* Move setting of tx retry limits on a separate function
 (we 'll clean up this AR5210 mess later)
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * On turbo mode increase PHY settling times, note that
 we only increase switch settling time on AR5212 as indicated
 by initvals.

 * A few cleanups: Move frame control settings for AR5210 from
 reset_tx_queue to tweak_initvals and remove phy_scal settings
 from tweak_initvals (we tweak them alread on set_sleep_clock).
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Now that we properly set rx/tx latencies for AR5311 remove
 that old buggy part of code left inside ath5k_hw_tweak_initval_settings
 that was never executed (you can't have an RF5112 radio on a mac older
 than AR5212). Also use a magic value for 5311 PHY_SCAL value.
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Handle all usec parameters in one function. It's much cleaner
 this way.
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Prepare for half/quarter/turbo support, introduce a new
 ah_bwmode parameter and get rid of ah_turbo. Bwmode stands
 for "bandwidth mode" and can have 4 values, default (20MHz),
 turbo (40MHz), half rate (10MHz), and quarter rate (5MHz).
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Since we only use ath5k_hw_stop_tx_dma to stop the beacon
 queue, introduce a new function ath5k_hw_stop_beacon_queue so
 that we can use that instead and have better control. In the future
 we can add more beacon queue specific stuff there (maybe tweak
 beacon timers or something), for now just call ath5k_hw_stop_tx_dma.

 * Also since we don't call ath5k_hw_stop_rx/tx_dma from outside
 dma.c, make them static.
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Make sure we are not trying to set RXDP while RX is active,
 for now ignore the return value.
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * DCU early termination should be used to quickly flush QCU
 according to docs so don't enable it for all queues, enable
 it only when stopping each queue and disable it when we are
 done.
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Increase timeouts on ath5k_hw_stop_tx_dma and also wait for
 tx queue to stop before checking for pending frames

 * Add a new debug level to debug dma start/stop
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Since we stop rx/tx dma and pcu durring reset there is no need to
 call ath5k_hw_stop_rx/tx_dma before, also there is no need to call
 them durring stop_locked since we can use ath5k_hw_dma_stop for
 both.
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Stop PCU receive logic (DRU) durring reset
 We need to be sure pcu is not active when trying to stop rx dma
 right now this is done on ath5k_reset (base.c) but later we are
 going to clean it up.
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * Add a new function to stop rx/tx dma and use in when reset starts
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

 * No functional changes

 * Clean up reset:
 Introduce init functions for each unit and call them instead
 of having everything inside ath5k_hw_reset (it's just c/p for
 now so nothing changes except calling order -I tested it with
 various cards and it's ok-)

 * Further cleanups:
 ofdm_timings belongs to phy.c
 rate_duration belongs to pcu.c
 clock functions are general and belong to reset.c (more to follow)

 * Reorder functions for better organization:
 We start with helpers and other functions follow in categories,
 init functions are last
Signed-off-by: NNick Kossifidis <mickflemm@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This feature is to mitigate the problem of certain 3
stream chips that exceed the PCIe power requirements.An EEPROM flag
controls which chips have APM enabled which is basically read from
miscellaneous configuration element of the EEPROM header.

This workaround will reduce power consumption by using 2 Tx chains for
Single and Double stream rates (5 GHz only).All self generated frames
(regardless of rate) are sent on 2 chains when this feature is
enabled(Chip Limitation).

Cc: Paul Shaw <paul.shaw@atheros.com>
Signed-off-by: NMohammed Shafi Shajakhan <mshajakhan@atheros.com>
Tested-by: NMohammed Shafi Shajakhan <mshajakhan@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The RX aggregation locking documentation was
wrong, which led Christian to also code the
timer timeout handling for it somewhat wrongly.

Fix the documentation, the two places that
need to hold the reorder lock across accesses
to the structure, and the debugfs code that
should just use RCU.

Also, remove acquiring the sta->lock across
reorder timeouts since it isn't necessary, and
change a few places to GFP_KERNEL because the
code path here doesn't need atomic allocations
as I noticed when reviewing all this.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Acked-by: NChristian Lamparter <chunkeey@googlemail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The recent tx path cleanups moved the software retry count tracking
from the ath_buf to the skb cb, however the actual counter update
referred to the wrong location, confusing block-ack window tracking.
Fix this by using the retries counter in the struct ath_frame_info.
Signed-off-by: NFelix Fietkau <nbd@openwrt.org>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This reverts commit 446fad5a.

