openwrt/staging/blogic.git
14 years agowl1271: Changed SDIO MMC host claiming
Teemu Paasikivi [Thu, 18 Mar 2010 10:26:28 +0000 (12:26 +0200)]
wl1271: Changed SDIO MMC host claiming

Changed driver to keep MMC host claimed while wlan interface is up.
This keeps MMC stack from powering the SDIO device down.

Signed-off-by: Teemu Paasikivi <ext-teemu.3.paasikivi@nokia.com>
Reviewed-by: Juuso Oikarinen <juuso.oikarinen@nokia.com>
Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com>
Signed-off-by: Luciano Coelho <luciano.coelho@nokia.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agowl1271: Cleaned up wlan power on/off functions
Teemu Paasikivi [Thu, 18 Mar 2010 10:26:27 +0000 (12:26 +0200)]
wl1271: Cleaned up wlan power on/off functions

Added method for wlan power control to io_ops struct and moved
wl1271_power_on and wl1271_power_off functions to wl1271_io.h.

Signed-off-by: Teemu Paasikivi <ext-teemu.3.paasikivi@nokia.com>
Reviewed-by: Juuso Oikarinen <juuso.oikarinen@nokia.com>
Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com>
Signed-off-by: Luciano Coelho <luciano.coelho@nokia.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agowl1271: Fix configuration of the TX opportunity value
Juuso Oikarinen [Thu, 18 Mar 2010 10:26:26 +0000 (12:26 +0200)]
wl1271: Fix configuration of the TX opportunity value

The per-queue TX opportunity value is configured by the mac80211 in units
of 32us. The firmware however wants it in us, so add the conversion.

Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com>
Reviewed-by: Kalle Valo <kalle.valo@nokia.com>
Signed-off-by: Luciano Coelho <luciano.coelho@nokia.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_htc: Support for AR9271 chipset.
Sujith [Wed, 17 Mar 2010 08:55:25 +0000 (14:25 +0530)]
ath9k_htc: Support for AR9271 chipset.

Features:

 * Station mode
 * IBSS mode
 * Monitor mode
 * Legacy support
 * HT support
 * TX/RX 11n Aggregation
 * HW encryption
 * LED
 * Suspend/Resume

For more information: http://wireless.kernel.org/en/users/Drivers/ath9k_htc

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Senthil Balasubramanian <senthilkumar@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: fix hardware deinit
Sujith [Wed, 17 Mar 2010 08:55:24 +0000 (14:25 +0530)]
ath9k_hw: fix hardware deinit

Without this you will get a panic if the device initialization
fails. Also, free ath_hw instance properly. ath9k_hw_deinit()
shouldn't do it.

Cc: stable@kernel.org
Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: add HTC init hardware call for special resets for AR9271
Sujith [Wed, 17 Mar 2010 08:55:23 +0000 (14:25 +0530)]
ath9k_hw: add HTC init hardware call for special resets for AR9271

AR9271 needs a full reset only upon the first reset, add
a call for the driver to enable these special resets. We
can optimize this out later without an export.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: always set the core clock for AR9271
Sujith [Wed, 17 Mar 2010 08:55:22 +0000 (14:25 +0530)]
ath9k_hw: always set the core clock for AR9271

When initializing the PLL on AR9271 we always need
to set the core clock to 117MHz. While at it remove
the baud rate settings for the serial device on the
AR9271, the default settings work well unless you
want to customize it.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: use the skip count for PA calibration on AR9271
Sujith [Wed, 17 Mar 2010 08:55:21 +0000 (14:25 +0530)]
ath9k_hw: use the skip count for PA calibration on AR9271

Periodic power amplifier offset calibration is skipped on ath9k
algorithmically, this is required on AR9271.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: restrict valid nf readings for AR9271 to -114
Sujith [Wed, 17 Mar 2010 08:55:20 +0000 (14:25 +0530)]
ath9k_hw: restrict valid nf readings for AR9271 to -114

Noisefloor values read on AR9271 are unreliable if they
are less than -114, set those statically to -116.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: fix noisefloor history buffer usage on AR9271
Sujith [Wed, 17 Mar 2010 08:55:19 +0000 (14:25 +0530)]
ath9k_hw: fix noisefloor history buffer usage on AR9271

Noisefloor calibration involves querying hardware for samples
and storing this information on a history buffer in hardware for
actual noisefloor calibration processing in hardware. The history
buffer supports collecting information for all Atheros hardware,
one history buffer slot for each chain on each channel used for
MIMO operation. For current hardware this means one history
buffer slot for each chain on both the control (or primary) channel
and the extension (or secondary) channel. We know which noisefloor
registers to poke for collecting noisefloor data through the
chainmask.

For AR9285 and AR9271 devices, both 1x1, the chaimmask is defined as
0x9 = 0b0001001. The first four bits represent each chain out of
a maximum of 4 chains [0-3] on the primary channel. The last four
bits represent each chain on the extension channel. A chainmask
of 0x9 therefore indicates chain 1 is active on both the primary
and the extension channel.

AR9271 only requires collecting and storing noisefloor history buffer
data for the first chain on both the control and extension channel
(nfarray[0] and nfarray[3]) so fix the code and avoid which reads
and writes to the history buffer for the other chains.

