An INTERFACE_DISABLED event received on an interface that is
currently operating a P2P GO means that the group session ended.
In such a case, if the interface was dynamically added remove it, and
if not, remove all the network blocks that are temporary, assuming
that if needed a new session will be started by an external entity.
The use case was triggering rfkill (both SW and HW). This case popped up
as part of a testing cycle, where after a toggle in the rfkill state,
the result was that the interface was not deleted, but on the other hand
the wpa_supplicant did not configure the kernel to re-start the AP
functionality again.
Signed-hostap: Ilan Peer <ilan.peer@intel.com>
On scan results event if a concurrent P2P scan was triggered previously,
scan results processing is canceled, p2p_find executed, and a new sta
scan is triggered (pending scan). However, this new sta scan does not
restore the scan_req value of the previous scan (whose scan result has
been canceled).
If we are currently connected to an AP and use ap_scan=2, the new
triggered scan will cause an associtation-without-scan in
wpa_supplicant_scan:
(ap_scan == 2 & scan_req != MANUAL_SCAN_REQ)
=> wpa_supplicant_assoc_try()
causing an association error and a disconnection.
This patch fixes this issue by restoring the previous scan_req value.
Signed-off-by: Loic Poulain <loicx.poulain@intel.com>
This is a mechanism used in Android to extend driver interface in vendor
specific ways. This is included only for the purpose of Android
compatibility. Proper interface commands should be used for any new
functionality.
Signed-hostap: Jouni Malinen <j@w1.fi>
Relative time shouldn't be calculated based on gettimeofday
because that clock can jump (e.g., when the time is adjusted
by the system administrator.)
On systems where that is available, use CLOCK_BOOTTIME (on
fairly recent Linux systems, this clock takes into account
the time spend suspended) or CLOCK_MONOTONIC (on Linux and
some POSIX systems, this clock is just freely running with
no adjustments.)
Reported-by: Holger Schurig <holgerschurig@gmail.com>
Signed-hostap: Johannes Berg <johannes.berg@intel.com>
Scan initiated from wps_nfc command context was ketp on
getting rescheduled due to an on-going scheduled scan. So
cancel sched_scan before issuing a reassociation scan.
Signed-hostap: Jithu Jance <jithu@broadcom.com>
The shorter 250 ms wait for the next scan request can be used also for
the case of persistent group re-invocation instead of just formation of
a new group. This speeds up the process and makes this more robust
especially in cases where the GO is using MCC.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
The QoS Map Set element was passed in full to the driver instead of just
the payload of the element. This resulted in the updated QoS Map being
rejected. Validate the element id/len and send only the payload to the
driver.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
Only force_freq was used in the wpas_p2p_set_own_freq_preference() call
which allowed the P2P module channel re-selection to ignore the
preference for using a channel we are already using. Fix this by setting
either force_freq or pref_freq as the preference based on which one is
set. This allows p2p_ignore_shared_freq parameter to be used whether to
prefer the shared frequency in this case.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
There is no need to use wpas_p2p_num_unused_channels() here in the
default configuration of p2p_ignore_sahred_freq=0, so re-order the
conditions to skip that operation. This is a bit more efficient and the
debug log is also a bit cleaner in the default case.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
p2p_ignore_shared_freq=1 was supposed to allow a MCC-capable device to
ignore a preference for using the same channel on multiple interfaces.
However, it was not used when inviting a peer to re-invoke a persistent
group. This case needs special handling since the peer's channel list is
not available to perform channel reselection and the operating channel
indicated in the Invitation Request frames ends up getting used as the
operating channel if the transmitted of that frames becomes the GO.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
p2p_ignore_shared_freq=1 was supposed to allow a MCC-capable device to
ignore a preference for using the same channel on multiple interfaces.
However, it was not used during processing of an Invitation Request. Fix
that case to use channel preference instead of channel forcing if free
channels are available. This allows p2p_ignore_shared_freq=1 case to
ignore the preference.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
It is confusing to talk about current operating channels being
unavailable for P2P when there are no current operating channels. Make
the debug message easier to understand.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
It is clearer if there is only a single loop of the channel list and
shared debug prints. In addition, the note about current operating
channels not being available is quite confusing if there are no
operating group, so make that part of the message conditional.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
Commit 0d08efa447 modified
wpas_p2p_setup_freqs() design to use number of MCC channels capability
from the driver. However, it resulted in regression on how the preferred
vs. forced channel selection is done in the case of a MCC device.
force_freq was set unconditionally even though this was supposed to be
done only if no additional channels are available. pref_freq needs to be
used when possible to avoid preventing connection.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
While starting from PNO start context, the scheduled scan was not
setting the flag wpa_s->scanning. This was resulting in the subsequent
SCAN command to proceed further and send command to nl80211/cfg80211.
The expected behavior of cancelling sched_scan was not happening here.
While sched_scan is in progress and a legacy scan comes on the
cli/socket, the sched_scan is cancelled and normal scan is allowed to
continue. However, sometimes sched_scan cancelled event comes a bit
delayed and we will send out the scan command before the wpa_s->scanning
is cleared. Instead, reschedule the incoming scan req if the
wpa_s->scanning shows that it is still in progress.
Signed-hostap: Jithu Jance <jithu@broadcom.com>
Add chan_switch to the control interface of wpa_supplicant and hostapd,
and also to wpa_cli and hostapd_cli.
