Jouni Malinen [Fri, 6 Feb 2015 10:08:28 +0000 (12:08 +0200)]
Document the wpa_msg_cb "global" parameter
Instead of an int variable with magic values 0, 1, 2, use an enum that
gives clearer meaning to the values now that the original boolean type
global argument is not really a boolean anymore.
Jouni Malinen [Wed, 10 Jun 2015 12:40:35 +0000 (15:40 +0300)]
Do not set own_disconnect_req flag if not connected
This fixes some issues where a disconnection event may get ignored if an
earlier operation to remove or disable a network resulted in an attempt
to disconnect event though no connection was in place.
wpa_s->current_ssid != NULL alone is not sufficient to determine that
there will be a driver event notifying completion of such disconnection
request. Set own_disconnect_req to 1 only if wpa_s->wpa_state is also
indicating that there is a connection or an attempt to complete one.
This showed up in a failure, e.g., when running the hwsim test case
scan_int followed by ap_vlan_wpa2_psk_radius_required where the latter
ended up not processing a connection failure event and getting stuck not
trying to run a new scan and connection attempt.
This adds support for Qt5 while still preserving support for building
with Qt4. The same source code builds with both Qt versions now. The
functionality should be identical.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jouni Malinen [Wed, 25 Mar 2015 09:41:26 +0000 (11:41 +0200)]
Fix pairwise cipher suite bitfields to the driver in mixed mode
Commit 95b6bca66de9dc0a2ddd9164ec052a7d5f58804b ('Add rsn_pairwise bits
to set_ieee8021x() driver_ops') modified cipher configuration to use
unconditionally wpa_pairwise | rsn_pairwise. While that works for many
cases, it does not handle the case of dynamic configuration changes over
the control interface where wpa_pairwise or rsn_pairwise values may not
get cleared when the wpa parameter is modified. Fix this inconsistency
by configuring the driver with only the bits that are valid for the
currently enabled WPA/WPA2 version(s).
Jouni Malinen [Wed, 10 Jun 2015 10:41:54 +0000 (13:41 +0300)]
Remove [MU-BEAMFORMEE] option from hostapd vht_capab parameter
The standard hardcodes the MU Beamformee Capable subfield is hardcoded
to 0 when transmitting by an AP, so there is no need to provide a
configuration parameter for setting this to one.
MAYANK HAARIT [Tue, 2 Jun 2015 05:47:33 +0000 (05:47 +0000)]
P2P: Add GO Intent of connecting device in GO Negotiation Request event
Add GO Intent information of connecting device in GO Negotiation Request
event which will help applications to decide its own GO intent value in
advance and can avoid failure cases when both devices use GO Intent 15
depending on application requirement.
Alan T. DeKok [Thu, 4 Jun 2015 20:53:06 +0000 (16:53 -0400)]
Set Acct-Session-Id from os_get_random() instead of os_get_time()
So that systems with bad clocks will send random session IDs,
instead of always ones starting at the same second.
If os_get_random() isn't available, use os_get_time(). But also
mix in now.tv_usec, so that the accounting session ID is more
likely to be globally and temporally unique.
Krishna Vamsi [Thu, 21 May 2015 13:59:31 +0000 (19:29 +0530)]
Add support to request a scan with specific SSIDs
Support a request to scan specific SSIDs given by user with the SCAN
command. The SSID list can be suffixed to the scan command as follows.
For example, if SSIDs "ABC" and "abc123" need to be specifically
scanned, the command should be "SCAN ssid 414243 ssid 616263313233". The
value of the SSID is passed in hexadecimal representation.
Jouni Malinen [Mon, 1 Jun 2015 15:56:19 +0000 (18:56 +0300)]
hostapd: Make sure band selection does not result in NULL dereference
Explicitly check for iface->current_mode before dereferencing it. While
this case may not happen in practice, it is better for the setup
functions to be more careful when doing the initial band selection.
