mac80211: simplify RX PN/IV handling

The current rx->queue value is slightly confusing.
It is set to 16 on non-QoS frames, including data,
and then used for sequence number and PN/IV checks.
Until recently, we had a TKIP IV checking bug that
had been introduced in 2008 to fix a seqno issue.
Before that, we always used TID 0 for checking the
PN or IV on non-QoS packets.

Go back to the old status for PN/IV checks using
the TID 0 counter for non-QoS by splitting up the
rx->queue value into "seqno_idx" and "security_idx"
in order to avoid confusion in the future. They
each have special rules on the value used for non-
QoS data frames.

Since the handling is now unified, also revert the
special TKIP handling from my patch
"mac80211: fix TKIP replay vulnerability".

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: LCN-PHY add place for new PHY support

LCN-PHY was found in 14e4:4727 card. It uses LCN/1 and 0x2064/1 radio.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: use AES_BLOCK_SIZE

mac80211 has a defnition of AES_BLOCK_SIZE and
multiple definitions of AES_BLOCK_LEN. Remove
them all and use crypto/aes.h.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: read radio ID on new cores

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: use radio ID reading code to older cores only

Newer ones need separated way

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: select BLOCKIO for BCMA

We want PIO as fallback for BCMA as well.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: change selecting ucode for newer cores

Older cores had unique PHY. This is not true anymore for newer ones.
For example core rev 16 can be LP, SSLPN or N (PHY).

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: bus: add helpers for getting/setting wldev from/in bus core

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: handle BCMA in bus switches

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: use switches for SSB specific code

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: fix CMAC races

Just like TKIP and CCMP, CMAC has the PN race.
It might not actually be possible to hit it now
since there aren't multiple ACs for management
frames, but fix it anyway.

Also move scratch buffers onto the stack.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: fix CCMP races

Since we can process multiple packets at the
same time for different ACs, but the PN is
allocated from a single counter, we need to
use an atomic value there. Use atomic64_t to
make this cheaper on 64-bit platforms, other
platforms will support this through software
emulation, see lib/atomic64.c.

We also need to use an on-stack scratch buf
so that multiple packets won't corrupt each
others scratch buffers.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: fix TKIP races, make API easier to use

Our current TKIP code races against itself on TX
since we can process multiple packets at the same
time on different ACs, but they all share the TX
context for TKIP. This can lead to bad IVs etc.

Also, the crypto offload helper code just obtains
the P1K/P2K from the cache, and can update it as
well, but there's no guarantee that packets are
really processed in order.

To fix these issues, first introduce a spinlock
that will protect the IV16/IV32 values in the TX
context. This first step makes sure that we don't
assign the same IV multiple times or get confused
in other ways.

Secondly, change the way the P1K cache works. I
add a field "p1k_iv32" that stores the value of
the IV32 when the P1K was last recomputed, and
if different from the last time, then a new P1K
is recomputed. This can cause the P1K computation
to flip back and forth if packets are processed
out of order. All this also happens under the new
spinlock.

Finally, because there are argument differences,
split up the ieee80211_get_tkip_key() API into
ieee80211_get_tkip_p1k() and ieee80211_get_tkip_p2k()
and give them the correct arguments.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: implement BCMA bus ops

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: make b43_wireless_init bus generic

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

Merge branch 'for-linville' of git://git./linux/kernel/git/luca/wl12xx

wl12xx: start/stop queues according to global per-AC counters

Split tx_queue_count to count per-AC skb's queued, instead of relying on
the skb-queue len. The skb queues used were only valid in STA-mode, as
AP-mode uses per-link queues.

This fixes a major regression in AP-mode, caused by the patch
"wl12xx: implement Tx watermarks per AC". With that patch applied, we
effectively had no regulation of Tx queues in AP-mode. Therefore a
sustained high rate of Tx could cause exhaustion of the skb memory pool.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

rt2x00: Implement tx_frames_pending mac80211 callback function.

Implementing this callback function will cause mac80211 refrain from
going to powersave state when there are still untransmitted TX frames
in the queues.
This would exactly mimic the behaviour of the legacy vendor driver which
also doesn't go in powersave mode if there are still TX frames that are not
transmitted.
This should make powersaving and rt2x00 a better couple.

Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rt2x00: Properly identify rt2800usb devices.

