staging: lustre: Use min3 macro

Replace comparsions between three expressions using
two min macros with min3 macro. Done using Coccinelle
Script:

@@
expression e1;
expression e2;
expression e3;
@@
(
- min(min(e1, e2), e3)
+ min3(e1, e2, e3)
|
- min(e1, min(e2, e3))
+ min3(e1, e2, e3)
)

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: remove item from TODO list

TODO file contains task to verify and correct function return
sites. Need to check for and implement correct usage of goto's when
there is work to be done before returning.

Remove task from TODO list after already having completed audit of
directory drivers/staging/dgnc.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: audit goto's in dgnc_tty

TODO file requests fix up of error handling.

Audit dgnc_mgmt.c and fix all return paths to be uniform and inline
with kernel coding style.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: audit goto's in dgnc_mgmt

TODO file requests fix up of error handling.

Audit dgnc_mgmt.c and fix all return paths to be uniform and inline
with kernel coding style.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: dgnc: audit goto's in dgnc_driver

TODO file requests fix up of error handling.

Audit dgnc_driver.c and fix all return paths to be uniform and inline
with kernel coding style.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchi: Remove ARM64 from TODO list

ARM64 for core vchiq which is the core of vc04_services should
now be work complete.  The driver compiles without any errors
or warnings, and works just as well as 32 bit mode.  The
necessary compatibility wrappers for the 32 bit ioctls have been
written and merged.

Since no more ARM64 specific changes should be needed, perhaps
it's best to remove it from the TODO list.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: use kernel preferred style for handling errors

This patch replaces NULL error values with error pointer values.

Signed-off-by: Aishwarya Pant <aishpant@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: replace kmalloc with kzalloc

This patch replaces kmalloc and memset with kzalloc

Signed-off-by: Aishwarya Pant <aishpant@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: add check to avoid null pointer dereference

This patch adds checks after memory allocation to avoid possible null
pointer dereferences.

Signed-off-by: Aishwarya Pant <aishpant@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: remove depends on ARM

Since all the arm64 specific issues have been fixed now
and the camera is working fine with a arm64 kernel, the
depends on ARM can be removed from Kconfig.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Fix bogus compiler warnings regarding constants

In debug logging code, the compiler is warning about imposible
situations and size of constants not matching the format specifier.

This change fixes all three instances of this.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Fix buffer overflow calculation on query of camera properties

The code that queries properties on the camera has a check
for buffer overruns if the firmware sends too much data.  This
check is incorrect, and during testing I was seeing stack corruption.

I believe this error can actually happen in normal use, just for
some reason it doesn't appear on 32 bit as often.  So perhaps
it's best for the check to be fixed.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Convert spinlock to mutex in handle mapping code

The handle mapping code that converts context pointers to handles uses
a spinlock.  Since the btree implementation can sleep while allocating
memory, turning on several kernel debugging options will result in
errors in the log.

Since this code path is never called in atomic context, perhaps it's
better to just use a mutex.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Convert struct mmal_buffer_header info fields to u32

The struct mmal_buffer_header has multiple fields used for informational
and debugging purposes.  These are safe to convert to u32.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Convert struct mmal_port info fields to u32

The struct mmal_port has a few informational fields. Convert these to
u32.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Convert client_context field to a 32 bit handle

The client_context field is passed around which is really just a pointer
to a msg_context.  A lookup table mechanism for msg_context was added
previously, so convert this field to a handle as well.

The firmware never interperates the client_context, just passed it back.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Convert delayed_buffer to u32

A delayed buffer field is passed between the firmware and the
kernel.  This field is never used either so it's safe to
change it to a u32.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Convert client_component field to u32

In the messages passed back and forth between the camera and
the firmware, a client_component field is passed.

This is a pointer to a structure that represents part of the
camera. Luckly, it's only used for debug logging, so simply
convert it to a u32.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-camera: Create struct mmal_es_format_local to mirror struct mmal_es_format

The struct struct mmal_es_format is passed between the firmware which has
pointers.  A local version of mmal_es_format is also used.

