staging: comedi: pcl816: tidy up multi-line comments

Reformat the multi-line comments in the kernel CodingStyle.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: pcl818: remove unused "boards constants"

These defines are not used. Remove them to get rid of the
checkpatch.pl issue about avoiding CamelCase.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: pcl818: prefer using the BIT macro

Fix the checkpatch.pl issues.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: update the MODULE_DESCRIPTION

Change the MODULE_DESCRIPTION to something more useful than the
generic "Comedi low-level driver".

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: remove unnecessary include

This driver does not use interrupts. Remove the include.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: tidy up multiq3_encoder_insn_read()

Encoders are not a "normal" subdevice in comedi. For aesthetics, tidy
up this function and add a couple comments to clarify the function and
explain the strange munging of the data.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: remove unnecessary define

This define is not needed. Remove it.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: allow user to reset encoder channels

Currently this driver resets all the encoder channels when the driver
is first attached. Add a (*insn_config) to the subdevice to allow the
use to reset the channels manually.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: document the encoder chip commands

Add some comments to document the commands that are sent to the
encoder chips.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: tidy up control register bit defines

For aesthetics, move these defines closer to the register define and
use the BIT macro to define the bits.

Add macros to set the bits needed to select the channel for the analog
output, analog input, and encoder (counter) subdevices as well as the
realtime clock registers.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: introduce multiq3_set_ctrl()

According to the programming manual. the 'SH' and 'CLK' bits in the
control register need to be kept high at all times. Clarify this by
introducing a helper function to set the control register.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: tidy up status register bit defines

For aesthetics, move these defines closer to the register define and
use the BIT macro to define the bits.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: tidy up register map defines

For auesthetics, rename the register map defines. Convert the offset
values to hex and add some whitespace.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: tidy up subdevice init

For aesthetics, add some whitespace to the subdevice init and rearrange
the initialization a bit.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: tidy up multiq3_ai_insn_read()

For aesthetics, use the proper symbol when reading the A/D data register
to get the 16-bit sample data.

Use the comedi_offset_munge() to do the 2's complement to offset binary
munging of the sample data.

Tidy up the function a bit.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: multiq3: tidy up multi-line comments

Reformat the multi-line comments in the kernel CodingStyle.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: avoid calculating pacer divisor multiple times

The pacer will either be used to trigger each conversion or to start each
scan. Modify the (*do_cmdtest) so that one one divisor calculation is needed.
Save the divisor value in the private data so that the (*do_cmd) does not
have to do the calculation again.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: ai async command requires a pacer

The ai (*do_cmd) assumes that either the convert_src is TRIG_TIMER or
the scan_begin_src is TRIG_TIMER and always programs the pacer clock
with the appropriate trigger argument.

Add a Step 2b check to the (*do_cmdtest) to ensure that at least one
of the triggers is TRIG_TIMER.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: cleanup Step 3 of ai (*do_cmdtest)

Step 3 of the (*do_cmdtest) trivially validates the async command arguments.
The validations also modify the arguments if they are invalid so that the
user gets valid values if the test fails.

Reorder the checks so that if any of the checks fail proper values are
used for subsequent checks.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: fix daqp_ai_cancel()

Make sure interrupts are disabled and the event flags are cleared when
an analog input async command is canceled.

Remove the unnecessary calls to stop any running conversions in the
(*insn_read) and (*do_cmd) functions. The comedi core will only call
these functions if the subdevice is not busy (it has already been
canceled).

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: handle shared interrupt

This is a PCMCIA driver and interrupts are always shared. Detect if
the hardware did not produce the interrupt and return IRQ_NONE so
that other drivers might handle it.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: remove 'interrupt_mode'

The interrupt handler is now only used for the ai async command.
Remove the unnecessary 'interrupt_mode' from the private data and
tidy up the interrupt handler.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: only hook up cmd support if we have an IRQ

Interrupts are only needed by this driver to support the analog input async
commands. Don't hook up the command support if pcmcia_request_irq() fails
instead of completely failing the (*auto_attach).

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: don't use interrupts for ai (*insn_read)

The comedi (*insn_read) functions are supposed to do simple one-shot
reading of an analog input channel. Currently this driver enables
interrupts and uses wait_for_completion_interruptible() to allow the
interrupt routine to let the analog input (*insn_read) know that
the end-of-conversion has occured.