Amit Khatri [Thu, 28 May 2015 12:19:46 +0000 (17:49 +0530)]
P2P: Add manufacturer info into D-Bus peer interface
Allow the Manufacturer information for a found P2P peer device to be
fetched through the D-Bus interface similarly to p2p_peer <mac address>
on ctrl interface.
P2P: Fix a potential memory leak in a P2P+NFC corner case
Avoid memory leak due to previous allocation for ssid->ssid. It does not
look like this is hit in normal cases, but at might be possible for the
SSID to get set if the peer is present in previous scan results and WPS
code ends up copying the SSID from there.
Krishna Vamsi [Fri, 8 May 2015 22:32:36 +0000 (15:32 -0700)]
Add libwpa_client build option to use a dynamic library
Add support to compile libwpa_client.so. This can be used by external
programs to interact with the wpa_supplicant control interface. Also
rename the static version of this library to libwpa_client.a to be
consistent with the name used previous in Android builds.
Krishna Vamsi [Thu, 28 May 2015 14:17:41 +0000 (19:47 +0530)]
P2P: Restart group formation timer upon receiving new Inv Req
A new Invitation Request might be received on a just started persistent
group if the previous Invitation Response sent isn't received at the
peer (GO of the persistent group). When the peer sends an Invitation
Request again, treat it as start of group formation and restart the
group formation timer at this point of time.
Sunil Dutt [Thu, 28 May 2015 15:11:12 +0000 (20:41 +0530)]
Postpone updating of wpa_s->current_bss till association event
In the case of driver-based BSS selection, a bssid_hint in the connect
request might not result in to the selection of the same BSS. Thus,
postpone the updation of the wpa_s->current_bss till the association
event is received unless the BSSID is forced. This fixes issues where
wpa_s->current_bss may end up being updated to point to the BSS that
wpa_supplicant provided as a hint for a roaming case, but then not
restored if the driver decides to "return" to the current BSS instead.
This could result in some operations (e.g., WNM BSS TM response) not
working properly due to incorrect BSS being identified in
wpa_s->current_bss.
Peng Xu [Fri, 8 May 2015 17:53:08 +0000 (10:53 -0700)]
Extend hw_mode to support any band for offloaded ACS case
When device supports dual band operations with offloaded ACS, hw_mode
can now be set to any band (hw_mode=any) in order to allow ACS to select
the best channel from any band. After a channel is selected, the hw_mode
is updated for hostapd.
Ilan Peer [Tue, 12 May 2015 14:39:58 +0000 (17:39 +0300)]
P2P: Use p2p_init_wpa_s for P2P command redirection over D-Bus
D-Bus used the p2p_dev member in struct wpa_supplicant to track the
interface used for P2P Device operations. However, this is not needed as
the interface used for P2P Device operations is stored in struct
wpa_global->p2p_init_wpa_s.
Replace all the redirections to wpa_s->p2p_dev with redirections to
wpa_s->global->p2p_init_wpa_s and removed this member as it is no
longer needed.
Jouni Malinen [Mon, 25 May 2015 15:21:05 +0000 (18:21 +0300)]
P2P: Fix D-Bus PresenceRequest to use group interface
This command is for a specific P2P group, not for the P2P Device
instance and as such, wpa_s pointer must not be replaced with the
interface that is used for P2P Device management.
Ilan Peer [Tue, 12 May 2015 14:40:01 +0000 (17:40 +0300)]
P2P: Fix wpas_remove_persistent_peer() to use P2P mgmt interface
The function used wpa_s->parent->conf to iterate the P2P networks and
update the configuration file. However, wpa_s->parent is not
necessarily the interface used to manage the P2P Device operations.
Fix this by accessing the configuration file of the interface
initialized to manage the P2P Device operations.
Ilan Peer [Tue, 12 May 2015 14:40:00 +0000 (17:40 +0300)]
P2P: Fix wpas_p2p_add_persistent_group_client() to use P2P mgmt interface
The function used wpa_s->parent->conf to iterate the P2P networks and
update the configuration file. However, wpa_s->parent is not
necessarily the interface used to manage the P2P Device operations.
Fix this by accessing the configuration file of the interface initialized
to managed the P2P Device operations.
