staging: comedi: kcomedilib: Fixed coding style issue

Fixed checkpatch.pl warning in kcomedilib_main.c:
 - Block comments use * on subsequent lines

Signed-off-by: Nayeemahmed Badebade <itachi.opsrc@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: goldfish: Prefer kernel type u32 over uint32_t

Prefer kernel type u32 over uint32_t to maintain uniformity.

Signed-off-by: Ravi Teja Darbha <ravi2j@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: goldfish: Fix NULL comparison style

Fixed NULL comparison style as suggested by checkpatch.pl with --strict
option.

Signed-off-by: Ravi Teja Darbha <ravi2j@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: remove unnecessary le32_to_cpu

The values passed to le32_to_cpu are already in the correct byte order.
This fixes four "cast to restricted __le32" sparse warnings.

Signed-off-by: Michał Bartoszkiewicz <mbartoszkiewicz@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: ste_rmi4: Remove unnecessary MODULE_ALIAS()

The driver has a I2C device id table that is used to create the modaliases
which already contains "synaptics_rmi4_ts". So the alias is not needed.

Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723au: remove unimplemented function declarations

Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Acked-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Cc: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8192e: r8192E_dev.c: Remove unused variable

This patch discards the variable as it is not used anywhere throughout
the kernel.

Build tested it.

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

Staging: rtl819u: ieee80211: Remove unused variable

This patch discards the variable count as it is not used anywhere
throughout the kernel.

Build tested it.

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

Staging: rtl8723au: core: Remove unused variable

This patch discards the variable as it is not used anywhere throughout
the kernel.

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

Staging: rtl8192e: rtl_core.c: Remove unused variable

This patch discards the variable ResetThreshold as it is not used
anywhere throughout the kernel.

Build tested it.

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

staging: panel: panel.c: Fixed coding style issues

Fixed warnings reported by checkpatch.pl:
 - Block comments use a trailing */ on a separate line
 - Block comments use * on subsequent lines

Signed-off-by: Nayeemahmed Badebade <itachi.opsrc@gmail.com>
Acked-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: Fix module autoload for OF platform driver

This platform driver has a OF device ID table but the OF module
alias information is not created so module autoloading won't work.

Signed-off-by: Luis de Bethencourt <luis@debethencourt.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: mt29f_spinand: Export OF module alias information

The SPI core always reports the MODALIAS uevent as "spi:<modalias>"
regardless of the mechanism that was used to register the device
(i.e: OF or board code) and the table that is used later to match
the driver with the device (i.e: SPI id table or OF match table).

So drivers needs to export the SPI id table and this be built into
the module or udev won't have the necessary information to autoload
the needed driver module when the device is added.

But this means that OF-only drivers needs to have both OF and SPI id
tables that have to be kept in sync and also the dev node compatible
manufacturer prefix is stripped when reporting the MODALIAS. Which can
lead to issues if two vendors use the same SPI device name for example.

To avoid the above, the SPI core behavior may be changed in the future
to not require an SPI device table for OF-only drivers and report the
OF module alias. So, it's better to also export the OF table even when
is unused now to prevent breaking module loading when the core changes.

Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/fbtft : Add missing whitespace around operators

Add blank spaces around operators where recommended
by checkpatch.pl

Signed-off-by: Anish Bhatt <anish@chelsio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/fbtft : Fix multiple/missing blank line issues

Remove or add blank lines as recommended by checkpatch.pl

Signed-off-by: Anish Bhatt <anish@chelsio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/fbtft : Remove unicode characters

Remove stray unicode quotes around name

Signed-off-by: Anish Bhatt <anish@chelsio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/fbtft : Remove repeated set_addr_win debug messages

fbtft_par_dbg(DEBUG_SET_ADDR_WIN.. ) is repeated in every
set_addr_win() handler, this could be replicated by using the kernel
function tracer instead.