Since the noisefloor varies depending on the number of chains your
device supports also initialize the noisefloor history buffer with
reasonable values seen on 1x1 devices such as AR9285.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: Fix full sleep setup for AR9271
Sujith [Wed, 17 Mar 2010 08:55:18 +0000 (14:25 +0530)]
ath9k_hw: Fix full sleep setup for AR9271

After telling the AR9271 to go into full sleep we do not need
to clear the RTC reset signal.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: fix TX descriptor setup for AR9271
Sujith [Wed, 17 Mar 2010 08:55:17 +0000 (14:25 +0530)]
ath9k_hw: fix TX descriptor setup for AR9271

The TX descriptors setup for AR971 requires the same
setup as AR9285, so use that.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: skip chip tests for AR9271
Sujith [Wed, 17 Mar 2010 08:55:16 +0000 (14:25 +0530)]
ath9k_hw: skip chip tests for AR9271

The chip test is not required for AR9271 on the host driver
code as the firmware will do the test internally on its own.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: add GPIO setup code for AR9271
Sujith [Wed, 17 Mar 2010 08:55:15 +0000 (14:25 +0530)]
ath9k_hw: add GPIO setup code for AR9271

Assign the proper number of GPIO pins for AR9271.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k_hw: update initialization values for AR9271
Sujith [Wed, 17 Mar 2010 08:55:14 +0000 (14:25 +0530)]
ath9k_hw: update initialization values for AR9271

Update the register initialization values for AR9271.
This is based on our last review from our systems team.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k: decrease size of ath9k.ko
Ming Lei [Mon, 1 Mar 2010 15:13:29 +0000 (23:13 +0800)]
ath9k: decrease size of ath9k.ko

From e74b075cdb143d45be9b371ee8a8e2dcfc15ab34 Mon Sep 17 00:00:00 2001
From: Ming Lei <tom.leiming@gmail.com>
Date: Sat, 27 Feb 2010 23:50:54 +0800
Subject: [PATCH] ath9k: decrease size of ath9k.ko

The patch defines the fields of 'valid_single_stream' and 'valid' in
struct ath_rate_table as char type, so decrease the size of ath9k.ko
about 2KB.

old ath9k.ko
[tom@tom-lei ath9k]$ size ath9k.ko
   text    data     bss     dec     hex filename
  69344    3080     168   72592   11b90 ath9k.ko

new ath9k.ko
[tom@tom-lei ath9k]$ size ath9k.ko
   text    data     bss     dec     hex filename
  67304    3080     168   70552   11398 ath9k.ko

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoMerge branch 'wireless-next-2.6' of git://git.kernel.org/pub/scm/linux/kernel/git...
John W. Linville [Tue, 23 Mar 2010 20:46:43 +0000 (16:46 -0400)]
Merge branch 'wireless-next-2.6' of git://git./linux/kernel/git/iwlwifi/iwlwifi-2.6

14 years agoiwlwifi: code cleanup for connectivity recovery
Wey-Yi Guy [Fri, 5 Mar 2010 22:22:46 +0000 (14:22 -0800)]
iwlwifi: code cleanup for connectivity recovery

Split the connectivity check and recovery routine into separated
functions based on the types
   1. iwl_good_ack_health() - check for ack count
   2. iwl_good_plcp_health() - check for plcp error

Based on the type of errors being detected, different recovery methods
will be used to bring the system back to normal operational state.

Because different NIC has different HW and uCode, the behavior is also
different; these functions thus now form part of the ops infrastructure,
so we can have more control on how to monitor and recover from error condition
case per device.

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: Recover TX flow failure
Wey-Yi Guy [Thu, 4 Mar 2010 21:38:59 +0000 (13:38 -0800)]
iwlwifi: Recover TX flow failure

Monitors the tx statistics to detect the drop in throughput.
When the throughput drops, the ratio of the actual_ack_count and the
expected_ack_count also drops.  At the same time, the aggregated
ba_timeout (the number of ba timeout retries) also rises.  If the
actual_ack_count/expected_ack_count ratio is 0 and the number of ba
timeout retries rises to BA_TIMEOUT_MAX, no tx packets can be delivered.
Reloading the uCode and bring the system back to normal operational
state.

Signed-off-by: Trieu 'Andrew' Nguyen <trieux.t.nguyen@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: move plcp check to separated function
Wey-Yi Guy [Thu, 4 Mar 2010 21:38:58 +0000 (13:38 -0800)]
iwlwifi: move plcp check to separated function

Move the plcp error checking into stand alone function and pointed by ops
to accommodate devices not needing this recovery.

Signed-off-by: Trieu 'Andrew' Nguyen <trieux.t.nguyen@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: Recover TX flow stall due to stuck queue
Wey-Yi Guy [Tue, 2 Mar 2010 01:23:50 +0000 (17:23 -0800)]
iwlwifi: Recover TX flow stall due to stuck queue

Monitors the internal TX queues periodically.  When a queue is stuck
for some unknown conditions causing the throughput to drop and the
transfer is stop, the driver will force firmware reload and bring the
system back to normal operational state.

The iwlwifi devices behave differently in this regard so this feature is
made part of the ops infrastructure so we can have more control on how to
monitor and recover from tx queue stall case per device.

Signed-off-by: Trieu 'Andrew' Nguyen <trieux.t.nguyen@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: clean up driver names for 1000/5000/6000
Shanyu Zhao [Sat, 6 Mar 2010 01:05:20 +0000 (17:05 -0800)]
iwlwifi: clean up driver names for 1000/5000/6000

Align the driver names with official product names for 1000/5000/6000
series. This change mainly affects the debug messages show up in system
log. No functional changes.

Signed-off-by: Shanyu Zhao <shanyu.zhao@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: sta_id cannot be invalid in rs_initialize_lq
Johannes Berg [Thu, 21 Jan 2010 15:07:44 +0000 (07:07 -0800)]
iwlwifi: sta_id cannot be invalid in rs_initialize_lq

The only caller of this is iwl_rs_rate_init
which is only called with a valid sta_id.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: implement new mac80211 station add/remove calls
Reinette Chatre [Tue, 23 Feb 2010 00:24:47 +0000 (16:24 -0800)]
iwlwifi: implement new mac80211 station add/remove calls

mac80211 recently implemented two new callbacks that are used to request
station add/remove from the driver. The benefot from these new callbacks
are that they enable the driver to sleep while performing this work.