Luckly, the two versions are always memberwise copied from each other
so simply have different structures for the local and msg versions.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: firmware: Convert sscanf calls to strtoul

Also convert the fw_update_type and fw_timeout variables to
unsigned and update the printf specifier to %u.

The FW_MGMT_IOC_SET_TIMEOUT_MS ioctl takes an unsigned int
and checkpatch was complaining about not checking the sscanf
return values.

Signed-off-by: Michael Sartain <mikesart@fastmail.com>
Acked-by: Viresh Kumar <viresh.kumar at linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: firmware: Change long long unsigned to unsigned long long

Fixes checkpatch warning:
  type 'long long unsigned int' should be specified
  in [[un]signed] [short|int|long|long long] order

Signed-off-by: Michael Sartain <mikesart@fastmail.com>
Acked-by: Viresh Kumar <viresh.kumar at linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: firmware: Remove extra braces from single line if

Fixes checkpatch warning:
  braces {} are not necessary for any arm of this statement

Signed-off-by: Michael Sartain <mikesart@fastmail.com>
Acked-by: Viresh Kumar <viresh.kumar at linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: firmware: Remove trailing semicolon from FW_TIMEOUT_DEFAULT

Fixes checkpatch warning:
  macros should not use a trailing semicolon

Signed-off-by: Michael Sartain <mikesart@fastmail.com>
Acked-by: Viresh Kumar <viresh.kumar at linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Refactor conditionals

Refactor conditionals to reduce one level of indentation and improve
code readability.

Signed-off-by: Narcisa Ana Maria Vasile <narcisaanamaria12@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Remove error message on kmalloc() failure

Remove 'Out of memory' message because kmalloc already prints a message
in case of error.

Signed-off-by: Narcisa Ana Maria Vasile <narcisaanamaria12@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ks7010: removed code in comments.

Commenting Code Is a Bad Idea.
Comments are their to explain the code and how the code achieves its
goal and as codes in the comments  does not explain what the code is
doing so there is no use of commenting them.
So in this patch codes in the comments are removed.

Signed-off-by: Arushi Singhal <arushisinghal19971997@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: Remove useless cast.

Variable dac_data is already declared as of type u8. Again explicit type
casting of dac_data to u8, is not required. Hence this patch removes it
by using the following coccinelle script.

@@
type T;
T *ptr;
T p;
@@
(
- (T *)(&p)
+ &p
|
- (T *)ptr
+ ptr
|
- (T *)(ptr)
+ ptr
|
- (T)(p)
+ p
)

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Use ARRAY_SIZE macro

Use ARRAY_SIZE to calculate the size of an array.
The semantic patch used can be found here:
https://github.com/coccinelle/coccinellery/blob/master/arraysize/array.cocci

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wlan-ng: add byte order annotation to struct p80211_caphdr

Fixes the following sparse warnings around line 3514 in hfa384x_usb.c:

warning: incorrect type in assignment (different base types)
   expected unsigned int [unsigned] [usertype] version
   got restricted __be32 [usertype] <noident>

Signed-off-by: Aapo Vienamo <aapo.vienamo@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_utils: Don't include headers twice

There is no need to include types.h and vmalloc.h twice.

This issue has been found by make includecheck.

Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Clean up tests if NULL returned on failure

Some functions like kmalloc/kzalloc return NULL on failure.
When NULL represents failure, !x is commonly used.

This was done using Coccinelle:
@@
expression *e;
identifier l1;
@@

e = \(kmalloc\|kzalloc\|kcalloc\|devm_kzalloc\)(...);
...
- e == NULL
+ !e

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: sm750fb: Removed unnecessary parentheses.

