GitHub/moto-9609/android_kernel_motorola_exynos9610.git
11 years agomac80211: support drivers reporting VHT RX
Johannes Berg [Fri, 9 Nov 2012 14:07:02 +0000 (15:07 +0100)]
mac80211: support drivers reporting VHT RX

Add support to mac80211 for having drivers report
received VHT MCS information.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agonl80211/cfg80211: add VHT MCS support
Johannes Berg [Fri, 9 Nov 2012 13:56:41 +0000 (14:56 +0100)]
nl80211/cfg80211: add VHT MCS support

Add support for reporting and calculating VHT MCSes.

Note that I'm not completely sure that the bitrate
calculations are correct, nor that they can't be
simplified.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: convert to channel definition struct
Johannes Berg [Fri, 9 Nov 2012 10:39:59 +0000 (11:39 +0100)]
mac80211: convert to channel definition struct

Convert mac80211 (and where necessary, some drivers a
little bit) to the new channel definition struct.

This will allow extending mac80211 for VHT, which is
currently restricted to channel contexts since there
are no drivers using that which makes it easier. As
I also don't care about VHT for drivers not using the
channel context API, I won't convert the previous API
to VHT support.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agonl80211/cfg80211: support VHT channel configuration
Johannes Berg [Thu, 8 Nov 2012 22:14:50 +0000 (23:14 +0100)]
nl80211/cfg80211: support VHT channel configuration

Change nl80211 to support specifying a VHT (or HT)
using the control channel frequency (as before) and
new attributes for the channel width and first and
second center frequency. The old channel type is of
course still supported for HT.

Also change the cfg80211 channel definition struct
to support these by adding the relevant fields to
it (and removing the _type field.)

This also adds new helper functions:
 - cfg80211_chandef_create to create a channel def
   struct given the control channel and channel type,
 - cfg80211_chandef_identical to check if two channel
   definitions are identical
 - cfg80211_chandef_compatible to check if the given
   channel definitions are compatible, and return the
   wider of the two

This isn't entirely complete, but that doesn't matter
until we have a driver using it. In particular, it's
missing
 - regulatory checks on the usable bandwidth (if that
   even makes sense)
 - regulatory TX power (database can't deal with it)
 - a proper channel compatibility calculation for the
   new channel types

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agocfg80211: pass a channel definition struct
Johannes Berg [Thu, 8 Nov 2012 20:25:48 +0000 (21:25 +0100)]
cfg80211: pass a channel definition struct

Instead of passing a channel pointer and channel type
to all functions and driver methods, pass a new channel
definition struct. Right now, this struct contains just
the control channel and channel type, but for VHT this
will change.

Also, add a small inline cfg80211_get_chandef_type() so
that drivers don't need to use the _type field of the
new structure all the time, which will change.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agonl80211: add documentation for channel type
Johannes Berg [Thu, 8 Nov 2012 18:20:56 +0000 (19:20 +0100)]
nl80211: add documentation for channel type

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agocfg80211: remove remain-on-channel channel type
Johannes Berg [Thu, 8 Nov 2012 17:31:02 +0000 (18:31 +0100)]
cfg80211: remove remain-on-channel channel type

As mwifiex (and mac80211 in the software case) are the
only drivers actually implementing remain-on-channel
with channel type, userspace can't be relying on it.
This is the case, as it's used only for P2P operations
right now.

Rather than adding a flag to tell userspace whether or
not it can actually rely on it, simplify all the code
by removing the ability to use different channel types.
Leave only the validation of the attribute, so that if
we extend it again later (with the needed capability
flag), it can't break userspace sending invalid data.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: fix managed mode channel flags handling
Johannes Berg [Mon, 26 Nov 2012 10:57:41 +0000 (11:57 +0100)]
mac80211: fix managed mode channel flags handling

If ieee80211_prep_channel() decides that HT should be
disabled (because the HT IEs from the AP were invalid)
it will set the IEEE80211_STA_DISABLE_HT to not send
HT capabilities to the AP when associating. If this
happens during authentication, the flag will be lost
and we send HT frames, even if the channel config was
set up for non-HT. This can lead to issues.

Fix this by always resetting the ifmgd flags to zero
when the channel context is released so that the flag
resetting in ieee80211_mgd_assoc() isn't necessary.

To make the code a bit easier move the call to release
the channel in ieee80211_set_disassoc() to the end of
the function together with the flag resetting (which
needs to be at the end to avoid timers setting flags.)

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: remove mesh config macros from mesh_plink.c
Marco Porsch [Thu, 22 Nov 2012 02:40:32 +0000 (18:40 -0800)]
mac80211: remove mesh config macros from mesh_plink.c

Use shortcut pointer instead where it is appropriate.

Signed-off-by: Marco Porsch <marco.porsch@etit.tu-chemnitz.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: refactor ieee80211_set_qos_hdr
Marco Porsch [Thu, 22 Nov 2012 02:40:31 +0000 (18:40 -0800)]
mac80211: refactor ieee80211_set_qos_hdr

Return early if not a QoS Data frame.
Give proper documentation.