Simplify the function by using the comedi_timeout() helper to check
the aux status register to see when the conversion is finished.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: tidy up aux register bitss

For aesthetics, and use the BIT macro to define the bits and
define some macros for the timer mode and d/a update bits.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: document the 16-bit registers

All the registers are 8-bit except for the analog output and timer
registers. For clarity, add some comments.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: tidy up command register bits

For aesthetics, move the bit defines for this register and use the
BIT macro to define the bits. Define a macro for the scanrate bits
and use it to remove the CamelCase. Rename the defines to match the
register.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: tidy up status register bits

For aesthetics, move the bit defines for this register and use the
BIT macro to define the bits.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: tidy up control register bits

For aesthetics, move the bit defines for this register and use the
BIT macro to define the bits. Rename the defines to match the
register.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: tidy up scanlist register bits

For aesthetics, move the bit defines for this register and use the
BIT macro to define the bits.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: rename register offsets

For aesthetics, rename the register map defines and convert the
values to hex. Also, move the private data definition and
DAQP_FIFO_SIZE define after the register definitions.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: tidy up multi-line comments

Reformat the multi-line comments in the kernel CodingStyle.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: introduce daqp_clear_events()

Introduce a helper function for the common code used to clear any
pending interrupts.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: redefine pacer clock options

For aesthetics, define a macro for the pacer clock options and rename
the CamelCase.

Remove the unnecessary local variables used to set the control register.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: spaces preferred around that '<<'

Fix the checkpatch.pl issues about:
CHECK: spaces preferred around that '<<' (ctx:VxV)

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: use comedi_timeout() in ao (*insn_write)

The data link between the D/A data port and the D/A converter is a serial
link. The serial link requires about 8ms to complete a transfer. Use the
comedi_timeout() helper to ensure that there is not a previous transfer
still happening before trying to write new data to the channel.

For aesthetics, save the readback value after updating the hardware.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: use comedi_offset_munge() in ao (*insn_write)

Use the comedi_offset_munge() helper to do the offset binary to two's
complement conversion when writing a new analog output value.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: introduce daqp_ai_get_sample()

Introduce a helper function to get a two's complement sample from
the FIFO and munge it to the offset binary format that comedi uses.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: there are 4 digital inputs

This board has 4 digital inputs not 1. Fix the subdevice init.

The digital input lines are shared with other functions. Add a
comment about this.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: quatech_daqp_cs: there are 4 digital outputs

This board has 4 digital outputs not 1. Fix the subdevice init.

The digital output lines are used for the external channel selection
when the expansion mode is enabled. Add a comment about this.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: pcm3724: update the MODULE_DESCRIPTION

Change the MODULE_DESCRIPTION to something more usefull than the
generic "Comedi low-level driver".

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: pcm3724: tidy up register map defines

For aesthetics, rename the register map defines so they have namespace
associated with the driver and use the BIT macro to define the bits.

Add the missing defines for the registers and use them to remove the
magic values from the driver.

Add a comment about the extra registers to explain why this driver
doesn't just use the standard 8255 driver instead.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: pcm3724: tidy up multi-line comments

Reformat the multi-line comments in the kernel CodingStyle.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging: make emxx_udc.c explicitly non-modular

The Kconfig currently controlling compilation of this code is:

drivers/staging/emxx_udc/Kconfig:config USB_EMXX
drivers/staging/emxx_udc/Kconfig:       bool "EMXX USB Function Device Controller"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.

The .remove function was declared __exit, so it wouldn't have been
available for a sysfs bind/unbind anyway, so lets be explicit here and
use ".suppress_bind_attrs = true" to prevent root from doing something
silly.

We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.

Cc: Vincenzo Scotti <vinc94@gmail.com>
Cc: Haneen Mohammed <hamohammed.sa@gmail.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Roberta Dobrescu <roberta.dobrescu@gmail.com>
Cc: Tapasweni Pathak <tapaswenipathak@gmail.com>
Cc: Peter Chen <peter.chen@freescale.com>
Cc: Robert Baldyga <r.baldyga@samsung.com>
Cc: Chris Rorvick <chris@rorvick.com>
Cc: "Gujulan Elango, Hari Prasath (H.)" <hgujulan@visteon.com>
Cc: devel@driverdev.osuosl.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging: make android timed_output.c explicitly non-modular

The Kconfig currently controlling compilation of this code is:

drivers/staging/android/Kconfig:config ANDROID_TIMED_OUTPUT
drivers/staging/android/Kconfig:        bool "Timed output class driver"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

Since module_init translates to device_initcall in the non-modular
case, the init ordering remains unchanged with this commit.

We replace module.h with init.h and export.h ; the latter since this
file does actually export some symbols.

We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.

