]> git.ipfire.org Git - thirdparty/hostap.git/log
thirdparty/hostap.git
7 years agotests: Fix ap_cipher_tkip_countermeasures_sta2
Jouni Malinen [Sun, 19 Nov 2017 14:02:07 +0000 (16:02 +0200)] 
tests: Fix ap_cipher_tkip_countermeasures_sta2

hostapd implementation was changed to use a valid Status Code when
rejecting the connection. This test case was forgotten at the time, but
it needs a matching change to allow the new value (1 instead of 14).

Signed-off-by: Jouni Malinen <j@w1.fi>
7 years agotests: DPP protocol testing - invalid Config Attr Obj in Conf Req
Jouni Malinen [Sun, 19 Nov 2017 12:12:55 +0000 (14:12 +0200)] 
tests: DPP protocol testing - invalid Config Attr Obj in Conf Req

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing for invalid Config Attrib Object value
Jouni Malinen [Sun, 19 Nov 2017 12:12:10 +0000 (14:12 +0200)] 
DPP: Protocol testing for invalid Config Attrib Object value

Extend dpp_test to cover a case where Config Attrib Object value is
invalid in Configuration Request frame.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol testing - invalid Status in Auth Resp/Conf
Jouni Malinen [Sun, 19 Nov 2017 10:53:50 +0000 (12:53 +0200)] 
tests: DPP protocol testing - invalid Status in Auth Resp/Conf

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing for invalid DPP Status value
Jouni Malinen [Sun, 19 Nov 2017 10:53:04 +0000 (12:53 +0200)] 
DPP: Protocol testing for invalid DPP Status value

Extend dpp_test to cover cases where DPP Status value is invalid in
Authentication Response/Confirm frames.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Use helper functions to build Bootstrap Key Hash attributes
Jouni Malinen [Sun, 19 Nov 2017 10:41:57 +0000 (12:41 +0200)] 
DPP: Use helper functions to build Bootstrap Key Hash attributes

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Replace custom undefined attr with DPP Status in after-wrapped data
Jouni Malinen [Sun, 19 Nov 2017 10:32:00 +0000 (12:32 +0200)] 
DPP: Replace custom undefined attr with DPP Status in after-wrapped data

This has the same impact and is needed for some testing needs.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Use a helper function to build DPP Status attribute
Jouni Malinen [Sun, 19 Nov 2017 10:27:14 +0000 (12:27 +0200)] 
DPP: Use a helper function to build DPP Status attribute

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut DPP/QR (mutual) initiator (response pending)
Jouni Malinen [Sun, 19 Nov 2017 10:07:07 +0000 (12:07 +0200)] 
tests: sigma_dut DPP/QR (mutual) initiator (response pending)

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol testing for invalid I/R-bootstrap key hash
Jouni Malinen [Sun, 19 Nov 2017 09:33:22 +0000 (11:33 +0200)] 
tests: DPP protocol testing for invalid I/R-bootstrap key hash

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing for writing invalid I/R Bootstrap Key Hash
Jouni Malinen [Sun, 19 Nov 2017 09:32:02 +0000 (11:32 +0200)] 
DPP: Protocol testing for writing invalid I/R Bootstrap Key Hash

Extend dpp_test to cover cases where Initiator/Responder Bootstrap Key
Hash value in DPP Authentication frames is invalid (flip one bit).

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol testing - invalid I/R-proto key in Auth Req/Resp
Jouni Malinen [Sat, 18 Nov 2017 22:12:37 +0000 (00:12 +0200)] 
tests: DPP protocol testing - invalid I/R-proto key in Auth Req/Resp

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing capability to generate invalid Protocol Key
Jouni Malinen [Sat, 18 Nov 2017 22:11:44 +0000 (00:11 +0200)] 
DPP: Protocol testing capability to generate invalid Protocol Key

This extends dpp_test to allow invalid Initiator/Responder Protocol Key
to be written into the Authentication Request/Response frame.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut controlled AP and PSKHEX
Jouni Malinen [Sat, 18 Nov 2017 15:48:27 +0000 (17:48 +0200)] 
tests: sigma_dut controlled AP and PSKHEX

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoStronger GTK derivation routine
Jouni Malinen [Sat, 18 Nov 2017 15:08:11 +0000 (17:08 +0200)] 
Stronger GTK derivation routine

If the build include SHA384, use that to derive GTK from GMK. In
addition, add more random bytes bytes to the PRF-X() context data for
longer GTK to reduce dependency on the randomness of the GMK.