Signed-off-by: Marco Porsch <marco.porsch@etit.tu-chemnitz.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: move Mesh Capability field definition to ieee80211.h
Marco Porsch [Thu, 22 Nov 2012 02:40:30 +0000 (18:40 -0800)]
mac80211: move Mesh Capability field definition to ieee80211.h

Signed-off-by: Marco Porsch <marco.porsch@etit.tu-chemnitz.de>
[prefix with IEEE80211_]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: in ADHOC print debug message for every Auth message
Antonio Quartulli [Sun, 25 Nov 2012 22:24:27 +0000 (23:24 +0100)]
mac80211: in ADHOC print debug message for every Auth message

The debug message has to be printed also for an Auth message with
auth_sequence != 1. This helps understanding whether the two Auth
messages are exchanged correctly or not.

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: in ADHOC don't update last_rx if sta is not authorized
Antonio Quartulli [Sun, 25 Nov 2012 22:13:42 +0000 (23:13 +0100)]
mac80211: in ADHOC don't update last_rx if sta is not authorized

It does not make sense to keep a station alive if it is not authorised
at all. If IBSS/RSN is used it could also be the case that something
went wrong during the keys exchange and the stations ended up in a not
recoverable state.

By not updating last_rx we are giving the station a chance to be
deleted and to start the key exchange once again from scratch.

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agocfg80211: change function signature of cfg80211_get_p2p_attr()
Arend van Spriel [Sun, 25 Nov 2012 18:13:28 +0000 (19:13 +0100)]
cfg80211: change function signature of cfg80211_get_p2p_attr()

The function cfg80211_get_p2p_attr() can fail and returns
a negative error code. However, the return type is unsigned
int. The largest positive number is determined by desired_len
variable in the function, which is u16. So changing the return
type to int to allow easy error checking. Also change the type
for the attribute to enum for improved type checking.

Signed-off-by: Arend van Spriel <arend@broadcom.com>
[fix indentation, don't use u8 attr variable]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agocfg80211: use DS or HT operation IEs to determine BSS channel
Johannes Berg [Fri, 23 Nov 2012 13:23:30 +0000 (14:23 +0100)]
cfg80211: use DS or HT operation IEs to determine BSS channel

Currently, mac80211 checks the DS params IE if present and
uses it for the (primary) BSS channel, instead of the one
that the frame was received on. This is particularly useful
in the 2.4 GHz band since a frame is often received on one
of the adjacent channels due to overlap.

Move this code to cfg80211 so other drivers also do this.

Additionally, on 5 GHz, in particular with some (possibly)
upcoming changes in 802.11ai and duplicate transmissions
when wider channels are used, something similar happens.
So if present, also use the (primary) channel information
contained in the HT operation IE.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: disable HT advertising unless AP supports it
Johannes Berg [Thu, 22 Nov 2012 13:34:05 +0000 (14:34 +0100)]
mac80211: disable HT advertising unless AP supports it

If the AP doesn't support HT, or more importantly if
it does but we have to disable it because its IEs are
broken, don't advertise HT support in our association
request. Otherwise, we configure our channel to be a
20 MHz non-HT channel but the AP might still think we
support HT, or even 40 MHz.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: rename IEEE80211_STA_DISABLE_11N to HT
Johannes Berg [Thu, 22 Nov 2012 13:32:09 +0000 (14:32 +0100)]
mac80211: rename IEEE80211_STA_DISABLE_11N to HT

Since the 11n spec amendment was rolled into the
2012 version, "11n" no longer makes sense. Use
"HT" instead.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: fix RX chains configuration
Johannes Berg [Thu, 22 Nov 2012 13:17:23 +0000 (14:17 +0100)]
mac80211: fix RX chains configuration

If the driver doesn't support 40 MHz channels, then
mac80211 erroneously sets number of RX chains to one
although the number of chains is independent of the
support for 40 MHz channels.

Fix this by checking the 40 MHz support only for the
code that sets the 40 MHz channel not the complete
HT code block.

This also means the HT20 channel type will always be
set in the changed code block so there's no need to
set it in case we override the AP due to invalid IEs
in the probe response/beacon.

The indentation is a bit quirky, but I'm rewriting
this code for VHT support so this will change again
very soon.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: fix radiotap vendor area skipping
Johannes Berg [Thu, 22 Nov 2012 09:20:58 +0000 (10:20 +0100)]
mac80211: fix radiotap vendor area skipping

The radiotap vendor area in the skb head must be skipped
and accounted for in a few functions until it is removed.
I missed this in my patch, so a few places use this data
as though it was the 802.11 header, fix these places.

Reported-by: Wojciech Dubowik <Wojciech.Dubowik@neratec.com>
Tested-by: Wojciech Dubowik <Wojciech.Dubowik@neratec.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: Add debugfs callbacks for station addition/removal
Sujith Manoharan [Tue, 20 Nov 2012 03:16:02 +0000 (08:46 +0530)]
mac80211: Add debugfs callbacks for station addition/removal

Provide drivers with hooks to create debugfs files when
a new station is added. This would help drivers to take
advantage of mac80211's station list infrastructure and not maintain
tedious station management code internally.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
[ifdef inline wrapper functions]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: fix channel context suspend/reconfig handling
Johannes Berg [Mon, 19 Nov 2012 21:19:08 +0000 (22:19 +0100)]
mac80211: fix channel context suspend/reconfig handling

Sujith reported warnings with suspend/resume due to
channel contexts. When I looked into it, I realised
that the code was completely broken as it unassigned
the channel contexts when suspending, which actually
means they are destroyed.

Eliad Peller then pointed out that we also need to
remove the channel contexts from the driver. When I
looked into this, I also noticed that the code isn't
handling the virtual monitor interface correctly (if
it exists.)

Fix this by calling just the driver methods (if they
are implemented) instead of using the channel context
management code. Also add reconfiguration for the
virtual monitor interface.