Removed parentheses on the right hand side of assignment, as they are
not required. The following coccinelle script was used to fix this
issue:

@@
local idexpression id;
expression e;
@@

id =
-(
e
-)

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rts5208: Remove unnecessary cast on void pointer

The following Coccinelle script was used to detect this:
@r@
expression x;
void* e;
type T;
identifier f;
@@
(
  *((T *)e)
|
  ((T *)x)[...]
|
  ((T*)x)->f
|
- (T*)
  e
)

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gdm724x: Replace ternary operator with min macro

Use macro min() to get the minimum of two values for
brevity and readability. The macro MUX_TX_MAX_SIZE
has a value of 2048 which is well within the integer
limits. This check was done manually.

Found using Coccinelle:
@@ type T; T x; T y; @@
(
- x < y ? x : y
+ min(x,y)
|
- x > y ? x : y
+ max(x,y)
)

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: Fix sparse warnings in r8192U_dm.c

Fix these warnings:
drivers/staging//rtl8192u/r8192U_dm.c:2307:49: warning: cast from restricted __le16
drivers/staging//rtl8192u/r8192U_dm.c:2308:44: warning: cast from restricted __le16
drivers/staging//rtl8192u/r8192U_dm.c:2309:44: warning: cast from restricted __le16

Signed-off-by: Matthieu Simon <gmatthsim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: fix typos in macros

Rename macro definition and usage to represent correct spelling of DEFAULT:
ODM_REG_RX_DEFUALT_A_11N => ODM_REG_RX_DEFAULT_A_11N
ODM_REG_RX_DEFUALT_B_11N => ODM_REG_RX_DEFAULT_B_11N

Signed-off-by: Sebastian Haas <sehaas@deebas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: fix typos in comments

Fix typos reported by checkpatch.pl

Signed-off-by: Sebastian Haas <sehaas@deebas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: Clean up tests if NULL returned on failure

Some functions like kmalloc/kzalloc return NULL on failure.
When NULL represents failure, !x is commonly used.

This was done using Coccinelle:
@@
expression *e;
identifier l1;
@@

e = \(kmalloc\|kzalloc\|kcalloc\|devm_kzalloc\)(...);
...
- e == NULL
+ !e

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8192u: clean up some white space issues

This patch fixes two coding style errors, reported by the checkpatch
script.

Signed-off-by: Elia Geretto <elia.f.geretto@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192e: Indent the code

Indent the code in order to follow the rules and to
increase the readability of the code.

Signed-off-by: Georgiana Rodica Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192e: Remove useless type conversion

Some type conversions like casting a pointer/non-pointer to a pointer of same type,
casting to the original type using addressof(&) operator, etc. are not needed.
Therefore, remove them. Done using coccinelle:

@@
type t;
t *p;
t a;
@@
(
- (t)(a)
+ a
|
- (t *)(p)
+ p
|
- (t *)(&a)
+ &a
)

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxd2099: Remove redundant code

Removed redundant code marked with #if 0 to fix the following
checkpath.pl issue: CHECK: if this code is redundant consider
removing it.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Reviewed-by: Daniel Baluta <daniel.baluta@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxd2099: Change kzalloc parameter

Changed the structure tag 'struct cxd' with the variable '*ci' to fix the
following checkpath.pl issue: CHECK: Prefer kzalloc(sizeof(*ci)...) over
kzalloc(sizeof(struct cxd)).

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxd2099: Remove multiple blank lines

Removed multiple blank lines to fix the checkpath.pl issue:
CHECK: Please don't use multiple blank lines.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: cxd2099: Add spaces around arithmetical operators

Added spaces around multiple arithmetical operators
to fix the checkpatch.pl issue: CHECK: spaces preferred around that 'operator'.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: Remove unnecessary parentheses

Fix the checkpatch warnings of kind:

Unnecessary parentheses around priv->apTD1Rings[0]

Remove the parentheses around variables
being referenced.

Signed-off-by: Sreya Mittal <sreyamittal5@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: media: atomisp: Match alignment with open parenthesis.

This patch matches alignment with open parenthesis, removes extra space
before '||' operators and fixes the checkpatch issue.

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: media: atomisp: Remove useless cast.