GMK is 256 bits of random data and it was used with SHA256, so the
previous design was likely sufficient for all needs even with 128 bits
of additional randomness in GTK derivation. Anyway, adding up to 256
bits of new randomness and using SHA384 can be helpful extra protection
particularly for the cases using GCMP-256 or CCMP-256 as the group
cipher.

Signed-off-by: Jouni Malinen <j@w1.fi>
7 years agotests: Use group 26 instead of 25 in mesh test cases
Jouni Malinen [Sat, 18 Nov 2017 11:57:59 +0000 (13:57 +0200)] 
tests: Use group 26 instead of 25 in mesh test cases

This allows mesh_sae_groups_invalid and
wpas_mesh_secure_sae_group_negotiation to be run with BoringSSL (group
25 not available anymore).

Signed-off-by: Jouni Malinen <j@w1.fi>
7 years agotests: Skip fils_sk_pfs_25 with BoringSSL
Jouni Malinen [Sat, 18 Nov 2017 11:55:41 +0000 (13:55 +0200)] 
tests: Skip fils_sk_pfs_25 with BoringSSL

It looks like BoringSSL has dropped support for group 25 (192-bit Random
ECP Group).

Signed-off-by: Jouni Malinen <j@w1.fi>
7 years agotests: Skip ap_wpa2_eap_fast_eap_sim if no EAP-FAST support
Jouni Malinen [Sat, 18 Nov 2017 11:49:19 +0000 (13:49 +0200)] 
tests: Skip ap_wpa2_eap_fast_eap_sim if no EAP-FAST support

Signed-off-by: Jouni Malinen <j@w1.fi>
7 years agoDPP: Fix dpp_test_gen_invalid_key() with BoringSSL
Jouni Malinen [Sat, 18 Nov 2017 11:22:17 +0000 (13:22 +0200)] 
DPP: Fix dpp_test_gen_invalid_key() with BoringSSL

Unlike OpenSSL, BoringSSL returns an error from
EC_POINT_set_affine_coordinates_GFp() is not on the curve. As such, need
to behave differently here depending on which library is used.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: Skip DPP tests with Brainpool curves when using BoringSSL
Jouni Malinen [Sat, 18 Nov 2017 11:07:44 +0000 (13:07 +0200)] 
tests: Skip DPP tests with Brainpool curves when using BoringSSL

BoringSSL does not include these EC curves.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Build bootstrapping key DER encoding using custom routine
Jouni Malinen [Sat, 18 Nov 2017 10:19:43 +0000 (12:19 +0200)] 
DPP: Build bootstrapping key DER encoding using custom routine

While the OpenSSL version of i2d_EC_PUBKEY() seemed to be able to use
the POINT_CONVERSION_COMPRESSED setting on the EC key, that did not seem
to work with BoringSSL. Since this is not exactly robust design, replace
use of i2d_EC_PUBKEY() with a custom routine that enforces the DPP rules
on SubjectPublicKeyInfo (compressed format of the public key,
ecPublicKey OID, parameters present and indicating the curve by OID).

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Use a helper function to DER encode bootstrapping key
Jouni Malinen [Sat, 18 Nov 2017 10:14:21 +0000 (12:14 +0200)] 
DPP: Use a helper function to DER encode bootstrapping key

This routine was previously implemented twice using i2d_EC_PUBKEY().
There is no need to duplicate that implementation and especially since
it looks like this implementation needs to be replaced for BoringSSL,
start by using a shared helper function for both locations so that there
is only a single place that uses i2d_EC_PUBKEY() to build the special
DPP bootstrapping key DER encoding.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoBoringSSL: Add AES support with 192-bit keys
Jouni Malinen [Fri, 17 Nov 2017 19:03:04 +0000 (21:03 +0200)] 
BoringSSL: Add AES support with 192-bit keys

BoringSSL restored the previously removed AES-192 ECB support in ("Add
AES-192 ECB.") commit. Since this is needed for DPP with the P-384
curve, restore support for this through EVP_aes_192_ecb().