Cc: "Arve Hjønnevåg" <arve@android.com>
Cc: Riley Andrews <riandrews@android.com>
Cc: devel@driverdev.osuosl.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging: make android lowmemorykiller.c explicitly non-modular

The Kconfig currently controlling compilation of this code is:

drivers/staging/android/Kconfig:config ANDROID_LOW_MEMORY_KILLER
drivers/staging/android/Kconfig:        bool "Android Low Memory Killer"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

Since module_init translates to device_initcall in the non-modular
case, the init ordering remains unchanged with this commit.

We replace module.h with init.h and moduleparam.h ; the latter since
this file was previously implicitly relying on getting that header.

We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.

Cc: "Arve Hjønnevåg" <arve@android.com>
Cc: Riley Andrews <riandrews@android.com>
Cc: devel@driverdev.osuosl.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging: make android sw_sync.c explicitly non-modular

The Kconfig currently controlling compilation of this code is:

staging/android/Kconfig:config SW_SYNC
staging/android/Kconfig:        bool "Software synchronization objects"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

Since module_init translates to device_initcall in the non-modular
case, the init ordering remains unchanged with this commit.

We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.

Cc: "Arve Hjønnevåg" <arve@android.com>
Cc: Riley Andrews <riandrews@android.com>
Cc: devel@driverdev.osuosl.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging: make android ashmem.c explicitly non-modular

The Kconfig currently controlling compilation of this code is:

drivers/staging/android/Kconfig:config ASHMEM
drivers/staging/android/Kconfig:        bool "Enable the Anonymous Shared Memory Subsystem"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

Since module_init translates to device_initcall in the non-modular
case, the init ordering remains unchanged with this commit.

We replace module.h with init.h and export.h ; the latter since this
file uses the global THIS_MODULE.

We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.

Cc: "Arve Hjønnevåg" <arve@android.com>
Cc: Riley Andrews <riandrews@android.com>
Cc: devel@driverdev.osuosl.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging: make android tegra_ion.c properly tristate

The Kconfig currently controlling compilation of this code is:

drivers/staging/android/ion/Kconfig:config ION_TEGRA
drivers/staging/android/ion/Kconfig:    tristate "Ion for Tegra"

...which led me to incorrectly conclude this file was built modular
earlier.  However the above CONFIG is just used to enter the dir and
once we do enter that dir, we see the build is unconditional:

drivers/staging/android/ion/Makefile:obj-$(CONFIG_ION_TEGRA) += tegra/
drivers/staging/android/ion/tegra/Makefile:obj-y += tegra_ion.o

...meaning that it currently is not being built as a module by anyone.

However, given that the Kconfig did explicitly choose tristate, and that
the dummy ion driver is (functionally) tristate, I chose to make the
Makefile do the right thing for it to build as a module.

After this change, on an ARM allmodconfig, we see:

  CC [M]  drivers/staging/android/ion/tegra/tegra_ion.o

so it does build OK as a module.  I can't vouch for the modular
functionality however, so consider this compile tested only.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Arve Hjønnevåg" <arve@android.com>
Cc: Riley Andrews <riandrews@android.com>
Cc: Stephen Warren <swarren@wwwdotorg.org>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Alexandre Courbot <gnurou@gmail.com>
Cc: Markus Elfring <elfring@users.sourceforge.net>
Cc: devel@driverdev.osuosl.org
Cc: linux-tegra@vger.kernel.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/staging: make android ion_page_pool.c explicitly non-modular

The Kconfig currently controlling compilation of this code is:

drivers/staging/android/ion/Kconfig:menuconfig ION
drivers/staging/android/ion/Kconfig:    bool "Ion Memory Manager"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

Since module_init translates to device_initcall in the non-modular
case, the init ordering remains unchanged with this commit.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Arve Hjønnevåg" <arve@android.com>
Cc: Riley Andrews <riandrews@android.com>
Cc: devel@driverdev.osuosl.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: olpc_dcon: Remove braces

Braces in single statement blocks are not needed.