Reported-by: Sujith Manoharan <sujith@msujith.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: make remain_on_channel() op pass vif param
Eliad Peller [Mon, 19 Nov 2012 15:05:09 +0000 (17:05 +0200)]
mac80211: make remain_on_channel() op pass vif param

Drivers (e.g. wl12xx) might need to know the vif
to roc on (mainly in order to configure the
rx filters correctly).

Add the vif to the op params, and update the current
users (iwlwifi) to use the new api.

Signed-off-by: Eliad Peller <eliad@wizery.com>
[fix hwsim]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agocfg80211: Add TDLS event to allow drivers to request operations
Jouni Malinen [Fri, 16 Nov 2012 20:49:57 +0000 (22:49 +0200)]
cfg80211: Add TDLS event to allow drivers to request operations

The NL80211_CMD_TDLS_OPER command was previously used only for userspace
request for the kernel code to perform TDLS operations. However, there
are also cases where the driver may need to request operations from
userspace, e.g., when using security on the AP path. Add a new cfg80211
function for generating a TDLS operation event for drivers to request a
new link to be set up (NL80211_TDLS_SETUP) or an existing link to be
torn down (NL80211_TDLS_TEARDOWN). Drivers can optionally use these
events, e.g., based on noticing data traffic being sent to a peer
station that is seen with good signal strength.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: support radiotap vendor namespace RX data
Johannes Berg [Fri, 16 Nov 2012 09:09:08 +0000 (10:09 +0100)]
mac80211: support radiotap vendor namespace RX data

In some cases, in particular for experimentation, it
can be useful to be able to add vendor namespace data
to received frames in addition to the normal radiotap
data.

Allow doing this through mac80211 by adding fields to
the RX status descriptor that describe the data while
the data itself is prepended to the frame.

Also add some example code to hwsim, but don't enable
it because it doesn't use a proper OUI identifier.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: introduce IEEE80211_NUM_TIDS and use it
Johannes Berg [Wed, 14 Nov 2012 22:22:21 +0000 (23:22 +0100)]
mac80211: introduce IEEE80211_NUM_TIDS and use it

Introduce IEEE80211_NUM_TIDS in the generic 802.11
header file and use it in place of STA_TID_NUM and
NUM_RX_DATA_QUEUES which are both really the number
of TIDs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: use CMAC_PN_LEN
Johannes Berg [Wed, 14 Nov 2012 22:15:11 +0000 (23:15 +0100)]
mac80211: use CMAC_PN_LEN

Instead of hardcoding its value (6), use the constant.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: check add_chanctx callback before use in ieee80211_reconfig
Arend van Spriel [Mon, 19 Nov 2012 11:01:05 +0000 (12:01 +0100)]
mac80211: check add_chanctx callback before use in ieee80211_reconfig

During testing our mac80211 driver a fatal error occurred which was
signalled to mac80211. Upon performing the reconfiguration of the
device a WARN_ON was triggered. This warning checked the return value
of drv_add_chanctx(). However, this returns -EOPNOTSUPP when the
driver does not provide the callback. As the callback is optional
better check it is defined before calling drv_add_chanctx().

Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: support RX_FLAG_MACTIME_END
Thomas Pedersen [Tue, 13 Nov 2012 18:46:27 +0000 (10:46 -0800)]
mac80211: support RX_FLAG_MACTIME_END

Allow drivers to indicate their mactime is at RX completion and adjust
for this in mac80211. Also rename the existing RX_FLAG_MACTIME_MPDU to
RX_FLAG_MACTIME_START to clarify its intent. Based on similar code by
Johannes Berg.

Signed-off-by: Thomas Pedersen <thomas@cozybit.com>
[fix docs, atheros drivers]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agonl/cfg80211: advertise OBSS scan requirement
Amitkumar Karwar [Sat, 10 Nov 2012 01:51:30 +0000 (17:51 -0800)]
nl/cfg80211: advertise OBSS scan requirement

wpa_supplicant will do OBSS scan for drivers that implement
auth/assoc API. Drivers that implement nl80211 connect API
(rather than auth/assoc) may need wpa_supplicant to do this
as well.

Add a new feature flag to inform it (wpa_s) that a driver
needs wpa_supplicant to do OBSS scans.

Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: reassign channel contexts before stations
Johannes Berg [Wed, 7 Nov 2012 11:40:41 +0000 (12:40 +0100)]
mac80211: reassign channel contexts before stations

Since channel contexts are usually present before
stations can be added to an interface, reassign
before stations them in reconfiguration as well.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: clarify interface iteration and make it configurable
Johannes Berg [Tue, 6 Nov 2012 19:23:30 +0000 (20:23 +0100)]
mac80211: clarify interface iteration and make it configurable

During hardware restart, all interfaces are iterated even
though they haven't been re-added to the driver, document
this behaviour. The same also happens during resume, which
is even more confusing since all of the interfaces were
previously removed from the driver. Make this optional so
drivers relying on the current behaviour can still use it,
but to let drivers that don't want this behaviour disable
it.