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoOpenSSL: Debug message if requested AES key length is not supported
Jouni Malinen [Fri, 17 Nov 2017 18:57:56 +0000 (20:57 +0200)] 
OpenSSL: Debug message if requested AES key length is not supported

This makes it clearer why some AES operations fail especially with
BoringSSL where the 192-bit case is not supported.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoBoringSSL: Add DPP special cases regardless of claimed version number
Jouni Malinen [Fri, 17 Nov 2017 18:44:42 +0000 (20:44 +0200)] 
BoringSSL: Add DPP special cases regardless of claimed version number

It looks like BoringSSL claims to have OPENSSL_VERSION_NUMBER for a
1.1.0 version, but it does not provide ECDSA_SIG_set0() or
ECDSA_SIG_get0(). For now, add the helper functions regardless of the
version BoringSSL claims to be. Similarly, include the X509_ALGOR_get0()
workaround unconditionally for BoringSSL.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoBoringSSL: Implement crypto_ecdh_init()
Jouni Malinen [Fri, 17 Nov 2017 18:41:25 +0000 (20:41 +0200)] 
BoringSSL: Implement crypto_ecdh_init()

BoringSSL does not provide some of the OpenSSL API that was used here,
so update this to use similar design to what was already done with DPP
key derivation.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoBoringSSL: Comment out SSL_set_default_passwd_cb*() calls
Jouni Malinen [Fri, 17 Nov 2017 18:34:17 +0000 (20:34 +0200)] 
BoringSSL: Comment out SSL_set_default_passwd_cb*() calls

It looks like BoringSSL claims to have OPENSSL_VERSION_NUMBER for a
1.1.0 version, but it does not provide SSL_set_default_passwd_cb*(). For
now, comment out this regardless of the version BoringSSL claims to be.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoBoringSSL: Comment out SSL_set1_sigalgs_list() call
Jouni Malinen [Fri, 17 Nov 2017 18:30:37 +0000 (20:30 +0200)] 
BoringSSL: Comment out SSL_set1_sigalgs_list() call

It looks like BoringSSL claims to have OPENSSL_VERSION_NUMBER for a
1.1.0 version, but it does not provide SSL_set1_sigalgs_list(). For now,
comment out this regardless of the version BoringSSL claims to be.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoBoringSSL: Define RSA_bits() helper
Jouni Malinen [Fri, 17 Nov 2017 18:24:46 +0000 (20:24 +0200)] 
BoringSSL: Define RSA_bits() helper

It looks like BoringSSL claims to have OPENSSL_VERSION_NUMBER for a
1.1.0 version, but it does not provide RSA_bits(). For now, add this
backwards compatibility wrapper for BoringSSL regardless of the version
it claims to be.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut DPP Configurator and self-configuration
Jouni Malinen [Fri, 17 Nov 2017 17:14:21 +0000 (19:14 +0200)] 
tests: sigma_dut DPP Configurator and self-configuration

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: Fix spelling of sigma_dut DPPConfEnrolleeRole argument
Jouni Malinen [Fri, 17 Nov 2017 17:13:50 +0000 (19:13 +0200)] 
tests: Fix spelling of sigma_dut DPPConfEnrolleeRole argument

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: Fix ap_config_reload_on_sighup test
Sriram R [Fri, 17 Nov 2017 09:43:36 +0000 (15:13 +0530)] 
tests: Fix ap_config_reload_on_sighup test

Use absolute path name for configuration file to ensure the file can be
succesfully reloaded and read on SIGHUP signal. This is needed when
running the test case on host (i.e., not using a VM).

Signed-off-by: Sriram R <srirrama@qti.qualcomm.com>
7 years agotests: sigma_dut and group cipher override on AP
Jouni Malinen [Fri, 17 Nov 2017 10:44:50 +0000 (12:44 +0200)] 
tests: sigma_dut and group cipher override on AP

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut and AP enabling two pairwise ciphers
Jouni Malinen [Fri, 17 Nov 2017 10:33:08 +0000 (12:33 +0200)] 
tests: sigma_dut and AP enabling two pairwise ciphers

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: hostapd group_cipher override
Jouni Malinen [Fri, 17 Nov 2017 10:33:33 +0000 (12:33 +0200)] 
tests: hostapd group_cipher override

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoAllow group cipher selection to be overridden
Jouni Malinen [Fri, 17 Nov 2017 10:31:41 +0000 (12:31 +0200)] 
Allow group cipher selection to be overridden

The new hostapd configuration parameter group_cipher can now be used to
override the automatic cipher selection based on enabled pairwise
ciphers. It should be noted that selecting an unexpected group cipher
can result in interoperability issues and this new capability is mainly
for testing purposes.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agowlantest: Search bss/sta entry more thoroughly for 4-address frames
Jouni Malinen [Wed, 15 Nov 2017 00:12:20 +0000 (02:12 +0200)] 
wlantest: Search bss/sta entry more thoroughly for 4-address frames

Previous design worked for the case where only one of the devices was
beaconing, but failed in one direction to find the PTK if both devices
beaconed. Fix this by checking the A1/A2 fields in both directions if
the first pick fails to find the sta entry.