Signed-off-by: Anish Bhatt <anish@chelsio.com>
Suggested-by: Greg KH <gregkh@linuxfoundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/fbtft : Use BIT() macro when possible

Based on checkpatch.pl recommendations, (1 << x)
is replaced by BIT(x)

Signed-off-by: Anish Bhatt <anish@chelsio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: fb_watterott: define backlight_ops statically

instead of devm_kzalloc'ing them

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: fb_ssd1351: define backlight_ops statically

instead of devm_kzalloc'ing them

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: fbtft_request_gpios: reduce nesting

Returning immediately if no platform_data or platform_data->gpios is
specified reduceis code nesting

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: fbtft-core: define backlight_ops statically

instead of devm_kzalloc'ing them

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: remove unused bl_ops from fbtft_unregister_backlight

The only usage of bl_ops variable in fbtft_unregister_backlight function
was assigment of a value to that variable, therefore the assignment and
the variable itself can be safely removed

Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/fbtft : Remove FSF mailing address

checkpatch.pl recommends that this is no longer required.
Also replaces ASCII-art copyright notice with simple text

Signed-off-by: Anish Bhatt <anish@chelsio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix quoted string split across lines code style issue

Quoted strings should not be split to help text grep in the source.
All quoted strings that were split have thus been merged to one unique
quoted string each to follow the code style.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix missing blank line after declarations code style issue

Adds whitespaces to separate the variables declarations and the
function content.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix unnecessary whitespace code style issue

Whitespaces are not necessary before a quoted newline. Remove those.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix unnecessary else after return code style issue

An else statement is not useful after a return.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix unnecessary parentheses code style issue

Two sets of parentheses were used to contain the same statement.
In those cases, one of them has been removed, as unnecessary.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix unnecessary check before kfree code style issue

kfree(NULL) is safe and the checks were not required.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: remove forward declarations in .c file

Checkpatch was giving a "externs should be avoided in .c files" because
of these forward declarations. As these were not useful in this case,
they have been removed.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix unecessary braces code style issue

braces {} are not necessary for any arm of a statement containing
one statement on each side.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: remove return statement of void function

void function return statement was not useful in this case.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: include linux/uaccess.h instead of asm/uaccess.h

Use #include <linux/uaccess.h> instead of <asm/uaccess.h>

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: clean C99 // comments

Replace C99 // comments by /* comments */ to follow the
kernel code style. Remove some unuseful comments.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: whitespace neatening to fix consistent spacing code style errors

Clean-up the file by using a cleaner spacing around symbols and
words. Mostly use the automatic checkpatch whitespacing fixes.
This takes care of the consistent spacing errors reported by
checkpatch.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix missing struct leading to consistent spacing code style error

A missing struct keyword in variable declaration triggered a
need consistent spacing around '*' code style error. The struct
keyword thus has been added everywhere for the rtl8192_rx_info
struct, and therefore its typedef removed as not needed anymore.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix else following close brace code style error

Fix "else should follow close brace" checkpatch error.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix code indent using spaces code style error

Fix "code indent should use tabs where possible" checkpatch error

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix space before close parenthesis code style error

A space existed before the close parenthesis of an if statement. This patch
removes it to follow the kernel code style.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: r8192U_core: fix switch and case indent code style error

Some switch and case were not be at the same indent level.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: most: Use NULL instead of zero

This patch fixes the warning generated by sparse
"Using plain integer as NULL pointer" by using NULL
instead of zero.

Signed-off-by: Ronit halder <ronit.crj@gmail.com>
Acked-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: NULL comparison style

checkpatch complains when a variable comparison to NULL is written as:
variable == NULL or variable != NULL.

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

staging: most: fix Makefile

The Makefile is including "drivers/media/video". But there is no such
directory in kernel tree. Since it is aim-v4l2 this might have been
"drivers/media/v4l2-core", but the Kconfig already mentions that it
depends on VIDEO_V4L2. So no need to mention that again in the Makefile.

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

staging: most: remove unused variable

The variable conf was only assigned the value but was never used.

