GitHub/exynos8895/android_kernel_samsung_universal8895.git
10 years agoBluetooth: Remove connection interval parameters from hci_conn_params_set
Marcel Holtmann [Tue, 1 Jul 2014 10:11:06 +0000 (12:11 +0200)]
Bluetooth: Remove connection interval parameters from hci_conn_params_set

The connection interval parameter of hci_conn_params_set are always used
with the controller defaults. So just let hci_conn_params_add set the
controller default and not bother resetting them to controller defaults
every time the hci_conn_params_set is called.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Use hci_conn_params_add within hci_conn_params_set
Marcel Holtmann [Tue, 1 Jul 2014 10:11:05 +0000 (12:11 +0200)]
Bluetooth: Use hci_conn_params_add within hci_conn_params_set

The hci_conn_params_add function provides the default allocation of
connection parameters. To avoid code duplication, use that code from
hci_conn_params_set to allocate or lookup parameter struct.

As a benefit the connection latency and supervision timeout parameters
are no longer reset to default when calling hci_conn_params_set.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Change hci_conn_params_add to return the parameter struct
Marcel Holtmann [Tue, 1 Jul 2014 10:11:04 +0000 (12:11 +0200)]
Bluetooth: Change hci_conn_params_add to return the parameter struct

When adding new connection parameters, it is useful to return either
the existing struct or the newly created one.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Use bool for smp_ltk_encrypt return value
Marcel Holtmann [Tue, 1 Jul 2014 08:59:24 +0000 (10:59 +0200)]
Bluetooth: Use bool for smp_ltk_encrypt return value

The return value of smp_ltk_encrypt is simple boolean, so just use
bool and make the code a bit more readable.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Fallback to SCO on error code 0x10 (Connection Accept Timeout)
Nick Pelly [Mon, 30 Jun 2014 05:55:01 +0000 (11:25 +0530)]
Bluetooth: Fallback to SCO on error code 0x10 (Connection Accept Timeout)

This is to support the Motorola HF850 carkit which reports the error
code 0x10 for an eSCO attempt, even though it advertises eSCO support.

With this patch we will retry with a SCO connection, which succeeds.

Signed-off-by: Nick Pelly <npelly@google.com>
Signed-off-by: Kiran Kumar Raparthy <kiran.kumar@linaro.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Allow L2CAP getpeername() for BT_CONFIG state
Johan Hedberg [Tue, 1 Jul 2014 09:07:23 +0000 (12:07 +0300)]
Bluetooth: Allow L2CAP getpeername() for BT_CONFIG state

We have all the necessary remote information for getpeername() when we
are in the BT_CONFIG state so this should be allowed. This is
particularly important for LE sockets where changing the security level
will temporarily move the socket into BT_CONFIG state.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: constify seq_operations
Fabian Frederick [Mon, 30 Jun 2014 17:26:23 +0000 (19:26 +0200)]
Bluetooth: constify seq_operations

bt_seq_ops is only used with __seq_open_private as
const struct seq_operations *

Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Connection parameters check helper
Andre Guedes [Thu, 26 Jun 2014 00:52:52 +0000 (21:52 -0300)]
Bluetooth: Connection parameters check helper

This patch renames l2cap_check_conn_param() to hci_check_conn_params()
and moves it to hci_core.h so it can reused in others files. This helper
will be reused in the next patch.

Signed-off-by: Andre Guedes <andre.guedes@openbossa.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Start background scanning only when controller is ready
Marcel Holtmann [Mon, 30 Jun 2014 14:04:12 +0000 (16:04 +0200)]
Bluetooth: Start background scanning only when controller is ready

When the controller is not active or in init/setup phase, do not
try to start or stop background scanning.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Add default connection parameters before pairing
Marcel Holtmann [Mon, 30 Jun 2014 10:34:40 +0000 (12:34 +0200)]
Bluetooth: Add default connection parameters before pairing

When trying to pair a new Bluetooth Low Energy device, then make sure
that the default connections parameters are in place before trying to
establish the first connection to that device. With the connection
parameters structure allocated, the slave preferred values can now
easily be tracked and all future connections will use the correct
values from that start decreasing connection establishment time.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Provide function to create and set connection parameters
Marcel Holtmann [Mon, 30 Jun 2014 10:34:39 +0000 (12:34 +0200)]
Bluetooth: Provide function to create and set connection parameters

In some cases it is useful to not overwrite connection parametes and
instead just create default ones if they don't exist. This function
does exactly that. hci_conn_params_add will allow to create new
default connection parameters. hci_conn_params_set will set the
values and also create new parameters if they don't exist.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Expose default supervision timeout setting via debugfs
Marcel Holtmann [Mon, 30 Jun 2014 10:34:38 +0000 (12:34 +0200)]
Bluetooth: Expose default supervision timeout setting via debugfs

The controller has a default value for the supervision timeout. Expose
this via debugfs for testing purposes.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Expose default connection latency setting via debugfs
Marcel Holtmann [Mon, 30 Jun 2014 10:34:37 +0000 (12:34 +0200)]
Bluetooth: Expose default connection latency setting via debugfs

The controller has a default value for the connection latency. Expose
this via debugfs for testing purposes.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Provide defaults for LE connection latency and timeout
Marcel Holtmann [Mon, 30 Jun 2014 10:34:36 +0000 (12:34 +0200)]
Bluetooth: Provide defaults for LE connection latency and timeout

Store the connection latency and supervision timeout default values
with all the other controller defaults. And when needed use them
for new connections.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Add Device Added and Device Removed management events
Marcel Holtmann [Sun, 29 Jun 2014 20:28:34 +0000 (22:28 +0200)]
Bluetooth: Add Device Added and Device Removed management events

When devices are added or removed, then make sure that events are send
out to all other clients so that the list of devices can be easily
tracked. This is especially important when external clients are
adding or removing devices within the auto-connection list.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Replace le_auto_conn debugfs with device_list entry
Marcel Holtmann [Sun, 29 Jun 2014 14:15:49 +0000 (16:15 +0200)]
Bluetooth: Replace le_auto_conn debugfs with device_list entry

Since the auto-connection handling has gained offical management
command support, remove the le_auto_conn debugfs entry.