In addition, select the proper rsc value (rsc_tods vs. rsc_fromds) based
on A2 (TA) value for ToDS+FromDS frames to avoid reporting incorrect
replay issues.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoFix block comment style in QCA vendor attribute definition
Edayilliam Jayadev [Wed, 8 Nov 2017 05:52:40 +0000 (11:22 +0530)] 
Fix block comment style in QCA vendor attribute definition

Fix the block comment style issue introduced as part of commit
fbfceef3af7f8d454926548f6a7ebe7f9bb4848a ("Add QCA vendor commands for
spectral scan").

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoAdd QCA_NL80211_VENDOR_SUBCMD_PEER_FLUSH_PENDING
tinlin [Tue, 14 Nov 2017 03:37:03 +0000 (11:37 +0800)] 
Add QCA_NL80211_VENDOR_SUBCMD_PEER_FLUSH_PENDING

Add sub-command QCA_NL80211_VENDOR_SUBCMD_PEER_FLUSH_PENDING to flush
pending packets in firmware. The attributes are listed in enum
qca_wlan_vendor_attr_flush_pending. The QCA_WLAN_VENDOR_ATTR_PEER_ADDR
specifies the peer MAC address and the QCA_WLAN_VENDOR_ATTR_AC specifies
the access category of the pending packets.

Signed-off-by: Lin Tingting <tinlin@qti.qualcomm.com>
7 years agoAdd new QCA vendor attribute for LL stats
Zhang Qian [Wed, 8 Nov 2017 08:49:04 +0000 (16:49 +0800)] 
Add new QCA vendor attribute for LL stats

A new vendor attribute QCA_WLAN_VENDOR_ATTR_LL_STATS_WMM_AC_PENDING_MSDU
is added for vendor sub-command QCA_NL80211_VENDOR_SUBCMD_LL_STATS_GET.
This attribute is for pending MSDUs corresponding to respective AC.

Signed-off-by: Zhang Qian <zhangq@qti.qualcomm.com>
7 years agoatheros: Process SAE authentication frames using EVENT_RX_MGMT
Ashok Kumar Ponnaiah [Tue, 14 Nov 2017 16:47:07 +0000 (18:47 +0200)] 
atheros: Process SAE authentication frames using EVENT_RX_MGMT

This adds support for SAE in AP mode with the atheros driver interface.
EVENT_RX_MGMT includes SAE processing while EVENT_AUTH would require
more changes to make this work.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoatheros: Generate EVENT_TX_STATUS events for management frames
Ashok Kumar Ponnaiah [Tue, 14 Nov 2017 16:44:05 +0000 (18:44 +0200)] 
atheros: Generate EVENT_TX_STATUS events for management frames

This is needed for DPP functionality.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agocommon: Avoid conflict with __bitwise macro from linux/types.h
Beniamino Galvani [Mon, 30 Oct 2017 10:14:40 +0000 (11:14 +0100)] 
common: Avoid conflict with __bitwise macro from linux/types.h

Undefine the __bitwise macro before defining it to avoid conflicts
with the one from linux/types.h; the same is done some lines above
when __CHECKER__ is defined. Fixes the following warning:

  In file included from ../src/l2_packet/l2_packet_linux.c:15:0:
  hostap/src/utils/common.h:438:0: warning: "__bitwise" redefined
   #define __bitwise

  In file included from /usr/include/linux/filter.h:9:0,
                   from ../src/l2_packet/l2_packet_linux.c:13:
  /usr/include/linux/types.h:21:0: note: this is the location of the previous definition
   #define __bitwise __bitwise__

Signed-off-by: Beniamino Galvani <bgalvani@redhat.com>
7 years agoDPP: Fix compiler warning of testing code
Masashi Honma [Thu, 9 Nov 2017 20:13:21 +0000 (05:13 +0900)] 
DPP: Fix compiler warning of testing code

../src/common/dpp.c: In function 'dpp_test_gen_invalid_key':
../src/common/dpp.c:5531:10: warning: return makes integer from pointer without a cast [-Wint-conversion]
   return NULL;
          ^

Signed-off-by: Masashi Honma <masashi.honma@gmail.com>
7 years agowlantest: Do not ignore RSN/WPA/OSEN element before full BSS info
Jouni Malinen [Tue, 14 Nov 2017 10:55:48 +0000 (12:55 +0200)] 
wlantest: Do not ignore RSN/WPA/OSEN element before full BSS info

wlantest used to ignore RSN/WPA/OSEN element in (Re)Association Request
frame if no Beacon frame had been seen from the AP before the
association exchange. This could result in not being able to derive keys
properly. Work around this by skipping that step if the BSS entry is not
yet complete.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agowlantest: Remove unnecessary duplication of tk_len from STA entries
Jouni Malinen [Tue, 14 Nov 2017 10:54:45 +0000 (12:54 +0200)] 
wlantest: Remove unnecessary duplication of tk_len from STA entries