This is a big patch since a few things need to be coordinated in this move.
First we need to decouple station management from rate scaling, which
caused a lot of code to be moved and/or deleted. Next we needed to tie in
with mac80211's station management callback and let it direct our station
management as well as trigger the rate scaling initialization.

Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: only add broadcast station once
Reinette Chatre [Fri, 19 Feb 2010 06:58:32 +0000 (22:58 -0800)]
iwlwifi: only add broadcast station once

Currently the broadcast station is added after every RXON command. Change
this to only add the broadcast station when interface is added by mac80211.
With this we need some extra work to ensure broadcast station is always
present since station table is cleared when RXON without ASSOC bit set is
sent. To deal with this we re-add all driver known stations to uCode after
such an RXON command is sent.

We also do some cleanup and remove the various calls to clear the station
table. We now only clear the station table in two scenarios:
- only clear uCode portion of station table when RXON command without ASSOC
bit is sent
- clear uCode and driver portion when interface goes down or is removed.

We need to do this clearing when interface goes down to deal with the
device restart/reconfigure routines which do not remove the interface, but
do add the interface during reconfiguration.

Previously the keys were also cleared when station table in driver is
cleared, this is not done anymore since mac80211 will take care that keys
are set and cleared correctly.

There is a known issue with this change. Associating with different AP
without bringing interface down fails with a firmware error. This is
because of the lack of full station notification support and the later
patches in this series that complete the station notification support will
fix this.

Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlagn: Add Flow-handler (FH) register dump for 4965.
Ben Cahill [Tue, 2 Mar 2010 20:48:25 +0000 (12:48 -0800)]
iwlagn: Add Flow-handler (FH) register dump for 4965.

FH register dump debug feature was recently added for 5000 and later;
4965 can use it also.

Signed-off-by: Ben Cahill <ben.m.cahill@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agorename new rfkill sysfs knobs
florian@mickler.org [Sat, 13 Mar 2010 12:31:05 +0000 (13:31 +0100)]
rename new rfkill sysfs knobs

This patch renames the (never officially released) sysfs-knobs
"blocked_hw" and "blocked_sw" to "hard" and "soft", as the hardware vs
software conotation is misleading.

It also gets rid of not needed locks around u32-read-access.

Signed-off-by: Florian Mickler <florian@mickler.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agomac80211: (really) fix rates setup on IBSS merge
Bruno Randolf [Wed, 3 Mar 2010 09:45:42 +0000 (18:45 +0900)]
mac80211: (really) fix rates setup on IBSS merge

when an IBSS merge happened, the supported rates for the newly added station
were left empty, causing the rate control module to be initialized with only
the basic rates.

the section of the ibss code which deals with updating supported rates for
an already existing station failed to inform the rate control module about the
new rates. as both minstrel and pid don't have an update function i just use
the init function.

also remove unnecessary (unsigned long long) casts and edit debug message.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoRevert "mac80211: fix rates setup on IBSS merge"
John W. Linville [Tue, 16 Mar 2010 19:02:35 +0000 (15:02 -0400)]
Revert "mac80211: fix rates setup on IBSS merge"

I accidentally merged an incomplete version of the patch...

This reverts commit b4d59a9317e41faec3d0b6a03f0454d1e8abb710.

Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agort2x00: Fix TX status reporting for rt2800pci.
Alban Browaeys [Sun, 28 Feb 2010 16:14:40 +0000 (17:14 +0100)]
rt2x00: Fix TX status reporting for rt2800pci.

After testing, we found that TX_STA_FIFO_MCS is the last MCS value
tried. If the transmission failed, 8 frames have been transmitted. If the
transmission succeed, we can easily compute the number of retry.  This patch fix
the way status is reported to mac80211 rate control. It has 2 bugs :

1. mcs can contain the short preamble flag and it will lead to wrong
computations.

2. minstrel nearly always say that 54 Mbits is the best rate, even if we are
very far from the AP

Signed-off-by: Benoit Papillault <benoit.papillault@free.fr>
Signed-off-by: Alban Browaeys <prahal@yahoo.com>
Acked-by: Ivo van Doorn <ivdoorn@gmail.com>
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agort2x00: txdone implementation supporting hw encryption.
Alban Browaeys [Fri, 26 Feb 2010 22:19:59 +0000 (23:19 +0100)]
rt2x00: txdone implementation supporting hw encryption.

This is an implementation that support WCID being the encryption key.
Wireless Cli Id was set to be the encryption key in rt2800pci_write_tx_desc
and read (TX_STA_FIFO_WCID) as the current queue entry index.

Signed-off-by: Benoit Papillault <benoit.papillault@free.fr>
Signed-off-by: Alban Browaeys <prahal@yahoo.com>
Acked-by: Ivo van Doorn <ivdoorn@gmail.com>
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agort2x00: fix warning when building rt2800pci with just soc support
Helmut Schaa [Mon, 15 Mar 2010 16:22:26 +0000 (17:22 +0100)]
rt2x00: fix warning when building rt2800pci with just soc support

Fix compile warning "rt2800pci.c:1248: warning: 'rt2800pci_device_table'
defined but not used" when building rt2800pci with only soc support
(without pci).

Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoipw2x00: replace "ieee80211" with "libipw" where appropriate
Pavel Roskin [Fri, 12 Mar 2010 05:01:22 +0000 (00:01 -0500)]
ipw2x00: replace "ieee80211" with "libipw" where appropriate

"ieee80211" was the old name of the common library for ipw2100 and
ipw2200.  It was renamed to "libipw", but some occurrences of the old
name remained.

Rename alloc_ieee80211() to alloc_libipw() and free_ieee80211() to
free_libipw().  Adjust comments and label names.  Change prefixes in
diagnostic messages.

