The external authentication command and event does not need to copy the
BSSID/SSID values into struct external_auth since those values are used
before returning from the call. Simplify this by using const u8 * to
external data instead of the array with a copy of the external data.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Extend commit 5ff39c1380 ("SAE: Support external authentication
offload for driver-SME cases") to support external authentication
with drivers that implement AP SME by notifying the status of
SAE authentication to the driver after SAE handshake as the
driver acts as a pass through for the SAE Authentication frames.
Signed-off-by: Srinivas Dasari <dasaris@codeaurora.org>
This extends driver interface to nl80211 by introducing the following
changes,
1. Register for Authenication frames in driver-based AP SME mode.
2. Advertise NL80211_ATTR_EXTERNAL_AUTH_SUPPORT in set_ap when
offloaded SAE authentication is supported.
3. Extend the NL80211_CMD_EXTERNAL_AUTH interface to also send PMKID
so that the drivers can respond to the PMKSA cached connection
attempts from the stations avoiding the need to contact user space
for all PMKID-based connections.
4. Send external auth status to driver only if it is a driver based
SME solution.
Signed-off-by: Srinivas Dasari <dasaris@codeaurora.org>
beacon_set_done did not get reset to zero on disabling interface using
DISABLE control interface command and the subsequent ENABLE command will
caused configuration of Beacon/Probe Response/Association Response frame
IEs twice. The unnecessary two step configuration can be avoided by
resetting beacon_set_done on DISABLE so that ENABLE can bring up the
interface in a single step with fully updated IEs.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Binary presentations of element and scalar can be written directly to
the allocated commit message buffer instead of having to first write
them into temporary buffers just to copy them to the actual message
buffer.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
RFC 5931 has these conditions as MUST requirements, so better follow
them explicitly even if the rand,mask == 0 or rand+mask == 0 or 1 cases
are very unlikely to occur in practice while generating random values
locally.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This adds an explicit check for 0 < x,y < prime based on RFC 5931,
2.8.5.2.2 requirement. The earlier checks might have covered this
implicitly, but it is safer to avoid any dependency on implicit checks
and specific crypto library behavior. (CVE-2019-9498 and CVE-2019-9499)
Furthermore, this moves the EAP-pwd element and scalar parsing and
validation steps into shared helper functions so that there is no need
to maintain two separate copies of this common functionality between the
server and peer implementations.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
When processing an EAP-pwd Commit frame, the server's scalar and element
(elliptic curve point) were not validated. This allowed an adversary to
bypass authentication, and act as a rogue Access Point (AP) if the
crypto implementation did not verify the validity of the EC point.
Fix this vulnerability by assuring the received scalar lies within the
valid range, and by checking that the received element is not the point
at infinity and lies on the elliptic curve being used. (CVE-2019-9499)
The vulnerability is only exploitable if OpenSSL version 1.0.2 or lower
is used, or if LibreSSL or wolfssl is used. Newer versions of OpenSSL
(and also BoringSSL) implicitly validate the elliptic curve point in
EC_POINT_set_affine_coordinates_GFp(), preventing the attack.
Signed-off-by: Mathy Vanhoef <mathy.vanhoef@nyu.edu>
When processing an EAP-pwd Commit frame, verify that the peer's scalar
and elliptic curve element differ from the one sent by the server. This
prevents reflection attacks where the adversary reflects the scalar and
element sent by the server. (CVE-2019-9497)
The vulnerability allows an adversary to complete the EAP-pwd handshake
as any user. However, the adversary does not learn the negotiated
session key, meaning the subsequent 4-way handshake would fail. As a
result, this cannot be abused to bypass authentication unless EAP-pwd is
used in non-WLAN cases without any following key exchange that would
require the attacker to learn the MSK.
Signed-off-by: Mathy Vanhoef <mathy.vanhoef@nyu.edu>
When processing an EAP-pwd Commit frame, the peer's scalar and element
(elliptic curve point) were not validated. This allowed an adversary to
bypass authentication, and impersonate any user if the crypto
implementation did not verify the validity of the EC point.