For debugging purposes replace it a simple device_list debugfs
entry that allows listing of the current internal auto-connection
list used for passive scanning.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Add support for Add/Remove Device management commands
Marcel Holtmann [Sun, 29 Jun 2014 17:44:03 +0000 (19:44 +0200)]
Bluetooth: Add support for Add/Remove Device management commands

This allows adding or removing devices from the background scanning
list the kernel maintains. Device flagged for auto-connection will
be automatically connected if they are found.

The passive scanning required for auto-connection will be started
and stopped on demand.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Use LE connection parameters if known
Marcel Holtmann [Sun, 29 Jun 2014 14:43:27 +0000 (16:43 +0200)]
Bluetooth: Use LE connection parameters if known

When the LE connection parameters for connection latency and
supervision timeout are known, then use then. If they are not
know fallback to defaults.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Store latency and supervision timeout in connection params
Marcel Holtmann [Sun, 29 Jun 2014 14:43:26 +0000 (16:43 +0200)]
Bluetooth: Store latency and supervision timeout in connection params

When the slave updates the connection parameters, store also the
connection latency and supervision timeout information in the
internal list of connection parameters for known devices.

Having these values available allowes the auto-connection
procedure to use the correct values from the beginning without
having to request an update on every connection establishment.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Update background scanning from hci_conn_params_clear
Marcel Holtmann [Sun, 29 Jun 2014 11:41:51 +0000 (13:41 +0200)]
Bluetooth: Update background scanning from hci_conn_params_clear

When calling hci_conn_params_clear function, it should update the
background scanning properly and not require a separate call to
update it.

For the case when the function is used during unregister of a
controller, an extra safe guard is but in place.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Clear pending connections from hci_conn_params_clear
Marcel Holtmann [Sun, 29 Jun 2014 11:41:50 +0000 (13:41 +0200)]
Bluetooth: Clear pending connections from hci_conn_params_clear

When hci_conn_params_clear is called, it is always followed by a
call to hci_pend_le_conns_clear. So instead of making this explicit
just make sure it is always called. This makes this function similar
on how hci_conn_params_add and hci_conn_params_del work.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Move hci_pend_le_conn_* functions to different location
Marcel Holtmann [Sun, 29 Jun 2014 11:41:49 +0000 (13:41 +0200)]
Bluetooth: Move hci_pend_le_conn_* functions to different location

The hci_pend_le_conn_* function should be placed before their actual
users. So move them before hci_conn_params_* functions.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Don't use non-resolvable private address for passive scanning
Marcel Holtmann [Sun, 29 Jun 2014 10:20:15 +0000 (12:20 +0200)]
Bluetooth: Don't use non-resolvable private address for passive scanning

The usage of non-resovlable private addresses for passive scanning is
a bad idea. Passive scanning will not send any SCAN_REQ and thus using
your identity address for passive scanning is not a privacy issue.

It is important to use the identity address during passive scanning
since that is the only way devices using direct advertising will be
reported correctly by the controller. This is overlooked detail in
the Bluetooth specification that current controllers are not able
to report direct advertising events for other than their current
address.

When remote peers are using direct advertising and scanning is done
with non-resolvable private address these devices will not be found.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Restrict access for raw-only controllers
Marcel Holtmann [Sun, 29 Jun 2014 10:13:05 +0000 (12:13 +0200)]
Bluetooth: Restrict access for raw-only controllers

Bluetooth controllers that are marked for raw-only usage can only be
used with user channel access. Any other operation should be rejected.

This simplifies the whole raw-only support since it now depends on
the fact that the controller is marked with HCI_QUIRK_RAW_DEVICE and
runtime raw access is restricted to user channel operation.

The kernel internal processing of HCI commands and events is designed
around the case that either the kernel has full control over the device
or that the device is driven from userspace. This now makes a clear
distinction between these two possible operation modes.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Add support for Get Clock Info mgmt command
Johan Hedberg [Sat, 28 Jun 2014 14:54:07 +0000 (17:54 +0300)]
Bluetooth: Add support for Get Clock Info mgmt command

This patch implements support for the Get Clock Information mgmt
command. This is done by performing one or two HCI_Read_Clock commands
and creating the response from the stored values in the hci_dev and
hci_conn structs.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Add tracking of local and piconet clock values
Johan Hedberg [Sat, 28 Jun 2014 14:54:06 +0000 (17:54 +0300)]
Bluetooth: Add tracking of local and piconet clock values

This patch adds support for storing the local and piconet clock values
from the HCI_Read_Clock command response to the hci_dev and hci_conn
structs. This will be later used in another patch to implement support
for the Get Clock Info mgmt command.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Use kzalloc instead of kmalloc for pending mgmt commands
Johan Hedberg [Sat, 28 Jun 2014 14:54:05 +0000 (17:54 +0300)]
Bluetooth: Use kzalloc instead of kmalloc for pending mgmt commands

By using kzalloc we ensure that there are no struct members, such as the
user_data pointer, left uninitialized.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Increment management interface revision
Marcel Holtmann [Sat, 28 Jun 2014 10:36:10 +0000 (12:36 +0200)]
Bluetooth: Increment management interface revision

This patch increments the management interface revision due to the
changes with the debug key command and other fixes.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Send HCI_Read_Clock_Offset before disconnecting
Marcel Holtmann [Fri, 27 Jun 2014 12:32:16 +0000 (14:32 +0200)]
Bluetooth: Send HCI_Read_Clock_Offset before disconnecting