Ilan Peer [Tue, 12 May 2015 14:39:59 +0000 (17:39 +0300)]
P2P: Fix wpas_p2p_set_own_freq_preference() to use P2P mgmt interface
wpas_p2p_set_own_freq_preference() accessed wpa_s->parent->conf to test
if p2p_ignore_shared_freq is set, but wpa_s->parent is not necessarily
the interface used to manage the P2P Device operations.
Fix this by accessing the configuration file of the interface
initialized to manage the P2P Device operations.
Ben Rosenfeld [Tue, 12 May 2015 14:39:57 +0000 (17:39 +0300)]
P2P: Use the P2P Device management interface in wpas_p2p_remove_client()
As wpas_p2p_remove_client() is not necessarily called from the interface
used to manage the P2P Device operations, when removing a client, use
the P2P management interface to iterate over the saved networks and
remove the relevant entries form the P2P GO network blocks.
Signed-off-by: Ben Rosenfeld <ben.rosenfeld@intel.com>
Sunil Dutt [Sat, 9 May 2015 09:25:03 +0000 (14:55 +0530)]
P2P: Advertize cross connection to WLAN AP on a non-P2P interface
Commit 1c2aa04c96626f5b000d167bb5274a8c39b7dac2 ('P2P: Do not add P2P
IEs on P2P disabled interface') removed the P2P IEs from association on
non-P2P interface. However, an AP functioning as a P2P manager needs the
cross connection capability of the station (P2P Device). This needs to
be done to meet the P2P specification requirements even if the station
interface has p2p_disabled=1 in case P2P in general is enabled.
Jouni Malinen [Sun, 24 May 2015 08:58:45 +0000 (11:58 +0300)]
EAP-EKE: Add Session-Id
While RFC 6124 does not define how Session-Id is constructed for
EAP-EKE, there seems to be consensus among the authors on the
construction. Use this Type | Nonce_P | Nonce_S construction based on
the following email:
From: Yaron Sheffer <yaronf.ietf at gmail.com>
To: ietf at ietf.org
Date: Wed, 17 Nov 2010 13:13:42 +0200
Expanding on my previous response, I suggest to resolve Bernard's
concern by adding the following text:
5.6 EAP Key Generation
EAP-EKE can be used for EAP key generation, as defined by [RFC 5247].
When used in this manner, the values required to establish the key
hierarchy are defined as follows:
- Peer-Id is the EAP-EKE ID_P value.
- Server-Id is the EAP-EKE ID_S value.
- Session-Id is the concatenated Type | Nonce_P | Nonce_S, where Type is
the method type defined for EAP-EKE in [Sec. 4.1], a single octet.
Jouni Malinen [Sun, 24 May 2015 07:47:21 +0000 (10:47 +0300)]
Increase DH key size in the hostapd.conf example
OpenSSL is moving to use 2048-bit DH key size as the default with
dhparam. Increase the value in the hostapd.conf to match that to reduce
likelihood of ending up using a shorter key.
ACS: Scan only channels specified in the channel list
The ACS code part of hostapd scans all the channels even if the channel
list is specified in the hostapd.conf. Limit the ACS scan channels to
the list specified in the config file.
Jouni Malinen [Fri, 8 May 2015 15:21:54 +0000 (18:21 +0300)]
tests: Fix persistent_group_per_sta_psk for P2P Device case
The dev[1] <--> dev[2] data connectivity test was using incorrect
function. dev[2] is also using a P2P group and as such, can have a
different group interface.
Jouni Malinen [Fri, 8 May 2015 15:15:56 +0000 (18:15 +0300)]
tests: Fix go_neg_with_bss_connected with P2P Device
Group interface name was fetched from the results of an incorrect group
formation and because of this, group removal failed in case P2P Device
is used and dev[1] ends up getting different group ifname for the
groups.
Jouni Malinen [Fri, 8 May 2015 14:59:52 +0000 (17:59 +0300)]
tests: Add P2P Device support for persistent_group
The network operations need to use the global control interface to be
performed on the interface that stores the network profiles for
persistent groups.