Also convert all API users, keeping the old semantics
except in hwsim, where the new normal ones are desired.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: call driver method when restart completes
Johannes Berg [Tue, 6 Nov 2012 18:18:13 +0000 (19:18 +0100)]
mac80211: call driver method when restart completes

When the driver requests a restart (reconfiguration) it
gets all the normal method calls, but can't really tell
why they're happening. Call a new restart_complete op
in the driver when the restart completes, so it could
keep its own state about the restart and clear it there.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: remove unused tracepoint
Johannes Berg [Thu, 8 Nov 2012 17:16:48 +0000 (18:16 +0100)]
mac80211: remove unused tracepoint

Clearly the tracepoint drv_offchannel_tx was
forgotten when that functionality was removed,
remove it now.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
11 years agomac80211: add debugfs file for HW queues
Johannes Berg [Thu, 8 Nov 2012 14:29:28 +0000 (15:29 +0100)]
mac80211: add debugfs file for HW queues

Add a debugfs file showing which HW queues were
allocated to a virtual interface, including the
CAB queue for AP interfaces.

Change-Id: I486924e961b6ad6785a79db09620919ee644e703
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: fix TX error path
Johannes Berg [Wed, 7 Nov 2012 13:02:30 +0000 (14:02 +0100)]
mac80211: fix TX error path

One error path in ieee80211_subif_start_xmit() will
double-free the SKB. Set it to NULL to prevent that.

This issue was introduced by my channel context
changes.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: include export.h in aes_cmac
Emmanuel Grumbach [Wed, 7 Nov 2012 09:13:58 +0000 (11:13 +0200)]
mac80211: include export.h in aes_cmac

This is needed since this file exports functions.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: use kstrtoull return value
Johannes Berg [Mon, 5 Nov 2012 19:30:39 +0000 (20:30 +0100)]
mac80211: use kstrtoull return value

If kstrtoull() returns an error code (a value
smaller than zero), use it since it can be an
error other than -EINVAL.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: fix race in TKIP MIC test debugfs file
Johannes Berg [Mon, 5 Nov 2012 19:27:57 +0000 (20:27 +0100)]
mac80211: fix race in TKIP MIC test debugfs file

Accessing sdata->vif.bss_conf.bssid without any
protection here is racy, use u.mgd.associated
instead and lock the correct mutex for it.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: use mac_pton
Johannes Berg [Mon, 5 Nov 2012 19:24:38 +0000 (20:24 +0100)]
mac80211: use mac_pton

Instead of implementing practically the same
function (hwaddr_aton) use mac_pton.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agonl80211: send the NL80211_ATTR_SSID in nl80211_send_iface()
Antonio Quartulli [Wed, 7 Nov 2012 11:52:20 +0000 (12:52 +0100)]
nl80211: send the NL80211_ATTR_SSID in nl80211_send_iface()

The userspace may want to know what is the current ssid that a given
interface is using. This patch enables nl80211 to send the
NL80211_ATTR_SSID attribute in nl80211_send_iface().

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agocfg80211: store the ssid into wirless_dev in AP mode
Antonio Quartulli [Wed, 7 Nov 2012 11:52:19 +0000 (12:52 +0100)]
cfg80211: store the ssid into wirless_dev in AP mode

Store the configured ssid in wdev->ssid when starting an AP

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: pass P2P powersave parameters to driver
Johannes Berg [Mon, 29 Oct 2012 19:08:01 +0000 (20:08 +0100)]
mac80211: pass P2P powersave parameters to driver

While connected to a GO, parse the P2P NoA attribute
and pass the CT Window and opportunistic powersave
parameters to the driver.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agowireless: add utility function to get P2P attribute
Johannes Berg [Mon, 29 Oct 2012 18:48:40 +0000 (19:48 +0100)]
wireless: add utility function to get P2P attribute

Parsing the P2P attributes can be tricky as their
contents can be split across multiple (vendor) IEs.
Thus, it's not possible to parse them like IEs (by
returning a pointer to the data.) Instead, provide
a function that copies the attribute data into a
caller-provided buffer and returns the size needed
(useful in case the buffer was too small.)

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agocfg80211: allow registering more than one beacon listener
Ben Greear [Fri, 26 Oct 2012 21:49:25 +0000 (14:49 -0700)]
cfg80211: allow registering more than one beacon listener

The commit:

commit 5e760230e42cf759bd923457ca2753aacf2e656e
Author: Johannes Berg <johannes.berg@intel.com>
Date:   Fri Nov 4 11:18:17 2011 +0100

    cfg80211: allow registering to beacons

allowed only a single process to register for beacon events
per wiphy.  This breaks cases where a user may want two or
more VIFs on a wiphy and run a seperate hostapd process on
each vif.

This patch allows multiple beacon listeners, fixing the
regression.

Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: implement set_mcast_rate() callback
Antonio Quartulli [Fri, 2 Nov 2012 12:27:49 +0000 (13:27 +0100)]
mac80211: implement set_mcast_rate() callback

This new callback can be used to tune the rate to be used to send
multicast frames.

In the current state the multicast rate can be specified on IBSS/MESH
joining only. This makes it impossible to select a custom multicast
rate when then join command is sent by an external program (e.g.
wpa_supplicant)

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agonl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command
Antonio Quartulli [Fri, 2 Nov 2012 12:27:48 +0000 (13:27 +0100)]
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command

This command triggers a new callback: set_mcast_rate(). It enables
the user to change the rate used to send multicast frames for vif
configured as IBSS or MESH_POINT

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agowireless: add peer-to-peer related definitions
Arend van Spriel [Mon, 5 Nov 2012 14:29:09 +0000 (15:29 +0100)]
wireless: add peer-to-peer related definitions

The Peer-to-Peer IE is vendor-specific IE identified by WiFi Alliance
OUI and specific P2P OUI type. The payload of this IE consists of
so-called P2P attributes. This patch adds definitions for processing
these attributes.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: send deauth only with channel context
Johannes Berg [Mon, 29 Oct 2012 08:46:31 +0000 (09:46 +0100)]
mac80211: send deauth only with channel context

When userspace asks to deauthenticate and we're just
authenticated (or still authenticating) send a deauth
frame instead of deleting the auth request.