When the connection is in master role and it is going to be
disconnected based on the disconnection timeout, then send
the HCI_Read_Clock_Offset command in an attempt to update the
clock offset value in the inquiry cache.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Remove unneeded hci_conn_disconnect abstraction
Marcel Holtmann [Fri, 27 Jun 2014 11:45:09 +0000 (13:45 +0200)]
Bluetooth: Remove unneeded hci_conn_disconnect abstraction

The abstraction of disconnect operation via hci_conn_disconnect is not
needed and it does not add any readability. Handle the difference of
AMP physical channels and BR/EDR/LE connection in the timeout callback.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Remove reason parameter from hci_amp_disconn function
Marcel Holtmann [Fri, 27 Jun 2014 11:45:08 +0000 (13:45 +0200)]
Bluetooth: Remove reason parameter from hci_amp_disconn function

The hci_amp_disconn function is a local function and there is no
need for a reason parameter. That one can be retrieved from the
hci_conn object easily.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Remove unnecessary hcon->smp_conn variable
Johan Hedberg [Fri, 27 Jun 2014 11:23:07 +0000 (14:23 +0300)]
Bluetooth: Remove unnecessary hcon->smp_conn variable

The smp_conn member of struct hci_conn was simply a pointer to the
l2cap_conn object. Since we already have hcon->l2cap_data that points to
the same thing there's no need to have this second variable. This patch
removes it and changes the single place that was using it to use
hcon->l2cap_data instead.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Fix missing check for SMP session in smp_user_confirm_reply
Johan Hedberg [Fri, 27 Jun 2014 11:23:06 +0000 (14:23 +0300)]
Bluetooth: Fix missing check for SMP session in smp_user_confirm_reply

The smp_user_confirm_reply() function is called whenever user space
sends a user confirmation reply mgmt command. In case of a misbehaving
user space, or if the SMP session was removed by the time the command
comes it is important that we return an appropriate error and do not try
to access the non-existent SMP context. This patch adds the appropriate
check for the HCI_CONN_LE_SMP_PEND flag before proceeding further.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Remove unnecessary hci_dev_unlock for smp_user_confirm_reply
Johan Hedberg [Fri, 27 Jun 2014 11:23:05 +0000 (14:23 +0300)]
Bluetooth: Remove unnecessary hci_dev_unlock for smp_user_confirm_reply

Now that the SMP context has it's own crypto handle it doesn't need to
lock the hci_dev anymore for most operations. This means that it is safe
to call smp_user_confirm_reply with the lock already held.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Update SMP crypto functions to take the SMP context
Johan Hedberg [Fri, 27 Jun 2014 11:23:04 +0000 (14:23 +0300)]
Bluetooth: Update SMP crypto functions to take the SMP context

Passing the full SMP context instead of just the crypto context lets us
use the crypto handle from the context which in turn removes the need to
lock the hci_dev. Passing the SMP context instead of just the crypto
handle allows a bit more detailed logging which is helpful in
multi-adapter scenarios.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Add dedicated AES instance for each SMP context
Johan Hedberg [Fri, 27 Jun 2014 11:23:03 +0000 (14:23 +0300)]
Bluetooth: Add dedicated AES instance for each SMP context

Many places have to be extra careful to not hold the hdev lock when
calling into the SMP code. This is because the SMP crypto functions use
the crypto handle that's part of the hci_dev struct. Giving the SMP
context its own handle helps simplifying the locking logic and removes
the risk for deadlocks.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Fix missing hdev locking in smp_cmd_ident_addr_info
Johan Hedberg [Fri, 27 Jun 2014 11:23:02 +0000 (14:23 +0300)]
Bluetooth: Fix missing hdev locking in smp_cmd_ident_addr_info

The hdev lock must be held before calling into smp_distribute_keys. Also
things such as hci_add_irk() require the lock. This patch fixes the
issue by adding the necessary locking into the smp_cmd_ident_addr_info
function.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Use macro instead of hard-coded value
Andre Guedes [Wed, 25 Jun 2014 19:44:45 +0000 (16:44 -0300)]
Bluetooth: Use macro instead of hard-coded value

This patch replaces the hard-coded value in hci_bdaddr_is_rpa() helper
by the corresponding macro ADDR_LE_DEV_RANDOM.

Signed-off-by: Andre Guedes <andre.guedes@openbossa.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Convert hci_conn->link_mode into flags
Johan Hedberg [Tue, 24 Jun 2014 14:03:50 +0000 (17:03 +0300)]
Bluetooth: Convert hci_conn->link_mode into flags

Since the link_mode member of the hci_conn struct is a bit field and we
already have a flags member as well it makes sense to merge these two
together. This patch moves all used link_mode bits into corresponding
flags. To keep backwards compatibility with user space we still need to
provide a get_link_mode() helper function for the ioctl's that expect a
link_mode style value.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Remove ssp_debug_mode debugfs option
Marcel Holtmann [Tue, 24 Jun 2014 11:13:04 +0000 (13:13 +0200)]
Bluetooth: Remove ssp_debug_mode debugfs option

The ssp_debug_mode debugfs option for developers is no longer
needed. Support for using Secure Simple Pairing (SSP) debug
mode is exposed by the management interface now.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Add support for mode 0x02 for mgmt_set_debug_keys
Johan Hedberg [Tue, 24 Jun 2014 11:00:28 +0000 (14:00 +0300)]
Bluetooth: Add support for mode 0x02 for mgmt_set_debug_keys

This patch adds a new valid mode 0x02 for the mgmt_set_debug_keys
command. The 0x02 mode sets the HCI_USE_DEBUG_KEYS flag which makes us
always use debug keys for pairing.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Add a new HCI_USE_DEBUG_KEYS flag
Johan Hedberg [Tue, 24 Jun 2014 11:00:27 +0000 (14:00 +0300)]
Bluetooth: Add a new HCI_USE_DEBUG_KEYS flag