Fix this vulnerability by assuring the received scalar lies within the
valid range, and by checking that the received element is not the point
at infinity and lies on the elliptic curve being used. (CVE-2019-9498)
The vulnerability is only exploitable if OpenSSL version 1.0.2 or lower
is used, or if LibreSSL or wolfssl is used. Newer versions of OpenSSL
(and also BoringSSL) implicitly validate the elliptic curve point in
EC_POINT_set_affine_coordinates_GFp(), preventing the attack.
Signed-off-by: Mathy Vanhoef <mathy.vanhoef@nyu.edu>
Explicitly verify that own and peer commit scalar/element are available
when trying to check SAE confirm message. It could have been possible to
hit a NULL pointer dereference if the peer element could not have been
parsed. (CVE-2019-9496)
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Try to avoid showing externally visible timing or memory access
differences regardless of whether the derived pwd-value is smaller than
the group prime.
This is related to CVE-2019-9494.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This is an initial step towards making the FFC case use strictly
constant time operations similarly to the ECC case.
sae_test_pwd_seed_ffc() does not yet have constant time behavior,
though.
This is related to CVE-2019-9494.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
These groups have significant probability of coming up with pwd-value
that is equal or greater than the prime and as such, need for going
through the PWE derivation loop multiple times. This can result in
sufficient timing different to allow an external observer to determine
how many rounds are needed and that can leak information about the used
password.
Force at least 40 loop rounds for these MODP groups similarly to the ECC
group design to mask timing. This behavior is not described in IEEE Std
802.11-2016 for SAE, but it does not result in different values (i.e.,
only different timing), so such implementation specific countermeasures
can be done without breaking interoperability with other implementation.
Note: These MODP groups 22, 23, and 24 are not considered sufficiently
strong to be used with SAE (or more or less anything else). As such,
they should never be enabled in runtime configuration for any production
use cases. These changes to introduce additional protection to mask
timing is only for completeness of implementation and not an indication
that these groups should be used.
This is related to CVE-2019-9494.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Make the non-failure path in the function proceed without branches based
on r_odd and in constant time to minimize risk of observable differences
in timing or cache use. (CVE-2019-9494)
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The QR test result can provide information about the password to an
attacker, so try to minimize differences in how the
sae_test_pwd_seed_ecc() result is used. (CVE-2019-9494)
Use heap memory for the dummy password to allow the same password length
to be used even with long passwords.
Use constant time selection functions to track the real vs. dummy
variables so that the exact same operations can be performed for both QR
test results.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This algorithm could leak information to external observers in form of
timing differences or memory access patterns (cache use). While the
previous implementation had protection against the most visible timing
differences (looping 40 rounds and masking the legendre operation), it
did not protect against memory access patterns between the two possible
code paths in the masking operations. That might be sufficient to allow
an unprivileged process running on the same device to be able to
determine which path is being executed through a cache attack and based
on that, determine information about the used password.
Convert the PWE finding loop to use constant time functions and
identical memory access path without different branches for the QR/QNR
cases to minimize possible side-channel information similarly to the
changes done for SAE authentication. (CVE-2019-9495)
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Get rid of the branches that depend on the result of the Legendre
operation. This is needed to avoid leaking information about different
temporary results in blinding mechanisms.
This is related to CVE-2019-9494 and CVE-2019-9495.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
These functions can be used to help implement constant time operations
for various cryptographic operations that must minimize externally
observable differences in processing (both in timing and also in
internal cache use, etc.).
This is related to CVE-2019-9494 and CVE-2019-9495.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This helps in reducing measurable timing differences in operations
involving private information. BoringSSL has removed BN_FLG_CONSTTIME
and expects specific constant time functions to be called instead, so a
bit different approach is needed depending on which library is used.
