The previous mechanism of unloaded the providers from tls_deinit() did
not work correctly for some cases. In particular, it was possible for
hostapd to end up unloading both providers and not being able to recover
from this if TLS server was not enabled.
Address this more cleanly by introducing a new crypto_unload() function
that will be called when the process is exiting.
Fixes: 097ca6bf0b ("OpenSSL: Unload providers on deinit")
Signed-off-by: Jouni Malinen <j@w1.fi>
The DH file parameters are applicable only for the TLS server, so this
parameter did not really have any impact to functionality. Remove it to
get rid of useless code and confusing documentation for the network
block configuration.
Signed-off-by: Jouni Malinen <j@w1.fi>
Commit 0b5e98557e ("FIPS: Use OpenSSL CMAC implementation instead of
aes-omac1.c") added this implementation initially only for the FIPS
builds. However, there does not seem to be any remaining need to avoid
depending on the OpenSSL library implementation for builds, so move to
that implementation unconditionally to reduce the binary size a bit.
Signed-off-by: Jouni Malinen <j@w1.fi>
When receiving a beacon report request with the mode set to active,
channels that are marked as NO_IR were not added to the scan request.
However, active mode just mean that active scan is allowed, but not
that it is a must, so these channels should not be omitted.
Include channels that are marked as NO_IR in the scan request even
if the mode is set to active.
Signed-off-by: Avraham Stern <avraham.stern@intel.com>
The id and cred_id variables are reset to 0 every time the
wpa_config_read() function is called, which is fine as long as it is
only called once. However, this is not the case when using both the -c
and -I options to specify two config files.
This is a problem because the GUI, since commit eadfeb0e93 ("wpa_gui:
Show entire list of networks"), relies on the network IDs being unique
(and increasing), and might get into an infinite loop otherwise.
This is solved by simply making the variables static.
Signed-off-by: Naïm Favier <n@monade.li>
force_kdk_derivation is defined within CONFIG_TESTING_OPTIONS, so need
to use matching condition when accessing it.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
On the 6 GHz band the following is not allowed (see IEEE Std
802.11ax-2021, 12.12.2), so do not allow association with an AP using
these configurations:
- WEP/TKIP pairwise or group ciphers
- WPA PSK AKMs
- SAE AKM without H2E
In addition, do not allow association if the AP does not advertise a
matching RSNE or does not declare that it is MFP capable.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
So far, this option was only present in the Makefiles. Document it as
being available for configuration since the WFA program has already been
launched.
Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
The default behavior in wpa_supplicant is to disable use of TLSv1.3 in
EAP-TLS unless explicitly enabled in network configuration. The new
CONFIG_EAP_TLSV1_3=y build parameter can be used to change this to
enable TLSv1.3 by default (if supported by the TLS library).
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
It would have been possible for the authentication attemps to go into a
loop if the AP/Authenticator/authentication server were to believe EAP
authentication succeeded when the local conclusion in Supplicant was
failure. Avoid this by timing out authentication immediately on the
second consecutive EAP authentication failure.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
To support the STA testbed role, the STA has to disable transmitting
EAPOL-Key group msg 2/2 of Group Key Handshake. Add test parameter to
disable sending EAPOL-Key group msg 2/2 of Group Key Handshake.
Signed-off-by: Veerendranath Jakkam <quic_vjakkam@quicinc.com>
STA needs to check AP's information after receive reassociation
response. STA uses connected AP's Beacon/Probe Response frame to compare
with Reassociation Response frame of the target AP currently. However,
if one AP supports OCV and the other AP doesn't support OCV, STA will
fail to verify RSN capability, then disconnect. Update current_bss to
the target AP before check, so that STA can compare correct AP's RSN
information in Reassociation Response frame.
Signed-off-by: Xin Deng <quic_deng@quicinc.com>
Add AP mode commands for ACL and BTM into wpa_cli similarly to the way
these were already available in hostapd_cli.
Signed-off-by: Chaoli Zhou <quic_zchaoli@quicinc.com>
Support installing the updated allow/deny list to the driver if it
supports ACL offload. Previously, only the not-offloaded cases were
updated dynamically.
Signed-off-by: Chaoli Zhou <quic_zchaoli@quicinc.com>
Extend AP mode ACL control interface commands to work from
wpa_supplicant in addition to the previously supported hostapd case.
Signed-off-by: Chaoli Zhou <quic_zchaoli@quicinc.com>
Extend AP mode BTM control interface commands to work from
wpa_supplicant in additiona to the previously support hostapd case.
Signed-off-by: Chaoli Zhou <quic_zchaoli@quicinc.com>
Enable BTM capability for AP mode only and do not affect P2P GO mode.
This can be used for AP band steering when using wpa_supplicant to
control AP mode operations.
Signed-off-by: Chaoli Zhou <quic_zchaoli@quicinc.com>
dpp_configurator_params changes were taken into use in the
non-TCP/Controller case immediately on change, but that was not the case
for the Controller where this was updated only when explicitly starting
it. Change this to update dpp_configurator_params for the Controller as
well even if it is already running.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Since commit 200c7693c9 ('Make WEP functionality an optional build
parameter'), WEP support is optional and, indeed, off by default.