Ben Rosenfeld [Tue, 5 May 2015 09:37:04 +0000 (12:37 +0300)]
tests: Update group_ifname after group start
After P2P-GROUP-STARTED event, use group_form_result in order to update
the group_ifname for the device. This is needed when using P2P Device
for managing P2P operations which results in a separate group interface
being used.
Signed-off-by: Ben Rosenfeld <ben.rosenfeld@intel.com>
Ben Rosenfeld [Tue, 5 May 2015 09:36:58 +0000 (12:36 +0300)]
tests: Change list_networks() to use global interface with P2P
list_networks() always used the wlanX control interface to query for the
current list of networks. However, when a dedicated P2P Device is used,
the global control interface should be used when checking persistent
group network profiles.
Fix this by adding an optional parameter indicating that the P2P
networks are requested, and in such a case use the global control
interface.
In addition update test_p2p_persistent to use the argument when needed.
Signed-off-by: Ben Rosenfeld <ben.rosenfeld@intel.com>
Ilan Peer [Tue, 5 May 2015 09:36:56 +0000 (12:36 +0300)]
tests: Skip some tests in P2PS when a dedicated P2P Device is used
Some tests in test_p2ps.py test a scenario where a separate P2P
group interface is not used. However, this is not a valid case
when a dedicated P2P Device interface is used, as in such a case
a separate group interface must be used.
Handle this by skipping such tests in case a dedicated P2P Device is
used.
Ben Rosenfeld [Tue, 5 May 2015 09:36:54 +0000 (12:36 +0300)]
tests: Remove all P2P networks in call to reset
When resetting a device, remove all the P2P networks to prevent
unexpected behavior in following tests. This is needed for the case
where P2P Device interface is used.
Signed-off-by: Ben Rosenfeld <ben.rosenfeld@intel.com>
Ben Rosenfeld [Tue, 5 May 2015 09:36:51 +0000 (12:36 +0300)]
tests: Change set_country() to use the global interface to get event
This is required for cases that multiple interfaces are used and the
event can happen on any of them, for example when a dedicated P2P Device
interface is used.
Signed-off-by: Ben Rosenfeld <ben.rosenfeld@intel.com>
Ilan Peer [Mon, 4 May 2015 12:50:46 +0000 (15:50 +0300)]
P2P: Fix association with an AP/P2P GO that is not a P2P manager
Do not add a P2P IE when a station interface is trying to associate
to an AP or P2P GO that publishes a P2P IE but does not include
a P2P manageability attribute.
This addresses an interoperability issue that was reported in
https://bugzilla.kernel.org/show_bug.cgi?id=96471, where a P2P GO
rejects association from a station interface without a specified
reason.
Sunil Dutt [Wed, 6 May 2015 11:25:10 +0000 (16:55 +0530)]
Add QCA vendor subcmd for Link Property Query
Link Property query vendor command shall facilitate the information
of the Wi-Fi link. MAC address of the Wi-Fi peer is given as an input
for querying the link properties.
The L (Length) and M (More) flags needs to be cleared before deciding
whether the locally generated response requires fragmentation. This
fixes an issue where these flags from the server could have been invalid
for the following message. In some cases, this could have resulted in
triggering the wpabuf security check that would terminate the process
due to invalid buffer allocation.
Jouni Malinen [Sat, 2 May 2015 16:26:06 +0000 (19:26 +0300)]
EAP-pwd server: Fix Total-Length parsing for fragment reassembly
The remaining number of bytes in the message could be smaller than the
Total-Length field size, so the length needs to be explicitly checked
prior to reading the field and decrementing the len variable. This could
have resulted in the remaining length becoming negative and interpreted
as a huge positive integer.
In addition, check that there is no already started fragment in progress
before allocating a new buffer for reassembling fragments. This avoid a
potential memory leak when processing invalid message.