The main operation that needs protection against side channel attacks is
BN_mod_exp() that depends on private keys (the public key validation
step in crypto_dh_derive_secret() is an exception that can use the
faster version since it does not depend on private keys).
crypto_bignum_div() is currently used only in SAE FFC case with not
safe-prime groups and only with values that do not depend on private
keys, so it is not critical to protect it.
crypto_bignum_inverse() is currently used only in SAE FFC PWE
derivation. The additional protection here is targeting only OpenSSL.
BoringSSL may need conversion to using BN_mod_inverse_blinded().
This is related to CVE-2019-9494 and CVE-2019-9495.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
These wpa_supplicant network profile parameters could be used to specify
a single match string that would be used against the dNSName items in
subjectAltName or CN. There may be use cases where more than one
alternative match string would be useful, so extend these to allow a
semicolon delimited list of values to be used (e.g.,
"example.org;example.com"). If any of the specified values matches any
of the dNSName/CN values in the server certificate, consider the
certificate as meeting this requirement.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Incorrect gen->type value was used to check whether subjectAltName
contained dNSName entries. This resulted in all domain_match and
domain_suffix_match entries failing to find a match and rejecting the
server certificate. Fix this by checking against the correct type
definition for dNSName.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Previously the EAP user database had to include a wildcard entry for ERP
to work since the keyName-NAI as User-Name in Access-Request would not
be recognized without such wildcard entry (that could point to any EAP
method). This is not ideal, so add a separate check to allow any stored
ERP keyName-NAI to be used for ERP without any requirement for the EAP
user database to contain a matching entry.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Inclusion of common/dpp.h into hostapd/main.c brought in an undesired
unconditional dependency on OpenSSL header files even for builds where
DPP is not enabled. Fix this by making the dpp.h contents, and in
particular the inclusion of openssl/x509.h, conditional on CONFIG_DPP.
Fixes: 87d8435cf9 ("DPP: Common configurator/bootstrapping data management")
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The rules defining which DH groups are suitable for SAE use were
accepted into IEEE 802.11 REVmd based on this document:
https://mentor.ieee.org/802.11/dcn/19/11-19-0387-02-000m-addressing-some-sae-comments.docx
Enforce those rules in production builds of wpa_supplicant and hostapd.
CONFIG_TESTING_OPTIONS=y builds can still be used to select any o the
implemented groups to maintain testing coverage.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
identity_buf may be NULL here. Handle this case explicitly by printing
"N/A" instead relying on snprintf converting this to "(null)" or some
other value based on unexpected NULL pointer.
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
While commit 1c156e783d ("Fixed tls_prf() to handle keys with
odd length") added support for keys with odd length, the function
never reached this code as the function would return earlier in
case the key length was odd. Fix this by removing the first check
for the key length.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
This commit introduces an attribute
QCA_WLAN_VENDOR_ATTR_ROAM_AUTH_REASON to carry the roam reason code
through QCA_NL80211_VENDOR_SUBCMD_KEY_MGMT_ROAM_AUTH event.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Avoid duplicated code in each user of dpp_build_conf_req() by moving the
common encapsulation case into this helper function.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
When connected using FT-SAE key mgmt, use PMK from PMKSA cache as XXKey
for PMK-R0 and PMK-R1 derivations. This fixes an issue where FT key
hierarchy could not be established due to missing (not yet configured)
XXKey when using SAE PMKSA caching for the initial mobility domain
association.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
In the case of the driver not supporting full AP mode STA state (i.e.,
not adding a STA entry before association), the QoS parameters are not
allowed to be modified when going through (re)association exchange for a
STA entry that has not been removed from the kernel. cfg80211 would
reject such command to update STA flags, so do not add the WMM parameter
in this case.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Ethernet frames have minimum length of 64 octets and shorter frames may
end up getting arbitrary padding in the end. This would result in the
FT/RRB receiver rejecting the frame as an incorrectly protected one.
Work around this by padding the message so that it is never shorter than
the minimum Ethernet frame.
