Commit graph

9400 commits

Author SHA1 Message Date
Andrei Otcheretianski
223ccebfa3 P2PS: Save wps_prov_info on the responder side
When device A sends PD response to device B, device A should save
wps_prov_info for device B. Not doing so would result in a redundant and
incorrect PD flow, e.g., when upon PROV-DISC-DONE event device B starts
a GO and device A should join it.

Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
2015-07-07 20:26:07 +03:00
Andrei Otcheretianski
2fc866d1bc P2PS: Validate WPS config methods more accurately in PD Request
In case of a P2PS PD, allow keypad, display, and P2PS WPS config
methods. For a legacy PD, allow keypad, display, and pushbutton methods.

Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
2015-07-07 20:24:01 +03:00
Max Stepanov
ea210b9f8a P2P: Refactor p2p_process_prov_disc_resp() function
Add 'else if' to P2PS status verification to prevent a redundant
condition checking. The first 'if' condition is true only if
  status == P2P_SC_SUCCESS || status == P2P_SC_SUCCESS_DEFERRED.
while the second condition checks:
  status != P2P_SC_SUCCESS &&
  status != P2P_SC_FAIL_INFO_CURRENTLY_UNAVAILABLE &&
  status != P2P_SC_SUCCESS_DEFERRED
Thus, the two conditions are mutually exclusive and 'else if' can be
used if this case.

Signed-off-by: Max Stepanov <Max.Stepanov@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
2015-07-07 20:21:44 +03:00
Max Stepanov
20f4c3d76b P2P: Delete redundant comparison in p2p_process_prov_disc_resp()
Delete redundant comparison of msg.wps_config_methods with
dev->req_config_methods in p2p_process_prov_disc_resp() since it's
already done early in this function. Also, the second comparison
doesn't make too much sense: it can happen after a possible
p2p_reset_pending_pd() call setting dev->req_config_methods to 0.

Signed-off-by: Max Stepanov <Max.Stepanov@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
2015-07-07 20:20:43 +03:00
Max Stepanov
82d6113502 P2PS: Fix adv_id and adv_mac params of P2P-PROV-DISC-FAILURE
When wpa_supplicant receives a PD Response with reject status it
generated P2P-PROV-DISC-FAILURE event without adv_id and adv_mac
parameters. Fix this by adding these parameters to the
wpas_prov_disc_fail() function call.

Signed-off-by: Max Stepanov <Max.Stepanov@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
2015-07-07 19:40:49 +03:00
Dan Williams
e8492c8145 hostap: Fix send_mlme() after 'freq' parameter addition
This fixes the incomplete driver_hostap.c change from commit
5d180a7739 ('drivers: Add freq parameter
to send_mlme() function') that did not take into account the internal
callers.

Signed-off-by: Dan Williams <dcbw@redhat.com>
2015-07-07 19:40:49 +03:00
Jouni Malinen
e0236b152c tests: IEEE 802.1X and HELD state
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 19:25:31 +03:00
Jouni Malinen
c8a15753eb Send EAPOL-Start on HELD->CONNECTING transition
Previously, only CONNECTING->CONNECTING case ended up sending out an
EAPOL-Start frame to avoid sending the unnecessary initial EAPOL-Start.
However, this optimization prevented new EAPOL-Start from being
initiated when leaving the HELD state. Allow that case to trigger
immediate EAPOL-Start transmission to speed up connection.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 19:25:31 +03:00
Mike Gerow
471c810bc1 pkcs11: Don't ask for a new PIN on TLS handshake failure
The only time the PIN should fail is when we initialize the TLS
connection, so it doesn't really make sense to get rid of the PIN just
because some other part of the handshake failed.

This is a followup to commit fd4fb28179
('OpenSSL: Try to ensure we don't throw away the PIN unnecessarily').

Signed-off-by: Mike Gerow <gerow@google.com>
2015-07-07 19:25:31 +03:00
Jouni Malinen
1f560ff08f tests: Add Undefined Behavior Sanitizer (ubsan) configuration
This allows hostapd and wpa_supplicant to be built for hwsim test cases
with ubsan functinality from the recent gcc/clang compiler versions.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 18:23:04 +03:00
Jouni Malinen
66f1e078e7 DATA_TEST_TX: Avoid ubsan warning on 0x80<<24 not fitting in int
Use unsigned constant instead of signed to avoid warning with the LSB
being set in an int.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
75352270fa Avoid misaligned IPv4 header accesses in DATA_TEST_* commands
The IPv4 header after the Ethernet header is not 32-bit aligned and the
previous version ended up accessing 32-bit members at misaligned
addresses.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
476a634d60 Avoid ubsan warning on 0x80<<24 not fitting in int in WPA_GET_BE32/LE32
Use a typecast to make this shift unsigned so that the MSB fits within
the range of allowed values.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
cf6fd19b34 ndisc_snoop: Avoid misaligned read of IPv6 address
The IPv6 address in the frame buffer may not be 32-bit aligned, so use a
local copy to align this before reading the address with 32-bit reads
(s6_addr32[]).

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
e9ed7d9898 EAP-TTLS: Avoid ubsan warning on 0x80<<24 not fitting in int
Use a typecast to make this unsigned so that the MSB fits within the
range of allowed values.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
0f5fb8a467 radiotap: Avoid ubsan warning on 1<<31
Use the BIT() macro and unsigned int to avoid ubsan warning on 1<<31 not
fitting in an int.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
fc880b11ed NFC: Avoid misaligned read of an NDEF field
The 32-bit version of payload length field may not be 32-bit aligned in
the message buffer, so use WPA_GET_BE32() to read it instead of ntohl().