Sitecom WLA4000 (USB ID 0x0df6:0x0060) is an RT3072 chipset.
Sitecom WLA5000 (USB ID 0x0df6:0x0062) is an RT3572 chipset.

Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rt2x00: Add device ID for RT539F device.

Reported-by: Wim Vander Schelden <wim@fixnum.org>
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rt2x00: Reduce window of a queue's tx lock.

Currently a lot of actions that can be done without the queue's tx lock
being held are done inside the locked area.
Move them out to have a leaner and meaner code that operates while the
tx lock is being held.

Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rt2x00: Don't use queue entry as parameter when creating TX descriptor.

The functions that create the tx descriptor structure do not operate on
a queue entry at all. Signal this fact in the code by not providing a
queue entry as a parameter, but the rt2x00 device structure and the skb
directly.

This patch is a preparation for reducing the time a queue is locked for
a tx operation.

Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rt2x00: Serialize TX operations on a queue.

The rt2x00 driver gets frequent occurrences of the following error message
when operating under load:
phy0 -> rt2x00queue_write_tx_frame: Error - Arrived at non-free entry in the
non-full queue 2.

This is caused by simultaneous attempts from mac80211 to send a frame via
rt2x00, which are not properly serialized inside rt2x00queue_write_tx_frame,
causing the second frame to fail sending with the above mentioned error
message.

Fix this by introducing a per-queue spinlock to serialize the TX operations
on that queue.

Reported-by: Andreas Hartmann <andihartmann@01019freenet.de>
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

bcma: detect PCI core working in hostmode

We must not init it like clientmode one, it would break device (tested
by Hauke on BCM4718). Add stub hostmode driver for now.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mwifiex: modify SDIO aggregation Tx/Rx buffer size

The SDIO aggregation buffer size has been modified to an optimum
value which gives good throughput results.

Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ipw2100: Fix command list for debugging

There is a stray "undefined" string in the array, get rid of it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: "John W. Linville" <linville@tuxdriver.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: support GTK rekey offload

This adds the necessary mac80211 APIs to support
GTK rekey offload, mirroring the functionality
from cfg80211.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

cfg80211/nl80211: support GTK rekey offload

In certain circumstances, like WoWLAN scenarios,
devices may implement (partial) GTK rekeying on
the device to avoid waking up the host for it.

In order to successfully go through GTK rekeying,
the KEK, KCK and the replay counter are required.

Add API to let the supplicant hand the parameters
to the driver which may store it for future GTK
rekey operations.

Note that, of course, if GTK rekeying is done by
the device, the EAP frame must not be passed up
to userspace, instead a rekey event needs to be
sent to let userspace update its replay counter.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: allow driver to iterate keys

When in suspend/wowlan, devices might implement crypto
offload differently (more features), and might require
reprogramming keys for the WoWLAN (as it is the case
for Intel devices that use another uCode image). Thus
allow the driver to iterate all keys in this context.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwlwifi: remove unnecessary read of PCI_CAP_ID_EXP

The PCIE capability offset is saved during PCI bus walking.  It will
remove an unnecessary search in the PCI configuration space if this
value is referenced instead of reacquiring it.

Signed-off-by: Jon Mason <jdmason@kudzu.us>
Acked-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

Merge branch 'for-linville' of git://git./linux/kernel/git/luca/wl12xx

ssb: use pci_dev->subsystem_{vendor,device}

The SSB code reads PCI subsystem IDs from the PCI configuration registers while
they are already stored by the PCI subsystem in the 'subsystem_{vendor|device}'
fields of 'struct pci_dev'...

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ssb: use pci_dev->revision

The SSB code reads PCI revision ID from the PCI configuration register while
it's already stored by the PCI subsystem in the 'revision' field  of 'struct
pci_dev'...

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: Add RX side DMA memory barrier

This adds a memory barrier to ensure the writes to the ring memory
are committed before the DMA ring pointer is updated.
We do a similar thing on the TX side already.

Signed-off-by: Michael Buesch <m@bues.ch>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

NFC: add Documentation/networking/nfc.txt

Signed-off-by: Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
Signed-off-by: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

NFC: pn533: add NXP pn533 nfc device driver

Signed-off-by: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
Signed-off-by: Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

NFC: add the NFC socket raw protocol

This socket protocol is used to perform data exchange with NFC
targets.