Unfortunately, this padding is apparently not enough with all Ethernet
devices and it is still possible to see extra two octet padding at the
end of the message even if larger frames are used (e.g., showed up with
128 byte frames). For now, work around this by trying to do AES-SIV
decryption with two octets shorter frame (ignore last two octets) if the
first attempt fails.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Merge the practically copy-pasted implementations in wpa_supplicant and
hostapd into a single shared implementation in dpp.c for managing
configurator and boostrapping information. This avoid unnecessary code
duplication and provides a convenient location for adding new global DPP
data.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The memcpy calls added for exposing the PMK from wpa_auth module could
end up trying to copy the same memory buffer on top of itself.
Overlapping memory areas are not allowed with memcpy, so this could
result in undefined behavior. Fix this by making the copies conditional
on the updated value actually coming from somewhere else.
Fixes: b08c9ad0c7 ("AP: Expose PMK outside of wpa_auth module")
Signed-off-by: Jouni Malinen <j@w1.fi>
The offset update for copying KEK2 from the extended PTK was overriding
the offset instead of incrementing it (a likely copy-paste error from
the first offset assignment based on KCK). This resulted in KEK2 being
set to incorrect segment of PTK. Fix this by updating the offset
properly so that KEK2 is copied from the correct place at the end of the
PTK.
Fixes: 2f37387812 ("FILS: Add more complete support for FT-FILS use cases")
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
SSL_use_certificate_chain_file() is not available in the current
BoringSSL even though the defined OPENSSL_VERSION_NUMBER is large enough
to claim that this function would be present in the OpenSSL API.
Fall back to using SSL_use_certificate_file() with BoringSSL to fix the
build.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This code was after the FILS handling that would have encrypted the
frame. While FILS and OWE are never used together, the OWE handling
should really be before the FILS handling since no IEs can be added
after the FILS encryption step. In addition, the Diffie-Hellman
Parameter element is not a Vendor Specific element, so it should be
before some of the Vendor Specific elements even though it is not
defined in IEEE 802.11.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Use Diffie-Hellman key exchange to derivate additional material for
PMK-to-PTK derivation to get PFS. The Diffie-Hellman Parameter element
(defined in OWE RFC 8110) is used in association frames to exchange the
DH public keys. For backwards compatibility, ignore missing
request/response DH parameter and fall back to no PFS in such cases.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
DPP allows Diffie-Hellman exchange to be used for PFS in PTK derivation.
This requires an additional Z.x (x coordinate of the DH shared secret)
to be passed to wpa_pmk_to_ptk(). This commit adds that to the function
and updates all the callers to pass NULL,0 for that part in preparation
of the DPP specific changes to start using this.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The previous OWE implementation on the AP side rejected any
(Re)Association Request frame with the Diffie-Hellman Parameter element
if AKM was not OWE. This breaks compatibility with DPP PFS, so relax
that rule to allow DPP AKM to be used as well. While this commit alone
does not add support for PFS, this allows interoperability between
non-PFS implementation on the AP and a newer PFS implementation on the
STA.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Allow an additional context value to be passed to TLS exporter as
specified in RFC 5705 section 4.
This does not yet implement it for the internal TLS implementation.
However, as currently nothing uses context yet, this will not break
anything right now. WolfSSL maintainers also stated that they are not
going to add context support yet, but would look into it if/when this is
required by a published draft or a standard.
Signed-off-by: Ervin Oro <ervin.oro@aalto.fi>
This allows devices supporting DPP protocol version 2 or newer to
provision networks that enable both the legacy (PSK/SAE) and DPP
credentials.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Share a single parsing implementation for both hostapd and
wpa_supplicant to avoid code duplication. In addition, clean up the
implementation to be more easily extensible.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
X509_get_subject_name() in OpenSSL 1.0.2 does not mark its argument as a
const pointer, so need to type cast this to avoid a build warning.
Fixes: 841205a1ce ("OpenSSL: Add 'check_cert_subject' support for TLS server")
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The queue_len * 50 ms wait time was too large with the retransmission
timeouts used in the mesh case for SAE. The maximum wait of 750 ms was
enough to prevent successful completion of authentication after having
hit the maximum queue length. While the previous commit is enough to
allow this to complete successfully in couple of retries, it looks like
a smaller wait time should be used here even if it means potentially
using more CPU.