On the other hand, if we've just disassociated and
therefore deleted all our state already, drop the
deauth request because we no longer have a channel
context to send it on.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: use a counter for remain-on-channel cookie
Johannes Berg [Fri, 26 Oct 2012 14:13:06 +0000 (16:13 +0200)]
mac80211: use a counter for remain-on-channel cookie

Instead of using the pointer which can be re-used
fairly quickly due to allocator patterns and then
makes debugging difficult, maintain a counter and
use its value. Since it's a 64-bit value it can't
really wrap, but catch that case anyway since it
most likely points to a bug somewhere.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: combine status/drop reporting
Johannes Berg [Fri, 26 Oct 2012 13:53:06 +0000 (15:53 +0200)]
mac80211: combine status/drop reporting

The TX status reporting is done for both the
nl80211 report as well as the socket option.
The socket option is also reported when an
skb is dropped to guarantee that the copy in
the IDR tree is freed and status is reported
to userspace.

However, when a frame is dropped, no nl80211
status is reported. This can cause userspace
to stop making progress while waiting for a
status notification.

Combine the nl80211 and socket option status
reporting into a new function and call it in
both places -- when the status comes in from
the driver and when the skb is dropped.

While at it, also simplify the code in the
nl80211 portion a bit.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211_hwsim: print per interface TX power
Johannes Berg [Wed, 24 Oct 2012 09:54:31 +0000 (11:54 +0200)]
mac80211_hwsim: print per interface TX power

Just for debugging, print the interface TX
power whenever it changes.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211: handle TX power per virtual interface
Johannes Berg [Wed, 24 Oct 2012 08:59:25 +0000 (10:59 +0200)]
mac80211: handle TX power per virtual interface

Even before channel contexts/multi-channel, having a
single global TX power limit was already problematic,
in particular if two managed interfaces connected to
two APs with different power constraints. The channel
context introduction completely broke this though and
in fact I had disabled TX power configuration there
for drivers using channel contexts.

Change everything to track TX power per interface so
that different user settings and different channel
maxima are treated correctly. Also continue tracking
the global TX power though for compatibility with
applications that attempt to configure the wiphy's
TX power globally.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agocfg80211: allow per interface TX power setting
Johannes Berg [Wed, 24 Oct 2012 08:17:18 +0000 (10:17 +0200)]
cfg80211: allow per interface TX power setting

The TX power setting is currently per wiphy (hardware
device) but with multi-channel capabilities that doesn't
make much sense any more.

Allow drivers (and mac80211) to advertise support for
per-interface TX power configuration. When the TX power
is configured for the wiphy, the wdev will be NULL and
the driver can still handle that, but when a wdev is
given the TX power can be set only for that wdev now.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agonl80211: move "can set channel" check
Johannes Berg [Wed, 24 Oct 2012 08:04:58 +0000 (10:04 +0200)]
nl80211: move "can set channel" check

Setting the wdev to NULL when the channel can't be
set for that interface type (to treat the channel
setting for the wiphy/monitor) currently works, but
is confusing in the code if netdev/wdev aren't both
set/unset in the same way. Move the check whether
the channel can be set to where it's needed so that
wdev and netdev are always both assigned or NULL.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agomac80211_hwsim: allow using channel contexts
Johannes Berg [Fri, 27 Jul 2012 17:48:26 +0000 (19:48 +0200)]
mac80211_hwsim: allow using channel contexts

To use mac80211_hwsim for testing channel contexts it
has to support them, and for that it has to support
hw scan and hw-remain-on-channel.

Since it's pure software, the off-channel activities
are really not off-channel but listening and sending
on a second channel. Also, the multi-channel isn't
really doing TDM, it's just on both channels at the
same time.

For testing purposes, you can specify the number of
concurrent channels with a module parameter, it is
set to one by default. When set to two or more, the
userspace API for wmediumd is disabled as it has no
provisions for multi-channel yet.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
12 years agoMerge remote-tracking branch 'wireless-next/master' into mac80211-next
Johannes Berg [Tue, 30 Oct 2012 08:09:48 +0000 (09:09 +0100)]
Merge remote-tracking branch 'wireless-next/master' into mac80211-next

12 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless
John W. Linville [Mon, 29 Oct 2012 20:05:51 +0000 (16:05 -0400)]
Merge branch 'master' of git://git./linux/kernel/git/linville/wireless

Conflicts:
drivers/net/wireless/mwifiex/cfg80211.c

12 years agossb: handle BCM43222 in pmu code.
Hauke Mehrtens [Thu, 25 Oct 2012 16:39:29 +0000 (18:39 +0200)]
ssb: handle BCM43222 in pmu code.