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

staging: most: use NULL pointer

sparse was complaining that an integer is used as NULL pointer. Fix it
by using NULL.

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

staging: most: make functions static

split_arg_list() and audio_set_pcm_format() are being called from the
same file and is not referenced from outside, so make them as static.

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

staging: most: remove unused functions

These functions were only defined but not used anywhere.

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

staging: most: out of memory error

If kzalloc fails it will print lots of debugging information in the log,
no need to have another in the code.

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

staging: most: remove multiple blank line

Multiple blank lines are not recommended in the kernel coding style.

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

staging: most: bool comparison style

Mentioning true or false in the if comparison is error prone and also
not according to the coding style.

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

staging/rtl8723au: Use %pM format specifier to print mac address

printk() supports %pM format specifier for printing 6-byte MAC/FDDI
addresses in hex notation small buffers, let's use it intead of %x:%x...

Signed-off-by: Alexander Kuleshov <kuleshovmail@gmail.com>
Acked-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/rtl8188eu: Use %pM format specifier to print mac address

printk() supports %pM format specifier for printing 6-byte MAC/FDDI
addresses in hex notation small buffers, let's use it intead of %x:%x...

Signed-off-by: Alexander Kuleshov <kuleshovmail@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: simplify 'memset' of 2D array

This patch simplifies the 'memset' done on a static 2D array.

Signed-off-by: Hari Prasath Gujulan Elango <hgujulan@visteon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: change address to fixed value

The linux_wlan_init_test_config() is called once when net driver is loaded.
And because the pointer type of the pstrWFIDrv is changed with the interger
type, this patch replaces it with designated value instead of pointer.

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

staging: wilc1000: add an argument for Handle_SetWfiDrvHandler

Similar to functions of same layer, this patch add an argument for
Handle_SetWfiDrvHandler function. As a result, the redundant typecasting is
removed.

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

staging: wilc1000: use the real data type

This patch changes the type of gu8FlushedJoinReqDrvHandler with his real
data type becasue typecasting is not necessary. In result, typecasting
which is not necessary and some building warnings is removed.

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

staging: wilc1000: rename WILC_MsgQueueDestroy to wilc_mq_destroy

This patch replaces WILC_MsgQueueDestroy to wilc_mq_destroy to
shorten function name and avoid CamelCase.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename WILC_MsgQueueRecv to wilc_mq_recv

This patch replaces WILC_MsgQueueRecv with wilc_mq_recv
to shorten function name and avoid CamelCase.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename WILC_MsgQueueSend to wilc_mq_send

This patch replaces WILC_MsgQueueSend with wilc_mq_send to
shorten function name and avoid CamelCase.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: rename WILC_MsgQueueCreate to wilc_mq_create

This patch replaces WILC_MsgQueueCreate with wilc_mq_create to
shorten function name and avoid CamelCase.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: replace WILC_ErrNo by int type

This patch replaces WILC_ErrNo with int type.
WILC_ErrNo typedef is also removed.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: wilc_msgqueue.c: fix bool comparison style

This patch changes bool comparison style found by checkpatch.pl
CHECK: Using comparison to true is error prone

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: wilc_msgqueue.c: remove kfree NULL check

This patch removes kfree NULL check.
WARNING: kfree(NULL) is safe and this check is probably not required.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: wilc_msgqueue.c: use ! operator instead of NULL comparison

This patch uses ! operator instead of NULL comparison.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: wilc_msgqueue.c: remove unnecessary parentheses

This patch remove unnecessary parentheses found by checkpatch.pl
CHECK: Unnecessary parentheses around pHandle->hSem

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: wilc_msgqueue.c: remove blank line before a close brace

This patch removes blank line before a close brace "}"
CHECK: Blank lines aren't necessary before a close brace '}'

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: wilc_msgqueue.c: remove multiple blank lines

This patch removes multiple blank lines.
CHECK: Please don't use multiple blank lines

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: wilc_msgqueue.c: remove blank line after open brace

This patch removes a blank line for open brace "{"
CHECK: Blank lines aren't necessary after an open brace '{'

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: wilc_msgqueue.c: remove braces for single statement

This patch removes braces for single statement blocks.
WARNING: braces {} are not necessary for single statement blocks

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: add a blank line after declaration

This patch adds a blank line after declaration found by checkpatch.pl
WARNING: Missing a blank line after declarations.