The change had broken the packet injection on monitoring mode.
Signed-off-by: NRajkumar Manoharan <rmanoharan@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Signed-off-by: NVasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

It is not ideal to enable btcoex based on subsys id as it is
not unique, they are so random. It is also a pain keeping
all of them in a table to enable btcoex for a particular
hw. Going back to the old idea.
Signed-off-by: NVasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This implements the new off-channel TX API
in mac80211 with a new work item type. The
operation doesn't add a new work item when
we're on the right channel and there's no
wait time so that for example p2p probe
responses will be transmitted without delay.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

With p2p, it is sometimes necessary to transmit
a frame (typically an action frame) on another
channel than the current channel. Enable this
through the CMD_FRAME API, and allow it to wait
for a response. A new command allows that wait
to be aborted.

However, allow userspace to specify whether or
not it wants to allow off-channel TX, it may
actually want to use the same channel only.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Minor changes in LUT
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

use different kill_ack_mask and kill_cts_mask when detect SCO
is active.
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Change to offical name for 100 devices:
   "Intel(R) Centrino(R) Wireless-N 100"

Change to offical name for 130 devices:
   "Intel(R) Centrino(R) Wireless-N 130"
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Change to offical name for 6000g2b devices:
   "Intel(R) Centrino(R) Wireless-N 1030"
   "Intel(R) Centrino(R) Advanced-N 6230"
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Change to offical name for 6000g2a devices:
   "Intel(R) Centrino(R) Advanced-N 6205"
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Change to offical name for 6050g2 devices:
   "Intel(R) Centrino(R) Wireless-N 6150"
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Instead of hardcoding the numbers that must
match mac80211, use the constants. Not that
this means we could change the constants,
but at least this way it's clearer.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Our hardware has reliable TX status, but we're
not currently advertising that to mac80211.
Since the packet loss monitoring will depend
on it, advertise it.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Since aggregation queues are station-specific, the
device will not reject packets in them but rather
will stop the appropriate aggregation queues when
a station goes to sleep. I forgot to account for
this in the driver, so if a station went to sleep
that had aggregation enabled, traffic would stop
indefinitely.

Fix this by only accounting frames queued on the
normal AC queues for associated station.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Through races, a packet may be enqueued for
transmission to a station while that station
is going to sleep, in which case the warning
here triggers. Instead of warning, check the
condition -- if this packet is not a PS-poll
response then we still enqueue it but it will
be rejected by the device since the station
is marked as asleep.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

The valid tx/rx antenna information is part of EEPROM, so use it
to configure the device.

For few cases, the EEPROM did not reflect the correct antenna, but
it is too late to modify the EEPROM, so overwrite with .cfg parameters
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

was not used, remove it
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

It is no longer used anywhere
Signed-off-by: NFelix Fietkau <nbd@openwrt.org>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Categorize AR7010 & AR9287 devices based on driver_info
of usb_device_id, instead of PIDs. This avoids per-device cases
and minimize code changes for new device addition.
Signed-off-by: NRajkumar Manoharan <rmanoharan@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

AR9287 devices (PCI/USB) use different eeprom start location
to read nvram. New devices might endup with same devid. So use
driver_info to set offset, instead of devid. driver_info is
valid for HTC devices alone which is filled in usb_device_id.
Signed-off-by: NRajkumar Manoharan <rmanoharan@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Added driver_info to identify AR7010, R9287 HTC devices.
Signed-off-by: NRajkumar Manoharan <rmanoharan@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The driver_info stores the device category information which
is used to load appropriate device firmware, select firmware offset
and eeprom starting location. The driver_info is accessed across
ath9k_htc and ath9k_hw. Hence placed under common structure.
Signed-off-by: NRajkumar Manoharan <rmanoharan@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

For 6000g2b and up, adding advance power management support
for better power consumption
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

If shadow register is enable, modify the power management
command to inform uCode
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>