The 'iorange' local variable is only used in the request_region()
call. Remove the variable and just use the board->io_range directly.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The comedi core initializes the dev->board_name before calling the
driver (*attach) function. There is not reason to reinitialize it
in the driver.

Use the dev->board_name when doing the request_resource() instead
of the open-coded string.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The printk's during the board attach are just added noise. Remove
them.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

For aesthetic reasons, move the boardinfo table near the struct
definition. Remove the unnecessary comments in the struct definition.
Reformat the boardinfo in C99 format and add some whitespace to
help readability.

Rename the boardinfo struct and the boardinfo table so they have
namespace associated with the driver.

Remove the ACL7225_SIZE and P16R16DIO_SIZE defines. They are only
used in the boardinfo table and the open-coded values provide the
information more clearly.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This driver adds support for the Television Encoder integrated
on i.MX53 SoCs (TVEv2).

Currently only the VGA output mode is supported, which only uses
the TVDAC to generate RGB levels. HSYNC and VSYNC signals are
routed directly from the IPU signal generator pins through IOMUXC.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Add WRG and WCLK opcodes to the display controller microcode,
and allow multi instruction codes.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This adds display interface timings for the Television Encoder
connected to IPU DI1 on i.MX53 and adds some configuration
glue code to select which IPU signal generators / pins are to
be used for HSYNC/VSYNC signals.

The default configuration is pin2/pin3 for hsync/vsync. The
VGA connector on i.MX53-QSB uses pin7/pin8, and the analog
part of the DVI-I connector on MBa53 connects to pin4/pin6.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

24-bit GBR order is needed on the display interface connected
to the Television Encoder (TVEv2) on i.MX53.
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

There are actually overflow bug and typo. And bug was never happened due to the
typo.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: NSamuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge tag 'iio-for-3.10d' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next

Jonathan writes:

	Fourth set of changes for iio in the 3.10 cycle.  This time just a fix an
	a cleanup.

	The fix is to the dummy driver for some inconsistent specification of
	parameters available vs parameters supplied.

	The cleanup is for some silly memcmp usage.

Correct the mismatch bewteen calibscale and calibbias, and export
the in_accel_calibscale sysfs entry

Updated to apply to current tree.
Signed-off-by: NJin Feng <jin88.feng@gmail.com>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

Calling memcmp() to check the value of the first byte in a string is overkill.
Just use buf[0] == '1' or buf[0] != '1' as appropriate.
Signed-off-by: NTony Luck <tony.luck@intel.com>
Acked-by: NLars-Peter Clausen <lars@metafoo.de>
Signed-off-by: NJonathan Cameron <jic23@kernel.org>

The dev->board_name is now initialized by the comedi core before calling
the(*attach) or (*auto_attach) function in a driver. As long as the driver
does no additional probing, it's no longer necessary initialize the board_name.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: NIan Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The dev->board_name is always initialized before calling the(*attach)
or (*auto_attach) function. It's no longer necessary to validate the
pointer in comedi_pci_enable().
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: NIan Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The dev->board_name is always initialized before calling the(*attach)
or (*auto_attach) function. The "BUG" check in comedi_device_postconfig()
is no longer necessary.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: NIan Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The comedi (*attach) and (*auto_attach) functions are used to attach
legacy and PnP type devices to the comedi subsystem. If we can set the
dev->board_name before doing the attach, the drivers will not have to
worry about doing it.

Drivers that do additional probing can still change the dev->board_name
if necessary.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: NIan Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The struct comedi_device pointer in this file, and the rest of the
comedi subsystem, is typically called 'dev'. Rename the local variable
'comedi_dev' in comedi_auto_config() for consistency.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Reviewed-by: NIan Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

While removing, the device needs to unregister
the sensor from thermal framework. Before
calling the call back the driver needs to check
if the call back is registered. This patch
fix the check by checking the right callback.
Signed-off-by: NEduardo Valentin <eduardo.valentin@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

When builtin compiled, there is a chance for this driver
be probed before cpufreq driver is up and running. In this
case, the cpucooling device can be wrong initialized.