The length of the TK is available within struct wpa_ptk, so there is no
need to try to maintain it separately in wlantest.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoReject PMK-to-PTK derivation with unsupported cipher
Jouni Malinen [Tue, 14 Nov 2017 10:50:30 +0000 (12:50 +0200)] 
Reject PMK-to-PTK derivation with unsupported cipher

There should be no wpa_pmk_to_ptk() calls with the cipher argument
indicating a cipher that is not allowed as a pairwise cipher. However,
it looks like that was possible to happen with wlantest. Check for this
corner case explicitly to avoid generating confusing debug logs.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut DPP/QR (mutual) responder (response pending sequence)
Jouni Malinen [Mon, 13 Nov 2017 17:16:35 +0000 (19:16 +0200)] 
tests: sigma_dut DPP/QR (mutual) responder (response pending sequence)

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut DPP/QR responder (channel list override)
Jouni Malinen [Mon, 13 Nov 2017 16:44:24 +0000 (18:44 +0200)] 
tests: sigma_dut DPP/QR responder (channel list override)

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol testing - Auth Conf replaced by Res
Jouni Malinen [Mon, 13 Nov 2017 10:55:34 +0000 (12:55 +0200)] 
tests: DPP protocol testing - Auth Conf replaced by Res

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Testing capability to send unexpected Authentication Response
Jouni Malinen [Mon, 13 Nov 2017 10:47:30 +0000 (12:47 +0200)] 
DPP: Testing capability to send unexpected Authentication Response

This is for protocol testing to check what happens if the Responser
receives an unexpected Authentication Response instead of Authentication
Confirm.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Reject unexpected Req/Resp message based on Auth/PKEX role
Jouni Malinen [Mon, 13 Nov 2017 10:54:20 +0000 (12:54 +0200)] 
DPP: Reject unexpected Req/Resp message based on Auth/PKEX role

This prevents issues where an unexpected message in the DPP
Authentication exchange or PKEX could result in undefined behavior.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP Authentication Response retries
Jouni Malinen [Mon, 13 Nov 2017 10:33:45 +0000 (12:33 +0200)] 
tests: DPP Authentication Response retries

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Retransmit DPP Authentication Response frame if it is not ACKed
Jouni Malinen [Mon, 13 Nov 2017 10:34:17 +0000 (12:34 +0200)] 
DPP: Retransmit DPP Authentication Response frame if it is not ACKed

This extends wpa_supplicant DPP implementation to retransmit DPP
Authentication Response frame every 10 seconds up to 5 times if the peer
does not reply with DPP Authentication Confirm frame.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Stop authentication exchange of DPP_STOP_LISTEN
Jouni Malinen [Mon, 13 Nov 2017 10:12:08 +0000 (12:12 +0200)] 
DPP: Stop authentication exchange of DPP_STOP_LISTEN

Previously, this command stopped listen operation immediately, but if
there was an ongoing authentication exchange, a new listen operation was
started. This is not really expected behavior, so stop the
authentication exchange first with this command to avoid restarting
listen operation.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP with Initiator in either role
Jouni Malinen [Mon, 13 Nov 2017 09:29:21 +0000 (11:29 +0200)] 
tests: DPP with Initiator in either role

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Allowed initiator to indicate either role
Jouni Malinen [Mon, 13 Nov 2017 09:29:45 +0000 (11:29 +0200)] 
DPP: Allowed initiator to indicate either role

The new role=either parameter can now be used with DPP_AUTH_INIT to
indicate that the initiator can take either the Configurator or Enrollee
role.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP bootstrapping info channel list
Jouni Malinen [Sun, 12 Nov 2017 10:17:32 +0000 (12:17 +0200)] 
tests: DPP bootstrapping info channel list

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Support multiple channels for initiating DPP Authentication
Jouni Malinen [Sun, 12 Nov 2017 10:17:54 +0000 (12:17 +0200)] 
DPP: Support multiple channels for initiating DPP Authentication

This extends wpa_supplicant to iterate over all available channels from
the intersection of what the peer indicates and the local device
supports when initiating DPP Authentication. In addition, retry DPP
Authentication Request frame up to five times if no response is
received.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Share a helper function for PKEX final steps
Jouni Malinen [Sun, 12 Nov 2017 07:43:13 +0000 (09:43 +0200)] 
DPP: Share a helper function for PKEX final steps

