The CAB (Content after Beacon) queue is used for beacon-triggered
transmission of buffered multicast frames. If lots of multicast frames
were buffered and this queue fills up, it drowns out all regular
traffic. To limit the damage that buffered traffic can do, try to limit
the queued data to becaon_interval / 8.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This adds support for PS-Poll and U-APSD driver-buffered frames (part of
an aggregation session).
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Use the UAPSD hardware queue to get PS-Poll responses out as fast as
possible and without backoff.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Waiting for all subframes of an existing aggregation session to drain
before allowing mac80211 to start a new one is fragile and deadlocks
caused by this behavior have been observed.
Since mac80211 has proper synchronization for aggregation session
start/stop handling, a better approach to session handling is to simply
allow mac80211 to start a new session at any time. This requires
changing the code to discard any packets outside of the BlockAck window
in the A-MPDU software retry code.
This patch implements the above and also simplifies the code.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
When aggregation stop is requested, don't run the mac80211 aggregation
stop callback yet, while the session is still blocked.
Also, when aggregation flush is requested, don't run the callback at all.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
In some cases it can be useful to change the MAC address of a virtual
interface to something that's completely different from the EEPROM
stored MAC address. In this case it is a bad idea to use the EEPROM MAC
address for calculating the BSSID mask, as that would make it too wide.
In one case a few devices have been observed to send ACKs for many
packets on the channel not directed at them, which results in a neat
Denial of Service attack on the channel.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
As the comment in ath_get_next_rx_buf indicates, if a descriptor with
the done bit set follows one with the done bit cleared, both descriptors
should be discarded, however the driver is not doing that yet.
To fix this, use the rs->rs_more flag as an indicator that the following
frame should be discarded. This also helps with the split buffer case:
if the first part of the frame is discarded, the following parts need to
be discarded as well, since they contain no valid header or usable data.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
On joining an existing IBSS network, beaconing has to start
only after a TSF sync has happened by receiving a beacon from
the BSS. In creator mode, beaconing can start immediately after
a HW reset has been done.
Now that mac80211 notifies the driver of the mode type (creator/joiner)
via ieee80211_bss_conf->ibss_creator, make use of it to properly setup
the HW beacon timers.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The ath9k hardware reports whenever an frame was part
of an A-MPDU. MAC80211 already provides the necessary
API to pass this additional information along to
whomever needs it.
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The sample data received through the spectral scan can be either in big or
little endian byteorder. This information isn't stored in the output file.
Therefore it is not possible for the analyzer software to find the correct byte
order.
It is relative common to get the data from a low end AP in big endian mode and
transfer it to another computer in little endian mode to analyze it. Therefore,
it would be better to store it in network (big endian) byte order.
The extension of the 8 bit bins for each bin to 16 bit is not necessary. This
operation can be done in userspace or on a different machine. Instead the
max_exp defining the amount of shifting required for each bin is exported to
userspace.
The change of the output format requires a change of the type in the sample
tlv to allow the userspace program to correctly detect the bin format.
Reported-by: Zefir Kurtisi <zefir.kurtisi@neratec.com>
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
[siwu@hrz.tu-chemnitz.de: squashed patches, update commit message, rebase, fix endianess bug]
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Export the various parameters to userspace.
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dropping packets from aggregation sessions is usually not a good idea, as
it might upset the synchronization of the BlockAck receive window of the
remote node. The use of the retry_tx parameter to reset/tx-drain functions
also seemed a bit arbitrary.
This patch removes this parameter altogether and ensures that pending tx
frames are not dropped for no good reason.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The commit "ath9k: fix rx flush handling" added a deadlock that happens
because ath_rx_tasklet is called in a section that has already taken the
rx buffer lock.
It seems that the only purpose of the rxbuflock was a band-aid fix to the
reset vs rx tasklet race, which has been properly fixed in the commit
"ath9k: add a better fix for the rx tasklet vs rx flush race".
Now that the fix is in, we can safely remove the lock to avoid such issues.
Cc: stable@vger.kernel.org
Reported-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Right now the rx flush is not doing anything useful on AR9003+, as it only
works if the buffers in the rx FIFO have not been purged yet, as is done
by ath_stoprecv.
To fix this, always call ath_flushrecv from within ath_stoprecv before
the FIFO is emptied, but still after the hw receive path has been stopped.
This ensures that frames received (and ACKed by the hardware) shortly before
a reset will be seen by the software, which should improve A-MPDU session
stability.