To pave the way for actively using debug keys for pairing this patch
adds a new HCI_USE_DEBUG_KEYS flag for the purpose. When the flag is set
we issue a HCI_Write_SSP_Debug mode whenever HCI_Write_SSP_Mode(0x01)
has been issued as well as before issuing a HCI_Write_SSP_Mode(0x00)
command.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Fix ignoring debug keys in mgmt_load_link_keys
Johan Hedberg [Tue, 24 Jun 2014 11:00:26 +0000 (14:00 +0300)]
Bluetooth: Fix ignoring debug keys in mgmt_load_link_keys

We should never allow user space to feed back debug keys to the kernel.
If the user desires to use debug keys require setting the appropriate
debug keys mode and performing a new pairing.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Convert hcon->flush_key to a proper flag
Johan Hedberg [Tue, 24 Jun 2014 10:15:53 +0000 (13:15 +0300)]
Bluetooth: Convert hcon->flush_key to a proper flag

There's no point in having boolean variables in the hci_conn struct
since it already has a flags member. This patch converts the flush_key
member into a proper flag.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Don't store debug keys if flag for them is not set
Johan Hedberg [Tue, 24 Jun 2014 10:15:51 +0000 (13:15 +0300)]
Bluetooth: Don't store debug keys if flag for them is not set

Instead of waiting for a disconnection to occur to remove a debug key
simply never store it in the list to begin with. This means we can also
remove the debug keys check when looking up keys in
hci_link_key_request_evt().

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Rename HCI_DEBUG_KEYS to HCI_KEEP_DEBUG_KEYS
Johan Hedberg [Tue, 24 Jun 2014 10:15:50 +0000 (13:15 +0300)]
Bluetooth: Rename HCI_DEBUG_KEYS to HCI_KEEP_DEBUG_KEYS

We're planning to add a flag to actively use debug keys in addition to
simply just accepting them, which makes the current generically named
DEBUG_KEYS flag a bit confusing. Since the flag in practice affects
whether the kernel keeps debug keys around or not rename it to
HCI_KEEP_DEBUG_KEYS.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Move mgmt event sending out from hci_add_link_key()
Johan Hedberg [Tue, 24 Jun 2014 10:15:49 +0000 (13:15 +0300)]
Bluetooth: Move mgmt event sending out from hci_add_link_key()

There are two callers of hci_add_link_key(). The first one is the HCI
Link Key Notification event and the second one the mgmt code that
receives a list of link keys from user space. Previously we've had the
hci_add_link_key() function being responsible for also emitting a mgmt
signal but for the latter use case this should not happen. Because of
this a rather awkward new_key paramter has been passed to the function.

This patch moves the mgmt event sending out from the hci_add_link_key()
function, thereby making the code a bit more understandable.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Update hci_add_link_key() to return pointer to key
Johan Hedberg [Tue, 24 Jun 2014 10:15:48 +0000 (13:15 +0300)]
Bluetooth: Update hci_add_link_key() to return pointer to key

By returning the added (or updated) key we pave the way for further
refactoring (in subsequent patches) that allows moving the mgmt event
sending out from this function (and thereby removal of the awkward
new_key parameter).

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Remove redundant calls to h5_reset_rx
Loic Poulain [Mon, 23 Jun 2014 13:49:46 +0000 (15:49 +0200)]
Bluetooth: Remove redundant calls to h5_reset_rx

h5_reset_rx is unconditionally called at the end of
h5_complete_rx_pkt, no need to call it anymore after
that.

Signed-off-by: Loic Poulain <loic.poulain@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Add support LE slave connection update procedure
Marcel Holtmann [Mon, 23 Jun 2014 10:18:51 +0000 (12:18 +0200)]
Bluetooth: Add support LE slave connection update procedure

When the current LE connection parameters of a slave connection do not
match up with the controller defined values, then trigger the connection
update procedure to allow adjusting them.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Set default min/max connection interval for LE slaves
Marcel Holtmann [Mon, 23 Jun 2014 10:14:51 +0000 (12:14 +0200)]
Bluetooth: Set default min/max connection interval for LE slaves

For all incoming LE connections, the minimum and maximum connection
interval is a value that should be copied from the controller default
values. This allows to properly check if the resulting connection
interval of a newly established connection is in the range we are
expecting.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Track LE connection parameter update event
Marcel Holtmann [Mon, 23 Jun 2014 09:40:05 +0000 (11:40 +0200)]
Bluetooth: Track LE connection parameter update event

When the LE controller changes its connection parameters, it will send
a connection parameter update event. Make sure that the new set of
parameters are stored in hci_conn struct and thus will properly update
the previous values retrieved from the connection complete event.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Store current LE connection parameters in hci_conn struct
Marcel Holtmann [Mon, 23 Jun 2014 09:40:04 +0000 (11:40 +0200)]
Bluetooth: Store current LE connection parameters in hci_conn struct

The LE connection parameters are needed later on to be able to decide
if it is required to trigger connection update procedures. So when the
connection has been established successfully, store the current used
parameters in hci_conn struct.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: 6LoWPAN: Remove network devices when unloading
Jukka Rissanen [Wed, 18 Jun 2014 13:37:11 +0000 (16:37 +0300)]
Bluetooth: 6LoWPAN: Remove network devices when unloading

When the module is unloaded, unregister the network device
so that the system does not try to access non-existing device.

Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: 6LoWPAN: Count module usage
Jukka Rissanen [Wed, 18 Jun 2014 13:37:10 +0000 (16:37 +0300)]
Bluetooth: 6LoWPAN: Count module usage

Count how many 6LoWPAN connections there exists so that we
do not unload the module if there are still connections alive.

Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: 6LoWPAN: Create a kernel module
Jukka Rissanen [Wed, 18 Jun 2014 13:37:09 +0000 (16:37 +0300)]
Bluetooth: 6LoWPAN: Create a kernel module

Instead of adding the 6LoWPAN functionality to Bluetooth module,
we create a separate kernel module for it.

Usage:

In the slave side do this:

$ modprobe bluetooth_6lowpan
$ echo 62 > /sys/kernel/debug/bluetooth/6lowpan_psm
$ hciconfig hci0 leadv

In the master side do this:

$ modprobe bluetooth_6lowpan
$ echo 62 > /sys/kernel/debug/bluetooth/6lowpan_psm
$ echo 'connect E0:06:E6:B7:2A:73 1' > \
                  /sys/kernel/debug/bluetooth/6lowpan_control

The 6LoWPAN functionality can be controlled by psm value. If it
is left to 0, then the module is disabled and all the 6LoWPAN
connections are dropped if there were any. In the above example,
the psm value is just an example and not a real value for
6LoWPAN service. The real psm value is yet to be defined in
Bluetooth specification.

The 6lowpan controlling interface is a temporary solution
until the specifications are ready.

Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: 6LoWPAN: Use connected oriented channel instead of fixed one
Jukka Rissanen [Wed, 18 Jun 2014 13:37:08 +0000 (16:37 +0300)]
Bluetooth: 6LoWPAN: Use connected oriented channel instead of fixed one

Create a CoC dynamically instead of one fixed channel for communication
to peer devices.

Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Provide L2CAP ops callback for memcpy_fromiovec
Jukka Rissanen [Wed, 18 Jun 2014 13:37:07 +0000 (16:37 +0300)]
Bluetooth: Provide L2CAP ops callback for memcpy_fromiovec

The highly optimized TX path for L2CAP channels and its fragmentation
within the HCI ACL packets requires to copy data from user provided
IO vectors and also kernel provided memory buffers.

This patch allows channel clients to provide a memcpy_fromiovec callback
to keep this optimized behavior, but adapt it to kernel vs user memory
for the TX path. For all kernel internal L2CAP channels, a default
implementation is provided that can be referenced.

In case of A2MP, this fixes a long-standing issue with wrongly accessing
kernel memory as user memory.

This patch originally by Marcel Holtmann.

Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Use separate dbg_flags to special debugfs options
Marcel Holtmann [Sat, 21 Jun 2014 02:53:17 +0000 (04:53 +0200)]
Bluetooth: Use separate dbg_flags to special debugfs options

All the special settings configured via debugfs are either developer
only options or temporary solutions. To not clutter the standard flags,
move them to their own dbg_flags entry.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Fix checking for master LTKs
Johan Hedberg [Wed, 18 Jun 2014 11:09:40 +0000 (14:09 +0300)]
Bluetooth: Fix checking for master LTKs

When the rename of STK_SLAVE to simply STK happened we missed this place
in the ltk_type_master function. Now, checking for master is as simple
as checking whether the type is SMP_LTK. The helper function is kept
around for better readability in the (right now three) callers and for
simpler extension with new key types in the future.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Tested-by: Lukasz Rymanowski <lukasz.rymanowski@tieto.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Fix validating IO capability values in mgmt commands
Johan Hedberg [Tue, 17 Jun 2014 12:14:48 +0000 (15:14 +0300)]
Bluetooth: Fix validating IO capability values in mgmt commands

The valid range of IO capabilities for the Set IO Capability and Pair
Device mgmt commands is 0-4 (4 being the KeyboarDisplay capability for
SMP). We should return an invalid parameters error if user space gives
us a value outside of this range.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Remove trailing whitespaces from Kconfig
Stephan Gabert [Mon, 16 Jun 2014 16:52:25 +0000 (18:52 +0200)]
Bluetooth: Remove trailing whitespaces from Kconfig

Fixed a coding style issue. Removed trailing whitespaces in
drivers/bluetooth/Kconfig.

Signed-off-by: Stephan Gabert <stephan.gabert@fau.de>
Signed-off-by: Nicolas Pfeiffer <nicolas.pfeiffer@fau.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Add flexible buffer byte order swapping function
Johan Hedberg [Mon, 16 Jun 2014 16:25:19 +0000 (19:25 +0300)]
Bluetooth: Add flexible buffer byte order swapping function

Since the SMP code needs to swap ordering of variable length buffers add
a convenience function that can be used for any length buffer.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Convert SMP flags into an enum
Johan Hedberg [Mon, 16 Jun 2014 16:25:18 +0000 (19:25 +0300)]
Bluetooth: Convert SMP flags into an enum

There's no reason to have explicit values for these flags. Convert them
to an enum to be consistent with other similar flags.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Remove unused LTK authentication defines
Johan Hedberg [Mon, 16 Jun 2014 16:25:17 +0000 (19:25 +0300)]
Bluetooth: Remove unused LTK authentication defines

These defines were probably put in to track authenticated vs
unauthenticated LTKs, however since the LTK struct has a separate
boolean authenticated member these were never used.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Remove HCI prefix from SMP LTK defines
Johan Hedberg [Mon, 16 Jun 2014 16:25:16 +0000 (19:25 +0300)]
Bluetooth: Remove HCI prefix from SMP LTK defines

The LTK type has really nothing to do with HCI so it makes more sense to
have these in smp.h than hci.h. This patch moves the defines to smp.h
and removes the HCI_ prefix in the same go.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Remove unnecessary SMP STK define
Johan Hedberg [Mon, 16 Jun 2014 16:25:15 +0000 (19:25 +0300)]
Bluetooth: Remove unnecessary SMP STK define