The BCM43222 with the chipid 43222 or 0xa8d6 in hex do not need any
special handling in the pmu code. This prevents some error messages
being shown.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agossb: add PCI ID 0x4350
Hauke Mehrtens [Thu, 25 Oct 2012 16:39:28 +0000 (18:39 +0200)]
ssb: add PCI ID 0x4350

14e4:4350 is found on a Broadcom BCM43222.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agoath9k: Dump BTCOEX tuning parameters
Rajkumar Manoharan [Thu, 25 Oct 2012 11:46:54 +0000 (17:16 +0530)]
ath9k: Dump BTCOEX tuning parameters

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agoath9k_hw: validate MCI stuck after RTC wakeup
Rajkumar Manoharan [Thu, 25 Oct 2012 11:46:53 +0000 (17:16 +0530)]
ath9k_hw: validate MCI stuck after RTC wakeup

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agoath9k_hw: Fix concurrent tx on lower tx power
Rajkumar Manoharan [Thu, 25 Oct 2012 11:46:52 +0000 (17:16 +0530)]
ath9k_hw: Fix concurrent tx on lower tx power

Whenever WLAN receives scheduling msg from BT, it reduces tx power
based on RSSI level. And then BT starts simultaneous transmission
along with WLAN. Sometimes HW MAC compares tx power that is used
prior to power reduction which is causing BT transmission to defer.

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agoath9k_hw: Enable hw PLL power save for AR9565
Rajkumar Manoharan [Thu, 25 Oct 2012 11:46:51 +0000 (17:16 +0530)]
ath9k_hw: Enable hw PLL power save for AR9565

This reduced the power consumption to half in full and network sleep.

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agoath9k_hw: Enable hw PLL power save for AR9462
Rajkumar Manoharan [Thu, 25 Oct 2012 11:41:31 +0000 (17:11 +0530)]
ath9k_hw: Enable hw PLL power save for AR9462

This reduced the power consumption to half in full and network sleep.

Cc: stable@vger.kernel.org
Cc: Paul Stewart <pstew@chromium.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agoar5523: Add new driver
Pontus Fuchs [Tue, 23 Oct 2012 18:33:57 +0000 (20:33 +0200)]
ar5523: Add new driver

This driver is for the AR5523 chipset from Atheros. It was created
in 2007 by Christoph Hellwig but it was never finished. I found it a
couple of months ago and after some polishing it's working pretty
fine.

The driver was written with the FreeBSD driver (uath) as reference,
which was written with the reverse-engineered windows driver as
reference, hence the feature set is very limited. Station mode
only, no HW crypto offload.

Signed-off-by: Pontus Fuchs <pontus.fuchs@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: store IEs per virtual interface
Arend van Spriel [Mon, 22 Oct 2012 20:55:41 +0000 (13:55 -0700)]
brcmfmac: store IEs per virtual interface

For AP feature the IEs are stored in global structure. For future
functionality like P2P-GO it needs to be stored per virtual interface
so better store it in the virtual interface structure.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: remove unnecessary macro usage in brcmf_cfg80211_resume()
Arend van Spriel [Mon, 22 Oct 2012 20:55:40 +0000 (13:55 -0700)]
brcmfmac: remove unnecessary macro usage in brcmf_cfg80211_resume()

The macro wiphy_to_cfg() is a bit redundant as the function already
has a pointer variable to brcmf_cfg80211_info structure.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: add virtual interface support in brcmf_cfg80211_suspend()
Arend van Spriel [Mon, 22 Oct 2012 20:55:39 +0000 (13:55 -0700)]
brcmfmac: add virtual interface support in brcmf_cfg80211_suspend()

With multiple interfaces suspend will need to iterate over all and
bring down the link.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: use memset when setting a broadcast mac address
Arend van Spriel [Mon, 22 Oct 2012 20:55:38 +0000 (13:55 -0700)]
brcmfmac: use memset when setting a broadcast mac address

The driver had a global constant ether_bcast, which was copied
whenever a broadcast mac address was needed. This patch does a
memset(dest, 0xFF, ETH_ALEN) instead and consequently removes
the global ether_bcast.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: rename check_sys_up() to check_vif_up()
Arend van Spriel [Mon, 22 Oct 2012 20:55:37 +0000 (13:55 -0700)]
brcmfmac: rename check_sys_up() to check_vif_up()

The function now return the status for a virtual interface so
it seems better rename the function to understand what it does.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: remove unused enumeration wl_prof_list
Arend van Spriel [Mon, 22 Oct 2012 20:55:36 +0000 (13:55 -0700)]
brcmfmac: remove unused enumeration wl_prof_list

The enumeration wl_prof_list is no longer used so it can be safely
removed.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: cleanup brcmf_cfg80211_profile structure
Arend van Spriel [Mon, 22 Oct 2012 20:55:35 +0000 (13:55 -0700)]
brcmfmac: cleanup brcmf_cfg80211_profile structure

A couple of unused fields have been removed and kernel-doc info has
been added to the structure.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: remove debugfs functionality from wl_cfg80211.c
Arend van Spriel [Mon, 22 Oct 2012 20:55:34 +0000 (13:55 -0700)]
brcmfmac: remove debugfs functionality from wl_cfg80211.c

In wl_cfg80211.c debugfs directory was created to expose
dtim_period and beacon_interval. However, this can be easily
obtained using iw so it is removed from the driver.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: separate connection status from scanning status
Arend van Spriel [Mon, 22 Oct 2012 20:55:33 +0000 (13:55 -0700)]
brcmfmac: separate connection status from scanning status

The connection status is to be kept per virtual interface and
the scanning status is for device. So they need to be separated
for multiple interface support.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: use vif struct to check_sys_up() function
Arend van Spriel [Mon, 22 Oct 2012 20:55:32 +0000 (13:55 -0700)]
brcmfmac: use vif struct to check_sys_up() function