Drop the processing wait time to queue_len * 10 ms so that the maximum
wait time is 150 ms if the queue is full.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The previous design of simply queuing all SAE commit messages was not
exactly good at allowing recovery from a flooding state if the valid
peer used frequent retransmissions of the SAE message. This could
happen, e.g., with mesh BSSs using SAE. The frequent retransmissions and
restarts of SAE authentication combined with SAE confirm messages
bypassing the queue ended up in not being able to finish SAE exchange
successfully.
Fix this by modifying the queuing policy to queue SAE confirm messages
if there is a queued SAE commit message from the same peer so that the
messages within the same exchange do not get reordered. In addition,
replace queued SAE commit/confirm message if a new matching message is
received from the same peer STA. This is useful for the case where the
peer restarts SAE more quickly than the local end has time to process
the queued messages.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Use this new message from Enrollee to Configurator to indicate result of
the config object provisioning if both devices support protocol version
2 or newer.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Do not set sm->timer_tick_enabled if the eloop_register_timeout() call
fails so that the next attempt to enable the timer in
eapol_enable_timer_tick() can try to recover from unexpected eloop
failures. This should not really be needed in practical use cases, but
certain out-of-memory test cases can trigger allocation failure in
eloop_register_timeout() and if that happens, the previous EAPOL
supplicant state machine implementation got pretty much completely stuck
for any operation needing the timer.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Verify that the AP uses matching PMKR1Name in (Re)Association Response
frame when going through FT initial mobility domain association using
FILS. Thise step was missing from the initial implementation, but is
needed to match the IEEE 802.11ai requirements for explicit confirmation
of the FT key hierarchy (similarly to what is done in FT 4-way handshake
when FILS is not used).
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Derive PMKR1Name during the FILS authentication step, verify that the
station uses matching PMKR1Name in (Re)Association Request frame, and
add RSNE[PMKR1Name] into (Re)Association Response frame when going
through FT initial mobility domain association using FILS. These steps
were missed from the initial implementation, but are needed to match the
IEEE 802.11ai requirements for explicit confirmation of the FT key
hierarchy (similarly to what is done in FT 4-way handshake when FILS is
not used).
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This gets rid of a confusing error message "FILS: Failed to add PMKSA
cache entry based on ERP" for cases where PMKSA caching is disabled in
hostapd (disable_pmksa_caching=1). Functionality remains unchanged,
i.e., no cache entry was added before this change either.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
NL80211_CMD_GET_KEY response may return the actual key in addition to
the last used sequence number that we need. That might result in a key
being left in unused heap memory after the buffer is freed.
Explicitly clear the message payload with the possibly included key
material from heap memory before returning from the handler function
(and having libnl free the nlmsg) when key information is obtained from
the driver using the NL80211_CMD_GET_KEY command.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Add a check in nl80211 driver layer to not include PMK while sending
NL80211_CMD_DEL_PMKSA explicitly. Though it is taken care already in
supplicant layer by setting the pmk_len to zero, it would be good
to have a check in nl80211 layer in order to avoid future accidental
inclusions of keying material in commands that do not need them.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Send out the new Protocol Version attribute in Authentication
Request/Response messages and determine the peer version based on this
attribute.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
hostapd configuration parameters fragm_threshold and rts_threshold were
documented to disable the threshold with value -1 and not change driver
configuration if the parameter is not included. However, -1 was mapped
into not changing the driver value, so the explicit disabling part did
not work.
Replace the default values for these to be -2 so that explicitly set
configuration value -1 can be distinguished from the case of not
including the parameter. Map the -1 value to a driver request to disable
the threshold. Ignore any error from this operation just in case to
avoid breaking functionality should some drivers not accept the (u32) -1
value as a threshold value request to disable the mechanism.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This subcommand is used to update Zigbee state and specified WLAN
durations to enhance success ratio of Zigbee joining network. The
attributes defined in enum qca_mpta_helper_vendor_attr are used to
deliver these parameters to the driver.