Signed-off-by: Chaehyun Lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: replace MALLOC_WILC_BUFFER() macro to avoid possible memory leak

The MALLOC_WILC_BUFFER() macro was using a return statement, and didn't
take care of possible memory leaks and subsequent bugs when it was failing
after succeeding some allocations. This patch corrects this behavior.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: remove FREE_WILC_BUFFER()

It was just a wrapper around kfree(), so call that instead.

Signed-off-by: Raphaël Beamonte <raphael.beamonte@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: fix coding style checks in sw_sync.c

replace comparison "obj" to NULL with "!obj"

Signed-off-by: Trung Thanh Le <trungthanh1608@yahoo.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ashmem.c: 'Alignment' style issue

fix two CHECK issues by checkpatch.pl with --strict:
Alignment should match open parenthesis

Signed-off-by: Peng Sun <sironhide0null@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ashmem.c: 'No space necessary after' style issue

fix two CHECK issues by checkpatch.pl with --strict:
No space is necessary after a cast

Signed-off-by: Peng Sun <sironhide0null@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: android: ashmem.c: 'spaces preferred around that -' style issue

fix a CHECK style issue by checkpatch.pl with --strict:
spaces preferred around that '-'

Signed-off-by: Peng Sun <sironhide0null@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: replace typedef struct tagSRxDesc

with struct vnt_rx_desc and all members the same.

volatile is removed from pointers as this generates warning
message.

Only the first four members of vnt_rx_desc need to be volatile.

Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: replace typedef struct tagDEVICE_RD_INFO

with struct vnt_rd_info

volatile is removed because it will generate a warning
(in any case this member is not) and renaming rd_info.

Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: replace typedef struct tagRDES1

with struct vnt_rdes1 and members
wReqCount -> req_count
wReserved -> reserved

Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: replace typedef struct tagRDES0

with struct vnt_rdes0 replacing members as follows
wResCount -> res_count
f15Reserved -> f15_reserved
f1Owner -> owner

big endian
f8Reserved1 -> f8_reserved1
f7Reserved -> f7_reserved

Narrowing endian differences to inside structure.

Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: usbduxfast: remove unnecessary clearing of private data

The clearing of the private data members in the (*detatch) is not
necessary. The comedi core is going to kfree the private data as soon
as the (*detach) finishes.

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: usbduxfast: absorb usbduxfast_attach_common()

This function is only called by the (*auto_attach) to do the final
setup of the analog input subdevice.

For aesthetics absorb it into usbduxfast_auto_attach() and reorder
the initialization of the subdevice to better follow the "norm" in
comedi drivers.

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: usbduxfast: remove unnecessary semaphore down/up

The semaphore down/up in usbduxfast_attach_common() is not necessary.
This function is only called as part of the (*auto_attach) and does
not talk to the USB device.

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: usbduxfast: document strange ai maxdata

The subdevice 'maxdata' is typically a mask of the valid bits that can
be returned by a subdevice, (1 << bits) - 1. The analog inputs of this
device have a resolution of 12-bits so the 'maxdata' should be 0x0fff.
But, this hardware can produce a value of 0x1000 indicating an overflow
from the ADC. The comedilib library's comedi_to_phys() function will
then return NAN when this value is read from the hardware.

Add a comment to clarify the strage 'maxdata' 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: usbduxfast: use a common exit path in ai (*do_cmd)

For aesthetics, use 'goto' and a common exit point in this function so
that the up of the semaphore is not overlooked.

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: usbduxfast: tidy up the conversion timing validation

Cleanup the validation of the cmd->convert_arg to use the same calculation
in the (*do_cmdtest) and the (*do_cmd).