Cc: stable@vger.kernel.org
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Ensure that the rx tasklet is no longer running when entering the reset path.
Also remove the distinction between flush and no-flush frame processing.
If a frame has been received and ACKed by the hardware, the stack needs to see
it, so that the BA receive window does not go out of sync.
Cc: stable@vger.kernel.org
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Adds the spectral scan feature for ath9k. AR92xx and AR93xx chips
are supported for now. The spectral scan is triggered by configuring
a mode through a debugfs control file. Samples can be gathered via
another relay debugfs file.
Essentially, to try it out:
echo chanscan > /sys/kernel/debug/ieee80211/phy0/ath9k/spectral_scan_ctl
iw dev wlan0 scan
cat /sys/kernel/debug/ieee80211/phy0/ath9k/spectral_scan0 > samples
echo disable > /sys/kernel/debug/ieee80211/phy0/ath9k/spectral_scan_ctl
This feature is still experimental.
The special "chanscan" mode is used to perform spectral scan while
mac80211 is scanning for channels. To allow this,
sw_scan_start/complete() ops have been added.
The patch contains code snippets and information from Zefir Kurtisi and
information provided by Adrian Chadd and Felix Fietkau.
Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Acked-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The 'stations' debugfs file has multiple issues. It doesn't scale
to an arbitrary number of associated stations and allocating
64K is not elegant either. Now that changes have been made in
mac80211 to support dynamic creation/deletion of driver-specific
debugfs files on station addition/removal, remove this file and
make use of the mac80211 hooks (which will be done in a sebsequent
patch).
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The 'xmit' debugfs file has become big and unwieldy, fix
multiple issues with its usage:
* Store TX counters/statistics only for the 4 Access Categories.
Use IEEE80211_NUM_ACS instead of ATH9K_NUM_TX_QUEUES.
* Move various utility macros to debug.h, they can be reused
elsewhere.
* Remove tx_complete_poll_work_seen.
* Remove code that accesses various internal queue-specific
variables without any locking whatsoever. HW/SW queue details
will be handled in a subsequent patch.
* Do not print internal values like txq_headidx and txq_headidx.
They were mostly unused anyway, considering code like:
PRX("txq_tailidx: ", txq_headidx);
* Handle 'txprocdesc' for EDMA too.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Use the macros provided by mac80211 and remove redundant
declarations inside the drivers.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
On lower WLAN signal strength, WLAN downlink traffic might suffer
from retransmissions. At the mean time, playing SCO/A2DP profiles
is affecting WLAN stability. In such scenario, by stomping SCO/A2DP
BT traffic completely for a BTCOEX period, gives WLAN traffic an
oppertunity to recover PHY rate. It also improves WLAN stability at
lower RSSI without sacificing BT traffic.
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The debugfs file for dumping btcoex parameters unconditionally
assumes a MCI-based device. This will not work for older btcoex
chips. Fix this by branching out the routine into separate
functions.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The simulataneous transmission of both WLAN and BT might cause
increase in power levels. To avoid regulatory violation, WLAN tx
power will be adjusted according to BT power index based on avaliability
of BT scheduling messages. WLAN tx power reduction might affect its
performance. So WLAN tx power is only be lowered when the signal strength
is good enough. Otherwise concurrent tx will be disabled and WLAN uses
it default power levels. Also concurrent tx is disabled whenever WLAN is
moving to off-channel which might be used by BT.
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
As of now the ANI cycle is executed only when the chip is awake.
On idle state case, the station wakes up from network sleep for
beacon reception. Since most of the time, ANI cycle is not syncing
with beacon wakeup, ANI cycle is ignored. Approx 5 mins once, the
calibration is performed. This could affect the connection stability
when the station is idle for long. Even though the OFDM and CCK phy
error rates are too high, ANI is unable to tune its immunity level
as quick enough due to rare execution.
Here the experiment shows that OFDM and CCK levels are at default
even on higher phy error rate.
listenTime=44 OFDM:3 errs=121977/s CCK:2 errs=440818/s ofdm_turn=1
This change ensures that ANI calibration will be exectued atleast
once for every 10 seconds. The below result shows improvements and
immunity levels are adopted quick enough.
listenTime=557 OFDM:4 errs=752/s CCK:4 errs=125/s ofdm_turn=0
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Force chain 1 to be used for CCK rates since the target power
table stored in EEPROM is too high to transmit with both chains.