The distributions are now catching up and disabling WEP in their builds.
Unfortunately, there's no indication prior to an attempt to connect to a
WEP network that it's not going to work. Add a capability to communicate
that.
Unlike other capabilities, this one is negative. That is, it indicates
lack of a WEP support as opposed to its presence. This is necessary
because historically there has been no capability to indicate presence
of WEP support and therefore NetworkManager (and probably others) just
assumes it's there.
Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
Acked-by: Davide Caratti <davide.caratti@gmail.com>
CurrentAuthMode should be set as a real auth type when authentication is
in progress. wpa_supplicant has a property "State" which indicates the
authentication stage already. I think setting auth mode as "INACTIVE" in
all auth progress stages is not a good idea, because sometimes we need
to handle this connection according to the auth type even when
authentication is not complete. For example, NetworkManager may recall
ask-password-dialog when auth mode is "wpa-psk" and "sae", try next
access point when auth mode is "EAP-xx" when password is incorrect.
Since "CurrentAuthMode" is set as "INACTIVE" in all not fully completed
situations, we do not know how to handle it.
Signed-off-by: Aris Aachen <chenyunxiong@unionitech.com>
Signed-off-by: ArisAachen <chenyunxiong@uniontech.com>
This is mainly for testing purposes to allow a Configurator to the curve
between provisioning cases. This would not work for real deployement
cases unless every Enrollee were reconfigured.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Allow the Configurator to be configured to use a specific curve for the
netAccessKey so that it can request the Enrollee to generate a new key
during the configuration exchange to allow a compatible Connector to be
generated when the network uses a different curve than the protocol keys
used during the authentication exchange.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Allow PKEX v1-only or v2-only behavior to be specific for the Responder
role. This is mainly for testing purposes.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Commit a41a29192e ("build: Pull common fragments into a build.rules
file") introduced a regression into wpa_supplicant build process. The
build target libwpa_client.so is not built regardless of whether the
option CONFIG_BUILD_WPA_CLIENT_SO is set or not. This happens because
this config option is used before it is imported from the configuration
file. Moving its use after including build.rules does not help: the
variable ALL is processed by build.rules and further changes are not
applied. Similarly, option CONFIG_NO_WPA_PASSPHRASE also does not work
as expected: wpa_passphrase is always built regardless of whether the
option is set or not.
Re-enable these options by adding both build targets to _all
dependencies.
Fixes: a41a29192e ("build: Pull common fragments into a build.rules file")
Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Couple of the WPS/P2P/RADIUS-PSK cases were freeing heap memory
allocations without explicitly clearing the PSK value. Add such clearing
for these to avoid leaving the PSK in memory after it is not needed
anymore.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This is needed to avoid leaving wpa_state to SCANNING if network
introduction fails and a new association is not started.
This was found with the following test case sequence:
dpp_conn_status_connector_mismatch scan_trigger_failure
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
A previously set netrole (e.g., from DPP_LISTEN or DPP_AUTH_INIT) could
have been used in a following DPP_CHIRP or DPP_RECONFIG operation. This
could result in trying to request incorrect configuration and likely
rejection from the Configurator. Fix this by clearing the netrole when
starting these operations.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
It was possible for at least the wpa_s->last_ssid to be left pointing to
a removed network which could result in processing the following
association as a reassociation-within-an-ESS even when it was moving to
a different ESS. This could result in unexpected behavior. This was
found with the following test case sequence:
sigma_dut_ap_psk_sae_ft sae_h2e_password_id ap_wps_pk_oom sigma_dut_client_privacy
Move clearing of wpa_s->last_ssid and wpa_s->current_ssid into
wpas_notify_network_removed() to catch all cases similarily to the way
wpa_s->next_ssid was already cleared.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
The offchannel frame TX wait was stopped whenever processing
DPP_STOP_LISTEN in most cases. However, there was a corner case on the
Responder side where this operation was skipped after PKEX was completed
successful and the Authentication Request frame had not yet been
received from the Initiator.
While this does not normally cause any significant issue, this could
result in unexpected behavior especially in test cases that run multiple
DPP PKEX operations in a row since the start of a new TX operation might
get delayed while waiting for the previous TX-wait to complete.
This was found with the following test case sequence:
dpp_reconfig_retries dpp_pkex_alloc_fail
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Allow macsec_csindex to be configured and select the cipher suite when
the participant acts as a key server.