Signed-off-by: stonez <stonez@codeaurora.org>
Peer rate statistics is per-peer cached data in the driver. These
statistics needs to be flushed to a user space application on
synchronous/asynchronous events. This command is used as an event from
the driver to flush per-peer cached statistics to the application.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This patch added 'check_cert_subject' support to match the value of
every field against the DN of the subject in the client certificate. If
the values do not match, the certificate verification will fail and will
reject the user.
This option allows hostapd to match every individual field in the right
order, also allow '*' character as a wildcard (e.g OU=Development*).
Note: hostapd will match string up to 'wildcard' against the DN of the
subject in the client certificate for every individual field.
Signed-off-by: Paresh Chaudhary <paresh.chaudhary@rockwellcollins.com>
Signed-off-by: Jared Bents <jared.bents@rockwellcollins.com>
Signed-off-by: Jouni Malinen <j@w1.fi>
These parameters were using the u8*/len style types even though they
were used as char* strings without an explicit length field. Make this
char* instead of u8* to avoid confusion and unnecessary type casting.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This gets rid of some static analyzer warnings about uninitialized
variables being used in comparisons or write operations.
Signed-off-by: Jouni Malinen <j@w1.fi>
Include only one of hostapd_mgmt_rx() and hostapd_action_rx() functions
in the build. Previously, NEED_AP_MLME builds (i.e., cases where hostapd
AP MLME implementation is included) included both of these functions and
both were tried in sequence. In addition to being difficult to
understand, that could result in unexpected behavior if
hostapd_mgmt_rx() rejected a frame and return 0 to allow
hostapd_action_rx() to attempt to process the frame.
All the operations included in hostapd_action_rx() are supposed to be
available through the hostapd_mgmt_rx() call in handle_action() and
those should result in the exact same Category/Action-based handler
function to be called in the end. As such, this should not result in
different behavior. And if there is a difference, that would be pointing
at a hidden bug that would need to be fixed anyway. Furthermore, builds
without NEED_AP_MLME would not have any difference in behavior or
contents of the binary either.
Signed-off-by: Jouni Malinen <j@w1.fi>
There is no need to go through the following handler calls in
hostapd_action_rx() after having found the matching WLAN_ACTION_WNM
handler.
Signed-off-by: Jouni Malinen <j@w1.fi>
hostapd_action_rx() was pointing at incorrect field (Action vs.
Category) for the wpa_ft_action_rx() call and the length check for SA
Query Action frames. This resulted in those frames getting dropped as
invalid (FT) or ignored as truncated (SA Query). Fix this by pointing to
the correct place at the beginning of the frame body.
This issue had a long history. These were broken during cleanup in
commit dbfb8e82ff ("Remove unnecessary EVENT_RX_ACTION") which
actually fixed the initial reason for the error accidentally. It was
just that that error was needed to cancel out another earlier error..
One of the errors came from misuse of the EVENT_RX_ACTION API in commit
deca6eff74 ("atheros: Add new IEEE 802.11r driver_ops"). That pointed
struct rx_action data/len to cover the Action frame from the Category
field to the end of the frame body while the API was documented to cover
Action field to the end of the frame body. This error was cancelled by
another error in commit 88b32a99d3 ("FT: Add FT AP support for drivers
that manage MLME internally") that called wpa_ft_action_rx() with the
struct rx_action::data field as the second argument. That argument needs
to point to the Category field, but that struct rx_action field was
supposed to point to the Action field.
Number of the Action frame handlers added into hostapd_action_rx() had
been fixed more or less accidentally after this in various other
commits, but the FT and SA Query handlers had ended up maintaining the
incorrect operations. This is now fixing those.
This seems to fix at least some cases of FT-over-DS with drivers that
use driver-based AP MLME. Such drivers might use internal SA Query
processing, so it is not clear whether that part actually fixes any real
issues.