Explicit type casting of variable, with same type as that of variable,
is not required. The following coccinelle script is used to remove it.

@@
type T;
T *ptr;
T p;
@@
(
- (T *)(&p)
+ &p
|
- (T *)ptr
+ ptr
|
- (T *)(ptr)
+ ptr
|
- (T)(p)
+ p
)

This patch also removes unnecessary parentheses.

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: Remove max_u64 function

Removed the max_u64 function to find the maximum value of
two unsigned 64 bit numbers to use the linux macro max
instead.

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lustre: Remove unnecessary cast on void pointer

The following Coccinelle script was used to detect this:
@r@
expression x;
void* e;
type T;
identifier f;
@@
(
  *((T *)e)
|
  ((T *)x)[...]
|
  ((T*)x)->f
|
- (T*)
  e
)

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: Remove unnecessary cast on void pointer

The following Coccinelle script was used to detect this:
@r@
expression x;
void* e;
type T;
identifier f;
@@
(
  *((T *)e)
|
  ((T *)x)[...]
|
  ((T*)x)->f
|
- (T*)
  e
)

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: nvec: Remove unnecessary cast on void pointer

The following Coccinelle script was used to detect this:

@r@
expression x;
void* e;
type T;
identifier f;
@@
(
  *((T *)e)
|
  ((T *)x)[...]
|
  ((T*)x)->f
|
- (T*)
  e
)

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8712: Remove unnecessary cast on void pointer

The following Coccinelle script was used to detect this:
@r@
expression x;
void* e;
type T;
identifier f;
@@
(
  *((T *)e)
|
  ((T *)x)[...]
|
  ((T*)x)->f
|
- (T*)
  e
)

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8712: rtl871x_mp_ioctl.h - style fix

Fixed style of block comments
Found using checkpatch

Signed-off-by: Derek Robson <robsonde@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/fbtft: Hush checkpatch.pl warning about unnecessary line continuations

Use a single string literal as the fmt argument passed to dev_err()
instead of multiple string literals split with an embedded backslash
character.

Signed-off-by: Alexander Kapshuk <alexander.kapshuk@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: fb_sh1106: use new helper helper fbtft_write_buf_dc

Make use of new helper fbtft_write_buf_dc.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: use helper fbtft_write_buf_dc in fbtft-bus.c

Make use of new helper fbtft_write_buf_dc.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: add helper fbtft_write_buf_dc

Factor out setting DC GPIO and writing buffer content to helper
fbtft_write_buf_dc.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: fb_sh1106: use own implementation of write_register

The default implementation of write_register keeps DC low for the first
byte only. SH1106 requires DC to be low for all bytes of a multi-byte
command. To deal with this limitation we currently use a separate call
to write_reg for each single command byte what is not really efficient.

Therefore override the default implementation of write_register with an
own one which keeps DC low for all bytes.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: improve a syslog message in fbtft_register_framebuffer

There are drivers like fb_sh1106 which allocate a txbuf less than 1 KiB.
In this case the message would include "0 KiB buffer memory" what
doesn't make sense and in general I doubt that there is any benefit in
informing the user about allocation of a very small buffer.

Therefore omit this message part if the buffer has less than 1 KiB.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

speakup: add unicode variant of /dev/softsynth

This adds /dev/softsynthu, along /dev/softsynth, which emits output in
UTF-8 encoding, thus allowing to support 16bit characters.  Most of the
code is shared, only the read function has to behave differently in
latin1 and in unicode mode.  Since Linux only supports 16bit characters,
we can just hardcode the UTF-8 encoding.

Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: Chris Brannon <chris@the-brannons.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

speakup: convert screen reading to 16bit characters

This adds 16bit character support to most of the screen reading by
extending characters to u16 throughout the code.