Found by checkpatch.pl

Signed-off-by: Ksenija Stanojevic<ksenija.stanojevic@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: simplify a trivial if-return sequence

Signed-off-by: Parshuram Thombare <thombarepr@yahoo.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgap: remove unused configuration

The IO configuration was not used to configure the board. It was only
read from the configuration file. Stop reading it and also remove the
other related variables defined for it.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: remove parenthesis

checkpatch was warning us about extra unneeded parenthesis.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: alignment style

checkpatch was warning us that the alignment should match the open
parenthesis.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: remove space after cast

Space is not necessary after typecast.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: remove blankline after brace

Blank lines are not needed after opening braces. checkpatch was giving
us warnings about this.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: remove blankline before brace

Blank lines are not needed before closing braces. checkpatch was giving
warning about this.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: remove multiple blank lines

checkpatch warns us about multiple blank lines which are not needed.
Remove them.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: move open brace in line with control flow statement

Move open braces to be in line with "if" control flow statements.
Addresses checkpatch.pl:
ERROR: that open brace { should be on the previous line

Signed-off-by: Alison Schofield <amsfield22@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: wilc1000: linux_wlan: Remove NULL check before kfree

kfree on NULL pointer is a no-op.

The semantic patch used -

// <smpl>
@@ expression E; @@
- if (E != NULL) { kfree(E); }
+ kfree(E);
@@ expression E; @@
- if (E != NULL) { kfree(E); E = NULL; }
+ kfree(E);
+ E = NULL;
// </smpl>smpl>

Signed-off-by: Shraddha Barke <shraddha.6596@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: wilc1000: wilc_wfi_cfgoperations: Remove NULL check before kfree

kfree on NULL pointer is a no-op.

The semantic patch used-

// <smpl>
@@ expression E; @@
- if (E != NULL) { kfree(E); }
+ kfree(E);
@@ expression E; @@
- if (E != NULL) { kfree(E); E = NULL; }
+ kfree(E);
+ E = NULL;
// </smpl>smpl>

Signed-off-by: Shraddha Barke <shraddha.6596@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: wilc1000: wilc_msgqueue: Use kmemdup instead of kmalloc and memcpy

Replace kmalloc and memcpy with kmemdup.
Problem found using coccicheck

Signed-off-by: Shraddha Barke <shraddha.6596@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove braces around single statement blocks

Remove excess braces as suggested by checkpatch.pl:
WARNING: braces {} are not necessary for single statement blocks

Signed-off-by: Alison Schofield <amsfield22@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: linux_wlan_spi: include header

A check using 'sparse' shows warnings in linux_wlan_spi.c:

drivers/staging/wilc1000/linux_wlan_spi.c:43:19: warning: symbol 'wilc_spi_dev' was not declared. Should it be static?
drivers/staging/wilc1000/linux_wlan_spi.c:71:19: warning: symbol 'wilc_bus' was not declared. Should it be static?
drivers/staging/wilc1000/linux_wlan_spi.c:95:5: warning: symbol 'linux_spi_init' was not declared. Should it be static?
drivers/staging/wilc1000/linux_wlan_spi.c:195:5: warning: symbol 'linux_spi_write' was not declared. Should it be static?
drivers/staging/wilc1000/linux_wlan_spi.c:320:5: warning: symbol 'linux_spi_read' was not declared. Should it be static?
drivers/staging/wilc1000/linux_wlan_spi.c:365:5: warning: symbol 'linux_spi_write_read' was not declared. Should it be static?
drivers/staging/wilc1000/linux_wlan_spi.c:402:5: warning: symbol 'linux_spi_set_max_speed' was not declared. Should it be static?

Let's avoid it by including "linux_wlan_spi.h" header.

Signed-off-by: Stanislav Kholmanskikh <kholmanskikh.s.s@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: linux_wlan_spi: correct types

linux_spi_write(), linux_spi_read(), linux_spi_write_read():

    nwi->io_func.u.spi.spi_tx = linux_spi_write;
    nwi->io_func.u.spi.spi_rx = linux_spi_read;
    nwi->io_func.u.spi.spi_trx = linux_spi_write_read;

are expected to accept arguments of 'u8 *', u32 types:

    struct {
        int (*spi_max_speed)(void);
        int (*spi_tx)(u8 *, u32);
        int (*spi_rx)(u8 *, u32);
        int (*spi_trx)(u8 *, u8 *, u32);
    } spi;

However, linux_spi_read() and linux_spi_write_read() do not do this,
they use 'unsigned char *' and 'unsigned long' instead.

Changed the types of their arguments to satisfy the expectations.