Signed-off-by: Jouni Malinen <j@w1.fi>
The struct hostapd_eap_user changes with a new allocated variable were
not covered in the RADIUS server code. Fix this by using eap_user_free()
instead of custom memory freeing operation in radius_server.c.
The hwsim tests with salted password (ap_wpa2_eap_pwd_salt_sha1,
ap_wpa2_eap_pwd_salt_sha256, ap_wpa2_eap_pwd_salt_sha512) triggered
these memory leaks.
Fixes: d52ead3db7 ("EAP-pwd server: Add support for salted password databases")
Signed-off-by: Masashi Honma <masashi.honma@gmail.com>
Copy the Finite Cyclic Group field value from the request to the
response Authentication frame if we end up rejecting the request due to
unsupported group.
IEEE Std 802.11-2016 has conflicting statements about this behavior.
Table 9-36 (Presence of fields and elements in Authentication frames)
indicates that the Finite Cyclic Group field is only included with
status code values 0 (success) and 76 (anti-clogging token request)
while SAE protocol description implying that the Finite Cyclic Group
field is set to the rejected group (12.4.8.6.3 and 12.4.8.6.4).
The standard language needs to cleaned up to describe this
unambiguously, but since it looks safe to add the field into the
rejection case and since there is desire to have the field present to be
able to implement what exactly is stated in 12.4.8.6.4, it looks
reasonable to move ahead with the AP mode implementation change. There
is no change in wpa_supplicant for now to modify its behavior based on
whether this field is present, i.e., wpa_supplicant will continue to
work with both the old and new hostapd behavior for SAE group
negotiation.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
tmp2 (y^2) was derived once in each iteration of the loop and only freed
after all the loop iterations. Fix this by freeing the temporary value
during each iteration.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The new hostapd configuration parameter wps_cred_add_sae=1 can be used
to request hostapd to add SAE configuration whenever WPS is used to
configure the AP to use WPA2-PSK and the credential includes a
passphrase (instead of PSK). This can be used to enable WPA3-Personal
transition mode with both SAE and PSK enabled and PMF enabled for PSK
and required for SAE associations.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Add a new QCA vendor specific feature capability indication for the
device to indicate the support of TWT.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The current/default behavior of set blacklist BSSID QCA vendor command
is a mandate to the driver - do not consider this BSSID for connect/roam
till reset.
There are use cases where this need not be a mandate and thus could
provide the flexibility for the driver to consider this BSSID if there
are no better ones. Such use cases can use this new flag attribute to
only hint the blacklist of a BSSID to the driver.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Do not start SAE authentication from scratch if a STA starts a new
attempt for the same group while we still have previously generated PWE
available. Instead, use the previously generated PWE as-is and skip
anti-clogging token exchange since the heavy processing is already
completed. This saves unnecessary processing on the AP side in case the
STA failed to complete authentication on the first attempt (e.g., due to
heavy SAE load on the AP causing a timeout) and makes it more likely for
a valid STA to be able to complete SAE authentication during a DoS
attack.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Do not start SAE authentication from scratch when the AP requests
anti-clogging token to be used. Instead, use the previously generated
PWE as-is if the retry is for the same AP and the same group. This saves
unnecessary processing on the station side in case the AP is under heavy
SAE authentiation load.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Add a 16-bit token index into the anti-clogging token. This can be used
to enforce only a single use of each issued anti-clogging token request.
The token value is now token-index |
last-30-octets-of(HMAC-SHA256(sae_token_key, STA-MAC-address |
token-index)), i.e., the first two octets of the SHA256 hash value are
replaced with the token-index and token-index itself is protected as
part of the HMAC context data.
Track the used 16-bit token index values and accept received tokens only
if they use an index value that has been requested, but has not yet been
used. This makes it a bit more difficult for an attacker to perform DoS
attacks against the heavy CPU operations needed for processing SAE
commit since the attacker cannot simply replay the same frame multiple
times and instead, needs to request each token separately.
