This avoids race conditions with UML time-travel where some cases ended
up not reporting the DPP-TX-STATUS event and that resulted in discarding
events that are more important to verify for correct behavior.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
One of the sigma_dut testing cases missed the t.join() call to make
surte the separate thread terminated. This could result in confusing
"unexpected stdout output" in a middle of an unrelated test case.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Stop and restart P2P_FIND on dev[1] to avoid timing issues where the
wait on dev[1] makes the test case fail even though this was trying to
verify that dev[0] is reacting quickly enough.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Wait a bit between the DPP-SCAN-PEER-QR-CODE event and reporting of the
QR Code to avoid a race condition.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Avoid a race condition that could result in the relay (hostapd)
maintaining STA association with PMF and discarding of the GAS frames.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Terminate the group explicitly before cleanup to avoid a race condition
between an automatically started scan to find the GO again and the scan
started to clear the scan cache.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Wait a bit to allow stopping of the DPP_LISTEN operation to be completed
before issuing the DPP_QR_CODE command that would send the frame that
these test cases expect to not get ACKed.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Wait some time between the first DPP Authentication Response that
indicates the response is not yet available and the second DPP
Authentication Response to make this sequence a bit more realistic and
less likely to hit race conditions with UML time-travel.
Signed-off-by: Jouni Malinen <j@w1.fi>
Wait for hostapd to complete STA authorization before running the
connectivity test between two associated STAs.
Signed-off-by: Jouni Malinen <j@w1.fi>
It can take significant amount of time to find the peer in this type of
a case where an AP connection is forced on a different channel while
going through p2p_find on both devices.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
DPP initiator will try three channels in this sequence and it can take
very close to the previously used five second timeout before being able
to try on the actual operating channel of the AP. This could result in
the test case failing unnecessarily. Increase the timeout to avoid this.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Avoid potential race condition in ap_cipher_tkip_countermeasures_ap by
sending the first test frame only after the AP has completed processing
the 4-way handshake and also wait a bit to allow the group key handshake
to be completed.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
If a tests hits a WARN_ON_ONCE then the retriggering will make it appear
to pass, since the warning will not happen again. Make this more
reliable by resetting the states at the beginning of each test.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
The kernel has an asynchronous work to enable TX, which hasn't always
run by the time we get to TX tests. Do a sysfs read from the carrier
file before TX, on newer kernels this synchronises the needed state.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
It looks like the lifetime_in_memory test cases can hit a read failure
(Errno 5 - Input/output error) every now and then, so skip memory areas
that report that, but go through all readable process memory.
Signed-off-by: Jouni Malinen <j@w1.fi>
Wait for AP/GO to complete processing before taking the next step in a
test instead of waiting just for STA. This avoids race conditions with
UML time-travel.
Signed-off-by: Jouni Malinen <j@w1.fi>
The remaining lifetime of the PMKSA entry is decreasing, so need to
allow it to differ between the old and new entries.
Signed-off-by: Jouni Malinen <j@w1.fi>
Avoid race condition with TX status reporting for the broadcast
Deauthentication frame. This could be delivered after the SET command
for ext_mgmt_frame_handled=1 and as such, could result in an
MGMT-TX-STATUS event that would confuse the next step in the test.
Signed-off-by: Jouni Malinen <j@w1.fi>
Wait for the externally triggered scan to actually start before issuing
the own scan command to avoid a race condition with UML time-travel.
Signed-off-by: Jouni Malinen <j@w1.fi>
Do not leave dev[0] in state where it is waiting for
NL80211_CMD_ASSOCIATE to complete since that might deliver an
ASSOC_TIMED_OUT event to the next test case.
Signed-off-by: Jouni Malinen <j@w1.fi>
This test case has a race condition between the first and second
initiated DPP exchange. New kernel scheduling with UML time-travel is
making the first exchange complete before the first one can be started.
Making this fully robust would likely need ext_mgmt_frame_handling=1 and
more complex test script, but as an initial workaround, split URI
parsing from dev[2] to happen before dev[1] is started so that only a
single DPP_AUTH_INIT command is needed during the race window.
In addition, detect the race condition and SKIP the test instead of
reporting FAIL.
Signed-off-by: Jouni Malinen <j@w1.fi>
Wait for hostapd to complete processing before taking the next step in a
test instead of waiting just for wpa_supplicant. This avoids race
conditions with UML time-travel.
Signed-off-by: Jouni Malinen <j@w1.fi>
Even though the radio work has started here, kernel might not have
started ROC (i.e., the REMAIN_ON_CHANNEL event has not yet been
received). Wait a bit to increase likelihood of everything being ready
for the following DPP operation.
Signed-off-by: Jouni Malinen <j@w1.fi>
At least with UML time-travel, it is possible for the response frame to
be received before the TX status of the request is delivered. Such a
sequence resulted in run_req_beacon() dropping the response event and
the following step failing due to a missing event. Avoid this by caching
the received response in run_req_beacon() and returning them to the
caller.
Signed-off-by: Jouni Malinen <j@w1.fi>
Wait some time before requesting disconnection to allow hostapd to
complete 4-way handshake processing. Wait some time after disconnection
has been completed on the STA before trying to use SAE again with the AP
so that hostapd has a chance to complete disconnection with UML
time-travel.
Signed-off-by: Jouni Malinen <j@w1.fi>
The underlying driver can reject only specific links in an association
request. In that case we will only ignore the corresponding BSS. Add a
test for this flow by simulating an association failure on the second
link specifically.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Test ML association, automatic link discovery using ML probe requests
and marking BSSs as failed after an ML association failure, if the
corresponding link also rejected the association.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Also check that we can discover BSSs requests using ML probe requests.
