The sensor_template_group structures are never modified, so declare them as
const.

Done with the help of Coccinelle.
Signed-off-by: NJulia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

NCT6792 and NCT6793 are mostly register compatible to NCT6791, but
temperature sources are different and difficult to manage with a single
temperature label array. Introduce separate temperature label arrays
for those chips to reflect the differences.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Both NCT6791D and NCT6792D permit selection of a 'virtual' temperature
register as temperature source. The virtual temperature registers are
registers 0xea to 0xef in bank 0 and can be written by software.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

NCT6793D is register compatible with NCT6792D.

Also move nct6775_sio_names[] closer to enum kinds to simplify
adding new chips.
Tested-by: NGrazvydas Ignotas <notasas@gmail.com>
Reviewed-by: NJean Delvare <jdelvare@suse.de>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

The STEP_UP_TIME and STEP_DOWN_TIME registers are swapped for all chips but
NCT6775.
Reported-by: NGrazvydas Ignotas <notasas@gmail.com>
Reviewed-by: NJean Delvare <jdelvare@suse.de>
Cc: stable@vger.kernel.org # v3.10+
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

The following error message is seen when loading the nct6775 driver
with DEBUG_LOCK_ALLOC enabled.

BUG: key ffff88040b2f0030 not in .data!
------------[ cut here ]------------
WARNING: CPU: 0 PID: 186 at kernel/locking/lockdep.c:2988
				lockdep_init_map+0x469/0x630()
DEBUG_LOCKS_WARN_ON(1)

Caused by a missing call to sysfs_attr_init() when initializing
sysfs attributes.
Reported-by: NAlexey Orishko <alexey.orishko@gmail.com>
Reviewed-by: NJean Delvare <jdelvare@suse.de>
Cc: stable@vger.kernel.org # v3.12+
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Auxiliary fan monitoring is not enabled on ASRock Z77 Pro4-M
with BIOS version 2.00 if booted in UEFI Ultra-FastBoot mode.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

After a suspend/resume cycle it is not guaranteed that the hardware monitoring
device is still enabled. Ensure that this is the case after resume.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Get rid of #ifdef CONFIG_PM by using SIMPLE_DEV_PM_OPS and declaring suspend
and resume functions with __maybe_unused.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

checkpatch complains about
	WARNING: Missing a blank line after declarations

Add missing blank lines. Also reorder variables length-wise where appropriate
if a function header is touched anyway.
Reviewed-by: NJean Delvare <jdelvare@suse.de>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

NCT6792D is similar to NCT6791D. Only beep control and temperature
monitoring registers are different.
Reviewed-by: NJean Delvare <jdelvare@suse.de>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

A platform_driver does not need to set an owner, it will be populated by the
driver core.
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

num_attr_groups is only used in nct6775_probe(), make it to be local variable.
Signed-off-by: NAxel Lin <axel.lin@ingics.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

This driver also supports NCT6106D and NCT6791D, thus update module description
and Kconfig accordingly.
Signed-off-by: NAxel Lin <axel.lin@ingics.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Call platform_device_unregister() rather than platform_device_put() to
unregister successfully registered platform devices.
Signed-off-by: NAxel Lin <axel.lin@ingics.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Signed-off-by: NJean Delvare <khali@linux-fr.org>

After a suspend/resume cycle, the NCT6791 is back to its original BIOS
programming. In this state, HWMON IO access may be locked.
Re-enable it during resume.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Unlike other chips supported by this driver, the NCT6791 only has a single
set of registers to configure weighted fan control. Enable it only for the
single channel supporting it.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

The number of SMIOVT registers on NCT6779 and NCT6791 is limited to 2.
As result, the driver may not report some of the temperatures used
for fan control. This can result in some of the pwmX_temp_sel or
pwm2_weight_temp_sel attributes to wrongly return 0.
Fortunately, the chip has registers to monitor those temperatures.
Add them to the list of temperatures to report.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

We don't actually use "j" for anything.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

drivers/hwmon/nct6775.c:3866:1-3: WARNING: PTR_RET can be used

 Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR

Generated by: coccinelle/api/ptr_ret.cocci

CC: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

smatch complains about a potential out-of-bounds access to the
temp_offset array. That doesn't happen in practice, but it doesn't
hurt to add an explicit check either. This prevents potential problems
in the future (for example if the number of 'fixed' temperature
sensors is increased to add support for another chip).
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

smatch complains that we don't use the return value from find_temp_source().
Valid point, only find_temp_source() doesn't return a valid error code.
Have it return a valid error code and use it.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Hibernation uses its own set of callback functions, even if the code
is the same as the code used for suspend/restore.
Signed-off-by: NHarald Judt <h.judt@gmx.at>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Use the wrapper function for retrieving the platform data instead of
accessing dev->platform_data directly.
Signed-off-by: NJingoo Han <jg1.han@samsung.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Smatch complains that we have a array overflow:

	drivers/hwmon/nct6775.c:1456 nct6775_update_device()
		error: buffer overflow 'data->temp' 4 <= 4

Guenter Roeck says that the array should have been made larger in
7cbbd6aee6 (Add support for critical low/high temperature limits on
NCT6106).  This patch does that.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Plan going forward is to attach all device attributes to the hwmon device and no
longer to the platform device. With that change, accessing platform data outside
the probe function will be more difficult. To avoid the problem, change code
to no longer rely on it.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Signed-off-by: NDavid Bartley <andareed@gmail.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Static attribute allocation is large and very repetitive.
Allocate attributes and attribute groups dynamically instead.
This reduces the size of the driver source by more than 600 lines,
and object size by more than 20k (more than 30%).
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

NCT6775 does not support alarms for fans 4 and 5. Drop the attributes.

cc: stable@vger.kernel.org # 3.10+
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Driver displays wrong alarms for temperature attributes.

Turns out that temperature alarm bits are not fixed, but determined
by temperature source mapping. To fix the problem, walk through
the temperature sources to determine the correct alarm bit associated
with a given attribute.

Cc: stable@vger.kernel.org # 3.10+
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Overtemperature and hysteresis registers only exist for primary
temperature registers, not for alternates, so do not assign
those registers when initializing alternates.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

Add space around binary operators (CodingStyle, chapter 3.1).
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>

nct6775_sio_names should be a constant pointer to an array of
constant strings.
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>