Keep /proc/net/ieee80211 under the original name to avoid breaking user
interface.

Move the affected EXPORT_SYMBOL macros to their proper places.

Signed-off-by: Pavel Roskin <proski@gnu.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoorinoco: disable support for prism chipset by default
Pavel Roskin [Fri, 12 Mar 2010 04:13:28 +0000 (23:13 -0500)]
orinoco: disable support for prism chipset by default

The hostap driver provides better support for Prism chipset.

Signed-off-by: Pavel Roskin <proski@gnu.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agowl1251: reduce eeprom read wait time
Grazvydas Ignotas [Thu, 11 Mar 2010 15:45:07 +0000 (17:45 +0200)]
wl1251: reduce eeprom read wait time

4sec wait is way too pessimistic, TI driver uses 40ms here,
and testing shows that is ebough, so let's also use that.
While at it, add useful sounding comment from the TI driver.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Acked-by: Kalle Valo <kalle.valo@iki.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agowl1251: fix ELP_CTRL register accesses when using SDIO
Grazvydas Ignotas [Thu, 11 Mar 2010 15:44:57 +0000 (17:44 +0200)]
wl1251: fix ELP_CTRL register accesses when using SDIO

For some unknown reason ELP_CTRL can't be accesed using
sdio_memcpy_* functions (any attemts to do so result in timeouts):

 wl1251: ERROR sdio write failed (-110)
 wl1251: ERROR sdio read failed (-110)
 wl1251: WARNING WLAN not ready

To fix this, add special IO functions for ELP_CTRL access that are
using sdio_readb/sdio_writeb. Similar handling is done in TI
reference driver from Android code drop.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Cc: Bob Copeland <me@bobcopeland.com>
Acked-by: Kalle Valo <kalle.valo@iki.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agowl1251: make local symbols static
Grazvydas Ignotas [Fri, 12 Mar 2010 10:28:41 +0000 (12:28 +0200)]
wl1251: make local symbols static

Make local functions and data static, also constify
some structures. While at it, clean up unneeded includes.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Cc: Bob Copeland <me@bobcopeland.com>
Acked-by: Kalle Valo <kalle.valo@iki.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agomac80211: optimize tx status processing
Felix Fietkau [Thu, 11 Mar 2010 15:28:24 +0000 (16:28 +0100)]
mac80211: optimize tx status processing

When a cooked monitor interface is active, ieee80211_tx_status()
generates a radiotap header for every single frame, even if it wasn't
injected and thus won't be sent to a monitor interface.
This patch reduces cpu utilization by moving the cooked monitor check a
bit earlier, before it generates the rtap header.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoDRIVER: Libertas: Fixed coding style in rx.c
Stewart Malik [Thu, 11 Mar 2010 09:58:29 +0000 (20:28 +1030)]
DRIVER: Libertas: Fixed coding style in rx.c

Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: remove unused beacon timer code
John W. Linville [Wed, 10 Mar 2010 16:08:50 +0000 (11:08 -0500)]
ath5k: remove unused beacon timer code

Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: remove dead source in ath5k_combine_linear_pcdac_curves
John W. Linville [Wed, 10 Mar 2010 16:08:49 +0000 (11:08 -0500)]
ath5k: remove dead source in ath5k_combine_linear_pcdac_curves

This code was commented-out when it was added about a year ago and
remains unchanged -- seems as if we don't need it...

Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: remove some dead functions
John W. Linville [Wed, 10 Mar 2010 16:08:48 +0000 (11:08 -0500)]
ath5k: remove some dead functions

"ath5k: remove stale function declarations, make some functions static"
commented-out some unused functions.  This removes them.

Signed-off-by: John W. Linville <linville@tuxdriver.com>
Acked-by: Bob Copeland <me@bobcopeland.com>
14 years agoairo : Print of firmware version
matthieu castet [Sun, 28 Feb 2010 14:42:54 +0000 (15:42 +0100)]
airo : Print of firmware version

For the firmware version 5.30.17 the log file shows:
Firmware version 5.30.11

The variable softSubVer is binary.

Signed-off-by: Matthieu CASTET <castet.matthieu@free.fr>
Signed-off-by: Jose Alonso <joalonsof@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoairo : fix printing status info
matthieu castet [Sun, 28 Feb 2010 14:31:21 +0000 (15:31 +0100)]
airo : fix printing status info

For some status, reason is encoded in the low byte, but airo_print_status forgot tp mask low byte in status parsing.

This make it only work when reason is 0.

Signed-off-by: Matthieu CASTET <castet.matthieu@free.fr>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agossb: Export ssb_chipco_gpio_control - needed by N PHY code
Larry Finger [Wed, 10 Mar 2010 16:41:32 +0000 (10:41 -0600)]
ssb: Export ssb_chipco_gpio_control - needed by N PHY code

The latest changes in the N PHY core require the symbol
ssb_chipco_gpio_control to be exported.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: fixing retries in ath5k_hw_setup_4word_tx_desc
Andrew Blaich [Mon, 1 Mar 2010 15:30:40 +0000 (10:30 -0500)]
ath5k: fixing retries in ath5k_hw_setup_4word_tx_desc

The rate control algorithm, default is Minstrel for ath5k, determines
the number of retries to use for each rate.  However, there exists in
ath5k_hw_setup_4word_tx_desc (which is called for AR5212 like devices)
a set number of retries defined by AR5K_TUNE_HWTXTRIES.  The set
number of tries is added to the tx_tries0 variable setup by the rate
control algorithm.  This changes the number of retries the rate
control algorithm considers necessary.  By removing the
AR5K_TUNE_HWTXTRIES from the retry calculation the rate control
algorithm is given control over the number of retries.