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
e01281fccc trace: Use explicit alignment requirements to avoid misalignment
64-bit builds with CONFIG_WPA_TRACE=y resulted in the wpabuf pointers
getting misaligned (only 32-bit aligned) and that would result in reads
and writes of unaligned size_t values. Avoid this by indicating explicit
alignment requirement for wpabuf_trace to 8 octets (i.e., there will be
extra four octets of padding in case of 64-bit builds).

Similarly, struct os_alloc_trace resulted in some potential misalignment
cases, e.g., when CONFIG_ACS=y uses a 'long double' variable within
struct hostapd_channel_data. Avoid misalignment issues with explicit
alignment indication.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
f97e3ce473 drivers: Use unsigned arguments for sta_set_flags()
Since BIT() is now returning unsigned int, these functions need to match
that to avoid compiler warnings.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
32d6463fe7 Make BIT() unsigned int instead of int
This is needed to avoid ubsan warnings on BIT(31).

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
bd47c80564 Include stddef.h to use the standard offsetof()
src/utils/list.h ended up defining a local version of offsetof() due to
stddef.h not getting included. This resulted in unnecessary warnings
from ubsan related to "dereferencing" of a NULL pointer.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
028b19675d RADIUS DAS: Avoid compiler warning on abs()
The input parameter ended up being converted to long int instead of int,
so use an explicit typecase to get rid of the compiler warning.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
597e8afeb4 ERP: Avoid mixing of enum types
Use explicit typecasting to avoid implicit conversion warnings in cases
where enum eap_erp_type is used in functions taking an EapType argument.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
96e8c13da7 D-Bus: Fix dont_quote const declaration
Commit 38279bdb35 ('D-Bus: Coding style
cleanup') handled the dbus_new_handlers.c change properly, but misplaced
the second 'const' in dbus_old_handlers.c in a way that resulted in
duplicated const rather than marking the actual value const.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
34d3eaa8ee tests: WPA2-PSK-FT AP over DS protocol testing
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
cf671d54bd tests: WPA2-PSK-FT and OOM
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
7b741a5383 tests: WPA2-PSK-FT AP with GCMP-256 cipher
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
05a90d78dc FT: Allow CCMP-256 and GCMP-256 as group ciphers
The FT-specific check for valid group cipher in wpa_ft_gen_req_ies() was
not up-to-date with the current list of supported ciphers. Fix this by
using a generic function to determine validity of the cipher. In
practice, this adds support for using CCMP-256 and GCMP-256 as the group
cipher with FT.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
348bc4e0eb tests: IEEE 802.1X connection and EAPOL-Key protocol tests
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
2ff4928967 tests: IEEE 802.1X and EAPOL-Start retransmissions
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-07 16:25:06 +03:00
Jouni Malinen
85774b70a1 tests: EAP server and OOM
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-01 00:42:09 +03:00
Jouni Malinen
405c621cdb tests: WPA2-Enterprise connection using MAC ACL
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-07-01 00:34:27 +03:00
Jouni Malinen
df7ad0fa11 tests: EAP-TLS and unknown OCSP signer
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-30 22:05:37 +03:00
Jouni Malinen
d2a1047e67 tests: Invalid OCSP data (parsing failure)
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-30 22:01:05 +03:00
Jouni Malinen
7be5ec991d tests: Invalid openssl_ciphers value
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-30 21:55:59 +03:00
Jouni Malinen
f24b97972b OpenSSL: Merge error returns
These similar error cases can use a single return statement.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-30 21:48:22 +03:00
Jouni Malinen
84d6a17a27 TLS: Remove unused tls_capabilities()
This mechanism to figure out TLS library capabilities has not been used
since commit fd2f2d0489 ('Remove
EAP-TTLSv1 and TLS/IA') (Sep 2011).

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-30 21:40:48 +03:00
Jouni Malinen
6eddd5303c tests: EAP-FAST/MSCHAPv2 and server OOM
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-30 21:36:12 +03:00
Jouni Malinen
b3ff3decf6 tests: DH parameter file DSA conversion and error cases
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
0c83ae0469 tests: EAP-TLS with PKCS12 that includes additional certificates
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
b1fb42751a tests: EAP-TLS and OOM
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
b197a8194b tests: EAP-TLS and server checking CRL
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
786722763d ms_funcs: Merge similar return cases
There is no need to have separate return statements for these corner
cases that are unlikely to be hit in practice.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
359636170f hw_features: Merge similar return case in check_40mhz_2g4()
There is no need to have separate return statements for these corner
cases that are unlikely to be hit in practice.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
486f4e3c83 tests: EAP-SIM and OOM
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
f1ab79c34c tests: WPA2-Enterprise connection using EAP-IKEv2 and OOM
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
8c4e4c01f6 tests: WPA2-Enterprise connection using EAP-PSK and OOM
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
10ba4ae4d3 tests: SAE and no random numbers available
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
bf7071bb0c tests: WPA2-PSK AP and no random numbers available
Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00
Jouni Malinen
aac1efec34 Reject the initial 4-way handshake if initial GTK setup fails
This makes the AP/Authenticator design more robust against unexpected
failures in random number generation.

Signed-off-by: Jouni Malinen <j@w1.fi>
2015-06-29 23:23:56 +03:00