staging: comedi: das16: move some functions to improve readability

For aesthetics, move some of the functions to make the code easier
to follow.

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: das16: tidy up das16_ai_insn_read()

Factor out the loop that waits for the ai conversion to complete.

Tidy up the function a bit.

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: das16: introduce das16_ai_{en, dis}able() helpers

Introduce a couple helper functions to enable and disable the
analog input interrupt/dma conversions.

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: das16: remove the register map comments

This is just extra cruft. 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: das16: remove unused static const variable

This variable is not used in the driver. 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: das16: remove unused DAS16_SIZE

This define is not used in the driver. 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: das16: add a 'can_burst' flag to the private data

If the boardinfo 'size' is > 0x400 it indicates that the board has
additional registers that allow bursting of analog input data.

To better indicate this add a flag to the private data.

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: das16: remove 'volatile' on private data variables

As indicated by checkpatch.pl, "WARNING: Use of volatile is usually
wrong: ...". The variables in the private data that are marked
volatile don't need to be. Remove the volatile.

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: das16: remove DEBUG_PRINT macro

This macro outputs some printk() debug info that is just added noise.
Remove it as well as the DEBUG stuff.

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: das16: remove reg_dump()

This function just prints some development debug information. There is
no reason to leave this in the final 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: das16: remove 'ai_unipolar' and 'ai_singleended'

These members in the private data are only used during the board attach.

Move the reading of the status register to the board attach and just
check the value where needed when initializing the analog input subdevice.

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: das16: remove unused struct definition

This struct is not used by the driver. 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: das16: tidy up the comedi_lrange tables

Cleanup the whitespace in the tables.

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: das16: cleanup analog output subdevice

All the boards supported by this driver that have analog outputs use
the same function to handle the (*insn_write) for the subdevice. They
all also have 12-bit resolution (maxdata = 0x0fff).

Remove the 'ao' and 'ao_nbits' members from the boardinfo and replace
them with a simple bit-field flag 'has_ao'.

Tidy up the subdevice init a bit as well as the (*insn_write) function.

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: das16: all supported board have digital outputs

All the boards supported by this driver have digital outputs and all
of then use the same function to handle the (*insn_bits) for the
subdevice.

Remove the 'do_' member from the boardinfo and always initialize the
digital output subdevice during the board attach.

Tidy up the subdevice init a bit as well as the (*insn_bits) function.
Remove SDF_READABLE from the subdev_flags. This is a pure digital output
subdevice and the outputs are not readable.

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: das16: all supported board have digital inputs

All the boards supported by this driver have digital inputs and all
of then use the same function to handle the (*insn_bits) for the
subdevice.

Remove the 'di' member from the boardinfo and always initialize the
digital input subdevice during the board attach.

Tidy up the subdevice init a bit as well as the (*insn_bits) function.
The (*insn_bits) function does not need to clear data[0].

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: das16: change boardinfo 'ai_nbits' to 'ai_maxdata'

The subdevice functions that used this boardinfo can use the subdevice
's->maxdata' value instead. Tidy up those functions.

Change the boardinfo so that the calculation is not needed during the
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: das16: all supported board have analog inputs

All the boards supported by this driver have analog inputs and all
of then use the same function to handle the (*insn_read) for the
subdevice.

Remove the 'ai' member from the boardinfo and always initalize the
analog input subdevice during the board attach.

Tidy up the subdevice init a bit.

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: das16: fix DMA init during board attach

DMA support is optional in this driver. To enable it, the user
passes the desired DMA channel during the board attach. A timer
is then used by the driver to periodically get the analog input
data from the DMA.

Since the DMA support is optional, modify the init code so that
it only fails if the DMA channel is unavailable or the DMA buffers
can not be allocated. Don't fail just because the user passed an
invalid DMA channel (only DMA channels 1 and 3 are valid).

Remove the printk() noise about the DMA. Change the printk() when
the request_dma() fails into a dev_err().