Signed-off-by: Andrew Blaich <ablaich@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath9k: fix rate control tx status handling for A-MPDU
Felix Fietkau [Mon, 1 Mar 2010 12:32:11 +0000 (13:32 +0100)]
ath9k: fix rate control tx status handling for A-MPDU

Currently the rate control tx status update gets called for every
subframe of an A-MPDU, and ath9k marks the frame with the relevant
status update with an internal flag. This not suitable for rate control
algorithms using the standard mac80211 rate control API, so fix this by
using IEEE80211_TX_STAT_AMPDU for marking the correct frames that
should be processed.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agowl1271: Changed wl1271_sdio to be selectable only on ARM
Teemu Paasikivi [Wed, 10 Mar 2010 11:22:31 +0000 (13:22 +0200)]
wl1271: Changed wl1271_sdio to be selectable only on ARM

As wl1271_sdio implementation depends on ARM GPIO impelementation it is
not directly usable on other architectures at the moment. Added
ARM dependency to kernel configuration option.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Teemu Paasikivi <ext-teemu.3.paasikivi@nokia.com>
Reviewed-by: Luciano Coelho <luciano.coelho@nokia.com>
Signed-off-by: Luciano Coelho <luciano.coelho@nokia.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: Minor EEPROM documentation updates
Bruno Randolf [Tue, 9 Mar 2010 07:56:21 +0000 (16:56 +0900)]
ath5k: Minor EEPROM documentation updates

Here are some minor updates for EEPROM, mostly documentation and some small
fixes which have no effect at the moment.

- fixed_bias is not available for B mode.

- AR5K_EEPROM_[RT]X_CHAIN_DIS is 3 bit. this is MIMO and will not be used in
  ath5k, but just to be correct.

- AR5K_EEPROM_JAP_MID_EN added, and shift of following flags adapted.

- added some documentation for EEPROM values and some comments.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: IQ calibration for AR5211 is slightly different
Bruno Randolf [Tue, 9 Mar 2010 07:56:15 +0000 (16:56 +0900)]
ath5k: IQ calibration for AR5211 is slightly different

according to the HAL sources the calculation of the Q value is slightly
different for AR5211 chips.

i couldn't test this since IQ calibration never finishes on older parts. this
is a different problem...

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: add debugfs file frameerrors
Bruno Randolf [Tue, 9 Mar 2010 07:56:00 +0000 (16:56 +0900)]
ath5k: add debugfs file frameerrors

add a debugfs file to see different RX and TX errors as reported in our status
descriptors. this can help to diagnose driver problems.

statistics can be cleared by writing 'clear' into the frameerrors file.

example:

# cat /sys/kernel/debug/ath5k/phy0/frameerrors
RX
---------------------
CRC     27      (11%)
PHY     3       (1%)
FIFO    0       (0%)
decrypt 0       (0%)
MIC     0       (0%)
process 0       (0%)
jumbo   0       (0%)
[RX all 245]

TX
---------------------
retry   2       (9%)
FIFO    0       (0%)
filter  0       (0%)
[TX all 21]

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: remove ah_gpio_npins
Bruno Randolf [Tue, 9 Mar 2010 07:55:55 +0000 (16:55 +0900)]
ath5k: remove ah_gpio_npins

it's never used and we have a newer implementation in gpio.c.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: remove ah_mac_revision
Bruno Randolf [Tue, 9 Mar 2010 07:55:49 +0000 (16:55 +0900)]
ath5k: remove ah_mac_revision

it's not used, and we have ah_mac_srev.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: remove ah_magic
Bruno Randolf [Tue, 9 Mar 2010 07:55:44 +0000 (16:55 +0900)]
ath5k: remove ah_magic

it's never used. probably a leftover from the old OpenHAL days...

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: remove double opmode definition
Bruno Randolf [Tue, 9 Mar 2010 07:55:38 +0000 (16:55 +0900)]
ath5k: remove double opmode definition

opmode (operating mode) was defined in struct ath5k_hw and struct ath5k_softc.
remove it from ath5k_hw and use only from ath5k_softc (sc->opmode).

(btw: what's the meaning of opmode when we have multiple interfaces?)

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: preserve antenna settings
Bruno Randolf [Tue, 9 Mar 2010 07:55:28 +0000 (16:55 +0900)]
ath5k: preserve antenna settings

save antenna settings and preserve across resets.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: add antenna statistics and debugfs file for antenna settings
Bruno Randolf [Tue, 9 Mar 2010 07:55:17 +0000 (16:55 +0900)]
ath5k: add antenna statistics and debugfs file for antenna settings

keep statistics about which antenna was used for TX and RX. this is used only
for debugging right now, but might have other applications later.

add a new file 'antenna' in debugfs (/sys/kernel/debug/ath5k/phy0/antenna) to show
antenna use statistics and antenna diversity related register values. it can
also be used to set the antenna mode until we have proper support for that in
iw:
  - echo diversity > antenna: use default antenna mode (RX and TX diversity)
  - echo fixed-a > antenna: use fixed antenna A for RX and TX
  - echo fixed-b > antenna: use fixed antenna B for RX and TX
  - echo clear > antenna: reset antenna statistics

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agorndis_wlan: Implement cfg80211 PMKSA API
Jussi Kivilinna [Thu, 4 Mar 2010 16:27:36 +0000 (18:27 +0200)]
rndis_wlan: Implement cfg80211 PMKSA API

Add support for cfg80211 set_pmksa/del_pmksa/flush_pmksa. Updating PMKID
entry list is done on driver side since NDIS API requires full list update.

v2:
 - fixed to use new netdev_dbg/warn/etc instead of old devdbg/warn/etc
 - fixed false padding from struct ndis_80211_bssid_info

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agorndis_wlan: get max_num_pmkids from device
Jussi Kivilinna [Thu, 4 Mar 2010 16:27:24 +0000 (18:27 +0200)]
rndis_wlan: get max_num_pmkids from device

Extend rndis_wlan_get_caps() to get 802.11 capabilities and maximum
supported number of PMKIDs by device.

v2: fixed to use new netdev_dbg/warn/etc instead of old devdbg/warn/etc

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agorndis_wlan: remove unused variables from priv structure
Jussi Kivilinna [Thu, 4 Mar 2010 16:27:12 +0000 (18:27 +0200)]
rndis_wlan: remove unused variables from priv structure

Some variables were left unused after cfg80211 conversion. Remove those and related deadcode.