This is needed to avoid regulatory violation.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Ensure that led pin is filled and set to OFF before starting
the driver. With recent changes, drv_start is being called even
before led_init is being completed. This is causing led is always
OFF on driver load when the interface is UP. This patch splits the
led init and fills the led pin before register hw.
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Its more correct to convert btcoex_period to 'us' while
comparing with btcoex_no_stomp which is in 'us'.
Did not find any functionality issues being fixed,
as the generic hardware timer triggers are usually
refreshed with the newer duty cycle.
Cc: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Remove the control.sta pointer from ieee80211_tx_info to free up
sufficient space in the TX skb control buffer for the upcoming
Transmit Power Control (TPC).
Instead, the pointer is now on the stack in a new control struct
that is passed as a function parameter to the drivers' tx method.
Signed-off-by: Thomas Huehn <thomas@net.t-labs.tu-berlin.de>
Signed-off-by: Alina Friedrichsen <x-alina@gmx.net>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
[reworded commit message]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Using work_pending() to defer certain operations when
a HW-reset work has been queued is racy since the check
would return false when the work item is actually in
execution. Use SC_OP_HW_RESET instead to fix this race.
Also, unify the reset debug statistics maintenance.
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Currently, there are problems with how ANI is handled in
multi-VIF scenarios. This patch addresses them by unifying
the start/stop logic.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Remove unused variables, use a helper function to choose
the slot and reset beaconing status at one place.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Setup the beacon queue parameters after disabling
interrupts. Also, remove the redundant call in conf_tx()
for IBSS mode since the queue would be configured
with the appropriate cwmin/cwmax values when beaconing
is enabled.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
In the tx_last_beacon() callback, mac80211's beaconing
status can be used instead. The beacon tasklet doesn't require
it because it is disabled when removing a slot.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
* The beaconing status routine is not required, since in
multi-VIF cases the HW beacon parameters should not be
re-configured.
* Remove SC_OP_TSF_RESET - when a beaconing interface comes
up the first time, the TSF has to be reset.
* Simplify ath9k_allow_beacon_config().
* Handle setting/clearing the SWBA interrupt properly.
* Remove the TSF mangling in IBSS mode, it is not required.
* General code cleanup.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This patch revamps interface addition and deletion and simplifies
slot allocation. There is no need to setup the beacon buffer
in add/remove interface, remove this and use simple APIs for
assigning/deleting slots.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
If the aggregate size exceeds the TXOP limit, it leads to lots of unnecessary
hardware and software retries.
The previous 4ms frame limit table was completely undocumented, the commit
that updated it only vaguely referenced and equation from the standard,
but I've been unable to replicate its results.
Fix this by using a formula based on the code in ath_pkt_duration, which is
more likely to be correct for this case.
Reported-by: Dave Täht <dave.taht@gmail.com>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Prepare for using different queue size defaults for each AC.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
add suspend/resume/set_wakeup callbacks to the driver
*suspend
- bail out only if all the conditions for configuring WoW.
is fine, currently multivif case is not handled
- check for associated state.
- map wow triggers from user space data.
- add deauth/disassoc pattern and user defined pattern,
for the later a list is maintained.
- store the interrupt mask before suspend, enabled beacon
miss interrupt for WoW.
- configure WoW in the hardware by calling ath9k_hw_wow_enable.
*resume
- restore the interrupts based on the interrupt mask
stored before suspend.
- call ath9k_hw_wow_wakeup to configure/restore the hardware.
- after wow wakeup clear away WoW events and query the
WoW wakeup reason from the status register
*set_wakeup
- to call 'device_set_wakeup_enable' from cfg80211/mac80211
when wow is configured and as per Rafael/Johannnes the
right way to do so rather in the driver suspend/resume
call back
Cc: Senthil Balasubramanian <senthilb@qca.qualcomm.com>
Cc: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Cc: vadivel@qca.qualcomm.com
Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
*add structures, macros and variables for WoW, so that the driver
can make use of it.
*maintain a list for user enabled patterns and masks
*track pattern slots for the hardware limitation on the
maximum number of patterns that can be stored.
*track interrupts enabled before WoW suspend, so
that can be reconfigured after resume
*have macros to parse user defined wow configurations to
hardware code
Cc: Senthil Balasubramanian <senthilb@qca.qualcomm.com>
Cc: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Cc: vadivel@qca.qualcomm.com
Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com>
Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Felix Fietkau