Signed-off-by: leiwei <quic_leiwei@quicinc.com>
There is no need to maintain the DPP authentication state if config
request processing fails, so clear state also in the GAS server request
handler similarly to the other failure cases.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
The hwsim test cases are trying to clear this parameter between test
cases, but that was not really done correctly for many of the sigma_dut
test cases. Instead of fixing the text scripts to do this more
carefully, it seems to be simpler to just force the FLUSH command to
clear this.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This is needed since the gas_query_req() operation could remain waiting
indefinitely for the response if the Configurator keeps sending out
comeback responses with additional delay. The DPP technical
specification expects the Enrollee to continue sending out new Config
Requests for 60 seconds, so this gives an extra 60 second time after the
last expected new Config Request for the Configurator to determine what
kind of configuration to provide.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Instead of depending on the TX-wait-response-time to be sufficient to
cover the full GAS exchange, start an ongoing listen operation on the
negotiation channel (if no such listen operation is already in place) to
allow the configuration exchange to take longer amount of time. This is
needed for cases where the conf=query is used to request Configurator
parameters from upper layers and that upper layer processing (e.g., user
interaction) takes significant amount of time.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Once the GO/AP interface initialization is completed, check if the
operating frequency set in the wpa_supplicant group interface structure
is different than the one set in the hostapd interface structure
associated with the group interface. If yes, update the frequency in the
wpa_supplicant group interface and network configuration to the
frequency set in the hostapd interface structure.
The frequency set in the hostapd interface is the correct/final
frequency wpa_supplicant configured in the kernel/driver. This is done
because wpa_supplicant may switch the initially requested primary and
secondary frequencies to get a secondary frequency with no beacons (to
avoid interference or 20/40 MHz coex logic). And the updated frequency
is informed by the driver only after the interface setup is completed
through the channel switch event - EVENT_CH_SWITCH. But wpa_supplicant
updates the frequency to applications through the P2P_GROUP_STARTED
event which is triggered before the EVENT_CH_SWITCH event. To send the
correct frequency to applications the frequency must be updated before
sending the P2P_GROUP_STARTED event.
Bug: 191272346
Test: Manual - Verified that GO frequency is updated and reported
correctly to Nearby application.
Signed-off-by: Sunil Ravi <sunilravi@google.com>
This provides an alternative mechanism for upper layer components to
control configuration parameters to be used by the local Configurator.
Instead of the previously used design where the Configurator parameters
had to be provided before initiating the DPP Authentication exchange,
the new alternative approach allows the DPP Authentication exchange to
be started before any Configurator parameters have been determined and
wpa_supplicant will then request the parameters once the DPP
Configuration Request has been received from the Enrollee. This allows
the Config Request information to be used at upper layers to determine
how the Enrollee should be configured.
For example for an Initiator:
CTRL: DPP_QR_CODE <URI from Responder/Enrollee>
CTRL: DPP_AUTH_INIT peer=1 conf=query
<3>DPP-CONF-NEEDED peer=1 src=02:00:00:00:00:00 net_role=sta name="Test" opclass=81,82,83,84,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130 mud_url=N/A
(upper layer processing; potentially including user interaction)
CTRL: DPP_CONF_SET peer=1 conf=sta-sae ssid=736165 pass=70617373776f7264
<3>DPP-CONF-SENT
For example for a Responder:
CTRL: SET dpp_configurator_params conf=query
CTRL: DPP_LISTEN 2412 role=configurator
<3>DPP-CONF-NEEDED peer=2 src=02:00:00:00:01:00 net_role=sta name="Test" opclass=81,82,83,84,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130 mud_url=N/A
(upper layer processing; potentially including user interaction)
CTRL: DPP_CONF_SET peer=2 conf=sta-sae ssid=736165 pass=70617373776f7264
<3>DPP-CONF-SENT
For example for an Initiator that can act both as a Configurator and an
Enrollee in a case where the Initiator becomes the Enrollee:
CTRL: DPP_AUTH_INIT peer=1 role=either conf=query
<3>DPP-CONF-RECEIVED
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Extend the GAS server functionality to allow a request handler to return
the initial comeback delay with a later callback instead of having to
indicate the comeback delay when returning from the handler function.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
The TCP code path did not handle the postponed connection attempt on TX
status and the following result message from the Enrollee to the
Configurator. Fix this by adding TCP-versions of these operations to
match the way wpa_supplicant implemented this for the Public Action
frames.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This extends dpp_test functionality to allow DPP Network Introduction
exchanges to use an incorrect value in the Protocol Version attribute.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Use a separate ver=<1|2> parameter to DPP_PKEX_ADD instead of
overloading init=1 with version indication. This allows additional
options for forcing v1-only and v2-only in addition to automatic mode
(start with v2 and fall back to v1, if needed).
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Use automatic PKEX version negotiation as the initiator by starting with
PKEXv2 and if no response is received, trying again with PKEXv1. For
now, this is enabled only in wpa_supplicant CONFIG_DPP3=y builds.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
The rx_freq of Public Action frame was not maintained by the GO and the
GO always sent the response on the operating channel. This causes
provision discovery failure when a P2P Device is sending a PD Request on
a 2.4 GHz social channel and the GO is responding on a 5 GHz operating
channel.
Save the rx_freq and use it for GO to sent the response. This extends
commit c5cc7a59ac ("Report offchannel RX frame frequency to hostapd")
to cover additional frame types.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