Signed-off-by: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
Signed-off-by: Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

NFC: add NFC socket family

Signed-off-by: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
Signed-off-by: Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

NFC: add nfc generic netlink interface

The NFC generic netlink interface exports the NFC control operations
to the user space.

Signed-off-by: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
Signed-off-by: Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

NFC: add nfc subsystem core

The NFC subsystem core is responsible for providing the device driver
interface. It is also responsible for providing an interface to the control
operations and data exchange.

Signed-off-by: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
Signed-off-by: Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: allow driver to impose WoWLAN restrictions

If the driver can't support WoWLAN in the current
state, this patch allows it to return 1 from the
suspend callback to do the normal deconfiguration
instead of using suspend/resume calls. Note that
if it does this, resume won't be called.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ssb: PCI revision ID register is 8-bit wide

The SSB code reads PCI revision ID register as 16-bit entity while the register
is actually 8-bit only (the next 8 bits are the programming interface register).
Fix the read and make the 'rev' field of 'struct ssb_boardinfo' 8-bit as well,
to match the register size.

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rtlwifi: rtl8192de: Fix error exit from hw_init

In routine rtl92de_hw_init(), there are two places where a failure is
not handled correctly.

Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rtlwifi: rtl8192de: Replace loops calling udelay with single mdelay

There are a number of loops to implement delays. These are replaced with
single calls to mdelay().

The need for a fix was noted by Andrew Morton <akpm@linux-foundation.org>.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

carl9170: use carl9170 queue enums

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rtlwifi: rtl8192de: Remove comparison of boolean with true

Tests of a boolean against "true" are not needed as non-zero is sufficient..

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rtlwifi: rtl8192{ce,cu,se} Remove irq_enabled

This should be unnecessary if synchronize_irq is used.

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rtlwifi: rtl8192de: Remove irq_enabled boolean

Prepare rtl8192de for the removal of irq_enaqbled.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

drivers/net/wireless/rtlwifi/rtl8192de/phy.c: fix udelay() usage

ERROR: "__bad_udelay" [drivers/net/wireless/rtlwifi/rtl8192de/rtl8192de.ko] undefined!

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: fix smatch complains

mlme.c
l.757 ieee80211_dynamic_ps_enable_work(11)
variable dereferenced before check 'sdata'

mesh_pathtbl.c
l.650 mesh_path_del(20) double lock 'bottom_half'
l.663 mesh_path_del(33) double unlock 'bottom_half'

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

carl9170: allow PSM if the 5 GHz band is selected

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

carl9170: enable IEEE80211_HW_NEED_DTIM_PERIOD

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

carl9170: import 1.9.4 firmware headers

This patch imports all shared header changes
from carl9170fw.git.

 * update copyright boilerplate

 * add some more strategic __aligned(4).

 * WoWLAN

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

net, wireless: Don't return uninitialized in __cfg80211_stop_sched_scan()

If the 'driver_initiated' function argument to
__cfg80211_stop_sched_scan() is not 0 then we'll return an
uninitialized 'err' from the function.

Signed-off-by: Jesper Juhl <jj@chaosbits.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: Skip tailroom reservation for full HW-crypto devices with race fix

Based on inputs from Johannes Berg <johannes@sipsolutions.net>
from http://article.gmane.org/gmane.linux.kernel.wireless.general/68193
and http://article.gmane.org/gmane.linux.kernel.wireless.general/71702

In xmit path, devices that do full hardware crypto (including
MMIC and ICV) need no tailroom. For such devices, tailroom
reservation can be skipped if all the keys are programmed into
the hardware (i.e software crypto is not used for any of the
keys) and none of the keys wants software to generate Michael
MIC and IV.

v2: Added check for IV along with MMIC.
Reported-by: Fabio Rossi <rossi.f@inwind.it>
Tested-by: Fabio Rossi <rossi.f@inwind.it>
Signed-off-by: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
Cc: Mohammed Shafi Shajakhan <mshajakhan@atheros.com>
v3: Fixing races to avoid WARNING: at net/mac80211/wpa.c:397
    ccmp_encrypt_skb+0xc4/0x1f0
Reported-by: Andreas Hartmann <andihartmann@01019freenet.de>
Tested-by: Andreas Hartmann <andihartmann@01019freenet.de>
v4: Added links with message ID

