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>

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>

Check the BT PSPoll flag when fill PM command to uCode
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Adding additional power management option available for the device.
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Bit 7 of BT config flag is used to enable/disable PSPoll sync.
Make the name to match it.
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Using static const generally increases object text and decreases data size.
It also generally decreases overall object size.

   text	   data	    bss	    dec	    hex	filename
  48644	     57	  12120	  60821	   ed95	drivers/net/wireless/b43/phy_n.o.new
  48661	     57	  12120	  60838	   eda6	drivers/net/wireless/b43/phy_n.o.old
  37906	     86	   7904	  45896	   b348	drivers/net/wireless/iwlwifi/iwl-agn-lib.o.new
  37937	     86	   7904	  45927	   b367	drivers/net/wireless/iwlwifi/iwl-agn-lib.o.old
  37781	    523	   6752	  45056	   b000	drivers/net/wireless/iwlwifi/iwl-3945.o.new
  37781	    523	   6752	  45056	   b000	drivers/net/wireless/iwlwifi/iwl-3945.o.old

Changed b43_nphy_write_clip_detection to take a const u16 *
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

If only 3945 is selected, and is a module, build
fails because iwl-legacy.c won't be compiled. Fix
this by adding it to the build correctly.

This doesn't happen for 4965 because it is a bool
option, not tristate, since it's built into the
AGN module.
Reported-by: NRandy Dunlap <randy.dunlap@oracle.com>
Tested-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Subtract of jiffies is fine even if one variable overwrap.
Signed-off-by: NStanislaw Gruszka <sgruszka@redhat.com>
Acked-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

We can simplify length calculation in iwlagn_tx_skb, that function
is enough complex, without fuzz it more than necessary.
Signed-off-by: NStanislaw Gruszka <sgruszka@redhat.com>
Acked-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Signed-off-by: NStanislaw Gruszka <sgruszka@redhat.com>
Acked-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Have the STATUS_READY bit set before commit_rxon call to avoid fail
to send tx power to uCode.
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

When setting rxon chain and filter, no need to commit RXON when the
chain flag is not changed. This reduces the number of RXON commands
we send down to uCode.
Signed-off-by: NShanyu Zhao <shanyu.zhao@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

EEPROM contain the SKU information for the device, use it.
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

RXON command without association bit can clear the QoS info in the
uCode. Therefore, before sending the associated RXON, we need to
send the QoS command just in case.
Signed-off-by: NShanyu Zhao <shanyu.zhao@intel.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

When the DTIM is not 1, then the slot timing is in
some cases required to be calclulated based on the
DTIM interval instead of the beacon interval, fix
that.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Instead of unconditionally sending unassoc RXON,
before any assoc RXON, re-send only the AP STA
entry which is required after the BSSID has been
programmed into the device to set up internal
filters in the microcode properly.

This fixes some issues that we correlated with
sending a lot of RXON commands to the device.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

For reasons that aren't entirely clear to me,
we sometimes get here during hardware reset
without the interface being set. Don't crash,
but keep a warning.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Currently, when a PAN queue needs to be
stopped, we erroneously stop queue number 5
(for example) with mac80211 -- which doesn't
even exist!

To avoid that problem, recalculate the swq_id
for all queues when setting up the queues,
and don't use the default identity mapping
that is acceptable for devices which don't
support PAN.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Previously, we used the swq_id's mechanism
to have AC and HW queue different only for
aggregation queues. To be able to fix a bug
with iPAN simply always build the swq_id as
ac | (hwq << 2) and remove the flag bit.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Instead of passing the txq->swq_id, pass the
txq struct directly to make sure that in the
future nobody will pass an invalid number.
Only three places actually change from using
the txq_id or the skb's queue_mapping to now
using txq->swq_id as well.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Even if this check were to happen, using the
txq_id here (which is a HW queue) would lead
to confusion in mac80211. Luckily, it doesn't
seem like this can ever happen.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Simply remove the unused variable swq_id.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Some variables are misnamed in the FIFO setup
code, fix that.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Set the default led mode for different devices.

For the newer devices such as 6000g2a, 6000g2b and newer,
the default led mode is On/Off instead of blinking.

The led_mode still can be control through module parameter
  0: system default
  1: On/Off
  2: blinking
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Change the default BT_KILL_ACK_MASK and BT_KILL_CTS_MASK for BT coex
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Use dynamic aggregation threshold if bt traffic load is high
to reduce the impact on aggregated frame.
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

Setting the max/min/def value for BT dynamic aggregation threshold.
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

For newer devices, uCode provide both "number of frames sent"
and "number of frames acked" information inside the compressed_ba
packet. So instead of figure the success/failure information through
the bitmap, use those information which is much betrer approach.
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>

This patch moves code out from wireless drivers where two different
functions are defined in three code locations for the same purpose and
provides a common function to sign extend a 32-bit value.
Signed-off-by: NAndreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

For 6000 series devices and up, enable automatic update MAC's register
for better power usage in PSP mode
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Disconnected antenna algorithm is used to find out which antennas are
disconnected. It should be disabled for devices that support advanced
bluetooth coexist.
Signed-off-by: NShanyu Zhao <shanyu.zhao@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Disconnected antenna algorithm is seperated into its own function from chain noise
calibration routine for better code management.
Signed-off-by: NShanyu Zhao <shanyu.zhao@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

When the HT information is changed due to
BSS changes (like legacy stations joining)
we need to recalculate HT RXON parameters.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

During the RXON rewrite, this code got lost.
When we've just associated, we need to enable
all calibrations and see if some were already
finished. Add back the missing code.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The RXON rework resulted in a massive loss of
throughput because we weren't programming the
device completely correctly -- the BSSID has
to be programmed into the device before the
AP station is uploaded. To fix this, simply
always send the unassoc RXON, i.e. even when
it was already unassoc so that the BSSID and
some other parameters are updated properly.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Garen noticed that this was wrong. Fix
the calibration -- default to multiple
chains and fall back to single where
possible.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

"mac80211: Fix WMM driver queue configuration"
inadvertedly broke iwlwifi, because now mac80211
configures the QoS settings before assoc, and
therefore before HT. Thus, iwlwifi no longer told
the device about the HT setting, which it needs
to -- and thus throughput went down a lot. Fix
this by resending the QoS command to the device
not only when QoS/WMM settings change, but also
when HT changes.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

In iwl_rx_queue_update_write_ptr function
replace iwl_write_direct32 with iwl_write32 when not in power save mode.
We don't have to go through grab nic access as the NIC is already awake.
Signed-off-by: NTomas Winkler <tomas.winkler@intel.com>
Signed-off-by: NWey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>