Remove the, then unnecessary, redundant check in the (*do_cmd).

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: usbduxfast: TRIG_EXT is only valid for cmd->start_src

According to the code, and verified with the author Bernd Porr, the
TRIG_EXT trigger is only valid for the cmd->start_src. Remove this
trigger from the cmd->scan_begin_src and cmd->convert_src.

The only valid scan_begin_src is then TRIG_FOLLOW  and the convert_src
is only TRIG_TIMER. Remove the redundant unique trigger checks in Step
2b of the (*do_cmdtest) as well as any other unnecessary checks of
those trigger sources.

Also remove the incorrect comment for a cmd->stop_src of TRIG_EXT.

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: usbduxfast: TRIG_EXT only works for 1 or 16 channels

Move this check from the (*do_cmd) to the (*do_cmdtest).

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: usbduxfast: fix multi-line comment

This comment does not need to be multi-line. Fix it to remove the
checkpatch warning.

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: usbduxfast: use preferred kernel types

As suggested by checkpatch.pl:

CHECK: Prefer kernel type 'u8' over 'uint8_t'
CHECK: Prefer kernel type 's8' over 'int8_t'

For consistency, also change the uint16_t and int16_t types to
u16 and s16.

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: usbduxfast: remove impossible 'if (!devpriv)' checks

The private data 'devpriv' is allocated in the (*auto_attach) before
the subdevices are initialized and the (*auto_attach) has to complete
successfully before the core will call any of the subdevice operations.
Because of this the (!devpriv) checks in the subdevice code can never
fail. Remove them.

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: usbduxfast: only chanlist of 1, 2, 3, or 16 are supported

The analog input (*do_cmd) only supports channel lists of 1, 2, 3, or 16
channels. Add a check for this to usbduxfast_ai_check_chanlist() and remove
the check from usbduxfast_ai_cmd().

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: usbduxfast: chanlist check is Step 5 of (*do_cmdtest)

The channel list should be checked in Step 5 of the (*do_cmdtest) not
as part of the (*do_cmd). Factor the check out of usbduxfast_ai_cmd().

Tidy up the factored out code. The channel number 'i' will never be
greater than NUMCHANNELS due to the subdevice setup and the checks
done in the code. The up/down of the semaphore is also not needed
because the (*do_cmdtest) never actually tries to access 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: s526: fix the block comments

Tidy up the block comments to fix the checkpatch.pl warnings.

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: s526: remove #if 0'ed out code block

This #if 0'ed out code is not necessary. It's actually just a copy
of the psuedo code in the S526 Hardware Manual that shows an example
of a counter application. Remove 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: s526: remove unnecessary defines

Remove a couple unused defines,

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: s526: remove s526_ai_insn_config()

This (*insn_config) does not follow the comedi core API. It also
would not work as expected.

It appears to be trying to configure the analog input subdevice so
that the (*insn_read) would read multiple channels (data[0]) and
optionally enable the 15us delay (data[1]) needed for the multiplexor
to change channels between samples.

Unfortunately, the comedi core expects (*insn_read) operations to
return 1 or more samples for a single channel, which is what the
(*insn_read) in this driver does.

The (*insn_config) is also enabling the analog input end-of-conversion
interrupt. This isn't needed, and might be a problem since the driver
does not currently request and interrupt. The enable bit does not
need to be set for the end-of-conversion to occur in the interrupt
status register.

Remove the (*insn_config) and modify the (*insn_read) to automatically
handle the 15us delay when needed.

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: s526: tidy up s526_ai_rinsn()

Rename this function to follow the normal naming in comedi drivers.

Use the comedi_offset_munge() helper to munge the hardware two's
complement data to the comedi offset binary format.

Change the final return to insn->n to clarify the return value without
the need for the comment.

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: s526: wait for analog output conversions to complete

It's possible for the user to write more than one value to an analog
output channel with the (*insn_write) operation. Use comedi_timeout()
to check the interrupt status register to ensure that each data
conversion is complete before writing the next 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>