Move the timer initialization so it's only setup if DMA is available.
It's not needed otherwise. Also, only hook up the subdevice command
support functions if DMA is available. This allows removing a couple
sanity checks in the command support.

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: das16: remove DMA irq support

As noted in the driver, a timer is always used by this driver to handle
the DMA because samples could be dropped while waiting for the DMA done
interrupt.

Remove the irq setup code as well as 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: das16: "timer_mode" is always used for DMA

As noted in the driver, a timer is always used by this driver to handle the
DMA due to buggy DMA controllers and the lack of a hardware fifo in some of
the supported boards.

Remove all the non-timer mode specific code.

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: olpc_dcon: provide detailed FB_OLPC_DCON help section

The FB_OLPC_DCON help section is to short according to checkpatch.pl.
We want more information about the controller type, its task, its video
pipeline position and so on.

There are no style issues, remove checkpatch.pl TODO entry.

Signed-off-by: Jens Frederich <jfrederich@gmail.com>
Acked-by: Andres Salomon <dilinger@queued.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8192u/ieee80211: add missing single_release()

The debug file is opened with single_open(), but there's no
single_release().

Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8192e: add missing single_release()

The debug file is opened with single_open(), but there's no
single_release().

Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gdm724x: depend on CONFIG_TTY

The driver uses the tty layer, so explicitly say that, to prevent
randomconfig build errors.

Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: reformat documentation

Reformat the high-level documentation to human readable plain text by
removing HTML and other tags.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: delete CVS keyword markers

Delete CVS keyword markers.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: convert @return

Convert "@return" to "Returns: ".

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: convert @param

Convert "@param name" inside comments to "@name:".

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: delete duplicated documentation from .h file

Delete duplicated function call documentation from the .h file. These
functions are just internal anyway, and the .h file will be completely
removed in the future when the driver will be made a single self-contained
.c file.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: delete @INTERNAL

Delete "@INTERNAL" from comments.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: reformat long comments

Reformat long comments according to multi-line comment style.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: fix whitespace issues

Fix most of the whitespace issues reported by checkpatch:
- use tabs for indentation
- brace placement vs. newlines
- (foo_t*) -> (foo_t *)
- i=0 -> i = 0

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: delete __cvmx_usb_complete_to_string()

Delete a redundant function.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: replace cvmx_le16_to_cpu with le16_to_cpu

Replace cvmx_le16_to_cpu() with le16_to_cpu().

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: replace cvmx_wait_usec with udelay

Just use udelay().

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: replace long udelays with mdelays

Replace long udelays with mdelays.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: eliminate cvmx_likely/unlikely

Just use likely/unlikely.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: utilize Linux error codes

Use generic Linux error codes to replace cvmx_usb_status_t.

The conversion table:

Before After
cvmx_usb_status_t int
CVMX_USB_SUCCESS 0
CVMX_USB_INVALID_PARAM -EINVAL
CVMX_USB_NO_MEMORY -ENOMEM
CVMX_USB_BUSY -EBUSY
CVMX_USB_TIMEOUT -ETIMEDOUT
CVMX_USB_INCORRECT_MODE Deleted (was never used)

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: delete __cplusplus

Delete __cplusplus.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: delete redundant example code

Delete redundant example code found in comments. It's already there in
the driver proper.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: delete redundant debug flags

Delete redudant debug flags and dead code behind them.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: cvmx-usb: delete driver-specific tracing code

Delete call, parameter, and return value tracing.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: use correct board type for EdgeRouter Lite

Use CVMX_BOARD_TYPE_UBNT_E100 for EdgeRouter Lite.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon-usb: depend on CAVIUM_OCTEON_SOC

The USB driver should depend on SOC instead of CPU.

Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: LeakyBucket: Fixed a pointer asterisk placement issue

Fixed a pointer asterisk placement issue

Signed-off-by: Lilis Iskandar <veeableful@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: LeakyBucket: Fixed code indent issues

Fixed code indent issues

Signed-off-by: Lilis Iskandar <veeableful@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: LeakyBucket: Fixed C99 comments

Fixed C99 comments

Signed-off-by: Lilis Iskandar <veeableful@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: LeakyBucket: Fixed brace issues

Fixed brace issues

Signed-off-by: Lilis Iskandar <veeableful@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: bcm: LeakyBucket: Fixed spacing/tabing issues

This one fixes spacing/tabbing issues.

Signed-off-by: Lilis Iskandar <veeableful@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: nvec: re-enable the clock on resume

On resume the slave controller is reinitialized. The tegra i2c master
controller disables the clock at the end of the initialiation, propably
to save some power, and enables it again on each transfer. We don't
do this yet and also forgot to enable the clock on resume. Fix this
copy-paste error by not disabling the clock after initialization.

This didn't striked us yet because suspend/resume hasn't landed in mainline
yet, but will soon.

Signed-off-by: Marc Dietrich <marvin24@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gdm724x: Modified function name conflict set_endian

Conflict resolution for function name set_endian when building for powerpc
The name changed to gdm_set_endian

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Won Kang <wonkang@gctsemi.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: dt9812: remove duplicated include from dt9812.c

Remove duplicated include.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gdm724x: remove duplicated include from gdm_lte.c

Remove duplicated include.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: xillybus: remove duplicated include from xillybus_core.c

Remove duplicated include.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: Fix sparse warnings about use of obsolete initializer

This patch fixes a lot of sparse warnings about use of obsolete
struct and array initializer.

Signed-off-by: Emil Goode <emilgoode@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: Fix non-ANSI sparse warnings

This patch fixes a lot of non-ANSI sparse warnings by adding void to
parameterless functions.

Signed-off-by: Emil Goode <emilgoode@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge 3.11-rc3 into staging-next

We want these fixes here.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Linux 3.11-rc3

Merge tag 'pinctrl-for-v3.11-2' of git://git./linux/kernel/git/linusw/linux-pinctrl

Pull pin control fixes from Linus Walleij:
 - Driver fixes for AM33xx, SIRF and PFC pin controllers
 - Fix a compile warning from the pinctrl single-register driver
 - Fix a little nasty memory leak

* tag 'pinctrl-for-v3.11-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
  pinctrl: fix a memleak when freeing maps
  pinctrl: pinctrl-single: fix compile warning when no CONFIG_PM
  pinctrl: sh-pfc: fix SDHI0 VccQ regulator on sh73a0 with DT
  arm/dts: sirf: fix the pingroup name mismatch between drivers and dts
  pinctrl: sirf: add usp0_uart_nostreamctrl pin group for usp-uart without flowctrl
  pinctrl: sirf: fix the pin number and mux bit for usp0
  pinctrl: am33xx dt binding: correct include path

Merge tag 'trace-fixes-3.11-rc2' of git://git./linux/kernel/git/rostedt/linux-trace

Pull tracing fixes from Steven Rostedt:
 "Oleg is working on fixing a very tight race between opening a event
  file and deleting that event at the same time (both must be done as
  root).

  I also found a bug while testing Oleg's patches which has to do with a
  race with kprobes using the function tracer.

  There's also a deadlock fix that was introduced with the previous
  fixes"

* tag 'trace-fixes-3.11-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: Remove locking trace_types_lock from tracing_reset_all_online_cpus()
  ftrace: Add check for NULL regs if ops has SAVE_REGS set
  tracing: Kill trace_cpu struct/members
  tracing: Change tracing_fops/snapshot_fops to rely on tracing_get_cpu()
  tracing: Change tracing_entries_fops to rely on tracing_get_cpu()
  tracing: Change tracing_stats_fops to rely on tracing_get_cpu()
  tracing: Change tracing_buffers_fops to rely on tracing_get_cpu()
  tracing: Change tracing_pipe_fops() to rely on tracing_get_cpu()
  tracing: Introduce trace_create_cpu_file() and tracing_get_cpu()

Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "This is five bug fixes, two of which fix long standing problems
  causing crashes (sd and mvsas).  The remaining three are hung (isci
  race) or lost (qla2xxx, isci) devices"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  [SCSI] isci: fix breakage caused by >16byte CDB patch
  [SCSI] mvsas: Fix kernel panic on tile due to unaligned data access
  [SCSI] sd: fix crash when UA received on DIF enabled device
  [SCSI] qla2xxx: Properly set the tagging for commands.
  [SCSI] isci: Fix a race condition in the SSP task management path

vt6655/trivial: replace numeric with standard PM state macros

Use standard PM state macros PCI_Dx instead of numeric 0/1/2..

Signed-off-by: Yijing Wang <wangyijing@huawei.com>
Cc: Forest Bond <forest@alittletooquiet.net>
Cc: Jiri Kosina <trivial@kernel.org>
Cc: Devendra Naga <devendra.aaru@gmail.com>
Cc: Joe Perches <joe@perches.com>
Cc: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Cc: Bill Pemberton <wfp5p@virginia.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

zram: prevent data loss in error cases of function zram_bvec_write()

In function zram_bvec_write(), previous data at the index is
already freed by function zram_free_page().
When failed to compress or zs_malloc, there is no way to restore old data.
Therefore, free previous data when it's about to update.

Also, no need to check whether table is not empty outside of
function zram_free_page(), because the function properly checks inside.

Signed-off-by: Sunghan Suh <sunghan.suh@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: asus_oled: delete driver

No one has the hardware for it anymore, and there has not been any
development on it in a long time.

If someone shows up with the hardware, and wants to clean it up, this
can be easily reverted.

Reported-by: Lidza Louina <lidza.louina@gmail.com>
Cc: Jakub Schmidtke <sjakub@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: usbdux: consolidate usbduxsub_unlink_{in, out}urbs()

These functions are identical other than the which and how many urbs
are unlinked.

Consolidate the functions into one usbdux_unlink_urbs() function.

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: usbdux: move and rename the bulk transfer commands

For aesthetic reasons, move the defines used for the bulk transfer
commands out of the middle of the code and rename them so they are
a bit more readable.

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: usbdux: remove unused define

This define is not used, and it's actually incorrect. The usbdux device
only has 4 analog output channels.

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: usbdux: 'dac_commands' does not need to be kzalloc()'d

The 'dac_commands' buffer is used to pass the ao channel list from
usbdux_ao_cmd() to the urb callback, usbduxsub_ao_isoc_irq(). This
buffer does not need to be allocated.

Change it into a simple array of the correct size. Rename the variable
to 'ao_chanlist' to clarify what it actually is.

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: usbdux: remove unnecessary check in usbdux_ao_cmd()

This check is not necessary. The cmd->chanlist_len will always be
less than or equal to the subdevice len_chanlist that was setup
during the attach of the 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: usbdux: remove unnecessary check in usbdux_ai_cmd()

This check is not necessary. The cmd->chanlist_len will always be
less than or equal to the subdevice len_chanlist that was setup
during the attach of the 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: usbdux: simplify initializing the ao urb transfer_buffer

Remove the ugly casting of datap to the ao urb->transfer_buffer. The format
of the data in the buffer is simply.

  [0] = # of channels to update (s->async->cmd.chanlist.len)
  [1] = lsb of value for 1st channel to update
  [2] = msb of value for 1st channel to update
  [3] = 1st channel to update (depvriv->dac_commands[0])

If more than 1 channel is to be updated they are simply added to the buffer:

  [4] = lsb of value for 2st channel to update
  [5] = msb of value for 2st channel to update
  [6] = 1st channel to update (depvriv->dac_commands[1])

etc.