Note that this requires a new enough mac80211 which parses the ML probe
request and reports the contained BSSs.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Going through everything to test parsing of the basic ML element and the
RNR element is not really feasible for MBSSID mode. As such, add a unit
test to excercise parsing the available links from the RNR, basic ML
element and MBSSID-Index element if present.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Having the ability to trigger multiple failures in one test can be
useful. Add support to the test infrastructure to do this.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
These were either sending the command to the wrong
hostapd/wpa_supplicant instance or using the wrong command. This
currently causes the wait to just immediately stop, but with future
commits it would start failing.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Future commits change the related code and exception string. Avoiding
using the context is easy here and actually avoids two layers of
nesting.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
IEEE P802.11be/D4.0, 9.4.2.312.2.3 states that the AP MLD ID should only
be included in some ML probe responses. Beacon frames shouldn't include
AP MLD ID.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
With more channels now available in some regulatory domains, some scan
operations can take longer. Give them more time to complete in tests.
For two cases this required adding a timeout parameter to connect(),
which is then passed through.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Test functionality in cases where MBSSID is used with a WPA2-Personal
only BSS as the transmitted BSS and WPA3-Personal BSS as the
nontransmitted BSS.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Otherwise the station or AP might be doing router advertisements,
causing keep alive already without the mechnism that is being tested.
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
dpp_config_legacy_gen_two_conf_psk and dpp_config_legacy_gen_two_conf
tried to set a DPP parameter before having verified that CONFIG_DPP was
used in the build.
Signed-off-by: Jouni Malinen <j@w1.fi>
Starting a thread to initiate DPP before starting the responder through
sigma_dut can result in unexpected testing behavior since there may not
be enough time to get the responder enabled before timing out som
initiator actions. Wait a second at the beginning of the initiator
thread in dpp_init_conf() similarly to how this was handled in other
initiator-from-thread cases.
Signed-off-by: Jouni Malinen <j@w1.fi>
Wait for stdout/stderr in a more robust manner to avoid blocking the
pipes and kill the sigma_dut process if it fails to terminate cleanly.
Signed-off-by: Jouni Malinen <j@w1.fi>
Responder receives Authentication Request and Config Request in a
sequence and it is possible for the Config Request to be received before
MGMT_RX_PROCESS has been processed for Authentication Request in the
cases where the test script is in the middle of RX processing. This can
result in DPP-AUTH-SUCCESS being delivered only after the MGMT-RX event
for Config Reques which means that wait_auth_success() would lose that
MGMT-RX event.
Avoid this issue by caching the "extra" MGMT-RX event within
wait_auth_success() and having the caller verify if the Config Request
(GAS Initial Request) has already been received before waiting to
receive it.
This makes dpp_gas, dpp_gas_comeback_after_failure, and
dpp_gas_timeout_handling more robust.
Signed-off-by: Jouni Malinen <j@w1.fi>
UML time travel allows the deauthentication event to be processed more
quickly than the delivery of EAP-Success to the client through the test
script, so accept either sequence here.
Signed-off-by: Jouni Malinen <j@w1.fi>
connect_network() tried to make test log more readable with a
dump_monitor() call at the end of the function. However, this could end
up practically dropping an event that arrives more or less immediately
after CTRL-EVENT-CONNECTED. This could happen with UML time travel,
e.g., in suite_b_192_pmksa_caching_roam.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Wait for hostapd connection event before issue HS20_WNM_NOTIF to avoid a
race condition with UML time travel.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This event may be sent before CTRL-EVENT-CONNECTED, so modify the test
cases to wait directly for TRANSITION-DISABLE by skipping the separate
wait for CTRL-EVENT-CONNECTED.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
It is possible for the sigma_dut process to be scheduled in a manner
that ends up combining the status,RUNNING and status,COMPLETE lines into
a single TCP message. This was supposed to be handled in the
sigma_dut_cmd() implementations, but that design had been broken by code
refactoring that changed the indentation level incorrectly.
Fixes: d68946d510 ("tests: sigma_dut and DPP push button first on Enrollee")
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Use more specific condition for the allocation failure to allow
wpa_supplicant_trigger_scan() implementation to be modified without
making this test case fail.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Wait for allocation failure using wait_fail_trigger() instead of waiting
for a scan failure event since that failure event will go away with
implementation change.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Add SKB tracing (which shows now why/where a frame was dropped
in the stack), and also -T for stack trace at each event.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Give the AP some time to set up stations fully (in the
kernel) so that traffic forwarding will work.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
We can't do ANQP when the STA is connected but the AP hasn't fully set
up the STA yet, so wait on the AP side before continuing.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
We need to wait for the MGMT-RX event before disabling
ext_mgmt_frame_handling again, otherwise we might be disabling it and
hostapd only receives the deauth frame after we already disable it,
defeating the purpose of the test.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
With PMF, we cannot do even deauth unless we wait for the STA to have
fully connected on the AP side, the STA thinking it has isn't sufficient
since it immediately says so after M4. Add wait_sta() before disconnect,
and also before SA_QUERY.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Before querying the PMKSA cache, wait for the STA to have appeared on
the AP side, otherwise scheduling differences may have us asking when
the STA thinks it's connected but the AP hasn't fully processed that.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
All processes need to have a bit of time to mark the kernel STAs
authorized, otherwise traffic may fail. Give them some time, and also
use check_connectivity() in connectivity() since it's the same check,
just different arguments.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Due to scheduling changes, we don't always now succeed to reconnect in
exactly 1 second, it might take 1.01. Give it 1.1 for a bit more leeway,
it's not clear why it should be exactly 1 second anyway.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