Signed-off-by: Stanislav Kholmanskikh <kholmanskikh.s.s@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename tenuAuth_type of struct host_if_wep_attr

This patch renames tenuAuth_type of struct host_if_wep_attr to auth_type
to avoid CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename u8mode of struct host_if_wep_attr

This patch renames u8mode of struct host_if_wep_attr to mode to avoid
CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename u8Wepidx of struct host_if_wep_attr

This patch renames u8Wepidx of struct host_if_wep_attr to index to avoid
CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename u8WepKeylen of struct host_if_wep_attr

This patch renames u8WepKeylen of struct host_if_wep_attr to key_len to
avoid CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename pu8WepKey of struct host_if_wep_attr

This patch renames pu8WepKey of struct host_if_wep_attr to key in order
to avoid CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename strHostIFpmkidAttr of union host_if_key_attr

This patch renames strHostIFpmkidAttr of union host_if_key_attr to
pmkid to avoid CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename strHostIFwpaAttr of union host_if_key_attr

This patch renames strHostIFwpaAttr of union host_if_key_attr to wpa to
avoid CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename strHostIFwepAttr of union host_if_key_attr

This patch renames strHostIFwepAttr of union host_if_key_attr to wep to
avoid CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename u32SetCfgFlag of struct cfg_param_val

This patch renames u32SetCfgFlag of struct cfg_param_val to flag to
avoid CamelCase naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename pstrCfgParamVal in struct cfg_param_attr

This patch renames a variable of struct cfg_param_val in struct
cfg_param_attr, pstrCfgParamVal to cfg_attr_info to avoid CamelCase
naming convention.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: replace drvHandler and hWFIDrv with hif_drv

This patch replaces the variable names of struct host_if_drv used as the
functions' input parameter, drvHandler and hWFIDrv with hif_drv. In
addition, the local variable declared in many functions, pstrWFIDrv is
removed and hif_drv is directly used. A debug message printing pstrWFIDrv is
deleted while removing the local variable because it is not useful as well.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove wilc_wlan.c included in wilc_wfi_cfgoperations.c

This patch removes "wilc_wlan.c" from the wilc_wfi_cfgoperation.c file and
adds wilc_wlan.o into Makefile to compile it because there is few benefits.
This patch also adds "wilc_wfi_netdevice.h" in the wilc_wlan.c file to avoid
the compile errors.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: host_infterface.h: remove unused enum tenuWILC_StaFlag

This patch removes unused enum tenuWILC_StaFlag from the
host_interface.h.

Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tenuScanConnTimer

This patch removes typedef from the enum tenuScanConnTimer and
renames it to scan_conn_timer.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tenuConnDisconnEvent

This patch removes typedef from the enum tenuConnDisconnEvent and
renames it to conn_event.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename typedef from tenuCfgParam

This patch rename typedef from the enum tenuCfgParamand
rename it to cfg_param.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tenuScanEvent

This patch removes typedef from the enum tenuScanEvent and
rename it to scan_event.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tstrStatistics

This patch removes typedef from the struct tstrStatistics and
rename it to rf_info.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove unused struct cfg_param_t

This patch removes unused the struct cfg_param_t.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove unused struct WILC_WFIDrvHandle

This patch removes unused the struct WILC_WFIDrvHandle with related
commnets.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tstrHiddenNetworkInfo

This patch removes typedef from the struct tstrHiddenNetworkInfo with
related comments and renames it to hidden_net_info.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tstrFoundNetworkInfo

This patch removes typedef from the struct tstrFoundNetworkInfo and
rename it to found_net_info.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tenuHostIFstate

This patch removes typedef from the enum tenuHostIFstate and
rename it to host_if_state.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tstrWILC_UsrConnReq

This patch removes typedef from the struct tstrWILC_UsrConnReq with the
related comments and renames it to user_conn_req.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tstrWILC_UsrScanReq

This patch removes typedef from the struct tstrWILC_UsrScanReq with
related comments and renames it to user_scan_req.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove _tstrJoinParam from join_bss_param

This patch removes struct _tstrJoinParam from the struct join_bss_param
because it is not used inside the struct.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tstrJoinBssParam

This patch removes typedef from the struct tstrJoinBssParam
and renames it to join_bss_param.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove unused struct tstrBssTable

This patch removes struct tstrBssTable with related comment.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove unused struct tstrWidJoinReqExt

This patch removes struct tstrWidJoinReqExt because it's not used.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename the member variable, ps8WidVal of wid

This patch renames ps8WidVal of struct wid to val.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename the member variable, s32ValueSize of wid

This patch renames s32ValueSize of struct wid to size.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename the member variable, u16WIDid of wid

This patch renames u16WIDid of struct wid to id.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename the member variable, enuWIDtype of wid

This patch renames enuWIDtype of struct wid to type.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove typedef from tenuWIDtype

This patch remove typedef from the enum tenuWIDtype
and rename it to WID_TYPE.

Signed-off-by: Leo Kim <leo.kim@atmel.com>
Signed-off-by: Tony Cho <tony.cho@atmel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>