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agorndis_wlan: copy only useful data from rndis_command respond
Jussi Kivilinna [Thu, 4 Mar 2010 16:27:02 +0000 (18:27 +0200)]
rndis_wlan: copy only useful data from rndis_command respond

rndis_query_oid() uses full output buffer size to copy response buffer
from rndis_command()/device. This doesn't cause problems as response buffer
is sized based on output buffer but does copy extra unset bytes.
So change rndis_query_oid() so that only meaningful bytes are being copied.

Also in case of malfunctioning device/cable/etc returned data offset from
device might be wrong so bound check memory access correctly, so add
checks for this.

v2: fixed to use new netdev_dbg/warn/etc instead of old devdbg/warn/etc

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agominstrel: make the rate control ops reusable from another rc implementation
Felix Fietkau [Mon, 1 Mar 2010 21:21:40 +0000 (22:21 +0100)]
minstrel: make the rate control ops reusable from another rc implementation

This patch makes it possible to reuse the minstrel rate control ops
from another rate control module. This is useful in preparing for the
new 802.11n implementation of minstrel, which will reuse the old code
for legacy stations.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agominstrel: simplify and fix debugfs code
Felix Fietkau [Mon, 1 Mar 2010 21:17:38 +0000 (22:17 +0100)]
minstrel: simplify and fix debugfs code

This patch cleans up the debugfs read function for the statistics by
using simple_read_from_buffer instead of its own semi-broken hack.
Also removes a useless member of the minstrel debugfs info struct.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoath5k: Fix TX/RX padding for all frames
Benoit Papillault [Sat, 27 Feb 2010 22:05:26 +0000 (23:05 +0100)]
ath5k: Fix TX/RX padding for all frames

Currently, the padding position is based on
ieee80211_get_hdrlen_from_skb(). This is not correct since the HW does
padding on RX (and expect the same padding to be present on TX) at the
following position :

- management : 24 + 6 if 4-addr format
- control    : 24 + 6 if 4-addr format
- data       : 24 + 6 if 4-addr format + 2 if QoS
- invalid    : 24 + 6 if 4-addr format

whereas ieee80211_get_hdrlen_from_skb() is :

- management : 24
- control    : 16 except for ACK/CTS where it is 10
- data       : 24 + 6 if 4-addr format + 2 if QoS + 2 if QoS & order
- invalid    : 24

So, correct frames are not affected : management frames do not use
4-addr format, control frames have no body and invalid frames are ...
not valid by definition. However, in order to use monitor interface for
debugging purpose, one must be able to send/receive any frames, be it
correct or not. Such frames are affected by incorrect padding.

Moreover, since padding is added on TX, we need to remove it before
calling ieee80211_tx_status. This affect TX packets received by monitor
interfaces.

It has been tested between an ath5k based card (AR5212) and an ar9170usb
based card (netgear WNDA3100) using a frame generator and a monitor
interface for each card.

v2: Added ath5k_add_padding / ath5k_remove_padding

Signed-off-by: Benoit Papillault <benoit.papillault@free.fr>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agomac80211: give warning if building w/out rate ctrl algorithm
Andres Salomon [Fri, 26 Feb 2010 00:18:47 +0000 (19:18 -0500)]
mac80211: give warning if building w/out rate ctrl algorithm

I discovered that if EMBEDDED=y, one can accidentally build a mac80211 stack
and drivers w/ no rate control algorithm.  For drivers like RTL8187 that don't
supply their own RC algorithms, this will cause ieee80211_register_hw to
fail (making the driver unusable).

This will tell kconfig to provide a warning if no rate control algorithms
have been selected.  That'll at least warn the user; users that know that
their drivers supply a rate control algorithm can safely ignore the
warning, and those who don't know (or who expect to be using multiple
drivers) can select a default RC algorithm.

Signed-off-by: Andres Salomon <dilinger@collabora.co.uk>
Cc: stable@kernel.org
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoiwmc3200wifi: typo fix and code cleanup
Zhu Yi [Thu, 25 Feb 2010 06:15:31 +0000 (14:15 +0800)]
iwmc3200wifi: typo fix and code cleanup

Fix wrong IWM_RX_TICKET_DROP_REASON_MSK macro define, typo and other
small cleanups.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoiwmc3200wifi: mark some pmksa functions static
Zhu Yi [Thu, 25 Feb 2010 06:15:30 +0000 (14:15 +0800)]
iwmc3200wifi: mark some pmksa functions static

This fixes the sparse warnings.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoiwmc3200wifi: increase concatenated buffer
Zhu Yi [Thu, 25 Feb 2010 06:15:29 +0000 (14:15 +0800)]
iwmc3200wifi: increase concatenated buffer

Increase concatenated buffer from 8K to 32K to get better performance.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoiwmc3200wifi: protect rx_tickets and rx_packets[] lists
Zhu Yi [Thu, 25 Feb 2010 06:15:28 +0000 (14:15 +0800)]
iwmc3200wifi: protect rx_tickets and rx_packets[] lists

Protect rx_tickets and rx_packets[] lists with spinlocks to fix the
race condition for concurrent list operations. In iwmc3200wifi both
sdio_isr_worker and rx_worker workqueues can access the rx ticket
and packets lists at the same time under high rx load.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoiwmc3200wifi: add ftrace event tracing support
Zhu Yi [Thu, 25 Feb 2010 06:15:27 +0000 (14:15 +0800)]
iwmc3200wifi: add ftrace event tracing support