Generate the PKEX bootstrapping information and release the PKEX session
in a helper function that both the initiator and responder can use
instead of maintaining this functionality separately in two places.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut DPP protocol testing - Peer Discovery Request
Jouni Malinen [Mon, 6 Nov 2017 10:59:12 +0000 (12:59 +0200)] 
tests: sigma_dut DPP protocol testing - Peer Discovery Request

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol testing - network introduction
Jouni Malinen [Mon, 6 Nov 2017 10:40:47 +0000 (12:40 +0200)] 
tests: DPP protocol testing - network introduction

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing to allow missing attributes in peer discovery
Jouni Malinen [Mon, 6 Nov 2017 10:40:18 +0000 (12:40 +0200)] 
DPP: Protocol testing to allow missing attributes in peer discovery

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut and DPP protocol testing
Jouni Malinen [Sun, 5 Nov 2017 22:35:58 +0000 (00:35 +0200)] 
tests: sigma_dut and DPP protocol testing

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: sigma_dut controlled AP as DPP PKEX responder
Jouni Malinen [Sat, 4 Nov 2017 10:32:46 +0000 (12:32 +0200)] 
tests: sigma_dut controlled AP as DPP PKEX responder

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: hostapd as DPP Responder requiring mutual authentication
Jouni Malinen [Sat, 4 Nov 2017 09:29:21 +0000 (11:29 +0200)] 
tests: hostapd as DPP Responder requiring mutual authentication

This verifies the qr=mutual case with STATUS_RESPONSE_PENDING.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Add DPP_LISTEN and DPP_STOP_LISTEN to hostapd
Jouni Malinen [Sat, 4 Nov 2017 09:27:00 +0000 (11:27 +0200)] 
DPP: Add DPP_LISTEN and DPP_STOP_LISTEN to hostapd

This is an initial step in allowing non-operating channel operations
with DPP when hostapd is the Responder. For now, this is only used for
specifying role=configurator/enrollee and qr=mutual cases similarly to
the wpa_supplicant configuration for in Responder role. Request to use a
non-operating channel will be rejected.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoAdd QCA vendor commands for spectral scan
Edayilliam Jayadev [Wed, 25 Oct 2017 08:51:49 +0000 (14:21 +0530)] 
Add QCA vendor commands for spectral scan

Add the following vendor commands and their vendor attributes for
spectral scan.

 1) QCA_NL80211_VENDOR_SUBCMD_SPECTRAL_SCAN_GET_CONFIG
        Get current values of spectral parameters.
 2) QCA_NL80211_VENDOR_SUBCMD_SPECTRAL_SCAN_GET_DIAG_STATS
        Get stats for spectral scan debug.
 3) QCA_NL80211_VENDOR_SUBCMD_SPECTRAL_SCAN_GET_CAP_INFO
        Get the spectral hardware capability.
 4) QCA_NL80211_VENDOR_SUBCMD_SPECTRAL_SCAN_GET_STATUS
        Get the current status of spectral scan.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoP2P: Continue P2P_WAIT_PEER_(IDLE/CONNECT) sequence on a listen cancel
Sunil Dutt [Mon, 9 Oct 2017 14:32:20 +0000 (20:02 +0530)] 
P2P: Continue P2P_WAIT_PEER_(IDLE/CONNECT) sequence on a listen cancel

Some drivers may accept the remain-on-channel command, but instead of
indicating start event for remain-on-channel, just indicate that the
operation has been canceled immediately. This listen cancel from the
WAIT_PEER_CONNECT state ended up in discontinuation of further
WAIT_PEER_IDLE/WAIT_PEER_CONNECT state transitions. Hence, delay the
subsequent IDLE state by 100 ms.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoP2P: ACS offload for the autonomous GO
Sunil Dutt [Mon, 25 Sep 2017 15:56:52 +0000 (21:26 +0530)] 
P2P: ACS offload for the autonomous GO

This commit introduces the ACS functionality for the autonomous GO. The
optional parameter <freq> in p2p_group_add is enhanced to carry a value
"acs" with the intention to select the channels among any supported
band. freq = 2 / 5 carry the need to select the channels only in the
respective bands 2.4 / 5 GHz. This functionality is on top of the host
driver's capability to offload ACS, which is advertized through
WPA_DRIVER_FLAGS_ACS_OFFLOAD.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoAdd new QCA vendor attributes for MAC counters
Zhang Qian [Fri, 29 Sep 2017 02:36:45 +0000 (10:36 +0800)] 
Add new QCA vendor attributes for MAC counters

Add QCA_WLAN_VENDOR_ATTR_LL_STATS_EXT_REPORT_TIME
and QCA_WLAN_VENDOR_ATTR_LL_STATS_EXT_MEASUREMENT_TIME to record
timestamp and duration for the last MAC counters. Some user layer
application is delay sensitive. It needs to know time stamp and
measurment duration for the counters.