Signed-off-by: Yogesh Ashok Powar <yogeshp@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

wl12xx: always initialize AP-mode max power level

Sometimes we only get the Tx power level via op->config when the FW is
off. Record the received power level when this happens and use it to
initialize the firmware during boot.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: initialize max_power levels for 11a channels

Set the maximum FW supported power levels for 11a channels.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: implement cancel_hw_scan() callback

After implementing cancel_hw_scan(), there is no more need to
flush scan_complete_work on suspend.

scan_complete_work enqueues a new scan_work, that might lead
to the following warning (if the device was already suspended):

WARNING: at drivers/mmc/core/core.c:169 mmc_wait_for_req+0x1ac/0x2c8()
Backtrace:
[<c004b50c>] (dump_backtrace+0x0/0x110)
[<c04fda50>] (dump_stack+0x0/0x1c)
[<c008f700>] (warn_slowpath_common+0x0/0x6c)
[<c008f76c>] (warn_slowpath_null+0x0/0x2c)
[<c034b058>] (mmc_wait_for_req+0x0/0x2c8)
[<c034f804>] (mmc_io_rw_extended+0x0/0x1e4)
[<c0350b5c>] (sdio_io_rw_ext_helper+0x0/0x1ac)
[<c0350e00>] (sdio_memcpy_fromio+0x0/0x2c)
[<bf194464>] (wl1271_sdio_raw_read+0x0/0x1bc [wl12xx_sdio])
[<bf15ac14>] (wl1271_cmd_wait_for_event_or_timeout+0x0/0x19c [wl12xx])
[<bf15adb0>] (wl1271_cmd_wait_for_event+0x0/0x30 [wl12xx])
[<bf15b7dc>] (wl1271_cmd_disconnect+0x0/0x14c [wl12xx])
[<bf158fd0>] (wl1271_unjoin+0x0/0x48 [wl12xx])
[<bf159070>] (wl1271_op_config+0x0/0x490 [wl12xx])
[<bf105084>] (ieee80211_hw_config+0x0/0x14c [mac80211])
[<bf113120>] (ieee80211_recalc_idle+0x0/0x4c [mac80211])
[<bf109a70>] (__ieee80211_scan_completed+0x0/0x1ac [mac80211])
[<bf109fd4>] (ieee80211_scan_work+0x0/0x484 [mac80211])
[<c00a197c>] (worker_thread+0x0/0x1e8)
[<c00a5874>] (kthread+0x0/0x90)

Signed-off-by: Eliad Peller <eliad@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: add stop_scan command

define wl1271_scan_stop() to cancel an ongoing hw scan.

Signed-off-by: Eliad Peller <eliad@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: configure suspend/resume only if associated/started

Configure the device (to wowlan) only if the sta/ap
is associated/started.

Signed-off-by: Eliad Peller <eliad@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: one SDIO initialization is enough

Skip manual device power on in case runtime PM is enabled for
our device.

This eliminates a secondary & redundant SDIO init sequence.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: implement Tx watermarks per AC

Each AC is stopped when its queue is filled up to the high watermark,
and restarted when its queue it lower than the low watermark. This
ensures congested ACs are not able to starve other ACs.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: Increase low & high TX watermarks

The current TX watermark settings cause the driver to stop queues too
frequently. Among other things, this can have a negative impact on WMM
prioritization, since mac80211 sorts pending packets by their ACs.

Fix this by increasing the high watermark to 256 packets. Increase the
low watermark to 32 to minimize periods with queues being stopped.

Signed-off-by: Ido Yariv <ido@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: schedule TX packets according to FW occupancy

When selecting packets for transmission, prefer the ACs that are least
occupied in the FW. When packets for multiple ACs are present in the FW,
it decides which to transmit according to WMM QoS parameters.

With these changes, lower priority ACs should not be starved when higher
priority traffic is present.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: use 802.11 header location after relocation to frame start

When operating with TKIP encryption, the function wl1271_tx_fill_hdr()
relocates the 802.11 header to the start of the frame, and leaves room
for the security header.