We never store the "master" type of STKs since we request encryption
directly with them so we only need one STK type (the one that's
looked-up on the slave side). Simply remove the unnecessary define and
rename the _SLAVE one to the shorter form.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Fix missing NULL check for smp_chan_create() return value
Johan Hedberg [Mon, 16 Jun 2014 16:25:14 +0000 (19:25 +0300)]
Bluetooth: Fix missing NULL check for smp_chan_create() return value

The smp_chan_create function may return NULL, e.g. in the case of memory
allocation failure, so we always need to check for this.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
10 years agoBluetooth: Use struct delayed_work for HCI command timeout
Marcel Holtmann [Mon, 16 Jun 2014 10:30:56 +0000 (12:30 +0200)]
Bluetooth: Use struct delayed_work for HCI command timeout

Since the whole HCI command, event and data packet processing has been
migrated to use workqueues instead of tasklets, it makes sense to use
struct delayed_work instead of struct timer_list for the timeout
handling. This patch converts the hdev->cmd_timer to use workqueue
as well.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Use explicit header and body length for L2CAP SKB allocation
Marcel Holtmann [Sun, 8 Jun 2014 09:22:28 +0000 (11:22 +0200)]
Bluetooth: Use explicit header and body length for L2CAP SKB allocation

When allocating the L2CAP SKB for transmission, provide the upper layers
with a clear distinction on what is the header and what is the body
portion of the SKB.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Shrink size of struct l2cap_ctrl fields
Marcel Holtmann [Sun, 8 Jun 2014 08:26:57 +0000 (10:26 +0200)]
Bluetooth: Shrink size of struct l2cap_ctrl fields

The struct l2cap_ctrl fields are wasting an unsigned int when all the
bits can fit into an __u8 field.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Assign L2CAP socket priority when allocating SKB
Marcel Holtmann [Thu, 5 Jun 2014 13:22:51 +0000 (15:22 +0200)]
Bluetooth: Assign L2CAP socket priority when allocating SKB

The SKB for L2CAP sockets are all allocated in a central callback
in the socket support. Instead of having to pass around the socket
priority all the time, assign it to skb->priority when actually
allocating the SKB.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: Use const for struct l2cap_ops field
Marcel Holtmann [Sun, 8 Jun 2014 08:05:31 +0000 (10:05 +0200)]
Bluetooth: Use const for struct l2cap_ops field

The struct l2cap_ops field should not allow any modifications and thus
it is better declared as const.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
10 years agoBluetooth: ath3k: reduce pipe setting times in ath3k_load_fwfile()
Adam Lee [Thu, 5 Jun 2014 13:47:44 +0000 (21:47 +0800)]
Bluetooth: ath3k: reduce pipe setting times in ath3k_load_fwfile()

Invoking usb_sndbulkpipe() on same pipe for same purpose only once is
enough.

Signed-off-by: Adam Lee <adam8157@gmail.com>
Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
10 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth
Marcel Holtmann [Thu, 3 Jul 2014 14:39:39 +0000 (16:39 +0200)]
Merge git://git./linux/kernel/git/bluetooth/bluetooth

10 years agob43: add more bcma cores
Rafał Miłecki [Tue, 1 Jul 2014 14:33:57 +0000 (16:33 +0200)]
b43: add more bcma cores

This adds some cores with 0x2057 radio which will be supported soon as
well as core 40 that I missed in the earlier firmware patch.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: N-PHY: complete generic support for 0x2057 radio
Rafał Miłecki [Tue, 1 Jul 2014 14:19:08 +0000 (16:19 +0200)]
b43: N-PHY: complete generic support for 0x2057 radio

It doesn't include any device (radio revision) specific code yet, so it
isn't really usable. As the commit says, it's just some generic code.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: N-PHY: fixes for radio 0x2057
Rafał Miłecki [Tue, 1 Jul 2014 14:19:07 +0000 (16:19 +0200)]
b43: N-PHY: fixes for radio 0x2057

Enable initialization and update calibration code to fix:
b43-phy0 ERROR: Radio 0x2057 rcal timeout
b43-phy0 debug: Radio 0x2057 rccal timeout
b43-phy0 debug: Radio 0x2057 rccal timeout
b43-phy0 ERROR: Radio 0x2057 rcal timeout

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl818x_pci: fix pci probe returns success when it fails
Andrea Merello [Mon, 30 Jun 2014 16:19:40 +0000 (18:19 +0200)]
rtl818x_pci: fix pci probe returns success when it fails

There are several exit path from the PCI probe function.
Some of them, that are taken in case of errors, forget to set the "err"
variable, that is returned by the probe function.
This can lead to the kernel thinking the probe function succeeds while it
didn't, and this in turn causes extra calls to the "remove" function.

This patch fix this problem by ensuring "err" variable is assigned to a proper
non-zero value in each exit path.

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl818x_pci: handle broken PIO mapping
Andrea Merello [Mon, 30 Jun 2014 16:19:27 +0000 (18:19 +0200)]
rtl818x_pci: handle broken PIO mapping

All boards supported by this driver could work using PIO or MMIO for accessing
registers.
This driver tries to access HW by using MMIO, and, if this fails for somewhat
reason, the driver tries to fall back to PIO mode.

MMIO-mode is straightforward on all boards.
PIO-mode is straightforward on rtl8180 only.

On rtl8185 and rtl8187se boards not all registers are directly available in PIO
mode (they are paged).

On rtl8185 there are two pages and it is known how to switch page.
PIO mode works, except for only one access to a register out of default page,
recently added by me in the initialization code with patch:
rtl818x_pci: Fix rtl8185 excessive IFS after CTS-to-self
This can be easily fixed to work in both cases (MMIO and PIO).

On rtl8187se, for a number of reasons, there is much more work to do to fix PIO
access.
PIO access is currently broken on rtl8187se, and it never worked.

This patch fixes the said register write for rtl8185 and makes the driver to
fail cleanly if PIO mode is attempted with rtl8187se boards.

