep consistent with hostapd configuration parser.
Signed-off-by: Jouni Malinen <j@w1.fi>
If the network profile key_mgmt parameter was not set, wpa_supplicant
defaulted to enabling both WPA-PSK and WPA-EAP. This is not correct for
AP mode operations, so remove WPA-EAP in such a case to fix WPA-PSK
without explicit key_mgmt parameter.
Signed-off-by: Jouni Malinen <j@w1.fi>
wpa_supplicant was starting P2P group processing for all AP mode
interfaces in CONFIG_P2P=y builds. This is unnecessary and such
operations should be enabled only for actual GO interfaces.
Signed-off-by: Jouni Malinen <j@w1.fi>
Commit afb2e8b891 ('tests: Store P2P
Device ifname in class WpaSupplicant') did not take into account the
possibility of capa.flags not existing in get_driver_status() and broke
WEXT test cases. Fix this by checking that capa.flags is present before
looking at its value.
Signed-off-by: Jouni Malinen <j@w1.fi>