Add event tracer for iwmc3200wifi driver. When enabled, all the
commands and responses between the driver and firmware (also
including Tx/Rx frames) will be recorded in the ftrace ring buffer.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoiwmc3200wifi: remove "_safe" for some list_for_each_entry usage
Zhu Yi [Thu, 25 Feb 2010 06:15:26 +0000 (14:15 +0800)]
iwmc3200wifi: remove "_safe" for some list_for_each_entry usage

Use list_for_each_entry instead of list_for_each_entry_safe in
places iteration against list entry removal is not required.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoiwmc3200wifi: refuse to associate on unallowed channels
Zhu Yi [Thu, 25 Feb 2010 06:15:25 +0000 (14:15 +0800)]
iwmc3200wifi: refuse to associate on unallowed channels

We need to make sure we don't associate with APs on unallowed
channels (according to regulatory setting). This could happen
when the channel is not specified (auto-select) within the
connection request. In this case we get the AP's channel until
the firmware indicates the association succeeded later. We need
to verify the associated channel. If the channel is disabled by
regulatory, we have to disassociate with the AP.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoenhance sysfs rfkill interface
florian@mickler.org [Fri, 26 Feb 2010 11:01:34 +0000 (12:01 +0100)]
enhance sysfs rfkill interface

This commit introduces two new sysfs knobs.

/sys/class/rfkill/rfkill[0-9]+/blocked_hw: (ro)
hardblock kill state
/sys/class/rfkill/rfkill[0-9]+/blocked_sw: (rw)
softblock kill state

Signed-off-by: Florian Mickler <florian@mickler.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoDocument the rfkill sysfs ABI
florian@mickler.org [Wed, 24 Feb 2010 11:05:16 +0000 (12:05 +0100)]
Document the rfkill sysfs ABI

This moves sysfs ABI info from Documentation/rfkill.txt to the
ABI subfolder and reformats it.

This also schedules the deprecated sysfs parts to be removed in
2012 (claim file) and 2014 (state file).

Signed-off-by: Florian Mickler <florian@mickler.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoiwl3945: check ucode load error code
Abhijeet Kolekar [Fri, 26 Feb 2010 23:17:01 +0000 (15:17 -0800)]
iwl3945: check ucode load error code

Check successful completion of BSM loading. Give a chance to
load BSM again. If BSM loading is unsuccessful we should exit
and not initilize NIC. Before this checking even if the BSM loading failed
driver tries to initilize the NIC which should not be the case.

Signed-off-by: Abhijeet Kolekar <abhijeet.kolekar@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: clean up queue/fifo handling
Johannes Berg [Wed, 24 Feb 2010 09:57:19 +0000 (01:57 -0800)]
iwlwifi: clean up queue/fifo handling

4965 hardware has 7 queues reserved and the
remaining ones used for aggregation, 5000
and higher need to have 10 reserved. This
is not very clear in the code right now,
unfortunately.

Introduce a new IWL_TX_FIFO_UNUSED constant
and make the queue/FIFO mapping arrays able
to hold that value, and change the setup
code to reserve all queues in the arrays
(the queue number is the index) and use the
new unused constant to not map those queues
to any FIFO.

Additionally, clear up the AC/queue mapping
code to be more understandable. The mapping
is the identity mapping right now, but with
the mapping function I think it's easier to
understand what happens there.

Finally, HCCA isn't implemented at all and
I think newer microcode removed it, so let's
remove all mention of it in the code, some
comments remain for 4965.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Acked-by: Shanyu Zhao <shanyu.zhao@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: add internal short scan support for 3945
Wey-Yi Guy [Wed, 24 Feb 2010 16:28:30 +0000 (08:28 -0800)]
iwlwifi: add internal short scan support for 3945

Add internal short scan support for 3945 NIC, This allows 3945 NIC
to support radio reset request like the other series of NICs.

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove unnecessary parameter in scan function
Wey-Yi Guy [Tue, 23 Feb 2010 22:36:13 +0000 (14:36 -0800)]
iwlwifi: remove unnecessary parameter in scan function

small change to remove unnecessary "ret" parameter since it not being
used.

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove duplicate function
Wey-Yi Guy [Tue, 23 Feb 2010 22:05:29 +0000 (14:05 -0800)]
iwlwifi: remove duplicate function

"internal_scan" debugfs function serves the same propose of mode "0" of
"force_reset" debugfs function. Remove it to avoid duplicate.

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove needlessly exported symbols
Reinette Chatre [Fri, 19 Feb 2010 19:41:32 +0000 (11:41 -0800)]
iwlwifi: remove needlessly exported symbols

iwl_apm_stop_master and iwl_internal_short_hw_scan
are only used within iwlcore and thus do not need to be exported.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: change WEP key protection to use mutex
Johannes Berg [Fri, 19 Feb 2010 19:42:32 +0000 (11:42 -0800)]
iwlwifi: change WEP key protection to use mutex

For later station notification support we would like WEP key setting to be
done synchronously always. Currently all places from which WEP key is set
can sleep, but the usage of sta_lock prevents it to do so. Modify the
locking to use priv->mutex instead and thus enable this call to sleep.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove un-necessary parameters
Wey-Yi Guy [Thu, 18 Feb 2010 23:22:07 +0000 (15:22 -0800)]
iwlwifi: remove un-necessary parameters

small code cleanup to remove un-necessary parameters

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlagn: move sysfs flags and filter_flags files to debugfs
Johannes Berg [Thu, 18 Feb 2010 08:36:07 +0000 (00:36 -0800)]
iwlagn: move sysfs flags and filter_flags files to debugfs

These files are incompatible with some changes
I'm making, and don't really belong into sysfs
anyway as they can only be used for debugging.
Since writing them will probably crash the
firmware, keep only the reading part and move
them into debugfs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove STATUS_MODE_PENDING
Johannes Berg [Wed, 17 Feb 2010 12:54:08 +0000 (04:54 -0800)]
iwlwifi: remove STATUS_MODE_PENDING

Since rfkill integration, mac80211 can no
longer add an interface while the hardware
is not ready, so STATUS_MODE_PENDING can
never be set.