Signed-off-by: Zhang Qian <zhangq@qti.qualcomm.com>
7 years agotests: DPP Protocol testing for invalid Config Resp attribute values
Jouni Malinen [Fri, 3 Nov 2017 19:11:25 +0000 (21:11 +0200)] 
tests: DPP Protocol testing for invalid Config Resp attribute values

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing for invalid Config Resp attribute values
Jouni Malinen [Fri, 3 Nov 2017 19:11:00 +0000 (21:11 +0200)] 
DPP: Protocol testing for invalid Config Resp attribute values

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP Config Request/Response missing attributes
Jouni Malinen [Fri, 3 Nov 2017 18:59:35 +0000 (20:59 +0200)] 
tests: DPP Config Request/Response missing attributes

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Report Config Request/Response failure reasons on control interface
Jouni Malinen [Fri, 3 Nov 2017 18:58:53 +0000 (20:58 +0200)] 
DPP: Report Config Request/Response failure reasons on control interface

This provides more details of failures to upper layer components.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing to remove attributes from Config Req/Resp
Jouni Malinen [Fri, 3 Nov 2017 18:39:00 +0000 (20:39 +0200)] 
DPP: Protocol testing to remove attributes from Config Req/Resp

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol - I/R-auth tag mismatch in PKEX Commit-Reveal
Jouni Malinen [Fri, 3 Nov 2017 18:18:52 +0000 (20:18 +0200)] 
tests: DPP protocol - I/R-auth tag mismatch in PKEX Commit-Reveal

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing - invalid I/R-Auth value in PKEX Commit-Reveal
Jouni Malinen [Fri, 3 Nov 2017 18:18:24 +0000 (20:18 +0200)] 
DPP: Protocol testing - invalid I/R-Auth value in PKEX Commit-Reveal

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol - invalid Bootstrap Key in PKEX Commit-Reveal
Jouni Malinen [Fri, 3 Nov 2017 17:58:41 +0000 (19:58 +0200)] 
tests: DPP protocol - invalid Bootstrap Key in PKEX Commit-Reveal

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing - invalid Bootstrap Key value in PKEX Commit-Reveal
Jouni Malinen [Fri, 3 Nov 2017 17:54:48 +0000 (19:54 +0200)] 
DPP: Protocol testing - invalid Bootstrap Key value in PKEX Commit-Reveal

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol testing - invalid Status in PKEX Exchange Response
Jouni Malinen [Fri, 3 Nov 2017 17:41:24 +0000 (19:41 +0200)] 
tests: DPP protocol testing - invalid Status in PKEX Exchange Response

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Protocol testing - invalid Status value in PKEX Exchange Response
Jouni Malinen [Fri, 3 Nov 2017 17:33:17 +0000 (19:33 +0200)] 
DPP: Protocol testing - invalid Status value in PKEX Exchange Response

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Move PKEX z derivation on Responder to earlier phase
Jouni Malinen [Fri, 3 Nov 2017 17:24:41 +0000 (19:24 +0200)] 
DPP: Move PKEX z derivation on Responder to earlier phase

K and z can be derived already based on information available at the
time the PKEX Exchange Request is being processed, so move these there
from the PKEX Commit-Reveal Request processing since that matches the
DPP tech spec description close and allows PKEX exchange to be aborted
earlier if anything unexpected happens.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Fix a typo in a debug print
Jouni Malinen [Fri, 3 Nov 2017 17:02:07 +0000 (19:02 +0200)] 
DPP: Fix a typo in a debug print

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Explicitly check that PKEX Qr is not the point-at-infinity
Jouni Malinen [Fri, 3 Nov 2017 17:01:24 +0000 (19:01 +0200)] 
DPP: Explicitly check that PKEX Qr is not the point-at-infinity

This was already done for Qi, but the same needs to be done for Qr as
well.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP and PKEX with mismatching code limit
Jouni Malinen [Fri, 3 Nov 2017 16:55:32 +0000 (18:55 +0200)] 
tests: DPP and PKEX with mismatching code limit