Non-latin1 characters are assumed to be alphabetic type for now.

non-latin1 vt_notifier_call-provided characters are not ignored any
more, and the 16bit character returned by get_char is not truncated any
more. For simplicity, speak_char still only supports latin1 characters.
Its direct mode however does support 16bit characters, so in practice
this will not be a limitation, non-latin1 languages will be handled by
the synthesizer. spelling words does not support direct mode yet, for
simplicity for now it will ignore 16bit characters.

For simplicity again, speakup messages are left in latin1 for now.

Some coding style is fixed along the way.

Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: Okash Khawaja <okash.khawaja@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

speakup: extend synth buffer to 16bit unicode characters

This extends the synth buffer slots to 16bit, so as to hold 16bit
unicode characters.

synth_buffer_getc and synth_buffer_peek now return 16bit characters.
Speech synthesizers which do not support characters beyond latin1 can
use the synth_buffer_skip_nonlatin1() helper to skip the non-latin1
characters before getting or peeking. All synthesizers are made to use
it for now.

This makes synth_buffer_add take a 16bit character. For simplicity for
now, synth_printf is left to using latin1 formats and strings.
synth_putwc, synth_putwc_s, synth_putws and synth_putws_s helpers are
however added to put 16bit characters and strings.

Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: Chris Brannon <chris@the-brannons.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: Remove unneeded header file

module.h contains a call to moduleparam.h making the call to it
redundant and useless.

@ includesmodule @
@@

 #include <linux/module.h>

@ depends on includesmodule @
@@

- #include <linux/moduleparam.h>

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: Remove useless cast.

Variable ip_addr is already declared as pointer to u8. Again explicit type
casting of ip_addr to u8, is not required. Hence this patch removes it
by using the following coccinelle script.

@@
type T;
T *ptr,p;
@@
(
- (T *)(&p)
+ &p
|
- (T *)ptr
+ ptr
|
- (T *)(ptr)
+ ptr
|
- (T)(p)
+ p
)

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: removes redundant 'continue' in while loop conditional blocks

Removing 6 continue statements from a while loop. The
continue statements are redundant here since control
already returns to the beginning of the loop upon exit
of any of the conditional blocks. Found using Coccinelle.

Signed-off-by: Tahia Khan <tahia.khan@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: Fixes camel-casing in wilc_gnrl_info_received

Fixes checkpatch warning by renaming pu8Buffer to buffer
and u32Length to length in wilc_gnrl_info_received.

Signed-off-by: Tahia Khan <tahia.khan@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: Fixes camel-casing in wilc_network_info_received

Fixes checkpatch warning by renaming pu8Buffer to buffer
and u32Length to length in wilc_network_info_received.

Signed-off-by: Tahia Khan <tahia.khan@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: Fixes camel-casing in wilc_scan_complete_received

Fixes checkpatch warning by renaming pu8Buffer to buffer
and u32Length to length in wilc_scan_complete_received.

Signed-off-by: Tahia Khan <tahia.khan@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging:wilc1000:wilc_spi: Added blank line after function and modified comment style

Added blank line after function and modified comment style.

Signed-off-by: Georgios Emmanouil <geo.emmnl@gmail.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging:wilc1000:wilc_spi: Fixed spelling error

Fixed spelling error. 'unkmown' to 'unknown'.

Signed-off-by: Georgios Emmanouil <geo.emmnl@gmail.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging:wilc1000:wilc_spi: Fixed comment style to the preferred kernel comment style

Fixed comment style to the preferred kernel comment style.

Signed-off-by: Georgios Emmanouil <geo.emmnl@gmail.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging:wilc1000:wilc_sdio: Modified comment style to preferred kernel comment style

Modified comment style to preferred kernel comment style.

Signed-off-by: Georgios Emmanouil <geo.emmnl@gmail.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging:wilc1000:host_interface: Integrated two 'if' statements to a single 'if' statement

Removed unnecessary 'if' statement and integrated the condition to the
previous 'if' statement.

Signed-off-by: Georgios Emmanouil <geo.emmnl@gmail.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging:wilc1000:host_interface: Removed unnecessary blank line

Removed unnecessary blank line.