While this does not add significant extra processing/CPU need for the
attacker, this can be helpful in combination with the queued processing
of SAE commit messages in enforcing more delay during flooding of SAE
commit messages since the new anti-clogging token values are not
returned before the new message goes through the processing queue.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This allows better control of processing new SAE sessions so that other
operations can be given higher priority during bursts of SAE requests,
e.g., during a potential DoS attack. The receive commit messages are
queued (up to maximum of 15 entries) and processed from eloop callback.
If the queue has multiple pending entries, more wait time is used to go
through the each new entry to reduce heavy CPU load from SAE processing.
Enable anti-clogging token use also based on the pending commit message
queue and not only based on the already started sessions.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Change the AP mode default for SAE to enable only the group 19 instead
of enabling all ECC groups that are supported by the used crypto library
and the SAE implementations. The main reason for this is to avoid
enabling groups that are not as strong as the mandatory-to-support group
19 (i.e., groups 25 and 26). In addition, this disables heavier groups
by default.
In addition, add a warning about MODP groups 1, 2, 5, 22, 23, and 24
based on "MUST NOT" or "SHOULD NOT" categorization in RFC 8247. All the
MODP groups were already disabled by default and would have needed
explicit configuration to be allowed.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Pass the group order (if known/specified) to crypto_dh_derive_secret()
(and also to OpenSSL DH_generate_key() in case of Group 5) and verify
that the public key received from the peer meets 1 < pubkey < p and
pubkey^q == 1 mod p conditions.
While all these use cases were using only ephemeral DH keys, it is
better to use more explicit checks while deriving the shared secret to
avoid unexpected behavior.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This extension of VLAN assignment code had a bug in one of the code
paths where vlan_id could have been left uninitialized. This could
result in SAE authentication getting rejected in cases where VLAN
assignment is not used if the uninitialized stack memory had nonzero
value.
Fixes: dbfa691df4 ("VLAN assignment based on used WPA/WPA2 passphrase/PSK")
Signed-off-by: Jouni Malinen <j@w1.fi>
Use unsigned 1 (1U) instead of signed (1) when doing left shift that
could potentially need to use all bits of the 32-bit unsigned variable.
radius_server.c:2254:14: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
Signed-off-by: Jouni Malinen <j@w1.fi>
Avoid an unnecessary unsigned integer overflow warning due to loop index
j-- use.
hostapd.c:661:10: runtime error: unsigned integer overflow: 0 - 1 cannot be represented in type 'unsigned int'
Signed-off-by: Jouni Malinen <j@w1.fi>
ie.wps_ie is an array, so there is no point in checking whether it is
NULL.
driver_atheros.c:1221:9: error: address of array 'ie.wps_ie' will
always evaluate to 'true' [-Werror,-Wpointer-bool-conversion]
Signed-off-by: Jouni Malinen <j@w1.fi>
(&mgmt->u.deauth.reason_code + 1) is not exactly clean and now that we
have the u8 variable[] member in the struct after this field, use that
directly to avoid clang compiler warning:
ctrl_iface_ap.c:454:18: error: taking address of packed member
'reason_code' of class or structure 'ieee80211_mgmt::(anonymous
union)::(anonymous)' may result in an unaligned pointer value
[-Werror,-Waddress-of-packed-member]
Signed-off-by: Jouni Malinen <j@w1.fi>
FST_MAX_LLT_MS definition depended on undefined behavior with unsigned
integer overflow. Avoid that and also optimize the
FST_LLT_{MS_TO_VAL,VAL_TO_MS} macros to handle larger values without
overflowing 32-bit unsigned integers.
fst_session.c:1274:52: runtime error: unsigned integer overflow: 4294967295 * 32 cannot be represented in type 'unsigned int'
Signed-off-by: Jouni Malinen <j@w1.fi>
Split the check and decrementation into separate steps to avoid an
unnecessary UBSan warning.
hostapd.c:1895:14: runtime error: unsigned integer overflow: 0 - 1 cannot be represented in type 'size_t' (aka 'unsigned long')
Signed-off-by: Jouni Malinen <j@w1.fi>