Signed-hostap: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Build CSA settings and call the driver to perform the switch. Construct
Beacon, Probe Response, and (Re)Association Response frames both for CSA
period and for the new channel. These frames are built based on the
current configuration. Add CSA IE in Beacon and Probe Response frames.
Signed-hostap: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Add csa_settings struct which holds parameters for CSA. Change driver
interface for switch_channel(), so that it will receive this struct and
not only the new frequency as it was before. This allows wpa_supplicant
to provide all the required parameters (beacons, proberesp, assocresp,
CSA IE) which are required by cfg80211 implementation.
Signed-hostap: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
This was supposed to be a minimal sample of eloop wrapper, but it is
unclear whether this is of that much use and the file has not been kept
up-to-date. Remove this file to reduce maintenance effort. The other
eloop*.c files can be used as a starting point if something new is
needed.
Signed-hostap: Jouni Malinen <j@w1.fi>
This triggers re-transmission of CTRL-EVENT-STATE-CHANGE and
CTRL-EVENT-CONNECTED events on STATUS command for Android framework
specific processing.
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
Commit eb32460029 left an unneeded
sim_type argument to scard_init(). Remove that unnecessary argument to
clean up the implementation.
Signed-hostap: Masashi Honma <masashi.honma@gmail.com>
Add DBus methods for TDLS operations similar to those available
for the control interface. This includes Discover, Setup, and
Teardown commands. While here, add a method to query the TDLS
link status and add a DBus method for it.
Tested with CONFIG_TDLS enabled, on a TDLS-enabled host and
peer capable of TDLS:
dbus-send --system --dest=fi.w1.wpa_supplicant1 --print-reply \
/fi/w1/wpa_supplicant1/Interfaces/0 \
fi.w1.wpa_supplicant1.Interface.TDLSStatus string:<peer-mac-address>
yields: string "peer does not exist"
dbus-send --system --dest=fi.w1.wpa_supplicant1 --print-reply \
/fi/w1/wpa_supplicant1/Interfaces/0 \
fi.w1.wpa_supplicant1.Interface.TDLSDiscover string:<peer-mac-address>
yields no error
dbus-send --system --dest=fi.w1.wpa_supplicant1 --print-reply \
/fi/w1/wpa_supplicant1/Interfaces/0 \
fi.w1.wpa_supplicant1.Interface.TDLSSetup string:<peer-mac-address>
yields no error
dbus-send --system --dest=fi.w1.wpa_supplicant1 --print-reply \
/fi/w1/wpa_supplicant1/Interfaces/0 \
fi.w1.wpa_supplicant1.Interface.TDLSStatus string:<peer-mac-address>
yields: string "connected" after TDLS completes
dbus-send --system --dest=fi.w1.wpa_supplicant1 --print-reply \
/fi/w1/wpa_supplicant1/Interfaces/0 \
fi.w1.wpa_supplicant1.Interface.TDLSTeardown string:<peer-mac-address>
yields no error
dbus-send --system --dest=fi.w1.wpa_supplicant1 --print-reply \
/fi/w1/wpa_supplicant1/Interfaces/0 \
fi.w1.wpa_supplicant1.Interface.TDLSStatus string:<peer-mac-address>
yields: string "peer not connected"
Signed-hostap: Paul Stewart <pstew@chromium.org>
Request new scan only for the interface for which the original scan
request and results has come. Otherwise while sharing scan results along
with P2P interfaces, the new scan will be requested on P2P interfaces.
Signed-hostap: Jithu Jance <jithu@broadcom.com>
Since eloop_deplete_timeout() is doing practically same in
wpa_supplicant_req_scan(), revert the old mechanism from commit
7e1488494e to avoid unnecessary work in
this function.
Signed-hostap: Jouni Malinen <j@w1.fi>
During persistent group re-invocation, GO may end up using a different
channel as the operation channel compared to what was indicated in the
invitation frames. This may break the connection if the peer device ends
up scanning the GO only on the channel from the invitation frame. Fix
this by using the negotiated channel (if available) on the GO as the
operating channel instead of the channel that was provided in the
p2p_invite command to start negotiation.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
There are use cases requesting the host driver to initiate the TDLS
setup with the peer only when configured by the external applications.
Thus, enable this control by tdls_external_control=1 and pass the
requisite information for the specific TDLS operation to the driver
on the request from such use cases.
This operation mode expects the driver to initiate TDLS link
automatically based on signal strength and traffic to a peer and tear
down links whenever they are not used or suitable due to signal strength
etc. The list of peers with which such operations are to be performed
are provided with the TDLS driver operations.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
This is a better way of matching P2P groups based on the unique P2P
Device Address (e.g., from P2P Group ID) and SSID pair instead of using
the not necessarily unique P2P Interface Address.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
There is no need to parse the IE buffer again to find the SSID of the
BSS since that information is already stored in struct wpa_bss.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
If a BSS is disallowed temporarily with disallow_aps, the network
connection is going to fail. As such, there is not much point in
allowing Interworking network selection to try to connect with such BSS.
As such, do not consider disallowed networks for automatic network
selection and reject requests to connect to them through
INTERWORKING_CONNECT.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
Do not add multiple network blocks for the same network from a single
credential. INTERWORKING_CONNECT used to generate a new network block
for each instance regardless of what network blocks have already been
configured. While this allows the connection to go through, it is not
efficient to leave behind potentially large number of network blocks
with the same contents (or worse, changed contents). Address this by
removing an older network block for the same credential before adding a
new one.
Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