Jouni Malinen [Sat, 2 May 2015 16:23:04 +0000 (19:23 +0300)]
EAP-pwd peer: Fix Total-Length parsing for fragment reassembly
The remaining number of bytes in the message could be smaller than the
Total-Length field size, so the length needs to be explicitly checked
prior to reading the field and decrementing the len variable. This could
have resulted in the remaining length becoming negative and interpreted
as a huge positive integer.
In addition, check that there is no already started fragment in progress
before allocating a new buffer for reassembling fragments. This avoid a
potential memory leak when processing invalid message.
Jouni Malinen [Fri, 1 May 2015 13:40:44 +0000 (16:40 +0300)]
EAP-pwd server: Fix payload length validation for Commit and Confirm
The length of the received Commit and Confirm message payloads was not
checked before reading them. This could result in a buffer read
overflow when processing an invalid message.
Fix this by verifying that the payload is of expected length before
processing it. In addition, enforce correct state transition sequence to
make sure there is no unexpected behavior if receiving a Commit/Confirm
message before the previous exchanges have been completed.
Thanks to Kostya Kortchinsky of Google security team for discovering and
reporting this issue.
Jouni Malinen [Fri, 1 May 2015 13:37:45 +0000 (16:37 +0300)]
EAP-pwd peer: Fix payload length validation for Commit and Confirm
The length of the received Commit and Confirm message payloads was not
checked before reading them. This could result in a buffer read
overflow when processing an invalid message.
Fix this by verifying that the payload is of expected length before
processing it. In addition, enforce correct state transition sequence to
make sure there is no unexpected behavior if receiving a Commit/Confirm
message before the previous exchanges have been completed.
Thanks to Kostya Kortchinsky of Google security team for discovering and
reporting this issue.
AP WMM: Fix integer underflow in WMM Action frame parser
The length of the WMM Action frame was not properly validated and the
length of the information elements (int left) could end up being
negative. This would result in reading significantly past the stack
buffer while parsing the IEs in ieee802_11_parse_elems() and while doing
so, resulting in segmentation fault.
This can result in an invalid frame being used for a denial of service
attack (hostapd process killed) against an AP with a driver that uses
hostapd for management frame processing (e.g., all mac80211-based
drivers).
Thanks to Kostya Kortchinsky of Google security team for discovering and
reporting this issue.
WPS: Replace the httpread_debug design with standard debug prints
The debug information from httpread can be helpful in figuring out error
cases in general and as such, should be enabled by default. Get rid of
the hardcoded httpread_debug value that would require source code
changes to enable.
Verify that ncopy parameter to memcpy is not negative. While this is not
supposed to be needed, it is a good additional protection against
unknown implementation issues.
strtoul() return value may end up overflowing the int h->chunk_size and
resulting in a negative value to be stored as the chunk_size. This could
result in the following memcpy operation using a very large length
argument which would result in a buffer overflow and segmentation fault.
This could have been used to cause a denial service by any device that
has been authorized for network access (either wireless or wired). This
would affect both the WPS UPnP functionality in a WPS AP (hostapd with
upnp_iface parameter set in the configuration) and WPS ER
(wpa_supplicant with WPS_ER_START control interface command used).
Validate the parsed chunk length value to avoid this. In addition to
rejecting negative values, we can also reject chunk size that would be
larger than the maximum configured body length.
Thanks to Kostya Kortchinsky of Google security team for discovering and
reporting this issue.
Matti Gottlieb [Tue, 28 Apr 2015 11:01:04 +0000 (14:01 +0300)]
Fix sending ANQP request to an unknown BSS while associated
While being associated, if an ANQP request is received for a different
AP that doesn't exist in the BSS list, the ANQP request will be sent on
the frequency of the AP that we are currently associated to.
In such a case, it is possible that the ANQP request would be sent on
a channel different than that of the requested AP, potentially delaying
other requests/activities.
Avoid sending the ANQP request to an AP that is not in the BSS list.
Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com>
Fix memory leak introduced in commit 32a097fdd26b9401fbd22054a2a01ba2d71f139a ("wpa_cli: Keep track of
available networks") by tracking networks only when in interactive mode.
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>