A newly created GO might move to another channel before the client was
able to connect to it. This creates a situation where the client
searches the GO on the channel agreed upon during GO Negotiation or
invitation signaling, while the GO is on another channel. This in turn
might lead to delayed connection or connection failure and group
removal.
Fix this by disallowing a GO CS as long as there is some activity that
should delay the switch. If a GO move is not allowed, set a timeout to
re-attempt the move.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Re-factor the code, so channel optimizations would be also triggered
upon the following changes: channel updates from the kernel,
disallow_freq interface, etc.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Update hardware features for all interfaces inside the loop, don't treat
the calling wpa_s instance specially. Perform the P2P channel list
updates after the hardware features are updated. This will prevent P2P
from relying on stale information.
Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com>
Not all paths in wpas_p2p_init_go_params() verified that the candidate
frequency can be used for GO purposes. Fix this, and in addition
re-factor the code to put better emphasis on the frequency selection
priorities.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Remove the code that considers removing GOs from their current
channel due to frequency interference, as this is already handled
as part of the P2P channels update.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
A P2P GO interface that was instantiated after a GO Negotiation or
Invitation holds the intersection of frequencies between the GO and the
client. In case the GO is going to move to another frequency, allow it
to move only to a frequency that is also supported by the client.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Add a function to compute the group common frequencies, and
use it to update the group_common_frequencies as part of the
channel switch flows.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Upon any change in the currently used channels evaluate if a GO should
move to a different operating frequency, where the possible scenarios:
1. The frequency that the GO is currently using is no longer valid,
due to regulatory reasons, and thus the GO must be moved to some
other frequency.
2. Due to Multi Concurrent Channel (MCC) policy considerations, it would
be preferable, based on configuration settings, to prefer Same
Channel Mode (SCM) over concurrent operation in multiple channels.
The supported policies:
- prefer SCM: prefer moving the GO to a frequency used by some other
interface.
- prefer SCM if Peer supports: prefer moving the GO to a frequency
used by some other station interface iff the other station
interface is using a frequency that is common between the local and
the peer device (based on the GO Negotiation/Invitation signaling).
- Stay on the current frequency.
Currently, the GO transition to another frequency is handled by a
complete tear down and re-setup of the GO. Still need to add CSA flow to
the considerations.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Add ieee80211_freq_to_channel_ext() conversion function into
ieee802_11_common.c. This function converts freq to channel and
additionally computes operating class, based on provided HT and VHT
parameters.
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Set peer_group_removed only if peer_group_added has already been set.
This fixes an issue where a propertiesChanged event triggered by an
earlier test case was able to get dbus_p2p_group_termination_by_go
terminated too early. This happened, e.g., with sequence
"dbus_p2p_two_groups dbus_p2p_group_termination_by_go".
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Only run peerJoined() steps once to avoid trying to use GetAll() on an
already removed group and double-removal of a group. This did not make
the test case fail, but the exception is printed out in pretty confusing
way to stdout, so better get rid of it.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Remove the duplicated -ddKt command line argument to avoid setting
hostapd debug level to EXCESSIVE.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Add support to read/configure log_level using hostapd control interface
LOG_LEVEL command similarly to what was already supported in
wpa_supplicant.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This makes it possible to use these helper functions from hostapd as
well as the current use in wpa_supplicant.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This buffer is owned by the FST module, so mark it const in the
set_ies() callback to make it clearer which component is responsible for
modifying and freeing this.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
This fixes an issue where freed MB IEs buffer memory could potentially
have been accessed after an interface is detached from FST group.
Without this fix, if an interface is detached from FST group, it can use
MB IEs buffer previously set by fst_iface_set_ies(), although the buffer
was released by fst_iface_delete().
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
NO_EVENTS parameter was added to STATUS command by commit
a6ab82d7b4 ('Android: Add NO_EVENTS
parameter to status command'). This patch adds handling of the new
parameter in wpa_cli so that "status no_events" can be used to specify
this parameter.
Signed-off-by: Daichi Ueura <daichi.ueura@sonymobile.com>
Currently wpa_cli connects to global control interface if -i/-p
parameters are not specified. wpa_cli on global control interface
is not useful since the prefix like "IFNAME=wlan0 " needs to be
added to some commands like "IFNAME=wlan0 scan". And, specifying
-i/-p parameters every time is annoying. To improve efficiency of
debugging, this patch enables to make wpa_cli work without extra
parameters.
If you still want to connect to global control interface,
the following command can be used instead:
$ wpa_cli -g@android:wpa_wlan0 (or -gwlan0)
Signed-off-by: Daichi Ueura <daichi.ueura@sonymobile.com>
The mesh SAE auth often fails with master branch. By bisect I found
commit eb5fee0bf5 ('SAE: Add side-channel
protection to PWE derivation with ECC') causes this issue. This does not
mean the commit has a bug. This is just a CPU resource issue.
After the commit, sae_derive_pwe_ecc() spends 101(msec) on my PC (Intel
Atom N270 1.6GHz). But dot11RSNASAERetransPeriod is 40(msec). So
auth_sae_retransmit_timer() is always called and it can causes
continuous frame exchanges. Before the commit, it was 23(msec).
On the IEEE 802.11 spec, the default value of dot11RSNASAERetransPeriod
is defined as 40(msec). But it looks short because generally mesh
functionality will be used on low spec devices. Indeed Raspberry Pi B+
(ARM ARM1176JZF-S 700MHz) requires 287(msec) for new
sae_derive_pwe_ecc().
So this patch makes the default to 1000(msec) and makes it configurable.
This issue does not occur on infrastructure SAE because the
dot11RSNASAERetransPeriod is not used on it.