This checks the status that will soon be moved to virtual
interface data so preparing for that use the structure
brcmf_cfg80211_vif as parameter instead.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meulemen <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: store profile information per virtual interface
Arend van Spriel [Mon, 22 Oct 2012 20:55:31 +0000 (13:55 -0700)]
brcmfmac: store profile information per virtual interface

The profile information applies to an interface so each virtual
interface needs it. So it is removed from brcmf_cfg80211_info
and added to brcmf_cfg80211_vif structure.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: introduce brcmf_cfg80211_vif structure
Arend van Spriel [Mon, 22 Oct 2012 20:55:30 +0000 (13:55 -0700)]
brcmfmac: introduce brcmf_cfg80211_vif structure

This patch introduces the brcmf_cfg80211_vif structure which is
used to keep track of multiple virtual interfaces in the driver.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: remove brcmf_find_bssidx() function
Arend van Spriel [Mon, 22 Oct 2012 17:36:27 +0000 (10:36 -0700)]
brcmfmac: remove brcmf_find_bssidx() function

The function brcmf_find_bssidx() is no longer used so remove
it from the driver code.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: change parameter list for send_key_to_dongle()
Arend van Spriel [Mon, 22 Oct 2012 17:36:26 +0000 (10:36 -0700)]
brcmfmac: change parameter list for send_key_to_dongle()

The first two parameters given to send_key_to_dongle() are redundant
so they have been removed.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: use fwil for default configuration setup.
Hante Meuleman [Mon, 22 Oct 2012 17:36:25 +0000 (10:36 -0700)]
brcmfmac: use fwil for default configuration setup.

modify the setup code to use the refactored firmware interface layer.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: streamline header parse code of sdio glom read
Franky Lin [Mon, 22 Oct 2012 17:36:24 +0000 (10:36 -0700)]
brcmfmac: streamline header parse code of sdio glom read

Use brcmf_sdio_hdparser to handle header of super frame and sub
frame in glomming frame read.

Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: use struct brcmf_if as interface object for fwil functions
Arend van Spriel [Mon, 22 Oct 2012 17:36:23 +0000 (10:36 -0700)]
brcmfmac: use struct brcmf_if as interface object for fwil functions

The functions for communicating were given the net_device only because
its private data contained struct brcmf_if object. However, not all
firmware related interfaces will be associated with a net_device. To
accomodate provisioning firmware for such interfaces the struct
brcmf_if object will be passed to the fwil functions.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: add function converting ieee80211_channel to chanspec
Arend van Spriel [Mon, 22 Oct 2012 17:36:22 +0000 (10:36 -0700)]
brcmfmac: add function converting ieee80211_channel to chanspec

The firmware carries channel information in a different format
as the provided ieee80211_channel structure. Conversion is needed
when receiving requests from cfg80211 carrying ieee80211_channel
structures. This patch adds a utility function to do that.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: use bssidx from struct brcmf_if for bsscfg specific commands
Arend van Spriel [Mon, 22 Oct 2012 17:36:21 +0000 (10:36 -0700)]
brcmfmac: use bssidx from struct brcmf_if for bsscfg specific commands

The firmware interface has functions to send bsscfg specific commands
to the device. These functions currently have a bssidx parameter, but
that same information is stored in struct brcmf_if, which is in the
private data of the net_device parameter.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: rework driver initialization in brcmf_bus_start()
Arend van Spriel [Mon, 22 Oct 2012 17:36:20 +0000 (10:36 -0700)]
brcmfmac: rework driver initialization in brcmf_bus_start()

In brcmf_bus_start() a number of settings are sent to the device. For
this functions are used that bypass the common firmware interface.
By reordering the code in brcmf_bus_start() this bypass can be removed.

Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: extend struct brcmf_if with bssidx field
Arend van Spriel [Mon, 22 Oct 2012 17:36:19 +0000 (10:36 -0700)]
brcmfmac: extend struct brcmf_if with bssidx field

When the firmware notifies the driver about adding a new interface
it also provides an index for the bss associated with this interface.
This index will be needed for upcoming features like peer-to-peer.
By adding this index in struct brcmf_if it is easy to obtain as this
will be associated with the net_device private data.

Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: clean usb download code.
Hante Meuleman [Mon, 22 Oct 2012 17:36:18 +0000 (10:36 -0700)]
brcmfmac: clean usb download code.

reuse ioctl waiting method.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: remove redundant function brcmf_c_mkiovar_bsscfg
Hante Meuleman [Mon, 22 Oct 2012 17:36:17 +0000 (10:36 -0700)]
brcmfmac: remove redundant function brcmf_c_mkiovar_bsscfg

function brcmf_c_mkiovar_bsscfg became redundant with refactoring
of firmware interface layer.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: change testmode command to use new firmware interface layer
Hante Meuleman [Mon, 22 Oct 2012 17:36:16 +0000 (10:36 -0700)]
brcmfmac: change testmode command to use new firmware interface layer

switch to new firmware interface layer and remove redundant code.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: remove unused iswl variable.
Hante Meuleman [Mon, 22 Oct 2012 17:36:15 +0000 (10:36 -0700)]
brcmfmac: remove unused iswl variable.

Variable iswl always gets initialised to the same and support for
this var is not needed.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agobrcmfmac: refactor firmware interface layer.
Hante Meuleman [Mon, 22 Oct 2012 17:36:14 +0000 (10:36 -0700)]
brcmfmac: refactor firmware interface layer.