Thus, this patch makes sure this driver is probed only
when cpufreq driver is ready. Whenever there is no
cpufreq driver registered, the probe will return -EPROBE_DEFER.
Tested-by: NJ Keerthy <j-keerthy@ti.com>
Signed-off-by: NEduardo Valentin <eduardo.valentin@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Extrapolation rules have been revisited.
Signed-off-by: NEduardo Valentin <eduardo.valentin@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Update the constants to the correct hotspot extrapolation
equation constants. OMAP4 constants are revisited and correct.
Signed-off-by: NEduardo Valentin <eduardo.valentin@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch defines and utilizes the extrapolation constants for OMAP4430.
Signed-off-by: NEduardo Valentin <eduardo.valentin@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

TC1/TC2 are not needed anymore, API has been upgraded.
This is a TODO left-over.
Signed-off-by: NEduardo Valentin <eduardo.valentin@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Min/Max cooling state are defined by registration helper
function, if no specific limits are passed. No need to change
this code.
Signed-off-by: NEduardo Valentin <eduardo.valentin@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch removes out of the TODO list those already completed.

Here is the status and why they are removed:
 on ti-bandgap.c:
-- Add support to hwmon: REMOVED, no need to have hwmon interfaces as
   the control is done via thermal framework.
-- Test every exposed API to userland: DONE, via thermal fw APIs
   By now, no specific API is exposed by this driver
-- Revisit data structures and simplify them: DONE, all
   unused fields are flagged for future removal.
-- Once SCM-core api settles, update this driver accordingly: DONE,
   the BG driver can exist without SCM driver by ioremapping its own
   registers and doing its own locking.

 on ti-thermal-common.c/ti-thermal.h:
-- Revisit trips and its definitions: DONE, for now there is no
   need to change current definition. Alert based policy will be add
   in future.
-- Revisit trending: DONE, OMAP5 history buffer support has been
   implemented. Devices without history buffer will use thermal fw
   trending capability.

on omap5-thermal.c
-- Add support for GPU cooling: REMOVED: this will not be part
   of this driver. Must be done in a separated cooling device.

 generally:
-- make checkpatch.pl and sparse happy: DONE, sparse remaining
   warning is not an issue.
-- update documentation: DONE, kernel-doc for ti-bandgap is now
   available.
Signed-off-by: NEduardo Valentin <eduardo.valentin@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fix checkpatch warning about seq_printf.
Signed-off-by: NValentin Ilie <valentin.ilie@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

  need set '\0' at the end. or cause issue.

    it is called by c4_ioctl in drivers/staging/cxt1e1/linux.c
    all things need be initialized, before provide them to user mode.
    so we can not use strlcpy instead of strncpy.

  code style:
    all contents of the file use 4 spaces instead of '\t',
    so this patch has to follow, now.
Signed-off-by: NChen Gang <gang.chen@asianux.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The 'thisboard' macro relies on a local variable having a specific
name and yields a pointer derived from that local variable.

Replace the macro with local variables and use the comedi_board()
helper to get the pointer.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The 'boardtype' macro relies on a local variable having a specific
name and yields a struct derived from that local variable.

Replace the macro with local variables and use the comedi_board()
helper to get the struct as a pointer. Use pointer access when
using the variable.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The 'boardtype' macro relies on a local variable having a specific
name and yields a struct derived from that local variable.

Replace the macro with local variables and use the comedi_board()
helper to get the struct as a pointer. Use pointer access when
using the variable.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The 'thisboard' macro relies on a local variable having a specific
name and yields a pointer derived from that local variable.

Replace the macro with local variables and use the comedi_board()
helper to get the pointer.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The 'thisboard' macro relies on a local variable having a specific
name and yields a pointer derived from that local variable.

Replace the macro with local variables and use the comedi_board()
helper to get the pointer.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This function is not used. Just remove it.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This pr_err() is just added noise, the user can't do anything about it.
Just remove it.

Since this is the only pr_level() message in the driver, also remove
the pr_fmt() macro.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Many of the static functions in this driver have names that could
potentially clash with external symbols (tty_ioctl, tty_write, etc.).

Rename all the static functions so they have a 'serial2002_' prefix
to avoid any issues.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

For aesthetic reasons, hookup the comedi_device (*open) and (*close)
functions after everything else in the attach has succeeded.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

A return value of >=0 indicates a successful attach to the comedi core.
Return 0 since that is more common in the kernel.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>