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: usbdux: cleanup the private data 'outBuffer'

This buffer is used to cache the values that are written to the
analog output channels. Currently it only caches the single writes
to the channels using the (*insn_write) callback. The async command
writes are not cached. The buffer is also being kzalloc'ed during
the attach of the driver to a size much larger that required.

Rename the CamelCase buffer and change it to an array in the private
data of the correct size to cache the analog output channel values.

Modify the analog output urb callback so it updates the cached values
with those used for the asynchronous command to allow readback after
the command completes.

The sanity check of the index to dac_commands[] (i.e. the 'chan' being
written) is not needed. The chanlist_len will always be less than the
number of channels.

Also, fix the dev_err() message so it uses the proper device pointer.

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: usbdux: rename private data variables

The usbdux and usbduxsigma drivers are _very_ similar. For aesthetic
reasons, rename the private data variables in this driver to match
the names in the usbduxsigma driver so we can start sharing the
common code.

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: usbdux: clarify bipolar ai data in usbduxsub_ai_isoc_irq()

Use the comedi_range_is_bipolar() helper instead of checking the
'range' index against a magic number.

Also, use the s->maxdata to calculate the value needed to munge the
value for bipolar data instead of the magic number.

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: usbdux: move usbdux_firmware_upload()

For aesthetics, move this function closer to the (*auto_attach).

Also, rename some of the defined constants that are used by the
firmware upload.

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: usbdux: remove some unused defines

These defines are not used by the driver. 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: usbdux: remove the usb endpoint defines

The endpoint defines are each only used in one place and don't help
clarify the code. Remove the defines and just open code the values.

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: usbdux: use the stop helpers in the detach

Use the stop helpers instead of duplicating the code in the detach.

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: usbdux: tidy up unlink and stop helpers

For aesthetic reasons, pass the comedi_device pointer to the unlink
helpers instead of the private data pointer.

All the unlink helpers simply call usb_kill_urb() to cancel any pending
transfer requests. The usb passed to usb_kill_urb() can be NULL so the
extra sanity check is not required.

The unlink helpers will always return success so just make them void
functions.

Since the stop helpers will also always return success, make them void
functions as well.

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: usbdux: fix usbdux_pwm_start()

Add the missing down/up of the semaphore to prevent other commands
from being issued to the usb device while the pwn is being stopped.

Rename the local variable used for the private data pointer to the
comedi "norm".

Use memset() to initialize the urb transfer buffer.

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: usbdux: fix usbdux_pwm_cancel()

Add the missing down/up of the semaphore to prevent other commands
from being issued to the usb device while the pwn is being stopped.

Rename the local variable used for the private data pointer to the
comedi "norm".

Make sure to check that usbdux_pwm_stop() was successful before
sending command to the usb device to stop the pwm.

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: usbdux: tidy up usbdux_pwm_stop()

For aesthetic reasons, pass the comedi_device pointer to this function
instead of the private data pointer. Rename the local variable used
for the private data pointer to the comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the 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: usbdux: fix usbdux_counter_write()

Comedi (*insn_write) operations are supposed to write insn->n values.
Fix this function to work like the core expects.

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the 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: usbdux: fix usbdux_counter_read()

Comedi (*insn_read) operations are supposed to read and return insn->n
values. Fix this function to work like the core expects.

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.

Tidy up the exit path using goto to ensure that the semaphore is
released.

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: usbdux: tidy up usbdux_dio_insn_bits()

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.

Tidy up the exit path using goto to ensure that the semaphore is
released.

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: usbdux: tidy up usbdux_dio_insn_config()

Tidy up this function a bit.

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: usbdux: tidy up usbdux_ao_cmd()

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.

Make sure an ao command is not already running and return -EBUSY.

Tidy up the exit path using goto to ensure that the semaphore is
released.

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: usbdux: tidy up usbdux_ao_inttrig()

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.

Tidy up the exit path using goto to ensure that the semaphore is
released.