Some functions in the Tx path rely on the location of the header,
namely, for purposes of roaming in STA mode and connecting new stations
in AP mode. Call these functions only after the header is relocated.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: AP mode - support FW TX inactivity triggers

In AP mode we register for the MAX_TX_RETRY and INACTIVE_STA events.
Both are reported to the upper layers as a TX failure in the offending
stations.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

wl12xx: fix Tx security sequence number handling

Do not reset the security sequence number when issuing a join command or
interface is removed. Instead, reset the counter only during the unjoin
command.

Added the notion of counter wrap-around to the LSB number in
wl1271_tx_complete_packet.

Added post recovery padding to adjust for potential security number
progress during the recovery process by the firmware and avoid
potential interop issues in encrypted networks.

Signed-off-by: Oz Krakowski <ozk@ti.com>
Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>

iwlagn: remove the indirection for the rx write pointer

Not needed since the driver split.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: remove hcmd ops

All "agn" devices use the same hcmd functions, no need to call indirectly.
remove hcmd_ops

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: call bt_coex directlly

Call the 2-wire and advanced bt-coex function directly to avoid mistake

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: scd memory boundary

Assign memory boundary for SCD context, tx status and translation table

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: move the tx allocation funcs to the transport layer

These functions allocate all the Tx context. Only the simple tx_init is exported as API.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: add rx_free to transport layer

The transport layer ness to release all rx ressources. This function is an API for it.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: introduce transport layer and implement rx_init

The transport layer is responsible for all the queues, DMA rings etc...
This is the beginning of the separation of all the code that is tighly
related to HW design to the aforementioned transport layer.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: add module parameter to disable stuck queue watchdog timer

Add the parameter to disable stuck queue watchdog timer, different platforms
might have different timing. Provide the option to disable the timer to prevent
un-necessary firmware reload.

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: add correct firmware name for 135 series

135 series are WiFi/BT combo and require different uCode from 105 series.
[A
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: re-define the testmode cmd and attr enum

To make sure not having issues when adding new testmode commands or attributes
in the future, re-define the enum. no functional changes

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

iwlagn: verify mutex held for sync commands

Emmanuel noticed that there's no explicit checking
that prevents the driver from attempting to issue
multiple synchronous commands at the same time and
wrote a patch to check. However, his patch warns
only if a collision actually happened, an unlikely
thing since the driver mutex should be held for
synchronous command submissions.

So instead of checking that a collision happened
add a check that the mutex is held which ensures
that collisions can't happen.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

rtlwifi: potential forever loop in rtl92de_hw_init()

"i" should be an int here because we are trying to use it to count
to 10000.  The original code looks like it could hang in a forever
loop.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rtlwifi: rtl8192de: Fix build errors when using allyes configuration

After adding rtl8192de to linux-next, making the rtlwifi drivers be built-in
results in the following warnings:

  LD      drivers/net/wireless/rtlwifi/built-in.o
drivers/net/wireless/rtlwifi/rtl8192de/built-in.o: In function `rtl92ce_sw_led_on':
(.text+0x11fb6): multiple definition of `rtl92ce_sw_led_on'
drivers/net/wireless/rtlwifi/rtl8192ce/built-in.o:(.text+0xa326): first defined here
drivers/net/wireless/rtlwifi/rtl8192de/built-in.o:(.bss+0x0): multiple definition of `dm_digtable'
drivers/net/wireless/rtlwifi/rtl8192c/built-in.o:(.bss+0x0): first defined here
ld: Warning: size of symbol `dm_digtable' changed from 40 in drivers/net/wireless/rtlwifi/rtl8192c/built-in.o to 48 in drivers/net/wireless/rtlwifi/rtl8192de/built-in.o
drivers/net/wireless/rtlwifi/rtl8192de/built-in.o: In function `rtl92ce_sw_led_off':
(.text+0x11cfe): multiple definition of `rtl92ce_sw_led_off'
drivers/net/wireless/rtlwifi/rtl8192ce/built-in.o:(.text+0xa06e): first defined here

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: correct 0x2059 radio init

Sometimes additional steps are performed while initializing 2059 radio.
We did not find the condition yet, so make it always true for now.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k_htc: Add device ID for Sony UWA-BR100

for more details please take a look at:
http://comments.gmane.org/gmane.linux.drivers.ath9k.devel/6541
http://www.wikidevi.com/wiki/Sony_UWA-BR100