Signed-off-by: Georgios Emmanouil <geo.emmnl@gmail.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: bcm2835-audio: Align with parenthesis

Fix the checkpatch issue:
Alignment should match open parenthesis

Signed-off-by: Sreya Mittal <sreyamittal5@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: kobjects: Delete comparison

Comparison to NULL is not necessary because 'if' statement tests if
'synth' is NULL anyway.

Delete comparison in 'if' structure to fix the checkpath.pl issue :
CHECK: Comparison to NULL could be written 'synth'.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: kobjects: Rearrange headers

Put a part from the headers of some functions in a new line to have less than
80 characters.

Fixed the checkpath.pl issue:
CHECK: line over 80 characters.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: kobjects: Align the code

Fixed the checkpath.pl issue:
CHECK: Aligment should match open paranthesis.

Added and deleted spaces and tabs to align the code.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: kobjects: Delete blank line

Delet the blank line after an open brace '{' to fix the checkpath.pl issue:
CHECK: Blank lines aren't necessary after an open brace '{'.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: kobjects: Delete space

Added a space after multiple casts to fix the checkpath.pl issue:
CHECK: No space is necessary after a cast.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: Clean up tests if NULL returned on failure

Some functions like kmalloc/kzalloc return NULL on failure.
When NULL represents failure, !x is commonly used.

This was done using Coccinelle:
@@
expression *e;
identifier l1;
@@

e = \(kmalloc\|kzalloc\|kcalloc\|devm_kzalloc\)(...);
...
- e == NULL
+ !e

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:speakup:i18n.c: Align open paranthesis

Fixed the following checkpath.pl issue:
CHECK: alignment should match open paranthesis.

Deleted a tab and added spaces to align open paranthesis.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: Fixed coding style errors and aligned indents

Fixed indentation to use tabs and aligned all the fields to same level.

Signed-off-by: Tuomo Rinne <tuomo.rinne@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: Add __printf format/arg verification, fix fallout

Using __printf allows the compiler to verify formats and arguments.
Use it and fix the single misuse found.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup:indentation should use tabs

Indentation should always use tabs and never spaces.

Signed-off-by: Arushi Singhal <arushisinghal19971997@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: emxx_udc: Add comment for spinlock_t definition.

Members of nbu2ss_udc structure can change device state, maintain
completion state and control driver. Also provide access to read and
write to register. Hence, exclusive access to nbu2ss_udc is required.
The lock variable of type spinlock_t guarantees the exclusive access
and protects it.

In this patch, comment is added for spinlock_t definition, to fix the
following checkpatch issue:

CHECK: spinlock_t definition without comment

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: xgifb: Improved coding style

New variable is added to make the code more readable.

Signed-off-by: Arushi Singhal <arushisinghal19971997@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192u: Constify ieee80211_qos_parameters structure

Declare ieee80211_qos_parameters structure constant it is only passed
as src parameter to the function memcpy. The fields of
def_qos_parameters structure are never modified and hence it can be
declared as const.

Coccinelle Script:

@r1 disable optional_qualifier@
identifier i;
position p;
@@

static struct ieee80211_qos_parameters i@p ={...};

@ok1@
identifier r1.i;
position p;
expression e1,e2;
@@
memcpy(e1,&i@p,e2)

@bad@
position p!={r1.p,ok1.p};
identifier r1.i;
@@
i@p

@depends on !bad disable optional_qualifier@
identifier r1.i;
@@
static
+const
struct ieee80211_qos_parameters i={...};

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: ni_atmio: make device_ids const

The `device_ids[]` passed to `MODULE_DEVICE_TABLE()` should be `const`.
When the "ni_atmio" driver is built-in, gcc warns about `device_ids`
being defined but ununsed.  Make it `const`.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: allow it to be built-in

There is no particular reason why comedi has to be built as kernel
modules.  Remove the `depends on m` from the Kconfig file to allow it to
be built-in.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: comedi: comedi_fops: Change comedi_num_legacy_minors type

Change to unsigned to allow removal of negative value check in
init section. Use smaller data type since the max possible
value currently is 48.