Return -EBUSY instead if an ao command is already running.

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: usbdux: tidy up usbdux_ao_insn_write()

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.

Tidy up the exit path using goto to ensure that the semaphore is
released.

Return -EBUSY instead of 0 if the (*insn_write) cannot be done because
a command is running.

Tidy up the for() loop that writes the data. The dux_commands[1] and [4]
can be set outside the loop since they are constant. Use a local pointer
for dux_commands[2] to load the value to write. Only the last value needs
to be cached in the private data for read back.

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: usbdux: tidy up usbdux_ao_insn_read()

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the 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: usbdux: clarify bipolar ai data

Use the comedi_range_is_bipolar() helper instead of checking the
'range' index against a magic number.

Also, use the s->maxdata to calculate the value needed to munge the
value for bipolar data instead of the magic number.

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: usbdux: tidy up usbdux_ai_insn_read()

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.

Tidy up the exit path using goto to ensure that the semaphore is
released.

Return -EBUSY instead of 0 if the (*insn_read) cannot be done because
a command is running.

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: usbdux: tidy up usbdux_ai_cmd()

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.

Tidy up the exit path using goto to ensure that the semaphore is
released.

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: usbdux: tidy up usbdux_ai_inttrig()

Rename the local variable used for the private data pointer to the
comedi "norm".

Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.

Tidy up the exit path using goto to ensure that the semaphore is
released.

Return -EBUSY if an ai command is already running.

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: usbduxsigma: set *_cmd_running flags before submitting urbs

As pointed out by Ian Abbott, the *_cmd_running flags should be set
before submitting the urbs. There is a possible race condition where
an urb could complete and the flag is checked in the completion
routine before it's set.

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

staging: gdm724x: use GFP_ATOMIC under spin lock

A spin lock is taken here so we should use GFP_ATOMIC.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Acked-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: comedi: Kconfig: COMEDI_NI_ATMIO doesn't need to depend on ISAPNP

The COMEDI_NI_ATMIO configuration option causes the "ni_atmio" module to
be built, which handles various National Instruments ISA-PNP data
acquisition cards.  The configuration option currently depends on
ISAPNP.

The "ni_atmio" module is a "legacy" comedi driver with no
auto-configuration support and it doesn't register as a PNP card driver.
The only difference in initialization from the other comedi ISA card
drivers is that it will use the ISA-PNP functions to find, attach and
activate the card if no I/O base address has been specified by the user
in the COMEDI_DEVCONFIG ioctl.

The module compiles fine without the ISAPNP configuration option as the
<linux/isapnp.h> and <linux/pnp.h> headers provide dummy inline
functions if the ISAPNP and PNP options, respectively, are not enabled.

Remove the dependancy on the ISAPNP option since the module builds
without it and can be used without it.

(There is a very slim chance that comedi will support proper ISA PNP
drivers one day, in which case the ni_atmio driver could be converted to
use this support and not support manual configuration.  However, not all
the PnP IDs of boards supported by this driver are currently known to
us.)

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

staging: comedi: Kconfig: COMEDI_NI_ATMIO16D should not depend on ISAPNP

The COMEDI_NI_ATMIO16D configuration option causes the "ni_atmio16d"
module to be built, which handles the National Instruments AT-MIO-16 and
AT-MIO-16D cards.  The configuration option currently depends on ISAPNP,
but the module doesn't use any ISAPNP functions and the cards are legacy
ISA cards with base I/O port, IRQ and DMA channel configured by jumpers.
(The driver doesn't use DMA though.)

Remove the ISADNP dependancy from COMEDI_NI_ATMIO16D and fix up the help
text - the card names are missing a dash and they're not ISA-PNP.

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

staging: comedi: dt9812: Resolves sparse endian warnings.

Resolves warnings from the "sparse" checker of the form "warning: cast to
restricted __le??".

Signed-off-by: Shaun Laing <shaun@xresource.ca>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>