Reported-by: Thomas Novin <thomas@xyz.pp.se>
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>

ath9k: move few descriptor macros to ath9k.h

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: basic PHY init

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: add init tables

They were written from observing MMIO writes to registers 0x72 0x74 and
0x73 right after phy_write(0x017e) <- 0x3830 which finishes chennel
switching. RegExps were used to translate writes to arrays.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: init radio when enabling it

Masks and sets were found in MMIO dumps by using MMIO hacks. Shortly:
radio_write(0x0c51) <- 0x0070
radio_write(0x0c5a) <- 0x0003
radio_write(0x0146) <- 0x0003
radio_write(0x0546) <- 0x0003
radio_write(0x0946) <- 0x0003
radio_write(0x002e) <- 0x0078
radio_write(0x00c0) <- 0x0080
radio_write(0x002e) <- 0xff87
radio_write(0x00c0) <- 0xff7f
radio_write(0x0011) <- 0xfff7

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: replace radio routing magic numbers

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

wl12xx: AP-mode - use mac80211 indication about STA WME support

When adding a station, use the information given in the mac80211
populated ieee80211_sta structure to determine if it supports WME.
Provide this information to the FW.

This patch depends on "mac80211: propagate information about
STA WME support down".

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: propagate information about STA WME support down

Add a memeber to the ieee80211_sta structure to indicate whether the STA
supports WME.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rtlwifi: use PCI_VENDOR_ID_*

Use PCI_VENDOR_ID_* from pci_ids.h instead of creating #define locally.

Signed-off-by: Jon Mason <jdmason@kudzu.us>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rtlwifi: remove unnecessary read of PCI_CAP_ID_EXP

The PCIE capability offset is saved during PCI bus walking.  It will
remove an unnecessary search in the PCI configuration space if this
value is referenced instead of reacquiring it.

Also, remove unnecessary and unused #defines for PCI.

Signed-off-by: Jon Mason <jdmason@kudzu.us>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwlegacy: remove unnecessary read of PCI_CAP_ID_EXP

The PCIE capability offset is saved during PCI bus walking.  It will
remove an unnecessary search in the PCI configuration space if this
value is referenced instead of reacquiring it.

Signed-off-by: Jon Mason <jdmason@kudzu.us>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: implement lacking 0x908 PHY reg op

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: add channel switching tables for 2 GHz band

Tables were taken from observing writes in MMIO dumps.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: perform some tables ops on channel switching

Starring at MMIO dumps around PHY channel switching has led to finding
serie of 3 similar ops this patch implements.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: upload PHY values when switching channel

After calibrating radio you can find few PHY writes in MMIO dumps:
 phy_read(0x0009) -> 0x0000
phy_write(0x01ce) <- 0x03dd
phy_write(0x01cf) <- 0x03d9
phy_write(0x01d0) <- 0x03d5
phy_write(0x01d1) <- 0x0424
phy_write(0x01d2) <- 0x0429
phy_write(0x01d3) <- 0x042d
By comparing to N-PHY code we found out that they are PHY tables for
channel switching plus band info read at the beginning.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: HT-PHY: prepare place for HT-PHY tables

They are big arrays uploaded to the hardware on init, calibration, etc.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

Merge branch 'master' of git://git./linux/kernel/git/linville/wireless-2.6

Conflicts:
drivers/net/wireless/iwlwifi/iwl-tx.c

iwlagn: use PCI_DMA_* for pci_* operations

"iwlagn: map command buffers BIDI" uses the DMA_* enumerations for DMA
directions, even though the pci_* DMA API is still in use.  That patch
was undoubtedly developed on top of "iwlagn: don't use the PCI wrappers
for DMA operation", which is due in the next release.

Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: dynamic PS - don't enter PS when TX frames are pending

Use the tx_frames_pending() driver callback to determine if Tx frames are
pending for its internal queues. If so postpone the dynamic PS timeout
to avoid interrupting Tx traffic.

The commit e8306f989483e4b97a8b37dd268de6c8c6f35e75 enabled this
behavior for drivers with IEEE80211_HW_PS_NULLFUNC_STACK. We enable this
for all drivers supporting dynamic PS.

This patch helps improve performance in noisy environments.

Signed-off-by: Arik Nemtsov <arik@wizery.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>