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: PKEX counter t
Jouni Malinen [Fri, 3 Nov 2017 15:49:45 +0000 (17:49 +0200)] 
DPP: PKEX counter t

Add limit on number of failed attempts that could have used PKEX code.
If the limit (5) is reached, drop the PKEX state (including the code)
and report this on the control interface to indicate that a new code
needs to be entered due to possible attack.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP PKEX retry after incorrect PKEX code
Jouni Malinen [Fri, 3 Nov 2017 15:15:08 +0000 (17:15 +0200)] 
tests: DPP PKEX retry after incorrect PKEX code

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Terminate PKEX exchange on detection of a mismatching code
Jouni Malinen [Fri, 3 Nov 2017 15:13:17 +0000 (17:13 +0200)] 
DPP: Terminate PKEX exchange on detection of a mismatching code

Clean up the pending PKEX exchange if Commit-Reveal Request processing
indicates a mismatch in the PKEX code. Previously, the this case was
silently ignored and the session was left in pending state that
prevented new PKEX exchanges from getting initated. Now, a new attempt
is allowed to be initiated.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Use dpp_bn2bin_pad() helper to simplify code
Jouni Malinen [Fri, 3 Nov 2017 14:49:48 +0000 (16:49 +0200)] 
DPP: Use dpp_bn2bin_pad() helper to simplify code

Number of places writing BIGNUM values with left-padding were open
coding this helper functionality unnecessarily.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP and PKEX with mismatching curve
Jouni Malinen [Fri, 3 Nov 2017 14:45:28 +0000 (16:45 +0200)] 
tests: DPP and PKEX with mismatching curve

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: PKEX and STATUS_BAD_GROUP
Jouni Malinen [Fri, 3 Nov 2017 14:43:58 +0000 (16:43 +0200)] 
DPP: PKEX and STATUS_BAD_GROUP

Report mismatching finite cyclic group with PKEX Exchange Response using
STATUS_BAD_GROUP and provide more detailed error report over the control
interface on the peer device when this happens.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Remove obsolete TODO comment on discovery object
Jouni Malinen [Fri, 3 Nov 2017 10:09:15 +0000 (12:09 +0200)] 
DPP: Remove obsolete TODO comment on discovery object

The optional channel information was removed from the discovery object
in the DPP tech spec, so no need to maintain this TODO note anymore.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: WPA+WPA2-PSK/TKIP countermeasures (detected by mixed mode AP)
Jouni Malinen [Fri, 3 Nov 2017 08:52:24 +0000 (10:52 +0200)] 
tests: WPA+WPA2-PSK/TKIP countermeasures (detected by mixed mode AP)

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoFix Status Code in TKIP countermeasures case
Jouni Malinen [Fri, 3 Nov 2017 08:49:45 +0000 (10:49 +0200)] 
Fix Status Code in TKIP countermeasures case

The previously used WLAN_REASON_MICHAEL_MIC_FAILURE (14) value as a
response to Authentication frame or (Re)Association Request frame is not
correct since the resp value is encoded in the Status Code (not Reason
Code) field. Status Code 14 is WLAN_STATUS_UNKNOWN_AUTH_TRANSACTION
which is really what this value would have meant in the response frames.

There is no Michael MIC failure status code, so have to use the generic
"Unspecified failure" (1) reason code for these cases.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: DPP protocol testing - invalid Encrypted Key in PKEX Exchange
Jouni Malinen [Thu, 2 Nov 2017 22:49:35 +0000 (00:49 +0200)] 
tests: DPP protocol testing - invalid Encrypted Key in PKEX Exchange

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Testing capability to generate invalid PKEX encrypted key (M and N)
Jouni Malinen [Thu, 2 Nov 2017 22:42:54 +0000 (00:42 +0200)] 
DPP: Testing capability to generate invalid PKEX encrypted key (M and N)

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agotests: Missing attributes in DPP PKEX messages
Jouni Malinen [Thu, 2 Nov 2017 22:14:44 +0000 (00:14 +0200)] 
tests: Missing attributes in DPP PKEX messages

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Report PKEX failure reasons over control interface
Jouni Malinen [Thu, 2 Nov 2017 22:14:02 +0000 (00:14 +0200)] 
DPP: Report PKEX failure reasons over control interface

This provides more information to upper layer software to report failure
reasons on the UI.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
7 years agoDPP: Extend protocol testing to cover missing attributes in PKEX
Jouni Malinen [Thu, 2 Nov 2017 21:53:55 +0000 (23:53 +0200)] 
DPP: Extend protocol testing to cover missing attributes in PKEX

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>