Signed-off-by: Cheah Kok Cheong <thrust73@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ks7010: fix spelling mistake

Comment contains misspelled work 'Adress'.

Correct spelling: Adress -> Address

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ks7010: fix checkpatch CONSTANT_COMPARISON

Checkpatch emits WARNING: Comparisons should place the constant on the
right side of the test.

Move comparison constant to the right side of the test.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ks7010: fix checkpatch whitespace warnings

Checkpatch emits various warnings/errors pointing to misplaced
spaces.

- trailing whitespace
- please, no spaces at the start of a line
- please, no space before tabs
- Unnecessary space before function pointer arguments
- unnecessary whitespace before a quoted newline
- code indent should use tabs where possible

Remove all undesirable whitespace.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ks7010: eap_packet: Format comments to fit in line

Split lines to have less than 80 characters.

Fix the checkpatch.pl warning:
WARNING: line over 80 characters.

Signed-off-by: Tamara Diaconita <diaconita.tamara@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: ks7010: Unnecessary parentheses removed and improved coding style.

Unnecessary parentheses are removed as reported by checkpatch.pl
to make coder nicer and to improve readability.
Also coding style is improved as it's often nicer to read if
 &(foo[0]) is converted to foo like:
 memcpy(&(ap->bssid[0]), &(ap_info->bssid[0]), ETH_ALEN);
 memcpy(ap->bssid, ap_info->bssid, ETH_ALEN);

Signed-off-by: Arushi Singhal <arushisinghal19971997@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging, android: remove lowmemory killer from the tree

Lowmemory killer is sitting in the staging tree since 2008 without any
serious interest for fixing issues brought up by the MM folks. The main
objection is that the implementation is basically broken by design:
- it hooks into slab shrinker API which is not suitable for this
  purpose. lowmem_count implementation just shows this nicely.
  There is no scaling based on the memory pressure and no
  feedback to the generic shrinker infrastructure.
  Moreover lowmem_scan is called way too often for the heavy
  work it performs.
- it is not reclaim context aware - no NUMA and/or memcg
  awareness.

As the code stands right now it just adds a maintenance overhead when
core MM changes have to update lowmemorykiller.c as well. It also seems
that the alternative LMK implementation will be solely in the userspace
so this code has no perspective it seems. The staging tree is supposed
to be for a code which needs to be put in shape before it can be merged
which is not the case here obviously.

Signed-off-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_arm: Disable ability to dump memory by default

vc04_services has an ioctl interface to dump arbitrary memory
to a custom debug log.  This is typically only needed by
diagnostic tools, and can potentially be a security issue
if the devtmpfs node doesn't have adequate permissions set.

Since the ability to dump memory still has debugging value,
create a new build configuration and disable the feature
by default.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_arm: Add compatibility wrappers for ioctls

This patch adds compatibility wrappers for the ioctls
exposed by vchiq/vc04_services.  The compat ioctls are
completely implemented on top of the native ioctls.  No
existing lines are modified.

While the ideal approach would be to cleanup the existing
code, this path is simplier and easier to review. While
it does have a small runtime performance penality vs
seperating the existing code into wrapper+worker functions,
the penality is small since only the metadata is copied
back onto the 32 bit user mode stack.

The on top of approach is the approach used by several
existing performance critical subsystems of Linux such
as the DRM 3D graphics subsystem.

Testing:

1. A 32 bit chroot was created on a RPI 3 and vchiq_test
was built for armhf.  The usual tests were run such as
vchiq_test -f 10 and vchiq_test -p.

2. This patch was copied onto the shipping version of
the Linux kernel used for the RPI and that kernel was
built for arm64. That kernel was used to boot Raspbian.
Many of the builtin features are now functional such
as the "hello_pi" examples, and minecraft_pi.

Signed-off-by: Michael Zoran <mzoran@crowfest.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>