While doing this, I converted also a couple of printk(KERN_ERR) to dev_err(), in
order to make checkpatch happy.

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl8180: disable buggy rate fallback mechanism
Andrea Merello [Mon, 30 Jun 2014 16:19:10 +0000 (18:19 +0200)]
rtl8180: disable buggy rate fallback mechanism

Currently the driver configures mac80211 to provide two rates for each TX frame:
One initial rate and one alternate fallback rate, each one with its retry count.

HW does not support fully this: rtl8180 doesn't have support for rate scaling at
all, and rtl8185/rtl8187SE supports it in a way that does not fit with mac80211:
The HW does automatically fall back to the next lower rate, and only a lower
limit can be specified, so the HW may TX also on rates in between the two rates
specified by mac80211.  Furthermore only the total TX retry count can be
specified for each packet, while the number of TX attempts before scaling rate
can be configured only globally (not per each packet).

Currently the driver sets the HW auto rate fallback mechanism to quickly scale
rate after a couple of retries, and it uses the alternate rate requested by
mac80211 as fallback limit rate (and it does this even wrongly).

The HW indeed will behave differently than what mac80211 mandates, that is
probably undesirable, and the reported TX retry count may not refer to what
mac80211 thinks, and this could fool mac80211.

This patch makes the driver to declare to mac80211 to support only one rate
configuration for each packet, and it does disable the HW auto rate fallback
mechanism, relying only on SW and letting mac80211 to do all by itself.

This should ensure correct operation and fairness respect to mac80211.
Indeed here tests with iperf do not show significant performance differences.

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl8180: fix incorrect TX retry.
Andrea Merello [Mon, 30 Jun 2014 16:18:55 +0000 (18:18 +0200)]
rtl8180: fix incorrect TX retry.

HW is programmed with wrong retry count value for TX:

Mac80211 passes to driver the number of times the TX should be attempted.
The HW, instead, wants the number of time the TX should be retried if it fails
the first time (assuming we have to TX it at least one time).

This patch correct this.

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl818x_pci: add comment pointing to the rtl8187se reference code
Andrea Merello [Mon, 30 Jun 2014 16:18:36 +0000 (18:18 +0200)]
rtl818x_pci: add comment pointing to the rtl8187se reference code

Rtl8187se support has been added to the rtl818x_pci driver by extracting a lot
of information from a rtl8187se Linux staging driver included in the kernel at
the time rtl8187se support was added.
The rtl818x_pci main file has a comment that advertises this.

Recently this staging driver has been removed from the kernel, but I still feel
it can be useful as "reference" code (in case of bugs, or to implement
improvements in rtl818x_pci driver).

This one-line patch adds a comment in rtl818x_pci driver to point people
searching for that "reference code" to the last kernel version still containing
it (3.14).

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl818x_pci: Fix rtl8185 excessive IFS after CTS-to-self
Andrea Merello [Mon, 30 Jun 2014 16:18:25 +0000 (18:18 +0200)]
rtl818x_pci: Fix rtl8185 excessive IFS after CTS-to-self

Measuring time between _end_ of CTS-to-self and _end_ of datapacket (with a
prism54 board and mac80211 hacked to let the MAC timestamp stay untouched in the
radiotap header) resulted in about 300uS, while the datapacket itself should be
by far shorter (less than 100uS) and IFS should be SIFS (10uS).
This measure was confirmed whith a scope: about 250uS IFS has been seen between
the two packets.

This situation causes the CTS-to-self protection mechanism to work incorrectly
due to the NAV expiring during, or even before beginning, the packet
transmission, and it also causes the performances to be anyway reduced due to
time waste.

This problem has been seen at every packet TXed with CTS-to-self enabled on
rtl8185 board.
rtl8187se seems not affected (and rtl8180, being a 802.11b card, does not have
CTS-to-self mechaninsm).

This patch fixes this by adding a magic register write, making the board wait
for correct SIFS after CTS-to-self packet.

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl818x_pci: Fix BSSID register written incorrectly
Andrea Merello [Mon, 30 Jun 2014 16:17:48 +0000 (18:17 +0200)]
rtl818x_pci: Fix BSSID register written incorrectly

BSSID register was written with six byte-writes.
It seems that, similarly to what happens with MAC registers, they needs to be
written with one 16-bit and one 32-bit writes, otherwise the write does not work.

The byte write didn't work only on my rtl8185, while it worked on rtl8180 and
rtl8187se, BTW since there are probably a number of different ASIC revisions out
of there, I let the change to affect all cards.
It shouldn't hurt anyway.

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: treat LCNXN-PHY as extra N-PHY devices
Rafał Miłecki [Sun, 29 Jun 2014 19:46:45 +0000 (21:46 +0200)]
b43: treat LCNXN-PHY as extra N-PHY devices

LCNXN is simply a continuation of N, e.g. code handling LCNXN revs 0 and
1 is mostly the same as for N-PHY revs 7+.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agodrivers/net/wireless/ipw2x00/libipw_module.c: remove unnecessary null test before...
Fabian Frederick [Sat, 28 Jun 2014 12:18:52 +0000 (14:18 +0200)]
drivers/net/wireless/ipw2x00/libipw_module.c: remove unnecessary null test before kfree

Fix checkpatch warning:
WARNING: kfree(NULL) is safe this check is probably not required

Cc: Stanislav Yakovlev <stas.yakovlev@gmail.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: linux-wireless@vger.kernel.org
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: fix memory leaks and error handling in rsi_91x_usb
Alexey Khoroshilov [Thu, 26 Jun 2014 22:51:25 +0000 (02:51 +0400)]
rsi: fix memory leaks and error handling in rsi_91x_usb

The patch fixes a couple of issues:
- absence of deallocation of rsi_dev->rx_usb_urb[0] in the driver;
- potential NULL pointer dereference because of lack of checks for memory
  allocation success in rsi_init_usb_interface().