Also, remove another superfluous channel
sanity check and return the commit_rxon
return value in case it failed.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove alive start adhoc restart
Johannes Berg [Wed, 17 Feb 2010 09:43:41 +0000 (01:43 -0800)]
iwlwifi: remove alive start adhoc restart

This code will never trigger, because when
we call this during interface start, neither
priv->vif nor priv->iw_mode conditions will
be true, and when it happens during reset
then priv->vif is also NULL. Also, in both
cases the same code will be executed as part
of iwl_bss_info_changed(), which contains a
copy of this code.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: move 3945 clip groups to 3945 data
Johannes Berg [Wed, 17 Feb 2010 10:39:19 +0000 (02:39 -0800)]
iwlwifi: move 3945 clip groups to 3945 data

Only used on 3945, so it can be in the
3945-specific part of the data.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: clear up AC/FIFO debug output
Johannes Berg [Fri, 22 Jan 2010 12:06:41 +0000 (04:06 -0800)]
iwlwifi: clear up AC/FIFO debug output

Not all queues are mapped for frame transmission,
so saying "AC" is misleading. They are all mapped
to FIFOs, obviously, though, so use that term and
also say AC/CMD instead of just AC for the queue
type.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove sanity check
Johannes Berg [Thu, 21 Jan 2010 15:31:07 +0000 (07:31 -0800)]
iwlwifi: remove sanity check

iwl_mac_beacon_update can only be called
when the operating mode is IBSS, so it
doesn't need to verify that again.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove IBSS channel sanity check
Johannes Berg [Thu, 21 Jan 2010 15:32:20 +0000 (07:32 -0800)]
iwlwifi: remove IBSS channel sanity check

mac80211 will always enforce using a channel
that the driver allowed for IBSS use.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove priv->active_rate_basic
Johannes Berg [Thu, 21 Jan 2010 19:33:19 +0000 (11:33 -0800)]
iwlwifi: remove priv->active_rate_basic

This variable is assigned a default value,
but then assigned zero as soon as mac80211
calls a change channel (which will happen
right after the hw is started) and after
that it never changes again.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove never-changing priv->rates_mask variable
Johannes Berg [Thu, 21 Jan 2010 19:23:30 +0000 (11:23 -0800)]
iwlwifi: remove never-changing priv->rates_mask variable

It's always just IWL_RATES_MASK.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove frame dropping
Johannes Berg [Thu, 18 Feb 2010 18:56:57 +0000 (10:56 -0800)]
iwlwifi: remove frame dropping

With some of the new code in mac80211, public action
frames can be exchanged as non-injected frames even
while not associated.

Aside from that, dropping frames here is pointless
since we do deal with arbitrary frames that were
injected already, so let mac80211 make the decision
about which frames to allow or not.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove dead code from iwl_mac_reset_tsf
Johannes Berg [Thu, 21 Jan 2010 15:32:58 +0000 (07:32 -0800)]
iwlwifi: remove dead code from iwl_mac_reset_tsf

iwl_mac_reset_tsf will only ever be called in
IBSS mode, so checking for other modes is not
useful and the code that depends on that will
not be executed.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: remove superfluous channel check
Johannes Berg [Thu, 21 Jan 2010 15:26:26 +0000 (07:26 -0800)]
iwlwifi: remove superfluous channel check

mac80211 will not allow using channels not
marked as allowing IBSS for IBSS mode.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlagn: remove write-only variables
Johannes Berg [Tue, 19 Jan 2010 18:04:28 +0000 (10:04 -0800)]
iwlagn: remove write-only variables

Updating the variables last_rx_rssi, last_tsf
and last_beacon_time needs a lot of code but
they are not actually used in iwlagn (only in
3945) so we can move them to the 3945 specific
data.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: move ICT data to agn part of union
Johannes Berg [Thu, 21 Jan 2010 14:25:54 +0000 (06:25 -0800)]
iwlwifi: move ICT data to agn part of union

Since the ICT data is all AGN specific, it can
be the first data to create the _agn part of the
device-specific union in the priv struct.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlwifi: move 3945 specific data into union
Johannes Berg [Thu, 21 Jan 2010 14:09:28 +0000 (06:09 -0800)]
iwlwifi: move 3945 specific data into union

To be obviously correct, this patch only modifies
3945 files and the struct definition. A follow-up
patch will create the AGN part of the union.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agoiwlagn: move ICT code into separate file
Johannes Berg [Thu, 21 Jan 2010 14:21:10 +0000 (06:21 -0800)]
iwlagn: move ICT code into separate file

All the ICT ISR code is iwlagn specific, and doesn't
need to be in iwlcore. So create a new iwl-agn.h
header file that will hold agn specific function
declarations etc., and move the ICT code into a new
iwl-agn-ict.c file that is linked into iwlagn. This
also gets rid of exporting those symbols.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
14 years agob43: N-PHY: switch to chanspec ops
Rafał Miłecki [Sat, 27 Feb 2010 12:03:41 +0000 (13:03 +0100)]
b43: N-PHY: switch to chanspec ops

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agob43: N-PHY: implement chanspec setup
Rafał Miłecki [Sat, 27 Feb 2010 12:03:40 +0000 (13:03 +0100)]
b43: N-PHY: implement chanspec setup

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