Signed-off-by: Masashi Honma <masashi.honma@gmail.com>
On recent kernels, it seems that something changed (scheduler?)
that makes hwsim send the scan done event so quickly that iw isn't
scheduled back in to listen for it, causing iw to get stuck.
Work around this by using the scan trigger command (it'll be quick
enough so that we don't really need to wait) and the scan trigger
and dump commands where the results are required (and use a small
sleep there instead of waiting for the scan results.)
I'll try to fix this separately in iw later.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
This verifies P2P extended listen timing operations by confirming that a
peer is not discoverable during the provisioning step and that the peer
becomes discoverable after having removed the group during such
provisioning step. The latter case was broken until the 'P2P: Cancel
group formation when deleting a group during group formation' commit.
Signed-off-by: Jouni Malinen <j@w1.fi>
Otherwise P2P remains in provisioning state and continues to skip
extended listening forever.
Signed-off-by: Michael Olbrich <m.olbrich@pengutronix.de>
Allow proper update for listen_reg_class and listen_channel with
changed_parameters [CFG_CHANGED_P2P_LISTEN_CHANNEL] configuration for
command received through ctrl_interface. Without this, "set
p2p_listen_channel" and "set p2p_listen_reg_class" do not update the
listen channel. The D-Bus version was already setting these flags.
Signed-off-by: Purushottam Kushwaha <p.kushwaha@samsung.com>
Verify that Groups list for a P2P Peer gets updated properly on group
addition and removal (three different paths).
Signed-off-by: Jouni Malinen <j@w1.fi>
This variable is needed to figure out whether a wpa_supplicant interface
is for a P2P group that is (or was) connected to a specific GO. The
previous implementation was able to find such a case only when there was
an association with the GO. However, this may be needed even if there is
a temporary disconnection from the GO. Keep the GO device address
information over such temporary disconnections and only remove it on
group termination. This fixes an issue with D-Bus Peer PropertiesChanged
signals for the Groups property in case a P2P group gets removed due to
group idle timeout instead of explicit group termination command (local
request) or GO notification.
Signed-off-by: Jouni Malinen <j@w1.fi>
The P2P configuration file is wrongly set as STA configuration file,
even though a separate configuration file is mentioned with '-m' option.
Add initialization and deallocation of global.params->conf_p2p_dev to
fix this.
Signed-off-by: Gautam <gautams@broadcom.com>
The new CONFIG_NO_RC4=y build option can be used to remove all internal
hostapd and wpa_supplicant uses of RC4. It should be noted that external
uses (e.g., within a TLS library) do not get disabled when doing this.
This removes capability of supporting WPA/TKIP, dynamic WEP keys with
IEEE 802.1X, WEP shared key authentication, and MSCHAPv2 password
changes.
Signed-off-by: Jouni Malinen <j@w1.fi>
FIPS_mode_set(1) cannot be called multiple times which could happen in
some dynamic interface cases. Avoid this by enabling FIPS mode only
once. There is no code in wpa_supplicant to disable FIPS mode, so once
it is enabled, it will remain enabled.
Signed-off-by: Jouni Malinen <j@w1.fi>
omac1_aes_128() implementation within crypto_openssl.c is used in this
case and that cannot fail the memory allocation similarly to the
non-FIPS case and aes-omac1.c.
Signed-off-by: Jouni Malinen <j@w1.fi>
OpenSSL rejects the cipher string 'EXPORT' in FIPS mode in a way that
results in the locally generated error showing up before the EAP method
has been accepted.
Signed-off-by: Jouni Malinen <j@w1.fi>
MD4 is not allowed in such builds, so comment out md4_vector() from the
build to force compile time failures for cases that cannot be supported
instead of failing the MD¤ operations at runtime. This makes it easier
to detect and fix accidental cases where MD4 could still be used in some
older protocols.
Signed-off-by: Jouni Malinen <j@w1.fi>
In addition, replace some of the CHAP cases with PAP since that enables
more coverage without breaking the main test focus.
Signed-off-by: Jouni Malinen <j@w1.fi>
The PKCS12 file with default openssl options cannot be used with OpenSSL
1.0.1 in FIPS mode. Replace this with -descert version as a workaround.
Signed-off-by: Jouni Malinen <j@w1.fi>
Commit 94f1fe6f63 ('Remove master key
extraction from tls_connection_get_keys()') left only fetching of
server/client random, but did not rename the function and structure to
minimize code changes. The only name is quite confusing, so rename this
through the repository to match the new purpose.
Signed-off-by: Jouni Malinen <j@w1.fi>
tls_connection_get_keys() used to return TLS master secret, but that
part was removed in commit 94f1fe6f63
('Remove master key extraction from tls_connection_get_keys()'). Since
then, there is no real need for preventing this function from being used
in FIPS mode.
Signed-off-by: Jouni Malinen <j@w1.fi>
The bytes pointer was not reset back to the beginning of the buffer when
mixing in additional entropy from the crypto module. This resulted in
writing beyond the return buffer and not getting the required mixing of
the extra entropy for the actual return buffer.
Signed-off-by: Jouni Malinen <j@w1.fi>
It looks like the compiler version used in Android 5.0 warns about
potentially uninitialized oper_freq variable in these debug messages.
That is not really valid since this code path can be reached only if
found != 0 and in such a case, oper_freq is set. Anyway, it seems better
to avoid compiler warnings, so add an unnecessary initialization for
oper_freq for now.
Signed-off-by: Jouni Malinen <j@w1.fi>
MD5 is not allowed in such builds, so comment out md5_vector() from the
build to force compile time failures for cases that cannot be supported
instead of failing the MD5 operations at runtime. This makes it easier
to detect and fix accidental cases where MD5 could still be used in some
older protocols.
Signed-off-by: Jouni Malinen <j@w1.fi>