By the way, it makes rsi_probe() returning error code instead of 1
and fixes comments regarding returning values.

Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: GFP_ATOMIC is not needed in rsi_init_usb_interface()
Alexey Khoroshilov [Thu, 26 Jun 2014 22:51:24 +0000 (02:51 +0400)]
rsi: GFP_ATOMIC is not needed in rsi_init_usb_interface()

Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Calculate sleep duration
Rajkumar Manoharan [Thu, 26 Jun 2014 11:24:42 +0000 (16:54 +0530)]
ath9k: Calculate sleep duration

Right now sleep duration is configured as beacon interval. It should be
the multiple of beacon interval by listen period which helps to
reduce station power consumption.

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Increase max listen interval
Rajkumar Manoharan [Thu, 26 Jun 2014 11:24:41 +0000 (16:54 +0530)]
ath9k: Increase max listen interval

Earlier the listen interval is used to decide switching between
operating and off-channels during bgscan and to improve throughput,
the listen interval is reduced to 1. After optimiztion in scan
state machine, listen period is not used for decision making and
hence reverting it back to original value.

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Cache beacon config after association
Rajkumar Manoharan [Thu, 26 Jun 2014 11:24:40 +0000 (16:54 +0530)]
ath9k: Cache beacon config after association

The beacon configurations are not cached properly after the station
associates with AP. Not handling BEACON_INFO, is failing to update
dtim period and also it is causing below warning message.

WARNING: CPU: 1 PID: 0 at drivers/net/wireless/ath/ath9k/recv.c:548
ath_rx_tasklet+0xc89/0xca0 [ath9k]()
 Call Trace:
  [<c14669c9>] dump_stack+0x48/0x69
  [<c104f1a2>] warn_slowpath_common+0x82/0xa0
  [<fd38c2f9>] ? ath_rx_tasklet+0xc89/0xca0 [ath9k]
  [<fd38c2f9>] ? ath_rx_tasklet+0xc89/0xca0 [ath9k]

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath5k: capture CCK and OFDM restarts
Mathy Vanhoef [Fri, 13 Jun 2014 23:14:56 +0000 (01:14 +0200)]
ath5k: capture CCK and OFDM restarts

Treat frames that underwent a CCK or OFDM restart as frames with an invalid CRC.

Signed-off-by: Mathy Vanhoef <vanhoefm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath5k: support for FIF_FCSFAIL filter
Mathy Vanhoef [Fri, 13 Jun 2014 21:40:22 +0000 (23:40 +0200)]
ath5k: support for FIF_FCSFAIL filter

When the FIF_FCSFAIL filter flag is set, pass frames with CRC errors.

Signed-off-by: Mathy Vanhoef <vanhoefm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowireless: mwifiex: Use the proper interfaces
Thomas Gleixner [Thu, 12 Jun 2014 07:31:34 +0000 (08:31 +0100)]
wireless: mwifiex: Use the proper interfaces

Why is converting time formats so desired if there are proper
interfaces for this?

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Bing Zhao <bzhao@marvell.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: linux-wireless@vger.kernel.org
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoRevert "mwifiex: Use the proper interfaces"
Bing Zhao [Sat, 21 Jun 2014 00:07:42 +0000 (17:07 -0700)]
Revert "mwifiex: Use the proper interfaces"

This reverts commit a82fc3b4a2bceb7c6587249cb690342eb5065979.

Thomas corrected me on that I misunderstood Johannes' comment
for net_timedelta() and the ktime_get_real() usage inside
__net_timestamp().

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Johannes Berg <johannes@sipsolutions.net>
Cc: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: b43_phyops_a can be static
Fengguang Wu [Wed, 25 Jun 2014 19:46:51 +0000 (15:46 -0400)]
b43: b43_phyops_a can be static

CC: "Rafał Miłecki" <zajec5@gmail.com>
CC: "John W. Linville" <linville@tuxdriver.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Fixed warnings reported by static code analyzers.
Jahnavi Meher [Wed, 25 Jun 2014 10:18:15 +0000 (15:48 +0530)]
rsi: Fixed warnings reported by static code analyzers.

Fixed a warning related to incorrect return type and removed an
unnecessary semi colon.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Fixed errors and warnings reported by static code analyzers.
Jahnavi Meher [Wed, 25 Jun 2014 10:17:45 +0000 (15:47 +0530)]
rsi: Fixed errors and warnings reported by static code analyzers.

Fixed a potential buffer overflow in 'rsi_rates' and a sparse warning
related to difference in endianness in rsi_91x_mgmt.c.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agop54: use request_firmware_direct() for optional EEPROM override
Luis R. Rodriguez [Tue, 24 Jun 2014 22:39:43 +0000 (15:39 -0700)]
p54: use request_firmware_direct() for optional EEPROM override

The p54 driver uses request_firmware() twice, once for actual
firmware and then another time for an optional user overide on
EEPROM, 3826.eeprom. The custom EEPROM  is optional but if not
present we'll introduce an extra lag of 60 seconds with udev
present. Annotate we don't want udev nonsense here to avoid
the lag in case its not present.

This was found with the following SmPL patch.

@ firmware_not_critical @
expression cf;
expression config_file;
expression dev;
int ret;
identifier l;
statement S;
@@

- ret = request_firmware(&cf, config_file, dev);
+ ret = request_firmware_direct(&cf, config_file, dev);
if (ret < 0) {
... when != goto l;
    when != return ret;
    when any
} else {
...
release_firmware(cf);
...
}

Cc: Takashi Iwai <tiwai@suse.de>
Cc: Christian Lamparter <chunkeey@googlemail.com>
Cc: linux-wireless@vger.kernel.org
Cc: cocci@systeme.lip6.fr
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Acked-By: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>