Refactor the functions that are related to getting and setting
data to and and from the firmware.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agocarl9170: split up carl9170_handle_mpdu
Christian Lamparter [Mon, 22 Oct 2012 13:01:29 +0000 (15:01 +0200)]
carl9170: split up carl9170_handle_mpdu

carl9170_handle_mpdu is the final part of the
rx path of the driver. It splits the raw data
streams from the device into skb packets and
passes them on to mac80211. As a result of
continuous updates, it grew over the years when
new code was added by the following commits:
- report A-MPDU status
- fix HT peer BA session corruption
- A-MPDU frame type filter
- ...

This patch splits the routine into two stages.
The first stage only deals with the details
about extracting and verifying the data from
the incoming stream. Whereas the second stage
packs it into skbs and passes it on to mac80211.

Reported-by: Javier Lopez <jlopex@cozybit.com>
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agomwifiex: rx path enhancement to derive priv only once
Avinash Patil [Sat, 20 Oct 2012 02:19:21 +0000 (19:19 -0700)]
mwifiex: rx path enhancement to derive priv only once

We derive mwifiex_private structure which is per interface from
received skb's rx_info. Once priv is derived, same priv can be
propagated to other functions instead of callee deriving priv
from rx_info again.

Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agomwifiex: handle extended supported rates IE for AP
Avinash Patil [Sat, 20 Oct 2012 02:19:20 +0000 (19:19 -0700)]
mwifiex: handle extended supported rates IE for AP

During start_ap handler, some rates come as extended supported
rates IE - part of beacon tail IE. This patch adds support for
parsing them and adding to bss_rates TLV for bss_start command
to firmware.

Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
12 years agoRevert "mwifiex: retrieve correct max_power information in reg_notifier handler"
Amitkumar Karwar [Sat, 20 Oct 2012 02:19:19 +0000 (19:19 -0700)]
Revert "mwifiex: retrieve correct max_power information in reg_notifier handler"

This reverts commit 34202e28fe7fc8551313f9a035a8857db83de757

We made "34202ee.." because we didn't support custom regulatory rules
at that time. But now we use our own custom regulatory rules, so it
needs to be changed back.

Also, chan->max_power calculations in cfg80211 were broken. Hence we
started using chan->max_reg_power. Now it has got fixed in
following commit.

commit 5e31fc0815a4e2c72b1b495fe7a0d8f9bfb9e4b4
Author: Stanislaw Gruszka <sgruszka@redhat.com>
Date:   Tue Jul 24 08:35:39 2012 +0200

    wireless: reg: restore previous behaviour of chan->max_power calculations

Hence we will use chan->max_power instead of chan->max_reg_power.

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>
12 years agomwifiex: disable channel filtering for SSID specific scan from user
Amitkumar Karwar [Sat, 20 Oct 2012 02:19:18 +0000 (19:19 -0700)]
mwifiex: disable channel filtering for SSID specific scan from user

If MWIFIEX_DISABLE_CHAN_FILT bit in scan mode bitmap is set,
firmware will turn off the filtering of scan responses from
adjacent channels.

Currently the bit is set only for internal SSID specific scan
performed during association. We will set it for user requested
SSID specific scan as well.

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>
12 years agomwifiex: minor cleanup and a fix in scan semaphore usage
Amitkumar Karwar [Sat, 20 Oct 2012 02:19:17 +0000 (19:19 -0700)]
mwifiex: minor cleanup and a fix in scan semaphore usage

mwifiex_request_scan() takes care of synchronous internal scan
performed by driver during association.
Currently the semaphore acquired for the scan is unnecessarily
released at the end of different paths. Also, failure paths
returning error code other than "-1" are not considered.

We will release it at the end of routine to fix above issues.

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>
12 years agomwifiex: abort scan upon interface down
Amitkumar Karwar [Sat, 20 Oct 2012 02:19:16 +0000 (19:19 -0700)]
mwifiex: abort scan upon interface down

When the interface is down, we will abort scan by calling
cfg80211_scan_done() with abort option. This fixes WARN_ON
triggered by cfg80211 in wdev_cleanup_work().

Driver's internal variables/flags are cleared once we get
response for current scan command. Meanwhile we will block
new scan request from cfg80211.

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>
12 years agomwifiex: use LOW_PRIORITY scan flag provided in scan request
Amitkumar Karwar [Sat, 20 Oct 2012 02:19:15 +0000 (19:19 -0700)]
mwifiex: use LOW_PRIORITY scan flag provided in scan request

We will delay/abort scan operation based on traffic for low
priority scan.

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>
12 years agobcma: suspend/resume callbacks should be conditionally compiled on CONFIG_PM_SLEEP
Yuanhan Liu [Tue, 16 Oct 2012 14:59:02 +0000 (22:59 +0800)]
bcma: suspend/resume callbacks should be conditionally compiled on CONFIG_PM_SLEEP

This will fix warnings like following when CONFIG_PM_SLEEP is not set:

        warning: 'xxx_suspend' defined but not used [-Wunused-function]
        warning: 'xxx_resume' defined but not used [-Wunused-function]

Because
SET_SYSTEM_SLEEP_PM_OPS(suspend_fn, resume_fn)

Only references the callbacks on CONFIG_PM_SLEEP (instead of CONFIG_PM).

Cc: John W. Linville <linville@tuxdriver.com>
Cc: "Rafał Miłecki" <zajec5@gmail.com>
Cc: linux-wireless@vger.kernel.org
Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>